Datasets:

TaylorAI

/

pubmed_commercial

like 12

Follow

Taylor 12

PMC005xxxxxx/PMC5000599.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081201ijms-17-01201ArticleAnti-Diabetic Effect of Portulaca oleracea L. Polysaccharideandits Mechanism in Diabetic Rats Bai Yu 1*Zang Xueli 2Ma Jinshu 3Xu Guangyu 4Mihailidou Anastasia Susie Academic Editor1 Pharmaceutical College, Jilin Medical University, Jilin 132011, China2 Department of Pharmacy, Changchun Medical College, Changchun 130033, China; [email protected] Department of Pathology, China-Japan Union Hospital of Jilin University, Changchun 130033, China; [email protected] College of Pharmacy, Beihua University, Jilin 132013, China; [email protected]* Correspondence: [email protected]; Tel.: +86-432-6487-802925 7 2016 8 2016 17 8 120127 4 2016 19 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Diabetes mellitus (DM) is a metabolic syndrome caused by multiple genetic and environmental factors. Traditional Chinese medicine preparations have shown a comprehensive and function-regulating characteristic. Purslane (Portulaca oleracea L.) is an annual succulent herb. Currently, there have been some related reports on the treatment of diabetes with purslane. The current study was designed to separate and purify the polysaccharide, a systematic study of its physical and chemical properties, antioxidant activity, and anti-diabetic mechanism, in order to provide a theoretical basis for the development of drugs of purslane. A crude water soluble polysaccharide extracted from purslane was named CPOP (crude Portulaca oleracea L. polysaccharide). Effects of CPOP on bodyweight, glucose tolerance test (GTT), fasting blood glucose (FBG), fasting serum insulin (FINS), insulin sensitivity index (ISI), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), methane dicarboxylic aldehyde (MDA), and superoxygen dehydrogenises (SOD) were investigated. The results indicate that the oral administration of CPOP could significantly increase the body weight and significantly improve the glucose tolerance in diabetic rats. Meanwhile, CPOP could significantly reduce the FBG level, and elevate the FINS level and ISI value in diabetic rats. In addition, CPOP could significantly reduce TNF-α and IL-6 levels in diabetic rats; CPOP could also reduce MDA and SOD activities in the liver tissue of diabetic rats. These results suggest that the anti-diabetic effect of CPOP may be associated with its antioxidant and anti-inflammatory effects. purslanepolysaccharideanti-diabetic ==== Body 1. Introduction Type II diabetes mellitus, an endocrine and metabolic disease caused by the combined effects of polygenic and environmental factors, accounts for more than 90 percent of all diabetic patients [1]. The condition is characterized by a relative lack of insulin secretion and/or insulin resistance [2]. According to the World Health Organization, the number of diabetic patients across the world has reached 200 million, and is expected to exceed 336 million by year 2030 [3]. Oral hypoglycemic therapy with drugs, such as biguanides [4], thiazolidinediones (TZDs) [5], and glucosidase [6] inhibitors, is the primary therapeutic modality for type II diabetes. Despite their efficacy in maintaining glycemic control, oral hypoglycemic agents may not prevent the long-term complications of diabetes, such as nephropathy [7], and cardiovascular [8] disorders. Further, long-term use of these drugs is often associated with serious side effects, such as gastrointestinal disorders [9] associated with use of acarbose; granulocytopenia and hypoglycemia with glibenclamide; and lactic acidosis [10] associated with metformin therapy. In traditional Chinese medicine, diabetes is referred to as “Xiaokezheng” or “Xiaobing” and an abundant clinical experience of its treatment has accumulated over the years [11]. A total of 187 different traditional Chinese remedies for diabetes have been documented in the Compendium of Materia Medica, a famous Chinese medicine work. As compared to the modern chemical drugs, the traditional Chinese medicine is known to modulate physiological regulation that effectively prevents or delays the multi-systemic long-term complications of diabetes in addition to lowering blood sugar levels [12,13,14,15,16]. In addition, Chinese medicine typically has a lower propensity for severe toxicity and adverse reactions. Therefore, research into Chinese herbal remedies for diabetes has evoked considerable interest. Purslane is an annual succulent herb with succulent leaves that may grow prostrate or erect depending on light availability [17], which is distributed all over the world, and grows well in diverse geographical environments [18,19]. Purslane belongs to family Portulacaceae and is classified as a C4 plant, which is listed as one of the most useful medicinal plants and named “Global Panacea” by the World Health Organization [20]. It is a traditional Chinese herb now widely distributed throughout the world. The active ingredients include polysaccharides [21], fatty acids [22], flavonoids [22], coumarin [22], and alkaloids [23]. It is rich in antioxidant vitamins and omega-3 fatty acids [20] and can be used as a vegetable as well as for various curative purposes in health care, especially in preventing some cardiovascular diseases and maintaining a healthy immune system [24]. It is known to have antibacterial, anti-inflammatory, and antioxidant properties, and is known to regulate lipid and sugar metabolism in the body. The aqueous extract of Portulaca oleracea also prevents diabetic vascular inflammation, hyperglycemia, and diabetic endothelial dysfunction in type II diabetic db/db mice, suggesting its protective role against diabetes and related vascular complications [25]. The crude polysaccharide extract of this plant also lowers blood glucose and modulates the metabolism of blood lipids and glucose in alloxan-induced diabetic mice [26], whilst decreasing the levels of total cholesterol, triglycerides, and fasting blood glucose in type II diabetic mice [27]. Use of purslane polysaccharides for treatment of diabetes has not been rigorously evaluated. In this study, purslane polysaccharides was separated and purified, and a systematic study of its physical and chemical properties, antioxidant activity, as well as its anti-diabetic mechanism systematically examined. The objective is to provide a theoretical basis for the therapeutic use of purslane polysaccharides in the treatment of diabetes. 2. Results and Discussion 2.1. Polysaccharide Characterization The yield of CPOP (crude Portulaca oleracea L. polysaccharide) approximated 9.6% of the dry weight of raw material with 48.3% total carbohydrates (by phenol-sulfuric acid), 10.3% proteins, and 40.5% uronic acid. The molecular weight of CPOP was determined and calculated by high-performance size-exclusion chromatography. CPOP showed one main molecular weight distribution (7.3 × 103 Da) and two minor molecular weight distributions (11.9 × 103 and 9.3 × 104 Da) (Figure 1). The neutral monosaccharide composition was assessed by gas chromatography, which showed the presence of rhamnose, arabinose, xylose, mannose, glucose, and galactose in the ratio of 1:1.1:1.3:1.9:2.4:3.4:1 (Figure 2). 2.2. General Condition and Body Weight of Diabetic Rats The mental status of rats in the normal control (NC) group appeared normal and they showed good response and movement. Their fur was shiny and their body weights increased steadily. After the intraperitoneal injection of streptozotocin (STZ), rats in the model control group (MC) showed a tendency for increased intake of food and water, had an increased urine output and thin and soft stools. Further, these rats showed progressive loss of body weight, appeared depressed with sluggish response, lethargic movement, and untarnished fur. After oral administration of CPOP and glyburide, the mental state of diabetic rats appeared to improve significantly. The body weight of rats in the 200-CPOP and 400-CPOP groups increased significantly (p < 0.05 or p < 0.01) (Table 1). 2.3. Effects of CPOP (Crude Portulaca oleracea L. Polysaccharide) on Glucose Tolerance in Diabetic Rats The development of type II diabetes mellitus progresses from a state of normal glucose tolerance (NGT) to impaired glucose tolerance (IGT) and, finally, diabetes [28]. Thus, intervention during the state of IGTs appears to be a key to the prevention and treatment of type II diabetes mellitus. As shown in Table 2, after the overnight fast, blood sugar levels of rats that had been injected 20% sterile glucose solution in NC, MC, CPOP, and glyburide groups, reached their peak values at the first 30 min and, thereafter, gradually decreased; at the 120th min post-injection, the blood sugar levels in the NC group were restored to their normal basal level, while those in the MC, CPOP, and glyburide groups remained at a high level, indicating reduced glucose tolerance in rats in those groups. Compared with that in MC group, the fall in blood sugar levels in type II diabetic rats that had been injected with the glucose solution in the CPOP groups were significantly accelerated, suggesting a potential improvement in glucose tolerance induced by CPOP. Compared with those at the first 30 min, decreased rates of blood glucose levels at the 60th and 120th min were, respectively, 13.01% and 32.7% in rats in the MC group, while those in 100-CPOP, 200-CPOP, 400-CPOP, and glyburide groups at the 60th min were 21.67%, 27.10%, 32.01%, and 33.12%, respectively; those at the 120th min were 41.48%, 44.35%, 49.54%, and 52.39%, respectively. 2.4. Effects of CPOP on Fasting Blood Glucose (FBG), Fasting Serum Insulin (FINS) and Insulin Sensitivity Index (ISI) As illustrated in Figure 3 and Figure 4, compared with those in the NC group, Fasting Blood Glucose (FBG) values were significantly higher and Fasting Serum Insulin (FINS) levels significantly lower in the MC group (p < 0.01). Compared with those in the MC group, FBG levels were significantly lower, while FINS levels significantly higher in rats in CPOP groups (p < 0.05 or p < 0.01). As shown in Figure 5, compared with those in the MC group, Insulin Sensitivity Index (ISI) values were higher in the 100-CPOP, 200-CPOP and 400-CPOP groups, of which those in the 200-CPOP and 400-CPOP groups increased significantly (p < 0.05 or p < 0.01). 2.5. Effects of CPOP on Tumor Necrosis Factor-α (TNF-α) and Interleukin-6 (IL-6) Levels in Diabetic Rats In recent years, a large number of studies have implicated cytokine-mediated inflammation in the pathogenesis of type II diabetes mellitus [29,30]. Several inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), have been implicated in the causation of insulin resistance and appear to confer an increased risk of micro vascular complications in type II diabetes mellitus [31,32]. The TNF-α and IL-6 levels were significantly higher in the MC group, as compared to those in the NC group (p < 0.01). Further, as compared with those in MC group, TNF-α and IL-6 levels were significantly lower in the CPOP-treated diabetic rats on 28th day after oral administration of 100, 200, and 400 mg/kg CPOP (p < 0.05 or p < 0.01) (Figure 6 and Figure 7). 2.6. Effects of CPOP on Methane Dicarboxylic Aldehyde (MDA) Contents and Superoxygen Dehydrogenises (SOD) Activities in Diabetic Rats Although the molecular mechanisms of diabetes and its associated complications remain unclear, an increasing body of evidence appears to implicate the reactive oxygen species (ROS) in the pathogenesis of diabetes and its complications [33]. The strong correlation between diabetes and oxidative stress is well documented [34]. Superoxide dismutase, an endogenous antioxidase, serves to ameliorate the toxic effects of the oxygen radicals. Methane dicarboxylic aldehyde (MDA), an important lipid peroxide, is a sensitive index of metabolic levels of free radicals. The MDA content in rat liver tissue was significantly higher, while superoxygen dehydrogenises (SOD) activity was significantly lower in the MC group, as compared to those in the NC group (p < 0.01).On the 28th day after the intragastric administration of 100, 200, and 400 mg/kg CPOP, MDA content was lower, while the SOD activity was higher in the 100-CPOP, 200-CPOP, and 400-CPOP groups as compared to those in the MC group. Further, the difference in MDA content and SOD activity between the MC and CPOP groups was found to be statistically significant (p < 0.01) (Figure 8 and Figure 9). 2.7. Effects of CPOP on Protein Tyrosine Phosphatase 1B (PTP1B) Activities in Diabetic Rats Protein tyrosine phosphatase 1B (PTP1B), an ideal therapeutic target for type II diabetes, has become the intense pharmaceutical interest for treating type II diabetes over the past decade [35]. Considering that PTP1B directly dephosphorylates insulin receptors and the receptor substrate, thereby negatively regulating the insulin signaling pathway, it may also have a direct effect on its downstream pathways by affecting the activity or expression of insulin signal transduction molecules to inhibit insulin signal transduction, leading to insulin resistance [36,37]. We expected CPOP to be useful in the treatment of type II diabetes [38]. The active ingredient having PTP1B inhibition activity has been extracted from a variety of natural herbs and plants [39]. The results of this study found that the model group PTP1B levels were significantly higher (p < 0.01); compared with the model group, PTP1B content which of diabetic rats in the 100-CPOP, 200-CPOP, and 400-CPOP after 28 days decreased in a dose-dependent manner (p < 0.05 or p < 0.01) (Figure 10). These studies indicate that PTP1B inhibitors may be promising candidates for novel anti-diabetic drug development. 3. Materials and Methods 3.1. Materials and Chemicals Machixian (Portulaca oleracea L.), was purchased from Jilin Farmer’s Market and identified by Zhang Lihua at the College of Pharmacy, Beihua University, Jilin, China; Streptozotocin (STZ) was acquired from Sigma-Aldrich, Shanghai, China (80,082,038); glibenclamide was acquired from Zhejiang Nanyang Pharmaceutical Co., Ltd. (Hangzhou, China) (20121101); fasting glucose/fasting blood glucose (FBG) test kits were acquired from Sichuan Mike Technology Shares Limited Liability Company (Chengdu, China) (1,301,014); fasting insulin assay kit/fasting serum insulin (FINS) were acquired from Beijing Northern Institute of Biotechnology (Beijing, China) (130,219); tumor necrosis factor-α (TNF-α) detection kit was acquired from Beijing Northern Institute of Biotechnology (120,518); interleukin-6 (IL-6) detection kit was acquired from Wuhan Boster Biological Engineering Co., Ltd. (Wuhan, China) (20120316); Malondialdehyde (MDA) detection kit was acquired from Nanjing Jiancheng Institute of Biotechnology (Nanjing, China) (201,211,108); superoxide dismutase (SOD) detection kit was acquired from Nanjing Jiancheng Institute of Biotechnology (20,121,214). Trifluoroacetic acid (TFA), T-series dextrans (T-2000, T-70, T-40, T-20, and T-10) and the standard monosaccharides (rhamnose, fucose, arabinose, xylose, mannose, galactose, glucose, galacturonic acid) were acquired from Sigma Chemical Co. (St. Louis, MO, USA). All the other chemical reagents were of analytical grade. 3.2. Extraction of Polysaccharide of Portulaca oleracea L. The residual root and impurities of fresh Portulace oleracea L. (1000 g) were removed and the herb washed, dried at 60 °C, pulverized, and extracted twice with ethanol. The residues were extracted with hot water at 80 °C (1:20, w/v) three times, at 3 h each. The extracted solution was mixed and concentrated, and the associated proteins removed using the Sevag method [40]. The solution was concentrated and precipitated with four volumes of 95% (v/v) ethanol at 4 °C for 24 h in order to isolate the polysaccharide. The precipitate was washed with absolute ethanol, acetone and ether, respectively. Finally, the precipitate was suspended in water and lyophilized to yield the crude polysaccharide, referred to as crude Portulaca oleracea L. polysaccharide (CPOP). 3.3. Characterization of CPOP The total carbohydrate content in the CPOP fraction was determined by the phenol-sulfuric acid method and the glucose was taken as the standard [41]. The protein content was quantified according to the folin-phenol method, using Bovine Serum Albumin (BSA) as standard [42]. Total uronic acid contents were measured by m-hydroxydiphenyl method using galacturonic acid as the standard [43]. 3.4. Monosaccharide Composition Analysis The identification and quantification of the monosaccharides in CPOP was achieved by gas chromatography (GC). Polysaccharides (10 mg) were hydrolyzed with 2 M TFA at 100 °C for 2 h, followed by drying, reduction with NaBH4, and acetylation with Ac2O-NaOAc at 120 °C for 1 h [44]. The Ac2O was denatured with ice water, and the resulting alditol acetate extracted with chloroform and examined by GC. The following neutral monosaccharides were used as references: rhamnose, fucose, arabinose, xylose, mannose, galactose, and glucose. GC was performed on a Varian model 3300 instrument (Varian, Palo Alto, CA, USA) equipped with a DB-225 capillary column (Agilent, Beijing, China, 30 m × 0.25 mm inside diameter (i.d.)) and detected with a flame ionization detector (Agilent, 260 °C).The column temperature was increased from 150 to 200 °C at the rate of 4 °C/min and then held for 5 min. The quantification was carried out from the peaks area, using response factors from standard monosaccharide. 3.5. Molecular Weight Determination The molecular weight of CPOP was determined by high-performance size-exclusion chromatography (HPSEC) performed on a SHIMADZU HPLC system (SHIMADZU, Suzhou, China) equipped with one TSK-G3000PWXL column (TOSOH, Tokyo, Japan, 7.8 mm i.d. × 30.0 cm) and a SHIMADZU RID-10A detector. The mobile phase was 0.7% Na2SO4 and the flow rate was 0.5 mL/min. The sample was dissolved in the mobile phase and centrifuged (10,000 rpm; 35 min), and 20 µL of supernatant injected on each run. Dextran standards with different molecular weights (T-2000, T-70, T-40, T-20, and T-10) were used to calibrate the column and establish a standard curve. 3.6. Animals Special pathogen freeanimal (SPF) Sprague Dawley rats, each weighing 200 ± 20 g, were purchased from the Experimental Center at the Jilin University, Jilin, China. All experimental procedures were approved by the Animal Ethics Committee at the Beihua University, Jilin, China. The rats were housed in polypropylene cages under controlled conditions (22 ± 0.5 °C and a 12:12-hour light/dark cycle) and on a standard laboratory feed with free access to both food and water. The rats were acclimatized for one week before the start of the experiments. 3.7. Establishment of Rat Diabetic Model The rats were kept in an environment with a relative humidity of 40%–70% and 22 ± 1 °C temperature with free access to both water and food. After a 12 h fasting period, rats were administered 60 mg/kg newly-prepared STZ solution (dissolved in 0.1 M citrate buffer, pH 4.2) via intra-peritoneal injection. Those in the normal control group (NC) were injected the same volume of citrate buffer. At 72 h after STZ injection, blood samples were collected by tail bleeding. The fasting blood glucose levels were measured. Rats with fasting blood glucose levels of ≥11.1 mM were selected as the rat diabetic model. 3.8. Grouping and Administration The diabetic rats were randomly divided into control group, model group (MC), 100-CPOP group (100 mg/kg per body weight), 200-CPOP group (200 mg/kg per body weight), 400-CPOP group (400 mg/kg per body weight), and glyburide group (25 mg/kg per body weight). On day 2 after the modeling, oral administration of the above described drugs was started once a day for 28 days. Rats in the 100-CPOP, 200-CPOP, 400-CPOP, and glyburide groups were administered the corresponding agents, and those in the control group (NC) and the model group (MC) were administered identical volumes of saline. Pre- and post- treatment general condition of the rats and changes in body weight were documented. 3.9. Glucose Tolerance Test On day 21 of treatment, the rats were fasted for 12 h, and injected 20% sterile glucose solution (2 g/kg) intraperitoneally. The blood was obtained by tail bleeding and the serum separated. Blood glucose levels at 0, 30, 60, and 120 min after glucose injection were measured with the glucose oxidase method. 3.10. Detection and Calculation of FBG, FINS and ISI On day 28 of treatment, rats were anesthetized by intraperitoneal injection of 100 mg/kg urethane and blood specimens obtained from the aorta abdominalis. Specimens were centrifuged (3000 rpm, 10 min) to obtain the serum, which was kept in tubes and frozen for use. FBG levels were measured by the glucose oxidase method, FINS levels were determined by radioimmunoassay, and ISI values were calculated based on the following formula: ISI = ln (FBG × FINS)−1 3.11. Detection of Serum Interleukin-6 and Tumor Necrosis Factor-α Levels Enzyme-linked immuneosorbent assay (ELISA) for serum levels of IL-6 and TNF-α levels were performed as per the instructions accompanying the ELISA kits. 3.12. Methanedicarboxylic Aldehyde Content and Super Oxygen Dehydrogenase Activity in Liver On day 28 after administration, the rats were sacrificed after obtaining blood samples via the aorta abdominalis. The liver tissue was separated and an appropriate amount was dissolved in pre-cooled saline to prepare a homogenate containing 10% liver tissue. The homogenate was centrifuged and supernatant obtained. MDA content and SOD activity was measured with xanthine oxidase and thiobarbituric acid (TBA) condensation methods, respectively, as per the kit manufacturer’s instructions. 3.13. Detecting the Expression Levels of PTP1B in Liver by Western Blot Rat liver tissue was isolated, then frozen at −80 °C refrigerator and cell lysates were added. Protein concentration was determined by using BCA Protein Assay Kit (Thermo Scientific, Waltham, MA, USA). After electrophoresis, 60 mg proteins were transferred to polyvinylidene difluoride (PVDF) membranes. PTP1B antibody was added to the samples and then frozen overnight at 4 °C. The following day, the membranes were probed with a secondary antibody for 2 h. The bands were detected using enhanced chemiluminescence reagents. Protein bands were visualized by autoradiography and the intensities were analyzed by Image J (National Institutes of Health, v2.1.4.7, Bethesda, MD, USA, 2011). 3.14. Statistical Analysis All data analyses were performed with SPSS 17.0 software (SPSS, version 17.0, New York, NY, USA, 2010). Data is expressed as mean ± standard deviation. Inter-group differences were assessed by t-test; p < 0.05 or p < 0.01 was considered indicative of a statistically significant difference. 3.15. Ethical Consideration The study was approved by the Ethics Committee of School of Basic Medical Sciences, Jilin University, Jilin, China. 4. Conclusions A water-soluble polysaccharide named CPOP was extracted from purslane. CPOP contains 48.3% total carbohydrates (by phenol-sulfuric acid), 10.3% proteins, 40.5% uronic acid, and is mainly composed of rhamnose, arabinose, xylose, mannose, glucose, and galactose in the ratio of 1:1.1:1.3:1.9:2.4:3.4. In this study, intragastric administration of CPOP was associated with a significant increase in the body weight and a significant improvement in glucose tolerance in the diabetic rats. CPOP appeared to significantly reduce FBG levels and elevate FINS and ISI levels in diabetic rats. In addition, CPOP appeared to significantly reduce TNF-α and IL-6 levels in diabetic rats, suggesting an anti-inflammatory effect. CPOP treatment was also associated with reduced MDA content and increased SOD activity in the liver tissue of diabetic rats, indicating its antioxidant properties. These results suggest that anti-diabetic effect of CPOP may be mediated via its antioxidant and anti-inflammatory effects. Acknowledgments This work was supported by the National Natural Science Foundation of China (81401712). The science and technology research project of Education Department in Jilin province “The Twelfth Five Year Plan” science and technology research project (2014-505) and (2014-546). Author Contributions Yu Bai and Guangyu Xu conceived and designed the experiments; Yu Bai, Xueli Zang and Jinshu Ma performed the experiments; Yu Bai and Guangyu Xu analyzed the data; Jinshu Ma contributed reagents, materials and analysis tools; Yu Bai wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 High-performance size-exclusion chromatography (HPSEC) chromatogram of CPOP (crude Portulaca oleracea L. polysaccharide). 1: 9.3 × 104 Da, 2: 11.9 × 103 Da, 3: 7.3 × 103 Da. Figure 2 (a) Gas chromatography (GC) chromatogram of mixed standard monosaccharides derivatives; (b) GC chromatogram of hydrogen derivative from CPOP. Peak identification: 1: rhamnose, 2: fructose, 3: arabinose, 4: xylose, 5: mannose, 6: galactose, 7: glucose, 8: myo-inositol, 9: glucuronic acid, and 10: galacturonic acid. Figure 3 Effect of CPOP on Fasting Blood Glucose (FBG) in Streptozotocin (STZ)-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 4 Effect of CPOP on Fasting Serum Insulin (FINS) in STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 5 Effect of CPOP on Insulin Sensitivity Index (ISI) in STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 6 Effect of CPOP on tumor necrosis factor-α (TNF-α) levels in STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 7 Effect of CPOP on interleukin-6 (IL-6) levels in STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 8 Effect of CPOP on methane dicarboxylic aldehyde (MDA) contents in liver tissues of STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 9 Effects of CPOP on superoxygen dehydrogenises (SOD) activities in liver tissues of STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. Figure 10 Effects of CPOP on protein tyrosine phosphatase 1B (PTP1B) levels in liver tissues of STZ-induced diabetic rats. ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. GAPDH: glyceraldehyde-3-phosphate dehydrogenase. ijms-17-01201-t001_Table 1Table 1 Mean body weight of rats by study group (x¯ ± SD, N = 8). Group Dose (mg/kg) Baseline Mean Body Weight (g) Mean Body Weight on Day 14 (g) Mean Body Weight on Day 28 (g) NC – 208.31 ± 16.28 251.08 ± 26.21 299.23 ± 31.27 MC – 209.48 ± 14.85 187.92 ± 15.17 ** 150.42 ± 13.11 ** 100-CPOP 100 207.75 ± 13.61 219.47 ± 22.68 247.36 ± 23.14 200-CPOP 200 209.62 ± 16.06 227.71 ± 24.15 # 251.03 ± 22.47 ## 400-CPOP 400 208.83 ± 18.27 228.45 ± 19.21 ## 268.85 ± 25.98 ## Glyburide 25 209.29 ± 16.22 236.25 ± 20.52 ## 275.21 ± 24.35 ## ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). Data expressed as mean ± standard deviation (SD), NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP (crude Portulaca oleracea L. polysaccharide) is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. ijms-17-01201-t002_Table 2Table 2 Results of glucose tolerance test by study group (x¯ ± SD, N = 8). Group Dose (mg/kg) Mean Fasting Blood Sugar Level (mM) 0 min 30 min 60 min 120 min NC – 4.32 ± 0.88 13.65 ± 1.24 8.94 ± 1.47 4.89 ± 0.95 MC – 16.04 ± 2.15 ** 48.65 ± 3.29 ** 42.32 ± 4.18 ** 32.74 ± 2.82 ** 100-CPOP 100 14.42 ± 2.07 # 42.82 ± 2.81 ## 33.54 ± 3.23 ## 25.06 ± 2.67 ## 200-CPOP 200 12.36 ± 1.52 ## 33.98 ± 3.61 ## 24.77 ± 2.75 ## 18.91 ± 2.03 ## 400-CPOP 400 10.49 ± 1.84 ## 28.46 ± 3.05 ## 19.35 ± 2.36 ## 14.36 ± 1.42 ## Glyburide 25 8.37 ± 1.26 ## 22.16 ± 2.19 ## 14.82 ± 1.63 ## 10.55 ± 1.34 ## ** p < 0.01 vs. NC group (N = 8); # p < 0.05, ## p < 0.01 vs. MC group (N = 8). Data expressed as mean ± standard deviation (SD), NC is the normal control group; MC is the model control group; 100-CPOP is the 100 mg/kg per body weight CPOP group; 200-CPOP is the 200 mg/kg per body weight CPOP group; 400-CPOP is the 400 mg/kg per body weight CPOP group. ==== Refs References 1. Westerhaus B. Gosmanov A.R. Umpierrez G.E. Diabetes prevention: Can insulin secretagogues do the job? Prim. Care Diabetes 2011 5 73 80 10.1016/j.pcd.2010.11.002 21185798 2. Defronzo R.A. From the triumvirate to the ominous octet: A new paradigm for the treatment of type 2 diabetes mellitus Diabetes 2009 58 773 795 10.2337/db09-9028 19336687 3. Wild S. Roglic G. Green A. Sicree R. King H. Global prevalence of diabetes: Estimates for the year 2000 and projections for 2030 Diabetes Care 2004 27 1047 1053 10.2337/diacare.27.5.1047 15111519 4. Kitabchi A.E. Temprosa M. Knowler W.C. Role of insulin secretion and sensitivity in the evolution of type 2 diabetes in the diabetes prevention program: Effects of life style intervention and metformin Diabetes 2005 54 2404 2414 16046308 5. Inzucchi S.E. Oral antihyperglycemic therapy for type 2 diabetes: Scientific review JAMA 2002 287 360 372 10.1001/jama.287.3.360 11790216 6. Xiang A.H. Peters R.K. Kjos S.L. Goico J. Ochoa C. Marroquin A. Tan S. Hodis H.N. Azen S.P. Buchanan T.A. Pharmacological treatment of insulin resistance at two different stages in the evolution of type 2 diabetes: Impact on glucose tolerance and β-cell function J. Clin. Endocrinol. Metab. 2004 89 2846 2851 10.1210/jc.2003-032044 15181067 7. Mann J.F. Schmieder R.E. McQueen M. Dyal L. Schumacher H. Pogue J. Wang X. Maggioni A. Budaj A. Chaithiraphan S. Renal outcomes with telmisartan, ramipril, or both, in people at high vascular risk (the ONTARGET study): A multicentre, randomised, double-blind, controlled trial Lancet 2008 372 547 553 10.1016/S0140-6736(08)61236-2 18707986 8. Nissen S.E. Wolsk I.K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes N. Engl. J. Med. 2007 356 2457 2471 10.1056/NEJMoa072761 17517853 9. Scarpello J.H. Hodgson E. Howlett H.C. Effect of metformin on bile salt circulation and intestinal motility in type 2 diabetes mellitus Diabet. Med. 1998 15 651 656 10.1002/(SICI)1096-9136(199808)15:8<651::AID-DIA628>3.0.CO;2-A 9702467 10. Owen M.R. Doran E. Halestrap A.P. Evidence that metformin exerts its anti-diabetic effects through inhibition of complex 1 of the mitochondrial respiratory chain Biochem. J. 2000 348 607 614 10.1042/bj3480607 10839993 11. Zhen Z.Y. Wei L. Feng Z. Deciphering the therapeutic mechanisms of Xiao-Ke-An in treatment of type 2 diabetes in mice by a Fangjiomics approach Acta Pharmacol. Sin. 2015 36 699 707 25960133 12. Li W.L. Zheng H.C. Bukuru J. de Kimpe N. Natural medicines used in the traditional Chinese medical system for therapy of diabetes mellitus J. Ethnopharmacol. 2004 92 1 21 10.1016/j.jep.2003.12.031 15099842 13. Liu J.P. Zhang M. Wang W.Y. Grimsgaard S. Chinese herbal medicines for type 2 diabetes mellitus Cochrane Database Syst. Rev. 2004 3 CD003642 15266492 14. Xie W. Xing D. Sun H. Wang W. Ding Y. Du L. The effects of Ananas comosus L. leaves on diabetic-dyslipidemic rats induced by alloxan and a high-fat/high-cholesterol diet Am. J. Chin. Med. 2005 33 95 105 10.1142/S0192415X05002692 15844837 15. Xie W. Zhang Y. Wang N. Zhou H. Du L. Ma X. Shi X. Cai G. Novel effects of macrostemonoside A, a compound from Allium macrostemon Bung, on hyperglycemia, hyperlipidemia, and visceral obesity in high-fat diet-fed C57BL/6 mice Eur. J. Pharmacol. 2008 599 159 165 10.1016/j.ejphar.2008.09.042 18930725 16. Xie W.D. Zhao Y.N. Du L.J. Cai G.P. Scorpion in combination with Gypsum: Novel antidiabetic activities in streptozotocin-induced diabetic mice by up-regulating pancreatic PPARγ and PDX-1 expressions Evid. Based Complement. Altern. Med. 2011 2011 683561 10.1093/ecam/neq031 21799688 17. Chauhan B.S. Johnson D.E. Seed germination ecology of Portulaca oleracea L.: An important weed of rice and upland crops Ann. Appl. Biol. 2009 155 61 69 10.1111/j.1744-7348.2009.00320.x 18. Lee J. Chauhan B.S. Johnson D.E. Germination of fresh horse purslane (Trianthema portulacastrum) seeds in response to different environmental factors Weed Sci. 2011 59 495 499 10.1614/WS-D-11-00002.1 19. D’Andrea R.M. Andreo C.S. Lara M.V. Deciphering the mechanisms involved in Portulaca oleracea (C4) response to drought: Metabolic changes including crassulacean acid-like metabolism induction and reversal upon re-watering Physiol. Plant. 2014 152 414 430 10.1111/ppl.12194 24655215 20. Uddin M.K. Juraimi A.S. Hossain M.S. Purslane weed (Portulaca oleracea ): A prospective plant source of nutrition, omega-3 fatty acid, and antioxidant attributes Sci. World J. 2014 2014 10.1155/2014/951019 24683365 21. Zhao R. Zhang T. Zhao H. Effects of Portulaca oleracea L. polysaccharides on phenotypic and functional maturation of murine bone marrow derived dendritic cells Nutr. Cancer 2015 67 987 993 10.1080/01635581.2015.1060352 26219397 22. Zhou Y.X. Xin H.L. Rahman K. Wang S.-J. Peng C. Zhang H. Portulaca oleracea L.: A review of phytochemistry and pharmacological effects BioMed Res. Int. 2015 2015 10.1155/2015/925631 25692148 23. Meng Y. Ying Z. Xiang Z. Hao D. Zhang W. Zheng Y. Gao Y. Ying X. The anti-inflammation and pharmacokinetics of a novel alkaloid from Portulaca oleracea L. J. Pharm. Pharmacol. 2016 68 397 405 10.1111/jphp.12526 26888212 24. Rahdari P. Hosseini S.M. Tavakoli S. The studying effect of drought stress on germination, proline, sugar, lipid, protein and chlorophyll content in purslane (Portulaca oleracea L.) leaves J. Med. Plants Res. 2012 6 1539 1547 25. Lee A.S. Lee Y.J. Lee S.M. Yoon J.J. Kim J.S. Kang D.G. Lee H.S. Portulaca oleracea ameliorates diabetic vascular inflammation and endothelial dysfunction in db /db mice Evid. Based Complement. Altern. Med. 2012 2012 10.1155/2012/741824 22474522 26. Gong F. Li F. Zhang L. Li J. Zhang Z. Wang G. Hypoglycemic effects of crude polysaccharide from Purslane Int. J. Mol. Sci. 2009 10 880 888 10.3390/ijms10030880 19399226 27. El-Sayed M.I.K. Effects of Portulaca oleracea L. seeds in treatment of type-2 diabetes mellitus patients as adjunctive and alternative therapy J. Ethnopharmacol. 2011 137 643 651 10.1016/j.jep.2011.06.020 21718775 28. Winzell M.S. Ahren B. The high-fat diet-fed mouse: A model for studying mechanisms and treatment of impaired glucose tolerance and type 2 diabetes Diabetes 2004 53 215 219 10.2337/diabetes.53.suppl_3.S215 29. Leinonen E. Hurt-Camejo E. Wiklund O. Hultén L.M. Hiukka A. Taskinen M.R. Insulin resistance and adiposity correlate with acute-phase reaction and soluble cell adhesion molecules in type 2 diabetes Atherosclerosis 2003 166 387 394 10.1016/S0021-9150(02)00371-4 12535753 30. Spranger J. Kroke A. Mohlig M. Hoffmann K. Bergmann M.M. Ristow M. Boeing H. Pfeiffer A.F. Inflammatory cytokines and the risk to develop type 2 diabetes Diabetes 2003 52 812 817 10.2337/diabetes.52.3.812 12606524 31. Ruan H. Hacohen N. Golub T.R. van Parijs L. Lodish H.F. Tumor necrosis factor-α suppresses adipocytes-specific genes and activates expression of preadipocyte genes in 3T3-L1 adipocytes: Nuclear factor-κB activation by TNF-α is obligatory Diabetes 2002 51 1319 1336 10.2337/diabetes.51.5.1319 11978627 32. Naguib G. Al-Mashat H. Desta T. Graves D.T. Diabetes prolongs the inflammatory response to a bacterial stimulus through cytokine dysregulation J. Investig. Dermatol. 2004 123 87 92 10.1111/j.0022-202X.2004.22711.x 15191547 33. Lee H.B. Yu M.R. Yang Y.Q. Reactive oxygen species-regulated signaling pathways in diabetic nephropathy J. Am. Soc. Nephrol. 2003 14 241 245 10.1097/01.ASN.0000077410.66390.0F 34. Susztak K. Raff A.C. Schiffer M. Glucose-induced reactive oxygen species cause apoptosis of podocytes and podocyte depletion at the onset of diabetic nephropathy Diabetes 2006 55 225 233 10.2337/diabetes.55.01.06.db05-0894 16380497 35. Koren S. Fantus I.G. Inhibition of the protein tyrosine phosphatase PTP1B: Potential therapy for obesity, insulin resistance and type-2 diabetes mellitus Best Pract. Res. Clin. Endocrinol. Metab. 2007 21 621 640 10.1016/j.beem.2007.08.004 18054739 36. Wang N. Zhang D. Mao X. Zou F. Jin H. Ouyang J. Astragalus polysaccharides decreased the expression of PTP1B through relieving ER stress induced activation of ATF6 in a rat model of type 2 diabetes Mol. Cell. Endocrinol. 2009 307 89 98 10.1016/j.mce.2009.03.001 19524131 37. Tamrakar A.K. Maurya C.K. Rai A.K. PTP1B inhibitors for type 2 diabetes treatment: A patent review (2011–2014) Expert Opin. Ther. Pat. 2014 24 1101 1115 10.1517/13543776.2014.947268 25120222 38. Pan D. Wang L. Hu B. Zhou P. Structural characterization and bioactivity evaluation of an acidic proteoglycan extract from Ganoderma lucidum fruiting bodies for PTP1B inhibition and anti-diabetes Biopolymers 2014 101 613 623 10.1002/bip.22426 24127303 39. Davis J.A. Sharma S. Mittra S. Sujatha S. Kanaujia A. Shukla G. Katiyar C. Lakshmi B.S. Bansal V.S. Bhatnagar P.K. Antihyperglycemic effect of Annona squamosa hexane extract in type 2 diabetes animal model: PTP1B inhibition, a possible mechanism of action? Indian J. Pharmacol. 2012 44 326 332 10.4103/0253-7613.96304 22701240 40. Sun Y.X. Wang S.S. Li T.B. Li X. Jiao L. Zhang L. Purification, structure and immunobiological activity of a new water-soluble polysaccharide from the mycelium of Polyporus albicans (Imaz.) teng Bioresour. Technol. 2008 99 900 904 10.1016/j.biortech.2007.01.029 17368889 41. Dubois M. Gilles K.A. Hamilton J.K. Gilles K.A. Hamilton J.K. Rebers P.A. Smith F. Colorimetric method for determination of sugars and related substances Anal. Chem. 1956 28 350 356 10.1021/ac60111a017 42. Lowry O.H. Farr A.L. Rosebrough N.J. Farr A.L. Randall R.J. Protein measurement with the Folin-Phenol reagents J. Biol. Chem. 1951 193 265 275 14907713 43. Filisetti-Cozzi T.M. Carpita N.C. Measurement of uronic acids without interference from neutral sugars Anal. Biochem. 1991 197 157 162 10.1016/0003-2697(91)90372-Z 1952059 44. Lehrfeld J. Simultaneous gas-liquid chromatographic determination of aldonic acids and aldoses Anal. Chem. 1985 57 346 348 10.1021/ac00279a077

PMC005xxxxxx/PMC5000600.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081202ijms-17-01202ReviewRole of Lipids in the Onset, Progression and Treatment of Periodontal Disease. A Systematic Review of Studies in Humans Varela-López Alfonso 1Giampieri Francesca 2Bullón Pedro 3Battino Maurizio 2Quiles José L. 1*Verardo Vito Academic Editor1 Department of Physiology, Institute of Nutrition and Food Technology “Jose Mataix”, University of Granada, Biomedical Research Center, Avda. Conocimiento s/n, 18100 Armilla, Spain; [email protected] Dipartimento di Scienze Cliniche Specialistiche ed Odontostomatologiche (DISCO)-Sez. Biochimica, Facoltà di Medicina, Università Politecnica delle Marche, 60131 Ancona, Italy; [email protected] (F.G.); [email protected] (M.B.)3 Department of Stomalogy, Dental School, University of Sevilla, C/Avicena s/n, 41009 Sevilla, Spain; [email protected]* Correspondence: [email protected]; Tel.: +34-958-241-000 (ext. 20316)25 7 2016 8 2016 17 8 120201 6 2016 20 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The risk of different oral problems (root caries, tooth mobility, and tooth loss) can be increased by the presence of periodontal disease, which has also been associated with a growing list of systemic diseases. The presence of some bacteria is the primary etiology of this disease; a susceptible host is also necessary for disease initiation. In this respect, the progression of periodontal disease and healing of the periodontal tissues can be modulated by nutritional status. To clarify the role of lipids in the establishment, progression, and/or treatment of this pathology, a systematic review was conducted of English-written literature in PubMed until May 2016, which included research on the relationship of these dietary components with the onset and progression of periodontal disease. According to publication type, randomized-controlled trials, cohort, case-control and cross-sectional studies were included. Among all the analyzed components, those that have any effect on oxidative stress and/or inflammation seem to be the most interesting according to current evidence. On one hand, there is quite a lot of information in favor of a positive role of n-3 fatty acids, due to their antioxidant and immunomodulatory effects. On the other hand, saturated fat-rich diets increase oxidative stress as well the as intensity and duration of inflammatory processes, so they must be avoided. dietfatty acidspolyunsaturated fatty acids (PUFA)nutritionoral healthperiodontitis ==== Body 1. Introduction Periodontal disease is a multifactorial pathology featured by the breakdown of periodontal tissues [1]. There is an increasing proportion (from 30% to 65% over the last decades) of adults retaining their teeth until late in life [2], and, nowadays, periodontal disease is a serious problem in older adults [3]. Additionally, the risk of other oral problems (root caries, tooth mobility, and tooth loss) can be increased by the presence of periodontal disease, which has also been associated with a growing list of chronic systemic diseases and impaired cognition [4,5,6,7,8,9,10,11,12,13,14,15,16]. Traditionally, research on the effect of nutrition on oral disease has focused on the dietary effects on the risk of caries [17]. However, it is less understood how the diet affects the development and progression of periodontal disease. A bidirectional correlation has been reported among nutrition, dietary intake, and oral health [18,19]. On one hand, some studies have indicated that oral health status may have implications for dietary intake [20,21,22]. On the other hand, it is largely assumed that the progression of periodontal disease and healing of the periodontal tissues can be modulated by nutritional status. Actually, in spite of the fact that the presence of bacteria is the primary etiology of this disease, a susceptible host is also necessary for disease initiation [23]. Furthermore, many inflammatory conditions and/or diseases, such as type 2 diabetes mellitus (DM), cardiovascular diseases, and rheumatoid arthritis have been related to diet, all of which have been associated with periodontitis [24]. Likewise, nutritional status could affect the integrity of both hard and soft tissues in the oral cavity [23]. First, during teeth development, foods provide a nutritional or systemic effect that may affect the maturation of dentine and enamel. Then, when the teeth have erupted, foods can influence the maintenance of their structure through dietary and topical effects [25]. Regarding all of these implications, it has been hypothesized that “periodontal treatment could be enhanced with the addition of certain nutrients to periodontal therapy, providing a safe method to potentiate the clinical response following treatment” [26]. In particular, it has been observed that the lipid composition of the cell membranes and blood lipoprotein content can be modified through the diet [27,28,29], which has been linked to susceptibility to oxidative damage [30]. Likewise, the response against certain bacterial products can also be modulated by membrane lipid profile [31]. For these reasons, some modifications of dietary patterns affecting lipid profiles could be interesting to both preventing periodontal diseases and improving periodontal health, since all components affected by them are considered key aspects in the pathogenesis of periodontal disease [11]. There are some reviews which provide information about the association between periodontitis and diet components [3,24,25,32,33,34,35,36,37,38,39] although many of them were restricted to certain population groups [3], study types [3,24,34], and/or did not include lipids [3,24,26,35]. However, few attempts of systematic reviews have been carried out [1,3,24,34]. Overall, the systematic approach with the focused question is precluded by a paucity of nutritional interventions and a wide heterogeneity of designs. This paper systemically reviews the literature available on databases until May 2016 on the relationship of lipids with the development and progression of periodontal disease in humans, attending with special interest to dietary interventions and implications of each one for mechanisms involved in theses pathologies. 2. Results A total of 5564 publications were detected in the initial search. After applying inclusion and exclusion criteria, title and abstract screening left 220 available articles. After full-text reading, 13 studies were selected, with one that was duplicated and eliminated (Figure 1). The relationship between dietary lipids and periodontal disease has been addressed by many reports. These included both observational (Table 1) and experimental (Table 2) studies. Observational studies were represented by two cross-sectional [40,41], a case-control [42], and three cohort studies [43,44,45]. Among the selected cross-sectional studies, dietary fat intake was estimated in two studies [40,41]. In one of them, total dietary intake of carbohydrates, along with many other nutrients, was assessed by a food frequency questionnaire (FFQ) in adults from Japan using associated data from the 2005 National Health and Nutrition Examination Survey (NHANES). After excluding pregnant women and subjects younger than 20 years, the remaining individuals (n = 3043) were divided into two groups according to their individual maximum Community Periodontal Index (CPI) values (0–2 or 3–4) to subsequent analyses. Only total dietary fat intake, in terms of total amount and percentage of energy, was available, reporting that fat intake was lower in the group with higher Community Periodontal Index (CPI) (50.8 ± 21 vs. 55.3 ± 23.1 g and 23.2% ± 7% vs. 25.2% ± 7.3% Kcal, p < 0.001). After adjusting for confounders in the multivariate logistic regression analysis, only fat as percentage of calories was inversely associated with advanced periodontal disease (adjusted odds ratio (OR) = 0.984, p = 0.026) [40]. The other available study was performed using a subset of 9182 adult participants in the U.S. NHANES. This study showed a negative association of dietary intake of representative polyunsaturated fatty acids (PUFA) belonging to n-3 series with moderate to severe periodontitis. The adjusted ORs of periodontitis associated with the lowest tertile of n-3 PUFA intake compared with the highest were 0.78 (95% confidence interval (CI): 0.61–1.00, p = 0.009) for docosahexaenoic acid (DHA), 0.85 (95% CI: 0.67–1.08, p = 0.10) for eicosapetanoic acid (EPA), and 0.86 (95% CI: 0.60–1.23, p = 0.28) for γ-linoleic acid (GLA). These associations were changed in a low-way by multivariate adjustment, as well as after DHA, EPA, or GLA dietary supplements users’ exclusion [41]. Results from the single case-control study also had findings supportive of the previously mentioned study. In this study, serum levels of PUFA were compared between people with normal bone height or with a bone loss of more than 3 mm, as determined in a radiographic film. Results showed that patients with bone loss presented higher n-6 PUFA levels than the control group (8.04% ± 0.35% vs. 7.24% ± 0.51% mole; p = 0.03), whereas the opposite was observed for n-3 PUFA (2.04% ± 0.17% vs. 2.54% ± 0.41% mole; p = 0.01) [42]. All the cohort studies reviewed were based on a subset from the Niigata City (Japan) study. EPA and DHA intakes effect on periodontal disease events over 5 years were evaluated in one of these studies, but only DHA led to relevant results. Individuals who consumed the lowest intake (1st tertile) of DHA presented an incidence rate ratio (IRR) of 1.49 (95% CI: 1.01–2.21) with respect to the highest tertile after simultaneous adjustment for possible confounders [43]. In another study, dietary n-6 to n-3 PUFA ratio was considered as the main predictor to estimate its influence on periodontal disease events after 3 years of follow-up. A high dietary n-6 to n-3 PUFA ratio was significantly associated with a greater number of periodontal disease events in Poisson regression analysis [44]. In the last cohort study, saturated fatty acids (SFA) intake was evaluated with respect to the same outcome. Among non-smokers, it was found that high dietary intake of SFA was associated with more periodontal disease events. In more detail, the multivariate adjusted relative risk (RR) values were 1.00, 1.19 (95% CI: 0.72–1.97), 1.55 (95% CI: 0.95–2.52), and 1.92 (95% CI: 1.19–3.11), in the 1st, 2nd, 3rd, and 4th quartiles of SFA intake, respectively [45]. Among the collected studies, six dietary interventions were found [46,47,48,49,50]. The earliest study was conducted by Campan, et al. [46] on subjects without oral hygiene as a model of gingivitis, where subjects received supplements of fish oil (n-3 PUFA-rich) or olive oil (as placebo). After a short-period (8 days), Gingival Index (GI) was reduced by fish oil (p < 0.05), but there were no differences between experimental and control groups. Subsequently, other investigations have been aimed at treating patients with periodontitis. The first one [47], performed on adults with periodontitis, used refined olive oil mixed with corn oil as placebo, and fish oil or borage oil as a source of n-3 (EPA), or n-6 (GLA) PUFA, respectively. Different periodontal parameters were measured at baseline and after treatment (12 weeks). The Modified Gingival Index (MGI), Plaque Index (PI), periodontal probing depth (PPD), and β-glucuronidase levels in gingivocrevicular fluid (GCF). After treatment, subjects treated with borage oil displayed improvement of gingival inflammation (1.04 vs. 0.68, p = 0.016). Subjects taking either fish or borage oil alone displayed an improvement in PPD. Statistically significant differences were found only when borage oil and placebo were compared (−0.50 vs. 0.02, p = 0.044). No change was observed in GCF or β-glucuronidase levels. Four other interventions were done after a scaling and root planning (SRP) treatment [48,49,50,51]. In one of these studies [48], subjects with moderate and severe chronic periodontitis taking n-3 PUFA supplements (comprising a combination of EPA and DHA) were compared with others taking placebo. After 12 weeks of treatment, several periodontal disease outcomes improved, which included PPD (2.15 ± 0.53 vs. 2.77 ± 0.47 mm, p < 0.05), clinical attachment loss (CAL) (2.73% ± 0.98% vs. 3.72% ± 0.62%, p < 0.05), GI (2.23 ± 0.57 vs. 1.76 ± 0.63, p < 0.05); and sulcus bleeding index (SBI, 1.41 ± 0.30 vs. 1.80 ± 0.39, p < 0.05). In addition, most of them improved after 6 weeks. These were PPD (2.61 ± 0.52 vs. 3.19 ± 0.48 mm, p < 0.05), SBI (1.65 ± 0.28 vs. 2.06 ± 0.40, p < 0.05), and GI (1.32 ± 0.21 vs. 1.58 ± 0.33, p < 0.05). However, PI remained unchanged among groups. Another study was carried out for a longer period of time (12 months) on generalized chronic periodontitis patients. In this study, supplements of DHA and EPA had no effect on the percentage of sites with bleeding on probing (BOP), visible PI, PPD, or clinical attachment loss. However, sample size in this research (n = 15) was quite reduced [52]. In the next study, the combination of n-3 PUFA and aspirin was tested on advanced chronic periodontitis patients. After 3 and 6 months of treatment, a PPD reduction (−2 mm vs. −1.4 mm at 6 months, p < 0.05) and a gain of attachment in n-3 PUFA fed subjects with respect to both baseline and control group were confirmed by statistical analyses (75.3% vs. 28.6% sites with 1 to 3 mm gain, 0.8% vs. 0.1% sites with more of 4 mm gain at 6 months, p < 0.05). At 3 and 6 months, the n-3 group showed a significant reduction in salivary receptor activator of nuclear factor κB ligand (RANKL) and matrix metalloproteinase-8 (MMP-8) levels compared with baseline. Levels were also lower than the control group at 6 months (p < 0.01). Consumption of supplements containing n-3 PUFA and aspirin led to a significant variation in the frequency of pockets with PPD less than 4 mm (34.3% gain at 3 months and 39.1% at 6 months vs. −26.9% at 3 months and −31.5% at 6 months, p < 0.05) [53]. In the remaining study, periodontitis patients also received an aspirin treatment, but in this case, the treatment group was supplemented with DHA while the control group was treated with soy and corn oil supplements. Despite possible aspirin effect, DHA supplements decreased the mean PPD (−0.29 ± 0.13 mm, p = 0.03) and GI (−0.26 ± 0.13, p = 0.04), although there were no differences in PI or BOP. Along with periodontal health assessment, levels of different molecules related to inflammation were also measured in both GCF and plasma. Results analysis showed lower levels of GCF, high sensitivity C-reactive protein (hsCRP) (−5.3 ± 2.4 ng/mL, p = 0.03), and IL-1β (−20.1 ± 8.2 pg/mL, p = 0.02), but not IL-6 or systemic hsCRP. Moreover, in this study, the adherence to dietary modification was confirmed by measurement of DHA level in red blood cells (this fatty acid increased from 3.6% ± 0.9% to 6.2% ± 1.6% as a consequence of supplementation). 3. Discussion In spite of the information collected in this review, in many cases, there was little evidence supporting some degree of relationship between periodontal disease and lipid intake. This may be because inferences often come from observational studies, mostly cross-sectionals. Furthermore, many of the reviewed studies have been performed in certain age groups, or with particular diseases or physiological situations. So, translating these findings to the overall population is hampered. In addition, few articles investigated the mechanisms underlying the observed relationships. Moreover, although indices used to evaluate periodontal health are few and mostly accepted (such as Russell’s periodontal index, CPI, or GI), there is a wide variety of periodontitis definitions based on alveolar bone loss (ABL), PPD, CAL, or even bleeding (BOP). Consequently, comparisons among findings from different publications are not easy. Nutritional assessment in observational studies may also be inconvenient. When dietary intakes are estimated by 24-h [40,41] or 3-day recalls [42,43], consequent bias may exist, namely in investigations with small sample sizes. Alternatively, conclusions in many cases are derived only from macronutrient assessment in blood [42], but these values are not always directly related to dietary intake. Both methodological features could help to explain why some papers reported no associations or associations only related to certain age groups. In addition, observed associations could be (at least in part) a consequence of an excess in energy intake or to any nutritional deficiency that might occur when dietary intake is modified far from healthy dietary recommendations. On one hand, experimental studies on macronutrient effects often have a difficult interpretation, particularly those focused on the role of quantitative changes in diet. Namely, increases in lipid content usually lead to hypercaloric diets. Despite these difficulties, it is possible to draw some conclusions. Studies evaluating the possible effects of total amount of fat were only represented by a single study in Japanese adults [40]. It has been proposed that a high fat intake might lead to enhanced oxidative stress because of an overloading of the Krebs cycle, among other reasons [38,39]. Oxidative stress has been associated with periodontal disease exerting its effects by means of direct damage to cells. It has been related through the activation of redox-sensitive transcription factors, leading to the production of pro-inflammatory molecules (Figure 2) which enhance and propagate the inflammatory response, elevating the local levels of oxidative stress [38,39]. Unexpectedly, results from the study mentioned suggested a positive association between periodontal health and fat intake [40]. Several questions could help to explain this rare result. On one hand, a higher percentage of fat in the diet does not necessarily imply a higher intake of calories that can overload the Krebs cycle. On the other hand, dietary fat quality is very important. However, both aspects were not collected by the Japanese study. In this sense, potential effects of different fatty acid types (derived directly from the diet or as supplements) on periodontal disease have been investigated. PUFAs, particularly n-3 PUFAs, have received the most attention, although other fatty acid types have been also studied. Overall, studies conferred a positive role to n-3 PUFA over periodontal health [41,42,43,44], and negative roles to SFA [45]. In turn, results for n-6 PUFA are discussed because it is well known that some of these fatty acids have harmful associations [42,53]. When n-6 PUFAs have been studied directly in comparison with n-3 PUFAs, the former have always reported clearly negative effects [44]. Among studies in favor of these associations, there are some cohort studies, all of them based on subsets from the Niigata City (Japan) study [43,44]. In addition, some nutritional interventions on n-3 PUFA have demonstrated a beneficial effect on periodontitis [48,50]. This was expected because of anti-inflammatory properties have been widely attributed to n-3 PUFA, which are in fact able to down-regulate pro-inflammatory gene expression [53] and the production of pro-resolving lipid mediators, resolvins, maresins, and protectins [33]. Interestingly, it has also been reported that the n-3 PUFAs EPA and DHA have a broad range of antibacterial activities, including the inhibition of putative periodontal pathogens, including Porphyromonas gingivalis, Fusobacterium nucleatum, and Prevotella intermedia, among others [54]. The negative association between fat-rich diets and periodontitis found in the Japanese sample could be explained by the role of these fatty acids, since the population of Japan shows a much higher proportion of dietary n-3 PUFA [55] than is present in an average Western diet [56]. Concerning n-6 PUFAs, it is well known that their derived pro-inflammatory prostanoids directly compete with the anti-inflammatory n-3 PUFAs. This fact makes it easy to understand the importance of carefully considering the n-6-to-n-3 PUFA ratio in the diet. In relation to saturated fat, the concern is that an excess metabolically leads to an increased production of low density lipoprotein (LDL) cholesterol. This, when oxidized, forms oxidized LDL, which binds to toll-like receptor 4 (TLR-4) on neutrophils, again activating downstream pro-inflammatory cascades [38], among other things. It would be possible to conclude that macronutrients that have any effect on antioxidant status or inflammatory processes should be considered for the prevention or improvement of periodontal disease. In this sense, there is quite a body of evidence to consider regarding the positive role of including a high proportion of n-3 PUFA in the diet. On the other hand, those lipids or diets that increase oxidative stress or affect the immune system should be avoided to prevent periodontal disease or to achieve better results after periodontal therapies. These would include saturated fat-rich or hypercaloric diets. However, after analyzing all collected data, there is a need for more studies to confirm the effects of diet, mainly cohort studies with large sample sizes. Additionally, more consensus in periodontitis diagnosis and improvement of dietary intake estimations are needed to enhance our overall understanding. Indeed, problems in the last issue also focus on the importance of carrying out more sensitive experimental investigations and/or methodological approaches. In the future, the design of new clinical trials that combine dietary interventions as periodontal therapies might be the most useful tools to emphasize the importance of diet in these pathologies. Regardless, it would be appropriate to maintain an optimal dietary n-6:n-3 PUFA ratio. This is consistent with the recommendations of the 2011 European workshop on Periodontology that suggested that “the dental team should consider including advice to all patients on increasing levels of fish oils, fibre, fruit, and vegetables, and to reduce levels of refined sugars as part of a periodontal prevention/treatment regime and a general health benefit message” [39,57]. 4. Material and Methods 4.1. Selection Criteria Inclusion and exclusion criteria for the selection of studies to review were established prior to the start with the literature search. All relevant studies investigating the association between periodontal disease and dietary lipids in humans were included, even when they included only people belonging to certain age groups or subject to any special physiological condition or illness (pregnancy, menopause, DM, and others). For this reason, studies had to assess any periodontal health condition and dietary intake or clear nutritional status markers for the mentioned dietary components. Regarding study design, cross-sectional, cohort, and case-control studies were selected within those belonging to observational type. On the other hand, only randomized-controlled trials (RCTs) were accepted from experimental studies collected. 4.2. Information Source and Search Terms The electronic database of the National Library of Medicine, Washington, D.C. (MEDLINE: PubMed) was used to search potential publications for this review. Firstly, two themes were created and combined by using the Boolean operator “AND”. In turn, the operator “OR” was used to create each theme, combining search terms appearing as either explorer text words or Medical Subjects Headings (Mesh), when they existed and it was not contained within other also used. Asterisk (*) also was used after some terms to search for all terms that begin with the word. The selected periodontal disease related terms were: periodontal, periodontitis, gingivitis and it was also included the following outcome measurement related to periodontal disease: “alveolar bone loss” (Mesh) OR “periodontal diseases” (Mesh) OR “periodontal attachment loss” (Mesh) OR “periodontal index” (Mesh) OR “gingival hemorrhage” (Mesh) OR “periodontal disease” OR periodontitis OR “alveolar bone loss” OR “alveolar bone resorption” OR “tooth attachment” OR “tooth mobility” OR “gingivitis” OR “clinical attachment loss” OR “periodontal attachment level” OR “attachment loss” OR “periodontal pocket” OR “pocket depth” OR “probing depth” OR “bleeding on probing” OR “gingival bleeding” OR “gingival hemorrhage” OR “gingival index” OR “bleeding index” OR “periodontal index”. In addition to this, a second theme related to nutrition or diet was created. Search terms were the next: “food” (Mesh) OR “diet” (Mesh) OR “eating” (Mesh) OR “nutrition surveys” (Mesh) OR “nutrition assessment” (Mesh) OR “nutrition therapy” (Mesh) OR “nutrition processes” (Mesh) OR “nutritional status” (Mesh) OR nutrition* OR nutrition OR nutrient* OR nutrient OR food OR dietary OR diet* OR intake OR intakes OR consumption* OR consumption OR ingestion OR eating. 4.3. Search Strategy The search strategy was focused on English-written studies from inception of the database until May 2016. A comprehensive literature search was run independently by two of the review authors. Firstly, a screening adjusted for higher sensitivity (i.e., without restrictive search items) of title and abstract was carried out. Reviews that could mention studies related to the established selection criteria were collected, and a screening was done to find new additional articles that would follow the rest of the process; Secondly, full-texts were obtained and after full-text screening; those articles that did not accomplish the proposed selection criteria were excluded. Duplicated studies were removed; Moreover, full-text was screened in the web for articles whose title suggested that they were related to these review objectives; Finally, when there were disagreements or inconsistencies relative to the inclusion of some publications or data, all authors discussed the issue to eventually achieve mutual consensus. 4.4. Data Collection Process, Data Items, and Summary Measures Specific data about populations, interventions, study designs, as well as outcomes relevant for this review’s questions were extracted from articles. Data from eligible studies were independently evaluated. When they were available, mean, mean differences and standard deviations of outcome measurements (like clinical outcomes for periodontal disease, nutritional status, or dietary intakes), were included in our description of the results. In the same sense, association measurements were also collected, such as correlation or regression coefficients, IRRs, ORs, RRs and their corresponding 95% CI, as well as significance levels considered or p-values, but only if significant associations and/or differences were found. 4.5. Quality Assessment and Risk of Bias Subsequently, potentially-relevant publications were read in full to determine their quality, mainly through the risk of bias assessment. Several criteria depending on the study design were applied to determine methodological quality of each publication. In the case of observational studies, evaluation and strategies to deal with them were considered in addition to risk of bias. According to the prevention of confounding effects, the employment of randomization, only in clinical trials, matching, or restriction, were explored at the design level, while the use of multivariate analysis, stratification, or frequency matching was appraised at the data analysis level. As to the risk of bias, it four types of bias have been proposed: selection, detection, attrition, and reporting bias [58]. Other authors have simplified this classification into two categories: selection and information bias [59]. Regardless, critical issues for bias prevention are different between design types. In cross-sectional studies, three points were mainly evaluated: sample randomization; reliability and objectivity of variables assessment; as well as definition of group if comparisons were made. In case–control studies, objectivity and reliability of evaluations of subjects was also taken into account. For this study type, category definitions and exposures evaluation method also were included. Lastly, in cohort studies, exposure and status definition, measurement, differences in duration, reliability, and objectivity of outcome assessment were considered, as well as the presence of differences in intensity of medical surveillance, loss of follow-up, and missing data treatment. In addition, possible differences in outcomes or exposure to risk factors between subjects who drop out and those who stay in the study were taken into account. Inclusion criteria and/or sources of data or individuals’ background were considered for the establishment of sample external validity and population set represented. Additionally, grouping criteria in data analysis were taken into account, when they were presented, especially if they were selected post hoc from alternative options. On the other hand, intervention trials were assessed according to Cochrane guidelines, which take into account five type of bias: selection, performance, detection, attrition, and reporting bias [60]. Based in its recommendations, the following domains were evaluated: sequence generation (randomization), allocation concealment, blinding of participants, operators and/or examiners, incomplete outcome data, and selective reporting. With regard to the results of this assessment, degree of risk (high, low, or unclear) was established for each type of bias, as well as its possible magnitude and direction. The risk of bias and its possible effects were summarized for each outcome within each study. Finally, it was summarized for the review as a whole, where it was possible. Acknowledgments Alfonso Varela-López is recipient of a fellowship of FPU program from the Spanish Ministry of Education. Authors acknowledge to the University of Granada and the Autonomous Government of Andalusia for partial support of the research team. Conflicts of Interest The authors declare no conflict of interest. Abbreviations BOP bleeding on probing CAL clinical attachment loss CI confidence interval CPI Community Periodontal Index DHA docosahexaenoic acid DM diabetes mellitus EPA eicosapentanoic acid FFQ food frequency questionnaire GCF gingivocrevicular fluid GI Gingival Index GLA γ-linoleic acid hsCRP high sensitivity C-reactive protein IRR incidence rate ratio LDL low density lipoprotein Mesh Medical Subjects Headings MGI Modified Gingival Index MMP-8 matrix metalloproteinase-8 NHANES National Health and Nutrition Survey OR odds ratio PI Plaque Index PPD Periodontal probing depth PUFA polyunsaturated fatty acid RANKL receptor activator of nuclear factor κB ligand RCT randomized-controlled trial RR relative risk SBI Sulcus Bleeding Index SFA saturated fatty acids SRP scaling and root planning TLR-4 toll-like receptor 4 Figure 1 Screening protocol. Figure 2 Mechanisms for fat intake effects. Abbreviations: LDL: low-density lipoprotein, PUFA: polyunsaturated fatty acids, SFA: saturated fatty acids, TLR-4: toll-like receptor-4. ijms-17-01202-t001_Table 1Table 1 Observational studies on lipids association with periodontal disease. Reference; Study Type Subjects; Age Main Outcomes/Groups Compared Exposures Main Results/Conclusion Hamasaki, et al., 2016 [40]; CS 3043 NHANES participants (Japan) ≥20 years Adjusted OR of CPI = 3–4 Dietary intake of total fat (wt and %E) Negative association with dietary intake of fat in %E Naqvi, et al., 2010 [41]; CS 9182 NHANES 1999–2004 participants (USA); ≥20 years Adjusted OR of periodontitis 1 Dietary intake of FA Negative association with n-3 PUFA, DHA, EPA and GLA Requirand, et al., 2000 [42]; CC 105 patients (France) 41.1 ± 2.6 years/43.4 ± 6.6 years Suffered from bone loss ≤3 mm on several teeth vs. normal bone height-periodontium Serum levels of PUFA n-6 PUFA were higher in patients with bone loss, whereas n-3 PUFA were lower Iwasaki, et al., 2010 [43]; C 55 Niigata study participants (Japan); 74 years IRR of periodontal disease events 2 Dietary intakes of DHA, and EPA Negative association with DHA intake Iwasaki, et al., 2011 [44]; C 235 Niigata study participants (Japan); 75 years Adjusted RR of periodontal disease events 3 Dietary intakes energy- adjusted of n-6 and n-3 PUFA and n-6/n-3 PUFA ratio Positive association with n-6/n-3 PUFA ratio Iwasaki, et al., 2011 [45]; C 264 Niigata study participants (Japan); 75 years Adjusted RR of periodontal disease events 3 Energy-adjusted dietary intakes of SFA Positive association 1 PPD ≥ 4 mm and AL ≥ 3 mm in any mid-facial or mesial tooth; 2 CAL ≥ 6 mm in ≥2 teeth and PPD ≥ 5 mm in ≥1 site; 3 number of teeth with AL ≥ 3 mm/year. Abbreviations: %E: percentage of Energy; AL: attachment loss; C: cohort study; CC: case-control; CPI: Community Periodontal Index; CS: cross-sectional study; DHA: docosahexaenoic acid; EPA: eicosapentanoic acid; FA: fatty acid; GLA: γ-linoleic acid; IRR: incidence rate ration; NHANES: National Health and Nutrition Survey; OR: Odds ratio; PUFA: polyunsaturated fatty acids; RR: relative risk; SFA: saturated fatty acids; wt: weight; y: years; PPD: periodontal probing depth; CAL: clinical attachment loss. ijms-17-01202-t002_Table 2Table 2 Experimental studies on lipids effect on periodontal disease. Reference; Study Type Subjects; Age Experimental Treatments (Duration) Analytic Measurement Main Results/Conclusions Campan, et al., 1997 [46]; RCT 37 healthy volunteers with intensive oral hygiene for 14 days Oral hygiene abstention (29 days), in combination with supplementation with fish oil or olive oil as placebo (last 8 days) PI, GI, PBI, and gingival levels (only in 10 volunteers) of AA, EPA, DHA, DPA, PGE2 and LTB4 Fish oil supplements reduced GI, but there are no differences between experimental and control group. LTB4 was lower fish oil treated subjects Rosenstein, et al., 2003 [47]; RCT (DB) 30 subjects with periodontitis; 18–60 years Supplementation with EPA or borage oil, both, or a mixture of olive and corn oil as placebo (12 weeks) PI, MGI, BOP, PPD and CAL and salivary RANKL and MMP-8 Supplementation with borage oil or EPA improved PPD, but only borage oil effect was significant respect to placebo. Additionally, it also improved MGI Deore, et al., 2014 [48]; RCT (DB) 60 subjects with moderate and severe chronic periodontitis; 45.4 ± 4.9/44.5 ± 5.2 years Supplementation with n-3 PUFA or placebo; after SRP (6 or 12 weeks) ABL, P. gingivalis identification, serum FA profile Treatment reduced PPD and salivary RANKL and MMP-8 levels; and increased CAL Martinez, et al., 2014 [49]; RCT (DB) 15 patients with generalized chronic periodontitis (43.1 ± 6/46.1 ± 11.6 years) Supplementation with n-3 PUFA or placebo; after SRP (12 months) % BOP, visible plaque index, PPD and CAL No effect El-Sharkawy, et al., 2010 [50]; RCT (DB) 80 subjects with advanced chronic periodontitis; 30–70 years Supplementation with fish oil and aspirin or placebo; after SRP (3 or 6 months) PI, GI, OHIS, BOP, SBI, PPD, CAL and serum levels of CRP Supplementation with borage oil or EPA improved PPD, but only borage oil effect was significant respect to placebo. Additionally, it also improved MGI Naqvi, et al., 2014 [51]; RCT (DB) 46 subjects with moderate periodontitis; adults Supplementation with DHA or soy/corn oil capsules, in combination with aspirin (3 months) GI, PI, BOP, PPD, GCF levels of hsCRP, IL-6 and IL-1β, systemic inflammatory markers plasma levels, and erythrocytes fatty acids Supplementation with DHA decreased mean PPD and GI. This was accompanied by lower hsCRP and IL-1β levels in GCF Abbreviations: AA: arachidonic acid; ABL: alveolar bone loss; BOP: bleeding on probing; CAL: clinical attachment loss; CRP: C-reactive protein; d: days; DB: double-blind; DHA: docosahexaenoic acid; EPA: eicosapentanoic acid; FA: fatty acid; GCF: gingivocrevicular fluid; GI: gingival index; hsCRP: high sensitive C-reactive protein; IL-6: interleukin-6; IL-1β: interleukin-1β; LTB4: leukotriene B4; m: months; MGI: modified gingival index MMP-8: matrix metalloproteinase-8; OHIS: oral health index; PBI: papillary bleeding index; PGE2: prostaglandin E2; P. gingivalis: Porphyromonas gingivalis; PI: plaque index; PPD: periodontal probing depth, PUFA: polyunsaturated fatty acids; RANKL: receptor activator of nuclear factor κB ligand ; RCT: randomized-controlled trial; SBI: sulcus and bleeding index SRP: scaling and root planning; y: years; w: weeks. ==== Refs References 1. Amaral C.D.S.F. Vettore M.V. Leão A. The relationship of alcohol dependence and alcohol consumption with periodontitis: A systematic review J. Dent. 2009 37 643 651 10.1016/j.jdent.2009.04.011 19576673 2. Shay K. Ship J.A. The importance of oral health in the older patient J. Am. Geriatr. Soc. 1995 43 1414 1422 10.1111/j.1532-5415.1995.tb06624.x 7490396 3. Van der Putten G.-J. Vanobbergen J. de Visschere L. Schols J. de Baat C. Association of some specific nutrient deficiencies with periodontal disease in elderly people: A systematic literature review Nutrition 2009 25 717 722 10.1016/j.nut.2009.01.012 19539173 4. Battino M. Ferreiro M.S. Fattorini D. Bullon P. in vitro antioxidant activities of mouthrinses and their components J. Clin. Periodontol. 2002 29 462 467 10.1034/j.1600-051X.2002.290512.x 12060430 5. Battino M. Ferreiro M.S. Gallardo I. Newman H.N. Bullon P. The antioxidant capacity of saliva J. Clin. Periodontol. 2002 29 189 194 10.1034/j.1600-051X.2002.290301x.x 11940135 6. Battino M. Bompadre S. Leone L. Pugnaloni A. Rubini C. Ferreiro M.S. Gallardo I. Bullon P. The effect of Cyclosporine A chronic administration on the antioxidant pattern of rat liver mitochondria: Structural and functional consequences BioFactors 2003 18 271 275 10.1002/biof.5520180230 14695943 7. Battino M. Ferreiro M.S. Armeni T. Politi A. Bompadre S. Massoli A. Bullon P. in vitro antioxidant activities of antioxidant-enriched toothpastes Free Radic. Res. 2005 39 343 350 10.1080/10715760400023853 15788239 8. Battino M. Bompadre S. Politi A. Fioroni M. Rubini C. Bullon P. Antioxidant status (CoQ10 and Vit. E levels) and immunohistochemical analysis of soft tissues in periodontal diseases BioFactors 2005 25 213 217 10.1002/biof.5520250126 16873949 9. Bullon P. Pugnaloni A. Gallardo I. Machuca G. Hevia A. Battino M. Ultrastructure of the gingiva in cardiac patients treated with or without calcium channel blockers J. Clin. Periodontol. 2003 30 682 690 10.1034/j.1600-051X.2003.00358.x 12887336 10. Bullon P. Fioroni M. Goteri G. Rubini C. Battino M. Immunohistochemical analysis of soft tissues in implants with healthy and peri-implantitis condition, and aggressive periodontitis Clin. Oral Implants Res. 2004 15 553 559 10.1111/j.1600-0501.2004.01072.x 15355397 11. Bullon P. Morillo J.M. Ramirez-Tortosa M.C. Quiles J.L. Newman H.N. Battino M. Metabolic syndrome and periodontitis: Is oxidative stress a common link? J. Dent. Res. 2009 88 503 518 10.1177/0022034509337479 19587154 12. Bullon P. Gallardo I. Goteri G. Rubini C. Battino M. Ribas J. Newman H.N. Nifedipine and cyclosporin affect fibroblast calcium and gingiva J. Dent. Res. 2007 86 357 362 10.1177/154405910708600411 17384032 13. Bullon P. Quiles J.L. Morillo J.M. Rubini C. Goteri G. Granados-Principal S. Battino M. Ramirez-Tortosa M. Gingival vascular damage in atherosclerotic rabbits: Hydroxytyrosol and squalene benefits Food. Chem. Toxicol. 2009 47 2327 2331 10.1016/j.fct.2009.06.026 19545596 14. Bullon P. Cordero M.D. Quiles J.L. Morillo J.M. del Carmen Ramirez-Tortosa M. Battino M. Mitochondrial dysfunction promoted by Porphyromonas gingivalis lipopolysaccharide as a possible link between cardiovascular disease and periodontitis Free Radic. Biol. Med. 2011 50 1336 1343 10.1016/j.freeradbiomed.2011.02.018 21354301 15. Bullon P. Cordero M.D. Quiles J.L. Ramirez-Tortosa M.D.C. Gonzalez-Alonso A. Alfonsi S. García-Marín R. de Miguel M. Battino M. Autophagy in periodontitis patients and gingival fibroblasts: Unraveling the link between chronic diseases and inflammation BMC Med. 2012 10 1202 10.1186/1741-7015-10-122 23075094 16. Kaye E.K. n-3 Fatty acid intake and periodontal disease J. Am. Diet. Assoc. 2010 110 1650 1652 10.1016/j.jada.2010.08.017 21034877 17. Scardina G.A. Messina P. Good oral health and diet J. Biomed. Biotechnol. 2012 2012 10.1155/2012/720692 22363174 18. Sullivan D.H. Martin W. Flaxman N. Hagen J.E. Oral health problems and involuntary weight loss in a population of frail elderly J. Am. Geriatr. Soc. 1993 41 725 731 10.1111/j.1532-5415.1993.tb07461.x 8315182 19. Krall E. Hayes C. Garcia R. How dentition status and masticatory function affect nutrient intake J. Am. Dent. Assoc. 1998 129 1261 1269 10.14219/jada.archive.1998.0423 9766107 20. Budtz-Jørgensen E. Chung J.P. Rapin C.H. Nutrition and oral health Best Pract. Res. Clin. Gastroenterol. 2001 15 885 896 10.1053/bega.2001.0247 11866483 21. Marshall T.A. Warren J.J. Hand J.S. Xie X.-J. Stumbo P.J. Oral health, nutrient intake and dietary quality in the very old J. Am. Dent. Assoc. 2002 133 1369 1379 10.14219/jada.archive.2002.0052 12403539 22. Hung H.-C. Willett W. Ascherio A. Rosner B.A. Rimm E. Joshipura K.J. Tooth loss and dietary intake J. Am. Dent. Assoc. 2003 134 1185 1192 10.14219/jada.archive.2003.0353 14528990 23. Boyd L.D. Lampi K.J. Importance of nutrition for optimum health of the periodontium J. Contemp. Dent. Pract. 2001 2 36 45 12167932 24. Van der Velden U. Kuzmanova D. Chapple I.L.C. Micronutritional approaches to periodontal therapy J. Clin. Periodontol. 2011 38 Suppl. 11 142 158 10.1111/j.1600-051X.2010.01663.x 21323711 25. Schifferle R.E. Nutrition and periodontal disease Dent. Clin. N. Am. 2005 49 595 610 10.1016/j.cden.2005.03.008 15978243 26. Neiva R.F. Steigenga J. Al-Shammari K.F. Wang H.-L. Effects of specific nutrients on periodontal disease onset, progression and treatment J. Clin. Periodontol. 2003 30 579 589 10.1034/j.1600-051X.2003.00354.x 12834494 27. Huertas J.R. Battino M. Lenaz G. Mataix F.J. Changes in mitochondrial and microsomal rat liver coenzyme Q9 and Q10 content induced by dietary fat and endogenous lipid peroxidation FEBS Lett. 1991 287 89 92 10.1016/0014-5793(91)80022-U 1879539 28. Huertas J.R. Martínez-Velasco E. Ibáñez S. López-Frías M. Ochoa J.J. Quiles J.L. Parenti-Castelli G. Mataix J. Lenaz G. Virgin olive oil protect heart mitochondria from peroxidative damage during aging BioFactors 1999 9 337 343 10.1002/biof.5520090233 10416050 29. Ochoa-Herrera J.J. Huertas J.R. Quiles J.L. Mataix J. Dietary oils high in oleic acid, but with different non-glyceride contents, have different effects on lipid profiles and peroxidation in rabbit hepatic mitochondria J. Nutr. Biochem. 2001 12 357 364 10.1016/S0955-2863(01)00150-4 11516640 30. Lobb K. Chow C.K. Fatty acid classification and nomenclature Fatty Acids in Foods and Their Health Implications Chow C.K. CRC Press Boca Raton, FL, USA 2008 1 15 31. Calder P.C. n–3 Polyunsaturated fatty acids, inflammation, and inflammatory diseases Am. J. Clin. Nutr. 2006 83 S1505 S1519 32. Schifferle R.E. Periodontal disease and nutrition: Separating the evidence from current fads Periodontol. 2000 2009 50 78 89 10.1111/j.1600-0757.2008.00297.x 19388955 33. Neiva R.F. Al-Shammari K. Nociti F.H. Soehren S. Wang H.-L. Effects of vitamin-B complex supplementation on periodontal wound healing J. Periodontol. 2005 76 1084 1091 10.1902/jop.2005.76.7.1084 16018750 34. Kulkarni V. Bhatavadekar N.B. Uttamani J.R. The effect of nutrition on periodontal disease: A systematic review J. Calif. Dent. Assoc. 2014 42 302 311 25087348 35. Varela-López A. Bullón P. Giampieri F. Quiles J.L. Non-nutrient, naturally occurring phenolic compounds with antioxidant activity for the prevention and treatment of periodontal diseases Antioxidants 2015 4 447 481 10.3390/antiox4030447 26783837 36. Varela-López A. Quiles J.L. Cordero M. Giampieri F. Bullón P. Oxidative stress and dietary fat type in relation to periodontal disease Antioxidants 2015 4 322 344 10.3390/antiox4020322 26783708 37. Varela-López A. Battino M. Bullón P. Quiles J.L. Dietary antioxidants for chronic periodontitis prevention and its treatment: A review on current evidences from animal and human studies ARS Pharm. 2015 56 131 140 10.4321/S2340-98942015000300001 38. Raindi D. Nutrition and Periodontal Disease Dent. Update 2015 43 66 68 27024903 39. Milward M.R. Chapple I.L.C. The role of diet in periodontal disease Clin. Dent. Health 2013 52 18 21 40. Hamasaki T. Kitamura M. Kawashita Y. Ando Y. Saito T. Periodontal disease and percentage of calories from fat using national data J. Periodontal Res. 2016 10.1111/jre.12375 27028150 41. Naqvi A.Z. Buettner C. Phillips R.S. Davis R.B. Mukamal K.J. n-3 Fatty acids and periodontitis in US adults J. Am. Diet. Assoc. 2010 110 1669 1675 10.1016/j.jada.2010.08.009 21034880 42. Requirand P. Gibert P. Tramini P. Cristol J.P. Descomps B. Serum fatty acid imbalance in bone loss: Example with periodontal disease Clin. Nutr. 2000 19 271 276 10.1054/clnu.2000.0107 10952799 43. Iwasaki M. Yoshihara A. Moynihan P. Watanabe R. Taylor G.W. Miyazaki H. Longitudinal relationship between dietary ω-3 fatty acids and periodontal disease Nutrition 2010 26 1105 1109 10.1016/j.nut.2009.09.010 20097537 44. Iwasaki M. Taylor G.W. Moynihan P. Yoshihara A. Muramatsu K. Watanabe R. Miyazaki H. Dietary ratio of n-6 to n-3 polyunsaturated fatty acids and periodontal disease in community-based older Japanese: A 3-year follow-up study Prostaglandins Leukot. Essent. Fat. Acids 2011 85 107 112 10.1016/j.plefa.2011.04.002 21601439 45. Iwasaki M. Manz M.C. Moynihan P. Yoshihara A. Muramatsu K. Watanabe R. Miyazaki H. Relationship between saturated fatty acids and periodontal disease J. Dent. Res. 2011 90 861 867 10.1177/0022034511405384 21505204 46. Campan P. Planchand P.O. Duran D. Pilot study on n-3 polyunsaturated fatty acids in the treatment of human experimental gingivitis J. Clin. Periodontol. 1997 24 907 913 10.1111/j.1600-051X.1997.tb01210.x 9442428 47. Rosenstein E.D. Kushner L.J. Kramer N. Kazandjian G. Pilot study of dietary fatty acid supplementation in the treatment of adult periodontitis Prostaglandins Leukot. Essent. Fat. Acids 2003 68 213 218 10.1016/S0952-3278(02)00272-7 48. Deore G.D. Gurav A.N. Patil R. Shete A.R. Naiktari R.S. Inamdar S.P. Omega 3 fatty acids as a host modulator in chronic periodontitis patients: A randomised, double-blind, palcebo-controlled, clinical trial J. Periodontal Implant Sci. 2014 44 25 32 10.5051/jpis.2014.44.1.25 24616831 49. Martinez G.L. Koury J.C. Martins M.A. Nogueira F. Fischer R.G. Gustafsson A. Figueredo C.M.S. Serum level changes of long chain-polyunsaturated fatty acids in patients undergoing periodontal therapy combined with one year of omega-3 supplementation: A pilot randomized clinical trial J. Periodontal Implant Sci. 2014 44 169 177 10.5051/jpis.2014.44.4.169 25177518 50. El-Sharkawy H. Aboelsaad N. Eliwa M. Darweesh M. Alshahat M. Kantarci A. Hasturk H. van Dyke T.E. Adjunctive treatment of chronic periodontitis with daily dietary supplementation with omega-3 Fatty acids and low-dose aspirin J. Periodontol. 2010 81 1635 1643 10.1902/jop.2010.090628 20572767 51. Naqvi A.Z. Hasturk H. Mu L. Phillips R.S. Davis R.B. Halem S. Campos H. Goodson J.M. van Dyke T.E. Mukamal K.J. Docosahexaenoic acid and periodontitis in adults: A randomized controlled trial J. Dent. Res. 2014 93 767 773 10.1177/0022034514541125 24970858 52. Ramirez-Tortosa M.C. Quiles J.L. Battino M. Granados S. Morillo J.M. Bompadre S. Newman H.N. Bullon P. Periodontitis is associated with altered plasma fatty acids and cardiovascular risk markers Nutr. Metab. Cardiovasc. Dis. 2010 20 133 139 10.1016/j.numecd.2009.03.003 19500957 53. Chapple I.L.C. Potential mechanisms underpinning the nutritional modulation of periodontal inflammation J. Am. Dent. Assoc. 2009 140 178 184 10.14219/jada.archive.2009.0131 19188414 54. Choi J.-S. Park N.-H. Hwang S.-Y. Sohn J.H. Kwak I. Cho K.K. Choi I.S. The antibacterial activity of various saturated and unsaturated fatty acids against several oral pathogens J. Environ. Biol. 2013 34 673 676 24640241 55. Sugano M. Characteristics of fats in Japanese diets and current recommendations Lipids 1996 31 S283 S286 10.1007/BF02637092 8729135 56. Meyer B.J. Mann N.J. Lewis J.L. Milligan G.C. Sinclair A.J. Howe P.R.C. Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids Lipids 2003 38 391 398 10.1007/s11745-003-1074-0 12848284 57. Chapple I.L.C. Milward M.R. Ling-Mountford N. Weston P. Carter K. Askey K. Dallal G.E. de Spirt S. Sies H. Patel D. Adjunctive daily supplementation with encapsulated fruit, vegetable and berry juice powder concentrates and clinical periodontal outcomes: A double-blind RCT J. Clin. Periodontol. 2012 39 62 72 10.1111/j.1600-051X.2011.01793.x 22093005 58. Agabegi S.S. Stern P.J. Bias in research Am. J. Orthop. 2008 37 242 248 18587501 59. Vandenbroucke J.P. von Elm E. Altman D.G. Gøtzsche P.C. Mulrow C.D. Pocock S.J. Poole C. Schlesselman J.J. Egger M. STROBE Initiative Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and elaboration Ann. Intern. Med. 2007 147 W163 W194 10.7326/0003-4819-147-8-200710160-00010-w1 17938389 60. Higgins J.P.T. Altman D.G. Assessing risk of bias in included studies Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series Higgins J.P.T. Green S. John Wiley and Sons, Ltd. Chichester, UK 2008 187 241

PMC005xxxxxx/PMC5000601.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081203ijms-17-01203ArticleGenome-Wide Identification and Characterization of Carboxypeptidase Genes in Silkworm (Bombyx mori) Ye Junhong Li Yi Liu Hua-Wei Li Jifu Dong Zhaoming Xia Qingyou Zhao Ping *Zhu Kun Yan Academic EditorState Key Laboratory of Silkworm Genome Biology, Southwest University, Chongqing 400716, China; [email protected] (J.Y.); [email protected] (Y.L.); [email protected] (H.-W.L.); [email protected] (J.L.); [email protected] (Z.D.); [email protected] (Q.X.)* Correspondence: [email protected]; Tel.: +86-23-6825-088528 7 2016 8 2016 17 8 120310 5 2016 19 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The silkworm (Bombyx mori) is an economically-important insect that can secrete silk. Carboxypeptidases have been found in various metazoan species and play important roles in physiological and biochemical reactions. Here, we analyzed the silkworm genome database and characterized 48 carboxypeptidases, including 34 metal carboxypeptidases (BmMCP1–BmMCP34) and 14 serine carboxypeptidases (BmSCP1–BmSCP14), to better understand their diverse functions. Compared to other insects, our results indicated that carboxypeptidases from silkworm have more family members. These silkworm carboxypeptidases could be divided into four families: Peptidase_M2 carboxypeptidases, Peptidase_M14 carboxypeptidases, Peptidase_S10 carboxypeptidases and Peptidase_S28 carboxypeptidases. Microarray analysis showed that the carboxypeptidases had distinct expression patterns, whereas quantitative real-time PCR demonstrated that the expression level of 13 carboxypeptidases significantly decreased after starvation and restored after re-feeding. Overall, our study provides new insights into the functional and evolutionary features of silkworm carboxypeptidases. carboxypeptidase familyevolutionary analysisexpression patternstarvationBombyx mori ==== Body 1. Introduction In insects, food proteins are preliminary digested by midgut endopeptidases and then by exopeptidases into single free amino acids that are further absorbed by intestinal cells [1]. Endopeptidases, such as trypsin, chymotrypsin, elastase, thermolysin, pepsin, glutamyl endopeptidase, cathepsin B, cathepsin L and neprilysin, are proteolytic peptidases that break the peptide bonds of nonterminal amino acids. On the other hand, exopeptidases, such as aminopeptidases and carboxypeptidases, are applied to the N-terminal and C-terminal peptide bonds, respectively, of the protein to release single free amino acids [2]. Carboxypeptidases are classified into two sub-categories, according to their catalytic mechanism: serine carboxypeptidases with an active serine residue in the active site and metal carboxypeptidases with a metal ion in the active site [3,4]. In general, carboxypeptidases perform important physiological functions, such as food digestion, blood clotting, growth factor production and regulation of biological processes in tissues and organs [5,6,7]. Carboxypeptidases have been identified in many insects, including Simulium vittatum [8], Glossina morsitans [9], Helicoverpa armigera [10], Mamestra configurata [11], Anopheles gambiae [12], Aedes aegypti [13], Trichoplusia ni [1,14] and Triatoma brasiliensis [15]. In 1976, Ward first reported carboxypeptidase activity in the midgut of Tineola bisselliella [16]. In 1997, Broadway identified five types of carboxypeptidases in the midgut of Trichoplusia ni [14] and reported that carboxypeptidase A is eight-fold more active than carboxypeptidase B [1]. Eighteen carboxypeptidase genes were identified in Aedes aegypti, whereas after blood feeding, the expression level of 11 carboxypeptidase genes increased 40-fold in the midgut [7]. In blood-sucking insect populations, the expression of digestive carboxypeptidase genes is promoted by blood meal [8,12]. The silkworm is commonly known as an economically-important insect that secretes silk protein threads to build a cocoon. The study of protein digestion and nutrient absorption in silkworm may reveal the underlying mechanism of silk protein synthesis. Most of the silkworm endopeptidases are serine proteases. In 2010, 51 serine proteases and 92 serine proteases homologs were identified in the silkworm [17]. However, information on carboxypeptidases is still limited. The carboxypeptidase gene MF-CPA from silkworm molting fluid has been previously cloned, characterized [18] and identified in the embryos at the end of the organogenesis [19]. Since silkworm genome sequencing is completed, it is possible to identify the carboxypeptidases family members in the whole genome of silkworm, B. mori [20,21,22]. In the present work, we identified the silkworm carboxypeptidases family and analyzed its genomic organization, expression and molecular evolution in order to reveal additional unknown carboxypeptidases. Moreover, we investigated the characteristics of the silkworm carboxypeptidase family, including phylogeny relationships, as well as spatial and temporal expression profiles. Our results provided preliminary evidence to support the functional roles of carboxypeptidases in food digestion. 2. Results 2.1. Identification and Characterization of the Carboxypeptidase Family By searching the silkworm genome database (SilkDB), 48 members of the carboxypeptidase family were identified; these included, 34 metal carboxypeptidases (nine metal carboxypeptidases containing the Peptidase_M2 domain and twenty five metal carboxypeptidases containing the Peptidase_M14 domain) and 14 serine carboxypeptidases (five serine carboxypeptidases containing the Peptidase_S10 domain and nine serine carboxypeptidases containing the Peptidase_ S28 domain; Table 1). 2.2. Phylogenetic Analysis Carboxypeptidases are usually classified into two sub-categories based on their active sites: serine carboxypeptidases and metal carboxypeptidases [3]. To analyze the relationship between silkworm metal carboxypeptidases and those of other species, a phylogenetic tree was constructed with MEGA 6.0, using metal carboxypeptidase amino acid sequences from silkworm and Caenorhabditis elegans (NP_495012), Aedes aegypti (AAT36725.1, AAT36726.1, AAT36727.1, AAT36728.1, AAT36729.1, AAT36730.1, AAT36731.1, EAT39608, EAT37218, EAT46817, EAT44906, AAT36732.1, AAT36733.1, AAT36734.1, ABO21075.1, ABO21076.1, ABO21077.1 and EAT36173), Danaus plexippus (EHJ78972), Spodoptera frugiperda (ADB96010), Helicoverpa armigera (CAF25190) and Homo sapiens (NP_001859). Metal carboxypeptidases with different domains (Peptidase_M2 and Peptidase_M14) were clustered into different groups based on their evolutionary relationship (Figure 1A). Moreover, phylogenetic analyses showed that metal carboxypeptidases with the Peptidase_M14 domain in silkworm were clustered together with those of other species. We also constructed a phylogenetic tree using serine carboxypeptidase amino acid sequences from silkworm and Triatoma brasiliensis (ABU88379 and ABU88380), Triatoma infestans (AAZ43093), Coptotermes formosanus (AGM32338) and Riptortus pedestris (BAN20175). The results showed that serine carboxypeptidase duplication and divergence led to the separation of Peptidase_S10 carboxypeptidases and Peptidase_S28 carboxypeptidases (Figure 1B) and that silkworm Peptidase_S10 carboxypeptidases have a very close genetic relationship with those of the other insect species. 2.3. Expression Profiles Microarray data from SilkDB were used to analyze the expression profiles of carboxypeptidases in different tissues. As shown in Figure 2A, 44 carboxypeptidases had corresponding oligonucleotide probes, and a heat map was created based on signal intensity values. The majority of Peptidase_M14 carboxypeptidases was highly expressed in the midgut, whereas other high expression levels in the testis. The rest of carboxypeptidases were expressed in various tissues (Figure 2A). The expression profiles of four carboxypeptidase genes (BmMCP10, BmMCP16, BmMCP34 and BmSCP9) that lack specific oligonucleotide probes in the database were replenished by quantitative real-time PCR (qRT-PCR). BmSCP9 was mainly expressed in the fat body epidermis and sex gland. BmMCP10 was highly expressed in the midgut. BmMCP16 was expressed in the sex gland. However, the gene BmMCP34 was not expressed in all tissues at Day 3 of the fifth instar larval stage (Figure 2B). The expression profile of the midgut-specific carboxypeptidases was further examined by qRT-PCR in nine different tissues, including the head, epidermis, testis, ovary, midgut, silk gland, Malpighian tubules, hemocytes and fat body at Day 3 of the fifth instar larval stage (Figure 3). The results were consistent with those of the heat map. Moreover, we created two temporal expression pattern heat maps, one for each gender (Figure S1). Midgut-specific carboxypeptidases were expressed throughout the larval stage, whereas others were only expressed in the pupal stage. 2.4. Carboxypeptidase Expression Profile after Starvation As shown in Figure 4, we tested the influence of feeding, starvation and starvation-refeeding on the expression level of midgut-specific carboxypeptidases (Figure 4). Our results showed that these carboxypeptidases were significantly downregulated after starvation and restored their expression after re-feeding. The expression of carboxypeptidases in the feeding group was stable throughout the experimental period. The expression level of BmSCP12, BmSCP14, BmMCP13, BmMCP20, BmMCP22, BmMCP30 and BmMCP31 reached a peak between 48 and 72 h after feeding, whereas that of other carboxypeptidases, including BmSCP1, BmMCP23 and BmMCP27, reached a peak much earlier. BmMCP14 was highly expressed throughout the starvation and refeeding period, whereas the expression level of BmSCP12, BmMCP13, BmMCP27 and BmMCP31 was significantly increased at the beginning of starvation. This may be the stress response of silkworm midgut. Eleven carboxypeptidases were significantly downregulated after starvation; most of them (BmSCP12, BmSCP14, BmMCP13, BmMCP14, BmMCP20, BmMCP22 and BmMCP30) restored their expression level after refeeding, whereas the rest (BmSCP1, BmMCP23, BmMCP27 and BmMCP31) failed to reach the initial expression level. 3. Discussion Carboxypeptidases are widely found in members of the taxon Metazoa [23,24]. These enzymes are exopeptidases that generally catalyze different reactions based on their active sites. In the present study, 48 carboxypeptidases were identified in the silkworm genome. The silkworm possesses a higher number of carboxypeptidases than do other insects [7,25]; therefore, further study is needed to investigate their unknown functions. The size of silkworm carboxypeptidases ranges from 73 AA–1051 AA. It is generally considered that 40–50 residues are the lower limit of the functional domains and that protein sizes range from 40–50 residues to thousands of residues. We predicted the BmMCP33 (996 AA) and BmSCP11 (1051 AA) domains and found that the former has two Peptidase_M14 domains, whereas the latter is an endomembrane protein 70. Differences in the size of silkworm carboxypeptidases and the combination of the carboxypeptidase domain with other domains suggested that different carboxypeptidases might have different functions. According to the domain features, 34 metal carboxypeptidases were classified into two groups: Peptidase_M2 carboxypeptidases and Peptidase_M14 carboxypeptidases [3]. Carboxypeptidases that contain the Peptidase_M2 domain are known as angiotensin-converting enzymes (ACEs) [26,27]. ACEs are highly important for the regulation of blood pressure [28]. A. gambiae has nine ACE-like genes, but their functions remain unclear [27]. The M14 family is one of the most widely-studied metal carboxypeptidase subunits. The functions of Peptidase_M14 carboxypeptidases are various and diverse, including the digestion of food [29], the processing of bioactive peptides [30] and the metabolism of bacterial cell walls [31]. Fourteen serine carboxypeptidases were classified into two groups: Peptidase_S10 carboxypeptidases and Peptidase_S28 carboxypeptidases [3]. The Peptidase_S10 carboxypeptidase family is active only at acidic pH and is different from most of the other serine peptidase families [32]. There are two types of Peptidase_S10 carboxypeptidases; one (e.g., carboxypeptidase C) that shows preference for hydrophobic residues [33,34,35] and another (e.g., carboxypeptidase D) that shows preference for basic amino acids on either side of the scissile bond, but it is also able to cleave peptides with hydrophobic residues [33,36,37]. Carboxypeptidases in the family S28 suppress angiotensin II by the cleavage of the C-terminal-Pro Phe bond [38]. Additionally, recombinant carboxypeptidases in the family S28 associated with H-kininogen are able to activate plasma prekallikrein [39]. In general, serine carboxypeptidases are considered to play the role of lysosomes and participate in the turnover of proteins. In addition, some of them release amino acids from extracellular proteins and peptides [3]. Phylogenetic analysis of silkworm carboxypeptidases is presented in Figure 1. The tree of metal carboxypeptidase topologies indicated that the divergence and duplication of the Peptidase_M14 carboxypeptidase gene occurred before the separation of B. mori and A. aegypti. Eleven A. aegypti carboxypeptidase genes were induced in the midgut by blood-meal feeding [7], suggesting that B. mori carboxypeptidase genes might be also induced by food intake. The silkworm Peptidase_M14 carboxypeptidase is very conservative in the taxon Metazoa. Human digestive carboxypeptidases (NP_001859) hydrolyze the C-terminal peptide of dietary polypeptide chains [40]. Therefore, the silkworm Peptidase_M14 carboxypeptidase might also have a digestive function. Similarly, the B. mori Peptidase_S28 carboxypeptidase genes and the T. brasiliensis serine carboxypeptidase genes share orthologs in the main branch of the tree. T. brasiliensis use serine carboxypeptidase as a digestive enzyme, suggesting that the B. mori Peptidase_S28 carboxypeptidase might be also a digestive enzyme. Carboxypeptidases play key roles in various physiological and biochemical processes in many insects. In the present study, some Peptidase_M14 carboxypeptidases and the serine carboxypeptidases BmSCP1, BmSCP3, BmSCP12 and BmSCP14 were specifically expressed in the midgut of silkworm. In Anopheles culicifacies, the carboxypeptidase AcCP is specifically expressed in the midgut, whereas in T. brasiliensis, the serine carboxypeptidases tbscp-1 and tbscp-2 are highly expressed in the posterior midgut (small intestine) and lowly expressed in the salivary glands, fat body and anterior midgut (stomach) [15]. These results demonstrated that silkworm carboxypeptidases might participate in digestion in the midgut. Several carboxypeptidases were specifically expressed in the testis and might play important roles in the male reproductive development. Whereas others were widely expressed in various tissues and might perform different functions. For example, the molting fluid carboxypeptidase A (MF-CPA) is identified in the molting fluid of insects at the pupal ecdysis and molting pre-pupal stages. MF-CPA has been proposed to degrade proteins in old epidermal cells and to participate in the recycling of amino acids [18]. The insect digestive tract is divided into three parts: the foregut, midgut and hindgut. The midgut is the most advanced of digestive organs and the most important place for digestion and absorption in insects. The digestion of silkworm larvae includes mechanical digestion and chemical digestion. Under the action of the midgut digestive juice, macromolecules from mulberry, such as carbohydrates, proteins and lipids, are digested into small molecule compounds and absorbed by midgut epithelial cells. Then, compounds are transported to other organs through the hemolymph to provide energy for silkworm growth, development and other life activities. In the present study, starvation could regulate the expression levels of carboxypeptidases in the larval midgut, and re-feeding could restore them to the initial levels. These results suggested that the expression of midgut carboxypeptidases was induced by food intake. Similar results have been also reported in A. aegypti; 11 of A. aegypti carboxypeptidases are upregulated in response to blood meal feeding [7]. The expression profile of induced by starvation and re-feeding in our study was the same as the expression profile of a chymotrypsin-like serine protease in Spodoptera litura [41]. Here, the expression levels of BmSCP1 and BmMCP30 were higher after re-feeding compared to those during normal feeding. Harmonia axyridis can completely compensate the body sizes through accelerated growth [42]. In some animals, the compensatory growth is sometimes faster than the normal growth [43,44], and starvation is applied in animal rearing to obtain economic benefits [45]. In summary, 48 members of the silkworm carboxypeptidase family were identified and characterized in the present study. The expression patterns and two phylogenetic trees of carboxypeptidases were analyzed. We further explored the function of carboxypeptidases, especially of those that were specifically expressed in the midgut. Our findings provided a reference for future studies on Lepidoptera carboxypeptidases. 4. Materials and Methods 4.1. Biological Materials The silkworm strain Dazao (p50) was used in this study. The silkworms were reared on fresh mulberry leaves at 25 °C with 70%–80% relative humidity and a 16-h light/8-h dark cycle in a growth chamber of the State Key Laboratory of Silkworm Genome Biology. Samples from embryonic stages and larval tissues were isolated and stored in liquid nitrogen. 4.2. Identification of the Carboxypeptidase Family in Silkworm SilkDB [46] was used to predict the silkworm carboxypeptidase family. Carboxypeptidase genes from other species were downloaded from GenBank [47]. The BLAST alignment tool was downloaded from the ftp site of the National Center for Biotechnology Information [48]. Carboxypeptidases sequences from other species were used as queries to BLAST against the SilkDB with an E-value threshold of 10−6 [49]. Subsequently, SMART (Simple Modular Architecture Research Tool) [50] and Pfam [51] were used to validate each putative protein. 4.3. Bioinformatics and Phylogeny Analysis of the Silkworm Carboxypeptidase Family The open reading frame (ORF) of carboxypeptidases in B. mori was identified using ORF Finder [52]. The signal peptide was predicted by SignaIP 4.1 [53]. The molecular weight and isoelectric point were predicted using ProtParam [54]. The amino acid sequences of putative carboxypeptidase were aligned using ClustalX [55]. A phylogenetic trees of metal-carboxypeptidases and another of serine-type carboxypeptidases were constructed by the neighbor-joining method with 1000 bootstrap replicates using MEGA 6.0 [56,57]. 4.4. Expression Profiles of Silkworm Carboxypeptidase Genes via Whole-Genome Microarrays A genome-wide oligonucleotide microarray with more than 22,000 probes, including 44 carboxypeptidase-specific oligonucleotide probes, was established as previously described [58]. We identified four carboxypeptidase genes without specific oligonucleotide probes in the database. Microarray data revealed that carboxypeptidase genes had different expression patterns in the tissues of the fifth instar larva at Day 3. Next, diverse tissues, including testis, ovary, head, integument, fat body, midgut, hemocytes, Malpighian tubules, anterior/middle silk gland and posterior silk gland, were collected. To identify the developmental expression patterns, silkworm from 20 different time points (from Day 3 of the fifth instar larval stage to the moth stage) were collected from both genders. Gene expression levels were visualized using GeneCluster 3.0 (University of Tokyo, Tokyo, Japan) [59]. 4.5. Silkworm Starvation Experiment Newly molted fifth instar larvae were divided into three groups, to test whether carboxypeptidases were induced by starvation. Larvae in the feeding group were fed on mulberry leaves throughout the experimental period. Larvae in the starvation groups were starved for 6, 12, 24, 48 and 72 h. Larvae in the starvation-refeeding groups were starved for 12, 24 and 36 h and then fed for 12, 24 and 36 h, respectively [41,60]. The larval midguts in each group were collected for analysis. 4.6. RNA Extraction Total RNA was extracted from all tissues (testis, ovary, head, integument, fat body, midgut, hemocytes, Malpighian tubules, anterior/middle silk gland and posterior silk gland) at Day 3 of the fifth instar larval stage and starvation experiment samples using the Total RNA Kit II (Omega, Norcross, GA, USA), according to the manufacturer’s protocol. Total RNA (2 μg) was reverse transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA). To synthesize the first-strand cDNA, 2 μg of total RNA was mixed with 2 μL of 50 μM oligo (dT) in a total volume of 15 μL. The mixture was briefly spun, heated at 70 °C for 5 min and incubated on ice for 5 min. The mixture was then spun briefly and replaced on ice. After other components (5 μL 5× first strand synthesis buffer, 1 μL dNTP mix, 1 μL RNase inhibitor, 100 U M-MLV reverse transcriptase) were added to the mixture, the reaction mixture was spun briefly and incubated at 42 °C for 1.5 h. The cDNA was then incubated at 92 °C for 10 min and stored at −20 °C. 4.7. qRT-PCR qRT-PCR was performed using the Step-One-Plus™ Real-Time PCR system (Thermo-Fischer Scientific, Waltham, MA, USA) with SYBR® Premix Ex Taq™ II (TaKaRa, Shiga, Japan). PCR conditions were 94 °C for 30 s, followed by 40 cycles at 95 °C for 5 s and 60 °C for 30 s. PCR conditions were 94 °C for 30 s, followed by 40 cycles at 95 °C for 5 s and 60 °C for 30 s. All cDNA samples were normalized using the B. mori eukaryotic translation initiation factor 4A (BmeIF-4a, silkworm microarray probe ID, sw22934; sense primer, 5′-TTCGTACTGGCTCTTCTCGT-3′; antisense primer, 5′-CAAAGTTGATAGCAATTCCCT-3′) as the internal control. Each expression assay was repeated at least three times. The primer sequences of all genes are listed in Table S1. The relative gene expression level was determined by the 2−ΔΔCt method [61]. Statistical significance at p < 0.05 was determined by Student’s t-test using GraphPad [62]. Acknowledgments Our work was supported by the National Natural Science Foundation (Grant No. 31530071; 31472154), the Fundamental Research Funds for the Central Universities (Grant No. XDJK2016E018) and the PhD Start-up Foundation of Southwest University (Grant No. swu113113). Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1203/s1. Click here for additional data file. Author Contributions These studies were conceived of and designed by Junhong Ye, Qingyou Xia and Ping Zhao; Whole experiments were performed by Junhong Ye with help from Yi Li; Hua-Wei Liu contributed to the starvation experiment part; Data analysis and paper writing were done by Junhong Ye, Jifu Li and Zhaoming Dong. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Phylogenetic analysis of carboxypeptidase amino acids. (A) Neighbor-joining tree of the metal carboxypeptidase family. Red lines represent metal carboxypeptidases with the Peptidase_M14 domain, and blue lines represent metal carboxypeptidases with the Peptidase_M2 domain; (B) Neighbor-joining tree for the serine carboxypeptidase family. Red lines represent serine carboxypeptidases with the Peptidase_S10 domain, and blue lines represent serine carboxypeptidases with the Peptidase_S28 domain. The number at each branch represents the percentage of 1000 bootstrap iterations. Values below 50% were omitted. Abbreviations: Rha, Rhabditida; Pri, Primates; Lep, Lepidoptera; Dip, Diptera; Hem, Hemiptera; Dic, Dictyoptera. Figure 2 (A) Microarray analysis of carboxypeptidase expression in different silkworm tissues. Each column represents an organ or tissue, including testis, ovary, head, integument, fat body, midgut, hemocytes, Malpighian tubules, anterior/middle silk gland and posterior silk gland. Red represents high expression, and green represents low expression; (B) Expression patterns of the silkworm carboxypeptidases BmSCP9, BmMCP10 and BmMCP16. Abbreviations: He, head; Ep, epidermis; Te, testis; Ov, ovary; Mi, midgut; Si, silk gland; Ma, Malpighian tubules; Ha, hemocytes; Fa, fat body. Figure 3 Expression patterns of midgut-specific carboxypeptidases at Day 3 of the fifth instar. Abbreviations: He, head; Ep, epidermis; Te, testis; Ov, ovary; Mi, midgut; Si, silk gland; Ma, Malpighian tubules; Ha, hemocytes; Fa, fat body. *** p < 0.001. Figure 4 Expression patterns of midgut-specific carboxypeptidases after feeding, starvation and starvation-refeeding. Differences among groups were identified by Student’s t-test. * p < 0.05; ** p < 0.01; *** p < 0.001. Abbreviations: F, feeding; S, starvation; SR, starvation-refeeding. ijms-17-01203-t001_Table 1Table 1 Characterization of silkworm carboxypeptidases. Gene Name in SilkDB Chromosome Scaffold Position Strand Domain Size (AA) Probe MW/Da pI Signal Peptide BmMCP1 BGIBMGA002530-TA 9 nscaf2511 5048133–5052368 + Peptidase_M2 73 sw03136 8984.4 9.3 no BmMCP2 BGIBMGA006234-TA 6 nscaf2851 1576285–1581353 − Peptidase_M2 409 sw03231 46,754.2 4.89 no BmMCP3 BGIBMGA002527-TA 9 nscaf2511 4990965–5002524 + Peptidase_M2 649 sw14172 74,229.8 5.11 yes BmMCP4 BGIBMGA006539-TA 6 nscaf2853 6801913–6813068 + Peptidase_M2 647 sw11710 74,185.1 5.39 no BmMCP5 BGIBMGA002529-TA 9 nscaf2511 5030694–5035485 + Peptidase_M2 173 sw20193 – – no BmMCP6 BGIBMGA002531-TA 9 nscaf2511 5192136–5202359 + Peptidase_M2 153 sw11006 17,724 7.01 no BmMCP7 BGIBMGA003228-TA 2 nscaf2623 1029259–1049209 − Peptidase_M2 711 sw17925 82,533.6 9.06 no BmMCP8 BGIBMGA002526-TA 9 nscaf2511 4965783–4980141 + Peptidase_M2 648 sw00643 74,913.1 5.34 yes BmMCP9 BGIBMGA009693-TA 2 nscaf2964 820517–835616 − Peptidase_M2 535 sw19815 62,129.8 7.32 no BmMCP10 BGIBMGA009487-TA 14 nscaf2953 517470–526524 + Peptidase_M14 428 – 48,175.4 6.06 yes BmMCP11 BGIBMGA004799-TA 25 nscaf2818 2043455–2064130 − Peptidase_M14 468 sw03973 53,767.3 9.18 no BmMCP12 BGIBMGA004800-TA 25 nscaf2818 2024210–2029825 − Peptidase_M14 344 sw20598 38,598.6 6.55 no BmMCP13 BGIBMGA013275-TA 16 nscaf3063 3241878–3245274 + Peptidase_M14 346 sw18892 39,898 5.24 yes BmMCP14 BGIBMGA013276-TA 16 nscaf3063 3256098–3269527 + Peptidase_M14 445 sw05014 50,709.9 5.41 yes BmMCP15 BGIBMGA004801-TA 25 nscaf2818 1996486–2007642 − Peptidase_M14 427 sw18974 48,293.5 5.78 no BmMCP16 BGIBMGA008973-TA 3 nscaf2930 2150368–2156343 + Peptidase_M14 365 – 41,363.1 9.22 no BmMCP17 BGIBMGA001892-TA 19 nscaf2204 1727896–1729970 − Peptidase_M14 298 sw09352 33,757.5 6.37 no BmMCP18 BGIBMGA001891-TA 19 nscaf2204 1730646–1734010 − Peptidase_M14 418 sw14852 48,425.2 9.24 no BmMCP19 BGIBMGA008975-TA 3 nscaf2930 2171627–2175156 + Peptidase_M14 267 sw07649 30,547.4 5.31 no BmMCP20 BGIBMGA009477-TA 14 nscaf2953 493144–498033 − Peptidase_M14 429 sw02069 48,772.3 6.08 yes BmMCP21 BGIBMGA004797-TA 25 nscaf2818 2109792–2113829 − Peptidase_M14 383 sw11839 43,402.5 5.51 no BmMCP22 BGIBMGA009478-TA 14 nscaf2953 480242–485505 − Peptidase_M14 394 sw00546 44,578.6 5.39 no BmMCP23 BGIBMGA004798-TA 25 nscaf2818 2099617–2105069 − Peptidase_M14 395 sw03321 44,319.1 4.8 no BmMCP24 BGIBMGA006871-TA 10 nscaf2859 1283815–1295739 + Peptidase_M14 604 sw12842 68,595.4 6.45 yes BmMCP25 BGIBMGA008976-TA 3 nscaf2930 2176287–2190645 + Peptidase_M14 365 sw03831 41,702.1 6.15 no BmMCP26 BGIBMGA008910-TA 3 nscaf2930 1816975–1823616 − Peptidase_M14 479 sw07559 53,979.4 5.75 yes BmMCP27 BGIBMGA004830-TA 25 nscaf2818 155397–165470 − Peptidase_M14 670 sw03365 75,788.8 8.55 no BmMCP28 BGIBMGA006715-TA 10 nscaf2855 3350086–3361536 + Peptidase_M14 371 sw20772 42,903 5.3 no BmMCP29 BGIBMGA001890-TA 19 nscaf2204 1746161–1749490 − Peptidase_M14 353 sw20818 40,328.8 6.09 no BmMCP30 BGIBMGA009486-TA 14 nscaf2953 512395–515181 + Peptidase_M14 409 sw10048 46,594.4 5.81 no BmMCP31 BGIBMGA009476-TA 14 nscaf2953 546619–551421 − Peptidase_M14 397 sw12317 45,444.3 6.06 no BmMCP32 BGIBMGA000307-TA 22 nscaf1681 1661872–1673460 − Peptidase_M14 483 sw11693 54,330.3 5.73 yes BmMCP33 BGIBMGA012807-TA 16 nscaf3058 6657227–6684446 − Peptidase_M14 996 sw17444 11,1592 5.73 no BmMCP34 BGIBMGA012806-TA 16 nscaf3058 6684947–6690407 − Peptidase_M14 294 – 33,318.1 5.55 yes BmSCP1 BGIBMGA012452-TA 9 nscaf3045 2246700–2252046 − Peptidase_S28 352 sw12542 40,622 4.69 yes BmSCP2 BGIBMGA003579-TA 5 nscaf2674 1284051–1285310 − Peptidase_S28 419 sw13803 48,709.9 4.96 no BmSCP3 BGIBMGA008167-TA 24 nscaf2891 256973–274367 + Peptidase_S28 282 sw14786 31,433.3 4.83 no BmSCP4 BGIBMGA000549-TA 1 nscaf1690 3748863–3757879 − Peptidase_S28 439 sw20204 49,930.5 6.08 yes BmSCP5 BGIBMGA013534-TA 5 nscaf3075 927568–931683 + Peptidase_S28 383 sw14860 42,863.6 6 no BmSCP6 BGIBMGA003110-TA 4 nscaf2589 2567114–2581357 + Peptidase_S10 374 sw00555 43,079.3 6.05 no BmSCP7 BGIBMGA012773-TA 16 nscaf3058 8186538–8187953 − Peptidase_S10 471 sw04192 53,120.8 5.56 yes BmSCP8 BGIBMGA003111-TA 4 nscaf2589 2593763–2602101 + Peptidase_S10 402 sw17614 45,191.6 5.11 no BmSCP9 BGIBMGA003109-TA 4 nscaf2589 2546587–2560291 + Peptidase_S10 285 – 32,834.9 5.7 no BmSCP10 BGIBMGA013085-TA 16 nscaf3058 8193345–8194772 + Peptidase_S10 475 sw14878 54,378 8.46 yes BmSCP11 BGIBMGA003112-TA 4 nscaf2589 2615158–2633253 + Peptidase_S10 1051 sw01972 11,9508 5.66 no BmSCP12 BGIBMGA010348-TA 12 nscaf2990 809983–812392 + Peptidase_S10 403 sw06212 46,108.2 4.99 no BmSCP13 BGIBMGA006502-TA 6 nscaf2853 4266715–4273371 + Peptidase_S10 376 sw15928 42,942.2 8.8 no BmSCP14 BGIBMGA010349-TA 12 nscaf2990 814696–818250 + Peptidase_S10 365 sw06213 40,585.3 4.95 no ==== Refs References 1. Wang P. Li G. Kain W. Characterization and cDNA cloning of midgut carboxypeptidases from Trichoplusia ni Insect Biochem. Mol. Biol. 2004 34 831 843 10.1016/j.ibmb.2004.05.003 15262287 2. Vendrell J. Querol E. Avilés F.X. Metallocarboxypeptidases and their protein inhibitors: Structure, function and biomedical properties Biochim. Biophys. Acta 2000 1477 284 298 10.1016/S0167-4838(99)00280-0 10708864 3. Ferreira C. Rebola K.G. Cardoso C. Bragatto I. Ribeiro A.F. Terra W.R. Insect midgut carboxypeptidases with emphasis on S10 hemipteran and M14 lepidopteran carboxypeptidases Insect Mol. Biol. 2015 24 222 239 10.1111/imb.12151 25488368 4. Skidgel R.A. Erdös E.G. Cellular carboxypeptidases Immunol. Rev. 1998 161 129 141 10.1111/j.1600-065X.1998.tb01577.x 9553770 5. Fricker L.D. Activation and membrane binding of carboxypeptidase E J. Cell. Biochem. 1988 38 279 289 10.1002/jcb.240380407 3149282 6. Walter M.F. Zeineh L.L. Black B.C. Mcivor W.E. Wright T.R.F. Biessmann H. Catecholamine metabolism and in vitro induction of premature cuticle melanization in wild type and pigmentation mutants of Drosophila melanogaster Arch. Insect Biochem. Physiol. 1996 31 219 233 10.1002/(SICI)1520-6327(1996)31:2<219::AID-ARCH9>3.0.CO;2-U 8580497 7. Isoe J. Zamora J. Miesfeld R.L. Molecular analysis of the Aedes aEgypti carboxypeptidase gene family Insect Biochem. Mol. Biol. 2009 39 68 73 10.1016/j.ibmb.2008.09.006 18977440 8. Ramos A. Mahowald A. Jacobs-Lorena M. Gut-specific genes from the black fly Simulium vittatum encoding trypsin-like and carboxypeptidase-like proteins Insect Mol. Biol. 1993 1 149 163 10.1111/j.1365-2583.1993.tb00116.x 8269093 9. Yan J. Cheng Q. Li C.B. Aksoy S. Molecular characterization of three gut genes from Glossina morsitans morsitans: Cathepsin B, zinc-metalloprotease and zinc-carboxypeptidase Insect Mol. Biol. 2002 11 57 65 10.1046/j.0962-1075.2001.00308.x 11841503 10. Bown D.P. Midgut carboxypeptidase from Helicoverpa armigera (Lepidoptera: Noctuidae) larvae: Enzyme characterisation, cDNA cloning and expression Insect Biochem. Mol. Biol. 1998 28 739 749 10.1016/S0965-1748(98)00067-8 9807221 11. Hegedus D. Baldwin D. O’Grady M. Braun L. Gleddie S. Sharpe A. Lydiate D. Erlandson M. Midgut proteases from Mamestra configurata (Lepidoptera: Noctuidae) larvae: Characterization, cDNA cloning, and expressed sequence tag analysis Arch. Insect Biochem. Physiol. 2003 53 30 47 10.1002/arch.10084 12701112 12. Edwards M.J. Lemos F.J.A. Donnelly-DOman M. Jacobs-Lorena M. Rapid induction by a blood meal of a carboxypeptidase gene in the gut of the mosQuito Anopheles gambiae Insect Biochem. Mol. Biol. 1997 27 1063 1072 10.1016/S0965-1748(97)00093-3 9569647 13. Lavazec C. Bonnet S. Thiery I. Boisson B. Bourgouin C. cpbAg1 encodes an active carboxypeptidase B expressed in the midgut of Anopheles gambiae Insect Mol. Biol. 2005 14 163 174 10.1111/j.1365-2583.2004.00541.x 15796749 14. Broadway R.M. Dietary regulation of serine proteinases that are resistant to serine proteinase inhibitors J. Insect Physiol. 1997 43 855 874 10.1016/S0022-1910(97)00028-0 12770497 15. Waniek P.J. Araujo C.A. Momoli M.M. Azambuja P. Jansen A.M. Genta F.A. Serine carboxypeptidases of Triatoma brasiliensis (Hemiptera, Reduviidae): Sequence characterization, expression pattern and activity localization J. Insect Physiol. 2014 63 9 20 10.1016/j.jinsphys.2014.02.003 24548612 16. Ward C.W. Properties of the major carboxypeptidase in the larvae of the webbing clothes moth, Tineola bisselliella Biochim. Biophys. Acta 1976 429 564 572 10.1016/0005-2744(76)90304-1 4137 17. Zhao P. Wang G.H. Dong Z.M. Duan J. Xu P.Z. Cheng T.C. Xiang Z.H. Xia Q.Y. Genome-wide identification and expression analysis of serine proteases and homologs in the silkworm Bombyx mori BMC Genom. 2010 11 1 11 10.1186/1471-2164-11-405 20576138 18. Ote M. Mita K. Kawasaki H. Daimon T. Kobayashi M. Shimada T. Identification of molting fluid carboxypeptidase A (MF-CPA) in Bombyx mori Comp. Biochem. Physiol. 2005 141 314 322 10.1016/j.cbpc.2005.04.005 15936966 19. Li J.Y. Moghaddam S.H. Chen J.E. Chen M. Zhong B.X. Shotgun proteomic analysis on the embryos of silkworm Bombyx mori at the end of organogenesis Insect Biochem. Mol. Biol. 2010 40 293 302 10.1016/j.ibmb.2010.01.008 20138214 20. Xia Q. Zhou Z. Lu C. Cheng D. Dai F. Li B. Zhao P. Zha X. Cheng T. Chai C. A draft sequence for the genome of the domesticated silkworm (Bombyx mori ) Science 2004 306 1937 1940 15591204 21. Xia Q. Guo Y. Zhang Z. Li D. Xuan Z. Li Z. Dai F. Li Y. Cheng D. Li R. Complete resequencing of 40 genomes reveals domestication events and genes in silkworm (Bombyx ) Science 2009 326 433 436 10.1126/science.1176620 19713493 22. Xia Q. Li S. Feng Q. Advances in silkworm studies accelerated by the genome sequencing of Bombyx mori Annu. Rev. Entomol. 2013 59 513 536 10.1146/annurev-ento-011613-161940 24160415 23. Stein L.D. Bao Z. Blasiar D. Blumenthal T. Brent M.R. Chen N. Chinwalla A. Clarke L. Clee C. Coghlan A. The genome sequence of Caenorhabditis briggsae : A platform for comparative genomics PLoS Biol. 2003 1 1203 10.1371/journal.pbio.0000045 14624247 24. Sapio M.R. Fricker L.D. Crboxypeptidases in disease: Insights from peptidomic studies Proteom. Clin. Appl. 2014 8 327 337 10.1002/prca.201300090 24470285 25. Zhang K.X. Tan L. Pellegrini M. Zipursky S.L. Mcewen J.M. Rapid changes in the translatome during the conversion of growth cones to synaptic terminals Cell Rep. 2016 14 1258 1271 10.1016/j.celrep.2015.12.102 26832407 26. Houard X. Williams T.A. Michaud A. Dani P. Isaac R.E. Shirras A.D. Coates D. Corvol P. The Drosophila melanogaster -related angiotensin-I-converting enzymes acer and ance–Distinct enzymic characteristics and alternative expression during pupal development Eur. J. Biochem. 1998 257 599 606 10.1046/j.1432-1327.1998.2570599.x 9839949 27. Burnham S. Smith J.A. Lee A.J. Isaac R.E. Shirras A.D. The angiotensin-converting enzyme (ACE) gene family of Anopheles gambiae BMC Genom. 2005 6 1203 10.1186/1471-2164-6-172 16329762 28. Fraga-Silva R.A. Costa-Fraga F.P. Murça T.M. Moraes P.L. Augusto M.L. Lautner R.Q. Castro C.H. Soares C.M.A. Borges C.L. Ana Paula N. Angiotensin-converting enzyme 2 activation improves endothelial function Hypertension 2013 61 1233 1238 10.1161/HYPERTENSIONAHA.111.00627 23608648 29. Kumar A. Sharma A. Sharma R. Gakhar S.K. Identification, characterization and analysis of expression of gene encoding carboxypeptidase A in Anopheles culicifacies A (Diptera: Culicidae) Acta Trop. 2014 139 123 130 10.1016/j.actatropica.2014.05.022 24910441 30. Cool D.R. Loh Y.P. Carboxypeptidase E is a sorting receptor for prohormones: Binding and kinetic studies Mol. Cell. Endocrinol. 1998 139 7 13 10.1016/S0303-7207(98)00081-1 9705069 31. Hourdou M.L. Guinand M. Vacheron M.J. Michel G. Denoroy L. Duez C. Englebert S. Joris B. Weber G. Ghuysen J.M. Characterization of the sporulation-related γ-d -glutamyl-(l )meso-diaminopimelic-acid-hydrolysing peptidase I of Bacillus sphaericus NCTC 9602 as a member of the metallo(zinc) carboxypeptidase A family. Modular design of the protein Biochem. J. 1993 292 563 570 10.1042/bj2920563 8503890 32. Jung G. Ueno H. Hayashi R. Proton-relay system of carboxypeptidase Y as a sole catalytic site: Studies on mutagenic replacement of his 397 J. Biochem. 1998 124 446 450 10.1093/oxfordjournals.jbchem.a022133 9685740 33. Breddam K. Serine carboxypeptidases. A review Carlsberg Res. Commun. 1986 51 83 128 10.1007/BF02907561 34. Valls L.A. Hunter C.P. Rothman J.H. Stevens T.H. Protein sorting in yeast: The localization determinant of yeast vacuolar carboxypeptidase Y resides in the propeptide Cell 1987 48 887 897 10.1016/0092-8674(87)90085-7 3028649 35. Jackman H.L. Morris P.W. Deddish P.A. Skidgel R.A. Erdös E.G. Inactivation of endothelin I by deamidase (lysosomal protective protein) J. Biol. Chem. 1992 267 2872 2875 1737744 36. Breddam K. Sørensen S.B. Svendsen I. Primary structure and enzymatic properties of carboxypeptidase II from wheat bran Carlsberg Res. Commun. 1987 52 297 311 10.1007/BF02907172 37. Liao D.I. Breddam K. Sweet R.M. Bullock T. Remington S.J. Refined atomic model of wheat serine carboxypeptidase II at 2.2-A resolution Biochemistry 1992 31 9796 9812 10.1021/bi00155a037 1390755 38. Barbara L. Pryor K.A.D. Wu J.K. Frank M. Patel R.A. Craik C.S. Ellman J.A. Cummings R.T. Thornberry N.A. Catalytic properties and inhibition of proline-specific dipeptidyl peptidases II, IV and VII Biochem. J. 2003 371 525 532 12529175 39. Shariat-Madar Z. Mahdi F. Schmaier A.H. Recombinant prolylcarboxypeptidase activates plasma prekallikrein Blood 2004 103 4554 4561 10.1182/blood-2003-07-2510 14996700 40. Witt H. Beer S. Rosendahl J. Chen J.M. Chandak G.R. Masamune A. Bence M. Szmola R. Oracz G. Milan Macek J. Variants in CPA1 are strongly associated with early-onset chronic pancreatitis Nat. Genet. 2013 45 1216 1220 10.1038/ng.2730 23955596 41. Chun Z. Daohua Z. Sichun Z. Lin L. Shu T. Li Y. Songnian H. Qili F. A chymotrypsin-like serine protease cDNA involved in food protein digestion in the common cutworm, Spodoptera litura: Cloning, characterization, developmental and induced expression patterns, and localization J. Insect Physiol. 2010 56 788 799 20149796 42. Dmitriew C.D. Cooray M.C. Rowe L.R. Effects of early resource-limiting conditions on patterns of growth, growth efficiency, and immune function at emergence in a damselfly (Odonata: Coenagrionidae) Can. J. Zool. 2007 85 310 318 10.1139/Z07-004 43. Larsson A. Lewander K. Metabolic effects of starvation in the eel, Anguilla Anguilla L. Comp. Biochem. Physiol. Part A 1973 44 367 374 10.1016/0300-9629(73)90489-1 44. Jobling M. Effects of starvation on proximate chemical composition and energy utilizaton of plaice, Pleuronectes platessa L. J. Fish Biol. 2006 17 325 334 10.1111/j.1095-8649.1980.tb02766.x 45. MaChado C.R. Garofaloj M. Effects of starvation, refeeding, and insulin on energy-linked metabolic processes in catfish (Rhamdia hilarii ) adapted to a carbohydrate-rich diet Gen. Comp. Endocrinol. 1988 71 3429 3437 10.1016/0016-6480(88)90272-9 46. SilkDB Available online: http://www.silkdb.org/silkdb/ (accessed on 9 January 2016) 47. GenBank Available online: http://www.ncbi.nlm.nih.gov/Genbank/ (accessed on 10 January 2016) 48. The ftp site of the National Center for Biotechnology Information Available online:ftp://ftp.ncbi.nih.gov/blast/ (accessed on 10 January 2016) 49. Enright A.J. van Dongen S. Ouzounis C.A. An efficient algorithm for large-scale detection of protein families Nucleic Acids Res. 2002 30 1575 1584 10.1093/nar/30.7.1575 11917018 50. SMART Available online: http://smart.embl-heidelberg.de/ (accessed on 14 January 2016) 51. Pfam Available online: http://pfam.sanger.ac.UK/ (accessed on 14 January 2016) 52. ORF Finder Available online: http://www.ncbi.nlm.nih.gov/gorf/gorf.html (accessed on 15 January 2016) 53. SignaIP 4.1 Available online: http://www.cbs.dtu.dk/services/SignalP/ (accessed on 15 January 2016) 54. ProtParam Available online: http://web.expasy.org/protparam/ (accessed on 15 January 2016) 55. Thompson J.D. Higgins D.G. Gibson T.J. CLUSTAL W: Improving the sensitiv ity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice Nucleic Acids Res. 1994 22 4673 4680 10.1093/nar/22.22.4673 7984417 56. Tamura K. Stecher G. Peterson D. Filipski A. Kumar S. MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0 Mol. Biol. Evol. 2013 30 2725 2729 10.1093/molbev/mst197 24132122 57. Tamura K. Dudley J. Nei M. Kumar S. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0 Mol. Biol. Evol. 2007 24 1596 1599 10.1093/molbev/msm092 17488738 58. Xia Q. Cheng D. Duan J. Wang G. Cheng T. Zha X. Liu C. Zhao P. Dai F. Zhang Z. Microarray-based gene expression profiles in multiple tissues of the domesticated silkworm, Bombyx mori Genome Biol. 2007 10.1186/gb-2007-8-8-r162 17683582 59. Golub T.R. Slonim D.K. Tamayo P. Huard C. Gaasenbeek M. Mesirov J.P. Coller H. Loh M.L. Downing J.R. Caligiuri M.A. Molecular classification of cancer: Class discovery and class prediction by gene expression monitoring Science 1999 286 531 537 10.1126/science.286.5439.531 10521349 60. Broehan G. Kemper M. Driemeier D. Vogelpohl I. Merzendorfer H. Cloning and expression analysis of midgut chymotrypsin-like proteinases in the tobacco hornworm J. Insect Physiol. 2008 54 1243 1252 10.1016/j.jinsphys.2008.06.007 18634789 61. Livak K.J. Schmittgen T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2 method Methods 2001 25 402 408 10.1006/meth.2001.1262 11846609 62. GraphPad Available online: http://www.graphpad.com/quickcalcs/ttest1.cfm (accessed on 25 February 2016)

PMC005xxxxxx/PMC5000602.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081204ijms-17-01204ArticleGenome-Wide Identification and Expression Analysis of Two-Component System Genes in Tomato He Yanjun 1†Liu Xue 1†Ye Lei 1Pan Changtian 1Chen Lifei 1Zou Tao 12Lu Gang 12*Zhu Jianhua Academic Editor1 Department of Horticulture, Zhejiang University, Hangzhou 310058, China; [email protected] (Y.H.); [email protected] (X.L.); [email protected] (L.Y.); [email protected] (C.P.); [email protected] (L.C.); [email protected] (T.Z.)2 Key Laboratory of Horticultural Plant Growth, Development and Biotechnology, Agricultural Ministry of China, Hangzhou 310058, China* Correspondence: [email protected]; Tel.: +86-571-8898-2277† These authors contributed equally to this work. 26 7 2016 8 2016 17 8 120430 5 2016 15 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The two-component system (TCS), which comprises histidine kinases (HKs), phosphotransfers (HPs), and response regulator proteins (RRs), plays pivotal roles in regulating plant growth, development, and responses to biotic and abiotic stresses. TCS genes have been comprehensively identified and investigated in various crops but poorly characterized in tomato. In this work, a total of 65 TCS genes consisting of 20 HK(L)s, six HPs, and 39 RRs were identified from tomato genome. The classification, gene structures, conserved domains, chromosome distribution, phylogenetic relationship, gene duplication events, and subcellular localization of the TCS gene family were predicted and analyzed in detail. The amino acid sequences of tomato TCS family members, except those of type-B RRs, are highly conserved. The gene duplication events of the TCS family mainly occurred in the RR family. Furthermore, the expansion of RRs was attributed to both segment and tandem duplication. The subcellular localizations of the selected green fluorescent protein (GFP) fusion proteins exhibited a diverse subcellular targeting, thereby confirming their predicted divergent functionality. The majority of TCS family members showed distinct organ- or development-specific expression patterns. In addition, most of TCS genes were induced by abiotic stresses and exogenous phytohormones. The full elucidation of TCS elements will be helpful for comprehensive analysis of the molecular biology and physiological role of the TCS superfamily. tomatotwo-component systemphylogenyevolutionexpression profiles ==== Body 1. Introduction A two-component system (TCS) via phosphorylation between histidine and aspartic-acid residues was first identified in bacteria [1,2]. The TCS system in bacteria consists of a membrane-associated histidine protein kinase (HK) and a cytoplasmic response regulator (RR) with a receiver (REC) domain. HK proteins sense environmental signals and autophosphorylate the histidine residue (H) of the HK domain; the phosphate is then transferred to an aspartate residue (D) of the REC domain of the RR protein [1,2]. A complex TCS signaling system has been identified in eukaryotic species, including higher plants [3]. Plant TCS components typically consist of three signal elements: hybrid HKs, histidine-containing phosphotransfers (HPs), and RRs [4,5]. Arabidopsis cytokinin signaling is a typical representative of TCS signal systems. Three transmembrane histidine kinases, namely, AHK2, AHK3, and AHK4 function as cytokinin receptors and negatively respond to stresses in Arabidopsis cytokinin signaling [3,4,5,6]. These kinases perceive stimulus and are autophosphorylated at a conserved histidine residue in transmitter domain. The phosphory groups are then transferred to HPs at a conserved aspartate residue. Finally, HPs transmit the signal to receiver domain in type-B RRs, which could function as transcription factor to response to various environmental signals via numerous cis-elements in the promoter of type-A ARRs and further regulate the expression of downstream stress-related genes [3,4,5,6]. Take advantage of available whole genomic sequences, various TCS genes have been successfully identified and investigated in several plant species, including Arabidopsis [4], rice [7], maize [8], soybean [9], and Chinese cabbage [10]. Plant TCS elements play vital roles in responses to abiotic stress, particularly drought, high salinity, and high or low temperature. Most of TCS elements in Arabidopsis interact with ABA to participate in drought, salt, and low temperature stresses [11,12,13,14,15,16,17,18,19]. The expression level of AHP1, AHP2, and AHP3 in Arabidopsis are significantly repressed by heat stress [20]. In rice, OsAHP1/2 knockdown seedlings respond to salt and drought stresses in different patterns [21]. OsHK3 participates in ABA-induced antioxidant defense [22]. In soybean, the expressions of most TCS genes are sensitive to dehydration [23]. The roles of some tomato TCS members involved in stress responses have been studied. The pollens of tomato never-ripe (Nr) mutant, a histidine kinase mutant, are more sensitive to heat stress via affecting pollen carbohydrate metabolism [24]. Some phytochromes (PHYs), function as histidine kinases and participate in response to drought stress [25]. Tomato, as a model of fleshy fruit plant, is an economically important fruit crop grown worldwide. The reproductive development of tomato is susceptible to various adverse environments, resulting in reduced yield and quality. TCS plays important roles in signal transduction involved in stress responses and plant development. The tomato ethylene receptor and PHY subfamily, which both belong to HK(L) family have been identified and investigated [26,27]. Some tomato ethylene receptor and PHY elements have been proved to play important roles in plant reproduction development. Transgenic plants with reduced ethylene receptor LeETR4 expression levels enhance flower senescence, and affect fruit set [28]. Moreover, the PHY subfamily genes can modulate carotenoid levels and regulate the time required for phase transition during fruit ripening [29]. However, tomato TCS genes have not been systematically investigated. In this study, the putative TCS elements in tomato were identified through in silico analysis. The classification, chromosome distribution, and evolutionary relationships of the TCS gene family were predicted and analyzed. Subcellular localizations were predicted and verified based on the transformation of onion epidermal cells. The expression profiles of some identified TCS genes were determined using quantitative real-time PCR analysis (qRT-PCR) to assess their responses to different abiotic stresses and plant hormones. Our comprehensive analyses of the TCS elements in tomato may provide a framework for future studies to elucidate the function of the TCS family genes in stress tolerance and hormone response in tomato. 2. Results 2.1. Identification of the TCS Genes in Tomato BLASTP searches were performed in Sol GenomiSl Network (http://solgenomiSl.net/) to explore putative TCS members in tomato by employing 280 TCS protein sequences from Arabidopsis [4], rice [7], maize [8], soybean [9], wheat [30], and Chinese cabbage [10] as queries. A total of 211 non-redundant sequences including 71 HK(L)s, eight HPs, and 132 RRs putative hits in tomato genome database were identified. The putative TCS proteins were searched with HMMER 3.0 by using the global HMM profile of the TCS characteristic domains. A total of 55, seven, and 56 non-redundant putative HK(L)s, HPs, and RRs were identified, respectively. The sequences obtained by above two methods were compared to remove redundancy. The non-redundant proteins were filtered further using Pfam and SMART based on the presence of structural characteristics and conserved domains of TCS elements. Finally, 65 TCS members consisting of 20 HK(L)s, six HPs, and 39 RRs were confirmed in tomato. All tomato TCS members were named according to the homologous genes in Arabidopsis. This nomenclature has also been used in soybean and Chinese cabbage [9,10]. TCS genes have been intensively studied in some model plant species and important crops. TCS gene family in tomato contains 65 members, which is bigger than that of all reported species except Glycine max (98), and Brassica rapa (85) (Table 1). 2.2. HK Proteins in Tomato Twenty HK(L) proteins were identified in tomato and categorized into nine HKs and 11 HK-likes (HKLs) based on the presence of the conserved His-kinase transmitter (HK) domain. Nine HK proteins were further classified into five subgroups: three cytokinin receptor-like SlHKs, three ethylene receptor-like SlHKs, one CKI1-like SlHK, one CKI2/AHK5-like SlHKs, and one AHK1-like SlHK. All of these HKs possess a conserved HK domain that contains five conserved signature motifs, namely, H, N, G1, F, and G2 [4]. Additionally, all 11 SlHKLs were divided into three subgroups: five PHY-like SlHKLs, two PDK-like SlHKLs, and other four ethylene receptor-like SlHKLs, in which the H sites of the HK domain is replaced by other amino acids (Table S1). Three cytokinin receptor-like SlHKs, namely SlHK4, SlHK5, and SlHK6 exhibit conserved protein structure and high sequence identity (62%–64%) with Arabidopsis AHK4–AHK6. Gene structure analysis showed that SlHK4, SlHK5, and SlHK6 possess 9–10 introns. These SlHK proteins contain two conserved motifs (motifs 1 and 4), as identified by MEME, as well as four conserved domains, namely, HK, REC, CHASE, and transmembrane (TM) domains, as recognized by Pfam and SMART online tools (Figure 1 and Figure S1). Multiple sequence alignment showed that CHASE domains are highly conserved among tomato cytokinin receptors (Figure S2), and CHASE domain is crucial for proteins to recognize and bind cytokinin [4]. Tomato SlHK1 has 38% identity with CKI1 in Arabidopsis and CKI1, which is involved in cytokinin signaling and development of female gametophytes in Arabidopsis [5]. Tomato ethylene receptors and PHY members were identified in previous study [26,27]. Tomato ethylene receptors contain seven members (SlHK7-SlHK9 and SlHKL1-SlHKL4), which contain a C2H2-type zinc-finger (C2H2) domain as an ethylene-binding domain (Table S1). Among these genes, SlHKL1-SlHKL4 contain one or two introns, similar exon-intron architecture were also found in Arabidopsis homologous genes. However, SlHK7 and SlHK9 have five or six introns (Figure 1, Table S1). It should be noted that there is no intron in SlHK8 gene, although it has 90% amino sequence with SlHK7 (Table S2). In addition, we identified a new ethylene receptor EIN4-like gene, namely SlHKL2, which shows 59% sequence identity with Arabidopsis EIN4. SlHKL2 contains characteristic domains, namely, TM, GAF, HKL, and REC. Tomato PHY genes were previously named PHYA, PHYB1, PHYB2, PHYE, and PHYF [27]. The tomato PHYF is a homolog of Arabidopsis PHYC. In the present study, these genes which were renamed as SlHKL5–9, show 59% to 78% sequence similarity to their counterparts in Arabidopsis and all genes contain GAF, PHY, PAS, and HKL domains (Figure S2). The PHY members in Arabidopsis function as red or far-red light photoreceptors and participate in various photomorphogenic processes [33]. 2.3. HP Proteins in Tomato We identified six HP proteins in tomato with four authentic and two pseudo-HPs. Each of the six SlHP genes contains five or six introns except SlPHP2. All SlHP proteins exhibit 53% to 70% identity to their homologs, namely, AHP1, AHP4, and APHP1 in Arabidopsis. All tomato HPs contain two conserved motifs (motifs 1 and 2) (Figure 2, Table S1). HP proteins generally possess the Hpt domain with a signature motif of XHQXKGSSXS. However, in SlPHP1 and SlPHP2, the His of Hpt domain is replaced by Tyr and Asn, respectively (Figure S3). Although SlPHP1 lost the conserved Hpt domain, it shows a high identity (63%) to authentic HP (AHP4) in Arabidopsis. 2.4. RR Proteins in Tomato Thirty-nine RRs, including seven type-A RRs, 23 type-B RRs, one type-C RR, and eight pseudo-RRs (PRRs) were identified in tomato (Table S1). Tomato type-A RRs, including SlRR1–7, have much smaller average protein size compared with these of other RRs and share a high degree sequence identities (58% to 79%) with their counterparts (ARR3, ARR9, and ARR17) in Arabidopsis. They exhibit quite similar gene structure. All tomato type-A RRs possess four introns, except for SlRR3, which contains only one intron (Figure 3). These type-A RRs only contain one conserved REC domain and only correspond to motif 3, as identified by MEME (Figure 3 and Figure S4). Plant type-B RR proteins are usually featured by REC domain in N-end and Myb domain in C-end. However, nearly half of tomato type-B RRs only contain a REC domain, where the Myb domain may be lost during the evolution of the tomato RR family. Notably, except REC domain, SlRR29 protein also has a Trans-reg-C domain, which is only present in prokaryote TCS element (Figure S5). This finding indicates that eukaryote TCS elements are probably evolved from that of prokaryotes. Only one type-C RR, SlRR31, was identified in tomato; this protein shares 37% identity with Arabidopsis ARR22 and contains only one REC domain which is similar to the structure of type-A RR proteins. Tomato PRR subfamily was further classified into six clock and two type-B PRRs (Table S1). SlPRR1–6 belonging to clock PRRs, share 37% to 68% sequence identity with the homologous proteins, namely, APRR1, APRR5, and APRR7, in Arabidopsis. All six clock PRRs have two motifs (motifs 1 and 3), except SlPRR5, which lose its pseudo-REC domain. SlPRR1–6 contain a CCT domain, which plays an important role in regulating circadian rhythm and controlling flowering time [34]. Only two type-B PRRs, namely, SlPRR7 and SlPRR8, were identified to be homologous with APRR2 in Arabidopsis with 45% and 46% sequence identities, respectively. SlPRR7 and SlPRR8 are characterized by a pseudo-REC and a Myb domain (Figure 3). SlPRR8 regulates tomato plastid development and fruit ripening [35]. 2.5. Phylogenetic Analysis of Plant TCS Proteins All amino acid sequences of HK(L) proteins from Arabidopsis, rice, maize, soybean, Chinese cabbage, Lotus japonicus, Physcomitrella patens, wheat, and tomato were used to perform multiple alignments and generate phylogenetic trees for exploring the evolutionary relationships of these HK(L)s (Figure 4). The HK(L) proteins in these nine species were divided into seven distinct clades, the cytokinin receptor, ethylene receptor, PHY-like, CKI1-like, CKI2/AHK5-like, AHK1-like, and PDK-like subfamilies; this finding is similar to that reported in previous studies [5,10]. Tomato HL(L)s usually have much closer relationships to that of soybean and Lotus japonicas than that of other species, which both are leguminous crops. As expected, the members from Physcomitrella patens, the only moss in the nine species, are generally distinct from the members of other species in each subclade. Unlike the other subfamilies, ethylene receptor and PHY-like subfamily members show an alternating distribution of monocots and eudicots in the phylogenetic tree, hence, these two subfamilies likely occurred before the divergence of dicots and monocots. The phylogenetic tree divides HP proteins from the nine species into four clades, namely, I, II, III, and IV (Figure 5). Clade I is only occupied by HPs from dicots. Notably, the HPs from Physcomitrella patens are grouped into clade I and have closest relationship with SlHP3 in tomato. Meanwhile, the HPs in clade II are all from monocots. Clade III is further divided into two subclades consisting of the HP members from monocot and dicot, respectively. All the RR proteins from the nine species were grouped into type-A, type-B, type-C, and PRR subfamily (Figure 6). Generally, tomato RRs are phylogenetically closer to soybean and Lotus japonicus RRs. Phylogenetic analyses showed that type-A RRs from these nine species share a fairly close evolutionary relationship to each other. Type-B RRs, the biggest subgroup, could be further divided into six subgroups. Type-B I RRs are the most predominant RRs and contain the RRs from the nine species. Type-B IV and V subgroups only contain RRs from monocots. And type-B VI subgroup is exclusively occupied by tomato RRs (SlRR16–18 and SlRR25–28), as a tomato-special subgroup. Previous studies suggested that type-C RRs may be the oldest RR genes; type-A RR, as the youngest subgroup, may evolved from type-C RRs by mutations happened in their promoter sequences [36]. Type-C RRs share similar structure, whereas have lower similarity with type-A RR proteins. All monocot and eudicot type-C RRs show an alternating distribution in the phylogenetic tree; this finding indicates that these type-C RRs probably already existed before the divergence of monocotyledons and dicotyledons. The type-C RRs from Physcomitrella patens exhibit closer relationship with RRs from monocotyledons. Notably, Arabidopsis APRRs are clustered into a distinct subclade in the phylogenetic tree, whereas PRRs from the other species could be further divided into clock and type-B PRR subclades. The type-B PRRs are closely related to type-B RRs, which is consistent with the results of previous phylogenetic analyses [9,10]. 2.6. Genomic Distribution and Gene Duplication of Tomato TCS Members All identified tomato TCS family genes, except SlRR29, located on the scaffcold A00, are distributed on 12 tomato chromosomes (Figure 7). The HK(L)s are unevenly mapped on all tomato chromosomes except A03. Six SlHPs are located on tomato chromosome A01, A03, A06, A08, and A11, whereas 39 SlRRs are mapped on all the tomato chromosomes, except chromosome A09. The gene duplication events were analyzed in the tomato TCS gene family. The duplicate pairs result from segment duplication, including SlHKL1/SlHKL4, SlRR12/SlRR13, SlRR22/SlRR23, and SlPRR1/SlPRR2, respectively. On the other hand, the duplication genes of SlRR16, SlRR18, and SlRR25–SlRR28, which are clustered on chromosome A11, were identified as tandem duplicates. All these results suggest that segmental and tandem duplication probably contribute to the expansion of the tomato TCS gene, which differs from that in Arabidopsis, Chinese cabbage, and soybean [5,9,10]. The synonymous rate (Ks), non-synonymous rate (Ka), and Ka/Ks of these duplicates were calculated, and duplication time was speculated using the values of Ks (Table 2). The Ks of four segment duplicates range from 0.6 to 0.79. Thus, the divergent time ranges from 46.15 Mya to 60.77 Mya. The Ka/Ks values of the segment duplicates are less than 1, indicating that they underwent purify selection. Meanwhile, the tandem cluster of SlRR16, SlRR18, and SlRR25-SlRR28 were speculated to diverge from 5.97 Mya to 26.55 Mya. The Ka/Ks values of all these duplicates are less than 1, indicating that that purification selection occurred in these duplicates except in the duplicated pair of SlRR18/26. 2.7. Analysis of Cis-Elements in Putative Promoter Regions of TCS Genes in Tomato We identified and analyzed cis-regulatory elements in the putative promoter regions of the TCS genes in tomato (Table S3). Numerous cis-motifs are involved in responses to abiotic stresses (high or low temperature, wound, and drought) and hormone treatments (ethylene, MeJA, salicylic acid, and ABA). ABA (ABRE and CE3) and drought (MBS) related cis-elements were detected in 21 out of total 65 gene promoters. Only seven TCS members contain the low temperature-responsive elements (LTR). Interestingly, we identified 38 heat stress-responsive elements (HSE) in the putative promoter regions of tomato TCS genes. Consistently, it has been reported that the pollens of the SlHK9 mutant were more sensitive to heat stress [24]. 2.8. Subcellular Localization of TCS Proteins from Tomato Subcellular localization of some TCS proteins was analyzed using transiently expression via green fluorescent protein (GFP)-fusion proteins in the epidermal cells of onion. SlHK8, as an ethylene receptor, was predicted to be located in cytoplasm using SubLoc v1.0 website [37] (Table S1). But, in fact, besides cytoplasm, it was also detected to be located in nuclear, membrane, and cell walls, indicating that SlHK8 probably serves as membrane bound receptor. SlHP2 and SlHP3 were predicted to be located in cytoplasm and mitochondrion, respectively. Consistently, the fluorescence signal of SlHP2-GFP and SlHP3-GFP proteins was mainly detected in cytoplasm and membrane. It is worth noting that SlHP3-GFP fluorescence signal was also detected in the nucleus. SlRR8 is located in the nucleus, which is consistent with its function as a transcription factor. As a type-A RR, the fluorescence signal of the SlRR1-GFP protein was clearly detected in the nucleus, whereas a weak signal was also found in the cytoplasm (Figure 8). Similarly, all of the RR proteins in Arabidopsis were found to be localized exclusively in the nucleus except type-A ARRs, namely, ARR3 and ARR16, which are also localized to the cytoplasm [38]. Thus, the subcellular localizations of the five selected tomato TCS proteins are highly similar to that of the homologous proteins in Arabidopsis and display a diverse subcellular targeting, indicating their predicted divergent functionality. 2.9. Expression Profiles of Tomato TCS Genes in Various Organs The electronic expression profiles of 65 tomato TCS genes in various organs/tissues were downloaded from the tomato eFP browser at bar.utoronto.ca. Among them, the transcripts of 20 TCS genes were quite low in all detected organs/tissues. So we clustered the rest 45 gene expression profiles in various organs with MeV4.8. The heatmap indicated that the expression profiles of tomato TCS genes in leaf, root, flower, and fruit could be divided into four clades (Figure 9A). The genes in clade I highly expressed in tomato fruit, whereas most of them had a quite low expression level in leaf. In clade II, highest mRNA levels of 17 genes were detected in root. The genes from clade III are predominantly expressed in leaf. Notably, the transcripts of SlHP4 and SlPHP1, along with their homologous gene AHP4 in Arabidopsis [4], were specifically detected in the leaves. There were 11 genes were found express highest in flower and grouped into clade IV. Among them, the transcripts of SlHKL1, SlRR6/7, and SlPRR8 were only detected in flower. The expression profiles of tomato TCS genes in six fruit different developmental stage were further summarized and clustered (Figure 9B). They could be further grouped into five clades (Figure 9B). In clade I, the gene expression levels gradually increase during fruit development. Clade II members are expressed highly at the middle stage of fruit development. In contrast with that in clade I, the gene expression levels in clade V exhibit a decrease trend during fruit development with the lowest level at 10 d after breaker. As expected, all of type-A RRs, except SlRR6 and SlRR7, are clustered in clade V and exhibit a high expression level in early fruit, indicating that they may be involved in fruit development. The selected 45 tomato TCS genes were subjected to Gene Ontology (GO) enrichment analysis (Figure 10). GO analysis indicated that TCS genes are mainly associated with three molecular functions, namely, ethylene binding and receptor activity, kinases binding and activity, phosphotransfer and photoreceptor activity. Coordinately, these genes are mainly located in nucleus, intracellular, reticulum and endomembrane system, and intrcellar membrane-bounded organelle. However, they are involved in many biological processes, such as regulation of peptidyl-histidine phosphorylation, negative regulation of phosphorelay signal transduction system and ethylene-activated signaling pathway, lateral and secondary growth, cellular response to sucrose stimulus, regulation of chlorophyll and tetrapyrrole catabolic process, regulation of iron ion transport, and cellular response to cold. In additional, KEGG pathways that were significantly enriched in the TCS genes were shown in Table S4. As expected, the mainly enriched pathway mapped in these tomato TCS genes include plant hormone signal transduction (ko: 04075), circadian rhythm (ko: 04712), and two-component system (ko: 02020), which is consistent with the findings of function studies in Arabidopsis TCS [4]. 2.10. Expression Profiles of TCS Genes in Response to Exogenous Hormones and Abiotic Stresses Evaluation of the expression levels of 31 randomly selected TCS genes revealed that most of the detected TCS genes in tomato could be induced by exposure to exogenous trans-zeatin and ABA treatment (Figure 11). However, the expression patterns vary among distinct TCS genes. For ZT treatment, all the detected tomato type-A RRs, namely, SlRR1–SlRR5, are generally induced by ZT, particularly at 1 h after treatment. Similarly, the type-A RRs in other species were proven to be obviously upregulated by cytokinin [5,10]. All the detected tomato SlHKs display different response profiles to ZT treatment. SlHK8 are obviously downregulated. However, SlHK6–7 could be induced by ZT. Almost all of the detected SlHPs are generally induced, except SlHP2, which is obviously downregulated. Under ABA treatment, tomato HK transcripts are generally induced and their expression levels are maintained at a relatively high level at 8 h. By contrast, most of HPs are induced at the early stages but then decrease to a low level at 8 h except SlHP1. In addition, most of SlRRs positively respond to ABA except SlRR2–3, SlRR23, and SlPRR4–6. Almost all of the detected SlHPs are upregulated in response to drought stress. All of HPs are evidently induced from 1 h after drought treatment and generally maintain at a higher level, except that the transcript levels of SlPHP1 decrease 2 h after treatment. However, all the tested SlHKs except SlHK4/7 are downregulated after drought treatment. Except SlRR3, the expression of the other tomato type-A RRs are generally downregulated even though the transcript levels of SlRR4–6 increase at 8 h. Except SlPRR4–5, all SlPRRs are obviously upregualted by drought. For salt treatment (Figure 12), most of HPs, including SlHP1, SlHP2, SlHP3, and SlPHP1 are evidently downregulated in general after salt treatment, although SlHP1 is slightly induced at 1 h. Similarly, a majority of SlHKs including SlHK2, SlHK3, SlHK4, and SlHK5 are generally downregulated by salt stress, although S1HKs slightly decrease at 1 h. Notably, two ethylene receptors (SlHK7 and SlHK8) are significantly induced by salt stress but repressed at 2 h. The other ethylene receptor like gene, SlHKL2, is obviously repressed by drought. Interestingly, most TCS genes such as SlHP4, SlPRR1, SlPRR2, SlPRR4, and SlPRR5 exhibit similar response patterns to drought and salt. However, some genes including SlHK8, SlHP3, SlHP4, and SlRR23 exhibit an opposite expression patterns under two abiotic stresses. 3. Discussion In this study, a total of 65 TCS genes, including 20 HK(L)s, six HPs and 39 RRs, were identified from tomato genome. The number of TCS family members in tomato is slightly bigger than that of Arabidopsis (56), rice (52), and maize (59), but obviously fewer than that in Glycine max (98), and Brassica rapa (85) (Table 1). In detail, the number of HK(L) family members in tomato (20) is larger than that in Arabidopsis (17), rice (11), and maize (11), which is only less than that in soybean (36). It was worth mentioning that tomato contains the largest number of type-B RRs (23) in all identified species and is nearly twice as many as that in Arabidopsis (12). In tomato, four pairs of segment duplicates including SlHKL1 and SlHKL4, SlRR12 and SlRR13, SlRR22 and SlRR23, and SlPRR1 and SlPRR2 were found. And a tandem duplicate cluster of SlRR16, SlRR18, and SlRR25–SlRR28 were identified. These tandem duplicates with high similarties exhibit conserved protein and gene structure, which all have a REC domain. Furthermore they occupied a tomato-specific type-B VI subfamily in the phylogenetic tree. Segmental duplication and tandem duplication events both contribute to the expansion of the TCS gene family in tomato. By contrast, in Arabidopsis, Chinese cabbage, and soybean, segmental duplication was the main mechanism contributing to the duplication of TCS genes [5,9,10]. In Arabidopsis, 10 pairs of segmental duplicates were found which accounted for 35.71% of all Arabidopsis TCS genes [5,10]. In Chinese cabbage, 61 genes among all 85 TCSs were identified to be duplicated because of segmental duplication [10]. A total of 66 out of 98 soybean TCS genes were identified to be segment duplicates [9]. Tandem duplication was not found in the TCS genes from Arabidopsis and soybean. Only one pair of duplicated genes was identified in Chinese cabbage. These results suggested that, unlike that in tomato, segmental duplication might be the main mechanism contributing to the duplication of TCS genes in Arabidopsis, Chinese cabbage, and soybean [5,9,10]. In tomato TCS genes, all of the tandem duplication and half of segmental duplication occur in type-B RRs. Thus, the gene duplication of type-B RRs mainly contributes to the expansion of TCS. Previous studies demonstrated that tomato genome underwent two independent large-scale genome and/or segmental duplication events. One of these duplications was ancient and occurred around 170–235 Mya, immediately after the divergence of monocots and dicots. The other duplication was recent polyploidy duplication, which occurred approximately 90 Mya and is the estimated divergence time of tomato and Arabidopsis [39,40]. In this study, the Ks of segmental duplicates in the tomato TCS genes ranged from 0.6 to 0.79, which corresponded to the divergence time of 46 Mya to 60 Mya, suggesting that the gene duplication events occurred after the split of tomato and Arabidopsis. On the other hand, SlRR16, SlRR18, and SlRR25–28 in the tomato genome clustered together on chromosome A11, forming a tandem duplicate cluster. The calculated divergence time was varied from 5.97 Mya to 26.55 Mya. Tandem duplicates usually occurred more recently than segment duplicates, which probably occurred because tandem duplications in plants were more likely to participate in stress responses and these tandem duplicates were not retained as long as nontandem duplicates [41] (Table 2). Expression analysis indicated that 45 TCS genes in tomato are predominantly expressed in the root, fruit, or flower, whereas the transcripts of other genes could not be detected in any tomato organs (Figure 9). Twelve genes including SlHK2–4, SlHKL3/5, SlHP2/3, SlRR9/10, and SlRR21–23, are predominantly expressed in the roots where the cytokinins are mainly synthesized. These genes probably play important roles in cytokinin signal transduction like their homologous genes in Arabidopsis [5]. This result is consistent with the finding in Chinese cabbage and soybean [10,23]. Most of TCS elements including HKLs and PRRs, exhibit preferential expression in tomato fruit. Notably, all ethylene receptors, except SlHKL3, are highly expressed during fruit ripening stage. Ethylene receptors SlHK7–9 and SlHKL1–4 all have a C2H2 domain, which could perceive ethylene signal. In fact, the function of some TCS elements in the development and ripening of tomato fruit has been widely studied [28,29,35]. Transgenic plants with reduced LeETR4 (SlHKL4) enhanced flower senescence and failed to fruit set [28]. Analysis on phytochrome phyA, phyB1, and phyB2 single, double, or triple mutants indicated that they participated in modulating the carotenoid formation and the time required for phase transitions during fruit ripening [29]. Additionally, tomato SlPRR8 was verified to regulate plastid development and fruit ripening [35]. Increasing evidence verified that TCS proteins are involved in responses to various abiotic stresses [10,17,18,19,23]. In this study, 31 tomato TCS genes were detected, and most genes appear to be regulated by drought- and salt-stresses (Figure 12). Most of tomato TCS genes negatively respond to salt stress, and similar results were found in soybean and Chinese cabbage [10,23]. However, 18 out of 31 genes are obviously induced by drought treatment, which differ from that in Arabidopsis, soybean and Chinese cabbage [10,17,18,19,23]. For examples, most of HPs and PRRs in tomato are upregulated by drought stress, but AHP and APRRs in Arabidopsis negatively responded to drought [18,19]. Similarly, tomato type-B RRs SlRR9, SlRR22, and SlRR23 are upregulated by drought but their homologous genes ARR1 and ARR12 negatively responded to drought [17]. In additional, we identified 23 out of total 65 genes containing dehydration-inducible ABRE, CE3, and/or MBS motifs in their promoter regions (Table S3). The expression profiles of 12 out of all these 23 genes including SlHK4/5/7, SlHP1, SlRR1/21/22, and SlPRR1/3/4/5/8 were analyzed by qRT-PCR. In detail, nine genes including SlHK4/7, SlHP1, SlRR21/22, and SlPRR1/3/8 are generally induced by drought, which are consistent with the promoter analyses, but the other three genes (SlHK5, SlRR1, and SlPRR5) are downregulated. The inconsistent results of SlHK5, SlRR1, and SlPRR5 between expression profiles and promoter analyses were also observed in soybean and Chinese cabbage [10,23]. Plant TCS elements were determined to play vital roles in responses to abiotic stresses, particularly drought, high salinity, and high or low temperature in Arabidopsis, rice, tomato, and soybean [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,42,43]. Tomato LE-ETR3 (Nr) participated in salt and heat stresses, and reducing expression of LE-ETR4 led to an enhanced hypersensitive response [24,42,43]. Meanwhile, tomato PHYA, PHYB1, and PHYB2 were verified to modulate drought stress responses [25]. The expression analyses here for TCS elements in this work provide an important implication on the function of these family genes under abiotic stresses. 4. Materials and Methods 4.1. Identification of TCS Genes in Tomato Protein sequences of all known plant TCS genes, particularly 56, 52, 51, 98, 62, and 85 members in the genome of Arabidopsis, rice, maize, soybean, Lotus japonicus, Physcomitrella patens, wheat, and Chinese cabbage, respectively, were downloaded from Phytozome [44] and then used as queries to perform BLASTP searches in the SGN database (http://solgenomics.net/) with E-value of 1 × 10−5 as the threshold [9,30]. Meanwhile, the tomato genome protein sequences were downloaded from SGN database and Hidden Markov Model (HMM) profiles of TCS characteristic domains, i.e., HisK (PF00512), HATPase (PF02518), HPt (PF01627), and REC (PF00072) were downloaded from Pfam (http://pfam.janelia.org/). Then we searched for TCS genes with HMMER 3.0 using the global HMM profile of these TCS characteristic domains with expected values less than 0.1. After removing redundant sequences, a total of 118 putative elements were identified. As a final quality check, each identified sequence using the two strategies above was subsequently confirmed using Pfam (http://pfam.janelia.org/) and SMART (http://smart.embl-heidelberg.de/) databases according to whether or not it possessed the structural characteristics and conserved domains of TCS elements, i.e., HisK, HATPase, REC, CHASE domain for cytokinin binding (CHASE), ethylene-binding domain (C2H4), and HPt domains. Tomato TCS homolog proteins in Arabidopsis were identified using BLASTP search against Arabidopsis databases of TAIR website (http://www.arabidopsis.org/) with default expected values. ExPASy [45] was used to calculate the molecular weights and isoelectric points (PIs) of putative tomato TCS proteins. Subcellular localizations were predicted using SubLoc v1.0 website [37]. 4.2. Gene Structure Construction, Motif Analysis, and Phylogenetic Analysis The exon–intron organizations of all tomato TCS genes were mapped using Gene Structure Display Server [46]. Each family motif was identified using the MEME program [47]. The predicted peptide sequences of the conserved domain in the TCS proteins were identified by employing the SMART database. Then multiple-sequence alignment for the predicted peptide sequences was generated using Clustal X v1.81 with default parameters [48]. The similarity of the tomato TCS proteins with those from Arabidopsis, rice, and tomato genome was calculated by DNAStar (Madison, WI, USA). Phylogenetic analysis was performed using MEGA 5.0 program by neighbor-joining (NJ) method with 1000 replicates of the bootstrap based on the full-length protein sequences [49]. 4.3. Chromosomal Localization and Evolutionary Analysis of TCS Genes All the TCS genes were assigned to the corresponding tomato chromosomes based on the SGN database. A pair of genes were identified as tandem duplicates if the genes both shared ≥40% amino acid sequence similarity and separated by fewer than five intervening genes [50]. PGDD [51] was adopted to perform synteny analysis and detect the segment duplications, as described in cucumber MADS gene family [50]. Full-length amino acid sequences were aligned using the ClustalW algorithm [52], and then Ks and Ka were calculated using the Codeml procedure of the PAML program [53]. Divergence time of the gene pairs was estimated using synonymous mutation rate of substitutions per synonymous site per year, as follows: T = Ks/2x (x = 6.56 × 10−9) [54]. 4.4. Analysis of Putative Promoter Regions TCS Genes in Tomato The upstream sequences (1.5 kb) of the initiation codon in TCS genomic DNA were obtained from Phytozome [44] as the putative promoter regions, and the cis-regulatory elements in the promoter regions were identified using PlantCARE website [55]. 4.5. Subcellular Localization The randomly selected TCS genes were amplified using gene-specific primers (Table S5) and cloned into the pFGC-EGFP plasmids by Xba I and BamH I restriction sites under the control of the 35S cauliflower mosaic virus promoter (35S CaMV). The pFGC:GFP empty vector was used as control. The recombinant vectors were transformed into onion epidermal cells by particle bombardment using the Biolistic PDS-1000/He gene gun system (Bio-Rad, Hercules, CA, USA) [56]. After 16–18 h of incubation in darkness, the onion epidermal cell was plasmolyzed in 0.3 g·mL−1 sucrose for 5 min and the fluorescence of GFP was photographed by a Leica DMLE camera (Leica, Wetzlar, Germany). 4.6. Tomato Plant Growth and Treatments Tomato (S. lycopersicum L.) cv. Micro-Tom from Tomato Genetics Resource Center (University of California, Davis, CA, USA) was used for expression analysis. The seedlings were grown in a growth chamber in temperature-controlled greenhouses of Zhejiang University under day/night temperatures of 28/20 ± 1 °C and light intensity of 250 μmol·m−2·s−1 with 16-h day length. Three-week-old tomato seedlings were used for abiotic stresses and exogenous hormone treatments. For cytokinin and ABA treatment, the seedlings were sprayed with 100 μM ZT and 100 μM ABA, respectively. The second true leaf on each plant was sampled at 0 (control), 1, 2, 4, and 8 h after spraying. To induce drought stress, the seedlings were transferred to filter paper and the leaves were collected at 0, 1, 2, 4, and 8 h. For high salt treatment, the nutrient solution was supplemented with 100 mM NaCl and the leaves were separately collected at 0, 1, 2, 4, and 8 h after treatment. All phytohormone and abiotic treatments were repeated three times and each treatment contained at least 20 seedlings. All materials were frozen at −75 °C until RNA isolation. 4.7. Expression Analysis of TCS Genes in Growth and Development Gene The electronic expression data of tomato TCS genes in various organs were obtained by gene locus from the tomato eFP browser at http://bar.utoronto.ca [57]. The electronic expression profiles of all detected tomato TCS genes expressed in leaves, roots, unopened flower buds, fully opened flowers, and the fruits at six developmental stages (1 cm, 2 cm, and 3 cm fruit, mature green fruit, breaker fruit, and fruit at 10 days after breaker) were summarized and used to generate the heatmap with Multiple Array Viewer [58]. Tomato TCS genes were extracted for GO functional enrichment analysis (http://geneontology.org/) and KEGG pathway enrichment analysis [59] with default parameters. 4.8. RNA Isolation and qRT–PCR The total RNA was extracted from the collected materials using TRIZOL reagent (Invitrogen, Karlsruhe, Germany) according to the manufacturer-recommended protocol. The first cDNA strand was generated from 1 μg of total RNA using the PrimerScript RT reagent kit (Takara, Otsu, Japan) according to the manufacturer’s instructions. Specific primers used in the qRT-PCR were designed by Primer 5 Software, and each primer was searched in the tomato database to ensure its specificity. The qRT-PCR reactions were performed on the CFX96 Real Time System machine (Bio-RAD, Hercules, CA, USA), programmed to heat for 30 s at 95 °C, followed by 40 cycles of 5 s at 95 °C and 45 s at 55 °C, and at the end, 1 cycle of 1 min at 95 °C, 30 s at 50 °C and 30 s at 95 °C. Two biological and three technical replicates for each sample were performed with 15 μL of reaction volume using the SYBR Premix Ex Taq kit (TOYOBO, Osaka, Japan). The tomato SlUbi3 gene (GenBank accession number X58253) was selected as an internal control [60]. The relative gene expression level was calculated using the 2−ΔΔCt method. Heatmap was generated by Multiple Array Viewer using the relative expression data of each gene [58]. 5. Conclusions In our study, 20 HK(L)s, six HPs and 39 RRs were identified from tomato genome. Gene classification, gene structures, conserved domains, chromosome distribution, phylogenetic relationship, synteny relationship, gene duplication events, and subcellular localizations of the TCS genes were predicted and analyzed in detail. The tomato TCS elements showed significant sequence and domain conservation except type-B RRs. Gene duplication events mainly occurred in the RR family of tomato TCS genes. Both segment duplication and tandem duplication contributed to gene expansion. The subcellular localization of selected proteins displayed a diverse subcellular targeting and probably played divergent roles. Most TCS genes are predominantly expressed in tomato reproductive organs particularly in fruit development. Meanwhile, promoter analyses and qRT-PCR results indicated that almost all of TCSs could respond to various stresses and exogenous hormone treatments. The identification of tomato TCS elements would provide a more comprehensive sight and solid foundation to elucidate their roles in mediating hormone cross-talk and stress responses in further. Acknowledgments The work was supported by the grants from the National Science Foundation of China (Grant numbers 31271633; 31471878), Public Welfare Projects of Zhejiang Province (2014C32012), and Zhejiang Province Key Science and Technology Innovation Team (2013TD05). Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1204/s1. Click here for additional data file. Author Contributions Yanjun He performed the experiments, analyzed the data, and drafted the manuscript. Xue Liu participated in qRT-PCR experiments and data analysis. Lei Ye and Changtian Pan collected the public dataset and assisted with data analysis. Lifei Chen and Tao Zou prepared the cucumber samples. Gang Lu conceived the study and its design and assisted with revisions to the manuscript. All authors read and consented to the final version of the manuscript. Conflicts of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Figure 1 Phylogenetic analysis, gene structure, and conserved motifs of all HK(L) genes in tomato. (A) The phylogenetic tree of HK(L) proteins. Predicted amino acid sequences of HK(L) proteins were aligned using the Clustal X v1.81 program. The phylogenetic tree was constructed using the neighbor-joining (NJ) method with 1000 bootstrap replicates as implemented in the MEGA 5.0; (B) Gene structure was analyzed using the Gene Structure Display Server online. The green boxes indicate the exons, and lines indicate the introns; (C) Schematic distribution of conserved motifs in the HK(L) proteins. Motif analysis was performed using MEME 4.0 software as described in the methods. The colored boxes represent different motifs in the corresponding position of each HK(L) protein. Figure 2 Phylogenetic analysis (A); gene structure (B); and conserved motifs (C) of the HP family members in tomato. For other details, see Figure 1. Figure 3 Phylogenetic analysis (A); gene structure (B); and conserved motifs (C) of RR genes in tomato. For other details, see Figure 1. Figure 4 Phylogenetic relationship of HK(L) proteins in Arabidopsis, rice, maize, Chinese cabbage, soybean, Lotus japonicus, Physcomitrella patens, wheat, and tomato. The phylogenetic trees were constructed using the NJ method with bootstrap 1000 tests by MEGA 5.0. The diverse subgroups of HK(L) proteins were marked by different colors. The bar represents the relative divergence of the sequences examined. Figure 5 Phylogenetic phylogenetic analysis of plant HP family genes in Arabidopsis, rice, maize, Chinese cabbage, soybean, Lotus japonicus, Physcomitrella patens, wheat, and tomato. For other details, see Figure 4. Figure 6 Phylogenetic relationship of RR proteins in Arabidopsis, rice, maize, Chinese cabbage, soybean, Lotus japonicus, Physcomitrella patens, wheat, and tomato. For other details, see Figure 4. Figure 7 Chromosomal distribution of TCS genes in tomato. The chromosome number is indicated at the top of each chromosome. The arrows indicate the sense (▲) and antisense (▼) strands. The pairs of genes with tandem duplication were highlighted with the yellow background. The duplicated gene pairs have been link by dark line. Figure 8 Subcellular localization of TCS proteins. Green fluorescent protein (GFP)-fusion proteins were transiently expressed in onion epidermis cells under the control of the 35S promoter. After 16–18 h of incubation, GFP signal was detected with a green fluorescence microscope. Fluorescence (up) and bright-field images (down) of plasmolyzed empty vector pFGC: GFP transgenic cell (A); Fluorescence (up) and bright-field images (down) of 35S::SlHK8-GFP (B); 35S::SlHP2-GFP (C); 35S::SlHP3-GFP (D); 35S::SlRR1-GFP (E); and 35S::SlRR8-GFP (F) transgenic cell. Scale bar was presented in bottom right. Figure 9 Heat map representation for organ-specific expression (A) and six fruit development stages-related expression (B) profiles of TCS genes in tomato. These electronic expression data were downloaded from the tomato eFP browser at bar.utoronto.ca. The heatmap was drawn by MeV4.8. The expression levels are presented using fold-change values transformed to Log2 format compared with control. The Log2 (fold-change values) and the color scale are shown at the top of heat map. Green, black, and red represent low, medium, and strong expression, respectively. Figure 10 The Gene Ontology (GO) analysis of TCS genes. The TCS genes were categorized into three groups: molecular function (A); biological process (B); and cell component (C). Figure 11 Heat map representation for the response patterns to exogenous trans-zeatin (ZT) (A) and ABA (B) of TCS genes in tomato. The second true leaves were collected at 0, 1, 2, 4, and 8 h after 100 μM ZT or 100 μM ABA treatment. The heatmap were manufactured by MeV4.8. The color scale representing the relative expression values is shown in the upper left of the heatmap. Figure 12 Heat map representation for the response patterns to drought (A) and salt (B) stresses of TCS genes in tomato. The second true leaves were collected at 0, 1, 2, 4, and 8 h after the onset of stress treatments. For other details, see Figure 10. ijms-17-01204-t001_Table 1Table 1 Summary of the two-component system (TCS) gene numbers identified in plants. Species HK(L) HP (Pseudo-HP) Type-A RR Type-B RR Type-C RR Pseudo RR Total Reference Arabidopsis thaliana 17 (9) 6 (1) 10 12 2 9 56 [4] Oryza sativa 11 (3) 5 (3) 13 13 2 8 52 [7] Lotus japonicus 14 7 7 11 1 5 40 [31] Glycine max 36 (15) 13 18 15 3 13 98 [9] Zea mays 11 (3) 9 (2) 16 9 3 11 59 [8] Physcomitrella patens 18 3 7 5 2 4 39 [32] Triticum aestivum 7 10 41 2 0 2 45 [30] Brassica rapa 20 (9) 8 (1) 21 17 4 15 85 [10] Solanum lycopersicum 20 (11) 6 (2) 7 23 1 8 65 – HK(L), HP, and RR represent His-kinase or like, phosphotransfer, and response regulator protein, respectively. ijms-17-01204-t002_Table 2Table 2 Ks, Ka, and Ka/Ks calculation and divergent time of the duplicated tomato TCS gene pairs. Duplicated Gene Pairs Ks Ka Ka/Ks Duplicated Type Purify Selection Time (Mya *) SlHKL1/SlHKL4 0.68 0.23 0.34 Segmental Yes 52.31 SlRR9/SlRR13 0.79 0.38 0.48 Segmental Yes 60.77 SlRR22/SlRR23 0.74 0.17 0.23 Segmental Yes 56.92 SlPRR1/SlPRR2 0.60 0.21 0.35 Segmental Yes 46.15 SlRR16/SlRR18 0.34 0.29 0.84 Tandem Yes 26.15 SlRR16/SlRR27 0.32 0.23 0.71 Tandem Yes 24.98 SlRR16/SlRR28 0.24 0.20 0.85 Tandem Yes 18.48 SlRR16/SlRR26 0.20 0.19 0.99 Tandem Yes 15.07 SlRR18/SlRR27 0.20 0.12 0.59 Tandem Yes 15.08 SlRR18/SlRR28 0.31 0.26 0.84 Tandem Yes 23.90 SlRR18/SlRR26 0.19 0.20 1.06 Tandem No 14.80 SlRR18/SlRR25 0.21 0.18 0.84 Tandem Yes 16.10 SlRR25/SlRR27 0.17 0.10 0.59 Tandem Yes 13.45 SlRR25/SlRR28 0.08 0.06 0.79 Tandem Yes 5.97 SlRR25/SlRR26 0.11 0.10 0.91 Tandem Yes 8.28 SlRR26/SlRR27 0.35 0.33 0.96 Tandem Yes 26.55 SlRR26/SlRR28 0.16 0.15 0.94 Tandem Yes 12.34 SlRR27/SlRR28 0.11 0.06 0.58 Tandem Yes 8.41 * Mya, million years ago. ==== Refs References 1. Mizuno T. Compilation of all genes encoding two-component phosphotransfer signal transducers in the genome of Escherichia coli DNA Res. 1997 4 16 18 10.1093/dnares/4.2.161 2. Stock A.M. Robinson V.L. Goudreau P.N. Two-component signal transduction Annu. Rev. Biochem. 2000 69 183 215 10.1146/annurev.biochem.69.1.183 10966457 3. Urao T. Yamaguchi-Shinozaki K. Shinozaki K. Two-component systems in plant signal transduction Trends Plant Sci. 2000 5 67 74 10.1016/S1360-1385(99)01542-3 10664616 4. Hwang I. Chen H.C. Sheen J. Two-component signal transduction pathways in Arabidopsis Plant Physiol. 2002 129 500 515 10.1104/pp.005504 12068096 5. Schaller G.E. Kieber J.J. Shiu S.H. Two-component signaling elements and histidyl-aspartyl phosphorelays Arabidopsis Book 2008 e0112 10.1199/tab.0112 22303237 6. Grefen C. Harter K. Plant two-component systems: Principles, functions, complexity and cross talk Planta 2004 219 733 742 10.1007/s00425-004-1316-4 15232695 7. Pareek A. Singh A. Kumar M. Kushwaha H.R. Lynn A.M. Singla-Pareek S.L. Whole-genome analysis of Oryza sativa reveals similar architecture of two-component signaling machinery with Arabidopsis Plant Physiol. 2006 142 380 397 10.1104/pp.106.086371 16891544 8. Chu Z. Ma Q. Lin Y. Tang X. Zhou Y. Zhu S. Fan J. Cheng B. Genome-wide identification, classification, and analysis of two-component signal system genes in maize Genet. Mol. Res. 2011 10 3316 3330 10.4238/2011.December.8.3 22194197 9. Mochida K. Yoshida T. Sakurai T. Yamaguchi-Shinozaki K. Shinozaki K. Tran L.S.P. Genome-wide analysis of two-component systems and prediction of stress-responsive two-component system members in soybean DNA Res. 2010 17 303 324 10.1093/dnares/dsq021 20817745 10. Liu Z. Zhang M. Kong L. Lv Y. Zou M. Lu G. Cao J. Yu X. Genome-wide identification, phylogeny, duplication, and expression analyses of two-component system genes in Chinese cabbage (Brassica rapa ssp. pekinensis ) DNA Res. 2014 21 379 396 10.1093/dnares/dsu004 24585003 11. Tran L.S.P. Urao T. Qin F. Maruyama K. Kakimoto T. Shinozaki K. Yamaguchi-Shinozaki K. Functional analysis of AHK1/ATHK1 and cytokinin receptor histidine kinases in response to abscisic acid, drought, and salt stress in Arabidopsis Proc. Natl. Acad. Sci. USA 2007 104 20623 20628 10.1073/pnas.0706547105 18077346 12. Wohlbach D.J. Quirino B.F. Sussman M.R. Analysis of the Arabidopsis histidine kinase ATHK1 reveals a connection between vegetative osmotic stress sensing and seed maturation Plant Cell 2008 20 1101 1117 10.1105/tpc.107.055871 18441212 13. Tran L.S.P. Shinozaki K. Yamaguchi-Shinozaki K. Role of cytokinin responsive two-component system in ABA and osmotic stress signalings Plant Signal. Behav. 2010 5 148 150 10.4161/psb.5.2.10411 20023424 14. Pham J. Liu J. Bennett M.H. Mansfield J.W. Desikan R. Arabidopsis histidine kinase 5 regulates salt sensitivity and resistance against bacterial and fungal infection New Phytol. 2012 194 168 180 10.1111/j.1469-8137.2011.04033.x 22256998 15. Kumar M.N. Jane W.N. Verslues P.E. Role of the putative osmosensor Arabidopsis histidine kinase1 in dehydration avoidance and low-water-potential response Plant Physiol. 2013 161 942 953 10.1104/pp.112.209791 23184230 16. Jeon J. Kim J. Arabidopsis response Regulator1 and Arabidopsis histidine phosphotransfer Protein2 (AHP2), AHP3, and AHP5 function in cold signaling Plant Physiol. 2013 161 408 424 10.1104/pp.112.207621 23124324 17. Nguyen K.H. Van Ha C. Nishiyama R. Watanabe Y. Leyva-González M.A. Fujita Y. Tran U.T. Li W. Tanaka M. Seki M. Arabidopsis type B cytokinin response regulators ARR1, ARR10, and ARR12 negatively regulate plant responses to drought Proc. Natl. Acad. Sci. USA 2016 113 3090 3095 10.1073/pnas.1600399113 26884175 18. Nishiyama R. Watanabe Y. Leyva-Gonzalez M.A. Van Ha C. Fujita Y. Tanaka M. Seki M. Yamaguchi-Shinozaki K. Shinozaki K. Arabidopsis AHP2, AHP3, and AHP5 histidine phosphotransfer proteins function as redundant negative regulators of drought stress response Proc. Natl. Acad. Sci. USA 2013 110 4840 4845 10.1073/pnas.1302265110 23487796 19. Nakamichi N. Kusano M. Fukushima A. Kita M. Ito S. Yamashino T. Saito K. Sakakibara H. Mizuno T. Transcript profiling of an Arabidopsis PSEUDO RESPONSE REGULATOR arrhythmic triple mutant reveals a role for the circadian clock in cold stress response Plant Cell Physiol. 2009 50 447 462 10.1093/pcp/pcp004 19131357 20. Miyata S. Urao T. Yamaguchi-Shinozaki K. Shinozaki K. Characterization of genes for two-component phosphorelay mediators with a single HPt domain in Arabidopsis thaliana FEBS Lett. 1998 437 11 14 10.1016/S0014-5793(98)01188-0 9804162 21. Sun L. Zhang Q. Wu J. Zhang L. Jiao X. Zhang S. Zhang Z. Sun D. Lu T. Sun Y. Two rice authentic histidine phosphotransfer proteins, OsAHP1 and OsAHP2 , mediate cytokinin signaling and stress responses in rice Plant Physiol. 2014 165 335 345 10.1104/pp.113.232629 24578505 22. Feng W. Qin T. Yao W. Wen D. Zhang A. Tan M. Jiang M. OsHK3 is a crucial regulator of abscisic acid signaling involved in antioxidant defense in rice J. Integr. Plant Biol. 2015 57 213 228 24912543 23. Le D.T. Nishiyama R. Watanabe Y. Mochida K. Yamaguchi-Shinozaki K. Shinozaki K. Tran L.S. Genome-wide expression profiling of soybean two-component system genes in soybean root and shoot tissues under dehydration stress DNA Res. 2011 8 265 272 10.1093/dnares/dsq032 21208938 24. Firon N. Pressman E. Meir S. Khoury R. Altahan L. Ethylene is involved in maintaining tomato (Solanum lycopersicum ) pollen quality under heat-stress conditions AOB Plants 2012 2012 pls024 10.1093/aobpla/pls024 23050072 25. D’Amico-Damiao V. Cruz F.J.R. Gavassi M.A. Santos D.M.M. Melo H.C. Carvalho R.F. Photomorphogenic modulation of water stress in tomato (Solanum lycopersicum L.): The role of phytochromes A, B1, and B2 J. Hortic. Sci. Biotechnol. 2015 90 25 30 26. Harry K. Denise T. The tomato ethylene receptor gene family: Form and function Physiol. Plant. 2002 115 336 341 12081525 27. Pratt L.H. Cordonnier-Pratt M.M. Kelmenson P.M. Lazarova G.I. Kubota T. Alba R.M. The phytochrome gene family in tomato (Solanum lycopersicum L.) Plant Cell Environ. 1997 20 672 677 10.1046/j.1365-3040.1997.d01-119.x 28. Tieman D.M. Taylor M.G. Ciardi J.A. Klee H.J. The Tomato Ethylene Receptors NR and LeETR4 Are Negative Regulators of Ethylene Response and Exhibit Functional Compensation within a Multigene Family Proc. Natl. Acad. Sci. USA 2000 97 5663 5668 10.1073/pnas.090550597 10792050 29. Gupta S.K. Sharma S. Santisree P. Kilambi H.V. Appenroth K. Sreelakshmi Y. Sharma R. Complex and shifting interactions of phytochromes regulate fruit development in tomato Plant Cell Environ. 2014 37 1688 1702 10.1111/pce.12279 24433205 30. Gahlaut V. Mathur S. Dhariwal R. Khurana J.P. Tyagi A.K. Balyan H.S. Gupta P.K. A multi-step phosphorelay two-component system impacts on tolerance against dehydration stress in common wheat Funct. Integr. Genom. 2014 14 1 10 10.1007/s10142-014-0398-8 25228409 31. Ishida K. Niwa Y. Yamashino T. Mizuno T. A genome-wide compilation of the two-component systems in Lotus japonicus DNA Res. 2009 16 237 247 10.1093/dnares/dsp012 19675111 32. Ishida K. Yamashino T. Nakanishi H. Mizuno T. Classification of the genes involved in the two-component system of the moss Physcomitrella patens Biosci. Biotechnol. Biochem. 2010 74 2542 2545 10.1271/bbb.100623 21150091 33. Rockwell N.C. Su Y.S. Lagarias J.C. Phytochrome structure and signaling mechanisms Annu. Rev. Plant Biol. 2006 57 837 858 10.1146/annurev.arplant.56.032604.144208 16669784 34. Hayama R. Coupland G. Shedding light on the circadian clock and the photoperiodic control of flowering Curr. Opin. Plant Biol. 2003 6 13 19 10.1016/S1369-5266(02)00011-0 12495746 35. Pan Y. Bradley G. Pyke K. Ball G. Lu C. Fray R. Marshall A. Jayasuta S. Baxter C. van Wijk R. Network inference analysis identifies an APRR2-like gene linked to pigment accumulation in tomato and pepper fruits Plant Physiol. 2013 161 1476 1485 10.1104/pp.112.212654 23292788 36. Pils B. Heyl A. Unraveling the Evolution of Cytokinin Signaling Plant Physiol. 2009 151 782 791 10.1104/pp.109.139188 19675156 37. SubLoc Avaliable online: http://www.bioinfo.tsinghua.edu.cn/SubLoc/eu_predict.htm (accessed on 3 July 2016) 38. Dortay H. Gruhn N. Pfeifer A. Schwerdtner M. Schmülling T. Heyl A. Toward an interaction map of the two-component signaling pathway of Arabidopsis thaliana J. Proteom. Res. 2008 7 3649 3660 10.1021/pr0703831 18642946 39. Consortium T.T.G. The tomato genome sequence provides insights into fleshy fruit evolution Nature 2012 485 635 641 22660326 40. Blanc G. Wolfe K.H. Widespread paleopolyploidy in model plant species inferred from age distributions of duplicate genes Plant Cell 2004 16 1667 1678 10.1105/tpc.021345 15208399 41. Hanada K. Zou C. Lehti-Shiu M.D. Shinozaki K. Shiu S.H. Importance of lineage-specific expansion of plant tandem duplicates in the adaptive response to environmental stimuli Plant Physiol. 2008 148 993 1003 10.1104/pp.108.122457 18715958 42. Ciardi J.A. Tieman D.M. Jones J.B. Klee H.J. Reduced expression of the tomato ethylene receptor gene LeETR4 enhances the hypersensitive response to Xanthomonas campestris pv. vesicatoria Mol. Plant Microb. Interact. 2001 14 487 495 10.1094/MPMI.2001.14.4.487 11310736 43. Poór P. Kovács J. Borbély P. Takács Z. Szepesi Á. Tari I. Salt stress-induced production of reactive oxygen-and nitrogen species and cell death in the ethylene receptor mutant Never ripe and wild type tomato roots Plant Physiol. Biochem. 2015 97 313 322 26512971 44. Phytozome Avaliable online: http://phytozome.jgi.doe.gov/pz/portal.html (accessed on 3 July 2016) 45. ExPASy Avaliable online: http://web.expasy.org/compute_pi/ (accessed on 3 July 2016) 46. Gene Structure Display Server Avaliable online: http://gsds.cbi.pku.edu.cn (accessed on 3 July 2016) 47. MEME Avaliable online: http://meme-suite.org/tools/meme (accessed on 3 July 2016) 48. Thompson J.D. Gibson T.J. Plewniak F. Jeanmougin F. Higgins D.G. The CLUSTAL_X windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools Nucleic Acids Res. 1997 25 4876 4882 10.1093/nar/25.24.4876 9396791 49. Tamura K. Peterson D. Peterson N. Stecher G. Nei M. Kumar S. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods Mol. Biol. Evol. 2011 28 2731 2739 10.1093/molbev/msr121 21546353 50. Hu L. Liu S. Genome-wide analysis of the MADS-box gene family in cucumber Genome 2012 55 245 256 10.1139/g2012-009 22376137 51. PGDD Avaliable online: http://chibba.agtec.uga.edu/duplication/ (accessed on 3 July 2016) 52. ClustalW Avaliable online: http://www.genome.jp/tools/clustalw/ (accessed on 3 July 2016) 53. PAML Avaliable online: http://www.bork.embl.de/pal2nal/ (accessed on 3 July 2016) 54. Yuan S. Xu B. Jing Z. Zheni X. Qiang C. Zhimin Y. Qingsheng C. Bingru H. Comprehensive analysis of CCCH-type zinc finger family genes facilitates functional gene discovery and reflects recent allopolyploidization event in tetraploid switchgrass BMC Genom. 2015 16 1 16 10.1186/s12864-015-1328-4 25765300 55. PlantCARE Avaliable online: http://bioinformatics.psb.ugent.be/webtools/plantcare/html/search_CARE.html (accessed on 3 July 2016) 56. Gan C. Gene gun accelerates DNA-coated particles to transform intact-cells Scientist 1989 3 25 57. Tomato eFP Browser Avaliable online: http://bar.utoronto.ca/efp_tomato/cgi-bin/efpWeb.cgi (accessed on 3 July 2016) 58. Saeed A.I. Bhagabati N.K. Braisted J.C. Wei L. Sharov V. Howe E.A. Li J. Thiagarajan M. White J.A. Quackenbush J. TM4 Microarray Software Suite Method Enzymol. 2006 411 134 193 59. KEGG Avaliable online: http://www.genome.jp/kaas-bin/kaas_main (accessed on 3 July 2016) 60. Wu J. Liu S. He Y. Guan X. Zhu X. Cheng L. Wang J. Lu G. Genome-wide analysis of SAUR gene family in Solanaceae species Gene 2012 509 38 50 10.1016/j.gene.2012.08.002 22903030

PMC005xxxxxx/PMC5000603.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081205ijms-17-01205ReviewEffects of Secondary Plant Metabolites on Microbial Populations: Changes in Community Structure and Metabolic Activity in Contaminated Environments Musilova Lucie 1*Ridl Jakub 2Polivkova Marketa 1Macek Tomas 1Uhlik Ondrej 1*Iriti Marcello Academic Editor1 Department of Biochemistry and Microbiology, Faculty of Food and Biochemical Technology, University of Chemistry and Technology Prague, Technicka 3, 166 28 Prague, Czech Republic; [email protected] (M.P.); [email protected] (T.M.)2 Department of Genomics and Bioinformatics, Institute of Molecular Genetics of the Czech Academy of Sciences, Videnska 1083, 142 20 Prague, Czech Republic; [email protected]* Correspondence: [email protected] (L.M.); [email protected] (O.U.); Tel.: +420-220-445-136 (L.M. & O.U.)29 7 2016 8 2016 17 8 120504 4 2016 15 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Secondary plant metabolites (SPMEs) play an important role in plant survival in the environment and serve to establish ecological relationships between plants and other organisms. Communication between plants and microorganisms via SPMEs contained in root exudates or derived from litter decomposition is an example of this phenomenon. In this review, the general aspects of rhizodeposition together with the significance of terpenes and phenolic compounds are discussed in detail. We focus specifically on the effect of SPMEs on microbial community structure and metabolic activity in environments contaminated by polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs). Furthermore, a section is devoted to a complex effect of plants and/or their metabolites contained in litter on bioremediation of contaminated sites. New insights are introduced from a study evaluating the effects of SPMEs derived during decomposition of grapefruit peel, lemon peel, and pears on bacterial communities and their ability to degrade PCBs in a long-term contaminated soil. The presented review supports the “secondary compound hypothesis” and demonstrates the potential of SPMEs for increasing the effectiveness of bioremediation processes. secondary plant metabolites (SPMEs)community structurecarbon flowbioremediation ==== Body 1. Introduction Plants as primary producers synthesize tremendous amounts of organic compounds while consuming carbon dioxide and light energy. The spectrum of synthesized compounds is dependent on the plant species, and physiological and environmental conditions. Some of the synthesized compounds are released into the rhizosphere, the soil directly surrounding roots [1,2], which is affected by those released chemicals [3,4,5,6]. Plants deposit approximately 11% of fixed carbon into the rhizosphere [7,8]. The released carbon may appear to represent a significant energy loss for the plant; however, it may actually be beneficial due to the stimulation of biological activity in the rhizosphere [9], including stimulation of rhizosphere bacteria [10], which provide the plant with increased nutrient solubility, fixed nitrogen, and/or competitive suppression of pathogens [11], as well as plant growth promoting molecules [12,13]. Exuded compounds can further change the properties of the surrounding soil and are important for obtaining nutrients, mediating biological interactions, or decreasing the toxicity of pollutants [14,15]. Plant exudates and decomposing litter contain secondary plant metabolites (SPMEs) among other compounds. Beyond their role in mediating plant–microbe interactions, it is hypothesized that SPMEs can stimulate microbial metabolism of pollutants present in the environment, which is termed the “secondary compound hypothesis” [16,17,18]. In a contaminated environment, indigenous microflora usually contain genetic determinants enabling the synthesis of degradative enzymes [19,20]; however, environmental conditions can also often limit natural decontamination processes [21]. Plant-released chemicals can potentially improve these conditions [17,22], induce required genes for degradation of pollutants [23], or serve as primary substrates during the cometabolism of pollutants [16,18]. Although several studies focusing on the effects of SPMEs on microbial diversity and activity towards pollutants have delivered supporting evidence, the introduction of metagenomics [24] and high-throughput sequencing techniques (for review see [25]) has opened up new possibilities for addressing this topic, which has not yet been fully exploited. Despite these advances, linking one specific molecule from exudates with its effect under complex environmental conditions and relationships still remains challenging. In this review we present the general aspects of rhizodeposition and observed effects of selected secondary metabolites on bacterial cultures. The effects of SPMEs are demonstrated through research that focuses on changes in bacterial community structure and metabolic activity. Further examination of communities in environments polluted by polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs) is presented. 2. Root Exudates and Their Effects on Present Microflora 2.1. Root Exudates: Carbon Gateway to the Rhizosphere The rate of all excreted carbon compounds differs along the root in space and time and is dependent on many factors [7]. Initially, photosynthesis was considered the main factor influencing the amount of rhizodeposited carbon, because higher rates of excretion were reported during the day than at night [26]. However, a closer link was later proposed [27], connecting the carbon excretion rate to the rate of transport of the photosynthate into the roots. Therefore, the physiological state of a plant and environmental conditions are considered to be the main drivers of rhizodeposition rates. Another major factor that influences the amount of carbon excreted is the age of the plant: in general, young seedlings release relatively large amounts of carbon, which decreases over time with the increasing age of the plant. For example, rice (Oryza sativa L.) seedlings retain about 50% of assimilated carbon in above-ground parts and about 27% is exuded from roots [28]. In rice plants during maturation and flowering, plant carbon management shifts and more carbon is retained in the above-ground parts and less than 4% is exuded [28]. A similar trend was observed for other plants such as wheat (Triticum sp.), perennial rye-grass (Lolium perenne L.), and mung beans (Vigna radiate (L.) R. Wilezek) [29,30,31]. On the other hand, the amount of carbon deposited into the rhizosphere can be increased even in elderly plants under stress conditions as was observed in barley (Hordeum vulgare L.): when grown under a low potassium concentration, barley plants considerably increased the amount of exuded organic carbon in slow-growing roots compared to fast-growing roots [32]. The effect of drought, a different type of stress, on crops has been closely studied and an increase of carbon excretion was reported under drought stress in agricultural monocultures [33], although this response is expected to be different for mixtures of species [34]. One potential mechanistic explanation for the increased carbon exudation under stress conditions was proposed to be the loss of integrity of root cell membranes and malfunction of metabolism in affected cells [35]. On average, plants excrete 10%–20% of total assimilated carbon into the rhizosphere over the course of their life span [36,37], with lower amounts of deposition being reported for hydroponic cultures [38] or sterile-grown plants [37]. Rhizodeposited compounds are products of both primary and secondary metabolism [15] divided into two groups; low molecular weight (LMW) compounds, which dominate the exudates [27], and high molecular weight (HMW) compounds (Figure 1). Low molecular weight compounds are composed of water-soluble compounds present in the plant cytoplasm at high concentrations [18]. Other non-carbonaceous LMW components include protons, inorganic ions, and water [39]. Because LMW compounds are transported via passive transport, they are more prevalent than HMW compounds, whose transport requires the input of energy in plant materials. High molecular weight compounds include mucilage, enzymes, growth regulators, vitamins, and many SPMEs (i.e., flavonoids and allied phenolics, terpenoids, and alkaloids) [15,18,40]. Plant secondary metabolites (Figure 2) can have the following mechanistic functions: increasing availability of nutrients or their increased input to the plant, e.g., increase of phosphate solubility or uptake of metals to plants; establishing both positive and negative ecological relationships; acting as hormones or effectors for cell differentiation [41]. General classification of plant chemicals distinguishes primary and secondary metabolites based on whether they have an essential role for metabolism and are ubiquitous in plants [42,43]. However, the given mechanistic functions of SPMEs demonstrate that these metabolites are essential for plant protection from environmental stress, therefore they are important for plants in spite of the wide variety of synthesized compounds among plants [44]. Accordingly, here we consider SPMEs as products distinct from primary metabolism belonging to the following groups based on structure: (i) phenolic compounds including flavonoids; (ii) terpenoids and steroids; and (iii) compounds containing nitrogen or sulfur, such as alkaloids, glucosinolates, and non-ribosomal peptides and proteins [45,46,47,48,49,50]. It should be noted that the role of primary and secondary metabolites can overlap, as is demonstrated in defense against biotic stress, which can be in Arabidopsis thaliana mediated by glutathione [51] as well as by various SPMEs derived from indole-like camalexin or glucosinolates [52]. Different organisms do not necessarily use the same molecules for accomplishing the same function such as in metals acquisition, for which the bacterium Pseudomonas aeruginosa synthesizes secondary metabolites pyochelin and pyoverdine [53], while some plant roots release citric acid for the same function [54]. Similarly, under high heavy metal content in the environment, plants tend to produce compounds such as phytochelatins [55] or proline [56] to chelate the metals.LMW carbon-containing (LMW-C) components of the root exudates are hypothesized to be the reason for the primary response in rhizosphere microorganisms [14,57], resulting in a “priming effect” on the microbial community—the increase of microbial biomass and soil organic matter decomposition after the input of fresh organic matter [58]. A common example of the priming effect is found in a study that amended multiple soil types with citric acid and found a marked increase in carbon dioxide production and a shift in the relative abundance of β-Proteobacteria [59] in amended soils. HMW SPMEs have also been shown to cause major shifts in soil microbial community structure [60,61,62,63]. Rhizosphere microorganisms respond to SPMEs in different ways; not only can SPMEs serve as carbon and/or energy sources, but they often bear antimicrobial activity or the ability to disrupt bacterial quorum sensing [64,65]—a bacterial system for determination of cell density, which then further influences the gene expression of affected cells. Although some SPMEs have antimicrobial activity and their presence in a system will decrease community richness, some portion of the bacterial community can typically use antimicrobial compounds as a carbon and energy source, as long as the bacteria are not susceptible to the present compounds and those compounds do not reach levels inhibitory to microbial growth [66]. It should be emphasized that not only do plants control the amount of compounds excreted, but also the composition of exudates [67]. For example, if the soil moisture content is very high, oxygen availability will decrease and create anoxic conditions which will then cause plant cells to potentially accumulate lactate and ethanol to phytotoxic levels, and therefore the plant will release larger quantities of these compounds in an effort to decrease their intracellular levels [68]. Exudation patterns of crested wheatgrass (Agropyron cristatum (L.) Gaertn. cv. CD-II) were studied under drought, flooding, and nutrient limitation conditions, which revealed that malic acid was predominant among exuded organic acids. Drought led to a significant increase in wheatgrass-exuded organic acids and, under low potassium conditions, an increased amount of exudates was detected. This increase was not caused, however, by exudation of organic acids, suggesting that the wheatgrass released different compounds than acids such as saccharides [69]. During cultivation under phosphate deficiency, bean plants (Phaseolus vulgaris L.) increased the amount of exuded phenolic compounds [70], while maize (Zea mays L. var. Surprise) increased γ-aminobutyric acid and carbohydrate exudation [71]. Plant metabolites can also enter the rhizosphere through decomposition of deposited litter and below-ground root turnover. Plants annually support the growth of fine roots during spring and summer; however, between 40% and 70% of these fine roots die off in autumn as the plant prepares for winter [72], and these decaying fine roots provide rhizosphere microflora with nutrients and SPMEs. In addition, the decaying roots release air channels in the soil, which allow the increased oxygen flow necessary for most enzymatic activity and can preferentially be used by new roots in subsequent growing seasons [73]. The amount of SPMEs released by root turnover can be substantial, as demonstrated by a study on mulberry plants, which released the same amount of phenolic compounds during root turnover as was exuded by soybeans throughout the growth season [72,74]. 2.2. Root Exudates: Effect on the Rhizosphere Microflora It has long been known that there are more microorganisms living in close proximity to plant roots compared to bulk soil; this difference can be up to several orders of magnitude [6,40,75,76,77]. This phenomenon has been termed “the rhizosphere effect” and is potentially due to the increased amount of carbon excreted from roots into the rhizosphere [1,78]. When further examined, differences in microbial densities have been observed in different parts of the rhizosphere and correspond to the amount of exudates released by the particular specialized root cells, which differ along the root. For example, in the wild oat (Avena fatua L.) root tips had the highest populations of living rhizosphere bacteria, followed by root hairs; the mature root was the least populated [79]. Changes in exudate patterns not only affect microbial density but also strongly affect the structure [80,81] and function of microbial communities. These shifts then alter plant metabolic pathways and regulations [82,83], which then can retroactively result in changes of exuded compounds [84]. These mutually affected relationships across microbial communities and plants arise from the need for competitive regulation, because both the plant and soil microorganisms depend on resources present in the same soil. Root exudates can play an important role during chemotaxis by serving as attractants or repellents for soil microbiota [75]. For example, a strain of Rhodococcus erythropolis has been found to be attracted to phenolic compounds exuded from Arabidopsis thaliana L. roots [85], and phenolic acid root exudates have been shown to play a very important role in the nodulation process by serving as attractants for Rhizobium species [86,87]. Chemotaxis of beneficial microflora is also important for disease suppression in some plants due to competitive colonization of plants by symbionts protecting the host plant from pathogenic bacterium. One example of this is the plant symbiont Pseudomonas fluorescens, which is attracted to citric and malic acid released by tomato (Lycopersicon esculentum L.) roots [88] and can protect the host plant from a pathogenic bacterium, Ralstonia solanacearum, which is attracted to its host by diverse amino and organic acids and different SPMEs [89]. However, the same combination of SPMEs, amino acids, and other exuded compounds that attracts the beneficial bacteria also serves to attract potentially pathogenic organisms [90]. In addition to SPMEs and attracted microorganisms, plant resistance to a disease can also be influenced abiotically by available nutrients such as essential ions or nitrogen [57]. 2.3. Roots and Associated Microorganisms: How to Study Interactions? As mentioned above, the rhizosphere is shaped by complex interactions between plants and microorganisms, and the study of these interactions requires interlacing many different experimental approaches [91] (for an overview and examples, see Table 1). Growing roots are the main drivers of changes in the rhizosphere and therefore methods for studying their growth and morphology in situ have been developed such as root windows installed along the soil profile beneath studied plants [92], with imaging systems collecting data over a certain time period [93]. However, this approach visualizes root development on the observed interface and does not provide information on the whole root system of the studied plant. In order to access the 3D structure of a root system, transparent culture media suitable for plant cultivation have been developed [94], examples of which include Phytagel™ and Nafion™ [95]. Although useful for visualizing root growth, these artificial and clear culture media have different composition and physical properties than soil, which can have artificial influences on root growth. For example, roots grow faster and thinner when the penetration resistance is low [96]. In order to overcome such limitations in soil, computed tomography (CT) [97,98], magnetic resonance (MRI) [99,100], and neutron radiography [101] have been successfully implemented in order to provide images of root system with high resolution (for an extensive review of imaging technologies see Downie et al. [102], Oburger et al. [94], and York et al. [6]). Implementation of high-resolution imaging techniques has also allowed for tracing nutrient transport in plants following their release into the rhizosphere. For example, MRI [103] and neutron tomography [102] have been successfully used to monitor water uptake by roots. Chemicals in the rhizosphere have been traced by 11C positron emission tomography (11C-PET) combined with MRI, which allowed for tracing newly synthesized photosynthate throughout the plant and rhizosphere [104]. Tracing of chemical compounds in the rhizosphere is also possible using optodes—selective optical sensors—which have been implemented in studying oxygen gradient, pH, and CO2 dynamics in different plants [105,106,107,108]. Tracing of exuded compounds, including their uptake by rhizosphere microflora, can be studied through fluorescence-based methods. One fluorescence-based method uses biosensors—genetically modified (GM) microorganisms that express fluorescent proteins or are able to emit bioluminescence in the presence or absence of a studied molecule. The main advantage of such sensors is specificity in detection of studied microorganisms and compounds; however, GM microorganisms are distinct from the wild type and therefore may behave in different ways [109] (for a review on biosensors and their environmental application see Jusoh et al. [110]). In the past, biosensors have been developed for studying carbon flow in plant exudates [111,112], during nodulation [113], and for tracing bacterial quorum sensing [114] and root colonization [115]. In addition, biosensors have been used for tracing chemical compounds like nitrate [116], nitrogen [117], phosphorus [118], arsenic [119], saccharides and amino acids [120], and iron [121]. The light signal produced by a biosensor can also be detected in situ using fluorescent microscopy. Fluorescence in situ hybridization (FISH) is a microscopy method used for quantification of populations bearing selected genes [122,123]. This method is especially useful for studying biocontrol of plant pathogens [124,125,126]. Isotope probing methods based on radioisotopes (radioisotope probing, RIP) or stable isotopes (stable isotope probing, SIP) have been shown useful for determination of rhizodeposition rates and carbon flow from the plant into the rhizosphere. These techniques require incubation of plants in atmosphere-containing 14CO2 [22] or 13CO2 [127], respectively, and exudates or microorganisms feeding on the now labeled exudates can be determined. These microorganisms capable of utilizing labeled substrates can then be detected based on separation of biologically important markers with an incorporated “heavy” or radioactive isotope and further analysis of those labeled biomarkers [128,129,130,131]. Examples of biological markers that can be labeled in such experiments are phospholipid-derived fatty acids (PLFA), DNA, RNA, or proteins that can be analyzed by various methods. For direct observation of microorganisms interacting with radiolabeled plant exudates, FISH combined with microradiophotography (FISH-MAR) can be used [122]. For analysis of the plant-associated microbial community structure based on DNA or RNA, metagenomics [132] or metatranscriptomics [133], respectively, can be exploited. Metagenomics studies the complete DNA material present in an environment [134] and is commonly combined with high-throughput sequencing, while metatranscriptomics focuses solely on all present RNA molecules and thus is more suitable for evaluation of microbial response to environmental stimuli [135]. In other words, metagenomics can answer questions about which microorganisms reside in the rhizosphere and what is their metabolic potential (i.e., which functional genes the microorganisms have), while metatranscriptomics answers the question of which genes are expressed under studied conditions. In addition, metagenomics and metatranscriptomics can be dived into two groups: (i) unselected metagenomics and metatranscriptomics, which both focus on random sequencing of all informational molecules in the samples; and (ii) targeted metagenomics and metatranscriptomics, which both focus on a reduced pool of informational molecules and can be targeted based on sequence (in this form, high-throughput sequencing is called amplicon sequencing) or function [136]. In addition to metagenomics and metatranscriptomics, two other methods for studying plant–microbe interactions have been developed namely: (i) metaproteomics studying all proteins in an environmental sample (for review see Hettich et al. [137]); and (ii) metabolomics dealing with small molecular weight molecules present during cell metabolic processes (e.g., saccharides, amino acids, fatty acids, vitamins, secondary metabolites (for review see van Dam et al., [138])). 3. Role of Secondary Metabolites in Biodegradation of Organic Contaminants 3.1. Secondary Metabolites: Structural Similarities to Organic Pollutants As soon as a contaminant enters the environment, the microbial community reacts and begins to adapt to its presence. The organisms able to use the contaminant as a carbon/energy source or terminal electron acceptor will prosper and become more dominant in the community [157]. In order to metabolize the present contaminant, microorganisms must synthesize specific enzymes, the production of which usually requires an inductor such as the contaminant [158,159,160,161]. Some contaminants cannot serve as the primary substrate and provide cells with energy, and therefore are degraded via cometabolism with another compound, which does serve as the primary substrate (Figure 3) [162,163]. Some plant-derived compounds present in the rhizosphere can serve as primary substrates in cometabolism or inducers of degradative enzymes due to their structural similarities to the contaminants (Figure 4) [18,164], and therefore may provide the energy needed for contaminant metabolism [16,165]. The presence of large amounts of SPMEs in the rhizosphere, possibly serving as primary substrates and/or enzyme inducers, can explain the increased amount of microorganisms capable of degradation of pollutants that can be found in the rhizosphere versus the bulk soil [18,166]. Yet the spectrum of exuded compounds differs among plants and therefore stimulation of degradation may not be equal for all compounds and microorganisms [72]. Donnelly et al. [16] were the first group to demonstrate the ability of SPMEs to support the growth of contaminant-degrading bacteria and to enhance the bacterial degradative activity towards polychlorinated biphenyls (PCBs). After their work, many additional experiments demonstrating the stimulation of degrading activity towards PCBs have been published [63,167,168,169]. The effect of plants or their metabolites is usually compared to the most efficient inductors known and applied in laboratory experiments, such as biphenyl, which is a commonly used inductor of the PCB-degradation pathway [170,171]. In addition to PCBs, other pollutants have been identified that are more efficiently transformed in the rhizosphere than in the bulk soil, including PAHs and other petroleum hydrocarbons (PHs), pesticides, detergents, and explosives [172]. 3.2. Phenolics: From Simple Phenolics to Flavonoids and Lignin, Their Biological Role and Effect on Biodegradation of Pollutants Phenolic compounds represent a very diverse group of SPMEs. Their biosynthesis proceeds in a manner similar to the production of aromatic amino acids via the shikimate pathway, which occurs only in plants and microorganisms [173]. The main metabolite in this pathway is shikimic acid, which serves as the precursor of amino acids as well as benzoic and cinnamic acid [174]. Further modifications of these acids lead to lignins, lignanes, phenylpropenes, and coumarines [174]. Derivatives of cinnamic acid can be reduced, which leads to the synthesis of phenylpropanoids, or hydroxylated and used for the synthesis of coumarines or psoralens [175]. Using the polyketide biosynthetic pathway, cinnamic acid is transformed into 4-coumaric acid, which condensates first with acetyl coenzyme A and then with three combined acetate units derived from malonyl-CoA, leading to the synthesis of stilbens, which can be further extended to flavonoids. Flavonoids are widespread and very often occur in the form of glycosides. Most flavonoids compose of phenylbenzopyrone core and differ in the number and position of hydroxyl groups [176]. When cinnamic acid is derivatized, caffeic, ferulic, or sinapinic acid is produced and can lead to the biosynthesis of lignin, among other compounds. The chemical composition of monomers used is specific to the plant species, but usually alcohols derived from cinnamic acid (e.g., coniferyl and sinapyl alcohol) are polymerized by peroxidases, which catalyze hydrogenation of the alcohols followed by creation of free radicals freely joining together and creating lignin. Lignin and lignocellulose are some of the most widespread components of plant biomass [177] and their microbial degradation plays an important role in the carbon cycle [178]. Lignocellulose is composed of cellulose, hemicellulose, and lignin, which are linked by both covalent bonds and non-covalent interactions. These non-covalent interactions lead to the creation of a complex three-dimensional structure, which is resistant to chemical and biological degradation [179]. Nevertheless, metabolic pathways involved in lignin depolymerization have evolved in microorganisms. Although some bacteria are involved in lignin biodegradation [180,181], degradation by ligninolytic white rot fungi has been investigated in closer detail [182]. The functional group of white rot fungi belong mostly to the Basidiomycetes class. The name “white rot” is derived from the white color, which appears on wood after the fungi degrade lignin while leaving cellulose behind. Microbial degradation of lignin proceeds extracellularly and is mediated by enzymes with low substrate specificity, such as peroxidases and laccases. The peroxidases participating in lignin degradation are expressed under nitrogen-limiting conditions [183,184] and are divided into three groups: lignin peroxidases (LiPs, EC 1.11.1.14), manganese peroxidases (MnPs, EC 1.11.1.13), and versatile peroxidases (VPs, EC 1.11.1.16). LiPs are typically present in the form of isoenzymes which are able to depolymerize lignin via oxidation of veratryl alcohol and creation of its radical cations, which then leads to oxidation of not only phenolic compounds but also a wide variety of compounds including even non-phenolic compounds like amines, aromatic ethers, or polycyclic aromatic compounds. MnPs are structurally similar to LiPs [185]; however, they depolymerize lignin by a slightly different mechanism, namely via the release of Mg3+-oxalacetate, which selectively oxidizes phenolic compounds and creates phenoxy radicals indirectly responsible for lignin oxidation. VPs have been reported to dispose of both the activities of LiPs and MnPs. Other examples of enzymes that can have a role in lignin degradation are cytochrome P-450 monooxygenases [186], enzymes oxidizing cellobiose, dehydrogenating amyl alcohol derivatives, or quinone reductases [177,187]. In addition to peroxidases, other proposed indirect mechanisms of lignin degradation are based on Fenton’s reaction (i.e., the creation of hydrogen peroxide with Fe(II) ions) and yield hydroxyl radicals. The hydroxyl radical is a strong non-selective oxidant and is involved in lignin modification and polysaccharide degradation [188]. The hydrogen peroxide necessary for the proposed mechanism has been shown to be produced by several white and brown rot fungi [177], either by the production of small phenolic compounds reducing iron and leading to H2O2 [188] or by enzymatic activity, e.g., aryl-alcohol oxidase (EC 1.1.3.7) or cellobiose dehydrogenase (EC 1.1.99.18) [177]. In some white rot fungi, ligninolytic culture conditions have often been associated with the degradation of PCBs of PAHs (for an overview see Table 2). Many studies reported the degradation of PCBs by Pleurotus ostreatus under both nitrogen-limiting and -rich conditions [189], and by Phanerochaete chrysosporium only under nitrogen-limiting conditions [190]. However, the degradation of PCBs in Phanerochaete chrysosporium was proved to be mediated by radical attack and not due to enzymatic activity [191]. Other examples of the contaminant-degrading capabilities associated with white rot fungi include reported degradation of certain PAHs by liquid culture of Irpex lacteus [192,193] or by Pleurotus ostreatus [194], which also stimulated the growth of Actinobacteria capable of the degradation of PAHs [195]. Later, MnPs were also reported to be involved in the degradation of PAHs [196]. In addition, the fungus Lentinus tigrinus [197] was described as capable to degrade PAHs using laccase under nitrogen-rich conditions and MnP under nitrogen-limiting conditions. The analysis of degradation products of PAHs indicated activity of cytochrome P-450 in the early stages of cultivation, suggesting that it might be used in the initial hydroxylation of those substrates in fungi. White rot fungi were also reported to be able to degrade other organic pollutants like polychlorinated dibenzo-p-dioxins [198], trinitrotoluene [199], lindane [200], and pentachlorophenol [201]. Some bacteria, such as actinomycetes, also possess the ability to degrade lignin [214]—although, unlike some fungi, they do not carry genes for complete depolymerization and biotransformation of lignin. In addition, the bacterial peroxidases are reported as less powerful for lignin oxidation [181]. One of the enzymes possibly used for bacterial extracellular oxidation is a dye-decolorizing heme peroxidase that is able to oxidize non-phenolic components of lignin. This peroxidase, among other degradative enzymes, has been found in Rhodococcus jostii RHA1 [215], a well-described potent degrader of organic contaminants [216]. In addition to water conduction and supporting structure, lignin is used for harmless storage of partially metabolized contaminants in plants. Ligninolytic activity of microorganisms leads to repeated release of lignin-building units including phenolic compounds as well as partially plant-metabolized contaminants, which are usually more polar and bioavailable [217] and can be further transformed. For instance, chlorobiphenyls have been shown to be hydroxylated in plant cells [218] and these hydroxy-derivatives have been shown to be further transformed by enzymes of the biphenyl degradation pathway [219,220,221,222], supporting the hypothesis that biphenyl-degrading bacteria play a role in the degradation of final products derived from lignin [171]. More recently, evidence was provided that the biphenyl catabolic pathway evolved in some bacteria to allow for the metabolism of SPMEs in soil [223]. The connection between phenolic compounds and the biphenyl catabolic pathway has been the focus of studies for decades (for an overview see Table 2). Donnelly et al. [16] tested the growth of three biphenyl-degrading bacteria, Cupriavidus necator H850 (formerly Alcaligenes eutrophus), Burkholderia xenovorans LB400 (formerly Pseudomonas sp.), and Rhodococcus globerulus P6 (formerly Corynebacterium sp. MB1), on 13 phenolic SPMEs serving as a sole carbon source. For Cupriavidus necator H850, several compounds including apigenin, catechin, or morin served as a better growth substrate than biphenyl, and the best cometabolic degradation of PCBs was observed while growing on naringin. For the strains LB400 and P6, the fastest growth was found during cultivation with biphenyl as the sole C-source; however, these strains were also able to utilize some phenolic compounds like myricetin, catechin, or chrysin as growth substrates. The highest PCB-degradation activity was observed for Burkholderia xenovorans LB400 during growth on myricetin as the sole C-source and for Rhodococcus globerulus P6 on coumarin. In addition, the strain Cupriavidus necator H850 has been shown to use salicylic acid as a growth substrate and inducer of PCB cometabolism [167]. In a more recent experiment, Toussaint et al. [85] studied the ability of root exudates from Arabidopsis thaliana to support the growth of Rhodococcus erythropolis U23A and induce PCB-degradation activity. Although the bacterium was able to utilize concentrated root exudates as a sole growth substrate, single flavonoids detected as main components in the exudates were not able to support the growth of the bacterium. Nevertheless, when Rhodococcus erythropolis U23A was grown on root exudates as a sole carbon source, it was able to cometabolically convert 4-chlorobiphenyl to 4-chlorobenzoic acid and metabolize three of 18 tested PCB-congeners. Flavanone was detected as the most abundant compound in the root exudates and its ability to induce the biphenyl degradation pathway in U23A was confirmed. This research demonstrated the ability of Arabidopsis thaliana root exudates to increase the level of biphenyl-catabolizing enzymes above the basal level. In a follow-up experiment [224], growth and induction conditions were optimized, and flavanone, flavone, and isoflavone were detected as significantly better inducers of biphenyl pathway than biphenyl itself. However, the primary growth substrate increased the efficiency of the expression of biphenyl degradation pathway, as can be demonstrated in the case of isoflavone, which was a strong inducer during the growth on sodium acetate, mannitol, and sucrose, but its induction capability was much weaker during the growth on glucose or mannose [223]. The ability of plant-derived phenolics to increase degradation of pollutants beyond PCBs has also been the subject of research [225,226,227,228,229] and can be demonstrated by an experiment by Scheublin et al. [211], who investigated the transcription profiles of a Gram-positive bacterium Arthrobacter chlorophenolicus A6 after the growth on leaves of common bean (Phaseolus vulgaris) and on tryptic soy agar (TSA) with and without the addition of 4-chlorophenol (4-CP). The authors discovered that a subset of cph genes for the degradation of 4-CP was expressed after the growth on the leaves. It was hypothesized that the genes encoding for enzymes transforming hydroquinone to 3-oxoadipate were induced by hydroquinone detected in leaf washes. 3.3. Terpenes: Biological Role and Effect on Biodegradation of Pollutants Terpenes are a class of natural compounds, mostly of plant origin, which are composed of two or more isoprene units. The biological synthesis of terpenes begins with the condensation of isopentenyl diphosphate and dimethylallyl diphosphate, which are both synthesized from three acetyl coenzyme A units at the beginning of the mevalonate pathway. Terpenes are classified based on the number of condensed isoprene units and are responsible for characteristic plant fragrances and essential oils in many fruits (e.g., citrus fruits) and herbs; they can also act as chemoattractants or repellents [230]. Generally, microbial uptake of terpenes is challenging due to their volatility, low water solubility, and common antimicrobial properties. Nevertheless, at the beginning of the 1960s a bacterium Pseudomonas sp. capable of growth using d-limonene, one of the most common terpenes, as a sole carbon source was described [231]. After this discovery, other bacteria capable of terpene utilization have been described [232,233,234,235]. To date, five different microbial biotransformation pathways for limonene have been proposed [236]. The main utilization pathway starts with the hydroxylation of C7 and yields in perillic alcohol, which is further oxidized to perillic acid and further metabolized via pathway similar to β-oxidation. The ability to metabolize d-limonene through perillic alcohol does not seem to be species-specific as it was described in micromycetes (Hormonema sp. [237]), yeasts (Yarrowia lipolytica [238]), and bacteria (Bacillus stearothermophilus BR388 [239] or Pseudomonas spp. [231,240]). One of the other biotransformation pathways results in the production of (3R)-3-isoprenyl-6-oxoheptanoyl-CoA, as was described in bacterium Rhodococcus erythropolis DCL14 [236]. This bacterium, similarly to Pseudomonas fluorescens [241], can metabolize limonene through limonene 1,2-epoxide and limonene 1,2-diol. Another hydroxylation leads to hydroxyl ketone, which undergoes transformation via a Baeyer–Villiger reaction mediated by an oxygenase. The resulting product is then metabolized through β-oxidation [236]. Other widespread terpenes are α- and β-pinene, which can be found in pine bark and needles, and represent byproducts formed during cellulose synthesis. The bacterium Pseudomonas sp. PIN has been implicated in the metabolism of pinene via limonene, although the exact pathway for this metabolism is unclear [232]. However, a large amount of perillic acid was detected, suggesting that Pseudomonas sp. PIN uses the same metabolic pathway for the transformation of limonene as Rhodococcus erythropolis DCL14. From the wide group of terpenes, limonene, carvone, and pinene have been the most intensely studied as possible primary substrates for cometabolic degradation of pollutants (for an overview see Table 2). Tandlich et al. [168] investigated the possible induction effect of limonene and carvone in bacterium Pseudomonas stutzeri, a known degrader of PCBs. The strain was grown on xylose or glycerol used as carbon sources; the terpenes were used in concentrations of 10 and 20 mg·L−1 and the degradation of formerly used PCB-mixture Delor 103 was analyzed. The results showed that, after induction, the bacterium degraded a higher amount of PCBs during the growth on xylose in comparison to the growth on glycerol, with the most promising system being cultivated on xylose and induced by carvone. Though PCB degradation during growth on glycerol was lower, a broader spectrum of congeners was degraded. When the bacterium Pseudomonas stutzeri was induced by limonene, it degraded an increased amount of higher-chlorinated congeners and the increase in degradation was dependent on the concentration of the inducer. In contrast to the Tandlich study, Gilbert et al. [207,208] found that carvone derived from mint (Mentha spicata) induced PCB degradation by Arthrobacter sp. B1B. Carvone in combination with surfactants has also been investigated in an effort to increase the bioavailability of Aroclor 1242—a commercial mixture of PCBs—for microbial degradation by bacterial strains Arthrobacter sp. B1B and Cupriavidus necator H850 in artificially contaminated soil [167]. Application of carvone-grown Arthrobacter sp. B1B in combination with (i) sorbitan trioleate in ratio 1:10; (ii) Cupriavidus necator H850 in ratio 1:1; or (iii) fructose in a 1:10 ratio led to a doubled degradation of PCBs in soil in comparison with non-bioamended controls after 18 weeks of repeated amendment. The differences in removal of PCBs were not significant among different amendments; however, sorbitan trioleate was demonstrated to support growth of the inoculum and increase the bioavailability of PCBs and degradation of higher-chlorinated congeners. Co-inoculation of both strains did not lead to enhanced removal of PCBs; nevertheless, an increase in degradation of several multiple ortho-substituted congeners was detected. Terpenes have also been demonstrated to induce cometabolism of pollutants other than PCBs. For example, degradation of 2,4-dichlorophenol (2,4-DCP)—product of 2,4-dichlorophenoxyacetic acid metabolization—was studied after the addition of limonene and α-pinene in three different soil types [210], one of which was sampled from grassland covered bog, and the others from below pine and oak trees. Mineralization of 2,4-DCP was comparable among samples, but α-pinene showed a better effect than limonene in soil derived from pine surroundings and bog, suggesting that the bacteria in those environments may be better adapted to α-pinene’s presence. As another example of chlorophenolics, pentachlorophenol (PCP) has been proposed to be metabolized by Arthrobacter sp. B1B induced by l-carvone [242] in the same manner as PCB mixtures [208]. Furthermore, cumene (isopropylbenzene) was demonstrated as a suitable growth substrate for enrichment cultures of bacteria capable of degradation of trichloroethylene (TCE) [213]—a widespread water contaminant originating from extensive use of chlorinated solvents. Cumene induced TCE degradation capability in the bacterium Rhodococcus gordoniae P3, which was detected both in pure liquid culture and soil [212]. Later, cometabolic biotransformation of TCE was demonstrated using other terpenes and terpenoids, namely R-carvone, S-carvone, linalool, and cumene as growth substrates for indigenous bacterial communities from a site contaminated by TCE in the UK [243]. 4. Complex Effect of Plant Metabolites on Bioremediation of Contaminated Soil The secondary compound hypothesis suggests that SPMEs released into the environment by root exudation or plant litter decomposition affect soil microbial populations and can stimulate their metabolic activities toward degradation of organic contaminants [16,17,18]. Although we have presented evidence supporting this hypothesis, these studies were based mostly on experiments with pure cultures. With the implementation of high-throughput sequencing technologies, researchers are able to assess the whole community structure in any environment; therefore, the method is suitable for the assessment of the role of SPMEs in the entire microbial community [244,245]. Through the development of SIP, researchers have gained the ability to directly link microbial community structure and function based on incorporation of 13C, 15N, or 17O derived from labeled compounds [128,129,246]. These microbial ecology techniques have contributed to a successful understanding of ongoing processes in contaminated sites, which is important for the improvement of bioremediation processes (for an overview of this section see Table 2). The effect of individual SPMEs on microbial community structure and degradation of PCBs in a long-term contaminated soil [245] was investigated in response to eight-week repeated soil amendment with limonene, naringin, and caffeic acid [63,247]. Bacterial diversity was reduced in all samples compared to the control/non-amended soil, with caffeic acid being associated with the largest reduction in community diversity. The metabolism of 4-chlorobiphenyl in the bacterial community revealed activity of Proteobacteria in all samples; however, differences at lower taxonomic levels were detected. In non-amended soil, active populations of Pseudomonas, Rhodanobacter, Azoarcus, Porphyrobacter, and Gemmatimonas were detected, from which only Azoarcus was detected in amended soils, namely in soil amended with limonene. Otherwise, the soils with limonene and naringin additions were dominated by Hydrogenophaga genus, while the soil amended with caffeic acid was found to be dominated by the genus Burkholderia. Additionally, patterns of degraded PCB congeners among the samples differed and, although caffeic acid-amended soil harbored less diverse bacteria, a broader spectrum of degraded congeners was detected. Importantly, degradation of usually very persistent higher-chlorinated biphenyls was detected. SPMEs are commonly present in soil as a complex mixture, with different SPMEs having differential impacts on the microbial community. One study examined the effects of a complex mixture of SPMEs from plant litter on the biodegradation of Aroclor 1242 in soil. Hernandez et al. [61] amended soil with either orange peel, ivy or eucalyptus leaves, or pine needles and assessed the disappearance of PCBs over six months. Under these amendment conditions, complete mineralization of present PCBs was achieved, while degradation of only lower chlorinated congeners was detected in the control sample. Additionally, all types of soil amendment led to a five-fold increased abundance of cultivable biphenyl-utilizing bacteria [16]. Following the study by Hernandez et al. [61], it is relevant to present new insights from a study focused on the effects of grapefruit peel, lemon peel, and pears on the changes in bacterial communities and their PCB-degradation activity in long-term contaminated soil (for a more detailed description of the methods used, see Appendix A). The plant litter used was rich in naringin, limonene, and caffeic acid, respectively [248]. Additionally, these SPMEs have already been associated with an enhancing of degradative activity towards PCBs [63]. Our results confirmed that the natural materials rich in selected SPMEs changed the community structure, as can be seen in Figure A1. This observation was further supported by obtaining phylogenetically different and more abundant biphenyl-utilizing isolates during cultivation, which is a commonly reported phenomenon [61,203]. Although the amount of cultivable biphenyl-utilizing bacteria increased, the diversity in soil decreased in the following order: the untreated soil was the most diverse, followed by soil samples treated by pears and grapefruit peel, with the soil amended with lemon peel being the least diverse. A decrease in species richness can in some cases lead to the inability of an ecosystem to keep providing its services (e.g., biomass turnover) [249] and was reported as significantly influenced by plant litter [250]. However, in this experiment, the ability of present microflora to degrade 4-chlorobiphenyl and benzoic acid was retained, as was demonstrated by SIP. Different bacteria deriving carbon from 4-chlorobiphenyl and benzoic acid were detected in each treatment. The incubation of soil with natural materials led to changes in the patterns of detected PCB congeners, suggesting an ongoing, preferential biodegradation process [247]. These results demonstrate a possible usage of lemon and grapefruit peels, which are typically a waste product from juice production, for bioremediation of contaminated sites. Both presented experiments demonstrate that SPMEs derived from plant litter affect soil microflora similarly as plant-exuded SPMEs with the possibility to stimulate degradative activities towards PCBs. As was described earlier, the root turnover is an important source of SPMEs in soil. The effect of 43 different plants’ root tissues on the removal of the PAHs pyrene and benzo[a]pyrene from soil was tested by Yi and Crowley in a series of experiments [209]. From all the tested plants, only four stimulated degradation, namely radish (Raphanus sativus L.), potato (Solanum tuberosum L.), carrot (Daucus carota L.), and celery (Apium graveolens L.). In these plants, the authors further focused on the compounds probably responsible for the stimulation of degradation activity in the soil. In celery, the most effective stimulator in the experiment, terpenes and derivatives of salicylic acid were identified as the specific compounds responsible for the increase in removal of PAHs, though the application of these SPMES in their pure form did not prove successful for degradation of PAHs, indicating that an unspecified and uninvestigated synergistic effect was causing the degradation activity. Therefore, a more generic approach was considered and linoleic acid (an unsaturated C18:2 fatty acid) was detected as the only common factor in the measured plants. After these findings, the authors conducted an experiment focused on the comparison of rhizodeposition by celery (Apium graveolens L.) and wheat (Triticum aestivum L.), celery root crushate, and the addition of linoleic acid or its sodium salt. In the pots with celery plants or root crushate, and in those with linoleic acid or sodium linoleate, the degradation of benzo[a]pyrene and pyrene was fast and comparable among the treatments. On the other hand, wheat did not enhance the removal of PAHs as their residual amounts were comparable to the unplanted control. Although the role of linoleic acid in the stimulation of degradation of PAHs was not the topic of the presented research, the authors suggested it might increase the number of degrading microorganisms, or more likely serve as a biosurfactant increasing the bioavailability of PAHs. Uhlik et al. [202] have also investigated the potentially stimulatory effects of growing certain plants in contaminated soils, namely the effect of horseradish (Armoracia rusticana P. Gaertn., B. Mey. et Scherb.) grown in soil contaminated with PCBs on bacterial communities in the rhizosphere deriving carbon from biphenyl. Changes in composition of bacterial communities after planting horseradish into the soil were detected; in the rhizosphere, only proteobacterial sequences were detected to derive carbon from biphenyl, while the bulk soil contained in addition sequences mostly belonging to Firmicutes. Looking for a suitable plant capable of stimulating microbial PCB degradation during remediation process, Leigh et al. [203] focused on autochthonous PCB-degrading bacteria associated with mature trees naturally colonizing a PCB-contaminated site and changes in their abundance in dependence to seasonal changes and soil depth. No significant differences were detected among the samples in the uppermost layer, which contained roots of widespread grasses and forbs in addition to the tree (Austrian pine—Pinus nigra J. F. Arnold, ash—Fraxinus excelsior L., two weeping birches—Betula pendula Roth, goat willow—Salix caprea L., or black locust—Robinia pseudoacacia L.) roots. In the middle soil layer, where only tree roots were present, again no significant differences were detected between June and August. However, in samples collected in November and May, a significantly higher number of PCB-degraders was isolated from the pine root zone. The deepest layer of willow root zone harbored the highest amount of PCB-degraders. These results suggest that Austrian pine and willow support the growth of PCB-metabolizing bacteria and can be suitable candidates for rhizoremediation. Therefore, Leigh et al. [251] studied the pine root zone microbiome in closer detail, focusing on bacteria deriving carbon from biphenyl and their functional genes. When the 13C-labeled metagenome was analyzed using GeoChip functional array [252], 28 different genes associated with the degradation of aromatic hydrocarbons were detected, revealing several genes of the β-ketoadipate pathway, which is common in soil bacteria during microbial degradation of many SPMEs. As was mentioned above, the amount and composition of root exudates is dependent on environmental conditions [26,33]. Therefore, some trees growing in higher latitudes have been reported as producing a higher amount of secondary metabolites than similar tree species growing at lower latitudes [253], which could possibly lead to increased stimulation of degradation capabilities of present microflora. In 1996, a bioremediation project was initiated in Alaska with soil contaminated by diesel and crude oil, vegetated with annual ryegrass (Lolium multiflorum Lam.) alone or in a mixture with red fescue (Festuca rubra L.), and fertilized. After two years, plots that had been vegetated and fertilized had significantly increased petroleum hydrocarbon loss when compared to unamend sites [205]. Fifteen years later, Leewis et al. [254] reexamined the site and described the long-term effects of phytoremediation and nutrient amendment on the site. In a preliminary screening, a decrease of 80%–95% in the last reported values of contaminants was detected, which suggested ongoing bioremediation processes at the site. In addition, with an increase in trees, a decrease in petroleum hydrocarbons concentration was observed, and the diesel contaminated site was colonized by a higher amount of plants. Interestingly, the non-native annual grasses were not found on the site anymore; only native trees and seedlings were detected with willow (Salix sp.), Alaskan birch (Betula neolaskana), white spruce (Picea glauca (Moench) Voss), and balsam poplar (Populus balsamifera L.) dominating the site. When they assessed microbial diversity, the researchers reported patterns dependent on the original applied soil treatments. 5. Conclusions and Future Perspectives Secondary metabolites present in plant fruits, leaves, or exudates alter the composition of bacterial communities and their metabolic pathways. These interactions can be potentially exploited to increase the effectiveness of bioremediation techniques, especially phytoremediation. Although some trees such as poplars and willows have been used for phytoremediation, the overall mechanism and roles of all participating organisms are not fully understood [206]. Furthermore, there are still gaps in understanding which plant-derived compounds stimulate the microbial degradation of specific contaminants. For instance, some preliminary results indicate that plant-derived compounds promote the bacterial degradation of cis-1,2-dichloroethylene (Fraraccio and Uhlik, unpublished data), which often accumulates as a degradation product of tetrachloroethylene. Such new information could open new avenues for bioremediation research investigating the link between plants and microbial degradative activity. In recent years, research has expanded beyond the rhizosphere to investigate the role that endophytes—microorganisms living inside plant tissue that do not cause visible harm to the host plant [255,256]—play in bioremediation. Several studies have demonstrated that colonization of plants by endophytes is beneficial due the plant-growth promoting effects of some bacteria. Colonization of plants by such bacteria is important for the growth in a contaminated environment as it leads to the increase of a plant’s resistance or to the decrease in accumulation of the pollutant in the plant due to microbial degradation [257,258,259]. In one case, the endophytes were described to have more degradative genes than were present in the rhizosphere [260], indicating that plants may intentionally attract bacteria harboring degradative genes and provide them with habitat and nutrients. However, very little is still known about SPMEs used for specific communication between plants and their associated bacteria or microbial species selection by plants in the dependence on environmental conditions. These questions could be addressed by SIP experiments designed to track the flow of carbon derived from the plant through the rhizosphere community, looking for patterns of expressed enzymes in both plants and microorganisms, and identifying the role of different metabolites maintaining the relationships. Acknowledgments The authors are grateful to Jachym Suman for his assistance with graphics, Miluse Hroudova for her assistance during pyrosequencing, Vlasta Dudkova for her assistance during analysis of PCB content, and Mary-Cathrine Leewis for her comments on the manuscript. This manuscript benefitted from the financial support of the Czech Ministry of Education, Youth and Sport project No. LH14004 and Czech Science Foundation project No. 14–32432S. Author Contributions Lucie Musilova performed the experiments and wrote the manuscript, Jakub Ridl pyrosequenced amplicons, Marketa Polivkova wrote the manuscript, Tomas Macek designed the experiment and wrote the manuscript, and Ondrej Uhlik designed the experiment, analyzed the data, and wrote the manuscript. Conflicts of Interest The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results. Abbreviations SPMEs secondary plant metabolites PAHs polycyclic aromatic hydrocarbons PCBs polychlorinated biphenyls LMW low molecular weight LMW-C low molecular weight carbon containing HMW high molecular weight LiP lignin peroxidase MnP manganese peroxidase CT computed tomography MRI magnetic resonance imaging PET positron emission tomography FISH fluorescence in situ hybridization FISH-MAR fluorescence in situ hybridization combined with microradiography RIP radioisotope probing PLFA phospholipid-derived fatty acids SIP stable isotope probing VP versatile peroxidase DCP dichlorophenol CP chlorophenol TNT trinitrotoluene Appendix A A.1. Description of Our Experiment The long-term PCB contaminated soil was collected in the dumpsite in Lhenice in 2009. The soil was mainly contaminated by the mixtures of PCBs Delor 103 and 106 with level of chlorination similar to the Aroclor 1242 and 1260, respectively [261]. The amount of natural materials used was calculated in order to achieve the concentration of limonene, naringin or caffeic acid corresponding to 2 g per g of the soil, as was used in our previous study [63]. In order to avoid any effect caused by food additives, the natural products were washed with detergent prior to homogenization. The homogenized plant materials were added to 1.5 kg of soil per sample repeatedly every six weeks for 18 weeks. After the end of the incubation time, bacterial diversity was assessed using pyrosequencing of 16S rRNA gene amplicons in the same way as described elsewhere [262] and using cultivation on biphenyl as a sole carbon source [263]. From samples collected at the same time point, microcosms for stable isotope probing with 13C-4-chloro-biphenyl and 13C-benzoate were constructed and incubated for 14 days and microbial diversity was assessed in the same way as total communities [262]. Figure A1 Total community composition in soils after amendment with natural materials represented as bacterial phyla or classes in the case of Proteobacteria. Bulk soil was not amended by any material. Figure 1 Carbon flow in plants—carbon dioxide is assimilated by plants and used for synthesis of metabolites, which are used in anabolism or released by rhizodeposition into the rhizosphere. Root exudates further affect soil properties and residing microbiota. Figure 2 General overview of SPME classification and functions. Figure 3 Schematic visualization of cometabolism, differences between primary substrates and cometabolites, and role of secondary plant metabolites (SPMEs) in cometabolism of pollutants. Figure 4 A few examples of structural similarities between contaminants and plant secondary metabolites (SPMEs), adapted from Singer et al. [18]. (a) pyrene; (b) confusarine; (c) 3,8-dichlorodibenzo-p-dioxin; (d) xanthone; (e) 4-chlorobiphenyl; (f) naringin. ijms-17-01205-t001_Table 1Table 1 An overview of methods used for describing roots and plant–microbe interactions in the rhizosphere. Object of Study Method Reference Example of Use root growth and morphology observation windows + imaging system [92,93] [92,93] transparent culture media, e.g., PhytagelTM or NafionTM [94,95] [95] computed tomography (CT) [97,139] [97,139] magnetic resonance (MRI) [99,140] [99,140] neutron radiography [101] [101] nutrient transport magnetic resonance (MRI) [103] [103] neutron tomography [102] [102] 11C-positron emission tomography (11C-PET) [104] [141] optode [106,107,108] [105,106,107,108] radioisotope labelling (RIP) [22] [22] stable isotope labelling (SIP) [127] [142,143] interactions plant–microbes biosensors [110] [111,112,113,114,115,116,117,118,119,120,121] fluorescence in situ hybridization [122] [122,124,125,126] metagenomics [134] [63,144,145,146] metatranscriptomics [135] [147,148,149,150] metaproteomics [151] [152,153,154] metabolomics [155] [155,156] ijms-17-01205-t002_Table 2Table 2 Examples of treatments that led to changes in microbial community structure and activity towards contaminants. Contaminant Treatment Observed Effect Reference PCBs limonene reduction in diversity of bacterial community; [63] community dominated by Hydrogenophaga; Azoarcus and Hydrogenophaga dominated utilization of 4-chloro-13C-biphenyl; naringin reduction of diversity of bacterial community; Hydrogenophaga dominated utilization of 4-chloro-13C-biphenyl; caffeic acid largest reduction in diversity of bacterial community; Burkholderia dominated utilization of 4-chloro-13C-biphenyl; degradation of higher-chlorinated PCBs PCBs orange peel complete mineralization of PCBs; [61] increased abundance of cultivable biphenyl-utilizing bacteria; ivy leaves complete mineralization of PCBs; increased abundance of cultivable biphenyl-utilizing bacteria; eucalyptus leaves complete mineralization of PCBs; increased abundance of cultivable biphenyl-utilizing bacteria PCBs grapefruit peel reduction in diversity of bacterial community; this paper Hydrogenophaga, Caulobacter, and Skermanella dominated utilization of 4-chloro-13C-biphenyl; Azotobacter dominated utilization of 4-chloro-13C-biphenyl; increased abundance of cultivable biphenyl-utilizing bacteria; lemon peel largest reduction in diversity of bacterial community; Nocardioides dominated utilization of 4-chloro-13C-biphenyl; Skermanella dominated utilization of 4-chloro-13C-biphenyl; increased abundance of cultivable biphenyl-utilizing bacteria; pears reduction in diversity of bacterial community; Azotobacter dominated utilization of 4-chloro-13C-biphenyl; increased abundance of cultivable biphenyl-utilizing bacteria PCBs horseradish Hydrogenophaga dominated utilization of 13C-biphenyl [202] PCBs Austrian pine increased abundance of cultivable biphenyl-utilizing bacteria [203] ash weeping birch goat willow black locust PCBs horseradish microbial populations of the root zone of each plant significantly differed from one another and/or from the bulk soil [204] black nightshade tobacco diesel and crude oil annual ryegrass enhanced bioremediation [205] red fescue diesel oil Alaskan willow willow had a significant role in structuring the total bacterial community and resulted in significant decreases in diesel range organics [206] PCBs nitrogen-rich fungus Pleurotus ostreatus disposes of PCB-degradation activity [189] nitrogen-limiting PCBs nitrogen-limiting fungus Phanerochaete chrysosporium disposes of PCB-degradation activity [190] PCBs naringin bacterium Cupriavidus necator H850 disposes of PCB-degrading activity while grown on the compounds as carbon sources [16] apigenin catechin morin salicylic acid [167] PCBs myricetin bacterium Burkholderia xenovorans LB400 disposes of PCB-degrading activity [16] catechin chrysin PCBs limonene bacterium Pseudomonas stutzeri disposes of PCB-degrading activity while grown on the compounds as carbon sources [168] carvone PCBs Mentha spicata (carvone) bacterium Arthrobacter sp. B1B disposes of PCB-degrading activity while grown on the compound as a carbon source [207,208] 4-chlorobiphenyl (PCB 3) Arabidopsis thaliana exudates (flavanone) bacterium Rhodococcus erythropolis U23A disposes of PCB-degrading activity while grown on the exudates as a carbon source [85] PAHs radish (terpenes, salicylic acid) enhanced bioremediation [209] potato carrot celery PAHs not specified fungus Phanerochaete chrysosporium disposes of PAH-degrading activity [194] PAHs not specified fungus Irpex lacteus disposes of PAH-degrading activity [192,193] PAHs nitrogen-rich fungus Lentinus tigrinus disposes of PAH-degradation activity [197] PCP not specified fungus Pleurotus ostreatus disposes of PCP-degradation activity [201] PCP not specified fungus Irpex lacteus disposes of PCP-degradation activity [201] PCP not specified fungus Trametes versicolor disposes of PCP-degradation activity [201] PCP not specified fungus Bjerkandera adusta disposes of PCP-degradation activity [201] PCP carvone bacterium Arthrobacter sp. B1B disposes of PCP-degradation activity [210] 4-chlorophenol (4-CP) leaves of Phaseolus vulgaris bacterium Arthrobacter chlorophenolicus A6 disposes of 4-CP-degradation activity [211] TCE cumene bacterium Rhodococcus gordoniae P3 disposes of TCE-degradation activity [212] TCE cumene bacterium Pseudomonas sp. JR1 disposes of TCE-degradation activity [213] TCE cumene bacterium Rhodococcus erythropolis BD1 disposes of TCE-degradation activity [213] TNT not specified fungus Pleurotus ostreatus disposes of TNT-degradation activity [199] TNT not specified fungus Phanerochaete sordida disposes of TNT-degradation activity [199] TNT not specified fungus Phlebia brevispora disposes of TNT-degradation activity [199] TNT not specified fungus Cyathus stercoreus disposes of TNT-degradation activity [199] lindane intermediate nitrogen concentration fungus Pleurotus ostreatus disposes of lindane-degradation activity [200] dioxins not specified fungus Panellus stipticus 99–334 disposes of dibenzo-p-dioxins-degradation activity [198] PCBs, polychlorinated biphenyls; PAHs, polyaromatic hydrocarbons; PCP, pentachlorophenol; TCE, trichloroethylene; TNT, trinitrotoluene. ==== Refs References 1. Hartmann A. Rothballer M. Schmid M. Lorenz Hiltner, a pioneer in rhizosphere microbial ecology and soil bacteriology research Plant Soil 2008 312 7 14 10.1007/s11104-007-9514-z 2. Bakker P.A.H.M. Berendsen R.L. Doornbos R.F. Wintermans P.C.A. Pieterse C.M.J. The rhizosphere revisited: Root microbiomics Front. Plant Sci. 2013 4 165 10.3389/fpls.2013.00165 23755059 3. Smalla K. Wieland G. Buchner A. Zock A. Parzy J. Kaiser S. Roskot N. Heuer H. Berg G. Bulk and rhizosphere soil bacterial communities studied by denaturing gradient gel electrophoresis: Plant-dependent enrichment and seasonal shifts revealed Appl. Environ. Microbiol. 2001 67 4742 4751 10.1128/AEM.67.10.4742-4751.2001 11571180 4. Berg G. Smalla K. Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere FEMS Microbiol. Ecol. 2009 68 1 13 10.1111/j.1574-6941.2009.00654.x 19243436 5. Schreiter S. Eltlbany N. Smalla K. Microbial communities in the rhizosphere analyzed by cultivation-independent DNA-based methods Principles of Plant–Microbe Interactions: Microbes for Sustainable Agriculture Lugtenberg B. Springer International Publishing New York, NY, USA 2015 289 298 6. York L.M. Carminati A. Mooney S.J. Ritz K. Bennett M.J. The holistic rhizosphere: Integrating zones, processes, and semantics in the soil influenced by roots J. Exp. Bot. 2016 10.1093/jxb/erw108 26980751 7. Jones D.L. Nguyen C. Finlay R.D. Carbon flow in the rhizosphere: Carbon trading at the soil–root interface Plant Soil 2009 321 5 33 10.1007/s11104-009-9925-0 8. Lakshmanan V. Selvaraj G. Bais H.P. Functional soil microbiome: Belowground solutions to an aboveground problem Plant Physiol. 2014 166 689 700 10.1104/pp.114.245811 25059708 9. Nguyen C. Rhizodeposition of organic C by plant: Mechanisms and controls Sustainable Agriculture Lichtfouse E. Navarrete M. Debaeke P. Véronique S. Alberola C. Springer Houten, The Netherlands 2009 97 123 10. Lu Y. Zhou Y. Nakai S. Hosomi M. Zhang H. Kronzucker H.J. Shi W. Stimulation of nitrogen removal in the rhizosphere of aquatic duckweed by root exudate components Planta 2014 239 591 603 10.1007/s00425-013-1998-6 24271005 11. Pieterse C.M.J. de Jonge R. Berendsen R.L. The soil-borne supremacy Trends Plant Sci. 2016 21 171 173 10.1016/j.tplants.2016.01.018 26853594 12. Egamberdieva D. Abdiev A. Khaitov B. Synergistic interactions among root-associated bacteria, rhizobia and chickpea under stress conditions Plant-Environment Interaction: Responses and Approaches to Mitigate Stress Azooz M.M. Ahmad P. John Wiley & Sons, Ltd. Chichester, UK 2016 250 261 13. Barret M. Morrissey J.P. O’Gara F. Functional genomics analysis of plant growth-promoting rhizobacterial traits involved in rhizosphere competence Biol. Fertil. Soils 2011 47 729 743 10.1007/s00374-011-0605-x 14. Bais H.P. Weir T.L. Perry L.G. Gilroy S. Vivanco J.M. The role of root exudates in rhizosphere interactions with plants and other organisms Annu. Rev. Plant Biol. 2006 57 233 266 10.1146/annurev.arplant.57.032905.105159 16669762 15. Bertin C. Yang X.H. Weston L.A. The role of root exudates and allelochemicals in the rhizosphere Plant Soil 2003 256 67 83 10.1023/A:1026290508166 16. Donnelly P.K. Hegde R.S. Fletcher J.S. Growth of PCB-degrading bacteria on compounds from photosynthetic plants Chemosphere 1994 28 981 988 10.1016/0045-6535(94)90014-0 17. Chaudhry Q. Blom-Zandstra M. Gupta S. Joner E.J. Utilising the synergy between plants and rhizosphere microorganisms to enhance breakdown of organic pollutants in the environment Environ. Sci. Pollut. Res. Int. 2005 12 34 48 10.1065/espr2004.08.213 15768739 18. Singer A.C. Crowley D.E. Thompson I.P. Secondary plant metabolites in phytoremediation and biotransformation Trends Biotechnol. 2003 21 123 130 10.1016/S0167-7799(02)00041-0 12628369 19. Urbance J.W. Cole J. Saxman P. Tiedje J.M. BSD: The biodegradative strain database Nucleic Acids Res. 2003 31 152 155 10.1093/nar/gkg032 12519970 20. Muangchinda C. Chavanich S. Viyakarn V. Watanabe K. Imura S. Vangnai A.S. Pinyakong O. Abundance and diversity of functional genes involved in the degradation of aromatic hydrocarbons in Antarctic soils and sediments around Syowa Station Environ. Sci. Pollut. Res. 2015 22 4725 4735 10.1007/s11356-014-3721-y 25335763 21. Meckenstock R.U. Elsner M. Griebler C. Lueders T. Stumpp C. Aamand J. Agathos S.N. Albrechtsen H.-J. Bastiaens L. Bjerg P.L. Biodegradation: Updating the concepts of control for microbial cleanup in contaminated aquifers Environ. Sci. Technol. 2015 49 7073 7081 10.1021/acs.est.5b00715 26000605 22. Pausch J. Tian J. Riederer M. Kuzyakov Y. Estimation of rhizodeposition at field scale: Upscaling of a 14 C labeling study Plant Soil 2013 364 273 285 10.1007/s11104-012-1363-8 23. Marihal A.K. Jagadeesh K.S. Plant–microbe interaction: A potential tool for enhanced bioremediation Plant Microbe Symbiosis: Fundamentals and Advances Arora K.N. Springer India New Delhi, India 2013 395 410 24. Handelsman J. Rondon M.R. Brady S.F. Clardy J. Goodman R.M. Molecular biological access to the chemistry of unknown soil microbes: A new frontier for natural products Chem. Biol. 1998 5 R245 R249 10.1016/S1074-5521(98)90108-9 9818143 25. Metzker M.L. Sequencing technologies—The next generation Nat. Rev. Genet. 2010 11 31 46 10.1038/nrg2626 19997069 26. Kuzyakov Y. Raskatov A. Kaupenjohann M. Turnover and distribution of root exudates of Zea mays Plant Soil 2003 254 317 327 10.1023/A:1025515708093 27. Jones D.L. Hodge A. Kuzyakov Y. Plant and mycorrhizal regulation of rhizodeposition New Phytol. 2004 163 459 480 10.1111/j.1469-8137.2004.01130.x 28. Lu Y. Watanabe A. Kimura M. Input and distribution of photosynthesized carbon in a flooded rice soil Glob. Biogeochem. Cycles 2002 16 32-1 32-8 10.1029/2002GB001864 29. Keith H. Oades J.M. Martin J.K. Input of carbon to soil from wheat plants Soil Biol. Biochem. 1986 18 445 449 10.1016/0038-0717(86)90051-9 30. Kuzyakov Y. Kretzschmar A. Stahr K. Contribution of Lolium perenne rhizodeposition to carbon turnover of pasture soil Plant Soil 1999 213 127 136 10.1023/A:1004566027237 31. Zang H. Yang X. Feng X. Qian X. Hu Y. Ren C. Zeng Z. Rhizodeposition of nitrogen and carbon by mungbean (Vigna radiata L.) and its contribution to intercropped oats (Avena nuda L.) PLoS ONE 2015 10 1205 10.1371/journal.pone.0121132 25821975 32. Liljeroth E. Bååth E. Mathiasson I. Lundborg T. Root exudation and rhizoplane bacterial abundance of barley (Hordeum vulgare L.) in relation to nitrogen fertilization and root growth Plant Soil 1990 127 81 89 10.1007/BF00010839 33. Dijkstra F.A. Morgan J.A. Blumenthal D. Follett R.F. Water limitation and plant inter-specific competition reduce rhizosphere-induced C decomposition and plant N uptake Soil Biol. Biochem. 2010 42 1073 1082 10.1016/j.soilbio.2010.02.026 34. Sanaullah M. Chabbi A. Rumpel C. Kuzyakov Y. Carbon allocation in grassland communities under drought stress followed by 14 C pulse labeling Soil Biol. Biochem. 2012 55 132 139 10.1016/j.soilbio.2012.06.004 35. Neumann G. Römheld V. The release of root exudates as affected by the plant physiological status The Rhizosphere: Biochemistry and Organic Substances at the Soil-Plant Interface 2nd ed. Pinton R. Varanini Z. Nannipieri P. CRC Press Boca Raton, FL, USA 2007 23 72 36. Salt D.E. Smith R.D. Raskin I. Phytoremediation Annu. Rev. Plant Physiol. 1998 49 643 668 10.1146/annurev.arplant.49.1.643 15012249 37. Hütsch B.W. Augustin J. Merbach W. Plant rhizodeposition—An important source for carbon turnover in soils J. Plant Nutr. Soil Sci. 2002 165 397 407 10.1002/1522-2624(200208)165:4<397::AID-JPLN397>3.0.CO;2-C 38. Farrar J. Hawes M. Jones D. Lindow S. How roots control the flux of carbon to the rhizosphere Ecology 2003 84 827 837 10.1890/0012-9658(2003)084[0827:HRCTFO]2.0.CO;2 39. Martin B.C. George S.J. Price C.A. Ryan M.H. Tibbett M. The role of root exuded low molecular weight organic anions in facilitating petroleum hydrocarbon degradation: Current knowledge and future directions Sci. Total Environ. 2014 472 642 653 10.1016/j.scitotenv.2013.11.050 24317170 40. Jones D.L. Organic acids in the rhizosphere—A critical review Plant Soil 1998 205 25 44 10.1023/A:1004356007312 41. Demain A.L. Fang A. The natural functions of secondary metabolites History of Modern Biotechnology I Fiechter A. Springer Heidelberg, Germany 2000 Volumn 69 1 39 42. Dixon R.A. Natural products and plant disease resistance Nature 2001 411 843 847 10.1038/35081178 11459067 43. Pichersky E. Gang D.R. Genetics and biochemistry of secondary metabolites in plants: An evolutionary perspective Trends Plant Sci. 2000 5 439 445 10.1016/S1360-1385(00)01741-6 11044721 44. Walton N.J. Brown D.E. Chemicals from Plants: Perspectives on Plant Secondary Products Imperial College Press London, UK 1999 45. Crozier A. Jaganath I.B. Clifford M.N. Phenols, polyphenols and tannins: An overview Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet Crozier A. Clifford M.N. Ashihara H. Blackwell Publishing, Ltd. Oxford, UK 2007 1 24 46. Humphrey A.J. Beale M.H. Terpenes Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet Crozier A. Clifford M.N. Ashihara H. Blackwell Publishing, Ltd. Oxford, UK 2007 47 101 47. Christensen L.P. Brandt K. Acetylenes and psoralens Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet Crozier A. Clifford M.N. Ashihara H. Blackwell Publishing, Ltd. Oxford, UK 2007 137 173 48. Mithen R. Sulphur-containing compounds Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet Crozier A. Clifford M.N. Ashihara H. Blackwell Publishing, Ltd. Oxford, UK 2007 25 46 49. Zulak K.G. Liscombe D.K. Ashihara H. Facchini P.J. Alkaloids Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet Crozier A. Clifford M.N. Ashihara H. Blackwell Publishing, Ltd. Oxford, UK 2007 102 136 50. Iriti M. Faoro F. Ozone-induced changes in plant secondary metabolism Climate Change and Crops Singh S.N. Springer Heidelberg, Germany 2009 245 268 51. Noctor G. Mhamdi A. Chaouch S. Han Y.I. Neukermans J. Marquez-Garcia B. Queval G. Foyer C.H. Glutathione in plants: An integrated overview Plant Cell Environ. 2012 35 454 484 10.1111/j.1365-3040.2011.02400.x 21777251 52. Schlaeppi K. Bodenhausen N. Buchala A. Mauch F. Reymond P. The glutathione-deficient mutant pad2–1 accumulates lower amounts of glucosinolates and is more susceptible to the insect herbivore Spodoptera littoralis Plant J. 2008 55 774 786 10.1111/j.1365-313X.2008.03545.x 18466300 53. Brandel J. Humbert N. Elhabiri M. Schalk I.J. Mislin G.L.A. Albrecht-Gary A.-M. Pyochelin, a siderophore of Pseudomonas aeruginosa : Physicochemical characterization of the iron(iii), copper(ii) and zinc(ii) complexes Dalton Trans. 2012 41 2820 2834 10.1039/c1dt11804h 22261733 54. Chen Y.X. Lin Q. Luo Y.M. He Y.F. Zhen S.J. Yu Y.L. Tian G.M. Wong M.H. The role of citric acid on the phytoremediation of heavy metal contaminated soil Chemosphere 2003 50 807 811 10.1016/S0045-6535(02)00223-0 12688495 55. Yadav S.K. Heavy metals toxicity in plants: An overview on the role of glutathione and phytochelatins in heavy metal stress tolerance of plants S. Afr. J. Bot. 2010 76 167 179 10.1016/j.sajb.2009.10.007 56. Hayat S. Hayat Q. Alyemeni M.N. Wani A.S. Pichtel J. Ahmad A. Role of proline under changing environments: A review Plant Signal. Behav. 2012 7 1456 1466 10.4161/psb.21949 22951402 57. Erb M. Lu J. Soil abiotic factors influence interactions between belowground herbivores and plant roots J. Exp. Bot. 2013 64 1295 1303 10.1093/jxb/ert007 23505310 58. Fontaine S. Mariotti A. Abbadie L. The priming effect of organic matter: A question of microbial competition? Soil Biol. Biochem. 2003 35 837 843 10.1016/S0038-0717(03)00123-8 59. Eilers K.G. Lauber C.L. Knight R. Fierer N. Shifts in bacterial community structure associated with inputs of low molecular weight carbon compounds to soil Soil Biol. Biochem. 2010 42 896 903 10.1016/j.soilbio.2010.02.003 60. Slater H. Gouin T. Leigh M.B. Assessing the potential for rhizoremediation of PCB-contaminated soils in northern regions using native tree species Chemosphere 2011 84 199 206 10.1016/j.chemosphere.2011.04.058 21596420 61. Hernandez B.S. Koh S.C. Chial M. Focht D.D. Terpene-utilizing isolates and their relevance to enhanced biotransformation of polychlorinated biphenyls in soil Biodegradation 1997 8 153 158 10.1023/A:1008255218432 62. Singer A.C. Smith D. Jury W.A. Hathuc K. Crowley D.E. Impact of the plant rhizosphere and augmentation on remediation of polychlorinated biphenyl contaminated soil Environ. Toxicol. Chem. 2003 22 1998 2004 10.1897/02-471 12959523 63. Uhlik O. Musilova L. Ridl J. Hroudova M. Vlcek C. Koubek J. Holeckova M. Mackova M. Macek T. Plant secondary metabolite-induced shifts in bacterial community structure and degradative ability in contaminated soil Appl. Microbiol. Biotechnol. 2013 97 9245 9256 10.1007/s00253-012-4627-6 23250224 64. Gao M. Teplitski M. Robinson J.B. Bauer W.D. Production of substances by Medicago truncatula that affect bacterial quorum sensing Mol. Plant Microbe Interact. 2003 16 827 834 10.1094/MPMI.2003.16.9.827 12971606 65. Koh C.-L. Sam C.-K. Yin W.-F. Tan L.Y. Krishnan T. Chong Y.M. Chan K.-G. Plant-derived natural products as sources of anti-quorum sensing compounds Sensors 2013 13 6217 6228 10.3390/s130506217 23669710 66. Bernier S.P. Surette M.G. Concentration-dependent activity of antibiotics in natural environments Front. Microbiol. 2013 4 10.3389/fmicb.2013.00020 23422936 67. Badri D.V. Vivanco J.M. Regulation and function of root exudates Plant Cell Environ. 2009 32 666 681 10.1111/j.1365-3040.2009.01926.x 19143988 68. Xia J.H. Roberts J.K.M. Improved cytoplasmic pH regulation, increased lactate efflux, and reduced cytoplasmic lactate levels are biochemical traits expressed in root tips of whole maize seedlings acclimated to a low-oxygen environment Plant Physiol. 1994 105 651 657 12232232 69. Henry A. Doucette W. Norton J. Bugbee B. Changes in crested wheatgrass root exudation caused by flood, drought, and nutrient stress J. Environ. Qual. 2007 36 904 912 10.2134/jeq2006.0425sc 17485723 70. Juszczuk I. Wiktorowska A. Malusá E. Rychter A. Changes in the concentration of phenolic compounds and exudation induced by phosphate deficiency in bean plants (Phaseolus vulgaris L.) Plant Soil 2004 267 41 49 10.1007/s11104-005-2569-9 71. Carvalhais L.C. Dennis P.G. Fedoseyenko D. Hajirezaei M.-R. Borriss R. von Wirén N. Root exudation of sugars, amino acids, and organic acids by maize as affected by nitrogen, phosphorus, potassium, and iron deficiency J. Plant Nutr. Soil Sci. 2011 174 3 11 10.1002/jpln.201000085 72. Leigh M.B. Fletcher J.S. Fu X. Schmitz F.J. Root turnover: An important source of microbial substrates in rhizosphere remediation of recalcitrant contaminants Environ. Sci. Technol. 2002 36 1579 1583 10.1021/es015702i 11999069 73. Han E. Kautz T. Perkons U. Uteau D. Peth S. Huang N. Horn R. Köpke U. Root growth dynamics inside and outside of soil biopores as affected by crop sequence determined with the profile wall method Biol. Fertil. Soils 2015 51 847 856 10.1007/s00374-015-1032-1 74. D’Arcy-Lameta A. Study of soybean and lentil root exudates Plant Soil 1986 92 113 123 10.1007/BF02372272 75. Bulgarelli D. Schlaeppi K. Spaepen S. Themaat E.V.L.V. Schulze-Lefert P. Structure and functions of the bacterial microbiota of plants Annu. Rev. Plant Biol. 2013 64 807 838 10.1146/annurev-arplant-050312-120106 23373698 76. Schlaeppi K. Dombrowski N. Oter R.G. van Themaat E.V.L. Schulze-Lefert P. Quantitative divergence of the bacterial root microbiota in Arabidopsis thaliana relatives Proc. Natl. Acad. Sci. USA 2014 111 585 592 10.1073/pnas.1321597111 24379374 77. Bulgarelli D. Garrido-Oter R. Münch P.C. Weiman A. Dröge J. Pan Y. McHardy A.C. Schulze-Lefert P. Structure and function of the bacterial root microbiota in wild and domesticated barley Cell Host Microbe 2015 17 392 403 10.1016/j.chom.2015.01.011 25732064 78. Hartmann A. Schmid M. van Tuinen D. Berg G. Plant-driven selection of microbes Plant Soil 2009 321 235 257 10.1007/s11104-008-9814-y 79. DeAngelis K.M. Brodie E.L. DeSantis T.Z. Andersen G.L. Lindow S.E. Firestone M.K. Selective progressive response of soil microbial community to wild oat roots ISME J. 2008 3 168 178 10.1038/ismej.2008.103 19005498 80. Dzantor E.K. Woolston J.E. Momen B. PCB dissipation and microbial community analysis in rhizosphere soil under substrate amendment conditions Int. J. Phytoremediat. 2002 4 283 295 10.1080/15226510208500088 81. İnceoğlu Ö. van Overbeek L.S. Salles J.F. van Elsas J.D. Normal operating range of bacterial communities in soil used for potato cropping Appl. Environ. Microbiol. 2013 79 1160 1170 10.1128/AEM.02811-12 23220956 82. Fan B. Carvalhais L.C. Becker A. Fedoseyenko D. von Wirén N. Borriss R. Transcriptomic profiling of Bacillus amyloliquefaciens FZB42 in response to maize root exudates BMC Microbiol. 2012 12 1 13 10.1186/1471-2180-12-116 22221383 83. Zyśko A. Sanguin H. Hayes A. Wardleworth L. Zeef L.A.H. Sim A. Paterson E. Singh B.K. Kertesz M.A. Transcriptional response of Pseudomonas aeruginosa to a phosphate-deficient Lolium perenne rhizosphere Plant Soil 2012 359 25 44 10.1007/s11104-011-1060-z 84. Smith D.L. Subramanian S. Lamont J.R. Bywater-Ekegärd M. Signaling in the phytomicrobiome: Breadth and potential Front. Plant Sci. 2015 6 10.3389/fpls.2015.00709 26442023 85. Toussaint J.-P. Pham T. Barriault D. Sylvestre M. Plant exudates promote PCB degradation by a rhodococcal rhizobacteria Appl. Microbiol. Biotechnol. 2012 95 1589 1603 10.1007/s00253-011-3824-z 22202970 86. Blum U. Staman K. Flint L. Shafer S. Induction and/or selection of phenolic acid-utilizing bulk-soil and rhizosphere bacteria and their influence on phenolic acid phytotoxicity J. Chem. Ecol. 2000 26 2059 2078 10.1023/A:1005560214222 87. Sugiyama A. Yazaki K. Root exudates of legume plants and their involvement in interactions with soil microbes Secretions and Exudates in Biological Systems Vivanco M.J. Baluška F. Springer Berlin & Heidelberg, Germany 2012 27 48 88. De Weert S. Vermeiren H. Mulders I.H.M. Kuiper I. Hendrickx N. Bloemberg G.V. Vanderleyden J. de Mot R. Lugtenberg B.J.J. Flagella-driven chemotaxis towards exudate components is an important trait for tomato root colonization by Pseudomonas fluorescens Mol. Plant Microbe Interact. 2002 15 1173 1180 10.1094/MPMI.2002.15.11.1173 12423023 89. Yao J. Allen C. Chemotaxis is required for virulence and competitive fitness of the bacterial wilt pathogen Ralstonia solanacearum J. Bacteriol. 2006 188 3697 3708 10.1128/JB.188.10.3697-3708.2006 16672623 90. Meng F. Yao J. Allen C. A MotN mutant of Ralstonia solanacearum is hypermotile and has reduced virulence J. Bacteriol. 2011 193 2477 2486 10.1128/JB.01360-10 21421761 91. Neumann G. George T.S. Plassard C. Strategies and methods for studying the rhizosphere—The plant science toolbox Plant Soil 2009 321 431 456 10.1007/s11104-009-9953-9 92. Takenaka C. Miyahara M. Ohta T. Maximov T.C. Response of larch root development to annual changes of water conditions in eastern Siberia Polar Sci. 2016 10 160 166 10.1016/j.polar.2016.04.012 93. Davis J.P. Haines B. Coleman D. Hendrick R. Fine root dynamics along an elevational gradient in the southern Appalachian mountains, USA For. Ecol. Manag. 2004 187 19 33 10.1016/S0378-1127(03)00226-3 94. Oburger E. Schmidt H. New methods to unravel rhizosphere processes Trends Plant Sci. 2016 21 243 255 10.1016/j.tplants.2015.12.005 26776474 95. Downie H. Holden N. Otten W. Spiers A.J. Valentine T.A. Dupuy L.X. Transparent soil for imaging the rhizosphere PLoS ONE 2012 7 1205 10.1371/journal.pone.0044276 22984484 96. Chimungu J.G. Loades K.W. Lynch J.P. Root anatomical phenes predict root penetration ability and biomechanical properties in maize (Zea mays ) J. Exp. Bot. 2015 66 3151 3162 10.1093/jxb/erv121 25903914 97. Mooney S.J. Pridmore T.P. Helliwell J. Bennett M.J. Developing X-ray computed tomography to non-invasively image 3-D root systems architecture in soil Plant Soil 2012 352 1 22 10.1007/s11104-011-1039-9 98. Mairhofer S. Zappala S. Tracy S. Sturrock C. Bennett M.J. Mooney S.J. Recovering complete plant root system architectures from soil via X-ray µ-computed tomography Plant Methods 2013 9 10.1186/1746-4811-9-8 23514198 99. Stingaciu L. Schulz H. Pohlmeier A. Behnke S. Zilken H. Javaux M. Vereecken H. In situ root system architecture extraction from magnetic resonance imaging for water uptake modeling Vadose Zone J. 2013 12 10.2136/vzj2012.0019 100. Metzner R. Eggert A. van Dusschoten D. Pflugfelder D. Gerth S. Schurr U. Uhlmann N. Jahnke S. Direct comparison of MRI and X-ray CT technologies for 3D imaging of root systems in soil: Potential and challenges for root trait quantification Plant Methods 2015 11 1 11 10.1186/s13007-015-0060-z 25649124 101. Oswald S.E. Menon M. Carminati A. Vontobel P. Lehmann E. Schulin R. Quantitative imaging of infiltration, root growth, and root water uptake via neutron radiography Vadose Zone J. 2008 7 1035 1047 10.2136/vzj2007.0156 102. Downie H.F. Adu M.O. Schmidt S. Otten W. Dupuy L.X. White P.J. Valentine T.A. Challenges and opportunities for quantifying roots and rhizosphere interactions through imaging and image analysis Plant Cell Environ. 2015 38 1213 1232 10.1111/pce.12448 25211059 103. Van Dusschoten D. Metzner R. Kochs J. Postma J.A. Pflugfelder D. Buehler J. Schurr U. Jahnke S. Quantitative 3D analysis of plant roots growing in soil using magnetic resonance imaging Plant Physiol. 2016 10.1104/pp.15.01388 26729797 104. De Schepper V. Bühler J. Thorpe M. Roeb G. Huber G. van Dusschoten D. Jahnke S. Steppe K. 11 C-PET imaging reveals transport dynamics and sectorial plasticity of oak phloem after girdling Front. Plant Sci. 2013 4 200 10.3389/fpls.2013.00200 23785380 105. Larsen M. Santner J. Oburger E. Wenzel W.W. Glud R.N. O2 dynamics in the rhizosphere of young rice plants (Oryza sativa L.) as studied by planar optodes Plant Soil 2015 390 279 292 10.1007/s11104-015-2382-z 26166902 106. Koop-Jakobsen K. Wenzhöfer F. The dynamics of plant-mediated sediment oxygenation in Spartina anglica rhizospheres—A planar optode study Estuaries Coasts 2015 38 951 963 10.1007/s12237-014-9861-y 107. Blossfeld S. Schreiber C.M. Liebsch G. Kuhn A.J. Hinsinger P. Quantitative imaging of rhizosphere pH and CO2 dynamics with planar optodes Ann. Bot. 2013 112 267 276 10.1093/aob/mct047 23532048 108. Faget M. Blossfeld S. von Gillhaussen P. Schurr U. Temperton V.M. Disentangling who is who during rhizosphere acidification in root interactions: Combining fluorescence with optode techniques Front. Plant Sci. 2013 4 392 10.3389/fpls.2013.00392 24137168 109. Marschner P. Crowley D. Rengel Z. Rhizosphere interactions between microorganisms and plants govern iron and phosphorus acquisition along the root axis—Model and research methods Soil Biol. Biochem. 2011 43 883 894 10.1016/j.soilbio.2011.01.005 110. Jusoh W. Wong L.S. Exploring the potential of whole cell biosensor: A review in environmental applications Int. J. Chem. Environ. Biol. Sci. 2014 2 52 56 111. Porteous F. Killham K. Meharg A. Use of a lux -marked rhizobacterium as a biosensor to assess changes in rhizosphere C flow due to pollutant stress Chemosphere 2000 41 1549 1554 10.1016/S0045-6535(00)00072-2 11057680 112. Puglisi E. Fragoulis G. Del Re A.A.M. Spaccini R. Piccolo A. Gigliotti G. Said-Pullicino D. Trevisan M. Carbon deposition in soil rhizosphere following amendments with compost and its soluble fractions, as evaluated by combined soil–plant rhizobox and reporter gene systems Chemosphere 2008 73 1292 1299 10.1016/j.chemosphere.2008.07.008 18768204 113. Tobar R.M. Azcón-Aguilar C. Sanjuán J. Barea J.M. Impact of a genetically modified Rhizobium strain with improved nodulation competitiveness on the early stages of arbuscular mycorrhiza formation Appl. Soil Ecol. 1996 4 15 21 10.1016/0929-1393(96)00104-7 114. DeAngelis K.M. Firestone M.K. Lindow S.E. Sensitive whole-cell biosensor suitable for detecting a variety of N -acyl homoserine lactones in intact rhizosphere microbial communities Appl. Environ. Microbiol. 2007 73 3724 3727 10.1128/AEM.02187-06 17400771 115. Liu X. Germaine K.J. Ryan D. Dowling D.N. Whole-cell fluorescent biosensors for bioavailability and biodegradation of polychlorinated biphenyls Sensors 2010 10 1377 1398 10.3390/s100201377 22205873 116. DeAngelis K.M. Ji P. Firestone M.K. Lindow S.E. Two novel bacterial biosensors for detection of nitrate availability in the rhizosphere Appl. Environ. Microbiol. 2005 71 8537 8547 10.1128/AEM.71.12.8537-8547.2005 16332845 117. Jensen L.E. Kragelund L. Nybroe O. Expression of a nitrogen regulated lux gene fusion in Pseudomonas fluorescens DF57 studied in pure culture and in soil FEMS Microbiol. Ecol. 1998 25 23 32 10.1111/j.1574-6941.1998.tb00457.x 118. Kragelund L. Christoffersen B. Nybroe O. de Bruijn F.J. Isolation of lux reporter gene fusions in Pseudomonas fluorescens DF57 inducible by nitrogen or phosphorus starvation FEMS Microbiol. Ecol. 1995 17 95 106 10.1111/j.1574-6941.1995.tb00134.x 119. Kuppardt A. Vetterlein D. Harms H. Chatzinotas A. Visualisation of gradients in arsenic concentrations around individual roots of Zea mays L. Using agar-immobilized bioreporter bacteria Plant Soil 2010 329 295 306 10.1007/s11104-009-0154-3 120. Jaeger Iii C.H. Lindow S.E. Miller W. Clark E. Firestone M.K. Mapping of sugar and amino acid availability in soil around roots with bacterial sensors of sucrose and tryptophan Appl. Environ. Microbiol. 1999 65 2685 2690 10347061 121. Loper J.E. Henkels M.D. Availability of iron to Pseudomonas fluorescens in rhizosphere and bulk soil evaluated with an ice nucleation reporter gene Appl. Environ. Microbiol. 1997 63 99 105 8979343 122. Amann R. Fuchs B.M. Single-cell identification in microbial communities by improved fluorescence in situ hybridization techniques Nat. Rev. Microbiol. 2008 6 339 348 10.1038/nrmicro1888 18414500 123. Wagner M. Haider S. New trends in fluorescence in situ hybridization for identification and functional analyses of microbes Curr. Opin. Biotechnol. 2012 23 96 102 10.1016/j.copbio.2011.10.010 22079351 124. Cao Y. Zhang Z. Ling N. Yuan Y. Zheng X. Shen B. Shen Q. Bacillus subtilis SQR 9 can control Fusarium wilt in cucumber by colonizing plant roots Biol. Fertil. Soils 2011 47 495 506 10.1007/s00374-011-0556-2 125. Khabbaz S.E. Zhang L. Cáceres L.A. Sumarah M. Wang A. Abbasi P.A. Characterisation of antagonistic Bacillus and Pseudomonas strains for biocontrol potential and suppression of damping-off and root rot diseases Ann. Appl. Biol. 2015 166 456 471 10.1111/aab.12196 126. Buddrus-Schiemann K. Schmid M. Schreiner K. Welzl G. Hartmann A. Root colonization by Pseudomonas sp. DSMZ 13134 and impact on the indigenous rhizosphere bacterial community of barley Microb. Ecol. 2010 60 381 393 10.1007/s00248-010-9720-8 20644925 127. Haichar F.E.Z. Heulin T. Guyonnet J.P. Achouak W. Stable isotope probing of carbon flow in the plant holobiont Curr. Opin. Biotechnol. 2016 41 9 13 10.1016/j.copbio.2016.02.023 27019410 128. Uhlik O. Leewis M.C. Strejcek M. Musilova L. Mackova M. Leigh M.B. Macek T. Stable isotope probing in the metagenomics era: A bridge towards improved bioremediation Biotechnol. Adv. 2013 31 154 165 10.1016/j.biotechadv.2012.09.003 23022353 129. Radajewski S. McDonald I.R. Murrell J.C. Stable-isotope probing of nucleic acids: A window to the function of uncultured microorganisms Curr. Opin. Biotechnol. 2003 14 296 302 10.1016/S0958-1669(03)00064-8 12849783 130. Nikolausz M. Palatinszky M.A. Rusznyak A. Richnow H.H. Kappelmeyer U. Kastner M. Novel approach using substrate-mediated radiolabelling of RNA to link metabolic function with the structure of microbial communities FEMS Microbiol. Lett. 2007 274 154 161 10.1111/j.1574-6968.2007.00821.x 17590229 131. Aelion C.M. Norman R.S. Isotopic labeling in environmental and biodegradation studies Environmental Isotopes in Biodegradation and Bioremediation Aelion C.M. Höhener P. Hunkeler D. Aravena R. CRC Press Boca Raton, FL, USA 2009 327 349 132. Xu J. Microbial ecology in the age of genomics and metagenomics: Concepts, tools, and recent advances Mol. Ecol. 2006 15 1713 1731 10.1111/j.1365-294X.2006.02882.x 16689892 133. Turner T.R. Ramakrishnan K. Walshaw J. Heavens D. Alston M. Swarbreck D. Osbourn A. Grant A. Poole P.S. Comparative metatranscriptomics reveals kingdom level changes in the rhizosphere microbiome of plants ISME J. 2013 7 2248 2258 10.1038/ismej.2013.119 23864127 134. Handelsman J. Metagenomics: Application of genomics to uncultured microorganisms Microbiol. Mol. Biol. Rev. 2004 68 669 685 10.1128/MMBR.68.4.669-685.2004 15590779 135. Gilbert J.A. Hughes M. Gene expression profiling: Metatranscriptomics High-Throughput Next Generation Sequencing: Methods and Applications Kwon M.Y. Ricke C.S. Humana Press Totowa, NJ, USA 2011 195 205 136. Suenaga H. Targeted metagenomics: A high-resolution metagenomics approach for specific gene clusters in complex microbial communities Environ. Microbiol. 2012 14 13 22 10.1111/j.1462-2920.2011.02438.x 21366818 137. Hettich R.L. Pan C. Chourey K. Giannone R.J. Metaproteomics : Harnessing the power of high performance mass spectrometry to identify the suite of proteins that control metabolic activities in microbial communities Anal. Chem. 2013 85 4203 4214 10.1021/ac303053e 23469896 138. Van Dam N.M. Bouwmeester H.J. Metabolomics in the rhizosphere: Tapping into belowground chemical communication Trends Plant Sci. 2016 21 256 265 10.1016/j.tplants.2016.01.008 26832948 139. Mairhofer S. Zappala S. Tracy S.R. Sturrock C. Bennett M. Mooney S.J. Pridmore T. Rootrak: Automated recovery of three-dimensional plant root architecture in soil from X-ray micro-computed tomography images using visual tracking Plant Physiol. 2012 158 10.1104/pp.111.186221 22190339 140. Metzner R. Dusschoten D. Bühler J. Schurr U. Jahnke S. Belowground plant development measured with magnetic resonance imaging (MRI): Exploiting the potential for non-invasive trait quantification using sugar beet as a proxy Front. Plant. Sci. 2014 5 10.3389/fpls.2014.00469 25278947 141. Beer S. Streun M. Hombach T. Buehler J. Jahnke S. Khodaverdi M. Larue H. Minwuyelet S. Parl C. Roeb G. Design and initial performance of PlanTIS: A high-resolution positron emission tomograph for plants Phys. Med. Biol. 2010 55 635 10.1088/0031-9155/55/3/006 20071758 142. Yuan H. Zhu Z. Liu S. Ge T. Jing H. Li B. Liu Q. Lynn T.M. Wu J. Kuzyakov Y. Microbial utilization of rice root exudates: 13 C labeling and PLFA composition Biol. Fertil. Soils 2016 52 615 627 10.1007/s00374-016-1101-0 143. Wang J. Chapman S.J. Yao H. Incorporation of 13 C-labelled rice rhizodeposition into soil microbial communities under different fertilizer applications Appl. Soil Ecol. 2016 101 11 19 10.1016/j.apsoil.2016.01.010 144. Gschwendtner S. Engel M. Lueders T. Buegger F. Schloter M. Nitrogen fertilization affects bacteria utilizing plant-derived carbon in the rhizosphere of beech seedlings Plant Soil 2016 1 13 10.1007/s11104-016-2888-z 145. De Santi C. Altermark B. Pierechod M.M. Ambrosino L. de Pascale D. Willassen N.-P. Characterization of a cold-active and salt tolerant esterase identified by functional screening of arctic metagenomic libraries BMC Biochem. 2016 17 1 13 10.1186/s12858-016-0057-x 26782782 146. Unno Y. Shinano T. Metagenomic analysis of the rhizosphere soil microbiome with respect to phytic acid utilization Microbes Environ. 2013 28 120 127 10.1264/jsme2.ME12181 23257911 147. Chapelle E. Mendes R. Bakker P.A.H.M. Raaijmakers J.M. Fungal invasion of the rhizosphere microbiome ISME J. 2016 10 265 268 10.1038/ismej.2015.82 26023875 148. Chaparro J.M. Badri D.V. Vivanco J.M. Rhizosphere microbiome assemblage is affected by plant development ISME J. 2014 8 790 803 10.1038/ismej.2013.196 24196324 149. Yergeau E. Sanschagrin S. Maynard C. St-Arnaud M. Greer C.W. Microbial expression profiles in the rhizosphere of willows depend on soil contamination ISME J. 2014 8 344 358 10.1038/ismej.2013.163 24067257 150. Newman M.M. Lorenz N. Hoilett N. Lee N.R. Dick R.P. Liles M.R. Ramsier C. Kloepper J.W. Changes in rhizosphere bacterial gene expression following glyphosate treatment Sci. Total Environ. 2016 553 32 41 10.1016/j.scitotenv.2016.02.078 26901800 151. Hettich R.L. Sharma R. Chourey K. Giannone R.J. Microbial metaproteomics: Identifying the repertoire of proteins that microorganisms use to compete and cooperate in complex environmental communities Curr. Opin. Microbiol. 2012 15 373 380 10.1016/j.mib.2012.04.008 22632760 152. Becher D. Bernhardt J. Fuchs S. Riedel K. Metaproteomics to unravel major microbial players in leaf litter and soil environments: Challenges and perspectives Proteomics 2013 13 2895 2909 10.1002/pmic.201300095 23894095 153. Lin W. Wu L. Lin S. Zhang A. Zhou M. Lin R. Wang H. Chen J. Zhang Z. Lin R. Metaproteomic analysis of ratoon sugarcane rhizospheric soil BMC Microbiol. 2013 13 1 13 10.1186/1471-2180-13-135 23286760 154. Brooks J.M. Benson D.R. Comparative metabolomics of root nodules infected with Frankia sp. strains and uninfected roots from Alnus glutinosa and Casuarina cunninghamiana reflects physiological integration Symbiosis 2016 1 10 10.1007/s13199-016-0379-x 155. Ghaste M. Mistrik R. Shulaev V. Applications of Fourier transform ion cyclotron resonance (FT-ICR) and orbitrap based high resolution mass spectrometry in metabolomics and lipidomics Int. J. Mol. Sci. 2016 17 816 10.3390/ijms17060816 27231903 156. O’Leary B.M. Neale H.C. Geilfus C.-M. Jackson R.W. Arnold D.L. Preston G.M. Early changes in apoplast composition associated with defence and disease in interactions between Phaseolus vulgaris and the halo blight pathogen Pseudomonas syringae pv. Phaseolicola Plant Cell Environ. 2016 10.1111/pce.12770 27239727 157. Oger M.P. Mansouri H. Nesme X. Dessaux Y. Engineering root exudation of lotus toward the production of two novel carbon compounds leads to the selection of distinct microbial populations in the rhizosphere Microb. Ecol. 2004 47 96 1039 10.1007/s00248-003-2012-9 15259274 158. Costura R.K. Alvarez P.J.J. Expression and longevity of toluene dioxygenase in Pseudomonas putida F1 induced at different dissolved oxygen concentrations Water Res. 2000 34 3014 3018 10.1016/S0043-1354(00)00035-X 159. Szokol J. Rucka L. Simcikova M. Halada P. Nesvera J. Patek M. Induction and carbon catabolite repression of phenol degradation genes in Rhodococcus erythropolis and Rhodococcus jostii Appl. Microbiol. Biotechnol. 2014 98 8267 8279 10.1007/s00253-014-5881-6 24938209 160. Watanabe T. Inoue R. Kimura N. Furukawa K. Versatile transcription of biphenyl catabolic bph operon in Pseudomonas pseudoalcaligenes KF707 J. Biol. Chem. 2000 275 31016 31023 10.1074/jbc.M003023200 10900199 161. Kim D. Kim S.W. Choi K.Y. Lee J.S. Kim E. Molecular cloning and functional characterization of the genes encoding benzoate and p -hydroxybenzoate degradation by the halophilic Chromohalobacter sp. strain HS-2 FEMS Microbiol. Lett. 2008 280 235 241 10.1111/j.1574-6968.2008.01067.x 18248426 162. Frascari D. Zanaroli G. Danko A.S. In situ aerobic cometabolism of chlorinated solvents: A review J. Hazard. Mater. 2015 283 382 399 10.1016/j.jhazmat.2014.09.041 25306537 163. Rentz J.A. Alvarez P.J.J. Schnoor J.L. Benzo[a ]pyrene co-metabolism in the presence of plant root extracts and exudates: Implications for phytoremediation Environ. Pollut. 2005 136 477 484 10.1016/j.envpol.2004.12.034 15862401 164. Zorádová-Murínová S. Dudášová H. Lukáčová L. Čertík M. Šilharová K. Vrana B. Dercová K. Adaptation mechanisms of bacteria during the degradation of polychlorinated biphenyls in the presence of natural and synthetic terpenes as potential degradation inducers Appl. Microbiol. Biotechnol. 2012 94 1375 1385 10.1007/s00253-011-3763-8 22159613 165. Leigh M.B. Methods for rhizoremediation research: Approaches to experimental design and microbial analysis Phytoremediation and Rhizoremediation. Theoretical Background 9th ed. Macková M. Dowling D. Macek T. Springer Dordrecht, The Netherlands 2006 33 55 166. Singer A.C. Thompson I.P. Bailey M.J. The tritrophic trinity: A source of pollutant-degrading enzymes and its implications for phytoremediation Curr. Opin. Microbiol. 2004 7 239 244 10.1016/j.mib.2004.04.007 15196490 167. Singer A.C. Gilbert E.S. Luepromchai E. Crowley D.E. Bioremediation of polychlorinated biphenyl-contaminated soil using carvone and surfactant-grown bacteria Appl. Microbiol. Biotechnol. 2000 54 838 843 10.1007/s002530000472 11152078 168. Tandlich R. Brezna B. Dercova K. The effect of terpenes on the biodegradation of polychlorinated biphenyls by Pseudomonas stutzeri Chemosphere 2001 44 1547 1555 10.1016/S0045-6535(00)00523-3 11545520 169. Ionescu M. Beranova K. Dudkova V. Kochankova L. Demnerova K. Macek T. Mackova M. Isolation and characterization of different plant associated bacteria and their potential to degrade polychlorinated biphenyls Int. Biodeterior. Biodegrad. 2009 63 667 672 10.1016/j.ibiod.2009.03.009 170. Furukawa K. Miyazaki T. Cloning of a gene cluster encoding biphenyl and chlorobiphenyl degradation in Pseudomonas pseudoalcaligenes J. Bacteriol. 1986 166 392 398 3009395 171. Furukawa K. Suenaga H. Goto M. Biphenyl dioxygenases: Functional versatilities and directed evolution J. Bacteriol. 2004 186 5189 5196 10.1128/JB.186.16.5189-5196.2004 15292119 172. Macek T. Mackova M. Kas J. Exploitation of plants for the removal of organics in environmental remediation Biotechnol. Adv. 2000 18 23 34 10.1016/S0734-9750(99)00034-8 14538117 173. Maeda H. Dudareva N. The shikimate pathway and aromatic amino acid biosynthesis in plants Annu. Rev. Plant Biol. 2012 63 73 105 10.1146/annurev-arplant-042811-105439 22554242 174. Dewick P.M. The shikimate pathway: Aromatic amino acids and phenylpropanoids Medicinal Natural Products 3rd ed. Dewick P.M. John Wiley & Sons, Ltd. Chichester, UK 2009 137 186 175. Duke J.A. Handbook of Phytochemical Constituent Grass, Herbs and Other Economic Plants 1st ed. CRC Press Boca Raton, FL, USA 1992 176. Winkel-Shirley B. Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology Plant Physiol. 2001 126 485 493 10.1104/pp.126.2.485 11402179 177. Dashtban M. Schraft H. Syed T.A. Qin W. Fungal biodegradation and enzymatic modification of lignin Int. J. Biochem. Mol. Biol. 2010 1 36 50 21968746 178. Ruiz-Dueñas F.J. Martínez Á.T. Microbial degradation of lignin: How a bulky recalcitrant polymer is efficiently recycled in nature and how we can take advantage of this Microb. Biotechnol. 2009 2 164 177 10.1111/j.1751-7915.2008.00078.x 21261911 179. Jeffries T. Biodegradation of lignin and hemicelluloses Biochemistry of Microbial Degradation Ratledge C. Springer Houten, The Netherlands 1994 233 277 180. Sainsbury P.D. Mineyeva Y. Mycroft Z. Bugg T.D.H. Chemical intervention in bacterial lignin degradation pathways: Development of selective inhibitors for intradiol and extradiol catechol dioxygenases Bioorg. Chem. 2015 60 102 109 10.1016/j.bioorg.2015.05.002 25984987 181. Brown M.E. Chang M.C.Y. Exploring bacterial lignin degradation Curr. Opin. Chem. Biol. 2014 19 1 7 10.1016/j.cbpa.2013.11.015 24780273 182. Kameshwar A.k.S. Qin W. Recent developments in using advanced sequencing technologies for the genomic studies of lignin and cellulose degrading microorganisms Int. J. Biol. Sci. 2016 12 156 171 10.7150/ijbs.13537 26884714 183. Stewart P. Kersten P. Vanden Wymelenberg A. Gaskell J. Cullen D. Lignin peroxidase gene family of Phanerochaete chrysosporium : Complex regulation by carbon and nitrogen limitation and identification of a second dimorphic chromosome J. Bacteriol. 1992 174 5036 5042 1629160 184. Li D. Alic M. Brown J.A. Gold M.H. Regulation of manganese peroxidase gene transcription by hydrogen peroxide, chemical stress, and molecular oxygen Appl. Environ. Microbiol. 1995 61 341 345 7887613 185. Huang Z. Liers C. Ullrich R. Hofrichter M. Urynowicz M.A. Depolymerization and solubilization of chemically pretreated powder river basin subbituminous coal by manganese peroxidase (MnP) from Bjerkandera adusta Fuel 2013 112 295 301 10.1016/j.fuel.2013.04.081 186. Hirosue S. Tazaki M. Hiratsuka N. Yanai S. Kabumoto H. Shinkyo R. Arisawa A. Sakaki T. Tsunekawa H. Johdo O. Insight into functional diversity of cytochrome P450 in the white-rot basidiomycete Phanerochaete chrysosporium : Involvement of versatile monooxygenase Biochem. Biophys. Res. Commun. 2011 407 118 123 10.1016/j.bbrc.2011.02.121 21362401 187. Cullen D. Recent advances on the molecular genetics of ligninolytic fungi J. Biotechnol. 1997 53 273 289 10.1016/S0168-1656(97)01684-2 9177046 188. Morel M. Meux E. Mathieu Y. Thuillier A. Chibani K. Harvengt L. Jacquot J.-P. Gelhaye E. Xenomic networks variability and adaptation traits in wood decaying fungi Microb. Biotechnol. 2013 6 248 263 10.1111/1751-7915.12015 23279857 189. Cvancarova M. Kresinova Z. Filipova A. Covino S. Cajthaml T. Biodegradation of PCBs by ligninolytic fungi and characterization of the degradation products Chemosphere 2012 88 1317 1323 10.1016/j.chemosphere.2012.03.107 22546633 190. Yadav J.S. Quensen J.F. Tiedje J.M. Reddy C.A. Degradation of polychlorinated biphenyl mixtures (Aroclors 1242, 1254, and 1260) by the white rot fungus Phanerochaete chrysosporium as evidenced by congener-specific analysis Appl. Environ. Microbiol. 1995 61 2560 2565 7618867 191. Krcmar P. Ulrich R. Degradation of polychlorinated biphenyl mixtures by the lignin-degrading fungus Phanerochœte chrysosporium Folia Microbiol. 1998 43 79 84 10.1007/BF02815549 9569633 192. Cajthaml T. Erbanova P. Sasek V. Moeder M. Breakdown products on metabolic pathway of degradation of benz[a ]anthracene by a ligninolytic fungus Chemosphere 2006 64 560 564 10.1016/j.chemosphere.2005.11.034 16403417 193. Cajthaml T. Erbanova P. Kollmann A. Novotný C. Sasek V. Mougin C. Degradation of PAHs by ligninolytic enzymes of Irpex lacteus Folia Microbiol. 2008 53 289 294 10.1007/s12223-008-0045-7 18759111 194. Wang C. Sun H. Li J. Li Y. Zhang Q. Enzyme activities during degradation of polycyclic aromatic hydrocarbons by white rot fungus Phanerochaete chrysosporium in soils Chemosphere 2009 77 733 738 10.1016/j.chemosphere.2009.08.028 19751947 195. Byss M. Elhottova D. Triska J. Baldrian P. Fungal bioremediation of the creosote-contaminated soil: Influence of Pleurotus ostreatus and Irpex lacteus on polycyclic aromatic hydrocarbons removal and soil microbial community composition in the laboratory-scale study Chemosphere 2008 73 1518 1523 10.1016/j.chemosphere.2008.07.030 18782639 196. Baborova P. Moder M. Baldrian P. Cajthamlova K. Cajthaml T. Purification of a new manganese peroxidase of the white-rot fungus Irpex lacteus , and degradation of polycyclic aromatic hydrocarbons by the enzyme Res. Microbiol. 2006 157 248 253 10.1016/j.resmic.2005.09.001 16256312 197. Covino S. Svobodova K. Kresinova Z. Petruccioli M. Federici F. D’Annibale A. Cvancarova M. Cajthaml T. In vivo and in vitro polycyclic aromatic hydrocarbons degradation by Lentinus (Panus ) tigrinus CBS 577.79 Bioresour. Technol. 2010 101 3004 3012 10.1016/j.biortech.2009.12.020 20056409 198. Sato A. Watanabe T. Watanabe Y. Harazono K. Fukatsu T. Screening for basidiomycetous fungi capable of degrading 2,7-dichlorodibenzo-p -dioxin FEMS Microbiol. Lett. 2002 213 213 217 10.1111/j.1574-6968.2002.tb11308.x 12167540 199. Donnelly K.C. Chen J.C. Huebner H.J. Brown K.W. Autenrieth R.L. Bonner J.S. Utility of four strains of white-rot fungi for the detoxification of 2,4,6-trinitrotoluene in liquid culture Environ. Toxicol. Chem. 1997 16 1105 1110 10.1002/etc.5620160603 200. Rigas F. Dritsa V. Marchant R. Papadopoulou K. Avramides E.J. Hatzianestis I. Biodegradation of lindane by Pleurotus ostreatus via central composite design Environ. Int. 2005 31 191 196 10.1016/j.envint.2004.09.024 15661282 201. Rüttimann-Johnson C. Lamar R.T. Binding of pentachlorophenol to humic substances in soil by the action of white rot fungi Soil Biol. Biochem. 1997 29 1143 1148 10.1016/S0038-0717(96)00296-9 202. Uhlik O. Jecna K. Mackova M. Vlcek C. Hroudova M. Demnerova K. Paces V. Macek T. Biphenyl-metabolizing bacteria in the rhizosphere of horseradish and bulk soil contaminated by polychlorinated biphenyls as revealed by stable isotope probing Appl. Environ. Microbiol. 2009 75 6471 6477 10.1128/AEM.00466-09 19700551 203. Leigh M.B. Prouzova P. Mackova M. Macek T. Nagle D.P. Fletcher J.S. Polychlorinated biphenyl (PCB)-degrading bacteria associated with trees in a PCB-contaminated site Appl. Environ. Microbiol. 2006 72 2331 2342 10.1128/AEM.72.4.2331-2342.2006 16597927 204. Ridl J. Kolar M. Strejcek M. Strnad H. Stursa P. Paces J. Macek T. Uhlik O. Plants rather than mineral fertilization shape microbial community structure and functional potential in legacy contaminated soil Front. Microbiol. 2016 7 10.3389/fmicb.2016.00995 27446035 205. Reynolds C. Koenen B. Carnahan J. Walworth J. Bhunia P. Rhizosphere and nutrient effects on remediating subarctic soils Situ On-Site Bioremediat. 1997 4 297 302 206. Leewis M.-C.C.E. Uhlik O. Fraraccio S. McFarlin K. Kottara A. Glover C. Macek T. Leigh M.B. Differential impacts of willow and mineral fertilizer on bacterial communities and biodegradation in diesel fuel oil-contaminated soil Front. Microbiol. 2016 7 10.3389/fmicb.2016.00837 27313574 207. Gilbert E.S. Crowley D.E. Plant compounds that induce polychlorinated biphenyl biodegradation by Arthrobacter sp. strain B1B Appl. Environ. Microbiol. 1997 63 1933 1938 9143124 208. Gilbert E.S. Crowley D.E. Repeated application of carvone-induced bacteria to enhance biodegradation of polychlorinated biphenyls in soil Appl. Microbiol. Biotechnol. 1998 50 489 494 10.1007/s002530051325 9830100 209. Yi H. Crowley D.E. Biostimulation of PAH degradation with plants containing high concentrations of linoleic acid Environ. Sci. Technol. 2007 41 4382 4388 10.1021/es062397y 17626440 210. McLoughlin E. Rhodes A.H. Owen S.M. Semple K.T. Biogenic volatile organic compounds as a potential stimulator for organic contaminant degradation by soil microorganisms Environ. Pollut. 2009 157 86 94 10.1016/j.envpol.2008.07.029 18819735 211. Scheublin T.R. Deusch S. Moreno-Forero S.K. Müller J.A. van der Meer J.R. Leveau J.H.J. Transcriptional profiling of gram-positive Arthrobacter in the phyllosphere: Induction of pollutant degradation genes by natural plant phenolic compounds Environ. Microbiol. 2014 16 2212 2225 10.1111/1462-2920.12375 24373130 212. Suttinun O. Lederman P.B. Luepromchai E. Application of terpene-induced cell for enhancing biodegradation of TCE contaminated soil Songklanakarin J. Sci. Technol. 2004 26 131 142 213. Dabrock B. Riedel J. Bertram J. Gottschalk G. Isopropylbenzene (cumene)—A new substrate for the isolation of trichloroethene-degrading bacteria Arch. Microbiol. 1992 158 9 13 10.1007/BF00249058 1444717 214. Kirby R. Actinomycetes and lignin degradation Advances in Applied Microbiology Laskin A.I. Bennett J.W. Gadd G.M. Sariaslani S. Academic Press San Diego, CA, USA 2005 Volumn 58 125 168 215. Ahmad M. Roberts J.N. Hardiman E.M. Singh R. Eltis L.D. Bugg T.D.H. Identification of DypB from Rhodococcus jostii RHA1 as a lignin peroxidase Biochemistry 2011 50 5096 5107 10.1021/bi101892z 21534568 216. McLeod M.P. Warren R.L. Hsiao W.W.L. Araki N. Myhre M. Fernandes C. Miyazawa D. Wong W. Lillquist A.L. Wang D. The complete genome of Rhodococcus sp. RHA1 provides insights into a catabolic powerhouse Proc. Natl. Acad. Sci. USA 2006 103 15582 15587 10.1073/pnas.0607048103 17030794 217. Tehrani R. Van Aken B. Hydroxylated polychlorinated biphenyls in the environment: Sources, fate, and toxicities Environ. Sci. Pollut. Res. 2014 21 6334 6345 10.1007/s11356-013-1742-6 23636595 218. Rezek J. Macek T. Mackova M. Triska J. Ruzickova K. Hydroxy-PCBs, methoxy-PCBs and hydroxy-methoxy-PCBs: Metabolites of polychlorinated biphenyls formed in vitro by tobacco cells Environ. Sci. Technol. 2008 42 5746 5751 10.1021/es800445h 18754503 219. Francova K. Mackova M. Macek T. Sylvestre M. Ability of bacterial biphenyl dioxygenases from Burkholderia sp. LB400 and Comamonas testosteroni B-356 to catalyse oxygenation of ortho -hydroxychlorobiphenyls formed from PCBs by plants Environ. Pollut. 2004 127 41 48 10.1016/S0269-7491(03)00257-4 14553993 220. Pham T.T.M. Sondossi M. Sylvestre M. Metabolism of doubly para -substituted hydroxychlorobiphenyls by bacterial biphenyl dioxygenases Appl. Environ. Microbiol. 2015 81 4860 4872 10.1128/AEM.00786-15 25956777 221. Sondossi M. Barriault D. Sylvestre M. Metabolism of 2,2′- and 3,3′-dihydroxybiphenyl by the biphenyl catabolic pathway of Comamonas testosteroni B-356 Appl. Environ. Microbiol. 2004 70 174 181 10.1128/AEM.70.1.174-181.2004 14711640 222. Tehrani R. Lyv M.M. Kaveh R. Schnoor J.L. van Aken B. Biodegradation of mono-hydroxylated PCBs by Burkholderia xenovorans Biotechnol. Lett. 2012 34 2247 2252 10.1007/s10529-012-1037-x 22918793 223. Pham T.T.M. Tu Y. Sylvestre M. Remarkable ability of Pandoraea pnomenusa B356 biphenyl dioxygenase to metabolize simple flavonoids Appl. Environ. Microbiol. 2012 78 3560 3570 10.1128/AEM.00225-12 22427498 224. Pham T.T. Pino Rodriguez N.J. Hijri M. Sylvestre M. Optimizing polychlorinated biphenyl degradation by flavonoid-induced cells of the rhizobacterium Rhodococcus erythropolis U23A PLoS ONE 2015 10 1205 10.1371/journal.pone.0126033 25970559 225. Chen S.H. Aitken M.D. Salicylate stimulates the degradation of high-molecular weight polycyclic aromatic hydrocarbons by Pseudomonas saccharophila P15 Environ. Sci. Technol. 1998 33 435 439 10.1021/es9805730 226. Kang J.H. Kondo F. Bisphenol a degradation by bacteria isolated from river water Arch. Environ. Contam. Toxicol. 2002 43 265 269 10.1007/s00244-002-1209-0 12202920 227. Kamath R. Schnoor J.L. Alvarez P.J.J. Effect of root-derived substrates on the expression of nah -lux genes in Pseudomonas fluorescens HK44: Implications for PAH biodegradation in the rhizosphere Environ. Sci. Technol. 2004 38 1740 1745 10.1021/es0306258 15074683 228. Wojcieszyńska D. Guzik U. Greń I. Perkosz M. Hupert-Kocurek K. Induction of aromatic ring: Cleavage dioxygenases in Stenotrophomonas maltophilia strain KB2 in cometabolic systems World J. Microbiol. Biotechnol. 2011 27 805 811 10.1007/s11274-010-0520-6 21475727 229. Greń I. Wojcieszyńska D. Guzik U. Perkosz M. Hupert-Kocurek K. Enhanced biotransformation of mononitrophenols by Stenotrophomonas maltophilia KB2 in the presence of aromatic compounds of plant origin World J. Microbiol. Biotechnol. 2009 26 289 295 10.1007/s11274-009-0172-6 230. Crowell P.L. Prevention and therapy of cancer by dietary monoterpenes J. Nutr. 1999 129 775S 778S 10082788 231. Ballal N.R. Bhattacharyya P.K. Rangachari P.N. Perillyl alcohol dehydrogenase from a soil pseudomonad Biochem. Biophys. Res. Comm. 1966 23 473 478 10.1016/0006-291X(66)90752-2 4289759 232. Yoo S.K. Day D.F. Bacterial metabolism of α- and β-pinene and related monoterpenes by Pseudomonas sp. strain PIN Process Biochem. 2002 37 739 745 10.1016/S0032-9592(01)00262-X 233. De Carvalho C.C.C.R. Parreño-Marchante B. Neumann G. da Fonseca M.M.R. Heipieper H.J. Adaptation of Rhodococcus erythropolis DCL14 to growth on n -alkanes, alcohols and terpenes Appl. Microbiol. Biotechnol. 2005 67 383 388 10.1007/s00253-004-1750-z 15856218 234. Bicas J.L. Fontanille P. Pastore G.M. Larroche C. A bioprocess for the production of high concentrations of R -(+)-α-terpineol from R -(+)-limonene Process Biochem. 2010 45 481 486 10.1016/j.procbio.2009.11.007 235. Eaton R.W. Sandusky P. Biotransformations of (+/−)-geosmin by terpene-degrading bacteria Biodegradation 2010 21 71 79 10.1007/s10532-009-9282-y 19578827 236. Van der Werf M.J. Swarts H.J. de Bont J.A.M. Rhodococcus erythropolis DCL14 contains a novel degradation pathway for limonene Appl. Environ. Microbiol. 1999 65 2092 2102 10224006 237. Van Dyk M.S. van Rensburg E. Moleleki N. Hydroxylation of (+)limonene, (−)α-pinene and (−)β-pinene by a Hormonema sp. Biotechnol. Lett. 1998 20 431 436 10.1023/A:1005399918647 238. Ferrara M.A. Almeida D.S. Siani A.C. Lucchetti L. Lacerda P.S.B. Freitas A. Tappin M.R.R. Bon E.P.S. Bioconversion of R -(+)-limonene to perillic acid by the yeast Yarrowia lipolytica Braz. J. Microbiol. 2013 44 1075 1080 10.1590/S1517-83822014005000008 24688495 239. Chang H.C. Oriel P. Bioproduction of perillyl alcohol and related monoterpenes by isolates of Bacillus stearothermophilus J. Food Sci. 1994 59 660 662 10.1111/j.1365-2621.1994.tb05588.x 240. Mars A. Gorissen J. van den Beld I. Eggink G. Bioconversion of limonene to increased concentrations of perillic acid by Pseudomonas putida GS1 in a fed-batch reactor Appl. Microbiol. Biotechnol. 2001 56 101 107 10.1007/s002530100625 11499915 241. Redondo-Nieto M. Barret M. Morrisey J.P. Germaine K. Martínez-Granero F. Barahona E. Navazo A. Sánchez-Contreras M. Moynihan J.A. Giddens S.R. Genome sequence of the biocontrol strain Pseudomonas fluorescens F113 J. Bacteriol. 2012 194 1273 1274 10.1128/JB.06601-11 22328765 242. Crowley D.E. Focht D.D. Gilbert E.S. Hernandez B.S. Composition and Method for Degradation of Polychlorinated Biphenyl Compounds U.S. Patent US5968360 A 19 10 1999 243. Brown J.R.M. Thompson I.P. Paton G.I. Singer A.C. Enhanced biotransformation of tce using plant terpenoids in contaminated groundwater Lett. Appl. Microbiol. 2009 49 769 774 10.1111/j.1472-765X.2009.02738.x 19843209 244. Radajewski S. Webster G. Reay D.S. Morris S.A. Ineson P. Nedwell D.B. Prosser J.I. Murrell J.C. Identification of active methylotroph populations in an acidic forest soil by stable-isotope probing Microbiology 2002 148 2331 2342 10.1099/00221287-148-8-2331 12177327 245. Mackova M. Uhlik O. Lovecka P. Viktorova J. Novakova M. Demnerova K. Sylvestre M. Macek T. Bacterial degradation of polychlorinated biphenyls Geomicrobiology: Molecular and Environmental Perspective Loy A. Mandl M. Barton L.L. Springer Dordrecht, The Netherlands 2010 347 366 246. Chen Y. Murrell J.C. When metagenomics meets stable-isotope probing: Progress and perspectives Trends Microbiol. 2010 18 157 163 10.1016/j.tim.2010.02.002 20202846 247. Musilová L. Uhlík O. Strejček M. Dudková V. Macková M. Macek T. Enchancement of Polychlorinated Biphenyls Bioremediation by Soil Enrichment with Plant Secondary Metabolites Proceedings of the 5th International Symposium on Biosorption and Bioremediation Prague, Czech Republic 25–29 June 2012 Lovecká P. Nováková M. Prouzová P. Uhlík O. ICT Prague Press Prague, Czec, Republic 2012 135 138 248. Duke J. Dr. Duke’s Phytochemical and Ethnobotanical Databases Available online: https://phytochem.nal.usda.gov/phytochem/search (accessed on 10 June 2016) 249. Bell T. Newman J.A. Silverman B.W. Turner S.L. Lilley A.K. The contribution of species richness and composition to bacterial services Nature 2005 436 1157 1160 10.1038/nature03891 16121181 250. Wardle D.A. Bonner K.I. Nicholson K.S. Biodiversity and plant litter: Experimental evidence which does not support the view that enhanced species richness improves ecosystem function Oikos 1997 79 247 258 10.2307/3546010 251. Leigh M.B. Pellizari V.H. Uhlík O. Sutka R. Rodrigues J. Ostrom N.E. Zhou J. Tiedje J.M. Biphenyl-utilizing bacteria and their functional genes in a pine root zone contaminated with polychlorinated biphenyls (PCBs) ISME J. 2007 1 134 148 10.1038/ismej.2007.26 18043623 252. He Z. Gentry T.J. Schadt C.W. Wu L. Liebich J. Chong S.C. Huang Z. Wu W. Gu B. Jardine P. Geochip: A comprehensive microarray for investigating biogeochemical, ecological and environmental processes ISME J. 2007 1 67 77 10.1038/ismej.2007.2 18043615 253. Stark S. Julkunen-Tiitto R. Holappa E. Mikkola K. Nikula A. Concentrations of foliar quercetin in natural populations of white birch (Betula pubescens ) increase with latitude J. Chem. Ecol. 2008 34 1382 1391 10.1007/s10886-008-9554-8 18946705 254. Leewis M.-C. Reynolds C.M. Leigh M.B. Long-term effects of nutrient addition and phytoremediation on diesel and crude oil contaminated soils in subarctic Alaska Cold Reg. Sci. Technol. 2013 96 129 137 10.1016/j.coldregions.2013.08.011 24501438 255. Petrini O. Fungal endophytes of tree leaves Microbial Ecology of Leaves Andrews J.H. Hirano S.S. Springer New York, NY, USA 1991 179 197 256. Compant S. Saikkonen K. Mitter B. Campisano A. Mercado-Blanco J. Editorial special issue: Soil, plants and endophytes Plant Soil 2016 1 11 10.1007/s11104-016-2927-9 257. Van Aken B. Yoon J.M. Schnoor J.L. Biodegradation of nitro-substituted explosives 2,4,6-trinitrotoluene, hexahydro-1,3,5-trinitro-1,3,5-triazine, and octahydro-1,3,5,7-tetranitro-1,3,5-tetrazocine by a phytosymbiotic Methylobacterium sp. associated with poplar tissues (Populus deltoides × nigra DN34) Appl. Environ. Microbiol. 2004 70 508 517 14711682 258. Moore F.P. Barac T. Borrernans B. Oeyen L. Vangronsveld J. van der Lelie D. Campbell C.D. Moore E.R.B. Endophytic bacterial diversity in poplar trees growing on a BTEX-contaminated site: The characterisation of isolates with potential to enhance phytoremediation Syst. Appl. Microbiol. 2006 29 539 556 10.1016/j.syapm.2005.11.012 16919907 259. Weyens N. van der Lelie D. Taghavi S. Vangronsveld J. Phytoremediation: Plant-endophyte partnerships take the challenge Curr. Opin. Biotechnol. 2009 20 248 254 10.1016/j.copbio.2009.02.012 19327979 260. Siciliano S.D. Fortin N. Mihoc A. Wisse G. Labelle S. Beaumier D. Ouellette D. Roy R. Whyte L.G. Banks M.K. Selection of specific endophytic bacterial genotypes by plants in response to soil contamination Appl. Environ. Microbiol. 2001 67 2469 2475 10.1128/AEM.67.6.2469-2475.2001 11375152 261. Pavlikova D. Macek T. Mackova M. Pavlik M. Monitoring native vegetation on a dumpsite of PCB-contaminated soil Int. J. Phytoremediat. 2007 9 71 78 10.1080/15226510601139433 18246716 262. Uhlik O. Wald J. Strejcek M. Musilova L. Ridl J. Hroudova M. Vlcek C. Cardenas E. Mackova M. Macek T. Identification of bacteria utilizing biphenyl, benzoate, and naphthalene in long-term contaminated soil PLoS ONE 2012 7 1205 10.1371/journal.pone.0040653 22808223 263. Uhlík O. Strejček M. Junková P. Šanda M. Hroudová M. Vlček C. Macková M. Macek T. Matrix-assisted laser desorption ionization (MALDI)-time of flight mass spectrometry- and MALDI biotyper-based identification of cultured biphenyl-metabolizing bacteria from contaminated horseradish rhizosphere soil Appl. Environ. Microbiol. 2011 77 6858 6866 10.1128/AEM.05465-11 21821747

PMC005xxxxxx/PMC5000604.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081206ijms-17-01206ArticleFull-Length cDNA Cloning, Molecular Characterization and Differential Expression Analysis of Lysophospholipase I from Ovis aries Liu Nan-Nan 1†Liu Zeng-Shan 1†Hu Pan 1†Zhang Ying 1†Lu Shi-Ying 1Li Yan-Song 1Yang Yong-Jie 2Zhang Dong-Song 3Zhou Yu 1Ren Hong-Lin 1*da Silva Mateus Webba Academic EditorTikkanen Ritva Academic Editor1 Key Laboratory of Zoonosis Research, Ministry of Education/Institute of Zoonosis/College of Veterinary Medicine, College of Animal Sciences, Jilin University, Xi An Da Lu 5333, Changchun 130062, China; [email protected] (N.-N.L.); [email protected] (Z.-S.L.); [email protected] (P.H.); [email protected] (Y.Z.); [email protected] (S.-Y.L.); [email protected] (Y.-S.L.); [email protected] (Y.Z.)2 Department of Food Science, College of Agriculture, Yanbian University, Yanji 133002, China; [email protected] Animal Husbandry and Veterinary Unit of Xiangyang Town, Liuhe 135305, China; [email protected]* Correspondence: [email protected]; Tel.: +86-431-8783-5735† These authors contributed equally to this work. 28 7 2016 8 2016 17 8 120603 5 2016 19 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Lysophospholipase I (LYPLA1) is an important protein with multiple functions. In this study, the full-length cDNA of the LYPLA1 gene from Ovis aries (OaLypla1) was cloned using primers and rapid amplification of cDNA ends (RACE) technology. The full-length OaLypla1 was 2457 bp with a 5′-untranslated region (UTR) of 24 bp, a 3′-UTR of 1740 bp with a poly (A) tail, and an open reading frame (ORF) of 693 bp encoding a protein of 230 amino acid residues with a predicted molecular weight of 24,625.78 Da. Phylogenetic analysis showed that the OaLypla1 protein shared a high amino acid identity with LYPLA1 of Bos taurus. The recombinant OaLypla1 protein was expressed and purified, and its phospholipase activity was identified. Monoclonal antibodies (mAb) against OaLypla1 that bound native OaLypla1 were generated. Real-time PCR analysis revealed that OaLypla1 was constitutively expressed in the liver, spleen, lung, kidney, and white blood cells of sheep, with the highest level in the kidney. Additionally, the mRNA levels of OaLypla1 in the buffy coats of sheep challenged with virulent or avirulent Brucella strains were down-regulated compared to untreated sheep. The results suggest that OaLypla1 may have an important physiological role in the host response to bacteria. The function of OaLypla1 in the host response to bacterial infection requires further study in the future. Lysophospholipase IOvis ariesBrucellatissue distributiondifferential expression ==== Body 1. Introduction Lysophospholipase I (LYPLA1), also known as acyl-protein thioesterase 1 (APT1), is a widely distributed enzyme with phospholipase A2 [1], lysophospholipase [1,2] and acyl-protein thioesterase [3,4,5] activity. Phospholipids are important components in cellular membranes and are involved in signal transduction, mediator production and eicosanoid formation in both normal and disease states [6]. Lysophospholipids (LPLs) are important molecules in phospholipid metabolism and have received much attention as they are critical for cell survival and function [6,7,8]. LPLs are bioactive second messengers that modulate gene expression and are involved in multiple processes, such as stimulation of growth, phagocytosis of macrophages [6,9,10,11,12], activation of T lymphocytes [13,14], activation of many immune-related proteins [15], and promotion of anti-tumor and bactericidal activities [6]. LPLs are strictly regulated because increased levels of lysophospholipids are associated with many diseases [6,16]. The lysophospholipases (LYPLAs) are considered to be safeguards to ensure that normal levels lysophospholipids are maintained [6]. Multiple enzymes display LYPLA activity, and LYPLA1 belongs to the family of LYPLAs [4,16]. Palmitoylation involves the attachment of a 16-carbon fatty acid palmitate via a thioester bond to specific cysteine residues of target proteins and is an important post-translational modification that is critical for protein localization and function [5]. In contrast to other lipid modifications, such as isoprenylation and myristoylation, palmitoylation is a unique reversible post-translational modification [17] that allows proteins to rapidly shuttle between intracellular membrane compartments [18,19] and can be dynamically regulated by specific extracellular stimuli [18]. This modification is important for regulating protein subcellular localization, stability, trafficking, translocation to lipid rafts, aggregation, and interaction with effectors and other protein functions [17]. Recent studies have shown that palmitoylation is involved in endocytosis, reproduction, cell growth, fat and sugar homeostasis and signal transduction at the synapse [18]. Palmitoylation/depalmitoylation cycles are potential novel regulatory networks [20]. The multiple functions of palmitoylation suggest that it should be studied in detail. Palmitoylation is catalyzed by palmitoylation acyltransferases (PATs) while only a few enzymes catalyze depalmitoylation reactions [5,21]. LYPLA1 was identified as an acyl-protein thioesterase (APT), which is a depalmitoylation enzyme, in addition to palmitoyl protein thioesterases (PPT) [3], and can catalyze the removal of mislocalized palmitoylated proteins from endomembranes by depalmitoylation [21,22,23]. The PPT is localized predominantly in the lysosome [24]. However, LYPLA1 is principally localized in the cytosol, while some is present in the plasma membrane, the nuclear membrane and the endoplasmic reticulum (ER) [21]. Brucellosis is a zoonotic disease found worldwide that is caused by Brucella, resulting in infectious abortion and fever [25]. Brucella infections are chronic, and the interaction between the host and the Brucella pathogen is continuous [26]. Virulent Brucella strains invade the macrophages through lipid rafts and then reside in an acidified compartment, which fuses with components of the early endosomal pathway [27,28]. The macrophages kill the majority of Brucella cells at an early stage of infection [28,29], and the remaining Brucella cells establish and maintain a persistent intracellular infection in host cells using many virulence factors and strategies. Brucella can also induce the loss of the robust antigen-processing capacity of professional phagocytes and prevent phagosome-lysosome fusion and programmed cell death of infected macrophages, favoring pathogen survival and replication [26,30,31,32,33]. The host transcriptional responses against infection by Brucella have been characterized in several studies [28,34,35]. It is crucial to understand the host response to Brucella. In our lab, a time course suppression subtractive hybridization (SSH) cDNA library of buffy coats from sheep (Ovis aries) infected with different virulent Brucella strains was constructed to analyze the modulation of transcriptional profiles of hosts exposed to Brucella infection, and the differentially transcribed genes were screened. Among these genes, a partial cDNA of LYPLA1 (OaLypla1) containing a full-length 3′-UTR was found to show differential expression in buffy coats from different virulent Brucella-infected sheep. As a result, the OaLypla1 gene was chosen as a target candidate gene to further study the response of the host to Brucella infection. LYPLA1 is a protease with diverse biological functions that catalyzes multiple different reactions. It was reported that LYPLA1 is down-regulated in macrophages after LPS stimulation [36]. However, the expression profiles of LYPLA1 gene of the host infected with bacteria and whether LYPLA1 takes part in the host immune response after the bacterial infection had not been investigated before. Thus, more information about the OaLypla1 is required to better understand the potential relationship between the expression profiles of OaLypla1 gene and bacteria infection. In this study, we identified the full-length cDNA sequence of a novel LYPLA1 gene from Ovis aries (O. aries) for the first time. The recombinant OaLypla1 protein was expressed and purified, and its phospholipase activity was assessed. The tissue distribution of OaLypla1 was determined, and differential expression profiles of OaLypla1 in the buffy coats of sheep following challenge with different virulent Brucella strains were observed. Furthermore, we generated a monoclonal antibody (mAb) that reacts with the native OaLypla1. The results from this study may facilitate further study of the functions of OaLypla1 in the host response to infection with Brucella. 2. Results 2.1. Characterization of OaLypla1 cDNA The full-length cDNA sequence of the OaLypla1 gene was obtained using 5′-RACE and deposited in GenBank (accession number KJ000742). The full-length OaLypla1 cDNA was 2457 bp with a 5′-UTR of 24 bp, an ORF of 693 bp and a 3′-UTR of 1740 bp with a poly (A) tail downstream of a polyadenylation signal (AATAAA). The full-length nucleotide sequence and the deduced amino acid sequence of OaLypla1 cDNA are shown in Figure 1. The predicted OaLypla1 protein consisted of 230 amino acid residues with a predicted molecular weight of 24,625.78 Da and a theoretical isoelectric point of 6.77. The deduced OaLypla1 protein contained a 117GFSQG121 amino acid sequence corresponding to the GXSXG motif, which was located in an identical position to the amino acid sequence of the human LYPLA1 protein. A search using the BLASTn program in NCBI showed that the OaLypla1 cDNA had 96% identity with the LYPLA1 cDNA from Bos taurus (GenBank accession number: BC105143). Using the BLASTP program, the deduced amino acid of O. aries OaLypla1 was shown to exhibit high homology with the LYPLA1 proteins of other species, such as Bos taurus (99% identity), Pongo abelii (95% identity), Homo sapiens (95% identity), Macaca mulatta (94% identity), Cricetulus griseus (93% identity), Xenopus tropicalis (82% identity), and Dicentrarchus labrax (78% identity). Multiple sequence alignment analysis of OaLypla1 was conducted using the known LYPLA1 proteins from several vertebrates to determine the level of amino acid conservation. The results showed that highly conserved acids were observed in the entire protein sequence as shown in Figure 2. The deduced amino acid sequence of LYPLA1 from O. aries had the Ser119, Asp174 and His208 triad that formed the catalytic site for LYPLA1 proteins from the mouse [37] and human [2]. All of these proteins shared the GXSXG motif sequence (152GFSQG156), which had a similar position and was found in the active site of serine proteases, esterases and lipases [6]. To determine the phylogenetic relationships of OaLypla1, the amino acid sequences of LYPLA1 proteins from different species were selected. The phylogenetic tree was constructed by the neighbor joining method and revealed that the deduced amino acid sequence of OaLypla1 clustered with the ruminant subgroup near LYPLA1 from Bos taurus (Figure 3), suggesting that the gene cloned from O. aries belongs to the LYPLA1 family. 2.2. Protein Expression The coding sequence of OaLypla1 was cloned and inserted into the pET-30a vector. The recombinant proteins OaLypla1H (with a His6-tag) and OaLypla1W (with no tag) were over-expressed in E. coli BL21 (DE3). The results of SDS-PAGE analysis showed an approximately 25 kDa band in the induced cells, indicating that the recombinant OaLypla1H and OaLypla1W were successfully expressed in the transformed E. coli BL21 (DE3) cells following IPTG induction (Figure 4A). The recombinant OaLyplaH protein was purified and detected by SDS-PAGE and Western blots using a commercial His tag antibody and showed the same molecular weight of approximately 25 kDa, including the molecular weight of His6-tag (Figure 4B,C). 2.3. Activity Assay The phospholipase activity was determined by the egg yolk/agarose diffusion test. Phospholipase A is able to degrade micellar lecithins and cephalins into dissolvable lyso compounds and fatty acids, which leads to a clearing of egg yolk suspension in an agarose gel plate. Transparent rings will develop around the holes in which the phospholipase A is added after the egg yolk suspension is incorporated into the agarose gels [38]. The results of the egg yolk/agarose diffusion test showed that transparent rings were visualized around the holes after purified recombinant OaLypla1H was added. The diameters of the transparent rings were larger with increased concentrations of OaLypla1H. There were no transparent rings in the BSA and OaPDCD10 groups (Figure 5). 2.4. Specificity of the Monoclonal Antibody (mAb) A hybridoma cell line secreting mAbs against OaLypla1 was obtained as described in the methods and named OaLypla1-4A3. The results of the Western blotting analyses showed that the mAbs were able to bind the recombinant OaLypla1 and native OaLypla1 proteins extracted from kidney (Figure 6). 2.5. Tissue Distribution of OaLypla1 To determine the tissue expression profiles of OaLypla1, qPCR and Western blot analyses were carried out to examine the tissue distribution of OaLypla1 with the primers (Table 1) and the mAb prepared as described above. As shown in Figure 7A, the OaLypla1 mRNA was detected in the liver, spleen, lung, kidney and WBCs, with the highest expression in the kidney. OaLypla1 protein was detected in the liver, spleen, lung and kidney but not in the WBCs (Figure 7B) by Western blotting. 2.6. OaLypla1 Expression Profiles after Challenge with Virulent and Avirulent Brucella Strains The expression profiles of OaLypla1 in the buffy coats of O. aries challenged with virulent and avirulent Brucella strains were analyzed using qPCR, as shown in Figure 8. Compared with the untreated control group, the transcription levels of OaLypla1 were down-regulated in both the BmF-challenged group and the S2-challenged group from 3 to 75 dpc. In the S2-challenged group, the level of OaLypla1 transcription increased from 3 to 14 dpc, reached a peak at 14 dpc, and then decreased, with the lowest level at 75 dpc. In the BmF-challenged group, the OaLypla1 levels remained at a low level compared with the S2-challenged group. However, at 21 dpc, OaLypla1 in the BmF-challenged group was much higher than in the S2-challenged group. 3. Discussion Lysophospholipases are critical enzymes to regulate the multifunctional lysophospholipids. In this study, we focused on the OaLypla1 gene screened from a time course SSH cDNA library constructed in our lab before. The full-length cDNA of OaLypla1 was cloned and sequenced for the first time. The OaLypla1 gene was characterized at the molecular level and the recombinant protein was produced to identify its functional activities in vitro. Further, the tissue-specific expression of OaLypla1 was determined, and the bacterial stress responses of OaLypla1 were investigated after sheep were infected with Brucella. This was the first report about the LYPLA1 gene from sheep (Ovis aries). It is reported that LYPLA1 of murine represents a member of the serine hydrolase family with Ser119, Asp174, and His208 composing the catalytic triad [39], and site-directed mutagenesis indicated that mutation of each residue to Ala abolished LYPLA1 activity [37,39]. The protein sequence analysis showed that OaLypla1 has the same catalytic triad with LYPLA1 from murine, rat and human. To confirm the bioactivity of the lipase catalytic center 117GFSQG121 in the OaLypla1 protein, the phospholipase activity of the recombinant OaLypla1 was identified using the egg yolk/agarose diffusion test in this study. This is a sensitive and simple test that has been used frequently for phospholipase A activity assay in recent years [40,41,42,43,44]. The purified recombinant OaLypla1 was also used to generate an anti-OaLypla1 mAb that was confirmed to specifically bind to the recombinant and native OaLypla1 proteins. The tissue distribution of the OaLypla1 gene at both the mRNA and protein levels was measured using qPCR and Western blots, respectively. OaLypla1 was detected in all of the examined tissues, showing the highest mRNA expression in the kidney. The result was similar to the previous reports [2,4]. OaLypla1 was constitutively expressed in tissues at the transcriptional level, suggesting that OaLypla1 is a ubiquitously expressed gene and is a critical molecule that could potentially be involved in numerous physiological functions. OaLypla1 protein was not detected in WBCs, which was consistent with the low OaLypla1 mRNA abundance in WBCs. As described in a previous study, gene expression levels cannot be determined only by mRNA abundance based on the mRNA–protein correlation [45], and post-transcriptional processing would also affect the expression levels of genes. The different mRNA and protein abundances resulting from the same gene emphasize the importance of integrative analysis of transcription and translation [46]. LYPLA1 has been cloned from rat [4], murine [37], and human [2]. The lysophospholipase, phospholipase A2 and thioesterase activities were identified [21]. However, there is no investigation about the relationship between LYPLA1 and infection of bacteria. The LYPLA1 gene from Ovis aries cloned in this study may take apart in the immune response of the host towards the infection of bacteria. Brucellosis is a major zoonotic disease worldwide. Brucella can evade the host immune response to survive and reproduce in host cells. Currently, live attenuated vaccines are used to eradicate brucellosis in cattle, sheep and goats. Infection by both the live attenuated vaccines and by the virulent strain will cause similar immune responses, such as stimulation of Th1 responses [47]. As a consequence, it is impossible to distinguish between vaccinated animals and infected ones using the available serological tests at present [48]. It was reported that the expression profiles and releases of LYPLA1 are related to immunological stimulation and decreased LYPLA1 levels likely contribute to macrophage responses to pro-inflammatory stimuli. Along with the mRNA and protein levels of LYPLA1 decreased in LPS-stimulated RAW 264.7 cells, LYPLA1 released from RAW 264.7 cells into the culture medium was significantly increased. Inhibition of circulating LYPLA1 activity may be an effective treatment strategy for inflammation [36]. In the present study, the expression profiles of OaLypla1 were analyzed using qPCR, and the results confirmed that the transcriptional levels of OaLypla1 in WBCs were down-regulated by infection of Brucella compared with the uninfected sheep. Down-regulation of OaLypla1 expression in WBCs from the infected groups showed that LYPLA1 was perhaps involved in the immune response to the stimulation, and indicated that the transcription of the LYPLA1 gene is inducible, which is consistent with the previous report [36]. There are hardly any studies about the relationship between the expression profile of LYPLA1 and bacterial infection. LYPLA1 can hydrolyze lysophospholipids and participate in the phosphatidylcholine (PLC) pathway [39,49]. Thus, we speculated that the decreased LYPLA1 might result in the increasing level of LPLs, which are able to attract and activate macrophages and T or B cells, influence their interactions with other types of cells, and promote and modulate immune responses [50]. The exact mechanism of this phenomenon needs to be further studied. In addition, we also determined that the transcription patterns of the OaLypla1 gene in buffy coats were different between sheep infected with the avirulent Brucella suis S2 vaccine strain and those infected with the virulent Brucella melitensis field strain. Whether OaLypla1 can be a biomarker distinguishing between virulent Brucella infection and the avirulent Brucella vaccine inoculation and the relationship between OaLypla1 and brucellosis need to be further studied in the future. 4. Materials and Methods 4.1. Animals and Cells All of the sheep used in this research were purchased from Sangang farm (Jilin Province, China) and had no infectious diseases. Female BALB/c mice and sheep received food and water ad libitum [51]. All the animal experiments were carried out abiding by the provisions of EU animal management practices (24 November 1986), and approved by the Animals Ethics Committee of Jilin University of China in accordance with the Jilin university ethnic committee guideline for the Care and Use of Laboratory Animal (No. SCXK 2015-0004, 7 January 2015). The avirulent strain of Brucella (S2) was purchased from Harbin Pharmaceutical Group Bioengineering Co., Ltd. (Haerbin, China). The isolated and identified virulent strain of Brucella, myeloma SP2/0 cells, Escherichia coli DH5α competent cells and E. coli BL21 (DE3) cells were provided by the Key Laboratory of Zoonosis Research, Ministry of Education, Jilin University (Jilin, China). 4.2. Total RNA Isolation Total RNA was isolated as described previously [48]. Briefly, the anticoagulant-containing blood collected from O. aries was centrifuged at 800× g for 15 min at 4 °C to extract the buffy coats. The buffy coats were used for total RNA extraction using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. DNA and protein contamination was removed using recombinant DNase I (RNase-free) (TaKaRa, Dalian, China) and an RNeasy MinElute Cleanup Kit (Qiagen, Hilden, Germany). 4.3. 5′-RACE and Sequence Assembly To obtain the full-length cDNA sequence, the purified total RNA and a SMARTer™ RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA) were used. A set of gene-specific primers was designed (Table 1) and named GSP (gene-specific primer) and NGSP (nested gene-specific primer). The PCR amplifications were performed using the following protocol. Briefly, 20 µL reaction volumes containing 5′-RACE-Ready cDNA as the template, primers (0.4 µL of 10 µM GSP and 2.0 µL UPM provided by the kit), 0.4 µL 10 mM dNTPs, and 0.4 µL 50× Advantage Polymerase mix were prepared. The conditions were 94 °C for 3 min; 20 cycles of 30 s at 94 °C, 30 s at 54.1 °C, 3 min at 72 °C; then a final extension at 72 °C for 5 min. The products were diluted 1:50 and used as template cDNA for the nested PCR. The nested PCR amplification was similar to the first PCR amplification except for the primers and template. The PCR conditions were 3 min at 94 °C; 25 cycles of 30 s at 94 °C, 30 s at 61.1 °C, 3 min at 72 °C; then the last extension at 72 °C for 5 min. The PCR products were visualized on a 1% agarose gel stained with ethidium bromide. After purification and ligation into the pMDTM18-T vector (TaKaRa, Dalian, China), the target DNA products were transformed into E. coli DH5α competent cells and sequenced by Shanghai Sangon Biological Engineering Technology & Service Co., Ltd. (Shanghai, China). The obtained sequences and the partial cDNA sequence identified from the SSH cDNA library were assembled to obtain the full-length cDNA sequence of OaLypla1. 4.4. Sequence Verification and Analysis A homology search for the assembled sequence of OaLypla1 was performed using the BLAST search programs at NCBI [52]. The molecular weight of the putative protein was predicted using the Expert Protein Analysis System [53]. Characteristic domains or motifs were identified using the Motif scan program [54]. The amino acid sequences of LYPLA1 from various species were retrieved from NCBI and analyzed using ClustalW version 1.83. The phylogenetic tree was constructed based on the amino sequence alignment with the neighbor joining method from the MEGA version 4.1 program. 4.5. Cloning of OaLypla1 The open reading frame (ORF) of the Oalylpla1 transcript was amplified using the forward primer (20)KLS with a NdeI recognition site and the reverse primers (20)KLHisA (for the 6× His tag fusion) and (20)KLWA (for no fusion) with a XhoI recognition site at their 5′-ends, as listed in Table 1. The recombinant plasmids constructed in 4.3 were used as the template with Ex Taq (TaKaRa, Dalian, China), and the PCR reactions were performed at 94 °C for 30 s; then at 94 °C for 30 s, 60 °C for 30 s, 72 °C for 60 s for 32 cycles; and the final extension at 72 °C for 10 min. The PCR products were separated, purified and ligated into the pMDTM19-T Simple vector and transformed into E. coli DH5α cells. The cells with recombinant plasmids were confirmed with PCR using M13 forward and reverse primers, and the plasmids were then sequenced. 4.6. Expression and Purification of the Recombinant OaLypla1 Protein The recombinant plasmids were digested with NdeI and XhoI restriction enzymes, and the products were purified and ligated into the restriction enzyme-digested pET-30a vectors. The recombinant expression plasmids were transformed into E. coli BL21 (DE3), and the cells carrying the plasmid were named pET-30a-LYPLA1-H and pET-30a-LYPLA1-W. A single transformant colony was grown overnight in 5 mL Luria-Bertani (LB) medium containing 50 µg/mL kanamycin at 37 °C, and the protein expression of OaLypla1H (OaLypla1 with His6-tag) and OaLypla1W (OaLypla1 with no His6-tag) was separately induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) at a final concentration of 1.5 mM at 37 °C for 6.5 h. Total protein lysates extracted from the induced pET-30a-LYPLA1-H and pET-30a-LYPLA1-W cells were analyzed using 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Protein expression of OaLypla1H in 200 mL fresh LB medium was induced by adding IPTG at a final concentration of 1.5 mM at 37 °C for 6.5 h. The induced pET-30a-LYPLA1-H cells were harvested and resuspended in binding buffer (20 mM sodium phosphate, 30 mM imidazole, 0.5 M NaCl, pH 7.4) [48]. The supernatant was collected by centrifugation at 12,000× g for 30 min at 4 °C after the resuspended cells were sonicated. Then, the supernatant was loaded onto a HisTrapTM FF crude (GE Healthcare, Pittsburgh, PA, USA) resin, and after washed with binding buffer for 10 times of column volume, the OaLypla1H protein was eluted with elution buffer (20 mM sodium phosphate, 0.5 M NaCl, 0.5 M imidazole, pH 8.0). The purified protein was dialyzed using 0.01 M PBS (pH 8.0). The purified protein OaLypla1H was run on 12% SDS-PAGE with a protein marker (Thermo, Waltham, MA, USA) and stained with Coomassie brilliant blue R250. The molecular mass and the purity of the purified protein were assessed. Western blotting was performed with a commercial anti-His tag antibody (Abcam, Cambridge, MA, USA) to demonstrate that the recombinant OaLypla1H protein had been expressed and purified. 4.7. Phospholipase Activity Assay The phospholipase activity of the OaLypla1H protein was analyzed as described in previous reports [38,55,56,57]. Briefly, the egg yolk was diluted 1:4 with 0.85% NaCl, and the supernatant (Buffer A) was collected by centrifugation at 3500 rpm for 2 min. Agarose (0.6 g) was dissolved in 100 mL 0.05 M NaAc (pH 7.5) at 120 °C for 10 min. Buffer A (3 mL) and 0.01 M CaCl2 (1 mL) were added to the 0.6% agarose NaAc solution when the temperature cooled to 50 °C, and the solution was then poured into glass plates (Ф 150 mm), and a hole was punched after the agarose solidified. A 0.85% NaCl solution (50 µL) containing different concentrations of purified OaLypla1H (0.05–1.2 mg/mL) was added into the holes of the agarose plates, and 50 µL of 0.85% NaCl containing 1.6 mg/mL of the bovine serum albumin (BSA) and 0.8 mg/mL of the recombinant programmed cell death 10 of Ovis aries (OaPDCD10) with a 6× His tag [58] were applied as negative controls. The plates were incubated at 37 °C for 24 h. The diameters of visible transparent circles appearing as a result of the phospholipase activity of OaLypla1H were measured. 4.8. Preparation of the mAb against the OaLypla1 Protein Eight- to ten-week-old female BALB/C mice were immunized in the footpad with 100 µg of OaLypla1H emulsified with an equal volume of complete Freund’s adjuvant (CFA, Sigma, St. Louis, MO, USA) or incomplete Freund’s adjuvant (IFA, Sigma, St. Louis, MO, USA) based on a previous report [59]. The mice were housed with adequate food and water. Four days after the fourth immunization, the spleen cells isolated from the immunized mice were fused with myeloma cells (SP2/0) at a ratio of 10:1 in the presence of PEG1000. Then, the fused cells were cultured in 96-well cell culture plates using 20% (v/v) FBS/HAT (Sigma, St. Louis, MO, USA) medium, which was changed once per 4 days. Two weeks later, the hybridoma cells secreting the anti-OaLypla1 mAb were screened and cloned by limiting dilution at least three times. The culture supernatant of the positive hybridoma cells was used for the next experimental study. 4.9. Specificity Analysis The specificity of the mAb was analyzed using Western blotting as described in a previous report [48]. Untagged OaLypla1W, OaLypla1H with a His6-tag and whole proteins from sheep kidneys were used to confirm the specificity of the mAb binding by Western blotting. Briefly, protein samples were extracted from bacteria or sheep kidneys, separated with 12% SDS-PAGE, and probed with the anti-OaLypla1 mAb (dilution 1:20) after transfer to a PVDF membrane (Millipore, Billerica, MA, USA). Then, the horseradish peroxidase-labeled goat anti-mouse IgG (dilution 1:2000) was incubated with the PVDF membrane. The membrane detection was performed with a BeyoECL Plus kit (Beyotime, Shanghai, China) using the ECL detection system (DNR, Jerusalem, Israel). 4.10. Tissue Distribution Tissue distribution of OaLypla1 in the liver, spleen, lung, kidney and white blood cells (WBCs) of sheep was analyzed by quantitative real-time PCR (qPCR) and Western blotting. Total RNA was extracted from different tissues and organs of 3 healthy sheep using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. One microgram of total RNA was used to synthesize cDNA using a PrimeScriptTM RT reagent kit with gDNA Eraser (Perfect Real Time) (TaKaRa, Dalian, China). The qPCR reaction was carried out in a total volume of 20 µL containing 10 µL of FastStart Universal SYBR Green Master (ROX) (Roche, Basel, Switzerland), 0.6 µL of the forward primer (10 µM), 0.6 µL of the reverse primer (10 µM), and 1.5 µL of 5-fold diluted cDNAs. The qPCR cycling protocol was 95 °C for 10 min and 40 cycles of 15 s at 95 °C, 62 °C for 35 s and 72 °C for 32 s. Samples were normalized with β-actin, and the relative transcription level of OaLypla1 was calculated using the 2−ΔΔCt method. Each assay was repeated in triplicate. The total proteins from different tissues (liver, spleen, lung, kidney and WBCs) were extracted using RIPA lysis buffer (Beyotime, Shanghai, China) with 10 µL PMSF according to the manufacturer’s instructions. The total protein concentration was measured by the Bradford method [60] using a Bradford protein assay kit (Bio-Rad, Hercules, CA, USA). Total proteins (125 µg) from different sheep tissues and organs were used for Western blot analysis with the generated anti-OaLypla1 mAb following a standard protocol [48], and β-actin was used as an internal control. Optical densities of OaLypla1 and β-actin were calculated using Quantity One software (Bio-Rad, Hercules, CA, USA). Significant differences were determined by one-way analysis of variance (ANOVA) using the SPSS 13.0 software (IBM, Armonk, NY, USA). 4.11. Relative Transcript Level Analysis of OaLypla1 in Buffy Coats after Challenge with Virulent and Avirulent Brucella Strains For differential expression analysis of OaLypla1, qPCR was employed with a pair of gene-specific primers (OaLypla1S and OaLypla1A) listed in Table 1, and β-actin (GenBank accession number U39357) was used as an internal control [48]. As described in a previous study [61], 9 sheep were randomly divided into three groups (n = 3). Three sheep were challenged with a virulent Brucella melitensis field strain (BmF) as the BmF-challenged group, and three sheep were inoculated with an avirulent Brucella suis S2 vaccine strain (S2) as the S2-inoculated group. Each sheep was injected with a total dose of 2.2 × 109 cfu of bacteria. In addition, three other sheep were treated with the same volume of sterile 0.85% NaCl as the normal control group. The buffy coat samples from three individuals in each group were obtained at 3, 7, 14, 21, 30, 40, 50, 60 and 75 days post-challenge (dpc). Extracting the total RNAs and performing the qPCR for OaLypla1 differential expression analysis in buffy coats were performed as described in 4.10 above. Each assay was repeated in triplicate. 5. Conclusions This study identified and characterized a novel full-length cDNA of the LYPLA1 gene from Ovis aries (OaLypla1), and the corresponding protein was expressed, purified and characterized. OaLypla1 was widely expressed in different tissues at both the mRNA and protein levels. qPCR analysis showed that the expression of OaLypla1 in the white blood cells of Ovis aries was down-regulated after infected with Brucella. This study provides fundamental data for further investigations exploring the relationship between Brucella infection and the expression patterns of OaLypla1 and the possible functions of OaLypla1 during Brucella infection. Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 30901070), the Science & Technology Development Project of Jilin Province, China (No. 20150204078NY) and the Graduate Innovation Fund of Jilin University (No. 2015041). Author Contributions Hong-Lin Ren conceived and designed the experiments; Nan-Nan Liu, Zeng-Shan Liu, Pan Hu and Ying Zhang performed the experiments; Yong-Jie Yang, Shi-Ying Lu, Yan-Song Li, and Yu Zhou analyzed the data; Dong-Song Zhang contributed materials and analysis tools; and Nan-Nan Liu, Zeng-Shan Liu, Pan Hu, Ying Zhang and Hong-Lin Ren wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 The full-length cDNA and the corresponding amino acid sequence of OaLypla1. The polyadenylation signal sequence (AATAAA) is shown in bold and underlined. The lipase motif GXSXG is boxed, and the catalytic triad is marked by asterisks (*). The arrows represent the places and lengths of the primers (5’-3’) used in this study. Figure 2 Multiple alignment analysis of the amino acid sequences of LYPLA1s from different vertebrates. The conserved amino acid residues of LYPLA1s are indicated by asterisks (*) above the column. Conserved substitutions are shown by colons (:), and dots (.) indicate semi-conserved amino acids. GenBank GI numbers of LYPLA1 protein sequences are given as follows: Ovis aries LYPLA1: gi (592882195); Bos taurus LYPLA1: gi (77736321); Cricetulus griseus LYPLA1: gi (537237418); Homo sapiens LYPLA1: gi (5453722); Macaca mulatta LYPLA1: gi(388453011); Mus musculus LYPLA1: gi (6678760); Oryctolagus cuniculus LYPLA1: gi (157954426); Pongo abelii LYPLA1: gi (197099340); Rattus norvegicus LYPLA1: gi (6981362); and Xenopus (Silurana) tropicalis LYPLA1: gi(54020910). The different colors are used to differ the proteinogenic 20 amino acids. Red: R and K; Brick red: G; Orange: C; Yellow: P; Green: N, S, T, and Q; Indigo: H and Y; Blue: M, A, L, I, V, F and W; Purple: D and E. Figure 3 Phylogenetic relationship of LYPLA1 proteins from different species. The phylogenetic tree was constructed using MEGA 4.1 with the ClustalW algorithm. The number on the nodes indicates bootstrap values from 1000 replications. Figure 4 Expression and purification analysis of the recombinant OaLypla1 protein. (A) Expression analysis of the recombinant OaLypla1 by SDS-PAGE. M: protein marker (Thermo, Waltham, MA, USA); lane 1: total proteins from the uninduced pET-30a-LYPLA1-W cells; lane 2: total proteins from the induced pET-30a-LYPLA1-W cells (expressing OaLypla1 with no His6-tag); lane 3: total proteins from the induced pET-30a-LYPLA1-H cells (expressing OaLypla1 with a His6-tag) showing the shifted band due to the His6-tag; (B) Purification analysis of the recombinant OaLypla1H by SDS-PAGE. M: protein marker (Thermo, Waltham, MA, USA); lane 1: total proteins from the uninduced pET-30a-LYPLA1-H cells; lane 2: total proteins from the induced pET-30a-LYPLA1-H cells; lane 3: the purified recombinant protein OaLypla1H; (C) Expression and purification analysis of the recombinant OaLypla1H by Western blotting. M: protein marker (Thermo, Waltham, MA, USA); lane 1: total proteins from the uninduced pET-30a-LYPLA1-H cells; lane 2: total proteins from the induced pET-30a-LYPLA1-H cells; lane 3: the purified recombinant protein OaLypla1H. Figure 5 Phospholipase activity assay of the OaLypla1 protein. (A) Egg yolk/agarose diffusion test: a, 0.8 mg/mL OaPDCD10 with a His6-tag; b, 1.6 mg/mL BSA; and c–h, 0.05, 0.1, 0.2, 0.4, 0.8 and 1.2 mg/mL of OaLypla1H; (B) Relative area ratio. The results were calculated as a measure of the relative area ratio with the area of the hole in the center as 100%. Figure 6 Immunoassay specificity of the monoclonal antibody against OaLypla1. (A) Immunoassay specificity against the recombinant OaLypla1. M: protein marker (Thermo, Waltham, MA, USA); lane 1: total proteins from the uninduced pET-30a-LYPLA1-W cells; lane 2: total proteins from the induced pET-30a-LYPLA1-W cells; lane 3: total proteins from the induced pET-30a-LYPLA1-H cells; (B) Immunoassay specificity against the native OaLypla1. M: protein marker (Thermo, Waltham, MA, USA); lane 1: total proteins from the induced pET-30a-LYPLA1-H cells as a positive control; lane 2: total proteins extracted from the kidney of O. aries. The pET-30a-LYPLA1-W cells were induced to express the recombinant OaLypla1 with no His6-tag, and the pET-30a-LYPLA1-H cells expressed OaLypla1 fused with a His6-tag. Figure 7 Tissue distribution of OaLypla1. (A) The tissue distribution of OaLypla1 determined by quantitative real-time PCR. The mRNA levels in tissues were normalized with β-actin; (B) OaLypla1 protein detected in the tissues using Western blotting. Statistical differences among liver, spleen, lung, kidney, and white blood cells were determined by one-way analysis of variance (ANOVA) using SPSS 13.0 software (IBM, Armonk, NY, USA). Data are presented as the mean relative expression ± SD (n = 3). An asterisk indicates a statistically significant difference (* p < 0.05). WBCs: white blood cells. Figure 8 Differential expression of OaLypla1 from buffy coats of O. aries challenged with different virulent Brucella strains. N: the normal sheep group; S: the group inoculated with the avirulent Brucella suis S2 strain; and B: the group challenged with the virulent field strain Brucella melitensis. Relative expression was calculated by the 2−∆∆Ct method using O. aries β-actin as an endogenous control. Statistical differences among the groups were determined by a one-way analysis of variance (ANOVA) using SPSS 13.0 software. Data are presented as the mean relative expression ± SD (n = 3) (* p < 0.05, ** p < 0.01). ijms-17-01206-t001_Table 1Table 1 Primers used in this study. Primer Object Sequence (5′-3′) GSP 5′-RACE 5′-TCTTGCCATAAGTTAGATCTTGCTG-3′ NGSP 5′-RACE 5′-GTCACTTCCATCATCAAATAGCACC-3′ (20)KLS ORF amplification 5′-CATATGTGCGGCAATAACATGTCGGC-3′ (20)KLHisA ORF amplification 5′-CTCGAGGTCAATGGGAGGTAGGAGCTTATC-3′ (20)KLWA ORF amplification 5′-CTCGAGTCAGTCAATGGGAGGTAGGAGCTTATC-3′ β-actin-S qPCR 5′-CCCAAGGCCAACCGTGAGAAGATGA-3′ β-actin-A qPCR 5′-CGAAGTCCAGGGCCACGTAGCAGAG-3′ OaLypla1S qPCR 5′-CCTATTGGTGGCGTGAACAGAGAC-3′ OaLypla1A qPCR 5′-GAACTGTGCATCATGCCTGCGTAG-3′ The letters marked by the single underline in the primer sequences stood for the restriction sites of Nde I (CATATG) and Xho I (CTCGAG). ==== Refs References 1. Portilla D. Crew M.D. Grant D. Serrero G. Bates L.M. Dai G. Sasner M. Cheng J. Buonanno A. cDNA cloning and expression of a novel family of enzymes with calcium-independent phospholipase A2 and lysophospholipase activities J. Am. Soc. Nephrol. 1998 9 1178 1186 9644627 2. Wang A. Yang H.C. Friedman P. Johnson C.A. Dennis E.A. A specific human lysophospholipase: cDNA cloning, tissue distribution and kinetic characterization Biochim. Biophys. Acta 1999 1437 157 169 10.1016/S1388-1981(99)00012-8 10064899 3. Duncan J.A. Gilman A.G. A cytoplasmic acyl-protein thioesterase that removes palmitate from G protein alpha subunits and p21RAS J. Biol. Chem. 1998 273 15830 15837 10.1074/jbc.273.25.15830 9624183 4. Sugimoto H. Hayashi H. Yamashita S. Purification, cDNA cloning, and regulation of lysophospholipase from rat liver J. Biol. Chem. 1996 271 7705 7711 8631810 5. Blaskovic S. Adibekian A. Blanc M. van der Goot G.F. Mechanistic effects of protein palmitoylation and the cellular consequences thereof Chem. Phys. Lipids 2014 180 44 52 10.1016/j.chemphyslip.2014.02.001 24534427 6. Wang A. Dennis E.A. Mammalian lysophospholipases Biochim. Biophys. Acta 1999 1439 1 16 10.1016/S1388-1981(99)00063-3 10395961 7. Moolenaar W.H. Hla T. SnapShot: Bioactive lysophospholipids Cell 2012 148 378 378.e2 10.1016/j.cell.2012.01.013 22265422 8. Sevastou I. Kaffe E. Mouratis M.A. Aidinis V. Lysoglycerophospholipids in chronic inflammatory disorders: The PLA2 /LPC and ATX/LPA axes Biochim. Biophys. Acta 2013 1831 42 60 10.1016/j.bbalip.2012.07.019 22867755 9. Sakai M. Miyazaki A. Hakamata H. Kobori S. Shichiri M. Horiuchi S. Endocytic uptake of lysophosphatidylcholine mediated by macrophage scavenger receptor plays a major role in oxidized low density lipoprotein-induced macrophage growth J. Atheroscler Thromb. 1996 2 81 86 10.5551/jat1994.2.81 9225214 10. Sakai M. Miyazaki A. Hakamata H. Sasaki T. Yui S. Yamazaki M. Shichiri M. Horiuchi S. Lysophosphatidylcholine plays an essential role in the mitogenic effect of oxidized low density lipoprotein on murine macrophages J. Biol. Chem. 1994 269 31430 31435 7989310 11. Ngwenya B.Z. Yamamoto N. Activation of peritoneal macrophages by lysophosphatidylcholine Biochim. Biophys. Acta 1985 839 9 15 10.1016/0304-4165(85)90175-8 3978122 12. Burdzy K. Munder P.G. Fischer H. Westphal O. Increase in the Phagocytosis of Peritoneal Macrophages by Lysolecithin Z. Naturforschung B 1964 19 1118 1120 13. Asaoka Y. Oka M. Yoshida K. Sasaki Y. Nishizuka Y. Role of lysophosphatidylcholine in T-lymphocyte activation: Involvement of phospholipase A2 in signal transduction through protein kinase C Proc. Natl. Acad. Sci. USA 1992 89 6447 6451 10.1073/pnas.89.14.6447 1631142 14. Asaoka Y. Oka M. Yoshida K. Nishizuka Y. Lysophosphatidylcholine as a possible second messenger synergistic to diacylglycerol and calcium ion for T-lymphocyte activation Biochem. Biophys. Res. Commun. 1991 178 1378 1385 10.1016/0006-291X(91)91046-F 1872854 15. Liu-Wu Y. Hurt-Camejo E. Wiklund O. Lysophosphatidylcholine induces the production of IL-1 β by human monocytes Atherosclerosis 1998 137 351 357 10.1016/S0021-9150(97)00295-5 9622278 16. Wang A. Johnson C.A. Jones Y. Ellisman M.H. Dennis E.A. Subcellular localization and PKC-dependent regulation of the human lysophospholipase A/acyl-protein thioesterase in WISH cells Biochim. Biophys. Acta 2000 1484 207 214 10.1016/S1388-1981(00)00020-2 10760470 17. Aicart-Ramos C. Valero R.A. Rodriguez-Crespo I. Protein palmitoylation and subcellular trafficking Biochim. Biophys. Acta 2011 1808 2981 2994 10.1016/j.bbamem.2011.07.009 21819967 18. Iwanaga T. Tsutsumi R. Noritake J. Fukata Y. Fukata M. Dynamic protein palmitoylation in cellular signaling Prog. Lipid Res. 2009 48 117 127 10.1016/j.plipres.2009.02.001 19233228 19. Tsutsumi R. Fukata Y. Noritake J. Iwanaga T. Perez F. Fukata M. Identification of G protein α subunit-palmitoylating enzyme Mol. Cell. Biol. 2009 29 435 447 10.1128/MCB.01144-08 19001095 20. Frenal K. Kemp L.E. Soldati-Favre D. Emerging roles for protein S-palmitoylation in Toxoplasma biology Int. J. Parasitol. 2014 44 121 131 10.1016/j.ijpara.2013.09.004 24184909 21. Hirano T. Kishi M. Sugimoto H. Taguchi R. Obinata H. Ohshima N. Tatei K. Izumi T. Thioesterase activity and subcellular localization of acylprotein thioesterase 1/lysophospholipase 1 Biochim. Biophys. Acta 2009 1791 797 805 10.1016/j.bbalip.2009.05.001 19439193 22. Rocks O. Gerauer M. Vartak N. Koch S. Huang Z.P. Pechlivanis M. Kuhlmann J. Brunsveld L. Chandra A. Ellinger B. The palmitoylation machinery is a spatially organizing system for peripheral membrane proteins Cell 2010 141 458 471 10.1016/j.cell.2010.04.007 20416930 23. Vartak N. Papke B. Grecco H.E. Rossmannek L. Waldmann H. Hedberg C. Bastiaens P.I. The autodepalmitoylating activity of APT maintains the spatial organization of palmitoylated membrane proteins Biophys. J. 2014 106 93 105 10.1016/j.bpj.2013.11.024 24411241 24. Verkruyse L.A. Hofmann S.L. Lysosomal targeting of palmitoyl-protein thioesterase J. Biol. Chem. 1996 271 15831 15836 8663305 25. Boschiroli M.L. Foulongne V. O’Callaghan D. Brucellosis: A worldwide zoonosis Curr. Opin. Microbiol. 2001 4 58 64 10.1016/S1369-5274(00)00165-X 11173035 26. Lin Y. Xiang Z. He Y. Ontology-based representation and analysis of host-Brucella interactions J. Biomed. Semant. 2015 6 37 10.1186/s13326-015-0036-y 26445639 27. Roop R.M. Bellaire B.H. Valderas M.W. Cardelli J.A. Adaptation of the brucella to their intracellular niche Mol. Microbiol. 2004 52 621 630 10.1111/j.1365-2958.2004.04017.x 15101970 28. He Y. Reichow S. Ramamoorthy S. Ding X. Lathigra R. Craig J.C. Sobral B.W. Schurig G.G. Sriranganathan N. Boyle S.M. Brucella melitensis triggers time-dependent modulation of apoptosis and down-regulation of mitochondrion-associated gene expression in mouse macrophages Infect. Immun. 2006 74 5035 5046 10.1128/IAI.01998-05 16926395 29. Celli J. de Chastellier C. Franchini D.M. Pizarro-Cerda J. Moreno E. Gorvel J.P. Brucella evades macrophage killing via VirB-dependent sustained interactions with the endoplasmic reticulum J. Exp. Med. 2003 198 545 556 10.1084/jem.20030088 12925673 30. Pei J. Wu Q. Kahl-McDonagh M. Ficht T.A. Cytotoxicity in macrophages infected with rough Brucella mutants is type IV secretion system dependent Infect. Immun. 2008 76 30 37 10.1128/IAI.00379-07 17938217 31. Fernandez-Prada C.M. Zelazowska E.B. Nikolich M. Hadfield T.L. Roop R.M. II Robertson G.L. Hoover D.L. Interactions between Brucella melitensis and human phagocytes: Bacterial surface O -Polysaccharide inhibits phagocytosis, bacterial killing, and subsequent host cell apoptosis Infect. Immun. 2003 71 2110 2119 10.1128/IAI.71.4.2110-2119.2003 12654833 32. Gross A. Terraza A. Ouahrani-Bettache S. Liautard J.P. Dornand J. In vitro Brucella suis infection prevents the programmed cell death of human monocytic cells Infect. Immun. 2000 68 342 351 10.1128/IAI.68.1.342-351.2000 10603407 33. Buchmeier N.A. Heffron F. Inhibition of macrophage phagosome-lysosome fusion by Salmonella typhimurium Infect. Immun. 1991 59 2232 2238 2050395 34. Eskra L. Mathison A. Splitter G. Microarray analysis of mRNA levels from RAW264.7 macrophages infected with Brucella abortus Infect. Immun. 2003 71 1125 1133 10.1128/IAI.71.3.1125-1133.2003 12595423 35. Rossetti C.A. Drake K.L. Adams L.G. Transcriptome analysis of HeLa cells response to Brucella melitensis infection: A molecular approach to understand the role of the mucosal epithelium in the onset of the Brucella pathogenesis Microbes Infect. 2012 14 756 767 10.1016/j.micinf.2012.03.003 22484383 36. Satou M. Nishi Y. Yoh J. Hattori Y. Sugimoto H. Identification and characterization of acyl-protein thioesterase 1/lysophospholipase I as a ghrelin deacylation/lysophospholipid hydrolyzing enzyme in fetal bovine serum and conditioned medium Endocrinology 2010 151 4765 4775 10.1210/en.2010-0412 20685872 37. Wang A. Deems R.A. Dennis E.A. Cloning, expression, and catalytic mechanism of murine lysophospholipase I J. Biol. Chem. 1997 272 12723 12729 10.1074/jbc.272.19.12723 9139730 38. Habermann E. Hardt K.L. A sensitive and specific plate test for the quantitation of phospholipases Anal. Biochem. 1972 50 163 173 10.1016/0003-2697(72)90495-2 4342994 39. Wang A. Loo R. Chen Z. Dennis E.A. Regiospecificity and catalytic triad of lysophospholipase I J. Biol. Chem. 1997 272 22030 22036 10.1074/jbc.272.35.22030 9268342 40. Gonzalez-Morales L. Diego-Garcia E. Segovia L. Gutierrez Mdel C. Possani L.D. Venom from the centipede Scolopendra viridis Say: Purification, gene cloning and phylogenetic analysis of a phospholipase A2 Toxicon 2009 54 8 15 10.1016/j.toxicon.2009.03.003 19285520 41. Ponce D. Lopez-Vera E. Aguilar M.B. Sanchez-Rodriguez J. Preliminary results of the in vivo and in vitro characterization of a tentacle venom fraction from the jellyfish Aurelia aurita Toxins (Basel) 2013 5 2420 2433 10.3390/toxins5122420 24322597 42. Fortes-Dias C.L. Ortolani P.L. Fernandes C.A. Lobo K.R. Amaral de Melo L. Borges M.H. Pazin W.M. Neto Mde O. Fernandez R.M. Fontes M.R. Insights on the structure of native CNF, an endogenous phospholipase A2 inhibitor from Crotalus durissus terrificus, the South American rattlesnake Biochim. Biophys. Acta 2014 1844 1569 1579 10.1016/j.bbapap.2014.05.001 24820993 43. Gomes H.L. Andrich F. Fortes-Dias C.L. Perales J. Teixeira-Ferreira A. Vassallo D.V. Cruz J.S. Figueiredo S.G. Molecular and biochemical characterization of a cytolysin from the Scorpaena plumieri (scorpionfish) venom: Evidence of pore formation on erythrocyte cell membrane Toxicon 2013 74 92 100 10.1016/j.toxicon.2013.07.023 23933196 44. Lino R.C. Martins F.I. Florentino I.F. Nascimento M.V. Galdino P.M. Andrade C.H. Rezende K.R. Menegatti R. Costa E.A. Anti-inflammatory effect of (E)-4-(3,7-dimethylocta-2,6-dienylamino)phenol, a new derivative of 4-nerolidylcatechol J. Pharm. Pharmacol. 2013 65 133 141 10.1111/j.2042-7158.2012.01582.x 23215696 45. Nie L. Wu G. Zhang W.W. Correlation between mRNA and protein abundance in Desulfovibrio vulgaris: A multiple regression to identify sources of variations Biochem. Biophys. Res. Commun. 2006 339 603 610 10.1016/j.bbrc.2005.11.055 16310166 46. Tian Q. Stepaniants S.B. Mao M. Weng L. Feetham M.C. Doyle M.J. Yi E.C. Dai H. Thorsson V. Eng J. Integrated genomic and proteomic analyses of gene expression in Mammalian cells Mol. Cell. Proteom. 2004 3 960 969 10.1074/mcp.M400055-MCP200 15238602 47. Dorneles E.M. Teixeira-Carvalho A. Araujo M.S. Sriranganathan N. Lage A.P. Immune response triggered by Brucella abortus following infection or vaccination Vaccine 2015 33 3659 3666 10.1016/j.vaccine.2015.05.057 26048781 48. Liu N.N. Liu Z.S. Lu S.Y. Hu P. Li Y.S. Feng X.L. Zhang S.Y. Wang N. Meng Q.F. Yang Y.J. Full-length cDNA cloning, molecular characterization and differential expression analysis of peroxiredoxin 6 from Ovis aries Vet. Immunol. Immunopathol. 2015 164 208 219 10.1016/j.vetimm.2015.01.006 25712755 49. Schmitz G. Ruebsaamen K. Metabolism and atherogenic disease association of lysophosphatidylcholine Atherosclerosis 2010 208 10 18 10.1016/j.atherosclerosis.2009.05.029 19570538 50. Graler M.H. Goetzl E.J. Lysophospholipids and their G protein-coupled receptors in inflammation and immunity Biochim. Biophys. Acta 2002 1582 168 174 10.1016/S1388-1981(02)00152-X 12069825 51. Feng X.L. Lu S.Y. Liu D. Li L. Wu X.Z. Song J. Hu P. Li Y.S. Tang F. Li Z.H. Direct competitive immunosorbent assay for detection of MEHP in human urine Chemosphere 2013 92 150 155 10.1016/j.chemosphere.2013.02.062 23561568 52. NCBI Available online: http://www.ncbi.nlm.nih.gov (accessed on 30 March 2016) 53. Compute pI/Mw tool Available online: http://web.expasy.org/compute_pi/ (accessed on 30 March 2016) 54. Motif Scan Available online: http://hits.isb-sib.ch/cgi-bin/PFSCAN (accessed on 30 March 2016) 55. Lobo de Araujo A. Radvanyi F. Determination of phospholipase A2 activity by a colorimetric assay using a pH indicator Toxicon 1987 25 1181 1188 10.1016/0041-0101(87)90136-X 3433293 56. Ben Bacha A. Al-Daihan S.K. Mejdoub H. Purification, characterization and bactericidal activities of phospholipase A2 from the dromedary intestine Int. J. Biol. Macromol. 2013 57 156 164 10.1016/j.ijbiomac.2013.03.020 23500664 57. Abousalham A. Verger R. Egg yolk lipoproteins as substrates for lipases Biochim. Biophys. Acta 2000 1485 56 62 10.1016/S1388-1981(00)00033-0 10802249 58. Yang Y.J. Liu Z.S. Lu S.Y. Li C. Hu P. Li Y.S. Liu N.N. Tang F. Xu Y.M. Zhang J.H. Molecular cloning, expression and characterization of programmed cell death 10 from sheep (Ovis aries ) Gene 2015 558 65 74 10.1016/j.gene.2014.12.040 25541025 59. Zhou Y. Zhang Y.Y. Shen Q.F. Lu S.Y. Ren H.L. Li Y.S. Liu Z.S. Pan F.G. Meng X.M. Zhang J.H. Development of a novel antibody probe useful for domoic acid detection Biosens. Bioelectron. 2009 24 3159 3163 10.1016/j.bios.2009.03.023 19376694 60. Bradford M.M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding Anal. Biochem. 1976 72 248 254 10.1016/0003-2697(76)90527-3 942051 61. Tang F. Ren H.L. Xu Y.M. Zou D.Y. Liu N.N. Li Y.S. Zhou Y. Song J. Li Z.H. Zhang Y.Y. Kinetics and cross-reactivity of the antibody in sheep inoculated with virulent and avirulent Brucella J. Anim. Vet. Adv. 2012 11 1564 1569 10.3923/javaa.2012.1564.1569

PMC005xxxxxx/PMC5000605.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081207ijms-17-01207ArticleChicken-Specific Kinome Array Reveals that Salmonella enterica Serovar Enteritidis Modulates Host Immune Signaling Pathways in the Cecum to Establish a Persistence Infection Kogut Michael H. 1Swaggerty Christina L. 1Byrd James Allen 1Selvaraj Ramesh 2Arsenault Ryan J. 3*Woo Patrick C. Y. Academic Editor1 Southern Plains Agricultural Resarch Center, United States Department of Agriculture, Agricultural Research Service, College Station, TX 77845, USA; [email protected] (M.H.K.); [email protected] (C.L.S.); [email protected] (J.A.B.)2 Ohio Agricultural Research Center, The Ohio State University, Wooster, OH 44691, USA; [email protected] Department of Animal and Food Sciences, University of Delaware, Newark, DE 19716, USA* Correspondence: [email protected]; Tel.: +1-302-831-757927 7 2016 8 2016 17 8 120721 5 2016 08 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Non-typhoidal Salmonella enterica induces an early, short-lived pro-inflammatory response in chickens that is asymptomatic of clinical disease and results in a persistent colonization of the gastrointestinal (GI) tract that transmits infections to naïve hosts via fecal shedding of bacteria. The underlying mechanisms that control this persistent colonization of the ceca of chickens by Salmonella are only beginning to be elucidated. We hypothesize that alteration of host signaling pathways mediate the induction of a tolerance response. Using chicken-specific kinomic immune peptide arrays and quantitative RT-PCR of infected cecal tissue, we have previously evaluated the development of disease tolerance in chickens infected with Salmonella enterica serovar Enteritidis (S. Enteritidis) in a persistent infection model (4–14 days post infection). Here, we have further outlined the induction of an tolerance defense strategy in the cecum of chickens infected with S. Enteritidis beginning around four days post-primary infection. The response is characterized by alterations in the activation of T cell signaling mediated by the dephosphorylation of phospholipase c-γ1 (PLCG1) that inhibits NF-κB signaling and activates nuclear factor of activated T-cells (NFAT) signaling and blockage of interferon-γ (IFN-γ) production through the disruption of the JAK-STAT signaling pathway (dephosphorylation of JAK2, JAK3, and STAT4). Further, we measured a significant down-regulation reduction in IFN-γ mRNA expression. These studies, combined with our previous findings, describe global phenotypic changes in the avian cecum of Salmonella Enteritidis-infected chickens that decreases the host responsiveness resulting in the establishment of persistent colonization. The identified tissue protein kinases also represent potential targets for future antimicrobial compounds for decreasing Salmonella loads in the intestines of food animals before going to market. Salmonellakinomeinterferon-γphospholipase cJAK-STAT pathway ==== Body 1. Introduction The Centers for Disease Control and Prevention continues to address multistate foodborne outbreaks that have impacted the health of the nation over the last 10 years [1]. One area of concern is the need for reduction of Salmonella as a foodborne pathogen. Despite control efforts that cost over a half a billion dollars annually, foodborne illnesses due to Salmonella continue to impact the consumer. Poultry are commonly identified as a major source of Salmonella. Asymptomatic carrier states are poorly understood. “Normal infections” include infection of chicks through an oral route and is characterized by a translocation through the intestinal epithelial cells followed with a splenic infection [2]. While asymptomatic carriers can be infected by Salmonella Enteritidis (SE) and Salmonella Typhmurium (ST), these bacteria can survive in the gastrointestinal tract of birds for months without showing clinical signs [3]. These Salmonella carriers have an infected gastrointestinal tract without showing clinical signs while excreting high concentrations of Salmonella into the environment [3,4,5,6,7,8,9,10]. These healthy carriers can be a risk to affect other birds by horizontal transmission or affect newly hatched chicks. Despite the importance of Salmonella as a human pathogen, relatively little is known about the host immune response or virulence mechanisms of persistent asymptomatic infections in the avian intestine. The most fundamental question to answer is how do these organisms manage to escape clearance for so long in the presence of the host immune response? Upon infection with Salmonella, an up-regulation of the innate inflammatory response is generated and is characterized by pro-inflammatory cytokines and granulocyte (heterophils in chickens) influx within hours [11,12,13,14]. Yet, this intestinal inflammatory response is somehow dampened facilitating pathogen survival and persistent infection [15] for up to 10 weeks or more [3,16]. One must keep in mind that this dampened inflammatory response in chickens may be a host-developed mechanism to minimize immune-mediated damage to the intestine at a time when the gut microbiome is being established (disease tolerance). Disease tolerance has recently been described as a “distinct host defense strategy” [17,18,19,20,21,22]. Thus, a diminished immune response provides conceivable advantages to both the host and bacterium during a persistent infection in chickens. The mechanisms involved in this down-regulation of the mucosal immune response are currently unknown. However, one can speculate that mucosal disease tolerance is required to establish a persistent infection. During host tolerance, defined as coping with a pathogenic encounter without a consequent reduction in health [17,18,19,20,21,22], the host’s strategy is to avoid a harmful excessive inflammatory response [23,24]. However, this strategy may enable pathogen persistence, such as that observed with Salmonella infections of poultry [15,16]. We, and others, have recently demonstrated the development of a Th2, anti-inflammatory response in the cecum of chickens that begins at least four days after an initial infection with Salmonella and continues for weeks [11,25,26,27,28]. Moreover, we have noted a significant increase in CD4+CD25+ (T regulatory) cells in the cecum that corresponds to this shift from a pro-inflammatory to an anti-inflammatory environment [29]. T regulatory cells (Tregs) have been linked to play crucial roles sustaining a balance between the host immune response and immunological tolerance in many infections in mammals [30,31,32,33,34]. Further, a role for Tregs during a persistent Salmonella infection was recently described using a mouse model of persistence [35]. More recently, we also found alterations in the tissue phenotype of the cecum of the Salmonella-infected animals that is distinguished by metabolic signatures indicative of metabolic reprogramming with a shift from anabolic to catabolic reactions [28]. It is during this phase that we speculate that Salmonella takes advantage of a reduction of host response to infection to begin to establish a persistent cecal colonization [28]. In this study, our hypothesis was that Salmonella enterica serovar Enteritidis (S. Enteritidis) induces a disease tolerance host defense mechanism in chickens that allows the bacteria to colonize persistently the cecum of poultry. To test the hypothesis, we analyzed a time-course of chicken-specific kinomic immune changes and interferon-γ (IFN-γ) mRNA transcription in avian cecal tissue during a persistent infection by S. Enteritidis. Using these techniques, we were able to identify specific phosphorylation based immune post-translational signaling changes during a chronic Salmonella colonization in chickens that provide confirmation for the transition from an early mucosal pro-inflammatory response to the development of a disease tolerant mucosal response. 2. Results 2.1. S. Enteritidis Infection Infection state of the chickens was confirmed by culturing the cecal contents and feces from each bird for S. Enteritidis with and without enrichment. Cultures showed that at least 75% of the chickens in the inoculated group were culture positive for S. Enteritidis throughout the experiment while Salmonella was never isolated from the birds in the control group at any given time point (Table 1 and Table 2). Four birds from each group at each time point were selected, infected birds were selected based on a consistent high level of S. Enteritidis colonization. 2.2. Kinome Arrays Chicken-specific kinome arrays custom-designed for the study of chicken immune signal transduction pathways were used [36]. Kinome analysis was carried out on the cecal samples from non-infected and infected chickens. The results from four animals from each group (S. Enteritidis-infected and non-infected) and time point were combined to provide a representative result. To remove any non-specific or baseline phosphorylation signal from the analysis data from each time point was corrected using the matched uninfected controls. The kinome data were subjected to pathway overrepresentation analysis to determine which cellular pathways/processes are activated under the infected and non-infected conditions. To ensure that the identified pathways represent conserved and consistent biological responses, input data were limited to peptides with a consistent pattern of differential phosphorylation across the four biological replicates in at least one of the treatment sets as well as significant changes (p ≤ 0.05) in phosphorylation level relative to the non-infected control treatment. These select data from the four animals were merged to generate a representative data set for each treatment condition. All peptides that showed significant phosphorylation changes relative to control (p ≤ 0.05) for each time point were input into the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database [37]. Using STRING functionality, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway results were generated for each dataset. The STRING generated KEGG pathway results showed a number of pathways altered by the S. Enteritidis infection at a statistically significant level (p ≤ 0.05 false discovery rate (FDR) corrected). Of particular interest were those pathways that contained peptides that were significantly differentially phosphorylated at multiple times over the course of the study. A subset of these pathways are shown in Table 3. Of particular note, the T cell signaling and JAK-STAT signaling pathways were dramatically altered by the infection. Both these pathways had multiple peptide phosphorylation events altered at multiple time points post-infection. In total 49 differentially phosphorylated peptides were observed in chickens within these two different pathways on the 4th day post-infection with S. Enteritidis (Table 3) signifying a dramatic local post-translational modification of the proteins within the infected cecum. Of the 49 peptides that were differentially phosphorylated, 26 belong to the T cell signaling pathway, and 23 to the JAK-STAT pathway. Only 33 total peptides were found to be differentially phosphorylated over days 7–14 post-infection within these two specific pathways (Table 3). 2.3. Phosphoryaltion Events within Specific Pathways 2.3.1. T Cell Receptor Signaling Pathway The transcription factor family Nuclear factor of activated T-cells (NFAT), play a crucial role in regulating the transcription of cytokines and other genes critical for immune response. Members of the NFAT family were found to be significantly phosphorylated in the ceca of S. Enteritidis-infected chickens (Table 4). This is a significant finding because inactivated NFAT proteins in the cytoplasm of a cell are in their phosphorylated form. Following T cell receptor (TCR) stimulation, cytoplasmic NFAT proteins are dephosphorylated and translocate from the cytoplasm to the nucleus where they regulate transcription of key cytokine genes. Thus, based on the findings here the increased phosphorylation of NFAT inactivates the proteins preventing its translocation to the nucleus and thus decreasing pro-inflammatory cytokine production. Simultaneously, we found that Iκκ-β, NF-κB1 and NF-κB1A were significantly dephosphorylated after 4–10 days of infection with S. Enteritidis (Table 4). NF-κB is a transcription factor that is phosphorylated when activated by various intra- and extra-cellular stimuli then translocates into the nucleus and stimulates the expression of genes involved in a variety of immune functions. Further analysis of the T cell receptor signaling pathway revealed two other significant changes in phosphorylation events: (1) phospholipase C-γ1 (PLCG1) was significantly dephosphorylated in the S. Enteritidis-infected cecal tissue at four days post-infection when compared to the non-infected control cecal tissue; and (2) a significant dephosphorylation of MAPKs, including MEK1, ERK1, MAP3K8, and p38. 2.3.2. JAK-STAT Signaling Pathway The JAK-STAT signaling cascade is represented quite comprehensively on the kinome array, it is possible to investigate the effects of a persistent cecal infection by S. Enteritidis on the principle signaling mechanism for a wide variety of cytokines and growth factors. At 4 days-post-infection, a differentiated series of phosphorylation events occurred at the receptor level in the infected birds when compared to the non-infected control birds (Table 5). First, a significant increase in phosphorylation of the IFN-α receptor (IFNAR1; p ≤ 0.003), and IL-2 receptor IL-2RB; (p ≤ 0.0003) were found. Simultaneously, there is a significant decrease on the phosphorylation of the IL-4 receptor (IL-4R; p ≤ 0.006), IFN-γ receptor (IFNGR1; p ≤ 0.006), IL-6 receptor gp130; (p ≤ 0.01), and IL-7 receptor (IL-7R; p ≤ 0.0008). Lastly, there was a five-fold increase in the phosphorylation of the IL-10 receptor (IL-10R-A; p ≤ 0.02) at four days post-infection was also elevated at 10 days post-infection (Table 5). Furthermore, the development of a persistent cecal infection in chickens by S. Enteritidis also appears to target the JAK kinases for degradation. However, JAK2 and JAK3 appeared to be targets for dephosphorylation where both had a three-fold decrease in phosphorylation at four days post-infection (Table 5). JAK3 appears to be a specific target since the dephosphorylation continued through day 10 post-infection where an 18-fold decrease was observed. Lastly, the persistent infection by S. Enteritidis also appeared to target the specific JAK substrate STAT4 (Table 5). We measured a significant two-fold (four days) to seven-fold (10 days) decrease in the phosphorylation of the STAT4 transcription factor in the ceca of the infected birds when compared to the cecal tissues from the non-infected birds. STAT1, 3, 5B and 6 had increased phosphorylation on day four post-infection, but all had a reduced phosphorylation 10–14 days post-infection. 2.4. Validation of Kinome Analysis with Antibody Array An often used methodof validating kinome peptide array data is by using phosphospecific antibodies. For example, performing a Western blot using phosphospecific antibodies that correspond to the phosphosites on the peptide array. If the phosphospecific antibody binds and the peptide array shows the same phosphosite has been phosphorylated there is confirmation of the array data. This type of validation is similar to how transcriptome data from a cDNA microarray is validated through the use of quantitative real-time PCR. In a variation of the standard validation procedure we chose to employ an antibody microarray, which contains many phosphospecific antibodies immobilized in an array format [27]. Though there is a scarcity of chicken specific antibodies, many of the central proteins of interest found in the peptide array results were relatively well conserved between humans and chickens, providing confidence that there would be significant observed binding through cross-reactivity of the antibodies. To illustrate the conservation of phosphosites the percent orthology between the chicken and human 15 amino acid phosphorylation target sites determined by NCBI Protein Blast analysis is shown in Table 6. Following the data normalization, the results showed similar peptide phosphorylation events to those observed with the peptide arrays (Table 6). 2.5. Altered Expression of IFN-γ Transcription As has been reported previously, during the early acute infection (within 24 h) by paratyphoid strains of Salmonella chickens up-regulate pro-inflammatory cytokines mRNA expression in the cecum [9,38,39,40]. In the present studies, we profiled the IFN-γ mRNA expression in the cecum of chickens 2, 4, 7, 10, and 14 days post-infection with S. Enteritidis and compared the results to the non-infected control birds. IFN-γ mRNA expression in the S. Enteritidis infected ceca from chickens was up-regulated two to seven days post-infection when compared to the non-infected birds expression in the cecum (Figure 1). However, there was a significant and dramatic nine-fold decrease in IFN-γ mRNA expression from day two post-infection to day four post-infection. The fold-change in IFN-γ mRNA expression remained unchanged through day 14 post-infection (Figure 1). 3. Discussion Relatively little is known about how and why Salmonella enterica persist in the avian intestine, specifically the interactions between the virulence mechanisms and host immune response. The persistent colonization of the gut, the carrier state, is established, and the Salmonella is able to stay in the ceca for months without triggering clinical signs of infection [4,5,7,8]. A persistent, chronic, subclinical Salmonella infection of the intestinal tract is important to continued bacterial propagation and the contamination of poultry as it is nearly impossible to detect and isolate infected birds [41]. We, and others, have speculated that the bacterium is involved in redirecting, or subverting, the host response toward disease tolerance [11,26,27,28,42]. The present study was designed to address the question of immune tolerance induction during a persistent paratyphoid Salmonella infection in chickens. Host responses to infectious agents are often regulated through phosphorylation. However, proteomic mechanisms of Salmonella acute infection biology and host responses to the bacteria have been investigated only in murine models [43,44,45,46,47]. Until recently, studies in poultry have been limited to the genomic responses of the host to infection (reviews in [11,48,49]). Our recent development of chicken-specific peptide arrays for kinome analysis of host phosphorylation-based cellular signaling responses provided us with the opportunity to develop a more detailed understanding of the chicken host-pathogen interactions with Salmonella [50,51]. Based on the findings here, our kinomic analysis demonstrate a phenotypic change in the avian cecum as it orchestrates the dynamics of immune signaling pathways, cytokine secretion, transcription factor expression, and the launch of a different immune microenvironment during the establishment of a persistent Salmonella infection and a return to intestinal homeostasis. Four days post-infection (pi) Salmonella induces an immune transition from an acute pro-inflammatory response to an established infection and a dampened or eliminated innate response [52]. By 4 days pi, we have described a substantial down-regulation of the expression pro-inflammatory cytokines that coincides with the up-regulation of the expression of anti-inflammatory cytokines [27,28]. Further, by day four pi a dramatic increase in Tregs (CD4+CD25+) in the cecum and remains elevated through 14 days pi [29]. This coordinated production of pro- versus anti-inflammatory responses is a central mechanism of an effective early inflammatory response and later return to tissue immune homeostasis. Finally, we used a kinomics approach to uncover the mechanisms used by S. Enteritidis to impact the avian inflammatory responses and determine host signaling events altered by the bacteria to create the conditions for a persistent infection. Our results identified multiple changes to the host kinome during the establishment of a persistent Salmonella infection in the avian cecum. This immune analysis that compared the immune responses between the S. Enteritidis-infected avian cecum and non-infected cecum provides novel information on host cellular signaling cascades that are altered during the establishment of Salmonella persistence (Table 5 and Table 6) [27,28]. Additionally, the relative lack of differential phosphorylation events found in the signaling pathways between the infected and non-infected ceca 7–14 days pi indicate that a level of immune homeostasis had been achieved and that the Salmonella were no longer being recognized as infectious agents and were now part of the commensal population. Further experiments are underway to further characterize and contrast this homeostasis to that of the non-infected controls. Here we have further described a series of phosphorylation-mediated changes in the ceca of chickens during the development of a persistent Salmonella infection. The most significant differences in host immune kinase activities in infected animals occurred within four days pi. These changes were localized to select pathways, specifically the T cell receptor and the JAK-STAT signaling pathways, which were altered by the persistent colonization of the cecum by S. Enteritidis. Stimulation of the T cell receptor results in the activation of the TCR signal transduction pathway. This pathway activates the transcription factors nuclear factor κB (NF-κB), nuclear factor of activated T-cells (NFAT), and activator protein 1 (AP-1), that induce expression of cytokine genes [53]. The results of this study clearly point to changes in the activity of all three of the central transcription factors, specifically at 4 days pi (Table 4). First, we found no significant effect on the phosphorylation of the AP-1 transcription factors between the S. Enteritidis-infected and non-infected tissues. The AP-1 pathway is dependent on activated of mitogen-activated protein kinases (MAPKs), such as extracellular signal–regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38, which promote the synthesis, phosphorylation, and activation of the Fos and Jun proteins that together comprise the AP-1 transcription factor [53]. However, we found a significant dephosphorylation of MAPKs, including MEK1, ERK1, MAP3K8, and p38 involved in the T cell receptor signaling cascade; thus, pointing to the lack of involvement of AP-1-induced genes during a persistent S. Enteritidis infection in the chicken (Table 4). Second, NFAT (phosphorylated) and NF-κB (dephosphorylated) were significantly differentially phosphorylated in the ceca of S. Enteritidis-infected chickens (Table 4). The central question is whether there is a common thread that could account for this differential response of these transcription factors. This thread appears to be phospholipase C-γ1 (PLCG1) that was significantly dephosphorylated in the S. Enteritidis-infected cecal tissue at 4 days pi when compared to the non-infected control cecal tissue. Activation of both NF-κB and NFAT requires the activity of PLC-γ1, which generates the second messengers diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP3). DAG leads to activation of protein kinase C θ (PKCθ), which, in turn, activates the inhibitor of κB (IκB) kinase (IKK) complex, resulting in the phosphorylation and degradation of IκBα and the translocation of the NF-κB p50:p65 heterodimer to the nucleus [54]. IP3 induces an increase in the concentration of cytoplasmic calcium (Ca2+) and activation of the Ca2+-dependent phosphatase calcineurin, which results in the rapid activation of NFAT, which is followed by its translocation to the nucleus [55]. To our knowledge, our report is the first to implicate Salmonella targeting of PLCG1 to manipulate the NF-κB and NFAT pathways to inhibit pro-inflammatory responses. What bacterial factors may be involved in dephosphorylating PLCG1 are unknown at this time and will be the focus of future experiments. NFAT proteins, a family of transcription factors, are critical to the transcription of cytokine genes and other genes that are critical for the control of inflammation and regulation of the immune response [55,56]. Further, NFAT must ultimately bind to additional transcription factors, such as AP-1 to form transcriptional complexes that regulate gene expression that are inducibly transcribed by immune-system cells [55,57]. NFAT functions to regulate the interaction of the innate immune cells with acquired immunity and to promote anti-inflammatory programs (reviewed by [58]). Thus, the increased phosphorylation of NFAT peptides would suggest the initiation of anti-inflammatory signals. NF-κB is a transcription factor whose activity is triggered in response to infectious agents and pro-inflammatory cytokines via the IκB kinase (IKK) complex and plays a key role in regulating the pro-inflammatory response [59,60]. Therefore, dephosphorylation of both IKK and NF-κB would result in a down-regulation of pro-inflammatory cytokines as we observed in the present experiments. As a result, the T cell receptor signaling pathway analysis data provide evidence that the establishment of a persistent infection by S. Enteritidis in the avian cecum appears to be partially due to the targeting of signaling cascades that inhibit the transcription of pro-inflammatory responses and induce the beginning of a transition from TH1/TH17 cells to the development of Tregs [61,62,63,64]. Further pathway analysis of the kinome data indicated differential phosphorylation of the JAK-STAT pathway, a signaling cascade that provides a direct mechanism to translate an extracellular signal into a transcriptional response, in S. Enteritidis-infected cecal tissue. The JAK-STAT system consists of three main components: (1) a receptor; (2) Janus kinase (JAK); and (3) Signal Transducers and Activator of Transcription (STAT) [65]. Based on the results from them present experiments (Table 6), the IFN-α, IL-2, IL-4, and IL-10 receptors were phosphorylated; whereas, IFN-γ, IL-7, and IL-6 cytokine family (gp130) receptors were dephosphorylated. IFN-γ is characteristic of a Th1 response whereas IL-4 is a signature cytokine of Th2 responses. IL-7 is involved in early T cell development and IL-6 is a pro-inflammatory cytokine involved in stimulating an immune response during infection. IL-2 is normally produced by T cells during an immune response and involved in growth, proliferation, and differentiation of T cells to become “effector” T cells [66,67]. When combined with the cytokine expression observed previously [27,28], the down-regulation of IFN-γ mRNA transcription shown here (Figure 1) provides a clear pattern of down-regulation of the pro-inflammatory cytokines (IL-6, IL-1β, IFN-γ) and an up-regulation of anti-inflammatory cytokines, IL-10 and TGF-β4 [27,28]. We speculate there is a profound immune transition from an active inflammatory response where the immune system was working to reduce the number of bacteria to an environment of homeostasis where the immune response is allowing for a persistent state of infection in the S. Enteritidis-infected cecal tissue. Cytokine receptor proteins lack enzymatic activity, thus are dependent upon JAKs to initiate signaling upon binding of their ligands. The JAK family has four members: JAK1, JAK2, JAK3 and tyrosine kinase 2 (TYK2) [68]. TYK2 is the only JAK family member that was activated (phosphorylated) in the ceca from the S. Enteritidis-infected chickens (Table 6). Although primarily involved in IL-12 and type I-IFN signaling, TYK2 is activated by IL-10 [69]. Most importantly, based on these experiments, the development of a persistent cecal infection by S. Enteritidis triggers a dephosphorylation of both JAK2 and JAK3 proteins (Table 6). JAK2 is an essential tyrosine kinase for modulating the immune response and whose activation contributes to the severe inflammatory response in sepsis [70,71]. Inhibition of JAK2 prevents NF-κB activation; thus “rescuing” mice from polymicrobial sepsis [72]. Therefore, we can conclude that the dephosphorylation of both JAK2 and NF-κB found via our kinomic analysis is indicative of a negative regulation of a pro-inflammatory response; in this case brought about by the establishment of a persistent Salmonella infection. Further experiments are required to confirm this hypothesis. JAK3 is predominantly expressed in hematopoietic lineage such as NK cells, T cells and B cells and intestinal epithelial cells [73,74,75]. JAK3 is the only JAK family member involved in all phases of T cell biology: development, proliferation, and differentiation [76,77,78]. For T cell differentiation, JAK3, along with IL-4, steer Th2 cell differentiation [78], but inhibition of JAK3 generates the induction of Tregs [79,80]. Therefore, the dephosphorylation of JAK2 and JAK3 found in the present studies would result in a change in the functional immune phenotype of the cecal environment that benefits the establishment of a tolerant mucosal immune response against the bacterial colonization. Although these studies cannot confirm what provoked this dephosphorylation of JAK2 and JAK3, we speculate that the mechanism is a specific action of the Salmonella organism as it begins to establish its long term colonization. These results are the first to infer that Salmonella have evolved a time-dependent strategy that blocks responsiveness of the JAK proteins that down-regulates the host response to infection. STAT4 is a decisive factor in host resistance to a variety of viral, bacterial, and protozoan pathogens while serving as the central regulator of IFN-γ production during inflammation [81]. Intestinal IFN-γ mRNA expression levels are a prevailing indicator of a reduced immune response associated with persistence of Salmonella in the chicken gastrointestinal tract [42]. Furthermore, the ratio between STAT1 and STAT4 are crucial for IFN-γ production during viral and Salmonella infections [82,83]. Herein, we found a reduced IFN-γ mRNA expression during the establishment of the persistent Salmonella infection (Figure 1) and an increased phosphorylation of STAT1 and dephosphorylation of STAT4 (Table 6). Our results are in agreement with two recent studies where N-ethyl-N-nitrosourea-induced mutations of mice resulted in increased STAT1 phosphorylation, suppressed STAT4 expression, and altered IFN-γ production that led to the increased susceptibility of the animals to S. Typhimurium infection [83,84]. IFN-γ has been shown to play a fundamental role in the resolution of intestinal Salmonella infection [13,42,85]. Further, our observation of a dramatic decrease in IFN-γ mRNA expression at day four p.i. is in agreement with previously reported results by other laboratories [13,26]. 4. Materials and Methods 4.1. Experimental Animals Experiments were conducted according to the regulations established by the United States Department of Agriculture Animal Care and Use Committee. Broiler chickens used in this study were obtained from a commercial breeder and were all of the same genetic background and were not vaccinated at any time. Chicks were placed in floor pens containing wood shavings, provided supplemental heat, water, and a balanced, unmedicated corn and soybean meal-based chick starter diet ad libitum that met or exceeded the levels of critical nutrients recommended by the National Research Council [86]. Salmonella was not detected in the feed or from the paper tray liners. 4.2. S. Enteritidis Challenge A poultry isolate of Salmonella enterica serovar Enteritidis (S. Enteritidis; (ID 9711771, part 24)) was obtained from the National Veterinary Services Laboratory (Ames, IA, USA), and was selected for resistance to nalidixic acid and novobiocin and maintained in tryptic soy broth (Difco Laboratories, Sparks, MD, USA) containing antibiotics (20 µg/mL nalidixic acid and 25 µg/mL novobiocin; Sigma Chemical Co.; St. Louis, MO, USA). A stock culture was prepared in sterile PBS and adjusted to a concentration of 1 × 109 colony forming units (CFU/mL). The viable cell concentration of the challenge dose for each experiment was determined by colony counts on XLT4 agar base plates with XLT4 supplement (Difco) and nalidixic acid and novobiocin (XLT-NN). 4.3. Experimental Design One-day-old broiler chickens were randomly distributed into two experimental groups: non-infected control and infected. Each group contained 200 birds fed a balanced, unmedicated corn and soybean meal-based diet. Four days post-hatch, all chickens were orally challenged with 1 mL of either 5 × 106 CFU/mL S. Enteritidis or mock challenged with 1 mL sterile PBS. Four, 7, 10, and 14 days after challenge, 50 chickens from each group were killed by cervical dislocation, cecal contents were analyzed for S. Enteritidis colonization, 10 of these chickens were used for: (a) cecal tonsils for quantitative real-time PCR (qRT-PCR); and (b) cecal tissue was flash frozen in liquid nitrogen and stored for use in the peptide and antibody arrays. All experiments were conducted three times. Therefore, the ceca from a total of 30 chickens for each of the 2 groups (10 chickens each in 3 experiments) were used to prepare the mRNA for the qRT-PCR IFN-γ assay described below. RNA from each bird (n = 10) was isolated and assayed separately and not pooled. Each RNA sample was replicated 3 times for IFN-γ expression per experiment. 4.4. Sample Collection for Peptide and Antibody Arrays At 4, 7, 10, and 14 days post infection, both ceca were removed from each of 10 birds from each group (non-infected and infected) and immediately flash frozen in liquid nitrogen to preserve kinase enzymatic activity. Samples were taken from liquid nitrogen and transferred to a −80 °C freezer until further experimental procedures were conducted. 4.5. Kinome Array At each of the time points and under each condition (infected and uninfected), 4 cecal samples from 4 different animals were taken from storage for analysis (32 samples total). Infected birds were selected based on a consistent high level of S. Enteritidis colonization. Cecal tissue samples were weighed to obtain a consistent 40 mg sample for the array protocol. Samples were homogenized by a hand-held Qiagen TissueRuptor (Valencia, CA, USA) in 100 μL of lysis buffer (20 mM Tris–HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM Ethylene glycol tetraacetic acid (EGTA), 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM Na3VO4, 1 mM NaF, 1 μg/mL leupeptin, 1 g/mL aprotinin and 1 mM Phenylmethylsulphonyl fluoride (all products from Sigma Aldrich (St. Louis, MO, USA), unless indicated). Following homogenization, the peptide array protocol was carried out as per Jalal et al. [87], with alterations described in Arsenault et al. [51,88]. 4.6. Antibody Array The antibody array assay kit was procured from Full Moon BioSystems (Sunnyvale, CA, USA). This technique was used as an alternative to procuring phosphospecific antibodies individually and performing several western blot assays. The protocol was carried out as per manufacturer’s instructions (Antibody Array User’s Guide Rev 11.3) with the following alteration to the homogenization step: instead of using the bead and vortex homogenization indicated in the kit, the hand-held Qiagen Tissue Ruptor was used. 4.7. Data Analysis: Kinome and Antibody Arrays Data normalization and PCA analysis was performed for both the peptide and antibody microarrays as per Li et al. [89] using the PIIKA2 online platform (http://saphire.usask.ca/saphire/piika/index.html). Briefly, the array data were analyzed by subtracting the background intensity from the foreground intensity, variance stabilization normalization was conducted to bring all of the arrays onto the same scale, and then t-test, clustering and pathway analysis were performed. This consistent analysis method facilitated a more direct comparison between the two distinct array datasets and allowed for a statistically robust analysis of the phosphorylation events being measured. Geneontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was performed by uploading the statistically significant peptide lists to the Search Tool for the Retrieval of Interacting Genes (STRING) [36]. 4.8. Sample Collection for Bacterial Contents The ceca from each chicken was removed aseptically, and the contents (0.25 g) were serially diluted to 1:100, 1:1000, or 1:10,000 and spread onto XLT-NN plates. The plates were incubated at 37 °C for 24 h, and the number of NN-resistant S. Enteritidis cells per gram of cecal contents was determined. The data from each experimental group were pooled from three separate trials for statistical analysis. 4.9. Sample Collection for mRNA Chickens from each experimental group were euthanized at 4, 7, 10, and 14 days post-infection. A 25-mg piece of tissue was removed from the cecal tonsils. The tissue was washed in PBS and placed in a 2-mL microcentrifuge tube with 1 mL of RNAlater (Qiagen, Inc., Valencia, CA, USA) and stored at −20 °C until processed. 4.10. RNA Isolation Tissues (50 mg) were removed from RNAlater and transferred to pre-filled 2 mL tubes containing Triple-Pure™ 1.5 mm zirconium beads. RLT lysis buffer (600 μL) from the RNeasy mini kit (Qiagen, Valencia, CA, USA) was added and the tissue was homogenized for 1–2 min at 4000 rpm in a Bead Bug microtube homogenizer (Benchmark Scientific, Inc., Edison, NJ, USA). Total RNA was extracted from the homogenized lysates according to the manufacturer’s instructions, eluted with 50 μL RNase-free water, and stored at −80 °C until qRT-PCR analyses performed. RNA was quantified and the quality was evaluated using a spectrophotometer (NanoDrop Products, Wilmington, DE, USA). The data from these three repeated experiments were pooled for presentation and statistical analysis. Total RNA (300 ng) from each sample was prepared. 4.11. Quantitative Real-Time PCR The primer and probe sets for IFN-γ and 28S rRNA were designed using the Primer Express software program (Applied Biosystems, Foster City, CA, USA). IFN-γ mRNA expression was quantitated using a well-described method. Primers and probes for IFN-γ and 28S rRNA-specific amplification have been described [25,47] and are provided in Table 7. The qRT-PCR was performed using the TaqMan fast universal PCR master mix and one-step RT-PCR master mix reagents [27,28] (Applied Biosystems). Amplification and detection of specific products were performed using the Applied Biosystems 7500 Fast real-time PCR system with the following cycle profile: one cycle of 48 °C for 30 min and 95 °C for 20 s and 40 cycles of 95 °C for 3 s and 60 °C for 30 s. Quantification was based on the increased fluorescence detected by the 7500 Fast sequence detection system due to hydrolysis of the target-specific probes by the 5 = nuclease activity of the rTth DNA polymerase during PCR amplification. Normalization was carried out against 28S rRNA, which was used as a housekeeping gene. To correct for differences in RNA levels between samples within the experiment, the correction factor for each sample was calculated by dividing the mean threshold cycle (Ct) value for 28S rRNA-specific product for each sample by the overall mean Ct value for the 28S rRNA-specific product from all samples. The corrected cytokine mean was calculated as follow: (average of each replicate × cytokine slope)/(28S slope × 28S correction factor). Fold changes in mRNA levels were calculated from mean 40 Ct values by the formula 2(40 Ct infected group − 40 Ct in non-infected control). 4.12. Statistical Analysis: mRNA Expression The mean and standard error of the mean were calculated and differences between groups were determined by analysis of variance. Significant differences were further separated using Duncan’s multiple range test [27]. Fold changes in RNA levels were calculated from mean 40 Ct values using formula 2(40 Ct infected group − 40 Ct in non−infected control). A p value of ≤0.05 was considered statistically significant. 5. Conclusions Collectively, we have outlined a series of altered phosphorylation events in multiple signaling pathways in the cecum of S. Enteritidis-infected chickens that induces an immunological tolerogenic response beginning around three to four days post-primary infection. The tolerance is characterized by alterations in T cell signaling pathway and blockage of IFN-γ protection through the disruption of the JAK-STAT signaling pathway. Further, the tolerance response induces a reduction in pro-inflammatory cytokine mRNA expression and an increase in anti-inflammatory cytokine mRNA expression. Acknowledgments This study was supported by USDA-ARS intramural funding. Author Contributions Michael H. Kogut and Ryan J. Arsenault conceived and designed the experiments; Michael H. Kogut, Ryan J. Arsenault, Christina L. Swaggerty, James Allen Byrd and Ramesh Selvaraj performed the experiments; Michael H. Kogut; Ryan J. Arsenault and Christina L. Swaggerty analyzed the data; Michael H. Kogut wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Expression of IFN-γ mRNA in the ceca from experimental chickens with persistent colonization by Salmonella Enteritidis. The expression of IFN-γ mRNA expression was determined by quantitative RT-PCR. Data represent the fold-change in mRNA expression in the cecum from infected chickens when compared to the mRNA expression in the cecum from non-infected chickens. Data represent the mean ± SEM from three separate experiments. * = significantly different from the non-infected controls. Different lower case letters = significantly different from infected chickens at 2 days post-infection (p ± 0.05). ijms-17-01207-t001_Table 1Table 1 Number of chickens positive for Salmonella Enteritidis ceca colonization for 2 weeks following challenge. Treatment Groups Percent Positive for Salmonella Enteritidis Cecal Colonization (Total Positive/Total Challenged) Days post-challenge 4 7 10 14 Non-infected control 0 0 0 0 (0/50) (0/50) (0/50) (0/50) Infected 100 100 90 83 (50/50) (50/50) (45/50) (41/50) ijms-17-01207-t002_Table 2Table 2 Cecal Salmonella Enteritidis CFUs for 2 weeks following challenge. Treatment Groups CFU of Salmonella Enteritidis in Cecum (log 10) Days post-challenge 4 7 10 14 Non-infected control 0 0 0 0 Infected 5.398 ± 1.112 5.708 ± 1.341 4.342 ± 00.859 3.476 ± 1.472 ijms-17-01207-t003_Table 3Table 3 KEGG Pathways generated by Search Tool for the Retrieval of Interacting Genes/Proteins (STRING). 4 Days 7 Days 10 Days 14 Days GO ID Pathway # Peptides p-Value (FDR) # Peptides p-Value (FDR) # Peptides p-Value (FDR) # Peptides p-Value (FDR) Reference hsa04660 T cell receptor signaling pathway 26 3.09 × 10−18 - N/S 9 2.9 × 10−5 11 6.93 × 10−7 Here hsa05130 Pathogenic Escherichia coli infection - N/S 2 N/S 4 N/S 4 N/S - hsa04250 TGF-β4 signaling pathway 7 0.016 - N/S - N/S - N/S [27] hsa04310 Wnt signaling pathway 13 0.0004 3 N/S - N/S 6 0.024 [27] hsa05217 Basal cell carcinoma - N/S - N/S 2 0.338 - N/S - hsa04150 mTOR signaling pathway 13 4.83 × 10−10 2 N/S 5 1.32 × 10−3 - N/S [28] hsa04630 JAK-STAT signaling pathway 23 4.13 × 10−12 1 3.8 × 10−2 6 2.14 × 10−4 6 2.9 × 10−4 Here Peptides that displayed a significant change in phosphorylation state were input into the STRING database for each time point. Generated pathways involved in immune activation/suppression that displayed p-value of less than 0.05 (FDR corrected) are listed. # Peptides refers to number of peptides within the given pathway that were present within the peptide array data set. N/S indicates that the pathway is non-significant. ijms-17-01207-t004_Table 4Table 4 Peptides from the T cell receptor signaling pathway that displayed a statistically significant change in phosphorylation. T Cell Receptor Signaling Pathway Days Post Infection 4 7 10 14 Peptide Fold Change p-Value Fold Change p-Value Fold Change p-Value Fold Change p-Value Akt1 1.52 0.03 - - - - - - Akt3 1.80 0.04 - - - - - - Cbl Y728 −1.97 0.03 - - - - −1.23 0.03 Cbl Y773 1.30 0.02 - - - - - - CDC42 1.69 0.01 - - - - - - IKK-β −2.87 5.39 × 10−5 - - −1.91 0.04 - - FYN - - - - −2.29 0.03 - - GRB2 - - - - −1.84 0.04 - - GSK-3β −2.25 0.002 - - −2.45 0.02 - - HRAS −1.97 0.009 - - - - - - ITK −2.87 0.007 - - - - 1.61 0.04 Jun S59 3.60 0.0008 - - - - - - Jun S69 −3.39 0.0004 - - - - - - MEK1 −3.23 0.03 - - - - - - MEK2 1.25 0.01 - - - - - - MAP2K2 −1.55 0.01 - - - - - - MAP3K14 2.87 0.005 - - - - 1.10 0.03 MAPK3K7 2.71 0.02 - - 4.48 0.01 - - MAP3K8 −1.40 0.02 −2.13 0.01 - - - - p38 MAPK (MAPK11) −1.97 0.02 - - - - - p38 MAPK (MAPK14) - - - - - - −1.72 0.04 ERK1 −4.27 0.001 - - 1.61 0.01 - - NFATC1 - - - - - 1.96 0.01 NAFATC2 1.78 0.04 - - - - - - NFATC3 2.24 0.01 - - - - - - NFκB1 −3.12 0.02 - - - −1.65 0.02 NFκB1A −2.59 0.001 - - −1.54 0.01 −1.57 0.01 PAK1 S198 −1.41 0.04 - - - - - PAK1 T212 5.36 0.001 - - - - 1.05 0.02 PAK1 T422 −2.64 0.01 - - - - - - PI3KR1 3.08 0.02 - - - - - - PLCG1 −2.79 0.001 - - - - - - PRKCQ −1.47 0.003 - - - - - - PTPRC - - - - −2.97 0.02 - - RAF1 −2.29 0.003 - - - - 1.36 0.01 SOS1 3.15 0.01 - - 1.76 0.04 1.94 0.03 Peptides that displayed a p-value of less than 0.05 are listed. ijms-17-01207-t005_Table 5Table 5 Peptides from the JAK-STAT signaling pathway that displayed a statistically significant change in phosphorylation. JAK-STAT Signaling Pathway DAYS POST-INFECTION 4 7 10 14 Peptide Fold Change p-Value Fold Change p-Value Fold Change p-Value Fold Change p-Value AKT1 1.52 0.03 - - - - - - AKT3 1.80 0.04 - - - - - - Cbl 1.29 0.02 - - - - - - IFNAR1 1.88 0.003 - - - - - - IFNGR1 −1.53 0.01 - - - - - - IL-10R-A 5.10 0.03 - - 2.72 0.03 - - IL-2RB 7.89 0.0003 - - - - - - IL4R 1.37 0.01 - - 5.19 0.003 - - IL-6R −1.81 0.01 - - - - - - IL7R −6.10 0.001 - - - - - - Jak2 −3.11 0.004 - - - - −1.69 0.05 Jak3 −3.69 0.002 −1.68 0.02 −18.74 0.0006 - - PIK3R1 3.08 0.02 - - - - - - PIM1 −3.09 0.03 - - - - 1.26 0.02 SOS1 3.15 0.01 - - 1.76 0.04 1.65 0.003 STAT1 2.48 0.03 - - - - −2.17 0.009 STAT3 S728 2.44 0.03 - - - - 1.94 0.03 STAT3 Y706 1.99 0.04 - - - - - - STAT4 −2.73 0.05 - - −7.52 0.04 - - STAT5B Y699 3.33 0.028599 - - - - - - STAT5B Y740 2.01 0.04 - - −1.29 0.01 STAT6 4.07 0.003 - - −1.58 0.007 - - TYK2 1.49 0.02 - - - - - - Peptides that displayed a p-value of less than 0.05 are listed. ijms-17-01207-t006_Table 6Table 6 Antibody array results. Antibody Array Peptide Array % Homology ID Fold Change p-Value ID Fold Change p-Value AMPK (Phospho-Thr174) 2.05 0.02 AMPK1 S173 4.23 0.03 100 ATF2 (Phospho-Ser112/94) −2.13 0.02 ATF2 T72 −2.88 0.005 Calmodulin (Phospho-Thr79/Ser81) −1.48 0.01 Calmodulin T80 −1.53 0.04 100 Calmodulin Y100 −1.56 0.003 CAMK2-β/γ/Δ (Phospho-Thr287) 1.17 0.01 CAMK2-alpha T305 2.79 0.01 100 CDC25C (Phospho-Thr48) −1.54 0.03 Cdc25A T510 −1.99 0.002 100 Ezrin (Phospho-Thr566) 2.14 0.03 Ezrin Y477 2.53 0.02 100 FAK (Phospho-Ser910) 2.04 0.03 FAK Y397 4.18 0.04 100 FLT3 (Phospho-Tyr842) −1.20 0.02 FLT3 Y452 −1.81 0.03 79 HSP27 (Phospho-Ser15) 1.12 0.02 HSP27 S15 −4.35 0.01 67 c-Jun (Phospho-Tyr170) 4.27 0.04 Jun S59 3.60 0.001 85 MEK1 (Phospho-Thr291) −1.22 0.03 MEK1 S222 −3.23 0.04 100 MEK-2 (Phospho-Thr394) −1.58 0.04 MEK2 S220 −1.55 0.01 100 MSK1 (Phospho-Ser376) −1.28 0.01 MSK1 S366 −3.26 0.008 100 P38 MAPK (Phospho-Thr180) −1.39 0.05 P38-alpha Y181 −1.97 0.02 100 PAK1 (Phospho-Thr122) 1.44 0.04 PAK1 T212 5.36 0.001 80 PKC delta (Phospho-Tyr52) −1.36 0.04 PKCD Y311 −1.15 0.001 100 PLCG1 (Phospho-Tyr783) −1.20 0.03 PLCG1 Y675 −2.79 0.001 83 SMAD 2 (Phospho-Thr220) 1.48 0.04 SMAD2 S245 3.43 0.01 100 Smad2 S255 4.53 0.01 Smad 2/3 (Phospho-Thr8) 1.09 0.03 Smad3 T180 1.44 0.03 Src (Phospho-Tyr418) −1.61 0.005 Src Y416 −1.44 0.03 100 Src Y527 −2.12 0.02 STAT3 (Phospho-Ser717) 1.98 0.03 STAT3 S728 2.44 0.03 100 STAT3 Y706 1.99 0.04 Trk (Phospho-Tyr515) 1.13 0.04 TrKA Y490 −1.52 0.02 84 TrKA Y674 −1.68 0.03 TrKA Y785 −2.04 0.0 XIAP (Phospho-Ser87) 1.17 0.04 XIZP S87 2.12 0.002 60 Statistically significant (p ≤ 0.05) phosphospecific antibody array results of Salmonella Entertidis cecal samples. Four days post-infection samples were compared to non-infected control samples to find changes in infected cecal tissue over time. Antibodies due to being bound to phosphorylated protein had a statistically significant difference in fluorescent signal are shown. Fold Change Antibody Array is the change in fluorescent signal when comparing the infected samples to control samples. Homology indicates the % similarity between human and chicken at the 15 amino acid region flanking the phosphorylation residue. Fold Change Peptide Array is the change in fluorescent signal as indicated by the peptide array. N/A indicates the exact phosphorylation target residue on the antibody array was not present on the peptide array or not significantly differentially phosphorylated. ijms-17-01207-t007_Table 7Table 7 Real-time quantitative RT-PCR probes and primers for 28S and IFN-γ. RNA Target Probe/Primer Sequence Accession Number a 28S Probe d 5′-(FAM)-AGGACCGCTACGGACCTCCACCA-(TAMRA)-3′ X59733 F b 5′-GGCGAAGCCAGAGGAAACT-3′ R c 5′-GACGACCGATTGCACGTC-3′ IFN-γ Probe d 5′-(FAM)-TGGCCAAGCTCCCGATGAACGA-(TAMRA)-3′ YO7922 F 5′-GTGAAGAAGGTGAAAGATATATCATGGA-3′ R 5′-GCTTTGCGTGGATTCTCA-3′ a Genomic DNA sequence; b Forward; c Reverse; d 5-carboxyfluorescein. ==== Refs References 1. Centers for Disease Control and Prevention (CDC) National Salmonella Surveillance Overview 2011 Available online: http://www.cdc.gov/ncezid/dfwed/PDFs/salmonella-annual-summary-2009-508c.pdf (accessed on 25 April 2016) 2. Desmidt R. Ducatelle R. Haesebrouck F. Pathogenesis of Salmonella enteritidis phage type four after experimental infection of young chickens Vet. Microbiol. 1997 56 99 109 10.1016/S0378-1135(96)01350-8 9228686 3. Barrow P.A. Huggins M.B. Lovell M.A. Simpson J.M. Observations on the pathogenesis of experimental Salmonella typhimurium infection in chickens Res. Vet. Sci. 1987 42 194 199 3296063 4. Barrow P.A. The paratyphoid Salmonellae Rev. Sci. Tech. 2000 19 351 375 10.20506/rst.19.2.1225 10935268 5. Gopinath S. Carden S. Monack D. Shedding light on Salmonella carriers Trends Immunol. 2012 20 320 327 10.1016/j.tim.2012.04.004 22591832 6. Santos R.L. Raffatellu M. Bevins C.L. Adams L.G. Tükel C. Tsolis R.M. Bäumler A.J. Life in the inflamed intestine, Salmonella style Trends Microbiol. 2009 17 498 506 10.1016/j.tim.2009.08.008 19819699 7. Smith H.W. Tucker J.F. The virulence of salmonella strains for chickens: Their excretion by infected chickens J. Hyg. (Lond.) 1975 84 479 488 10.1017/S0022172400027017 8. Barrow P.A. Lovell M.A. Experimental infection of egg-laying hens with Salmonella enteritidis Avian Pathol. 1991 20 339 352 9. Wathanage G.S. Kaiser P. Wigley P. Powers C. Mastroeni P. Brooks H. Barrow P. Smith A. Maskell D. McConnell I. Rapid expression of chemokines and pro-inflammatory cytokines in newly hatched chickens infected with Salmonella enterica serovar Typhimurium Infect. Immun. 2004 72 2152 2159 10.1128/IAI.72.4.2152-2159.2004 15039338 10. Perron G.G. Quessy S. Bell G. A reservoir of drug-resistant pathogenic bacteria in asymptomatic hosts PLoS ONE 2008 3 1207 10.1371/journal.pone.0003749 19015729 11. Wigley P. Immunity to bacterial infection in the chicken Dev. Comp. Immunol. 2013 41 413 417 10.1016/j.dci.2013.04.008 23648643 12. Beal R.K. Wigley P. Powers C. Hulme S.D. Barrow P.A. Smith A.L. Age at primary infection with Salmonella enterica serovar Typhimurium in the chicken influences persistence of infection and subsequent immunity ot re-challenge Vet. Immunol. Immunopathol. 2004 100 151 164 10.1016/j.vetimm.2004.04.005 15207453 13. Setta A. Barrow P.A. Kaiser P. Jones M.A. Immune dynamics following infection of avian macrophages and epithelial cells with typhoidal and non-typhoidal Salmonella enterica serovars: Bacterial invasion and persistence, nitric oxide and oxygen production, differential host gene expression, NF-κB signaling and cell cytotoxicity Vet. Immunol. Immunopathol. 2012 146 212 224 22475571 14. Kogut M.H. Tellez G.I. McGruder E.D. Hargis B.M. Williams J.D. Corrier D.E. DeLoach J.R. Heterophils are decisive components in the early responses of chickens to Salmonella enteritidis infections Microb. Pathog. 1994 16 141 151 10.1006/mpat.1994.1015 8047002 15. Monack D.M. Mueller A. Falkow S. Persistent bacterial infections: The interface of the pathogen and the host immune system Nat. Rev. Microbiol. 2004 2 747 765 10.1038/nrmicro955 15372085 16. Barrow P.A. Simpson J.M. Lovell M.A. Intestinal colonization in the chicken by food-poisoning salmonella serotypes; Microbial characteristics associated with faecal excretion Avian Path. 1988 17 571 588 10.1080/03079458808436478 18766717 17. Holt P.S. Gast R.K. Porter R.E. Stone H.D. Hyporesponsiveness of the systemic and mucosal humoral immune systems in chickens infected with Salmonella enterica serovar Enteritidis at one day of age Poult. Sci. 1999 78 1510 1517 10.1093/ps/78.11.1510 10560822 18. Raberg L. Sim D. Read A.F. Disentangling genetic variation for resistance and tolerance to infectious diseases in animals Science 2007 318 812 814 10.1126/science.1148526 17975068 19. Schneider D.S. Ayres J.S. Two ways to survive infection: What resistance and tolerance can teach us about treating infectious diseases Nat. Rev. Immunol. 2008 8 889 895 10.1038/nri2432 18927577 20. Soares M.P. Gozzelino R. Weiss S. Tissue damage control in disease tolerance Trends Immunol. 2014 35 483 494 10.1016/j.it.2014.08.001 25182198 21. Read A.F. Graham A.L. Raberg L. Animal defenses against infectious agents: Is damage control more important than pathogen control PLoS Biol. 2008 6 1207 10.1371/journal.pbio.1000004 19222305 22. Ayres J.S. Schneider D.S. Tolerance of infections Annu. Rev. Immunol. 2012 30 271 294 10.1146/annurev-immunol-020711-075030 22224770 23. Medzhitov R. Schneider D.S. Soares M.P. Disease tolerance as a defense strategy Science 2012 335 936 941 10.1126/science.1214935 22363001 24. Merrell D.S. Falkow S. Frontal and stealth attack strategies in microbial pathogenesis Nature 2004 430 250 256 10.1038/nature02760 15241423 25. Schmid-Hempel P. Immune defense, parasite evasion strategies and their relevance for “macroscopic phenomena” such as virulence Philos. Trans. R. Soc. Lond. B Biol. Sci. 2009 364 85 98 10.1098/rstb.2008.0157 18930879 26. Chausse A.-M. Grepinet O. Bottreau E. Robert V. Hennequet-Antier C. Lalmanach A.-C. Lcardonnel J. Beaumont C. Velge P. Susceptibility to Salmonella carrier-state: A possible Th2 response in susceptible chicks Vet. Immunol. Immunopathol. 2014 159 16 28 10.1016/j.vetimm.2014.03.001 24694400 27. Kogut M.H. Arsenault R.J. A role for the non-canonical Wnt-β-catenin and TGF-β signaling pathways in the induction of tolerance during the establishment of a Salmonella enterica serovar Enteritidis persistent cecal infection in chickens Front. Vet. Infect. Dis. 2015 2 33 10.3389/fvets.2015.00033 26664962 28. Kogut M.H. Genovese K.G. He H. Arsenault R.J. AMPK and mTOR: Sensors and regulators of immunmetabolic changes during Salmonella infection in the chicken Poult. Sci. 2016 95 345 353 10.3382/ps/pev349 26706353 29. Shanmugasundaram R. Kogut M.H. Arsenault R.J. Swaggerty C.L. Cole K.Y. Reddish M.J. Selvaraj R.K. Effect of Salmonella infection on cecal tonsil regulatory T cell properties in chickens Poult. Sci. 2015 94 1828 1835 10.3382/ps/pev161 26049799 30. Belkaid Y. Regulatory T cells and infection: A dangerous necessity Nat. Rev. Immunol. 2007 7 875 888 10.1038/nri2189 17948021 31. Belkaid Y. Tarbell K. Regulatory T cells in the control of host-microorganism interactions Annu. Rev. Immunol. 2009 27 551 589 10.1146/annurev.immunol.021908.132723 19302048 32. Mendez S. Reckling S.K. Piccirillo C.A. Sacks D. Belkaid Y. Role for CD4+ CD25+ regulatory T cells in reactivation of persistent leishmaniasis and control of concomitant immunity J. Exp. Med. 2004 200 201 210 10.1084/jem.20040298 15263027 33. Hesse M. Piccirillo C.A. Belkaid Y. Prufer J. Mentink-Kane M. Leusink M. Cheever A.W. Shevach E.M. Wynn T.A. The pathogenesis of schistosomiasis is controlled by cooperating IL-10-producing innate effector and regulatory T cells J. Immunol. 2004 172 3157 3166 10.4049/jimmunol.172.5.3157 14978122 34. Hisaeda H. Maekawa Y. Iwakawa D. Okada H. Himeno K. Kishihara K. Tsukumo S. Yasutomo K. Escape of malaria parasites from host immunity requires CD4+ CD25+ regulatory T cells Nat. Med. 2004 10 29 30 10.1038/nm975 14702631 35. Johanns T.M. Erelt J.M. Rowe J.H. Way S.S. Regulatory T cell suppressive potency dictates the balance between bacterial proliferation and clearance during persistent Salmonella infection PLoS Path. 2010 6 1207 10.1371/journal.ppat.1001043 20714351 36. Arsenault R. Kogut M.H. He H. Combined CpG and poly I:C stimulation of monocytes results in unique signaling activation not observed with the individual ligands Cell Signal. 2013 25 2246 2254 23876795 37. Szklarczyk D. Franceschini A. Kuhn M. Simonovic M. Roth A. Minguez P. Doerks T. Stark M. Muller J. Bork P. The STRING database in 2011: Functional interaction networks of proteins, globally integrated and scored Nucleic Acids Res. 2011 39 Suppl. S1 D561 D568 10.1093/nar/gkq973 21045058 38. Kaiser P. Rothwell L. Galyov E.E. Barrow P.A. Burnside J. Wigley P. Differential cytokine expression in avian cells in response to invasion by Salmonella typhimurium , Salmonella enteritidis and Salmonella gallinarum Microbiology 2000 146 3217 3226 10.1099/00221287-146-12-3217 11101679 39. Fasina Y.O. Holt P.S. Moran E.T. Moore R.T. Conner D.E. McKee S.R. Intestinal cytokine response of commercial source broiler chicks to Salmonella typhimurium infection Poult. Sci. 2008 87 1335 1346 10.3382/ps.2007-00526 18577613 40. Crhanova M. Hradecka H. Faldynova M. Mutulova M. Havlickova H. Sisak F. Rychlik I. Immune response of chicken gut to natural colonization by gut microflora and to Salmonella enterica serovar Enteritidis infection Infect. Immun. 2011 79 2755 2763 10.1128/IAI.01375-10 21555397 41. Lawley T.D. Bouley D.A. Hoy Y.E. Gerke C. Relman D.A. Monack D.M. Host transmission of Salmonella enterica serovar Typhimurium is controlled by virulence factors and indigenous intestinal microbiota Infect. Immun. 2008 76 403 416 10.1128/IAI.01189-07 17967858 42. Sadeyen J.-R. Trotereau J. Velge P. Marty J. Beaumont C. Barrow P.A. Bumstead N. Lalmanach A.-C. Salmonella carrier state in chicken: Comparison of immune response genes between susceptible and resistant animals Microb. Infect. 2004 6 1278 1286 43. Kuijl C. Savage N.D.L. Marsman M. Tuin A.W. Janssen L. Egan D.A. Keterma M. van den Nieuwendijk R. van den Eeden S.J.F. Geluk A. Intracellular bacterial growth is controlled by a kinase network around PKB/Akt1 Nature 2007 450 725 730 10.1038/nature06345 18046412 44. Rogers L.D. Brown N.F. Fang Y. Pelech S. Foster L.J. Phosphoproteomic analysis of Salmonella -infected cells identifies key kinase regulators and SopB-dependent host phosphorylation events Sci. Signal. 2011 4 191 10.1126/scisignal.2001668 21934108 45. Roppenser B. Kwon H. Canadien V. Xu R. Devreotes P.N. Grinstein S. Brumell J.H. Multiple host kinases contribute to Akt activation during Salmonella infection PLoS ONE 2013 8 1207 10.1371/journal.pone.0071015 23990921 46. Imami K. Bhavsar A.P. Yu H. Brown N.F. Rogers L.D. Finlay B.B. Foster L.J. Global impact of Salmonella pathogenicity island 2-secreted effectors on the host phosphoproteome Mol. Cell. Proteom. 2013 12 1632 1643 10.1074/mcp.M112.026161 23459991 47. Hannemann S. Gao B. Galan J.E. Salmonella modulation of host cell gene expression promotes its intracellular growth PLoS Path. 2013 9 1207 10.1371/journal.ppat.1003668 24098123 48. Chappell L. Kaiser P. Barrow P. Jones M.A. Johnston C. Wigley P. The immunobiology of avian systemic salmonellosis Vet. Immunol. Immunopathol. 2009 128 53 59 10.1016/j.vetimm.2008.10.295 19070366 49. Barrow P.A. Jones M.A. Smith A.L. Wigley P. The long view: Salmonell —The last forty years Avian Pathol. 2012 41 413 420 10.1080/03079457.2012.718071 23025669 50. Arsenault R. Kogut M.H. Chicken-specific peptide arrays for kinome analysis: Flight for the flightless Curr. Top. Biotechnol. 2013 7 79 89 51. Arsenault R. Napper S. Kogut M.H. Salmonella enterica Typhimurium infection causes metabolic changes in chicken muscle involving AMPK, fatty acid and insulin/mTOR signaling Vet. Res. 2013 44 35 23682635 52. Ruby T. McLaughlin L. Gopinath S. Monack D. Salmonella’s long-term relationship with its host FEMS Microbial. Rev. 2012 36 600 615 10.1111/j.1574-6976.2012.00332.x 22335190 53. Smith-Garvin J.E. Koretzky G.A. Jordan M.S. T cell activation Annu. Rev. Immunol. 2009 27 591 619 10.1146/annurev.immunol.021908.132706 19132916 54. Vallabhapurapu S. Karin M. Regulation and function of NF-κB transcription factors in the immune system Annu. Rev. Immunol. 2009 27 693 733 10.1146/annurev.immunol.021908.132641 19302050 55. Baine I. Abe B.T. Macian F. Regulation of T-cell tolerance by calcium/NFAT signaling Immunol. Rev. 2009 231 225 240 10.1111/j.1600-065X.2009.00817.x 19754900 56. Rao A. NF-ATp: A transcription factor required for the coordinate induction of several cytokine genes Immunol. Today 1994 15 274 281 10.1016/0167-5699(94)90007-8 8068174 57. Chen L. Glover J.N. Hogan P.G. Rao A. Harrison S.C. Structure of the DNA-binding domains from NFAT, Fos and Jun bound specifically to DNA Nature 1998 392 42 48 10.1038/32100 9510247 58. Zanoni I. Grancci F. Regulation and dysregulation of innate immunity by NFAT signaling downstream of pattern recognition receptors (PRRs) Eur. J. Immunol. 2012 42 1924 1931 10.1002/eji.201242580 22706795 59. Dev A. Iywe S. Razani B. Cheng G. NF-κB and innate immunity Curr. Top. Microbiol. Immunol. 2011 349 115 143 20848362 60. Hacker H. Karin M. Regulation and function of IKK and IKK-related kinases Sci. STKE 2006 2006 e13 10.1126/stke.3572006re13 17047224 61. Oh-hora M. Rao A. The calcium/NFAT pathway: Role in development and function of regulatory T cells Microb. Infect. 2009 11 612 619 10.1016/j.micinf.2009.04.008 19375514 62. Vaeth M. Schliesser U. Muller G. Reissig S. Satch K. Tuettenberg A. Jonulet H. Waisman A. Muller M.R. Serfling E. Dependence on nuclear factor of activated T-cells (NFAT) levels discriminates conventional T cells from Foxp3+ regulatory T cells Proc. Natl. Acad. Sci. USA 2012 109 16258 16263 10.1073/pnas.1203870109 22991461 63. Huang Y.-H. Sojka D.K. Fowell D.J. Regulatory T cells selectively attenuate, not terminate, T cell signaling by disrupting NF-κB nuclear accumulation in CD4 T cells J. Immunol. 2012 188 947 951 10.4049/jimmunol.1101027 22227565 64. Luo C.T. Li M.O. Transcriptional control of regulatory T cell development and function Trends Immunol. 2013 34 531 539 10.1016/j.it.2013.08.003 24016547 65. Aaronson D.S. Horvath C.M. A road map for those who don’t know JAK-STAT Science 2002 296 1653 1655 10.1126/science.1071545 12040185 66. Cantrell D.A. Smith K.A. The interleukin-2 T-cell system: A new cell growth model Science 1984 224 1312 1316 10.1126/science.6427923 6427923 67. Smith K.A. Interleukin-2: Inception, impact, and implications Science 1988 240 1169 1176 10.1126/science.3131876 3131876 68. Stark G.R. Kerr I.M. Williams B.R. Silverman R.H. Schreiber R.D. How cells respond to interferons Annu. Rev. Biochem. 1998 67 227 264 10.1146/annurev.biochem.67.1.227 9759489 69. Shaw M.H. Freeman G.J. Scott M.F. Fox B.A. Bzik D.J. Belkaid Y. Yap G.S. TYK2 negatively regulates adaptive Th1 immunity by mediating IL-10 signaling and promoting IFN-γ-dependent IL-10 reactivation J. Immunol. 2006 176 7263 7271 10.4049/jimmunol.176.12.7263 16751369 70. Kim J.-H. Kim S.-J. Lee I.-S. Lee M.-S. Uematsu S. Akira S. Oh K.I. Bacterial endotoxin induces the release of high mobility group box 1 via the IFN-β signaling system J. Immunol. 2009 182 2458 2466 10.4049/jimmunol.0801364 19201901 71. Liu H. Yao Y.-M. Yu Y. Dong N. Yin H.-N. Sheng Z.-Y. Role of Janus kinase/signal transducer and activator of transcription pathway in regulation of expression and inflammation-promoting activity of high mobility group box protein 1 in rat peritoneal macrophages Shock 2007 27 55 60 10.1097/01.shk.0000233197.40989.31 17172981 72. Pena G. Cai B. Deitch E.A. Ulloa L. JAK2 inhibition prevents innate immune responses and rescues animals from sepsis J. Mol. Med. 2010 88 851 859 10.1007/s00109-010-0628-z 20393690 73. Mishra J. Karanki S.S. Kumar N. Identification of molecular switch regulating interactions of Janus kinase 3 with cytoskeletal proteins J. Biol. Chem. 2012 287 41386 41391 10.1074/jbc.C112.363507 23012362 74. Mishra J. Waters C.M. Kumar N. Molecular mechanism of interleukin-2-induced mucosal homeostasis Am. J. Physiol. Cell Physiol. 2012 302 C735 C747 10.1152/ajpcell.00316.2011 22116305 75. Kumar N. Mishra J. Narang V.S. Waters C.M. Janus kinase 3 regulates interleukin 2-induced mucosal wound repair through tyrosine phosphorylation of villin J. Biol. Chem. 2007 282 30341 30345 10.1074/jbc.C600319200 17537734 76. Nosaka T. van Deursen J.M.A. Tripp R.A. Thierfelder W.E. Witthuhn B.A. McMickle A.P. Doherty P.C. Grosveld G.C. Ihle J.N. Defective lymphoid development in mice lacking JAK3 Science 1995 270 800 802 10.1126/science.270.5237.800 7481769 77. Park S.Y. Saijo K. Takahashi T. Osawa M. Areas H. Hirayama N. Miyake K. Nakaucji H. Shirasawa T. Saito T. Developmental defects of lymphoid cells in JAK3 kinase-deficient mice Immunity 1995 3 771 782 10.1016/1074-7613(95)90066-7 8777722 78. Ghoreschi K. Laurence A. O’Shea J.J. Janus kinases in immune cell signaling Immunol. Rev. 2009 228 273 287 10.1111/j.1600-065X.2008.00754.x 19290934 79. Guo Z. Khattar M. Schroder P.M. Miyahara Y. Wang G. He X. Chen W. Stepkowski S.M. A dynamic dual role of IL-2 signaling in the two-step differentiation process of adaptive regulatory T cells J. Immunol. 2013 190 3153 3162 10.4049/jimmunol.1200751 23427250 80. Cetkovic-Cvrlje M. Olson M. Ghate K. Targeting Janus tyrosine kinase 3 (JAK3) with an inhibitor induces secretion of TGF-β4 by CD4+ T cells Cell. Mol. Immunol. 2012 9 350 360 10.1038/cmi.2012.20 22728763 81. Kaplan M.H. STAT4: A critical regulator of inflammation in vivo Immunol. Res. 2005 32 231 241 10.1385/IR:31:3:231 16106075 82. Miyagi T. Gil M.P. Wang X. Louten J. Chu W.M. Biron C.A. High basal STAT4 balanced by STAT1 induction to control type 1 interferon effects in natural killer cells J. Exp. Med. 2007 204 2383 2396 10.1084/jem.20070401 17846149 83. Richer E. Prendergast C. Zhang D.-E. Qureshi S.T. Vidal S.M. Malo D. N -ethyl-N -nitrosurea-induced mutation in ubiquitin-specific peptidase 18 causes hyperactivation of IFN-αβ signaling and suppresses STAT4-induced IFN-γ production, resulting in increased susceptibility to Salmonella Typhimurium J. Immunol. 2010 185 3593 3601 10.4049/jimmunol.1000890 20693420 84. Eva M.M. Yuki E.E. Dauphinee S.M. Schwartzentruber J.A. Pyzik M. Paquet M. Lathrop M. Majewski J. Vidal S.M. Malo D. Altered IFN-γ-mediated immunity and transcriptional expression patterns in N -ethyl-N -nitrosurea-induced STAT4 mutants confer susceptibility to acute typhoid-like disease J. Immunol. 2014 192 259 270 10.4049/jimmunol.1301370 24285835 85. Kogut M.H. Rothwell L. Kaiser P. IFN-γ priming of chicken heterophils upregulates the expression of proinflammatory and Th1 cytokine mRNA following receptor-meditated phagocytosis of Salmonella enterica serovar Enteritidis J. Int. Cytokine Res. 2005 25 73 81 10.1089/jir.2005.25.73 15695928 86. National Research Council Nutritional Requirements of Poultry 9th ed. National Academy Press Washington, DC, USA 1994 87. Jalal S. Arsenault R. Potter A.A. Babiuk L.A. Griebel P.J. Napper S. Genome to kinome: Species-specific peptide arrays for kinome analysis Sci. Signal. 2009 2 pl1 10.1126/scisignal.254pl1 19155530 88. Arsenault R.J. Li Y. Bell K. Doig K. Potter A. Griebel P.J. Kusalik A. Napper S. Mycobacterium avium subsp. Paratuberculosis inhibits interferon γ-induced signaling in bovine monocytes: Insights into the cellular mechanisms of Johne’s disease Infect. Immun. 2012 80 3039 3048 10.1128/IAI.00406-12 22689821 89. Li Y. Arsenault R.J. Trost B. Slind J. Griebel P.J. Napper S. Kusalik A. A systematic approach for analysis of peptide array kinome data Sci. Signal. 2012 5 pl2 10.1126/scisignal.2002429 22510468

PMC005xxxxxx/PMC5000606.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081208ijms-17-01208ReviewOne-Carbon Metabolism in Prostate Cancer: The Role of Androgen Signaling Corbin Joshua M. 1Ruiz-Echevarría Maria J. 2*Yang Li Academic Editor1 Department of Pathology, Oklahoma University Health Sciences Center, Oklahoma City, OK 73104, USA; [email protected] Department of Pathology, Oklahoma University Health Sciences Center and Stephenson Cancer Center, Oklahoma City, OK 73104, USA* Correspondence: [email protected]; Tel.: +1-405-271-187127 7 2016 8 2016 17 8 120822 6 2016 18 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Cancer cell metabolism differs significantly from the metabolism of non-transformed cells. This altered metabolic reprogramming mediates changes in the uptake and use of nutrients that permit high rates of proliferation, growth, and survival. The androgen receptor (AR) plays an essential role in the establishment and progression of prostate cancer (PCa), and in the metabolic adaptation that takes place during this progression. In its role as a transcription factor, the AR directly affects the expression of several effectors and regulators of essential catabolic and biosynthetic pathways. Indirectly, as a modulator of the one-carbon metabolism, the AR can affect epigenetic processes, DNA metabolism, and redox balance, all of which are important factors in tumorigenesis. In this review, we focus on the role of AR-signaling on one-carbon metabolism in tumorigenesis. Clinical implications of one-carbon metabolism and AR-targeted therapies for PCa are discussed in this context. one-carbon metabolismandrogen receptorepigeneticsmethylationpolyamine metabolismtranssufluration ==== Body 1. Introduction Prostate cancer (PCa) is the most frequently diagnosed non-skin cancer and the fifth leading cause of cancer death in men worldwide [1]. Clinically, PCa is a heterogeneous disease, ranging from an indolent disease, requiring no treatment, to highly aggressive PCa that develops into metastatic disease. Despite this heterogeneity, prostate tumor growth is, almost always, dependent upon the androgen receptor (AR) pathway [2,3,4], explaining the efficacy of androgen deprivation therapies (ADT) or anti-androgens for the treatment of hormone-naïve PCa [5,6]. However, most patients relapse following ADT and the disease progresses to castration-resistant prostate cancer (CRPC), which is lethal [7,8,9]. Central to the development of CRPC is the reactivation/adaptation of AR signaling to function under low androgen levels. Therefore, the AR and the processes downstream of the AR remain as targets for therapeutic intervention throughout the different stages of the disease. Recent results indicate that the AR drives a distinct transcriptional program in CRPC, and that changes in AR activity are critical to drive disease progression [10,11]. Efforts to identify clinically relevant, AR-modulated, transcriptional networks have established a link between the AR and cellular metabolism, consistent with the changes in metabolism that occur with disease progression [12,13]. Recent data indicate that expression of the constitutively active AR-V7 variant in CRPC has novel metabolic functions that may be specifically targeted [14]. In PCa, the one-carbon metabolism pathway is modulated by the AR. This pathway is comprised of several connected pathways that promote the folate-mediated transfer of one-carbon units necessary for essential cellular processes including DNA synthesis and repair and the maintenance of redox status. Because one-carbon metabolism is also the major source of methyl groups, as a modulator of this pathway, the AR also plays critical roles in histone and DNA methylation and in epigenetic mechanisms that are known to be relevant in oncogenesis [15,16,17]. Studies in PCa cell lines demonstrate AR-regulation of one-carbon metabolism enzymes, and altered cellular methylation potential in response to androgens [18,19,20,21]. In PCa clinical samples, accumulation of sarcosine, a methylated metabolite of the one-carbon pathway, correlates with disease progression [20]. Changes in several other metabolites also correlate with PCa risk [22]. These findings illustrate the role of the AR in PCa tumorigenesis by controlling metabolism, and the value of integrating metabolomic profiling and gene expression analysis for the identification of new biomarkers and therapeutic targets. In this review, we will focus on the role of the AR on one-carbon metabolism and the implications for disease progression. The first two sections focus on the relevance of one-carbon metabolism and its link to cancer. The third section outlines how AR-signaling modulates the expression and activity of enzymes involved in one-carbon metabolism, and how it affects methylation-mediated epigenetic processes in PCa. The final section discusses targeting one-carbon metabolism in PCa, and the potential effects of current AR-targeting therapeutic modalities on one-carbon metabolism. 2. The One-Carbon Metabolism Network One-carbon metabolism involves a complex network with two central cycles: (1) the folate cycle; and (2) the methionine cycle (Figure 1). In the folate cycle, tetrahydrofolate (THF) acts as a carbon carrier donor for the synthesis of purines and thymidilates, which are vital for DNA synthesis and repair. The transfer of methyl groups from 5-methylTHF to homocysteine to form methionine links the two cycles. Methionine is then converted to S-adenosyl-methionine (SAM), the universal methyl donor for protein and DNA methyltransferase reactions. By donating a methyl group, SAM is converted to S-adenosyl-homocysteine (SAH), and subsequently to homocysteine to close the cycle [17,23,24,25]. In addition to being recycled back to methionine, homocysteine can also be shunted to the transsulfuration pathway where it is converted into cystathionine, a precursor of glutathione, an important cofactor in oxidation/reduction (redox) reactions that regulate the cellular redox state. SAM can also contribute to the synthesis of polyamines, which are small organic cations that regulate multiple biological processes, including, translation and proliferation, linking the methionine cycle with polyamine synthesis [26,27]. Since one-carbon metabolism regulates essential processes including DNA synthesis and repair, epigenetic methylation reactions, redox homeostasis, and protein synthesis, the balanced flux through these four pathways (folate cycle, methionine cycle, transsulfuration pathway, polyamine synthesis) is essential for cellular homeostasis. In fact, disruptions in that balance contribute to the pathogenesis of many diseases, including cancer [28]. Balance within the one-carbon metabolism network is maintained in part by interactions involving substrates and enzymes from these pathways (Figure 2). SAM inhibits methylene-tetrahydrofolate reductase (MTHFR), the enzyme that catalyzes formation of 5-methylTHF, a necessary cofactor to regenerate methionine and, ultimately, SAM levels [17]. 5-methylTHF is an inhibitor of glycine N-methyltransferase (GNMT), the enzyme that catalyzes formation of sarcosine from glycine, which eventually donates methyl groups back to the THF in a reaction catalyzed by sarcosine dehydrogenase (SARDH) [29,30]. SAM also stimulates cystathionine beta-synthase (CBS), the enzyme that shuttles homocysteine into the transsulfuration pathway [31,32]. Additionally, folate regulates enzymes involved in polyamine metabolism [33,34]. These interactions maintain an exquisite balance between one-carbon metabolism and its associated pathways to maintain cellular homeostasis. 3. One-Carbon Metabolism in Cancer Cancer creates a demand and dependency on one-carbon metabolism. Proliferation of tumor cells not only requires increased DNA synthesis, but can also result in increased levels of reactive radical oxygen species (ROS), which are cytotoxic unless neutralized [35,36]. Methyl group availability for methyltransferases that modulate gene expression via epigenetic mechanisms is influenced by flux within the folate cycle and methionine cycles [15,16]. In addition, synthesis of polyamines, which have been suggested to have oncogenic functions through regulating protein synthesis and proliferation [37,38], is SAM-dependent. Several enzymes within the folate cycle are potentially oncogenic and are dysregulated in cancer. Serine hydroxymethyltransferase (SHMT) and glycine decarboxylase (GLDC) donate methyl groups to the folate pathway in sequential steps via the catabolism of serine and glycine, respectively [15]. SHMT, in concert with GLDC, drives tumorigenesis possibly by fueling the folate cycle and driving proliferation [39]. Thymidylate synthase (TS), another enzyme involved in the folate cycle, catalyzes the methylation of deoxyuracil-monophosphate to deoxythymidine-monophosphate, in a 5,10-methylene-THF-dependent reaction that is necessary for DNA synthesis and repair. The overexpression of TS is sufficient to induce a tumorigenic phenotype in NIH3T3 cells in vivo, and elevated TS expression correlates with a poor prognosis in multiple cancer types [40,41,42,43,44,45,46]. Furthermore, the TS inhibitor, 5-fluorouracil (5-FU), is used in the treatment of multiple cancers, especially colon cancer [47]. Paradoxically, although folate is necessary for cancer cell proliferation, multiple studies have reported a positive correlation between folate deficiency and disease risk for multiple cancers, especially breast and colon cancers [48,49,50,51]. Additionally, higher folate intake reduces the increased breast cancer risk associated with elevated alcohol consumption; this relationship may be due in part to the antagonistic effect of alcohol on folate absorption, metabolism and transport [51]. Aberrant uracil incorporation and chromosomal breaks can both be induced by folate deficiency, thus providing a potential mechanism by which folate deficiency can contribute to tumorigenesis [52,53]. Additionally, the MTHFR C677T polymorphism may be associated with increased breast cancer risk [49,54,55,56,57,58]. The C677T polymorphism reduces MTHFR activity, thus lowering 5-methyl-THF levels and decreasing methionine regeneration [59]. Not only can folate deficiency contribute to mutations during replication [53,60], but folate deficiency or MTHFR polymorphisms may also decrease methionine regeneration and SAM levels, thereby, reducing the ability of the cell to maintain DNA and histone methylation. Importantly, cancer cells often exhibit global DNA hypomethylation, a phenotype that may be linked to genomic instability [52,61,62]. In contrast, folate depletion blocked tumor progression in vivo and induced genetic instability in cells in vitro, in the Transgenic Adenocarcinoma of the Mouse Prostate (TRAMP) model for PCa [63,64]. Further, folate supplementation has been shown to drive tumor growth in some mouse and rat cancer models [65,66,67]. However, the timing of folate supplementation in disease progression is likely critical, as studies indicate that folate may be both protective against neoplastic lesion formation and a promoter of growth within established lesions [67,68,69]. These studies highlight a widely supported “double-edged sword” hypothesis for the role of the folate cycle in cancer: Folate depletion may contribute to initial transformation by inducing global DNA hypomethylation and subsequent genomic instability, while higher folate levels may promote the growth of transformed cells by enabling an increased rate of DNA synthesis [68,69]. Even in the presence of global DNA hypomethylation, many cancer cells contain gene specific hypermethylation, a silencing mechanism. The tumor suppressor Rb was the first gene found to be silenced by DNA hypermethylation during tumorigenesis [70]. Since then, numerous tumor suppressor genes have found to be silenced by DNA hypermethylation in cancer. Unlike DNA mutations, epigenetic aberrations—including DNA methylation—can be reversed by inhibiting the enzymes responsible for the epigenetic marks. This is one reason why targeting epigenetic enzymes has gained traction in cancer therapy [71]. Histone methylation is a SAM-dependent epigenetic process. Several methyl histone marks are dysregulated in many cancer types, and depending on the target residue, these methylated histones can contribute to gene activation or repression [72,73,74,75,76,77]. The enhancer of zeste homolog 2 (EZH2), DOT1L and mixed-lineage leukemia (MLL) methyltransferases are among the histone methyltransferases (HMTs) found to play important roles in driving a tumorigenic epigenome, which is similar to that of stem cells [78,79,80,81,82,83,84,85,86] HMTs use SAM as a methyl donor, and many HMTs are inhibited by SAH (Figure 2), a byproduct of methyltransferase reactions; therefore, one-carbon metabolism flux has a profound impact on the activity of these enzymes [87,88,89]. SAM not only serves as a cofactor for methyltransferases, but it is also shunted from one-carbon metabolism and utilized in polyamine synthesis. Polyamines have been implicated in cancer, and their oncogenic function may be linked to their roles in protein synthesis and cell cycle regulation [27,37,90]. Ornithine decarboxylase (ODC) catalyzes the formation of putrescine from ornithine, a rate-limiting step in the polyamine biosynthetic pathway. ODC is a MYC-regulated oncogene that is critical for cell cycle progression, in part by promoting MYC-induced p21 degradation [91,92]. Another shunt from the methionine cycle is the transsulfuration pathway, which is important for cellular redox homeostasis. The high intracellular oxygen levels required for aerobic respiration create an environment that produces highly reactive ROS. While physiological levels of ROS are essential for cell survival, an excess of ROS can have a wide range of detrimental effects, including DNA and protein damage. To prevent damage, the cell tightly regulates a series of antioxidant systems to restore redox homeostasis. One of the major antioxidants made within cells is glutathione, which is a product of the transsulfuration shunt of one-carbon metabolism. Reduced glutathione acts a cofactor for redox and conjugation reactions catalyzed by glutathione peroxidases and glutathione transferases to reduce hydrogen peroxide, a reactive product of initial superoxide neutralization, and neutralize toxins and carcinogens. Interestingly, in multiple cancers, glutathione peroxidases and glutathione transferases are silenced by DNA hypermethylation suggesting that the reduced activity of the enzymes drives tumorigenesis, likely through increased DNA damage [93,94,95,96,97,98]. However, the overexpression of glutathione peroxidases and glutathione transferases, along with elevated levels of reduced glutathione, has been observed to correlate with therapy resistance in multiple cancers [99,100,101]. This evidence suggests that the glutathione-dependent reduction and neutralization reactions may have complex pro-tumor and anti-tumor effects by improving survival and reducing DNA damage. Interestingly, elevated homocysteine may promote oxidative stress by inhibiting the expression and activity of glutathione peroxidases. Elevated plasma homocysteine levels, a condition that may also be associated with folate deficiency, is often seen in the setting of malignancy [48,102,103,104,105,106,107]. In addition to being a metabolite that is utilized in glutathione synthesis, homocysteine regulates the activity of enzymes that use glutathione as a cofactor. By controlling glutathione synthesis and utilization, changes in one-carbon metabolism flux can have a profound impact on redox metabolism, and therefore, potentially tumorigenesis and cancer progression. Taken together, alterations in one-carbon metabolism may contribute to tumorigenesis by fueling DNA synthesis, changing the DNA and histone methylomes, promoting protein translation, driving cell cycle progression and modulating redox balance. These changes can in turn promote sustained proliferation, induce tumorigenic gene expression changes, contribute to genomic instability, and promote survival—all important processes in tumorigenesis and cancer progression. 4. Androgen Signaling Modulates One-Carbon Metabolism and Epigenetics In the prostate, androgens and the AR regulate the activity/expression of several enzymes involved in the one-carbon metabolism pathways, specifically enzymes involved in SAM homeostasis (GNMT and SARDH) and the entry into the transsulfuration (CBS) and polyamine synthesis (ODC) pathways (Figure 1 and Table 1). This suggests that the changes in the AR activity that occur during PCa progression may have profound effects on global one-carbon metabolism and the epigenetics of this disease. In this section, we review the role of androgens/AR signaling in these checkpoints of the one-carbon metabolism network, with an emphasis on the effect on gene expression and focusing on the best characterized genes. Based on the impact of the one-carbon metabolism in epigenetics, we will also discuss the effect of androgen signaling on the activity/expression of methyltransferases and epigenetic processes in PCa 5. Androgen Signaling Regulates the Expression of Enzymes Involved in the One-Carbon Metabolism Network The AR is a nuclear receptor that is essential for prostate differentiation and homeostasis and for PCa initiation and progression. Binding of androgen, its major ligand, triggers a conformational change that promotes AR homodimerization and translocation to the nucleus, where it binds to the regulatory regions of its target genes, affecting their transcription [108]. Studies directed to identify AR transcriptional networks in different models of PCa have demonstrated an involvement of the AR in global metabolism by directly targeting enzymes involved in several metabolic processes [12,13,109,110]. Below we focus on several specific AR targets involved in one-carbon metabolism and their role in PCa. 5.1. GNMT, SARDH and Sarcosine Metabolism GNMT catalyzes the transfer of a methyl group from SAM to glycine to form SAH and sarcosine. The reverse reaction involves the oxidative demethylation of sarcosine into glycine, and it is catalyzed by mitochondrial SARDH or peroxisomal PIPOX [19,111]. It has been proposed that the “sarcosine cycle” and GNMT in particular regulate the SAM:SAH ratio, and therefore the methylation potential of the cell [111]. Methyltransferases are inhibited by SAH [87], GNMT is allosterically inhibited by 5-methylTHF [30], and SAM inhibits MTHFR and therefore formation of 5-methylTHF [111]. When SAM levels are low, this regulatory loop promotes release of the inhibition of MTHFR, resulting in de novo synthesis of 5-methylTHF and therefore ensuring inhibition of GNMT so that SAM will be saved for physiologically essential methylation reactions. High levels of SAM block formation of 5-methylTHF, releasing the inhibition of GNMT, which will convert excess SAM into sarcosine [111]. Because of the relevance of GNMT and the sarcosine cycle in methylation, changes in their expression or activity can have profound effects in essential cellular processes. The AR and the TMPRSS2-ERG fusion product (present in over 50% of localized PCa and whose expression is controlled by the AR) are known to coordinately regulate GNMT and SARDH expression [20,21]. Therefore, as expression/activity of these transcription factors changes with disease progression, so does the methylation potential of the cell. In fact, the role of GNMT and SARDH in PCa has gained recent interest, as both are dysregulated during tumorigenesis and control the metabolism of sarcosine. Sarcosine is a metabolite that increases during PCa progression to metastasis, and has been proposed as a potential non-invasive urine biomarker [20]. Using PCa cell lines, Sreekumar et al. [20] demonstrated that the enzymes involved in sarcosine metabolism act as regulators of cell invasion and are therefore potential therapeutic targets for prostate cancer. The addition of sarcosine or knockdown of SARDH in benign prostate epithelial cells enhanced their invasiveness. Recently, we demonstrated that sarcosine metabolism, not merely its concentration, and thus one-carbon availability, is responsible for the changes in invasion observed in PCa cells [18]. While controversy remains regarding whether the levels of GNMT in clinical PCa samples are downregulated [112] or upregulated [113], it is clear that dysregulation of GNMT may reflect changes in AR activity and ERG fusion status during PCa establishment and progression. Metabolomic analyses indicate that androgen supplementation results in elevated amino acid metabolism and increased methylation activity in PCa cells [114,115]. Interestingly, in breast cancer, the expression of sarcosine-related enzymes has been shown to vary according to cancer subtype [115]. A parallel with GNMT could be established with studies conducted on Nicotinamide N-methyltransferase (NNMT) [116]. NNMT, which catalyzes the transfer of a methyl group from SAM to nicotinamide to generate 1-methylnicotinamide (1-MNA) and SAH, and its products, are overexpressed in several aggressive cancer cell lines (e.g., ovarian, lung, and kidney) and in clinical samples [117]. Similar to sarcosine, 1-MNA does not have a known physiological role, but has been proposed to act as a sink for methyl groups, reducing the SAM:SAH ratio and the methylation potential of the cell [116]. The authors demonstrated that NNMT overexpression led to decreased methylation of proteins including histones, and associated changes in gene expression. It is possible that when GNMT is overexpressed and SARDH is underexpressed or its activity is decreased (as previously postulated for aggressive behavior in PCa; [20]), overproduction of sarcosine can exert a similar “methyl sink” effect. In this regard, we have previously demonstrated that the transmembrane protein with epidermal growth factor and two follistatin domains 2 (TMEFF2) is a tumor suppressor that cooperates with SARDH to modulate one-carbon metabolism in PCa cells [18,118] suggesting that additional factors may play a role in the activity of these enzymes. Metabolic changes in a TMEFF2 transgenic mouse model support this conclusion [119]. 5.2. CBS and the Transsulfuration Pathway As discussed above, homocysteine can enter the transsulfuration pathway in a reaction that involves condensation with serine, resulting in cystathionine. In mammals, this first and committed step of the pathway is catalyzed by CBS. The second step, the hydrolysis of cystathionine to cysteine, is catalyzed by the enzyme γ-cystathionase [120]. Cysteine is a limiting factor for glutathionine synthesis, but can also be catabolized via other routes, including a non-oxidative route that produces hydrogen sulfide (H2S). H2S plays a role in the regulation of many physiological processes, such as the cellular stress response, inflammation and energy metabolism [121,122,123,124], and it modulates AR activity [125]. Based on its roles in homocysteine homeostasis and H2S and glutathione generation, altered CBS activity/expression contributes to numerous diseases, including cancer [126,127,128]. The activity of CBS is stimulated by SAM binding [31,32,129], so that homocysteine metabolism can be directed towards remethylation when methionine/SAM levels are low, and towards the transsulfuration pathway when SAM levels are high. Studies using LNCaP, an androgen-dependent prostate cancer cell line, suggest that CBS expression may be downregulated by androgens via a currently unknown posttranscriptional mechanism and that this effect is accompanied by a decrease in glutathionine levels [130,131]. Reduced levels of CBS have also been reported in the metastatic PCa cell line PC3. However, this cell line does not express the AR, and the low levels of CBS did not seem to correlate with the cancer phenotype [132]. In addition, lower levels of plasma cysteine have been observed as a result of prostate tumor progression in mouse xenografts [133]. The above findings that suggest an impaired flux through the transsulfuration pathway in PCa are not supported by clinical metabolomic data. In a study analyzing metabolite levels in serum of patients who developed recurrent disease after primary treatment vs. patients that remained recurrence-free, the levels of homocysteine and cystathionine were significantly higher in the recurrent group than in the recurrence-free group [134]. Increased levels of homocysteine and methylated metabolites, with concomitant decrease in SAM, were observed in androgen-responsive PCa cells when compared with PCa cells that were non-responsive to androgens [114]. The levels of H2S are also significantly higher in patients with localized PCa than in patients with benign prostatic hyperplasia or healthy individuals [135]. These results suggest an androgen-mediated increase of methylation activity and an increased flux through the transsulfuration pathway in PCa and with the progression to aggressive disease. Reconciling these seemingly opposite results requires determining the role of transsulfuration metabolites in cancer, analyzing differences in methylation potential across individuals, and establishing the role of SAM and androgen signaling changes with disease progression in the one-carbon metabolism and transsulfuration pathways. As we discussed earlier, the AR plays a role in regulation of GNMT. Thus, in modulating GNMT activity, the AR indirectly control homocysteine levels and the SAM:SAH ratio, critical to methylation reactions and to the level of CBS. H2S inhibits the activity of the AR [125] providing a feedback loop by which excess cysteine and, therefore H2S, modulates AR activity and the methylation and transsulfuration pathways. In hepatic and lymphocytic cells, androgens have been demonstrated to regulate expression of glutathione S-transferase Pi (GSTP), an enzyme with a role in detoxification, by catalyzing the conjugation of many compounds to reduced glutathione [136,137,138]. Consequently, the AR can play a role in detoxification not only by regulating CBS levels, and thus glutathione, but also by regulating/modulating the activity of enzymes that act downstream of glutathione. Changes in ROS are known to have a role in the etiology and progression of PCa [139]. 5.3. ODC, SAM and Polyamine Synthesis The relevance of polyamines to cellular physiology is illustrated by the fact that knockout of several enzymes of the pathway are embryonic lethal in the mouse [140] and dysregulation of polyamine metabolism leads to disease [141]. Increased levels of polyamine synthesis and ODC levels have been associated with cancer and other hyperproliferatives diseases [37,91,142,143,144]. ODC catalyzes the initial and rate limiting step in the biosynthesis of polyamines, a conversion of ornithine to putrescine. Sequential reactions catalyzed by spermidine and spermine synthase convert putrescine into spermidine and spermine, respectively. These reactions require dcSAM, which is obtained from the decarboxylation of SAM in a reaction catalyzed by SAM-decarboxylase (AMD1; Figure 1). The prostate has exceptionally high levels of polyamines, which are synthesized in the epithelium for normal growth and for secretion into the seminal fluid [26,38,145,146,147,148]. The high level is due, in part, to the high expression of ODC and AMD1 [145,146,149,150,151]. Both enzymes, together with spermidine synthase, are induced transcriptionally by androgens/AR signaling in the prostate in a coordinated way [152,153,154,155,156]. Moreover, ODC is higher in PCa than in benign tissue, tissue from patients with benign prostate hyperplasia (BPH), or tissue from normal volunteers [149,157], indicating that changes that occur to the AR during PCa progression affect enzyme levels and polyamine synthesis. Providing further evidence for this notion, androgen-blocking therapies, inhibit production of spermine and spermidine [158,159]. The high polyamine requirements observed in the prostate, which are increased in PCa, sensitize the cells to folate levels [160]. Blocking polyamine synthesis by inhibiting AMD1 increases SAM levels and reduces the sensitivity to low levels of folate [160]. Interestingly, mild folate deficiency does not negatively impact polyamine levels, but does affect DNA methylation and cell growth, suggesting that maintaining polyamine pools is favored over maintaining SAM pools [63,160]. Due to the high demand for polyamines in the prostate and in PCa, changes in AR-mediated polyamine biosynthesis enzyme levels can create an imbalance in SAM levels and nucleotide pools, having profound effects on DNA damage, DNA methylation, and other epigenetic changes, leading to tumorigenesis and/or playing a role in disease progression (see below). 6. The Role of Androgen Signaling on Methyltransferases and the Epigenetics of PCa DNA and histone methylation are important epigenetic mechanisms that contribute to initiation and progression of PCa [75,161,162,163,164]. Based on the link between these epigenetic mechanisms and one-carbon metabolism, in this section we briefly review the role of methylation in PCa and discuss how the AR modulates the epigenetics of PCa, indirectly controlling one-carbon metabolism and directly affecting the expression and activity of methyltransferases. 6.1. DNA and Histone Methyltransferases in PCa In PCa, changes in DNA methylation are detected before the cancer becomes invasive and are maintained throughout disease progression [165,166]. These observations underscore the relevance of epigenetic mechanisms to PCa and suggest that epigenetic changes are early events that may even be responsible for PCa tumor initiation. The best-characterized epigenetic alteration in PCa is gene-specific DNA hypermethylation [167,168]. Aberrant hypermethylation of numerous genes including cell cycle control genes, detoxification and genes involved in apoptosis and DNA repair [166,167,168,169,170,171,172,173,174,175,176] and the AR itself [172,177,178,179,180] has been described. Correspondingly, expression and activity of DNA methyltransferase 1 (DNMT1), the methyltransferase that is primarily responsible for maintaining the DNA methylation pattern, is higher in localized, metastatic, and hormone-resistant PCa samples than in benign prostate hyperplasia (BPH) or normal tissue. DNMT1 level can predict disease recurrence after prostatectomy [167,168,175,180,181,182,183]. Changes in the level of DNMT1 with disease progression have also been reported in studies using the TRAMP mouse model [184]. Using this model, it was also demonstrated that inhibition of DNMT1 by 5-azacitidine treatment prevented tumorigenesis [185], underscoring the relevance of DNMT1 and hypermethylation to the establishment and progression of PCa. Expression of other enzymes involved in regulating DNA methylation (DNMT3, MBD4) is also increased in PCa and metastatic disease [74,186]. It is important to point out that global and gene specific hypomethylation changes are also associated with increased Gleason score and metastatic disease [161,187]. In Alzheimer’s disease, demethylase activity is affected by one-carbon metabolism (SAM:SAH ratio) [188], however, to our knowledge, similar studies have not been conducted in PCa. Histone methylation changes are also common in PCa. Studies using immunohistochemical methods have reported an overexpression of H3K27me3 global levels in metastatic prostate tumors compared with non-malignant prostate tissues [72]. Although other histone methylation changes have been reported in PCa, changes in H3K27 methylation are receiving more attention since EZH2, the histone methyltransferase responsible for H3K27 methylation, is overexpressed during prostate tumorigenesis and is associated with biochemical recurrence in patients with PCa [78,80,189,190,191,192,193]. Upregulation of EZH2 is associated with repression of tumor suppressor genes, high proliferation rates, and increased tumor aggressiveness in PCa [78]. It is also directly involved in DNA methylation through interaction with DNA methyltransferases [190,194], and can target genes for de novo methylation in cancer [195]. Although this review focusses on processes that are affected by one-carbon metabolism, changes in demethylases are also relevant to PCa. Several reviews have been recently published [196,197,198]. 6.2. Androgen Signaling Regulates the Expression and Activity of Methyltransferases in PCa DNA and histone methyltransferases utilize SAM as substrate leading to SAH production, an inhibitor of methyltransferases [199]. Therefore, methylation reactions are largely dictated by the SAM:SAH ratio and the level of expression of the methyltransferases. For in depth coverage of the effect of SAM levels on activity and specificity of methyltransferases, the reader is referred to an excellent review by Mentch and Locasale [16]. The AR and androgen signaling play a role in controlling methylation, modulating the expression of methyltransferases and/or their activity. As discussed above, the AR has important roles in regulating GNMT and the metabolism of sarcosine and the enzymes involved in the diversion of methyl groups into the transsulfuration and polyamine synthesis pathways. Increased GNMT expression leads to increased levels of the methyltransferase inhibitor SAH [200]. Therefore, changes in AR activity indirectly affect methyltransferase activity by modulating the SAM:SAH ratio. In clinical samples of PCa, increased GNMT expression significantly correlates with high Gleason score and reduced disease-free survival [113]. These effects could partly be due to inhibition of DNA and histone methylation. Global DNA hypomethylation has been correlated with high Gleason score and metastatic PCa [161,187]. In addition to this indirect effect, the AR has a direct effect on methyltransferase activity by binding to these enzymes and, in some cases, promoting their recruitment into specific regions on the chromatin. For example, the AR interacts with and recruits EHZ2, increasing H3K27 methylation and epigenetic silencing and leading to oncogenic transformation [201,202]. Similarly, using the protein Menin as a bridge, the AR recruits MLL, a SET-like H3K4 histone methyltransferase [84], promoting AR-mediated transcription [203]. Interaction of the AR with demethylases has also been described. For example, JHDM2A, a H3K9 demethylase, binds to and is recruited to AR target genes upon androgen stimulation, resulting in H3K9 demethylation and transcriptional activation [204]. Similarly, the AR can directly interact with LSD1 on many AR-repressed genes. LSD1 is a lysine demethylase that has a repressive function by demethylating H3K4me1 and H3K4me2 in response to androgen [205]. Interestingly, the AR is a target for LSD1. Since DNA/chromatin methylation influences AR activity, these examples illustrate the fact that by modulating methyltransferase/demethylase activity and/or expression, the AR can also control its own expression and/or activity (Figure 3). Changes in the one-carbon metabolism affecting methyltransferases (SAM:SAH ratio) also modulate DNA and chromatin methylation affecting the activity and/or expression of the AR. In addition, direct post-translational modification of the AR and/or co-activators by methyltransferases also occurs. The AR is directly methylated by the histone methyltransferase SET9 on lysine K632 resulting in enhanced transcriptional activity [206]. Interestingly, in CRPC, EZH2 functions as a coactivator for transcription factors including the AR. The activating function of EZH2 requires the methyltransferase domain, and it has been suggested that it functions by altering the AR-associated lysine methylation [207]. Finally, AR signaling can regulate expression of enzymes involved in histone methylation. It has been reported that androgens modulate expression of EZH2 in a concentration-dependent manner (EZH2 is repressed at 1 nM or higher). This effect requires a functional AR and is mediated by the binding of retinoblastoma (RB) and p130-associated proteins to the EZH2 promoter. While both mechanisms seem to be synergistic, their androgen dependence varies. RB-E2F1 are themselves regulated by androgens in PCa cells. p130 and its partner proteins bind to the EZH2 promoter in androgen-treated, but not in control treated cells [208,209]. Finally, expression of EZH2 can be repressed by miRNA101, which is regulated by androgens [210,211]. In summary, AR/androgen signaling has an important role in PCa epigenetics, both indirectly by controlling expression of key enzymes involved in one-carbon metabolism and associated pathways, and directly by controlling the expression and activity of DNA and histone methyltransferases. These effects can ultimately affect AR expression, which is also epigenetically controlled by DNA and histone methylation, or activity. These observations emphasize the precise link between the AR and one-carbon metabolism, and the potential effects that changes in AR signaling, that can occur with disease progression, may have on essential cellular processes (Figure 3). 7. Therapeutic Approaches to Prostate Cancer: Targeting the One-Carbon Metabolism The accelerated proliferation of cancer cells places a robust demand on one-carbon metabolism, which can be exploited for anticancer therapies. The antifolate, aminopterin, originally used by Sydney Farber to treat pediatric patients with acute lymphoblastic leukemia, was the first successful anticancer chemotherapeutic agent [212]. Today, multiple drugs targeting enzymes within the folate cycle are FDA-approved to treat a variety of cancer types [15]; however, these drugs have had mixed reports for the treatment of PCa. While early studies indicated that the antifolate, MTX, might have been beneficial in the treatment of CRPC, subsequent studies failed to support the original findings [213,214,215]. Because AR inhibition during ADT decreases polyamine synthesis, which may in turn increase methyl group availability in the folate cycle, it has been suggested that MTX may be more beneficial in the treatment of PCa at earlier stages of the disease [63,160]. Other branches of the one-carbon metabolism network have been explored as therapeutic targets. As we discussed previously, the natural polyamines, putrescine, spermine and spermidine are ubiquitous molecules; however, their requirements are particularly high in rapidly growing tissues during normal growth and development, and in tumors [37,216,217,218]. Several reports have described increased polyamine levels in the blood and/or urine of cancer patients [219,220,221,222,223] and elevated levels correlate with more advanced disease and worse prognosis [216,224,225,226,227]. Increased polyamine levels are associated with increased cell proliferation, decreased apoptosis and increased expression of genes affecting tumor invasion and metastasis [37,228]. More recently, it has been shown that increased polyamine levels indirectly lead to immunosuppressive conditions facilitating tumor spread [228]. Changes in polyamine levels have been reported in PCa [143,218,225,229,230,231]. Underscoring the clinical relevance of polyamines to prostate cancer, preclinical data suggest that inhibition of polyamine synthesis blocks the progression of the disease [232,233,234,235,236,237]. All together these observations validate the polyamine pathway as chemopreventive and chemotherapeutic for PCa. Several trials have focused on targeting the polyamine pathway as a strategy for chemoprevention in patients at risk for aggressive PCa using difluoromethylornithine (DFMO), an inhibitor of ODC [238,239]. The results of those trials indicated that DMFO treatment results in decreased levels of putrescine, decreased rate of prostate growth, and a trend towards decreased PSA doubling time. A recent clinical trial demonstrated that DFMO caused nearly complete depletion of putrescine (97.6%) but not of spermidine and spermine (73.6% and 50.8%, respectively) [150], and while very well tolerated [143], it seemed to be largely ineffective as a chemotherapeutic agent. The lack of effectiveness could be in part due to compensatory mechanisms such as increased polyamine uptake from circulation, or upregulation of other enzymes involved in the pathway. Supporting this, it was shown that polyamine reduced diet induced or maintained the quality of life of patients with CRPC [151]. In addition, studies in cell lines and xenografts indicate increased efficacy when using DMFO in combination with polyamine transport inhibitors [234]. Increased levels of SAM-dc [142] and spermine synthase [150] have been observed in patients with PCa. Other pathway inhibitors including polyamine analogs [142,143,240] or SAM-dc inhibitors [143] have been previously pursued in clinical trials; however, they have demonstrated high toxicity or only partial responses. In addition to drugs targeting one-carbon metabolism itself, methyltransferase inhibitors are also used to treat a variety of cancers, and several inhibitors are currently being investigated for cancer therapy [241]. 5′Azacitidine is a DNA methyltransferase inhibitor that is commonly used to treat myelodysplastic syndromes [242]. Importantly, epigenetic alterations, including DNA methylation, have been found to play an important role in therapy resistance, and 5′Azacitidine, and other demethylating agents, have been shown to be effective in combination therapy to improve chemosensitivity in other cancer types [243,244,245,246,247,248]. In PCa, for example, 5′Azacitidine improved chemosensitivity to docetaxel in patients with metastatic CRPC in phase I/II clinical trials [248]. Histone methyltransferases are also prime targets in epigenetic cancer therapy. EZH2, MLL, and DOT1L are potentially attractive targets in PCa, as all three modulate the activity of the AR [203,207,249]. MI-503 (MLL inhibitor) inhibits AR activity, and both DZNeP and MI-503 inhibit CRPC growth in mouse xenograft models [203,250]. Because EZH2 is overexpressed in metastatic CRPC and it drives a transcriptional signature that is associated with this stage of the disease, the potential use of EZH2 inhibitors in the treatment of CRPC is of particular interest [207,251]. Furthermore, EZH2 seems to have a role in both AR positive and AR negative CRPC, making EZH2 a versatile potential target in advanced PCa [207,251]. It is possible that therapeutics targeting one-carbon metabolism could work synergistically with direct methyltransferase inhibition to block the oncogenic functions of EZH2 and/or other methyltransferases in CRPC; however, this hypothesis remains to be tested. 8. Summary and Conclusions The one-carbon metabolism network integrates several pathways that, together, play central roles in the biosynthesis of nucleic acids and lipids, amino acid and vitamin metabolism, the maintenance of redox status, methylation reactions and polyamine biogenesis. Because of the relevance of these pathways to cell growth and proliferation, they are critical not only for cellular homeostasis but also for tumorigenesis, and are therefore significant therapeutic targets. The tight dependency among the pathways of the one-carbon metabolism network imposes an exquisite regulation to allow rapid responses to changes in cellular demands. The AR and androgen signaling regulate key enzymes involved in these pathways, including the ones that control the methylation potential of the cell and the entrance into the glutathione and polyamine biosynthetic pathways. Therefore, changes that occur in the AR levels or activity will have profound effects on the activity and output of the one-carbon metabolism network and downstream processes (Figure 4). ADT designed to decrease the levels of circulating androgens, or AR-directed therapies, are the mainstay treatments against advanced PCa, and are also used as adjuvants for local treatment of high risk disease. Because of their effect on AR signaling, these therapies affect the balance of the one-carbon metabolism. For example, it has been described that neo-adjuvant androgen blockade using an LHRH agonist, together with an anti-androgen, leads to decreased spermine and spermidine levels of the normal glands [158]. While in some instances the ADT-mediated effect on one-carbon metabolism may be beneficial, i.e., lowering the high levels of polyamines observed in cancer cells may help decrease their proliferative capacity, it is conceivable that it may also have a detrimental outcome. For instance, the blockade of polyamine synthesis would alter the flux of methyl groups toward other branches of the one-carbon metabolism network including the folate cycle, which potentially may lead to reduced sensitivity to the anti-folate methotrexate as discussed previously [63,160]. In addition, ADT or AR blockade would reduce the levels of GNMT, leading to increased SAM/SAH ratios and methyltransferase activity, a condition that maybe conducive to aggressive PCa (i.e., increased EZH2 levels in CRPC [78,191,193]). Finally, since the AR negatively regulates expression of CBS, an AR-signaling blockade would increase the flux towards the transsulfuration pathway, an effect that has been linked with increased therapeutic resistance [99,100,101]. Taken together these observations point to potential detrimental effects of ADT on one-carbon metabolism flux, and suggest that combination drug therapy in a precise order and timing may be helpful in the design of future clinical trials, and critical for successful treatment of PCa patients. Acknowledgments The authors are thankful to Kathy J. Kyler for editing and reviewing this manuscript. We thank the Stephenson Cancer Center at the University of Oklahoma, Oklahoma City, OK for funding support. The authors wish to apologize to all the colleagues whose work has not been cited in this manuscript due to space limitations. Conflicts of Interest The authors declare no conflict of interest. Figure 1 One-carbon metabolism and associated pathways. One-carbon metabolism involves the transfer of methyl groups to various substrates and cofactors within the folate and methionine cycles, and the polyamine biosynthetic and transsulfuration pathways. Methyl groups are utilized in the synthesis of nucleotides, and polyamines, as well as, DNA and protein methylation reactions. Enzymes are depicted in bold, while metabolites/substrates/cofactors are in regular font. Enzyme abbreviations are as follows: DHFR: Dihydrofolate reductase; SARDH: Sarcosine Dehydrogenase; SHMT: Serine hydroxymethyltransferase; GLDC: Glycine decarboxylase; GNMT: Glycine-N-methyltransferase; MTHFR: Methylene tetrahydrofolate reductase; MS: Methionine synthase; MAT: Methionine adenosyltransferase; AMD1: Adenosylmethionine decarboxylase; ODC: Ornithine decarboxylase; AHCY: S-adenosylhomocysteine hydrolase; CBS: Cystathionine beta-synthase. Figure 2 Regulation of the one-carbon metabolism maintains a balanced flux between the folate and methionine cycles and associated pathways. Metabolites produced within the folate and methionine pathways regulate the activity of the enzymes within the one-carbon metabolism network to maintain the balance of methyl groups and metabolites within the folate and methionine cycles and associated pathways and to allow for changes in response to cellular demands or growth conditions. See text for details. Enzymes are in bold, and substrates/cofactors are depicted in regular font. Black arrows indicate the directionality of reactions, red lines indicate inhibition, and the green arrow indicates activation. Enzyme abbreviations are as follows: SARDH: Sarcosine dehydrogenase; MTHFR: Methylene tetrahydrofolate reductase; CBS: Cystathionine beta-synthase; MTFs: Methyltransferases; GNMT: Glycine-N-methyltransferase. Figure 3 The impact of AR in prostate epigenetics. AR/androgen signaling can control the prostate epigenome: (1) indirectly by controlling expression of key enzymes involved in one-carbon metabolism and therefore the methylation potential of the cell (broken lines); or (2) directly by controlling the expression and activity of DNA and histone methyltransferases (solid lines). Figure 4 The AR impacts one-carbon metabolism and downstream processes by modulating the expression of specific associated enzymes. Green and red arrows/lines indicate reactions that are respectively activated and repressed by the AR via the modulation of enzyme expression. Enzymes include ODC1 (involved in polyamine synthesis), GNMT (catalyzes the conversion of glycine to sarcosine) and CBS (involved in glutathione synthesis). Metabolites increased and decreased by the AR in this manner are indicated in green and red, respectively. Black color depicts the one-carbon metabolism pathways and some other metabolites derived from it. Cellular processes potentially affected by AR-mediated one-carbon metabolism modulation are indicated in blue. By inhibiting AR, ADT (grey box) likely blocks many of these metabolic alterations in PCa cells, thus producing potentially beneficial and deleterious effects as discussed in the text. ijms-17-01208-t001_Table 1Table 1 Androgen responsive genes of the one carbon metabolism network. * Data from the androgen responsive gene database (available online: http://argdb.fudan.edu.cn/index_info.php). MTHFD: methylenetetrahydrofolate dehydrogenase; SMS: spermine synthase; SAT: spermine/spermidine N1-acetyl transferase; CTH: cystathione gamma-lyase or gamma-cystathionase; GPX: glutathione peroxidase; GSR: glutathione reductase. Other names as in legend of Figure 1 and Figure 2. Androgen Responsive Genes * Folate Methionine Polyamine Transulfuration SHMT MAT ODC CBS SARDH AHCY SMS CTH GLDC SAT GPX DHFR AMD1 GSR MTHFD MTHFR GNMT ==== Refs References 1. Torre L.A. Bray F. Siegel R.L. Ferlay J. Lortet-Tieulent J. Jemal A. Global cancer statistics, 2012 CA Cancer J. Clin. 2015 65 87 108 10.3322/caac.21262 25651787 2. De Winter R.J.A. Janssen P.J. Sleddens H.M. Verleun-Mooijman M.C. Trapman J. Brinkmann A.O. Santerse A.B. Schroder F.H. van der Kwast T.H. Androgen receptor status in localized and locally progressive hormone refractory human prostate cancer Am. J. Pathol. 1994 144 735 746 7512791 3. Chodak G.W. Kranc D.M. Puy L.A. Takeda H. Johnson K. Chang C. Nuclear localization of androgen receptor in heterogeneous samples of normal, hyperplastic and neoplastic human prostate J. Urol. 1992 147 798 803 1371552 4. Sadi M.V. Walsh P.C. Barrack E.R. Immunohistochemical study of androgen receptors in metastatic prostate cancer. Comparison of receptor content and response to hormonal therapy Cancer 1991 67 3057 3064 10.1002/1097-0142(19910615)67:12<3057::AID-CNCR2820671221>3.0.CO;2-S 1710537 5. Hammerer P. Madersbacher S. Landmarks in hormonal therapy for prostate cancer BJU Int. 2012 110 Suppl. 1 23 29 10.1111/j.1464-410X.2012.11431.x 23046037 6. Scher H.I. Beer T.M. Higano C.S. Anand A. Taplin M.E. Efstathiou E. Rathkopf D. Shelkey J. Yu E.Y. Alumkal J. Antitumour activity of MDV3100 in castration-resistant prostate cancer: A phase 1–2 study Lancet 2010 375 1437 1446 10.1016/S0140-6736(10)60172-9 20398925 7. Albertsen P.C. Hanley J.A. Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer JAMA 2005 293 2095 2101 10.1001/jama.293.17.2095 15870412 8. Merseburger A.S. Haas G.P. von Klot C.A. An update on enzalutamide in the treatment of prostate cancer Ther. Adv. Urol. 2015 7 9 21 10.1177/1756287214555336 25642291 9. Pienta K.J. Bradley D. Mechanisms underlying the development of androgen-independent prostate cancer Clin. Cancer Res. 2006 12 1665 1671 10.1158/1078-0432.CCR-06-0067 16551847 10. Sharma N.L. Massie C.E. Ramos-Montoya A. Zecchini V. Scott H.E. Lamb A.D. MacArthur S. Stark R. Warren A.Y. Mills I.G. The androgen receptor induces a distinct transcriptional program in castration-resistant prostate cancer in man Cancer Cell 2013 23 35 47 10.1016/j.ccr.2012.11.010 23260764 11. Wang Q. Li W. Zhang Y. Yuan X. Xu K. Yu J. Chen Z. Beroukhim R. Wang H. Lupien M. Androgen receptor regulates a distinct transcription program in androgen-independent prostate cancer Cell 2009 138 245 256 10.1016/j.cell.2009.04.056 19632176 12. Barfeld S.J. Itkonen H.M. Urbanucci A. Mills I.G. Androgen-regulated metabolism and biosynthesis in prostate cancer Endocr. Relat. Cancer 2014 21 T57 T66 10.1530/ERC-13-0515 24497572 13. Massie C.E. Lynch A. Ramos-Montoya A. Boren J. Stark R. Fazli L. Warren A. Scott H. Madhu B. Sharma N. The androgen receptor fuels prostate cancer by regulating central metabolism and biosynthesis EMBO J. 2011 30 2719 2733 10.1038/emboj.2011.158 21602788 14. Shafi A.A. Putluri V. Arnold J.M. Tsouko E. Maity S. Roberts J.M. Coarfa C. Frigo D.E. Putluri N. Sreekumar A. Differential regulation of metabolic pathways by androgen receptor (AR) and its constitutively active splice variant, AR-V7, in prostate cancer cells Oncotarget 2015 6 31997 32012 26378018 15. Locasale J.W. Serine, glycine and one-carbon units: Cancer metabolism in full circle Nat. Rev. Cancer 2013 13 572 583 10.1038/nrc3557 23822983 16. Mentch S.J. Locasale J.W. One-carbon metabolism and epigenetics: Understanding the specificity Ann. N. Y. Acad. Sci. 2016 1363 91 98 10.1111/nyas.12956 26647078 17. Tibbetts A.S. Appling D.R. Compartmentalization of mammalian folate-mediated one-carbon metabolism Annu. Rev. Nutr. 2010 30 57 81 10.1146/annurev.nutr.012809.104810 20645850 18. Green T. Chen X. Ryan S. Asch A.S. Ruiz-Echevarria M.J. TMEFF2 and SARDH cooperate to modulate one-carbon metabolism and invasion of prostate cancer cells Prostate 2013 73 1561 1575 10.1002/pros.22706 23824605 19. Khan A.P. Rajendiran T.M. Ateeq B. Asangani I.A. Athanikar J.N. Yocum A.K. Mehra R. Siddiqui J. Palapattu G. Wei J.T. The role of sarcosine metabolism in prostate cancer progression Neoplasia 2013 15 491 501 10.1593/neo.13314 23633921 20. Sreekumar A. Poisson L.M. Rajendiran T.M. Khan A.P. Cao Q. Yu J. Laxman B. Mehra R. Lonigro R.J. Li Y. Metabolomic profiles delineate potential role for sarcosine in prostate cancer progression Nature 2009 457 910 914 10.1038/nature07762 19212411 21. Ottaviani S. Brooke G.N. O’Hanlon-Brown C. Waxman J. Ali S. Buluwela L. Characterisation of the androgen regulation of glycine N -methyltransferase in prostate cancer cells J. Mol. Endocrinol. 2013 51 301 312 10.1530/JME-13-0169 23997240 22. Johansson M. van Guelpen B. Vollset S.E. Hultdin J. Bergh A. Key T. Midttun O. Hallmans G. Ueland P.M. Stattin P. One-carbon metabolism and prostate cancer risk: Prospective investigation of seven circulating B vitamins and metabolites Cancer Epidemiol. Biomark. Prev. 2009 18 1538 1543 10.1158/1055-9965.EPI-08-1193 19423531 23. Appling D.R. Compartmentation of folate-mediated one-carbon metabolism in eukaryotes FASEB J. 1991 5 2645 2651 1916088 24. Fox J.T. Stover P.J. Folate-mediated one-carbon metabolism Vitam. Horm. 2008 79 1 44 18804690 25. Scotti M. Stella L. Shearer E.J. Stover P.J. Modeling cellular compartmentation in one-carbon metabolism Wiley Interdiscip. Rev. Syst. Biol. Med. 2013 5 343 365 10.1002/wsbm.1209 23408533 26. Williams-Ashman H.G. Canellakis Z.N. Polyamines in mammalian biology and medicine Perspect. Biol. Med. 1979 22 421 453 10.1353/pbm.1979.0013 471697 27. Saini P. Eyler D.E. Green R. Dever T.E. Hypusine-containing protein eIF5A promotes translation elongation Nature 2009 459 118 121 10.1038/nature08034 19424157 28. Stover P.J. One-carbon metabolism-genome interactions in folate-associated pathologies J. Nutr. 2009 139 2402 2405 10.3945/jn.109.113670 19812215 29. Luka Z. Methyltetrahydrofolate in folate-binding protein glycine N -methyltransferase Vitam. Horm. 2008 79 325 345 18804700 30. Wagner C. Briggs W.T. Cook R.J. Inhibition of glycine N -methyltransferase activity by folate derivatives: Implications for regulation of methyl group metabolism Biochem. Biophys. Res. Commun. 1985 127 746 752 10.1016/S0006-291X(85)80006-1 3838667 31. Prudova A. Bauman Z. Braun A. Vitvitsky V. Lu S.C. Banerjee R. S -adenosylmethionine stabilizes cystathionine β-synthase and modulates redox capacity Proc. Natl. Acad. Sci. USA 2006 103 6489 6494 10.1073/pnas.0509531103 16614071 32. Pey A.L. Majtan T. Sanchez-Ruiz J.M. Kraus J.P. Human cystathionine β-synthase (CBS) contains two classes of binding sites for S -adenosylmethionine (SAM): Complex regulation of CBS activity and stability by SAM Biochem. J. 2013 449 109 121 10.1042/BJ20120731 22985361 33. Bjelakovic G. Pavlovic D. Jevtovic T. Stojanovic I. Sokolovic D. Bjelakovic G.B. Nikolic J. Basic J. Vitamin B-12 and folic acid effects on polyamine metabolism in rat liver Pteridines 2006 17 90 94 10.1515/pteridines.2006.17.3.90 34. Sun D. Wollin A. Stephen A.M. Moderate folate deficiency influences polyamine synthesis in rats J. Nutr. 2002 132 2632 2637 12221222 35. Ott M. Gogvadze V. Orrenius S. Zhivotovsky B. Mitochondria, oxidative stress and cell death Apoptosis 2007 12 913 922 10.1007/s10495-007-0756-2 17453160 36. Ozben T. Oxidative stress and apoptosis: Impact on cancer therapy J. Pharm. Sci. 2007 96 2181 2196 10.1002/jps.20874 17593552 37. Gerner E.W. Meyskens F.L. Polyamines and cancer: Old molecules, new understanding Nat. Rev. Cancer 2004 4 781 792 10.1038/nrc1454 15510159 38. Heby O. Role of polyamines in the control of cell proliferation and differentiation Differentiation 1981 19 1 20 10.1111/j.1432-0436.1981.tb01123.x 6173280 39. Zhang W.C. Shyh-Chang N. Yang H. Rai A. Umashankar S. Ma S. Soh B.S. Sun L.L. Tai B.C. Nga M.E. Glycine decarboxylase activity drives non-small cell lung cancer tumor-initiating cells and tumorigenesis Cell 2012 148 259 272 10.1016/j.cell.2011.11.050 22225612 40. Rahman L. Voeller D. Rahman M. Lipkowitz S. Allegra C. Barrett J.C. Kaye F.J. Zajac-Kaye M. Thymidylate synthase as an oncogene: A novel role for an essential DNA synthesis enzyme Cancer Cell 2004 5 341 351 10.1016/S1535-6108(04)00080-7 15093541 41. Suzuki M. Tsukagoshi S. Saga Y. Ohwada M. Sato I. Enhanced expression of thymidylate synthase may be of prognostic importance in advanced cervical cancer Oncology 1999 57 50 54 10.1159/000012000 10394125 42. Shintani Y. Ohta M. Hirabayashi H. Tanaka H. Iuchi K. Nakagawa K. Maeda H. Kido T. Miyoshi S. Matsuda H. New prognostic indicator for non-small-cell lung cancer, quantitation of thymidylate synthase by real-time reverse transcription polymerase chain reaction Int. J. Cancer 2003 104 790 795 10.1002/ijc.11014 12640689 43. Pestalozzi B.C. Peterson H.F. Gelber R.D. Goldhirsch A. Gusterson B.A. Trihia H. Lindtner J. Cortes-Funes H. Simmoncini E. Byrne M.J. Prognostic importance of thymidylate synthase expression in early breast cancer J. Clin. Oncol. 1997 15 1923 1931 9164203 44. Nomura T. Nakagawa M. Fujita Y. Hanada T. Mimata H. Nomura Y. Clinical significance of thymidylate synthase expression in bladder cancer Int. J. Urol. 2002 9 368 376 10.1046/j.1442-2042.2002.00479.x 12165018 45. Mizutani Y. Wada H. Yoshida O. Fukushima M. Nonomura M. Nakao M. Miki T. Significance of thymidylate synthase activity in renal cell carcinoma Clin. Cancer Res. 2003 9 1453 1460 12684419 46. Karlberg M. Ohrling K. Edler D. Hallstrom M. Ullen H. Ragnhammar P. Prognostic and predictive value of thymidylate synthase expression in primary colorectal cancer Anticancer Res. 2010 30 645 651 20332484 47. Longley D.B. Harkin D.P. Johnston P.G. 5-Fluorouracil: Mechanisms of action and clinical strategies Nat. Rev. Cancer 2003 3 330 338 10.1038/nrc1074 12724731 48. Zhang D. Wen X. Wu W. Guo Y. Cui W. Elevated homocysteine level and folate deficiency associated with increased overall risk of carcinogenesis: Meta-analysis of 83 case-control studies involving 35,758 individuals PLoS ONE 2015 10 1208 10.1371/journal.pone.0123423 25985325 49. Chen J. Gammon M.D. Chan W. Palomeque C. Wetmur J.G. Kabat G.C. Teitelbaum S.L. Britton J.A. Terry M.B. Neugut A.I. One-carbon metabolism, MTHFR polymorphisms, and risk of breast cancer Cancer Res. 2005 65 1606 1614 10.1158/0008-5472.CAN-04-2630 15735051 50. Giovannucci E. Epidemiologic studies of folate and colorectal neoplasia: A review J. Nutr. 2002 132 2350S 2355S 12163691 51. Chen P. Li C. Li X. Li J. Chu R. Wang H. Higher dietary folate intake reduces the breast cancer risk: A systematic review and meta-analysis Br. J. Cancer 2014 110 2327 2338 10.1038/bjc.2014.155 24667649 52. Kim Y.I. Pogribny I.P. Basnakian A.G. Miller J.W. Selhub J. James S.J. Mason J.B. Folate deficiency in rats induces DNA strand breaks and hypomethylation within the p53 tumor suppressor gene Am. J. Clin. Nutr. 1997 65 46 52 8988912 53. Blount B.C. Mack M.M. Wehr C.M. MacGregor J.T. Hiatt R.A. Wang G. Wickramasinghe S.N. Everson R.B. Ames B.N. Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: Implications for cancer and neuronal damage Proc. Natl. Acad. Sci. USA 1997 94 3290 3295 10.1073/pnas.94.7.3290 9096386 54. Ergul E. Sazci A. Utkan Z. Canturk N.Z. Polymorphisms in the MTHFR gene are associated with breast cancer Tumour Biol. J. Int. Soc. Oncodev. Biol. Med. 2003 24 286 290 10.1159/000076460 55. Kumar P. Yadav U. Rai V. Methylenetetrahydrofolate reductase gene c677t polymorphism and breast cancer risk: Evidence for genetic susceptibility Meta Gene 2015 6 72 84 10.1016/j.mgene.2015.08.008 26629412 56. Lu Q. Jiang K. Li Q. Ji Y.J. Chen W.L. Xue X.H. Polymorphisms in the MTHFR gene are associated with breast cancer risk and prognosis in a chinese population Tumour Biol. J. Int. Soc. Oncodev. Biol. Med. 2015 36 3757 3762 10.1007/s13277-014-3016-4 25566964 57. Maruti S.S. Ulrich C.M. Jupe E.R. White E. MTHFR C677T and postmenopausal breast cancer risk by intakes of one-carbon metabolism nutrients: A nested case-control study Breast Cancer Res. 2009 11 R91 10.1186/bcr2462 20030812 58. Wang Y. Yang H. Duan G. Mthfr gene a1298c polymorphisms are associated with breast cancer risk among chinese population: Evidence based on an updated cumulative meta-analysis Int. J. Clin. Exp. Med. 2015 8 20146 20156 26884927 59. Frosst P. Blom H.J. Milos R. Goyette P. Sheppard C.A. Matthews R.G. Boers G.J. den Heijer M. Kluijtmans L.A. van den Heuvel L.P. A candidate genetic risk factor for vascular disease: A common mutation in methylenetetrahydrofolate reductase Nat. Genet. 1995 10 111 113 10.1038/ng0595-111 7647779 60. Bester A.C. Roniger M. Oren Y.S. Im M.M. Sarni D. Chaoat M. Bensimon A. Zamir G. Shewach D.S. Kerem B. Nucleotide deficiency promotes genomic instability in early stages of cancer development Cell 2011 145 435 446 10.1016/j.cell.2011.03.044 21529715 61. Ehrlich M. DNA hypomethylation in cancer cells Epigenomics 2009 1 239 259 10.2217/epi.09.33 20495664 62. Poirier L.A. Methyl group deficiency in hepatocarcinogenesis Drug Metab. Rev. 1994 26 185 199 10.3109/03602539409029790 8082564 63. Bistulfi G. Foster B.A. Karasik E. Gillard B. Miecznikowski J. Dhiman V.K. Smiraglia D.J. Dietary folate deficiency blocks prostate cancer progression in the tramp model Cancer Prev. Res. 2011 4 1825 1834 10.1158/1940-6207.CAPR-11-0140 21836022 64. Bistulfi G. Vandette E. Matsui S. Smiraglia D.J. Mild folate deficiency induces genetic and epigenetic instability and phenotype changes in prostate cancer cells BMC Biol. 2010 8 1208 10.1186/1741-7007-8-6 20092614 65. Deghan Manshadi S. Ishiguro L. Sohn K.J. Medline A. Renlund R. Croxford R. Kim Y.I. Folic acid supplementation promotes mammary tumor progression in a rat model PLoS ONE 2014 9 1208 66. Lindzon G.M. Medline A. Sohn K.J. Depeint F. Croxford R. Kim Y.I. Effect of folic acid supplementation on the progression of colorectal aberrant crypt foci Carcinogenesis 2009 30 1536 1543 10.1093/carcin/bgp152 19541855 67. Song J. Medline A. Mason J.B. Gallinger S. Kim Y.I. Effects of dietary folate on intestinal tumorigenesis in the apcmin mouse Cancer Res. 2000 60 5434 5440 10.1016/S0016-5085(00)83190-X 11034085 68. Kim Y.I. Folate: A magic bullet or a double edged sword for colorectal cancer prevention? Gut 2006 55 1387 1389 10.1136/gut.2006.095463 16966698 69. Ulrich C.M. Potter J.D. Folate and cancer—Timing is everything JAMA 2007 297 2408 2409 10.1001/jama.297.21.2408 17551134 70. Greger V. Passarge E. Hopping W. Messmer E. Horsthemke B. Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma Hum. Genet. 1989 83 155 158 10.1007/BF00286709 2550354 71. Yoo C.B. Jones P.A. Epigenetic therapy of cancer: Past, present and future Nat. Rev. Drug Discov. 2006 5 37 50 10.1038/nrd1930 16485345 72. Ellinger J. Kahl P. von der Gathen J. Heukamp L.C. Gütgemann I. Walter B. Hofstädter F. Bastian P.J. von Ruecker A. Müller S.C. Global histone H3K27 methylation levels are different in localized and metastatic prostate cancer Cancer Investig. 2012 30 92 97 10.3109/07357907.2011.636117 22149091 73. Barlesi F. Giaccone G. Gallegos-Ruiz M.I. Loundou A. Span S.W. Lefesvre P. Kruyt F.A. Rodriguez J.A. Global histone modifications predict prognosis of resected non small-cell lung cancer J. Clin. Oncol. 2007 25 4358 4364 10.1200/JCO.2007.11.2599 17906200 74. Bianco-Miotto T. Chiam K. Buchanan G. Jindal S. Day T.K. Thomas M. Pickering M.A. O’Loughlin M.A. Ryan N.K. Raymond W.A. Global levels of specific histone modifications and an epigenetic gene signature predict prostate cancer progression and development Cancer Epidemiol. Biomark. Prev. 2010 19 2611 2622 10.1158/1055-9965.EPI-10-0555 20841388 75. Ellinger J. Kahl P. von der Gathen J. Rogenhofer S. Heukamp L.C. Gutgemann I. Walter B. Hofstadter F. Buttner R. Muller S.C. Global levels of histone modifications predict prostate cancer recurrence Prostate 2010 70 61 69 10.1002/pros.21038 19739128 76. Seligson D.B. Horvath S. McBrian M.A. Mah V. Yu H. Tze S. Wang Q. Chia D. Goodglick L. Kurdistani S.K. Global levels of histone modifications predict prognosis in different cancers Am. J. Pathol. 2009 174 1619 1628 10.2353/ajpath.2009.080874 19349354 77. Fraga M.F. Ballestar E. Villar-Garea A. Boix-Chornet M. Espada J. Schotta G. Bonaldi T. Haydon C. Ropero S. Petrie K. Loss of acetylation at LYS16 and trimethylation at LYS20 of histone H4 is a common hallmark of human cancer Nat. Genet. 2005 37 391 400 10.1038/ng1531 15765097 78. Bachmann I.M. Halvorsen O.J. Collett K. Stefansson I.M. Straume O. Haukaas S.A. Salvesen H.B. Otte A.P. Akslen L.A. EZH2 expression is associated with high proliferation rate and aggressive tumor subgroups in cutaneous melanoma and cancers of the endometrium, prostate, and breast J. Clin. Oncol. 2006 24 268 273 10.1200/JCO.2005.01.5180 16330673 79. Simon J.A. Lange C.A. Roles of the EZH2 histone methyltransferase in cancer epigenetics Mutat. Res. 2008 647 21 29 10.1016/j.mrfmmm.2008.07.010 18723033 80. Van Leenders G.J.L.H. Dukers D. Hessels D. van den Kieboom S.W.M. Hulsbergen C.A. Witjes J.A. Otte A.P. Meijer C.J. Raaphorst F.M. Polycomb-group oncogenes EZH2, BMI1, and RING1 are overexpressed in prostate cancer with adverse pathologic and clinical features Eur. Urol. 2007 52 455 463 10.1016/j.eururo.2006.11.020 17134822 81. Hyland P.L. McDade S.S. McCloskey R. Dickson G.J. Arthur K. McCance D.J. Patel D. Evidence for alteration of EZH2, BMI1, and KDM6A and epigenetic reprogramming in human papillomavirus type 16 E6/E7-expressing keratinocytes J. Virol. 2011 85 10999 11006 10.1128/JVI.00160-11 21865393 82. Lee J.Y. Kong G. Dot1l: A new therapeutic target for aggressive breast cancer Oncotarget 2015 6 30451 30452 26427043 83. Wong M. Polly P. Liu T. The histone methyltransferase DOT1L: Regulatory functions and a cancer therapy target Am. J. Cancer Res. 2015 5 2823 2837 26609488 84. Dou Y. Hess J.L. Mechanisms of transcriptional regulation by MLL and its disruption in acute leukemia Int. J. Hematol. 2008 87 10 18 10.1007/s12185-007-0009-8 18224408 85. Krivtsov A.V. Armstrong S.A. MLL translocations, histone modifications and leukaemia stem-cell development Nat. Rev. Cancer 2007 7 823 833 10.1038/nrc2253 17957188 86. Orzan F. Pellegatta S. Poliani P.L. Pisati F. Caldera V. Menghi F. Kapetis D. Marras C. Schiffer D. Finocchiaro G. Enhancer of zeste 2 (EZH2) is up-regulated in malignant gliomas and in glioma stem-like cells Neuropathol. Appl. Neurobiol. 2011 37 381 394 10.1111/j.1365-2990.2010.01132.x 20946108 87. Kerr S.J. Competing methyltransferase systems J. Biol. Chem. 1972 247 4248 4252 4338482 88. Shyh-Chang N. Locasale J.W. Lyssiotis C.A. Zheng Y. Teo R.Y. Ratanasirintrawoot S. Zhang J. Onder T. Unternaehrer J.J. Zhu H. Influence of threonine metabolism on S -adenosylmethionine and histone methylation Science 2013 339 222 226 10.1126/science.1226603 23118012 89. Huang W.Y. Yang P.M. Chang Y.F. Marquez V.E. Chen C.C. Methotrexate induces apoptosis through p53/p21-dependent pathway and increases e-cadherin expression through downregulation of HDAC/EZH2 Biochem. Pharmacol. 2011 81 510 517 10.1016/j.bcp.2010.11.014 21114963 90. Mandal S. Mandal A. Johansson H.E. Orjalo A.V. Park M.H. Depletion of cellular polyamines, spermidine and spermine, causes a total arrest in translation and growth in mammalian cells Proc. Natl. Acad. Sci. USA 2013 110 2169 2174 10.1073/pnas.1219002110 23345430 91. O’Brien T.G. Megosh L.C. Gilliard G. Soler A.P. Ornithine decarboxylase overexpression is a sufficient condition for tumor promotion in mouse skin Cancer Res. 1997 57 2630 2637 9205069 92. Nilsson J.A. Keller U.B. Baudino T.A. Yang C. Norton S. Old J.A. Nilsson L.M. Neale G. Kramer D.L. Porter C.W. Targeting ornithine decarboxylase in MYC-induced lymphomagenesis prevents tumor formation Cancer Cell 2005 7 433 444 10.1016/j.ccr.2005.03.036 15894264 93. Lee O.J. Schneider-Stock R. McChesney P.A. Kuester D. Roessner A. Vieth M. Moskaluk C.A. El-Rifai W. Hypermethylation and loss of expression of glutathione peroxidase-3 in barrett’s tumorigenesis Neoplasia 2005 7 854 861 10.1593/neo.05328 16229808 94. Zhang X. Yang J.J. Kim Y.S. Kim K.Y. Ahn W.S. Yang S. An 8-gene signature, including methylated and down-regulated glutathione peroxidase 3, of gastric cancer Int. J. Oncol. 2010 36 405 414 20043075 95. Yu Y.P. Yu G. Tseng G. Cieply K. Nelson J. Defrances M. Zarnegar R. Michalopoulos G. Luo J.H. Glutathione peroxidase 3, deleted or methylated in prostate cancer, suppresses prostate cancer growth and metastasis Cancer Res. 2007 67 8043 8050 10.1158/0008-5472.CAN-07-0648 17804715 96. Peng D.F. Razvi M. Chen H. Washington K. Roessner A. Schneider-Stock R. El-Rifai W. DNA hypermethylation regulates the expression of members of the Mu-class glutathione S -transferases and glutathione peroxidases in barrett’s adenocarcinoma Gut 2009 58 5 15 10.1136/gut.2007.146290 18664505 97. Jhaveri M.S. Morrow C.S. Methylation-mediated regulation of the glutathione S -transferase p1 gene in human breast cancer cells Gene 1998 210 1 7 10.1016/S0378-1119(98)00021-3 9524203 98. Lee W.H. Morton R.A. Epstein J.I. Brooks J.D. Campbell P.A. Bova G.S. Hsieh W.S. Isaacs W.B. Nelson W.G. Cytidine methylation of regulatory sequences near the Pi-class glutathione S -transferase gene accompanies human prostatic carcinogenesis Proc. Natl. Acad. Sci. USA 1994 91 11733 11737 10.1073/pnas.91.24.11733 7972132 99. Townsend D.M. Tew K.D. The role of glutathione-S -transferase in anti-cancer drug resistance Oncogene 2003 22 7369 7375 10.1038/sj.onc.1206940 14576844 100. Godwin A.K. Meister A. O’Dwyer P.J. Huang C.S. Hamilton T.C. Anderson M.E. High resistance to cisplatin in human ovarian cancer cell lines is associated with marked increase of glutathione synthesis Proc. Natl. Acad. Sci. USA 1992 89 3070 3074 10.1073/pnas.89.7.3070 1348364 101. Kramer R.A. Zakher J. Kim G. Role of the glutathione redox cycle in acquired and de novo multidrug resistance Science 1988 241 694 697 10.1126/science.3399900 3399900 102. Chen N. Liu Y. Greiner C.D. Holtzman J.L. Physiologic concentrations of homocysteine inhibit the human plasma gsh peroxidase that reduces organic hydroperoxides J. Lab. Clin. Med. 2000 136 58 65 10.1067/mlc.2000.107692 10882228 103. Durmaz A. Dikmen N. Homocysteine effects on cellular glutathione peroxidase (GPX-1) activity under in vitro conditions J. Enzym. Inhib. Med. Chem. 2007 22 733 738 10.1080/14756360601164929 18237028 104. Handy D.E. Zhang Y. Loscalzo J. Homocysteine down-regulates cellular glutathione peroxidase (GPX1) by decreasing translation J. Biol. Chem. 2005 280 15518 15525 10.1074/jbc.M501452200 15734734 105. Tastekin D. Erturk K. Bozbey H.U. Olmuscelik O. Kiziltan H. Tuna S. Tas F. Plasma homocysteine, folate and vitamin B12 levels in patients with lung cancer Exp. Oncol. 2015 37 218 222 26422108 106. Lubos E. Loscalzo J. Handy D.E. Homocysteine and glutathione peroxidase-1 Antioxid. Redox Signal. 2007 9 1923 1940 10.1089/ars.2007.1771 17822368 107. Lin J. Lee I.M. Song Y. Cook N.R. Selhub J. Manson J.E. Buring J.E. Zhang S.M. Plasma homocysteine and cysteine and risk of breast cancer in women Cancer Res. 2010 70 2397 2405 10.1158/0008-5472.CAN-09-3648 20197471 108. Matsumoto T. Sakari M. Okada M. Yokoyama A. Takahashi S. Kouzmenko A. Kato S. The androgen receptor in health and disease Annu. Rev. Physiol. 2013 75 201 224 10.1146/annurev-physiol-030212-183656 23157556 109. Tsouko E. Khan A.S. White M.A. Han J.J. Shi Y. Merchant F.A. Sharpe M.A. Xin L. Frigo D.E. Regulation of the pentose phosphate pathway by an androgen receptor-mtor-mediated mechanism and its role in prostate cancer cell growth Oncogenesis 2014 3 e103 10.1038/oncsis.2014.18 24861463 110. Tennakoon J.B. Shi Y. Han J.J. Tsouko E. White M.A. Burns A.R. Zhang A. Xia X. Ilkayeva O.R. Xin L. Androgens regulate prostate cancer cell growth via an AMPK-PGC-1α-mediated metabolic switch Oncogene 2014 33 5251 5261 10.1038/onc.2013.463 24186207 111. Luka Z. Mudd S.H. Wagner C. Glycine N -methyltransferase and regulation of S -adenosylmethionine levels J. Biol. Chem. 2009 284 22507 22511 10.1074/jbc.R109.019273 19483083 112. Huang Y.C. Lee C.M. Chen M. Chung M.Y. Chang Y.H. Huang W.J. Ho D.M. Pan C.C. Wu T.T. Yang S. Haplotypes, loss of heterozygosity, and expression levels of glycine N -methyltransferase in prostate cancer Clin. Cancer Res. 2007 13 1412 1420 10.1158/1078-0432.CCR-06-1551 17332283 113. Song Y.H. Shiota M. Kuroiwa K. Naito S. Oda Y. The important role of glycine N -methyltransferase in the carcinogenesis and progression of prostate cancer Mod. Pathol. 2011 24 1272 1280 10.1038/modpathol.2011.76 21572396 114. Putluri N. Shojaie A. Vasu V.T. Nalluri S. Vareed S.K. Putluri V. Vivekanandan-Giri A. Byun J. Pennathur S. Sana T.R. Metabolomic profiling reveals a role for androgen in activating amino acid metabolism and methylation in prostate cancer cells PLoS ONE 2011 6 1208 10.1371/journal.pone.0021417 21789170 115. Yoon J.K. Kim D.H. Koo J.S. Implications of differences in expression of sarcosine metabolism-related proteins according to the molecular subtype of breast cancer J. Transl. Med. 2014 12 1 11 10.1186/1479-5876-12-149 24387290 116. Ulanovskaya O.A. Zuhl A.M. Cravatt B.F. NNMT promotes epigenetic remodeling in cancer by creating a metabolic methylation sink Nat. Chem. Biol. 2013 9 300 306 10.1038/nchembio.1204 23455543 117. Zhang J. Wang Y. Li G. Yu H. Xie X. Down-regulation of nicotinamide N -methyltransferase induces apoptosis in human breast cancer cells via the mitochondria-mediated pathway PLoS ONE 2014 9 1208 10.1371/journal.pone.0089202 24558488 118. Chen X. Overcash R. Green T. Hoffman D. Asch A.S. Ruiz-Echevarria M.J. The tumor suppressor activity of the transmembrane protein with epidermal growth factor and two follistatin motifs 2 (TMEFF2) correlates with its ability to modulate sarcosine levels J. Biol. Chem. 2011 286 16091 16100 10.1074/jbc.M110.193805 21393249 119. Corbin J.M. Overcash R.F. Wren J.D. Coburn A. Tipton G.J. Ezzell J.A. McNaughton K.K. Fung K.M. Kosanke S.D. Ruiz-Echevarria M.J. Analysis of TMEFF2 allografts and transgenic mouse models reveals roles in prostate regeneration and cancer Prostate 2016 76 97 113 10.1002/pros.23103 26417683 120. Reed M.C. Thomas R.L. Pavisic J. James S.J. Ulrich C.M. Nijhout H.F. A mathematical model of glutathione metabolism Theor. Biol. Med. Model. 2008 5 1 16 10.1186/1742-4682-5-8 18182112 121. Li J.J. Li Q. Du H.P. Wang Y.L. You S.J. Wang F. Xu X.S. Cheng J. Cao Y.J. Liu C.F. Homocysteine triggers inflammatory responses in macrophages through inhibiting CSE-H2s signaling via DNA hypermethylation of CSE promoter Int. J. Mol. Sci. 2015 16 12560 12577 10.3390/ijms160612560 26047341 122. Kabil O. Banerjee R. Enzymology of h2s biogenesis, decay and signaling Antioxid. Redox Signal. 2014 20 770 782 10.1089/ars.2013.5339 23600844 123. Kajimura M. Fukuda R. Bateman R.M. Yamamoto T. Suematsu M. Interactions of multiple gas-transducing systems: Hallmarks and uncertainties of Co, No, and H2 S gas biology Antioxid. Redox Signal. 2010 13 157 192 10.1089/ars.2009.2657 19939208 124. Mustafa A.K. Gadalla M.M. Snyder S.H. Signaling by gasotransmitters Sci. Signal. 2009 2 re2 10.1126/scisignal.268re2 19401594 125. Zhao K. Li S. Wu L. Lai C. Yang G. Hydrogen sulfide represses androgen receptor transactivation by targeting at the second zinc finger module J. Biol. Chem. 2014 289 20824 20835 10.1074/jbc.M114.559518 24942741 126. Schalinske K.L. Smazal A.L. Homocysteine imbalance: A pathological metabolic marker Adv. Nutr. 2012 3 755 762 10.3945/an.112.002758 23153729 127. Bhattacharyya S. Saha S. Giri K. Lanza I.R. Nair K.S. Jennings N.B. Rodriguez-Aguayo C. Lopez-Berestein G. Basal E. Weaver A.L. Cystathionine β-synthase (CBS) contributes to advanced ovarian cancer progression and drug resistance PLoS ONE 2013 8 1208 10.1371/journal.pone.0079167 24236104 128. Szabo C. Coletta C. Chao C. Módis K. Szczesny B. Papapetropoulos A. Hellmich M.R. Tumor-derived hydrogen sulfide, produced by cystathionine-β-synthase, stimulates bioenergetics, cell proliferation, and angiogenesis in colon cancer Proc. Natl. Acad. Sci. USA 2013 110 12474 12479 10.1073/pnas.1306241110 23836652 129. Janošík M. Kery V. Gaustadnes M. Maclean K.N. Kraus J.P. Regulation of human cystathionine β-synthase by S -adenosyl-l -methionine: Evidence for two catalytically active conformations involving an autoinhibitory domain in the C-terminal region Biochemistry 2001 40 10625 10633 10.1021/bi010711p 11524006 130. Prudova A. Albin M. Bauman Z. Lin A. Vitvitsky V. Banerjee R. Testosterone regulation of homocysteine metabolism modulates redox status in human prostate cancer cells Antioxid. Redox Signal. 2007 9 1875 1881 10.1089/ars.2007.1712 17854288 131. Guo H. Gai J.-W. Wang Y. Jin H.-F. Du J.-B. Jin J. Characterization of hydrogen sulfide and its synthases, cystathionine β-synthase and cystathionine γ-lyase, in human prostatic tissue and cells Urology 2012 79 483.e481 483.e485 10.1016/j.urology.2011.10.013 22310774 132. Zhang W. Braun A. Bauman Z. Olteanu H. Madzelan P. Banerjee R. Expression profiling of homocysteine junction enzymes in the nci60 panel of human cancer cell lines Cancer Res. 2005 65 1554 1560 10.1158/0008-5472.CAN-04-1554 15735045 133. Al-Awadi F. Yang M. Tan Y. Han Q. Li S. Hoffman R.M. Human tumor growth in nude mice is associated with decreased plasma cysteine and homocysteine Anticancer Res. 2008 28 2541 2544 19035276 134. Stabler S. Koyama T. Zhao Z. Martinez-Ferrer M. Allen R.H. Luka Z. Loukachevitch L.V. Clark P.E. Wagner C. Bhowmick N.A. Serum methionine metabolites are risk factors for metastatic prostate cancer progression PLoS ONE 2011 6 1208 10.1371/journal.pone.0022486 21853037 135. Chwatko G. Forma E. Wilkosz J. Głowacki R. Jóźwiak P. Różański W. Bryś M. Krześlak A. Thiosulfate in urine as a facilitator in the diagnosis of prostate cancer for patients with prostate-specific antigen less or equal 10 ng/mL Clin. Chem. Lab. Med. 2013 51 1825 1831 10.1515/cclm-2013-0069 23843582 136. Hatayama I. Satoh K. Sato K. Developmental and hormonal regulation of the major form of hepatic glutathione s-transferase in male mice Biochem. Biophys. Res. Commun. 1986 140 581 588 10.1016/0006-291X(86)90771-0 3778469 137. Ikeda H. Serria M.S. Kakizaki I. Hatayama I. Satoh K. Tsuchida S. Muramatsu M. Nishi S. Sakai M. Activation of mouse pi-class glutathione S -transferase gene by Nrf2(Nf-E2-related factor 2) and androgen Biochem. J. 2002 364 563 570 10.1042/bj20011756 12023900 138. Imperlini E. Mancini A. Spaziani S. Martone D. Alfieri A. Gemei M. del Vecchio L. Buono P. Orru S. Androgen receptor signaling induced by supraphysiological doses of dihydrotestosterone in human peripheral blood lymphocytes Proteomics 2010 10 3165 3175 10.1002/pmic.201000079 20677326 139. Khandrika L. Kumar B. Koul S. Maroni P. Koul H.K. Oxidative stress in prostate cancer Cancer Lett. 2009 282 125 136 10.1016/j.canlet.2008.12.011 19185987 140. Pendeville H. Carpino N. Marine J.C. Takahashi Y. Muller M. Martial J.A. Cleveland J.L. The ornithine decarboxylase gene is essential for cell survival during early murine development Mol. Cell. Biol. 2001 21 6549 6558 10.1128/MCB.21.19.6549-6558.2001 11533243 141. Casero R.A. Pegg A.E. Polyamine catabolism and disease Biochem. J. 2009 421 323 338 10.1042/BJ20090598 19589128 142. Nowotarski S.L. Woster P.M. Casero R.A. Jr. Polyamines and cancer: Implications for chemotherapy and chemoprevention Expert Rev. Mol. Med. 2013 15 e3 10.1017/erm.2013.3 23432971 143. Casero R.A. Marton L.J. Targeting polyamine metabolism and function in cancer and other hyperproliferative diseases Nat. Rev. Drug Discov. 2007 6 373 390 10.1038/nrd2243 17464296 144. Giardiello F.M. Hamilton S.R. Hylind L.M. Yang V.W. Tamez P. Casero R.A. Jr. Ornithine decarboxylase and polyamines in familial adenomatous polyposis Cancer Res. 1997 57 199 201 9000553 145. Pegg A.E. Lockwood D.H. Williams-Ashman H.G. Concentrations of putrescine and polyamines and their enzymic synthesis during androgen-induced prostatic growth Biochem. J. 1970 117 17 31 10.1042/bj1170017 5420953 146. Blackshear P.J. Manzella J.M. Stumpo D.J. Wen L. Huang J.K. Oyen O. Young W.S. 3rd High level, cell-specific expression of ornithine decarboxylase transcripts in rat genitourinary tissues Mol. Endocrinol. 1989 3 68 78 10.1210/mend-3-1-68 2915649 147. Pegg A.E. Polyamine metabolism and its importance in neoplastic growth and a target for chemotherapy Cancer Res. 1988 48 759 774 3123052 148. Janne J. Alhonen L. Leinonen P. Polyamines: From molecular biology to clinical applications Ann. Med. 1991 23 241 259 10.3109/07853899109148056 1930914 149. Mohan R.R. Challa A. Gupta S. Bostwick D.G. Ahmad N. Agarwal R. Marengo S.R. Amini S.B. Paras F. MacLennan G.T. Overexpression of ornithine decarboxylase in prostate cancer and prostatic fluid in humans Clin. Cancer Res. 1999 5 143 147 9918212 150. Rhodes D.R. Barrette T.R. Rubin M.A. Ghosh D. Chinnaiyan A.M. Meta-analysis of microarrays: Interstudy validation of gene expression profiles reveals pathway dysregulation in prostate cancer Cancer Res. 2002 62 4427 4433 12154050 151. Cipolla B.G. Havouis R. Moulinoux J.P. Polyamine reduced diet (PRD) nutrition therapy in hormone refractory prostate cancer patients Biomed. Pharmacother. 2010 64 363 368 10.1016/j.biopha.2009.09.022 20106631 152. Janne O.A. Crozat A. Palvimo J. Eisenberg L.M. Androgen-regulation of ornithine decarboxylase and S -adenosylmethionine decarboxylase genes J. Steroid Biochem. Mol. Biol. 1991 40 307 315 10.1016/0960-0760(91)90196-C 1958536 153. Fjosne H.E. Strand H. Sunde A. Dose-dependent induction of ornithine decarboxylase and S -adenosyl-methionine decarboxylase activity by testosterone in the accessory sex organs of male rats Prostate 1992 21 239 245 10.1002/pros.2990210307 1437858 154. Cyriac J. Haleem R. Cai X. Wang Z. Androgen regulation of spermidine synthase expression in the rat prostate Prostate 2002 50 252 261 10.1002/pros.10052 11870804 155. Crozat A. Palvimo J.J. Julkunen M. Janne O.A. Comparison of androgen regulation of ornithine decarboxylase and S -adenosylmethionine decarboxylase gene expression in rodent kidney and accessory sex organs Endocrinology 1992 130 1131 1144 1537280 156. Fjosne H.E. Strand H. Ostensen M.A. Sunde A. Ornithine decarboxylase and S -adenosylmethionine decarboxylase activity in the accessory sex organs of intact, castrated, and androgen-stimulated castrated rats Prostate 1988 12 309 320 10.1002/pros.2990120404 3393492 157. Bai G. Kasper S. Matusik R.J. Rennie P.S. Moshier J.A. Krongrad A. Androgen regulation of the human ornithine decarboxylase promoter in prostate cancer cells J. Androl. 1998 19 127 135 9570735 158. Cohen R.J. Fujiwara K. Holland J.W. McNeal J.E. Polyamines in prostatic epithelial cells and adenocarcinoma: The effects of androgen blockade Prostate 2001 49 278 284 10.1002/pros.10023 11746274 159. Lloyd S.M. Arnold J. Sreekumar A. Metabolomic profiling of hormone-dependent cancers: A bird’s eye view Trends Endocrinol. Metab. 2015 26 477 485 10.1016/j.tem.2015.07.001 26242817 160. Bistulfi G. Diegelman P. Foster B.A. Kramer D.L. Porter C.W. Smiraglia D.J. Polyamine biosynthesis impacts cellular folate requirements necessary to maintain S -adenosylmethionine and nucleotide pools FASEB J. 2009 23 2888 2897 10.1096/fj.09-130708 19417083 161. Yegnasubramanian S. Haffner M.C. Zhang Y. Gurel B. Cornish T.C. Wu Z. Irizarry R.A. Morgan J. Hicks J. DeWeese T.L. DNA hypomethylation arises later in prostate cancer progression than CPG island hypermethylation and contributes to metastatic tumor heterogeneity Cancer Res. 2008 68 8954 8967 10.1158/0008-5472.CAN-07-6088 18974140 162. Jerónimo C. Bastian P.J. Bjartell A. Carbone G.M. Catto J.W.F. Clark S.J. Henrique R. Nelson W.G. Shariat S.F. Epigenetics in prostate cancer: Biologic and clinical relevance Eur. Urol. 2011 60 753 766 10.1016/j.eururo.2011.06.035 21719191 163. Labbe D.P. Zadra G. Ebot E.M. Mucci L.A. Kantoff P.W. Loda M. Brown M. Role of diet in prostate cancer: The epigenetic link Oncogene 2015 34 4683 4691 10.1038/onc.2014.422 25531313 164. Seligson D.B. Horvath S. Shi T. Yu H. Tze S. Grunstein M. Kurdistani S.K. Global histone modification patterns predict risk of prostate cancer recurrence Nature 2005 435 1262 1266 10.1038/nature03672 15988529 165. Perry A.S. Watson R.W. Lawler M. Hollywood D. The epigenome as a therapeutic target in prostate cancer Nat. Rev. Urol. 2010 7 668 680 10.1038/nrurol.2010.185 21060342 166. Nelson W.G. de Marzo A.M. Yegnasubramanian S. Epigenetic alterations in human prostate cancers Endocrinology 2009 150 3991 4002 10.1210/en.2009-0573 19520778 167. Yegnasubramanian S. Kowalski J. Gonzalgo M.L. Zahurak M. Piantadosi S. Walsh P.C. Bova G.S. de Marzo A.M. Isaacs W.B. Nelson W.G. Hypermethylation of CPG islands in primary and metastatic human prostate cancer Cancer Res. 2004 64 1975 1986 10.1158/0008-5472.CAN-03-3972 15026333 168. Jeronimo C. Henrique R. Hoque M.O. Mambo E. Ribeiro F.R. Varzim G. Oliveira J. Teixeira M.R. Lopes C. Sidransky D. A quantitative promoter methylation profile of prostate cancer Clin. Cancer Res. 2004 10 8472 8478 10.1158/1078-0432.CCR-04-0894 15623627 169. Maruyama R. Toyooka S. Toyooka K.O. Virmani A.K. Zöchbauer-Müller S. Farinas A.J. Minna J.D. McConnell J. Frenkel E.P. Gazdar A.F. Aberrant promoter methylation profile of prostate cancers and its relationship to clinicopathological features Clin. Cancer Res. 2002 8 514 519 11839671 170. Florl A.R. Steinhoff C. Müller M. Seifert H.H. Hader C. Engers R. Ackermann R. Schulz W.A. Coordinate hypermethylation at specific genes in prostate carcinoma precedes line-1 hypomethylation Br. J. Cancer 2004 91 985 994 10.1038/sj.bjc.6602030 15292941 171. Padar A. Sathyanarayana U.G. Suzuki M. Maruyama R. Hsieh J.T. Frenkel E.P. Minna J.D. Gazdar A.F. Inactivation of cyclin D2 gene in prostate cancers by aberrant promoter methylation Clin. Cancer Res. 2003 9 4730 4734 14581343 172. Yamanaka M. Watanabe M. Yamada Y. Takagi A. Murata T. Takahashi H. Suzuki H. Ito H. Tsukino H. Katoh T. Altered methylation of multiple genes in carcinogenesis of the prostate Int. J. Cancer 2003 106 382 387 10.1002/ijc.11227 12845678 173. Ellinger J. Bastian P.J. Jurgan T. Biermann K. Kahl P. Heukamp L.C. Wernert N. Müller S.C. von Ruecker A. Cpg island hypermethylation at multiple gene sites in diagnosis and prognosis of prostate cancer Urology 2008 71 161 167 10.1016/j.urology.2007.09.056 18242387 174. Lodygin D. Epanchintsev A. Menssen A. Diebold J. Hermeking H. Functional epigenomics identifies genes frequently silenced in prostate cancer Cancer Res. 2005 65 4218 4227 10.1158/0008-5472.CAN-04-4407 15899813 175. Perry A.S. Foley R. Woodson K. Lawler M. The emerging roles of DNA methylation in the clinical management of prostate cancer Endocr. Relat. Cancer 2006 13 357 377 10.1677/erc.1.01184 16728568 176. Li L.C. Epigenetics of prostate cancer Front. Biosci. J. Virtual Libr. 2007 12 3377 3397 10.2741/2320 177. Sasaki M. Tanaka Y. Perinchery G. Dharia A. Kotcherguina I. Fujimoto S. Dahiya R. Methylation and inactivation of estrogen, progesterone, and androgen receptors in prostate cancer J. Natl. Cancer Inst. 2002 94 384 390 10.1093/jnci/94.5.384 11880477 178. Reibenwein J. Pils D. Horak P. Tomicek B. Goldner G. Worel N. Elandt K. Krainer M. Promoter hypermethylation of GSTP1, AR, and 14–3-3σ in serum of prostate cancer patients and its clinical relevance Prostate 2007 67 427 432 10.1002/pros.20533 17192910 179. Chen M.F. Chen W.C. Chang Y.J. Wu C.F. Wu C.T. Role of DNA methyltransferase 1 in hormone-resistant prostate cancer J. Mol. Med. 2010 88 953 962 10.1007/s00109-010-0640-3 20544341 180. Patra S.K. Patra A. Zhao H. Dahiya R. DNA methyltransferase and demethylase in human prostate cancer Mol. Carcinog. 2002 33 163 171 10.1002/mc.10033 11870882 181. Zhang W. Jiao H. Zhang X. Zhao R. Wang F. He W. Zong H. Fan Q. Wang L. Correlation between the expression of DNMT1, and GSTP1 and APC, and the methylation status of GSTP1 and APC in association with their clinical significance in prostate cancer Mol. Med. Rep. 2015 12 141 146 10.3892/mmr.2015.3402 25738352 182. Valdez C.D. Kunju L. Daignault S. Wojno K.J. Day M.L. The e2f1/dnmt1 axis is associated with the development of ar negative castration resistant prostate cancer Prostate 2013 73 1776 1785 10.1002/pros.22715 24038143 183. Zhang Q. Chen L. Helfand B.T. Jang T.L. Sharma V. Kozlowski J. Kuzel T.M. Zhu L.J. Yang X.J. Javonovic B. Tgf-beta regulates DNA methyltransferase expression in prostate cancer, correlates with aggressive capabilities, and predicts disease recurrence PLoS ONE 2011 6 1208 184. Kinney S.R. Moser M.T. Pascual M. Greally J.M. Foster B.A. Karpf A.R. Opposing roles of DNMT1 in early- and late-stage murine prostate cancer Mol. Cell. Biol. 2010 30 4159 4174 10.1128/MCB.00235-10 20584988 185. McCabe M.T. Low J.A. Daignault S. Imperiale M.J. Wojno K.J. Day M.L. Inhibition of DNA methyltransferase activity prevents tumorigenesis in a mouse model of prostate cancer Cancer Res. 2006 66 385 392 10.1158/0008-5472.CAN-05-2020 16397253 186. Kobayashi Y. Absher D.M. Gulzar Z.G. Young S.R. McKenney J.K. Peehl D.M. Brooks J.D. Myers R.M. Sherlock G. DNA methylation profiling reveals novel biomarkers and important roles for DNA methyltransferases in prostate cancer Genome Res. 2011 21 1017 1027 10.1101/gr.119487.110 21521786 187. Brothman A.R. Swanson G. Maxwell T.M. Cui J. Murphy K.J. Herrick J. Speights V.O. Isaac J. Rohr L.R. Global hypomethylation is common in prostate cancer cells: A quantitative predictor for clinical outcome? Cancer Gen. Cytogenet. 2005 156 31 36 10.1016/j.cancergencyto.2004.04.004 15588853 188. Fuso A. Nicolia V. Cavallaro R.A. Scarpa S. DNA methylase and demethylase activities are modulated by one-carbon metabolism in Alzheimer’s disease models J. Nut. Biochem. 2011 22 242 251 10.1016/j.jnutbio.2010.01.010 20573497 189. Rhodes D.R. Sanda M.G. Otte A.P. Chinnaiyan A.M. Rubin M.A. Multiplex biomarker approach for determining risk of prostate-specific antigen-defined recurrence of prostate cancer J. Natl. Cancer Inst. 2003 95 661 668 10.1093/jnci/95.9.661 12734317 190. Hoffmann M.J. Engers R. Florl A.R. Otte A.P. Müller M. Schulz W.A. Expression changes in EZH2, but not in BMI-1, SIRT1, DNMT1 or DNMT3b, are associated with DNA methylation changes in prostate cancer Cancer Biol. Ther. 2007 6 1403 1412 10.4161/cbt.6.9.4542 18637271 191. Varambally S. Dhanasekaran S.M. Zhou M. Barrette T.R. Kumar-Sinha C. Sanda M.G. Ghosh D. Pienta K.J. Sewalt R.G.A.B. Otte A.P. The polycomb group protein EZH2 is involved in progression of prostate cancer Nature 2002 419 624 629 10.1038/nature01075 12374981 192. Laitinen S. Martikainen P.M. Tolonen T. Isola J. Tammela T.L.J. Visakorpi T. EZH2, Ki-67 and MCM7 are prognostic markers in prostatectomy treated patients Int. J. Cancer 2008 122 595 602 10.1002/ijc.23145 17943722 193. Yu J. Rhodes D.R. Tomlins S.A. Cao X. Chen G. Mehra R. Wang X. Ghosh D. Shah R.B. A polycomb repression signature in metastatic prostate cancer predicts cancer outcome Cancer Res. 2007 67 10657 10663 10.1158/0008-5472.CAN-07-2498 18006806 194. Viré E. Brenner C. Deplus R. Blanchon L. Fraga M. Didelot C. Morey L. van Eynde A. Bernard D. Vanderwinden J.M. The polycomb group protein EZH2 directly controls DNA methylation Nature 2006 439 871 874 10.1038/nature04431 16357870 195. Schlesinger Y. Straussman R. Keshet I. Farkash S. Hecht M. Zimmerman J. Eden E. Yakhini Z. Ben-Shushan E. Reubinoff B.E. Polycomb-mediated methylation on LYS27 of histone H3 pre-marks genes for de novo methylation in cancer Nat. Genet. 2007 39 232 236 10.1038/ng1950 17200670 196. Crea F. Sun L. Mai A. Chiang Y.T. Farrar W.L. Danesi R. Helgason C.D. The emerging role of histone lysine demethylases in prostate cancer Mol. Cancer 2012 11 1 10 10.1186/1476-4598-11-52 22216762 197. Berry W.L. Janknecht R. KDM4/JMJD2 histone demethylases: Epigenetic regulators in cancer cells Cancer Res. 2013 73 2936 2942 10.1158/0008-5472.CAN-12-4300 23644528 198. Franci G. Ciotta A. Altucci L. The Jumonji family: Past, present and future of histone demethylases in cancer Biomol. Concepts 2014 5 209 224 10.1515/bmc-2014-0010 25372754 199. Deguchi T. Barchas J. Inhibition of transmethylations of biogenic amines by S -adenosylhomocysteine. Enhancement of transmethylation by adenosylhomocysteinase J. Biol. Chem. 1971 246 3175 3181 5574392 200. Wang Y.-C. Tang F.-Y. Chen S.-Y. Chen Y.-M. Chiang E.-P.I. Glycine-N methyltransferase expression in HepG2 cells is involved in methyl group homeostasis by regulating transmethylation kinetics and DNA methylation J. Nutr. 2011 141 777 782 10.3945/jn.110.135954 21411609 201. Zhao J.C. Yu J. Runkle C. Wu L. Hu M. Wu D. Liu J.S. Wang Q. Qin Z.S. Yu J. Cooperation between polycomb and androgen receptor during oncogenic transformation Genome Res. 2012 22 322 331 10.1101/gr.131508.111 22179855 202. Chng K.R. Chang C.W. Tan S.K. Yang C. Hong S.Z. Sng N.Y. Cheung E. A transcriptional repressor co-regulatory network governing androgen response in prostate cancers EMBO J. 2012 31 2810 2823 10.1038/emboj.2012.112 22531786 203. Malik R. Khan A.P. Asangani I.A. Cieślik M. Prensner J.R. Wang X. Iyer M.K. Jiang X. Borkin D. Escara-Wilke J. Targeting the MLL complex in castration resistant prostate cancer Nat. Med. 2015 21 344 352 10.1038/nm.3830 25822367 204. Yamane K. Toumazou C. Tsukada Y. Erdjument-Bromage H. Tempst P. Wong J. Zhang Y. JHDM2A, a JMJC-containing H3K9 demethylase, facilitates transcription activation by androgen receptor Cell 2006 125 483 495 10.1016/j.cell.2006.03.027 16603237 205. Cai C. He H.H. Chen S. Coleman I. Wang H. Fang Z. Chen S. Nelson P.S. Liu X.S. Brown M. Androgen receptor gene expression in prostate cancer is directly suppressed by the androgen receptor through recruitment of lysine-specific demethylase 1 Cancer Cell 2011 20 457 471 10.1016/j.ccr.2011.09.001 22014572 206. Gaughan L. Stockley J. Wang N. McCracken S.R.C. Treumann A. Armstrong K. Shaheen F. Watt K. McEwan I.J. Wang C. Regulation of the androgen receptor by set9-mediated methylation Nucleic Acids Res. 2011 39 1266 1279 10.1093/nar/gkq861 20959290 207. Xu K. Wu Z.J. Groner A.C. He H.H. Cai C. Lis R.T. Wu X. Stack E.C. Loda M. Liu T. Ezh2 oncogenic activity in castration-resistant prostate cancer cells is polycomb-independent Science 2012 338 1465 1469 10.1126/science.1227604 23239736 208. Bohrer L.R. Chen S. Hallstrom T.C. Huang H. Androgens suppress ezh2 expression via retinoblastoma (RB) and p130-dependent pathways: A potential mechanism of androgen-refractory progression of prostate cancer Endocrinology 2010 151 5136 5145 10.1210/en.2010-0436 20881251 209. Hofman K. Swinnen J.V. Verhoeven G. Heyns W. E2f activity is biphasically regulated by androgens in lncap cells Biochem. Biophys. Res. Commun. 2001 283 97 101 10.1006/bbrc.2001.4738 11322773 210. Cao P. Deng Z. Wan M. Huang W. Cramer S.D. Xu J. Lei M. Sui G. MicroRNA-101 negatively regulates EZH2 and its expression is modulated by androgen receptor and HIF-1α/HIF-1β Mol. Cancer 2010 9 108 10.1186/1476-4598-9-108 20478051 211. Varambally S. Cao Q. Mani R.S. Shankar S. Wang X. Ateeq B. Laxman B. Cao X. Jing X. Ramnarayanan K. Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer Science 2008 322 1695 1699 10.1126/science.1165395 19008416 212. Farber S. Diamond L.K. Temporary remissions in acute leukemia in children produced by folic acid antagonist, 4-aminopteroyl-glutamic acid N. Engl. J. Med. 1948 238 787 793 10.1056/NEJM194806032382301 18860765 213. Loening S.A. Beckley S. Brady M.F. Chu T.M. deKernion J.B. Dhabuwala C. Gaeta J.F. Gibbons R.P. McKiel C.F. McLeod D.G. Comparison of estramustine phosphate, methotrexate and cis-platinum in patients with advanced, hormone refractory prostate cancer J. Urol. 1983 129 1001 1006 6343629 214. Saxman S. Ansari R. Drasga R. Miller M. Wheeler B. McClean J. Einhorn L. Phase III trial of cyclophosphamide versus cyclophosphamide, doxorubicin, and methotrexate in hormone-refractory prostatic cancer. A hoosier oncology group study Cancer 1992 70 2488 2492 10.1002/1097-0142(19921115)70:10<2488::AID-CNCR2820701016>3.0.CO;2-9 1423178 215. Jones W.G. Fossa S.D. Verbaeys A.C. Droz J.P. Klijn J.G. Boven E. de Pauw M. Sylvester R. Low-dose fortnightly methotrexate in advanced prostate cancer. The eortc genito-urinary tract cancer cooperative group Eur. J. Cancer 1990 26 646 10.1016/0277-5379(90)90099-F 2144757 216. Linsalata M. Caruso M.G. Leo S. Guerra V. D’Attoma B. di Leo A. Prognostic value of tissue polyamine levels in human colorectal carcinoma Anticancer Res. 2002 22 2465 2469 12174945 217. Canizares F. Salinas J. de las Heras M. Diaz J. Tovar I. Martinez P. Penafiel R. Prognostic value of ornithine decarboxylase and polyamines in human breast cancer: Correlation with clinicopathologic parameters Clin. Cancer Res. 1999 5 2035 2041 10473083 218. Heston W.D. Prostatic polyamines and polyamine targeting as a new approach to therapy of prostatic cancer Cancer Surv. 1991 11 217 238 1841753 219. Durie B.G. Salmon S.E. Russell D.H. Polyamines as markers of response and disease activity in cancer chemotherapy Cancer Res. 1977 37 214 221 830407 220. Kubota S. Okada M. Yoshimoto M. Murata N. Yamasaki Z. Wada T. Imahori K. Ohsawa N. Takaku F. Urinary polyamines as a tumor marker Cancer Detect. Prev. 1985 8 189 192 4064039 221. Russell D.H. Clinical relevance of polyamines Crit. Rev. Clin. Lab. Sci. 1983 18 261 311 10.3109/10408368209085073 6339165 222. Sakai S. Ito Y. Koide T. Tei K. Hara A. Sawada H. Detection of urinary polyamine by a new enzymatic differential assay. (III). Studies on urinary polyamines in patients with malignant genitourinary diseases Hinyokika Kiyo 1986 32 343 350 2425608 223. Chatel M. Darcel F. Quemener V. Hercouet H. Moulinoux J.P. Red blood cell polyamines as biochemical markers of supratentorial malignant gliomas Anticancer Res. 1987 7 33 38 3566181 224. Cipolla B. Guille F. Moulinoux J.P. Bansard J.Y. Roth S. Staerman F. Corbel L. Quemener V. Lobel B. Erythrocyte polyamines and prognosis in stage D2 prostatic carcinoma patients J. Urol. 1994 151 629 633 7508520 225. Cipolla B. Guille F. Moulinoux J.P. Quemener V. Staerman F. Corbel L. Lobel B. Polyamines and prostatic carcinoma: Clinical and therapeutic implications Eur. Urol. 1993 24 124 131 7689970 226. Weiss T.S. Bernhardt G. Buschauer A. Thasler W.E. Dolgner D. Zirngibl H. Jauch K.W. Polyamine levels of human colorectal adenocarcinomas are correlated with tumor stage and grade Int. J. Colorectal Dis. 2002 17 381 387 10.1007/s00384-002-0394-7 12355213 227. Bergeron C. Bansard J.Y. Le Moine P. Bouet F. Goasguen J.E. Moulinoux J.P. Le Gall E. Catros-Quemener V. Erythrocyte spermine levels: A prognostic parameter in childhood common acute lymphoblastic leukemia Leukemia 1997 11 31 36 10.1038/sj.leu.2400531 9001415 228. Soda K. The mechanisms by which polyamines accelerate tumor spread J. Exp. Clin. Cancer Res. 2011 30 1 9 10.1186/1756-9966-30-95 21208462 229. Cipolla B. Moulinoux J.P. Quemener V. Havouis R. Martin L.A. Guille F. Lobel B. Erythrocyte polyamine levels in human prostatic carcinoma J. Urol. 1990 144 1164 1166 1700142 230. Cheng L.L. Wu C.-L. Smith M.R. Gonzalez R.G. Non-destructive quantitation of spermine in human prostate tissue samples using HRMAS 1 H NMR spectroscopy at 9.4 T FEBS Lett. 2001 494 112 116 10.1016/S0014-5793(01)02329-8 11297745 231. McDunn J.E. Li Z. Adam K.P. Neri B.P. Wolfert R.L. Milburn M.V. Lotan Y. Wheeler T.M. Metabolomic signatures of aggressive prostate cancer Prostate 2013 73 1547 1560 10.1002/pros.22704 23824564 232. Gupta S. Ahmad N. Marengo S.R. MacLennan G.T. Greenberg N.M. Mukhtar H. Chemoprevention of prostate carcinogenesis by α-difluoromethylornithine in tramp mice Cancer Res. 2000 60 5125 5133 11016639 233. Kee K. Foster B.A. Merali S. Kramer D.L. Hensen M.L. Diegelman P. Kisiel N. Vujcic S. Mazurchuk R.V. Porter C.W. Activated polyamine catabolism depletes acetyl-coa pools and suppresses prostate tumor growth in tramp mice J. Biol. Chem. 2004 279 40076 40083 10.1074/jbc.M406002200 15252047 234. Devens B.H. Weeks R.S. Burns M.R. Carlson C.L. Brawer M.K. Polyamine depletion therapy in prostate cancer Prostate Cancer Prostatic Dis. 2000 3 275 279 10.1038/sj.pcan.4500420 12497077 235. Kadmon D. Chemoprevention in prostate cancer: The role of difluoromethylornithine (DFMO) J. Cell. Biochem. Suppl. 1992 16H 122 127 10.1002/jcb.240501228 1289667 236. Moulinoux J.P. Quemener V. Cipolla B. Guille F. Havouis R. Martin C. Lobel B. Seiler N. The growth of MAT-LyLu rat prostatic adenocarcinoma can be prevented in vivo by polyamine deprivation J. Urol. 1991 146 1408 1412 1942311 237. Danzin C. Jung M.J. Grove J. Bey P. Effect of alpha-difluoromethylornithine, an enzyme-activated irreversible inhibitor of ornithine decarboxylase, on polyamine levels in rat tissues Life Sci. 1979 24 519 524 10.1016/0024-3205(79)90173-5 431333 238. Meyskens F.L. Jr. Simoneau A.R. Gerner E.W. Chemoprevention of prostate cancer with the polyamine synthesis inhibitor difluoromethylornithine Recent Results Cancer Res. 2014 202 115 120 24531785 239. Simoneau A.R. Gerner E.W. Nagle R. Ziogas A. Fujikawa-Brooks S. Yerushalmi H. Ahlering T.E. Lieberman R. McLaren C.E. Anton-Culver H. The effect of difluoromethylornithine on decreasing prostate size and polyamines in men: Results of a year-long phase IIB randomized placebo-controlled chemoprevention trial Cancer Epidemiol. Biomarkers Prev. 2008 17 292 299 10.1158/1055-9965.EPI-07-0658 18268112 240. Streiff R.R. Bender J.F. Phase I study of N1-N11-diethylnorspermine (DENSPM) administered TID for 6 days in patients with advanced malignancies Investig. New Drugs 2001 19 29 39 10.1023/A:1006448516938 11291831 241. Wagner T. Jung M. New lysine methyltransferase drug targets in cancer Nat. Biotechnol. 2012 30 622 623 10.1038/nbt.2300 22781684 242. Kaminskas E. Farrell A. Abraham S. Baird A. Hsieh L.S. Lee S.L. Leighton J.K. Patel H. Rahman A. Sridhara R. Approval summary: Azacitidine for treatment of myelodysplastic syndrome subtypes Clin. Cancer Res. 2005 11 3604 3608 10.1158/1078-0432.CCR-04-2135 15897554 243. Fu S. Hu W. Iyer R. Kavanagh J.J. Coleman R.L. Levenback C.F. Sood A.K. Wolf J.K. Gershenson D.M. Markman M. Phase 1b–2a study to reverse platinum resistance through use of a hypomethylating agent, azacitidine, in patients with platinum-resistant or platinum-refractory epithelial ovarian cancer Cancer 2011 117 1661 1669 10.1002/cncr.25701 21472713 244. Iwata H. Sato H. Suzuki R. Yamada R. Ichinomiya S. Yanagihara M. Okabe H. Sekine Y. Yano T. Ueno K. A demethylating agent enhances chemosensitivity to vinblastine in a xenograft model of renal cell carcinoma Int. J. Oncol. 2011 38 1653 1661 10.3892/ijo.2011.999 21479360 245. Kiziltepe T. Hideshima T. Catley L. Raje N. Yasui H. Shiraishi N. Okawa Y. Ikeda H. Vallet S. Pozzi S. 5-Azacytidine, a DNA methyltransferase inhibitor, induces ATR-mediated DNA double-strand break responses, apoptosis, and synergistic cytotoxicity with doxorubicin and bortezomib against multiple myeloma cells Mol. Cancer Ther. 2007 6 1718 1727 10.1158/1535-7163.MCT-07-0010 17575103 246. Mao M. Tian F. Mariadason J.M. Tsao C.C. Lemos R. Jr. Dayyani F. Gopal Y.N. Jiang Z.Q. Wistuba II Tang X.M. Resistance to braf inhibition in braf-mutant colon cancer can be overcome with pi3k inhibition or demethylating agents Clin. Cancer Res. 2013 19 657 667 10.1158/1078-0432.CCR-11-1446 23251002 247. Ramachandran K. Gordian E. Singal R. 5-Azacytidine reverses drug resistance in bladder cancer cells Anticancer Res. 2011 31 3757 3766 22110197 248. Singal R. Ramachandran K. Gordian E. Quintero C. Zhao W. Reis I.M. Phase I/II study of azacitidine, docetaxel, and prednisone in patients with metastatic castration-resistant prostate cancer previously treated with docetaxel-based therapy Clin. Genitourin. Cancer 2015 13 22 31 10.1016/j.clgc.2014.07.008 25178642 249. Yang L. Lin C. Jin C. Yang J.C. Tanasa B. Li W. Merkurjev D. Ohgi K.A. Meng D. Zhang J. Lncrna-dependent mechanisms of androgen-receptor-regulated gene activation programs Nature 2013 500 598 602 10.1038/nature12451 23945587 250. Wee Z.N. Li Z. Lee P.L. Lee S.T. Lim Y.P. Yu Q. EZH2-mediated inactivation of IFN-γ-JAK-STAT1 signaling is an effective therapeutic target in MYC-driven prostate cancer Cell Rep. 2014 8 204 216 10.1016/j.celrep.2014.05.045 24953652 251. Beltran H. Prandi D. Mosquera J.M. Benelli M. Puca L. Cyrta J. Marotz C. Giannopoulou E. Chakravarthi B.V. Varambally S. Divergent clonal evolution of castration-resistant neuroendocrine prostate cancer Nat. Med. 2016 22 298 305 10.1038/nm.4045 26855148

PMC005xxxxxx/PMC5000607.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081209ijms-17-01209ArticleLiposomes Loaded with Hydrophobic Iron Oxide Nanoparticles: Suitable T2 Contrast Agents for MRI Martínez-González Raquel Estelrich Joan Busquets Maria Antònia *Sivakov Vladimir Academic EditorSalifoglou Athanasios Academic EditorDepartment of Pharmacy, Pharmaceutical Technology and Physical Chemistry, IN2UB, Faculty of Pharmacy, Avda Joan XXIII, 27-31, 08028 Barcelona, Spain; [email protected] (R.M.-G.); [email protected] (J.E.)* Correspondence: [email protected]; Tel.: +34-934-024-55627 7 2016 8 2016 17 8 120903 6 2016 20 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).There has been a recent surge of interest in the use of superparamagnetic iron oxide nanoparticles (SPIONs) as contrast agents (CAs) for magnetic resonance imaging (MRI), due to their tunable properties and their low toxicity compared with other CAs such as gadolinium. SPIONs exert a strong influence on spin-spin T2 relaxation times by decreasing the MR signal in the regions to which they are delivered, consequently yielding darker images or negative contrast. Given the potential of these nanoparticles to enhance detection of alterations in soft tissues, we studied the MRI response of hydrophobic or hydrophilic SPIONs loaded into liposomes (magnetoliposomes) of different lipid composition obtained by sonication. These hybrid nanostructures were characterized by measuring several parameters such as size and polydispersity, and number of SPIONs encapsulated or embedded into the lipid systems. We then studied the influence of acyl chain length as well as its unsaturation, charge, and presence of cholesterol in the lipid bilayer at high field strength (7 T) to mimic the conditions used in preclinical assays. Our results showed a high variability depending on the nature of the magnetic particles. Focusing on the hydrophobic SPIONs, the cholesterol-containing samples showed a slight reduction in r2, while unsaturation of the lipid acyl chain and inclusion of a negatively charged lipid into the bilayer appeared to yield a marked increase in negative contrast, thus rendering these magnetoliposomes suitable candidates as CAs, especially as a liver CA. magnetoliposomeshydrophobic Super Paramagnetic Iron Oxide Nanoparticles (SPIONs)Magnetic Resonance Imaging (MRI)relaxivitiescontrast agents ==== Body 1. Introduction Magnetic resonance imaging (MRI) is one of the most powerful noninvasive imaging techniques in diagnostic radiology, due to its high soft tissue contrast, spatial resolution, and penetration depth [1,2,3,4]. The magnetic fields employed MRI for clinical diagnosis are of 3 T, but preclinical studies with small animal models, which need the highest possible resolution, rely on very high field strengths (>7 T). Despite the relatively high quality of the MR images of the soft tissues, in some cases it is not possible to have enough image contrast to diagnose the pathology of interest. In these cases, a contrast agent (CA) is needed. The CA improves the contrast-to-noise ratio in MRI by shortening the spin-lattice T1 and/or spin-spin T2 relaxation times of the water protons within the tissues/regions of interest, thus enhancing the image contrast. Therefore, what is imaged is not the agent of contrast, but rather its effect on the relaxivity of the adjacent water protons, predominantly through the dipolar interaction [5]. The increase of MRI contrast produced by the magnetic nanoparticles is dependent on their composition, size, surface properties, and of the extent of aggregation in the biological milieu [6,7]. The efficiency of a CA depends on its r1 and r2 relaxivity as well as their ratio. The higher the ratio of r2/r1, the better the efficiency of a T2 CA and vice versa for a T1 CA [8,9]. Paramagnetic substances, such as gadolinium (Gd), are positive contrast agents (T1 CA). They enhance the MR signal intensity. However, they present two clinical limitations: Gd complexes have a certain degree of toxicity, and their efficiency at higher magnetic fields decreases. Due to these limitations, the research focus has shifted to negative CA such as superparamagnetic iron oxide nanoparticles (SPIONs). As compared to gadolinium compounds, SPIONs show the advantages of tunable size and shape, as well as possibility of surface modification and more effectiveness at lower concentrations because of their superparamagnetic property. SPIONs decrease the MR signal intensity of the regions where they are delivered and thus those regions appear darker in the image. Magnetite (Fe3O4) and maghemite (γ-Fe2O3) are two types of SPIONs, easily prepared in the laboratory, that have been used for biomedical applications since they meet the following criteria: (1) chemical stability under physiological conditions; (2) low toxicity and (3) high magnetic moments [10]. There are several formulations of SPIONs that have been approved by the U. S. Food and Drug Administration (FDA) and the European Medicines Agency (EMEA) for clinical use as MRI CA [11]. However, the majority of the approved compounds are, at present, out of the market [12]. The SPIONs for biomedical applications are usually prepared by hydrolytic methods, mainly by alkaline co-precipitation of stoichiometric amounts of Fe(II) and Fe(III). Such methods yield hydrophilic but usually polydisperse nanoparticles, and, in consequence, the nanoparticles can form large agglomerates in physiological media. Therefore, small well-defined SPIONs with a narrow distribution are of great interest, since magnetic properties change drastically with particle size. The thermal decomposition of metal precursors in organic media, a non-hydrolytic method, produces high-quality SPIONs with uniform size and high crystallinity. However, these SPIONs are hydrophobic. To achieve the necessary stability in aqueous media, the modification of their surface is required. Several methods based on the modification of the surface have been developed; ligand exchange with water-dispersible ligands and the encapsulation of ligands are two of the most representative strategies of such modification [7,13]. Considering the fact that phospholipids present several advantages such as biocompatibility, biodegradability, and reduced toxicity, liposomes can be used as a coating for the SPIONs. Combining nanoparticles with liposomes is a highly elaborative methodology in the emerging fields of nanomedicine and nanobiotechnology, since liposomes can carry on either hydrophobic or hydrophilic nanoparticles. SPIONs can be hybridized with liposomes to get magnetoliposomes (MLs; for a review on MLs, see [14,15]). MLs have been used as CA for molecular imaging and as a theranostic tool [16,17,18,19,20], but mainly as an efficient MRI CA with enhancing T2 contrast, although some groups have combined the T1 and T2 MRI CA in an unique system to obtain bimodal CA [16]. The magnetic properties of SPIONs depend on various factors, such as size, shape, composition, and crystallinity [11]. In this way, the proton relaxivity shows a strong dependence on the particles size. Moreover, according to the Koening–Kellar model [21] the types of surface functionalization and hydrophilicity influence the longitudinal (r1) and the transversal (r2) relaxivities. Therefore, the coating of magnetic nanoparticles greatly modifies the MRI contrast, either hampering the diffusion of water molecules, or forming hydrogen bonds with water molecules, thus increasing the residence time of water [22]. Hence, the measured proton relaxation rates depend strongly on the hydrophilic nature of the coating layer. Since liposomes can be made with different lipid formulations and can present several sizes and physical structures depending on the method of preparation, the coatings and structures interacting with SPIONs will be different. To improve the knowledge about the magnetic relaxation associated with the contrast produced by MLs, we have studied the impact that the sonication process has on the relaxivity of liposomes obtained by this method. To achieve this, we have used hydrophilic and hydrophobic magnetic particles encapsulated in liposomes. Liposomes were made of six formulations, differing in the fatty-acid chain length, the presence or absence of cholesterol (CHOL), and the presence or absence of negative charge (afforded by phosphatidylserine, (PS)). After sonication, the relaxivity properties of such hybrid nanoparticles were determined at 7 T. 2. Results and Discussion 2.1. Characteristics Size of different samples of MLs was determined by dynamic light scattering (DLS). A single population (monomodal distribution) constituted all MLs samples, with a z-diameter ranging from 220 nm to 335 nm. Polydispersity index was between 0.200 and 0.300. The size and polydispersity of such populations are strongly dependent on the sonication process. For making the comparison among the different samples possible, the conditions of the sonication process were kept invariably constant. After gel exclusion chromatography (GEC) purification, the lipid and the iron content were determined and the encapsulation efficiency was calculated as μmol of iron by mmol of lipid (Table 1). The volume accessible to the ferrofluid can explain the great difference observed in the encapsulation efficiency of liposomes containing hydrophilic or hydrophobic ferrofluid. While hydrophilic ferrofluid can be distributed either inside the aqueous interior of liposomes or in the external medium, the hydrophobic one can only be located in or near the lipid double layer. Consequently, the hydrophilic ferrofluid is encapsulated in liposomes to a lesser extent than the hydrophobic magnetic core. This also implies that the GEC purification is applied exclusively to the liposomal formulations containing hydrophilic nanoparticles. On the other hand, the number of magnetic nanoparticles encapsulated in each liposome N can be calculated with the data of the final concentrations of magnetite and lipid (Table 1). N was determined from the ratio of the total number of magnetic nanoparticles NMNP and lipid vesicles NVES. (1) N=NMNPNVES NMNP and NVES have been obtained from nanoparticle size and concentration, respectively through the following equations: (2) NMNP=MNPdNPVNP (3) NVES=NVOLNlipid where MNP is the total mass of Fe3O4 as obtained from colorimetric determination, dNP is the density of magnetite (dNP = 5.1 g·cm−3). VNP is the volume of a single nanoparticle taking a diameter of 5 nm and considering nanoparticles as perfect spheres (VNP = 65.45 nm3). Assuming that liposomes are unilamellar, for a large spherical liposome higher than 200 nm, the inner and the outer layers have the same surface and thus contain the same amount of lipids. From the average radius of each liposomal sample, the total area can be calculated. Dividing this total area by the cross-sectional area of 1,2-Dioleyl-sn-glicerol-3 phosphatidylcholine (DOPC) molecules (ao = 0.72 nm2) [23], 1,2-Dimyristoyl-sn-glicerol-3 phosphatidylcholine (DMPC) molecules (ao = 0.59 nm2) [24], PS molecules (ao = 0.70 nm2) [25] and CHOL molecules (ao = 0.35 nm2) [26], the number of lipid molecules per vesicle (Nlipid) can be calculated. An average cross-sectional area was used when liposomes were formed by more than one lipid (Table 1). Multiplying the lipid concentration of samples after GEC step by the Avogadro constant, the number of total lipids molecules (NVOL) by unit of volume can be obtained. From NMNP and NVES, the number of encapsulated magnetic particles by liposome can be determined (Equation (1)). Table 1 shows that the number of encapsulated magnetic particles is higher in liposomes containing hydrophobic particles than those encapsulating hydrophilic ones. When hydrophilic nanoparticles are used, their size is increased by the presence of the attached water molecules leading to an average hydrodynamic diameter of 20 nm. In consequence, the number of encapsulated nanoparticles inside the aqueous volume of a liposome is limited. Moreover, this encapsulation is based on random trapping by bilayer membrane closure. Thus, sorting as a function of intravesicle loading is usually a tricky problem. In the case of hydrophobic nanoparticles, these have a higher affinity to the acyl-chains confined in the hydrophobic interior of the phospholipid bilayer. However, we must take into account that the magnetic particles are 5 nm in size, and the average thickness of the bilayer ranges from 3.5 to 5 nm. For this reason, hydrophobic nanoparticles can sometimes project out of the bilayer (Figure 1A) or be embedded within two neighboring bilayers (Figure 1B). A third possibility is the formation of micelle-like assemblies (Figure 1C) by adsorption of phospholipids on the nanoparticle. The formation of the original MLs followed this strategy [27], although the size of the nanoparticles used was higher. A size of 5 nm, corresponding to the employed nanoparticles, involves a high curvature, and this would induce the creation of a large free volume between the acyl-chains of the monolayer. As can be observed in microscopy images, no free particles are observed. Hence, we only contemplated the two first possibilities. The embedding of a hydrophobic particle in a hydrophobic membrane is energetically favorable since it is the difference between a favorable Gibbs energy change produced by moving a hydrophobic particle from pure water into the bilayer and the energy needed to deform the hydrophobic membrane [28,29]. In Figure 2 TEM images of ML structures are showed. Some liposomes are in close contact, perhaps by merging of neighboring bilayers. Bilayer merging was probably driven by the interfacial activity of the nanoparticles, the energy gain by a hydrophobic surface partitioning from water into a hydrophobic environment. Cryo-TEM was used to verify nanoparticle loading and its effect on liposomal structure (Figure 3). Figure 3A,B confirm that hydrophobic nanoparticles are incorporated into the lipid bilayer. The high contrast of magnetic nanoparticles made visualization of the bilayer extremely difficult. However, the disposition of the nanoparticles in circular structures not would be observed if nanoparticles were not embedded into the bilayer. It is known that nanoparticles can disrupt vesicle formation under certain conditions, for example, when the diameter of hydrophobic nanoparticles is well above the thickness of the bilayer. In such a case, the lipid of liposomes tends to adsorb around the nanoparticles resulting in micelle-like structures instead of vesicles [29,30]. Contrarily, if the nanoparticles were of smaller diameter, the adsorption of lipid around the particles would lead to an excessive curvature. To avoid this, nanoparticles embed into the bilayer. Other studies that have also incorporated iron oxide nanoparticles into the bilayers [31,32] shown this kind of interaction. Theoretical studies have defined the mechanism of particle embedding as a result between the reduction of the energy of Gibbs obtained by moving a hydrophobic sphere from water into a hydrophobic milieu (ΔGsolv) and the increase of energy of Gibbs produced in the deformation of the bilayer (ΔGdef) [28]. In a system similar to the one used by us (nanoparticles of 5.5 nm of diameter and liposomes of dioleylphosphatidylcholine [29]) it was calculated that ΔGsolv is nearly an order of magnitude greater than ΔGdef. In our micrographs, the majority of nanoparticles are concentrated on one side of the vesicle giving some structures with appearance of Janus-type particles. This behavior has been also observed when hydrophobic gold nanoparticles were self-assembled with liposomes [33]. These authors indicated that the presence of nanoparticles in the bilayer results in a membrane deformation and separation, which establishes a void volume around the nanoparticle within the bilayer. Consequently, the nanoparticles cluster into the lipid bilayer in order to reduce the void space that surrounds them. This would explain the formation of the Janus-type nanoparticle vesicle-hybrid. Concerning the hydrophilic nanoparticles (Figure 3C,D), we observed that the size of liposomes encapsulating such nanoparticles is lesser than when liposomes were prepared with hydrophobic nanoparticles. The sonication step, although similar for both types of liposomes, produce vesicle populations of different sizes. According Michel et al. [34], in those systems formed by hydrophilic nanoparticles and fluid liposomes, where the interactions between membranes and nanoparticles are sufficiently attractive, when the radius of the particle is much smaller than the radius of the vesicle, the particle can either decorate the vesicle surface or be engulfed inside it. In the case of hydrophilic nanoparticles, their size is too small (below a critical radius that depends on the adhesion energy as well as on the value of the bilayer mean bending modulus) to be internalized inside the vesicles. We can observe that because the total adhesion energy does not overcome the energy needed for invagination, a large number of nanoparticles stay outside the liposome and form clusters. The obtained value of the hydrodynamic size for these liposomes corresponds indeed to the generated clusters—and not to the individual liposomes—as indicated by the obtained images. 2.2. Magnetic Resonance (MR) Contrast Properties The MR contrast properties of the magnetoliposomes were evaluated in vitro using 0.5% agar phantoms (T1 ≈ 2630 ms and T2 ≈ 210 ms at 7 T) which simulate tumor tissues. The phantoms contained several types of hybrid nanoparticles (magnetoliposomes with different lipid composition and two different magnetic nanoparticles). The transversal (T2) relaxation times were measured with different amounts of magnetoliposomes. Although compounds bearing superparamagnetic nanoparticles are almost exclusively T2 contrast agents [35], for comparison, longitudinal (T1) relaxation times were also determined. Magnetic resonance relaxation behavior of water protons in the presence of contrast agent in phantoms was characterized by linear relationships between iron concentration and the inverse of proton relaxation times. The slopes of the straight lines indicated different longitudinal and transverse relaxivities. In this way, the relaxivities of non-liposomal magnetic nanoparticles were determined to be r1 = 0.96 mM−1·s−1 and r2 = 74.5 mM−1·s−1 (r2/r1 ≈ 78) for hydrophilic magnetic nanoparticles, and r1 = 0.80 mM−1·s−1 and r2 = 50.7 mM−1·s−1 (r2/r1 ≈ 63) for hydrophobic magnetic nanoparticles, the concentration being expressed in iron. It is well-known that the relaxivity ratio of r2/r1 is an important parameter to estimate the efficiency of T2 CA. The hydrophobicity/hydrophilicity of the coatings has an impact on the diffusion of water within the coating layer. Concerning the nanoparticles used, the hydrophilic ones afford relaxivities that are higher than those of hydrophobic ones. The presence of oleic acid excludes water molecules around the magnetic core, and, in consequence, extends the distance of water molecules from the core. As relaxivities are strongly affected by the distance between the aqueous medium and the magnetite core, the resulting relaxivities are lower. It is difficult to compare these values with those reported for the commercial Resovist and Feridex, which are in the range of 7–14, since these values correspond to a magnetic field strength of 1.5 T [9]. Unlike the conventional paramagnetic contrast agents, i.e., gadolinium chelates, superparamagnetic nanoparticles have strong magnetic field strength dependency, and, for this reason, comparison of relaxivities must be always carried out at the same magnetic field strength. In this way, the values of relaxivities of the nanoparticles used in this study can be compared with those obtained at the same magnetic field strength with other superaparamagnetic nanoparticles employed by our group. In this way, magnetic particles of approximately 12 nm of diameter coated with polyethylenglycol of 4000 Da [36] gave the following values: r1 = 0.68 mM−1·s−1 and r2 = 311.1 mM−1·s−1; a commercial ferrofluid of similar size, stabilized with dextran (Micromod), gave r1 = 0.68 mM−1·s−1 and r2 = 337.4 mM−1·s−1. The highest r2 relaxivity values of these samples in comparison with the nanoparticles used in this study are due to the larger size of the magnetic core (12 nm vs. 5 nm), since the T2 contrast is enhanced not only by the magnetic field strength but also by the radius of iron oxide core [35]. In other words, the capability of MRI signal enhancement by nanoparticles correlates directly with the particle size [37]. Table 2 shows both types of relaxivities for the liposomal formulations used. As expected, r1 is poorly sensitive to concentration of iron oxide nanoparticles at 7 T due to the reduced susceptibility to dipolar contributions at high field, as well as the presence of bulky surface groups hindering the surface accessibility of water to the magnetic cores [38]. Contrarily, due to the superparamagnetic nature of magnetite cores, the transversal relaxivities are highly sensitive to the presence of any substances around the magnetic core. Values of r2 are higher for all the as-synthesized formulations compared with magnetic nanoparticles alone. The chemical characteristic of any liposomal surface facilitates the adsorption of a water layer around the liposome and avoids the free diffusion of these molecules towards the magnetic core resulting in high relaxivity of liposomal formulations. The maximal value of r2 was obtained with liposomes of DMPC and hydrophilic nanoparticles. As we have observed in microscopic images, hydrophilic nanoparticles tend to cluster, reducing the access of water molecules to the nanoparticles’ surfaces and greatly increasing the microscale magnetic inhomogeneity of the sample. For this reason, the visualization of the effect of the different components will be limited to those formulations containing hydrophobic nanoparticles. First, we can observe that the lipid composition affects the MRI properties. Figure 4 shows the inverse of spin-spin proton relaxivities times for DMPC and DOPC liposomes containing the same hydrophobic nanoparticles. Relaxivity values of DOPC liposomes are almost 2-fold of those obtained with DMPC liposomes. In this case, the only difference between such formulations is the length of the lipid acyl chain and the degree of lipid saturation. In this way, the r2 relaxivity of liposomes with unsaturated phospholipids (DOPC) is higher compared to those with saturated lipids (DMPC). This fact can be due to the different accessibility of water to the magnetic core. The same tendency was observed when the formulation contained negative charge (PS) (Table 2). In general, the embedding of hydrophobic nanoparticles in liposomes leaded to an r2/r1 ratio higher than that obtained with non-embedded nanoparticles. As indicated previously, the nanoparticles become confined to a part of the bilayer (like a magnetic Janus-particle) so that each liposome acts as a highly magnetizable particle. After subjecting the particle to an external magnetic field, it is able to locally acquire a magnetic moment comparable to the sum of the magnetic moments of each individual encapsulated magnetic core. It is noteworthy to mention that magnetoliposomes with PS presented the highest values of r2/r1 ratio. We observed that the presence of PS in a liposomal formulation produced a reduction in size in comparison with the non-charged liposomes [39]. Assuming that this reduction is due to more compaction of the lipid chains, the movement of the water molecules will be also affected by the presence of PS. Finally, insertion of CHOL into the bilayer seems to reduce the relaxivity values. This tendency was also observed in a study using liposomes of soybean phosphatidylcholine and CHOL [40], and was also reported in magnetoliposomes of different lipid composition, but at lower magnetic field strengths [41]. Notwithstanding the above, considering the differences between the composition of the liposomes used in the above studies and the composition of those employed by us, it is probably not viable to make a direct comparison between these two kinds of formulations here and extract conclusions about the effect of CHOL. 2.3. Stability of Liposomal Samples The size of the magnetoliposomes was determined after their incubation with isotonic saline or cell culture medium for 24 and 48 h. No significant differences were observed between the samples at room temperature and those kept at 4 °C. Magnetoliposomes incubated with culture medium aggregated. Their size after 24 h of incubation was almost double their value at time 0. The aggregation of pristine liposomes in the presence of serum components is a well-known common feature. It is widely proven that biomolecules, especially proteins, are attached to the surface of nanoparticles after their incubation with a biological milieu. These biomolecules form a dynamic shell known as protein corona [42]. However, for biomedical applications it is necessary that the surface coating of the nanoparticles presents anti-biofouling properties so that the nanoparticles are efficiently directed to the region of interest. The decoration of nanoparticles with polyethylene glycol (PEG) has been recognized as the strategy of choice, but recent studies have reported that PEGylation cannot entirely avoid the protein adsorption, although the degree of corona formation is undoubtedly diminished [43]. It is important to point out that magnetoliposomes with embedded hydrophobic nanoparticles present surface regions of a certain hydrophobicity (the hydrophobic part of any Janus-type nanoparticle), and such regions can attract very few plasma proteins. Hence, the liposome surface can be surrounded by a differential display of proteins depending on the presence or absence of hydrophobic nanoparticles in the bilayer. Magnetoliposomes incubated with isotonic saline did not show significant changes in their size (Figure 5). Average diameters remained unaltered after 24–48 h of incubation, which confirmed the good colloidal stability of the liposomes in the presence of sodium chloride. 3. Materials and Methods 3.1. Chemicals and Materials All chemicals were reagent-grade and used without purification. SPIO magnetite nanoparticles (5 nm, 5 mg/mL) dispersed in toluene and SPIO magnetite nanoparticles (5 nm, 5 mg/mL) dispersed in water were purchased from Sigma-Aldrich (St. Louis, MO, USA). Based on the density of magnetite (5.1 g/cm3), 5 mg/mL is equivalent to 9.4 × 1016 particles/mL. CHOL, 1,2-Dimyristoyl-sn-glicerol-3 phosphatidylcholine (DMPC), and 1,2-Dioleyl-sn-glicerol-3 phosphatidylcholine (DOPC) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Phosphatidylserine (PS) was purchased from Lipid Products (Redhill, Surrey, UK). Ultrapure water at 18.2 mΩ was obtained from a Millipore purification system (Millipore, Bedford, MA, USA) and used in all experiments. 3.2. Preparation of Magnetoliposomes Magnetoliposomes (MLs) were prepared according Chen et al. [44]. The lipid compositions were DMPC, DMPC:CHOL (2:1, molar ratio), DMPC:PS (9:1, molar ratio), DOPC, DOPC:CHOL (2:1, molar ratio), and DOPC:PS (9:1, molar ratio), at 10 mM lipid concentration. The lipid molecule to nanoparticle (L/N) ratio was approximately of 10,000:1. For obtaining hydrophobic magnetoliposomes (O-MLs), lipid (367 μL) and hydrophobic magnetic nanoparticles (402 μL) were mixed and placed in a round-bottom flask. Organic solvents were removed by rotary evaporation at 37 °C and reduced pressure for 45 min. Once the solvent was removed, the resulting lipid/nanoparticles film was hydrated with 1 mL of water. For hydrophilic magnetoliposomes (H-MLs), lipid (367 μL) was previously evaporated, and afterwards, the lipid was hydrated with 1 mL of an aqueous suspension of hydrophilic suspension of magnetic nanoparticles (2.01 mg/mL). After complete hydration, both suspensions (O-MLs and H-MLs) were sonicated in two steps. First, a gentle sonication in a bath sonifier (Transonic Digitals, Elma, Germany) for 30 min, and, then, five sonication steps of 10 s carried out in an UP200St ultrasonic processor (Hielscher, Teltow, Germany). The separation of liposome-encapsulated and free SPIONs was accomplished by gel exclusion chromatography (GEC). To this aim, 250 μL of MLs were applied to a 2.5-mL syringe filled with Sepharose 4B CL. After elution with water, MLs were separated from non-encapsulated ferrofluid. To exclude the possibility that hydrophilic SPIONs aggregate and co-elute with liposomes during GEC, free SPION dispersions were treated by the identical procedure used for ML formation (without addition of lipids) and assessed by GEC. 3.3. Characterization The hydrodynamic diameter and the corresponding polydispersity index (PI) of MLs were determined by dynamic light scattering at a fixed scattering angle of 90° with a Zetasizer Nano (Malvern, UK) at 25 °C. MLs from the stock solution were dispersed in water to obtain approximately 0.1 g·L−1 solid content. Geometry of MLs was observed by transmission electron microscopy (TEM) and cryo-TEM. For TEM observations, a Jeol 1010 microscope (Jeol, Tokyo, Japan) operating at 80,000 kV was used. Samples were prepared by placing a drop of MLs onto a 400-mesh copper grid coated with carbon, and after staining with uranyl acetate they were allowed to dry in the air before placing into the microscope. Images were recorded with a Megaview camera. Acquisition was accomplished with the Soft-Imaging software (SIS, Münster, Germany). For cryo-TEM observations, grids were transferred to a Tecnai F20 (FEI, Eindhoven, The Netherlands) using a cryoholder (Gatan, Warrendale, PA, USA). Images were taken at 200 kV, at a temperature ranging from −175 to −170 °C and using low-dose imaging conditions with a 4096 × 4096 pixel CCD Eagle camera (FEI, Eindhoven, The Netherlands). 3.4. Quantification of Iron and Lipid Content The iron content of MLs was determined by visible spectrophotometry on the basis of the ferrous ion using o-phenanthroline [45]. The calibration curve was performed with a solution of Fe3O4 (Aldrich, Milwaukee, WI, USA) in 12 mmol·L−1 HCl. The phospholipid content was determined by the Steward-Marshall method [46]. The calibration curve was performed with the same different lipid mixtures in chloroform that the used in the study. Absorbance was measured in a Shimadzu UV-2401 PC UV-vis spectrophotometer (Shimadzu, Kyoto, Japan). 3.5. Determination of the Relaxivity MRI experiments were conducted on a 7.0 T BioSpec 70/30 horizontal animal scanner (Bruker BioSpin, Ettlingen, Germany). T1 relaxometry maps were acquired by using RARE (rapid acquisition with rapid enhancement) sequence applying nine repetition times. For the T2 relaxometry maps, MSME (multi-slice multi-echo) sequence was used with a repetition time = 4764.346 ms with 16 echo times. The relaxation rates, R1 = 1/T1 and R2 = 1/T2 for each sample were processed using the Paravision 5.1 software (Bruker, BioSpin, Ettlingen, Germany). The relaxivity for a MRI CA is defined as the increase of the relaxation rate of the solvent (water) induced by 1 mmol·L−1 of the active ion and it is calculated according to Equation (4): (4) ri(obs)=[1Ti(obs)−1Ti(water)]/cFe Herein, i = 1 or 2, and cFe is the analytical iron concentration as determined by the o-phenanthroline reaction. The relaxivities were computed using linear regression analysis to fit relaxation rates and molar iron concentrations. To avoid particle aggregation during the period in which the magnetic field is applied, MLs and SPIONs were prepared by diluting them in air bubble-free agar phantoms. Ultra-pure agar solution was made at 75 °C. To remove air bubbles, nitrogen gas was flushed through the agar, and vacuum suction was applied. Then, the suspension was allowed to cool slowly to approximately 37 °C, and agar was mixed with MLVs or ferrofluid, and the mixing was placed in a 96-wells culture plate. The iron concentrations ranged from 0.17 to 0.49 mM. 3.6. Stability of Liposomal Samples The stability of several liposomal samples over time was evaluated by determining the change in their hydrodynamic diameter for a period of 24–48 h. Aliquots (50 μL) of purified DOPC- and DMPC-liposomes containing either hydrophobic or hydrophilic nanoparticles were diluted with 3 mL of isotonic saline or cell culture medium (Dulbecco’s modified Eagle Medium (DMEM) with 10% fetal calf serum) and kept at room temperature and in a freezer in quiescent conditions. 4. Conclusions We have characterized superparamagnetic liposomes with saturated DMPC and unsaturated DOPC phospholipids, and with or without CHOL or PS. The incorporation of hydrophilic or hydrophobic magnetic particles in the DMPC- or DOPC-based liposomes resulted in hybrid structures with higher transverse relaxation rates (r2) compared with naked magnetic nanoparticles. The main reason for this was that the diffusion coefficient of water near the nanoparticles was reduced. The liposomal coating thus ensured a longer interaction between the water protons and the magnetic field at the surface of the magnetic core than in absence of coating. Cryogenic electron transmission microscopy revealed effective embedding of hydrophobic magnetic particles into the bilayer, producing asymmetric structures similar to Janus-type nanoparticles, since nanoparticles were observed to cluster in a part of the bilayer. This finding renders cluster-liposome hybrids particularly promising candidates for MRI applications. In regards to the hydrophobic nanoparticles, the relaxivity of liposomes with unsaturated phospholipids was higher than that of those with saturated lipids. CHOL led to a smaller reduction in relaxivity. The highest relaxivity was obtained for magnetoliposomes containing PS. In general, it is feasible to use magnetoliposomes as negative contrast agents for MRI, making these nanosystems good candidates to be taken into consideration for molecular imaging. The contrast is particularly enhanced when hydrophobic nanoparticles are used, since, in this case, due to the uniform size and high crystallinity of the nanoparticles, there was an additional clustering effect into the bilayer. Because of these facts, high values of r2 are obtained. Furthermore, the diffusivity of the water in the coating can be modulated by changing the liposome composition (charge, degree of saturation, length of the hydrocarbon chain, and presence or absence of CHOL). These hybrid structures can be administered intravenously and due to the absence of steric stabilizers, e.g., PEG, they are metabolized by cell of the mononuclear phagocyte system (MPS), accumulating in liver, spleen, and bone marrow. In this way, these magnetoliposomes can be used for tumor detection, especially in liver diseases. Magnetic nanoparticles decrease the magnetic resonance signal intensity due to the uptake by kupffer cells. Therefore, the image of the tissue appears dark. However, tumor lesions reduce the uptake and the image of the tissue appears bright relative to the surrounding tissue. Thus, magnetic particles produce a strong contrast between normal and abnormal liver, thereby enabling clear detection of the abnormal tissue. Acknowledgments Maria Antònia Busquets and Joan Estelrich are grateful for the financial support given by the Spanish Ministry of Economy and Competitiveness for project MAT2012-36270-C04-03. The authors also thank the Generalitat of Catalunya for funding project 2014SGR227. Author Contributions Maria Antònia Busquets and Joan Estelrich conceived and designed the experiments; and wrote the manuscript. Raquel Martínez-González was in charge of the experimental section. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Location of the hydrophobic magnetic particle in the liposome: (A) Projected out of the bilayer; (B) embedded within neighboring bilayers and; (C) formation of micelle-like assemblies. Hydrophobic nanoparticles are represented by a dark blue circle whereas the phospholipids are the light blue parts (polar head joined to a two acyl chains). In a water milieu, phospholipids can form bilayers (A,B) or monolayers (micelle-like-assemblies as depicted in C). Figure 2 TEM images of DMPC liposomes encapsulating hydrophobic magnetic nanoparticles (MNPs). (A) The size of liposomes is indicated (scale bar: 200 nm); (B) Magnetic nanoparticles of an average diameter of 5 nm are visualized in the DMPC liposomes (scale bar: 100 nm). Figure 3 Cryo-TEM images of DMPC liposomes with hydrophobic (A,B) and hydrophilic (C,D) MNPs (scale bar: 100 nm). Liposomes are partially loaded with hydrophobic nanoparticles forming Janus-type nanoparticle-liposome hybrids (A,B). Hydrophilic particles are assembled forming branched (C) and lineal clusters (D). Figure 4 (Upper) T2 weighted contrast and color maps for iron oxide nanoparticles alone and in DOPC or DMPC liposomes; (Lower) Relaxation rates T2−1 as a function of iron concentration of magnetoliposomes formed by hydrophobic iron oxide nanoparticles and DOPC or DMPC. Figure 5 Change in the average diameter of four liposomal formulations after incubation with isotonic saline. The bar shows the size immediately after the mixing of the liposomes with saline. The value of the size after 24 and/or 48 h is marked with a dot. Magnetic hydrophilic (H) and hydrophobic (O) nanoparticles; DMPC: 1,2-Dimyristoyl-sn-glicerol-3 phosphatidylcholine; DOPC: 1,2-Dioleyl-sn-glicerol-3 phosphatidylcholine. ijms-17-01209-t001_Table 1Table 1 Encapsulation efficiency, average of cross-sectional area, and number (N) of magnetic hydrophilic (H) and hydrophobic (O) nanoparticles encapsulated into liposomes of different lipid composition. MLs: Magnetoliposomes; DMPC: 1,2-Dimyristoyl-sn-glicerol-3 phosphatidylcholine; DOPC: 1,2-Dioleyl-sn-glicerol-3 phosphatidylcholine; CHOL: Cholesterol and PS: Phosphatidylserine. MLs Encapsulation Efficiency/μmol Magnetite Average Cross-Sectional Area/nm2 N H-DMPC 6.9 0.59 3 O-DMPC 43.6 0.59 16 O-DMPC-CHOL 22.8 0.51 11 O-DMPC-PS 20.4 0.63 10 H-DOPC 7.9 0.72 2 O-DOPC 274.0 0.72 17 O-DOPC:CHOL 49.0 0.60 15 O-DOPC-PS 17.7 0.71 9 ijms-17-01209-t002_Table 2Table 2 r1, r2 and r2/r1 ratio of magnetic hydrophilic (H) and hydrophobic (O) nanoparticles encapsulated into liposomes of different lipid composition. MLs: Magnetoliposomes. MLs r1/mM−1·s−1 r2/mM−1·s−1 r2/r1 H-DMPC 9.1 1282 140 O-DMPC 0.9 340 378 O-DMPC-CHOL 0.8 230 288 O-DMPC-PS 0.8 798 ~1000 H-DOPC 3.4 678 199 O-DOPC 0.9 630 700 O-DOPC:CHOL 0.9 281 312 O-DOPC-PS 0.9 995 ~1000 ==== Refs References 1. Brown M.A. Semelka R.C. MRI: Basic Principles and Applications 4th ed. Wiley-Blackwell Hoboken, NJ, USA 2010 2. Brindle K. New approaches for imaging tumour responses to treatment Nat. Rev. Cancer 2008 8 94 107 10.1038/nrc2289 18202697 3. Nakada T. Clinical application of high and ultra high-field MRI Brain Dev. 2007 29 325 335 10.1016/j.braindev.2006.10.005 17113259 4. Vande Velde G. Baekelandt V. Dresselaers T. Himmelreich U. Magnetic resonance imaging and spectroscopy methods for molecular imaging Q. J. Nucl. Med. Mol. Imaging 2009 53 565 585 20016450 5. Amiri H. Saeidi K. Borhani P. Manafirad A. Ghavami M. Zerbi V. Alzheimer’s disease: Pathophysiology and applications of magnetic nanoparticles as MRI theranostic agents ACS Chem. Neurosci. 2013 4 1417 1429 10.1021/cn4001582 24024702 6. Veiseh O. Gunn J.W. Zhang M. Design and fabrication of magnetic nanoparticles for targeted drug delivery and imaging Adv. Drug. Deliv. Rev. 2010 62 284 304 10.1016/j.addr.2009.11.002 19909778 7. Na H. Song I. Hyeon T. Inorganic nanoparticles for MRI contrast agents Adv. Mater. 2009 21 2133 2148 10.1002/adma.200802366 8. Gultepe E. Reynoso F.J. Jhaveri A. Kulkarni P. Nagesha D. Ferris C. Harisinghani M. Campbell R.B. Sridhar S. Monitoring of magnetic targeting to tumor vasculature through MRI and biodistribution Nanomedicine 2010 5 1173 1182 10.2217/nnm.10.84 21039195 9. Qin J. Laurent S. Jo Y.S. Roch A. Mikhaylova M. Bhujwalla Z.M. Muller R.N. Muhammed M. A high-performance magnetic resonance imaging T 2 contrast agent Adv. Mater. 2007 19 1874 1878 10.1002/adma.200602326 10. Filippousi M. Angelakeris M. Katsikini M. Paloura E. Efthimiopoulos I. Wang Y. Zamboulis D. van Tendeloo G. Surfactant effects on the structural and magnetic properties of iron oxide nanoparticles J. Phys. Chem. C 2014 118 16209 16217 10.1021/jp5037266 11. Stephen Z.R. Kievit F.M. Zhang M. Magnetite Nanoparticles for medical MR imaging Mater. Today 2011 14 330 338 10.1016/S1369-7021(11)70163-8 12. Bulte W. In vivo MRI cell tracking: Clinical studies Am. J. Roentgenol. 2009 193 314 325 10.2214/AJR.09.3107 19620426 13. Colombo M. Carregal-Romero S. Casula M.F. Gutiérrez L. Morales M.P. Böhm I.B. Heverhagen J.T. Prosperi D. Parak W.J. Biological applications of magnetic nanoparticles Chem. Soc. Rev. 2012 41 4306 4334 10.1039/c2cs15337h 22481569 14. Fattahi H. Laurent S. Liu F. Arsalani N. Vander Elst L. Muller R.N. Magnetoliposomes as multimodal contrast agents for molecular imaging and cancer nanotheragnostics Nanomedicine 2011 6 529 544 10.2217/nnm.11.14 21542690 15. Soenen S. vande Velde G. Ketkar-Atre A. Himmelreich U. de Cuyper M. Magnetoliposomes as magnetic resonance imaging contrast agents WIREs Nanomed. Nanotechnol. 2011 3 197 211 10.1002/wnan.122 25363747 16. De Cuyper M. The original magnetoliposomes: From the physicochemical Basis to theranostic nanomedicine Soft Nanoparticles for Biomedical Applications Callejas-Fernández J. Estelrich J. Quesada-Pérez M. Forcada J. Royal Society of Chemistry Cambridge, UK 2014 109 132 17. Bulte J.W. de Cuyper M. Magnetoliposomes as contrast agents Methods Enzymol. 2003 373 175 198 14714404 18. Martina M.S. Fortin J.P. Menager C. Clement O. Barratt G. Grabielle-Madelmont C. Gazeau F. Cabuil V. Lesieur S. Generation of superparamagnetic liposomes revealed as highly efficient MRI contrast agents for in vivo imaging J. Am. Chem. Soc. 2005 127 10676 10685 10.1021/ja0516460 16045355 19. Plassat V. Martina M.S. Barratt G. Menager C. Lesieur S. Sterically stabilized superparamagnetic liposomes for MR imaging and cancer therapy: Pharmacokinetics and biodistribution Int. J. Pharm. 2007 344 118 127 10.1016/j.ijpharm.2007.05.018 17583452 20. Frascione D. Diwoky C. Almer G. Opriessnig P. Vonach C. Gradauer K. Leitinger G. Mangge H. Stollberger R. Prassl R. Ultrasmall superparamagnetic iron oxide (USPIO)-based liposomes as magnetic resonance imaging probes Int. J. Nanomed. 2012 7 2349 2359 21. Koenig S.H. Kellar K.E. Theory of 1/T 1 and 1/T 2 NMRD profiles of solutions of magnetic nanoparticles Magn. Reson. Med. 1995 34 227 233 10.1002/mrm.1910340214 7476082 22. LaConte L.E. Nitin N. Zurkiya O. Caruntu D. O′Connor C.J. Hu X. Bao G. Coating thickness of magnetic iron oxide nanoparticles affects R2 relaxivity J. Magn. Reson. Imaging 2007 26 1634 1641 10.1002/jmri.21194 17968941 23. Tristam-Nagle S. Nagle J.F. Lipid bilayers: Thermodynamic structure, fluctuations and interactions Chem. Phys. Lipids 2004 127 3 14 10.1016/j.chemphyslip.2003.09.002 14706737 24. Kucerka N. Kiselev M.A. Balgavy P. Determination of bilayer thickness and lipid surface area in unilamellar dimyristoylphosphatidylcholine vesicles from small-angle neutron scattering curves: A comparison of evaluation methods Eur. J. Biophys. 2004 33 328 334 10.1007/s00249-003-0349-0 12955364 25. Eisenberg M. Gresalfi T. Riccio T. McLaughlin S. Adsorption of monovalent cations to bilayer membranes containing negative phospholipids Biochemistry 1979 18 5213 5223 10.1021/bi00590a028 115493 26. Armengol X. Estelrich J. Physical stability of different liposome compositions obtained by extrusion method J. Microencapsul. 1995 12 525 535 10.3109/02652049509006783 8544096 27. De Cuyper M. Joniau M. Magnetoliposomes. Formation and structural characterization Eur. Biophys. J. 1988 15 311 319 10.1007/BF00256482 3366097 28. Ginzburg V.V. Balijepalli S. Modeling the thermodynamics of the interaction of nanoparticles with cell membranes Nano Lett. 2007 7 3716 3722 10.1021/nl072053l 17983249 29. Wi H.S. Lee K.-B. Pak H.K. Interfacial energy consideration in the organization of a quantum dot-lipid mixed system J. Phys. Condens. Matter 2008 20 494211:1 494211:6 30. Roiter Y. Ornatska M. Rammohan A.R. Balakrishnan J. Heine D.R. Minko S. Interaction of lipid membrane with nanostructured surfaces Langmuir 2009 25 6287 6299 10.1021/la900119a 19466783 31. Amstad E. Kohlbrecher J. Muller E. Schweizer T. Textor M. Reimhult E. Triggered release from liposomes through magnetic actuation of iron oxide nanoparticle containing membranes Nano Lett. 2011 11 1664 1670 10.1021/nl2001499 21351741 32. Krack M. Hohenberg H. Kornowski A. Lindner P. Weller H. Forster S. Nanoparticle-loaded magnetophoretic vesicles J. Am. Chem. Soc. 2008 130 7315 7320 10.1021/ja077398k 18484723 33. Rasch M.R. Rossinyol E. Hueso J.L. Goodfellow B.W. Arbiol J. Korgel B.A. Hydrophobic gold nanoparticle self-assembly with phosphatidylcholine lipid: Membrane-loaded and janus vesicles Nano Lett. 2010 10 3733 3739 10.1021/nl102387n 20731366 34. Michel R. Gradzielski M. Experimental aspects of colloidal interactions in mixed systems of liposome and inorganic nanoparticle and their applications Int. J. Mol. Sci. 2012 13 11610 11642 10.3390/ijms130911610 23109874 35. Estelrich J. Sanchez-Martin M.J. Busquets M.A. Nanoparticles in magnetic resonance imaging: From simple to dual contrast agents Int. J. Nanomed. 2015 10 1727 1741 36. García-Jimeno S. Estelrich J. Ferrofluid based on polyethylene glycol-coated iron oxide nanoparticles: Characterization and properties Colloids Surf. A 2013 420 74 81 10.1016/j.colsurfa.2012.12.022 37. Huang J. Zhong X. Wang L. Yang L. Mao H. Improving the magnetic resonance imaging contrast and detection methods with engineered magnetic nanoparticles Theranostics 2012 2 86 102 10.7150/thno.4006 22272222 38. Saha I. Chaffee K.E. Duanmu C. Woods B.M. Stokes A.M. Buck L.E. Walkup L.L. Sattenapally N. Huggenvik J. Gao Y. pH-Sensitive MR Responses Induced by dendron-functionalized SPIONs J. Phys. Chem. C 2013 117 1893 1903 10.1021/jp306128v 23494078 39. Roy M.T. Gallardo M. Estelrich J. Bilayer distribution of phosphatidylserine and phosphatidylethanolamine in lipid vesicles Bioconjug. Chem. 1997 8 941 945 10.1021/bc9701050 9404670 40. Carvalho A. Goncalves M.C. Martins M.B. Meixedo D. Feio G. Relaxivities of magnetoliposomes: The effect of cholesterol Magn. Reson. Imaging 2013 31 610 612 10.1016/j.mri.2012.10.024 23333578 41. Skouras A. Mourtas S. Markoutsa E. de Goltstein M.C. Wallon C. Catoen S. Antimisiaris S.G. Magnetoliposomes with high USPIO entrapping efficiency, stability and magnetic properties Nanomedicine 2011 7 572 579 10.1016/j.nano.2011.06.010 21704597 42. Amiri H. Bordonali L. Lascialfari A. Wan S. Monopoli M.P. Lynch I. Laurent S. Mahmoudi M. Protein corona affects the relaxivity and MRI contrast efficiency of magnetic nanoparticles Nanoscale 2013 5 8656 8665 10.1039/c3nr00345k 23896964 43. Gao Z. Ma T. Zhao E. Docter D. Yang W. Stauber R.H. Gao M. Small is smarter: Nano MRI contrast agents—Advantages and recent achievements Small 2016 12 556 576 10.1002/smll.201502309 26680328 44. Chen Y. Bose A. Bothun G.D. Controlled release from bilayer-decorated magnetoliposomes via electromagnetic heating ACS Nano 2010 6 3215 3221 10.1021/nn100274v 20507153 45. Kiwada H. Sato J. Yamada S. Kato Y. Feasibility of magnetic liposomes as a targeting device for drugs Chem. Pharm. Bull. 1986 34 4253 4258 10.1248/cpb.34.4253 3829157 46. Steward-Marshall J.C. Colorimetric determination of phospholipids with ammonium ferrothiocyanate Anal. Biochem. 1980 104 10 14 6892980

PMC005xxxxxx/PMC5000608.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081210ijms-17-01210ArticleStage-Related Defense Response Induction in Tomato Plants by Nesidiocoris tenuis Naselli Mario 1Urbaneja Alberto 2Siscaro Gaetano 1Jaques Josep A. 3Zappalà Lucia 1Flors Víctor 3Pérez-Hedo Meritxell 23*Maffei Massimo Academic EditorBarbero Francesca Academic Editor1 Dipartimento di Agricoltura, Alimentazione e Ambiente (Di3A), University of Catania, Via Santa Sofia 100, 95123 Catania, Italy; [email protected] (M.N.); [email protected] (G.S.); [email protected] (L.Z.)2 Unidad Asociada de Entomología UJI-IVIA, Centro de Protección Vegetal y Biotecnología, Instituto Valenciano de Investigaciones Agrarias (IVIA), Carretera de Moncada-Náquera Km. 4.5, Moncada, 46113 Valencia, Spain; [email protected] (A.U.); [email protected] (M.P.-H.)3 Unitat Associada d’Entomologia UJI-IVIA, Departament de Ciències Agràries i del Medi Natural, Universitat Jaume I, UJI, Campus del Riu Sec, 12071 Castelló de la Plana, Spain; [email protected] (J.A.J.); [email protected] (V.F.)* Correspondence: [email protected]; Tel.: +34-96-342-41-1527 7 2016 8 2016 17 8 121024 5 2016 19 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The beneficial effects of direct predation by zoophytophagous biological control agents (BCAs), such as the mirid bug Nesidiocoris tenuis, are well-known. However, the benefits of zoophytophagous BCAs’ relation with host plants, via induction of plant defensive responses, have not been investigated until recently. To date, only the females of certain zoophytophagous BCAs have been demonstrated to induce defensive plant responses in tomato plants. The aim of this work was to determine whether nymphs, adult females, and adult males of N. tenuis are able to induce defense responses in tomato plants. Compared to undamaged tomato plants (i.e., not exposed to the mirid), plants on which young or mature nymphs, or adult males or females of N. tenuis fed and developed were less attractive to the whitefly Bemisia tabaci, but were more attractive to the parasitoid Encarsia formosa. Female-exposed plants were more repellent to B. tabaci and more attractive to E. formosa than were male-exposed plants. When comparing young- and mature-nymph-exposed plants, the same level of repellence was obtained for B. tabaci, but mature-nymph-exposed plants were more attractive to E. formosa. The repellent effect is attributed to the signaling pathway of abscisic acid, which is upregulated in N. tenuis-exposed plants, whereas the parasitoid attraction was attributed to the activation of the jasmonic acid signaling pathway. Our results demonstrate that all motile stages of N. tenuis can trigger defensive responses in tomato plants, although these responses may be slightly different depending on the stage considered. Bemisia tabaciEncarsia formosatomatoinduced plant responseindirect defensephytohormones ==== Body 1. Introduction Plants are able to defend themselves from arthropods, pathogens and, in general, from biotic and abiotic stress conditions [1,2,3]. To this end, plants activate a cascade of events that include transcriptome changes of some of the genes involved in the biosynthesis of phytohormones that lead—directly and indirectly—to defensive responses [4,5]. The main phytohormones responsible for these responses are jasmonic, salicylic, abscisic acids, and ethylene (JA, SA, ABA, and ET respectively) [1,2,3,6,7,8]. Depending on the herbivore’s feeding habits (chewing, phloem, or cell content feeders), different hormone-related signaling pathways are triggered [9,10,11,12,13]. For instance, it is known that insects with piercing-sucking mouthparts (especially phloem feeders like most of the Hemiptera) mostly induce the SA-mediated resistance pathway, whereas insects with chewing mouthparts predominantly trigger the JA pathway [14,15,16]. However, JA may also be induced by cell content feeders such as thrips (Thysanoptera: Thripidae) and spider mites (Acari: Tetranychidae) [8]. Zoophytophagous arthropods—which feed both on other arthropod as prey and on plants during the same developmental stages—can also activate the same defense mechanisms as strict herbivores [17,18,19,20,21]. It is well-known that zoophytophagy provides adaptive advantages, such as the ecological flexibility to consume both prey and plants, thereby allowing the survival of these predators on plants when prey is scarce [22,23,24]. This is the case for Nesidiocoris tenuis Reuter (Hemiptera: Miridae), a widely used biological control agent (BCA) which has been extremely effective in controlling some key tomato pests, including the tobacco whitefly Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) and the invasive South American pinworm Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) [25,26,27]. Apart from this predation-dependent beneficial effect, the activity of N. tenuis females (feeding and/or oviposition) on tomato plants activates the ABA and JA pathways, which make tomato plants less attractive to phytophagous B. tabaci and more attractive to the whitefly parasitoid Encarsia formosa (Gahan) (Hymenoptera: Aphelinidae), respectively [21]. In addition, herbivore-induced plant volatiles (HIPVs) from N. tenuis-exposed plants can induce plant defenses in neighboring, undamaged (not exposed to the mirids) plants via JA, which result in the attraction of parasitoids [21]. These effects on plant defensive responses might be a reasonable explanation of the achievement reached by N. tenuis in integrated pest management programs in tomatoes. Pérez-Hedo, et al. [20] showed that the females of three different zoophytophagous BCAs (N. tenuis, Macrolophus pygmaeus Rambur, and Dicyphus maroccanus Wagner) differ in their ability to induce defensive responses in tomato plants, resulting in varying degrees of attractiveness of the plants to pests and natural enemies. In the case of tomato plants exposed to and therefore presumably punctured by N. tenuis, these plants were less attractive to the whitefly B. tabaci and to the lepidopteran T. absoluta. In contrast, tomato plants exposed to M. pygmaeus and D. maroccanus were not able to repel B. tabaci and, more interestingly, became more attractive to T. absoluta. Pappas, et al. [28] showed that tomato plants exposed to adult females, fifth instar nymphs, and young virgin females of M. pygmaeus were able to induce plant resistance against the two-spotted spider mite Tetranychus urticae Koch (Acari: Tetranychidae). In tomato plants infested by T. urticae, the number of T. urticae eggs laid was lower when these tomato plants had been exposed previously to M. pygmaeus [28]. To date, it has been shown only that the feeding and oviposition activities of N. tenuis adult females induce defensive plant responses in tomato plants [20,21]. Nevertheless, under field conditions, it is usually common to find a mix of instars and/or stages of this and other mirids [29]. To know whether nymphal instars and males are also able to induce defensive plant responses, in this work we evaluated the response induced by the feeding (plus oviposition in the case of adult females) activity of different instars/stages of N. tenuis compared to undamaged plants. This response was assessed by means of behavioral bioassays in a Y-tube olfactometer using adults of the herbivore B. tabaci and of the entomophagous parasitoid E. formosa. In addition, ultra-performance liquid chromatography coupled to mass spectrometry (UPLC-MS) and quantitative gene expression of selected phytohormones and genes, were analyzed. 2. Results 2.1. Olfactory Responses Induced by N. tenuis-Punctured Plant When plants exposed to young (NI and NII) or mature nymphs (NIV and NV) of N. tenuis were compared to undamaged plants (i.e., not exposed to the mirid), a clear preference of B. tabaci for undamaged plants was observed (χ2 = 5.000, p = 0.0253 and χ2 = 7.200, p = 0.0073, respectively, Figure 1a and Figure 2a). Similarly, undamaged tomato plants were more attractive for B. tabaci than tomato plants exposed to either adult females or males of N. tenuis (χ2 = 9.800, p = 0.0017 and χ2 = 3.951, p = 0.0468, respectively, Figure 3a and Figure 4a). Unlike B. tabaci, all plants exposed to the mirids were found to be more attractive to the parasitoid E. formosa than to undamaged plants (χ2 = 4.267, p = 0.0389 for young nymphs—Figure 1a; χ2 = 9.600, p = 0.0019 for mature nymphs—Figure 2a; χ2 = 6.898, p = 0.0086 for adult males—Figure 3a; and χ2 = 11.640, p = 0.0006 for adult females—Figure 4a). When adult-female-exposed plants were compared with plants exposed to adult males, the former were more repellent to B. tabaci (χ2 = 5.000, p = 0.0253) and more attractive to E. formosa (χ2 = 10.90, p = 0.0010, Figure 5b) than those exposed to adult males. However, when plants exposed to either young or mature nymphs were exposed to B. tabaci, no significant differences were detected (χ2 = 1.852, p = 0.1736). However, plants exposed to mature nymphs were more attractive to the parasitoid E. formosa than those exposed to young nymphs (χ2 = 6.667; p = 0.0098, Figure 5a). 2.2. Phytohormones Analysis and Plant Gene Expression In the apical part of all plants exposed to N. tenuis, the endogenous levels of ABA and of the bioactive phytohormone JA-Isoleucine (JA–ILE) were higher than in the apical part of undamaged plants (ABA: t = 1.992, p = 0.0359 for young nymphs—Figure 1c; t = 2.206, p = 0.0260 for mature nymphs—Figure 2c; t = 6.065, p < 0.0001 for males—Figure 3c; t = 4.298, p = 0.0008 for females—Figure 4c; and jasmonate-isoleucine (JA-Ile): t = 1.815, p = 0.0498 for young nymphs—Figure 1c; t = 2.153, p = 0.0272 for mature nymphs—Figure 2c; t = 3.711, p = 0.0020 for males—Figure 3c; t = 3.344, p = 0.0037 for females—Figure 4c). Consistent with these observations, the analysis of the relative expression of genes involved in indirect defense showed transcriptional differences between N. tenuis-exposed plants and undamaged plants. The ASR1 gene (a marker for ABA) and the PIN2 gene (a marker for JA) were upregulated in all N. tenuis-exposed plants compared to undamaged plants (ASR1: t = 4.276, p = 0.0010 for young nymphs—Figure 1b; t = 5.227, p = 0.0003 for mature nymphs—Figure 2b; t = 3.239, p = 0.0059 for males—Figure 3b; t = 3.730, p = 0.0023 for females—Figure 4b; and PIN2: t = 2.374, p = 0.0225 for young nymphs—Figure 1b; t = 15.65, p < 0.0001 for mature nymphs—Figure 2b; t = 8.185, p < 0.0001 for males—Figure 3b; t = 3.033, p = 0.0081 for females—Figure 4b). 3. Discussion Our results are the first evidence that all motile stages of N. tenuis can induce defensive responses in tomato plants. All motile stages can damage plants by feeding on them, whereas adult N. tenuis females may also cause damage through oviposition. Our results also demonstrate the direct relationship between N. tenuis’ plant feeding and defense induction in tomato. Specifically, the feeding on tomato plants by N. tenuis NI, NII, NIV, NV, adult males, and adult females all resulted in reduced attractiveness for B. tabaci and enhanced attractiveness for E. formosa relative to undamaged plants. Olfactometer results confirm the positive correlation between ABA concentrations and induced repellence to whiteflies as well as between JA and induced attraction of E. formosa [20,21]. Interestingly, the stronger behavioral response observed in plants exposed to N. tenuis adult females in the olfactometer could be attributed to chemicals either emitted or triggered by N. tenuis eggs inserted into the plant tissues, as occurred in other phytophagous species [19,30]. Therefore, the presence of N. tenuis eggs on a plant could cause a synergistic effect with feeding, resulting in enhanced repellence for B. tabaci and enhanced attraction of E. formosa in olfactory bioassays. In sum, the dual activity of females (i.e., feeding and oviposition) could explain these results. However, further research is needed. Indeed, volatile organic compounds (VOCs) emitted by plants in response to herbivore attack—either by feeding and/or by endo/esophytic oviposition—are known to repel further herbivore attacks [30,31]. Thus, the higher intensity of the effect produced by N. tenuis adult females on exposed plants might explain the stronger attraction of E. formosa, as previously demonstrated in different biological models [32]. Therefore, the potential presence of egg elicitors has to be better investigated. Another issue is the absence of specific defense responses by the plant to the mirid eggs, but this could be explained by the high degree of adaptation of these mirids to plants, confirming their mutualistic relations and suggesting a co-evolutionary approach to understanding these interactions. Furthermore, the results suggest that the joint use of different instars of N. tenuis under greenhouse conditions is a better implementation strategy with this mirid, because the simultaneous presence of different cohorts avoids strong hormonal fluctuations in tomato plants, thereby reducing the negative impact on harvest. Abiotic stresses, such as water stress or desiccation, induce the activation of the ABA pathway [33,34,35]. Nevertheless, ABA has been reported both as an inducer of plant defense response to necrotrophic pathogens and as an inhibitor of biotrophic pathogens [8]. Apart from the work of Pérez-Hedo, et al. [21], where high levels of ABA in tomato plants were shown to repel the whitefly B. tabaci, little information is available in relation to the effects of ABA on arthropods. JA is known for inducing direct and indirect plant defense responses against arthropods, this phytohormone works together with ET in orchestrating these responses [1,8]. The direct defense consists of the production of secondary metabolites, such as proteinase inhibitors, that inhibit the development of insects on activated plants [36,37], whereas indirect defense has been recently observed in trophic interaction studies. Particularly, this latter phenomenon is due to the production and release of VOCs. The synthesis and emission of VOCs is triggered by JA synthesis and mediates the attraction or rejection of beneficial and phytophagous species, respectively [10,36,38,39]. All the above-mentioned phytohormonal activity for indirect plant defense—which is induced by herbivores—culminates in the production and release of HIPVs [40]. These HIPVs are the signals for plant–plant and plant–arthropod communication, and result in the attraction of natural enemies and the repellence of herbivores, demonstrating the key role played by plants in orchestrating tritrophic interactions [30]. Future research may be oriented toward the extraction and characterization of VOCs involved in repelling herbivores and attracting natural enemies. Such research would lead to a better understanding of the phenomenon and help researchers to acquire new elements for future practical applications. Our results highlight that in those crops where the release and subsequent conservation of N. tenuis is common practice (e.g., more than 80% of tomato greenhouses in southeastern Spain [41]), the persistence of this mirid on the plant throughout the growing season is an extra benefit in protecting them against pests. In southeastern Spain, it is common practice to release N. tenuis adults on to the seedlings at a ratio of 0.5–1 adults per plant, and this release occurs approximately seven days before transplanting [25]. During this period, these adults—which are fed with eggs of the Mediterranean flour moth Ephestia kuehniella Zeller (Lepidoptera: Pyralidae)—lay eggs on the tomato plants such that when these plants are transplanted to the greenhouse, they already carry eggs of N. tenuis. Nevertheless, this is not the only benefit of this pre-plant release. During this seedling period, N. tenuis also feeds on the tomato plants, thereby activating the defenses of these plants [21]. Recently, Pappas, et al. [42], using M. pygmaeus as a model, proposed the term “plant vaccination” to describe this type of defensive activation induced in seedlings, because tomato plants reaching the greenhouse are “ready-to-defend” against herbivory. In the case of M. pygmaeus, Pappas, et al. [28] showed that this induction can last up to two weeks after the tomato plant comes in contact with the mirid. We have confirmed that this induction period is similar or longer for N. tenuis (same authors, unpublished data) such that the tomato plant would be “vaccinated” in the nursery, and this effect would last until the newly emerged nymphs start to feed on the plant, possibly renewing this defensive induction (as our results suggest). This effect on plants by the mirid could potentially counterbalance its potential damage on fruit [43]. The vaccination effect should be verified through field evaluations to confirm that plant activation is possible throughout the seedling, establishment, and conservation periods of N. tenuis in the crop. 4. Materials and Methods 4.1. Plants and Insects Tomato plants Solanum lycopersicum cv. Optima (Seminis Vegetable Seeds, Inc., Almería, Spain) were sown in soil, and two weeks after germination, seedlings were individually transferred to pots (8 × 8 × 8 cm). Plants were maintained undisturbed at 25 ± 2 °C, while relative humidity and photoperiod were held constant at 65% ± 5% relative humidity and 14:10 h (Light:Dark). Pesticide-free tomato plants were used for the experiments at seven weeks of age (approximately 20 cm high). B. tabaci, E. formosa, and all instars and stages of N. tenuis were provided directly by Koppert Biological Systems, S.L. (Águilas, Murcia, Spain). Young nymphs consisted of a proportional mix of NI and NII, whereas mature nymphs corresponded to a proportional mix of NIV and NV. Adult females and males of N. tenuis were less than 4 days old, whereas adult females of B. tabaci and E. formosa were less than 2 days old. To obtain N. tenuis-punctured plants, four undamaged tomato plants were enclosed for 24 h in a 60 × 60 × 60 cm plastic cage (BugDorm-2; Mega View Science Co., Ltd., Taichung, Taiwan) and exposed to 80 N. tenuis of the corresponding nymphal instar or adult sex (20 individuals per plant). All motile individuals were removed from plants before the experiment. 4.2. Y-Tube Bioassays The olfactory preference of B. tabaci and E. formosa for different scent sources was tested using a Y-tube olfactometer (Analytical Research Systems, Gainesville, FL, USA). This Y-tube consisted of a Y-shaped glass tube (2.4 cm in diameter with a base of 13.5 cm in length) which was connected to two identical 5 L glass jars via plastic tubes. Each jar was connected to an air pump that produced a unidirectional airflow at a rate of 150 mL/min and contained a tested odor source (tomato plant). The experiments were conducted at 23 ± 2 °C, 60% ± 10% RH, and a light intensity of 2516 lux [44]. The first set of observations was conducted comparing the olfactory preference of B. tabaci and E. formosa adult females for each treatment of N. tenuis-punctured plants (young nymphs—NI and NII, mature nymphs—NIV and NV, adult males and females) relative to undamaged plants, while another series of observations was carried out comparing adult-male- versus adult-female-exposed plants and young-nymph- versus mature-nymph-exposed plants. The choice of each B. tabaci and E. formosa adult female was recorded when the insect walked a distance of 3 cm in the chosen arm; in the case that a female did not make a choice after 15 min, they were excluded from the analysis. Each individual was used only once. After testing five individuals, odor sources were interchanged to avoid any influence of asymmetries in the setup. Thirty to forty valid replicates were performed for each treatment. 4.3. Phytohormone Analysis Apical parts of punctured tomato plants were exposed to N. tenuis at different instar stages for 24 h, and samples from undamaged plants were stored at −80 °C and analyzed to compare phytohormone concentrations. The apical part was considered to be the first 5 cm of the plant formed by the apical stem and young leaves. The phytohormone profile was analyzed using ultra-performance liquid chromatography coupled with mass spectrometry (UPLC-MS) [6,21,45]. This method can detect the concentrations of the two phytohormones involved in the tomato plant defensive responses: ABA and JA–Ile. 4.4. Quantification of Plant Gene Expression The apical parts of the plants were used to quantify the gene expression of ASR1 (ABA stress ripening protein 1)—a marker gene for ABA—and PIN2 (proteinase inhibitors 2), a marker gene for JA. Immediately after collection, apical samples were ground in liquid nitrogen and a portion of them served for RNA extraction. Plant RNA Kit (Omega Bio-TekInc, Doraville, GA, USA) was used to extract total RNA (1.5 μg), and RNase-free DNase (Promega Corporation, Madison, WI, USA) was employed to eliminate genomic DNA contamination. Reverse transcription, primers, and the PCR SYBR Green reaction were carried out as previously described by Pérez-Hedo, et al. [21]. Quantitative PCR was performed with the Smart Cycler II (Cepheid, Sunnyvale, CA, USA) sequence detector using standard PCR conditions. Expression of EF1 (Elongation factor 1) was used for normalization as housekeeping gene. Table 1 describes the sequences of the gene-specific primers used. 4.5. Data Analyses Data sets obtained from the Y-tube olfactometer observations were analyzed using a chi-square (χ2) test in order to evaluate whether the response of insects to different scent sources deviated from a null model, where odor sources were chosen with equal frequency. Data concerning gene expression levels and phytohormone analyses between different instars of N. tenuis-exposed plants and undamaged plants were compared by one tailed t-test (p < 0.05). Acknowledgments The research leading to these results was partially funded by the Spanish Ministry of Economy and Competitiveness (AGL2011-30538-C03 and AGL2014-55616-C3), the Conselleria d’Agricultura, Pesca i Alimentació de la Generalitat Valenciana, and the Italian Ministry of Education, University and Research (PRIN project 2010CXXHJE_004). The authors thank Javier Calvo (KOPPERT BS) for the supply of insects and Servei Central d’Instrumentació Científica of the Universitat Jaume I for technical support. MN received a PhD grant from the University of Catania (Italy). MP-H received a postdoctoral fellowship (Juan de la Cierva program) from the Spanish Ministry of Economy and Competitiveness. Author Contributions Alberto Urbaneja and Meritxell Pérez-Hedo designed the research and, together with Mario Naselli analyzed the data. All authors performed research and wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Plant responses induced by young N. tenuis nymphs. (a) Response of B. tabaci and E. formosa females when they were given the choice between intact tomato plants and punctured tomato plants in a Y-tube olfactometer. Significant differences using χ2 test, p < 0.05 are marked with (*); (b) Expression of the defensive genes ASR1 and PIN2 (target genes induced by the phytohormones abscisic acid (ABA) and jasmonate-isoleucine (JA-Ile), respectively). Data are presented as the mean of the ratio between the concentration of the gene transcripts and that of the constitutive elongation factor 1 (EF1) gene. Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05); (c) ABA and JA-Ile levels in the apical part of tomato plants. Each of the presented results is the mean of the hormone concentration (ng/g) of five independent analyses ± SE (n = 5). Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05). Figure 2 Plant responses induced by mature N. tenuis nymphs. (a) Response of B. tabaci and E. formosa females when were given the choice between intact tomato plants and punctured tomato plants in a Y-tube olfactometer. Significant differences using χ2 test, p < 0.05 are marked with (*); (b) Expression of the defensive genes ASR1 and PIN2 (target genes induced by the phytohormones ABA and JA-Ile, respectively). Data are presented as the mean of the ratio between the concentration of the gene transcripts and that of the constitutive EF1 gene. Significant differences were obtained by comparing punctured plants to intact plants. Results from one tailed t-test are marked with (*) (p < 0.05); (c) ABA and JA-Ile levels in the apical part of tomato plants. Each of the presented results is the mean of the hormone concentration (ng/g) of five independent analyses ± SE (n = 5). Significant differences were obtained by comparing punctured plants to intact plants. Results from a one tailed t-test are marked with (*) (p < 0.05). Figure 3 Plant responses induced by N. tenuis males. (a) Response of B. tabaci and E. formosa females when were given the choice between intact tomato plants and punctured tomato plants in a Y-tube olfactometer. Significant differences using χ2 test, p < 0.05 are marked with (*); (b) Expression of the defensive genes ASR1 and PIN2 (target genes induced by the phytohormones ABA and JA-Ile, respectively). Data are presented as the mean of the ratio between the concentration of the gene transcripts and that of the constitutive EF1 gene. Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05); (c) ABA and JA-Ile levels in the apical part of tomato plants. Each of the presented results is the mean of the hormone concentration (ng/g) of five independent analyses ± SE (n = 5). Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05). Figure 4 Plant responses induced by N. tenuis females. (a) Response of B. tabaci and E. formosa females when were given the choice between intact tomato plants and punctured tomato plants in a Y-tube olfactometer. Significant differences using χ2 test, p < 0.05 are marked with (*); (b) Expression of the defensive genes ASR1 and PIN2 (target genes induced by the phytohormones ABA and JA-Ile, respectively). Data are presented as the mean of the ratio between the concentration of the gene transcripts and that of the constitutive EF1 gene. Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05); (c) ABA and JA-Ile levels in the apical part of tomato plants. Each of the presented results is the mean of the hormone concentration (ng/g) of five independent analyses ± SE (n = 5). Significant differences were obtained by comparing punctured plants to intact plants. Results from a one-tailed t-test are marked with (*) (p < 0.05). Figure 5 Response of B. tabaci and E. formosa females in a Y-tube olfactometer. (a) Comparison between mature-nymph-punctured tomato plants with young-nymph-punctured tomato plants; (b) Comparison between female-punctured tomato plants with male-punctured tomato plants. Significant differences using χ2 test, p < 0.05 are marked with (*). ijms-17-01210-t001_Table 1Table 1 Primers used for quantification of the RNA levels of the genes studied. Gene Forward Primer (5’→3’) Reverse Primer (5’→3’) EF1 5-GATTGGTGGTATTGGAACTGTC-3 5-AGCTTCGTGGTGCATCTC-3 ASR1 5-ACACCACCACCACCACCTGT-3 5-GTGTTTGTGTGCATGTTCTGGA-3 PIN2 5-GAAAATCGTTAATTTATCCCAC-3 5-ACATACAAACTTTCCATCTTTA-3 ==== Refs References 1. Ton J. Flors V. Mauch-Mani B. The multifaceted role of ABA in disease resistance Trends Plant Sci. 2009 14 310 317 10.1016/j.tplants.2009.03.006 19443266 2. Santino A. Taurino M. de Domenico S. Bonsegna S. Poltronieri P. Pastor V. Flors V. Jasmonate signaling in plant development and defense response to multiple (a)biotic stresses Plant Cell Rep. 2013 32 1085 1098 10.1007/s00299-013-1441-2 23584548 3. Dicke M. van Loon J.J.A. Chemical ecology of phytohormones: How plants integrate responses to complex and dynamic environments J. Chem. Ecol. 2014 40 653 656 10.1007/s10886-014-0479-0 25037238 4. Thaler J.S. Farag M.A. Pare P.W. Dicke M. Jasmonate-deficient plants have reduced direct and indirect defences against herbivores Ecol. Lett. 2002 5 764 774 10.1046/j.1461-0248.2002.00388.x 5. De Vos M. van Oosten V.R. van Poecke R.M.P. van Pelt J.A. Pozo M.J. Mueller M.J. Buchala A.J. Metraux J.P. van Loon L.C. Dicke M. Signal signature and transcriptome changes of arabidopsis during pathogen and insect attack Mol. Plant Microbe Interact. 2005 18 923 937 10.1094/MPMI-18-0923 16167763 6. Flors V. Ton J. van Doorn R. Jakab G. Garcia-Agustin P. Mauch-Mani B. Interplay between JA, SA and ABA signalling during basal and induced resistance against Pseudomonas syringae and Alternaria brassicicola Plant J. 2008 54 81 92 10.1111/j.1365-313X.2007.03397.x 18088307 7. Flokova K. Tarkowska D. Miersch O. Strnad M. Wasternack C. Novak O. Uhplc-ms/ms based target profiling of stress-induced phytohormones Phytochemistry 2014 105 147 157 10.1016/j.phytochem.2014.05.015 24947339 8. Bari R. Jones J.D.G. Role of plant hormones in plant defence responses Plant Mol. Biol. 2009 69 473 488 10.1007/s11103-008-9435-0 19083153 9. Kaloshian I. Walling L.L. Hemipterans as plant pathogens Annu. Rev. Phytopathol. 2005 43 491 521 10.1146/annurev.phyto.43.040204.135944 16078893 10. Howe G.A. Jander G. Plant immunity to insect herbivores Annu. Rev. Plant Biol. 2008 59 41 66 10.1146/annurev.arplant.59.032607.092825 18031220 11. Avila C.A. Arevalo-Soliz L.M. Jia L.L. Navarre D.A. Chen Z. Howe G.A. Meng Q.W. Smith J.E. Goggin F.L. Loss of function of fatty acid desaturase7 in tomato enhances basal aphid resistance in a salicylate-dependent manner Plant Physiol. 2012 158 2028 2041 10.1104/pp.111.191262 22291202 12. Alba J.M. Schimmel B.C.J. Glas J.J. Ataide L.M.S. Pappas M.L. Villarroel C.A. Schuurink R.C. Sabelis M.W. Kant M.R. Spider mites suppress tomato defenses downstream of jasmonate and salicylate independently of hormonal crosstalk New Phytol. 2015 205 828 840 10.1111/nph.13075 25297722 13. Nalam V.J. Keeretaweep J. Sarowar S. Shah J. Root-derived oxylipins promote green peach aphid performance on arabidopsis foliage Plant Cell Online 2012 24 1643 1653 10.1105/tpc.111.094110 22474183 14. Heil M. Indirect defence via tritrophic interactions New Phytol. 2008 178 41 61 10.1111/j.1469-8137.2007.02330.x 18086230 15. Coppola V. Coppola M. Rocco M. Digilio M.C. D’Ambrosio C. Renzone G. Martinelli R. Scaloni A. Pennacchio F. Rao R. Transcriptomic and proteomic analysis of a compatible tomato-aphid interaction reveals a predominant salicylic acid-dependent plant response BMC Genom. 2013 14 10.1186/1471-2164-14-515 23895395 16. Ponzio C. Gols R. Pieterse C.M.J. Dicke M. Ecological and phytohormonal aspects of plant volatile emission in response to single and dual infestations with herbivores and phytopathogens Funct. Ecol. 2013 27 587 598 10.1111/1365-2435.12035 17. Halitschke R. Hamilton J.G. Kessler A. Herbivore-specific elicitation of photosynthesis by mirid bug salivary secretions in the wild tobacco Nicotiana attenuata New Phytol. 2011 191 528 535 10.1111/j.1469-8137.2011.03701.x 21443673 18. Kessler A. Baldwin I.T. Herbivore-induced plant vaccination. Part I. The orchestration of plant defenses in nature and their fitness consequences in the wild tobacco Nicotiana attenuata Plant J. 2004 38 639 649 10.1111/j.1365-313X.2004.02076.x 15125770 19. De Puysseleyr V. Hofte M. de Clercq P. Ovipositing Orius laevigatus increase tomato resistance against Frankliniella occidentalis feeding by inducing the wound response Arthropod-Plant Interact. 2011 5 71 80 10.1007/s11829-010-9117-0 20. Pérez-Hedo M. Bouagga S. Jaques J.A. Flors V. Urbaneja A. Tomato plant responses to feeding behavior of three zoophytophagous predators (hemiptera: Miridae) Biol. Control. 2015 86 46 51 10.1016/j.biocontrol.2015.04.006 21. Pérez-Hedo M. Urbaneja-Bernat P. Jaques J.A. Flors V. Urbaneja A. Defensive plant responses induced by Nesidiocoris tenuis (hemiptera: Miridae) on tomato plants J. Pest Sci. 2015 88 543 554 10.1007/s10340-014-0640-0 22. Perdikis D. Lykouressis D.P. Economou L.P. The influence of temperature, photoperiod and plant type on the predation rate of Macrolophus pygmaeus on Myzus persicae BioControl 1999 44 281 289 10.1023/A:1009959325331 23. Sanchez J.A. Gillespie D.R. McGregor R.R. Plant preference in relation to life history traits in the zoophytophagous predator Dicyphus hesperus Entomol. Exp. Appl. 2004 112 7 19 10.1111/j.0013-8703.2004.00174.x 24. Urbaneja A. Tapia G. Stansly P. Influence of host plant and prey availability on developmental time and surviorship of Nesidiocoris tenius (het.: Miridae) Biocontrol Sci. Technol. 2005 15 513 518 10.1080/09583150500088777 25. Calvo F.J. Bolckmans K. Belda J.E. Release rate for a pre-plant application of Nesidiocoris tenuis for Bemisia tabaci control in tomato BioControl 2012 57 809 817 10.1007/s10526-012-9455-1 26. Zappalà L. Biondi A. Alma A. Al-Jboory I.J. Arnò J. Bayram A. Chailleux A. El-Arnaouty A. Gerling D. Guenaoui Y. Natural enemies of the South American moth, Tuta absoluta , in Europe, North Africa and Middle-East, and their potential use in pest control strategies J. Pest Sci. 2013 86 635 647 10.1007/s10340-013-0531-9 27. Zappala L. Siscaro G. Biondi A. Molla O. Gonzalez-Cabrera J. Urbaneja A. Efficacy of sulphur on Tuta absoluta and its side effects on the predator Nesidiocoris tenuis J. Appl. Entomol. 2012 136 401 409 10.1111/j.1439-0418.2011.01662.x 28. Pappas M.L. Steppuhn A. Geuss D. Topalidou N. Zografou A. Sabelis M.W. Broufas G.D. Beyond predation: The zoophytophagous predator Macrolophus pygmaeus induces tomato resistance against spider mites PLoS ONE 2015 10 1210 10.1371/journal.pone.0127251 25974207 29. Calvo J. Blockmans K. Stansly P.A. Urbaneja A. Predation by Nesidiocoris tenuis on Bemisia tabaci and injury to tomato BioControl 2009 54 237 246 10.1007/s10526-008-9164-y 30. Fatouros N.E. Lucas-Barbosa D. Weldegergis B.T. Pashalidou F.G. van Loon J.J.A. Dicke M. Harvey J.A. Gols R. Huigens M.E. Plant volatiles induced by herbivore egg deposition affect insects of different trophic levels PLoS ONE 2012 7 1210 10.1371/journal.pone.0043607 22912893 31. Heil M. Silva Bueno J.C. Within-plant signaling by volatiles leads to induction and priming of an indirect plant defense in nature Proc. Natl. Acad. Sci. USA 2007 104 5467 5472 10.1073/pnas.0610266104 17360371 32. Colazza S. McElfresh J.S. Millar J.G. Identification of volatile synomones, induced by Nezara viridula feeding and oviposition on bean spp., that attract the egg parasitoid trissolcus basalis J. Chem. Ecol. 2004 30 945 964 10.1023/B:JOEC.0000028460.70584.d1 15274441 33. Kahn T.L. Fender S.E. Bray E.A. Oconnell M.A. Characterization of expression of drought and abscisic acid-regulated tomato genes in the drought-resistant species Lycopersicon pennellii Plant Physiol. 1993 103 597 605 12231965 34. Maskin L. Gudesblat G.E. Moreno J.E. Carrari F.O. Frankel N. Sambade A. Rossi M. Iusem N.D. Differential expression of the members of the ASR gene family in tomato (Lycopersicon esculentum ) Plant Sci. 2001 161 739 746 10.1016/S0168-9452(01)00464-2 35. Ramirez V. Coego A. Lopez A. Agorio A. Flors V. Vera P. Drought tolerance in arabidopsis is controlled by the ocp3 disease resistance regulator Plant J. 2009 58 578 591 10.1111/j.1365-313X.2009.03804.x 19175769 36. Kessler A. Baldwin I.T. Plant responses to insect herbivory: The emerging molecular analysis Annu. Rev. Plant Biol. 2002 53 299 328 10.1146/annurev.arplant.53.100301.135207 12221978 37. Abe H. Shimoda T. Ohnishi J. Kugimiya S. Narusaka M. Seo S. Narusaka Y. Tsuda S. Kobayashi M. Jasmonate-dependent plant defense restricts thrips performance and preference BMC Plant Biol. 2009 9 10.1186/1471-2229-9-97 19635132 38. Agut B. Gamir J. Jacas J.A. Hurtado M. Flors V. Different metabolic and genetic responses in citrus may explain relative susceptibility to Tetranychus urticae Pest Manag. Sci. 2014 70 1728 1741 10.1002/ps.3718 24375985 39. Agut B. Gamir J. Jaques J.A. Flors V. Tetranychus urticae -triggered responses promote genotype-dependent conspecific repellence or attractiveness in citrus New Phytol. 2015 207 790 804 10.1111/nph.13357 25771705 40. Dicke M. Behavioural and community ecology of plants that cry for help Plant Cell Environ. 2009 32 654 665 10.1111/j.1365-3040.2008.01913.x 19021885 41. Pérez-Hedo M. Urbaneja A. The zoophytophagous predator Nesidiocoris tenuis : A successful but controversial biocontrol agent in tomato crops Advances in Insect Control and Resistance Management Horowitz A.R. Ishaaya I. Springer International Publishing Dordrecht, The Netherland 2016 in press 42. Pappas M.L. Steppuhn A. Broufas G.D. The role of phytophagy by predators in shaping plant interactions with their pests Commun. Integr. Biol. 2016 9 e1145320 10.1080/19420889.2016.1145320 27195065 43. Biondi A. Zappala L. di Mauro A. Garzia G.T. Russo A. Desneux N. Siscaro G. Can alternative host plant and prey affect phytophagy and biological control by the zoophytophagous mirid Nesidiocoris tenuis ? BioControl 2016 61 79 90 10.1007/s10526-015-9700-5 44. Pérez-Hedo M. Urbaneja A. Prospects for predatory mirid bugs as biocontrol agents of aphids in sweet peppers J. Pest Sci. 2015 88 65 73 10.1007/s10340-014-0587-1 45. Forcat S. Bennett M.H. Mansfield J.W. Grant M.R. A rapid and robust method for simultaneously measuring changes in the phytohormones ABA, JA and SA in plants following biotic and abiotic stress Plant Methods 2008 4 10.1186/1746-4811-4-16 18590529

PMC005xxxxxx/PMC5000609.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081211ijms-17-01211ArticleGlutathione Transferases Superfamily: Cold-Inducible Expression of Distinct GST Genes in Brassica oleracea Vijayakumar Harshavardhanan 1Thamilarasan Senthil Kumar 1Shanmugam Ashokraj 1Natarajan Sathishkumar 1Jung Hee-Jeong 1Park Jong-In 1Kim HyeRan 2Chung Mi-Young 3Nou Ill-Sup 1*Zhu Jianhua Academic Editor1 Department of Horticulture, Sunchon National University, 255, Jungang-ro, Suncheon 57922, Korea; [email protected] (H.V.); [email protected] (S.K.T.); [email protected] (A.S.); [email protected] (S.N.); [email protected] (H.-J.J.); [email protected] (J.-I.P.)2 Plant Systems Engineering Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), 125 Gwahangno, Daejeon 34141, Korea; [email protected] Department of Agricultural Education, Sunchon National University, 255, Jungang-ro, Suncheon 57922, Korea; [email protected]* Correspondence: [email protected]; Tel.: +82-61-750-3240; Fax: +82-61-750-538927 7 2016 8 2016 17 8 121108 6 2016 15 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Plants, as sessile organisms, can suffer serious growth and developmental consequences under cold stress conditions. Glutathione transferases (GSTs, EC 2.5.1.18) are ubiquitous and multifunctional conjugating proteins, which play a major role in stress responses by preventing oxidative damage by reactive oxygen species (ROS). Currently, understanding of their function(s) during different biochemical and signaling pathways under cold stress condition remain unclear. In this study, using combined computational strategy, we identified 65 Brassica oleracea glutathione transferases (BoGST) and characterized them based on evolutionary analysis into 11 classes. Inter-species and intra-species duplication was evident between BoGSTs and Arabidopsis GSTs. Based on localization analyses, we propose possible pathways in which GST genes are involved during cold stress. Further, expression analysis of the predicted putative functions for GST genes were investigated in two cold contrasting genotypes (cold tolerance and susceptible) under cold condition, most of these genes were highly expressed at 6 h and 1 h in the cold tolerant (CT) and cold susceptible (CS) lines, respectively. Overall, BoGSTU19, BoGSTU24, BoGSTF10 are candidate genes highly expressed in B. oleracea. Further investigation of GST superfamily in B. oleracea will aid in understanding complex mechanism underlying cold tolerance in plants. coldglutathione transferase (GST)Brassica oleraceacontrasting genotypestolerance ==== Body 1. Introduction Cold stress is detrimental to plant growth and development, thus affecting agricultural productivity worldwide. During low temperature conditions, cold-tolerant plants activate a tolerance mechanism called cold acclimation, which prevents ice formation within the vegetative cells. However, many agricultural crops lack this cold acclimation mechanism. In general, plants achieve tolerance to cold stress by modifying biochemical and physiological factors at the cellular and molecular level. Cold acclimation involves expression of a set of so-called cold-regulated (COR) genes, induction of which is mediated by effector genes and various transcription factors [1]. In addition to their basic role of protecting cells during cold stress conditions, COR genes regulate signal transduction of defense-related and secondary metabolite genes [2]. Cold stress responses involve cis-acting elements such as ABREs (abscisic acid response element), DREs (c-repeat/dehydration responsive elements), and LTREs (low-temperature-responsive elements), which are regulated by abscisic acid (ABA)-dependent or -independent pathways [3]. ABA and reactive oxygen species (ROS) play vital roles as second messengers via Ca2+ signaling, inducing COR genes such as ERD6 [1], LOS5, FRO1 [4], AP2/ERF [5,6], bZIP [7] and CBFs [8]. Notably, there is also an increase in secondary metabolite accumulation in the plant cell during cold stress. Cold conditions induce increased accumulation of sugar [9], polyamines [10], anthocyanins [11], and glucosinolates [12]. Increased activity of enzymatic (superoxide dismutase (SOD), GST, glutathione reductase (GR), and glutathione peroxidases (GPX)) and non-enzymatic antioxidants (GSH, tripeptide thiols, and vitamins) are also evident in cold-stressed tissues [13]. Additionally, post-translational modifications, such as sumoylation (conjugation of small ubiquitin-like modifier (SUMO) proteins to target proteins) carried out by SIZ1 gene products during stress, play roles in cold tolerance [14]. SIZ1 regulates the expression of CBF/DREB genes by inhibiting MYB genes, thereby enhancing low temperature tolerance [15]. Plant glutathione transferases (GSTs, EC 2.5.1.18) represent a complex superfamily of proteins with multiple roles, involving site-specific (G-site) tripeptide conjugation (glutathione, GSH, and γ-Glu-Cys-Gly) leading to reduction of a wide range of electrophilic and hydrophobic substrates and redox buffering. GSTs play a vital role in glucosinolate biosynthesis and metabolism [16], transport of anthocyanin [17,18], and xenobiotic metabolism [19]. Thus far, evolutionary analysis of GSTs found in eukaryote photosynthetic organisms elucidated 14 major classes in this superfamily [20,21]. Currently, in plants, GSTs are grouped into tau, phi, theta, zeta, lambda, DHAR (dehydroascorbate reductase), EF1G (elongation factor 1 γ), TCHQD (tetrachlorohydroquinone dehalogenase), GHR (glutathionyl hydroquinone reductase), GST2N (glutathione transferases with two thioredoxin), and mPGES2 (microsomal prostaglandin e synthase type 2) classes based on sequence similarity, immunological reactivity, kinetic properties, and structural conformation. Among these classes, tau, phi, DHAR, and lambda GSTs are plant-specific proteins. GST proteins in all phyla typically contain GST N-domain (thioredoxin-like domain, conjugation of GSH moiety, G-site) and GST C-domain (bind to hydrophobic substrates, H-site) except GST2N class, which contains two repeated GST N-domain and lack terminal GST C-domain. A serine residue in the N-terminal region acts as the active site of GST proteins [22], except in those of the lambda, DHAR, GHR, and mPGES2 classes wherein the serine is replaced with a cysteine residue in their active site [21,23]. GSTs were originally considered to serve mainly for xenobiotic detoxification until the discovery of their functions in preventing oxidative damage to cells due to biotic and abiotic stresses such as pathogen infection, ROS, UV radiation, and heavy metal toxicity [24,25,26]. Dixon et al. (2009) reported that tau and phi class GSTs have a wide range of substrate specificity in Arabidopsis, which is correlated with high tolerance against cold, dehydration, UV, oxidative stress, salt, and heavy metals [27]. In Arabidopsis, phi class glutathione transferases (GSTF2) has high affinity to heterocyclic compounds and aids in intracellular transport of defense-related genes [28]. DHAR maintains a reduced ascorbic acid pool within the cell, acting as an antioxidant protein [29], and shows up-regulation during abiotic stresses such as light and drought [21]. Further, lambda GST genes show increased expression under heavy metal stress compared to other GST genes [30]. Among the GST superfamily, few tau, phi and theta GSTs have been known for high glutathione peroxidase (GPOX) activity [22,31], and Euphorbia esula GSTT1 exhibits up-regulation during drought stress and treatment with xenobiotics [32]. Zeta GSTs are known for their roles in tyrosine catabolism [33] and aid during the germination stage under chilling and salt stress in Euphorbia esula [32]. Little information is interpreted related to GHR and mPGES2 classes roles during environmental conditions in plants, however microarray data of Arabidopsis show differential regulation during various abiotic stress [21]. Brassica oleracea includes many important crops in the Brassicaceae family, namely cabbage (B. oleracea var. capitata), broccoli (B. oleracea var. italica), cauliflower (B. oleracea var. botrytis), kale (B. oleracea var. acephala), Brussels sprouts (B. oleracea var. gemmifera), collard greens (B. oleracea var. acephala), kohlrabi (B. oleracea var. gongylodes), and Chinese broccoli (B. oleracea var. alboglabra). Most Brassica species contain high levels of proteins and diversified secondary metabolites [16] that have unique phytochemical characteristics including anti-carcinogenic properties in humans [34]. Cabbage and broccoli are the major agricultural crops of B. oleracea. In this study, a combined computational strategy comprising HMM (hidden Markov model) profile scan coupled with BLAST analysis of the sequenced B. oleracea genome data from Brassica databases was employed to identify GST genes, and further raw data were processed and screened for candidate genes based on coding and protein sequences. We identified and characterized 65 BoGST genes. Moreover, genome-wide expression analysis was performed in two contrasting inbred lines of B. oleracea, Bo106 (cold tolerant (CT)) and Bo107 (cold susceptible (CS)), during cold treatment (4 °C) to reveal possible candidate gene involved during cold tolerance. 2. Results and Discussion 2.1. Identification of GST Genes in B. oleracea The release of the draft genome of B. oleracea to public databases (Brad, Bolbase and EnsemblPlants) has made it possible to analyze gene families based on annotation from the Arabidopsis genome. Here, we identified a comprehensive set of GST genes from the B. oleracea genome using combined computational analysis comprising HMM profiling and BLAST analyses. A series of systematic analyses was performed to assemble the final set of protein sequence. Firstly, Arabidopsis GST proteins (55 proteins) were collected using locus IDs published in Dixon and Edwards 2010. Based on the 3D structure of GST proteins in Arabidopsis, Armstrong [35] reported that GST N- and C-domains are the basic domain constituents of GST proteins. To spot the functional domains of those sequences, domain analysis was carried out and domain sequences of those proteins were used as input to build a GST-specific HMM profile. Secondly, two peptide datasets, from Bolbase and EnsemblPlants databases, were used as queries against the GST-specific HMM profile. We obtained 107 and 1109 proteins, respectively, from the datasets. From the HMMER results, multiple hits of the same genes were inferred, signifying that the assembled draft genome of B. oleracea contains multiple copies or fractional copies of the same gene. Manually, we identified 89 putative annotated GST genes from Brassica databases. We evaluated all 1305 identified proteins (which includes redundancy) to identify GST genes based on domain occurrence. In total, 65 non-redundant GST proteins were identified based on the HMM results and annotation searches showing they contained only GST-specific domains. Finally, we verified the annotations of these 65 protein sequences using local BLASTP against the Arabidopsis genome and NCBI database using customized E values (1 × 10−3, 1 × 10−10, 1 × 10−30 and 1 × 10−50). The results from each analysis were in agreement, showing paralogous and orthologous genes among B. oleracea (37 proteins), B. napus (21 proteins), B. rapa (6 proteins), and Zea mays (1 proteins) (Table S1). Also, results obtained from GST-specific HMM profile retrieved form Pfam database were similar to that of Arabidopsis GST-specific HMM profile (data not shown). GSTs are studied in a range of plant species such as Arabidopsis [22], barley [36], soybean [37], maize [37], rice [38], tomato [39], and Populus trichocarpa [40], and the total numbers of GST genes in selected crops are summarized in Table 1. Results of BLAST searches of BoGST proteins against the Arabidopsis genome were similar to those for orthologous genes as annotated in the Brad database (data not shown). Our domain analysis results confirmed the presence of both GST N- (thioredoxin-like) and GST C-domains (hydrophobic or electrophilic binding) in all GST proteins except two proteins (Bol010024 and Bol015341) which contains two repeats of thioredoxin domain and lack terminal C-domain (Table S2). From our prediction results, we also found other specific and multi domains such as EF1G (elongation factor 1 γ, pfam00647), GSTA (glutathione S-transferase multi-complex domain, COG0625), MaiA (maleylacetoacetate isomerase, TIGR01262) and ECM4 (glutathionyl-hydroquinone reductase, COG0435). Similarly, domain analysis of AtGST and BrGST proteins (uncharacterized) using the SMART and conserved domains database (CDD) database generated similar results (data not shown). The thioredoxin-like domain comprises βαβαββα motifs that make three layers of β-sheet enclosed with two α-helixes on either side (α/β/α) [41], and there is a small peptide sequence between the N-terminus and C-terminus called the linker region (8–15 aa) that connects these two termini [42]. Mohsenzadeh et al. (2011) reported that the C-terminal region of GST confers the substrate selectivity between the GST classes [43]. Secondary structure prediction using the garnier program employing the GOR method in the EMBOSS server showed a higher percentage of α-helix than β-sheets (Table S3). Overall, the 65 BoGST proteins have open reading frames ranging from 570–1248 bp, with predicted protein lengths between 189 and 415 amino acids, and different numbers of exons in their gene structure. The predicted molecular weights, pIs, and stability indexes ranged between 21.37–46.57 (kDa), 4.96–9.5 and 27.33–60.31, respectively. Based on their stability index values, 41 and 24 proteins were characterized as stable and unstable proteins, respectively (Table S3). 2.2. Classification and Sequence Characterization of BoGST Genes To distinguish between mammalian GSTs and plant GSTs, a classification system was first proposed by Droog (1997) [45], and later modified [46] based on gene association within the genome. Finally, a unified nomenclature system for plant GSTs was adopted [22]. Plant GSTs are classified into eight major classes, namely tau (U), phi (F), theta (T), zeta (Z), lambda (L), DHAR, TCHQD, and EF1G. To investigate the evolutionary relationships of BoGST proteins, known GST proteins from GST superfamilies characterized in monocots (rice, maize, wheat, and barley) and dicots (Arabidopsis, soybean and wild soybean) were collected (Tables S4 and S5). Here, we classified the BoGSTs based on their evolutionary relationships with GSTs from other species. The two largest and most abundant classes, tau and phi, had 28 and 14 closely grouped BoGSTs, respectively, whereas three BoGSTs were separately placed in the lambda and EF1G classes and two BoGSTs in theta and zeta classes. Among the rest of the BoGSTs, four and one were clustered in DHAR, and TCHQD classes, respectively (Figure 1, Figure S1). An additional eight proteins (Bol001864, Bol004474, Bol012366, Bol012372, Bol024359, Bol010024, Bol015341, and Bo035968) were not placed into any of the above mentioned plant classes. Based on BLAST results and domain analysis, five proteins (Bol001864, Bol004474, Bol012366, Bol012372, and Bol024359) were annotated as glutathionyl hydroquinone reductase (GHR) and one protein (Bol035968) as microsomal prostaglandin E synthase type 2 (mPGES-2) with both GST N- and GST C-domain, so we named these classes of proteins as BoGHR and BomPGES2, respectively (data not shown). The remaining two proteins (Bol010024 and Bol015431) consist two repeats of GST N-domain and lack C-terminal, hence these two proteins were classified as GST2N class. A summary of all GST genes in B. oleracea is given in Table 2. In plants, various classes of GST genes respond to cold stress in different species. Specifically, tau GST members in Arabidopsis [47], phi GSTs in Solanum species [48], theta GSTs in Euphorbia [32], and zeta GSTs as well as the TCHQD gene in O. sativa [49,50] show cold-responsive expression. Members of a cold-induced antioxidant enzyme family, DHAR, show activity specifically for rapid H2O2 scavenging in the chloroplast through the water-water cycle to remove H2O2 from the cell [51]. Ahsan et al. (2008) reported that GST omega class genes are closely related to human glutathione S-transferase omega (GSTO) genes and are involved in heavy metal detoxification in rice roots [52]. However, the roles of other GST classes in plants needs to be elucidated during cold stress and other abiotic stresses. Mapping of enzymes or proteins in biochemical pathways can help in understanding their biological function. AtGSTZ1-1 shows maleylacetone isomerase (MAI) activity with a GSH-dependent reaction, involved in tyrosine metabolism in Arabidopsis [33]. Cytochrome P450 (CYP)-mediated detoxification of drugs and xenobiotics requires phase II detoxification enzymes, namely glutathione transferases, for the final degradation process [53]. In addition to these pathways, GSTs are known to play roles in secondary metabolite [54] and auxin metabolism [55]. We mapped the BoGST proteins in the kyoto encyclopedia of genes and genomes (KEGG) database using Blast2Go software [56] to identify their possible roles in the plant. Three major pathways were predicted for BoGST proteins, of which 54 were assigned to glutathione metabolism (map00480), drug metabolism (map00982) and xenobiotics metabolism via cytochrome P450 (map00980). Besides these pathways, BoGST proteins were related to phenylpropanoid biosynthesis (map00940, 18 proteins), ascorbate and aldarate metabolism (map00053, four proteins), arachidonic acid metabolism (map00590, three proteins), tyrosine metabolism (map00350, two proteins), pyruvate metabolism (map00620, four proteins), styrene degradation (map00643, two proteins), and aminoacyl-tRNA biosynthesis (map00970, one protein). Pathway distributions among the different classes of BoGSTs are detailed in Table 3. Overall, the previous pathway-related studies on GST genes in plants have reported similar predicted pathways as those for these BoGST genes in B. oleracea. Among various GST classes, intron–exon organization is well categorized. The total numbers of exons within the different classes are included Table 2, and individual exons for the genes are listed in Table S3, Figure S2a,b. Members of the largest class, the tau class, contain two exons in their gene structure except BoGSTU11 (1 exon) (Table S3). The presence of a single exon in this group may possibly be due to selection pressure or genomic duplication. Tau class GSTs were first identified as being encoded by an auxin-inducible gene and are involved in responses to wide range of stresses such as wounding, pathogen infection, herbicides, and extreme temperature. BoGST members of the phi, theta, zeta, TCHQD and mPGES2 classes contained conserved gene orientation with 3, 7, 9, 2, and 6 exons, respectively, in their genomic structure. By contrast, the lambda, DHAR, EF1G, GHR and GST2N gene structures varied within classes. The results of our predicted intron–exon analysis were similar to those for rice GST genes [38]. However, we observed slight deviation when compared to Arabidopsis intron–exon GST results [42], which may be due to the genome size and evolution within the Brassicaceae family. Zeta GST genes from human, carnation, and C. elegans share a common gene structure with three exons, demonstrating that the zeta class gene structure evolved before these lineages diverged, and hence represents the ancestral class [57]. During cold stress, cis-acting factors are involved in activation and overexpression of zeta GST proteins in transgenic rice line [58]. There is no previous report of LTRE (low-temperature-responsive element) regulatory elements in GST superfamily members, although there are other elements such as ABREs (ABA-responsive elements) in wheat GST genes [59], an ERE (ethylene-responsive element) in carnation GST1 [60] and AuxREs (auxin-responsive elements) in tobacco GST [61]. To investigate cis-elements in BoGST genes, the sequences 1000 bp upstream from the 5′-end of the coding region were analyzed for conserved DRE and LTRE motifs. Nine BoGST genes were found to have both putative elements, whereas 10 and 21 genes were predicted to contain either DRE or LTRE cis-acting elements, respectively (Table S6, Figure S3). BoGSTU1, BoGSTU11, BoGSTU16, BoGSTU19, BoGSTL3 and BoGHR3 contained more putative LTRE elements in the sense strand of the promoter region, whereas only BoDHAR1, BoGST2N-1 and BomPGES2-1 had LTRE elements in the antisense region of the promoter. Based on computational analysis revealed several regions of DRE and LTRE elements are present in various promoter regions of BoGST genes, which might be involved during various abiotic stresses. Moons et al (2003) reported that OsGSTU4 and OsGSTU3 proteins each contain one potential N-glycosylation site similar to that of sorghum [62], and that this site is necessary for the activity of GSTs. Prediction using NetNGlyc (Asn-Xaa-Ser/Thr), showed that 33 BoGSTs have a potential site for N-glycosylation (Table S3). In addition, protein motif searching identified 10 motifs in BoGST proteins, and motif conservation was higher within the classes than between them (Figure S4a,b). Motif 1, present in 57 BoGSTs, contained the serine active site residue at the 18th position (Figure S5). In rice, out of 79 GSTs, 72 were found to contain a serine residue as their active site in their own motif region [50]. Determination of the sub-cellular localization of a protein is an important step toward learning the function of the protein. Little information regarding GST localization is available in the literature, and due to the lack of targeting peptide sequences in their N-terminal regions [27], most GST proteins are thought to be localized in the cytoplasm. Similarly, we found that most BoGST proteins were predicted to be cytoplasmic by Protcomp (61.5%, 40 proteins), tabulated in Table S3. Understanding the sub-cellular localization of GST proteins will aid in finding other possible proteins associated with GST, as well as determining their associated reactions with other proteins and substrates/ligands. 2.3. Chromosomal Distribution, Gene Duplication and Syntenic Regions Analysis of gene distribution in the chromosome showed that BoGST genes were spread throughout the genome. The most genes were found in scaffold regions (Cun, 11 genes) and C06 (nine genes), whereas C01 had the fewest GST genes (one gene). Among the tau GSTs, 13 were located in C04 and C06, whereas as a total of eight phi GSTs were in C03 and C09 (Figure 2). Clustering of GST genes was present only in the largest gene classes tau and phi with six clusters (highest) and three clusters, respectively, whereas we observed one cluster (lowest) in the scaffold regions with lambda GSTs. Similar gene distribution and clustering was reported in Arabidopsis [22]. To examine the sequence similarity of BoGST proteins within and between the classes, multiple sequence alignment was performed using DiAlign [63]. As expected, the similarity within the classes was higher than that between classes. GST proteins such as BoGSTU2, BoGSTU6, BoGSTU10, BoGSTU11, BoGSTU23, BoGSTU28, BoGSTF7, BoGSTL3, and BoGHR1 showed less than 60% similarity within their class, compared to not more than 40% between the classes (Tables S7–S10). These results raise the possibility of diversified roles, such as in secondary metabolite metabolism and substrate specificity, within members of a class due to relatively high levels of divergence between the proteins. Thirteen pairs of phi BoGST proteins shared more than 82% similarity, and three pairs of tau BoGSTs had similarity higher than 80%, which indicate high rate of gene duplication within these classes. Such gene duplication might be due to evolutionary pressure imposed on these GST gene family or whole genome triplication event in Brassicaceae family. As mentioned above, zeta GSTs represent the oldest known class and share high similarity among members from different kingdoms, e.g., humans and carnation share 38% identity, whereas humans and C. elegans have 49% similarity. These levels of similarity indicate that significant stretches of sequences have been conserved over a long evolutionary period, thus suggesting common biological function in all organisms [64]. The Brassica and Arabidopsis lineages diverged 20–24 million years ago [65]. To pinpoint the conserved blocks related to GSTs within the Arabidopsis and B. oleracea genomes, syntenic mapping was examined (Figure 3). Within the B. oleracea genome, C06 had more syntenic regions (18.8%) than other chromosomes, whereas C01, with the least (0.1%) GST genes showed fewer syntenic regions (Tables S11 and S12). Syntenic regions of BoGST genes shared 48.3% similarity (highest) in observed regions with Arabidopsis Chr1 (Chromosome 1) and 5.3% (lowest) with Chr4 (chromosome 4). In Arabidopsis, Chr1 contains 26 of 55 AtGSTs, which is the reason for high sequence conservation of BoGST genes with Chr1, is suggestive of genome evolution of duplicated genes from the same chromosome. Surprisingly, in Arabidopsis, Chr5 contains more GST genes than Chr4, but there were fewer syntenic regions than in Chr4 [22]. Genome duplication and gene losses during evolution of Arabidopsis to Brassica are likely responsible for this distribution of synteny between the two species. 2.4. Gene-Specific and Organ-Specific Expression Analysis in Non-Treated Samples Gene-specific primer pairs for 65 BoGSTs (Table S13) were used for reverse transcription-PCR (RT-PCR) analysis of class-wise expression patterns of BoGSTs in different organs (root, leaf, flower buds) of healthy non-treated B. oleracea (line Bo106). From semi-quantitative RT-PCR analysis among 65 genes, 43 genes were found to express in any one of the investigated tissues (Figure 4). As stated earlier, similarity within the classes was very high, which made it impossible to analyze gene-specific expression of the remaining genes. Of all the expressed genes in organ tissues, 33 genes were expressed in all organs ubiquitously. Notably, BoGSTU6, BoGSTU14 transcripts expressed exclusively in the roots, thus suggesting it is a root-specific gene that could be possibly involved in root associated biological reactions. In rice, tau class OsGSTU3 and OsGSTU4 genes are expressed only in root organs and also show high activity against various abiotic stresses [66]. Among the rest, members of all GST classes were expressed in all organs expect EF1G class, BoGSTF10, BoDHAR3, which were not expressed in root samples. Specifically, BoGSTF1 was expressed in leaf and flower bud, consistent with its reported function in aliphatic GSL biosynthesis in B. oleracea leaves [16]. Overall, the organ-specific expression patterns of BoGSTs were in parallel with those found in organ-specific microarray expression analysis of AtGSTs [22] and OsGSTs [50]. These results shows that GST genes are predominantly expressed in all organs suggest that GSTs may have regulatory functions within developing plant cells. 2.5. Differential Expression Pattern under Cold Stress GST genes are among cold-inducible genes in Arabidopsis [47], rice [58], Solanum sp. [48], Euphorbia esula [32], and S. bicolor [67]. Here, cold transcriptome data of the two contrasting lines (cold tolerant (CT) and cold susceptible (CS)) were used for BLAST searches coupled with data mining for our BoGST sequences and revealed a total of 33 unigenes differentially expressed during cold stress (Figure 5a,b; Table S14). For further expression profiling analysis based on Arabidopsis orthologous genes, cold microarray profiles for aerial parts were downloaded using the AtGenExpress visualization tool to investigate cold stress responses of GST genes. A total of 48 orthologous AtGST gene expression profiles were obtained for 65 corresponding BoGST genes. Relative gene expression using cluster analysis revealed that different class of GST genes showed up (red) and down (green) regulation at different time points of cold stress (Figure 5c). Based on our data and previous reports on GST proteins during stress conditions, a simple schematic model for GST roles in cold stress is proposed in Figure 6; details for a few genes in different pathways during cold stress are tabulated in Table 4. During cold stress, plant cells sense cold via changes in membrane fluidity and protein conformations, which elicit primary signals, such as ABA, Ca2+ (calcium ions), and NO− (nitric oxide) [68], and secondary signals like ROS, stomatal closure, and light perception. The best-known cold response pathway in plants is mediated by transcription factors that bind ABRE and DRE elements in promoters, and further induce COR genes (COR15, COR47, RD22, and RD29A). In addition, AP2/ERF (apetala2/ethylene response factor) elements are also involved in induction of cold-related genes [5,8,69]. During ROS formation (oxidative burst), GST and GPX proteins are known to be highly induced, and in turn detoxify lipid peroxides, DNA degradation products, and ROS. Further, increased cellular levels of ROS, which are second messengers during cold stress conditions, lead to PCD (programmed cell death) [54]. Besides these functions, ROS are also known to induce MAPKK (mitogen-activated protein kinase kinase) proteins, which further induce defense-related genes via MAPKs, and a conjugation process involving GSH by GST enzyme was further localized in vacuole by transporters (ATP-binding cassette (ABC) transporter cascade) for degradation. In the chloroplast, two types of ROS degradation take place: (i) via the ascorbate-glutathione system mediated by DHAR class members [70]; and (ii) conjugation of GSH aided by GST proteins [71]. ROS degradation in chloroplasts is strongly affected by light perception signal transduction by photosystem, Cys, Met and GSH biosynthesis [70]. Targeting of GST proteins to specific organelles within a plant cell during stress conditions reveals likely roles for those proteins. Based on the expression profile and localization of GST proteins, with especially tau and theta class members targeted inside the nucleus and showing differential gene expression (DGE) in cluster analysis (Figure 5c), some members of the tau and theta class are candidates to play roles as transcription factors [22,55]. By contrast, DHAR protein activity is high in thylakoid membranes of the chloroplast [27] and expression analysis during cold stress showed short-term and long-term induction of genes within the class. In general, cold tolerance of plants positively correlates with levels of anthocyanins, which probably protect chlorophyll from over-excitement during cold stress [13]. However in the phi class, GSTF12 is regulated by transcription factors R (bHLH family) and C1 (R2R3-MYB protein family), and is involved in translocation of anthocyanins to vacuoles via ABC transporters, which suggests a possible role in membrane stabilization and ROS scavenging during cold stress [72]. Additionally, TCHQD members are plasma membrane proteins whose function is uncharacterized [27], but interact with DHAR, tau, theta and lambda classes in protein–protein interaction network predicted using STRING database (data not shown), suggesting a possible role in signal perception. Further to validate the expression of BoGSTs in response to cold, we used leaf samples of two contrasting B. oleracea lines, CS and CT, for RT-PCR and qRT-PCR experiments. RT-PCR results for all GST genes expressed in the organ were analyzed in cold-treated samples, and all genes were successfully amplified except BoGSTU14 in leaves of both lines, and BoGSTU5 in leaf samples of the CS line (Figure S6). Considering previously published results, in silico localization analysis and deduced pathways for GST genes during cold stress, we selected BoGST genes for further analysis that possibly function as transcription factors (eight genes), H2O2 reduction in chloroplast (three genes), and signal perception in the plasma membrane (one gene), as well as co-regulated genes in anthocyanin sequestration to the vacuole (two genes). For these genes, qRT-PCR analysis revealed that transcript levels differed in leaf samples along the time course of cold stress (Figure 7). Of the eight genes analyzed based on possible roles at TFs in the GST superfamily (Figure 7a), BoGSTU2 and BoGSTT2 showed up-regulation in both inbred lines, showing the highest gene expression at 6 h of cold treatment, followed by down-regulation. Their similar expression patterns in both lines suggest that these two genes are not affected by any varying internal factors. In the CT inbred line, BoGSTU1, BoGSTU3, BoGSTU6, BoGSTU19 and BoGSTU24 were positively differentially regulated in response to cold stress; however, high transcript levels of these genes occurred at different times, followed by gradual decreases over the time course. Notably, BoGSTU3 showed continuous upregulation except at 24 h during cold stress. Moreover, BoGSTT1 showed no significant expression during stress time. In the CS inbred line, BoGSTU1, BoGSTU19, BolGSTU24, and BoGSTT1 showed down-regulation along the time period, whereas no significant expression change was observed in BoGSTU3, BoGSTU6 and BoGSTU19 genes during cold treatment. For theta class GST proteins, a myb-like transcription factor regulates gene expression during oxidative stress [22]. Similarly, in drought and cold, GSTT1 in Euphorbia esula showed higher expression than under control conditions [32]. By contrast, parA (tau class) in tobacco possibly functions in transcription regulation in addition to its GST activity [27,55]. In Arabidopsis, AtGSTU7 showed changes in expression within 3 h of different stresses [47]; additionally, AtGSTU17 acts as negative regulator in drought- and salt-mediated signal transduction [73]. Overall, among putative TFs in the GST superfamily, genes positively expressed were induced early in cold stress, and thus are likely to be positively involved in regulation of cold-related genes, secondary metabolite biosynthesis and metabolism and ROS reduction. Investigation of TCHQD revealed contrasting mRNA transcript levels in CT and CS lines, showing down-regulation and up-regulation respectively, consistent with this gene acting as a negative regulator during cold stress (Figure 7b). However, in rice, OsTCHQD1 accumulated after 3 h of cold treatment, significantly lower when compared to drought and salt stress [50], and it is also involved in reduction of pesticides [27]. In the GST superfamily, few genes encode antioxidant enzymes that reduce ROS produced due to cold stress in chloroplasts. Evaluating transcripts of antioxidant enzymes in B. oleracea revealed that BoDHAR2 was up-regulated in both inbred lines, showing high accumulation at 24 h of cold stress. This activation of DHAR genes occurs after high accumulation of ROS levels inside chloroplasts and these genes are also induced after prolonged cold stress (Figure 7c). In the CT line, BoDHAR3 showed up-regulation with a peak at 6 h, whereas BoGSTL2 showed no significant expression during the stress period. Different expression patterns were observed for BoGSTL2 (down-regulation from 3 h) and BoDHAR3 (no response) in the CS line. In wheat, DHAR shows up-regulation in contrasting seasonal varieties during prolonged cold [51], and the level of transcripts was also elevated in transgenic Arabidopsis lines compared to wild type under salt stress [74]. In addition, lambda GST activity is increased in response to heavy metals in Arabidopsis [30], although there are no reports on other abiotic stresses. Anthocyanin biosynthesis-related genes are induced during cold stress, conferring tolerance to the plants. Anthocyanin biosynthesis-related genes are highly induced during cold stress in kale, a B. oleracea member [75], and GST genes are also highly induced regulated during high light (HL) stress along with anthocyanin biosynthesis genes in Arabidopsis [76]. In Arabidopsis, especially GSTF12 promotes the transport of anthocyanins to vacuoles for increased tolerance during cold stress. Examination of two orthologous genes of AtGSTF12 in B. oleracea revealed that in the CT line, BoGSTF9 showed static expression and BoGSTF10 was significantly up-regulated over time during cold stress, whereas the BoGSTF10 failed to accumulate in the CS line, which might lead to the cold-susceptibility of the plant (Figure 7d). Two-way ANOVA statistical analysis support that expression levels of genes were significant (i.e., significant at 0.1% level of significance) at different time point within and between genotypes (Table S15). In sum, BoGSTU24 exhibited strongly contrasting expression patterns in CT and CS line, showing up-regulation and down-regulation, respectively, during cold stress, whereas BoGSTU19 showed opposite pattern. In CT lines, high accumulation of GST transcripts was observed at or after 6 h of cold stress, although in the CS line, the levels were highest at or after 1 h of cold stress. These findings suggest that there are differences in GST gene induction between the two inbred lines, which may be due to induced formation of ROS during cold stress. However, the phenotype and genotype of the two inbred line also affect the transcript levels of GST genes as well as ROS levels. In silico analysis of BoGST superfamily members supported the transcript expression study of BoGSTs in B. oleracea. Overall, these findings indicate that BoGSTU19, BolGSTU24, and BoGSTF10 are potential genes up-regulated during cold conditions. Further investigation on their functional behavior might help in understanding the cold tolerance mechanism conferred by GST superfamily genes. There are also reports on GST genes induced through the ABA pathway [29], and the GST superfamily may thus also be involved in conferring tolerance to other abiotic stress such as salt, drought, and wounding. 3. Materials and Methods 3.1. GST Sequence Retrieval A search was conducted based on annotation for glutathione transferase (GST) genes, and corresponding Bo (B. oleracea) coding (CDS) and protein sequences were retrieved from BRAD [77,78], Bolbase database [79,80] and EnsemblPlants database [81,82]. Furthermore, a complementary method that exploits advanced probabilistic methods, called HMM-profiling, was implemented to increase the accuracy in identifying candidate genes within a genome. For this method, we defined Arabidopsis GST amino acid sequences as a primary source of GST-specific domains (GST N- and GST C-domains) in the HMM analysis. Sequences of 55 proteins from Arabidopsis with GST-specific domains were aligned using Clustal Omega [83,84]. We used those aligned sequences (Stockholm alignment format) as an input for HMMBUILD program in HMMER 3.1b2 software [85] to construct our GST-specific HMM profile. This user-defined GST-HMM profile was used as model to search against the B. oleracea genome acquired from the Bolbase and Ensemble databases using the HMMSEARCH program. Further, the results were subjected to domain analysis using the SMART database [86,87] and CDD (Conserved Domain Database) [88,89] to remove sequences with false domains or partial domain architecture of classic GST proteins. Screening and post-processing of the results were done on the basis of default cutoff values and on the presence of GST-specific domains in their protein structure. The retrieved proteins from the B. oleracea genome were revalidated using local BLASTP searches against the NCBI database for confirmation of putative GST functions. The results obtained from GST-specific HMM profile using Arabidopsis GSTs were confirmed using another GST-specific profile retrieved from Pfam database (GST_N-PF02798 and GST_C-PF00043). 3.2. In Silico Approach for Identification and Characterization of GST Genes To understand the evolutionary relationship among BoGST proteins, processed GST proteins were aligned using CLUSTALW [90,91] with other known GST sequences from Arabidopsis, rice, maize, barley, soybean, wild soybean, and wheat (Table S5) with BLOSUM matrix employing default parameters and the alignment was condensed manually. A molecular phylogenetic tree was constructed using the ML (Maximum likelihood) procedure with the JTT (Jones, Taylor, Thornton) matrix-based amino acid substitution method in MEGA6.06 [92] and 1000 bootstrap replications to access tree topology and reliability. Primary analysis of the predicted molecular weights, pIs, and stability indexes was done using ProtParam [93,94]. Further, N-glycosylation sites were predicted using NetNGlyc 1.0 server [95]. Subcellular localization prediction of predicted BoGST proteins was performed using Protcomp 9.0 from Softberry [96]. Secondary structures of GST proteins were predicted using the garnier script tool from EMBOSS-6.6.0 [97]. Motif analysis of proteins was performed using MEME (Multiple Em for Motif Elicitation v4.10.1) [98] with the following parameters: (1) number of motifs = 10; (2) Motif width ≥6 and ≤50. Gene Structure Display Server (GSDS) web tool [99,100] was used to determine the number of introns and exons, using GFF3 (General file format) and aligning CDS and genomic region of the GST genes. Prediction of putative cis-acting regulatory elements in BoGST genes, using the regions about 1000-bp upstream from the translation initiation site (ATG), was carried out using PlantCARE [101] and PLACE [102], and manually validated as reported by Ibraheem et al. [103]. 3.3. Chromosomal Location and Syntenic Regions of BoGSTs BoGST gene information, chromosome, gene position, strand, and syntenic regions between B. oleracea and Arabidopsis were retrieved using the gene locus search option from Bolbase database [80]. Chromosomal positions of GST genes were drawn using MapChart 2.3 [104] software program. GST genes of B. oleracea were aligned against the B. oleracea and Arabidopsis genome using SyMap v3.4 [105] to obtain syntenic regions within the genome. Subsequently, the derived syntenic regions within the genome and syntenic regions between B. oleracea GST genes and the Arabidopsis genome from the database were used as input for Circos [106] software for visualization of syntenic regions of GST genes. 3.4. Sampling and Preparation of Plant Material To study the expression patterns of BoGST genes, two contrasting lines, Bo106 (cold tolerant (CT)) and Bo107 (cold susceptible (CS)) [8,107] previously referred to as BN106 (cold tolerant (CT)) and BN107 (cold susceptible (CS)) in the reference [5], were grown at the Department of Horticulture, Sunchon National University, Korea. Seeds of the two lines were aseptically inoculated in half-strength murashige and skoog (MS) medium in a growth chamber. The growth chamber was maintained at 25 ± 1 °C for 16 h light/8 h dark conditions. For organ-specific analysis, samples from fresh roots, leaves, and flower buds were removed from three weeks old healthy plants, frozen immediately in liquid nitrogen, and stored at −80 °C until RNA extraction. Three-week-old seedlings were subjected to cold treatment (4 °C) with three replications. Samples of cold-treated leaves were excised at different time points (0, 0.5, 1, 3, 6, 12, 24, and 48 h), frozen immediately in liquid nitrogen, and stored at −80 °C until RNA extraction. Frozen organ samples and cold-treated samples were subjected to total RNA extraction using an RNeasy Mini Kit (Qiagen, Valencia, CA, USA), subsequently RNA cleanup by DNase I treatment, (Takara Bio, Inc., Shiga, Japan). Isolated RNA was quantified using an ND-1000 Spectrophotometer and NanoDrop v3.7 software (NanoDrop Technologies, Wilmington, DE, USA). Synthesis of cDNA from RNA extracts was performed with Superscript III® First-strand Synthesis Supermix kit (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. 3.5. Qualitative and Quantitative PCR Expression Analysis Initial analysis of expression patterns was carried out using microarray data of Arabidopsis with orthologous loci downloaded from the AtGenExpress visualization tool (AVT) [108] during cold stress (leaf samples). Expression cluster analysis of GST genes were performed with the Cluster program [109] and results were visualized using GenePattern software [56,110]. Additionally, cold transcriptome data of B. oleracea for Bo106 (CT) and Bo107 (CS) lines were downloaded from the NCBI database [111] using TSA (Transcriptome Shotgun assembly; GAQY00000000) and SRA (Sequence Read Archive; SRS490050). Further, qualitative expression analysis using RT-PCR was conducted using one-step EmeraldAmp GT PCR Master Mix (Takara, Bio, Inc., Shiga, Japan). Gene-specific primer pairs for BoGSTs were used for RT-PCR and the actin gene from B. oleracea (JQ435879) was used as a housekeeping control gene (Table S13). RT-PCR was performed using 50 ng cDNA (1 µL) from organ and cold-treated samples as template in a master mix consisting of 2 µL primer pairs (10 pmol each of forward and reverse primer), 8 µL sterile water, and 9 µL Emerald master mix, in a total volume of 20 µL. PCR conditions were set as follows: initial denaturation 94 °C, succeeded by 35 cycles (30 cycles for organ samples) of denaturation at 94 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 45 s, with a final extension of 5 min at 72 °C for cold-treated samples. PCR products were visualized using 1.5% agarose gels (Duchefa Biochemie, Haarlem, The Netherlands). Real-Time quantitative PCR (qRT-PCR) was executed using 1 µL cDNA in a 20-µL reaction volume with iTaqTM SYBR® Green Super-mix with ROX (Foster City, CA, USA). Class-wise gene-specific primers for qRT-PCR were employed in this experiment (Table S13). Thermal-cycler conditions were set as follows: 5 min at 95 °C, followed by 40 cycles at 95 °C for 10 s, 60 °C for 10 s, 72 °C for 20 s, and then melting at 72 °C for 60 s and 97 °C for 1 s. The fluorescence was assessed following the last step of each cycle. Product amplification, detection, and data inspection were carried out using LightCycler96 (Roche, Basel, Switzerland). Relative gene expression levels were calculated using the ∆∆Ct method. Actin was used as housekeeping gene. 3.6. Data Statistics Statistical data analysis was performed for the relative gene expression levels from three biological replicates under each treatment (time-point) × genotype (inbred line) combinations. The log-transformed values were analyzed by two-way analysis of variance (ANOVA) following a generalized linear model using the MINITAB 16 (Minitab Inc., State College, PA, USA) statistical software. To separate the means under each treatment, a Tukey’s pairwise comparison test was performed. 4. Conclusions This is the first report on genome-wide characterization of GSTs in B. oleracea. In short, using a combined computational strategy, we identified 65 BoGSTs in the B. oleracea genome and characterized them based on domain, gene, and protein structures, sequence similarities, and expression patterns in response to cold stress conditions. Using two contrasting lines, BoGST genes were found to possess potential functions against cold stress in B. oleracea. Overall, the roles of GST along with GSH conjugation in various pathways and degradation process are important to consider for engineering of these candidates gene in recombinant DNA technology for the development of suitable and elite transgenic cultivars that can withstand various abiotic stresses. Acknowledgments This research was supported by Golden Seed Project (Center for Horticultural Seed Development, No. 213003-04-4-SB110), Ministry of Agriculture, Food and Rural Affairs (MAFRA), Ministry of Oceans and Fisheries (MOF), Rural Development Administration (RDA) and Korea Forest Service (KFS). Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1211/s1. Click here for additional data file. Author Contributions Ill-Sup Nou, Mi-Young Chung, and Jong-In Park conceived of the study. Harshavardhanan Vijayakumar carried out the experimental work. Harshavardhanan Vijayakumar and Senthil Kumar Thamilarasan drafted the manuscript. Senthil Kumar Thamilarasan, Ashokraj Shanmugam and Sathishkumar Natarajan performed in-silico analysis. Hee-Jeong Jung provided support for data analysis. HyeRan Kim and Senthil Kumar Thamilarasan contributed in manuscript preparation. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations ABA Abscisic acid BoGST Brassica oleracea Glutathione transferases CS Cold susceptible CT Cold tolerant DHAR Dehydro ascorbate reductase GHR Glutathionyl Hydroquinone reductase GPX Glutathione peroxidases GR Glutathione reductase GSH Glutathione GST Glutathione transferases GST2N GST protein with two repeats of thioredoxin domain mPGES2 Microsomal Prostaglandin E Synthase type 2 ROS Reactive oxygen species SOD Superoxide dismutase Figure 1 Phylogenetic trees and classification of glutathione transferase (GST) proteins of B. oleracea using Arabidopsis, rice, barley, wheat, sorghum, maize, and soybean published GST proteins. The unrooted phylogenetic trees was constructed based on multiple sequence alignment using ClustalW followed by maximum-likelihood method using JTT model MEGA6.06 software. Highlighted genes with diamond symbol, orange color represents BoGSTs and blue color represents orthologous GST genes. Figure 2 Chromosomal distribution of GST genes in B. oleracea genome plotted using Mapchart software. The red box indicates genes that are clustered within their GST class. Figure 3 Comparative genome mapping of orthologous and paralogous GST’s genes between B. oleracea and Arabidopsis chromosomes. A high level of conserved syntenic regions between the two species was evident. Figure 4 BoGST mRNAs showed distinct expression patterns in B. oleracea organs (R—Root, L—Leaf, and Fb—Flower bud). Transcripts specific for each GST genes amplified by reverse transcription-PCR, visualized using 1.5% agarose gel. Figure 5 The Venn diagram shows differentially expressed GST genes from cold transcriptome data of CT and CS lines: (a) up-regulated genes and (b) down-regulated genes. Expression profile cluster analysis of BoGST superfamily using Arabidopsis orthologous microarray data under cold stress: (c) aerial (leaf) sample. Expression cluster with red color indicates up-regulated genes and green color indicates down-regulated genes. Figure 6 Simplified overview of GST genes involved in cold pathway, colored diamond shape represents the possible role of different GST classes as mentioned in Table 4. Potential deleterious compound are shown in red color, black solid arrow indicates proven experimental evidence available in literature, black dotted arrow indicates possible predicted function based on localization and protein–protein interaction and the question mark indicates the unknown cellular reaction during cold stress. Role of GSH: photosystem in light perception are proposed pathways for GST genes during cold stress. TFs, transcription factor; Transcription factor R bHLH family; C1, R2R3-MYB protein family; ABA, abscisic acid; ABRE, abscisic acid responsive element; COR, cold responsive element; NO−, nitric oxide; cADPR, cyclic adenosine triphosphate ribose; Ca2+, calcium ion signal; CDPK, calcium dependent protein kinases; ICE1, inducer of CBF expression 1; CBF/DREB1, C-repeat binding factor/dehydration responsive-element binding; DRE, dehydration responsive element; GAOX, gibberellic acid oxidase; GA, gibberellic acid; MYB, myeloblastic transcription factor; ROS, reactive oxygen species; GST, glutathione transferase; PCD, programmed cell death; R-SG, conjugated compound with GSH; MAPK, mitogen-activated protein kinase; MAPKK, MAPK kinase; Cys, cysteine; Met, methionine; GSH, glutathione; GSSG, reduced glutathione; DHAR, dehydroascorbate reductase; DHA, dehydroascorbate; MDHAR- mono dehydroascorbate reductase; Asc, ascorbate; AsPX, ascorbate peroxidase; H2O2, hydrogen peroxide; GR, glutathione reductase; GRX, glutaredoxin. Figure 7 Relative quantitative (RQ) expression analysis of 14 BoGST genes which are involved in: (a) transcription activation; (b) signal perception in plasma membrane; (c) H2O2 reduction in chloroplast; and (d) co-regulated genes in anthocyanin sequestration to the vacuoles after cold stress treatment in Brassica oleracea. X-axis represents Time Course (0, 1, 3, 6, 12, 24, and 48 h) and Y-axis represents relative mRNA expression. Graph with orange line is CT line (Bo106), blue line is CS line (Bo107). For each gene, data are represented as relative expression levels to the levels measured at 0 h (2−ΔΔCt). Graph shows the mean of three biological replicates ± standard deviation. (a–e) lowercase letters represent significant differences between different time courses (one-way ANOVA, Tukey’s Test, p < 0.05). ijms-17-01211-t001_Table 1Table 1 Number of GST genes content in Brassica oleracea, Arabidopsis thaliana, Hordeum vulgare, Populus trichocarpa, Solanum lycopersicum, Oryza sativa, Zea mays and Glycine max. Plant Brassica oleracea Arabidopsis thaliana Hordeum vulgare Populus trichocarpa Solanum lycopersicum Oryza sativa Zea mays Glycine max GST Family Number Number Number Number Number Number Number Number Tau 28 28 50 58 56 40 28 20 Phi 14 13 21 9 5 16 12 4 Theta 2 2 1 2 4 2 N/A a N/A a Zeta 2 2 5 2 2 3 2 1 Lambda 3 3 2 3 5 N/A a N/A a N/A a DHAR 4 4 2 3 6 N/A a N/A a N/A a TCHQD 1 1 1 1 1 N/A a N/A a N/A a EF1G 3 2 2 3 1 N/A a N/A a N/A a Others 8 N/A a N/A a N/A a 1 N/A a N/A a N/A a Total 65 55 84 81 81 61 42 25 Reference [44] [36] [40] [39] [38] [37] [37] a, not available; GST, glutathione transferase; DHAR, dehydroascorbate reductase; TCHQD, tetrachlorohydroquinone dehalogenase; EF1G, elongation factor 1 γ. ijms-17-01211-t002_Table 2Table 2 Characterization of GST genes in Brassica oleracea. Sr. No. Class No. of GST Genes Nucleotide Length Range (bp) ORF Range (bp) No. of Exons Protein Length Range (aa) Mol. Wt. Range (kDa) pI Range Average Domain Range GST N-Region GST C-Region EF1G Region 1 Tau 28 705–4093 570–942 1–2 189–313 21.56–35.13 4.96–8.85 72–75 110–146 - 2 Phi 14 780–1198 603–777 3 200–258 22.59–28.84 5.13–8.21 60–75 114–118 - 3 Theta 2 1406–1470 726–738 7 241–245 27.38–27.66 9.36–9.5 75 128 - 4 Zeta 2 1994–2097 591–714 9 196–237 22.26–26.35 5.29–6.91 44–77 118–119 - 5 Lambda 3 1430–1776 708–906 8–9 187–301 21.37–34.39 5.08–8.82 77–68 88–121 - 6 DHAR 4 851–1421 633–774 3–6 210–257 23.22–28.63 5.76–8.28 56–72 118–121 - 7 TCHQD 1 1071 801 2 266 31.46 9.26 72 99 - 8 EF1G 3 1902–2326 1239–1248 6–7 412–415 46.4–46.57 5.56–5.64 71–81 107–120 106–108 9 GHR 5 1290–1576 954–1212 3–5 317–403 36.43–44.9 6.32–8.2 88–106 111–141 - 10 GST2N 2 2196–2401 1011–1017 11–12 336–338 36.93–36.95 8.81–9.26 76–77 - - 11 mPGES2 1 1464 942 6 313 35.13 8..85 72 146 - Sr. No., serial number; bp, base pair; Mol. Wt., molecular weight; aa, amino acid; kDa, Kilodalton; pI, Iso-electric point; GST, Glutathione transferase; N-, N-terminal; C-, C-terminal; EF1G, Elongation factor 1 γ. ijms-17-01211-t003_Table 3Table 3 Organization of GST super-family in B. oleracea based on pathway analysis using KEGG database. Class Abbreviation Genes Predicted Function Tau GSTU 28 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism, Pyruvate metabolism, Phenylpropanoid biosynthesis Phi GSTF 14 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism, Pyruvate metabolism, Phenylpropanoid biosynthesis, Arachidonic acid metabolism Theta GSTT 2 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism, Phenylpropanoid biosynthesis Zeta GSTZ 2 Drug metabolism-cytochrome P450, Glutathione metabolism, Styrene degradation, Tyrosine and Pyruvate metabolism Lambda GSTL 3 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism DHAR DHAR 4 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism, Pyruvate metabolism, Phenylpropanoid biosynthesis, Ascorbate and aldarate metabolism, Aminoacyl-tRNA biosynthesis TCHQD TCHQD 1 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism EF1G EF1G 3 N/A a GHR GHR 5 Drug and Xenobiotics metabolism-cytochrome P450, Glutathione metabolism GST2N GST2N 2 N/A a mPGES2 mPGES2 1 N/A a a, Not available. ijms-17-01211-t004_Table 4Table 4 Predicted BoGST genes involving in various pathways during cold stress in B. oleracea. Sr. No. Image Possible Role Localization GST Family Possible GST Class Involvement 1 TFs Nucleus Tau and Theta BoGSTU1, BoGSTU2, BoGSTU3, BoGSTU6, BoGSTU10, BoGSTU17, BoGSTU18, BoGSTU19, BoGSTU24, BoGSTU25, BoGSTT1, BoGSTT2 2 Gene induction Nucleus Unknown Unknown classes 3 H2O2 Reduction Chloroplast DHAR BoDHAR1, BoDHAR2, BoDHAR3, BoDHAR4 4 H2O2 Reduction Chloroplast Lambda BoGSTL1, BoGSTL2 5 GSH conjugation Cytoplasm Unknown Unknown classes 6 Anthocyanin-transporting gene Cytoplasm Phi BoGSTF9, BoGSTF10 7 Signaling protein Plasma membrane TCHQD BoTCHQD1 ==== Refs References 1. Thomashow M.F. Plant cold acclimation: Freezing tolerance genes and regulatory mechanisms Annu. Rev. Plant Physiol. Plant Mol. Biol. 1999 50 571 599 10.1146/annurev.arplant.50.1.571 15012220 2. Yamaguchi-Shinozaki K. Shinozaki K. Transcriptional regulatory networks in cellular responses and tolerance to dehydration and cold stresses Annu. Rev. Plant Biol. 2006 57 781 803 10.1146/annurev.arplant.57.032905.105444 16669782 3. Yamaguchi-Shinozaki K. Shinozaki K. Organization of cis-acting regulatory elements in osmotic- and cold-stress-responsive promoters Trends Plant Sci. 2005 10 88 94 10.1016/j.tplants.2004.12.012 15708346 4. Lee B. Lee H. Xiong L. Zhu J.-K. A mitochondrial complex I defect impairs cold-regulated nuclear gene expression Plant Cell 2002 14 1235 1251 10.1105/tpc.010433 12084824 5. Ahmed N.U. Jung H.-J. Park J.-I. Cho Y.-G. Hur Y. Nou I.-S. Identification and expression analysis of cold and freezing stress responsive genes of Brassica oleracea Gene 2015 554 215 223 10.1016/j.gene.2014.10.050 25445291 6. Du C. Hu K. Xian S. Liu C. Fan J. Tu J. Fu T. Dynamic transcriptome analysis reveals AP2/ERF transcription factors responsible for cold stress in rapeseed (Brassica napus L.) factors responsible for cold stress in rapeseed (Brassica napus L.) Mol. Genet. Genom. 2016 291 1053 1067 10.1007/s00438-015-1161-0 26728151 7. Hwang I. Jung H.-J. Park J.-I. Yang T.-J. Nou I.-S. Transcriptome analysis of newly classified bZIP transcription factors of Brassica rapa in cold stress response Genomics 2014 104 194 202 10.1016/j.ygeno.2014.07.008 25075938 8. Thamilarasan S.K. Park J.-I. Jung H.-J. Nou I.-S. Genome-wide analysis of the distribution of AP2/ERF transcription factors reveals duplication and CBFs genes elucidate their potential function in Brassica oleracea BMC Genom. 2014 15 1211 10.1186/1471-2164-15-422 24888752 9. Livingston D.P. Premakumar R. Tallury S.P. Carbohydrate partitioning between upper and lower regions of the crown in oat and rye during cold acclimation and freezing Cryobiology 2006 52 200 208 10.1016/j.cryobiol.2005.11.001 16359655 10. Kovács Z. Simon-Sarkadi L. Szucs A. Kocsy G. Differential effects of cold, osmotic stress and abscisic acid on polyamine accumulation in wheat Amino Acids 2010 38 623 631 10.1007/s00726-009-0423-8 19960214 11. Christie P.J. Alfenito M.R. Walbot V. Impact of low-temperature stress on general phenylpropanoid and anthocyanin pathways: Enhancement of transcript abundance and anthocyanin pigmentation in maize seedlings Planta 1994 194 541 549 10.1007/BF00714468 12. Baskar V. Gururani M.A. Yu J.W. Park S.W. Engineering glucosinolates in plants: Current knowledge and potential uses Appl. Biochem. Biotechnol. 2012 168 1694 1717 10.1007/s12010-012-9890-6 22983743 13. Janska A. Maršík P. Zelenková S. Ovesná J. Cold stress and acclimation-what is important for metabolic adjustment? Plant Biol. 2010 12 395 405 10.1111/j.1438-8677.2009.00299.x 20522175 14. Chinnusamy V. Zhu J. Zhu J.-K. Cold stress regulation of gene expression in plants Trends Plant Sci. 2007 12 444 451 10.1016/j.tplants.2007.07.002 17855156 15. Miura K. Jin J.B. Lee J. Yoo C.Y. Stirm V. Miura T. Ashworth E.N. Bressan R.A. Yun D.-J. Hasegawa P.M. SIZ1-mediated sumoylation of ICE1 controls CBF3/DREB1A expression and freezing tolerance in Arabidopsis Plant Cell 2007 19 1403 1414 10.1105/tpc.106.048397 17416732 16. Liu S. Liu Y. Yang X. Tong C. Edwards D. Parkin I.A.P. Zhao M. Ma J. Yu J. Huang S. The Brassica oleracea genome reveals the asymmetrical evolution of polyploid genomes Nat. Commun. 2014 5 3930 10.1038/ncomms4930 24852848 17. Kitamura S. Shikazono N. Tanaka A. TRANSPARENT TESTA 19 is involved in the accumulation of both anthocyanins and proanthocyanidins in Arabidopsis Plant J. 2004 37 104 114 10.1046/j.1365-313X.2003.01943.x 14675436 18. Alfenito M.R. Souer E. Goodman C.D. Buell R. Mol J. Koes R. Walbot V. Functional complementation of anthocyanin sequestration in the vacuole by widely divergent glutathione S -transferases Plant Cell 1998 10 1135 1149 10.1105/tpc.10.7.1135 9668133 19. Dixon D.P. Cummins I. Cole D.J. Edwards R. Glutathione-mediated detoxification systems in plants Curr. Opin. Plant Biol. 1998 1 258 266 10.1016/S1369-5266(98)80114-3 10066594 20. Oakley A. Glutathione transferases: A structural perspective Drug Metab. Rev. 2011 43 138 151 10.3109/03602532.2011.558093 21428697 21. Lallement P.-A. Brouwer B. Keech O. Hecker A. Rouhier N. The still mysterious roles of cysteine-containing glutathione transferases in plants Front. Pharmacol. 2014 5 192 10.3389/fphar.2014.00192 25191271 22. Dixon D.P. Edwards R. Glutathione transferases Arabidopsis Book 2010 8 e0131 10.1199/tab.0131 22303257 23. Edwards R. Dixon D.P. Plant glutathione transferases Methods Enzymol. 2005 401 169 186 16399386 24. Agrawal G.K. Jwa N.-S. Rakwal R. A pathogen-induced novel rice (Oryza sativa L.) gene encodes a putative protein homologous to type II glutathione S -transferases Plant Sci. 2002 163 1153 1160 10.1016/S0168-9452(02)00331-X 25. Mueller L.A. Goodman C.D. Silady R.A. Walbot V. AN9, a petunia glutathione S -transferase required for anthocyanin sequestration, is a flavonoid-binding protein Plant Physiol. 2000 123 1561 1570 10.1104/pp.123.4.1561 10938372 26. Kampranis S.C. Damianova R. Atallah M. Toby G. Kondi G. Tsichlis P.N. Makris A.M. A novel plant glutathione S -transferase/peroxidase suppresses Bax lethality in yeast J. Biol. Chem. 2000 275 29207 29216 10.1074/jbc.M002359200 10859306 27. Dixon D.P. Hawkins T. Hussey P.J. Edwards R. Enzyme activities and subcellular localization of members of the arabidopsis glutathione transferase superfamily J. Exp. Bot. 2009 60 1207 1218 10.1093/jxb/ern365 19174456 28. Dixon D.P. Sellars J.D. Edwards R. The Arabidopsis phi class glutathione transferase AtGSTF2: Binding and regulation by biologically active heterocyclic ligands Biochem. J. 2011 438 63 70 10.1042/BJ20101884 21631432 29. Moons A. Regulatory and functional interactions of plant growth regulators and plant glutathione S -transferases (GSTs) Vitam. Horm. 2005 72 155 202 16492471 30. Dixon D.P. Davis B.G. Edwards R. Functional divergence in the glutathione transferase superfamily in plants: Identification of two classes with putative functions in redox homeostasis in Arabidopsis thaliana J. Biol. Chem. 2002 277 30859 30869 10.1074/jbc.M202919200 12077129 31. Dixon D.P. Lapthorn A. Edwards R. Plant glutathione transferases Genome Biol. 2002 3 3004.1 3004.10 10.1186/gb-2002-3-3-reviews3004 32. Anderson J.V. Davis D.G. Abiotic stress alters transcript profiles and activity of glutathione S -transferase, glutathione peroxidase, and glutathione reductase in Euphorbia esula Physiol. Plant. 2004 120 421 433 10.1111/j.0031-9317.2004.00249.x 15032839 33. Dixon D.P. Cole D.J. Edwards R. Characterisation of a zeta class glutathione transferase from Arabidopsis thaliana with a putative role in tyrosine catabolism Arch. Biochem. Biophys. 2000 384 407 412 10.1006/abbi.2000.2125 11368331 34. Keck A.-S. Finley J.W. Cruciferous vegetables: Cancer protective mechanisms of glucosinolate hydrolysis products and selenium Integr. Cancer Ther. 2004 3 5 12 10.1177/1534735403261831 15035868 35. Armstrong R.N. Structure, catalytic mechanism, and evolution of the glutathione transferases Chem. Res. Toxicol. 1997 10 2 18 10.1021/tx960072x 9074797 36. Rezaei M.K. Shobbar Z.-S. Shahbazi M. Abedini R. Zare S. Glutathione S -transferase (GST) family in barley: Identification of members, enzyme activity, and gene expression pattern J. Plant Physiol. 2013 170 1277 1284 10.1016/j.jplph.2013.04.005 23664583 37. McGonigle B. Keeler S.J. Lau S.C. Koeppe M.K. Keefe D.P.O. A Genomics Approach to the comprehensive analysis of the glutathione S -transferase gene family in soybean and maize Plant Physiol. 2000 124 1105 1120 10.1104/pp.124.3.1105 11080288 38. Soranzo N. Sari Gorla M. Mizzi L. de Toma G. Frova C. Organisation and structural evolution of the rice glutathione S -transferase gene family Mol. Genet. Genom. 2004 271 511 521 10.1007/s00438-004-1006-8 15069639 39. Csiszar J. Horvath E. Vary Z. Galle A. Bela K. Brunner S. Tari I. Glutathione transferase supergene family in tomato: Salt stress-regulated expression of representative genes from distinct GST classes in plants primed with salicylic acid Plant Physiol. Biochem. 2014 78 15 26 10.1016/j.plaphy.2014.02.010 24607575 40. Lan T. Yang Z.-L. Yang X. Liu Y.-J. Wang X.-R. Zeng Q.-Y. Extensive functional diversification of the populus glutathione S -transferase supergene family Plant Cell 2009 21 3749 3766 10.1105/tpc.109.070219 19996377 41. Sheehan D. Meade G. Foley V.M. Dowd C.A. Structure, function and evolution of glutathione transferases: Implications for classification of non-mammalian members of an ancient enzyme superfamily J. Biochem. 2001 16 1 16 10.1042/bj3600001 42. Frova C. The plant glutathione transferase gene family: Genomic structure, functions, expression and evolution Physiol. Plant. 2003 119 469 479 10.1046/j.1399-3054.2003.00183.x 43. Mohsenzadeh S. Esmaeili M. Moosavi F. Shahrtash M. Plant glutathione S -transferase classification, structure and evolution Afr. J. Biotechnol. 2011 10 8160 8165 44. Wagner U. Edwards R. Dixon D.P. Mauch F. Probing the diversity of the Arabidopsis glutathione S -transferase gene family Plant Mol. Biol. 2002 49 515 532 10.1023/A:1015557300450 12090627 45. Droog F. Plant glutathione S -transferases, a tale of theta and tau J. Plant Growth Regul. 1997 95 107 10.1007/PL00006984 46. Edwards R. Dixon D.P. Walbot V. Plant glutathione S -transferases: Enzymes with multiple functions in sickness and in health Trends Plant Sci. 2000 5 193 198 10.1016/S1360-1385(00)01601-0 10785664 47. Sappl P.G. Carroll A.J. Clifton R. Lister R. Whelan J. Harvey Millar A. Singh K.B. The Arabidopsis glutathione transferase gene family displays complex stress regulation and co-silencing multiple genes results in altered metabolic sensitivity to oxidative stress Plant J. 2009 58 53 68 10.1111/j.1365-313X.2008.03761.x 19067976 48. Seppänen M.M. Cardi T. Borg Hyökki M. Pehu E. Characterization and expression of cold-induced glutathione S -transferase in freezing tolerant Solanum commersonii , sensitive S. tuberosum and their interspecific somatic hybrids Plant Sci. 2000 153 125 133 10.1016/S0168-9452(99)00252-6 10717318 49. Tsuchiya T. Takesawa T. Kanzaki H. Nakamura I. Genomic structure and differential expression of two tandem-arranged GSTZ genes in rice Gene 2004 335 141 149 10.1016/j.gene.2004.03.020 15194197 50. Jain M. Ghanashyam C. Bhattacharjee A. Comprehensive expression analysis suggests overlapping and specific roles of rice glutathione S -transferase genes during development and stress responses BMC Genom. 2010 11 1211 10.1186/1471-2164-11-73 20109239 51. Baek K.-H. Skinner D.Z. Alteration of antioxidant enzyme gene expression during cold acclimation of near-isogenic wheat lines Plant Sci. 2003 165 1221 1227 10.1016/S0168-9452(03)00329-7 52. Ahsan N. Lee D.G. Alam I. Kim P.J. Lee J.J. Ahn Y.O. Kwak S.S. Lee I.J. Bahk J.D. Kang K.Y. Comparative proteomic study of arsenic-induced differentially expressed proteins in rice roots reveals glutathione plays a central role during As stress Proteomics 2008 8 3561 3576 10.1002/pmic.200701189 18752204 53. Kumar S. Jin M. Weemhoff J. Cytochrome P450-mediated phytoremediation using transgenic plants: A need for engineered cytochrome P450 enzymes J. Pet. Environ. Eng. 2012 29 997 1003 10.4172/2157-7463.1000127 25298920 54. Marrs K.A. The functions and regulation of glutathione S -transferases in plants Annu. Rev. Plant Physiol. Plant Mol. Biol. 1996 47 127 158 10.1146/annurev.arplant.47.1.127 15012285 55. Takahashi Y. Hasezawa S. Kusaba M. Nagata T. Expression of the auxin-regulated parA gene in transgenic tobacco and nuclear localization of its gene products Planta 1995 196 111 117 10.1007/BF00193224 7767232 56. Blast2GO Available online: https://www.blast2go.com/ (accessed on 11 January 2016) 57. Blackburn A.C. Woollatt E. Sutherland G.R. Board P.G. Characterization and chromosome location of the gene GSTZ1 encoding the human Zeta class glutathione transferase and maleylacetoacetate isomerase Cytogenet. Cell Genet. 1998 83 109 114 10.1159/000015145 9925947 58. Takesawa T. Ito M. Kanzaki H. Kameya N. Nakamura I. Over-expression of ζ glutathione S -transferase in transgenic rice enhances germination and growth at low temperature Mol. Breed. 2002 9 93 101 10.1023/A:1026718308155 59. Pandey B. Sharma P. Pandey D.M. Varshney J. Sheoran S. Singh M. Singh R. Sharma I. Chatrath R. Comprehensive computational analysis of different classes of glutathione S -transferases in Triticum aestivum L. Plant Omics J. 2012 5 518 531 60. Itzhaki H. Maxson J.M. Woodson W.R. An ethylene-responsive enhancer element is involved in the senescence-related expression of the carnation glutathione-S -transferase (GST1) gene Proc. Natl. Acad. Sci. USA 1994 91 8925 8929 10.1073/pnas.91.19.8925 8090746 61. Van der Zaal B.J. Droog F.N. Pieterse F.J. Hooykaas P.J. Auxin-sensitive elements from promoters of tobacco GST genes and a consensus as-1-like element differ only in relative strength Plant Physiol. 1996 110 79 88 10.1104/pp.110.1.79 8587996 62. Gronwald J.W. Plaisance K.L. Isolation and characterization of glutathione S -transferase isozymes from sorghum Plant Physiol. 1998 117 877 892 10.1104/pp.117.3.877 9662530 63. Genomatix Software Suit Available online: http://www.genomatix.de/cgi-bin/dialign/dialign.pl (accessed on 23 January 2016) 64. Board P.G. Baker R.T. Chelvanayagam G. Jermiin L.S. Zeta, a novel class of glutathione transferases in a range of species from plants to humans Biochem. J. 1997 328 929 935 10.1042/bj3280929 9396740 65. Ziolkowski P.A. Kaczmarek M. Babula D. Sadowski J. Genome evolution in Arabidopsis/Brassica: Conservation and divergence of ancient rearranged segments and their breakpoints Plant J. 2006 47 63 74 10.1111/j.1365-313X.2006.02762.x 16824180 66. Moons A. Osgstu3 and osgtu4, encoding tau class glutathione S -transferases, are heavy metal- and hypoxic stress-induced and differentially salt stress-responsive in rice roots FEBS Lett. 2003 553 427 432 10.1016/S0014-5793(03)01077-9 14572664 67. Chi Y. Cheng Y. Vanitha J. Kumar N. Ramamoorthy R. Ramachandran S. Jiang S.-Y. Expansion mechanisms and functional divergence of the glutathione S -transferase family in sorghum and other higher plants DNA Res. 2011 18 1 16 10.1093/dnares/dsq031 21169340 68. Chinnusamy V. Zhu J. Zhu J.K. Gene regulation during cold acclimation in plants Physiol. Plant. 2006 126 52 61 10.1111/j.1399-3054.2006.00596.x 69. Jung H.-J. Dong X. Park J.-I. Thamilarasan S.K. Lee S.S. Kim Y.-K. Lim Y.-P. Nou I.-S. Hur Y. Genome-wide transcriptome analysis of two contrasting brassica rapa doubled haploid lines under cold-stresses using Br135K oligomeric chip PLoS ONE 2014 9 1211 10.1371/journal.pone.0106069 25167163 70. Tausz M. Šircelj H. Grill D. The glutathione system as a stress marker in plant ecophysiology: Is a stress-response concept valid? J. Exp. Bot. 2004 55 1955 1962 10.1093/jxb/erh194 15234995 71. Zagorchev L. Seal C.E. Kranner I. Odjakova M. A central role for thiols in plant tolerance to abiotic stress Int. J. Mol. Sci. 2013 14 7405 7432 10.3390/ijms14047405 23549272 72. Petrussa E. Braidot E. Zancani M. Peresson C. Bertolini A. Patui S. Vianello A. Plant flavonoids-biosynthesis, transport and involvement in stress responses Int. J. Mol. Sci. 2013 14 14950 14973 10.3390/ijms140714950 23867610 73. Chen J.-H. Jiang H.-W. Hsieh E.-J. Chen H.-Y. Chien C.-T. Hsieh H.-L. Lin T.-P. drought and salt stress tolerance of an Arabidopsis glutathione S -transferase U17 knockout mutant are attributed to the combined effect of glutathione and abscisic acid Plant Physiol. 2012 158 340 351 10.1104/pp.111.181875 22095046 74. Ushimaru T. Nakagawa T. Fujioka Y. Daicho K. Naito M. Yamauchi Y. Nonaka H. Amako K. Yamawaki K. Murata N. Transgenic Arabidopsis plants expressing the rice dehydroascorbate reductase gene are resistant to salt stress J. Plant Physiol. 2006 163 1179 1184 10.1016/j.jplph.2005.10.002 17032619 75. Zhang B. Hu Z. Zhang Y. Li Y. Zhou S. Chen G. A putative functional MYB transcription factor induced by low temperature regulates anthocyanin biosynthesis in purple kale (Brassica Oleracea var. acephala f. tricolor ) Plant Cell Rep. 2012 31 281 289 10.1007/s00299-011-1162-3 21987119 76. Vanderauwera S. Zimmermann P. Van Breusegem F. Langebartels C. Gruissem W. Inze D. Breusegem F. Van Genome-wide analysis of hydrogen peroxide-regulated gene expression in Arabidopsis reveals a high light-induced transcriptional cluster involved in anthocyanin biosynthesis Plant Physiol. 2005 139 806 821 10.1104/pp.105.065896 16183842 77. Cheng F. Liu S. Wu J. Fang L. Sun S. Liu B. Li P. Hua W. Wang X. BRAD, the genetics and genomics database for Brassica plants BMC Plant Biol. 2011 11 1211 10.1186/1471-2229-11-136 21995777 78. Brassica Database Available online: http://brassicadb.org/brad/ (accessed on 20 November 2015) 79. Yu J. Zhao M. Wang X. Tong C. Huang S. Tehrim S. Liu Y. Hua W. Liu S. Bolbase: A comprehensive genomics database for Brassica oleracea BMC Genom. 2013 14 1211 10.1186/1471-2164-14-664 24079801 80. Bolbase Database Available online: http://www.ocri-genomics.org/bolbase/ (accessed on 20 November 2015) 81. Kersey P.J. Allen J.E. Christensen M. Davis P. Falin L.J. Grabmueller C. Hughes D.S.T. Humphrey J. Kerhornou A. Khobova J. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega Mol. Syst. Biol. 2011 7 539 21988835 82. EnsemblPlants Database Available online: http://plants.ensembl.org/Brassica_oleracea/Info/Index (accessed on 20 November 2015) 83. Eddy S.R. A new generation of homology search tools based on probabilistic inference Genome Inform. 2009 23 205 211 20180275 84. Clustal Omega Available online: http://www.ebi.ac.uk/Tools/msa/clustalo/ (accessed on 9 December 2015) 85. Schultz J. Milpetz F. Bork P. Ponting C.P. SMART, a simple modular architecture research tool: Identification of signaling domains Proc. Natl. Acad. Sci. USA 1998 95 5857 5864 10.1073/pnas.95.11.5857 9600884 86. Marchler-Bauer A. Derbyshire M.K. Gonzales N.R. Lu S. Chitsaz F. Geer L.Y. Geer R.C. He J. Gwadz M. Hurwitz D.I. CDD: NCBI’s conserved domain database Nucleic Acids Res. 2015 43 D222 D226 10.1093/nar/gku1221 25414356 87. SMART Database Available online: http://smart.embl-heidelberg.de/ (accessed on 26 November 2015) 88. Thompson J.D. Higgins D.G. Gibson T.J. CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice Nucleic Acids Res. 1994 22 4673 4680 10.1093/nar/22.22.4673 7984417 89. Conserved Domain Database Available online: http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml (accessed on 26 November 2015) 90. Tamura K. Stecher G. Peterson D. Filipski A. Kumar S. MEGA6: Molecular evolutionary genetics analysis version 6.0 Mol. Biol. Evol. 2013 30 2725 2729 10.1093/molbev/mst197 24132122 91. CLUSTALW Available online: http://www.genome.jp/tools/clustalw/ (accessed on 9 December 2015) 92. Gasteiger E. Hoogland C. Gattiker A. Duvaud S. Wilkins M.R. Appel R.D. Bairoch A. Protein identification and analysis tools on the ExPASy server The Proteomics Protocols Handbook Humana Press New York, NY, USA 2005 571 607 93. Gupta R. Jung E. Brunak S. Prediction of N -glycosylation sites in human proteins 2004 in press 94. ProtParam Available online: http://web.expasy.org/protparam/ (accessed on 14 February 2016) 95. Bailey T.L. Williams N. Misleh C. Li W.W. MEME: Discovering and analyzing DNA and protein sequence motifs Nucleic Acids Res. 2006 34 369 373 10.1093/nar/gkl198 16845028 96. Softberry Available online: http://linux1.softberry.com/berry.phtml (accessed on 14 February 2016) 97. EMBOSS-6.6.0 Available online: http://emboss.sourceforge.net/apps/ (accessed on 4 March 2016) 98. Hu B. Jin J. Guo A.-Y. Zhang H. Luo J. Gao G. GSDS 2.0: An upgraded gene feature visualization server Bioinformatics 2014 31 1296 1297 10.1093/bioinformatics/btu817 25504850 99. Rombauts S. Déhais P. Van Montagu M. Rouzé P. PlantCARE, a plant cis-acting regulatory element database Nucleic Acids Res. 1999 27 295 296 10.1093/nar/27.1.295 9847207 100. Gene Structure Display Server (GSDS) Web Tool Available online: http://gsds.cbi.pku.edu.cn/index.php (accessed on 5 March 2016) 101. Higo K. Ugawa Y. Iwamoto M. Korenaga T. Plant cis-acting regulatory DNA elements (PLACE) database: 1999 Nucleic Acids Res. 1999 27 297 300 10.1093/nar/27.1.297 9847208 102. Ibraheem O. Botha C.E.J. Bradley G. In silico analysis of cis-acting regulatory elements in 5′ regulatory regions of sucrose transporter gene families in rice (Oryza sativa Japonica) and Arabidopsis thaliana Comput. Biol. Chem. 2010 34 268 283 10.1016/j.compbiolchem.2010.09.003 21036669 103. Voorrips R.E. MapChart: Software for the graphical presentation of linkage maps and QTLs J. Hered. 2002 93 77 78 10.1093/jhered/93.1.77 12011185 104. Soderlund C. Bomhoff M. Nelson W.M. SyMAP v3.4: A turnkey synteny system with application to plant genomes Nucleic Acids Res. 2011 39 1 9 10.1093/nar/gkr123 20805246 105. Krzywinski M. Schein J. Birol I. Connors J. Gascoyne R. Horsman D. Jones S.J. Marra M.A. Circos: An information aesthetic for comparative genomics Genome Res. 2009 19 1639 1645 10.1101/gr.092759.109 19541911 106. Kayum M.A. Park J.I. Ahmed N.U. Jung H.J. Saha G. Kang J.G. Nou I.S. Characterization and stress-induced expression analysis of Alfin-like transcription factors in Brassica rapa Mol. Genet. Genom. 2015 290 1299 1311 10.1007/s00438-015-0993-y 25618423 107. Reich M. Liefeld T. Gould J. Lerner J. Tamayo P. Mesirov J.P. GenePattern 2.0 Nat. Genet. 2006 38 500 501 10.1038/ng0506-500 16642009 108. AtGenExpress Visualization Tool (AVT) Available online: http://jsp.weigelworld.org/expviz/expviz.jsp (accessed on 18 March 2016) 109. Cluster Program Available online: http://bonsai.hgc.jp/~mdehoon/software/cluster/software.htm (accessed on 21 March 2016) 110. GenePattern Software Available online: http://genepattern.broadinstitute.org/gp/pages/index.jsf (accessed on 21 March 2016) 111. NCBI Database Available online: http://www.ncbi.nlm.nih.gov/ (accessed on 24 November 2015)

PMC005xxxxxx/PMC5000610.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081212ijms-17-01212ArticleDesign of Phosphonated Imidazolium-Based Ionic Liquids Grafted on γ-Alumina: Potential Model for Hybrid Membranes Pizzoccaro Marie-Alix 1Drobek Martin 1Petit Eddy 1Guerrero Gilles 2*Hesemann Peter 2Julbe Anne 1*Taubert Andreas Academic Editor1 Institut Européen des Membranes, UMR-5635 CNRS-UM-ENSCM, Université de Montpellier (CC047), Place Eugène Bataillon, F-34095 Montpellier cedex 5, France; [email protected] (M.-A.P.); [email protected] (M.D.); [email protected] (E.P.)2 Institut Charles Gerhardt, UMR-5253 CNRS-UM-ENSCM, Université de Montpellier (CC1701), Place Eugène Bataillon, F-34095 Montpellier cedex 5, France; [email protected]* Correspondence: [email protected] (G.G.); [email protected] (A.J.); Tel.: +33-467-144-223 (G.G.); +33-467-149-142 (A.J.)27 7 2016 8 2016 17 8 121224 6 2016 20 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Imidazolium bromide-based ionic liquids bearing phosphonyl groups on the cationic part were synthesized and grafted on γ-alumina (γ-Al2O3) powders. These powders were prepared as companion samples of conventional mesoporous γ-alumina membranes, in order to favor a possible transfer of the results to supported membrane materials, which could be used for CO2 separation applications. Effective grafting was demonstrated using energy dispersive X-ray spectrometry (EDX), N2 adsorption measurements, fourier transform infrared spectroscopy (FTIR), and special attention was paid to 31P and 13C solid state nuclear magnetic resonance spectroscopy (NMR). imidazolium-based ionic liquidsγ-aluminaphosphonate coupling agentgraftingsolid state NMRhybrid membrane ==== Body 1. Introduction In competition with amines, ionic liquids are known to interact strongly and reversibly with CO2, making supported ionic liquid (IL) materials versatile solids for use in adsorptive or membrane CO2 separation applications [1]. The most common systems are composed of ILs either impregnated or confined in matrices, which can be porous or non-porous (i.e., polymer, ceramic or hybrid matrices). These materials can have applications, for example, in batteries [2,3,4], as electrolytes [5] or as CO2 separation systems [6,7]. Imidazolium-based ILs grafted onto the surface of porous supports are promising systems for a range of applications, including catalysis [8,9], chromatography [10,11,12] and gas separation [13,14]. These types of systems have been defined by Fehrmann et al. [15], as supported ionic liquids (SILs), which refer to either inert or catalytically-active covalently-bound monolayers of ILs. In these materials, the IL does not act like the bulk liquid anymore, but as a surface modifier. As reported by the authors, tailoring the chemical nature of the support, as well as its microstructure (pore size, size distribution, surface area, etc.), govern the grafting of the IL and its distribution on the support surface. Covalent linking of ILs on a ceramic oxide support appears as an attractive strategy to fine-tune solids with outstanding properties for CO2 adsorption and with improved long-term stability. Ionic liquids can be grafted on mesoporous silica-based supports, such as MCM-41 [16] or SBA-15 [17], and they also can be incorporated within a silica hybrid matrix [18]. These mixed ionic-mineral phases have been the most widely-investigated systems for applications such as heterogeneous catalysis [13,19,20], gas separation [13,14] or CO2 sorption [21]. γ-alumina (γ-Al2O3) is a commonly-used ceramic support, and its hydroxylated surface is attractive for anchoring or grafting active species for either gas separation or heterogeneous catalysis [22]. In addition, this material can be cast easily as a continuous membrane film and was thus selected in this work as a relevant support for grafting imidazolium-based ionic liquids. Controlling the chemical grafting of ILs in the pores of a porous material is much more challenging than their simple impregnation in a porous support, yielding the supported ionic liquid phase (SILP). Obviously, the choice of the functionalized imidazolium-based IL is a key parameter, but the characterization of the grafting reaction is a tricky task. The efficiency of the grafting step needs to be quantified, and the spatial proximity between the grafted sites needs to be determined. Several functionalized imidazolium-based ILs have been reported in the literature with functional groups, such as trimethoxysilyl, thiol-, ether-, carboxylic acid-, amino- and hydroxyl-groups [23]. Each of these functional groups is adapted for grafting on a pre-functionalized support. Vangeli et al. [14] selected the trimethoxysilyl group to react with the hydroxyl groups of silica-based materials pre-treated with a piranha solution. The grafting reaction has been performed in two steps: (i) grafting of a silylated precursor and (ii) quaternization with 1-methylimidazole, yielding the imidazolium species. Despite the detection of carbon by elemental analysis and the measured decrease of support specific surface area after grafting, the demonstration of both the quaternization reaction and anchoring configuration were rather unclear. The chemical modification of γ-alumina powders with organosilanes has been largely investigated in the literature [24,25]. As an alternative, grafting reactions could also be realized with phosphonate or phosphinate coupling functions. Randon et al. [26] have linked phosphoric acid and alkyl phosphonic acid to the surface of both titania and zirconia membranes in order to improve their performance for the ultrafiltration of bovine serum albumin (BSA) proteins. Caro et al. [27] modified a γ-Al2O3 membrane top-layer with alkyl/aryl phosphonic acids, thus resulting in an organophobic behavior. Guerrero et al. [28,29] studied the grafting of phenylphosphonic acid and its ester derivatives on both γ-Al2O3 and TiO2 powders. The surface bonding modes were investigated by both diffuse reflectance infrared spectroscopy (DRIFT) and 31P solid-state magic angle spinning nuclear magnetic resonance spectroscopy (MAS NMR). The same authors also patented a process for modifying an inorganic substrate with organophosphorus coupling agents, relevant for antibacterial applications. In this work, imidazolium-based ILs with phosphonyl functional groups were used for their intrinsic antimicrobial properties [30]. The aim of the present work was to develop an optimized γ-Al2O3/imidazolium-based ILs system able to serve as a preliminary study for developing efficient gas separation hybrid membrane in which the IL will be effectively grafted on the pore surface. The approach involves the synthesis and characterizations of both the γ-Al2O3 support and the functionalized IL, followed by the investigation of the grafting reaction and the quantitative analysis of both the grafting step and the derived hybrid material. Two organophosphorus functionalized imidazolium-based ILs were selected (Figure 1): the 1-methyl-3–(3–(diethylphosphinyl)propyl)-imidazolium bromide (ImPE) and the 1-methyl-3–(3–((trimethoxysilyl)phosphinyl)propyl)-imidazolium bromide (ImTMSP). The synthesis of ImPE was carried out following the procedure described by Mu et al. [31], while the synthesis of ImTMSP was performed for the first time by adapting the work of McKenna et al. [32]. The modification of the γ-Al2O3 surface under either standard or forcing conditions was investigated by energy dispersive X-ray spectrometry (EDX) and N2 adsorption measurements. In addition, key information was derived from fourier transform infrared spectroscopy (FTIR) and 31P, 13C solid state NMR analysis. 2. Results and Discussion γ-alumina (γ-Al2O3) was prepared from boehmite using a sol-gel process described by Leenaars et al. [33] and followed by a 3-h thermal treatment in air at 600 °C. As revealed by 1H MAS NMR, hydroxyl groups are present on the alumina surface, and they are involved in the grafting reaction mechanism by condensation with ester functions (P-OX) of the organophosphonate coupling agent [34]. The surface of γ-Al2O3 powders was modified by treatment with an organic or aqueous grafting solution containing n-fold excess of either ImPE or ImTMSP ionic liquid. The quantity of IL used corresponds to the amount needed for a full surface coverage of the γ-Al2O3 particles (0.6 mmol, assuming an area of 25 Å2 per ionic liquid molecule). Depending on the ILs used, different reaction conditions were applied (Table 1). In order to evidence the spectroscopic characteristics of physisorbed phases or unreacted species on the surface of γ-Al2O3, a grafting experiment was first performed with ImPE in “physisorption conditions”. Secondly, in standard reaction conditions, grafting with ImPE was achieved during several days in an alcoholic solvent, while grafting with ImTMSP was realized during either one night or three days in dry methylene chloride. Otherwise, as reported for the grafting of diethyl phenylphosphonate coupling agents on γ-Al2O3, the use of forcing reaction conditions (i.e., excess of coupling agent relative to full surface coverage, and high temperature) did not lead to dissolution-precipitation mechanism (no formation of bulk aluminum phosphonate phases) and improved the surface grafting density [28,35]. Therefore, finally, forcing reaction conditions were also tested with ImPE in aqueous medium for one night by increasing the reaction temperature up to 130 °C. After the grafting treatment, samples were centrifuged, washed with an ethanol-water solution to remove unreacted and physisorbed species and dried at 70 °C under vacuum before analysis (see Materials and Methods). The characteristics of unmodified and grafted γ-Al2O3 powders are summarized in Table 2. Average phosphorus weight percentages (wt % P) obtained from EDX showed the presence of phosphorus in all of the grafted samples. The proportions do not exceed a full surface coverage (i.e., 3.2 wt % P), which is coherent with the surface reactions of the ILs coupling agents on γ-Al2O3. The sample modified with a six-fold excess of ImPE in forcing reaction condition (ImPE3) exhibited a slight increase of P contents by comparison with the sample obtained with a six-fold excess in standard condition (ImPE1). This result suggests that a higher temperature with a short reaction time tends to increase the rate of the surface modification reaction. For ImPE samples in forcing reaction conditions, an enhancement of ILs coupling agent excess in solution leads to an increase of the γ-Al2O3 surface coverage. In the case of ImTMSP samples, the P contents measured on modified γ-Al2O3 were in the range of the values obtained for the samples modified by ImPE. Nevertheless, no heat activation was needed with ImTMSP, suggesting a difference of reactivity, while the –PO(OSiMe3) function is known to be more reactive [28]. Finally, the efficiency of grafting reactions with ImTMSP did not seem to depend on reaction times, but was rather sensitive to the excess value of the IL coupling agent in solution. Nitrogen adsorption experiments did not reveal any important variation of the specific surface area values between the crude γ-Al2O3 powder and the grafted samples series, which is consistent with only the surface modification. The results of adsorption measurements gave also access to the BET constant (CBET) related to the affinity of the solid with N2 molecules ans so which is characteristic of the adsorbate/material surface interactions, as reported by Galarneau et al. [36]. The decrease of the CBET value reflects a reduction of the enthalpy of adsorption of N2 molecules on the surface and thus gives qualitative information about surface modification. All of the grafted samples showed a lower CBET constant than the starting γ-Al2O3 powders (Table 2). For each kind of coupling agent, the increase of the weight percentage of phosphorus measured on the modified samples correlates with a decrease of the CBET constant. In addition, from the weight percentage of phosphorus and the specific surface area, we can estimate the grafting density (P nm−2) on the surface of the γ-Al2O3 powders, assuming an area of 25 Å2 per phosphonate molecule (Table 2). Therefore, a full surface coverage of the alumina particles should not exceed 4 P atoms by nm2. In all of the grafted samples, values from 0.6 to 1.4 of the grafting density were obtained, suggesting that the surface coverage does not exceed about 30% of the full monolayer in all cases. In comparison with the literature, it has been demonstrated that the grafting on Degussa γ-Al2O3 in organic media with the diethyl phenylphosphonate coupling molecule resulted in a 50% surface coverage [28]. Moreover, in the same study, the authors have also evidenced that the grafting with bis(trimethylsilyl)ester phenylphosphonate coupling agent led to a higher percentage of phosphorus atoms, consistent with the formation of bulk aluminum phosphonate phases. It was also pointed out that the use of the dialkyl ester derivatives in organic media allowed the control of the grafting and excluded the formation of phosphonate phases even under prolonged heating. The partial surface coverage obtained in this study with the diethyl imidazolium phosphonate coupling molecule could result from a possible steric hindrance effect on the γ-Al2O3 surface (due to both the alkyl chain and the imidazolium ring) and also from the low reactivity of the coupling function. Furthermore, the results obtained with ImTMSP showed that the control of the reaction condition parameters can allow the incorporation of coupling agent quantities consistent with sole surface coverage without any evidence of the formation of bulk aluminum phosphonate phases. Thus, additional grafting parameters have to be tested (i.e., grafting duration, concentration, temperature, etc.) in order to optimize the reaction conditions and maximize the γ-alumina surface coverage. The reaction of organophosphorus derivatives on the γ-Al2O3 surface is supposed to involve both coordination of the oxygen of the phosphoryl groups (P=O) to Lewis acid sites and the condensation reactions of P-OX functions (X could be Et or SiMe3) with Al-OH surface groups. According to the literature, there are several possible bonding modes for phosphonate coupling molecules on an oxide surface [34]. In the case of phosphonyl imidazolium-based ILs, the possible bonding modes can be mono-, bi- or tri-dentate (Figure 2). The FTIR spectra between 1400 and 800 cm−1 of the two organophosphonate functionalized imidazolium bromide-based ILs are presented in Figure 3I. ImPE and ImTMSP showed P=O stretching vibrations at 1230 cm−1 and 1251 cm−1, respectively, and C−H deformation vibrations at 956 cm−1 for ImPE and 1035 cm−1 for ImTMSP. Asymmetric and symmetric P-O-C stretching vibrations are present only for ImPE at 1042 and 958 cm−1 [28,37]. In addition, the spectra of ImTMSP present a P-O-Si deformation vibration at 833 cm−1. Starting from the ionic liquid structure, DFT calculation is useful to estimate the different vibration modes of the coupling agents (Figure S10 in the Supplementary Information) and to identify some of the deformation bands, such as the =C-H imidazolium band and the -CH2- spacer alkyl chain band present at about 1165 cm−1 for both ILs. In all grafted samples, we can notice the disappearance of the phosphoryl (P=O) stretching bands near 1230 and 1251 cm−1, suggesting that the phosphoryl oxygen is strongly bonded to Lewis acid surface sites by coordination (Figure 3IIa–c). Moreover, the IR spectra are dominated by an absorption band at 1171 cm−1, typical of the =C-H and –CH2– deformation bands of the imidazolium ring and the spacer alkyl chain. The PO regions of the grafted samples between 950 and 1250 cm−1 differ depending on both the IL and the applied reaction parameters. The IR spectra of samples treated in forcing reaction conditions with ImPE (Figure 3IIb) present a strong absorption band at 1065 cm−1, tentatively ascribed to the P-O-Al stretching vibration [37]. It can be noted that this band became gradually broader when increasing the quantity of grafted species. Therefore, the presence of weak absorption bands at about 1040 and 950 cm−1 (region of P-O-C stretching bands) does not preclude the existence of some P-OEt residual groups. The FTIR spectrum of the sample modified with ImPE under standard reaction conditions (Figure 3IIa) shows a broader absorption band around 1050 cm−1 corresponding to the P-O group stretching mode. This band could be attributed to P-O surface species in organophosphonate/metal oxide systems according to Quiñones et al. [38]. In addition, the presence of strong residual P-O-C stretching bands at about 1040 cm−1 cannot be excluded. The IR spectra of samples prepared with ImTMSP (Figure 3IIc) in standard conditions are quite similar to the ImPE1 spectrum. The most important difference comes from the presence of residual P-O-Si deformation vibration between ~1000 and 800 cm−1, which suggests that all of the coupling functions have not reacted with the alumina surface [28]. To conclude, IR spectroscopy of samples grafted in standard reaction conditions clearly reveals the presence of residual P-O-C or P-O-Si vibrations, stating that phosphonate units are preferentially linked to the alumina surface through bidentate (or monodentate) binding modes. In the case of samples prepared in forcing conditions, weak residual P-O-C stretching modes may be present on the infrared spectra, indicating that the dominating bonding mode of the phosphonate groups seems to involve tridentate PO3 units. 13C cross polarization magic angle spinning (CP-MAS) liquid NMR spectra of ImPE and ImTMSP display both the specific chemical shifts (Figure 4Ia,b) of the 1-methyl-3-propylimidazolium group, numbered from C1 to C7, and those of the coupling functions, numbered from C8 to C9 for –POCH2CH3 and C8 only for –POSiMe3. In comparison, the 13C CP-MAS solid state NMR spectra of the grafted samples ImPE4 (Figure 4IIa) and ImTMSP4 (Figure 4IIb) show a slightly upfield shift for all of the atoms, principally due to the spatial proximity and chemical bonds with the γ-Al2O3 surface. Whatever the coupling function, the integrity of the organic molecule structure was conserved during the grafting process. In both grafted samples’ spectra, we can notice the presence of weak peaks attributed to residual -POCH2CH3 and -POSiMe3 functions, confirming the conclusions derived from the FTIR spectra (Figure 3) and supporting our hypothesis concerning the presence of multimodal bonding modes (i.e., tridentate, bidentate, monodentate). Additional information was provided by the 31P MAS NMR spectra of the grafted γ-Al2O3. The powder treated under physisorption reaction conditions allows identification of the chemical shift corresponding to physisorbed species, with a sharp resonance at 32.1 ppm (Figure S11). The 31P MAS NMR spectra of all of the grafted samples (Figure 5a,b) did not reveal this peak, indicating that only grafted phosphonate species were present on the alumina surface. Moreover, we did not notice any additional upfield sharp resonance peak resulting from a dissolution/precipitation phenomenon and the formation of aluminum phosphonate bulk phases. This was confirmed by powder X-ray diffraction patterns (XRD) highlighting the amorphous structure of all of the grafted samples (Figure S12). The 31P MAS NMR spectra of the γ-Al2O3 powder modified with ImPE under forcing reaction conditions are displayed in Figure 5a (ImPE2, ImPE3 and ImPE4). All three spectra present a broad resonance centered at about 23.6 ppm. The simulation of the spectra indicated the presence of at least three sites (signals at 32.4, 23.6 and 17.9 ppm (Figure 5, Table 3)) revealing the presence of multiple bonding modes (Figure 2) for the phosphonate units as already discussed for the IR spectra. As reported by Brodard-Severac et al. [39], the interaction of the P=O groups with surface Lewis or Brønsted acidic sites should lead to a downfield shift. On this basis, the signal at 32.4 ppm, integrating from 7% to 13%, could be tentatively ascribed to the minor monodentate bonding mode (Figure 2). 13C CP MAS NMR indicated the presence of residual P-OEt functions, and IR spectra showed also weak residual P-O-C stretching modes, stating that the dominating bonding mode of the phosphonate groups seems to involve tridentate PO3 units. Therefore, we propose to ascribe the major signal at 23.6 ppm, integrating from 55% to 58%, to tridentate phosphonate PO3 units grafted on the γ-Al2O3 surface. The third resonance at 17.9 ppm, integrating from 29% to 38%, was then attributed to grafted phosphonate functions in a bidentate mode. For both ImPE3 and ImPE4, it is interesting to notice that the increase of the proportion of this bonding mode correlates well with the increasing intensity of the IR stretches of residual P-OEt functions. Consequently, by increasing the n-fold excess of coupling agents during the grafting reaction in forcing reaction conditions, both tridentate and bidentate bonding modes of the phosphonate units were favored. The 31P MAS NMR spectrum of the γ-Al2O3 powder modified with ImPE under standard reaction conditions is also displayed in Figure 5a (ImPE1). It consists of a broad resonance centered at about 21.6 ppm with an important downfield asymmetrical shape. The simulation of the ImPE1 spectrum using a minimum number of resonances with a Gaussian–Lorentzian shape indicates the presence of at least three sites (Figure 5a, Table 3) at 31.6, 22.1 and 18.1 ppm, evidencing the presence of multiple bonding modes, as for ImPE2 to 4 (Figure 2). On the basis of the IR results, a higher number of residual P-OEt functions was detected when using standard rather than forcing reaction conditions. This implies an increasing proportion of monodentate and/or bidentate bonding modes of the phosphonate units. Thus, by comparison with the ImPE2 to 4 samples, we noticed a raising in the integration of the signals at 31.6 (18%) and 18.1 (45%) ppm, respectively ascribed to monodentate and bidentate bonding modes (Figure 2), which is in a good agreement with FTIR data. The signal corresponding to tridentate bonding mode at 22.1 ppm became minor with 37% integration. The above results suggest that the grafting of γ-Al2O3 with diethyl phosphonate coupling agent strongly depends on the surface modification reaction conditions. Soft standard reaction conditions mainly promote bidentate and monodentate bonding modes on the surface with a minority of tridentate phosphonate units. Conversely, forcing conditions lead rather to tridentate bonding modes on the alumina surface, with a smaller proportion of the other bonding modes. Figure 5b displays the 31P MAS NMR spectra of γ-Al2O3 grafted under standard conditions (25 °C) using ImTMSP at different reaction times and quantities of IL (two- and six-fold excess). The spectra of ImTMSP1 to 4 samples are qualitatively similar and present a broader peak in comparison with the spectra of γ-alumina grafted with ImPE in forcing reaction conditions, centered at 22.1 ppm. In all cases, the NMR signals present an asymmetrical shape. According to the simulated spectra, three chemical shifts at 25.6, 22.1 and 18.1 ppm were identified with a major resonance for the latter (Table 4). As for the ImPE sample in standard condition with the presence of P-OEt groups, the IR spectra of ImTMSP1 to 4 revealed important residual P-OSiMe3 functions, implying an increasing proportion of monodentate and/or bidentate bonding modes of the phosphonate units. The liquid state 31P NMR spectrum of ImTMSP (Figure S8) indicated an upfield chemical shift at 24.7 ppm (to be compared with 29.8 ppm for pure ImPE (Figure S5)) in good agreement with the P-OEt to P-OSiMe3 conversion [29]. On this basis, the signal at 25.6 ppm, integrating from 9% to 18%, could be tentatively ascribed to the minor monodentate bonding mode with two P-OSiMe3 functions (Figure 2). As for the ImPE1 sample in standard reaction conditions, the major resonances attributed to phosphonate units in bidentate bonding mode correspond to the upfield chemical shifts located at 18.2 ppm and integrating from 47% to 67%. The last signal at 22.1 ppm could be ascribed to tridentate phosphonate units. By using ImTMSP as the coupling agent, we cannot correlate unambiguously the influence of the reaction parameters (excess of coupling agent, reaction duration) with the proportion of the different bonding modes. The above results suggest that the reaction of γ-Al2O3 with either ImPE or ImTMSP in respective standard conditions promotes the grafting of phosphonate units with mainly a bidentate configuration. An increase of the temperature of the grafting reaction with ImPE favors the tridentate bonding mode of the phosphonate units on the γ-Al2O3. 3. Materials and Methods 3.1. Starting Materials Triethyl phosphite (98%), 1-methylimidazole (≥99%) and bromotrimethylsilane (BrSiMe3) (>97%) were purchased from Sigma-Aldrich (Saint-Quentin-Fallavier, France) and were used as received. 1,3-dibromopropane (98%) was provided by Fisher Chemical (Illkirch, France). Boehmite (Pural type) with a high crystallinity and surface area (249 m2/g) was supplied by CTI S.A. (Salindres, France). A batch of γ-Al2O3 powder was prepared by a sol-gel process based on colloid chemistry in aqueous media with a specific surface area of 220 m2·g−1. The γ-alumina powder batch was separated in small samples containing equal amounts of powder (400 mg) into an argon glovebox. This step avoided the presence of water on the alumina surface and yielded comparable conditions for the grafting reactions. ImPE was obtained as a yellow oil in a high yield by the corresponding nucleophilic substitution of 1-methylimidazole with diethyl(3-bromopropyl)phosphonate in tetrahydrofuran (THF) (δ 31P = 29.80 (CDCl3)). ImTMSP was synthetized in a round-bottomed flask by the reaction of the ionic liquid ImPE with BrSiMe3 (3 equiv) in dry methylene chloride (CH2Cl2) (δ 31P = 24.70 (DMSO d6)). THF and CH2Cl2 were provided by Sigma‑Aldrich and dried with the PureSolv, Innovative Technology device. 3.2. Grafting Reactions Standard Reaction Conditions: The “standard” reaction conditions are summarized in Table 1. Typical experiments are described below. Grafting solution with ImPE was prepared by dissolving n-fold excess of the IL in the selected solvent. Five milliliters of the grafting solution and 400 mg of the γ-Al2O3 powder stored under argon were mixed in a glass bottle closed with a Teflon cap. The suspension was heated at 90 °C for 12 days. After cooling down to room temperature, the suspension was then centrifuged at 8500 rpm for 5 min using a Sigma 3-16P centrifuge equipped with a Sigma 12150-H rotor. The supernatant was removed, and the remaining powder was re-dispersed in 5 mL of a (1:1) ethanol-water solution to remove the physisorbed species from the surface, and the new suspension was stirred at room temperature for 5 min. The ethanol-water solution supernatant was removed after centrifugation (8500 rpm, 5 min), and this washing step was repeated twice. The resulting paste was then dried under vacuum (5 to 10 mbar) at 70 °C for ~16 h to afford the sample ImPE1 as a powder. Grafting with ImTMSP was performed directly in the grafting round-bottomed flask in dry CH2Cl2. Typically, ImTMSP (1.2 to 3.6 mmol, corresponding to a 2- or 6-fold excess relative to the amount necessary for a full surface coverage on the γ-Al2O3 particles) was dissolved in 15 mL of dry CH2Cl2 under stirring, and 400 mg of γ-Al2O3 powder stored under argon were dispersed in the grafting solution. The suspension was kept under stirring at 25 °C under argon for time periods ranging from 17 h to 3 days. The suspension was then centrifuged at 8500 rpm for 5 min and the supernatant removed. The remaining paste was re-dispersed in 5 mL of CH2Cl2, and the new suspension was stirred at room temperature for 5 min. After centrifugation, the CH2Cl2 supernatant was removed and the washing step repeated once. Then, the resulting pastes were washed with a (1:1) ethanol-water solution and dried in the same conditions as above, to afford the samples ImTMSP1 to 4 as powders. Forcing Reaction Conditions: The “forcing” reaction conditions are summarized in Table 1. Typical experiments are described below. Grafting ImPE solutions was prepared in water by dissolving the pure ImPE at different proportions (2-, 6- or 12-fold excess). Ten milliliters of the grafting solution and 400 mg of the γ-Al2O3 powder stored under argon were dispersed in an autoclave, which was closed with a Teflon cap. The autoclave was sealed and the suspension heated at 130 °C for 17 h. The resulting grafted powders were washed and dried as previously described for ImPE samples grafted under standard reaction conditions to afford the samples ImPE2 to 4 as powders. 3.3. Characterization The BET specific surface areas and the CBET constants of the samples were obtained from nitrogen adsorption experiments at 77 K by using a Tristar instrument (Micromeritics) for the grafted powders and a ASAP 2020 (Micromeritics) for the γ-Al2O3 powders. Prior to measurements, samples were degassed under vacuum overnight at 100 °C for the grafted powders and 300 °C for the γ-Al2O3 powder. The weight percentage of phosphorus in the samples was determined by EDX using a Zeiss scanning electron microscope (SEM) EVO HD15 at 10 kV equipped for EDX analysis with the AZtecEnergy analysis software (Oxford instruments. Abindong, UK). Samples were prepared as pellets for the analysis and deposited on double-sided carbon tape. FTIR spectra were obtained with a Perkin-Elmer Spectrum 2 spectrophotometer and were recorded in the 4000 to 400 cm−1 range using 32 scans at a nominal resolution of 4 cm−1 in ATR mode (spectrum of γ-Al2O3 as a background spectrum). Solution NMR experiments: 13C and 31P NMR spectra were recorded using a Bruker 300-MHz NMR spectrometer at frequencies of 150.86 and 242.94 MHz, respectively. Solid state NMR experiments: Solid state NMR experiments were performed using a Varian VNMRS 600 MHz (14.1 T) NMR spectrometer. A 3.2-mm Varian T3 HX MAS probe was used for 1H, 13C and 31P experiments. The operating frequencies for 1H, 13C and 31P were 599.95, 150.86 and 242.86 MHz, respectively. All NMR experiments were performed under temperature regulation in order to ensure that the temperature inside the rotor is 20 °C. 13C CP-MAS solid state NMR spectra were recorded at a spinning frequency of 12 kHz MAS. Concerning the CP-MAS experiments, a contact time of 1 ms was fixed, the acquisition time to 30 ms, and the 1H channel was decoupling on this period. A recycle delay of 2 s was used with a number of scans of 11,450, which permit to obtain a signal-to-noise ratio between 30 and 35. 13C chemical shifts were referenced to external adamantane at 38.5 ppm. 31P MAS solid state NMR spectra were recorded at a spinning frequency of 20 kHz. The single pulse experiments were performed with a ~90° solid pulse of 3 μs and 1H decoupling during acquisition. A recycle delay of 45 s was employed (corresponding in both cases to full relaxation of 31P) with a number of scans of 56 for obtaining a signal-to-noise ratio between 53 and 79. 31P chemical shifts were referenced to external hydroxyapatite at 2.80 ppm (used as a secondary reference). 4. Conclusions One important outcome of this study bears on the possibility to perform the grafting with imidazolium bromide-based ILs bearing phosphonate functions (ImPE or ImTMSP) on γ-Al2O3 powders, either in dry methylene chloride solvent (for ImTMSP) or in aqueous and alcoholic solvents (for ImPE). Compared to previous studies published in the literature describing the grafting of phenylphosphonic acid or its bis(trimethylsilyl)ester derivative, no bulk aluminum phosphonate phases were evidenced in the present work. Moreover, this study confirmed that the use of the diethyl imidazolium phosphonate coupling molecule allowed the control of the grafting reaction by using either prolonged heating or high temperature. Surprisingly, the same behavior was demonstrated with the bis(trimethylsilyl)imidazolium ester derivative. FTIR and solid state NMR spectroscopy (31P, 13C) demonstrated that γ-Al2O3 surface modification with diethyl phosphonate coupling agent strongly depends on the grafting conditions. Soft standard reaction conditions mainly promote bidentate and monodentate bonding modes on the surface, with a minority of tridentate phosphonate units. Conversely, the forcing reaction conditions mainly lead to the formation of tridentate bonding modes on the alumina surface, with a smaller proportion of the other ones. In addition, the grafting of ImTMSP in standard conditions seems also to promote the alumina surface modification by phosphonate units in a mostly bidentate configuration. However, in all cases, the surface coverage does not exceed about 30% of the full monolayer. The results obtained in this study for ImPE and ImTMSP could result from a possible sterically-hindered effect on the γ-Al2O3 surface (due to both the alkyl chain and the imidazolium ring) and to the low reactivity of the coupling function. Furthermore, the results obtained with ImTMSP suggest that additional grafting parameters have still to be tested (i.e., grafting duration, concentration, temperature, etc.) in order to optimize the reaction conditions and maximize the γ-Al2O3 surface coverage. This work allowed establishing the optimized synthesis and characterization protocols for the development of imidazolium phosphonate-grafted γ-Al2O3 hybrid materials with controlled bonding modes and grafting rates. As a perspective on this fundamental work, the preparation and testing of the corresponding gas separation hybrid membranes will now be investigated. Acknowledgments Philippe Gaveau (Ingénieur de Recherche CNRS) and Bertrand Ribière (Assistant ingénieur CNRS) from the Institute Charles Gerhardt in Montpellier are sincerely acknowledged for their useful technical contributions in respectively solid state NMR and EDX analysis. Supplementary Materials The synthesis procedures and characterizations of γ-Al2O3 powder, the bulk 1-methyl-3–(3–(diethylphosphinyl)propyl)-imidazolium (ImPE), 1-methyl-3–(3–((trimethoxysilyl)phosphinyl)propyl)-imidazolium bromide (ImTMSP), as well as the details on XRD analysis, the synthesis procedure of the physisorbed sample, FTIR and DFT spectra of the ionic liquids, including DFT detailed calculations. The following are available online at Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1212/s1. Click here for additional data file. Author Contributions Gilles Guerrero and Anne Julbe were the idea source and writers of the manuscript. Marie-Alix Pizzoccaro is the co-writer and responsible for the synthesis and characterizations. Eddy Petit contributed to FTIR spectra attributions and the DFT calculus. Martin Drobek and Peter Hesemann contributed to this article by bringing their scientific expertise, on respectively inorganic (synthesis and characterizations of the γ-Al2O3) and organic chemistry (synthesis of ionic liquids). Conflicts of Interest The authors declare no conflict of interest. Figure 1 Representation of the structures of ImPE (1-methyl-3–(3–(diethylphosphinyl)propyl)-imidazolium bromide) (a) and ImTMSP (1-methyl-3–(3–((trimethoxysilyl)phosphinyl)propyl)-imidazolium bromide) (b). Figure 2 Schematic representation of some possible bonding modes for phosphonate coupling imidazolium-based ionic liquids on a γ-Al2O3 surface (X could be Et or SiMe3). Figure 3 Experimental FTIR spectra of ImPE (Ia,b) and ImTMSP (Ic) and respective grafting on γ-Al2O3 prepared under different reaction conditions: (IIa) with ImPE under standard condition (ImPE1), (IIb) with ImPE under forcing conditions (130 °C, 17 h) at different concentration (ImPE2 to ImPE4: two- to 12-fold excess) and (IIc) with ImTMSP under standard conditions (25 °C, six‑fold excess). Figure 4 13C CP-MAS NMR spectra of pure ionic liquids (I) and the γ-Al2O3 grafted samples (II) prepared under the different reaction conditions. Comparison between: (a) pure ImPE and ImPE4 (130 °C, 17 h, 12-fold excess); (b) pure ImTMSP and ImTMSP4 (six-fold excess, 72 h). Figure 5 31P MAS NMR spectra of γ-Al2O3 grafted samples prepared under different reaction conditions: (a) standard conditions with ImPE (90 °C, 288 h) (ImPE1) and forcing conditions with ImPE (130 °C, 17 h) at different concentrations (ImPE2 to ImPE4: two- to 12-fold excess); (b) standard conditions with ImTMSP (25 °C) at various reaction times and concentrations: twofold excess, ImTMSP1 (17 h); ImTMSP2 (72 h); and six-fold excess, ImTMSP3 (17 h); ImTMSP4 (72 h). ijms-17-01212-t001_Table 1Table 1 Standard and forcing conditions for grafting the γ-alumina powders with organophosphonate imidazolium-based ionic liquids (ILs). IL Solvent (mL) T (°C) Time (h) N-Fold Excess N (ILs) mmol Sample Standard conditions ImPE 2-butanol (5) 90 288 6 3.6 ImPE1 ImTMSP CH2Cl2 (14) 25 17 2 1.2 ImTMSP1 ImTMSP CH2Cl2 (14) 25 72 2 1.2 ImTMSP2 ImTMSP CH2Cl2 (14) 25 17 6 3.6 ImTMSP3 ImTMSP CH2Cl2 (14) 25 72 6 3.6 ImTMSP4 Forcing conditions ImPE water (10) 130 17 2 1.2 ImPE2 ImPE water (10) 130 17 6 3.6 ImPE3 ImPE water (10) 130 17 12 7.2 ImPE4 ijms-17-01212-t002_Table 2Table 2 Characteristics of the γ-alumina powders before and after grafting reactions. Sample CBET wt% P a P nm−2 b γ-Al2O3 82 0 / ImPE1 60 0.90 ± 0.05 0.9 ImPE2 63 0.62 ± 0.02 0.6 ImPE3 59 1.12 ± 0.04 1.1 ImPE4 55 1.42 ± 0.03 1.4 ImTMSP1 70 0.92 ± 0.04 0.9 ImTMSP2 66 0.92 ± 0.02 0.9 ImTMSP3 64 1.16 ± 0.10 1.1 ImTMSP4 63 1.00 ± 0.10 1.0 a From EDX analysis; b average number of coupling molecules per nm2. ijms-17-01212-t003_Table 3Table 3 Parameters used for the simulation of 31P MAS NMR spectra of γ-Al2O3 grafted with ImPE under either standard (ImPE1) or forcing (ImPE2, ImPE3 and ImPE4) reaction conditions. Sample ImPE1 ImPE2 ImPE3 ImPE4 δ (ppm) 31.6 22.1 18.1 32.4 23.6 17.9 32.4 23.6 17.9 32.4 23.6 17.9 Width (ppm) 7.5 7.6 11.5 9.1 6.9 10.0 7.4 7.4 13.0 7.4 8.0 14.4 Integration (%) 18 37 45 13 58 29 7 55 38 7 57 36 ijms-17-01212-t004_Table 4Table 4 Parameters used for the simulation of 31P MAS NMR spectra of γ-Al2O3 grafted with ImTMSP. Sample ImTMSP1 ImTMSP2 ImTMSP3 ImTMSP4 δ (ppm) 25.6 22.1 18.2 25.6 22.1 18.2 25.6 22.1 18.2 25.6 22.1 18.2 Width (ppm) 9.6 5.5 12 9 5.8 12.6 9 4.9 11.2 9 4.9 10.8 Integration (%) 18 35 47 9 30 62 10 23 67 13 29 58 ==== Refs References 1. Dai Z. Noble R.D. Gin D.L. Zhang X. Deng L. Combination of Ionic Liquids with Membrane Technology: A New Approach for CO2 Separation J. Membr. Sci. 2016 497 1 20 10.1016/j.memsci.2015.08.060 2. Nair J.R. Porcarelli L. Bella F. Gerbaldi C. Newly Elaborated Multipurpose Polymer Electrolyte Encompassing RTILs for Smart Energy-Efficient Devices ACS Appl. Mater. Interfaces 2015 7 12961 12971 10.1021/acsami.5b02729 26020809 3. Park H. Choi Y.S. Kim Y. Hong W.H. Song H. 1D and 3D Ionic Liquid–Aluminum Hydroxide Hybrids Preparedvia an Ionothermal Process Adv. Funct. Mater. 2007 17 2411 2418 10.1002/adfm.200600935 4. Porcarelli L. Gerbaldi C. Bella F. Nair J.R. Super Soft All-Ethylene Oxide Polymer Electrolyte for Safe All-Solid Lithium Batteries Sci. Rep. 2016 6 19892 10.1038/srep19892 26791572 5. Zehbe K. Kollosche M. Lardong S. Kelling A. Schilde U. Taubert A. Ionogels Based on Poly(methyl methacrylate) and Metal-Containing Ionic Liquids: Correlation between Structure and Mechanical and Electrical Properties Int. J. Mol. Sci. 2016 17 391 10.3390/ijms17030391 26999112 6. Scovazzo P. Kieft J. Finan D.A. Koval C. Dubois D. Noble R. Gas separations using non-hexafluorophosphate [PF6 ]− anion supported ionic liquid membranes J. Membr. Sci. 2004 238 57 63 10.1016/j.memsci.2004.02.033 7. Albo J. Yoshioka T. Tsuru T. Porous Al2 O3 /TiO2 tubes in combination with 1-ethyl-3-methylimidazolium acetate ionic liquid for CO2 /N2 separation Sep. Purif. Technol 2014 122 440 448 10.1016/j.seppur.2013.11.024 8. Mehnert C.P. Supported ionic liquid catalysis Chem. Eur. J. 2004 11 50 56 10.1002/chem.200400683 15515066 9. Van Doorslaer C. Wahlen J. Mertens P. Binnemans K. de Vos D. Immobilization of molecular catalysts in supported ionic liquid phases Dalton Trans. 2010 39 8377 8390 10.1039/c001285h 20419187 10. Qiu H. Takafuji M. Liu X. Jiang S. Ihara H. Investigation of π–π and ion-dipole interactions on 1-allyl-3-butylimidazolium ionic liquid-modified silica stationary phase in reversed-phase liquid chromatography J. Chromatogr. A 2010 1217 5190 5196 10.1016/j.chroma.2010.06.013 20580007 11. Pino V. Afonso A.M. Surface-bonded ionic liquid stationary phases in high-performance liquid chromatography-A review Anal. Chim. Acta 2012 714 20 37 10.1016/j.aca.2011.11.045 22244134 12. Wang Y. Tian M. Bi W. Row K.H. Application of Ionic Liquids in High Performance Reversed-Phase Chromatography Int. J. Mol. Sci. 2009 10 2591 2610 10.3390/ijms10062591 19582220 13. Perdikaki A.V. Vangeli O.C. Karanikolos G.N. Stefanopoulos K.L. Beltsios K.G. Alexandridis P. Kanellopoulos N.K. Romanos G.E. Ionic Liquid-Modified Porous Materials for Gas Separation and Heterogeneous Catalysis J. Phys. Chem. C 2012 116 16398 16411 10.1021/jp300458s 14. Vangeli O.C. Romanos G.E. Beltsios K.G. Fokas D. Kouvelos E.P. Stefanopoulos K.L. Kanellopoulos N.K. Grafting of imidazolium based ionic liquid on the pore surface of nanoporous materials—Study of physicochemical and thermodynamic properties J. Phys. Chem. B 2010 114 6480 6491 10.1021/jp912205y 20411960 15. Fehrmann R. Haumann M. Riisager A. Introduction Supported Ionic Liquids: Fundamentals and Applications 1st ed. Fehrmann R. Riisager A. Haumann M. Wiley-VCH Verlag GmbH & Co. KGaA Weinheim, Germany 2014 1 9 16. Gadenne B. Hesemann P. Moreau J.J.E. Supported ionic liquids: Ordered mesoporous silicas containing covalently linked ionic species Chem. Commun. 2004 1768 1769 10.1039/b405036c 15278177 17. Motos-Pérez B. Roeser J. Thomas A. Hesemann P. Imidazolium functionalized SBA-15 type silica: Efficient organocatalysts for Henry and cycloaddition reactions Appl. Organomet. Chem. 2013 27 290 299 10.1002/aoc.2974 18. Nguyen T.P. Hesemann P. Moreau J.J.E. I-Silica: Nanostructured silica hybrid materials containing imidazolium groups by hydrolysis-polycondensation of disilylated bis-N,N ′-alkyl-imidazolium halides Microporous Mesoporous Mater. 2011 142 292 300 10.1016/j.micromeso.2010.12.014 19. Zhang Q. Luo J. Wei Y. A silica gel supported dual acidic ionic liquid: An efficient and recyclable heterogeneous catalyst for the one-pot synthesis of amidoalkyl naphthols Green Chem. 2010 12 2246 2254 10.1039/c0gc00472c 20. Kotadia D.A. Soni S.S. Silica gel supported-SO3 H functionalized benzimidazolium based ionic liquid as a mild and effective catalyst for rapid synthesis of 1-amidoalkyl naphthols J. Mol. Catal. A Chem. 2012 353–354 44 49 10.1016/j.molcata.2011.11.003 21. Zhu J.M. Xin F. Sun Y.C. Dong X.C. Phosphonium-based ionic liquids grafted onto silica for CO2 sorption Theor. Found. Chem. Eng. 2014 48 787 792 10.1134/S0040579514060141 22. Julbe A. Farrusseng D. Guizard C. Porous ceramic membranes for catalytic and reactors and overview and new ideas J. Membr. Sci. 2001 181 3 20 10.1016/S0376-7388(00)00375-6 23. Xin B. Hao J. Imidazolium-based ionic liquids grafted on solid surfaces Chem. Soc. Rev. 2014 43 7171 7187 10.1039/C4CS00172A 25000475 24. Alami-Younssi S. Kiefer C. Larbot A. Persin M. Sarrazin J. Grafting γ-alumina microporous membranes by organosilanes: Characterisation by pervaporation J. Membr. Sci. 1998 143 27 36 10.1016/S0376-7388(97)00333-5 25. Leger C. De Lira H.L. Paterson R. Preparation and properties of surface modified ceramic membranes. Part III. Gas permeation of 5 nm alumina membranes modified by trichloro-octadecylsilane J. Membr. Sci. 1996 120 187 195 10.1016/0376-7388(96)00143-3 26. Randon J. Blanc P. Paterson R. Modification of ceramic membrane surfaces using phosphoric acid and alkyl phosphonic acids and its effects on ultrafiltration of BSA protein J. Membr. Sci. 1995 98 119 129 10.1016/0376-7388(94)00183-Y 27. Caro J. Noack M. Kölsch P. Chemically modified ceramic membranes Microporous Mesoporous Mater. 1998 22 321 332 10.1016/S1387-1811(98)00107-3 28. Guerrero G. Mutin P.H. Vioux A. Organically modified aluminas by grafting and sol–gel processes involving phosphonate derivatives J. Mater. Chem. 2001 11 3161 3165 10.1039/b104411g 29. Guerrero G. Mutin P.H. Vioux A. Anchoring of Phosphonate and Phosphinate Coupling Molecules on Titania Particles Chem. Mater. 2001 13 4367 4373 10.1021/cm001253u 30. Mutin P.H. Guerrero G. Almaric J. Preparation of An Inorganic Substrate Having Antimicrobial Properties Patent No. US 8586758 B2 19 11 2013 CNRS (FR); Université Montpellier II (FR) 31. Mu Z. Zhou F. Zhang S. Liang Y. Liu W. Preparation and Characterization of New Phosphonyl-Substitued Imidazolium ionic Liquids Helv. Chim. Acta 2004 87 2549 2555 10.1002/hlca.200490227 32. McKenna C. Schmidhauser J. Functional Selectivity in Phosphonate Ester Dealkylation with Bromotrimethylsilane J. Chem. Soc. Chem. Commun. 1979 17 739 10.1039/c39790000739 33. Leenaars A.F.M. Keizer K. Bruggraaf A.J. The reparation and characterization of alumina membrane with ultra-fine pores J. Mater. Sci. 1984 19 1077 1088 10.1007/BF01120016 34. Guerrero G. Alauzun J.G. Granier M. Laurencin D. Mutin P.H. Phosphonate coupling molecules for the control of surface/interface properties and the synthesis of nanomaterials Dalton Trans. 2013 42 12569 12585 10.1039/c3dt51193f 23887382 35. Mutin P.H. Guerrero G. Vioux A. Hybrid material from organophosphorus Coupling Molecules J. Mater. Chem. 2005 15 3761 3768 10.1039/b505422b 36. Galarneau A. Abid Z. Said B. Didi Y. Szymanska K. Jarzębski A. Tancret F. Hamaizi H. Bengueddach A. Di Renzo F. Synthesis and Textural Characterization of Mesoporous and Meso-/Macroporous Silica Monoliths Obtained by Spinodal Decomposition Inorganics 2016 4 9 10.3390/inorganics4020009 37. Thissen P. Vega A. Peixoto T. Chabal J.Y. Environment-Controlled Tethering by Aggregation and Growth of Phosphonic Acid Monolayers on Silicon Oxide Langmuir 2012 28 8046 8051 22554237 38. Quiñones R. mariea S. Gawalt E. Study of the Formation of Self-Assembled Monolayers on Nitinol Langmuir 2007 23 10123 10130 10.1021/la701110p 17725369 39. Brodard-Severac F. Guerrero G. Maquet J. Florian P. Gervais C. Mutin P.H. High-Field 17 O MAS NMR Investigation of Phosphonic Acid Monolayers on Titania Chem. Mater. 2008 20 5191 5196 10.1021/cm8012683 40. Taoufik M. Szeto K.C. Merle N. Del Rosal I. Maron L. Trébosc J. Tricot G. Gauvin R.M. Delevoye L. Heteronuclear NMR spectroscopy as a surface-selective technique: A unique look at the hydroxyl groups of γ-alumina Chem. Eur. J. 2014 20 4038 4046 10.1002/chem.201304883 24604836 41. Hewitt D.W. Newland G.L. Organophosphorus compounds. P-Arylated perhydro-1,2-azaphosphorines Aust. J. Chem. 1977 30 579 10.1071/CH9770579

PMC005xxxxxx/PMC5000611.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081213ijms-17-01213ReviewAquaporins in Health and Disease: An Overview Focusing on the Gut of Different Species Pelagalli Alessandra 12*Squillacioti Caterina 3Mirabella Nicola 3Meli Rosaria 4Ishibashi Kenichi Academic Editor1 Department of Advanced Biomedical Sciences, University of Naples “Federico II”, Via Pansini 5, 80131 Naples, Italy2 Institute of Biostructures and Bioimages, National Research Council, Via De Amicis 95, 80131 Naples, Italy3 Department of Veterinary Medicine and Animal Productions, University of Naples “Federico II”, Via Veterinaria 1, 80137 Naples, Italy; [email protected] (C.S.); [email protected] (N.M.)4 Department of Pharmacy, University of Naples “Federico II”, Via D. Montesano 49, 80131 Naples, Italy; [email protected]* Correspondence: [email protected]; Tel./Fax: +39-081-7442-09227 7 2016 8 2016 17 8 121318 5 2016 14 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Aquaporins (AQPs) play a pivotal role in gut homeostasis since their distribution and function is modulated both in physiological and in pathophysiological conditions. The transport of water and solutes through gut epithelia is essential for osmoregulation and digestive and absorptive functions. This passage is regulated by different AQP isoforms and characterized by their peculiar distribution in the gastrointestinal tract. To date, AQP localization has been identified in the gut and associated organs of several mammalian species by different techniques (immunohistochemical, western blotting, and RT-PCR). The present review describes the modulation of AQP expression, distribution, and function in gut pathophysiology. At the same time, the comparative description of AQP in animal species sheds light on the full range of AQP functions and the screening of their activity as transport modulators, diagnostic biomarkers, and drug targets. Moreover, the phenotype of knockout mice for several AQPs and their compensatory role and the use of specific AQP inhibitors have been also reviewed. The reported data could be useful to design future research in both basic and clinical fields. aquaporinsgutphysiologypathologyhuman and animal species ==== Body 1. Introduction Since the discovery of aquaporins (AQPs) by Peter Agre in 1992 [1], research in this channel protein field has continuously evolved, resulting in enhanced knowledge of their physiopathological role. Aquaporins show a wide distribution both in different organisms (bacteria, plants, and animals) and tissues, and a selective activity in conducting water molecules in and out of cells and preventing the passage of ions and other solutes [2]. Functionally AQPs are divided into three subfamilies: (a) orthodox AQPs (AQP1, 2, 4, and 5), which are selectively permeable for water; (b) aquaglyceroporins (AQP3, 7, 9, and 10), which are permeable to water as well as to glycerol, urea, and/or other small solutes; and (c) unorthodox aquaporins (AQP6, 8, 11, and 12), with peculiar intracellular localization [3,4] and functions [5,6,7]. Thirteen proteins have been identified, differing in size from 27 kDa (AQP8) to 37 kDa (AQP7), with diverse water permeabilities [8,9]. During the last two decades their role in gut physiology as proteins regulating multiple processes including the transfer of water as well as ions, solutes, and nutrients and feces constitution has attracted particular attention [10]. The study of the specific characteristics of aquaporin vs. glyceroporins is based on their different peculiarities, such as pore gating. This has provided new insights into structure–function relationships, as well as mechanisms of regulation, and into their diverse physiologic roles (for a review, see [11]). AQP organization, demonstrated by crystallographic studies as well, allows these channels to be defined as small intrinsic proteins with a specific permeability established by pore diameter due to the assembly of the different constituents (domains) [12]. According to this concept, AQPs play a major role in gut pathophysiology in the new clinical and therapeutic approaches to several diseases [13]. Large quantities of water molecules may transfer along the epithelia by different pathways (paracellular, transcellular, or both) according to the osmotic gradient resulting from the passive passage of ions and solutes [14]. Water transport occurs from the blood to the lumen or in the opposite direction, allowing other mechanisms related to it, i.e., secretion of hormones or factor release. Moreover, water transfer in gut lumen and its removal, such as feces formation, define a different role and compartmentalization of aquaporins throughout the entire intestinal tract [15]. Here we review the role of AQPs in the gut, focusing on their specific involvement and modification both in physiological and in pathological conditions, with particular reference to the differences between humans and animals. 2. AQP Structure and Distribution in Gut The coordinate passage of molecules and the absorption and secretion of electrolyte and fluids across the intestinal epithelium are important processes for gut homeostasis [16]. In fact, the gut represents an intelligent sensitive organ in continuous communication with the external environment with several regulatory activities due to different specialized immune cells and factors (i.e., nutrients, hormones, microbiota). To date, several intestinal AQPs have been identified [17], although their tissue characterization has not yet been completely defined. Both crystallographic, for the definition of structural features, and immunohistochemical studies have been performed. According to the particular organization of the intestinal tract (small intestine, large intestine, and liver), its morpho-anatomy and its different functions, specific AQPs have been located and water transcellular regulatory activities identified. Notably, AQPs were found in red blood cells for the first time in 1986, and subsequently they were identified as AQP1 [18]. In Table 1 the gut distribution of several AQPs, their particular selective permeability, and cellular localization are summarized and related citations reported. Moreover, in Figure 1 AQP cell polarity is shown, reporting channel protein distribution in the gastrointestinal tract and associated organs. 2.1. Small Intestine The small intestine along its specific tracts (duodenum, jejunum, and ileum) has been intensively investigated, showing the expression of at least nine AQPs (AQP1, 2, 3, 5, 7, 8, 9, 10, and 11) [21,41]. It is well known that the proximal segments of intestine are characterized by osmotic permeability and by secretive activities. In particular, at the duodenum level AQ1, 3, 7, 10, and 11 are mainly expressed. Among these, AQP1 is the best-known and most widely expressed along the capillary endothelium of the ileum mucosa and submucosa [20]. A structural molecular study on AQP1 showed that 50% of the protein is expressed as a glycosylated form by a 5.4 kDa polylactosaminyl oligosaccharide at residue N42 in the first extracellular loop [42]. Immunohistochemical studies evidenced both AQP2 and AQP3 distribution in the entire small intestine, albeit with a more limited tissue expression of AQP2 than of AQP3 [25]. In fact, AQP2 represents one of the less-studied proteins since only one paper reported its expression at the intestinal level, probably indicating its marginal role in this tissue [25]. AQP5 was defined by Parvin et al. [30] as an exocrine-type water channel for its granule secretory activity. It has been well characterized by immunohistochemical and western blotting alongside the rat duodenum, together with AQP1, showing a classic protein profile characterized by a band at 27 kDa (AQP1) and another at 28 kDa (AQP5), and their glycosylated forms (35- to 50-kDa). Research into AQP7 and AQP8 immunolabeling and PCR studies demonstrated their protein expression on epithelial cells of the rat small intestine [43]. Similarly, RT-PCR studies revealed AQP8 mRNA expression not only in the rat jejunum, but also in liver hepatocytes and pancreas acinar cells, demonstrating a peculiar distribution of this AQP [44]. Other data from Elkjær et al. [45] indicate that AQP8 is also distributed in intracellular compartments, suggesting its role in osmotic function between cytoplasma and vescicular compartments in several tissues including small intestine. By contrast, the AQP9 expression has been evidenced by multiple analytical techniques (immunohistochemical, RT-PCR, and western blotting) particularly along the epithelial cells in the ileum and duodenum region [26,35], probably in goblet cells. Moreover, AQP10 has been discovered and identified in the duodenum, jejunum, and ileum of humans [36,37,46]. The authors suggested that water passes the apical membrane of the epithelia principally through AQP10 and partly through AQP8, and the basolateral membrane through AQP3 [46]. Conversely, genetic studies demonstrated that the mouse AQP10 gene contains several mutations that lead to proteins without functional activity [38]. Such data were supported by knockout mice studies leading to the determination of a minimum role of some AQPs in gastro-intestinal (GI) tracts [2,47]. Thus, the absence of AQP10 in mice suggests that this AQP could play a special role in glycerol absorption in humans [47,48]. Recently, AQP11 has been discovered and well characterized in the human brain, showing a lower similarity to other well-known mammalian AQPs and aquaglyceroporins [49]. This protein has been described in the human duodenum [21,31]. 2.2. Large Intestine With respect to the small intestine, the distribution of AQPs in the large intestine also mirrors their involvement in the different processes set up herein. In the large intestine six AQPs have been identified (AQP1, 2, 3, 4, 7, and 8), not always distributed evenly in the same tracts (cecum, colon, and rectum) [21,41]. Their presence has been demonstrated by RT-PCR and immunohistochemical study [45,48]. A study by Laforenza et al. [21] showed that large quantities of AQP1 were expressed in colonic mucosa. Other authors demonstrated that AQP2 is present in the rat distal colon and that its role can be modulated by vasopressin [50]. Furthermore, AQP3 localization has been observed in the large intestine at the level of the stratified and basolateral epithelia in the distal colon [26,28,29,51] and in the rectum, whereas no signal detection was found in the caecum [52]. AQP4 has also been found in humans, albeit with a very low immunoreactive signal [21]. Moreover, functional studies on AQP4-knockout mice demonstrated that AQP4 deletion resulted in the reduction of water permeability in the proximal but not the distal colon, showing its role in transcellular water movement across surface colonocytes. Interestingly, colonocytes play little or no role in fecal dehydration and colonic fluid secretion [53]. AQP7 and AQP8 were also demonstrated to be expressed in the cecum, proximal and distal colon and rectum, albeit with a reduced intensity compared with the small intestine [54]. In particular, studies on the colons of AQP8-knockout mouse revealed their involvement in the regulation of different enzymes implicated in carbohydrate metabolism, using semi-quantitative, fluorescence-stained, two-dimensional gel electrophoresis (2-DE) coupled with nano LC-Ms/Ms [55]. 2.3. Liver Given the close organ association between the intestine and the liver, created by bile, hormones, inflammatory mediators, and products of digestion and absorption [56], the contribution of AQPs in their functions must be considered. In particular, the liver strongly expresses at least six AQPs (AQP1, 3, 7, 8, 9, and 11) [22,39,41,57]. Immunohistochemical studies revealed the expression of AQPs in different hepatic cell types, such as in cholangiocytes (AQP1 and AQP7), endothelial cells (AQP1), Kupffer cells (AQP3), and hepatocytes, (AQP7, 8, and 9) [22]. Recently, Ishibashi et al. [39] reviewed mammalian superaquaporins (AQP11 and AQP12), focusing on their roles, which are only speculated by the phenotypes of their null mutants. For example, the water transport through the superaquaporin inside the cell will be important for the cell organelle function; the facilitated vesicle-to-plasma membrane fusion will be controlled by water transport through vesicular AQPs, as suggested by AQP12 null mice [40]. AQP11 is also expressed in the liver and its knockout produced intracellular vacuoles in the hepatocyte around the portal area, which was more pronounced by fasting in the liver of AQP11 knockout mice [58]. 2.4. Pancreas and Gallbladder In recent years, the involvement of AQPs also in the pancreas and gallbladder has been investigated in relationship to their role in the secretion and reabsorption of water in pancreatic juice and in the formation of gall bladder stones, respectively. As shown in Table 1, different AQPs have been also identified in the pancreas (AQP1, 5, 8, and 12) [23,33]. Immunohistochemical and functional studies have partially clarified their physiological role. In particular, AQP1 was strongly expressed in centroacinar cells and both in apical and basolateral domains of intercalated and intralobular duct epithelia. Moreover, AQP5 was observed in the apical membrane of intercalated duct cells and in the duct-associated mucoid glands [23]. Differently, AQP8 immunoreactivity was shown in the apical plasma membrane domains of human acinar cells near the zymogen granules [23], confirming previous data reported in rats [33]. Immunohistochemical analysis revealed AQP12 staining at the basal side of the intracellular organelles of acinar cells close to the nucleus but not in either the duct cells or the islet cells [40]. The presence and specific distribution of AQPs in gallbladder have been investigated using immunohistochemical analysis and protein analysis demonstrating their localization in cell plasma membranes and intracellular vesicles of the gallbladder epithelium of humans and mice [59,60]. In particular, immunofluorescence and immunohistochemical studies showed strong AQP1 and AQP8 signal at the apical membrane of the mouse gallbladder epithelium [24,34]. Moreover, other data show a similar distribution in the gallbladders of wild-type and AQP1 null mice, with comparable epithelial thickness and cell density [24]. 3. Gut AQP Function The function of the digestive system is a complex of metabolic reactions involving the small and large intestine as well as the liver, pancreas, and gallbladder. Indeed, along the intestinal wall, the presence of different specialized cells and that of gut microbiota allow multiple processes to be set in train that terminate with the production of feces. Finally, gut function comprises not only nutrient absorption and secretion, but also processes such as homeostasis, regulation of resistance to disease, and production of factors involved in cell growth and repair [16]. In recent decades, the presence of different AQPs distributed through the gut has been convincingly demonstrated. These studies conducted in humans and rodents highlighted an interesting aspect regarding their possible involvement in different gut processes, namely participation in the plasticity and adaptability of the gut in relation to diet. In normal conditions in a human, about 1.5–2 L of water is absorbed daily by the colon, while the maximal capacity of the intestine to absorb fluids may be as high as 5–6 L per day [61]. The exact mechanism by which fluids are transported in the epithelia of the gastrointestinal tract has largely focused on whether water passes through cells (transcellular) or between cells (paracellular). According to the majority of the studies, the transport at the level of gastrointestinal tract is in most cases transcellular [62], even if it is still believed that absorption of water in the small intestine occurs primarily in a paracellular manner [63] or by cotransporters. The importance of the paracellular passage is also supported by research data obtained in AQP5 knockout mice, reporting that in these animals the decrease in transcellular water transport in parotid glands can be associated with an increase of paracellular permeability to ions [64,65]. A significant decrease in tight junction proteins, claudin-7, and occludin was observed in the AQP5−/− knockout mice, even if the molecular pathways remain elusive, these studies indicated that AQP5 could function to link paracellular and transcellular pathways. On the other hand, the hypothesis of water transport via co-transporters has been demonstrated for water uptake in intestinal epithelial cells which, as reported by the authors, have a very low expression of AQPs [66]. In the duodenum, water is secreted and its transport is affected by gastrointestinal hormones (i.e., gastrin, vasoactive intestinal peptide (VIP), and others), neurotransmitters, and histamine, which contribute to water and electrolyte balance [30]. Most of the intestinal water is absorbed in an isosmotic fashion by the small intestine and only in part by the large intestine, according to their different electrical resistance characteristics—as demonstrated for AQP7, which is involved in the physiological mechanisms of fluid absorption and secretion [54]. The discovery of a specific water channel and highly conserved AQPs on epithelial cells in the gastrointestinal tract have identified their role in rapid water movements. However, it is important to consider the cell structure and function along the different tracts in order to examine the functional activity of the different AQPs. According to their transepithelial electrical resistance characteristics, the epithelia from different portions of the intestinal tract may be classified into three categories: leaky (i.e., small intestine), moderately tight (i.e., colon and gastric antrum), and tight (i.e., gastric fundus) [67]. In particular, AQP1 has been demonstrated to be associated with AQP5 (in the duodenum, as well as in the pancreas) [23,30], with an activity probably related to water secretion [30]. Characterization of the columnar absorptive cells in the distal colon demonstrated that these cells express the three subunits of the epithelial Na+ channel in the apical membrane [68] and Na+-K+-ATPase in the basolateral membrane in a similar way to what occurs in the kidney [69]. The renin angiotensin aldosterone system plays a pivotal role in the regulation of water and sodium reabsorption in the kidneys. In particular, vasopressin plays a hormonal function in the mechanism of water homeostasis acting through AQPs. In particular, AQP2 is regulated by antidiuretic hormone (ADH), when plasma osmolarity increases thanks to the endings of magnocellular neurons in the posterior pituitary [69]. AQP1 is expressed in dietary fat processing [70], including cholangiocytes in the liver regulating bile production and pancreatic microvascular endothelium, where it plays a role in pancreatic fluid production [71]. Its activity has been demonstrated also in bile and pancreatic juice, thanks to the evidence that malabsorption problems were observed in AQP1 null-mice [70]. By contrast, AQP3 activity, well known in both the oral cavity and stomach, has also been evidenced in the distal colon: it is expressed and localized along the epithelial cells, at the level of lumen and crypts, suggesting its importance in water transport to the cells involved in the formation of intestinal contents and feces [52]. This data has also been confirmed by studies using HgCl2 as an AQP3 inhibitor, used to investigate the role of AQP3 in the regulation of fecal water content [72]. A proliferative activity for this AQP has been demonstrated in enterocytes, evidencing its possible therapeutic use in Crohn’s disease [7]. The role of AQP3 would appear to be related to transcellular water reabsorption, i.e., water transfer from the lumen to the interstitium according to the osmotic gradient [73,74]. AQP7 and AQP8, expressed along the large intestine, seem to play a role in water trafficking from lumen to the interstitium by a transcellular route [32,44]. It is particularly evident for AQP8, confirmed also by higher protein content in intracellular vesicles [45]. Functional studies have demonstrated that both the AQPs’ distribution patterns and expression levels could be modulated by feeding conditions, food preference, and developmental regulation. In particular, a correlation between protein expression and feeding was only observed for AQP6, showing that this protein is upregulated by feeding [75]. In another study conducted in European sea bass (Dicentrarchus labrax), short- and long-term fasting influenced metabolic activities and liver AQP9 expression, suggesting that nutritional status could modulate AQP’s role in hepatic glycerol uptake [76]. In addition, in our previous study conducted on buffalo calves fed with colostrum for one week, a different pattern of AQP1, 4, and 5 expression was observed with respect to the control group fed milk. These data indicate that a specific regulation of AQP function and distribution could be achieved according to the nutritional level or the food preference [77,78,79]. 4. Liver AQP Function The liver is a well-known metabolic organ sensible to nutrients and hormones. The involvement of AQPs in the liver must also be considered on the basis of their identification and different distribution in several hepatic cell types, as recently demonstrated [22,41]. The cited authors have charted in the human liver and in both human and mouse hepatocytes the presence of several AQPs providing interesting gene regulation by known drugs/hormones (i.e., dexamethasone, forskolin, rosiglitazone, and ghrelin). Recently, Laforenza et al. [41] have clarified the relationship between liver aquaglyceroporin expression and adipose regulation. The authors evidence the important AQP role in physiological conditions, in obesity and type 2 diabetes, identifying these proteins as potential therapeutic targets for metabolic disorders. The AQP9 hepatic role has been extensively studied, confirming its involvement in glycerol metabolism as well as in gluconeogenesis [80,81,82]. In particular, some studies have also clarified the role of protein deletion or protein functional modification using genetic model or specific inhibitors. For example, in livers of female AQP9+/+ mice, the immunohistochemical protein signal was more intense in hepatocytes close to the central vein (perivenous hepatocytes), whereas in male mice the label appeared to be more uniformly distributed in all the hepatocytes. As expected, AQP9−/− mice did not show any immunostaining [83]. In the meantime, studies on primary hepatocyte culture treated with a specific novel AQP inhibitor, HTS13286, revealed the metabolic importance of t AQP9, which plays a role in the control of glycerol uptake after hyperglycemia induction [84]. However, additional studies will be required to understand the transcriptional regulation of AQPs in the liver under pathophysiological conditions. Recently, new data and knowledge regarding AQP involvement in bile formation as well as in the development of bile secretory failure were reviewed [85]. During active choleresis, hepatocytes increase vesicle trafficking and bile passage, enhancing canalicular membrane water permeability. AQP8 modulates membrane water permeability, providing a molecular mechanism for the osmotically-coupled transport of solute and water during bile formation. Cholestasis is a pathologic condition defined as an impairment of normal bile formation, which may or may not be associated with bile flow obstruction. Estrogens are known to cause intrahepatic cholestasis in women during pregnancy and many drugs, such as oral contraceptives or postmenopausal replacement therapy, can induce this disorder [85]. A murine model of cholestasis, induced by 17α-ethinylestradiol (EE), has been used to investigate alterations in the expression of hepatocyte membrane transporters in this pathology. Our data reported in Figure 2A,B show a clear reduction in AQP8 expression in this experimental model. Generally speaking, in the liver, as in the other tissues, the regulation in AQP activity must take into account their structure at the C-terminal level for the interaction with other proteins, as well as the Ca2+-binding sites, an N-terminal conformational switch, and trafficking in the inner molecule. It must be added that some AQPs require an activation mechanism for their permeability regulation [86]. 5. Pancreas and Gallbladder AQP Functions Different studies have been performed to evaluate the exact functional contribution of AQPs in the pancreas and gallbladder to gain better comprehension of their role in pathological conditions. Based on limited published data, it is possible to confirm that water transport in the pancreas occurs both by paracellular and transcellular pathways [14]. In particular, the evaluation of AQPs’ role as water channels in the osmotic permeability of the acinar cell membrane was examined with the help of an Hg2+ inhibitor, which is known to effectively block most of the AQP isoforms, including AQP8 [87]. In particular, the treatment of the pancreas secretory vesicles with this inhibitor caused vesicle swelling, confirming the involvement of AQP1 in rapid gating of water in zymogen granules [88]. In the meantime, studies with knock-out (KO) mice (for AQP1 and AQP8) showed that pancreatic secretion is not significantly affected by AQP1 deletion [70]. Similarly, in another study, AQP12 deletion did not affect the overall pancreatic exocrine function in mice under a normal breeding environment, while AQP12-KO mice showed a more severe pathology resulting from CCK-8 analog-induced pancreatitis than wild type (WT) mice [40]. The confirmation of the role of some AQPs in pancreas exocrine function has been obtained by studies on liver X receptors (LXRs) β−/− mice, where a pancreatic exocrine insufficiency has been associated with a reduction in AQP1 expression [89]. Regarding the gallbladder function, different studies revealed the presence of AQP1 and AQP8, confirming their role as a protein directly involved in water transport across the apical membrane of gallbladder epithelium [24,34]. Moreover, a study on gallbladder AQP1-deficient mice demonstrated a strong reduction of water permeability in these animals, indicating that AQP1 provides the principal route for osmotic water transport by a transcellular rather than a paracellular pathway [24]. Functional gallbladder studies revealed changes of AQP expression and the absorptive function of this organ in mice fed a lithogenic diet, suggesting their involvement in water and electrolyte transport [90]. 6. AQPs in Gut Pathophysiology Altered expression of AQPs have been identified as co-factors in the etiopathogenesis of some gastroenteric disorders [91,92]. Diarrhea represents a common pathology and is widely studied using different animal models. It is characterized by two important conditions: (1) transepithelial hypersecretion of fluid in the gastrointestinal (GI) tract and (2) defects in water absorption in the colon. Both are important factors that suggest the definite involvement of AQPs. In a model of attaching and effacing pathogen-induced diarrhea, an evident alteration in AQP distribution (especially AQP2 and AQP3) has been evidenced in colonocytes [92]. In the same way, AQP4 and AQP8 are shown to decrease significantly in a mouse model of colitis after exposure to dextran sodium sulfate (DSS), confirming data obtained from clinical investigations in inflammatory bowel disease (IBD) patients [91]. Further studies in 2008 [93] and in 2010 [94] clarified that the decrease in AQP expression in enterocyte membranes might be the cause of restricted water re-absorption, leading to diarrhea generation. Modification of the AQP pattern seems to be attributed to a translocation mechanism mediated, at least in part, by EspF and EspGt [95]. Recent studies on inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), have suggested a particular role played by AQPs [96]. Aquaporin expression, especially AQP4, 7, and 8, was examined in a murine model of colitis and in patients with IBD or infection colitis [89]. The expression of AQP4 and AQP8 mRNA and protein was slight but significantly decreased, while AQP7 was more variable. Colitis can be aggravated by stress, like sleep deprivation, and improved by anti-inflammatory agents including melatonin [97]. In microarrays and real-time PCR on the mouse colon, mRNA of adiponectin and AQP8 were downregulated by sleep deprivation and upregulated by melatonin. Our unpublished data obtained on DSS-induced colitis in mice show a slight but significant reduction of AQP1 expression in cecum tracts (data not shown), determined using an antibody that recognized a specific band at 28 kDa and two other bands at 55 and 65 kDa, respectively. These last two bands, as previously evidenced by other authors in the avian small intestine, could presumably be associated to glycosylated forms [98]. At the same time, pharmacological treatment with sodium butyrate, a short-term fatty acid (SCFA) that is considered a fuel for intestinal epithelial cells, reverted the pathologic conditions, reducing the effect of colitis by normalizing AQP1 expression. As is well recognized, this SCFA modulates different processes in the gastrointestinal tract such as electrolyte and water absorption [99]. These data for the first time confirm the involvement of AQP1 in colitis, suggesting further studies to clarify the mechanisms involved in AQP dysfunction. These findings provide new information regarding the role of these membrane channels in IBD, focusing on their dysregulation in gut permeability and bidirectional water/glycerol transit induced by chronic inflammation. A role for AQPs was also investigated in celiac disease, another intestinal inflammatory disorder induced in genetically susceptible subjects by gluten ingestion [31]. Studies in this regard have demonstrated a dramatic reduction in both AQP mRNA and protein expression combined with a reduced activity of principal solute transporters in the villus (i.e., SGLT1, PEPT1, and NHE3). Moreover, in a model of allergic diarrhea, a disease resulting in immunological and microbiological homeostasis alterations, decreased expression of AQP4 and AQP8 was observed [100]. These data provide insights into the intestinal role of AQPs, defining a new mechanism to wash the food allergen out of the gastrointestinal tract. AQPs have also been studied in other diseases where fluid flux alterations may contribute to increased susceptibility to injury in the small intestine (i.e., shock after early ischemic injury) [101], introducing a new concept of the intestinal mucosa barrier. Accordingly, AQP3 expression alterations could be associated with modifications of transcellular and paracellular water transport causing intestinal and endotoxin translocation [102]. 7. AQP Expression and Distribution in the Gut of Other Species 7.1. Small Intestine To date, several aquaporin isoforms have been identified in the gut of different animal species. Expression and localization of aquaporin isoforms in the digestive system are summarized in Table 2. AQP1-immunoreactivity (IR) was found in endothelial cells of capillaries, small vessels, and central lacteals in the villi of the small intestine in the pig, rat, mouse, and buffalo [10,26,60,70,77,103,104]. In addition, AQP1-IR was also found in the enterocytes of the crypts in the buffalo small intestine [77], in the apical and basolateral membranes of Brunner’s gland cells in the rat duodenum [30], and in enteric neurons of the buffalo, rat, and sheep small intestine [77,105,106]. RT-PCR and Northern blot analysis confirmed AQP1 mRNA expression in the small intestine of buffalo, pigs, and rodents [10,77,103]. Moreover, AQP3-IR has been found only in the basolateral membrane of the epithelial cells in the villous tip of the rat small intestine [26,107,108], while in humans AQP3-IR is localized also in the enterocytes, goblet cells, and Paneth cells of the crypts [20,25,31]. AQP3 mRNA expression, using Northern blot, in situ hybridization analysis, and real-time RT-PCR, was found in the rat small intestine [107,108,109], suggesting a specific role for AQP3 as a water output modulator after an absorption process via the transcellular pathway [107], which involves SGLT1 or a combination of other transporters. AQP4-IR is distributed at the basolateral membrane of the cryptic cells located at the bottom of the crypt of the rat, guinea pig, buffalo, and porcine small intestine [10,78,110,111]. AQP4 was also found in the enterocytes along villi and in the Brunner glands [78,110,111] and revealed by Western blot analysis in the rat, guinea pig, mouse, and buffalo small intestine [78,104,109,110]. AQP4 mRNA expression was confirmed in the rat and buffalo small intestine [78,109]. In the rat duodenum, AQP5-IR is present in the apical and lateral membranes of Brunner’s gland secretory epithelium [19,112] and increases in the apical membranes of these cells after stimulation by vasoactive intestinal peptide [30]. Rat AQP5 mRNA has also been found in the duodenum [19]. The presence of AQP5 in the proximal region of the small intestine suggests its secretory role induced by several mediators [113]. Additionally, a study on the gastrointestinal tract of the chicken demonstrated the presence of ck-AQP5 in the crypt cells of the jejunum, ileum, and rectum, while it was found to be absent in the epithelial cells lining the villi. Abundance of ck-AQP5 mRNA and protein was higher in the jejunum, decreasing towards the colon [114]. In addition, AQP5-IR and mRNA were found in the enterocytes and endocrine cells of the buffalo small intestine [78]. In the rat small intestine, AQP6-IR and AQP7-IR as well as proteins and relative mRNAs were present in the apical part of the epithelial cells of the villus [54,75]. Moreover, AQP8-IR was localized in the subapical site of rat duodenum and proximal jejunum enterocytes [44,45,108,115]. On the contrary, in the human duodenum AQP8 does not seem to be expressed [31] although it was found in the human ileum [96]. Using semi-quantitative and real-time RT-PCR, AQP8 was detected in the rat duodenum, proximal jejunum [115], and ileum [108,109]. AQP9-IR was also localized in the goblet cell basolateral membrane of the rat small intestine [35]. Recently, a study on bovine and ovine duodenum showed that AQP10 gene was a pseudogene in these species [116], confirming a previous study conducted on humans and mice [38]. Pseudogenes are generally produced by gene duplications, which preserve the original function or could acquire a new function to survive in an unsafe environment in the form of pseudogenes. The authors suggest that in the case of bovine AQP10 pseudogene its function could be compensated for by other aquaglyceroporins. 7.2. Large Intestine The AQP1 isoform is localized in the capillary and small vessel endothelium of the rat and buffalo large intestine [26,77] and in the enterocytes of the buffalo [77]. Among birds, AQP1 has been found in the lower intestinal tract of the Passer domesticus; in particular, its epithelial distribution was limited to the distal rectum [98]. In the rat colon, AQP2-IR was localized in apical membrane epithelium, and RT-PCR, in situ hybridization, immunoblotting, and immunocytochemistry confirmed its expression in colonic crypts and, to a lesser extent, in surface absorptive epithelial cells [50]. Along the rat large intestine AQP3-IR was evidenced in the distal colon and rectum and localized at the basolateral membrane of the absorptive epithelial cells directly facing the lumen and at the neck of crypts [26]. In addition, in the rat proximal and distal colon, AQP3 protein and mRNA expression were also detected [27,107,108,109]. AQP4-IR was localized on the basolateral membrane of epithelial cells isolated from the rat, buffalo, and porcine colon [53,79,111,119]. In addition, AQP4 expression was found in the endothelium and enteric neurons of the buffalo and rodent colon [79,119]. This localization suggests the involvement of the enteric nervous system in body fluid homeostasis by monitoring changes in osmotic pressure and controlling water movement across the mucosa. The AQP4 protein and relative mRNA were also revealed in the buffalo and rat large intestine [79,109]. AQP5-IR, protein, and relative mRNA were shown in the endocrine cells of the buffalo large intestine [79] while AQP7-IR was distributed in the surface epithelial cells of the crypt of the rat colon and cecum [54,109]. Other data gave additional information about the importance of retrograde peristalsis for water conservation in murine species. In particular, AQP8-IR was localized in the subapical site of the absorptive epithelial cells of the rat colon, suggesting its role in fecal dehydration [43,45,115,123]. AQP8 staining was also observed in the intracellular compartment of the surface epithelial cells of rat proximal colon and rectum [108,115]. Moreover, in rat AQP8 mRNA was detected in the proximal and distal colon, rectum, pancreas, and liver [108,115] and AQP9 was localized at the basolateral membrane of the goblet cells [35]. 7.3. Liver, Pancreas, and Gallbladder Some AQPs were also expressed in organs anatomically and physiologically related to the gut. In particular, AQP1, 8, and 9 were evidenced in the liver and pancreas of the rat and pig [26,117]. AQP1 was also localized in the capillary and small vessel endothelium of the liver and pancreas of the rat [26], while AQP9 showed different tissue distribution between rats and pigs. This different tissue distribution could suggest for AQP9 a specific role according to the animal species, as already observed in humans and rat [124]. In the rat, AQP8 was detected in the apical membrane and cytoplasm of hepatocytes [44,45,115,117,120] and in the apical regions of pancreatic acinar cells [33,44]. AQP9 was localized in sinusoidal surface membranes of the rat and pig hepatocytes [45,121,122,125]. As reported in Table 1 and below, there are few studies in human and mice that reported AQP expression and role, while other species are overlooked. 8. Conclusions The detection of AQPs along the gut and their widespread distribution in human and animal species suggest their role in maintaining fluid homeostasis. Recent data on their involvement in the pathologies of digestive tract have further defined their mechanisms of action and possible applications in therapy using specific modulators. In recent years, new knowledge regarding barrier integrity and how different conditions (trauma, shock, etc.) can induce multiple intestinal reactions (intestinal cytokine response, translocation of intestinal bacteria, systemic inflammatory response syndrome) has stimulated research in the field of AQPs. In the meantime, new research approaches based either on genetic studies by the use of knockout mice and functional studies by the use of specific AQPs inhibitors have opened a pathway to new possibilities for clinical therapy. In particular, AQPs inhibitors could be used as probes to assess their function in several disease models, and without the need for RNA silencing or knockout models, which have additional limitations and drawbacks, such as adaptive changes in phenotypes. Pharmacological inhibition of AQP water permeability in epithelia, with consequent reduced fluid transport, was also reviewed [126], suggesting AQP modulation as a potential therapeutic target for human diseases involving water imbalance such as congestive heart failure, hypertension, and glaucoma. Even if this research area is underdeveloped, two recent reviews by Verkman et al. (2014) [13] and Beitz et al. (2015) [127] provide an overview on AQP-related disorders and pharmacological intervention in the therapeutic modulation of aquaporin functionality, identifying protein structural and chemical aspects of AQP modulator design. Indeed, the modulation of AQP functions is desirable in other several pathophysiological situations not only in the gut, such as cancer, heart failure, nephrogenic diabetes insipidus, and Sjögren’s syndrome. Acknowledgments The work was supported in part by research funds from the University of Naples “Federico II”, Naples, Italy. Author Contributions All authors contributed equally to this work. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Main AQPs’ cell polarity and channel protein distribution in the gastrointestinal tract and associated organs. (?) = unknown role or function. For the specific intracellular localization of different AQPs see text. Figure 2 (A) AQP8 protein expression in the liver of untreated (CON, control animals receiving propylene glycol as drug vehicle, s.c.) and cholestatic mice (EE). Cholestasis was induced by ethinyl estradiol (5 mg/kg/die for five days, s.c.). Male 10-week-old BALB/c mice were killed at day 5, 1 h after drug treatment; (B) Representative image of densitometric analysis of AQP8 protein band (28 kDa) is shown (n = 6). All data are expressed as mean ± SEM. Equal loading was confirmed by β-actin staining. Statistical analysis was performed by Student’s t-test analysis. * p < 0.05 vs. control group. ijms-17-01213-t001_Table 1Table 1 Gut distribution, selective permeability, and tissue localization of AQPs in human and murine tissues. AQP Isoform Selective Permeability Gut Distribution Cellular Localization References Small Intestine Large Intestine Liver Pancreas Gallbladder Duodenum Jejunum Ileum Caecum Colon AQP1 H2O + n.d. + + + + + + capillary endothelium of the mucosa and submucosa (duodenum, ileum), crypt epithelium (colon), acinar cells (pancreas), apical and intercellular membrane epithelium (gallbladder) Matsuzaki, T., 2003 [19]; Mobasheri, A., 2004 [20]; Laforenza, U., 2012 [21]; Gregoire, F., 2015 [22]; Burghardt. B, 2003, [23]; Li, L., 2009 [24] AQP2 H2O + + + n.d. n.d. n.d. n.d. n.d. mucosa Mobasheri, A., 2005 [25] AQP3 H2O, glycerol, urea + + + n.d. + + n.d. n.d. basolateral membrane of epithelial cells at luminal surface (intestine), Kupffer cells, and hepatocytes Matsuzaki, T., 2004 [26]; Ishibashi, K., 1995 [27]; Silberstein, C., 1999 [28]; Ikarashi, N., 2013 [29]; Mobasheri, A., 2005 [25]; Laforenza, U., 2012 [21]; Gregoire, F., 2015 [22] AQP4 H2O + n.d. n.d. n.d. + n.d. n.d. n.d. Basolateral membrane of crypt cells and of epithelial cells at luminal surface Matsuzaki, T., 2004 [26] AQP5 H2O + n.d. n.d. n.d. n.d. n.d. + n.d. apical membrane of duodenal gland secretory cells, intercalated duct cells (pancreas) Matsuzaki, T., 2004 [26]; Parvin, M.N., 2005 [30]; Burghardt. B, 2003 [23] AQP7 H2O, glycerol, urea + + + n.d. + n.d. n.d. n.d. Superficial epithelial cells Laforenza, U., 2010, 2012 [21,31]; Gregoire, F., 2015 [22] AQP8 H2O n.d. n.d. n.d. + + + + + subapical site of absorptive epithelial cells, hepatocytes and cholangiocytes, apical region of acinar cells (pancreas), apical membrane epithelium (gallbladder) Laforenza, U., 2010, 2012 [21,31]; Matsuzaki, T., 2004 [26]; Fisher, H., 2001 [32]; Koyama, Y., 1999 [10]; Gregoire, F., 2015 [22]; Burghardt. B, 2003 [23]; Hurley, P.T., 2001 [33]; Ambe, P.C., 2016 [34] AQP9 H2O, glycerol, urea + n.d. + n.d. n.d. + n.d. n.d. basolateral membrane of goblet cells, cholangiocytes Matsuzaki, T., 2004 [26]; Okada, S., 2003 [35]; Gregoire, F., 2015 [22]; Mobasheri, A., 2004 [36] AQP10 H2O, glycerol, urea + + + n.d. n.d. n.d. n.d. n.d. enterocytes, gastroenterohepatic (GEP) endocrine cells, and pseudogenes in mice Li, H., 2005 [37]; Morinaga, T., 2002 [38] AQP11 Unknown + + + n.d. + + n.d. n.d. enterocytes, hepatocytes Laforenza, U., 2010, 2012 [21,31]; Ishibashi, K., 2014 [39] AQP12 Unknown n.d. n.d. n.d. n.d. n.d. n.d. + n.d. acinar cells (pancreas) Ohta, E., 2009 [40] n.d. = not determined; + presence. ijms-17-01213-t002_Table 2Table 2 Gut distribution of AQPs in different animal species (n.d. = not determined; + presence). AQP Isoform Animal Species Gut Distribution Cellular Localization References Small Intestine Large Intestine Liver AQP1 Rat + + + endothelium of capillaries, small vessels and lacteals of the villi; apical and basolateral membrane of cells of Brunner’s gland ; enteric neurons; hepatic sinusoids Nielsen, S., 1993 [60]; Koyama, Y., 1999 [10]; Matsuzaki, T., 2004 [26]; Parvin, M.N., 2005 [30]; Nagahama, M., 2006 [105]; Talbot, N.C., 2003 [117] Mouse + n.d. n.d. endothelium of capillaries and small vessels Ma, T., 2001 [70] Buffalo + + n.d. endothelium of capillaries and small vessels; enterocytes of the crypts; enteric neurons De Luca, A., 2015 [77] Pig + n.d. + endothelium of lacteals of the villi, liver bile duct Jin, S.Y., 2006 [103]; Talbot, N.C., 2003 [117] Sheep + n.d. n.d. enteric neurons Arciszewski, M.B., 2011 [106] Birds (Passer domesticus) n.d. + n.d. epithelial cells of the distal rectum Casotti, G., 2007 [98] AQP2 Rat n.d. + n.d. apical membrane of surface columnar epithelial cells Gallardo, P., 2001 [50] AQP3 Rat + + n.d. basolateral membrane of the epithelial cells in the villous tip Matsuzaki, T., 1999,2004 [26,52]; Ishibashi, K., 1995 [27]; Ramirez-Lorca, R., 1999 [107]; Zhao, G.X., 2016 [108] AQP4 Rat + + n.d. basolateral membrane of the cryptic cells and surface of colon epithelial cells; enteric neurons of the colon Koyama, Y., 1999 [10]; Frigeri, A., 1995 [118]; Wang, K.S., 2000 [53]; Thi, M.M., 2008 [119] Mouse + + n.d. ileal and colon mucosa Cao, M., 2014 [104] Buffalo + + n.d. enterocytes of the crypts; endothelium Squillacioti, C., 2015 [78]; Pelagalli, A., 2015 [79] Pig + + n.d. enterocytes along the villi and in the bottom of the crypts and in the Brunner’s gland Arciszewski, M.B., 2015 [111] Guinea pig + n.d. n.d. enterocytes of the crypts Jiang, L., 2014 [110] AQP5 Rat + n.d. n.d. apical and lateral membranes of the secretory cells of Brunner’s gland Parvin, M.N., 2005 [30]; Matsuzaki, T., 2004 [26] Chicken + + n.d. enterocytes of the crypts Ramirez-Lorca, R., 2006 [114] Buffalo + + n.d. enterocytes of the crypts; endocrine cells Squillacioti, C., 2015 [78]; Pelagalli, A., 2015 [79] AQP6 Rat + n.d. n.d. apical region of the enterocytes in the villi Laforenza, U., 2009 [75] AQP7 Rat + + n.d. apical region of the enterocytes in the villi; epithelial cells of the colon and caecum Laforenza, U., 2005 [54] AQP8 Rat + + + apical region of the enterocytes in the villi and of the epithelial cells of the colon; hepatocytes Calamita, G., 2001 [115]; Elkejer, M.L., 2001 [45]; Tani, T., 2001 [44]; Garcia, F., 2001 [120]; Huebert, R.C., 2002 [57] AQP9 Rat + + + basolateral membrane of the goblet cells, hepatocytes Okada, S., 2003 [35]; Talbot, N.C., 2003 [117]; Nicchia, G.P., 2001 [121]; Huebert, R.C., 2002 [57] Pig n.d. n.d. + hepatocytes Talbot, N.C., 2003 [117]; Caperna, T.J., 2007 [122] ==== Refs References 1. Agre P. Sasaki S. Chrispeels J. Aquaporins: A family of water channel proteins Am. J. Physiol. Ren. Physiol. 1993 265 F461 2. Verkman A.S. Mitra A.K. Structure and function of aquaporin water channels Am. J. Physiol. Ren. Physiol. 2000 278 F13 F28 3. Heymann J.B. Engel A. Aquaporins: Phylogeny, structure, and physiology of water channels News Physiol. Sci. 1999 14 187 193 11390849 4. Takata K. Matsuzaki T. Tajika Y. Aquaporins: Water channel proteins of the cell membrane Prog. Histochem. Cytochem. 2004 39 1 83 10.1016/j.proghi.2004.03.001 15242101 5. Agre P. King L. Yasui M. Guggino W. Ottersen O. Fujiyoshi Y. Engel A. Nielsen S. Aquaporin water channels—From atomic structure to clinical medicine J. Physiol. 2002 542 3 16 10.1113/jphysiol.2002.020818 12096044 6. Ishibashi K. Aquaporin superfamily with unusual NPA boxes: S -aquaporins (superfamily, sip-like and subcellular-aquaporins) Cell. Mol. Biol. 2006 52 20 27 17543217 7. Rojek A. Praetorius J. Frøkiaer J. Nielsen S. Fenton R.A. A current view of the mammalian aquaglyceroporins Annu. Rev. Physiol. 2008 70 301 327 10.1146/annurev.physiol.70.113006.100452 17961083 8. Kreida S. Tornroth-Horsefield S. Structural insights into aquaporin selectivity and regulation Curr. Opin. Struct. Biol. 2015 33 126 134 10.1016/j.sbi.2015.08.004 26342685 9. Kitchen P. Day R.E. Salman M.M. Conner M.T. Bill R.M. Conner A.C. Beyond water homeostasis: Diverse functional roles of mammalian aquaporins Biochim. Biophys. Acta 2015 1850 2410 2421 10.1016/j.bbagen.2015.08.023 26365508 10. Koyama Y. Yamamoto T. Tani T. Nihei K. Kondo D. Funaki H. Yaoita E. Kawasaki K. Sato N. Hatakeyama K. Expression and localization of aquaporins in rat gastrointestinal tract Am. J. Physiol. 1999 276 C621 C627 10069989 11. Krane C.M. Goldstein D.L. Comparative functional analysis of aquaporins/glyceroporins in mammals and anurans Mamm Genome 2007 18 452 462 10.1007/s00335-007-9041-5 17653793 12. Schenk A.D. Hite R.K. Engel A. Fujiyoshi Y. Walz T. Chapter five—Electron crystallography and aquaporins Meth. Enzymol. 2010 483 91 119 20888471 13. Verkman A.S. Anderson M.O. Papadopoulos M.C. Aquaporins: Important but elusive drug targets Nat. Rev. Drug Discov. 2014 13 259 277 10.1038/nrd4226 24625825 14. Masyuk A.I. Marinelli R.A. La Russo N.F. Water transport by epithelia of the digestive tract Gastroenterology 2002 122 545 562 10.1053/gast.2002.31035 11832467 15. Cohly H.H.P. Isokpehi R. Rajnarayanan R.V. Compartmentalization of aquaporins in the human intestine Int. J. Environ. Res. Public Health 2008 5 115 119 10.3390/ijerph5020115 18678926 16. Kagnoff M.F. The intestinal epithelium is an integral component of a communications network J. Clin. Investig. 2014 124 2841 2843 10.1172/JCI75225 24983425 17. Thiagarajah J. Verkman A. Water transport in the gastrointestinal tract Physiology of the Gastrointestinal Tract 5th ed. Johnson L.R. Ghishan F.K. Kaunitz J.D. Academic Press New York, NY, USA 2006 1757 1780 18. Benga G. The first discovered water channel protein, later called aquaporin 1: Molecular characteristics, functions and medical implications Mol. Asp. Med. 2012 33 518 534 10.1016/j.mam.2012.06.001 22705445 19. Matsuzaki T. Tajika Y. Suzuki T. Aoki T. Hagiwara H. Takata K. Immunolocalization of the water channel, aquaporin-5 (AQP5), in the rat digestive system Arch. Histol. Cytol. 2003 66 307 315 10.1679/aohc.66.307 14692686 20. Mobasheri A. Marples D. Expression of the AQP-1 water channel in normal human tissues: A semiquantitative study using tissue microarray technology Am. J. Physiol. Cell Physiol. 2004 286 C529 C537 10.1152/ajpcell.00408.2003 14592814 21. Laforenza U. Water channel proteins in the gastrointestinal tract Mol. Asp. Med. 2012 33 642 650 10.1016/j.mam.2012.03.001 22465691 22. Gregoire F. Lucidi V. Zerrad‑Saadi A. Virreira M. Bolaky N. Delforge V. Lemmers A. Donckie V. Devière J. Demetter P. Analysis of aquaporin expression in liver with a focus on hepatocytes Histochem. Cell Biol. 2015 144 347 363 10.1007/s00418-015-1341-3 26126651 23. Burghardt B. Elkjær M.L. Kwon T.H. Rácz G.Z. Varga G. Steward M.C. Nielsen S. Distribution of aquaporin water channels AQP1 and AQP5 in the ductal system of the human pancreas Gut 2003 52 1008 1016 10.1136/gut.52.7.1008 12801959 24. Li L. Zhang H. Ma T. Verkman A.S. Very high aquaporin-1 facilitated water permeability in mouse gallbladder Am. J. Physiol. Gastrointest. Liver Physiol. 2009 296 G816 G822 10.1152/ajpgi.90680.2008 19179619 25. Mobasheri A. Wray S. Marples D. Distribution of AQP2 and AQP3 water channels in human tissue microarrays J. Mol. Histol. 2005 36 1 14 10.1007/s10735-004-2633-4 15703994 26. Matsuzaki T. Tajika Y. Ablimit A. Aoki T. Hagiwara H. Takata K. Aquaporins in the digestive system Med. Electron. Microsc. 2004 37 71 80 10.1007/s00795-004-0246-3 15221647 27. Ishibashi K. Sasaki S. Saito F. Ikeuchi T. Marumo F. Structure and chromosomal localization of a human water channel (AQP3 ) gene Genomics 1995 27 352 354 10.1006/geno.1995.1055 7558005 28. Silberstein C. Kierbel A. Amodeo G. Zotta E. Bigi F. Berkowski D. Ibarra C. Functional characterization and localization of AQP3 in the human colon Braz. J. Med. Biol. Res. 1999 32 1303 1313 10.1590/S0100-879X1999001000018 10510269 29. Ikarashi N. The elucidation of the function and the expression control mechanism of aquaporin-3 in the colon Yakugaku Zasshi 2013 133 955 961 10.1248/yakushi.13-00173 23995803 30. Parvin M.N. Kurabuchi S. Murdiastuti K. Yao C. Tanaka C.K. Akamatsu T. Kanamori N. Hosoi K. Subcellular redistribution of AQP5 by vasoactive intestinal polypeptide in the Brunner’s gland of the rat duodenum Am. J. Physiol. Gastrointest. Liver Physiol. 2005 288 G1283 G1291 10.1152/ajpgi.00030.2004 15650134 31. Laforenza U. Miceli E. Gastaldi G. Scaffino M.F. Ventura U. Fontana J.M. Orsenigo M.N. Corazza G.R. Solute transporters and aquaporins are impaired in celiac disease Biol. Cell 2010 102 457 467 10.1042/BC20100023 20415666 32. Fischer H. Stenling R. Rubio C. Lindblom A. Differential expression of Aquaporin 8 in human colonic epithelial cells and colorectal tumors BMC Physiol. 2001 1 1213 10.1186/1472-6793-1-1 11231887 33. Hurley P.T. Ferguson C.J. Kwon T.H. Andersen M.L. Norman A.G. Steward M.C. Nielsen S. Case R.M. Expression and immunolocalization of aquaporin water channels in rat exocrine pancreas Am. J. Physiol. Gastrointest. Liver Physiol. 2001 280 G701 G709 11254497 34. Ambe P.C. Godde D. Zirngibl H. Storke S. Aquaporin-1 and 8 expression in the gallbladder mucosa might not be associated with the development of gallbladder stones in humans Eur. J. Clin. Investig. 2016 46 227 233 10.1111/eci.12586 26707370 35. Okada S. Misaka T. Matsumoto I. Watanabe H. Abe K. Aquaporin-9 is expressed in a mucus-secreting goblet cell subset in the small intestine FEBS Lett. 2003 540 157 162 10.1016/S0014-5793(03)00256-4 12681500 36. Mobasheri A. Shakibaei M. Marples D. AQP10 in human intestine needs two important references: Immunohistochemical localization of aquaporin 10 in the apical membranes of the human ileum: A potential pathway for luminal water and small solute absorption Histochem. Cell Biol. 2004 121 463 471 10.1007/s00418-004-0657-1 15221416 37. Li H. Kamiie J. Morishita Y. Yoshida Y. Yaoita E. Ishibashi K. Yamamoto T. Expression and localization of two isoforms of AQP10 in human small intestine Biol. Cell 2005 97 823 829 10.1042/BC20040091 15898950 38. Morinaga T. Nakakoshi M. Hirao A. Imai M. Ishibashi K. Mouse aquaporin 10 gene (AQP10 ) is a pseudogene Biochem. Biophys. Res. Commun. 2002 294 630 634 10.1016/S0006-291X(02)00536-3 12056815 39. Ishibashi K. Tanaka Y. Morishita Y. The role of mammalian superaquaporins inside the cell Biochim. Biophys. Acta 2014 1840 1507 1512 10.1016/j.bbagen.2013.10.039 24189537 40. Ohta E. Itoh T. Nemoto T. Kumagai J. Ko S.B. Ishibashi K. Ohno M. Uchida K. Ohta A. Sohara E. Pancreas-specific aquaporin 12 null mice showed increased susceptibility to caerulein-induced acute pancreatitis Am. J. Physiol. Cell Physiol. 2009 297 C1368 C1378 10.1152/ajpcell.00117.2009 19726746 41. Laforenza U. Bottino C. Gastaldi G. Mammalian aquaglyceroporin function in metabolism Biochim. Biophys. Acta 2016 1858 1 11 10.1016/j.bbamem.2015.10.004 26456554 42. Van Hoek A.N. Wiener M.C. Verbavatz J.M. Brown D. Townsend R.R. Lipniunas A.S. Verkman P.H. Purification and structure-function analysis of Native, PNGase F- and endo-β-galactosidase-treated CHIP28 water channels Biochemistry 1995 34 2212 2219 10.1021/bi00007a015 7532004 43. Tritto S. Gastaldi G. Zelenin S. Grazioli M. Orsenigo M.N. Ventura U. Laforenza U. Zelenina M. Osmotic water permeability of rat intestinal brush border membrane vesicles: Involvement of aquaporin-7 and aquaporin-8 and effect of metal ions Biochem. Cell Biol. 2007 85 675 684 10.1139/O07-142 18059526 44. Tani T. Koyama Y. Nihei K. Hatakeyama S. Ohshiro K. Yoshida Y. Yaoita E. Sakai Y. Hatakeyama K. Yamamoto T. Immunolocalization of aquaporin-8 in the rat digestive organs and testis Arch. Histol. Cytol. 2001 64 159 168 10.1679/aohc.64.159 11436986 45. Elkjær M.L. Nejsum L.N. Gresz V. Kwon T.H. Jensen U.B. Frøkiær J. Nielsen S. Immunolocalization of aquaporin-8 in rat kidney, gastrointestinal tract, testis, and airways Am. J. Physiol. Ren. Physiol. 2001 281 F1047 F1057 10.1152/ajprenal.0158.2001 46. Hatakeyama S. Yoshida Y. Tani T. Koyama Y. Nihei K. Ohshiro K. Kamiie J.I. Yaoita E. Suda T. Hatakeyama K. Cloning of a new aquaporin (AQP10) abundantly expressed in duodenum and jejunum Biochem. Biophys. Res. Commun. 2001 287 814 819 10.1006/bbrc.2001.5661 11573934 47. Verkman A.S. Physiological importance of aquaporins: Lessons from knockout mice Curr. Opin. Nephrol. Hypertens. 2000 9 517 522 10.1097/00041552-200009000-00010 10990371 48. Ma T. Verkman A.S. Aquaporin water channels in gastrointestinal physiology J. Physiol. 1999 517 317 326 10.1111/j.1469-7793.1999.0317t.x 10332084 49. Gorelick D.A. Praetorius J. Tsunenari T. Nielsen S. Agre P. Aquaporin-11: A channel protein lacking apparent transport function expressed in brain BMC Biochem. 2006 7 1 14 10.1186/1471-2091-7-14 16401342 50. Gallardo P. Cid L.P. Vio C.P. Sepulveda F.V. Aquaporin-2, a regulated water channel, is expressed in apical membranes of rat distal colon epithelium Am. J. Physiol. Gastrointest. Liver Physiol. 2001 281 G856 G863 11518698 51. Thiagarajah J. Zhao D. Verkman A.S. Impaired enterocyte proliferation in aquaporin-3 deficiency in mouse models of colitis Gut 2007 56 1529 1535 10.1136/gut.2006.104620 17573386 52. Matsuzaki T. Suzuki T. Koyama H. Tanaka S. Takata K. Water channel protein AQP3 is present in epithelia exposed to the environment of possible water loss J. Histochem. Cytochem. 1999 47 1275 1286 10.1177/002215549904701007 10490456 53. Wang K.S. Ma T. Filiz F. Verkman S. Bastidas J.A. Colon water transport in transgenic mice lacking aquaporin-4 water channels Am. J. Physiol. Gastrointest. Liver Physiol. 2000 279 G463 G470 10915657 54. Laforenza U. Gastaldi G. Grazioli M. Cova E. Tritto S. Faelli A. Calamita G. Ventura U. Expression and immunolocalization of aquaporin-7 in rat gastrointestinal tract Biol. Cell 2005 97 605 613 10.1042/BC20040090 15943587 55. Magdeldina S. Lia H. Yoshidaa Y. Enanya S. Zhanga Y. Xua B. Fujinakaa H. Yaoitaa E. Yamamoto T. Comparison of two dimensional electrophoresis mouse colon proteomes before and after knocking out Aquaporin 8 J. Proteom. 2010 73 2031 2040 10.1016/j.jprot.2010.06.010 20619372 56. Bajaj J.S. Hylemon P.B. Younossi Z. The intestinal microbiota and liver disease Am. J. Gastroenterol. Suppl. 2012 1 9 14 10.1038/ajgsup.2012.3 57. Huebert R.C. Splinter P.L. Garcia F. Marinelli R.A. LaRusso N.F. Expression and localization of aquaporin water channels in rat hepatocytes J. Biol. Chem. 2002 277 22710 22717 10.1074/jbc.M202394200 11932260 58. Rojek A. Füchtbauer E.M. Füchtbauer A. Jelen S. Malmendal A. Fenton R.A. Nielsen S. Liver-specific Aquaporin 11 knockout mice show rapid vacuolization of the rough endoplasmic reticulum in periportal hepatocytes after amino acid feeding Am. J. Physiol. Gastrointest. Liver Physiol. 2013 304 G501 G515 10.1152/ajpgi.00208.2012 23275615 59. Calamita G. Ferri D. Bazzini C. Mazzone A. Bottà G. Liquori G.E. Paulmich M. Portincasa P. Meyer G. Svelto M. Expression and subcellular localization of the AQP8 and AQP1 water channels in the mouse gall-bladder epithelium Biol. Cell 2005 97 415 423 10.1042/BC20040137 15859952 60. Nielsen S. Smith B.L. Christensen E.I. Agre P. Distribution of the aquaporin CHIP in secretory and resorptive epithelia and capillary endothelia Proc. Natl. Acad. Sci. USA 1993 90 7275 7279 10.1073/pnas.90.15.7275 8346245 61. Debongnie J.C. Phillips S.F. Capacity of the human colon to absorb fluid Gastroenterology 1978 74 698 703 631507 62. Day R.E. Kitchen P. Owen D.S. Bland C. Marshall L. Conner A.C. Bill R.M. Conner M.T. Human aquaporins: Regulators of transcellular water flow Biochim. Biophys. Acta 2014 1840 1492 1506 10.1016/j.bbagen.2013.09.033 24090884 63. Read N.W. Intestinal transport of fluid and electrolytes (physiology and pathophysiology) Gastroenterology: Clinical Science and Practice Bouchier I.A.D. Allan R.N. Hodgson H.J.F. Keighley M.R.B. W.B. Saunders London, UK 1993 447 459 64. Murakami M. Murdiastuti K. Hosoi K. Hill A.E. AQP and the control of fluid transport in a salivary gland J. Membr. Biol. 2006 210 91 103 10.1007/s00232-005-0848-2 16868676 65. Fischbarg J. Diecke F.P. Iserovich P. Rubashkin A. The role of the tight junction in paracellular fluid transport across corneal endothelium. electro-osmosis as a driving force J. Membr. Biol. 2006 210 117 130 10.1007/s00232-005-0850-8 16868674 66. Loo D.D.F. Wright E.M. Zeuthen T. Water pumps J. Physiol. 2002 542 53 60 10.1113/jphysiol.2002.018713 12096049 67. Powell D.W. Barrier function of epithelia Liver Physiol. 1981 241 G275 G288 68. Duc C. Farman N. Canessa C.M. Bonvalet J.P. Rossier B.C. Cell-Specific expression of epithelial sodium channel α, β, and γ subunits in aldosterone-responsive epithelia from the rat: Localization by in situ hybridization and immunocytochemistry J. Cell Biol. 1994 127 1907 1921 10.1083/jcb.127.6.1907 7806569 69. Donato V. Lacquaniti A. Cernaro V. Lorenzano G. Trimboli D. Buemi A. Lupica R. Buemi M. From water to aquaretics: A legendary route Cell. Physiol. Biochem. 2014 33 1369 1388 10.1159/000358704 24853354 70. Ma T. Jayaraman S. Wang K.S. Bastidas J.A. Verkman A.S. Defective dietary fat processing in transgenic mice lacking aquaporin-1 water channels Am. J. Physiol. 2001 280 C126 C134 71. Verkman A.S. Aquaporin water channels and endothelial cell function J. Anat. 2002 200 617 627 10.1046/j.1469-7580.2002.00058.x 12162729 72. Kon R. Ikarashi N. Hayakawa A. Haga Y. Fueki A. Kusunoki Y. Tajima M. Ochiai W. Machida Y. Sugiyama K. Morphine-induced constipation develops with increased aquaporin-3 expression in the colon via increased serotonin secretion Toxicol. Sci. 2015 145 337 347 10.1093/toxsci/kfv055 25766885 73. Fushimi K. Uchida S. Hara Y. Hirata Y. Marumo F. Sasaki S. Cloning and expression of apical membrane water channel of rat kidney collecting tubule Nature 1993 361 549 552 10.1038/361549a0 8429910 74. Ishibashi K. Sasaki S. Fushimi K. Uchida S. Kuwahara M. Saito H. Furukawa T. Nakajima K. Yamaguchi Y. Gojobori T. Molecular cloning and expression of a member of the aquaporin family with permeability to glycerol and urea in addition to water expressed at the basolateral membrane of kidney collecting duct cells Proc. Natl. Acad. Sci. USA 1994 91 6269 6273 10.1073/pnas.91.14.6269 7517548 75. Laforenza U. Gastaldi G. Polimeni M. Tritto S. Tosco M. Ventura U. Scaffino M.F. Yasui M. Aquaporin-6 is expressed along the rat gastrointestinal tract and upregulated by feeding in the small intestine BMC Physiol. 2009 9 1213 10.1186/1472-6793-9-18 19811639 76. Gambardella C. Gallusa L. Amaroli A. Terova G. Masini M.A. Ferrando S. Fasting and re-feeding impact on leptin and aquaglyceroporin 9 in the liver of European sea bass (Dicentrarchus labrax ) Aquaculture 2012 354 1 6 10.1016/j.aquaculture.2012.04.043 77. De Luca A. Vassalotti G. Pelagalli A. Pero M.E. Squillacioti C. Mirabella N. Lombardi P. Avallone L. Expression and localization of aquaporin-1 along the intestine of colostrum suckling buffalo calves Anat. Histol. Embryol. 2015 44 391 400 10.1111/ahe.12157 25348329 78. Squillacioti C. De Luca A. Pero M.E. Vassalotti G. Lombardi P. Avallone L. Mirabella N. Pelagalli A. Effect of colostrum and milk on small intestine expression of AQP4 and AQP5 in newborn buffalo calves Res. Vet. Sci. 2015 103 149 155 10.1016/j.rvsc.2015.10.004 26679810 79. Pelagalli A. Squillacioti C. De Luca A. Pero M.E. Vassalotti G. Lombardi P. Avallone L. Mirabella N. Expression and localization of aquaporin 4 and aquaporin 5 along the large intestine of colostrum-suckling buffalo calves Anat. Histol. Embryol. 2015 10.1111/ahe.12210 26429198 80. Rojek A.M. Skowronski M.T. Füchtbauer E.M. Füchtbauer A.C. Fenton R.A. Agre P. Frøkiaer J. Nielsen S. Defective glycerol metabolism in aquaporin 9 (AQP9) knockout mice Proc. Natl. Acad. Sci. USA 2007 104 3609 3614 10.1073/pnas.0610894104 17360690 81. Maeda N. Hibuse T. Funahashi T. Role of aquaporin-7 and aquaporin-9 in glycerol metabolism; involvement in obesity Handb. Exp. Pharmacol. 2009 190 233 249 19096781 82. Maeda N. Implications of aquaglyceroporins 7 and 9 in glycerol metabolism and metabolic syndrome Mol. Asp. Med. 2012 33 665 675 10.1016/j.mam.2012.02.004 22425521 83. Hashem M.A. Biochemical and expression studies on Aquaporin 9 (AQP9) in wild and AQP9 knockout mice Vet. Arhiv. 2010 80 93 112 84. Jelen S. Wacker S. Aponte-Santamaria C. Skott M. Rojek A. Johanson U. Kjellbom P. Nielsen S. de Groot B.L. Rutzler M. Aquapoprin-9 protein is the primary route of hepatocyte glycerol uptake for glycerol gluconeogenesis in mice J. Biol. Chem. 2011 286 44319 44325 10.1074/jbc.M111.297002 22081610 85. Lehmann G.L. Larocca M.C. Soria L.R. Marinelli R.A. Aquaporins: Their role in cholestatic liver disease World J. Gastroenterol. 2008 14 7059 7067 10.3748/wjg.14.7059 19084912 86. Borgnia M. Nielsen S. Engel A. Agre P. Cellular and molecular biology of the aquaporin water channels Annu. Rev. Biochem. 1999 68 425 458 10.1146/annurev.biochem.68.1.425 10872456 87. Ishibashi K. Kuwahara M. Kageyama Y. Tohsaka A. Marumo F. Sasaki S. Cloning and functional expression of a second new aquaporin abundantly expressed in testis Biochem. Biophys. Res. Commun. 1997 237 714 718 10.1006/bbrc.1997.7219 9299432 88. Cho S.J. Sattar A.K. Jeong E.H. Satchi M. Cho J.A. Dash S. Mayes M.S. Stromer M.H. Jena B.P. Aquaporin 1 regulates GTP-induced rapid gating of water in secretory vesicles Proc. Natl. Acad. Sci. USA 2002 99 4720 4724 10.1073/pnas.072083499 11917120 89. Gabbi C. Kim H.J. Hultenby K. Bouton D. Toresson G. Warner M. Gustafsson J.A. Pancreatic exocrine insufficiency in LXRβ−/− mice is associated with a reduction in aquaporin-1 expression Proc. Natl. Acad. Sci. USA 2008 105 5052 15057 10.1073/pnas.0808097105 18362349 90. Van Erpecum K.J. Wang D.Q.H. Moschetta A. Ferri D. Svelto M. Portincasa P. Hendrickx J.J. Schipper M. Calamita G. Gallbladder histopathology during murine gallstone formation: Relation to motility and concentrating function J. Lipid Res. 2006 47 32 41 10.1194/jlr.M500180-JLR200 16224116 91. Hardin J.A. Wallace L.E. Wong J.F. O’Loughlin E.V. Urbanski S.J. Gall D.G. Mac Naughton W.K. Beck P.L. Aquaporin expression is downregulated in a murine model of colitis and in patients with ulcerative colitis, Crohn’s disease and infectious colitis Cell Tissue Res. 2004 318 313 323 10.1007/s00441-004-0932-4 15338270 92. Guttmann J.A. Samji F.N. Li Y. Deng W. Lin A. Finlay B.B. Aquaporins contribute to diarrhoea caused by attaching and effacing bacterial pathogens Cell Microbiol. 2007 9 131 141 10.1111/j.1462-5822.2006.00773.x 16889624 93. Guttman J.A. Finlay B.B. Subcellular alterations that lead to diarrhea during bacterial pathogenesis Trends Microbiol. 2008 16 535 542 10.1016/j.tim.2008.08.004 18819802 94. Hodges K. Gill R. Infectious diarrhea. Cellular and molecular mechanisms Gut Microbes 2010 1 4 21 10.4161/gmic.1.1.11036 21327112 95. Lapointe T.K. O’Connor P.M. Buret A.G. The role of epithelial malfunction in the pathogenesis of enteropathogenic E. coli -induced diarrhea Lab. Investig. 2009 89 964 970 10.1038/labinvest.2009.69 19620958 96. Ricanek P. Lunde L.K. Frye S.A. Stoen M. Nygård S. Morth J.P. Rydning A. Vatn M.H. Moghaddam M.A. Tønjum T. Reduced expression of aquaporins in human intestinal mucosa in early stage inflammatory bowel disease Clin. Exp. Gastroenterol. 2015 8 49 67 10.2147/CEG.S70119 25624769 97. Chung S.H. Park Y.S. Kim O.S. Kim J.H. Baik H.W. Hong Y.O. Kim S.S. Shin J.H. Jun J.H. Jo Y. Melatonin attenuates dextran sodium sulfate induced colitis with sleep deprivation: Possible mechanism by microarray analysis Dig. Dis. Sci. 2014 59 1134 1141 10.1007/s10620-013-3013-2 24429513 98. Casotti G. Waldron T. Misquith G. Powers D. Slusher L. Expression and localization of an aquaporin-1 homologue in the avian kidney and lower intestinal tract Comp. Biochem. Physiol. 2007 147 355 362 10.1016/j.cbpa.2007.01.005 17320442 99. Berni Canani R. Di Costanzo M. Leone L. Pedata M. Meli R. Calignano A. Potential beneficial effects of butyrate in intestinal and extra intestinal diseases World J. Gastroenterol. 2011 17 1519 1528 10.3748/wjg.v17.i12.1519 21472114 100. Yamamoto T. Kuramoto H. Kadowaki M. Downregulation in aquaporin 4 and aquaporin 8 expression of the colon associated with the induction of allergic diarrhea in a mouse model of food allergy Life Sci. 2007 81 115 120 10.1016/j.lfs.2007.04.036 17574630 101. Zhi X. Tao J. Li Z. Jiang B. Feng J. Yang L. Xu H. Xu Z. MiR-874 promotes intestinal barrier dysfunction through targeting AQP3 following intestinal ischemic injury FEBS Lett. 2014 588 757 763 10.1016/j.febslet.2014.01.022 24462679 102. Zhang W. Xu Y. Chen Z. Xu Z. Xu H. Knockdown of aquaporin 3 is involved in intestinal barrier integrity impairment FEBS Lett. 2011 585 3113 3119 10.1016/j.febslet.2011.08.045 21907710 103. Jin S.Y. Liu Y.L. Xu L.N. Jiang Y. Wang Y. Yang B.X. Yang H.T. Ma H. Cloning and characterization of porcine aquaporin 1 water channel expressed extensively in gastrointestinal system World J. Gastroenterol. 2006 12 1092 1097 10.3748/wjg.v12.i7.1092 16534851 104. Cao M. Yang M. Ou Z. Li D. Geng L. Chen P. Chen H. Gong S. Involvement of aquaporins in a mouse model of rotavirus diarrhea Virol. Sin. 2014 29 211 217 10.1007/s12250-014-3469-z 25160756 105. Nagahama M. Ma N. Semba R. Naruse S. Aquaporin 1 immunoreactive enteric neurons in the rat ileum Neurosci. Lett. 2006 395 206 210 10.1016/j.neulet.2005.10.092 16309835 106. Arciszewski M.B. Stefaniak M. Zacharko-Siembida A. Całka J. Aquaporin 1 water channel is expressed on submucosal but not myenteric neurons from the ovine duodenum Ann. Anat. 2011 193 81 85 10.1016/j.aanat.2010.11.003 21168316 107. Ramírez-Lorca R. Vizuete M.L. Venero J.L. Revuelta M.J. Cano A.A. Ilundáin M. Echevarría M. Localization of aquaporin-3 mRNA and protein along the gastrointestinal tract of Wistar rats Pflugers Arch. 1999 438 94 100 10.1007/s004240050884 10370092 108. Zhao G.X. Dong P.P. Peng R. Li J. Zhang D.Y. Wang J.Y. Shen X.Z. Dong L. Sun J.Y. Expression, localization and possible functions of aquaporins 3 and 8 in rat digestive system Biotech. Histochem. 2016 16 1 8 10.3109/10520295.2016.1144079 26983346 109. Bottino C. Vázquez M. Devesa V. Laforenza U. Impaired aquaporins expression in the gastrointestinal tract of rat after mercury exposure J. Appl. Toxicol. 2016 36 113 120 10.1002/jat.3151 25854323 110. Jiang L. Li J. Liu X. Burnstock G. Xiang Z. Expression of aquaporin-4 water channels in the digestive tract of the guinea pig J. Mol. Histol. 2014 45 229 241 10.1007/s10735-013-9545-0 24122228 111. Arciszewski M.B. Matysek M. Sienkiewicz W. Immunohistochemical localization of aquaporin 4 (AQP4) in the porcine gastrointestinal tract Acta Vet. Brno 2015 84 321 326 10.2754/avb201584040321 112. Parvin M.N. Tsumura K. Akamatsu T. Kanamori N. Hosoi K. Expression and localization of AQP5 in the stomach and duodenum of the rat Biochim. Biophys. Acta 2002 1542 116 124 10.1016/S0167-4889(01)00172-0 11853885 113. Moore B.A. Morris G.P. Stephen V. A novel in vitro model of Brunner’s gland secretion in the guinea pig duodenum Am. J. Physiol. Gastrointest. Liver Physiol. 2000 278 G477 G485 10712268 114. Ramírez-Lorca R. Muñoz-Cabello A.M. Toledo-Aral J.J. Ilundáin A.A. Echevarría M. Aquaporins in chicken: Localization of ck-AQP5 along the small and large intestine Comp. Biochem. Physiol. Mol. Integr. Physiol. 2006 143 269 277 10.1016/j.cbpa.2005.12.007 16418008 115. Calamita G. Mazzone A. Bizzoca A. Cavalier A. Cassano G. Thomas D. Svelto M. Expression and immunolocalization of the aquaporin-8 water channel in rat gastrointestinal tract Eur. J. Cell Biol. 2001 80 711 719 10.1078/0171-9335-00210 11824790 116. Tanaka Y. Morishita Y. Ishibashi K. Aquaporin 10 is a pseudogene in cattle and their relatives Biochem. Biophys. Rep. 2015 1 16 21 117. Talbot N.C. Garrett W.M. Caperna T.J. Analysis of the expression of aquaporin-1 and aquaporin-9 in pig liver tissue: Comparison with rat liver tissue Cells Tissues Organs 2003 174 117 128 10.1159/000071152 12835575 118. Frigeri A. Gropper M.A. Umenishi F. Kawashima M. Brown D. Verkman A.S. Localization of MIWC and GLIP water channel homologs in neuromuscular, epithelial and glandular tissues J. Cell Sci. 1995 108 2993 3002 8537439 119. Thi M.M. Spray D.C. Hanani M. Aquaporin-4 Water Channels in Enteric Neurons J. Neurosci. Res. 2008 86 448 456 10.1002/jnr.21496 17893913 120. García F. Kierbel A. Larocca M.C. Gradilone S.A. Splinter P. La Russo N.F. Marinelli R.A. The water channel aquaporin-8 is mainly intracellular in rat hepatocytes, and its plasma membrane insertion is stimulated by cyclic AMP J. Biol. Chem. 2001 276 12147 12152 10.1074/jbc.M009403200 11278499 121. Nicchia G.P. Frigeri A. Nico B. Ribatti D. Svelto M. Tissue distribution and membrane localization of aquaporin-9 water channel: Evidence for sex-linked differences in liver J. Histochem. Cytochem. 2001 49 1547 1556 10.1177/002215540104901208 11724902 122. Caperna T.J. Shannon A.E. Richards M.P. Garrett W.M. Talbot N.C. Identification and characterization of aquaporin-9 (AQP9) in porcine hepatic tissue and hepatocytes in monolayer culture Domest. Anim. Endocrinol. 2007 32 273 286 10.1016/j.domaniend.2006.03.008 16857339 123. Laforenza U. Cova E. Gastaldi G. Tritto S. Grazioli M. LaRusso N.F. Splinter P.L. D’Adamo P. Tosco M. Ventura U. Aquaporin-8 is involved in water transport in isolated superficial colonocytes from rat proximal colon J. Nutr. 2005 135 2329 2336 16177191 124. King L.S. Agre P. Man is not a rodent: Aquaporins in the airways Am. J. Respir. Cell Mol. Biol. 2001 24 221 223 10.1165/ajrcmb.24.3.f202 11245620 125. Nihei K. Koyama Y. Tani T. Yaoita E. Ohshiro K. Adhikary L.P. Kurosaki I. Shirai Y. Hatakeyama K. Yamamoto T. Immunolocalization of aquaporin-9 in rat hepatocytes and Leydig cells Arch. Histol. Cytol. 2001 64 81 88 10.1679/aohc.64.81 11310508 126. Tradtrantip L. Tajima M. Li L. Verkman A.S. Aquaporin water channels in transepithelial fluid transport J. Med. Investig. 2009 56 179 184 10.2152/jmi.56.179 20224178 127. Beitz E. Golldack A. Rothert M. von Bülow J. Challenges and achievements in the therapeutic modulation of aquaporin functionality Pharmacol. Ther. 2015 155 22 35 10.1016/j.pharmthera.2015.08.002 26277280

PMC005xxxxxx/PMC5000612.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081214ijms-17-01214ReviewAcute Generalized Exanthematous Pustulosis: Pathogenesis, Genetic Background, Clinical Variants and Therapy Feldmeyer Laurence *Heidemeyer Kristine Yawalkar Nikhil Jackson Chris Academic EditorDepartment of Dermatology, Inselspital, Bern University Hospital, University of Bern, Bern 3010, Switzerland; [email protected] (K.H.); [email protected] (N.Y.)* Correspondence: [email protected]; Tel.: +41-31-632-6622; Fax: +41-31-632-223127 7 2016 8 2016 17 8 121425 4 2016 13 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Acute generalized exanthematous pustulosis (AGEP) is a severe, usually drug-related reaction, characterized by an acute onset of mainly small non-follicular pustules on an erythematous base and spontaneous resolution usually within two weeks. Systemic involvement occurs in about 20% of cases. The course is mostly benign, and only in rare cases complications lead to life-threatening situations. Recent studies highlight the importance of genetic variations in interleukin-36 receptor antagonist gene (IL-36RN) in the pathogenesis of this disease. The physiopathology of AGEP remains unclear, but an involvement of innate and acquired immune cells together with resident cells (keratinocytes), which recruit and activate neutrophils via production of cytokines/chemokines such as IL-17, IL-36, granulocyte-macrophage colony-stimulating factor (GM-CSF), tumor necrosis factor alpha (TNFα) and chemokine (C-X-C motif) ligand 8 (CXCL8)/IL-8, has been postulated. Treatment is based on the removal of the causative drug, supportive care, infection prevention and use of potent topical or systemic steroids. acute generalized exanthematous pustulosisdermatologyskindrug reaction ==== Body 1. Introduction Cutaneous adverse reactions to drugs are common and encompass a variety of mild to severe and life-threatening reactions. Acute generalized exanthematous pustulosis (AGEP) represents a severe, usually drug-related skin reaction characterized by acute formation of sterile pustules on an erythematous background, fever and neutrophilia. 2. Background and Epidemiology While the clinical picture of drug-induced pustular eruptions in patients without any history of psoriasis had already been described in 1968 by Baker and Ryan, the term AGEP was introduced by Beylot et al. in 1980 [1,2]. AGEP is a rare adverse drug reaction with an incidence of one to five cases per million per year [3], but it might be underreported. It can occur at any age and seems to be more frequent in women [4]. 3. Aetiology Although many causative factors leading to AGEP have been described, it is, in over 90% of cases, associated with the ingestion of drugs [5,6]. Aminopenicillins, pristinamycin, sulphonamides, quinolones, hydroxychloroquine, terbinafin and diltiazem are the most frequent causative drugs [7]. In particular cases, AGEP is induced by bacterial, viral or parasitic infections (e.g., parvovirus B19 [8,9], mycoplasma [10,11], cytomegalovirus [12], coxsackie B4 [13], Chlamydia pneumoniae [14], Escherichia coli [15], and echinococcus [16]), spider bites [17], herbal medications [18], lacquer [18], mercury [19] and even psoralen combined with ultraviolet A (PUVA) treatment [20]. Finally venoms, foods and xenobiotics have also been suspected to induce the reaction [21]. 4. Genetic Background The genetic predisposition for the development of AGEP is not known. There seems to be a correlation between mutations in the IL-36RN gene, encoding the interleukine-36 receptor antagonist (IL-36Ra), and the development of generalized pustular eruptions after drug intake. IL-36Ra has an anti-inflammatory function and blocks the proinflammatory cytokines IL-36α, IL-36β and IL-36γ. Mutations in the IL-36RN gene can result in uncontrolled IL-36 signalling and increased downstream production of further proinflammatory cytokines and chemokines [22]. However, it is still unclear if mutations in IL-36RN lead to AGEP or, rather, to a drug-induced generalized pustular psoriasis (GPP), as it is described in some cases [23,24]. 5. Pathogenesis AGEP has been classified as a T cell-related sterile neutrophilic inflammatory response (type IVd reaction) [25,26,27]. The activation, proliferation and migration of drug-specific cluster of differentiation (CD) 4 and CD8 T cells play an important role in the development of AGEP (Figure 1), as supported by the use of patch tests [17,18,19,20] and in vitro tests [21,22]. It is supposed that drug-specific cytotoxic T cells and cytotoxic proteins such as granzyme B and perforin induce the apoptosis of keratinocytes, leading to subcorneal vesicles [27,28]. Recently, it has also been shown that, besides in toxic epidermal necrolysis (TEN), granulysin is also expressed by CD4 and CD8 T cells and natural killer (NK) cells in different drug reactions including AGEP, suggesting that granulysin may also play a role in the pathogenesis of AGEP [28]. Furthermore, in vitro tests have shown that drug-specific T cells in AGEP patients produced significantly more chemokine (C-X-C motif) ligand 8 (CXCL8)/IL-8, a potent neutrophil chemotactic chemokine [26,27,28,29]. CXCL8/IL-8 is thought to play a central role in the formation of pustules by recruitment of neutrophils. The increased levels of IL-17 and IL-22 as well as granulocyte-macrophage colony-stimulating factor (GM-CSF) in AGEP patients may also participate in the strong neutrophilic activity by the synergistic effect on the production of CXCL8/IL-8 and the prevention of apoptosis of the neutrophils [28,29]. Recent studies also described a higher level of IL-17 expression by neutrophils, mast cells (MC), and macrophages, and a lower level by T cells, in AGEP patients, indicating that innate cells may also be involved in the pathogenesis of AGEP [29]. Furthermore, a deficiency in the IL36-Ra in some AGEP patients seems to play a role, leading to the increased expression of various proinflammatory cytokines and chemokines such as IL-1, IL-6, IL-12, IL-23, IL-17, tumor necrosis factor alpha (TNFα) and CXCL8/IL-8, which can further enhance neutrophilic recruitment and activation [22,23,30]. In some AGEP patients, IL-5 expressed by infiltrating T cells may lead to the eosinophilia presented in about 30% of AGEP patients [26]. An elevated expression of TNFα in AGEP patients has been reported [31]. 6. Clinical Features and Variants Characteristically, patients with AGEP develop an acute rash with pinhead-sized pustules on an erythematous oedematous base, starting in the main folds (axillary, inguinal and submammary areas) and spreading quickly (within a few hours) on the trunk and limbs (Figure 2). The time period from drug ingestion to reaction onset is usually within 48 h, with antibiotics having a median of 24 h [7]. There is an itching or sometimes burning sensation [3,32]. Mucosal involvement, especially orally, is reported in about 20%–25% of patients but mostly in a limited extension and only on one mucosal region [5]. Systemic inflammation signs in the acute phase of the disease include fever (>38.0 °C), leucocytosis (>10,000/mL), elevated levels of C-reactive protein (CRP) and mostly increased levels of neutrophils (>7000/mL). As mentioned above, 30% of patients also present an eosinophilia and in 75% of cases a hypocalcaemia, probably related to hypoalbuminemia, is found [5,33]. Multiorgan involvement has been reported in 17% of cases [33]. Skin eruptions are sometimes accompanied by lymphadenopathy and occasionally by hepatocellular dysfunction and cholestasis as well as nephritis. Lung and bone marrow might also be involved, leading to respiratory failure and neutropenia, respectively [33]. One case with phenytoin-induced AGEP and cerebellar symptoms has been reported, while it is unknown whether cerebellar symptoms were related to the drug reaction or to phenytoin toxicity [34]. AGEP usually shows a mild course but high fever or cutaneous superinfection can complicate the process and lead to severe illness and sometimes life-threatening situations, especially in patients of poor general condition. The reported mortality is under 5% [3,4]. Usually there is a spontaneous resolution of skin lesions within two weeks with a very typical collarette-shaped desquamation in the region of previous pustulosis [3,4]. Besides the normal presentation of AGEP, several atypical variants and overlap syndromes have been described. For example, an overlap of AGEP and drug reaction with eosinophilia and systemic symptoms (DRESS) [35,36], or TEN [36,37,38,39,40,41,42,43,44], as well as an AGEP case with targetoid lesions [45,46] have been reported. A dozen of localized reactions have been reported, and referred to as acute localized exanthematous pustulosis (ALEP) [47,48,49,50]. In about 50% of patients, additional skin symptoms such as erythematous oedema of the hand and face, purpura, vesicles and blisters have been reported. Differential diagnosis of other pustular eruptions (such as bacterial or fungal infections, neutrophilic dermatoses, etc.) can mostly be excluded easily by clinical picture, history and histopathological findings. Acute GPP can present with the same clinical picture and may be difficult to distinguish, as the histopathological findings can sometimes not clearly distinguish between these two diseases (Table 1). The most important factor for the diagnosis is the quicker resolution time seen in AGEP. Recent studies described similarities in the pathogenesis of AGEP and GPP, like mutations in the IL-36Ra and an elevated expression of IL-17 [51]. DRESS typically presents with a morbilliform rash spreading from the face to the rest of the body, but might present with pustules as well. DRESS develops with a longer latent period of two to six weeks and mucous membranes and systemic involvement are more common [52]. Stevens-Johnson syndrome (SJS) and TEN are characterized by epidermal detachment and histological full-thickness epidermal necrosis [53]. Severe cases of AGEP, especially those with mucosal involvement, might be difficult to distinguish from these entities, and overlap forms have been described. 7. Histopathological Findings A histopathological examination should be performed to distinguish AGEP from other pustular eruptions. The skin biopsy should include a pustule. Typically, the biopsy shows spongiform subcorneal and/or intraepithelial pustules, an oedematous papillary dermis and perivascular infiltrates with neutrophils and some eosinophils (Figure 3). In some cases, necrotic keratinocytes and leucocytoclastic vasculitis can also be found. Histopathologically, it can be difficult to differentiate AGEP from GPP. The presence of eosinophils, necrotic keratinocytes, a mixed interstitial and mid-dermal perivascular infiltrate and absence of tortuous or dilated blood vessels favors AGEP, while the presence of psoriasiform acanthosis is characteristic of GPP [24,54]. 8. Diagnosis Diagnosis of AGEP can be made clinically with the support of histopathological findings as well as patch tests. The EuroSCAR study group presented a standardized scoring system in 2001, the AGEP validation score, including the morphology of skin lesions, the presence of fever, the clinical course, and the laboratory and histopathological findings [4]. To identify the responsible drug in case of polymedication, a patch test can be performed after complete skin resolution. The sensibility of the patch test in AGEP is higher than in other drug reactions such as SJS or TEN (58% positive in AGEP vs. 24% positive in SJS/TEN). A positive result often shows small pustules at the location of testing [55,56,57]. 9. Therapy Discontinuation of the causative agent is the main objective. Due to the mostly benign and self-limiting course, a supportive treatment based on topical steroids and disinfectant solutions during the pustular phase and rehydrating lotions during the desquamative phase, as well as antipyretics, is usually sufficient. In very extensive rashes, the intake of systemic steroids for a short time can be discussed [4], although there is no evidence that they reduce disease duration: their use is empirical and not supported by any randomized study. In one report, there was no difference between different treatment regimens regarding the course and duration of the disease or the length of fever [6,18,58]. Two cases of steroid-induced AGEP have been reported [58]. 10. Conclusions Recent research has improved our understanding of the pathophysiology of the disease but so far no markers have been identified that can predict which patients will develop the disease. The establishment of precise diagnostic criteria in AGEP is definitely a fundamental basis for clinical trials of quality [4]. There are currently no published randomized trials for a topical or systemic therapy for AGEP. Such trials are difficult to perform in a rare disease but necessary to provide a definitive answer. It is important to sensitize dermatologists and internists to this diagnosis as we feel that this disease is underreported. Indeed, the rash is usually not life-threatening and has a high rate of spontaneous resolution, so patients are not always referred to a dermatologist. Acknowledgments This work was supported by a Swiss Cancer Research Foundation grant (BIL KFS-3344-02-2014) to L.F. No funding was received for covering the costs to publish in open access. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Putative pathogenic mechanisms in acute generalized exanthematous pustulosis (AGEP). (A) In cases with drug involvement, the initial phase involves stimulation of drug-specific T cells and (B) their migration to the skin; (C) These T cells, possibly together with natural killer T (NKT) cells/natural killer (NK) cells are activated in the skin, where they induce apoptosis of keratinocytes through cytotoxic proteins and Fas/Fas ligand (FasL) interactions resulting in the formation of subcorneal vesicles; (D) Furthermore, these T cells together with subsequently activated bystander and inflammatory cells (keratinocytes, dendritic cells (DC), MC, neutrophils) release various cytokines and chemokines; (E) which predominately lead to neutrophilic inflammation and the formation of pustules. Figure 2 (A) Widespread rash with numerous pinhead-sized pustules on an erythematous oedematous base; (B) Flexural accentuation with characteristic collarette-shaped desquamation is typically observed in AGEP. Figure 3 Typical histological features of AGEP are indicated. ijms-17-01214-t001_Table 1Table 1 Factors favoring the diagnosis of AGEP over pustular psoriasis. AGEP Generalized Pustular Psoriasis History of psoriasis (family/personal) usually lacking often present Distribution pattern initially predominance in the folds more generalized Onset of pustules fast (hours or few days after use of medication) slower Duration of pustules Shorter (rapid resolution in a few days, max. 15 days, after drug suspension) longer Size of pustules tiny (pinhead) larger Duration of eruption/fever shorter (resolution in a few days after drug suspension) longer History of drug reaction usual uncommon Recent drug administration very frequent less frequent Arthritis rare about 30% Histology single-cell necrosis of keratinocytes, edema of papillary dermis, vasculitis, exocytosis of eosinophils papillomatosis, acanthosis, tortuous or dilated vessels ==== Refs References 1. Beylot C. Bioulac P. Doutre M.S. Acute generalized exanthematic pustuloses (four cases) Ann. Dermatol. Vénéréol. 1980 107 37 48 6989310 2. Baker H. Ryan T.J. Generalized pustular psoriasis. A clinical and epidemiological study of 104 cases Br. J. Dermatol. 1968 80 771 793 10.1111/j.1365-2133.1968.tb11947.x 4236712 3. Sidoroff A. Acute generalized exanthematous pustulosis Chem. Immunol. Allergy 2012 97 139 148 22613859 4. Sidoroff A. Halevy S. Bavinck J.N. Vaillant L. Roujeau J.C. Acute generalized exanthematous pustulosis (AGEP)—A clinical reaction pattern J. Cutan. Pathol. 2001 28 113 119 10.1034/j.1600-0560.2001.028003113.x 11168761 5. Roujeau J.C. Bioulac-Sage P. Bourseau C. Guillaume J.C. Bernard P. Lok C. Plantin P. Claudy A. Delavierre C. Vaillant L. Acute generalized exanthematous pustulosis. Analysis of 63 cases Arch. Dermatol. 1991 127 1333 1338 10.1001/archderm.1991.01680080069004 1832534 6. Chang S.-L. Huang Y.-H. Yang C.-H. Hu S. Hong H.-S. Clinical manifestations and characteristics of patients with acute generalized exanthematous pustulosis in Asia Acta Derm. Venereol. 2008 88 363 365 18709306 7. Sidoroff A. Dunant A. Viboud C. Halevy S. Bavinck J.N.B. Naldi L. Mockenhaupt M. Fagot J.-P. Roujeau J.-C. Risk factors for acute generalized exanthematous pustulosis (AGEP)-results of a multinational case-control study (EuroSCAR) Br. J. Dermatol. 2007 157 989 996 10.1111/j.1365-2133.2007.08156.x 17854366 8. Ofuji S. Yamamoto O. Acute generalized exanthematous pustulosis associated with a human parvovirus B19 infection J. Dermatol. 2007 34 121 123 10.1111/j.1346-8138.2006.00230.x 17239150 9. Lee D. Kang J.N. Hwang S.H. Lee Y.S. Kim H. Seo J.K. Sung H.S. Acute generalized exanthematous pustulosis induced by parvovirus b19 infection Ann. Dermatol. 2014 26 399 400 10.5021/ad.2014.26.3.399 24966644 10. Lim C.S.-H. Lim S.-L. Acute generalized exanthematous pustulosis associated with asymptomatic Mycoplasma pneumoniae infection Arch. Dermatol. 2009 145 848 849 10.1001/archdermatol.2009.127 19620579 11. Taguchi K. Oka M. Bito T. Nishigori C. Acute generalized exanthematous pustulosis induced by Mycoplasma pneumoniae infection J. Dermatol. 2016 43 113 114 10.1111/1346-8138.13151 26427049 12. Haro-Gabaldón V. Sánchez-Sánchez-Vizcaino J. Ruiz-Avila P. Gutiérrez-Fernández J. Linares J. Naranjo-Sintes R. Acute generalized exanthematous pustulosis with cytomegalovirus infection Int. J. Dermatol. 1996 35 735 737 10.1111/j.1365-4362.1996.tb00653.x 8891829 13. Feio A.B. Apetato M. Costa M.M. Sá J. Alcantara J. Acute generalized exanthematous pustulosis due to Coxsackie B4 virus Acta Méd. Port. 1997 10 487 491 9341042 14. Manzano S. Guggisberg D. Hammann C. Laubscher B. Acute generalized exanthematous pustulosis: First case associated with a Chlamydia pneumoniae infection Arch. Pédiatrie Organe Off. Soc. Fr. Pédiatrie 2006 13 1230 1232 10.1016/j.arcped.2006.06.004 16919427 15. Klein N. Hartmann M. Helmbold P. Enk A. Acute generalized exanthematous pustulosis associated with recurrent urinary tract infections Hautarzt Z. Für Dermatol. Venerol. Verwandte Geb. 2009 60 226 228 10.1007/s00105-008-1604-1 18773184 16. Cannistraci C. Parola I.L.L.A. RiganO R. Bassetti F. Ortona E. Santucci B. Picardo M. Siracusano S. Acute generalized exanthematous pustulosis in cystic echinococcosis: Immunological characterization Br. J. Dermatol. 2003 148 1245 1249 10.1046/j.1365-2133.2003.05346.x 12828756 17. Davidovici B.B. Pavel D. Cagnano E. Rozenman D. Halevy S. EuroSCAR RegiSCAR study group Acute generalized exanthematous pustulosis following a spider bite: Report of 3 cases J. Am. Acad. Dermatol. 2006 55 525 529 10.1016/j.jaad.2006.05.010 16908368 18. Choi M.J. Kim H.S. Park H.J. Park C.J. Lee J.D. Lee J.Y. Kim H.O. Park Y.M. Clinicopathologic manifestations of 36 Korean patients with acute generalized exanthematous pustulosis: A case series and review of the literature Ann. Dermatol. 2010 22 163 169 10.5021/ad.2010.22.2.163 20548906 19. Belhadjali H. Mandhouj S. Moussa A. Njim L. Amri M. Zakhama A. Zili J. Mercury-induced acute generalized exanthematous pustulosis misdiagnosed as a drug-related case Contact Dermat. 2008 59 52 54 10.1111/j.1600-0536.2007.01306.x 18598307 20. Bonnetblanc J.M. Combeau A. Dang P.M. Hydroxychloroquine-puvatherapy photoinduced acute generalized exanthematous pustulosis Ann. Dermatol. Vénéréol. 1995 122 604 605 8745685 21. Raison-Peyron N. “Cutaneous adverse drug reactions” are not always drug-induced Eur. J. Dermatol. 2013 23 439 442 24007777 22. Gabay C. Towne J.E. Regulation and function of interleukin-36 cytokines in homeostasis and pathological conditions J. Leukoc. Biol. 2015 97 645 652 10.1189/jlb.3RI1014-495R 25673295 23. Navarini A.A. Valeyrie-Allanore L. Setta-Kaffetzi N. Barker J.N. Capon F. Creamer D. Roujeau J.C. Sekula P. Simpson M.A. Trembath R.C. Rare variations in IL36RN in severe adverse drug reactions manifesting as acute generalized exanthematous pustulosis J. Investig. Dermatol. 2013 133 1904 1907 10.1038/jid.2013.44 23358093 24. Halevy S. Kardaun S.H. Davidovici B. Wechsler J. EuroSCAR RegiSCAR study group The spectrum of histopathological features in acute generalized exanthematous pustulosis: A study of 102 cases Br. J. Dermatol. 2010 163 1245 1252 10.1111/j.1365-2133.2010.09967.x 20698849 25. Britschgi M. Pichler W.J. Acute generalized exanthematous pustulosis, a clue to neutrophil-mediated inflammatory processes orchestrated by T cells Curr. Opin. Allergy Clin. Immunol. 2002 2 325 331 10.1097/00130832-200208000-00006 12130947 26. Britschgi M. Steiner U.C. Schmid S. Depta J.P. Senti G. Bircher A. Burkhart C. Yawalkar N. Pichler W.J. T-cell involvement in drug-induced acute generalized exanthematous pustulosis J. Clin. Investig. 2001 107 1433 1441 10.1172/JCI12118 11390425 27. Schmid S. Kuechler P.C. Britschgi M. Steiner U.C. Yawalkar N. Limat A. Baltensperger K. Braathen L. Pichler W.J. Acute generalized exanthematous pustulosis: Role of cytotoxic T cells in pustule formation Am. J. Pathol. 2002 161 2079 2086 10.1016/S0002-9440(10)64486-0 12466124 28. Schlapbach C. Zawodniak A. Irla N. Adam J. Hunger R.E. Yerly D. Pichler W.J. Yawalkar N. NKp46+ cells express granulysin in multiple cutaneous adverse drug reactions Allergy 2011 66 1469 1476 10.1111/j.1398-9995.2011.02677.x 21819408 29. Kabashima R. Sugita K. Sawada Y. Hino R. Nakamura M. Tokura Y. Increased circulating Th17 frequencies and serum IL-22 levels in patients with acute generalized exanthematous pustulosis J. Eur. Acad. Dermatol. Venereol. 2011 25 485 488 10.1111/j.1468-3083.2010.03771.x 20569282 30. Marrakchi S. Guigue P. Renshaw B.R. Puel A. Pei X.Y. Fraitag S. Zribi J. Bal E. Cluzeau C. Chrabieh M. Interleukin-36-receptor antagonist deficiency and generalized pustular psoriasis N. Engl. J. Med. 2011 365 620 628 10.1056/NEJMoa1013068 21848462 31. Meiss F. Helmbold P. Meykadeh N. Gaber G. Marsch W.C. Fischer M. Overlap of acute generalized exanthematous pustulosis and toxic epidermal necrolysis: Response to antitumour necrosis factor-alpha antibody infliximab: Report of three cases J. Eur. Acad. Dermatol. Venereol. 2007 21 717 719 10.1111/j.1468-3083.2006.02026.x 17448012 32. Beylot C. Doutre M.S. Beylot-Barry M. Acute generalized exanthematous pustulosis Semin. Cutan. Med. Surg. 1996 15 244 249 10.1016/S1085-5629(96)80037-X 9069592 33. Hotz C. Valeyrie-Allanore L. Haddad C. Bouvresse S. Ortonne N. Duong T.A. Ingen-Housz-Oro S. Roujeau J.C. Wolkenstein P. Chosidow O. Systemic involvement of acute generalized exanthematous pustulosis: A retrospective study on 58 patients Br. J. Dermatol. 2013 169 1223 1232 10.1111/bjd.12502 23855377 34. Shingade P.U. Wankhede V. Kataria P.S. Sonone N. Rare case of phenytoin induced acute generalized exanthematous pustulosis with cerebellar syndrome Indian J. Dermatol. 2014 59 210 10.4103/0019-5154.127715 24700960 35. Gey A. Milpied B. Dutriaux C. Mateus C. Robert C. Perro G. Taieb A. Ezzedine K. Jouary T. Severe cutaneous adverse reaction associated with vemurafenib: DRESS, AGEP or overlap reaction? J. Eur. Acad. Dermatol. Venereol. 2016 30 178 179 10.1111/jdv.12685 25175726 36. Bouvresse S. Valeyrie-Allanore L. Ortonne N. Konstantinou M.P. Kardaun S.H. Bagot M. Wolkenstein P. Roujeau J.C. Toxic epidermal necrolysis, DRESS, AGEP: Do overlap cases exist? Orphanet J. Rare Dis. 2012 10.1186/1750-1172-7-72 23009177 37. Cohen A.D. Cagnano E. Halevy S. Acute generalized exanthematous pustulosis mimicking toxic epidermal necrolysis Int. J. Dermatol. 2001 40 458 461 10.1046/j.1365-4362.2001.01241.x 11679003 38. Scheinfeld N. Wesson K. Perry P. Weinberg J. Acute generalized exanthematous pustulosis resembling toxic epidermal necrolysis caused by famotidine Acta Derm. Venereol. 2003 83 76 77 12636039 39. Peermohamed S. Haber R.M. Acute generalized exanthematous pustulosis simulating toxic epidermal necrolysis: A case report and review of the literature Arch. Dermatol. 2011 147 697 701 10.1001/archdermatol.2011.147 21690532 40. Goh T.K. Pang S.M. Thirumoorthy T. Goh S.G.N. Acute generalised exanthematous pustulosis and toxic epidermal necrolysis induced by carbamazepine Singap. Med. J. 2008 49 507 510 41. Lateef A. Tan K.B. Lau T.C. Acute generalized exanthematous pustulosis and toxic epidermal necrolysis induced by hydroxychloroquine Clin. Rheumatol. 2009 28 1449 1452 10.1007/s10067-009-1262-4 19727917 42. Worsnop F. Chong H. Ostlere L. Natkunarajah J. Acute generalized exanthematous pustulosis mimicking toxic epidermal necrolysis in patients with psoriasis: A coincidence? Clin. Exp. Dermatol. 2015 40 688 689 10.1111/ced.12590 25677856 43. Smith V.M. Pollock B. Wilkinson S.M. Rose R.F. Acute generalized exanthematous pustulosis mimicking toxic epidermal necrolysis in patients with psoriasis Clin. Exp. Dermatol. 2016 40 688 689 10.1111/ced.12827 26931079 44. Moling O. Perino F. Piccin A. Acute generalized exanthematous pustulosis with overlap features of toxic epidermal necrolysis/Stevens-Johnson syndrome Int. J. Dermatol. 2014 53 e27 e28 10.1111/j.1365-4632.2012.05465.x 23452000 45. Lin J.H. Sheu H.M. Lee J.Y.-Y. Acute generalized exanthematous pustulosis with erythema multiforme-like lesions Eur. J. Dermatol. 2002 12 475 478 12370139 46. Poeschl M.D. Hurley M.Y. Goyal S.D. Vidal C.I. Targetoid eruptions: Acute generalized exanthematous pustulosis Am. J. Dermatopathol. 2014 36 827 828 10.1097/DAD.0000000000000167 25243398 47. Corral de la Calle M. Martín Díaz M.A. Flores C.R. Vidaurrazaga C. Acute localized exanthematous pustulosis secondary to levofloxacin Br. J. Dermatol. 2005 152 1076 1077 10.1111/j.1365-2133.2005.06568.x 15888181 48. Kim S.W. Lee U.H. Jang S.J. Park H.S. Kang Y.S. Acute localized exanthematous pustulosis induced by docetaxel J. Am. Acad. Dermatol. 2010 63 e44 e46 10.1016/j.jaad.2009.05.007 20633784 49. Liang C.P. Yang C.S. Shen J.L. Chen Y.J. Sorafenib-induced acute localized exanthematous pustulosis in a patient with hepatocellular carcinoma Br. J. Dermatol. 2011 165 443 445 10.1111/j.1365-2133.2011.10377.x 21495998 50. Tresch S. Cozzio A. Kamarashev J. Harr T. Schmid-Grendelmeier P. French L.E. Feldmeyer L. T cell-mediated acute localized exanthematous pustulosis caused by finasteride J. Allergy Clin. Immunol. 2012 129 589 594 10.1016/j.jaci.2011.07.033 21868079 51. Kakeda M. Schlapbach C. Danelon G. Tang M.M. Cecchinato V. Yawalkar N. Uguccioni M. Innate immune cells express IL-17A/F in acute generalized exanthematous pustulosis and generalized pustular psoriasis Arch. Dermatol. Res. 2014 306 933 938 10.1007/s00403-014-1488-0 25030504 52. Husain Z. Reddy B.Y. Schwartz R.A. DRESS syndrome: Part I. Clinical perspectives J. Am. Acad. Dermatol. 2013 68 693.e1 693.e14 10.1016/j.jaad.2013.01.033 23602182 53. Schwartz R.A. McDonough P.H. Lee B.W. Toxic epidermal necrolysis: Part I. Introduction, history, classification, clinical features, systemic manifestations, etiology, and immunopathogenesis J. Am. Acad. Dermatol. 2013 69 173.e1 173.e13 10.1016/j.jaad.2013.05.003 23866878 54. Kardaun S.H. Kuiper H. Fidler V. Jonkman M.F. The histopathological spectrum of acute generalized exanthematous pustulosis (AGEP) and its differentiation from generalized pustular psoriasis J. Cutan. Pathol. 2010 37 1220 1229 10.1111/j.1600-0560.2010.01612.x 20738458 55. Wolkenstein P. Chosidow O. Fléchet M.L. Robbiola O. Paul M. Dumé L. Revuz J. Roujeau J.C. Patch testing in severe cutaneous adverse drug reactions, including Stevens-Johnson syndrome and toxic epidermal necrolysis Contact Dermat. 1996 35 234 236 10.1111/j.1600-0536.1996.tb02364.x 56. Barbaud A. Drug patch testing in systemic cutaneous drug allergy Toxicology 2005 209 209 216 10.1016/j.tox.2004.12.024 15767038 57. Barbaud A. Collet E. Milpied B. Assier H. Staumont D. Avenel-Audran M. Grange A. Amarger S. Girardin P. Guinnepain M.-T. A multicentre study to determine the value and safety of drug patch tests for the three main classes of severe cutaneous adverse drug reactions Br. J. Dermatol. 2013 168 555 562 10.1111/bjd.12125 23136927 58. Buettiker U. Keller M. Pichler W.J. Braathen L.R. Yawalkar N. Oral prednisolone induced acute generalized exanthematous pustulosis due to corticosteroids of group A confirmed by epicutaneous testing and lymphocyte transformation tests Dermatology 2006 213 40 43 10.1159/000092837 16778426

PMC005xxxxxx/PMC5000613.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081215ijms-17-01215ArticleA Machine Learning Approach for Hot-Spot Detection at Protein-Protein Interfaces Melo Rita 12Fieldhouse Robert 3Melo André 4Correia João D. G. 1Cordeiro Maria Natália D. S. 4Gümüş Zeynep H. 3Costa Joaquim 5Bonvin Alexandre M. J. J. 6Moreira Irina S. 26*González-Díaz Humberto Academic Editor1 Centro de Ciências e Tecnologias Nucleares, Instituto Superior Técnico, Universidade de Lisboa, Estrada Nacional 10 (ao km 139,7), 2695-066 Bobadela LRS, Portugal; [email protected] (R.M.); [email protected] (J.D.G.C.)2 CNC—Center for Neuroscience and Cell Biology; Rua Larga, Faculdade de Medicina, Polo I, 1ºandar, Universidade de Coimbra, 3004-504 Coimbra, Portugal3 Department of Genetics and Genomics and Icahn Institute for Genomics and Multiscale Biology, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA; [email protected] (R.F.); [email protected] (Z.H.G.)4 REQUIMTE (Rede de Química e Tecnologia), Faculdade de Ciências da Universidade do Porto, Departamento de Química e Bioquímica, Rua do Campo Alegre, 4169-007 Porto, Portugal; [email protected] (A.M.); [email protected] (M.N.D.S.C.)5 CMUP/FCUP, Centro de Matemática da Universidade do Porto, Faculdade de Ciências, Rua do Campo Alegre, 4169-007 Porto, Portugal; [email protected] Bijvoet Center for Biomolecular Research, Faculty of Science—Chemistry, Utrecht University, Utrecht 3584CH, The Netherlands; [email protected]* Correspondence: [email protected]; Tel.: +351-239-820-19027 7 2016 8 2016 17 8 121524 5 2016 18 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Understanding protein-protein interactions is a key challenge in biochemistry. In this work, we describe a more accurate methodology to predict Hot-Spots (HS) in protein-protein interfaces from their native complex structure compared to previous published Machine Learning (ML) techniques. Our model is trained on a large number of complexes and on a significantly larger number of different structural- and evolutionary sequence-based features. In particular, we added interface size, type of interaction between residues at the interface of the complex, number of different types of residues at the interface and the Position-Specific Scoring Matrix (PSSM), for a total of 79 features. We used twenty-seven algorithms from a simple linear-based function to support-vector machine models with different cost functions. The best model was achieved by the use of the conditional inference random forest (c-forest) algorithm with a dataset pre-processed by the normalization of features and with up-sampling of the minor class. The method has an overall accuracy of 0.80, an F1-score of 0.73, a sensitivity of 0.76 and a specificity of 0.82 for the independent test set. protein-protein interfaceshot-spotsmachine learningSolvent Accessible Surface Area (SASA)evolutionary sequence conservation ==== Body 1. Introduction Among all of the cellular components of living systems, proteins are the most abundant and the most functionally versatile. The specific interactions formed by these macromolecules are vital in a wide-range of biological pathways [1]. Protein-protein interactions involved in both transient and long-lasting networks of specific complexes play important roles in many biological processes [2,3,4]. Characterizing the critical residues involved in these interactions by both experimental and computational methods is therefore crucial to a proper understanding of living systems. Furthermore, only by gaining a complete understanding at atomistic detail can new methods be developed to modulate their binding [5,6]. Protein-protein interfaces often involve a large number of residues. However, it is generally recognized that small regions of a few residues, termed “Hot-Spots (HS)”, are essential for maintaining the integrity of the interface. The development of techniques to identify and characterize protein-based interfaces has become widespread. Experimental Alanine Scanning Mutagenesis (ASM) continues to be a valuable technique for both detecting and analyzing protein-binding interfaces. The contribution of a residue to the binding energy is measured by the binding free energy difference (ΔΔGbinding) between the wild-type (WT) and mutant complex upon mutation of a specific residue to alanine [7]. Bogan and Thorn [8] defined the residues with ΔΔGbinding ≥ 2.0 kcal·mol−1 as HS; and the residues with ΔΔGbinding < 2.0 kcal·mol−1 as Null-Spots (NS). Experimental methods for identifying HS are based on molecular biology techniques that are accurate, but still complex, time-consuming and expensive [9]. Highly efficient computational methods for predicting HS can provide a viable alternative to experiments. Molecular Dynamics (MD) simulations can be used to predict changes in the binding strength of protein complexes by calculating the free energy difference from an initial to a final state [10,11]. However, due to the complexity and typical large size of protein-protein complexes, these methods are still computationally expensive. Recently, machine learning approaches trained on various features of experimentally-determined HS residues have been developed in order to predict HS in new protein complexes [6,12,13,14]. In previous work, we have investigated feature-based methods combining Solvent Accessible Surface Area (SASA) descriptors calculated from static structures and MD ensembles and trained predictors using a Support Vector Machine (SVM) algorithm [15]. However, we only applied these to a small number of complexes, and the prediction performance was hampered by a high number of false positives. More recently, we added an extra feature (residue evolutionary sequence conservation) on a significantly larger dataset. In that study, we explored additional Machine Learning (ML) techniques, which led us to develop a more accurate and time-efficient HS detection methodology. This resulted in new HS predictor models for both protein-protein and protein-nucleic acid interactions, and we implemented the best performing models into two web tools [14]. In this study, we significantly expand both the number of studied protein-protein complexes and the number of 3D complex structure-based features used for prediction, including: interface size, the type of interaction between residues at the interface of the complex and the number of different types of residues at the interface. To the evolutionary sequence-based features, we added the Position-Specific Scoring Matrix (PSSM), for a total of 79 features. We have further tested a total of 27 algorithms from a simple linear-based function to support-vector machine models with different cost functions. The best predictor, based on a conditional inference random forest (c-forest) algorithm, achieves an overall performance characterized with an F1-score of 0.73, an accuracy of 0.80, a sensitivity of 0.76 and a specificity of 0.82. To the best of our knowledge, these values are higher than all other available prediction techniques. 2. Results In the current study, we have used the Classification And Regression Training (Caret) Package [16] from the R software [17], which provides a unified interface with a large number of built-in classifiers, in order to train an HS predictor. The dataset used for this purpose includes 545 amino acids from 53 complexes (140 HS and 405 NS). We calculated the percentage of the different types of amino acids within the NS set (Ser: 7.4; Gly: 1.5; Pro: 2.0; Val: 3.2; Leu: 2.7; Ile: 5.2; Met: 1.0; Cys: 0.7; Phe: 4.7; Tyr: 5.9; Trp: 4.9; His: 4.4; Lys 8.9; Arg: 10.6; Gln: 5.4; Asn: 6.2; Glu: 9.9; Asp: 7.2; Thr: 8.1) and within the HS set (Ser: 2.1; Gly: 2.9; Pro: 2.9; Val: 3.6; Leu: 7.1; Ile: 4.3; Met: 0.0; Cys: 0.0; Phe: 6.4; Tyr: 20.0; Trp: 5.7; His: 2.1; Lys 7.1; Arg: 6.4; Gln: 2.1; Asn: 5.0; Glu: 7.1; Asp: 10.7; Thr: 4.3). For both sets, there is a natural expected tendency for a higher percentage of large hydrophobic or charged residues at the interfaces, in particular Tyr. Although different patterns could influence the training of a robust classifier, we have previously successfully constructed models that were bias-free for all different amino acids [14]. We randomly split this dataset (see for details Supplementary Information Table S1) into a training set consisting of 70% of data (382 mutations) and an independent test set (163 mutations, 30%). This is a standard division scheme demonstrated to give a good result. All 27 classification models (listed in the Methods Section) were tested using 10-fold cross-validation repeated 10 times in order to avoid overfitting and to obtain the model’s generalization error. This means that the training set was split randomly into ten isolated parts, using nine of the ten parts to train the model and taking the remaining fold of data to test the final performance of the model. This process was repeated ten times. The performance of the five best algorithms for each tested condition was independently evaluated on the test set to ensure an unbiased assessment of the accuracy of the final model. The 79 features used in this work have different scales (i.e., the range of the raw data varies significantly), and therefore, we have performed feature normalization or data standardization of the predictor variables at the training set by centering the data, i.e., subtracting the mean and normalizing it by dividing by the standard deviation. The same protocol was followed for the test set taking into account the use of the training mean and standard deviation to ensure a good estimation of the model quality and generalization power. As we have a high-dimensional dataset (79 features), we have also applied Principal Components Analysis (PCA) to reduce the dimensionality of the data. PCA works by establishing an orthogonal transformation of the data to convert a set of possible correlated variables into a set of linearly-uncorrelated ones, the so-called principal components. One of the main concerns when applying classification to the detection of HS is the natural imbalance of the data. As expected, the number of HS is lower than the number of NS at a protein-protein interface, as indicated by the presence of 185 HS and 360 NS in the main dataset. In ML classification methods, the disparity of the frequencies of the observed classes may have a very negative impact on the models’ performance. To overcome this problem, we have tried two different subsampling techniques for the training set: down-sampling and up-sampling. In the first, there is a random sub-setting of all classes at the training set with their class frequency matching the least prevalence class (HS), whereas in the up-sampling, the opposite is happening with random sampling (with the replacement) of the minority class (HS) to reach the same size as the majority class (NS). Different conditions were thus established: (i) Scaled; (ii) Scaled Up; (iii) Scaled Down; (iv) PCA; (v) PCA Down; and (vi) PCA Up. Various statistical metrics (described in detail in the Methods Section) were adopted to evaluate the performance of the algorithms tested: Area Under the Receiver Operator Curve (AUROC), accuracy, True Positive Rate (TPR), True Negative Rate (TNR), Positive Predictive Value (PPV), False Positive Rate (FPR), False Negative Rate (FNR) and F1-score. Figure 1 illustrates the workflow followed in this study. The results for the training set for the best five algorithms for each of the six conditions studied are listed in Table 1. All statistical metrics obtained for the complete set of algorithms can be found in Supplementary Information Table S2, in which a more straightforward comparison by type of method can be made. The best classifiers seem to be almost constant in all six different pre-processing conditions, including one neuronal network (avNNET: model averaged Neural Network) and two tree-based methods (C5.0 Tree, C5.0 Rules). The fourth and fifth classifiers vary from nnet (neuronal network), to c-forest, GBM (stochastic gradient boosting machine) and svmRadialSigma (support vector machines with the Radial basis function kernel). The up-sampling of the HS class seems to improve the classifier performance presenting AUROC values higher than 0.80 in the majority of the cases. The performance of a classifier on the training set from which it was constructed gives a poor estimate of its accuracy in new cases. Furthermore, overfitting on algorithms without regularization terms (such as decision trees and neural networks) is harder to address on the training set. Therefore, the true predictive accuracy of the classifier was estimated on a separate test set corresponding to 30% of the main dataset. Table 2 summarizes the performance on the independent test set for the best classifiers shown in Table 1. From all of methods, c-forest, trained on the normalized up-scaling set, had the highest performance metrics on both training and test sets. It was therefore chosen as a final model. In our analysis of this classifier (Figure 2), we observed that the key features are structural ones: specifically, relSASAi, ΔSASAi, the number of contacts established by the interfacial residues at 4 Å and the number of LEU, VAL and HIS residues at the interface. All of these features were calculated using built-in functions of the VMD package [18] and in-house scripts. To validate the accuracy of the best predictor, we performed the HS predictions with other methods reported in the literature, such as Robetta [19], KFC2-A (Knowledge-based FADE and Contacts) [20], KFC2-B [20] and CPORT (Consensus Prediction Of interface Residues in Transient complexes)(not specialized in HS prediction, but instead, a protein-protein interface predictor) [21] on the same training and test sets. The comparison among these ML methods (Table 3) demonstrates that our new method achieves the best performance with F1-scores/AUROC values of 0.73/0.78 on the test set against 0.39/0.62, 0.56/0.66, 0.42/0.67 and 0.43/0.54 for Robetta, KFC2-A, KFC2-B and CPORT, respectively. 3. Discussion Machine learning is an area of artificial intelligence that is data driven with a focus on the development of computational techniques for making inferences or predictions. It has become widely used in a variety of areas due to its reduced application time and high performance. Over the past few years, a few algorithms have been applied for the specific problem in this study: the detection of hot-spots at protein-protein interfaces [13,14,15,22,23,24,25,26,27,28,29,30,31,32,33,34,35]. Here, neural networks and tree-based methods were highlighted as some of the high performance classifiers. Neural networks are inspired by biological nervous systems transmitting the information by a vast network of interconnecting processing elements (neurons). Decision trees organize the knowledge extracted from a hierarchy by using simple tests over the features of the training set. Both have been shown in the past to be promising ML algorithms in the bioinformatics field. Random forests were also shown to be able to predict the impact of each variable in high dimensional problems even in the presence of complex interactions [36]. In particular, c-forest [36], an implementation of the random forest and bagging ensemble method that uses conditional inference trees as base learners, achieved the top performance (Table 2) with a high F1-score of 0.93 on the training set using a 10 repeated 10-fold cross-validation. The values in the independent test (F1 score 0f 0.73) were also very high compared to the ones currently reported in the literature and surpassing all of the other methods tested in this study (Table 3; SBHD (Sasa-Based Hot-spot Detection) 0.61, Robetta 0.39, KFC2-A 0.56, KFC2-B 0.42 and CPORT 0.42). One important aspect that seemed to improve the results compared to our previous approaches (SBHD) was the use of in-built R techniques to balance the training data: up-scaling of the data led to a substantial improvement of the F1-score and to a decrease of the FPR to about 0.19 on the independent test set. In this particular classifier, the first seven features with higher importance were all structure-based: two already used in previous versions of our algorithm (ΔSASAi and relSASAi, check Material and Methods) and five new ones (the number of residues at a 4 Å distance and the number of LEU, VAL, HIS and PRO residues at the interface). The PSSM value for the TYR residues, one of the most common residues as HS, was the first genomic-based feature to be ranked as important. 4. Material and Methods 4.1. Dataset Construction We constructed a database of complexes by combining information from the Alanine Scanning Energetics database (ASEdb) [37], the Binding Interface Database (BID) [38] and the SKEMPI (Structural database of Kinetics and Energetics of Mutant Protein Interactions) [39] and PINT (Protein-protein Interactions Thermodynamic Database) [40] databases, which provide both experimental ΔΔGbinding values for interfacial residues and tridimensional (3D) X-ray structure information. The protein sequences were filtered to ensure a maximum of 35% sequence identity for at least one protein in each interface. Crystal structures were retrieved from the Protein Data Bank (PDB) [41], and all water molecules, ions and other small ligands were removed. Our final dataset consists of 545 mutations from 53 different complexes. 4.2. Sequence/Structural Features From a structural point of view, we compiled 12 previously-used different SASA descriptors for all interfacial residues [14,15]: (i) compSASAi, the solvent accessible surface area of residue i in the complex form; (ii) monSASAi, the residue SASA in the monomer form; (iii) ΔSASAi, the SASA difference upon complexation (Equation (1)); (iv) relSASAi, the ratio between ΔSASA for each residue and the monSASAi value for the same residue (Equation (2)). A further four features (comp/resSASAi, mon/resSASAi, Δ/resSASAi and rel/resSASAi), defined by Equations (3)–(6), were determined applying amino acid standardization by dividing the previous features by the average protein resSASAr values as determined by Miller and colleagues [42,43], with r being the respective residue type. Four additional, amino-acid standardized features were calculated by replacing the values determined by Miller by our own protein averages aveSASAr for each amino acid type in its respective protein: comp/aveSASAi, mon/aveSASAi, Δ/aveSASAi and rel/aveSASAi, defined in Equations (7)–(10). (1) ΔSASAi=|SASAicomp−SASAimon| (2) SASAirel=ΔSASAiSASAimon (3) SASAicomp/res=SASAicompSASArres (4) SASAimon/res=SASAimonSASArres (5) SASAiΔ/res=ΔSASAiSASArres (6) SASAirel/res=relSASAiSASArres (7) SASAicomp/ave=SASAicompSASArave (8) SASAimon/ave=SASAimonSASArave (9) SASAiΔ/ave=ΔSASAiSASArave (10) SASAirel/ave=relSASAiSASArave As the SASA features described in Equations (3)–(10) are rather small, the results presented here were multiplied by a factor of 103. We further introduced two features directly related to the size of the interface: the total number of interfacial residues and the ΔSASAtotal (sum of the ΔSASAi of all residues at the protein-protein binding interfaces). Twenty other features were added by splitting the total number of interface residues into the 20 amino acid types. Four contact features were also calculated: (i) the number of protein-protein contacts within 2.5 Å and (ii) 4.0 Å distance cut-offs, respectively; (iii) the number of intermolecular hydrogen bonds; and (iv) the number of intermolecular hydrophobic interactions. In-house scripts using the VMD molecular package [18] were used for all of these calculations. We used in total 38 structural features in our study. To utilize evolutionary sequence conservation information, we used the ConSurf server [44] that calculates a conservation score for each amino acid at an interfacial position for a complex, based on known sequences in different organisms. We also computed, PSSM using BLAST [45,46], as well as the weighted observed percentages, introducing them as 40 new features for all interfacial residues. Positive values in this matrix appear for substitutions more frequent than expected by random chance, and negative values indicate that the substitution is not frequent. Therefore, a total of 41 evolutionary sequence-related features were added to the structural features, resulting in 79 features in total for this study. 4.3. Machine Learning Techniques We first pre-processed the dataset by eliminating missing values or NZV (Near Zero Variance) features. Next, as mentioned in the Results section, we normalized the dataset and performed PCA. The algorithms tested were: avNNet (model averaged Neural Network); bagEarth (bagged MARS (multivariate adaptive regression splines)); bagEarthGCV Bagged MARS using gCV pruning; bagFDA (bagged Flexible Discriminant Analysis); C5.0Rules (single C5.0 Ruleset); C5.0Tree (single C5.0 Tree); c-forest (conditional inference random forest); ctree (conditional inference tree); ctree2 (conditional inference tree); earth (multivariate adaptive regression spline); fda (flexible discriminant analysis); gaussprLinear (Gaussian process); GBM (stochastic gradient boosting machine); gcvEarth (multivariate adaptive regression splines); hdda (high dimensional discriminant analysis); knn (k-nearest neighbors); lda (linear discriminant analysis); lda2 (linear discriminant analysis); multinom (penalized multinomial regression); nnet (neuronal networks); nb (naive Bayes); pda2 (penalized discriminant analysis); svmLinear (Support Vector Machines with Linear Kernel); svmLinear2 (Support Vector Machines with Linear Kernel); svmPoly (Support Vector Machines with Polynomial Kernel); svmRadial (support vector machines with the Radial basis function kernel); svmRadialCost (support vector machines with the Radial basis function kernel); svmRadialSigma (support vector machines with the Radial basis function kernel); svmRadialWeights (support vector machines with class Weights). The validity and performance of the various methods was determined by measuring the Area Under the Receiver Operator Curve (AUROC), the accuracy (Equation (11)), True Positive Rate (TPR/recall/sensitivity, Equation (12)), True Negative Rate (TNR/specificity, Equation (13)), Positive Predictive Value (PPV/Precision, Equation (14)), Negative Predictive Value (NPV) (Equation (15)), False Positive Rate (FPR/fall-out, Equation (16)), False Negative Rate (FNR, Equation (17)) and F1-score (Equation (18)) over our dataset. (11) Accuracy=TP+TNTP+FP+FN+TN (12) TPR=TPTP+FN (13) TNR=TNFP+TN (14) PPV=TPTP+FP (15) NPV=FPFP+TN  (16) FPR=FPFP+TN=1−TNR (17) FNR=FNTP+FN=1−TPR (18) F1 score=2TP2TP+FP+FN  In the equations above, TP stands for True Positive (predicted hot-spots that are actual hot-spots), FP stands for False Positive (predicted hot-spots that are not actual hot-spots), FN stands for False Negative (non-predicted hot-spots that are actual hot-spots) and TN stands the True Negatives (correctly-predicted null-spots). 4.4. Comparison with Other Software We compared our results with some of the common methods in the literature: Robetta [19], KFC2-A [20] and KFC2-B [20] and CPORT [21]. 5. Conclusions In conclusion, we were thus able to train an accurate and robust predictor using c-forest, a random forest ensemble learning method, and up-sampling of the minor class (HS) for dataset balance. This new method can now be widely applied to the detection of HS in protein-protein interfaces. The code is available upon request, will be implemented as a web-server in the near future and made available for the scientific community at the HADDOCK GitHub repository (http:github.com/haddocking). Acknowledgments Rita Melo acknowledges support from the Fundação para a Ciência e a Tecnologia (FCT—SFRH/BPD/97650/2013). The Centre for Nuclear Sciences and Technologies (C2TN) of Instituto Superior Técnico (IST) authors gratefully acknowledge the FCT support through the UID/Multi/04349/2013 project. Irina S. Moreira acknowledges support by the FCT Investigator program—IF/00578/2014 (co-financed by European Social Fund and Programa Operacional Potencial Humano) and by a Marie Skłodowska-Curie Individual Fellowship MSCA-IF-2015 (MEMBRANEPROT 659826). Irina S. Moreira also acknowledges FEDER (Programa Operacional Factores de Competitividade—COMPETE 2020) and FCT–project UID/NEU/04539/2013. Zeynep H. Gümüş acknowledges support from the Center for Basic and Translational Research on Disorders of the Digestive System, Rockefeller University, through the generosity of the Leona M. and Harry B. Helmsley Charitable Trust and from the start-up funds of the Icahn School of Medicine at Mount Sinai. This work was supported in part through the computational resources and staff expertise provided by Scientific Computing at the Icahn School of Medicine at Mount Sinai. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1215/s1. Click here for additional data file. Author Contributions Rita Melo, Robert Fieldhouse and Irina S. Moreira performed the experiments. André Melo, João D. G. Correia, Maria Natália N. D. S. Cordeiro, Zeynep H. Gumus, Joaquim Costa, Alexandre M. J. J. Bonvin and Irina S. Moreira conceived of and designed the experiments. All authors analyzed the data and wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 The flowchart of the current work. Figure 2 Top 15 variables for the c-forest method. SASA, Solvent Accessible Surface Area; #, Number of residues ijms-17-01215-t001_Table 1Table 1 Statistical metrics attained for five algorithms with top performance for each of the studied conditions for the training set. Pre-Processing Metrics Algorithms Scaled Nnet avNNET C5.0 Tree C5.0 Rules svmRadialSigma AUROC 0.52 0.65 0.77 0.72 0.78 Accuracy 0.92 0.94 0.96 0.92 0.91 Sensitivity 0.92 0.88 0.88 0.85 0.80 Specificity 0.91 0.98 1.00 0.96 0.97 PPV 0.86 0.95 0.99 0.92 0.93 NPV 0.95 0.94 0.94 0.92 0.89 FPR 0.09 0.02 0.00 0.04 0.03 F1-score 0.89 0.92 0.93 0.89 0.86 Scaled_Down c-Forest avNNET C5.0Tree C5.0Rules GBM AUROC 0.79 0.70 0.73 0.71 0.80 Accuracy 0.91 0.95 0.96 0.90 1.00 Sensitivity 0.93 0.96 0.96 0.89 0.99 Specificity 0.90 0.93 0.95 0.91 1.00 PPV 0.90 0.93 0.95 0.9 1.00 NPV 0.92 0.96 0.96 0.89 0.99 FPR 0.1 0.07 0.05 0.09 0 F1-score 0.91 0.95 0.96 0.9 1.00 Scaled_Up c-Forest avNNET C5.0Tree C5.0Rules GBM AUROC 0.85 0.75 0.85 0.82 0.84 Accuracy 0.93 0.94 0.98 0.95 0.98 Sensitivity 0.93 0.96 0.99 0.96 0.97 Specificity 0.93 0.92 0.97 0.94 0.99 PPV 0.93 0.92 0.97 0.94 0.99 NPV 0.93 0.96 0.99 0.95 0.97 FPR 0.07 0.08 0.03 0.06 0.01 F1-score 0.93 0.94 0.98 0.95 0.98 PCA nnet avNNET C5.0Tree C5.0Rules svmRadialSigma AUROC 0.69 0.75 0.61 0.59 0.76 Accuracy 1.00 0.99 0.98 0.92 0.91 Sensitivity 1.00 0.97 0.98 0.91 0.76 Specificity 1.00 1.00 0.98 0.93 0.99 PPV 1.00 0.99 0.96 0.89 0.97 NPV 1.00 0.98 0.99 0.95 0.88 FPR 0 0 0.02 0.07 0.01 F1-score 1.00 0.98 0.97 0.90 0.85 PCA_Down nnet avNNET C5.0Tree C5.0Rules svmRadialSigma AUROC 0.70 0.78 0.67 0.67 0.75 Accuracy 0.87 0.91 0.97 0.91 0.91 Sensitivity 0.88 0.88 0.96 0.96 0.88 Specificity 0.87 0.93 0.99 0.87 0.93 PPV 0.87 0.92 0.99 0.88 0.93 NPV 0.88 0.89 0.96 0.95 0.89 FPR 0.13 0.07 0.01 0.13 0.07 F1-score 0.87 0.90 0.97 0.92 0.91 PCA_Up nnet avNNET C5.0Tree C5.0Rules svmRadialSigma AUROC 0.75 0.82 0.80 0.78 0.80 Accuracy 0.95 0.98 0.98 0.96 0.94 Sensitivity 0.94 0.97 0.99 0.96 0.92 Specificity 0.96 0.99 0.98 0.96 0.95 PPV 0.96 0.99 0.98 0.96 0.95 NPV 0.94 0.97 0.99 0.96 0.92 FPR 0.04 0.01 0.02 0.04 0.05 F1-score 0.95 0.98 0.98 0.96 0.94 avNNET: model averaged Neural Network; C5.0 Rules (single C5.0 Ruleset); C5.0 Tree (single C5.0 Tree); c-forest (conditional inference random forest); GBM (stochastic gradient boosting machine); nnet (neuronal network); svmRadialSigma (support vector machines with the Radial basis function kernel); Positive Predictive Value (PPV); Negative Predictive Value (NPV); False Positive Rate (FPR). ijms-17-01215-t002_Table 2Table 2 Statistical metrics attained for 5 algorithms with the top performance for each of the studied conditions for the independent test set. Pre-Processing Metrics Algorithms Scaled Nnet avNNET C5.0 Tree C5.0 Rules svmRadialSigma AUROC 0.71 0.68 0.68 0.72 0.70 Accuracy 0.74 0.71 0.71 0.74 0.73 Sensitivity 0.57 0.57 0.5 0.60 0.55 Specificity 0.83 0.79 0.83 0.82 0.83 PPV 0.65 0.6 0.62 0.65 0.64 NPV 0.78 0.77 0.75 0.79 0.77 FPR 0.43 0.43 0.4 0.4 0.45 F1-score 0.61 0.58 0.55 0.62 0.59 Scaled_Down c-forest avNNET C5.0 Tree C5.0 Rules GBM AUROC 0.75 0.68 0.63 0.71 0.73 Accuracy 0.76 0.69 0.64 0.72 0.75 Sensitivity 0.79 0.71 0.67 0.76 0.74 Specificity 0.74 0.69 0.62 0.70 0.75 PPV 0.63 0.55 0.49 0.59 0.62 NPV 0.87 0.81 0.77 0.84 0.84 FPR 0.21 0.29 0.33 0.24 0.26 F1-score 0.7 0.62 0.57 0.66 0.68 Scaled_Up c-forest AvNNET C5.0 Tree C5.0 Rules GBM AUROC 0.78 0.73 0.65 0.70 0.80 Accuracy 0.80 0.75 0.69 0.73 0.82 Sensitivity 0.76 0.66 0.48 0.59 0.76 Specificity 0.82 0.80 0.80 0.81 0.85 PPV 0.70 0.64 0.57 0.63 0.73 NPV 0.86 0.81 0.74 0.78 0.86 FPR 0.24 0.34 0.52 0.41 0.24 F1-score 0.73 0.65 0.52 0.61 0.75 PCA Nnet avNNET C5.0 Tree C5.0 Rules svmRadialSigma AUROC 0.65 0.73 0.68 0.71 0.71 Accuracy 0.67 0.75 0.7 0.74 0.74 Sensitivity 0.60 0.60 0.66 0.67 0.52 Specificity 0.71 0.84 0.72 0.77 0.86 PPV 0.54 0.67 0.57 0.62 0.67 NPV 0.77 0.79 0.79 0.81 0.76 FPR 0.4 0.4 0.34 0.33 0.48 F1-score 0.57 0.64 0.61 0.64 0.58 PCA_Down Nnet avNNET C5.0 Tree C5.0 Rules svmRadialSigma AUROC 0.70 0.68 0.59 0.61 0.69 Accuracy 0.71 0.69 0.61 0.63 0.70 Sensitivity 0.76 0.71 0.55 0.60 0.72 Specificity 0.68 0.69 0.64 0.64 0.69 PPV 0.56 0.55 0.46 0.48 0.56 NPV 0.84 0.81 0.72 0.74 0.82 FPR 0.24 0.29 0.45 0.4 0.28 F1-score 0.65 0.62 0.50 0.53 0.63 PCA_Up Nnet avNNET C5.0 Tree C5.0 Rules svmRadialSigma AUROC 0.67 0.75 0.56 0.61 0.69 Accuracy 0.7 0.77 0.59 0.63 0.71 Sensitivity 0.59 0.64 0.48 0.55 0.64 Specificity 0.76 0.84 0.65 0.68 0.75 PPV 0.58 0.69 0.43 0.48 0.59 NPV 0.77 0.81 0.69 0.73 0.79 FPR 0.41 0.36 0.52 0.45 0.36 F1-score 0.58 0.66 0.46 0.52 0.61 avNNet: model averaged Neural Network; C5.0 Rules (single C5.0 Ruleset); C5.0 Tree (single C5.0 Tree); c-forest (conditional inference random forest); GBM (stochastic gradient boosting machine); nnet (neuronal network); svmRadialSigma (support vector machines with the Radial basis function kernel). ijms-17-01215-t003_Table 3Table 3 Comparison of the statistical metrics attained for the best predictor in this work and some of the most common ones in the literature. Perfomance Algorithms c-Forest/ Up-Scaling Classes SBHD2 Robetta KFC2-A KFC2-B CPORT Training Test Training Test Training Test Training Test Training Test Training Test AUROC 0.85 0.78 0.74 0.69 0.62 0.62 0.72 0.66 0.60 0.67 0.54 0.54 Accuracy 0.93 0.80 0.70 0.71 0.66 0.66 0.76 0.71 0.70 0.73 0.49 0.49 Sensitivity 0.93 0.76 0.70 0.70 0.38 0.29 0.57 0.53 0.26 0.28 0.55 0.54 Specificity 0.93 0.82 0.70 0.71 0.85 0.88 0.85 0.81 0.93 0.96 0.45 0.47 PPV 0.93 0.70 0.55 0.56 0.61 0.60 0.67 0.59 0.65 0.80 0.34 0.35 NPV 0.93 0.86 0.82 0.82 0.68 0.67 0.79 0.77 0.71 0.72 0.66 0.66 F1-score 0.93 0.73 0.62 0.62 0.47 0.39 0.62 0.56 0.37 0.42 0.42 0.42 ==== Refs References 1. Sudarshan S. Kodathala S.B. Mahadik A.C. Mehta I. Beck B.W. Protein-protein interface detection using the energy centrality relationship (ECR) characteristic of proteins PLoS ONE 2014 9 1215 10.1371/journal.pone.0097115 24830938 2. Phizicky E.M. Fields S. Protein-protein interactions: Methods for detection and analysis Microbiol. Rev. 1995 59 94 123 7708014 3. Clackson T. Wells J.A. A hot spot of binding energy in a hormone-receptor interface Science 1995 267 383 386 10.1126/science.7529940 7529940 4. Uetz P. Giot L. Cagney G. Mansfield T.A. Judson R.S. Knight J.R. Lockshon D. Narayan V. Srinivasan M. Pochart P. A comprehensive analysis of protein-protein interactions in saccharomyces cerevisiae Nature 2000 403 623 627 10688190 5. Cho H. Wu M. Bilgin B. Walton S.P. Chan C. Latest developments in experimental and computational approaches to characterize protein–lipid interactions Proteomics 2012 12 3273 3285 10.1002/pmic.201200255 22997137 6. Moreira I.S. The role of water occulsion for the definition of a protein binding hot-spot Curr. Top. Med. Chem. 2015 15 2068 2079 10.2174/1568026615666150519103733 25986686 7. Cunningham B. Wells J. High-resolution epitope mapping of hgh-receptor interactions by alanine-scanning mutagenesis Science 1989 244 1081 1085 10.1126/science.2471267 2471267 8. Bogan A.A. Thorn K.S. Anatomy of hot spots in protein interfaces 1 J. Mol. Biol. 1998 280 1 9 10.1006/jmbi.1998.1843 9653027 9. Wan H. Li Y. Fan Y. Meng F. Chen C. Zhou Q. A site-directed mutagenesis method particularly useful for creating otherwise difficult-to-make mutants and alanine scanning Anal. Biochem. 2012 420 163 170 10.1016/j.ab.2011.09.019 22001375 10. Massova I. Kollman P.A. Computational alanine scanning to probe protein-protein interactions:  A novel approach to evaluate binding free energies J. Am. Chem. Soc. 1999 121 8133 8143 10.1021/ja990935j 11. Moreira I.S. Fernandes P.A. Ramos M.J. Computational alanine scanning mutagenesis—An improved methodological approach J. Comput. Chem. 2007 28 644 654 10.1002/jcc.20566 17195156 12. Bromberg Y. Rost B. Comprehensive in silico mutagenesis highlights functionally important residues in proteins Bioinformatics 2008 24 i207 i212 10.1093/bioinformatics/btn268 18689826 13. Darnell S.J. Page D. Mitchell J.C. An automated decision-tree approach to predicting protein interaction hot spots Proteins: Struct. Funct. Bioinform. 2007 68 813 823 10.1002/prot.21474 17554779 14. Munteanu C.R. Pimenta A.C. Fernandez-Lozano C. Melo A. Cordeiro M.N.D.S. Moreira I.S. Solvent accessible surface area-based hot-spot detection methods for protein–protein and protein–nucleic acid interfaces J. Chem. Inform. Model. 2015 55 1077 1086 10.1021/ci500760m 25845030 15. Martins J.M. Ramos R.M. Pimenta A.C. Moreira I.S. Solvent-accessible surface area: How well can be applied to hot-spot detection? Proteins: Struct. Funct. Bioinform. 2014 82 479 490 10.1002/prot.24413 24105801 16. Caret: Classification and Regression Training Available online: https://cran.r-project.org/web/packages/caret/index.html (accessed on 25 July 2016) 17. R Development Core Team R: A Language and Environment for Statistical Computing R Foundation for Statistical Computing Vienna, Austria 2010 18. Humphrey W. Dalke A. Schulten K. VMD: Visual molecular dynamics J. Mol. Graph. 1996 14 33 38 10.1016/0263-7855(96)00018-5 8744570 19. Kim D.E. Chivian D. Baker D. Protein structure prediction and analysis using the robetta server Nucleic Acids Res. 2004 32 W526 W531 10.1093/nar/gkh468 15215442 20. Zhu X. Mitchell J. KFC2: A knowledge-based hot spot prediction method based on interface solvation, atomic density and plasticity features Proteins 2011 79 2671 2683 10.1002/prot.23094 21735484 21. De Vries S.J. Bonvin A.M.J.J. Cport: A consensus interface predictor and its performance in prediction-driven docking with haddock PLoS ONE 2011 6 1215 10.1371/journal.pone.0017695 21464987 22. Oshima H. Yasuda S. Yoshidome T. Ikeguchi M. Kinoshita M. Crucial importance of the water-entropy effect in predicting hot spots in protein-protein complexes Phys. Chem. Chem. Phys. 2011 13 16236 16246 10.1039/c1cp21597c 21842056 23. Liu Q. Hoi S. Kwoh C. Wong L. Li J. Integrating water exclusion theory into betacontacts to predict binding free energy changes and binding hot spots BMC Bioinform. 2014 15 57 10.1186/1471-2105-15-57 24568581 24. Guharoy M. Chakrabarti P. Empirical estimation of the energetic contribution of individual interface residues in structures of protein–protein complexes J. Comput. Aided Mol. Des. 2009 23 645 654 10.1007/s10822-009-9282-3 19479323 25. Guharoy M. Pal A. Dasgupta M. Chakrabarti P. Price (protein interface conservation and energetics): A server for the analysis of protein-protein interfaces J. Struct. Funct. Genom. 2011 12 33 41 10.1007/s10969-011-9108-0 21519818 26. Chen H. Zhou H.-X. Prediction of interface residues in protein-protein complexes by a consensus neural network method: Test against NMR data Proteins 2005 61 21 35 10.1002/prot.20514 16080151 27. Chen P. Li J. Wong L. Kuwahara H. Huang J. Gao X. Accurate prediction of hot spot residues through physicochemical characteristics of amino acid sequences Proteins: Struct. Funct. Bioinform. 2013 81 1351 1362 10.1002/prot.24278 23504705 28. Darnell S.J. LeGault L. Mitchell J.C. KFC server: Interactive forecasting of protein interaction hot spots Nucleic Acids Res. 2008 36 W265 W269 10.1093/nar/gkn346 18539611 29. Deng L. Guan J. Wei X. Yi Y. Zhou S. Boosting prediction performance of protein-protein interaction hot spots by using structural neighborhood properties Res. Comput. Mol. Biol. Lecture Notes Comput. Sci. 2013 7821 333 344 30. Cho K. Kim D. Lee D. A feature-based approach to modeling protein–protein interaction hot spots Nucleic Acids Res. 2009 37 2672 2687 10.1093/nar/gkp132 19273533 31. Segura Mora J. Assi S.A. Fernandez-Fuentes N. Presaging critical residues in protein interfaces: A web server to chart hot spots in protein interfaces PLoS ONE 2010 5 1215 10.1371/journal.pone.0012352 20808810 32. Xia J. Zhao X. Song J. Huang D. Apis: Accurate prediction of hot spots in protein interfaces by combining protrusion index with solvent accessibility BMC Bioinform. 2010 11 1215 10.1186/1471-2105-11-174 20377884 33. Wang L. Liu Z. Zhang X. Chen L. Prediction of hot spots in protein interfaces using a random forest model with hybrid features Protein Eng. Des. Sel. 2012 25 119 126 10.1093/protein/gzr066 22258275 34. Xu B. Wei X. Deng L. Guan J. Zhou S. A semi-supervised boosting svm for predicting hot spots at protein-protein interfaces BMC Syst. Biol. 2012 6 10.1186/1752-0509-6-S2-S6 23282146 35. Ozbek P. Soner S. Haliloglu T. Hot spots in a network of functional sites PLoS ONE 2013 8 1215 10.1371/journal.pone.0074320 24023934 36. Strobl C. Malley J. Tutz G. An introduction to recursive partitioning: Rationale, application and characteristics of classification and regression trees, bagging and random forests Psychol. Methods 2009 14 323 348 10.1037/a0016973 19968396 37. Thorn K.S. Bogan A.A. ASEdb: A database of alanine mutations and their effects on the free energy of binding in protein interactions Bioinformatics 2001 17 284 285 10.1093/bioinformatics/17.3.284 11294795 38. Fischer T.B. Arunachalam K.V. Bailey D. Mangual V. Bakhru S. Russo R. Huang D. Paczkowski M. Lalchandani V. Ramachandra C. The binding interface database (BID): A compilation of amino acid hot spots in protein interfaces Bioinformatics 2003 19 1453 1454 10.1093/bioinformatics/btg163 12874065 39. Moal I.H. Fernández-Recio J. Skempi: A structural kinetic and energetic database of mutant protein interactions and its use in empirical models Bioinformatics 2012 28 2600 2607 10.1093/bioinformatics/bts489 22859501 40. Kumar M.D.S. Gromiha M.M. Pint: Protein–protein interactions thermodynamic database Nucleic Acids Res. 2006 34 D195 D198 10.1093/nar/gkj017 16381844 41. Bernstein F.C. Koetzle T.F. Williams G.J. Meyer E.F. Brice M.D. Rodgers J.R. Kennard O. Shimanouchi T. Tasumi M. The protein data bank. A computer-based archival file for macromolecular structures Eur. J. Biochem. 1977 80 319 324 10.1111/j.1432-1033.1977.tb11885.x 923582 42. Miller S. Janin J. Lesk A.M. Chothia C. Interior and surface of monomeric proteins J. Mol. Biol. 1987 196 641 656 10.1016/0022-2836(87)90038-6 3681970 43. Miller S. Lesk A.M. Janin J. Chothia C. The accessible surface area and stability of oligomeric proteins Nature 1987 328 834 836 10.1038/328834a0 3627230 44. Ashkenazy H. Erez E. Martz E. Pupko T. Ben-Tal N. Consurf 2010: Calculating evolutionary conservation in sequence and structure of proteins and nucleic acids Nucleic Acids Res. 2010 38 W529 W533 10.1093/nar/gkq399 20478830 45. Altschul S. Gish W. Miller W. Myers E. Lipman D. Basic local alignment search tool J. Mol. Biol. 1990 215 403 410 10.1016/S0022-2836(05)80360-2 2231712 46. Camacho C. Coulouris G. Avagyan V. Ma N. Papadopoulos J. Bealer K. Madden T.L. Blast+: Architecture and applications BMC Bioinform. 2009 10 1 9 10.1186/1471-2105-10-421 20003500 47. Papageorgiou A.C. Shapiro R. Acharya K.R. Molecular recognition of human angiogenin by placental ribonuclease inhibitor—An x-ray crystallographic study at 2.0 angstrom resolution Embo J. 1997 16 5162 5177 10.1093/emboj/16.17.5162 9311977 48. Huang M. Syed R. Stura E.A. Stone M.J. Stefanko R.S. Ruf W. Edgington T.S. Wilson I.A. The mechanism of an inhibitory antibody on TF-initiated blood coagulation revealed by the crystal structures of human tissue factor, fab 5g9 and tf·5g9 complex1 J. Mol. Biol. 1998 275 873 894 10.1006/jmbi.1997.1512 9480775 49. Buckle A.M. Schreiber G. Fersht A.R. Protein-protein recognition: Crystal structural analysis of a barnase-barstar complex at 2.0-.Ang. Resolution Biochemistry 1994 33 8878 8889 10.1021/bi00196a004 8043575 50. Crystal structure of the E. Coli colicin E9 dnase domain with its cognate immunity protein im9 Available online: http://www.rcsb.org/pdb/explore.do?structureId=1bxi (accessed on 26 July 2016) 51. Scheidig A.J. Hynes T.R. Pelletier L.A. Wells J.A. Kossiakoff A.A. Crystal structures of bovine chymotrypsin and trypsin complexed to the inhibitor domain of alzheimer's amyloid beta-protein precursor (APPI) and basic pancreatic trypsin inhibitor (BPTI): Engineering of inhibitors with altered specificities Protein Sci.: Publ. Protein Soc. 1997 6 1806 1824 10.1002/pro.5560060902 9300481 52. Banner D.W. D′Arcy A. Chène C. Winkler F.K. Guha A. Konigsberg W.H. Nemerson Y. Kirchhofer D. The crystal structure of the complex of blood coagulation factor viia with soluble tissue factor Nature 1996 380 41 46 10.1038/380041a0 8598903 53. Braden B.C. Fields B.A. Ysern X. Dall'Acqua W. Goldbaum F.A. Poljak R.J. Mariuzza R.A. Crystal structure of an fv–fv idiotope–anti-idiotope complex at 1.9 å resolution J. Mol. Biol. 1996 264 137 151 10.1006/jmbi.1996.0629 8950273 54. Fuentes-Prior P. Iwanaga Y. Huber R. Pagila R. Rumennik G. Seto M. Morser J. Light D.R. Bode W. Structural basis for the anticoagulant activity of the thrombin-thrombomodulin complex Nature 2000 404 518 525 10.1038/35006683 10761923 55. Kwong P.D. Wyatt R. Robinson J. Sweet R.W. Sodroski J. Hendrickson W.A. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody Nature 1998 393 648 659 9641677 56. Malby R.L. Tulip W.R. Harley V.R. McKimm-Breschkin J.L. Laver W.G. Webster R.G. Colman P.M. The structure of a complex between the NC10 antibody and influenza virus neuraminidase and comparison with the overlapping binding site of the NC41 antibody Structure 1994 2 733 746 10.1016/S0969-2126(00)00074-5 7994573 57. Bhat T.N. Bentley G.A. Boulot G. Greene M.I. Tello D. Dall'Acqua W. Souchon H. Schwarz F.P. Mariuzza R.A. Poljak R.J. Bound water molecules and conformational stabilization help mediate an antigen-antibody association Proc. Natl. Acad. Sci. USA 1994 91 1089 1093 10.1073/pnas.91.3.1089 8302837 58. Padlan E.A. Silverton E.W. Sheriff S. Cohen G.H. Smithgill S.J. Davies D.R. Structure of an antibody antigen complex: Crystal-structure of the HyHEL-10 Fab-lysozyme complex Proc. Natl. Acad. Sci. USA 1989 86 5938 5942 10.1073/pnas.86.15.5938 2762305 59. Deisenhofer J. Crystallographic refinement and atomic models of a human Fc fragment and its complex with fragment-B of protein-A from staphylococcus-aureus at 2.9- and 2.8-ANG resolution Biochemistry 1981 20 2361 2370 10.1021/bi00512a001 7236608 60. Kobe B. Deisenhofer J. A structural basis of the interactions between leucine-rich repeats and protein ligands Nature 1995 374 183 186 10.1038/374183a0 7877692 61. Emsley J. Knight C.G. Farndale R.W. Barnes M.J. Liddington R.C. Structural basis of collagen recognition by integrin α2β 1 Cell 2000 101 47 56 10.1016/S0092-8674(00)80622-4 10778855 62. Kirsch T. Sebald W. Dreyer M.K. Crystal structure of the BMP-2-BRIA ectodomain complex Nat. Struct. Biol. 2000 7 492 496 10881198 63. Kvansakul M. Hopf M. Ries A. Timpl R. Hohenester E. Structural basis for the high-affinity interaction of nidogen-1 with immunoglobulin-like domain 3 of perlecan Embo J. 2001 20 5342 5346 10.1093/emboj/20.19.5342 11574465 64. Kamada K. Hanaoka F. Burley S.K. Crystal structure of the maze/mazf complex: Molecular bases of antidote-toxin recognition Mol. Cell 2003 11 875 884 10.1016/S1097-2765(03)00097-2 12718874 65. Sauereriksson A.E. Kleywegt G.J. Uhl M. Jones T.A. Crystal-structure of the C2 fragment of streptococcal protein-G in complex with the Fc domain of human-IgG Structure 1995 3 265 278 10.1016/S0969-2126(01)00157-5 7788293 66. Kuszewski J. Gronenborn A.M. Clore G.M. Improving the packing and accuracy of nmr structures with a pseudopotential for the radius of gyration J. Am. Chem. Soc. 1999 121 2337 2338 10.1021/ja9843730 67. Zhang E. St Charles R. Tulinsky A. Structure of extracellular tissue factor complexed with factor VIIa inhibited with a BPTI mutant J. Mol. Biol. 1999 285 2089 2104 10.1006/jmbi.1998.2452 9925787 68. Radisky E.S. Kwan G. Lu C.J.K. Koshland D.E. Binding, proteolytic, and crystallographic analyses of mutations at the protease-inhibitor interface of the subtilisin BPN’/chymotrypsin inhibitor 2 complex Biochemistry 2004 43 13648 13656 10.1021/bi048797k 15504027 69. Hage T. Sebald W. Reinemer P. Crystal structure of the interleukin-4/receptor alpha chain complex reveals a mosaic binding interface Cell 1999 97 271 281 10.1016/S0092-8674(00)80736-9 10219247 70. Fields B.A. Malchiodi E.L. Li H.M. Ysern X. Stauffacher C.V. Schlievert P.M. Karjalainen K. Mariuzza R.A. Crystal structure of a t-cell receptor β-chain complexed with a superantigen Nature 1996 384 188 192 10.1038/384188a0 8906797 71. Nishida M. Nagata K. Hachimori Y. Horiuchi M. Ogura K. Mandiyan V. Schlessinger J. Inagaki F. Novel recognition mode between vav and grb2 sh3 domains Embo J. 2001 20 2995 3007 10.1093/emboj/20.12.2995 11406576 72. Gamble T.R. Vajdos F.F. Yoo S.H. Worthylake D.K. Houseweart M. Sundquist W.I. Hill C.P. Crystal structure of human cyclophilin a bound to the amino-terminal domain of HIV-1 capsid Cell 1996 87 1285 1294 10.1016/S0092-8674(00)81823-1 8980234 73. Barinka C. Parry G. Callahan J. Shaw D.E. Kuo A. Bdeir K. Cines D.B. Mazar A. Lubkowski J. Structural basis of interaction between urokinase-type plasminogen activator and its receptor J. Mol. Biol. 2006 363 482 495 10.1016/j.jmb.2006.08.063 16979660 74. Abergel C. Monchois V. Byrne D. Chenivesse S. Lembo F. Lazzaroni J.-C. Claverie J.-M. Structure and evolution of the ivy protein family, unexpected lysozyme inhibitors in gram-negative bacteria Proc. Natl. Acad. Sci. USA 2007 104 6394 6399 10.1073/pnas.0611019104 17405861 75. Nam T.-W. Il Jung H. An Y.J. Park Y.-H. Lee S.H. Seok Y.-J. Cha S.-S. Analyses of MLc-IIBGLc interaction and a plausible molecular mechanism of Mlc inactivation by membrane sequestration Proc. Natl. Acad. Sci. USA 2008 105 3751 3756 10.1073/pnas.0709295105 18319344 76. Meenan N.A.G. Sharma A. Fleishman S.J. MacDonald C.J. Morel B. Boetzel R. Moore G.R. Baker D. Kleanthous C. The structural and energetic basis for high selectivity in a high-affinity protein-protein interaction Proc. Natl. Acad. Sci. USA 2010 107 10080 10085 10.1073/pnas.0910756107 20479265 77. Pelletier H. Kraut J. Crystal-structure of a complex between electron-transfer partners, cytochrome-c peroxidase and cytochrome-c Science 1992 258 1748 1755 10.1126/science.1334573 1334573 78. Prasad L. Waygood E.B. Lee J.S. Delbaere L.T.J. The 2.5 angstrom resolution structure of the jei42 fab fragment hpr complex J. Mol. Biol. 1998 280 829 845 10.1006/jmbi.1998.1888 9671553 79. Ghosh M. Meiss G. Pingoud A.M. London R.E. Pedersen L.C. The nuclease a-inhibitor complex is characterized by a novel metal ion bridge J. Biol. Chem. 2007 282 5682 5690 10.1074/jbc.M605986200 17138564 80. Schutt C.E. Myslik J.C. Rozycki M.D. Goonesekere N.C.W. Lindberg U. The structure of crystalline profilin beta-actin Nature 1993 365 810 816 10.1038/365810a0 8413665 81. Misaghi S. Galardy P.J. Meester W.J.N. Ovaa H. Ploegh H.L. Gaudet R. Structure of the ubiquitin hydrolase uch-l3 complexed with a suicide substrate J. Biol. Chem. 2005 280 1512 1520 10.1074/jbc.M410770200 15531586 82. Sundquist W.I. Schubert H.L. Kelly B.N. Hill G.C. Holton J.M. Hill C.P. Ubiquitin recognition by the human tsg101 protein Mol. Cell 2004 13 783 789 10.1016/S1097-2765(04)00129-7 15053872 83. Huang L. Hofer F. Martin G.S. Kim S.H. Structural basis for the interaction of ras with raigds Nat. Struct. Biol. 1998 5 422 426 10.1038/nsb0698-422 9628477 84. Hart P.J. Deep S. Taylor A.B. Shu Z.Y. Hinck C.S. Hinck A.P. Crystal structure of the human TβR2 ectodomain-TGF-β3 complex Nat. Struct. Biol. 2002 9 203 208 10.1038/nsb766 11850637 85. Bravo J. Li Z. Speck N.A. Warren A.J. The leukemia-associated AML1 (Runx1)-CBFβ complex functions as a DNA-induced molecular clamp Nat. Struct. Mol. Biol. 2001 8 371 378 10.1038/86264 11276260 86. Gouet P. Chinardet N. Welch M. Guillet V. Cabantous S. Birck C. Mourey L. Samama J.P. Further insights into the mechanism of function of the response regulator chey from crystallographic studies of the chey-chea(124–257) complex Acta Crystallogr. Sect. D-Biol. Crystallogr. 2001 57 44 51 10.1107/S090744490001492X 11134926 87. Schneider E.L. Lee M.S. Baharuddin A. Goetz D.H. Farady C.J. Ward M. Wang C.-I. Craik C.S. A reverse binding motif that contributes to specific protease inhibition by antibodies J. Mol. Biol. 2012 415 699 715 10.1016/j.jmb.2011.11.036 22154938 88. Hanson W.M. Domek G.J. Horvath M.P. Goldenberg D.P. Rigidification of a flexible protease inhibitor variant upon binding to trypsin J. Mol. Biol. 2007 366 230 243 10.1016/j.jmb.2006.11.003 17157870 89. Johnson R.J. McCoy J.G. Bingman C.A. Phillips G.N. Jr. Raines R.T. Inhibition of human pancreatic ribonuclease by the human ribonuclease inhibitor protein J. Mol. Biol. 2007 368 434 449 10.1016/j.jmb.2007.02.005 17350650 90. Bode W. Wei A.Z. Huber R. Meyer E. Travis J. Neumann S. X-ray crystal-structure of the complex of human-leukocyte elastase (pmn elastase) and the 3rd domain of the turkey ovomucoid inhibitor Embo J. 986 5 2453 2458 3640709 91. Read R.J. Fujinaga M. Sielecki A.R. James M.N.G. Structure of the complex of streptomyces-griseus protease-b and the 3rd domain of the turkey ovomucoid inhibitor at 1.8-a resolution Biochemistry 1983 22 4420 4433 10.1021/bi00288a012 6414511 92. Hammel M. Sfyroera G. Ricklin D. Magotti P. Lambris J.D. Geisbrecht B.V. A structural basis for complement inhibition by staphylococcus aureus Nat. Immunol. 2007 8 430 437 10.1038/ni1450 17351618 93. Iyer S. Wei S. Brew K. Acharya K.R. Crystal structure of the catalytic domain of matrix metalloproteinase-1 in complex with the inhibitory domain of tissue inhibitor of metalloproteinase-1 J. Biol. Chem. 2007 282 364 371 10.1074/jbc.M607625200 17050530 94. Zhang J.-l. Qiu L.-y. Kotzsch A. Weidauer S. Patterson L. Hammerschmidt M. Sebald W. Mueller T.D. Crystal structure analysis reveals how the chordin family member crossveinless 2 blocks BMP-2 receptor binding Dev. Cell 2008 14 739 750 10.1016/j.devcel.2008.02.017 18477456 95. Friedrich R. Fuentes-Prior P. Ong E. Coombs G. Hunter M. Oehler R. Pierson D. Gonzalez R. Huber R. Bode W. Catalytic domain structures of MT-SP1/matriptase, a matrix-degrading transmembrane serine proteinase J. Biol. Chem. 2002 277 2160 2168 10.1074/jbc.M109830200 11696548 96. Farady C.J. Egea P.F. Schneider E.L. Darragh M.R. Craik C.S. Structure of an Fab-protease complex reveals a highly specific non-canonical mechanism of inhibition J. Mol. Biol. 2008 380 351 360 10.1016/j.jmb.2008.05.009 18514224 97. Li Y.L. Li H.M. Smith-Gill S.J. Mariuzza R.A. Three-dimensional structures of the free and antigen-bound Fab from monoclonal antilysozyme antibody hyhel-63 Biochemistry 2000 39 6296 6309 10.1021/bi000054l 10828942 98. Reynolds K.A. Thomson J.M. Corbett K.D. Bethel C.R. Berger J.M. Kirsch J.F. Bonomo R.A. Handel T.M. Structural and computational characterization of the SHV-1 β-lactamase-β lactamase inhibitor protein interface J. Biol. Chem. 2006 281 26745 26753 10.1074/jbc.M603878200 16809340 99. Fujinaga M. Sielecki A.R. Read R.J. Ardelt W. Laskowski M. James M.N.G. Crystal and molecular-structures of the complex of α-chymotrypsin with its inhibitor turkey ovomucoid 3rd domain at 1.8 a resolution J. Mol. Biol. 1987 195 397 418 10.1016/0022-2836(87)90659-0 3477645

PMC005xxxxxx/PMC5000614.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081216ijms-17-01216ArticleDistinct Effects of Nalmefene on Dopamine Uptake Rates and Kappa Opioid Receptor Activity in the Nucleus Accumbens Following Chronic Intermittent Ethanol Exposure Rose Jamie H. 1Karkhanis Anushree N. 1Steiniger-Brach Björn 2Jones Sara R. 1*Singh Ashok K. Academic Editor1 Department of Physiology and Pharmacology Wake Forest School of Medicine, Winston-Salem, NC 27157, USA; [email protected] (J.H.R.); [email protected] (A.N.K.)2 H. Lundbeck A/S, Ottiliavej 9, 2500 Valby, Denmark; [email protected]* Correspondence: [email protected]; Tel.: +1-336-716-8533; Fax: +1-336-716-850127 7 2016 8 2016 17 8 121631 5 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The development of pharmacotherapeutics that reduce relapse to alcohol drinking in patients with alcohol dependence is of considerable research interest. Preclinical data support a role for nucleus accumbens (NAc) κ opioid receptors (KOR) in chronic intermittent ethanol (CIE) exposure-induced increases in ethanol intake. Nalmefene, a high-affinity KOR partial agonist, reduces drinking in at-risk patients and relapse drinking in rodents, potentially due to its effects on NAc KORs. However, the effects of nalmefene on accumbal dopamine transmission and KOR function are poorly understood. We investigated the effects of nalmefene on dopamine transmission and KORs using fast scan cyclic voltammetry in NAc brain slices from male C57BL/6J mice following five weeks of CIE or air exposure. Nalmefene concentration-dependently reduced dopamine release similarly in air and CIE groups, suggesting that dynorphin tone may not be present in brain slices. Further, nalmefene attenuated dopamine uptake rates to a greater extent in brain slices from CIE-exposed mice, suggesting that dopamine transporter-KOR interactions may be fundamentally altered following CIE. Additionally, nalmefene reversed the dopamine-decreasing effects of a maximal concentration of a KOR agonist selectively in brain slices of CIE-exposed mice. It is possible that nalmefene may attenuate withdrawal-induced increases in ethanol consumption by modulation of dopamine transmission through KORs. C57BL/6mousevoltammetryreleasepartial agonistdynorphindopaminealcohol ==== Body 1. Introduction Chronic alcohol use disorders are an enormous economic and financial burden in the United States [1]. A large body of literature has shown that chronic alcohol exposure down-regulates dopamine transmission in the nucleus accumbens (NAc) mice: [2]; rats: [3,4]; humans: [5,6], potentially leading to the negative affective states experienced during withdrawal [7,8]. It is plausible that attenuated dopamine neurotransmission following chronic ethanol exposure and withdrawal may be driven, at least in part, by increased function of inhibitory receptors on dopamine terminals in the NAc [2,9,10,11]. The dynorphin/Kappa opioid receptor (KOR) system is of particular interest due to its involvement in modulating ethanol drinking behaviors. For example, prodynorphin knockout mice showed lower preference for ethanol and consumed lower amounts of ethanol compared to wild-type mice [12]. Furthermore, multiple studies have shown increases in NAc KOR function following chronic ethanol exposure [9,10,13,14]. In fact, augmented NAc KOR function may mediate CIE-induced reductions in dopamine terminal function and increased ethanol consumption following extended ethanol exposure and withdrawal [15,16]. Furthermore, intra-NAc KOR blockade reduces relapse-like drinking behavior in rodents [16]. These data suggest that the dynorphin/KOR system may be a promising pharmacotherapeutic target to reduce relapse drinking in patients with alcohol dependence. Nalmefene (6-methylene naltrexone) is a pharmacotherapeutic agent approved in the European Union to combat heavy alcohol drinking in at-risk patients [17] (25 February 2013). Preclinical work showed that intra-NAc infusions of nalmefene reduced ethanol intake at lower doses in ethanol-dependent ethanol self-administering rats than in non-dependent animals [16], an effect that was credited to the partial agonist activity and high affinity of nalmefene at KORs [16,18]. Despite this behavioral evidence, the effects of nalmefene on dopamine terminal and NAc KOR function following chronic ethanol exposure are poorly understood, and elucidation of its effects may aid in understanding the basis of its clinical efficacy. To this end, ex vivo fast scan cyclic voltammetry (FSCV) was used to evaluate the effects of nalmefene on dopamine transmission and KOR function in the NAc core, 72 h following five weeks of chronic intermittent ethanol (CIE) exposure, in male C57BL/6 (C57) mice. 2. Results 2.1. CIE Exposure Reduced Dopamine Transmission in the NAc Core We maintained BECs at behaviorally and physiologically relevant levels (219.50 ± 39.31 mg/dL), as per [19]. Representative FSCV traces are overlaid in Figure 1A (Air: blue trace; CIE: red trace). Consistent with previous work from our laboratory [2,9] CIE reduced dopamine release (Figure 1B, t18 = 2.38 p < 0.05; Air: 1.06 ± 0.26 µM; CIE: 0.71 ± 0.40 µM) and increased rates of dopamine uptake (Figure 1C, t17 = 3.80, p < 0.05; Air: 1.52 ± 0.32 µM/s; CIE: 2.46 ± 0.68 µM/s) compared to air-exposed controls. 2.2. Nalmefene Slowed Dopamine Uptake Rates More in Brain Slices from CIE-Exposed Mice Than Controls To examine the effects of nalmefene on dopamine release and uptake following ethanol vapor and air exposure, increasing concentrations of this compound were bath-applied to accumbal brain slices. A comparison of percent change in dopamine uptake rate showed that nalmefene dose-dependently reduced uptake rates in both groups (Figure 2A, F4,8 = 5.29, p < 0.01), although this effect was greater in CIE exposed animals (F1,8 = 7.94, p < 0.05). No interaction between these factors was detected (F4,8 = 5.40, p > 0.05). When the absolute values of dopamine uptake rates after application of nalmefene were compared across the two groups, there were no differences (Figure 2B). Moreover, two-way RM ANOVA revealed a main effect of nalmefene concentration on dopamine release (Figure 2C, F4,9 = 48.34, p < 0.001), which was similar between inhalation groups (F1,9 = 7.27, p > 0.05). A separate group of brain slices were incubated in norBNI for 60 min to assess the effects of KOR blockade on dopamine transmission. A two-way ANOVA showed no effect of norBNI on dopamine release (Figure 2D, F1,24 = 0.12, p > 0.05), while a main effect of inhalation treatment on uptake rates (Figure 2E, F1,26 = 4.40, p < 0.05) was detected. Bonferroni post hoc analysis revealed a significant difference between the effects of norBNI on CIE-exposed mice compared to air-exposed controls (p < 0.05). 2.3. Nalmefene Reversed the Dopamine-Decreasing Effects of U50,488 in CIE-Exposed Mice Representative traces showing the effects of 0.3 µM U50,488 on dopamine release and uptake (Figure 3A, blue, air; Figure 3B, red, CIE) and 10.0 µM nalmefene reversal (black line, overlaid). The effects of CIE on KOR function were examined with increasing concentrations of U50,488. A RM two-way ANOVA revealed a main effect of KOR activation on dopamine release that was dose-dependent (Figure 3C, F3,10 = 31.69, p < 0.001), which was greater in brain slices from mice exposed to CIE (F1,10 = 6.26, p < 0.05). An interaction between these factors was also detected (F3,10 = 5.51, p < 0.01). Bonferroni post hoc analysis revealed a significant difference between inhalation groups at the 0.1 µM (p < 0.01) and 0.3 µM (p < 0.05) U50,488 concentrations. As nalmefene is a partial KOR agonist, it competes for the endogenous ligand binding site on the receptor [18,20]. To determine the ability of nalmefene to reverse the dopamine-decreasing effects of KOR activation with an exogenous ligand, 10.0 µM of the compound was added to the bath solution following the 0.3 µM concentration of U50,488. A two-way ANOVA, with U50,488 and nalmefene as factors revealed a main effect of drug (U50,488 vs. nalmefene; Figure 3D, F1,10 = 5.67, p < 0.05), as well as an interaction between drug and inhalation treatment (F1,10 = 6.19, p < 0.05). An effect of inhalation treatment alone was not detected (F1,10 = 2.46, p > 0.05). Bonferroni post hoc analysis revealed a significant increase in dopamine release due to nalmefene reversal in the CIE-exposed group compared to the 0.3 µM U50,488 concentration (p < 0.01). As nalmefene reduced dopamine uptake rates to a greater extent in brain slices from CIE-exposed mice compared to air-exposed controls, we examined the effects of a nalmefene challenge on U50,488-induced reductions in dopamine uptake rates. Surprisingly, a two way RM ANOVA revealed no effect of U50,488 on dopamine uptake rates (Figure 3E, F3,8 = 1.26, p > 0.05) in either inhalation group (F1,8 = 0.17, p > 0.05). Further, a two-way RM ANOVA revealed that nalmefene had no effect on dopamine uptake rates following the maximal (0.3 µM) concentration of U50,488 (Figure 3F, F1,16 = 0.04, p > 0.05). 3. Discussion The present study aimed to discern the pharmacological effects of nalmefene on dopamine terminal and KOR function following CIE exposure in mice (Figure 4). Congruent with previous work [2,3,4,9,11], CIE exposure reduced dopamine release, augmented rates of dopamine uptake and increased κ opioid system sensitivity to an agonist, promoting a hypodopaminergic state of the NAc. Nalmefene concentration-dependently reduced dopamine release similarly between inhalation conditions, but attenuated uptake rates more in brain slices from CIE-exposed mice compared to controls. Additionally, we found that a single concentration of nalmefene reversed the dopamine-decreasing effects of a maximal concentration of the KOR agonist U50,488 selectively in brain slices from CIE-exposed mice. These data are the first to demonstrate dopamine terminal modulation by nalmefene and point to mechanisms that may underlie nalmefene-induced reductions in ethanol intake following CIE exposure [15,16]. 3.1. Nalmefene Reduced Dopamine Release Equally in Both Inhalation Groups, but Attenuated Dopamine Uptake Rates More in Brain Slices of CIE-Exposed Mice Our data show similar dopamine-decreasing effects of nalmefene between inhalation groups, indicating that dynorphin tone may not be present in striatal brain slices. If dynorphin tone were to be present, nalmefene would compete with the endogenous ligand for receptor occupancy [21], resulting in antagonist-like effects, which could be measured voltammetrically as an increase in stimulated dopamine release [22]. Notably, control experiments using norBNI in this study showed no effect of KOR blockade on dopamine release in either inhalation group, further suggesting that dynorphin tone is not measurable in brain slices with the current technique. Therefore, it seems that nalmefene was acting as an agonist at KORs to reduce dopamine release [18]. In addition to being a partial KOR agonist, nalmefene also has antagonistic activity at mu (MOR) and delta (DOR) opioid receptors [18], making definitive designation of its effects to any one opioid receptor difficult. It has been shown using in vivo microdialysis that local activation of DORs and MORs results in an increase in extracellular dopamine [23]. However, DORs do not directly modulate dopamine afferents [24]. In fact, DOR activation-induced efflux of dopamine has been shown to occur via a mechanism involving the glutamate system as blockade of N-methyl-d-aspartate receptors in the presence of DOR agonist inhibited dopamine release [25]. Therefore, a direct DOR-driven modulation of presynaptic dopamine transmission is unlikely. Inhibitory MORs are primarily localized to GABAergic interneurons that feed onto dopamine terminals [26] and, therefore, it is possible that the dopamine decreasing effect of nalmefene observed in the current study could be due to a combination of MORs and KOR activity. However, a previous study showed that a MOR agonist reduced dopamine release evoked by single pulse stimulation, while increasing dopamine release evoked by phasic stimulation [27]. Since we used single pulse electrical stimulations in the current study and nalmefene is a MOR antagonist, it is unlikely that the reduction in dopamine release is via a mechanism involving the MORs. Overall however, nalmefene modulation of DORs/MORs, and its influence on dopamine transmission is not known in this context and some contribution of these receptors to the present findings cannot be completely ruled out. Although the effects of nalmefene on dopamine release were similar between inhalation groups, this compound concentration-dependently reduced dopamine uptake rates more in brain slices from CIE-exposed mice compared to controls. In this instance, it appears as though nalmefene is acting to reduce dopamine release and uptake through KORs. Notably, dopamine transporters have been consistently reported to be functionally upregulated following CIE exposure [2,3,4,9]. Here, we found that KOR blockade slowed uptake rates in brain slices from CIE-exposed mice compared to air-exposed controls using a single concentration of norBNI, which may be due to its documented effects on KOR-related intracellular signaling cascades [28]. Dopamine uptake has been previously reported to be altered by KOR agonists, given either acutely [29] or chronically [30,31]. A recent study showed that KORs exist both independently and in complex with dopamine transporters, and regulate dopamine transporter function via an ERK1/2-dependent pathway [29]. Therefore, we hypothesize that the physical or functional connection between dopamine transporters and KORs [29] is fundamentally altered following CIE, driving KOR antagonist-mediated reductions in uptake rates in the present experiments. However, using voltammetric methods in slices, we also found that KOR activation did not augment [29,30] or attenuate [32,33] dopamine uptake rates as reported previously, but eliminated nalmefene-induced reductions in dopamine uptake. Additional investigations into the interplay of U50,488 and nalmefene on KOR-induced alterations in dopamine uptake rates are needed to fully elucidate these findings. 3.2. Dopamine Release Is Restored by Nalmefene Following KOR Activation in Brain Slices from CIE-Exposed Mice To better understand the distinct effects of nalmefene on KOR function between the inhalation groups, a high concentration of nalmefene was applied to brain slices following a maximal concentration of U50,488. As earlier experiments suggested that dynorphin tone is undetectable in brain slices, the addition of U50,488 was necessary to examine any antagonistic effects of nalmefene on KORs. We found that nalmefene reversed the dopamine-decreasing effects of U50,488 in brain slices from CIE-exposed mice, suggesting that nalmefene competed with U50,488 for KOR occupancy. It is possible that this effect is due to CIE-induced receptor upregulation or functional supersensitivity of the kappa opioid system compared to air-exposed animals. In other work, CIE exposure increased levels of KOR binding in seizure-resistant mice [34] and augmented dynorphin-stimulated KOR activity in Wistar rats [13] compared to controls. Therefore, it is plausible that the observed reversal of the dopamine-decreasing effects of U50,488 in brain slices from CIE-exposed mice could be due to the antagonist capabilities of nalmefene interacting with the physical or functional upregulation of KORs. 3.3. Comparison of norBNI and Nalmefene Compounds that alleviate ethanol-induced reductions in dopamine transmission following chronic ethanol exposure, such as those that manipulate KORs, may assuage withdrawal symptoms. Previous work has shown that norBNI blocks withdrawal-induced increases in brain reward thresholds following chronic ethanol [7] and cocaine [35] exposure, and reduces ethanol withdrawal-induced increases in drinking via systemic [9,15] and intra-NAc [16] administration in rodents. Although one study showed increased ethanol intake in animals following norBNI administration [36] it should be noted that rats in that study had continuous access to ethanol and did not undergo withdrawal, unlike studies that show a norBNI-induced reduction in ethanol intake in intermittently exposed, dependent animals [9,15,16]. Moreover, conditioned place preference was absent in mice lacking the prodynorphin gene and mice administered norBNI [37]; however, in that study, conditioned place preference was not tested in a model of alcohol dependence and the mice likely did not experience withdrawal. These data indicate that KOR blockade alleviates withdrawal-induced reductions in dopamine system function in a direct and behaviorally-relevant manner. It is possible that norBNI induces these changes by reducing the effects of endogenous dynorphin on KORs or attenuating the intrinsic activity of functionally upregulated KORs [9,10,11], present work. For example, norBNI has been shown to reverse the low-dopamine state of the animals exposed to chronic stress, which then potentially results in a reduction in ethanol intake [38]. Similar to norBNI, systemic [15] and intra-NAc [16] nalmefene reduces withdrawal-induced increases in ethanol drinking in rodents and drinking in alcoholics [39,40]. As a partial agonist, nalmefene provides antagonist-like effects in the presence of a full agonist [21]. The ability of nalmefene to reverse the dopamine decreasing effects of U50,488 in brain slices from CIE-exposed mice compared to controls is likely innate to an upregulation in KOR function following CIE exposure [9,10,11], present work. Secondly, nalmefene reduced dopamine uptake to a greater extent in brain slices from CIE exposed animals. We hypothesize that CIE alters downstream signaling or physical interactions between receptors and dopamine transporters, as reported previously [29]. In summary, it is likely that norBNI is effective by increasing tonic levels of dopamine at baseline while nalmefene is effective via a reduction in dopamine uptake. 3.4. Behavioral Implications of the Effects of Nalmefene on Dopamine Terminal Function Intra-cerebroventricular [15] and intra-accumbal [16] nalmefene reduces ethanol intake at lower doses in ethanol dependent animals, compared to non-dependent animals, in part due to its high affinity and the unique action of this compound on KORs. It is plausible that nalmefene-induced reductions in ethanol drinking may be due to reductions in ethanol withdrawal-induced hypodopaminergia via attenuated uptake rates and increased dopamine release, selectively in brain slices from CIE-exposed mice ex vivo. In fact, dopamine transporter knockout mice, with inherently reduced rates of dopamine clearance compared to wild-type mice [41] consume less ethanol than their wild-type and heterozygous counterparts [42], providing evidence to support this hypothesis. The time-course of nalmefene-induced alterations in neurobiology is rapid and beneficial in the therapeutic application of this compound. In fact, ethanol intake in ethanol dependent rats was reduced with lower doses of nalmefene than non-dependent animals [15], and occurred with administration approximately 30 min prior to ethanol self-administration testing. In fact, one study showed that nalmefene was absorbed within one hour of administration in healthy subjects following single and multiple dosing schedules [43]. Thus, it is not surprising that alcohol dependent individuals with high or very high drinking risk level who are prescribed nalmefene take the compound orally when they predict they will encounter a high-risk situation (i.e., social environments where alcohol may be present), and consistently report reductions in overall alcohol consumption [40,44,45,46]. Use of nalmefene on a continuous schedule dose-dependently reduces alcohol intake over time [39,47] and attenuates relapse to heavy drinking [40,47] in treatment-seeking patients with alcohol dependence. Together, preclinical and clinical evidence strongly suggest that nalmefene reduces ethanol/alcohol intake on a rapid timescale, and data presented here suggest that KOR modulation of dopamine transmission contributes to the behavioral effects of systemic nalmefene administration. 4. Experimental Procedures 4.1. Subjects Male C57 mice (6–8 weeks old, Jackson Labs, Bar Harbor, ME, USA) were used for all experiments. Animals were allowed at least one week of habituation to the housing environment before CIE procedures began. All mice were individually housed and maintained on a 12-h light-dark cycle (lights off at 14:00), with a red-room light illuminated during the animals’ dark cycle. Standard rodent chow and water were available ad libitum, and replaced daily in the CIE exposed group. At 6–8 weeks old, the mice used in the present study are within the range of early adulthood [48,49]. Notably, mice ages considerably across the five weeks of air/CIE exposure (mice were 11–13 weeks of age at the time of sacrifice for ex vivo voltammetry experiments; both procedures are explained in detail below). At the time of sacrifice, mice are sexually mature and can be considered mature adults [48]. It is plausible that CIE exposure would result in differential neurobiological changes in young versus adult or old mice, but since the mice in this study are all considered “adult” in the literature, the differences are likely minimal. Animals were cared for according to the National Institutes of Health guidelines in Association for Assessment and Accreditation of Laboratory Animal Care, and all experimental protocols were approved by the Institutional Animal Care and Use Committee at Wake Forest University School of Medicine. 4.2. Chronic Intermittent Ethanol (CIE) Exposure This study utilized a CIE exposure protocol previously published [9]. Briefly, mice were exposed to ethanol vapor (CIE) or room air for 16 h/day, followed by 8 hours of room air. This procedure was repeated four times before a 72-h abstinence period (one cycle). Cycles were repeated five times. Although alterations in dopamine transmission are observed after a minimum of three cycles of exposure, the current study used five cycles of exposure to maintain continuity with a set of experiments examining the behavioral effects of KOR system manipulation following CIE exposure [9]. Approximately 30 min prior to chamber start time (17:00), air inhalation mice were systemically injected with pyrazole (1.0 mmol/kg; i.p.; Sigma-Aldrich, St. Louis, MO, USA), an alcohol dehydrogenase inhibitor, mixed with saline. Similarly, CIE exposure mice were systemically injected with pyrazole (1.0 mmol/kg; i.p.) mixed with ethanol (1.6 g/kg; i.p.). Mice metabolize ethanol very rapidly, and a combination of pyrazole, a loading dose of ethanol and continuous ethanol vapor exposure is required to maintain blood ethanol concentrations BECs in the desired range for 16 h in the ethanol chamber. BECs were measured the mornings following the first and final inhalation exposure of each cycle. Even though CIE exposure is a non-contingent method, it has been shown to drive augmented compulsive/anxiety-like behavior and an escalation of ethanol drinking, which suggests that ethanol dependence is achieved [9,50]. On the other hand, the other methods of alcohol exposure, such as drinking in the dark, do not usually produce dependence, which is routinely observed in human alcoholics. 4.3. Blood Ethanol Concentration (BEC) Measurement To ensure proper ethanol chamber function and physiologically relevant BECs, a submandibular vein blood draw was performed in ethanol vapor-exposed mice only. Less than 15 µL of blood was collected in BD microtainer tubes lined with lithium heparin (Becton Dickinson and Company, Franklin Lakes, NJ, USA). For BEC measurement, standards and samples were prepared with a commercially available alcohol dehydrogenase assay (Carolina Liquid Chemistries Corporation, Brea, CA, USA) and carefully pipetted into a 96 well plate. Plate analysis was done with SoftMax Pro Software, version 5 (Molecular Devices Corporation, Sunnyvale, CA, USA). Mean BECs were 219.50 ± 39.31 mg/dL. 4.4. Ex Vivo Fast Scan Cyclic Voltammetry (FSCV) FSCV was used to detect CIE-induced changes in dopamine release and uptake, as well as the effects of nalmefene, U50,488, and nor-binaltorphimine (norBNI), a KOR-specific agonist and antagonist, respectively, on dopamine dynamics and kinetics. Briefly, 300-µm-thick coronal brain slices, containing the NAc core were, prepared using a vibrating tissue slicer. Slices were incubated in oxygenated artificial cerebrospinal fluid and heated to 32 °C for approximately 60 min prior to experiment start. A bipolar stimulating electrode (Plastics One, Roanoke, VA, USA) and carbon fiber microelectrode (≈50 µm length, 7 µm radius (Goodfellow Corporation, Berwyn, PA, USA) were placed within 100 µm of each other on the surface of the slice. Dopamine efflux was induced by a single, rectangular, electrical pulse (4.0 ms; 350 µA, monophasic; interstimulus interval: 180 s), and detected by applying a triangular waveform every 100 ms to the recording electrode (−0.4 to +1.2 to −0.4 V vs. Ag/AgCl, 400 V/s). When baseline collections were stable for three consecutive stimulations, nalmefene (1.0–100.0 µM, generously provided by H. Lundbeck A/S), norBNI (1.0 µM, graciously provided by National Institute on Drug Abuse, NIDA) and U50,488 (0.01–0.3 µM, generously provided by NIDA) were cumulatively added to the bath. Following the maximal concentration of U50,488 (0.3 µM), a challenge concentration of nalmefene (10.0 μM) was added to the bath solution. In a separate set of experiments, a single dose of norBNI (1.0 µM) was applied to brain slices to identify the effects of KOR blockade on dopamine release and uptake measures. Clear current versus time plots were obtained using background current subtraction methods. Electrodes were calibrated immediately after experiments by recording their response (in nA) to a known concentration of dopamine (3.0 μM) using a flow-injection system. 4.5. Data Analysis Representative pre-drug traces of electrically-stimulated dopamine release were individually modeled. Data were collapsed across inhalation groups to obtain baseline dopamine release and uptake measures. Dopamine release was calculated as the amount of electrically-evoked dopamine released per stimulation, and Vmax was calculated as the maximal rate of uptake at the dopamine transporter. The apparent affinity of dopamine for the dopamine transporter (apparent Km) remained constant at 160 nM throughout analysis [51]. The effects of KOR ligands on dopamine release and uptake parameters were similarly analyzed. Demon Voltammetry and Analysis software [52] was used to collect and analyze all data. Graphs were created and statistical analyses were applied with GraphPad Prism (version 5, La Jolla, CA, USA). Student’s t-tests were used to analyze the effects of inhalation exposure on dopamine release and uptake rates. Repeated measures (RM) two-way analysis of variance (ANOVA) was used to determine the effects of increasing concentrations of U50,488 and nalmefene on dopamine release and uptake, with inhalation exposure and drug concentration as factors. Additionally, non-RM two-way ANOVAs, with inhalation exposure and drug concentration as factors, were used to determine the ability of nalmefene to reverse the dopamine-decreasing effects of the 0.3 µM U50,488 concentration, the effects of nalmefene on the maximal rate of uptake following 0.3 µM U50,488, as well as the effects of 1.0 µM norBNI on dopamine release and uptake parameters. When a significant main effect was detected, Bonferroni post hoc analysis was applied. 5. Conclusions Due to the high rate of recidivism to alcohol use disorders, the need for effective pharmacotherapies for this disorder is high. Modulation of KORs to restore dopamine system function during alcohol withdrawal is of interest therapeutically [53]. Overwhelming preclinical behavioral evidence [15,16] and clinical work in patients with alcohol dependence [39,40,44,45,47] indicate favorable therapeutic effects of nalmefene on alcohol consumption. Notably, the effects of nalmefene on drinking are due, in part, to its high affinity and partial agonist activity of this compound on KORs, particularly in the NAc [16]. We report that nalmefene reduced dopamine uptake rates and reversed the dopamine-decreasing effects of KOR activation, suggesting that nalmefene may augment dopamine transmission in vivo (Figure 4). Increased accumbal dopamine transmission by nalmefene would attenuate the hypodopaminergic state of this region through reductions in KOR activity and dopamine uptake rates. These mechanisms may underlie nalmefene-induced reductions in ethanol intake via increased dopaminergic function rodents: [9,15,16]; humans: [39,40,44,45,47]. These data promote insight into the pharmacological effects of nalmefene, and may provide a better understanding of its clinical efficacy. Acknowledgments H. Lundbeck A/S, AA014091 (Sara R. Jones); DA035558 (Jamie H. Rose); AA023874 (Anushree N. Karkhanis). Author Contributions Jamie H. Rose, Anushree N. Karkhanis, Björn Steiniger-Brach and Sara R. Jones, developed experiments. Jamie H. Rose executed experiments, analyzed and graphed all data. Jamie H. Rose, Anushree N. Karkhanis and Sara R. Jones wrote the paper. All authors listed sufficiently contributed to and edited the manuscript before submission. Conflict of Interest Part of this work was funded by H. Lundbeck A/S. This corporation provided nalmefene that was utilized in the present work. At the time the study was conducted Björn Steiniger-Brach was a full-time employee of H. Lundbeck A/S. Figure 1 Chronic intermittent ethanol (CIE) exposure reduced dopamine release and increased dopamine uptake in the nucleus accumbens (NAc) core. Representative FSCV traces are overlaid in (A) (Air: blue trace; CIE: red trace); (B) CIE reduced dopamine release in brain slices of the NAc from CIE-exposed mice compared to air-exposed controls; (C) CIE increased dopamine uptake rates (Vmax) in brain slices from CIE-exposed mice compared to controls. * p < 0.05, ** p < 0.05 (CIE: chronic intermittent ethanol). Figure 2 Nalmefene attenuates uptake rates in brain slices from CIE-exposed mice. (A) Nalmefene concentration-dependently attenuated the maximal rate of uptake rate (Vmax) in both groups, although this effect was greater in brain slices from CIE exposed animals compared to air-exposed mice; (B) nalmefene reduced the uptake rate in CIE-exposed animals to a level comparable to that found in air-exposed mice even; (C) nalmefene concentration-dependently decreased dopamine release similarly in the inhalation groups; (D) the KOR antagonist, norbinaltorphimine (norBNI) did not alter dopamine release in either inhalation group; (E) NorBNI reduced uptake rates in brain slices from CIE-exposed mice compared to controls. * p < 0.05 (CIE: chronic intermittent ethanol). Figure 3 Nalmefene reversed the dopamine-decreasing effects of U50,488 in CIE-exposed mice. (A) Representative traces of the effects of 0.3 µM U50,488 on dopamine release in brain slices from air (solid blue) and (B) CIE (solid red) exposed mice. The black line (overlaid) represents the effect of 10.0 µM nalmefene on these signals; (C) KOR activation with U50,488 dose-dependently reduced dopamine release more in brain slices from CIE-exposed mice compared to controls; (C,D) 10 µM nalmefene restored dopamine release following a maximal concentration of the KOR agonist U50,488 in brain slices from CIE-exposed mice to control levels; (E,F) U50,488 did not alter dopamine uptake rates across the concentration response curve, and nalmefene had no additional effect on uptake rates in either inhalation group. * p < 0.05, ** p < 0.01. (CIE: chronic intermittent ethanol; KOR: κ opioid receptor). Figure 4 Schematic diagram of chronic intermittent ethanol (CIE)-induced changes in kappa opioid receptors (KOR), dopamine transporters (DAT), dopamine release, and dynorphin and nalmefene effects on KOR function based on the current hypotheses. In comparison to air-exposed mice (Left), CIE-exposed mice (Middle) have lower dopamine release and greater uptake rates. This is potentially caused by increased KOR responses, as KORs inhibit dopamine release and increase DAT function when activated. The increased responses could be caused by increased levels of endogenous dynorphin released from the postsynaptic medium spiny neurons, an increase in receptor expression on the membranes of the presynaptic dopamine neurons, or both. (Right) shows a synapse from CIE-exposed animals in the presence of nalmefene. Nalmefene slows the KOR agonist-induced augmentation of uptake rates and inhibits the reduction in dopamine release in CIE-exposed animals, thus normalizing the dopamine release and uptake. ==== Refs References 1. Substance Abuse and Mental Health Service Administration The national survey on drug use and health report 2014 Available online: http://store.samhsa.gov/shin/content//NSDUH14-0904/NSDUH14-0904.pdf (accessed on 10 January 2015) 2. Karkhanis A.N. Rose J.H. Huggins K.N. Konstantopoulos J.K. Jones S.R. Chronic intermittent ethanol exposure reduces presynaptic dopamine neurotransmission in the mouse nucleus accumbens Drug Alcohol. Depend. 2015 150 24 30 10.1016/j.drugalcdep.2015.01.019 25765483 3. Budygin E.A. Oleson E.B. Mathews T.A. Läck A.K. Diaz M.R. McCool B.A. Jones S.R. Effects of chronic alcohol exposure on dopamine uptake in rat nucleus accumbens and caudate putamen Psychopharmacology (Berlin) 2007 193 495 501 10.1007/s00213-007-0812-1 17492432 4. Carroll M.R. Rodd Z.A. Murphy J.M. Simon J.R. Chronic ethanol consumption increases dopamine uptake in the nucleus accumbens of high alcohol drinking rats Alcohol 2006 40 103 109 10.1016/j.alcohol.2006.10.003 17307646 5. Volkow N.D. Wang G.J. Fowler J.S. Logan J. Hitzemann R. Ding Y.S. Pappas N. Shea C. Piscani K. Decreases in dopamine receptors but not in dopamine transporters in alcoholics Alcohol. Clin. Exp. Res. 1996 20 1594 1598 10.1111/j.1530-0277.1996.tb05936.x 8986209 6. Volkow N.D. Wang G.J. Telang F. Fowler J.S. Logan J. Jayne M. Ma Y. Pradhan K. Wong C. Profound decreases in dopamine release in striatum in detoxified alcoholics: Possible orbitofrontal involvement J. Neurosci. 2007 27 12700 12706 10.1523/JNEUROSCI.3371-07.2007 18003850 7. Schulteis G. Markou A. Cole M. Koob G.F. Decreased brain reward produced by ethanol withdrawal Proc. Natl. Acad. Sci. USA 1995 92 5880 5884 10.1073/pnas.92.13.5880 7597046 8. Boutros N. Semenova S. Markou A. Adolescent intermittent ethanol exposure diminishes anhedonia during ethanol withdrawal in adulthood Eur. Neuropsychopharmacol. 2014 24 856 864 10.1016/j.euroneuro.2014.01.022 24560005 9. Rose J.H. Karkhanis A.N. Jones S.R. Supersensitive kappa opioid receptors promote ethanol withdrawal-related behaviors and reduce dopamine signaling in the nucleus accumbens Int. J. Neuropsychopharmacol. 2016 19 10.1093/ijnp/pyv127 26625893 10. Siciliano C.A. Calipari E.S. Cuzon Carlson V.C. Helms C.M. Lovinger D.M. Grant K.A. Jones S.R. Voluntary ethanol intake predicts κ-opioid receptor supersensitivity and regionally distinct dopaminergic adaptations in macaques J. Neurosci. 2015 35 5959 5968 10.1523/JNEUROSCI.4820-14.2015 25878269 11. Siciliano C.A. Calipari E.S. Yorgason J.T. Lovinger D.M. Mateo Y. Jimenez V.A. Helms C.M. Grant K.A. Jones S.R. Increased presynaptic regulation of dopamine neurotransmission in the nucleus accumbens core following chronic ethanol self-administration in female macaques Psychopharmacology (Berlin) 2016 233 1435 1443 10.1007/s00213-016-4239-4 26892380 12. Blednov Y.A. Walker D. Martinez M. Harris R.A. Reduced alcohol consumption in mice lacking preprodynorphin Alcohol 2006 40 73 86 10.1016/j.alcohol.2006.12.002 17307643 13. Kissler J.L. Sirohi S. Reis D.J. Jansen H.T. Quock R.M. Smith D.G. Walker B.M. The one-two punch of alcoholism: Role of central amygdala dynorphins/kappa-opioid receptors Biol. Psychiatry 2014 75 774 782 10.1016/j.biopsych.2013.03.014 23611261 14. Walker B.M. Zorrilla E.P. Koob G.F. Systemic κ-opioid receptor antagonism by nor-binaltorphimine reduces dependence-induced excessive alcohol self-administration in rats Addict. Biol. 2011 16 116 119 10.1111/j.1369-1600.2010.00226.x 20579007 15. Walker B.M. Koob G.F. Pharmacological evidence for a motivational role of kappa-opioid systems in ethanol dependence Neuropsychopharmacology 2008 33 643 652 10.1038/sj.npp.1301438 17473837 16. Nealey K.A. Smith A.W. Davis S.M. Smith D.G. Walker B.M. κ-opioid receptors are implicated in the increased potency of intra-accumbens nalmefene in ethanol-dependent rats Neuropharmacology 2011 61 35 42 10.1016/j.neuropharm.2011.02.012 21338616 17. European Medicines Agency’s Committee for Medicinal Products for Human Use European Medicines Agency Recommends Approval of Medicine for Reduction of Alcohol Consumption 2012 Available online: http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/2012/12/WC500136277.pdf (accessed on 5 January 2015) 18. Bart G. Schluger J.H. Borg L. Ho A. Bidlack J.M. Kreek M.J. Nalmefene induced elevation in serum prolactin in normal human volunteers: Partial kappa opioid agonist activity? Neuropsychopharmacology 2005 30 2254 2262 10.1038/sj.npp.1300811 15988468 19. Griffin W.C. 3rd Lopez M.F. Becker H.C. Intensity and durations of chronic ethanol xposure is critical for subsequent escalation of voluntary ethanol drinking in mice Alcohol. Clin. Exp. Res. 2009 33 1893 1900 10.1111/j.1530-0277.2009.01027.x 19673744 20. Zhu B.T. Mechanistic explanation for the unique pharmacologic properties of receptor partial agonists Biomed. Pharmacother. 2005 59 76 89 10.1016/j.biopha.2005.01.010 15795100 21. Calvey T.N. Williams N.E. Drug Action, in Principles and Practice of Pharmacology for Anaesthetists 5th ed. Blackwell Publishing Ltd. Oxford, UK 2009 22. Spanagel R. Herz A. Shippinberg T.S. Opposing tonically active endogenous opioid systems modulate the mesolimbic dopaminergic pathway Proc. Natl. Acad. Sci. USA 1992 89 2046 2050 10.1073/pnas.89.6.2046 1347943 23. Hipólito L. Sánchez-Catalán M.J. Zanolini I. Polache A. Granero L. Shell/core differences in mu- and delta-opioid receptor modulation of dopamine efflux in nucleus accumbens Neuropharmacology 2008 55 183 189 10.1016/j.neuropharm.2008.05.012 18582908 24. Dilts R.P. Kalivas P.W. Autoradiographic localization of delta opioid receptors within the mesocorticolimbic dopamine system using radioiodinated [2-d -penicillamine, 5-d -penicillamine] enkephalin (125I-DPDPE) Synapse 1990 6 121 132 10.1002/syn.890060203 1978419 25. Fusa K. Takahashi I. Watanabe S. Aono Y. Ikeda H. Saigusa T. Nagase H. Suzuki T. Koshikawa N. Cools A.R. The non-peptidic delta opioid receptor agonist TAN-67 enhances dopamine efflux in the nucleus accumbens of freely moving rats via a mechanism that involves both glutamate and free radicals Neuroscience 2005 130 745 755 10.1016/j.neuroscience.2004.10.016 15590157 26. Svingos A.L. Moriwaki A. Wang J.B. Uhl G.R. Pickel V.M. Mu opioid receptors are localized to extrasynaptic plasma membranes of GABAergic neurons and their targets in the rat nucleus accumbens J. Neurosci. 1997 17 2585 2594 9065518 27. Britt J.P. McGehee D.S. Presynaptic opioid and nicotinic receptor modulation of dopamine overflow in the nucleus accumbens J. Neurosci. 2008 28 1672 1681 10.1523/JNEUROSCI.4275-07.2008 18272687 28. Bruchas M.R. Chavkin C. Kinase cascades and ligand-directed signaling at the kappa opioid receptor Psychopharmacology (Berlin) 2010 210 137 147 10.1007/s00213-010-1806-y 20401607 29. Kivell B. Uzelac Z. Sundaramurthy S. Rajamanickam J. Ewald A. Chefer V. Jaligam V. Bolan E. Simonson B. Annamalai B. Salvinorin A regulates dopamine transporter function via a kappa opioid receptor and ERK1/2-dependent mechanism Neuropharmacology 2014 86 228 240 10.1016/j.neuropharm.2014.07.016 25107591 30. Thompson A.C. Zapata A. Justice J.B. Jr. Vaughan R.A. Sharpe L.G. Shippenberg T.S. Kappa-opioid receptor activation modifies dopamine uptake in the nucleus accumbens and opposes the effects of cocaine J. Neurosci. 2000 20 9333 9340 11125013 31. Chefer V.I. Czyzyk T. Bolan E.A. Moron J. Pintar J.E. Shippenberg T.S. Endogenous kappa-opioid receptor systems regulate mesoaccumbal dopamine dynamics and vulnerability to cocaine J. Neurosci. 2005 25 5029 5037 10.1523/JNEUROSCI.0854-05.2005 15901784 32. Das D. Rogers J. Michael-Titus A.T. Comparative study of the effects of mu, delta and kappa opioid agonists on 3H-dopamine uptake in rat striatum and nucleus accumbens Neuropharmacology 1994 33 221 226 10.1016/0028-3908(94)90012-4 8035907 33. Chudapongse N. Kim S.Y. Kramer R.E. Ho I.K. Nonspecific effects of the selective kappa-opioid receptor agonist U-50,488H on dopamine uptake and release in PC12 cells J. Pharmacol. Sci. 2003 93 372 375 10.1254/jphs.93.372 14646257 34. Beadles-Bohling A.S. Wiren K.M. Alteration of kappa-opioid receptor system expression in distinct brain regions of a genetic model of enhanced ethanol withdrawal severity Brain Res. 2005 1046 77 89 10.1016/j.brainres.2005.03.043 15869750 35. Chartoff E. Sawyer A. Rachlin A. Potter D. Pliakas A. Carlezon W.A. Blockade of kappa opioid receptors attenuates the development of depressive-like behaviors induced by cocaine withdrawal in rats Neuropharmacology 2012 62 167 176 10.1016/j.neuropharm.2011.06.014 21736885 36. Mitchell J.M. Liang M.T. Fields H.L. A single injection of the kappa opioid antagonist norbinaltorphimine increases ethanol consumption in rats Psychopharmacology 2005 182 384 392 10.1007/s00213-005-0067-7 16001119 37. Nguyen K. Tseng A. Marquez P. Hamid A. Lutfy K. The role of endogenous dynorphin in ethanol-induced state-dependent CPP Behav. Brain Res. 2012 227 58 63 10.1016/j.bbr.2011.10.035 22074899 38. Karkhanis A.N. Rose J.H. Weiner J.L. Jones S.R. Early-Life Social Isolation Stress Increases Kappa Opioid Receptor Responsiveness and Downregulates the Dopamine System Neuropsychopharmacology 2016 41 2263 2274 10.1038/npp.2016.21 26860203 39. Mason B.J. Ritvo E.C. Morgan R.O. Salvato F.R. Goldberg G. Welch B. Mantero-Atienza E. A double-blind, placebo-controlled pilot study to evaluate the efficacy and safety of oral nalmefene HCl for alcohol dependence Alcohol. Clin. Exp. Res. 1994 18 1162 1167 10.1111/j.1530-0277.1994.tb00098.x 7847600 40. Karhuvaara S. Simojoki K. Virta A. Rosberg M. Löyttyniemi E. Nurminen T. Kallio A. Mäkelä R. Targeted nalmefene with simple medical management in the treatment of heavy drinkers: A randomized double-blind placebo-controlled multicenter study Alcohol. Clin. Exp. Res. 2007 31 1179 1187 10.1111/j.1530-0277.2007.00401.x 17451401 41. Jones S.R. Gainetdinov R.R. Wightman R.M. Caron M. Mechanisms of amphetamine action revealed in mice lacking the dopamine transporter J. Neurosci. 1998 18 1979 1986 9482784 42. Mittleman G. Call S.B. Cockroft J.L. Goldowitz D. Matthews D.B. Blaha C.D. Dopamine dynamics associated with, and resulting from, schedule-induced alcohol self-administration: Analyses in dopamine transporter knockout mice Alcohol 2011 45 325 339 10.1016/j.alcohol.2010.12.006 21354763 43. Ingman K. Hagelberg N. Aalto S. Någren K. Juhakoski A. Karhuvaara S. Kallio A. Oikonen V. Hietala J. Scheinin H. Prolonged central mu-opioid receptor occupancy after single and repeated nalmefene dosing Neuropsychopharmacology 2005 30 2245 2253 10.1038/sj.npp.1300790 15956985 44. Gual A. He Y. Torup L. van den Brink W. Mann K. ESENSE 2 Study Group A randomised, double-blind, placebo-controlled, efficacy study of nalmefene, as-needed use, in patients with alcohol dependence Eur. Neuropsychopharmacol. 2013 23 1432 1442 10.1016/j.euroneuro.2013.02.006 23562264 45. Mann K. Bladström A. Torup L. Gual A. van den Brink W. Extending the treatment options in alcohol dependence: A randomized controlled study of as-needed nalmefene Biol. Psychiatry 2013 73 706 713 10.1016/j.biopsych.2012.10.020 23237314 46. van der Brink W. Aubin H.J. Bladström A. Torup L. Gual A. Mann K. Efficacy of as-needed nalmefene in alcohol-dependent patients with at least a high drinking risk level: Results from a subgroup analysis of two randomized controlled 6-month studies Alcohol Alcohol. 2013 48 570 578 10.1093/alcalc/agt061 23873853 47. Mason B.J. Salvato F.R. Williams L.D. Ritvo E.C. Cutler R.B. A double-blind, placebo-controlled study of oral nalmefene for alcohol dependence Arch. Gen. Psychiatry 1999 56 719 724 10.1001/archpsyc.56.8.719 10435606 48. Flurkey K. Currer J.M. Harrison D.E. Chapter 20: Mouse models in aging research The Mouse in Biomedical Research, History, Wild Mice and Genetics 2nd ed. Fox J.G. Davisson M.T. Quimby F.W. Barthold S.W. Newcomer C.E. Smith A.L. Academic Press Salt Lake City, UT, USA 2007 637 672 49. Workman A.D. Charvet C.J. Clancy B. Darlington R.B. Finlay B.L. Modeling transformations of neurodevelopmental sequences across mammalian species J. Neurosci. 2013 33 7368 7383 10.1523/JNEUROSCI.5746-12.2013 23616543 50. Anderson R.I. Lopez M.F. Becker H.C. Stress-Induced Enhancement of Ethanol Intake in C57BL/6J Mice with a History of Chronic Ethanol Exposure: Involvement of Kappa Opioid Receptors Front. Cell. Neurosci. 2016 10 45 10.3389/fncel.2016.00045 26941607 51. Wu Q. Reith M.E. Wightman R.M. Kawagoe K.T. Garris P.A. Determination of release and uptake parameters from electrically evoked dopamine dynamics measured by real-time voltammetry J. Neurosci. Methods 2001 112 119 133 10.1016/S0165-0270(01)00459-9 11716947 52. Yorgason J.T. España R.A. Jones S.R. Demon voltammetry and analysis software: Analysis of cocaine-induced alterations in dopamine signaling using multiple kinetic measures J. Neurosci. Methods 2011 202 158 164 10.1016/j.jneumeth.2011.03.001 21392532 53. Lichtigfeld F.J. Gillman M.A. Role of dopamine mesolimbic system in opioid action of psychotropic analgesic nitrous oxide in alcohol and drug withdrawal Clin. Neuropharmacol. 1996 19 246 251 10.1097/00002826-199619030-00006 8726543

PMC005xxxxxx/PMC5000615.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081217ijms-17-01217ArticleCharacterization and Discrimination of Ancient Grains: A Metabolomics Approach Righetti Laura 1Rubert Josep 2*Galaverna Gianni 1Folloni Silvia 3Ranieri Roberto 3Stranska-Zachariasova Milena 2Hajslova Jana 2*Dall’Asta Chiara 1*Herrero Miguel Academic EditorSimó Carolina Academic EditorGarcia-Cañas Virginia Academic Editor1 Department of Food Science, University of Parma, Parco Area delle Scienze 95/A, 43124 Parma, Italy; [email protected] (L.R.); [email protected] (G.G.)2 Department of Food Analysis and Nutrition, Faculty of Food and Biochemical Technology, University of Chemistry and Technology, Prague, Technicka 3, 166 28 Prague 6, Czech Republic; [email protected] Open Fields Srl, Strada Consortile 2, Collecchio, 43044 Parma, Italy; [email protected] (S.F.); [email protected] (R.R.)* Correspondence: [email protected] (J.R.); [email protected] (J.H.); [email protected] (C.D.); Tel.: +420-220-444-387 (J.R.); +420-220-443-185 (J.H.); +390-521-905-431 (C.D.)27 7 2016 8 2016 17 8 121730 5 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Hulled, or ancient, wheats were the earliest domesticated wheats by mankind and the ancestors of current wheats. Their cultivation drastically decreased during the 1960s; however, the increasing demand for a healthy and equilibrated diet led to rediscovering these grains. Our aim was to use a non-targeted metabolomic approach to discriminate and characterize similarities and differences between ancient Triticum varieties. For this purpose, 77 hulled wheat samples from three different varieties were collected: Garfagnana T. turgidum var. dicoccum L. (emmer), ID331 T. monococcum L. (einkorn) and Rouquin T. spelta L. (spelt). The ultra high performance liquid chromatography coupled to high resolution tandem mass spectrometry (UHPLC-QTOF) metabolomics approach highlighted a pronounced sample clustering according to the wheat variety, with an excellent predictability (Q2), for all the models built. Fifteen metabolites were tentatively identified based on accurate masses, isotopic pattern, and product ion spectra. Among these, alkylresorcinols (ARs) were found to be significantly higher in spelt and emmer, showing different homologue composition. Furthermore, phosphatidylcholines (PC) and lysophosphatidylcholines (lysoPC) levels were higher in einkorn variety. The results obtained in this study confirmed the importance of ARs as markers to distinguish between Triticum species and revealed their values as cultivar markers, being not affected by the environmental influences. small grainsnon-targeted metabolomicsphenolic lipid compoundslipidomicsfoodomics ==== Body 1. Introduction Cereals represent one of the most important commodities providing basic nutrients to human diet, such as corn, rice, sorghum, or wheat, whose starchy grains are used as food. Cereals are annual plants, and cereal crops must be reseeded for each growing season. These cereal grasses, domesticated during the Neolithic period, formed the basis of systematic agriculture. In the particular case of Triticum species, they have been classified into hulled and free-threshing (“naked”) forms. Among the latter, bread and durum wheat are the most important Triticum species cultivated worldwide [1]. On the one hand, “hulled wheats”, which means that the kernel retains its husk during harvest, were the earliest domesticated wheats by mankind and are the ancestors of current wheats. Ancient wheat cultivation drastically decreased during the 1960s due to dietary and economic changes, as well as the introduction of bread and durum wheat, which are both higher yielding [2]. However, during the past years, the increasing demand for natural and organic products led to the rediscovery of ancient wheat species such as spelt (Triticum spelta L.), emmer (Triticum dicoccum L.), and einkorn (Triticum monococcum L.) [3]. This renewed interest is associated with the desire for a healthy and equilibrated diet, such as the Mediterranean diet. In fact, hulled wheat has been recognized as a dietetic and healthy cereal, and it is recommended in treatment of disease related to high blood cholesterol, colitis, and allergies [3]. A comparison of ancient and standard wheat highlighted that the ancient grains are characterized by a higher content of soluble dietary fiber, proteins, and lipids (mostly unsaturated fatty acids) [4]. In addition, ancient wheats provide a much greater proportion of rapidly digestible starch (RDS) and higher starch digestion index (SDI) compared to bread wheat [5,6]. Concerning trace elements, emmer, einkorn, and spelt mainly differed from wheat cultivars for higher contents of Li, Mg, P, Se, and Zn [7]. Another additional benefit might be connected with a relatively high concentration of antioxidant compounds, which can contribute to the excellent nutritional properties of the hulled wheats. Among these phenolic compounds, alkylresorcinols (ARs) represent one of the major groups that are found in high levels in the outer layers of the kernels [8]. The impact of ARs have been studied for wholegrain wheat and rye, because these layers are mostly removed during flour production [9]. Furthermore, the C17:0/C21:0 ARs homologue ratio has been proposed to differentiate between common and durum wheats [8,9]. Recently, the concentration of saturated ARs allowed the differentiation of Triticum species according to their degrees of ploidy [10]. In particular, the levels of all ARs homologues significantly differed between hexaploid (bread wheat and spelt), tetraploid (durum and emmer), and diploid (einkorn) species. Up to now, targeted methods, developed for quantification of a given class of metabolites, have been exclusively applied to investigate differences between ancient Triticum varieties [9,10]. Nevertheless, nowadays, advanced analytical tools have permitted the simultaneous analysis of hundreds of metabolites, allowing a better characterization of small molecules (up to 1200 Da), therefore, the composition of complex plant matrices can be investigated in-depth [11]. In fact, in the last decade, the applicability of metabolomics to food science and nutrition research has strongly emerged [11,12,13,14,15,16]. In the present study, a metabolomic untargeted method was developed to investigate a broad spectrum of ancient wheats compounds in order to determine the relative roles of genotype and environment in determining the metabolites composition. Identifying similarities and differences that permit to distinguish between ancient Triticum varieties may be useful for the determination of nutritional aspects and adulterations, since emmer and einkorn are more expensive than spelt. For this purpose, 77 hulled wheat samples were analyzed using a non-targeted metabolomics approach based on solid liquid extraction followed by a reversed phase liquid chromatography separation coupled to quadrupole-time-of-flight mass spectrometer (LC-QTOF), and multivariate data analysis. 2. Results 2.1. Multivariate Modeling To perform sample classification, at first all 77 chromatograms were independently aligned for both polarities (see Figure 1). This returned a primary dataset with 4191 and 3253 features for positive and negative modes, respectively. Afterward, data reduction was performed based on previous work [12]. The primary filtering step excluded the background peaks present in blank samples. Then, in order to remove signal redundancy, only monoisotopic peaks were considered. The third filtering step was performed by choosing all the molecular features present in at least 50% of the samples in one group. This last step removed 2051 peaks for positive mode and 1666 for negative mode, representing approximately 50% of the original dataset, leaving 686 and 490 peaks for positive and negative, respectively. At this point, the principal components analysis (PCA) models were built to investigate the metabolome, and therefore, differences between all three classes of wheat. The mechanism, already explained elsewhere [13], is based on the ability of the PC model to cluster samples in an unsupervised approach, since no information on group identity is used to construct the model. The PCA score plot obtained for positive and negative ionization modes are summarized in Figure 2. The first two principal components (PC) explained 50% of the total variance of the ESI(+) (32.9% and 17.1% for the PC1 and PC2, respectively) and 47.2% of the ESI(−) model (25.8% and 21.4% for the PC1 and PC2, respectively). Samples were arranged in three major groups, indicating a sample clustering according to the varieties: emmer, einkorn, and spelt. A more pronounced clustering, among sample classes, was obtained in the ESI(+) data, as it can be seen in the PCA score plot (Figure 2A), even if one sample from ID331 is mixed up with Garfagnana variety. One out of 77 samples fell outside the 95% confidence ellipse, as it is shown in the ESI(−) PCA score plot (Figure 2B). This is considered a “moderate” outlier, while samples out of the confidence interval value of 99% (critical limit) are “strong” outliers. For this reason, this outlier was kept into the data set. No clustering according to vegetative year, growing location, and farming condition was found. The differences between these three varieties were confirmed when partial least squares discriminant analysis (PLS-DA) (see Figure 3) and orthogonal partial least squares discriminant analysis (OPLS-DA) models were constructed. PLS-DA was performed to maximize differences and OPLS to highlight key variables and potential biomarkers. The quality of the models was excellent as shown in Table 1, where all the goodness of fit (R2) and the prediction ability (Q2) parameters are summarized. PLS-DA models highlighted highly quality parameters that were not significantly improved to OPLS-DA models, suggesting a low “structure noise” in the data set. OPLS-DA has the capacity to improve prediction ability because it separates out the structured noise, which is modeled separately. The high Q2 values obtained for both supervised models indicated excellent predictabilities and suggested that the metabolomics approach applied was able to reveal differences between the grain varieties studied. Moreover, in order to avoid the risk of overfitting, each generated model was validated by cross-validation tool [14], using the leave 1/3 out approach. Misclassification tables (see Table S1) indicate that 100% of ancient wheat lines (three out of three) were correctly classified in the ESI(–) data, while in ESI(+) OPLS-DA model the percentage of total correct classification was 98.7%, as one sample was not correctly predicted. 2.2. Discriminant Metabolites Identification In order to obtain relevant information regarding the metabolic differences between the varieties, a limited set of statistically meaningful metabolites had to be selected. In the present study, discriminant markers selection was performed merging the metabolites resulting from the PLS-DA loadings plot with those obtained using the Variable Influence in Projection (VIP threshold > 1.5). The identity of compounds that were found to be significant in class separation was confirmed by ultra high performance liquid chromatography coupled to high resolution tandem mass spectrometry (UHPLC-HRMS) analysis based on accurate MS and MS/MS data, as well as theoretical and experimental isotopic patterns were evaluated in-depth. Features were searched against the METLIN, KEGG, LIPIDMAPS and HMDB online databases [11]. At the same time, empiric formulae of the unknown compounds were calculated by Formula Finder option in Peak View software (version 2.2, SCIEX, Concord, ON, Canada) aiding to confirm or refuse potential structures. Subsequently, comparison of the fragmentation pathway of the proposed compound, found in the above-mentioned databases, with the fragments experimentally obtained confirmed the identity. All metabolites identified are summarized in Table 2 including tentative identification, pseudomolecolar ion, retention time, mass error (ppm), higher metabolite intensities associated with ancient grain varieties, and VIP values. For all metabolites identified calculated mass error (Δppm) was lower than 4 ppm. In the present work, seven statistically significant markers, belonging to the resorcinol’s class, were tentatively identified. The seven ARs were detected in negative ionization mode producing both a [M − H]¯ and the [M + HCOO]¯. Since these metabolites are commonly detected by GC-MS [17] or HPLC-UV [18] techniques, MS/MS spectra were not available in the online database. Thus, we tentatively identified them checking the exact mass (mass error less than 1.7 ppm), the match of experimental and theoretical isotope pattern in terms of spacing and relative intensities, and the most abundant fragment ion [M − C2H2O]¯ yielded from the resorcinol ring, resulting from the neutral loss of 42 Da (see Figure S1) [19]. For the lipid identification, LipidView software (version 1.3 beta, SCIEX) was employed. Diacylglycerols (DGs) and triacylglycerols (TGs) were detected in positive mode as ammonium adducts, giving a pseudomolecular ion [M + NH4]+. Identification of 1-palmitoyl-2-linoleoyl glycerol was based on the accurate m/z 610.5405 [M + NH4]+, theoretical and experimental isotopic patterns and on the product ions m/z 337.2737 and m/z 313.2737 corresponding to the loss of palmitic and linoleic acid, respectively. The mass spectrum of a TG contained two different fatty acids; 1,2-dipalmitoyl-3-linoleoyl glycerol (m/z 848.7708), and two DG ions (m/z 551.5034). Similarly, the MS/MS spectrum of 1-palmitoyl-2-oleoyl-3-eicosenoyl-glycerol (m/z 904.8339), as it contains three different fatty acid species, exhibited three DG ions (m/z 631.5660, m/z 605.5503 and m/z 577.5190) [15,16]. Phospholipids were detected in both ionization modes and confirmed by ESI(+) with a characteristic fragment ion of m/z 184.0739 for phosphatidylcholines (PC) and m/z 184.0739, m/z 104.1078, m/z 86.0974 m/z for lysophosphatidylcholines (lysoPCs) [20]. Lyso PC (16:0) fragmentation pattern that allows identification of the compounds, is depicted in Figure 1. 3. Discussion 3.1. Phenolic Compounds According to our results, ARs composition significantly differs between the studied varieties. In particular, two ARs, C21:0 and C19:0, turned out to be the most useful homologues to discriminate spelt from emmer and einkorn, as illustrated in the variable trend plot (Figure 4A). This is consistent with the results reported in the HEALTHGRAIN study [21], since, among the Triticum spp., spelt showed higher maximum values of ARs content ranging from 490 to 741 µg/g with C21:0 (approximately 47%) and C19:0 (36%) as the predominantly homologues found. Spelt wheat, Rouquin, showed a distribution of AR homologues similar to that of common wheat, in agreement with earlier studies [21,22], being both hexaploid species [10]. By contrast, C23:0 (see Figure 4B) and C25:0 had the highest influence to discriminate emmer variety, Garfagnana, showing the same homologue pattern of durum wheat, characterized by the influence of the longer homologues. These longer AR homologues, which were isolated from a cereal bran-milling fraction, have been found efficient inhibitors of 3-phosphoglycerate dehydrogenase. Note that 3-phosphoglycerate dehydrogenase is a key enzyme of triglyceride synthesis, in adipocytes [23]. Also for this reason, the intake of ARs is considered beneficial as it reduces the absorption of cholesterol, regulate metabolism of triacylglycerols and affect levels of lipid-soluble vitamins [24]. ARs with modified alkyl chains are also present in cereals. These are believed to differ from ARs only in side-chain unsaturation or oxidation. On average, 15%–20% ARs contain unsaturated hydrocarbon chains as well as ketone and hydroxyl groups [21,22,23]. In the present study, two alk(en)yl resorcinols, nonadecenyl-resorcinol, and heneicosenyl-resorcinol, were identified and contributed to the clustering and differentiation of spelt, Rouquin. These AR analogues are suggested to be more bioactive than normal saturated ones [24]. ARs are found mainly in the outer layers (bran fraction) of cereal grains, which means that they are largely missing in refined cereal flour and conventional cereal products. Taking into account that these ancient grain varieties are mostly consumed in the form of whole grain, ARs may be present in food in high enough concentrations to have a bioactive effect. ARs are absorbed in the small intestine of pig with an ileal recovery that varies between 21% and 40%, with no major difference between different chain-length homologues [25]. In fact, their metabolized forms have been found in human plasma and urine [26] suggesting that ARs might exert their biological effect in human after whole grain intake. Interestingly, our data suggests that sample clustering was not affected by growing location, organic or conventional farming and vegetative year, as it was previously reported [27]. Thus, the level of ARs metabolites was identified as a cultivar marker, strongly influenced by the genetic background, which is partially in line with Ziegler et al. [10]. In fact, they reported significant difference in the AR content of spelt grown in different location, whereas einkorn content did not differ among different location. This outcome indicated a strong genetic influence on the AR homologue profile, suggesting that the metabolomics approach applied could potentially allow the determination of ancient wheat adulterations. 3.2. Glycerophospholipids and Glycerolipids Among the statistically significant phospholipids, four molecular species (m/z 496.3399, m/z 520.3392, m/z 760.5851, m/z 758.5712) were found responsible for the separation of einkorn variety. Two PCs, PC (16:0/18:2) and PC (16:0/18:1) were tentatively identified and trend plot of PC (16:0/18:1) is illustrated in Figure 4C. These results are consistent with a previous study [4], as einkorn was reported to show a richer lipid profiling among the ancient varieties, a lipid content 50% higher than those of bread wheat. In fact, PC 34:2 together with lysoPC 16:0 and lysoPC 18:2 are the major PC species detected, representing 60%–70% of the total wheat PC [28]. Acyl carbon and double-bond configurations in phospholipids are probably combination of the major fatty acids, that in einkorn are reported to be linoleic (18:2), oleic (18:1), and palmitic (16:0) acids [4]. In bread wheat, linoleic acid is the prevalent fatty acid too, however palmitic acid is more abundant than oleic acid. Consequently, einkorn lipids profile has a higher content of monounsaturated fatty acids (MUFA), lower content of polyunsaturated fatty acids (PUFA), and lower saturated fatty acids (SFA) that, from a nutritional point of view, contribute to the prevention of cardiovascular diseases, since MUFA and PUFA reduce thrombosis and atherosclerosis risk, influencing lipid and cholesterol synthesis [4]. 4. Materials and Methods 4.1. Chemicals The deionized water used for the LC mobile phase was obtained from a Milli-Q® Integral system supplied by Merck (Darmstadt, Germany). High-performance LC (HPLC)-grade methanol, 2-propanol, dichloromethane, formic acid, and ammonium formate were supplied by Sigma-Aldrich (St. Louis, MO, USA). 4.2. Plant Material For this study, three ancient wheat species have been chosen: Garfagnana T. turgidum var. dicoccum L. (emmer), ID331 T. monococcum L. (einkorn), and Rouquin T. spelta L. (spelt). The most extensively cultivated species is T. turgidum ssp. dicoccum, which was largely grown in the hills and low mountain areas in Central and Southern Italy until the 19th Century, as reported by local tradition. The three varieties were cultivated in two locations in Emilia Romagna region, Parma and Bologna, in plots of 8.25 m2 with four replications. Grains were grown over two consecutive years (2013/2014 and 2014/2015) under two agricultural conditions: conventional (n = 23) and organic farming (n = 30) in Parma, whereas only conventional farming was applied in Bologna (n = 24). After harvesting, the whole grains were dried at ca. 10% humidity, stored at −20 °C and kept refrigerated until the analysis. Overall, 77 wheat samples were collected. 4.3. Metabolite Extraction Wheat samples were ground into a fine powder using a ball mill (MM 301 Retsch, Haan, Germany). An amount of 1 g of ground wheat was weighed into a 50 mL polypropylene centrifugation tube, followed by the addition of 10 mL of a mixture of methanol/dichloromethane (50:50, v/v). After brief shaking, the content was stirred for 30 min at 240 strokes/min by a shaker (IKA Laborartechnik, Stufen, Germany). The tube was centrifuged (13,416 g) for 7 min (Rotina 35 R, Hettich Zentrifugen, Germany), then 1 mL of the extract was evaporated to dryness under a gentle stream of nitrogen. Finally, the residues were re-dissolved in 1 mL of isopropanol/methanol/water (60:30:5, v/v) prior to UHPLC-Q-TOF analysis. During the sample preparation blanks were also prepared for analysis consisting of all the steps mentioned above except for the addition of sample. 4.4. Quality Control (QC) Samples Preparation In order to measure performance and system stability and assess the reproducibility of the sample treatment procedure, Quality Control samples (QC) were injected during the analyses. QCs (n = 2) were obtained by mixing equal volumes (50 μL) of all 77 sample extracts and following the same procedure as for the other samples. QCs were injected at the beginning of the run and after every 10 real samples. 4.5. Ultra-High-Performance Liquid Chromatography-High Resolution Mass Spectrometry UHPLC Dionex UltiMate 3000 RS system (Thermo Fisher Scientific, Waltham, MA, USA) coupled to a TripleTOF® 5600 quadrupole time-of-flight (TOF) mass spectrometer (SCIEX) was employed for untargeted analysis of wheat. The chromatographic separation was performed using an Acquity BEH C18 column (Waters, Milford, MA, USA) 100 mm × 2.1-mm inner diameter, 1.7-μm particle size maintained at 60 °C. The mobile phases for metabolic analysis were the same for negative and positive electrospray ionization (ESI) modes. Gradient elution was performed by using 5 mM ammonium formate in Milli-Q water/methanol (95:5, v/v) (solvent A) and 5 mM ammonium formate in isopropanol/methanol/Milli-Q water (65:30:5, v/v) (solvent B) both acidified with 0.1% formic acid. The following multistep elution gradient was used with both electrospray ionization (ESI) polarities: 0.0 min (10% solvent B; 0.40 mL·min−1) to 1.0 min (50% solvent B; 0.40 mL·min−1), subsequently 1–5 min (80% solvent B; 0.40 mL·min−1), 11.0 min, (100% solvent B; 0.50 mL·min−1). After a 4.5 min isocratic step, the system was re-equilibrated to initial conditions for 2.5 min (10% solvent B; 0.4 mL·min−1). The sample was permanently kept at 5 °C. The ion source was a DuoSpray™ with a separated ESI ion source and APCI. ESI was used for the sample measurement and APCI was used for exact mass calibration of the TripleTOF instrument. In ESI(+) mode, the source parameters for metabolic analysis were as follows: capillary voltage, +4500 V; nebulizing gas pressure, 60 psi; drying gas pressure, 50 psi; temperature, 550 °C; and declustering potential, 80 V. The capillary voltage in ESI(−) mode was −4000 V, and the other source settings were the same as for ESI(+). At the same time, a TOF MS method and information-dependent acquisition (IDA) method were used to collect MS and MS/MS spectra. The method consisted of a survey TOF MS experiment ranged from m/z 100 to 1200, in parallel, Product Ion (PI) spectra for the eight most intense ions of the survey spectra throughout the chromatographic run were recorded. Dynamic background subtraction was activated to acquire PI spectra of real eluted compounds, avoiding background ions. PI spectra were collected for ions ranging from m/z 50 to 1200. The PI spectra were recorded with collision energy of 35 V and collision energy spread of ±15 V was also set. In this way, both low-energy and high-energy fragment ions were present in a single spectrum. The total cycle time of the TOF MS and IDA methods was 0.55 s. An automatic m/z calibration was performed by the calibration delivery system for every five samples using a positive or negative APCI calibration solution (SCIEX) according to the batch polarity. Each set of samples for each polarity was preceded by three blank controls: Milli-Q water, methanol and a blank (extraction procedure without the sample). Finally, the same MS approach was applied in ESI(−) mode. The resolving power achieved was greater than 31,000 (m/z 321.0192) full width at half maximum (FWHM) with both polarities. The PI spectra were measured in high-sensitivity mode, which provides half resolving power. Instrument control and data acquisition were performed with Analyst 1.6 TF (SCIEX), and the qualitative analysis was performed using PeakView 2.2 (SCIEX) equipped with MasterView and Formula Finder and directly linked to the ChemSpider database, and LipidView software (version 1.3 beta, SCIEX) for lipid evaluation. The in-batch sequence of the samples was random (established on the basis of random number generation) to avoid any possible time-dependent changes during UHPLC-HRMS analysis, which would result in false clustering. To address overall process variability, metabolomics studies were augmented to include a set of eight sample technical replicates (10% of the samples set). Reproducibility analysis for compounds detected in these replicates provided a measure of variation for extraction, injection, retention time (RT), and mass accuracy. 4.6. Data Processing and Chemometrics Analysis Data processing has been performed based on previous work [12]. Briefly, MarkerView software (version 1.2.1, SCIEX) was used for data processing (data mining, alignment, filtering, normalization, and Principal Component Analysis (PCA)) of the UHPLC-HRMS records. Data mining was performed based on an automated algorithm using RT window and peak finding; retention time (RT) range 0.4–13 min and m/z range 100–1200. In the next step, RT and m/z alignment of the respective peaks was carried out using RT and m/z tolerances of 0.2 min and 0.02 Da, respectively. Two separate positive and negative ionization data matrices, comprising lists of molecular features (called also peaks by MarkerView) characterized for each sample by (i) RT; (ii) m/z value; (iii) respective intensity and (iv) charge state (monoisotopic and isotopic), were automatically obtained using MarkerView, and subsequently total area sum normalization was performed for each sample. Prior to the actual PCA, data matrices were pre-processed using the Pareto scaling (the square root of the standard deviation is used as the scaling factor). Orthogonal partial least squares discriminant analysis (OPLS-DA) employing the software SIMCA (v. 13.0, 2011, Umetrics, Umea, Sweden) was performed. The quality of the models was evaluated by the goodness-of-fit parameter (R2X), the proportion of the variance of the response variable that is explained by the model (R2Y) and the predictive ability parameter (Q2), which was calculated by a seven-round internal cross validation of the data using a default option of the SIMCA software. R2X and R2Y represent the fraction of the variance of X matrix and Y matrix, respectively, while Q2 suggests the predictive accuracy of the model. R2X, R2Y, and Q2 values close to 1 indicate an excellent model, and thus, from values higher than 0.5 indicate good quality of OPLS-DA models. In order to select the most significant and reliable variables, variable importance in the projection (VIP) was used. This parameter summarizes the importance of the X-variables, both for the X- and Y-models. In this research, VIP with the threshold >1.5 was used for selection of the most significant markers. VIP-values larger than 1 indicate important X-variables. To avoid risk of overfitting, as the results found after Multivariate Data Analysis (MVDA) are sensitive to chance-correlations, statistical models have to be validated. For this reason, supervised models, OPLS-DA and PLS-DA, were validated by cross-validation, using the leave one-third out approach. The data set was divided into three parts and one-third of samples were excluded to build a model with the remaining two-thirds of samples. Excluded samples, one-third of samples, were then predicted by this new model and the procedure was repeated until all samples had been predicted at least once. Each time the percentage of correctly classified samples was calculated by generating a misclassification table. 5. Conclusions In conclusion, differences in the metabolome of ancient grains were successfully detected using an untargeted UHPLC-HRMS metabolomics approach. Discriminant metabolites including alkylresorcinols, glycerophospholipids, and glycerolipids were identified allowing a metabolic characterization of ancient wheat grains. The results obtained in this study confirmed the importance of different AR homologues as markers to distinguish between Triticum species. Furthermore, all the 15 identified molecules were revealed to be cultivar markers, strongly influenced by the genetic background, since their abundance was not significantly affected by growing location, organic or conventional farming, and vegetative year. Acknowledgments The study presented was financially supported by LR 28/98—Emilia Romagna Region under the funding program 49, LR28/98, Call 2013—“AMicoGrano, Analysis of the incidence of Mycotoxins on modern and ancient Grains grown under organic and conventional regimes”. The study was undertaken within the following projects supported by the Ministry of Agricultural of the Czech Republic (QI111B044), by the Operational Programme Prague—Competitiveness (CZ.2.16/3.1.00/21537 and CZ.2.16/3.1.00/24503), and by the “National Program of Sustainability I”—NPU I (LO1601—No. MSMT-43760/2015). Josep Rubert thanks Generalitat Valenciana (Conselleria d’Educació, Cultura i Esport), for the VALi+d postdoctoral fellowship “Contractació de personal investigador en formació en fase postdoctoral 2014” (APOSTD/2014/120). The authors kindly thank Cristina Piazza and Roberto Reggiani from Azienda Agraria Sperimentale Stuard (Parma, Italy) for the support in selecting and harvesting the ancient wheat varieties, and Antonio Rossetti from OpenFields Srl (Collecchio, Parma, Italy) for the essential technical assistance. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1217/s1. Click here for additional data file. Author Contributions Chiara Dall’Asta and Gianni Galaverna designed the experiments and were responsible for the metabolomics study setup. LC-MS conducted analysis, statistical analysis of the data, metabolites identification: Laura Righetti and Josep Rubert. Interpretation of the data: Laura Righetti, Josep Rubert, Milena Stranska-Zachariasova, and Chiara Dall’Asta Sample collection: Silvia Folloni, Roberto Ranieri. Drafting the manuscript for important intellectual content: Laura Righetti, Josep Rubert, Gianni Galaverna, Milena Stranska-Zachariasova, Jana Hajslova, and Chiara Dall’Asta. Conflicts of Interest The authors have declared no conflict of interest. Abbreviations ARs alkylresorcinols QTOF quadrupole time of flight HRMS high resolution mass spectrometer QC quality control PCA principal component analysis PLS-DA projection on latent structure-discriminant analysis OPLS-DA orthogonal projection on latent structure-discriminant analysis DG diacylglycerol TG triacylglycerol PC phosphatidylcholine Lyso PC lysophosphatidylcholine Figure 1 Ultra high performance liquid chromatography coupled to high resolution tandem mass spectrometry base peak chromatograms of ancient wheat extract obtained using positive (A) and negative (B) ionization modes. Extracted ion chromatogram (XIC) of Lyso PC 16:0 ionized in positive ([M + H]+ m/z 496.3399) (C) and negative ([M + HCOO]− m/z 540.3332) (D) modes. Product ions acquired automatically by the information-dependent acquisition (IDA) method for the m/z 496.3399 (E) and m/z 540.3332 (F) parent ions. Blue arrows are thresholds and indicators in terms of RT and m/z values. Figure 2 Unsupervised principal components analysis (PCA) models built from positive (A) and negative (B) ionization data set. Red dots: Einkorn (ID331). Green dots: Emmer (Garfagnana). Blue dots: Spelt (Rouquin). Figure 3 (A,B) PLS-DA model built with positive ionization data (R2X = 0.578, R2Y = 0.942, Q2 = 0.916) and negative ionization data (R2X = 0.709, R2Y = 0.967, Q2 = 0.944). In both ionization modes these three varieties were clearly separated. Red dots: Einkorn (ID331). Green dots: Emmer (Garfagnana). Blue dots: Spelt (Rouquin). Figure 4 Variable trend plots of the most discriminant markers: nonadecanylresorcinol (C19:0), overexpressed in the spelt variety Spelt (Rouquin) (A), tricosylresorcinol (C23:0) marker having the highest influence to discriminate emmer variety (Garfagnana) (B) and PC (16:0/18:1), significantly higher in the einkorn variety (ID331) (C). ijms-17-01217-t001_Table 1Table 1 Statistical values for PCA, PLS-DA, OPLS-DA models. R2X (cum) and R2Y (cum) represent the variance of the x and y variables explained by the model, while Q2 is the cumulative predicted variation in the Y matrix. Statistical Parameters ESI(+) Models ESI(−) Models PCA PLS-DA OPLS-DA PCA PLS-DA OPLS-DA R2X (cum) 0.816 0.578 0.579 0.89 0.709 0.709 R2Y (cum) - 0.942 0.942 - 0.967 0.967 Q2 (cum) 0.663 0.916 0.917 0.778 0.944 0.956 Principal components analysis (PCA), partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA). ijms-17-01217-t002_Table 2Table 2 Identification of discriminant metabolites between the three wheat varieties. Biochemical Category Biochemical Class Tentative Identification Pseudomolecolar Ion m/z RT (min) Elemental Formula Mass Error (Δppm) Higher Metabolite Intensity in VIP Value Phenols Resorcinols Heptadecylresorcinol (C17:0) [M − H]− 347.2956 6.3 C23H40O2 1.7 spelt 1.5 Nonadecanylresorcinol (C19:0) [M − H]− 375.3269 6.9 C25H44O2 1.5 spelt 4.2 Nonadecenyl-resorcinol (C19:1) [M − H]− 373.3112 6.3 C25H42O2 1.4 spelt 2.2 Heneicosylresorcinol (C21:0) [M − H]− 403.3582 7.4 C27H48O2 1.4 spelt 2.9 Heneicosenyl-resorcinol (C21:1) [M − H]− 401.3425 6.9 C27H46O2 1.3 spelt 1.5 Tricosylresorcinol (C23:0) [M − H]− 431.3895 8 C29H52O2 1.3 emmer 3.2 Pentacosylresorcinol (C25:0) [M − H]− 459.4208 8.5 C31H56O2 1.2 emmer 3.1 Glycerophospholipids (GLP) Lysophosphatidylcholines (LysoPC) LysoPC 16:0 [M + H]+ 496.3399 4.5 C24H50NO7P 3.2 einkorn 4.3 LysoPC 18:2 [M + H]+ 520.3392 4.2 C26H50NO7P 1.2 einkorn 3.1 Phosphatidylcholines (PC) PC 16:0/18:1 [M + H]+ 760.5851 8.2 C42H82O8NP 1.6 einkorn 2.9 PC 16:0/18:2 [M + H]+ 758.5712 7.9 C42H80NO8P 2.3 einkorn 3.9 Phosphatidylinositols (PI) PI 16:0/18:1 [M + H]+ 835.5478 7.7 C43H81O13P 1.6 emmer 1.6 Glycerolipids (GL) Diacylglycerols (DG) DG 16:0/18:2 [M + NH4]+ 610.5405 8.8 C37H68O5 1.5 emmer 4.9 Triacylglycerols (TG) TG 16:0/16:0/18:2 [M + NH4]+ 848.7708 11.4 C53H98O6 1.8 spelt 3.1 TG 16:0/18:1/20:1 [M + NH4]+ 904.8339 12 C57H106O6 1.9 einkorn 3.8 Table columns: pseudomolecular ion = positive and negative ionization adduct; m/z = mass-to-charge ratio in daltons; RT = ion retention time in minutes; elemental formula = elemental composition of the neutral molecule; mass error ppm = Δ in ppm between the detected m/z and the theoretical m/z; higher metabolite intensity in = ion spectral intensity higher in emmer, einkorn, or spelt as indicated; VIP value = Variable Influence in Projection values. ==== Refs References 1. Shewry P.R. Wheat J. Exp. Bot. 2009 60 1537 1553 10.1093/jxb/erp058 19386614 2. Serpen A. Gokmen V. Karagoz A. Koksel H. Phytochemical quantification and total antioxidant capacities of Emmer (Triticum dicoccon Schrank) and Einkorn (Triticum monococcum L.) wheat landraces J. Agric. Food Chem. 2008 56 7285 7292 10.1021/jf8010855 18593174 3. Galterio G. Codianni P. Giusti A.M. Pezzarossa B. Cannella C. Assessment of the agronomic and technological characteristics of Triticum turgidum ssp. dicoccum Schrank and T. spelta L. Nahrung 2003 47 54 59 12653440 4. Hidalgo A. Brandolini A. Nutritional properties of einkorn wheat (Triticum monococcum L.) J. Sci. Food Agric. 2014 94 601 612 10.1002/jsfa.6382 24022812 5. Bonafaccia G. Galli V. Francisci R. Mair V. Skrabanja V. Kreft I. Characteristics of spelt wheat products and nutritional value of spelt wheat-based bread Food Chem. 2000 68 437 441 10.1016/S0308-8146(99)00215-0 6. Buvaneshwari G. Yenagi N.B. Hanchinal R.R. Naik R.K. Glycaemic responses to dicoccum products in the dietary management of diabetes Ind. J. Nutr. Diet. 2003 40 363 368 7. Piergiovanni A.R. Rizzi R. Pannaciulli E. Della Gatta C. Mineral composition in hulled wheat grains: A comparison between emmer (Triticum dicoccon Schrank) and spelt (T. spelta L.) accessions Int. J. Food Sci. Nutr. 1997 48 381 386 10.3109/09637489709028586 8. Landberg R. Kamal-Eldin A. Salmenkallio-Marttila M. Rouau X. Åman P. Localization of alkylresorcinols in wheat, rye and barley kernels J. Cereal Sci. 2008 48 401 406 10.1016/j.jcs.2007.09.013 9. Knodler M. Most M. Schieber A. Carle R. A novel approach to authenticity control of whole grain durum wheat (Triticum durum Desf.) flour and pasta, based on analysis of alkylresorcinol composition Food Chem. 2010 118 177 181 10.1016/j.foodchem.2009.04.080 10. Ziegler J.U. Steingass C.B. Longin C.F. Würschum T. Reinhold C. Schweiggert R.M. Alkylresorcinol composition allows the differentiation of Triticum spp. having different degrees of ploidy J. Cereal Sci. 2015 65 244 251 10.1016/j.jcs.2015.07.013 11. Rubert J. Zachariasova M. Hajslova J. Advances in high-resolution mass spectrometry based on metabolomics studies for food—A review Food Addit. Contam. Part A 2015 32 1685 1708 10.1080/19440049.2015.1084539 26365870 12. Rubert J. Lacina O. Fauhl-Hassek C. Hajslova J. Metabolic fingerprinting based on high-resolution tandem mass spectrometry: A reliable tool for wine authentication? Anal. Bioanal. Chem. 2014 406 6791 6803 10.1007/s00216-014-7864-y 24866710 13. Matthews S.B. Santra M. Mensack M.M. Wolfe P. Byrne P.F. Thompson H.J. Metabolite profiling of a diverse collection of wheat lines using ultraperformance liquid chromatography coupled with time-of-flight mass spectrometry PLoS ONE 2012 7 1217 10.1371/journal.pone.0044179 22957002 14. Rubingh C.M. Bijlsma S. Derks E. Bobeldijk I. Verheij E.R. Assessing the performance of statistical validation tools for megavariate metabolomics data Metabolomics 2006 2 53 61 10.1007/s11306-006-0022-6 24489531 15. Holcapek M. Jandera P. Zderadicka P. Hruba L. Characterization of triacylglycerol and diacylglycerol composition of plant oils using high-performance liquid chromatography-atmospheric pressure chemical ionization mass spectrometry J. Chromatogr. A 2003 1010 195 215 10.1016/S0021-9673(03)01030-6 12974290 16. Bird S.S. Marur V.R. Sniatynski M.J. Greenberg H.K. Kristal B.S. Serum lipidomics profiling using LC-MS and high-energy collisional dissociation fragmentation: Focus on triglyceride detection and characterization Anal. Chem. 2011 83 6648 6657 10.1021/ac201195d 21774539 17. Ross A.B. Kamal-Eldin A. Jung C. Shepherd M.J. Åman P. Gas chromatographic analysis of alkylresorcinols in rye (Secale cereale L.) grains J. Agric. Food Chem. 2001 81 1405 1411 10.1002/jsfa.956 18. Kulawinek M. Kozubek A. Quantitative determination of alkylresorcinols in cereal grains: Independence of the length of the aliphatic side chain J. Food Lipids 2008 15 251 262 10.1111/j.1745-4522.2008.00118.x 19. Chen J. He S. Mao H. Sun C. Pan Y. Characterization of polyphenol compounds from the roots and stems of Parthenocissus laetevirens by high-performance liquid chromatography/tandem mass spectrometry Rapid Commun. Mass Spectrom. 2009 23 737 744 10.1002/rcm.3937 19204932 20. Milne S. Ivanova P. Forrester J. Brown A.H. Lipidomics: An analysis of cellular lipids by ESI-MS Methods 2006 39 92 103 10.1016/j.ymeth.2006.05.014 16846739 21. Andersson A.M. Kamal-Eldin A. Fras A. Boros D. Amam P. Alkylresorcinols in wheat varieties in the HEALTHGRAIN diversity screen J. Agric. Food Chem. 2008 56 9722 9725 10.1021/jf8011344 18921971 22. Ross A.B. Shepherd M.J. Schüpphaus M. Sinclair V. Alfaro B. Kamal-Eldin A. Åman P. Alkylresorcinols in cereals and cereal products J. Agric. Food Chem. 2003 51 4111 4118 10.1021/jf0340456 12822955 23. Kozubek A. Tyman J.H.P. Resorcinolic lipids, the natural non-isoprenoid phenolic amphiphiles and their biological activity Chem. Rev. 1999 99 1 25 10.1021/cr970464o 11848979 24. Andersson U. Dey E.S. Holm C. Degerman E. Rye bran alkylresorcinols suppress adipocyte lipolysis and hormone-sensitive lipase activity Mol. Nutr. Food Res. 2011 55 S290 S293 10.1002/mnfr.201100231 21887818 25. Ross A.B. Shepherd M.J. Knudsen K.E.B. Glitsø L.V. Bowey E. Phillips J. Rowland I. Guo Z.X. Massy D.J.R. Aman P. Absorption of dietary alkylresorcinols in ileal-cannulated pigs and rats Br. J. Nutr. 2003 90 787 794 10.1079/BJN2003965 13129447 26. Hanhineva K. Brunius C. Andersson A. Marklund M. Juvonen R. Keski-Rahkonen P. Auriola S. Landberg R. Discovery of urinary biomarkers of whole grain rye intake in free-living subjects using nontargeted LC-MS metabolite profiling Mol. Nutr. Food Res. 2015 59 2315 2325 10.1002/mnfr.201500423 26264776 27. Zarnowski R. Suzuki Y. Pietr S.J. Alkyl- and alkenylresorcinols of wheat grains and their chemotaxonomic significance Z. Naturforsch. C 2004 59 190 196 10.1515/znc-2004-3-411 15241924 28. González-Thuillier I. Salt L. Chope G. Penson S. Skeggs P. Tosi P. Powers S.J. Ward J.L. Wilde P. Shewry P.R. Distribution of lipids in the grain of wheat (cv. Hereward) determined by lipidomic analysis of milling and pearling fractions J. Agric. Food Chem. 2015 63 10705 10710 10.1021/acs.jafc.5b05289 26582143

PMC005xxxxxx/PMC5000616.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081218ijms-17-01218ArticleThe Impact of Nonalcoholic Fatty Liver Disease on Renal Function in Children with Overweight/Obesity Pacifico Lucia 1*Bonci Enea 2Andreoli Gian Marco 1Di Martino Michele 3Gallozzi Alessia 1De Luca Ester 1Chiesa Claudio 4Tarantino Giovanni Academic Editor1 Policlinico Umberto I Hospital, Sapienza University of Rome, Viale Regina Elena 324, 00161 Rome, Italy; [email protected] (G.M.A.); [email protected] (A.G.); [email protected] (E.D.L.)2 Department of Experimental Medicine, Sapienza University of Rome, Viale Regina Elena 324, 00161 Rome, Italy; [email protected] Department of Radiological Sciences, Sapienza University of Rome, Viale Regina Elena 324, 00161 Rome, Italy; [email protected] Institute of Translational Pharmacology, National Research Council, Via Fosso del Cavaliere 100, 00133 Rome, Italy; [email protected]* Correspondence: [email protected]; Tel.: +39-06-4997-921527 7 2016 8 2016 17 8 121825 6 2016 21 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The association between nonalcoholic fatty liver disease (NAFLD) and chronic kidney disease has attracted interest and attention over recent years. However, no data are available in children. We determined whether children with NAFLD show signs of renal functional alterations, as determined by estimated glomerular filtration rate (eGFR) and urinary albumin excretion. We studied 596 children with overweight/obesity, 268 with NAFLD (hepatic fat fraction ≥5% on magnetic resonance imaging) and 328 without NAFLD, and 130 healthy normal-weight controls. Decreased GFR was defined as eGFR < 90 mL/min/1.73 m2. Abnormal albuminuria was defined as urinary excretion of ≥30 mg/24 h of albumin. A greater prevalence of eGFR < 90 mL/min/1.73 m2 was observed in patients with NAFLD compared to those without liver involvement and healthy subjects (17.5% vs. 6.7% vs. 0.77%; p < 0.0001). The proportion of children with abnormal albuminuria was also higher in the NAFLD group compared to those without NAFLD, and controls (9.3% vs. 4.0% vs. 0; p < 0.0001). Multivariate logistic regression analysis revealed that NAFLD was associated with decreased eGFR and/or microalbuminuria (odds ratio, 2.54 (confidence interval, 1.16–5.57); p < 0.05) independently of anthropometric and clinical variables. Children with NAFLD are at risk for early renal dysfunction. Recognition of this abnormality in the young may help to prevent the ongoing development of the disease. nonalcoholic fatty liver diseaserenal functionobesitychildren ==== Body 1. Introduction Concurrent with the epidemic of obesity across the world, nonalcoholic fatty liver disease (NAFLD) is becoming one of the most prevalent chronic liver disorders in both adults and children. It is now known that NAFLD is not only a risk factor for hepatic failure and hepatic carcinoma, but it is also associated with a spectrum of extrahepatic diseases generally linked to metabolic syndrome (MetS) such as type 2 diabetes, and cardiovascular disease [1,2]. Recent studies in the pediatric obese population have demonstrated that the prevalence of prediabetes and MetS is significantly increased in subjects with increased hepatic fat content, and that liver steatosis, independently of visceral and intramyocellular lipid content, is a key determinant of the impairment of liver, muscle, and adipose insulin sensitivity [3,4]. Several studies have reported associations between NAFLD and subclinical atherosclerosis and between NAFLD and cardiac function alterations, independently of established risk factors [5,6,7]. In addition, emerging evidence suggests that subjects with NAFLD have an increased risk of chronic kidney disease (CKD), defined by a decline in the estimated glomerular filtration rate (eGFR) and/or microalbuminuria and/or overt proteinuria [8,9,10,11,12]. However, no data are available in children regarding a possible association between NAFLD and impaired renal function. Recognition of the influence of NAFLD on renal function in the early age would enable us to better understand the association of NAFLD and CKD, since there is less potential for confusion with adult-onset complications. Thus, in this study we sought to determine whether children with overweight/obesity and NAFLD show signs of renal functional alterations, as assessed by eGFR and urinary albumin excretion, compared to children with overweight/obesity but without NAFLD as well as to healthy normal-weight controls. 2. Results 2.1. Clinical and Laboratory Data from the Study Population Clinical and laboratory data from the study population are presented in Table 1. None of the enrollees had type 2 diabetes mellitus. Patients with NAFLD were on average older than those without NAFLD and healthy controls, and had higher waist circumference (WC) as well as higher values for systolic and diastolic blood pressure (BP), higher triglycerides, aspartate aminotransferase (AST), alanine aminotransferase (ALT), uric acid, fasting glucose, insulin levels and homeostasis model assessment of insulin resistance (HOMA-IR) values, and lower high-density lipoprotein-cholesterol (HDL-C) concentrations. Patients with NAFLD had significantly lower whole-body insulin sensitivity index (WBISI) than those without NAFLD. Obese children with NAFLD and obese subjects without NAFLD had significantly higher eGFR compared to healthy controls (median, 115 (interquartile range, 104–134) and 115 (96–132) vs. 108 (100–118) mL/min/1.73 m2; p < 0.0001), whereas no differences were found between patients with and without NAFLD. However, a greater frequency of reduced eGFR (<90 mL/min/1.73 m2) was observed in obese subjects with NAFLD compared to obese children without liver involvement and healthy controls (17.5% vs. 6.7% vs. 0.77%, respectively; p < 0.0001). The proportion of children with microalbuminuria was also higher in the NAFLD group compared to obese children without liver involvement and healthy controls (9.3% vs. 4.0% vs. 0; p < 0.0001). None of the participants had eGFR < 60 mL/min/1.73 m2 or macroalbuminuria. Compared to healthy controls, the prevalence of hyperfiltration was higher in the obese cohort, regardless of liver involvement (Table 1). To analyze the variables associated with decreased eGFR and/or microalbuminuria, we performed a logistic regression analysis in the cohort of subjects with overweight/obesity. NAFLD (odds ratio (OR), 2.34; 95% confidence interval (CI), 1.31–4.16; p < 0.01) was associated with abnormal renal function independently of age, gender, and pubertal status. After further adjustment for body mass index-standard deviation (BMI-SD) score, WC, hypertension, low HDL-C values, elevated triglycerides, and glucose impairment, results did not substantially change (Table 2). 2.2. Findings in Children with Biopsy-Proven Nonalcoholic Fatty Liver Disease (NAFLD) To investigate the association of renal dysfunction further with advanced stages of NAFLD such as steatohepatitis (NASH), we analysed the data obtained in the small subgroup of 41 patients who underwent liver biopsy. Definite-NASH was diagnosed in 26 (63.4%) children, while not-NASH in 15 (36.5%). Compared to children without NASH, those with NASH had significantly lower eGFR (median, 88 (83–107) vs. 123 (110–130) mL/min/1.73 m2; p < 0.01). In addition, more children with NASH had eGFR of <90 mL/min/1.73 m2 and/or microalbuminuria than those without NASH (17/26 (65.4%) vs. 6/15 (40.0%); p < 0.01). 3. Discussion Early recognition of impaired renal function, in particular reduced GFR, is crucial to prevent serious complications [13]. Large epidemiologic studies have found a robust relationship between obesity and risk for CKD [14,15,16]. In a community-based sample of 2585 adult individuals with renal disease at baseline and a mean follow-up of 18.5 years, BMI was reported to determine a significant increase in the odds of developing kidney disease by 23% (OR, 1.23; 95% CI, 1.08–1.41) per standard deviation unit [14]. In 9685 adults participating to the Hypertension Detection and Follow-Up Program, free of CKD at baseline, the incidence of CKD was 28%, 31%, and 34%, respectively, in the ideal body mass index, overweight, and obese groups, after a follow-up of five years [15]. After adjustment for variables, such as age, gender, race, diabetes mellitus, mean baseline diastolic BP, and slope of diastolic BP, at baseline both overweight (OR, 1.21; 95% CI, 1.05 to 1.41) and obesity (OR, 1.40; 95% CI, 1.20 to 1.63) were associated with increased incident CKD odds at year 5 [15]. In addition, a retrospective cohort study of 320,252 adults, who were followed for 15 to 35 years, showed that a high BMI (≥25.0 kg/m2) determined who is at high risk of developing end-stage renal disease [16]. Taken together, these studies indicate that higher BMI in adults is a risk factor for the development of new onset kidney disease. Several possible pathophysiologic pathways may underlie this association. One possibility is that particular characteristics of obesity may account for the association between obesity and CKD. Indeed, obesity constitutes a complex syndrome involving metabolic traits and other factors that may interact with other environmental factors, leading to an increased risk for developing kidney disease. Clustering of these traits defines MetS, which has been reported to be consistently associated with CKD in cross-sectional studies [17,18]. NAFLD has been recently found to be an additional feature of MetS, with the main underlying cardiometabolic risk factors of the syndrome being abdominal obesity and insulin resistance [19,20]. Of note, insulin resistance is not only a metabolic determinant for the development of NAFLD but is also a predictor of incident CKD [21,22]. In addition, atherogenic dyslipidemia and type 2 diabetes are established risk factors for CKD [23,24]. As a consequence, many authors have concluded that NAFLD may have a pathogenic role in the development of CKD. The results of a recent meta-analysis have shown that (1) there is a positive relationship between NAFLD and an increased risk of CKD in adults; (2) the severity of liver disease is associated with an increased risk and severity of CKD; and (3) these relationships are maintained even after taking account of the well-known risk factors for CKD, and are independent of whole body/abdominal obesity and insulin resistance [8]. In our study, we investigated the influence of NAFLD on kidney function in a large pediatric population. This is the first study to demonstrate that overweight/obese children with NAFLD have a greater frequency of eGFR of <90 mL/min/1.73 m2 as well as of microalbuminuria than overweight/obese children without NAFLD. Furthermore, in the small number of children with biopsy-proven NAFLD we were able to show that the decline in renal function was greater in those with NASH. It is important to point out that subjects with obesity represent a particular population in whom early renal lesion consists of hyperfiltration. In fact, in line with previous studies [25,26,27], one of the main findings of this study was that children with overweight/obesity compared to normal-weight subjects had a higher prevalence of hyperfiltration, regardless of liver involvement. Glomerular hyperfiltration is well-recognized as an early renal injury occurring in a number of clinical conditions, including diabetes, hypertension, and obesity [28]. Hyperfiltration is hypothesized to be a precursor of intraglomerular hypertension responsible for albuminuria. GFR then declines progressively as albuminuria increases which may cause, in the long run, end-stage renal failure [28]. Thus, in obese patients with NAFLD, we should pay attention for minor impairment on renal function, since hyperfiltration may mask a pathological decline in renal function. The most plausible explanation for our findings is that the renal abnormalities in overweight/obese children with NAFLD may reflect the coexistence of underlying metabolic risk factors including higher BP, more dyslipidemia, and more insulin resistance compared to children without liver involvement. However, because in our study the presence of NAFLD remained significantly associated with decreased eGFR and/or microalbuminuria after taking account of traditional metabolic traits, we cannot rule out the possibility that NAFLD might at least in part contribute to the development of renal dysfunction independently of shared cardiometabolic risk factors. The strength of our study includes a large sample size and an extensive and complete analysis of metabolic variables. Nonetheless, some limitations require consideration. First, the cross-sectional design of the study precludes the establishment of causal relationship between NAFLD and abnormal kidney function. Second, we used an estimated GFR instead of a directly measured GFR to define renal function. The gold standard technique is clearance of inulin, but practical problems limit the application of this cumbersome methodology in children because of the necessity for steady-state infusion, and a urine sampling with a bladder catheter. Other tests for determining GFR are clearance of alternative exogenous markers such as iothalamate, which are also complex and difficult to do in routine clinical practice. Recent studies in children have reported current eGFR creatinine- and/or cystatin C-based equations to be reliable methods to assess kidney function, with some variations depending on the GFR ranges and the BMI classes [29,30,31]. The updated Schwartz formula has been shown to be accurate for estimating GFR when compared to inulin clearance as well as to iothalamate clearance in children and adolescents, with a wide range of renal function [29,30]. Moreover, obesity has not been found to affect GFR as estimated by Schwartz formula [31]. Finally, we measured creatinine concentration by kinetic colorimetric compensated technique, whereas in the updated Schwartz formula, it was determined by an enzymatic method. The two methods, however, are highly correlated [29]. In conclusion, our present study suggests that obese children with NAFLD are at risk for early renal dysfunction. Recognition of this abnormality in the young may be important because treatment to reverse the process is most likely to be effective if applied earlier in the disease process. 4. Materials and Methods 4.1. Study Subjects This observational cross-sectional study included 596 children and adolescents with overweight/obesity who were consecutively recruited at the outpatient Clinics (Hepatology, Lipid and Nutrition) of the Department of Pediatrics, Sapienza University of Rome, Italy, between 2007 and 2015. Two hundred and sixty eight subjects met the criteria for the diagnosis of NAFLD (i.e., hepatic fat fraction (HFF) ≥5% on magnetic resonance imaging (MRI)) [32]. In all enrollees, hepatic virus infections (hepatitis A–E and G, cytomegalovirus, and Epstein–Barr virus), autoimmune hepatitis, metabolic liver disease, α-1-antitrypsin deficiency, cystic fibrosis, Wilson’s disease, hemochromatosis, and celiac disease were excluded using appropriate tests [6,7]. In 41 of the NAFLD patients, due to persistent elevations in ALT concentrations, a liver biopsy was performed. The other 328 participants had HFF < 5% on MRI, normal levels of aminotransferases, and no evidence of chronic liver diseases (see above). Use of hepatotoxic drugs, as well as a history of type 1 or 2 diabetes, smoking and chronic alcohol intake were also exclusion criteria. None of the subjects had a history or known clinical, laboratory, and imaging signs of renal disease. The study also included a total of 130 apparently healthy normal-weight school students drawn from four randomly selected schools in the Rome area. All students were invited to take part in a pilot study whose objective was the prevention of cardiovascular disease in childhood. Eligibility criteria included age- and gender-specific BMI; no history of renal and liver diseases as well as of alcohol consumption and smoking; normal liver ultrasound, and normal biochemical values. All study subjects had a complete physical examination, as reported in detail elsewhere [5,6]. The degree of obesity was quantified using Cole’s least mean-square method, which normalizes the skewed distribution of BMI and expresses BMI as SD score [33]. The study protocol was reviewed and approved by the Ethics Committee of Policlinico Umberto I Hospital, Rome, Italy. Written informed consent was obtained from the parents, or guardians of the children included in this study, in accordance with principles of Helsinki Declaration. 4.2. Laboratory Mmeasurements Blood samples were taken from all study subjects, after an overnight fast, for estimation of glucose, insulin, urea nitrogen, creatinine, uric acid, total cholesterol, HDL-C, triglycerides, ALT, AST, and gamma-glutamyl transferase. An oral glucose tolerance test was performed for all overweight/obese children using 1.75 g/kg of glucose up to a maximum of 75 g. Two-hour post-load glucose and insulin were analyzed. Insulin resistance was calculated by the HOMA-IR. Insulin sensitivity was calculated by the WBISI with reduced time points according to the following formula: 10,000/√ (fasting glucose × fasting insulin × 2 h post-load glucose × 2 h post-load insulin) [34]. All analyses were performed on COBAS 6000 (Roche Diagnostics, Risch-Rotkreuz, Switzerland). Creatinine concentrations were measured by the kinetic colorimetric compensated Jaffé method using the Roche platform and the CREJ2–creatinine Jaffé Gen.2 assay (Roche Diagnostics, Identification number, 0769282), which was isotope-dilution mass spectrometry standardized, traceable to National Institute of Standards and Technology creatinine standard reference material (SRM 914 and SRM 967). Urinary albumin was determined on 24 h urine collections by the turbidimetric immunoassay ALBT2 (Roche Diagnostics, Identification number, 0767433). eGFR was calculated using the updated Schwartz formula: 0.413 × height (cm)/serum creatinine (mg/dL) [35]. 4.3. Liver Ultrasound Eexamination and Magnetic Resonance Imaging Liver ultrasound was performed by a single operator. Hepatic steatosis was diagnosed on the basis of the following features: a diffuse increase in echogenicity (a bright liver), liver to kidney contrast, deep beam attenuation, vascular blurring, and loss of definition of the diaphragm [36]. The amount of HFF was measured by MRI using the two-point Dixon method as modified by Fishbein [37], as previously described and validated [32,38]. 4.4. Liver Biopsy Liver biopsy was performed in 41 subjects because of persistent elevation in ALT. The clinical indication for biopsy was either to assess the presence of nonalcoholic steatohepatitis (NASH) or to determine the presence of other independent or competing liver diseases. The main histologic features of NAFLD were scored using the NASH Clinical Research Network criteria [39]. Biopsies were categorized into not-NASH and definite-NASH. 4.5. Definitions Overweight and obesity were defined according to age- and gender-specific cut-off points of BMI defined by the International Obesity Task Force criteria as proposed by Cole et al. [33]. Elevated BP was defined as systolic or diastolic BP ≥ 90th percentile for age, gender, and height [40]. Impaired fasting glucose was defined as glucose ≥5.6 mmol/L. High waist circumference (WC), high triglycerides, and low HDL-C were defined using the cut-off proposed by Cook et al. [41]. Insulin resistance was defined by 90th percentile of HOMA-IR for age and gender in our population of healthy normal-weight children. Abnormal albuminuria was defined as a 24-h urinary albumin excretion rate ≥30 mg (i.e., microalbuminuria was diagnosed if the 24-h albumin excretion rate was 30–299 mg and macroalbuminuria if the 24-h albumin excretion rate was ≥300 mg) [42]. As recommended by Kidney Disease Improving Global Outcomes (KDIGO) guidelines, eGFR categories were classified as follows: normal or high ≥90 mL/min/1.73 m2; mildly decreased, 60–89; mildly to moderately decreased, 45–59; moderately to severely decreased, 30–44; severely decreased, 15–29; and kidney failure <15 [42]. In the absence of an agreement in the literature, we defined glomerular hyperfiltration as eGFR > 95th percentile of that observed in our population of healthy normal-weight subjects (i.e., eGFR > 139 mL/min/1.73 m2). 4.6. Statistical Analysis Statistical analyses were performed using the SPSS package (version 22.0, SPSS Inc., Chicago, IL, USA). Data are reported as means and standard deviations for normally distributed variables, or as median and interquartile range for non-normally distributed variables. Differences between study groups in quantitative variables were evaluated by one-way analysis of variance (ANOVA) or Kruskal–Wallis test, as appropriate. Proportions were compared by the chi square test. Logistic regression analysis was used to assess the independent association of NAFLD with abnormal kidney function, after adjustment for age, gender, pubertal status, BMI-SD score, WC, hypertension, low HDL-C values, elevated triglycerides, and glucose impairment. Acknowledgments This study was supported by Sapienza University of Rome (Progetti di Ricerca Universitaria 2013/2014). Author Contributions Lucia Pacifico, Enea Bonci, Claudio Chiesa conceived and designed the study; Lucia Pacifico, Enea Bonci, Gian Marco Andreoli, Michele Di Martino, Alessia Gallozzi, and Ester De Luca collected and analyzed the data; Lucia Pacifico, Enea Bonci, Gian Marco Andreoli, Michele Di Martino, and Claudio Chiesa interpreted the data; Lucia Pacifico, Enea Bonci, and Claudio Chiesa wrote the manuscript. All authors read and approved the final version of the manuscript. Conflicts of Interest The authors declare no conflicts of interest. ijms-17-01218-t001_Table 1Table 1 Clinical and laboratory characteristics of the study population. Normal Weight NO NAFLD NAFLD p Value * No. patients 130 328 268 <0.0001 Age, years 10.6 (3.5) 10.1 (2.9) 11.2 (2.9) d <0.0001 Male sex, n (%) 61 (46.9) 151 (46.0) 166 (61.9) a,d <0.0001 BMI-SD score 0.17 (0.85) 1.85 (0.45) a 2.0 (0.45) a,d <0.0001 Waist circumference, cm 65 (10) 82 (12) a 92 (13) a,d <0.0001 Systolic BP, mmHg 102 (11) 107 (12) b 114 (12) a,d <0.0001 Diastolic BP, mmHg 63 (7) 65 (9) c 69 (8) a,d <0.0001 Total cholesterol, mg/dL 166 (145–186) 161 (139–187) 159 (137–181) 0.077 LDL-C 92 (72–118) 94 (76–115) 94 (74–111) 0.78 HDL-C, mg/dL 56 (50–83) 51 (44–60) a 46 (38–53) a,d <0.0001 Triglycerides, mg/dL 62 (50–83) 70 (50–99) 89 (58–127) a,d <0.0001 AST, U/L 22 (20–30) 23 (20–27) c 26 (21–35) a,d <0.0001 ALT, U/L 16 (13–20) 18 (14–23) b 31 (19–54) a,d <0.0001 Uric acid 0.21 (0.18–0.25) 0.25 (0.22–0.29) a 0.28 (0.24–0.34) a,d <0.0001 Glucose, mg/dL 83 (7) 83 (7) 85 (11) 0.002 Insulin, μU/mL 7.5 (4.3–10.5) 11.1 (7.5–15.4) a 15.2 (10.1–23.2) a,d <0.0001 HOMA-IR 1.58 (0.90–2.20) 2.30 (1.55–3.22) a 3.23 (2.05–5.0) a,d <0.0001 WBISI - 6.5 (4.5–9.0) 3.5 (2.4–5.6) d - eGFR, mL/min/1.73 m2 108 (100–118) 115 (104–134) a 115 (96–132) a <0.0001 eGFR < 90 mL/min/1.73 m2, n (%) 1 (0.77) 22 (6.7) b 47 (17.5) a,d <0.0001 eGFR > 139 mL/min/1.73 m2, n (%) 6 (4.6) 56 (17.0) a 46 (17.2) a 0.002 Microalbuminuria, n (%) 0 13 (4.0) a 25 (9.3) a,d <0.0001 Results are expressed as n (%), mean (standard devation), or median (interquartile ranges). * Anova or Kruskal-Wallis test; a p < 0.0001; b p < 0.01; c p < 0.05 vs. controls; d p < 0.0001 vs. obese children without NAFLD; NAFLD, nonalcoholic fatty liver disease; BMI-SD score, Body mass index- standard deviation score; BP, Blood pressure; LDL-C, Low density lipoprotein-cholesterol; HDL-C, High-density lipoprotein-cholesterol; AST, Aspartate aminotransferase; ALT, Alanine aminotransferase; HOMA-IR, Homeostasis model assessment of insulin resistance; WBISI, Whole-body insulin sensitivity index; eGFR, estimated glomerular filtration rate. ijms-17-01218-t002_Table 2Table 2 Associations of NAFLD with eGFR < 90 mL/min/1.73 m2 and/or microalbuminuria in children with overweight/obesity. Variables Odds Ratio (95% CI) p Value Adjusted model 1: age, gender, pubertal status 2.34 (1.31–4.16) 0.004 Adjusted model 2: model 1 plus BMI-SD score, WC, High BP, High TG, low HDL-C, and high FG 2.54 (1.16–5.57) 0.02 Adjusted model 3: model 1 plus BMI-SD score, WC, High BP, High TG, low HDL-C, and IR 2.30 (1.02–5.17) 0.04 CI, confidence interval; eGFR, estimated glomerular filtration rate; BMI-SD score, Body mass index- standard deviation score; WC, waist circumference; BP, Blood pressure; TG, triglycerides; HDL-C, High-density lipoprotein-cholesterol; FG, fasting glucose; IR, insulin resistance. ==== Refs References 1. Vanni E. Marengo A. Mezzabotta L. Bugianesi E. Systemic complications of nonalcoholic fatty liver disease: When the liver is not an innocent bystander Semin. Liver Dis. 2015 35 236 249 10.1055/s-0035-1562944 26378641 2. Chatterjee R. Mitra A. An overview of effective therapies and recent advances in biomarkers for chronic liver diseases and associated liver cancer Int. Immunopharmacol. 2015 24 335 345 10.1016/j.intimp.2014.12.024 25560752 3. Schwimmer J.B. Pardee P.E. Lavine J.E. Blumkin A.K. Cook S. Cardiovascular risk factors and the metabolic syndrome in pediatric nonalcoholic fatty liver disease Circulation 2008 118 277 283 10.1161/CIRCULATIONAHA.107.739920 18591439 4. D’Adamo E. Cali A.M. Weiss R. Santoro N. Pierpont B. Northrup V. Caprio S. Central role of fatty liver in the pathogenesis of insulin resistance in obese adolescents Diabetes Care 2010 33 1817 1822 10.2337/dc10-0284 20668154 5. Targher G. Day C.P. Bonora E. Risk of cardiovascular disease in patients with nonalcoholic fatty liver disease N. Engl. J. Med. 2010 363 1341 1350 10.1056/NEJMra0912063 20879883 6. Pacifico L. Anania C. Martino F. Cantisani V. Pascone R. Marcantonio A. Chiesa C. Functional and morphological vascular changes in pediatric nonalcoholic fatty liver disease Hepatology 2010 52 1643 1651 10.1002/hep.23890 20890890 7. Pacifico L. Di Martino M. De Merulis A. Bezzi M. Osborn J.F. Catalano C. Chiesa C. Left ventricular dysfunction in obese children and adolescents with nonalcoholic fatty liver disease Hepatology 2014 59 461 470 10.1002/hep.26610 23843206 8. Musso G. Gambino R. Tabibian J.H. Ekstedt M. Kechagias S. Hamaguchi M. Hultcrantz R. Hagström H. Yoon S.K. Charatcharoenwitthaya P. Association of non-alcoholic fatty liver disease with chronic kidney disease: A systematic review and meta-analysis PLoS Med. 2014 11 10.1371/journal.pmed.1001680 25050550 9. Sesti G. Fiorentino T.V. Arturi F. Perticone M. Sciacqua A. Perticone F. Association between noninvasive fibrosis markers and chronic kidney disease among adults with nonalcoholic fatty liver disease PLoS ONE 2014 9 1218 10.1371/journal.pone.0088569 24520400 10. Machado M.V. Gonçalves S. Carepa F. Coutinho J. Costa A. Cortez-Pinto H. Impaired renal function in morbid obese patients with nonalcoholic fatty liver disease Liver Int. 2012 32 241 248 10.1111/j.1478-3231.2011.02623.x 22098270 11. Targher G. Mantovani A. Pichiri I. Mingolla L. Cavalieri V. Mantovani W. Pancheri S. Trombetta M. Zoppini G. Chonchol M. Nonalcoholic fatty liver disease is independently associated with an increased incidence of chronic kidney disease in patients with type 1 diabetes Diabetes Care 2014 37 1729 1736 10.2337/dc13-2704 24696459 12. Pan L.L. Zhang H.J. Huang Z.F. Sun Q. Chen Z. Li Z.B. Yang S.Y. Li X.Y. Li X.J. Intrahepatic triglyceride content is independently associated with chronic kidney disease in obese adults: A cross-sectional study Metabolism 2015 64 1077 1085 10.1016/j.metabol.2015.06.003 26144271 13. Gansevoort R.T. Matsushita K. van der Velde M. Astor B.C. Woodward M. Levey A.S. de Jong P.E. Coresh J. Chronic Kidney Disease Prognosis Consortium Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts Kidney Int. 2011 80 93 104 10.1038/ki.2010.531 21289597 14. Fox C.S. Larson M.G. Leip E.P. Culleton B. Wilson P.W.F. Levy D. Predictors of new-onset kidney disease in a community-based population JAMA 2004 291 844 850 10.1001/jama.291.7.844 14970063 15. Kramer H. Luke A. Bidani A. Cao G. Cooper R. McGee D. Obesity and prevalent and incident CKD: The Hypertension Detection and Follow-Up Program Am. J. Kidney Dis. 2005 46 587 594 10.1053/j.ajkd.2005.06.007 16183412 16. Hsu C.Y. McCulloch C.E. Iribarren C. Darbinian J. Go A.S. Body mass index and risk for end-stage renal disease Ann. Intern. Med. 2006 144 21 28 10.7326/0003-4819-144-1-200601030-00006 16389251 17. Chen J. Muntner P. Hamm L.L. Jones D.W. Batuman V. Fonseca V. Whelton P.K. He J. The metabolic syndrome and chronic kidney disease in U.S. adults Ann. Intern. Med. 2004 140 167 174 10.7326/0003-4819-140-3-200402030-00007 14757614 18. Hoehner C.M. Greenlund K.J. Rith-Najarian S. Casper M.L. McClellan W.M. Association of the insulin resistance syndrome and microalbuminuria among nondiabetic native Americans. The Inter-Tribal Heart Project J. Am. Soc. Nephrol. 2002 13 1626 1634 10.1097/01.ASN.0000015762.92814.85 12039992 19. Speliotes E.K. Massaro J.M. Hoffmann U. Vasan R.S. Meigs J.B. Sahani D.V. Hirschhorn J.N. O’Donnell C.J. Fox C.S. Fatty liver is associated with dyslipidemia and dysglycemia independent of visceral fat: The Framingham Heart Study Hepatology 2010 51 1979 1987 10.1002/hep.23593 20336705 20. Grundy S.M. Metabolic syndrome pandemic Arterioscler. Thromb. Vasc. Biol. 2008 28 629 636 10.1161/ATVBAHA.107.151092 18174459 21. Bugianesi E. Moscatiello S. Ciaravella M.F. Marchesini G. Insulin resistance in nonalcoholic fatty liver disease Curr. Pharm. Des. 2010 16 1941 1951 10.2174/138161210791208875 20370677 22. Cheng H.T. Huang J.W. Chiang C.K. Yen C.J. Hung K.Y. Wu K.D. Metabolic syndrome and insulin resistance as risk factors for development of chronic kidney disease and rapid decline in renal function in elderly J. Clin. Endocrinol. Metab. 2012 97 1268 1276 10.1210/jc.2011-2658 22337909 23. Vlagopoulos P.T. Sarnak M.J. Traditional and non-traditional cardiovascular risk factors in chronic kidney disease Med. Clin. N. Am. 2005 89 587 611 10.1016/j.mcna.2004.11.003 15755469 24. Athyros V.G. Tziomalos K. Katsiki N. Doumas M. Karagiannis A. Mikhailidis D.P. Cardiovascular risk across the histological spectrum and the clinical manifestations of non-alcoholic fatty liver disease: An update World J. Gastroenterol. 2015 21 6820 6834 26078558 25. Wuerzner G. Pruijm M. Maillard M. Bovet P. Renaud C. Burnier M. Boshud M. Marked association between obesity and glomerular hyperfiltration: A cross-sectional study in an African population Am. J. Kidney Dis. 2010 56 303 312 10.1053/j.ajkd.2010.03.017 20538392 26. Xiao N. Jenkins T.M. Nehus E. Inge T.H. Michalsky M.P. Harmon C.M. Helmrath M.A. Brandt M.L. Courcoulas A. Moxey-Mims M. Kidney function in severely obese adolescents undergoing bariatric surgery Obesity 2014 22 2319 2325 10.1002/oby.20870 25376399 27. Franchini S. Savino A. Marcovecchio M.L. Tumini S. Chiarelli F. Mohn A. The effect of obesity and type 1 diabetes on renal function in children and adolescents Pediatr. Diabetes 2015 16 427 433 10.1111/pedi.12196 25131409 28. Palatini P. Glomerular hyperfiltration: A marker of early renal damage in prediabetes and pre-hypertension Nephrol. Dial. Transplant. 2012 27 1708 1714 10.1093/ndt/gfs037 22431709 29. Bacchetta J. Cochat P. Rognant N. Ranchin B. Hadj-Aissa A. Dubourg L. Which creatinine and cystatin C equations can be reliably used in children? Clin. J. Am. Soc. Nephrol. 2011 6 552 560 10.2215/CJN.04180510 21115623 30. Staples A. LeBlond R. Watkins S. Wong C. Brandt J. Validation of the revised Schwartz estimating equation in a predominantly non-CKD population Pediatr. Nephrol. 2010 25 2321 2326 10.1007/s00467-010-1598-7 20652327 31. Fadrowski J.J. Neu A.M. Schwartz G.J. Furth S.L. Pediatric GFR estimating equations applied to adolescents in the general population Clin. J. Am. Soc. Nephrol. 2011 6 1427 1435 10.2215/CJN.06460710 21566103 32. Pacifico L. Di Martino M. Catalano C. Panebianco V. Bezzi M. Anania C. Chiesa C. T1-weighted dual-echo MRI for fat quantification in pediatric nonalcoholic fatty liver disease World J. Gastroenterol. 2011 17 3012 3019 10.3748/wjg.v17.i25.3012 21799647 33. Cole T.J. Bellizzi M.C. Flegal K.M. Dietz W.H. Establishing a standard definition for child overweight and obesity worldwide: International survey BMJ 2000 320 1240 1243 10.1136/bmj.320.7244.1240 10797032 34. DeFronzo R.A. Matsuda M. Reduced time points to calculate the composite index Diabetes Care 2010 33 e93 10.2337/dc10-0646 20587713 35. Schwartz G.J. Muñoz A. Schneider M.F. Mak R.H. Kaskel F. Warady B.A. Furth S.L. New equations to estimate GFR in children with CKD J. Am. Soc. Nephrol. 2009 20 629 637 10.1681/ASN.2008030287 19158356 36. Hamer O.W. Aguirre D.A. Casola G. Lavine J.E. Woenckhaus M. Sirlin C.B. Fatty liver: Imaging patterns and pitfalls Radiographics 2006 26 1637 1653 10.1148/rg.266065004 17102041 37. Fishbein M.H. Gardner K.G. Potter C.J. Schmalbrock P. Smith M.A. Introduction of fast MR imaging in the assessment of hepatic steatosis Magn. Reson. Imaging 1997 15 287 293 10.1016/S0730-725X(96)00224-X 9201675 38. Pacifico L. Di Martino M. Anania C. Andreoli G.M. Bezzi M. Catalano C. Chiesa C. Pancreatic fat and β-cell function in overweight/obese children with nonalcoholic fatty liver disease World J. Gastroenterol. 2015 21 4688 4695 25914480 39. Kleiner D.E. Brunt E.M. van Natta M. Behling C. Contos M.J. Cummings O.W. Ferrell L.D. Liu Y.C. Torbenson M.S. Unalp-Arida A. Design and validation of a histological scoring system for nonalcoholic fatty liver disease Hepatology 2005 41 1313 1321 10.1002/hep.20701 15915461 40. National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents Pediatrics 2004 114 Suppl. 2 555 576 15286277 41. Cook S. Auinger P. Huang T.T. Growth curves for cardio-metabolic risk factors in children and adolescents J. Pediatr. 2009 155 Suppl. 6 e15 e26 10.1016/j.jpeds.2009.04.051 19732566 42. Levey A.S. de Jong P.E. Coresh J. Nahas M.E. Astor B.C. Matsushita K. Gansevoort R.T. Kasiske B.L. Eckardt K.U. The definition, classification, and prognosis of chronic kidney disease: A KDIGO controversities conference report Kidney Int. 2011 80 17 28 10.1038/ki.2010.483 21150873

PMC005xxxxxx/PMC5000617.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081219ijms-17-01219ArticleAntiproliferative and Pro-Apoptotic Effect of Novel Nitro-Substituted Hydroxynaphthanilides on Human Cancer Cell Lines Kauerova Tereza 1Kos Jiri 2Gonec Tomas 2Jampilek Josef 3Kollar Peter 1*Battino Maurizio Academic Editor1 Department of Human Pharmacology and Toxicology, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1946/1, 612 42 Brno, Czech Republic; [email protected] Department of Chemical Drugs, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1946/1, 612 42 Brno, Czech Republic; [email protected] (J.K.); [email protected] (T.G.)3 Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Comenius University, Odbojarov 10, 832 32 Bratislava, Slovakia; [email protected]* Correspondence: [email protected]; Tel.: +420-541-562-89228 7 2016 8 2016 17 8 121922 6 2016 21 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Ring-substituted hydroxynaphthanilides are considered as cyclic analogues of salicylanilides, compounds possessing a wide range of pharmacological activities, including promising anticancer properties. The aim of this study was to evaluate the potential anticancer effect of novel nitro-substituted hydroxynaphthanilides with a special focus on structure-activity relationships. The antiproliferative effect was assessed by Water Soluble Tetrazolium Salts-1 (WST-1) assay, and cytotoxicity was evaluated via dye exclusion test. Flow cytometry was used for cell cycle analysis and detection of apoptosis using Annexin V-FITC/PI assay. Protein expression was estimated by Western blotting. Our data indicate that the potential to cause the antiproliferative effect increases with the shift of the nitro substituent from the ortho- to the para-position. The most potent compounds, 3-hydroxy-N-(3-nitrophenyl)naphthalene-2-carboxamide (2), and 2-hydroxy-N-(4-nitrophenyl)-naphthalene-1-carboxamide (6) showed antiproliferative activity against THP-1 and MCF-7 cancer cells without affecting the proliferation of 3T3-L1 non-tumour cells. Compounds 2 and 6 induced the accumulation of THP-1 and MCF-7 cells in G1 phase associated with the downregulation of cyclin E1 protein levels, while the levels of cyclin B1 were not affected. Moreover, compound 2 was found to exert the pro-apoptotic effect on the THP-1 cells. These results suggest that hydroxynaphthanilides might represent a potential model structure for the development of novel anticancer agents. hydroxynaphthanilidessalicylanilidescell proliferationapoptosisanticancer effect ==== Body 1. Introduction Salicylanilide derivatives (N-substituted hydroxybenzamides) are known as multitarget agents that possess a wide spectrum of pharmacological activities. These compounds are largely investigated for their promising antibacterial and antimycobacterial effects [1,2,3,4,5]. Some salicylanilides, such as niclosamide or closantel, belong to the class of broad-spectrum anthelmintic agents [6]. Recently, using high-throughput screening, several studies uncovered an antitumor activity of niclosamide, thereby becoming widely studied as a potential anticancer agent [7]. It was proved to effectively induce growth inhibition in a broad spectrum of tumour cell lines together with a minimal toxicity on non-tumour cells [8,9]. On the molecular level, niclosamide inhibited multiple key oncogenic signalling pathways (e.g., Wnt/β-catenin, mTORC1, and NF-κB) [9,10,11,12]. In general, salicylanilide derivatives are presumed to share the structure similarity with the pharmacophore of 4-arylaminoquinazoline derivatives (e.g., gefitinib and erlotinib) that belong to the class of small molecule inhibitors of the protein kinase epidermal growth factor receptor (EGFR PTK) [13,14,15]. This fact led to the intensive research of salicylanilides anticancer properties, as their structure became an attractive model for the design of potent antitumor agents. Several studies were published, in which a series of newly-prepared salicylanilides showed antiproliferative activity against a spectrum of human cancer cell lines, such as promyelocytic leukaemia cells HL-60, chronic myelogenous leukaemia cells K562, human epithelial carcinoma cells A431, or breast carcinoma cells MCF-7. In addition, some salicylanilides have been recently reported to elicit cell cycle arrest or to induce apoptosis in human cancer cell lines [13,16,17,18]. Recently, several series of various ring-substituted hydroxynaphthanilides were designed and prepared as ring analogues of salicylanilides. Based on the principle of bioisosterism with quinoline-like compounds, the aromatic ring in the salicylanilide pharmacophore was extended by another to obtain the naphthalene scaffold in the structure [3,5]. Compounds containing a quinoline moiety exhibit various pharmacological effects, including anticancer activity [19], hence the hydroxynaphthanilides may possess promising pharmacological properties due to the connection of these two pharmacophores. The biological activity of salicylanilide pharmacophore could be modified by introducing appropriate substituents in the structure. In addition to a substitution pattern on the salicylic scaffold, SAR studies were focused also on substituents located on the aromatic ring of the anilide part in the structure. It was proved that the biological effects of salicylanilide derivatives are related to both the nature and the position of substituents. The electron parameters of anilide substituents could modify the conformational equilibrium between the closed-ring and open-ring forms of the structure and thus affect the biological activity of the whole molecule. That activity is usually referred to the presence of electron-withdrawing substituents on the anilide moiety [14,20]. In accordance with these findings, our previous results revealed the same relation between the toxicity of ring-substituted hydroxynaphthanilides to the THP-1 cancer cells and the presence of substituents with electron-withdrawing properties [3,4,5]. The SAR studies also found the presence of an electron-withdrawing nitro group to be one of the essential requirements for the anticancer effect of niclosamide [21]. Based on these findings, the substitution by a nitro moiety was determined to be appropriate for the potent anticancer effect of newly-designed hydroxynaphthanilides. Therefore, we have selected six newly-designed hydroxynaphthanilides, nitro-substituted in different positions on the anilide ring (Table 1), to evaluate their potential anticancer effects in the context of these structural differences. The aim of this work was to assess their antiproliferative activity in two cancer cell lines, THP-1 and MCF-7. Moreover, we also examined the effect on the growth of non-tumour cells 3T3-L1. In addition, changes in cell cycle distribution were evaluated, as well as their pro-apoptotic effect. 2. Results 2.1. Effect on Cell Proliferation and Viability Initially, we examined the effect of six nitro-substituted hydroxynaphthanilides on the proliferation of human leukaemia and breast carcinoma cell lines, using Water Soluble Tetrazolium Salts-1 (WST-1) assay. For such analyses, THP-1 and MCF-7 cells were treated with the compounds at concentrations ranging from 0.5 to 20 μM for 24 h. As shown in Figure 1a, compounds 2, 3, and 6 inhibit cell growth in both cell lines in a dose-dependent manner. The inhibitory effect of 2 and 6 was statistically significant (p < 0.001) starting from the concentration of 2.5 and 5 μM in THP-1 and MCF-7 cells, respectively. From the concentration-response curves, the IC50 values were determined. As summarized in Table 2, the IC50 values were found to be 3.06 μM in THP-1 and 4.61 μM in MCF-7 cells for compound 2, and 5.80 and 5.23 μM in THP-1 and MCF-7 cells, respectively, for compound 6. The strongest antiproliferative effect was observed in both THP-1 and MCF-7 cell lines after the treatment with compound 3 (IC50 1.05 and 1.65 μM, respectively). In contrast, neither compound 1 nor 4 (both ortho-substituted derivatives) was able to induce the inhibition of cell growth in THP-1 or MCF-7 cells at concentrations used in the assay. Compound 5 demonstrated antiproliferative activity only in MCF-7 cells, significant (p < 0.001) at concentrations of 10 and 20 μM (data not shown), however, a 50% reduction in cell growth was not achieved. The proliferation of THP-1 cells was not affected by this compound. After we found that compounds 2, 3, and 6 effectively inhibit the growth of both THP-1 and MCF-7 cancer cells at micromolar concentrations, we assessed additionally their effect on proliferation of non-tumour cell line, 3T3-L1, using WST-1 assay. While compounds 2 and 6 did not decrease cell growth at any of concentrations used, compound 3 affected the proliferation of 3T3-L1 cells in a dose-dependent manner (IC50 4.41 μM) (Figure 1b and Table 2). Subsequently, for the comparison of the antiproliferative and cytotoxic effects we assessed the cell viability after 24 h treatment with compounds 1–6 in both tumour cell lines using the dye exclusion test. In THP-1 cells, we obtained lower LC50 values: 7.91, 3.44, and 9.98 μM for compounds 2, 3, and 6, respectively. In general, less sensitivity towards the cytotoxic effect of tested compounds was observed in MCF-7 cells. Neither compound 2 nor 6 reduced cell viability under 50% in comparison with the control, while the strongest effect was induced by compound 3 (LC50 12.91 μM). 2.2. Effect on Distribution of Cells in Cell Cycle Phases The cell proliferation assays showed us the ability of selected compounds 2 and 6 to inhibit cancer cell growth. In order to determine at which stage of the cell cycle these compounds induce cell growth inhibition, flow cytometric analyses of cell cycle profiles in THP-1 and MCF-7 cell lines were performed. Cells were exposed to compounds 2 and 6 for 24 h at concentrations exerting significant inhibition of cell proliferation with no or very little concurrent effect on the cell viability. Therefore, THP-1 and MCF-7 cells were treated for 24 h with the compounds at concentrations of 2.5, 5, and 10 μM, respectively. In general, we detected a qualitatively similar effect on the distribution of cells in cell cycle phases following the treatment with compounds 2 and 6 in both leukaemia and breast carcinoma cells. Compounds 2 and 6 induced accumulation of cells in G1 phase in both THP-1 (Figure 2) and MCF-7 (Figure 3) cell lines. This was in concert with a simultaneous decrease in the number of cells observed in the S phase compared to the drug-free control, while the percentage of cells in the G2/M phase remained unchanged. Additionally, the cell cycle analysis allows determining the presence of a subdiploid cell population as a characteristic marker of cells with fractional DNA content. A significant increase (p < 0.001) of the sub-G1 peak was found only after the treatment with 5 µM of compound 2 in THP-1 cells, where an approximately eight-fold increase was observed compared to the drug-free control (Figure 4). In contrast, compound 2 did not induce any elevation of the sub-G1 peak in breast carcinoma cells. Similarly, no significant increase of sub-diploid population of THP-1 or MCF-7 cells caused by 24 h treatment with compound 6 in comparison with the control sample was detected. Next, based on the flow cytometric data that showed the accumulation of cells in the G1 phase upon the treatment with compounds 2 and 6, we examined their effect on the expression of regulatory proteins controlling G1/S and G2/M progression. Whereas total protein levels of cyclin B1 were not changed in THP-1 or MCF-7 cells, the treatment with both compounds 2 and 6 led to the dose-dependent decrease in expression of cyclin E1 (Figure 2c and Figure 3c). Importantly, the levels of cyclin E1 low molecular weight (LMW E1) isoform (42 kDa) were found to be significantly decreased in THP-1 cells. 2.3. Detection of Apoptosis by Annexin V-FITC/PI Assay To further examine possible pro-apoptotic effect of compound 2 on THP-1 cells, Annexin V-FITC/PI assay was performed using flow cytometry for the quantification of the early and late stages of apoptosis. Staining of cells by Annexin V-FITC conjugate reflects the externalization of phosphatidylserine on the outer surface of the cell membrane as one of the early indicators of apoptosis [22]. In order to obtain further insight into the mechanism of cell death induced by compound 2, we exposed THP-1 cells to a wider concentration range of 2.5, 5, and 10 µM and subsequently analysed the effect at three time-points of incubation (12, 18, and 24 h). The assay revealed that compound 2 induced a dose-dependent increase of the percentage of early apoptotic as well as late apoptotic/secondary necrotic leukaemia THP-1 cells. In correspondence with the previous detection of a subdiploid cell population compound 2, at concentrations of 2.5 µM, 5 µM, and 10 µM, elicited elevations of Annexin V/FITC-stained cell populations. As shown in Figure 5, this effect was observed even after 12 h of incubation; 10 µM of compound 2 increased significantly (p < 0.01) the proportion of early apoptotic cells to 9.37% in comparison to the percentage of control cells, 2.41%. The same concentration of compound 2 induced the elevation of the number of double-stained cells with incubation time, from 22.48% after 12 h to 41.88% after 24 h incubation. In general, the percentage of late apoptotic/secondary necrotic cells at higher concentrations of compound 2 (5 and 10 µM) prevailed over the early apoptotic cell population at all determined time points. Nevertheless, the different effect was observed after the treatment with two model compounds exerting the pro-apoptotic effect in THP-1 cells. As summarized in Figure 5, cisplatin was found to most effectively increase the rate of early apoptotic cells in a time-dependent manner up to 44.38% after 24 h exposure. While camptothecin increased significantly (p < 0.001) the percentage of both early and late apoptotic cells up to 21.28% and 24.10%, respectively, after 12 h, 24 h treatment led to a decrease of the early apoptotic population to 9.65%; in contrast, late apoptosis increased to 33.08%. 2.4. Analysis of Proteins Levels Involved in Apoptotic Pathways Most of the apoptotic signalling pathways are controlled by caspases that belong to a group of cysteine proteases [23]. To assess whether compound 2 affects these signalling cascades and which pathway is activated (intrinsic or extrinsic), the activities of caspase 3, caspase 9, and caspase 8 were evaluated using Western blot analysis. As summarized in Figure 6, after 24 h incubation, compound 2 induced cleavage of pro-caspase 3 dose-dependently; an approximately two-fold decrease of the inactive form upon the treatment with 10 μM compared to the control was detected. Similarly, a comparable two-fold increase of active caspase 3 level was observed after the exposure to the 10 μM concentration of compound 2 in comparison to the control. Additionally, a significant increase of cleaved caspase 9 levels was detected with the most pronounced effects at 10 μM. On the contrary, the level of active caspase 8 was not altered after the treatment with compound 2 in comparison to the control. 3. Discussion In the present study, we examined the anticancer effects of a series of newly-synthesized nitro-substituted hydroxynaphthanilide derivatives through the assessment of their antiproliferative activity and cytotoxicity. Our results showed the difference among the tested compounds in the antiproliferative activity. We found that the potency of cell growth inhibition correlates with the position of the electron-withdrawing nitro group on the anilide ring of the tested compounds. While ortho-substituted derivatives did not elicit any antiproliferative effect in both THP-1 and MCF-7 cancer cells, the shift of the nitro group to the meta- or para-position in compounds 2, 3, and 6, led to the cell growth inhibition. Thus, it can be assumed that, most likely, the antiproliferative activity of 3-hydroxynaphthalene-2-carboxanilide and 2-hydroxynaphthalene-1-carboxanilide derivatives increase depending on the position of the nitro group as follows: ortho < meta < para. This different activity could be possibly related to the steric effect of the anilide substituents. Recently, it was described that the presence of a substituent in the ortho position causes the twist of the whole aniline ring plane towards the naphthalene scaffold, while meta- and, especially, para-substituted derivatives have a practically linear molecule [24]. Moreover, not only the location of the substituent on the anilide moiety but also the position of the β-ring of naphthalene towards the phenolic and carboxanilide moietis affected the intensity of the antiproliferative effect of these compounds. In our study, stronger antiproliferative activity was observed in substituted 3-hydroxynaphthalene-2-carboxanilides when comparing the IC50 values of meta-substituted compounds 2 and 4 or para-substituted 3 and 6. The similar structure-activity relationship was determined for the cytotoxicity of the tested compounds. Nevertheless, compounds 2, 3, and 6 exerted stronger antiproliferative rather than cytotoxic effect in cancer cells; approximately 2–3-fold higher LC50 values compared to IC50 values were obtained in the assays on THP-1 cells. Even more pronounced difference was observed in MCF-7 cells, where the LC50 values were achieved only upon the treatment with compound 3, with an approximately seven-fold higher dose in comparison with IC50. To assess whether tested compounds also influence the growth of other than cancer cells, we have extended our antiproliferative analysis and employed non-tumour fibroblast cell line 3T3-L1. Compound 3 that exerted the most substantial antiproliferative and cytotoxic effects towards both cancer cell lines was also capable of inhibiting the growth of the non-tumour line. Interestingly, a different effect was observed upon the treatment with compounds 2 and 6, where such antiproliferative activity in non-tumour cells was not detected. Results of antiproliferative effects showed us that among all tested compounds, compounds 2 and 6 were the most potent and, thus, were chosen for further, more detailed analyses. One characteristic feature of cancer cells is the deregulation of the cell cycle, which leads to their uncontrolled proliferation. Therefore, the inhibition of cell cycle progression represents a common target of anticancer agents [25]. We performed the cell cycle analysis to reveal whether the antiproliferative effect of compounds 2 and 6 is reflected in the modification of cell cycle progression. Our results showed that both compounds were able to accumulate THP-1 and MCF-7 cancer cells in the G1 phase and to inhibit the transition of cells to the synthetic phase. We assume that this most likely reflects the antiproliferative effect observed in both cell lines (Figure 1a). The progression through the cell cycle is mediated by a family of cyclin-dependent kinases, the activity of which depends on the binding of the regulatory proteins, cyclins [26]. The observed accumulation of THP-1 and MCF-7 cells in the G1 phase after the treatment with compounds 2 and 6 was accompanied by a reduction of cyclin E1 level in a dose-dependent manner (Figure 2c). As the activator of CDK2, cyclin E1 is responsible for the G1/S phase progression and, thus, it is involved in surpassing the restriction point [27]. Many cancers typically overexpress cyclin E1, which is also proved in the MCF-7 cell line [28]. This might support our finding of only slight downregulation of cyclin E1 caused by the treatment of MCF-7 cells with compounds 2 and 6, although these compounds effectively inhibited the G1/S transition. Interestingly, besides the downregulation of cyclin E1 full-length form, we also detected a more pronounced reduction of LMW E1 isoform levels in THP-1 cells treated with compounds 2 and 6. LMW E1 isoforms are generated primarily in cancer cells, where they still remain fully functional. They have even higher potency to increase CDK2/E1 activity than the full-length form and, thus, they move the cells through the cell cycle more effectively than the full-length form [29,30]. Our previous study reported a similar detection of the decreased levels of cyclin E1 isoforms in THP-1 cells treated with geranylated flavanone tomentodiplacone B that coincided with an induced accumulation of cells in G1 phase [31]. While cyclin B1 is involved in the G2/M transition associated with CDK1 [26], we did not observe any change in the levels of cyclin B1 in THP-1 or MCF-7 cells after the exposure to compounds 2 and 6. These findings are supported by our flow cytometric data that did not indicate any significant difference in the proportion of cells in G2/M cell cycle phase upon the treatment with these compounds (Figure 2b and Figure 3b). Based on those results, we could suggest that compounds 2 and 6 most likely affect G1/S rather than the G2/M transition. The presence of cell nuclei with hypodiploid DNA content during the cell cycle analysis could indicate a possible presence of apoptotic cells [32]. The assessment of sub-G1 peak levels revealed different effects among the tested compounds; a significant increase was detected only in THP-1 cells upon the treatment with compound 2 (Figure 4). Based on these findings, we performed further analysis to prove its possible pro-apoptotic effect in the THP-1 cell line. Results of Annexin V-FITC/PI assay showed us that compound 2 induced the THP-1 cells to undergo an early stage of apoptosis even after 12 h exposure (Figure 5). Nevertheless, compound 2 accumulated more effectively (dose and time-dependently) in cells in the late apoptotic stage. These results correlate with the data obtained from the viability staining assay. In addition, two already known anticancer agents of a different mode of action, cisplatin, which is able to crosslink with the DNA and, thus, cause DNA damage [33], and camptothecin as the S-phase-specific inhibitor of the enzyme DNA topoisomerase-I [34], were added to the assay as model compounds with proved pro-apoptotic effects in THP-1 cells [35,36]. Although our results found all three compounds to significantly increase the number of cells positive for Annexin V-FITC staining, their effect led to different proportions of early and late apoptotic/secondary necrotic cells. While cisplatin induced a time-dependent substantial increase in the fraction of early apoptotic cells, camptothecin most likely elicited the time-dependent transfer of cells from early apoptotic to late apoptotic stages. These differences observed in the effect of three tested compounds enable us to presume a different mechanism of action of compound 2 in comparison with one of the two model anticancer agents. These findings prompted us to further investigate the involvement of compound 2 in the apoptotic pathways. The caspases regulate the process of apoptosis in a different manner. The activation of caspase 8 is realized through the extrinsic apoptotic pathway after the binding of a ligand to an appropriate death receptor. Subsequently, the active form interacts with effector caspase 3 and that results in its cleavage and activation. On the other hand, initiator caspase 9 is involved in the intrinsic, also known as the mitochondrial apoptosis pathway, and is activated after the leakage of the mitochondrial cytochrome c. This also leads to proteolytic cleavage of inactive procaspase 3 and to its activation. Therefore, it denotes the essential role of caspase 3 in both extrinsic and intrinsic pathways, as it also comprises a link between them [37,38]. After 24 h treatment, compound 2 was found to be capable of inducing an increase of active caspase 3 level, including the decreased level of inactive pro-caspase 3, both significantly at a concentration of 10 µM (Figure 6). At the same time, compound 2 caused also the cleavage of pro-caspase 9. On the contrary, no change in the level of the active form of caspase 8 was observed in comparison with the control, non-treated cells. These results indicate that compound 2 induces apoptosis in THP-1 cells by activating a caspase cascade. In addition, we could hypothesize that this compound might be preferably involved in the intrinsic apoptotic pathway. However, such specificity needs to be proved by additional analyses, and the mechanism of targeting apoptotic pathway remains unknown. 4. Materials and Methods 4.1. Chemicals and Reagents The tested nitro-substituted hydroxynaphthanilides 1−6 were prepared and supplied by the Department of Chemical Drugs, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Czech Republic. The synthesis and structural characterization of these compounds have been described previously [3,5]. Due to poor solubility in water, the compounds were dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO, USA), while the stock solutions were prepared freshly before each experiment. The final concentration of DMSO in the assays never exceeded 0.1% (v/v). Cisplatin and camptothecin were purchased from Sigma-Aldrich. RPMI 1640 and DMEM culture media, phosphate-buffered saline (PBS), foetal bovine serum (FBS) and antibiotics (penicillin and streptomycin) were obtained from HyClone Laboratories, Inc. (GE Healthcare, Logan, UT, USA). Mouse monoclonal antibodies against cyclin E1 (sc-247), caspase 3 (sc-7272) and caspase 9 (sc-17784) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal antibodies against cyclin B1 (ab2949) and caspase 8 (ab-25901) were purchased from Abcam (Cambridge, UK). All other reagents, unless specified elsewhere, were purchased from Sigma-Aldrich. 4.2. Cell Culture THP-1 human monocytic leukemia cell line, MCF-7 human breast adenocarcinoma cells and 3T3-L1 mouse embryonic fibroblast were purchased from the European Collection of Cell Cultures (ECACC, Salisbury, UK). Cells were routinely tested for the absence of mycoplasma (Hoechst 33258 staining method). THP-1 cells were maintained in RPMI 1640 culture medium containing 2 mM l-glutamine; MCF-7 and 3T3-L1 cells were cultured in DMEM medium. All of the culture media were supplemented with 10% heat-inactivated FBS and antibiotics (100 U/mL penicillin and 100 μg/mL streptomycin). Cells were maintained at 37 °C in a humidified atmosphere containing 5% CO2. 4.3. Analysis of Cell Proliferation and Viability Cell proliferation was evaluated using Cell Proliferation Reagent WST-1 (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium) (Roche Diagnostics, Mannheim, Germany) according to the manufacturer´s instructions. THP-1 (5 × 104 cells/100 μL culture medium per well), MCF-7 cells (1 × 104 cells/100 μL per well), and 3T3-L1 (2.5 × 103 cells/100 μL per well) were cultured in 96-well plates in triplicate. The measurement was performed using Synergy 2 Multi-Mode Microplate Reader (BioTek, Winooski, VT, USA) after 24 h incubation of cells with tested compounds dissolved in DMSO and subsequently in RPMI 1640 to final concentration ranging 0.5–20 μM in the assays. Cell viability was assessed by dye exclusion test. THP-1 (2 × 105 cells/mL per well) and MCF-7 cells (8 × 104 cells/mL per well) were incubated in 24-well plates with the indicated concentrations of compounds for 24 h. The number of viable cells was determined using hemocytometer after their staining with a solution of erythrosin B (0.1% erythrosin B (w/v) in PBS). The assays were conducted in triplicate. The IC50 and LC50 values were calculated from fitted concentration-response curves using GraphPad Prism 5.00 software (GraphPad Software, San Diego, CA, USA). 4.4. Cell Cycle Analysis THP-1 and sub-confluent MCF-7 cells were treated and subsequently incubated with indicated concentrations of compounds 2 and 6 for 24 h. After the incubation, cells were washed twice in PBS (pH 7.4), fixed in 70% ethanol and stored at −20 °C overnight. Fixed cells were collected by centrifugation, and supernatant was discarded. The cell pellet was washed twice with PBS and incubated with RNaseA (0.02 mg/mL) and 0.05% (v/v) Triton X-100 in PBS for 30 min at 37 °C. After the nuclei staining with propidium iodide (PI) (0.04 mg/mL), the cell cycle distribution was analysed using a flow cytometer Cell Lab Quanta SC (Beckman Coulter, Brea, CA, USA). The quantification of cell cycle distribution was carried out using software MultiCycle AV (Phoenix Flow System, San Diego, CA, USA). A total number of 2 × 104 cells was analysed per sample. 4.5. Detection of Apoptosis Using Annexin V-FITC/PI Assay Early and late stages of apoptosis were detected using Annexin V-FITC Kit—Apoptosis Detection Kit faccording to the manufacturer´s instructions. THP-1 cells were treated with increasing concentrations of compound 2 (2.5, 5, and 10 μM), cisplatin (10 μg/mL) and camptothecin (5 μM). At each time-point of incubation (12, 18, and 24 h) the cells were washed with ice-cold PBS prior to being resuspended at a concentration of 5 × 106 cells/mL in a total volume of 100 μL of 1× binding buffer. Annexin V-FITC solution (final concentration 0.25 μg/mL) and PI (final concentration 12.5 μg/mL) were added to each sample; the cell suspension was kept on ice and incubated for 15 min in the dark. After that, the analysis was carried out by flow cytometry. The data were evaluated using Kaluza Flow Cytometry Analysis 1.2. Per sample, a total number of 2 × 104 cells were analysed. 4.6. Western Blotting For Western blotting, cells were washed with PBS and lysed in lysis buffer (100 mM Tris-HCl, pH = 6.8; 20% glycerol; 1% SDS) containing protease and phosphatase inhibitor cocktails. Protein concentration was measured using Roti®-Quant universal (Carl Roth, Karsruhe, Germany) according to the manufacturer’s instructions. Cell lysates were supplemented with bromophenol blue (final concentration 0.01% (w/v)) and β-mercaptoethanol (final concentration 1% (v/v)) prior to being heated for 5 min at 95 °C. Equal amounts of protein (10 μg) were loaded into a 12% polyacrylamide gel, separated by SDS-polyacrylamide gel electrophoresis and subsequently electrotransferred onto nitrocellulose membranes. Reversible Ponceau S. staining was performed to assess equal sample loading. Then, the membranes were blocked with 5% non-fat dry milk in TBST (10 mM Tris-HCl pH = 7.5, 150 mM NaCl, 0.1% (v/v) Tween-20) and appropriate primary and secondary antibodies were used for immunodetection. The proteins were visualized by ECL Plus reagent according to the manufacturer’s instructions. The intensity of bands was semi-quantitatively analysed using the ImageJ software (National Institute of Mental Health, Bethesda, MD, USA). 4.7. Statistical Analysis All experimental data were expressed as the arithmetical mean ± standard deviation (SD). Statistical analysis was performed using one-way analysis of variance (ANOVA) followed by the Dunnett’s post test using GraphPad Prism 5.00 software. Statistical significance was assessed at levels of p < 0.05, p < 0.01 and p < 0.001. 5. Conclusions The present study provides the first description of the antiproliferative activity of nitro-substituted hydroxynaphthanilides in the context of structure-activity relationships. Our results indicate that the potency of ring-substituted hydroxynaphthanilides towards cell growth inhibition increases with positioning of the nitro group as follows: ortho < meta < para. The most promising compounds 2 and 6 exerted antiproliferative activity in THP-1 and MCF-7 cancer cells with single-digit micromolar IC50 values, while they had a minimal effect on the growth of 3T3-L1 non-tumour cells. Compounds 2 and 6 accumulated cancer cells THP-1 and MCF-7 in G1 cell cycle phase, which was accompanied by the observed down-regulation of cyclin E1 levels. Moreover, compound 2 was found to induce apoptosis in THP-1 cells via a caspase-mediated cascade. The results also indicate that apoptosis was probably induced through the intrinsic apoptotic pathway, although further analysis is still required to verify such assumption. According to the results, nitro-substituted hydroxynaphthanilides 2 and 6 can be considered as potential anticancer agents, and the structure of hydroxynaphthanilides is an appropriate model moiety for further design of compounds with potential anticancer properties. Acknowledgments This work was supported by the Internal Grant Agency of the University of Veterinary and Pharmaceutical Sciences Brno (323/2015/FaF). Author Contributions Peter Kollar conceived and designed the experiments; Tereza Kauerova performed the experiments and analyzed the data; Jiri Kos, Tomas Gonec and Josef Jampilek contributed reagents/materials/analysis tools; Tereza Kauerova and Peter Kollar wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Effect of compounds 2, 3, and 6 on cell proliferation and viability in THP-1, MCF-7 and 3T3-L1 cell lines. Cells were cultured with indicated concentrations of compounds 2, 3, and 6 for 24 h. (a) Proliferation of THP-1 and MCF-7 cells was determined using WST-1 assay; cell viability was assessed by erythrosin B exclusion test; (b) Proliferation of 3T3-L1 cells was determined using WST-1 assay. The results are shown as the means ± standard deviation (SD) of three independent experiments, each performed in triplicate. ** p < 0.01, *** p < 0.001, statistically significant difference in comparison with drug-free control (CTRL). Figure 2 Compounds 2 and 6 induce accumulation of THP-1 cells in the G1 phase. (a) Representative histograms of flow cytometric analysis of the DNA content in THP-1 cells after the incubation with indicated concentrations of compounds 2 and 6 for 24 h; (b) The distribution of THP-1 cells in the phases of the cell cycle upon the treatment with compounds 2 and 6 at 24 h. The results are expressed as the means ± SD of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001, statistically significant difference in comparison with control sample; (c) Expression of cell cycle regulators cyclin E1 and B1 in THP-1 cells treated by compounds 2 and 6 for 24 h, as determined by Western blot analysis. Protein levels of the samples were normalized according to the total protein stains. CTRL, control cells treated by the drug-free medium. Figure 3 Compounds 2 and 6 induce accumulation of MCF-7 cells in the G1 phase. (a) Representative histograms of flow cytometric analysis of the DNA content in MCF-7 cells after the incubation with indicated concentrations of compounds 2 and 6 for 24 h; (b) The distribution of MCF-7 cells in phases of the cell cycle upon the treatment with compounds 2 and 6 at 24 h. The results are expressed as the means ± SD of three independent experiments. *** p < 0.001, statistically significant difference in comparison with control sample; (c) Expression of cell cycle regulators cyclin E1 and B1 in MCF-7 cells treated by compounds 2 and 6 for 24 h, as determined by Western blot analysis. Protein levels of the samples were normalized according to the total protein stains. CTRL, control cells treated by the drug-free medium. Figure 4 Compound 2 causes a significant increase of hypodiploid sub-G1 peak in THP-1 cells. Quantification of the sub-G1 peak in THP-1 cells after the treatment by compounds 2 and 6 for 24 h. The results are expressed as the means ± SD of three independent experiments. *** p < 0.001, statistically significant difference in comparison with the drug-free control (CTRL). Figure 5 Detection of apoptosis after treatment with compound 2 in THP-1 cells at three points of incubation (12, 18, and 24 h). Cells were stained by Annexin V-FITC conjugate and PI; subsequent analysis was performed by flow cytometry. Cisplatin (10 µg/mL) and camptothecin (5 µM) were used as model compounds. (a) Representative dot plots of Annexin V-FITC/PI assay are shown. The particular quadrants represent proportion of cells that are I: viable; II: early apoptotic; III: late apoptotic/secondary necrotic; IV: necrotic; (b) Proportion of early apoptotic and late apoptotic/secondary necrotic THP-1 cells after the treatment by compound 2 and model compounds. The results are expressed as the means ± SD of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001, statistically significant difference in comparison with the drug-free control (CTRL). Figure 6 Levels of proteins involved in apoptotic pathways in THP-1 cells after 24 h treatment by compound 2. (a) Levels of caspase 3, caspase 8, and caspase 9 in THP-1 cells treated by compound 2 for 24 h, as determined by Western blot analysis. Data of typical immunoblot are reported; (b) Summary data of cleaved caspase 9 levels in THP-1 cells; (c) Summary data of pro-caspase 3 levels in THP-1 cells; (d) Summary data of cleaved caspase 3 levels in THP-1 cells. Protein levels of the samples were normalized according to the total protein stains. The results are expressed as the means ± SD of three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001, statistically significant difference in comparison with the drug-free control (CTRL). ijms-17-01219-t001_Table 1Table 1 Structures of tested compounds: (a) 3-hydroxy-N-(nitrophenyl)naphthalene-2-carboxamides; and (b) 2-hydroxy-N-(nitrophenyl)naphthalene-1-carboxamide. (a) (b) Compound R Compound R 1 2-NO2 4 2-NO2 2 3-NO2 5 3-NO2 3 4-NO2 6 4-NO2 ijms-17-01219-t002_Table 2Table 2 Antiproliferative and cytotoxic effects of tested compounds 1−6. IC50 and LC50 values were calculated using concentration-response curves generated from the results of WST-1 analysis and erythrosin B exclusion test, respectively. The values represent means ± SD of three independent experiments, each performed in triplicate. Compound THP-1 MCF-7 3T3-L1 IC50 (μM) LC50 (μM) IC50 (μM) LC50 (μM) IC50 (μM) 1 >20 >20 >20 >20 >20 2 3.06 ± 0.206 7.91 ± 0.240 4.61 ± 0.068 >20 >20 3 1.05 ± 0.199 3.44 ± 1.209 1.65 ± 0.938 12.91 ± 1.984 4.41 ± 0.293 4 >20 >20 >20 >20 >20 5 >20 >20 >20 >20 >20 6 5.80 ± 0.370 9.98 ± 0.349 5.23 ± 0.802 >20 >20 ==== Refs References 1. Kratky M. Vinsova J. Novotna E. Mandikova J. Wsol V. Trejtnar F. Ulmann V. Stolarikova J. Fernandes S. Bhat S. Salicylanilide derivatives block mycobacterium tuberculosis through inhibition of isocitrate lyase and methionine aminopeptidase Tuberculosis (Edinburgh) 2012 92 434 439 10.1016/j.tube.2012.06.001 22765970 2. Zadrazilova I. Pospisilova S. Masarikova M. Imramovsky A. Ferriz J.M. Vinsova J. Cizek A. Jampilek J. Salicylanilide carbamates: Promising antibacterial agents with high in vitro activity against methicillin-resistant staphylococcus aureus (MRSA) Eur. J. Pharm. Sci. 2015 77 197 207 10.1016/j.ejps.2015.06.009 26079401 3. Gonec T. Kos J. Zadrazilova I. Pesko M. Govender R. Keltosova S. Chambel B. Pereira D. Kollar P. Imramovsky A. Antibacterial and herbicidal activity of ring-substituted 2-hydroxynaphthalene-1-carboxanilides Molecules 2013 18 9397 9419 10.3390/molecules18089397 23924993 4. Gonec T. Kos J. Zadrazilova I. Pesko M. Keltosova S. Tengler J. Bobal P. Kollar P. Cizek A. Kralova K. Antimycobacterial and herbicidal activity of ring-substituted 1-hydroxynaphthalene-2-carboxanilides Bioorg. Med. Chem. 2013 21 6531 6541 10.1016/j.bmc.2013.08.030 24075143 5. Kos J. Zadrazilova I. Pesko M. Keltosova S. Tengler J. Gonec T. Bobal P. Kauerova T. Oravec M. Kollar P. Antibacterial and herbicidal activity of ring-substituted 3-hydroxynaphthalene-2-carboxanilides Molecules 2013 18 7977 7997 10.3390/molecules18077977 23884130 6. Olliaro P. Seiler J. Kuesel A. Horton J. Clark J.N. Don R. Keiser J. Potential drug development candidates for human soil-transmitted helminthiases PLoS Negl. Trop. Dis. 2011 5 1219 10.1371/journal.pntd.0001138 21695247 7. Mudduluru G. Walther W. Kobelt D. Dahlmann M. Treese C. Assaraf Y.G. Stein U. Repositioning of drugs for intervention in tumor progression and metastasis: Old drugs for new targets Drug Resist. Updat. 2016 26 10 27 10.1016/j.drup.2016.03.002 27180307 8. Osada T. Chen M. Yang X.Y. Spasojevic I. Vandeusen J.B. Hsu D. Clary B.M. Clay T.M. Chen W. Morse M.A. Antihelminth compound niclosamide downregulates Wnt signaling and elicits antitumor responses in tumors with activating APC mutations Cancer Res. 2011 71 4172 4182 10.1158/0008-5472.CAN-10-3978 21531761 9. Li Y. Li P.K. Roberts M.J. Arend R.C. Samant R.S. Buchsbaum D.J. Multi-targeted therapy of cancer by niclosamide: A new application for an old drug Cancer Lett. 2014 349 8 14 10.1016/j.canlet.2014.04.003 24732808 10. Fonseca B.D. Diering G.H. Bidinosti M.A. Dalal K. Alain T. Balgi A.D. Forestieri R. Nodwell M. Rajadurai C.V. Gunaratnam C. Structure-activity analysis of niclosamide reveals potential role for cytoplasmic pH in control of mammalian target of rapamycin complex 1 (mTORC1) signaling J. Biol. Chem. 2012 287 17530 17545 10.1074/jbc.M112.359638 22474287 11. Lu W. Lin C. Roberts M.J. Waud W.R. Piazza G.A. Li Y. Niclosamide suppresses cancer cell growth by inducing Wnt co-receptor LRP6 degradation and inhibiting the Wnt/β-catenin pathway PLoS ONE 2011 6 1219 10.1371/journal.pone.0029290 22195040 12. Jin Y. Lu Z. Ding K. Li J. Du X. Chen C. Sun X. Wu Y. Zhou J. Pan J. Antineoplastic mechanisms of niclosamide in acute myelogenous leukemia stem cells: Inactivation of the NF-kappaB pathway and generation of reactive oxygen species Cancer Res. 2010 70 2516 2527 10.1158/0008-5472.CAN-09-3950 20215516 13. Kang B.-R. Shan A.-L. Li Y.-P. Xu J. Lu S.-M. Zhang S.-Q. Discovery of 2-aryl-8-hydroxy (or methoxy)-isoquinolin-1(2H)-ones as novel EGFR inhibitor by scaffold hopping Bioorg. Med. Chem. 2013 21 6956 6964 10.1016/j.bmc.2013.09.027 24094432 14. Liechti C. Sequin U. Bold G. Furet P. Meyer T. Traxler P. Salicylanilides as inhibitors of the protein tyrosine kinase epidermal growth factor receptor Eur. J. Med. Chem. 2004 39 11 26 10.1016/j.ejmech.2003.09.010 14987830 15. Deng W. Guo Z. Guo Y. Feng Z. Jiang Y. Chu F. Acryloylamino-salicylanilides as EGFR PTK inhibitors Bioorg. Med. Chem. Lett. 2016 16 469 472 10.1016/j.bmcl.2005.06.088 16275081 16. Zhu X.F. Wang J.S. Cai L.L. Zeng Y.X. Yang D. SUCI02 inhibits the erbb-2 tyrosine kinase receptor signaling pathway and arrests the cell cycle in G1 phase in breast cancer cells Cancer Sci. 2006 97 84 89 10.1111/j.1349-7006.2006.00143.x 16367926 17. Imramovsky A. Jorda R. Pauk K. Reznickova E. Dusek J. Hanusek J. Krystof V. Substituted 2-hydroxy-N -(arylalkyl)benzamides induce apoptosis in cancer cell lines Eur. J. Med. Chem. 2016 68 253 259 10.1016/j.ejmech.2013.08.009 23981532 18. Zuo M. Zheng Y.W. Lu S.M. Li Y. Zhang S.Q. Synthesis and biological evaluation of N -aryl salicylamides with a hydroxamic acid moiety at 5-position as novel HDAC-EGFR dual inhibitors Bioorg. Med. Chem. 2012 20 4405 4412 10.1016/j.bmc.2012.05.034 22698782 19. Solomon V.R. Lee H. Quinoline as a privileged scaffold in cancer drug discovery Curr. Med. Chem. 2011 18 1488 1508 10.2174/092986711795328382 21428893 20. Guo L. Wang Q.L. Jiang Q.Q. Jiang Q.J. Jiang Y.B. Anion-triggered substituent-dependent conformational switching of salicylanilides. New hints for understanding the inhibitory mechanism of salicylanilides J. Org. Chem. 2007 72 9947 9953 10.1021/jo701823d 18020364 21. Walters Haygood C.L. Arend R.C. Gangrade A. Chettiar S. Regan N. Hassmann C.J. 2nd Li P.K. Hidalgo B. Straughn J.M. Jr. Buchsbaum D.J. Niclosamide analogs for treatment of ovarian cancer Int. J. Gynecol. Cancer 2015 25 1377 1385 10.1097/IGC.0000000000000506 26186072 22. Brumatti G. Sheridan C. Martin S.J. Expression and purification of recombinant annexin V for the detection of membrane alterations on apoptotic cells Methods 2008 44 235 240 10.1016/j.ymeth.2007.11.010 18314054 23. Kumar S. Caspase function in programmed cell death Cell Death Differ. 2007 14 32 43 10.1038/sj.cdd.4402060 17082813 24. Malik I. Bukovsky M. Andriamainty F. Galisinova J. Antimicrobial activity of meta-alkoxyphenylcarbamates containing substituted N -phenylpiperazine fragment Braz. J. Microbiol. 2012 3 959 965 24031913 25. Williams G.H. Stoeber K. The cell cycle and cancer J. Pathol. 2012 226 352 364 10.1002/path.3022 21990031 26. Malumbres M. Barbacid M. Cell cycle, CDKs and cancer: A changing paradigm Nat. Rev. Cancer 2009 9 153 166 10.1038/nrc2602 19238148 27. Diaz-Moralli S. Tarrado-Castellarnau M. Miranda A. Cascante M. Targeting cell cycle regulation in cancer therapy Pharmacol. Ther. 2013 138 255 271 10.1016/j.pharmthera.2013.01.011 23356980 28. Guo X. Hartley R.S. Hur contributes to cyclin E1 deregulation in MCF-7 breast cancer cells Cancer Res. 2006 66 7948 7956 10.1158/0008-5472.CAN-05-4362 16912169 29. Wingate H. Bedrosian I. Akli S. Keyomarsi K. The low molecular weight (LMW) isoforms of cyclin E deregulate the cell cycle of mammary epithelial cells Cell Cycle 2003 2 461 466 10.4161/cc.2.5.464 12963845 30. Akli S. Keyomarsi K. Cyclin E and its low molecular weight forms in human cancer and as targets for cancer therapy Cancer Biol. Ther. 2003 2 S38 S47 10.4161/cbt.201 14508079 31. Kollar P. Barta T. Zavalova V. Smejkal K. Hampl A. Geranylated flavanone tomentodiplacone B inhibits proliferation of human monocytic leukaemia (THP-1) cells Br. J. Pharmacol. 2011 162 1534 1541 10.1111/j.1476-5381.2010.01171.x 21175584 32. Kajstura M. Halicka H.D. Pryjma J. Darzynkiewicz Z. Discontinuous fragmentation of nuclear DNA during apoptosis revealed by discrete “sub-G1” peaks on DNA content histograms Cytom. A 2007 71 125 131 10.1002/cyto.a.20357 17252584 33. Dasari S. Tchounwou P.B. Cisplatin in cancer therapy: Molecular mechanisms of action Eur. J. Pharmacol. 2014 740 364 378 10.1016/j.ejphar.2014.07.025 25058905 34. Liu L.F. Desai S.D. Li T.K. Mao Y. Sun M. Sim S.P. Mechanism of action of camptothecin Ann. N. Y. Acad. Sci. 2000 922 1 10 10.1111/j.1749-6632.2000.tb07020.x 11193884 35. Stephenson D.A. Toltl L.J. Beaudin S. Liaw P.C. Modulation of monocyte function by activated protein C, a natural anticoagulant J. Immunol. 2006 177 2115 2122 10.4049/jimmunol.177.4.2115 16887970 36. Amran D. Sancho P. Fernandez C. Esteban D. Ramos A.M. de Blas E. Gomez M. Palacios M.A. Aller P. Pharmacological inhibitors of extracellular signal-regulated protein kinases attenuate the apoptotic action of cisplatin in human myeloid leukemia cells via glutathione-independent reduction in intracellular drug accumulation Biochim. Biophys. Acta 2005 1743 269 279 10.1016/j.bbamcr.2004.10.009 15843040 37. McIlwain D.R. Berger T. Mak T.W. Caspase functions in cell death and disease Cold Spring Harb. Perspect. Biol. 2013 5 a008656 10.1101/cshperspect.a008656 23545416 38. Degterev A. Boyce M. Yuan J. A decade of caspases Oncogene 2003 22 8543 8567 10.1038/sj.onc.1207107 14634618

PMC005xxxxxx/PMC5000618.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081220ijms-17-01220ArticleTheoretical and Kinetic Tools for Selecting Effective Antioxidants: Application to the Protection of Omega-3 Oils with Natural and Synthetic Phenols Guitard Romain Nardello-Rataj Véronique *Aubry Jean-Marie *Colombo Maria Laura Academic EditorUniv. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181–UCCS-Unité de Catalyse et Chimie du Solide, F-59000 Lille, France; [email protected]* Correspondence: [email protected] (V.N.-R.); [email protected] (J.-M.A.); Tel.: +33-3-2033-6369 (V.N.-R.); +33-3-2033-6364 (J.-M.A.)29 7 2016 8 2016 17 8 122003 6 2016 21 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Radical-scavenging antioxidants play crucial roles in the protection of unsaturated oils against autoxidation and, especially, edible oils rich in omega-3 because of their high sensitivity to oxygen. Two complementary tools are employed to select, among a large set of natural and synthetic phenols, the most promising antioxidants. On the one hand, density functional theory (DFT) calculations provide bond dissociation enthalpies (BDEs) of 70 natural (i.e., tocopherols, hydroxybenzoic and cinnamic acids, flavonoids, stilbenes, lignans, and coumarins) and synthetic (i.e., 2,6-di-tert-butyl-4-methylphenol (BHT), 3-tert-butyl-4-hydroxyanisol (BHA), and tert-butylhydroquinone (TBHQ)) phenols. These BDEs are discussed on the basis of structure–activity relationships with regard to their potential antioxidant activities. On the other hand, the kinetic rate constants and number of hydrogen atoms released per phenol molecule are measured by monitoring the reaction of phenols with 2,2-diphenyl-1-picrylhydrazyl (DPPH•) radical. The comparison of the results obtained with these two complementary methods allows highlighting the most promising antioxidants. Finally, the antioxidant effectiveness of the best candidates is assessed by following the absorption of oxygen by methyl esters of linseed oil containing 0.5 mmol L−1 of antioxidant and warmed at 90 °C under oxygen atmosphere. Under these conditions, some natural phenols namely epigallocatechin gallate, myricetin, rosmarinic and carnosic acids were found to be more effective antioxidants than α-tocopherol. natural and synthetic phenolsantioxidantbond dissociation enthalpy (BDE)2,2-diphenyl-1-picrylhydrazyl (DPPH•)omega-3 fatty acid methyl esters (FAMEs)linseed oilautoxidationstoichiometric number ==== Body 1. Introduction Omega-3 essential fatty acids have drawn attention of scientists for many years and studies have multiplied in recent decades, highlighting their virtues and mandatory character to the proper functioning of human bodies [1,2]. Nevertheless, due to their large number of unsaturations, omega-3 oils are highly oxidizable. This process plays an important role in the degradation of the organoleptic properties of food. All lipids containing unsaturated fatty acids such as vegetable oils, fish oils, animal fats, cell membranes and lipoproteins are concerned with lipid peroxidation. In recent decades, mechanistic studies of lipid peroxidation have known a renewed interest because of their implication in the field of nutrition. Unsaturated lipids (LH) are prone to autoxidation, which takes place in three main steps. The first one is the initiation step which consists of the loss of a hydrogen atom triggered by metal traces, light or heat (Equation (1)). The resulting lipid radical (L•) reacts with fundamental oxygen (3O2) in a second step to form a peroxyl radical (LOO•) (Equation (2)). During the propagation stage, LOO• reacts with LH to form fatty acid hydroperoxides (LOOH) which are primary oxidation products (Equation (3)). In a third step, i.e., the termination step, two radicals react together to form non-radical products (Equations (4)–(6)) [3,4]. Hydroperoxides are unstable compounds that lead to alcoxyl (LO•) and peroxyl (LOO•) radicals which further form other oxidized products such as alcohols, aldehydes and ketones. One possible decomposition of lipid hydroperoxides is known as the Russel mechanism in which the combination of two peroxyl radicals LOO• provides a ketone L(O), an alcohol LOH and singlet molecular oxygen 1O2 which can take place in biological systems (Equation (4)). Cyclisation mechanisms can also be involved in the formation of cyclic peroxides [5]. Initiation (1) LH→Initiator L• + H• Propagation (2) L• + 3O2→LOO• (3) LOO• +LH→LOOH+ L• Termination (4) LOO• + LOO•→L(O)+LOH+ 1O2 (5) LOO• + L•→LOOL (6) L• + L•→LL In order to reduce the damages of these free radicals on food and biological systems, scientists are searching effective and non-toxic antioxidants [6]. Different factors influence the antioxidant power of phenols [7]: (i) Low value of Bond Dissociation Enthalpy (BDE) of the phenolic bond favors the transfer of the phenolic hydrogen to free radicals (i.e., R•, RO• and ROO•) [8,9,10,11,12,13]; (ii) High value of ionization potential (IP) avoids the transfer of electron from phenols to oxygen. Consequently, the pro-oxidant potential of the antioxidant is reduced [7,11,14,15,16,17]; (iii) high solubility of the phenol into the protected medium improves the antioxidant power [18,19]. There are numerous experimental and theoretical investigations dealing with bond dissociation enthalpies (BDEs) of antioxidants [14,20,21,22,23]. Nevertheless, they are sometimes inconsistent with each other. Indeed, such data are basis set and solvent dependent. It is then crucial to have a reliable method that can accurately predict the BDEs of a large scope of phenols and build a predictive scale of their antioxidant power, supported by experimental data. In this paper, we determine the BDEs of 70 natural (i.e., tocopherols, derivatives of hydroxybenzoic and cinnamic acids, flavonols, flavones, flavanonols, flavanones, isoflavones, flavanols, stilbenes, lignans, and coumarins) and synthetic (i.e., 2,6-di-tert-butyl-4-methylphenol (BHT), 3-tert-butyl-4-hydroxyanisol (BHA), tert-butylhydroquinone (TBHQ), and propyl gallate (PG)) antioxidants by density functional theory (DFT) calculation. The method is referred to as B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) and allows the calculation of accurate BDEs in relative short time. On the other hand, kinetic rate constants and number of hydrogen atoms released per molecule of phenol have also been measured by monitoring the reaction of phenols with DPPH• radical. The comparison of the results obtained with those two complementary methods allows highlighting the most promising antioxidants. Finally, the antioxidant effectiveness of the best candidates has been assessed under more realistic conditions by following during the oxidation process the absorption of oxygen by fatty acid methyl esters (FAMEs) of linseed oil containing 0.5 mmol·L−1 of antioxidant. 2. Results 2.1. Bond Dissociation Enthalpies (BDE) of 70 Phenolic Antioxidants All of the (poly)phenols studied in this work are gathered by families in Table 1 and are classified from the lowest BDE to the highest BDE. The antioxidant power of 10 synthetic antioxidants, four tocopherols, eight hydroxybenzoic and eight hydroxycinnamic acids derivatives, 13 flavonols, two flavones, two flavanonols, four flavanones, three isoflavones, three catechins, two stilbenes, eugenol and isoeugenol, three phenols found in olive oil, one lignan, three coumarins, carnosic acid and carnosol are studied by DFT calculation. BDEs of all the O–H sites for each molecule have been calculated and results are described in supplementary materials (Table S1). Table 2 summarizes the calculated BDEs by the B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) method for the 70 phenols. Literature values are given in bracket and compared with our own values in supplementary materials (Figures S2–S4). Because of toxicity, some efficient antioxidants still used for polymers are no longer tolerated in food products. In recent years, the toxicity of BHT 7 and BHA 5 have extensively been studied and they are now very controversial [38]. Consequently, their ban in the near future is expected. One of the current alternatives is the natural (poly)phenols. BDEs of synthetic phenols are in the following order: 5-tert-butylpyrigallol 1 < pyrogallol 2 < hydroxyquinol 3 < propyl gallate 4 < BHA 5 < 4-tert-butylcatechol 6 < BHT 7 < TBHQ 8 < o-tert-butyl-p-cresol 9 < phloroglucinol 10. Tocopherols are monophenolic compounds derived from chromanol which are very soluble in oils making α-tocopherol 11 the most important antioxidant in edible fats and oils [39]. These phenols are frequently found in vegetable oils especially soybean, sunflower and palm oils. The four derivatives of tocopherol are distinguishable by the number and the position of the methyl substituents, which impact the BDEs. α-Tocopherol 11 has the lowest BDE compared to the β-, γ- and δ-tocopherols. Phenolic acids are another important class of antioxidants ubiquitous in food plants (i.e., fruit, vegetable, and coffee) [40]. There are simple phenolic acids based on hydroxybenzoic and hydroxycinnamic acids. BDEs of hydroxybenzoic acids are in the following order: Gallic acid 15 < protocatechuic acid 16 < syringic acid 17 < ellagic acid 18 < gentisic acid 19 < vanillic acid 20 < 4-hydroxybenzoic acid (PHBA) 21 < salicylic acid 22. Ellagic acid 18 is a particular combination of two molecules of Gallic acid and has a BDE of 78.4 kcal·mol−1. Furthermore, the hydroxycinnamic acid with the lowest BDE is rosmarinic acid 23. It is then followed by caffeic acid 24, chlorogenic acid 25, and sinapic acid 26. The others derivatives of hydroxycinnamic acids have highest BDEs (BDE > 80 kcal·mol−1). The class of flavonoids gathers more than 4000 different polyphenols found in leaves, stems, roots, fruits or seeds [41]. Their general chemical structure contains three rings A, B and C (Figure 1). The presence of carbonyls, double bonds and hydroxyl groups on the pyranyl ring C divides the flavonoids into different subclasses called flavonols, flavones, flavanonol, flavonones, isoflavone and flavanols. Substitution of A and B rings distinguishes the different phenolic antioxidants of each subclasses. The antioxidant activity of flavonoids depends on various factors [41]: (i) the metal-chelating potential that is strongly dependent on the arrangement of hydroxyls and carbonyl group around the molecule [42]; (ii) the presence of electron-donating substituents; and (iii) their ability to delocalize the unpaired electron leading to the formation of stable phenoxyl radical. Moreover, it has been shown that the phenolic ring B is the most active cycle [43]. Flavonols (i.e., gossypetin 31, myricetin 32, quercetin 34 and morin 41) have the 3-hydroxyflavone backbone which includes double bond and hydroxyl group on the pyranyl ring C. The flavonols with the lowest BDE are gossypetin 31 (66.6 kcal·mol−1), myricetin 32 (67.4 kcal·mol−1) and quercetin 34 (71.8 kcal·mol−1). BDEs of flavonols depend on the number of hydroxyl groups and their location on the structure of flavonols, which is discussed later. Flavones such as luteolin 44 and apigenin 45 are mainly found in cereals and herbs. They have the same chemical structure as flavonols without the hydroxyl group on the pyranyl ring C. Flavanonols (i.e., taxifolin 46 and aromadedrin 47) and flavanones (i.e., eriodictyol 48, homoeriodictyol 49, hesperetin 50 and naringenin 51) are other classes of flavonoids. They have the same chemical structure as flavonols but without the double bond on the pyranyl ring C and taxifolin 46 has a lower BDE than aromadedrin 47. Flavonones do not have double bond and hydroxyl group on the pyranyl site C. The flavanone with the lowest BDE is eriodictyol 48 (73.6 kcal·mol−1). Isoflavones (i.e., glycetin 52, genistein 53 and daidzein 54) are also studied. They are similar with flavones except that the B ring is bound to the C(3) position instead of the C(2). The three isoflavones studied have almost the same BDE. The OH group involved is located on the carbon C(4’). Finally, the last class of flavonoid studied is catechins, also called flavanols, which are abundant in tea (i.e., epigallocatechin gallate 55, gallocatechin 56 and catechin 57). The catechin with the lowest BDE is epigallocatechin gallate 55 (66.5 kcal·mol−1). The two investigated stilbenes (i.e., piceatannol 58 and resveratrol 59) are natural polyphenols present in many plants such as grapes. Piceatannol 58 differs from resveratrol 59 with an OH group at the C(3’) position which decreases the BDE. Eugenol 61 is a phenol found in clove essence oil whereas isoeugenol 60 is present in ylang-ylang essential oil. The position of the double bond influences the BDE leading to a higher value for eugenol 61. Hydroxytyrosol 62, catechol 63 and tyrosol 64 are antioxidants found in olive oil [44]. Hydroxytyrosol 62 is the phenol with the lowest BDE followed by catechol 63 and tyrosol 64 (81.0 kcal·mol−1). Sesamol 65 is a lignan found in sesame oil. It is a potent antioxidant and antiflammatory agent in various oxidative systems [45]. Lignans are phenyl propanoids derivated from phenylalanine and include also sesamin, sesamolin, sesaminol and sesamolinol [39]. The main coumarin called aesculetin 67 is found in tonka bean. Methyl and phenyl substituents can be grafted at the C(4) position but they have no impact on BDEs. Carnosol 69 and carnosic acid 70 are the two major components with rosmarinic acid 23 (already described) of rosemary extract (Rosmarinus officinalis L.) and are authorized in food in the form of extract [46]. For both compounds, same BDE (70.8 and 70.7 kcal·mol−1) was found. 2.2. Kinetic Rates of Hydrogen Transfer, Stoichiometric Numbersand Inhibition of FAMEs Linseed Oil Oxidation Thirty-two phenols (1, 4–9, 11, 15–17, 20–27, 32, 34, 55, 58–63, 65, 67, 69 and 70) have been selected to cover the different classes of antioxidants and confirm their antioxidant activity through the DPPH• (2,2-diphenyl-1-picrylhydrazyl) test and during the oxidation of FAMEs linseed oil (RapidOxy®). All these experimental methods have already been described in our previous works [27,29,47]. 2.2.1. Kinetic Rates of Hydrogen Transfer The DPPH• test [48,49,50,51,52,53] is commonly used to evaluate the antioxidant power of phenolic compounds. DPPH• is a stable radical with a maximum of absorption around 515–520 nm (purple). When antioxidants are mixed with this stable radical, there is a transfer of hydrogen from the antioxidant to the DPPH• radical which becomes yellow (Equation (7)) [47]. Thus, it is easy to follow the hydrogen transfer by UV-visible spectrometry. Toluene has been chosen as a solvent because it is inert towards radical reactions and cannot create hydrogen bonding. Indeed, the polarity of the solvent can drastically change the reactivity of polar antioxidants. The mechanism involved in this apolar aprotic solvent is called hydrogen atom transfer (HAT), which is the opposite of the sequential proton loss electron transfer (SPLET) mechanism that takes place in polar solvents [29]. (7) Kinetic rates (k) are obtained using either pseudo-first order kinetics (FOK) or second order kinetics (SOK) depending on the reactivity of the phenol under study [29,47]. Reactive phenols (1, 4, 5, 6, 8, 11, 17, 20, 26, 27, 60, 61, 62, 63, 65, 69 and 70) were mixed with DPPH• in stoichiometric amount leading to SOK (Equation (8)) [47]. Figure 2 shows an example of SOK reaction with hydroxytyrosol 62 in toluene. Kinetic rate constant is determined during the first 20 seconds of the reaction and only takes into account the reaction of phenolic hydrogen with DPPH•. (8) 1(A − Af) = 1(A0 − Af) − k(ε − ε′) t On the other hand, weakly reactive phenols (7, 9, 20 and 61) are introduced in excess at different concentrations with respect to DPPH•. Under these conditions, [ArOH]t ≈ [ArOH]0 and a pseudo-first order kinetics (FOK) describe the system (Equation (9)) [47]. Figure 3 shows the FOK reaction with eugenol 61 in toluene. (9) Ln (A −Af)(A0 − Af) = − k [ArOH]0 t To confirm that pseudo-first order kinetic (FOK) and second order kinetic (SOK) give the same results for the same phenolic antioxidants, vanillic acid 20 and eugenol 61 were studied using both conditions leading to similar rate constants k. Results are gathered in Table 3. 2.2.2. Stoichiometric Number (σexp) The second parameter highlighted by the DPPH• test is the number of hydrogen transferred from the phenol to the stable radical called the stoichiometric number (σexp). It is obtained via the final absorbance reached by DPPH• in the presence of a large excess of DPPH• with respect to the antioxidant concentration [54] (Equation (10)) [29,47]. (10) σexp=[DPPH•]0−[DPPH•]f[ArOH]0=A0−Af(ε−ε′)[ArOH]0 The mechanism of interaction between DPPH• radical and phenol takes place in two steps: (1) abstraction of the phenolic hydrogen; and (2) transfer of a second hydrogen or formation of dimers from the phenoxyl radical ArO•. All of the stoichiometric numbers (σexp) determined for the different phenols are summarized in Table 3. Toluene was replaced by ethyl acetate when antioxidants were not soluble. Figure 4 reports the result for catechol 63. 2.2.3. Inhibition of FAMEs Linseed Oil Oxidation FAMEs of linseed oil were synthesized by transesterification [55]. Gas chromatography–mass spectrometry (GC-MS) analysis shows that they were composed of 5.3% methyl palmitate (C16:0), 5.3% methyl stearate (C18:0), 33.1% methyl oleate (C18:1), 11.2% methyl linoleate (C18:2) and 45.1% methyl linolenate (C18:3, ω-3). The autoxidation of omega-3 oils in the presence of the different phenols has been kinetically investigated by measuring the oxygen consumption via RapidOxy® (Figure 5), which provides information on induction periods (IP) and oxidation rates (Rox) [47]. The efficiency of the antioxidants depends on their solubilization into the FAMEs. Indeed, a high solubility improves its protective effects of FAMEs against oxidation. As shown in Figure 6, the oxygen consumption during the oxidation process exhibits three steps: (i) the equilibration period corresponding to the increase of pressure following the increase of temperature after the achievement of the set pressure (450 kPa); (ii) the induction period defined by a slow decrease of the maximum pressure indicating that the antioxidant is effective; and (iii) the oxidation period characterized by a fast decrease of the oxygen consumption indicating the complete consumption of phenol which is no longer effective. The evolution of the pressure was then converted to a concentration of oxygen consumed in the liquid phase ∆[O2]t defined by Equation (11) [47], where Vtot and Vliq are the volumes of the cell and the FAMEs, respectively; Pmax is the maximum pressure obtained a few minutes after heating the cell; and Pt is the pressure measured at a given time. Oxidation rate (Rox) is defined as the rate when oxygen is consumed in the presence of antioxidants. It corresponds to the slope of the trend curve of [O2] consumed during the induction period. The two important parameters (i.e., induction period IP and oxidation rate Rox) are compiled in Table 3. (11) Δ[O2]t=(Pmax−Pt)RT×(Vtot−Vliq)Vliq According to the kinetic rates constants obtained with the DPPH• test, 5-tert-butyl-pyrogallol 1 (9480 M−1·s−1) and propyl gallate 4 (1240 M−1·s−1) are the most reactive synthetic phenols in toluene. With regard to natural phenolic antioxidants, α-tocopherol 11 exhibits the highest kinetic rate constants (2670 M−1·s−1) followed by carnosol 69 (1680 M−1·s−1), hydroxytyrosol 62 (1070 M−1·s−1) and carnosic acid 70 (640 M−1·s−1). Conversely, vanillic acid 20 (1.4 M−1·s−1), ferulic acid 27 (8.4 M−1·s−1) and eugenol 61 (3.9 M−1·s−1) are the least reactive phenols. With regard to stoichiometric numbers (σexp), three categories of antioxidants can be identified. First, there are the antioxidants capable of trapping more than three radical molecules (4, 15, 23, 32 and 55, σexp ≥ 3). Then, other phenols transfer 2 or 3 hydrogens to DPPH• radical (1, 5, 7, 8, 11, 16, 24, 25, 27, 34, 58, 61, 62, 63, 65, 67, 69 and 70, 2 ≤ σexp < 3). Finally, some compounds are not really active in the transfer of hydrogen since they trap less than two radicals per molecule of phenols (17, 26, 59 and 60, σexp < 2). The comparison between induction periods observed for all phenolic antioxidants reveals four categories of phenols. First of all, epigallocatechin gallate 55 is by far the most reactive phenol with an induction period of about 500 min. It is followed by piceatannol 58 (IP = 313 min). These two compounds belong to category A including “extremely effective” antioxidants. Then phenols of category B (1, 6, 23, 32 and 70) exhibit induction periods from 200 to 300 min and are considered “highly effective” antioxidants. Furthermore, phenols of category C (4, 5, 7, 11, 15, 24, 25, 34, 62, 63, 65, 67 and 69) have an IP between 200 and 100 min and are considered as “moderately effective” antioxidants. Finally, phenols 8, 9, 16, 17, 20, 21, 26, 27, 59, 60 and 61 of category D having an induction period lower than 100 min are “poorly effective” antioxidants. During the oxidation of FAMEs linseed oil, epigallocatechin gallate 55 and 5-tert-butyl-pyrogallol 1 show the lowest oxidation rate (0.08 and 0.06 mM−1·s−1 respectively). Myricetin 32 also provides low oxidation rates of 0.11 mM−1·s−1, whereas, conversely, vanillic acid 20 does not reduce the oxidation rate (1.03 mM−1·s−1). It is noteworthy that α-tocopherol 11, which is the phenol of reference, does not have the lowest oxidation rate (0.17 M−1·s−1). 3. Discussion 3.1. Bond Dissociation Enthalpies (BDE) of 70 Phenolic Antioxidants It is commonly admitted in the literature that BDEs of phenols are strongly influenced by the number, nature and position of the substituents linked to the phenol ring [11,14,20,22,23,29,47,56,57,58]. Nevertheless, as the results are dependent on the method of calculation used, it is difficult to compare literature values. As an example, the BDE of apigenin 45 was found to be 75.6 kcal·mol−1 by Pérez-Gonzalez et al. [20] and 82.2 kcal·mol−1 by Leopoldini et al. [22]. Furthermore, BDEs of 5-tert-butylpyrogallol 1, carnosol 69 and carnosic acid 70 have not been reported. Here, we use the B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) method to calculate the BDE of the 70 phenols investigated. It is noteworthy that our theoretical results are globally consistent with those obtained by Leopoldini et al. (R2 = 0.98) [22], Li et al. (R2 = 0.96) [23] and Pérez-Gonzalez et al. (R2 = 0.97) [20] (see the good correlations in Figures S2–S4 in supplementary materials). α-Tocopherol 11 exhibits a lower BDE than β-, γ- and δ-tocopherols. This low BDE (69.1 kcal·mol−1) results from different factors [27]: (1) the alkoxyl group in p-position; (2) the four alkyl substituents on the phenolic ring; (3) the molecular rigidity imposed by the pyran structure. Consequently, α-tocopherol 11 is expected to be the most powerful tocopherol. BDEs of hydroxybenzoic acid derivatives show that a substitution by two ortho-hydroxyl groups (Gallic acid, 15) allows a much stronger decrease of the BDE than a substitution by two ortho-methoxy groups (syringic acid, 17), which have BDEs of 70.2 and 78.1 kcal·mol−1, respectively. Moreover, this behavior is confirmed by comparing eriodictyol 48 (73.6 kcal·mol−1) and homoeriodictyol 49 (80.8 kcal·mol−1) (Figure 7). Nevertheless, ortho-carboxyl group (salicylic acid, 22) drastically increases the BDE compared to that of phenol itself (82.2 kcal·mol−1). Thanks to a better delocalization of the unpaired electron for the phenolic radical, hydroxycinnamic acids have lower BDE than hydroxybenzoic acids. As a consequence, caffeic acid 24 (72.1 kcal·mol−1) should have a better antioxidant power compared to protocatechuic acid 16 (75.5 kcal·mol−1). Based on this argument, isoeugenol 60 has a lower BDE than eugenol 61 of ≈4 kcal·mol−1 (Figure 8). BDE calculations of flavonoids highlight that the flavonol with the lowest BDE is gossypetin 31 (66.6 kcal·mol−1). As regards to its low BDE, it should be the most powerful flavonoid. The phenolic site involved is situated on the ring A as also demonstrated by Pérez-González et al. [20]. Except gossypetin 31, the O–H group (R(4’) position) on the B ring is always the most reactive site [22]. However, flavonoids without this hydroxyl group (R(4’) position) are exceptions to this rule. As examples, kaempferide (39) and galangin (43) have their most hydroxyl reactive site on the ring C (C(3) position). Flavones exhibit higher BDEs than flavonols. Indeed, luteolin 44 (73.1 kcal·mol−1) and apigenin 45 (82.1 kcal·mol−1) have higher BDE than quercetin 34 (71.8 kcal·mol−1) and kaempferol 42 (80.1 kcal·mol−1) respectively (Figure 9). This is due to the absence of OH group in the C ring. BDEs of O–H group for flavones in site R(4’) are about 10 kcal·mol−1 higher than for flavonols. Therefore, with equivalent substituents, flavones should be less reactive than flavonols through the HAT mechanism. BDEs of isoflavones studied are close to that of apigenin 45 (82.1 kcal·mol−1) suggesting that the location of the ring B does not alter the hydrogen transfer. Therefore, the antioxidant power of flavones and isoflavone should be similar when they have the same number of hydroxyl groups on the ring B. Flavanonols have higher BDEs than flavonols by comparing taxifolin 46 (73.2 kcal·mol−1) with quercetin 34 (71.8 kcal·mol−1) and kaempferol 42 (80.1 kcal·mol−1) with aromadedrin 47 (82.3 kcal·mol−1) (Figure 10). BDEs of O–H group for flavanonols in site R(4’) are about 2 kcal·mol−1 higher than for flavonols. Flavanones have higher BDEs than flavonols by comparing eriodictyol 48 (73.6 kcal·mol−1) and quercetin 34 (71.8 kcal·mol−1) (Figure 11). BDEs of the O-H group in site R(4’) is also about 2 kcal·mol−1 higher than for flavonols. That is a logical finding since the conjugation is broken due to the single bond. Therefore, the major effects are due to the neighboring groups. Substitution of the phenolic ring by two ortho-hydroxyl groups improves the stability of the central hydroxyl group and also that of the phenolic radical leading to a drastic decrease of BDEs compared to phenol (82.2 kcal·mol−1) [11,47] (Equation (12)). (12) It is the case for epigallocatechin gallate 55 (66.5 kcal·mol−1), 5-tert-butylpyrogallol 1 (66.6 kcal·mol−1), myricetin 32 (67.4 kcal·mol−1), pyrogallol 2 (68.0 kcal·mol−1), propyl gallate 4 (69.6 kcal·mol−1) and Gallic acid 15 (70.2 kcal·mol−1). Two ortho-hydroxyl groups (pyrogallol moieties) have a stronger impact on the decrease of the BDEs than only one ortho-hydroxyl function (catechol moieties). Indeed, pyrogallol structures (i.e., 5-tert-butylpyrogallol 1, pyrogallol 2, gallic acid 15, myricetin 32 and gallocatechin 55) have lower BDEs than their respective catechol compounds (i.e., 4-tert-butylcatechol 6, catechol 63, protocatechuic acid 16, quercetin 34 and catechin 57). Table 4 shows the comparison between these two types of antioxidants and highlights a systematic ΔBDE of ≈5 kcal·mol−1. Based on the BDEs of the studied phenols, a scale of predictive reactivity has been established from the lowest to the highest BDEs (Figure 12). It reveals four classes of antioxidants: (i) antioxidants with very low BDE from 65 to 70 kcal·mol−1; (ii) antioxidants with low BDE from 70 to 75 kcal·mol−1; (iii) antioxidants with medium BDE from 75 to 80 kcal·mol−1; and (iv) antioxidants with high BDE from 80 to 95 kcal·mol−1. The antioxidants with the lowest BDEs are expected to have the best antioxidant power. Very low BDEs (<70 kcal·mol−1) have been obtained for 5-tert-butylpyrogallol 1, myricetin 32, propyl gallate 4 and Gallic acid 15, which are pyrogallol derivatives. This class contains also α-tocopherol 11 with a BDE of 69.1 kcal·mol−1. Rosmarinic acid 23, carnosol 69 and carnosic acid 70 also exhibit a very low BDE. Indeed, they bear a catechol-type ring moiety, conjugated double bonds and alkyl substituents on the phenol rings which strongly contribute to lower the BDE. Then, catechol 63 itself and catechol-based derivatives with flavonol structure (i.e., quercetin 34), alkyl substituent (i.e., 4-tert-butyl-catechol 6, hydroxytyrosol 62) and conjugated double bonds (i.e., caffeic acid 24 and chlorogenic acid 25) have low BDEs. Moreover, monophenols with OCH3 groups (BHA 5), ortho-and para-alkyl substituents (BHT 7) and substituted hydroquinone (TBHQ 8) belong to this second class of antioxidants. Finally, there is also catechol-based derivative with electron-withdrawing group EWG (i.e., protocatechuic acid 16) and monophenol with dioxolane moiety (i.e., sesamol 65). The category of antioxidants with medium BDE includes monophenols with OCH3 groups (i.e., syringic acid 17, isoeugenol 60), conjugated double bond (resveratrol 59) and alkyl substituents (i.e., o-tert-butyl-p-cresol 9). Finally, vanillic acid 20, PHBA 21, ferulic acid 27, eugenol 61 and tyrosol 64 have high BDEs and are expected to be poorly reactive considering the HAT mechanism. Moreover, the simplest structure of hydroxycinnamic acid derivatives (o-, p- and m-coumaric acids 28, 29 and 30) and phloroglucinol 10 (phenol with two OH groups in meta position) have the highest BDEs. Vanillic acid 20, PHBA 21 and salicylic acid 22 have a higher BDE than phenol itself (kcal·mol−1) due to the effect of electron-withdrawing group (EWG). We can conclude that a powerful antioxidant must have pyrogallol (i.e., Gallic acid 15, myricetin 32, epigallocatechin gallate 55 and gallocatechin 56) or catechol moieties (i.e., rosmarinic acid 23, carnosol 69 and carnosic acid 70) conjugated with para-electron-donating substituents. They are the best natural alternatives to α-tocopherol 11 and synthetic phenolic antioxidants. 3.2. Kinetic Rates of Hydrogen Transfer, Stoichiometric Numbers and Inhibition of FAMEs Linseed Oil Oxidation The determination of the kinetic rates of hydrogen transfer for phenolic antioxidants is a way to experimentally confirm the antioxidant properties suggested by BDE calculations. The logarithm of the rate constants for the reaction of hydrogen transfer from phenol to the DPPH• radical is very well correlated with the calculated BDE of phenols (R2 = 0.96) confirming that the radical HAT mechanism occurs in toluene [29,47] (Figure 13). Nevertheless, due to the steric hindrance of the phenolic hydrogen, kinetic rates are sharply slowed down for BHA 5, BHT 7 and o-tert-butyl-p-cresol 9. Therefore, kinetic rates k are very low and do not follow the trend curve [27,47]. Log k decrease with increasing BDE has also been demonstrated by Foti et al. in heptane [34] and Marteau et al. in m-xylene [27]. Foti and co-workers have also proven that kinetics obtained with DPPH• are correlated with the reaction of phenols with peroxyl radicals ROO• [34,59]. Indeed, this test mimics the behavior of phenolic antioxidants during the inhibition of oils oxidation thanks to hydrogen transfer through a radical mechanism. Although the conditions and the method used to determine the kinetic rates of hydrogen transfer are different, our results obtained with the DPPH• test are consistent with the literature and with our scale of reactivity based on BDEs. These theoretical (BDE) and kinetic (DPPH• test) tools have allowed highlighting some promising effective antioxidants. Their potential antioxidant power has finally been evaluated against the oxidation of omega-3 oils derivatives (FAMEs). Figure 14 shows that the most efficient antioxidants are those with the lowest BDEs but no clear correlation between induction periods and BDEs could be obtained. There seems to be an exponential tendency (R2 = 0.86). Thereby, other parameters such as BDE influence the antioxidant power of phenols. The number of radicals trapped by molecule of antioxidant (σexp) has an important impact on the inhibition of oxidation and influences the induction period. These stoichiometric numbers were obtained with the DPPH• test and experiments point out that the most efficient antioxidants are those with the highest stoichiometric numbers (● σexp ≥ 3), whereas poorly effective phenols just trap fewer than two radicals per molecule (● σexp < 2). The trend displayed is clear: highly effective antioxidants are polyphenols characterized by high stoichiometric numbers. As an example, epigallocatechin gallate 55 traps more than five radicals per molecule of antioxidant and delays the oxidation process of about 500 minutes. Conversely, poorly effective antioxidants are those with low stoichiometric numbers as for syringic acid 17, isoeugenol 60 and sinapic acid 26. They transfer too few hydrogen atoms to be efficient on the delayed action of the oxidation process. Basically, moderately effective antioxidants (4, 5, 7, 11, 15, 24, 25, 34, 62, 63, 65, 67 and 69) trap two radicals per molecule. Nevertheless, there are some exceptions: Gallic acid 15 and propyl gallate 4 have higher stoichiometric numbers compared to moderately effective antioxidants. The transfer of all the hydrogen atoms from these phenols is probably too low to be competitive with the oxidation of FAMEs. Moreover, phenols with equal or higher stoichiometric numbers as moderately effective antioxidants (σexp ≥ 2) could be characterized as poorly effective antioxidants as attested by their low induction periods (i.e., o-tert-butyl-p-cresol 9, ferulic acid 27 and eugenol 61). Finally, piceatannol 58, which is also considered an extremely effective antioxidant, traps only two radicals per molecule of phenol. Accordingly, the number of radicals trapped by one molecule of phenol (σexp) is a crucial parameter for the protection of FAMEs against oxidation but the exceptions above-mentioned point out other essential factors as the BDE. Indeed, there is a very good correlation between oxidation rates (Rox) and BDEs (R2 = 0.97, Figure 15). We have previously described a correlation between BDEs and kinetic rates (DPPH• test) involving a transfer of hydrogen according to the radical HAT mechanism. Thereby, we assume that the mechanism involved during the inhibition of oxidation by phenolic antioxidants is also a radical mechanism. During the oxidation of FAMEs linseed oil, the lower the rate of oxygen consumption is, the more efficient the antioxidant is. Consequently, epigallocatechin gallate 55 and 5-tert-butyl-pyrogallol 1 are the most efficient antioxidants followed by myricetin 32. Contrary to the observation made with the DPPH• test, hindered phenols (i.e., BHA 5, BHT 7 and o-tert-butyl-p-cresol 9) are close to the correlation straight line. Their reactions with peroxyl radicals ROO•, which are less hindered than DPPH•, are not inhibited and they play their antioxidant role. Antioxidants with the lowest oxidation rate (Rox) are those with the highest stoichiometric numbers (● σexp ≥ 3) and conversely, poorly effective phenols are characterized by the highest Rox and lowest stoichiometric numbers (● σexp < 2). Consequently, the antioxidant power of phenols is determined by a combination of parameters: their BDE, the number of radicals trapped by one molecule of phenols and their ionization potential (IP) which has not been investigated here. As shown by Klein et al., the ionization potential of phenolic antioxidants has to be relatively high to be efficient in the protection of oxidized substrates [14]. Based on these characteristics, the four classes of antioxidants pointed out by the RapidOxy® experiments can be explained as follows: First of all, epigallocatechin gallate 55 is the most effective antioxidant due to its pyrogallol and galloyl moieties, which drastically decreases the BDE (66.5 kcal·mol−1) and increases the number of radicals traps by molecule (σexp = 5.4). Moreover, even if piceatannol 58 only traps two radicals, its low BDE allows decreasing the oxidation rate of FAMEs and strongly increasing the induction period. Then, phenols of category B (1, 6, 23, 32 and 70) are highly effective antioxidants. Pyrogallol structures (i.e., 5-tert-butyl-pyrogallol 1 and myricetin 32) and catechol moieties (i.e., 4-tert-butyl-pyrogallol 6, rosmarinic acid 23 and carnosic acid 70) possessing ortho-, para- or conjugated electron-donating groups (EDG) have low BDEs and stoichiometric number (σexp) between 2.0 and 3.0. Furthermore, phenols of category C (4, 5, 7, 11, 15, 24, 25, 34, 62, 63, 65, 67 and 69) are moderately effective antioxidants. Monophenols such as BHA 5, BHT 7, α-tocopherol 11 and sesamol 65 are able to transfer two hydrogens (σexp = 2). Moreover, all the catechol derivatives categorized as moderately effective antioxidants are just capable to transfer two hydrogens. Consequently, there is formation of ortho-quinone methide derivatives [47]. Gallic acid 15 and propyl gallate 4, identified as exception by their higher stoichiometric number (5 and 3.9, respectively), do not transfer their hydrogen enough quickly and the oxidation takes place at a rate of 0.32 and 0.26 mM·min−1. Therefore, even if these antioxidants could transfer more than two hydrogens, they are not highly efficient for the protection of omega-3 oils. Finally, phenols 8, 9, 10, 16, 17, 20, 21, 26, 27, 59, 60 and 61 belongs to category D and are considered as poorly effective antioxidants. Even if TBHQ 8 and protocatechuic acid 16 are catechol or hydroquinone derivatives capable to trap two radicals per molecule of phenol (σexp = 2), they are poorly reactive. Indeed, 16 reacts very slowly with the DPPH• radical and 8 could be subjected to thermal decomposition, volatilization or absorption by the food leading to a decrease of its antioxidant power [60]. o-tert-butyl-p-cresol 9 transfers more than two radicals (σexp = 2.5) per molecule but its high BDE (77.4 kcal·mol−1) is not in favor of an easy transfer of hydrogens. The other phenols included in this last category are monophenolic compounds with low kinetic rates of hydrogen transfer, high BDEs and stoichiometric number lower than 2.0. In conclusion, to estimate the efficiency of a phenolic antioxidant, it is necessary to combine both theoretical calculations of the BDEs and kinetic measurements (DPPH• test) of the rate constants and stoichiometric numbers. Through a systematic study based on 70 phenols, several efficient antioxidants better than α-tocopherol could be identified allowing a deeper understanding of the structure/activity relationships. The main rules that can be drawn are that antioxidants with low BDE, high kinetic rate of hydrogen transfer (k) and high number of radicals trapped by one molecule of phenols (σexp) are expected to be highly efficient providing that they act through the HAT mechanism. It appears that an efficient antioxidant should have pyrogallol or catechol moieties conjugated with para-electron-donating substituents. Apart from α-tocopherol 11, epigallocatechin gallate 55, piceatannol 58, myricetin 32, rosmarinic acid 23 and carnosic acid 70 are relevant alternatives to synthetic antioxidants such as propyl gallate 4, BHA 5 and BHT 7 for the preservation of omega-3 oils. 4. Materials and Methods 4.1. Reagents Catechol 63 (≥99%), 4-hydroxybenzoic acid 21 (PHBA, ≥99%), rosmarinic acid 23 (96%), quercetin 34 (≥98%), 2-tert-butyl-4-methylphenol 9 (99%), tert-butylhydroquinone 8 (TBHQ, 97%), 4-hydroxy-3-methoxybenzoic acid 20 (vanillic acid, 97%), sesamol 65 (98%), propyl gallate 4 (PG, ≥98%), isoeugenol 60 (98%), 3-tert-butyl-4-hydroxyanisol 5 (BHA, 98%), 3,4-dihydroxybenzoic acid 16 (protocatechuic acid, 97%), 2,6-di-tertbutyl-4-methylphenol 7 (BHT, ≥99%), 3,4-dihydroxycinnamic acid 24 (caffeic acid, 97%), ferulic acid 27 (99%), α-tocopherol 11 (≥96%), were from Sigma-Aldrich (Lyon, France). 3,5-dimethoxy-4-hydroxycinnamic acid 26 (sinapic acid, 98%), 5-tert-butylpyrogallol 1 (97%), syringic acid 17 (≥98%), eugenol 61 (99%), 6,7-dihydroxycoumarin 67 (aesculetin, ≥98%), were from Alfa Aesar (Karlsruhe, Germany). 3,4-dihydroxyphenyl ethanol 62 (hydroxytyrosol), myricetin 32 (≥98%), chlorogenic acid 25 (≥95%) and carnosic acid 70 (≥95%) were from Cayman Chemical Company (Ann Arbor, MI, USA). Gallic acid 15 (≥95%) was from Acros Organics (Geel, Belgium) and 4-tert-butylpyrogallol 6 (≥98%) was from Merck (France). Resveratrol 59 (≥98%) was from Tokyo Chemical Industry (TCI, Zwijndrecht, Belgium) and carnosol 69 was from Chromadex (Irvine, CA, USA). Solvents were of the purest grade commercially available from Sigma-Aldrich. The 2,2-diphenyl-1-picrylhydrazyl (DPPH•) radical was purchased from Sigma-Aldrich and kept at a temperature lower than 5 °C. Aluminum oxide, basic, Brockmann I, for chromatography, 50–200 μm, 60 Å was from Acros Organics (Geel, Belgium). Refined linseed oil was from Vandeputte Group, Mouscron, Belgium. FAME mix GLC-10 containing palmitic acid methyl ester (C16:0), stearic acid methyl ester (C18:0), oleic acid methyl ester (C18:1), linoleic acid methyl ester (C18:2) and linolenic acid methyl ester (C18:3) was from Supelco (Saint-Quentin Fallavier, France). 4.2. Calculation of the Bond Dissociation Enthalpies BDEs (O–H) The bond dissociation enthalpy or BDE is given by the difference between the enthalpy of the phenoxyl radical (plus that of the hydrogen atom) and that of the starting phenol as described by Equations (13) and (14). (13) ArO−H + X•→ArO• + X−H (14) BDE(ArO−H) = Hf0(ArO•) + Hf0(H•) − Hf0(ArO−H) The geometries of all the parent molecules were firstly optimized using the PM3 method and then the DFT one by using the B3LYP/6-311G(d,p) basis set. The first method was used to speed up the convergence of the optimization by the second one. The zero-point energy (ZPE) is taken into account to correct the BDE values. Geometries from this method were used as inputs to the final energy B3LYP/6-311G++(2d,2p) calculation. For species having several conformers, all of them were investigated. The conformer with the lowest electronic energy is retained. For radicals, the optimization also used the PM3 step plus the final UB3LYP/6-311G(d,p) method. The zero-point energy (ZPE) is also taken into account to correct the BDE values. Geometries were then used as inputs to the final UB3LYP/6-311G++(2d,2p) calculation. Calculations were performed in toluene. The method is described as B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p). 4.3. Determination of the Rate Constants for Hydrogen Transfer from Phenols to DPPH• Reactions of phenols with DPPH• are operating in toluene. Solutions of DPPH• were prepared in toluene at a concentration of approximately 5 × 10−3 mol·L−1. For phenols 7, 9, 16, 21 and 61, solutions were prepared in toluene at a concentration varying from 6 × 10−2 to 2 × 10−1 mol·L−1. Typically, 200–500 μL of the phenol solutions were added to 500 μL of DPPH• solution in a 50 mL glass reactor equipped with a UV fiber (from Varian equipped with a dip-probe; Varian, les Ulis, France) containing 20 mL of deoxygenated solvent maintained at 20 °C. The hydrogen transfer reaction from phenol to the DPPH• radical was accompanied by a change in the UV-visible spectrum and was monitored at 515 nm with a Varian spectrophotometer (Cary 50, 10 pts·s−1). The loss of DPPH• absorbance in the presence of an excess of phenol follows pseudo-first-order kinetics (FOK). The rate constants were determined for poorly reactive phenols 7, 9, 16, 21 and 61 for at least four different phenol concentrations by plotting kDPPH• versus [phenol]. In the case of other highly/moderately reactive phenols, the reaction with the DPPH• radical is very fast and the rate constant were determined by using stoichiometric conditions considering second order kinetics (SOK). For these phenols, solutions were prepared in toluene at a concentration of approximately 5 × 10−3 mol·L−1. Equipment for UV-visible analysis (Agilent, Les Ulis, France) and curve presenting the visualization of the lag time are presented in Figure S1. Values of the rate constants are given in the supplementary materials (Table S2). Under these conditions, ε and ε’ values are 11,788 L·mol−1·cm−1 and 24 L·mol−1·cm−1 for DPPH• and DPPH-H respectively [29]. 4.4. Determination of the Stoichiometric Number (σexp) for the Reaction of Phenolic Antioxidants with DPPH• Solutions of DPPH• are prepared in toluene at a concentration of ca. 1.5 × 10−4 M by sonicating the mixture until all DPPH• crystals were dissolved. The solutions are then maintained under argon at 20 °C. For phenols, solutions are also prepared in toluene at a concentration of 2.07 × 10−3 M by sonicating until all crystals are dissolved. Typically, 20 μL of the phenol solutions are added to 2.0 mL of a DPPH• solution in a UV cell stirred and maintained at 20 °C. The absorbance change is monitored at 515 nm by using the UV-Visible Cary 60 (Agilent, Les Ulis, France) every seconds or minutes. Final (Af) and initial (A0) absorbances are used to determine the stoichiometric number (σexp) according to Equation (15). Final absorbances are collected when constant values are reached during at least thirty minutes. Values of the stoichiometric numbers are summarized in Table 3 and detailed in the Supplementary Information (Table S3) [30]. (15) σexp=[DPPH•]0−[DPPH•]f[ArOH]0=A0−Af(ε−ε′)[ArOH]0 4.5. Synthesis of Antioxidant-Free Fatty Acid Methyl Esters (FAMEs) by Transesterification of Purified Linseed Oil Linseed oil was beforehand purified 3 times by alumina column chromatography to reach very low concentration of antioxidants naturally present in neat oil. The transesterification reaction of triglycerides of purified linseed oil with methanol into fatty acid methyl esters (FAMEs) is given in Equation (16). One liter of methanol was introduced into a 2 L three-necked equipped with a condenser and a gas bubbling. Sodium (10 g, 2 equiv.) was introduced piece by piece under argon followed by purified linseed oil (200 g, 1 equiv.). The reaction was performed during 12 h under magnetic stirrer. FAMEs were extracted with 3 × 300 mL of petroleum ether. The combined organic phases were evaporated under pressure. Isolated FAMEs were stored at −20 °C. (16) 4.6. Analysis of FAMEs Linseed Oil by GC-MS A Thermofisher (Courtaboeuf, France) GC Trace equipped with an AI 3000 injector connected to DSQ II simple quadrupole detector was used for the GC-MS analysis of FAMEs. Compound separation was achieved on a 30 m, DB5MS with 0.25 mm i.d. and 0.25 μm film thickness gas chromatographic column (J & W Scientific, Folsom, CA, USA). Carrier gas (ultra-pure helium) flow rate is 1.0 mL/min and the injector, the transfer line and the ions source were maintained at 250, 270 and 220 °C, respectively. The mass spectrometry (MS) detector was used in the electron ionization (EI) mode with an ionization voltage of 70 eV. The column was held at 130 °C for 0.5 min and then programmed at 0.3 °C·min−1 to 180 °C and maintained for 5 min. Then, the column was programmed at 3 °C·min−1 to 250 °C and maintained for 10 min. The compounds were injected in the Split mode with a ratio of 20. FAME mix GLC-10 (Sigma Aldrich, Lyon, France) was used to analyze and quantify the FAMEs composition. 4.7. Effect of the Phenolic Antioxidants on the Autoxidation of Fames Linseed Oil Two-milliliter FAMEs of linseed oil were introduced into the RapidOxy cell (25 mL) at room temperature. One hundred microliters of phenol (1, 4, 5, 6, 7, 8, 9, 11, 15, 16, 17, 20, 21, 22, 23, 24, 25, 26, 27, 32, 34, 55, 58, 59, 60, 61, 62, 63, 65, 67, 69 and 70) solutions (10−2 mol·L−1) were then added to reach a final antioxidant concentration of 5 × 10−4 mol·L−1. Antioxidants are solubilized in ethyl acetate and an ultrasound bath is used to homogenize solutions. Few amount of ethanol could be firstly used to pre-solubilize antioxidants not soluble in ethyl acetate. The cell was the closed and heated up to the temperature set (90 °C) under a pure oxygen pressure of 450 kPa. The O2 consumption was followed by monitoring the O2 pressure. The experiment was ended when the pressure reached 50% of the maximum pressure. The pressure decrease was converted into a concentration of oxygen per volumes of the FAMEs solution. Values of the induction periods and oxidation rates are summarized in Table 3 and detailed in the Supplementary Information (Table S4). Acknowledgments Chevreul Institute (FR 2638), Ministère de l’Enseignement Supérieur et de la Recherche, Région Nord–Pas de Calais, FEDER and the company Cargill are acknowledged for supporting and funding partially this work. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1220/s1. Click here for additional data file. Author Contributions The three authors equally contributed to this research article. This work is a part of the PhD thesis of Romain Guitard who was supervised by Véronique Nardello-Rataj and Jean-Marie Aubry. Conflicts of Interest The authors declare no conflict of interest. Abbreviations BHT Butylated hydroxytoluene BDE Bond dissociation enthalpy BHA Butylated hydroxyanisole DFT Density functional theory TBHQ tert-Butylhydroquinone EDG Electron-donating group PG Propyl gallate EWG Electron-withdrawing group FOK Pseudo-first-order kinetic LOO• Lipid peroxyl radical SOK Second order kinetic LOOH Lipid hydroperoxide DPPH 2,2-Diphenyl-1-picrylhydrazyl HAT Hydrogen atom transfer FAMEs Fatty acid methyl esters IP Induction period LH Unsaturated lipids Rox Oxidation rate Figure 1 General chemical structure of flavonoids. Figure 2 Evolution of the absorbance of DPPH• radical at 515 nm (1.25 × 10−4 mol·L−1) in the presence of hydroxytyrosol 62 (1.25 × 10−4 mol·L−1) in toluene at 20 °C. Linearization curve according to second order kinetics (SOK) (Equation (8)). Figure 3 (a) Change in absorbance at 515 nm of a solution of DPPH• (1.25 × 10−4 M) in the presence of an excess of eugenol 61 (2.25 × 10−3 M) in toluene, linearization of the logarithm of the absorbance using the final pseudo-first order kinetics (FOK) (Equation (9)) as a function time; and (b) regression constants of apparent kinetic rates as a function of initial concentrations of eugenol 61. Figure 4 Evolution of the absorbance of DPPH• radical (1.5 × 10−4 mol·L−1) at 515 nm in the presence of catechol 63 (2.07 × 10−5 mol·L−1) in toluene at 20 °C. Figure 5 RapidOxy® apparatus for measurement of oxygen consumption during the autoxidation process. Figure 6 Evolution of the oxygen pressure (left axis) and concentration of oxygen consumed (right axis) during the oxidation of fatty acid methyl esters (FAMEs) in the presence of rosmarinic acid (5 × 10−4 mol·L−1) at 90 °C Figure 7 Comparison between gallic acid (15) and syringic acid (17) and between eriodictyol (48) with homoeriodictyol (49). Figure 8 Comparison between caffeic acid (24) and protocatechuic acid (16) and between isoeugenol (60) and eugenol (61). Figure 9 Comparison between flavones (44 and 45) and flavonols (34 and 42). Figure 10 Comparison between flavanonols (46 and 47) and flavonols (34 and 42). Figure 11 Comparison between flavanones (48) and flavonols (34). Figure 12 Scale of expected effectiveness of all the phenolic antioxidants studied (1–70) according to their BDE calculated with B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) DFT method in vacuum. Figure 13 Logarithm of the rate constants (log k) for the reaction of phenolic antioxidants with DPPH• (ο hindered phenols and ● other phenols) in toluene as a function of their BDEs calculated with the B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) density functional theory (DFT) method. Figure 14 Induction periods (IP) as a function of the BDEs calculated with the B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) DFT method, stoichiometric numbers are indicated by: ● σexp ≥ 3, ● 3 < σexp ≤ 2 and ● σexp < 2 and phenols are categorized as: extremely effective (A); highly effective (B); moderately effective (C); and poorly effective (D). Figure 15 Oxidation rates as a function of the BDEs calculated with the B3LYP/6-311++G(2d,2p)//B3LYP/6-311G(d,p) DFT method, stoichiometric numbers are indicated by: ● σexp ≥ 3, ● 3 < σexp ≤ 2 and ● σexp < 2. ijms-17-01220-t001_Table 1Table 1 Names and numbers of the polyphenols studied in this work. N° Name N° Name N° Name Synthetic phenols Flavonols Catechins 1 5-Tert-butylpyrogallol 31 Gossypetin 55 Epigallocatechin gallate 2 Pyrogallol 32 Myricetin 56 Gallocatechin 3 Hydroxyquinol 33 Azaleatin 57 Catechin 4 Propyl gallate 34 Quercetin 5 BHA 35 Fisetin Stilbenes 6 4-Tert-butylcatechol 36 Laricitrin 58 Piceatannol 7 BHT 37 Syringetin 59 Resveratrol 8 TBHQ 38 Rhamnazin 9 o-Tert-butyl-p-cresol 39 Kaempferide Aromatic phenols 10 Phloroglucinol 40 Isorhamnetin 60 Isoeugenol 41 Morin 61 Eugenol Tocopherols 42 Kaempferol 11 α-Tocopherol 43 Galagin Phenols from olive oil 12 β-Tocopherol 62 Hydroxytyrosol 13 γ-Tocopherol Flavones 63 Catechol 14 δ-Tocopherol 44 Luteolin 64 Tyrosol 45 Apigenin Hydroxybenzoic acids Lignans 15 Gallic acid Flavanonols 65 Sesamol 16 Protocatechuic acid 46 Taxifolin 17 Syringic acid 47 Aromadedrin Coumarins 18 Ellagic acid 66 Methylesculetin 19 Gentisic acid Flavanones 67 Aesculetin 20 Vanillic acid 48 Eriodictyol 68 Nordalbergin 21 PHBA 49 Homoeriodictyol 22 Salicylic acid 50 Hesperetin Carnosic acid derivatives 51 Naringenin 69 Carnosol Hydroxycinnamic acids 70 Carnosic acid 23 Rosmarinic acid Isoflavones 24 Caffeic acid 52 Glycitein 25 Chlorogenic acid 53 Genistein 26 Sinapic acid 54 Daidzein 27 Ferulic acid 28 o-Coumaric acid 29 p-Coumaric acid 30 m-Coumaric acid ijms-17-01220-t002_Table 2Table 2 Bond dissociation enthalpies (BDEs) for the phenolic antioxidants studied (1–70). Synthetic phenolic antioxidants N° R(2) R(3) R(4) R(5) R(6) BDE (kcal·mol−1) 1 OH H C(CH3)3 H OH 66.6 (nd) 2 OH H H H OH 68.0 (77.7 [24]) 3 H H OH H OH 69.1 (70.4 [25]) 4 OH H C(O)OC3H7 H OH 69.6 (77.1 [23]) 5 C(CH3)3 H OCH3 H H 72.3 (80.7 [23]) 6 H H C(CH3)3 H OH 72.3 (81.1 [26]) 7 C(CH3)3 H CH3 H C(CH3)3 72.4 (79.9 [27]) 8 H H OH H C(CH3)3 74.3 (76.9 [28]) 9 H H CH3 H C(CH3)3 77.4 (78.1 [29]) 10 H OH H OH H 83.0 (87.7 [24]) Tocopherols N° R(2) R(3) R(5) BDE (kcal·mol−1) 11 CH3 CH3 CH3 69.1 (71.7 [30]) 12 CH3 H CH3 73.4 (77.7 [11]) 13 CH3 CH3 H 73.5 (78.2 [11]) 14 CH3 H H 75.4 (79.8 [11]) Derivatives of hydroxybenzoic acids N° R(2) R(3) R(4) R(5) R(2’) BDE (kcal·mol−1) 15 H OH OH OH H 70.2 (72.2 [22]) 16 H OH OH H H 75.5 (79.6 [24]) 17 H OCH3 OH OCH3 H 78.1 (82.7 [31]) 18 H OH OH OC(O)- -C6H(OH)2 78.4 (77.1 [32]) 19 OH H H OH H 79.5 (80.0 [32]) 20 H OCH3 OH H H 83.1 (87.0 [31]) 21 H H OH H H 84.7 (89.2 [24]) 22 OH H H H H 95.2 (93.0 [24]) Derivatives of hydroxycinnamic acids N° R(2) R(3) R(4) R(5) R(4’) BDE (kcal·mol−1) 23 H OH OH H C9O4H10 69.2 (75.3 [31]) 24 H OH OH H H 72.1 (73.6 [22]) 25 H OH OH H C6H2(OH)3CO2H 73.4 (78.7 [33]) 26 H OCH3 OH OCH3 H 75.4 (81.2 [34]) 27 H OCH3 OH H H 79.7 (84.5 [24]) 28 OH H H H H 80.1 (84.4 [24]) 29 H H OH H H 80.5 (84.9 [24]) 30 H OH H H H 84.4 (88.1 [35]) Flavonols N° R(2’) R(3’) R(4’) R(5’) R(5) R(7) R(8) BDE (kcal·mol−1) 31 H OH OH H OH OH OH 66.6 (65.5 [20]) 32 H OH OH OH OH OH H 67.4 (71.1 [36]) 33 H OH OH H OCH3 OH H 71.1 (66.1 [20]) 34 H OH OH H OH OH H 71.8 (72.3 [22]) 35 H H OH OH H OH H 72.3 (70.3 [36]) 36 H OCH3 OH OH OH OH H 72.5 (66.9 [20]) 37 H OCH3 OH OCH3 OH OH H 75.7 (63.8 [20]) 38 H OCH3 OH H OH OCH3 H 79.6 (65.2 [20]) 39 H H OCH3 H OH OH H 79.8 (73.8 [20]) 40 H OCH3 OH H OH OH H 79.8 (72.9 [20]) 41 OH H OH H OH OH H 79.8 (76.9 [36]) 42 H H OH H OH OH H 80.1 (80.9 [22]) 43 H H H H OH OH H 81.2 (76.0 [36]) Flavones N° R(3’) R(4’) R(5) R(7) BDE (kcal·mol−1) 44 OH OH OH OH 73.1 (74.5 [22]) 45 H OH OH OH 82.1 (82.9 [22]) Flavanonols N° R(3’) R(4’) R(5) R(7) BDE (kcal·mol−1) 46 OH OH OH OH 73.2 (74.7 [22]) 47 H OH OH OH 82.3 (75.7 [20]) Flavanones N° R(3’) R(4’) R(5) R(7) BDE (kcal·mol−1) 48 OH OH OH OH 73.6 (73.6 [36]) 49 OCH3 OH OH OH 80.8 (75.1 [20]) 50 OH OCH3 OH OH 82.2 (77.4 [36]) 51 H OH OH OH 82.4 (81.4 [36]) Isoflavones N° R(4’) R(5) R(6) R(7) BDE (kcal·mol−1) 52 OH H OCH3 OH 80.1 (78.0 [36]) 53 OH OH H OH 81.0 (78.0 [36]) 54 OH H H OH 81.9 (78.3 [36]) Catechins N° R(3’) R(4’) R(5’) R(3) R(5) R(7) BDE (kcal·mol−1) 55 OH OH OH C(O)C6H2(OH)3 OH OH 66.5 (69.0 [36]) 56 OH OH OH H OH OH 68.5 (63.7 [20]) 57 OH OH H H OH OH 74.4 (74.2 [22]) Stilbenes N° R(3‘) R(4’) R(3) R(5) BDE (kcal·mol−1) 58 OH OH OH OH 68.7 (62.9 [20]) 59 H OH OH OH 76.7 (70.3 [20]) Eugenol and Isoeugenol N° C(1)-C(2) C(2)-C(3) BDE (kcal·mol−1) 60 -CH=CH- -CH-CH3 76.6 (83.8 [27]) 61 -CH2-CH- -CH=CH2 80.2 (86.8 [27]) Antioxidants in olive oil N° R(2) R(4) BDE (kcal·mol−1) 62 OH CH2CH2OH 72.1 (73.5 [22]) 63 OH H 73.4 (76.4 [24]) 64 H CH2CH2OH 81.0 (87.8 [23]) Lignans N° R(4) BDE (kcal·mol−1) 65 OH 75.1 (80.6 [34]) Coumarins N° R(4) R(6) R(7) BDE (kcal·mol−1) 66 CH3 OH OH 72.0 (72.1 [37]) 67 H OH OH 72.5 (73.1 [37]) 68 C6H5 OH OH 72.6 (nd) Carnosol and carnosic acid N° R(1) R(2) BDE (kcal·mol−1) 69 / -C(O)O- 70.7 (nd) 70 -CO2H / 70.8 (nd) nd: not determined. ijms-17-01220-t003_Table 3Table 3 Rate constants (k) of hydrogen transfer from ArOH to DPPH• in toluene at 20 °C, stoichiometric numbers of H atoms (σexp) determined with an excess of DPPH• in toluene at 20 °C, induction periods (IP) and oxidation rates (Rox) are determined by the RapidOxy® experiments N° k (M−1·s−1) d Induction Period d Oxidation Rate d Stoichiometric Number SOK a FOK b IP (min) Rox (mM·min−1) σexp 0 e / / 0 1.23 / 1 9480 234 0.06 2.1 4 1240 162 0.26 3.9 5 184 167 0.35 2.0 6 776 220 0.37 2.5 7 0.18 131 0.44 2.0 8 600 45 0.53 2.0 9 0.36 56 0.77 2.5 11 2690 177 0.17 2.0 15 ns 178 0.32 5.0 c 16 ns 50 0.62 1.9 c 17 10.6 37 0.76 1.1 20 1.4 1.0 5 1.03 0 c 21 ns 6 1.20 0 c 23 ns 262 0.27 4.1 c 24 ns 148 0.36 2.0 c 25 ns 138 0.48 1.9 c 26 165 54 0.57 1.4 c 27 8.4 28 0.82 1.8 32 ns 262 0.11 3.4 c 34 ns 135 0.34 1.9 c 55 ns 476 0.08 5.4 c 58 ns 313 0.29 2.0 c 59 ns 67 0.68 0.9 c 60 38 49 0.72 0.9 61 3.9 2.7 27 0.93 2.1 62 1070 172 0.30 2.0 63 400 147 0.46 1.9 65 250 161 0.55 2.1 67 ns 112 0.50 2.1 69 1680 166 0.35 1.9 70 640 230 0.29 2.0 a, SOK: Second Order Kinetics ([DPPH•]0 = [ArOH]0); b, FOK: pseudo-First Order Kinetics ([ArOH]0 >> [DPPH•]0); c, ethyl acetate is used as a solvent; d, average on three values, ns: not soluble; e, blank with no antioxidant. ijms-17-01220-t004_Table 4Table 4 BDEs comparison between pyrogallol (15, 1, 2, 32 and 55) and catechol (16, 6, 63, 34 and 57) moieties. Pyrogallol Moieties Catechol Moieties ΔBDE (kcal·mol−1) N° BDE (kcal·mol−1) N° BDE (kcal·mol−1) 1 66.6 6 72.3 5.7 2 68.0 63 73.4 5.4 15 70.2 16 75.5 5.3 32 67.4 34 71.8 4.4 55 68.5 57 74.4 5.9 ΔBDE = BDE (catechol) − BDE (pyrogallol). ==== Refs References 1. Burr M.L. Lessons from the story of n-3 fatty acids Am. J. Clin. Nutr. 2000 71 397S 398S 10618004 2. Kryzhanovskii S.A. Vititnova M.B. Ω-3 polyunsaturated fatty acids and the cardiovascular system Hum. Physiol. 2009 35 491 501 10.1134/S036211970904015X 3. Kanner J. Rosenthal I. An assessment of lipid oxidation in foods Pure Appl. Chem. 1992 64 1959 1964 10.1351/pac199264121959 4. Miyamoto S. Martinez G.R. Medeiros M.H.G. di Mascio P. Singlet molecular oxygen generated from lipid hydroperoxides by the russell mechanism: Studies using 18(O)-labeled linoleic acid hydroperoxide and monomol light emission measurements J. Am. Chem. Soc. 2003 125 6172 6179 10.1021/ja029115o 12785849 5. Neff W.E. Frankel E.N. Photosensitized oxidation of methyl linolenate monohydroperoxides: Hydroperoxy cyclic peroxides, dihydroperoxides and hydroperoxy bis(cyclic peroxide)s Lipids 1984 19 952 957 10.1007/BF02534731 6. Halliwell B. Aeschbach R. Löliger J. Aruoma O.I. The characterization of antioxidants Food Chem. Toxicol. 1995 33 601 617 10.1016/0278-6915(95)00024-V 7628797 7. Zhang H.-Y. Sun Y.-M. Wang X.-L. Substituent effects on o-h bond dissociation enthalpies and ionization potentials of catechols: A dft study and its implications in the rational design of phenolic antioxidants and elucidation of structure-activity relationships for flavonoid antioxidants Chem. Eur. J. 2003 9 502 508 10.1002/chem.200390052 12532299 8. Van Acker S.A.B.E. Koymans L.M.H. Bast A. Molecular pharmacology of vitamin e: Structural aspects of antioxidant activity Free Radical Biol. Med. 1993 15 311 328 10.1016/0891-5849(93)90078-9 8406131 9. Zhang H.-Y. Theoretical methods used in elucidating activity differences of phenolic antioxidants J. Am. Oil Chem. Soc. 1999 76 745 748 10.1007/s11746-999-0170-8 10. Wright J.S. Carpenter D.J. McKay D.J. Ingold K.U. Theoretical calculation of substituent effects on the o-h bond strength of phenolic antioxidants related to vitamin e J. Am. Chem. Soc. 1997 119 4245 4252 10.1021/ja963378z 11. Wright J.S. Johnson E.R. DiLabio G.A. Predicting the activity of phenolic antioxidants:  Theoretical method, analysis of substituent effects, and application to major families of antioxidants J. Am. Chem. Soc. 2001 123 1173 1183 10.1021/ja002455u 11456671 12. Zhang H.-Y. Selection of theoretical parameter characterizing scavenging activity of antioxidants on free radicals J. Am. Oil Chem. Soc. 1998 75 1705 1709 10.1007/s11746-998-0320-4 13. Zhang H.-Y. On the o-h bond dissociation enthalpy of catechol New J. Chem. 2003 27 453 454 10.1039/b212161c 14. Klein E. Lukes V. Study of gas-phase O–H bond dissociation enthalpies and ionization potentials of substituted phenols-applicability of ab initio and DFT/B3LYP methods Chem. Phys. 2006 330 515 525 10.1016/j.chemphys.2006.09.026 15. Klein E. Lukes V. Cibulkova Z. Polovkova J. Study of N–H, O–H, and S–H bond dissociation enthalpies and ionization potentials of substituted anilines, phenols, and thiophenols J. Mol. Struct. Theochem 2006 758 149 159 10.1016/j.theochem.2005.10.015 16. Pratt D.A. DiLabio G.A. Brigati G. Pedulli G.F. Valgimigli L. 5-Pyrimidinols: Novel chain-breaking antioxidants more effective than phenols J. Am. Chem. Soc. 2001 123 4625 4626 10.1021/ja005679l 11457259 17. Bowry V.W. Ingold K.U. The unexpected role of vitamin E (α-tocopherol) in the peroxidation of human low-density lipoprotein Acc. Chem. Res. 1999 32 27 34 10.1021/ar950059o 18. Gotoh N. Noguchi N. Tsuchiya J. Morita K. Sakai H. Shimasaki H. Niki E. Inhibition of oxidation of low density lipoprotein by vitamin e and related compounds Free Radic. Res. 1996 24 123 134 8845913 19. Noguchi N. Okimoto Y. Tsuchiya J. Cynshi O. Kodama T. Niki E. Inhibition of oxidation of low-density lipoprotein by a novel antioxidant, BO-653, prepared by theoretical design Arch. Biochem. Biophys. 1997 347 141 147 10.1006/abbi.1997.0331 9344475 20. Perez-Gonzalez A. Rebollar-Zepeda A.M. Leon-Carmona J.R. Galano A. Reactivity indexes and O–H bond dissociation energies of a large series of polyphenols: Implications for their free radical scavenging activity J. Mex. Chem. Soc. 2012 56 241 249 21. Bakalbassis E.G. Lithoxoidou A.T. Vafiadis A.P. Theoretical calculation of accurate absolute and relative gas- and liquid-phase O–H bond dissociation enthalpies of 2-mono- and 2,6-disubstituted phenols, using dft/b3lyp J. Phys. Chem. A 2003 107 8594 8606 10.1021/jp034400v 22. Leopoldini M. Russo N. Toscano M. The molecular basis of working mechanism of natural polyphenolic antioxidants Food Chem. 2010 125 288 306 10.1016/j.foodchem.2010.08.012 23. Li M.-J. Liu L. Fu Y. Guo Q.-X. Accurate bond dissociation enthalpies of popular antioxidants predicted by the oniom-g3b3 method J. Mol. Struct. Theochem 2007 815 1 9 10.1016/j.theochem.2007.03.012 24. Hoelz L.V.B. Horta B.A.C. Araujo J.Q. Albuquerque M.G. Bicca de Alencastro R. da Silva J.F.M. Quantitative structure-activity relationships of antioxidant phenolic compounds J. Chem. Pharm. Res. 2010 2 291 306 25. Thavasi V. Leong L.P. Bettens R.P.A. Investigation of the influence of hydroxy groups on the radical scavenging ability of polyphenols J. Phys. Chem. A 2006 110 4918 4923 10.1021/jp057315r 16599462 26. Amorati R. Ferroni F. Lucarini M. Pedulli G.F. Valgimigli L. A quantitative approach to the recycling of α-tocopherol by coantioxidants J. Org. Chem. 2002 67 9295 9303 10.1021/jo026501f 12492331 27. Marteau C. Nardello-Rataj V. Favier D. Aubry J.M. Dual role of phenols as fragrances and antioxidants: Mechanism, kinetics and drastic solvent effect Flavour Frag. J. 2013 28 30 38 10.1002/ffj.3123 28. Nehru K. Jang Y. Oh S. Dallemer F. Nam W. Kim J. Oxidation of hydroquinones by a nonheme iron(iv)-oxo species Inorg. Chim. Acta 2008 361 2557 2561 10.1016/j.ica.2007.12.027 29. Marteau C. Guitard R. Penverne C. Favier D. Nardello-Rataj V. Aubry J.M. Boosting effect of ortho-propenyl substituent on the antioxidant activity of natural phenols Food Chem. 2016 196 418 427 10.1016/j.foodchem.2015.09.007 26593510 30. Leopoldini M. Marino T. Russo N. Toscano M. Antioxidant properties of phenolic compounds: H-atom versus electron transfer mechanism J. Phys. Chem. A 2004 108 4916 4922 10.1021/jp037247d 31. Nenadis N. Tsimidou M.Z. Contribution of dft computed molecular descriptors in the study of radical scavenging activity trend of natural hydroxybenzaldehydes and corresponding acids Food Res. Int. 2012 48 538 543 10.1016/j.foodres.2012.05.014 32. Chen Y. Xiao H. Zheng J. Liang G. Structure-thermodynamics-antioxidant activity relationships of selected natural phenolic acids and derivatives: An experimental and theoretical evaluation PLoS ONE 2015 10 1220 10.1371/journal.pone.0121276 25803685 33. Amorati R. Pedulli G.F. Cabrini L. Zambonin L. Landi L. Solvent and ph effects on the antioxidant activity of caffeic and other phenolic acids J. Agric. Food Chem. 2006 54 2932 2937 10.1021/jf053159+ 16608211 34. Foti M.C. Daquino C. Mackie I.D. DiLabio G.A. Ingold K.U. Reaction of phenols with the 2,2-diphenyl-1-picrylhydrazyl radical: Kinetics and dft calculations applied to determine aro-h bond dissociation enthalpies and reaction mechanism J. Org. Chem. 2008 73 9270 9282 10.1021/jo8016555 18991378 35. Pino E. Campos A.M. Lopez-Alarcon C. Aspee A. Lissi E. Free radical scavenging capacity of hydroxycinnamic acids and related compounds J. Phys. Org. Chem. 2006 19 759 764 10.1002/poc.1071 36. Amic D. Lucic B. Reliability of bond dissociation enthalpy calculated by the pm6 method and experimental teac values in antiradical qsar of flavonoids Bioorg. Med. Chem. 2010 18 28 35 10.1016/j.bmc.2009.11.015 19944611 37. Zhang H.-Y. Wang L.-F. Theoretical elucidation of structure-activity relationship for coumarins to scavenge peroxyl radical J. Mol. Struct.Theochem 2004 673 199 202 10.1016/j.theochem.2003.12.014 38. Saito M. Sakagami H. Fujisawa S. Cytotoxicity and apoptosis induction by butylated hydroxyanisole (bha) and butylated hydroxytoluene (bht) Anticancer Res. 2003 23 4693 4701 14981915 39. Choe E. Min D.B. Mechanisms of antioxidants in the oxidation of foods Compr. Rev. Food Sci. Food Saf. 2009 8 345 358 10.1111/j.1541-4337.2009.00085.x 40. Natella F. Nardini M. Di Felice M. Scaccini C. Benzoic and cinnamic acid derivatives as antioxidants:  Structure−activity relation J. Agric. Food Chem. 1999 47 1453 1459 10.1021/jf980737w 10563998 41. Musialik M. Kuzmicz R. Pawlowski T.S. Litwinienko G. Acidity of hydroxyl groups: An overlooked influence on antiradical properties of flavonoids J. Org. Chem. 2009 74 2699 2709 10.1021/jo802716v 19275193 42. Hudson B.J.F. Lewis J.I. Polyhydroxy flavonoid antioxidants for edible oils: Structural criteria for activity Food Chem. 1983 10 47 55 10.1016/S0308-8146(83)80001-6 43. Bors W. Heller W. Michel C. Saran M. Flavonoids as antioxidants: Determination of radical-scavenging efficiencies Methods Enzymol. 1990 186 343 355 2172711 44. Garcia Aranzazu M. Ruiz-Mendez V. Romero C. Brenes M. Effect of refining on the phenolic composition of crude olive oils J. Am. Oil Chem. Soc. 2006 83 159 164 10.1007/s11746-006-1189-8 45. Hsu D.-Z. Chu P.-Y. Chandrasekaran V.R.M. Liu M.-Y. Sesame lignan sesamol protects against aspirin-induced gastric mucosal damage in rats J. Funct. Foods 2009 1 349 355 10.1016/j.jff.2009.07.003 46. Erkan N. Ayranci G. Ayranci E. Antioxidant activities of rosemary (Rosmarinus officinalis L.) extract, blackseed (Nigella sativa L.) essential oil, carnosic acid, rosmarinic acid and sesamol Food Chem. 2008 110 76 82 10.1016/j.foodchem.2008.01.058 26050168 47. Guitard R. Paul J.F. Nardello V. Aubry J.M. Myricetin, rosmarinic and carnosic acid as superior natural antioxidant alternatives to a-tocopherol for the preservation of omega-3 oils Food Chem. 2016 submitted for publication 10.1016/j.foodchem.2016.06.038 48. Beghdad M.C. Benammar C. Bensalah F. Sabri F.-Z. Belarbi M. Chemat F. Antioxidant activity, phenolic and flavonoid content in leaves, flowers, stems and seeds of mallow (Malva sylvestris L.) from north western of algeria Afr. J. Biotechnol. 2014 13 486 491 49. Brand-Williams W. Cuvelier M.E. Berset C. Use of a free radical method to evaluate antioxidant activity Food Sci. Technol. 1995 28 25 30 10.1016/S0023-6438(95)80008-5 50. Antolovich M. Prenzler P.D. Patsalides E. McDonald S. Robards K. Methods for testing antioxidant activity Analyst 2002 127 183 198 10.1039/b009171p 11827390 51. Huang D. Ou B. Prior R.L. The chemistry behind antioxidant capacity assays J. Agric. Food Chem. 2005 53 1841 1856 10.1021/jf030723c 15769103 52. Mishra K. Ojha H. Chaudhury N.K. Estimation of antiradical properties of antioxidants using DPPH[rad] assay: A critical review and results Food Chem. 2012 130 1036 1043 10.1016/j.foodchem.2011.07.127 53. Goupy P. Bautista-Ortin A.-B. Fulcrand H. Dangles O. Antioxidant activity of wine pigments derived from anthocyanins: Hydrogen transfer reactions to the DPPH radical and inhibition of the heme-induced peroxidation of linoleic acid J. Agric. Food Chem. 2009 57 5762 5770 10.1021/jf900841b 19566083 54. Goupy P. Dufour C. Loonis M. Dangles O. Quantitative kinetic analysis of hydrogen transfer reactions from dietary polyphenols to the dpph radical J. Agric. Food Chem. 2003 51 615 622 10.1021/jf025938l 12537431 55. Encinar J.M. Pardal A. Martinez G. Transesterification of rapeseed oil in subcritical methanol conditions Fuel Process. Technol. 2012 94 40 46 10.1016/j.fuproc.2011.10.018 56. Lucarini M. Pedulli G.F. Free radical intermediates in the inhibition of the autoxidation reaction Chem. Soc. Rev. 2010 39 2106 2119 10.1039/b901838g 20514720 57. Brigati G. Lucarini M. Mugnaini V. Pedulli G.F. Determination of the substituent effect on the o-h bond dissociation enthalpies of phenolic antioxidants by the epr radical equilibration technique J. Org. Chem. 2002 67 4828 4832 10.1021/jo025755y 12098294 58. Lucarini M. Pedrielli P. Pedulli G.F. Cabiddu S. Fattuoni C. Bond dissociation energies of O-H bonds in substituted phenols from equilibration studies J. Org. Chem. 1996 61 9259 9263 10.1021/jo961039i 59. Foti M.C. Johnson E.R. Vinqvist M.R. Wright J.S. Barclay L.R.C. Ingold K.U. Naphthalene diols: A new class of antioxidants intramolecular hydrogen bonding in catechols, naphthalene diols, and their aryloxyl radicals J. Org. Chem. 2002 67 5190 5196 10.1021/jo020184v 12126405 60. Li J. Bi Y. Liu W. Sun S. Liu C. Ma S. Effect of acid value on tbhq and bht losses in heating oils: Identification of the esterification products of tbhq and free fatty acids J. Am. Oil Chem. Soc. 2014 91 1763 1771 10.1007/s11746-014-2510-6

PMC005xxxxxx/PMC5000619.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081221ijms-17-01221ArticleSynthesis, Crystal Study, and Anti-Proliferative Activity of Some 2-Benzimidazolylthioacetophenones towards Triple-Negative Breast Cancer MDA-MB-468 Cells as Apoptosis-Inducing Agents Abdel-Aziz Hatem A. 1*Eldehna Wagdy M. 2Ghabbour Hazem 34Al-Ansary Ghada H. 5Assaf Areej M. 6Al-Dhfyan Abdullah 7Zhang Ge Academic EditorMalemud Charles J. Academic Editor1 Department of Applied Organic Chemistry, National Research Center, Dokki, Giza 12622, Egypt2 Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Egyptian Russian University, Badr City, Cairo 11829, Egypt; [email protected] Department of Pharmaceutical Chemistry, College of Pharmacy, King Saud University, P.O. Box 2457, Riyadh 11451, Saudi Arabia; [email protected] Department of Medicinal Chemistry, Faculty of Pharmacy, Mansoura University, Mansoura 35516, Egypt5 Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Ain Shams University, Abbassia, Cairo 11566, Egypt; [email protected] Department of Biopharmaceutics and Clinical Pharmacy, Faculty of Pharmacy, The University of Jordan, Amman 11942, Jordan; [email protected] Stem Cell & Tissue Re-Engineering Program, Research Center, King Faisal Specialized Hospital & Research Center, MBC-03, P.O. Box 3354, Riyadh 11211, Saudi Arabia; [email protected]* Correspondence: [email protected]; Tel.: +966-146-77341; Fax: +966-146-7622029 7 2016 8 2016 17 8 122120 4 2016 11 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).On account of its poor prognosis and deficiency of therapeutic stratifications, triple negative breast cancer continues to form the causative platform of an incommensurate number of breast cancer deaths. Aiming at the development of potent anticancer agents as a continuum of our previous efforts, a novel series of 2-((benzimidazol-2-yl)thio)-1-arylethan-1-ones 5a–w was synthesized and evaluated for its anti-proliferative activity towards triple negative breast cancer (TNBC) MDA-MB-468 cells. Compound 5k was the most active analog against MDA-MB-468 (IC50 = 19.90 ± 1.37 µM), with 2.1-fold increased activity compared to 5-fluorouracil (IC50 = 41.26 ± 3.77 µM). Compound 5k was able to induce apoptosis in MDA-MB-468, as evidenced by the marked boosting in the percentage of florecsein isothiocyanate annexin V (Annexin V–FITC)-positive apoptotic cells (upper right (UR) + lower right (LR)) by 2.8-fold in comparison to control accompanied by significant increase in the proportion of cells at pre-G1 (the first gap phase) by 8.13-fold in the cell-cycle analysis. Moreover, a quantitative structure activity relationship (QSAR) model was established to investigate the structural requirements orchestrating the anti-proliferative activity. Finally, we established a theoretical kinetic study. synthesisX-rayanti-proliferativebreast cancer MDA-MB-468 cellsapoptosis ==== Body 1. Introduction Pertaining to the frequency of diagnosis worldwide, breast cancer is regarded as the second most frequently diagnosed cancer and the most frequently diagnosed tumor among women. Also, it is regarded as the fifth leading cause of cancer mortality [1]. In 2012, an estimated 1.67 million newly diagnosed breast cancer cases and 522,000 breast cancer deaths occurred worldwide [1]. The etiology of breast cancer is still unknown, although different risk factors have been established—to name just a few, first-degree relative’s breast cancer family history, mammographic density, benign breast disease, younger age at menarche, low parity, older age at first birth, older age at menopause, high postmenopausal body mass index, low premenopausal body mass index, and endogenous hormone levels have been established as risk factors for breast cancer [2,3]. Breast cancer is regarded as a diverse group of diseases with multiple intrinsic tumor subtypes that have various treatment modalities and long-term survival probabilities. The immunohistochemical expression of the estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor-2 (HER-2) forms the platform of characterization of clinically defined breast cancer subtypes [4]. In approximately 15%–20% of globally diagnosed breast cancers, the tumors do not express ER, PR, or HER-2. Such malignancies are designated as triple-negative breast cancer (TNBC) [5]. In the current medical era, TNBC, among all the breast cancer subgroups, has stood out as the greatest clinical challenge as these tumors have no clinically validated molecularly targeted therapy, are prevalent in younger women, associated with the worst prognosis, and often relapse rapidly. Also, TNBC are highly proliferative, poorly differentiated, often grade III carcinomas, genetically unstable, and preferentially metastasize to the brain and lungs [6,7,8,9,10]. Therefore, there is a critical need to develop potent and effective novel therapies to improve the outcomes of TNBC treatment. Because of their similarity to some naturally occurring nucleotides and their existence in several naturally occurring compounds, benzimidazole derivatives possess a wide range of biological activities and therapeutic effects [11,12,13]. In the field of medicinal chemistry, benzimidazole represents a highly privileged scaffold and has been copiously explored as an anti-proliferative agent targeting different breast cancer cells [14,15,16,17,18,19,20,21]. Surveying the literature revealed that different benzimidazole-based scaffolds were developed with significant activity toward the TNBC MDA-MB-468 cells in the anticancer drug screening program of the U.S. National Cancer Institute (NCI), according to their applied protocol against full NCI 60 human cell lines panel (Figure 1). Many researchers have reported the utility of 2-aryl benzimidazole derivatives as anti-proliferative agents against TNBC. Attaching a heterocyclic moiety, 5-tert-butyl-1H-pyrazol-3-yl, in position 1 and an aryl moiety, 4-chlorophenyl, in position 2 of benzimidazole core resulted in compound I (NSC: 751047) with good anti-proliferative activity against MDA-MB-468 (IC50 = 2.4 µM) [22] (Figure 1), while substitution of position 2 of 5-flouro and 5-methoxybenzimidazole with 1,2,4-oxadiazole moiety through a phenyl ring afforded compounds II and III, respectively, (NSC: 761109 and 761814) with IC50 values of 3.01 and 6.54 µM, respectively, against MDA-MB-468 [23] (Figure 1). Moreover, introduction of different aryl groups through a pyrazole linker at position 2 of the benzimidazole core, as in compounds IV and V (NSC: 768400 and 768399), achieved significant efficacy against MDA-MB-468 (IC50 values of 0.93 and 3.5 µM, respectively) [24] (Figure 1). Interestingly, linking different aryl moeities, through variable heterocyclic groups, via a three-atom linker, namely a propan-1-one group to position 2 of the benzimidazole core led to compounds VI–VIII (NSC: 761980, NSC: 759205 and NSC: 7604520) with low or sub-micromolar anti-proliferative activity against MDA-MB-468 (IC50 = 1.93, 0.79 and 0.69 µM, respectively) [25,26,27] (Figure 1). In addition, we recently introduced an efficacious benzimidazole-based scaffold as for development of potent antitumor agents that prove to have anti-proliferative activity not only toward the cancer stem cells but also toward the bulk of tumor cells of the colon HT-29 cell line [28]. The design of such a scaffold relies on linking different aryl or heteroaryl groups to position 2 of the benzimidazole core through a thio ethan-2-one linker. From the findings reported above, we came to the conclusion that linking the 2-position of benzimidazole scaffold to a terminal aryl or heteroaryl group directly or via variable spacers—an aryl, a heteroaryl, or a propan-1-one group—affords promising molecules that have significant anti-proliferative activity against TNBC. Regarding these points and as a continuation of our research program on the design and synthesis of effective antitumor candidates [29,30,31,32,33,34,35], it was thought worthwhile to extend our investigations around our study [28] to probe for benzimidazole derivatives having anti-proliferative activity towards TNBC. Our structure-based design was three-fold: (i) preserving benzimidazole structure with subsitution at 2-position; (ii) maintaining a terminal lipophilic group; and (iii) establishing a three-atom thio ethane-1-one linker to afford more flexibility for the designed molecules (Figure 2). Thus, the present work reports the synthesis of benzimidazoles 5a–w and their in vitro anti-proliferative activity against the TNBC MDA-MB-468 cell line. Moreover, the most active member in this study, 5k, was selected to be further investigated regarding its effects on cell cycle progression and potential apoptotic effect in the MDA-MB-468 cells, to acquire perception of the mechanism of the anti-proliferative activity of the prepared compounds. Eventually, a theoretical kinetic study was constituted. 2. Results 2.1. Synthetic Approach to Prepare the Target Derivatives The target compounds were prepared following our recently published procedure [28] via the reaction of compound 2 with different aromatic ketones 3a–w in glacial acetic acid in the presence of two equivalents conc. H2SO4 to afford a quantitative yield from the sulfate salts 4a–w. The prepared sulfate salts 4a–w were subsequently neutralized to afford the target 2-((benzimidazol-2-yl)thio)-1-arylethan-1-ones 5a–w in an excellent yield of 82%–96% (Scheme 1). Infrared (IR) spectra for compounds 5a–w displayed absorption bands attributable for the NH group in the range 3324–3460 cm−1, also a (C=O) band in the range of 1654–1690 cm−1. Also, their 1H-nuclear magnetic resonance (NMR) spectra displayed one singlet D2O-exchangeable signal due to the NH proton in the range of δ 12.52–12.80 ppm, whereas (–CH2–) protons appeared as singlet signals within δ 5.00 ppm. Furthermore, the 13C-NMR spectra of compounds 5a–w showed signals resonating in the region δ 182.16–193.90 ppm due to the carbon of carbonyl group, whereas the carbons of the (–CH2–) group appeared in the region of δ 38.46–43.97 ppm. 2.2. Single Crystal Analysis of Compounds 4u and 5v Crystals of compounds 4u and 5v were selected to analyze their single-crystal X-ray crystallographic after slow evaporation from solutions of ethanol. The instrument used is Bruker SMART APEX II D8 Venture diffractometer (Bruker, Karlsruhe, Germany) with graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) at 100 and 150 K, respectively. A direct method was applied to solve the structures that were subsequently refined with SHELXTL [36]. The positions of all the non-H-atoms were provided by E-maps. Using anisotropic temperature factors, the full-matrix least-squares refinement was carried out on F2’s for all non-H-atoms. Crystallographic data was deposited in the Cambridge Crystallographic Data Center and assigned the following deposition numbers: CCDC 1058838 and 1455648 for compounds 4u and 5v, respectively. In Figure 2 and Figure 3, the crystallographic structures of compounds 4u and 5v are represented, respectively. The exact structure is unambiguously defined by the single crystal X-ray study on both derivatives. The crystal structure of 4u confirmed two crystallographically independent cation molecules with one sulfate anion, in the presence of one molecule of ethanol in its asymmetric unit, as shown in Figure 3. The asymmetric unit of 5v contains one molecule only, as depicted in Figure 4. Table 1 listed the crystallographic data and the refinement for the crystals. Table 2 and Table 3 summarized some selected geometric parameters for 4u and 5v, respectively. Also, Figure S1 (in Supplementary Materials) displayed the molecular packing of compound 5v, while Table S1 showed the hydrogen-bond geometry (Å, °) for compound 5v. 2.3. Biological Evaluation of the Target Derivatives as Anti-Cancer Agents 2.3.1. In Vitro Anti-Proliferative Activity against MDA-MB-468 The WST-1 assay, as described by Ngamwongsatit et al. [37], was adopted to evaluate the anti-proliferative activity of the synthesized 2-((benzimidazol-2-yl)thio)-1-arylethan-1-ones 5a–w against human breast cancer cell line MDA-MB-468. 5-FU (florouracil) was selected as a positive control due to its broad spectrum of anticancer activity. The anti-proliferative activity was expressed as growth inhibitory concentration (IC50) values, which represent the compound concentrations required to produce a 50% inhibition of cell growth after 48 hours of incubation compared to untreated controls (Table 4). The obtained results of the tested benzimidazole derivatives 5a–w indicated that most of the prepared compounds showed good to moderate anti-proliferative activity against the tested MDA-MB-468 cancer cell line. Compound 5k merged as the most potent member against MDA-MB-468 (IC50 = 19.90 ± 1.37 µM) as it was 2.1 times more potent and efficacious than 5-fluorouracil (IC50 = 41.26 ± 3.77 µM). Moreover, analogs 5a, 5f, and 5j–t showed superior anti-proliferative activity (IC50 values ranging from 21.98 ± 1.91 to 26.60 ± 2.24 µM) compared to 5-fluorouracil, the reference drug, (IC50 = 41.26 ± 3.77 µM). In addition, compounds 5d, 5i, 5v, and 5w with IC50 = 31.97 ± 3.07, 32.80 ± 3.17, 30.34 ± 3.01 and 28.17 ± 2.24 µM, respectively, displayed good activity against MDA-MB-468. On the other hand, compounds 5b, 5c, 5g, 5h, and 5u were moderately active against MDA-MB-468 with IC50 values ranging from 50.78 ± 5.11 to 74.25 ± 6.23 µM. 2.3.2. Structure Activity Relationship Study (SAR Study) of the Target Compounds Observing the results in Table 4, valuable data could be extracted regarding the structure activity correlation of our compounds. Foremost, the effect of grafting diverse substituents on the terminal aryl moiety on the activities of the synthesized compounds 5a–t was closely investigated. Compound 5a bearing unsubstituted phenyl group showed good activity (IC50 = 22.31 ± 2.04 µM) in comparison to 5-fluorouracil (IC50 = 41.26 ± 3.77 µM), implying a doubling of the anti-proliferative activity. Introduction of fluorine atom, a classical bioisostere of the hydrogen atom, at the 4-position as in compound 5f resulted in comparable activity to the unsubstituted analogue 5a (IC50 = 24.96 ± 2.55 and 22.31 ± 2.04 µM, respectively). Interestingly, transferring the fluorine atom from the 4-position to the 2-position, 5e, resulted in an inactive derivative. Again, di-substitution with two fluorine atoms in the 2- and 4-positions, 5g, was not favorable to the activity (IC50 = 50.78 ± 5.11 µM) compared to the mono 4-F substituted derivative 5f (IC50 = 24.96 ± 2.55 µM). Incorporation of more bulky halogens as chlorine and bromine led to compounds 5d and 5h, respectively, with decreased activity (IC50 = 31.97 ± 3.07 and 53.83 ± 6.53 µM, respectively), suggesting that incorporation of a small halogen fluorine atom only in the 4-position is markedly advantageous to the activity. The order of activities of the halogenated derivatives 5d–h decreased in the order of 4-F > 4-Cl > 2,4-di-F > 4-Br > 2-F. Also, grafting an electron-withdrawing nitro group as in compound 5i resulted only in moderate improvement of the activity (IC50 = 32.80 ± 3.17 µM) compared to 5-FU (IC50 = 41.26 ± 3.77 µM), while introduction of methyl or amino groups, electron-donating groups, reduced the activity against MDA-MB-468, as shown in 5b and 5c analogs (IC50 = 74.25 ± 6.23 and 53.79 ± 5.02 µM, respectively). Contrariwise, substitution with electron-donating hydroxyl, methoxy, or ethoxy groups as in compounds 5j–t maintained the activity in the good range of activity regardless of their positions or numbers (IC50 = (19.90 ± 1.37)–(26.60 ± 2.24) µM). On the other hand, scrutinizing the anti-proliferative activity of compounds 5u–w gave us insight about the effect of exchanging the phenyl group of 5a for other aryl or heteroaryl moieties. Replacement of the phenyl ring of 5a with a naphthalin-2-yl group in compound 5u decreased the activity (IC50 = 66.45 ± 7.12 µM). Moreover, bioisosteric replacement of the phenyl moiety with 2-furyl or 2-thienyl groups (compounds 5v and 5w) moderately reduced the activity (IC50 = 30.34 ± 3.01 and 28.17 ± 2.24 µM, respectively). In conclusion, we can assume that incorporation of an unsubstituted phenyl group or its substitution with electron-donating hydroxyl, methoxy, or ethoxy groups is beneficial for activity against the MDA-MB-468 cell line, while introduction of heterocycles, such as 2-furyl or 2-thienyl, could not effectively replace the phenyl ring. 2.3.3. Cell-Cycle Analysis and Apoptotic Changes Investigation of Compound 5k Cell reproduction necessitates DNA replication with a concomitant nuclear division followed by cytoplasmic partitioning to sporulate two daughter cells. Such a successive routine is known as the “cell cycle” and involves four distinguishable phases. The G1 phase is a gap integrated between the M phase (nuclear division) and the S phase (DNA synthesis); another gap called G2 phase also occurs between S and M. These gaps permit the repair of DNA damage and replication errors [38]. To understand the mechanism behind the tumor suppression activity of the prepared compounds, the most active member in this study, 5k, was selected to be further investigated regarding its effects on cell cycle progression and its potential apoptotic effects in the MDA-MB-468 cell line. The MDA-MB-468 cells were treated with IC50 concentration of compound 5k for 24 h and its effect on the normal cell cycle was detected by fluorescence-activated cell sorting (FACS) analysis (Figure 5, Figure S2). Interestingly, exposure of MDA-MB-468 cells to 5k induced a remarkable augmentation in the proportion of cells at pre-G1 phase by 8.13-fold. The increase was accompanied by concomitant noteworthy mitigation in the percentage of cells at the G0/G1, S, and G2/M phases by 2.21-, 2.43-, and 11.83-fold in comparison to the control, respectively. 2.3.4. Evaluation of the Apoptotic Effect of Compound 5k by Fluorescein Isothiocyanate (FITC)-Labeled Annexin V (Annexin V–FITC) Assay The apoptotic effect of 5k was further evaluated by Annexin VFITC/PI (AV/PI) dual staining assay to examine the occurrence of phosphatidylserine externalization and also to understand whether it is due to physiological apoptosis or nonspecific necrosis. In this study MDA-MB-468 cells were treated with compound 5k for 48 h at 19.9 µM (IC50) to examine the apoptotic effect. It was observed that 5k showed significant apoptosis against MDA-MB-468 cells, as shown in Figure 6 and Figure 7. Results indicated that 5k showed 75.78% of apoptosis at 19.9 µM whereas 27.06% of apoptosis was observed in the control (untreated cells), comprising a 2.8-fold improvement compared to the control. This experiment suggests that 5k significantly induces apoptosis in MDA-MB-468 cells. 2.4. 2 Dimensional-Quantitative Structure Activity Relationship (2D-QSAR) Analysis for the Anti-Proliferative Activity of the Prepared Derivatives 5a–w 2.4.1. Elaboration of QSAR Model QSAR analysis of the anti-tumor activity of the prepared derivatives 5a–w was performed to establish a correlation between the biochemical data and the compound structures; moreover, it aids us in identifying the positive and negative structural features within the three scaffolds. DS 2.5 software (Discovery Studio 2.5, Accelrys, Co., Ltd., Accelrys, San Diego, CA, USA) was used to run the analysis. A set of 21 synthesized derivatives (5a–d, 5f–I, and 5k–w) was applied as a training set with their experimentally detected logIC50 against the MDA-MB-468 cancer cell line in the QSAR modeling. The two remaining synthesized members were used as an external test set to assess the predictive power and validate the established QSAR model. Various molecular descriptors for the training set molecules were calculated using the “Calculate Molecular Properties” module. 2D Descriptors entangled: topological descriptors, molecular properties, molecular property counts, AlogP, surface area, and volume. As for the 3D descriptors: dipole, principal moments of inertia, jurs descriptors, surface area, and volume, and shadow indices. To search for the best QSAR regression equation, genetic function approximation (GFA) was utilized, i.e., multiple linear regression modeling (MLR). 2.4.2. QSAR Study Results The best performing QSAR model is represented by Equation (1); Potency (LogIC50) against MDA-MB-468 cell line LogIC50 = 4.1269 + 0.0599 Num_ExplicitAtoms − 0.0139 Molecular_SAVol + 3.0316 CHI_V_3_C − 0.0920 Jurs_RPCS(1) Adopting Equation (1), QSAR model was graphically represented. This was accomplished by plotting the experimental values against the predicted bioactivity values logIC50 for the training set compounds, as shown in Figure 8. Also, the estimated and experimental activities data and the calculated descriptors of the training set compounds were summarized in Table 5. The Least-Squares method was used to build the models, r2 = 0.842, r2 (adj) = 0.803, r2 (pred) = 0.795, Least-Squared error = 0.004 for model 1, where r2 (adj) is r2 adjusted for the number of terms in the model; r2 (pred) is the prediction r2, equivalent to q2 from a leave-one-out cross-validation. 2.4.3. QSAR Validation Two of the prepared compounds (5e and 5j) were utilized to carry out the external validation of the determined QSAR equation. 5e and 5j were chosen as they exhibit mild and excellent activities. The observed activities versus those provided by QSAR study are presented in Table 6. 2.5. Theoretical Kinetic ADME Study of the Target Derivatives 5a–w A theoretical kinetic study carried out by Discovery Studio 2.5 software (Accelrys) was adopted to predict the ADME of the prepared derivatives 5a–w, Table 7. The lipophilicity was evaluated by calculating AlogP98, whereas the PSA_2D descriptor was adopted to estimate the polar surface area. Moreover, solubility level was predicted where all members of this study seemed to possess low solubility. In accordance with this anticipation, absorption levels implicate that they are well absorbed. Also, they are predicted to be non-inhibitors of CYP2D not to mention compounds 5b, 5q, and 5s, which are expected to inhibit CYP2D. LogP for compound 5k was determined experimentally and found to equal 3.75. It is worth mentioning that all compounds passed Lipinski’s rule of five. 3. Discussion In summary, a novel series of 2-((benzimidazol-2-yl)thio)-1-arylethan-1-one derivatives 5a–w has been synthesized. Their anti-proliferative activity against triple-negative breast cancer MDA-MB-468 cells was evaluated. Compound 5k was found to be the most active compound in this study with IC50 value of 19.90 ± 1.37 µM as it was 2.1 times more potent and efficacious than 5-fluorouracil (IC50 = 41.26 ± 3.77 µM). Also, analogs 5a, 5f, and 5j–t possessed excellent anti-proliferative activity with IC50 values ranging from 21.98 ± 1.91 to 26.60 ± 2.24 µM, which are better than the used reference drug. The preliminary SAR study showed that incorporation of unsubstituted phenyl group or its substitution with electron-donating hydroxyl, methoxy, or ethoxy groups are essential elements for the anti-tumor activity against MDA-MB-468, while introduction of heterocycles, such as 2-furyl or 2-thienyl, could not effectively replace the phenyl ring. In a cell-cycle analysis, compound 5k increased the percentage of MDA-MB-468 cells at pre-G1 by 8.13-fold and G2/M phase by 11.83-fold. Furthermore, treatment of MDA-MB-468 cells with 5k led to a marked increase in the percentage of annexin V–FITC-positive apoptotic cells (UR + LR) by 2.8-fold compared to the control. In addition, a QSAR model was established to investigate the structural requirements controlling activity against MDA-MB-468. Of note, the anticipated activities by the QSAR model were very near to the experimentally determined activities. Accordingly, this model could be conveniently applied for the prediction of more effective hits bearing the same structural framework. A theoretical kinetic study was constituted to anticipate the ADME of the prepared benzimidazoles. Moreover, single crystal X-ray diffraction has been included for compounds 4u and 5v. Through this work we planned to add a scientific contribution for the treatment of the resistant type; TNBC by exploring different 2-benzimidazole derivatives. Our design was inspired by previously reported active scaffolds. Among the designed and synthetized molecules, many of them showed 2-fold increase in activity compared to 5-FU; this led us to develop a fruitful SAR analysis that will be a guideline for our future work. 4. Experimental 4.1. Chemistry 4.1.1. General Melting points were determined using a Gallenkamp melting point apparatus (WeissTechnik, Loughborough, UK) and are uncorrected. Infrared (IR) Spectra were recorded as KBr disks using the Perkin Elmer FT-IR Spectrum BX apparatus (PerkinElmer, Boston, MA, USA). Mass spectra were measured on an Agilent TripleQuadrupole 6410 QQQ LC/MS equipped with an ESI (electrospray ionization) source (Agilent Technologies, Santa Clara, CA, USA). NMR Spectra were recorded on a Bruker NMR spectrometer (Bruker, Karlsruhe, Germany). 1H spectrum was run at 500 MHz and 13C spectrum was run at 125 MHz in deuterated dimethyl sulfoxide (DMSO-d6). Chemical shifts are expressed in δ values (ppm) using the solvent peak as internal standard. All coupling constant (J) values are given in hertz. The abbreviations used are as follows: s, singlet; d, doublet; m, multiplet. Elemental analyses were carried out at the Regional Center for Mycology and Biotechnology, Al-Azhar University, Cairo, Egypt. Analytical thin layer chromatography (TLC) on silica gel plates containing UV indicator (Merck KGaA, Darmstadt, Germany) was employed routinely to follow the course of reactions and to check the purity of products. All reagents and solvents were purified and dried by standard techniques. Compounds 4 & 5a, b, f, h, 4 & 5l–o, 4 & 5q, r, t, v, and w, are previously reported [28]. 4.1.2. Benzoimidazole-2-Thiol 2 Prepared according to the reported procedures [39]. 4.1.3. General Procedures for Synthesis of Sulfate Salts 4a–w Prepared according to the reported procedures [28]. 2-((2-(4-Aminophenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4c). White crystals, (yield 97%), m.p. 210–213 °C; IR (KBr, ν cm−1): 3450 (NH) and 1684 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.23 (s, 2H, CH2), 6.65 (d, 2H, H-3 and H-5 of 4-NH2C6H4, J = 8.5 Hz), 6.70 (s, 2H, NH2), 7.49–7.51 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.71–7.73 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.87 (d, 2H, H-2 and H-6 of 4-NH2C6H4, J = 8.5 Hz), 10.55 (s, 1H, NH), 12.64 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 41.55 (CH2), 113.40, 113.59, 114.66, 125.76, 130.92, 131.62, 132.74, 151.68, 189.11, 195.74 (C=O); ESI MS m/z: 665 [M + 1]+; Anal. Calcd. for C30H28N6O6S3: C, 54.20; H, 4.25; N, 12.64; Found C, 54.28; H, 4.21; N, 12.75. 2-((2-(4-Chlorophenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4d). White crystals, (yield 98%), m.p. 227–230 °C; IR (KBr, ν cm−1): 3400 (NH) and 1683 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.29 (s, 2H, CH2), 7.37–7.40 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.62–7.65 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.68 (d, 2H, H-3 and H-5 of 4-ClC6H4, J = 8.5 Hz), 8.08 (d, 2H, H-2 and H-6 of 4-ClC6H4, J = 8.5 Hz), 10.55 (s, 1H, NH), 12.61 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 56.50 (CH2), 113.84, 120.85, 122.77, 124.68, 129.53, 130.89, 134.12, 134.86, 139.48, 150.64, 192.07 (C=O); ESI MS m/z: 703 [M + 1]+, 704 [M + 2]+; Anal. Calcd. for C30H24Cl2N4O6S3: C, 15.21; H, 3.44; N, 7.96; Found C, 15.30; H, 3.49; N, 8.02. 2-((2-(2-Fluorophenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4e). White crystals, (yield 97%), m.p. 219–222 °C; IR (KBr, ν cm−1): 3435 (NH) and 1669 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.15 (s, 2H, CH2), 7.37–7.40 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.41–7.48 (m, 2H, Ar–H), 7.60–7.62 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.75–7.79 (m, 1H, Ar–H), 7.94 (t, 1H, H-6 of 2-FC6H4, J = 7.5 Hz), 10.64 (s, 1H, NH), 12.73 (s, 1H, NH). ESI MS m/z: 671 [M + 1]+; Anal. Calcd. for C30H24F2N4O6S3: C, 53.72; H, 3.61; N, 8.35; Found C, 53.67; H, 3.65; N, 8.38. 2-((2-(2,4-Difluorophenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4g). White crystals, (yield 96%), m.p. 206–209 °C; IR (KBr, ν cm−1): 3468 (NH) and 1680 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.27 (s, 2H, CH2), 7.30 (t, 1H, Ar–H, J = 8.5 Hz), 7.43–7.46 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.52 (t, 1H, Ar–H, J = 9.0 Hz), 7.65–7.69 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 8.03 (t, 1H, Ar–H, J = 9.0 Hz),10.52 (s, 1H, NH), 12.67 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 44.06 (CH2), 105.70, 105.91, 106.12, 113.09, 113.27, 113.76, 120.81, 125.25, 133.50, 133.85, 150.71, 189.03 (C=O). ESI MS m/z: 707 [M + 1]+; Anal. Calcd. for C30H22F4N4O6S3: C, 50.99; H, 3.14; N, 7.93; Found C, 50.91; H, 3.18; N, 8.01. 2-((2-(2-Hydroxyphenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4j). White crystals, (yield 98%), m.p. 238–240 °C; IR (KBr, ν cm−1): 3466 (NH) and 1684 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.30 (s, 2H, CH2), 7.02–7.11 (m, 4H, Ar–H), 7.38–7.42 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.67–7.70 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 9.59 (s, 1H, OH), 10.60 (s, 1H, NH), 12.87 (s, 1H, NH). ESI MS m/z: 667 [M + 1]+, 668 [M + 2]+; Anal. Calcd. for C30H26N4O8S3: C, 54.04; H, 3.93; N, 8.40; Found C, 54.12; H, 3.91; N, 8.46. 2-((2-(3-Hydroxyphenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4k). White crystals, (yield 97%), m.p. 228–230 °C; IR (KBr, ν cm−1): 3418 (NH) and 1670 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.34 (s, 2H, CH2), 7.11 (d, 1H, H-4 of 3-OHC6H4, J = 8.0 Hz), 7.40–7.44 (m, 2H, H-2 and H-3 of 3-OHC6H4), 7.47–7.49 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.54 (d, 1H, H-6 of 3-OHC6H4, J = 8.0 Hz), 7.70–7.72 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 9.73 (s, 1H, OH), 10.48 (s, 1H, NH), 12.61 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 41.58 (CH2), 113.69, 115.09, 120.03, 121.82, 125.59, 130.55, 133.19, 136.46, 151.08, 158.23, 192.52 (C=O). ESI MS m/z: 667 [M + 1]+; Anal. Calcd. for C30H26N4O8S3: C, 54.04; H, 3.93; N, 8.40; Found C, 54.13; H, 4.03; N, 8.46. 2-((2-(2,4-Dimethoxyphenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4p). White crystals, (yield 98%), m.p. 235–237 °C; IR (KBr, ν cm−1): 3377 (NH) and 1680 (C=O); 1H-NMR (DMSO-d6) δ ppm: 3.88 (s, 3H, OCH3), 4.00 (s, 3H, OCH3), 5.08 (s, 2H, CH2), 6.67 (d, 1H, H-5 of 2,4-(OCH3)2C6H3, J = 9.0 Hz), 6.74 (s, 1H, H-3 of 2,4-(OCH3)2C6H3), 7.43–7.44 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.65–7.66 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.79 (d, 1H, H-6 of 2,4-(OCH3)2C6H3, J = 9.0 Hz), 10.46 (s, 1H, NH), 12.37 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 45.31 (CH2), 56.33 (OCH3), 56.73 (OCH3), 98.92, 107.24, 113.37, 117.65, 121.38, 125.12, 133.06, 151.57, 162.02, 165.95, 191.78 (C=O). ESI MS m/z: 755 [M + 1]+; Anal. Calcd. for C34H34N4O10S3: C, 54.10; H, 4.54; N, 7.42; Found C, 54.16; H, 4.59; N, 7.38. 2-((2-(3,4,5-Trimethoxyphenyl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4s). White crystals, (yield 98%), m.p. 213–215 °C; IR (KBr, ν cm−1): 3427 (NH) and 1675 (C=O); 1H-NMR (DMSO-d6) δ ppm: 3.77 (s, 6H, 2OCH3), 3.89 (s, 12H, 4OCH3), 5.29 (s, 4H, 2CH2), 7.40–7.67 (m, 12H, Ar–H), 10.67 (s, 2H, 2NH), 12.49 (s, 2H, 2NH). ESI MS m/z: 815 [M + 1]+, 816 [M + 2]+; Anal. Calcd. for C36H38N4O12S3: C, 53.06; H, 4.70; N, 6.88; Found C, 52.96; H, 4.63; N, 6.80. 2-((2-(Naphthalen-2-yl)-2-oxoethyl)thio)-1H-benzo[d]imidazol-3-ium sulfate (4u). White crystals, (yield 96%), m.p. 246–250 °C; IR (KBr, ν cm−1): 3412 (NH) and 1670 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.38 (s, 2H, CH2), 7.31–7.33 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.57–7.60 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.66–7.74 (m, 2H, H-6 and H-7 of naphthalene), 8.04 (d, 2H, Ar–H, J = 8.5 Hz), 8.08 (d, 1H, Ar–H, J = 8.5 Hz), 8.17 (d, 1H, Ar–H, J = 8.0 Hz), 8.85 (s, 1H, H-1 of naphthalene), 10.54 (s, 1H, NH), 12.68 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 56.49 (CH2), 113.96, 123.93, 124.13, 127.72, 128.26, 129.01, 130.14, 131.27, 132.56, 132.83, 135.79, 136.26, 150.58, 193.14 (C=O). ESI MS m/z: 735 [M + 1]+; Anal. Calcd. for C38H30N4O6S3: C, 62.11; H, 4.11; N, 7.62; Found C, 62.18; H, 4.13; N, 7.57. 4.1.4. General Procedure for Preparation of the Target Derivatives 5a–w An aqueous solution (10 mL) of sodium bicarbonate was added to a stirred suspension of the adequate sulfate salts 4a–w (4 mmol) in water (20 mL). The mixture was stirred for 2 h at room temperature. The obtained solid was collected by filtration, washed several times with water, then dried and recrystallized from ethanol to furnish compounds 5a–w. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(4-aminophenyl)ethan-1-one (5c). White crystals (yield 85%), m.p. 175–178 °C; IR (KBr, ν cm−1): 3412 (NH) and 1680 (C=O); 1H-NMR (DMSO-d6) δ ppm: 4.87 (s, 2H, CH2), 6.21 (s, 2H, NH2), 6.59 (d, 2H, H-3 and H-5 of 4-NH2C6H4, J = 9.0 Hz), 7.10–7.14 (m, 4H, H-4, H-5, H-6 and H-7 of 2-mercaptobenzimidazole), 7.76 (d, 2H, H-2 and H-6 of 4-NH2C6H4, J = 8.5 Hz), 12.53 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 40.57 (CH2), 109.93, 113.03, 121.82, 122.78, 123.27, 131.42, 132.70, 150.41, 154.72, 190.66 (C=O); ESI MS m/z: 284 [M + 1]+; Anal. Calcd. for C15H13N3OS: C, 63.58; H, 4.62; N, 14.83; Found C, 63.81; H, 4.60; N, 14.89. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(4-chlorophenyl)ethan-1-one (5d). White crystals (yield 90%), m.p. 179–181 °C (reported: 189–191 °C [40]); IR (KBr, ν cm−1): 3410 (NH) and 1675 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.04 (s, 2H, CH2), 7.09–7.13 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.40–7.42 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.64 (d, 2H, H-3 and H-5 of 4-ClC6H4, J = 8.0 Hz), 8.08 (d, 2H, H-2 and H-6 of 4-ClC6H4, J = 8.5 Hz), 12.64 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 40.49 (CH2), 112.45, 121.87, 129.41, 130.82, 134.67, 139.06, 149.82, 193.09 (C=O); ESI MS m/z: 302.9 [M]+, 304.9 [M + 2]+. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(2-fluorophenyl)ethan-1-one (5e). White crystals (yield 89%), m.p. 145–148 °C; IR (KBr, ν cm−1): 3420 (NH) and 1654 (C=O); 1H-NMR (DMSO-d6) δ ppm: 4.93 (s, 2H, CH2), 7.09–7.12 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.37–7.49 (m, 4H, Ar–H), 7.70–7.74 (m, 1H, Ar–H), 8.08 (t, 1H, Ar–H, J = 7.5 Hz), 12.66 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 43.11 (CH2), 109.23, 117.29, 118.50, 121.69, 122.18, 124.61, 125.38, 131.09, 136.01, 149.80, 157.31, 160.50, 162.52, 191.90 (C=O); ESI MS m/z: 287 [M + 1]+, 288 [M + 2]+; Anal. Calcd. for C15H11FN2OS: C, 62.92; H, 3.87; N, 9.78; Found C, 63.09; H, 3.90; N, 9.84. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(2,4-difluorophenyl)ethan-1-one (5g). White crystals (yield 94%), m.p. 116–120 °C; IR (KBr, ν cm−1): 3420 (NH) and 1675 (C=O); 1H-NMR (DMSO-d6) δ ppm: 4.91 (s, 2H, CH2), 7.08–7.12 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.26 (t, 1H, Ar–H, J = 8.5 Hz), 7.39–7.41 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.47 (t, 1H, Ar–H, J = 9 Hz), 8.00 (q, 1H, Ar–H, J = 8.5 Hz), 12.61 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 42.95 (CH2), 105.54, 105.75, 105.96, 112.89, 113.04, 121.88, 133.35, 149.73, 190.71 (C=O); ESI MS m/z: 304 [M]+, 305 [M + 1]+; Anal. Calcd. for C15H10F2N2OS: C, 59.20; H, 3.31; N, 9.21; Found C, 59.46; H, 3.29; N, 9.32. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(2-hydroxyphenyl)ethan-1-one (5j). White crystals (yield 85%), m.p. 205–208 °C (reported: 201 °C [41]); IR (KBr, ν cm−1): 3408 (NH) and 1670 (C=O); 1H-NMR (DMSO-d6) δ ppm: 4.91 (s, 2H, CH2), 6.99–7.15 (m, 4H, Ar–H), 7.38–7.41 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.83–7.95 (m, 2H, Ar–H), 9.37 (s, 1H, OH), 12.53 (s, 1H, NH); ESI MS m/z: 285 [M + 1]+. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(3-hydroxyphenyl)ethan-1-one (5k). White crystals (yield 87%), m.p. 228–230 °C (reported: 224–227 °C [42]); IR (KBr, ν cm−1): 3336 (NH) and 1660 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.01 (s, 2H, CH2), 7.08–7.10 (m, 3H, H-5, H-6 of 2-mercaptobenzimidazole and H-4 of 3-OHC6H4), 7.37–7.45 (m, 4H, H-4, H-7 of 2-mercaptobenzimidazole and H-5, H-6 of 3-OHC6H4), 7.54 (s, 1H, H-2 of 3-OHC6H4), 9.89 (s, 1H, OH), 12.61 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 40.47 (CH2), 114.96, 119.86, 121.28, 121.61, 130.43, 137.23, 150.01, 158.14, 193.72 (C=O); ESI MS m/z: 285. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(2,4-dimethoxyphenyl)ethan-1-one (5p). White crystals (yield 92%), m.p. 208–211 °C; IR (KBr, ν cm−1): 3413 (NH) and 1655 (C=O); 1H-NMR (DMSO-d6) δ ppm: 3.92 (s, 3H, OCH3), 3.96 (s, 3H, OCH3), 4.80 (s, 2H, CH2), 6.65 (d, 1H, H-5 of 2,4-(OCH3)2C6H3, J = 9.0 Hz), 6.71 (s, 1H, H-3 of 2,4-(OCH3)2C6H3), 7.06–7.09 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.40–7.41 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.73 (d, 1H, H-6 of 2,4-(OCH3)2C6H3, J = 8.5 Hz), 12.60 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 43.97 (CH2), 56.20 (OCH3), 56.60 (OCH3), 98.88, 104.25, 106.96, 110.65, 114.19, 118.27, 121.63, 132.88, 150.15, 161.52, 165.36, 193.07 (C=O); ESI MS m/z: 329 [M + 1]+; Anal. Calcd. for C17H16N2O3S: C, 62.18; H, 4.91; N, 8.53; Found C, 62.40; H, 4.94; N, 8.45. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(3,4,5-trimethoxyphenyl)ethan-1-one (5s). White crystals (yield 88%), m.p. 233–235 °C; IR (KBr, ν cm−1): 3405 (NH) and 1670 (C=O); 1H-NMR (DMSO-d6) δ ppm: 3.77 (s, 3H, OCH3), 3.86 (s, 6H, OCH3), 5.04 (s, 2H, CH2), 7.12–7.43 (m, 6H, Ar–H), 12.68 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 39.37 (CH2), 56.60 (OCH3), 60.67 (OCH3), 106.58, 121.92, 131.12, 142.73, 149.87, 153.29, 192.94 (C=O); Anal. Calcd. for C18H18N2O4S: C, 60.32; H, 5.06; N, 7.82; Found C, 60.41; H, 5.03; N, 7.75. 2-((1H-Benzo[d]imidazol-2-yl)thio)-1-(naphthalen-2-yl)ethan-1-one (5u). White crystals (yield 92%), m.p. 160–162 °C; IR (KBr, ν cm−1): 3415 (NH) and 1673 (C=O); 1H-NMR (DMSO-d6) δ ppm: 5.20 (s, 2H, CH2), 7.10–7.13 (m, 2H, H-5 and H-6 of 2-mercaptobenzimidazole), 7.41–7.43 (m, 2H, H-4 and H-7 of 2-mercaptobenzimidazole), 7.64–7.72 (m, 2H, Ar–H), 8.02–8.07 (m, 3H, Ar–H), 8.15 (d, 1H, Ar–H, J = 8.0 Hz), 8.85 (s, 1H, H-1 of naphthalene), 12.80 (s, 1H, NH); 13C-NMR (DMSO-d6) δ ppm: 40.42 (CH2), 121.92, 124.19, 127.59, 128.20, 128.91, 129.42, 130.12, 131.11, 132.59, 133.19, 135.69, 149.99, 193.90 (C=O); ESI MS m/z: 319 [M + 1]+; Anal. Calcd. for C19H14N2OS: C, 71.68; H, 4.43; N, 8.80; Found C, 71.81; H, 4.40; N, 8.73. 4.2. Biological Evaluation 4.2.1. In Vitro Evaluation of the Anti-Proliferative Activity The synthesized derivatives 5a–w was evaluated for their anti-proliferative activity via the Stem Cell Therapy and Tissue Reengineering Program in the King Faisal Specialized Hospital and Research Center, Riyadh, Saudi Arabia. In vitro anti-proliferative activity was measured by the cell growth inhibition assay. This assay was conducted using a WST-1 reagent (Sigma-Aldrich Chemie Gmbh, Munich, Germany) for determination of the IC50 for each compound and the results are given in Table 4. MDA-MB-468 breast cancer cell line was purchased from the American Type Culture Collection (Manassas, Virginia, USA). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma-Aldrich Chemie Gmbh), supplemented with 10% FBS (Lonza, Visp, Switzerland), 100 IU/mL penicillin, 100 mg/mL streptomycin, and 2 mmol/L l-glutamine (Sigma). Cells were seeded into 96-well plates at 0.4 × 104/well and incubated overnight. The medium was replaced with a fresh one containing the desired concentrations of the test compounds. After 48 h, 10 µL of the WST-1 reagent were added to each well and the plates were re-incubated for 4 h at 37 °C. The amount of formazan was quantified using an ELISA reader (Thermo Fisher Scientific, Waltham, MA, USA) at 450 nm. The IC50 values were calculated according to the equation for Boltzmann sigmoidal concentration response curve using the nonlinear regression models (GraphPad, Prism Version 5, San Diego, CA, USA). The results reported are means of at least three separate experiments. Significant differences were analyzed by one-way analysis of variance (ANOVA) wherein the differences were considered to be significant at p < 0.05. 4.2.2. Cell Cycle Analysis The MDA-MB-468 cells were subjected to treatment with 19.90 µM of compound 5k for 24 h. Consequently, the cells were washed twice with ice-cold phosphate buffered saline (PBS). The treated cells were collected by centrifugation, fixed in ice-cold 70% (v/v) ethanol, washed with PBS, re-suspended with 0.1 mg/mL RNase, stained with 40 mg/mL PI, and analyzed by flow cytometry using FACScalibur (Becton Dickinson, BD, San Jose, CA, USA). The cell cycle distributions were calculated using CellQuest software (Becton Dickinson). 4.2.3. Annexin V–FITC Apoptosis Assay The MDA-MB-468 cells were seeded as described above and then incubated with 19.90 µM of compound 5k for 24 h. Cells were harvested, washed twice with PBS, and centrifuged. In brief, 105 of cells were treated with annexin V–FITC and propidium iodide (PI) using the apoptosis detection kit (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s protocol. Annexin V–FITC and PI binding were analyzed by flow cytometry on FACScalibur (BD Biosciences) without gating restrictions using 10,000 cells. Data were collected using logarithmic amplification of both the FL1 (FITC) and the FL2 (PI) channels. Quadrant analysis of co-ordinate dot plots was performed with CellQuest software. Unstained cells were used to adjust the photomultiplier voltage and for compensation setting adjustment to eliminate spectral overlap between the FL1 and the FL2 signal. Acknowledgments The authors would like to extend their sincere appreciation to the Deanship of Scientific Research at King Saud University (Riyadh, Saudi Arabia) for its funding of this research through the Research Group Project no. PRG-1436-038. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1221/s1. Click here for additional data file. Author Contributions Hatem A. Abdel-Aziz and Wagdy M. Eldehna conceived and designed the experiments; Wagdy M. Eldehna, Hazem Ghabbour, Ghada H. Al-Ansary carried out the experiments; Wagdy M. Eldehna and Ghada H. Al-Ansary analyzed and interpreted the data; Hazem Ghabbour carried out the single crystal analysis of compounds 4u and 5v and interpreted their data; Hatem A. Abdel-Aziz, Wagdy M. Eldehna and Ghada H. Al-Ansary prepared the manuscript; Areej M. Assaf and Abdullah Al-Dhfyan performed the biological screening. All authors have read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figures, Scheme and Tables Figure 1 Structures of some reported benzimidazoles I–VIII, by other research groups, with anti-proliferative activity against triple-negative breast cancer MDA-MB-468 cells [22,23,24,25,26,27]. Figure 2 Structure-based design of target benzimidazoles 5a–w as anti-triple-negative breast cancer (TNBC) agents. ijms-17-01221-sch001_Scheme 1Scheme 1 Synthesis of the target 2-((benzimidazol-2-yl)thio)-1-arylethan-1-ones 5a–w and their corresponding sulfate salts 4a–w. Reagents and conditions: (i) CS2/KOH/ethanol/reflux 2 h; (ii) Glacial acetic acid/reflux 0.5 h; (iii) Aqueous Na2CO3/r.t. (room temperature) 2 h. Figure 3 An ORTEP diagram of the final X-ray structure of compound 4u. Figure 4 An ORTEP diagram of the final X-ray structure of compound 5v. Figure 5 Bar chart shows percentage of MDA-MB-468 cells at each phase of cell cycle in control cells and cells treated with compound 5k. Figure 6 Fluorescein isothiocyanate (FITC)-labeled annexin V (Annexin V–FITC) staining. The cells were treated with dimethylsulfoxide (DMSO) as a control or with compound 5k at IC50 concentration for 24 h. The experiment was done in triplicate. Figure 7 Effect of compound 5k on the percentage of annexin V–FITC-positive staining in MDA-MB-468 cells. Data are presented as means ± S.D. a Indicates statistical difference from control at p < 0.0001. Figure 8 Predicted versus experimental logIC50 of the tested compounds against MDA-MB-468 cell line according to Equation (1) (r2 = 0.842). ijms-17-01221-t001_Table 1Table 1 Crystallographic data and refinements for compounds 4u and 5v. Compound 4u 5v Crystal data Chemical formula 2(C19H15N2OS) C2H6O·SO4 C13H10N2O2S Mr 780.94 258.29 Crystal system, space group Orthorhombic, P212121 Monoclinic, P21/c Temperature (K) 100 100 a, b, c (Å) 14.2409 (8), 15.8339 (9), 16.2043 (8) 11.4097 (7), 10.8336 (7), 10.1256 (7) β (°) 90.00 114.165 (2) V (Å3) 3653.9 (3) 1141.93 (13) Z 4 4 Radiation type Mo Kα Mo Kα µ (mm−1) 0.26 0.28 Crystal size (mm) 0.44 × 0.26 × 0.16 0.47 × 0.36 × 0.11 Data collection Diffractometer CCD area detector diffractometer Bruker APEX-II D8 venture diffractometer Absorption correction multi-scan, SADABS multi-scan, SADABS Tmin, Tmax 0.90, 0.92 0.89, 0.93 Number of measured, independent and observed [I > 2σ(I)] reflections 11,220, 11,220, 8647 9364, 1999, 1689 R int 0.090 Refinement R[F2 > 2ó(F2)], wR(F2), S 0.050, 0.121, 1.05 0.086, 0.250, 1.10 Number of reflections 11,220 1999 Number of parameters 494 163 Number of restraints 0 0 H-atom treatment by a mixture of independent and constrained refinement by a mixture of independent and constrained refinement Δρmax, Δρmin (e·Å−3) 0.76, −0.37 1.12, −0.87 CCDC number 1,058,838 1,455,648 ijms-17-01221-t002_Table 2Table 2 Selected geometric parameters (Å, °) for compound 4u. Bond Distances S1A–C7A 1.722 (3) O6–C21 1.437 (6) S1A–C8A 1.811 (3) N1A–C7A 1.333 (4) S1B–C7B 1.724 (3) N1A–C1A 1.406 (4) S1B–C8B 1.804 (3) N2A–C6A 1.387 (4) S2–O3 1.476 (3) N2A–C7A 1.338 (4) S2–O5 1.479 (3) N1B–C7B 1.342 (4) S2–O2 1.478 (3) N1B–C1B 1.398 (4) S2–O4 1.467 (2) N2B–C7B 1.339 (4) O1A–C9A 1.203 (6) N2B–C6B 1.381 (4) Bond Angles C7A–S1A–C8A 102.76 (15) N1A–C7A–N2A 109.8 (3) C7B–S1B–C8B 99.02 (15) S1A–C7A–N1A 130.1 (2) O3–S2–O5 108.55 (14) S1A–C8A–C9A 105.5 (2) O4–S2–O5 111.31 (14) O1A–C9A–C10A 121.0 (4) O3–S2–O4 110.49 (14) O1A–C9A–C8A 120.6 (3) O2–S2–O3 108.54 (16) N1B–C1B–C2B 131.1 (3) O2–S2–O4 109.46 (15) N1B–C1B–C6B 106.5 (3) O2–S2–O5 108.43 (14) N2B–C6B–C5B 131.8 (3) C1A–N1A–C7A 108.2 (3) N2B–C6B–C1B 107.0 (3) C6A–N2A–C7A 108.7 (3) S1B–C7B–N1B 128.7 (3) C1B–N1B–C7B 108.0 (3) S1B–C7B–N2B 121.3 (2) C6B–N2B–C7B 108.7 (3) N1B–C7B–N2B 109.9 (3) N1A–C1A–C2A 130.9 (3) S1B–C8B–C9B 107.8 (2) N1A–C1A–C6A 106.5 (3) O1B–C9B–C10B 122.0 (3) N2A–C6A–C1A 106.9 (3) O1B–C9B–C8B 120.4 (3) N2A–C6A–C5A 131.1 (3) O6–C21–C20 111.9 (4) ijms-17-01221-t003_Table 3Table 3 Selected geometric parameters (Å, °) for compound 5v. Bond Distances Cl1–C13 1.7361 (12) N1–C1 1.3831 (13) S1–C7 1.7521 (10) N1–C7 1.3686 (12) S1–C8 1.7949 (10) N2–C6 1.3947 (13) S1–C8 1.7949 (10) N2–C6 1.3947 (13) O1–C9 1.2146 (12) N2–C7 1.3157 (12) Bond Angles C6–S1–C7 102.2 (2) S1–C6–C5 113.5 (4) C3–O1–C4 106.5 (4) S1–C7–N2 117.3 (3) C7–N1–C8 103.8 (4) N1–C7–N2 114.4 (5) C7–N2–C13 106.7 (4) S1–C7–N1 128.4 (4) O1–C3–C2 111.2 (5) N1–C8–C9 130.2 (4) O1–C4–C5 116.2 (4) N1–C8–C13 109.6 (4) O1–C4–C1 109.8 (4) N2–C13–C8 105.5 (5) O2–C5–C6 123.4 (5) N2–C13–C12 132.2 (5) ijms-17-01221-t004_Table 4Table 4 In vitro anti-proliferative activity of compounds 5a–w against breast MDA-MB-468 cancer cell line. Compound Aryl IC50 (µM) a 5a C6H5 22.31 ± 2.04 5b 4-Me–C6H4 74.25 ± 6.23 5c 4-NH2–C6H4 53.79 ± 5.02 5d 4-Cl–C6H4 31.97 ± 3.07 5e 2-F–C6H4 NA b 5f 4-F–C6H4 24.96 ± 2.55 5g 2,4-di-F–C6H3 50.78 ± 5.11 5h 4-Br–C6H4 53.83 ± 6.53 5i 4-NO2–C6H4 32.80 ± 3.17 5j 2-OH–C6H4 22.41 ± 2.13 5k 3-OH–C6H4 19.90 ± 1.37 5l 4-OH–C6H4 21.98 ± 1.91 5m 2-OCH3–C6H4 22.32 ± 2.06 5n 4-OCH3–C6H4 25.35 ± 2.15 5o 3-OH-4-OCH3–C6H3 22.06 ± 1.93 5p 2,4-(OCH3)2–C6H3 26.60 ± 2.24 5q 3,4-(OCH3)2–C6H3 24.30 ± 2.13 5r 3,4-O-CH2–O–C6H3 22.79 ± 1.87 5s 3,4,5-(OCH3)3–C6H2 22.72 ± 2.03 5t 4-OC2H5–C6H4 23.91 ± 2.19 5u naphthalin-2-yl 66.45 ± 7.12 5v furan-2-yl 30.34 ± 3.01 5w thiophen-2-yl 28.17 ± 2.24 5-Fluorouracil 41.26 ± 3.77 a IC50 values are the mean ± S.D. of three separate experiments; b NA: Compounds having IC50 value > 100 µM. ijms-17-01221-t005_Table 5Table 5 Estimated IC50 data of the training set against MDA-MB-468 cell line and calculated descriptors governing IC50 according to Equation (1). Compound Experimental Activity (LogIC50) Predicted Activity (LogIC50) Residual Num_ExplicitAtoms Molecular_SAVol CHI_V_3_C Jurs_RPCS 5a 1.3485 1.4134 −0.0649 31 414.53 0.429 0.817 5b 1.8707 1.8742 −0.0035 34 430.44 0.596 0.834 5c 1.7307 1.6624 0.0683 33 426.68 0.525 0.736 5d 1.5047 1.5789 −0.0742 31 443.37 0.618 0.860 5f 1.3972 1.4763 −0.0791 31 423.74 0.492 0.807 5g 1.7057 1.4841 0.2216 31 432.96 0.536 0.752 5h 1.7310 1.7260 0.0050 31 462.96 0.756 0.844 5i 1.5159 1.4788 0.0371 33 441.92 0.541 0.928 5k 1.2989 1.3084 −0.0095 32 423.66 0.504 3.675 5l 1.3420 1.3186 0.0234 32 423.66 0.504 3.563 5m 1.3487 1.3758 −0.0271 35 445.34 0.477 0.720 5n 1.4040 1.4366 −0.0326 35 445.34 0.497 0.727 5o 1.3436 1.3587 −0.0151 36 454.48 0.553 2.662 5p 1.4249 1.3990 0.0259 39 476.16 0.545 0.630 5q 1.3856 1.4065 −0.0209 39 476.16 0.547 0.616 5r 1.3577 1.4848 −0.1271 34 444.90 0.528 0.633 5s 1.3564 1.3416 0.0148 43 506.98 0.579 0.296 5t 1.3786 1.3311 0.0475 38 465.86 0.497 0.710 5u 1.8225 1.8264 −0.0039 38 452.50 0.628 1.671 5v 1.4820 1.4312 0.0508 28 396.29 0.405 0.637 5w 1.4498 1.4865 −0.0367 28 417.77 0.523 0.655 ijms-17-01221-t006_Table 6Table 6 External validation of the established QSAR model. Compound Experimental Activity (LogIC50) Predicted Activity (LogIC50) Residual Num_ExplicitAtoms Molecular_SAVol CHI_V_3_C Jurs_RPCS 5e 2.1497 2.1497 0.00 31 423.74 0.473 0.795 5j 1.3504 1.3504 0.00 32 423.66 0.482 3.313 ijms-17-01221-t007_Table 7Table 7 Computer-aided ADME study of the prepared derivatives. Compound AlogP98 a PSA_2D b Solubility c Solubility Level d Absorption Level e CYP2D6 f CYP2D6 Probability g 5a 3.684 43.616 −4.830 2 0 0 0.316 5b 4.171 43.616 −5.324 2 0 1 0.554 5c 2.938 70.156 −4.307 2 0 0 0.376 5d 4.349 43.616 −5.560 2 0 0 0.376 5e 3.890 43.616 −5.158 2 0 0 0.336 5f 3.890 43.616 −5.152 2 0 0 0.376 5g 4.095 43.616 −5.478 2 0 0 0.326 5h 4.433 43.616 −5.636 2 0 0 0.287 5i 3.579 86.440 −5.034 2 0 0 0.257 5j 3.442 64.432 −4.355 2 0 0 0.326 5k 3.442 64.432 −4.363 2 0 0 0.336 5l 3.442 64.432 −4.366 2 0 0 0.386 5m 3.668 52.547 −4.911 2 0 0 0.376 5n 3.668 52.547 −4.887 2 0 0 0.396 5o 3.426 73.362 −4.486 2 0 0 0.425 5p 3.652 61.477 −4.959 2 0 0 0.306 5q 3.652 61.477 −4.949 2 0 1 0.524 5r 3.453 61.477 −5.112 2 0 0 0.346 5s 3.635 70.407 −5.013 2 0 1 0.554 5t 4.017 52.547 −5.102 2 0 0 0.366 5u 3.287 28.624 −4.870 2 0 0 0.306 5v 3.080 56.171 −4.348 2 0 0 0.099 5w 3.410 43.616 −4.643 2 0 0 0.425 a Lipophilicity descriptor; b Polar surface area; c Solubility parameter. (0–−2 = optimal, −2–−4 = good, −4–−6 = low, −6–−8 = very low); d Solubility level. (0 = extremely low, 1 = very low but possible, 2 = low, 3 = good, 4 = optimal); e Absorption level. (0 = good, 1 = moderate, 2 = low, 3 = very low); f CYP2D inhibition. (0 = non inhibitor, 1 = inhibitor); g CYP2D6 Probability: 0–0.5 = non inhibitor; 0.5–1 = inhibitor. ==== Refs References 1. International Agency for Research on Cancer IGlobocan: Estimated Cancer Incidence, Mortality, and Prevalence Worldwide in 2012 IARC Lyon, France 2014 2. Adami H. Hunter D. Trichopoulos D. Textbook of Cancer Epidemiology Oxford University Press New York, NY, USA 2002 301 373 3. Barnard M.E. Boeke C.E. Tamimi R.M. Established breast cancer risk factors and risk of intrinsic tumor subtypes Biochim. Biophys. Acta 2015 1856 73 85 10.1016/j.bbcan.2015.06.002 26071880 4. Atkinson R.L. Yang W.T. Rosen D.G. Landis M.D. Wong H. Lewis M.T. Creighton C.J. Sexton K.R. Hilsenbeck S.G. Sahin A.A. Cancer stem cell markers are enriched in normal tissue adjacent to triple negative breast cancer and inversely correlated with DNA repair deficiency Breast Cancer Res. 2013 15 R77 10.1186/bcr3471 24008095 5. Rahman M. Davis S.R. Pumphrey J.G. Bao J. Nau M.M. Meltzer P.S. Lipkowitz S. TRAIL induces apoptosis in triple-negative breast cancer cells with a mesenchymal phenotype Breast Cancer Res. Treat. 2009 113 217 230 10.1007/s10549-008-9924-5 18266105 6. Rakha E.A. Reis-Filho J.S. Ellis I.O. Basal-like breast cancer: A critical review J. Clin. Oncol. 2008 26 2568 2581 10.1200/JCO.2007.13.1748 18487574 7. Fadare O. Tavassoli F.A. Clinical and pathologic aspects of basal-like breast cancers Nat. Clin. Pract. Oncol. 2008 5 149 159 10.1038/ncponc1038 18212769 8. Smid M. Wang Y. Zhang Y. Sieuwerts A.M. Yu J. Klijn J.G. Foekens J.A. Martens J.W. Subtypes of breast cancer show preferential site of relapse Cancer Res. 2008 68 3108 3114 10.1158/0008-5472.CAN-07-5644 18451135 9. Maire V. Baldeyron C. Richardson M. Tesson B. Vincent-Salomon A. Gravier E. Marty-Prouvost B. De Koning L. Rigaill G. Dumont A. TTK/hMPS1 is an attractive therapeutic target for triple-negative breast cancer PLoS ONE 2013 8 1221 10.1371/journal.pone.0063712 23700430 10. Cleator S. Heller W. Coombes R.C. Triple-negative breast cancer: therapeutic options Lancet Oncol. 2007 8 235 244 10.1016/S1470-2045(07)70074-8 17329194 11. Bansal Y. Silakari O. The therapeutic journey of benzimidazoles: A review Bioorg. Med. Chem. 2012 20 6208 6236 10.1016/j.bmc.2012.09.013 23031649 12. Fei F. Zhou Z. New substituted benzimidazole derivatives: A patent review (2010–2012) Expert Opin. Ther. Pat. 2013 23 1157 1179 10.1517/13543776.2013.800857 23675911 13. Yadav G. Ganguly S. Structure activity relationship (SAR) study of benzimidazole scaffold for different biological activities: A mini-review Eur. J. Med. Chem. 2015 97 419 443 10.1016/j.ejmech.2014.11.053 25479684 14. Gümüş F. Algül Ö. Eren G. Eroğlu H. Diril N. Gür S. Özkul A. Synthesis, cytotoxic activity on MCF-7 cell line and mutagenic activity of platinum (II) complexes with 2-substituted benzimidazole ligands Eur. J. Med. Chem. 2003 38 473 480 10.1016/S0223-5234(03)00058-8 12767597 15. Bielawski K. Wolczynski S. Bielawska A. Inhibition of DNA topoisomerase I and II, and growth inhibition of MDA-MB-231 human breast cancer cells by bis-benzimidazole derivatives with alkylating moiety Pol. J. Pharmacol. 2004 56 373 378 15215569 16. Thimmegowda N.R. Swamy S.N. Kumar C.S.A. Kumar Y.C.S. Chandrappa S. Yip G.W. Rangappa K.S. Synthesis, characterization and evaluation of benzimidazole derivative and its precursors as inhibitors of MDA-MB-231 human breast cancer cell proliferation Bioorg. Med. Chem. Lett. 2008 18 432 435 10.1016/j.bmcl.2007.08.078 17981032 17. Taher A.T. Khalil N.A. Ahmed E.M. Synthesis of novel isatin-thiazoline and isatin-benzimidazole conjugates as anti-breast cancer agents Arch. Pharmacal Res. 2011 34 1615 1621 10.1007/s12272-011-1005-3 22076761 18. Romero-Castro A. León-Rivera I. Ávila-Rojas L.C. Navarrete-Vázquez G. Nieto-Rodríguez A. Synthesis and preliminary evaluation of selected 2-aryl-5(6)-nitro-1H -benzimidazole derivatives as potential anticancer agents Arch. Pharmacal Res. 2011 34 181 189 10.1007/s12272-011-0201-5 21380799 19. Bonham S. O’Donovan L. Carty M.P. Aldabbagh F. First synthesis of an aziridinyl fused pyrrolo[1,2-a ]benzimidazole and toxicity evaluation towards normal and breast cancer cell lines Org. Biomol. Chem. 2011 9 6700 6706 10.1039/c1ob05694h 21808774 20. Rahim A.S.A. Salhimi S.M. Arumugam N. Pin L.C. Yee N.S. Muttiah N.N. Keat W.B. Hamid S.A. Osman H. Mat I.B. Microwave-assisted synthesis of sec /tert -butyl 2-arylbenzimidazoles and their unexpected antiproliferative activity towards ER negative breast cancer cells J. Enzym. Inhib. Med. Chem. 2013 28 1255 1260 10.3109/14756366.2012.729828 23061895 21. Yoon Y.K. Ali M.A. Wei A.C. Choon T.S. Osman H. Parang K. Shirazi A.N. Synthesis and evaluation of novel benzimidazole derivatives as sirtuin inhibitors with antitumor activities Bioorg. Med. Chem. 2014 22 703 710 10.1016/j.bmc.2013.12.029 24387981 22. Abonia R. Cortés E. Insuasty B. Quiroga J. Nogueras M. Cobo J. Synthesis of novel 1,2,5-trisubstituted benzimidazoles as potential antitumor agents Eur. J. Med. Chem. 2011 46 4062 4070 10.1016/j.ejmech.2011.06.006 21719162 23. Kamal A. Reddy T.S. Vishnuvardhan M.V.P.S. Nimbarte V.D. Rao A.V.S. Srinivasulu V. Shankaraiah N. Synthesis of 2-aryl-1,2,4-oxadiazolo-benzimidazoles: Tubulin polymerization inhibitors and apoptosis inducing agents Bioorg. Med. Chem. 2015 23 4608 4623 10.1016/j.bmc.2015.05.060 26169762 24. Kamal A. Shaik A.B. Polepalli S. Kumar G.B. Reddy V.S. Mahesh R. Garimella S. Jain N. Synthesis of arylpyrazole linked benzimidazole conjugates as potential microtubule disruptors Bioorg. Med. Chem. 2015 23 1082 1095 10.1016/j.bmc.2015.01.004 25648686 25. Husain A. Rashid M. Mishra R. Parveen S. Shin D.-S. Kumar D. Benzimidazole bearing oxadiazole and triazolo-thiadiazoles nucleus: Design and synthesis as anticancer agents Bioorg. Med. Chem. Lett. 2012 22 5438 5444 10.1016/j.bmcl.2012.07.038 22840417 26. Rashid M. Husain A. Mishra R. Synthesis of benzimidazoles bearing oxadiazole nucleus as anticancer agents Eur. J. Med. Chem. 2012 54 855 866 10.1016/j.ejmech.2012.04.027 22608854 27. Husain A. Rashid M. Shaharyar M. Siddiqui A.A. Mishra R. Benzimidazole clubbed with triazolo-thiadiazoles and triazolo-thiadiazines: New anticancer agents Eur. J. Med. Chem. 2013 62 785 798 10.1016/j.ejmech.2012.07.011 23333063 28. Abdel-Aziz H.A. Ghabbour H.A. Eldehna W.M. Al-Rashood S.T.A. Al-Rashood K.A. Fun H.-K. Al-Tahhan M. Al-Dhfyan A. 2-((Benzimidazol-2-yl)thio)-1-arylethan-1-ones: Synthesis, crystal study and cancer stem cells CD133 targeting potential Eur. J. Med. Chem. 2015 104 1 10 10.1016/j.ejmech.2015.09.023 26413725 29. Aboul-Fadl T. Kadi A. Abdel-Aziz H.A. Novel N,N ′-Hydrazino-bis-isatin Derivatives with Selective Activity Against Multidrug-Resistant Cancer Cells U.S. Patent 20120252860 4 10 2012 30. Abdel-Aziz H.A. Ghabbour H.A. Eldehna W.M. Qabeel M.M. Fun H.-K. Synthesis, crystal structure and biological activity of cis /trans amide rotomers of (Z )-N ′-(2-oxoindolin-3-ylidene)formohydrazide J. Chem. 2014 2014 760434 10.1155/2014/760434 31. Fares M. Eldehna W.M. Abou-Seri S.M. Abdel-Aziz H.A. Aly M.H. Tolba M.F. Design, synthesis and in vitro antiproliferative activity of novel isatin-quinazoline hybrids Arch. Pharm. Chem. Life Sci. 2015 348 144 154 10.1002/ardp.201400337 25664631 32. Eldehna W.M. Ibrahim H.S. Abdel-Aziz H.A. Farrag N.N. Youssef M.M. Design, synthesis and in vitro antitumor activity of novel N -substituted-4-phenyl/benzylphthalazin-1-ones Eur. J. Med. Chem. 2015 89 549 560 10.1016/j.ejmech.2014.10.064 25462265 33. Eldehna W.M. Altoukhy A. Mahrous H. Abdel-Aziz H.A. Design, synthesis and QSAR study of certain isatin-pyridine hybrids as potential anti-proliferative agents Eur. J. Med. Chem. 2015 90 684 694 10.1016/j.ejmech.2014.12.010 25499988 34. Alafeefy A.M. Ahmad R. Abdulla M. Eldehna W.M. Al-Tamimi A.S. Abdel-Aziz H.A. Al-Obaid O. Carta F. Al-Kahtani A.A. Supuran C.T. Development of certain new 2-substituted-quinazolin-4-yl-aminobenzenesulfonamide as potential antitumor agents Eur. J. Med. Chem. 2016 109 247 253 10.1016/j.ejmech.2016.01.001 26774930 35. Eldehna W.M. Fares M. Ibrahim H.S. Aly M.H. Zada S. Ali M.M. Abou-Seri S.M. Abdel-Aziz H.A. Abou El Ella D.A. Indoline ureas as potential anti-hepatocellular carcinoma agents targeting VEGFR-2: Synthesis, in vitro biological evaluation and molecular docking Eur. J. Med. Chem. 2015 100 89 97 10.1016/j.ejmech.2015.05.040 26071861 36. Sheldrick G.M. A short history of SHELX Acta Crystallogr. Sect. A 2008 64 112 122 10.1107/S0108767307043930 18156677 37. Ngamwongsatit P. Banada P.P. Panbangred W. Bhunia A.K. WST-1-based cell cytotoxicity assay as a substitute for MTT-based assay for rapid detection of toxigenic Bacillus species using CHO cell line J. Microbiol. Methods 2008 73 211 215 10.1016/j.mimet.2008.03.002 18417231 38. Massagué J. G1 cell-cycle control and cancer Nature 2004 432 298 306 10.1038/nature03094 15549091 39. Mavrova A.T. Denkova P. Tsenov Y.A. Anichina K.K. Vutchev D.I. Synthesis and antitrichinellosis activity of some bis(benzimidazol-2-yl)amines Bioorg. Med. Chem. 2007 15 6291 6297 10.1016/j.bmc.2007.06.017 17600722 40. Loghmani-Khouzani H. Hajiheidari D. Synthesis of difluorinated β-ketosulfones and novel gem-difluoromethylsulfone-containing heterocycles as fluorinated building blocks J. Fluor. Chem. 2010 131 561 569 10.1016/j.jfluchem.2009.12.022 41. Turan-Zitouni G. Demirayak S. Erol K. Ozdemir M. Synthesis of some 2-[(benzazole-2-yl)thoiacetyl]phenol derivatives and preliminary investigation on their vasodilatory activity Farmaco 1994 40 755 757 7832978 42. Madsen P. Knudsen L.B. Wiberg F.C. Carr R.D. Discovery and structure-activity relationship of the first non-peptide competitive human glucagon receptor antagonists J. Med. Chem. 1998 41 5150 5157 10.1021/jm9810304 9857085

PMC005xxxxxx/PMC5000620.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081222ijms-17-01222ArticleMinocycline Loaded Hybrid Composites Nanoparticles for Mesenchymal Stem Cells Differentiation into Osteogenesis Tham Allister Yingwei 1†Gandhimathi Chinnasamy 1†Praveena Jayaraman 1Venugopal Jayarama Reddy 2Ramakrishna Seeram 2Kumar Srinivasan Dinesh 1*Hardy John G. Academic Editor1 Cellular and Molecular Epigenetics Lab, Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore 636921, Singapore; [email protected] (A.Y.T.); [email protected] (C.G.); [email protected] (J.P.)2 Center for Nanofibers and Nanotechnology, Department of Mechanical Engineering, National University of Singapore, Singapore 119260, Singapore; [email protected] (J.R.V.); [email protected] (S.R.)* Correspondence: [email protected]; Tel.: +65-6592-3055; Fax: +65-6515-0417† These authors contributed equally to this work. 28 7 2016 8 2016 17 8 122225 5 2016 15 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Bone transplants are used to treat fractures and increase new tissue development in bone tissue engineering. Grafting of massive implantations showing slow curing rate and results in cell death for poor vascularization. The potentials of biocomposite scaffolds to mimic extracellular matrix (ECM) and including new biomaterials could produce a better substitute for new bone tissue formation. A purpose of this study is to analyze polycaprolactone/silk fibroin/hyaluronic acid/minocycline hydrochloride (PCL/SF/HA/MH) nanoparticles initiate human mesenchymal stem cells (MSCs) proliferation and differentiation into osteogenesis. Electrospraying technique was used to develop PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH hybrid biocomposite nanoparticles and characterization was analyzed by field emission scanning electron microscope (FESEM), contact angle and Fourier transform infrared spectroscopy (FT-IR). The obtained results proved that the particle diameter and water contact angle obtained around 0.54 ± 0.12 to 3.2 ± 0.18 µm and 43.93 ± 10.8° to 133.1 ± 12.4° respectively. The cell proliferation and cell-nanoparticle interactions analyzed using (3-(4,5-dimethyl thiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt) MTS assay (Promega, Madison, WI, USA), FESEM for cell morphology and 5-Chloromethylfluorescein diacetate (CMFDA) dye for imaging live cells. Osteogenic differentiation was proved by expression of osteocalcin, alkaline phosphatase activity (ALP) and mineralization was confirmed by using alizarin red (ARS). The quantity of cells was considerably increased in PCL/SF/HA/MH nanoparticles when compare to all other biocomposite nanoparticles and the cell interaction was observed more on PCL/SF/HA/MH nanoparticles. The electrosprayed PCL/SF/HA/MH biocomposite nanoparticle significantly initiated increased cell proliferation, osteogenic differentiation and mineralization, which provide huge potential for bone tissue engineering. electrosprayingpolycaprolactonesilk fibroinhyaluronic acidminocycline hydrochloridenanoparticles ==== Body 1. Introduction The major obstacle faced by clinicians in orthopaedic and bone reconstructive surgery includes therapy for bone defects and fractures caused by tumour formation, trauma or diseases [1,2]. Autogenic and allogeneic bone implants are presently being used to repair fractures and improve bone growth in bone tissue engineering. However, grafting of massive implants has an inherent issue of slow curative level, cartilage development under poor vascularization and even causes apoptosis [3]. Biomaterials play an important role by providing appropriate substrates for cell growth and differentiation into the bone defect in addition to the structural and functional support for new tissue formation. The regeneration of bone deficiencies has reached some achievement when using injectable pastes and various scaffolding materials, there is a considerable space for development of new tissues in bone tissue engineering [4,5,6]. The potential benefits of nanoparticles are their average high surface to volume ratios, allowing better solubility thereby increasing bioactivity. The bioactive mesoporous nanoparticles will support cellular growth and bone repair, and also suitable for constructing macroporous devices to be useful in bone tissue regeneration [7]. Electrospray technique is one of the best methods to fabricate nano/microparticles. The basic principle in electrospraying technique is to set a high voltage to the biocomposite polymeric solution to flow out from the syringe in the form of particles. Advancement in electrospraying technology produced nanoparticles with more surface area-to-volume ratio and also physical and chemical properties more or less similar to the extracellular matrix (ECM) of natural bone. Human mesenchymal stem cells (MSCs) are multipotent cell source used for various biomedical applications, since they can be obtained from different sources such as bone marrow and adipose tissues [8,9]. MSCs have a multi-lineage differentiation potential leads to various cell types including osteoblasts, neuron-like cells, chondrocytes or fibroblasts [10,11]. Polycaprolactone (PCL) is a biodegradable polymer, generally used as an implantable material as it is easily degraded by hydrolysis of the ester linkages under biological environments [12]. Silk proteins are favourable materials for drug delivery and tissue engineering, due to their biocompatibility and biodegradability [13]. Silk fibroin (SF) protein is a Food and Drug Administration (FDA) approved naturally derived macromolecular protein that has been used to generate clinical sutures for several years [14]. Due to its tremendous tensile strength and biocompatibility, SF has been significantly used as a biomaterial in bone tissue engineering [15]. SF based tissue engineered bone transplants developed in bioreactors and fixed into calvarial deficiencies in mice revealed the ability to encourage bone formation within 30 days [16]. Consequently, the porous SF scaffolds supports the osteogenic-differentiation of MSCs, and such hMSCs cultured membranes have remarkably healed the femoral segmental defects in nude rats [17]. Hyaluronic acid (HA) is a naturally derived glycosaminoglycan used in biomedical applications such as post-surgical adhesion prevention and hydrophilic coatings. Addition of HA helps to increase cell proliferation and differentiation due to an increase in hydrophilicity and bone formation ability [18]. The combination of antibacterial agents with biomaterials is essential for repairing bone defects. Minocycline is a semi-synthetic, broad-spectrum bacteriostatic antibiotic, is energetic against aerobic, anaerobic, gram-positive and gram-negative bacteria, it possesses anti-collagenase activity, prevents bone infection and is able to promote proliferation and can improve bone growth, decrease connective tissue cessation and reduce bone resorption [19,20]. The present study, PCL was combined with SF, HA and Minocycline hydrochloride (MH), electrosprayed to obtain PCL/SF/HA/MH hybrid nanoparticles which has proved to be a promising biocomposite for MSCs proliferation, differentiation and mineralization holding great potential for bone tissue engineering. 2. Results and Discussion 2.1. Characterization of Nanoparticles The topography of nanoparticles plays an important role in the regulation of primary cell activities such as attachment and proliferation of cells in tissue engineering [21]. Field emission scanning electron microscope (FESEM) micrographs of nanoparticle showed nano-scaled and uniformed particles prepared under controlled conditions (Figure 1). The particle diameters were obtained in the range of 0.54 ± 0.12 to 3.2 ± 0.18 µm (Table 1). Contact angle measurement revealed that the addition of SF to PCL became hydrophilic because hydroxyl group in SF has the ability to form H bonds with H2O molecule for interpreting a significant increase in hydrophilicity as seen in PCL/SF with a contact angle of 75.4 ± 9.45° as compared to 133.1 ± 12.4° in PCL nanoparticles (Figure 2). Contact angle value of PCL/SF/HA/MH particles was significantly (p ≤ 0.001) reduced compared to pristine PCL nanoparticles. Upon addition of HA biomolecules, MH changes further hydrophilic with a contact angle of 64.42 ± 13.4 and 43.93 ± 30.8° for PCL/SF/HA and PCL/SF/HA/MH particles. The water absorbance ratio is directly proportional to the hydrophilic properties of the nanofibrous scaffolds which support prevention of dryness and the build-up of exudates on wounds [22]. The functional groups analyses of composite particles were studied using Fourier transform infrared spectroscopy (FT-IR) as presented in Figure 3. The peak characteristic of C–O–C symmetric stretching, C=O ester stretching, asymmetric and symmetric C–H alkane stretching on pristine PCL was observed at 1170, 1740, 2870 and 2950 cm−1 on the PCL. Furthermore, the characteristic peaks of amide I, II, and III of SF were also detected on the PCL/SF particles at 1654, 1530 and 1255 cm−1. The specific peak of hydrogen-bonded OH and NH stretching vibrations, asymmetric C=O and symmetric C–O stretching vibration of carboxyl groups in hyaluronic acid was noticed on PCL/SF/HA particle at 3420, 1617 and 1410 cm−1. The specific peak stretching vibration of amine, carboxyl group (C=O) and C–O stretching vibration of MH was detected at 3040, 1660 and 870 cm−1 on PCL/SF/HA/MH particles. 2.2. Particles Degradation and Minocycline Hydrochloride (MH) Release The degradation properties play an important role in biomolecule selection and strategy in tissue engineering [23,24,25,26]. Thus, the biocomposite scaffolds must meet definite model and suitable principles, with biocompatibility, tensile strength in particular cases of biomaterials. The ability of polymer biocomposites in a plethora of applications requires the optimal selection of matrix polymer chemistry, and matrix filler interaction for the degradation of materials [27]. The degradation of pristine PCL results from the hydroxylation and separation of excessive molecular mass chains, followed by altering carbon dioxide and H2O in the atmosphere of water. The structural variations of PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH particles on day 15, 30 and 45 are presented in Figure 4. After 15 days of degradation, for the PCL/SF, PCL/SF/HA, and PCL/SF/HA/MH particles and pore constructions remained similar and a small part of particles was scratch. The PCL/SF, PCL/SF/HA and PCL/SF/HA/MH nanoparticles were broken and finally, degenerate into small particles after day 30 and 45. This is caused by the hydrolysis and diffusion of small polymers from the samples, which effect in a disconnected construction and the diffusion of H2O molecule into the samples at initial degradation. Over a period of time the lengthy chains of polymers hydrolyzes and degrades the particles to breakdown into small parts. The weight loss was initiated by the degradation progress, which is evident that oligomeric substance able to dissolved in the medium from the polymer surface by the polymer chain hydrolysis. Degradation for 10–40 days of pristine PCL particles was 3.8% and 10.88% and the rate of degradation was much slower than PCL/SF particles (Figure 5). After day 30 and 40, addition of SF, the degradation rate PCL/SF, PCL/SF/HA and PCL/SF/HA/MH enriched and the weight loss was 10.65% and 13.33%, and 12.87% and 15.74%, and 14.8% and 17.65% respectively. Pristine PCL shows reduced rate of degradation then PCL/SF, PCL/SF/HA and PCL/SF/HA/MH particles. The crystallization of pristine PCL is greater as compared to electrosprayed PCL/SF, PCL/SF/HA and PCL/SF/HA/MH particles. The even drug release profile of MH loaded biocomposite particles is shown in Figure 6. MH reveals λmax at 260 nm, the drug release profile of MH were measured by drug-eluting membrane compositions of 5%, and 10% from the particles. Initial 100 h, burst releases were noticed about 10%–15%. A continuing sustained release was noticed within 150 and 500 h. The lower drug concentrations (0.04% and 0.07%), decreased release of drug was detected by the end of 30th day. The highest release ratio of MH was 5% and 10% w/w MH-loaded particles were around 52% ± 3% and 69% ± 5% at the end of 30th day. The progress of the experimental outcomes is the decrease of early burst release and the slow release due to the balance of drug inside the biocomposites structure. The initial burst released due to the influence of small amount of MH adsorbed on the sample surface. 2.3. Cell Morphology Scaffold properties are important determinants in regulating cell growth, morphology and cell signaling. Scaffold-cell interaction regulates early stage cellular activities namely cell adhesion and proliferation, which consequently influences differentiation and mineralization [28]. Cell culture after 5th, 10th and 15th days, the proliferation of MSCs was measured by (3-(4,5-dimethyl thiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt) (MTS) assay (Figure 7). The rate of proliferation on PCL/SF/HA/MH samples was significantly increased (p ≤ 0.001) than PCL nanoparticles. The obtained results proved that the chemical alteration of the nanoparticle composition by adding SF/HA increased cell infiltration and proliferation. Hydroxyl groups and amino groups present in these samples served as ligands, which stimulated the differentiation of MSCs into osteoblasts. However, the rate of proliferation was higher on PCL/SF/HA/MH compared to all other particles after 15 days. This can be identified owing to the presence of HA, which gives better surface roughness and more surface range for greater cell attachment, proliferation and differentiation. FESEM image of MSCs (Figure 8) showed that the standard cell structure on all the samples with more mineral growth on the PCL/SF/HA/MH particles. Cellular function such as adhesion and proliferation showing the preliminary stages of cell-biomaterial interaction, finally leads to differentiation and mineralization [29]. Furthermore, the gradual adhesion of cells in to the nanoparticles was observed and experienced enhanced cell-to-cell interaction with the formation of filapodia structure. Cells cultured on to the nanoparticles revealed that filapodial interactions were capable of nano-topographical features and their cytoskeletal growth was essential for cell proliferation and differentiation into osteoblasts. The observed results proved that structure of the cells was comparatively uniform in all nanoparticles. However, cells on PCL/SF/HA/MH showed better osteoblast morphology and mineralization. This is because SF and MH initiate proliferation of cells and hyaluronic acid acts as a chelating agent, which results in the mineralization of osteoblasts required for bone tissue engineering. The live cell images in composite particles were determined by cell morphology and 5-Chloromethylfluorescein diacetate (CMFDA) dye expression method as shows in Figure 9. The normal cell morphology was observed in PCL/SF and PCL/SF/HA particles to prove the good environment for cell growth and mineralization. The MSCs, which underwent osteogenic differentiation, showed cuboidal morphology on PCL/SF/HA/MH nanoparticles unlike all the other samples. 2.4. Mineralization An ideal composite particle should stimulate cell growth including both organic and inorganic constituents of natural bone materials. Alkaline phosphatase is an essential factor for bone matrix vesicles as it is involved in the production of apatite calcium phosphate and also a strong indicator of undeveloped osteoblast activity [30,31]. On day 15th, alkaline phosphatase activity (ALP) activity was significantly (p ≤ 0.001) increased in PCL/SF/HA/MH as compared to all other samples (Figure 10). This is because HA stimulates osteogenic differentiation of MSCs into new bone cells [32,33]. Alizarin red (ARS) staining showed a significant (p ≤ 0.001) increase of mineral deposition in PCL/SF/HA and PCL/SF/HA/MH compared to PCL on days 15 (Figure 11). Increased mineralization observed on day 10 and 15 of cell culture in PCL/SF/HA (22%, 46%), PCL/SF/HA/MH (26%, 49%), compared to PCL (10%, 20%). Qualitative analysis of mineralization was measured by alizarin red staining as shown in Figure 12. After 15th day of cells culture, PCL/SF/HA/MH (Figure 12e) interaction with cells showing more mineral deposition compared to PCL particles. Polylactic-co-glycolic acid attached HA/polyethylene glycol (PEG) scaffolds has effectively distributed Bone morphogenetic protein (BMP)-2 in vivo with gradual release from the scaffolds for up to a month. The developed tissue engineered porous scaffolds, which was later implanted in a confluent mouse osteoblastic cells (MC3T3-E1) sheet that provided desired cell growth for bone tissue engineering [34]. Presence of mineral deposits is a main indicator for matured osteoblasts that is useful for validating MSCs. Cell cultured on nanoparticles differentiated and reached mineralization level, where mineralized ECM is deposited on the particles. ALP activity indicates the presence of osteocalcin (OCN) expression during differentiation, because ALP positively regulates the production of ECM for mineral deposition [35,36]. Mineralization of MSCs is essential for the development of new bone and the observed results indicated that the addition of HA and MH enhanced mineralization, thereby improving mineral deposition for bone tissue engineering. 2.5. Expression of Osteocalcin Osteocalcin (OCN) plays an important role in regulating the mineralization, because it contains glutamic acid-rich regions which bind strongly to Ca2+ [37]. Figure 13f–j showed that the expression of green colour which indicates cluster of differentiation (CD90) (MSC specific marker protein), then the cells differentiates into osteoblasts to express OCN protein Figure 13k–o. Cell nucleus were stained with 4′,6-diamidino-2-phenylindole (DAPI) indicate blue colour in Figure 13a–e. The obtained results showed that the osteogenic differentiation of MSCs by the dual expression of both CD90 and OCN in Figure 13d–t. The results showed that the MSCs seeded on PCL/SF/HA/MH exhibited the distinct cuboidal morphology seen in osteoblasts and increased levels of OCN expression indicate extensive differentiation into osteoblasts of more MSCs as compared to all other nanoparticles. The obtained results proved that the MSCs underwent osteogenic differentiation and mineralization, by the stimulation of HA in the biocomposite PCL/SF/HA/MH particles for bone tissue regeneration. 3. Materials and Methods 3.1. Materials Polycaprolactone (PCL), hyaluronic acid, (HA) minocycline hydrochloride (MH), 1,1,1,3,3,3-hexafluoro-isopropanol (HFIP), methanol, Alizarin red-S, cetylpyridinium chloride were purchased from Sigma-Aldrich. Silk fibroin (SF) was purchased from Zhang Peng International Trading, Singapore. Dulbecco’s modified eagle’s medium (DMEM), nutrient mixture F-12, fetal bovine serum (FBS), antibiotics and trypsin-ethylene diamine tetra acetic acid (EDTA) were procured from GIBCO (Invitrogen, Carlsbad, CA, USA). CellTiter 96® Aqueous one solution was obtained from Promega, Madison, WI, USA. 3.2. Fabrication of Nanoparticle The electrospraying technique requires the optimization of several factors comprising high voltage power supply, distance between the needle tip and then collector plate, concentration of solutions and solution flow speed to make fine nanoparticles. Solutions of PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH were prepared for using electrospraying. 10% (w/v) solution of pristine PCL was prepared by using HFIP as a solvent. 10% (w/v) solution of PCL and SF (9:1) were prepared by using HFIP as a solvent. PCL/SF/HA and PCL/SF/HA/MH solutions were also prepared at the ratio of 8:1:1 and 7.5:1:1:0.5 in HFIP at the same concentration of 10%. For fabrication, solutions of PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH were separately fed into a 10 mL syringe connected to 22G × 1½ blunt needle and the needle was connected to a syringe pump at 1 mL/h flow rate with the high voltage of 16.0 to 16.5 kV. Under the influence of power supply the biocomposite solution was drained into particles, received on 15-mm cover slips, spread on a stainless steel plate enclosed with aluminium foil in 15 cm distance from the tip of the needle and later used for cell-culturing. The electrosprayed nanoparticles were dehydrated in vacuum pressure to remove any residual solvents. 3.3. Nanoparticles Characterization Surface morphology of the nanoparticle was observed under field emission scanning electron microscope (FESEM, FEI-QUANTA 200F, Roche, Woerden, Netherlands) at an accelerating voltage of 10 kV, later the nanoparticles were sputter coated with platinum (JEOL JFC-1200 Auto fine coater, JEOL LTD, Tokyo, Japan). Each hybrid composite particles, 5 samples were randomly selected to measure the particle size using ImageJ software (National Institutes of Health, Dune Sciences, Inc. Eugene, OR, USA). The hydrophilicity of the electrosprayed nanoparticle was determined by sessile drop water contact angle analysis using video contact angle optima surface study system (AST Products, Billerica, MA, USA). Functional groups of electrosprayed nanoparticle were analyzed using Avatar 380 FTIR spectrometer (Thermo Nicolet, Waltham, MA, USA) to determine the presence of functional groups. 3.4. Degradation and Drug Release Degradation studies of nanoparticles were conducted by in vitro method. The particles were weighed and kept in 15 mL of phosphate buffer solution (PBS) in an incubator at 37 °C. The PBS solution in the particle samples was changed once in two days. The degraded particles were washed carefully with H2O and then dehydrated through vacuum and their weights were measured. Weight loss ratio was measured by using the following formula. Weight loss %=W0-WdW0 × 100 (%) W0 is the actual weight before degradation; Wd is the dry weight after degradation. Structural modifications of particles before and after degradation were studied with FESEM at an accelerated voltage of 10 kV. The minocycline hydrochloride (MH) release profile of MH-loaded PCL/SF/HA particle was analysed using PBS, the MH-loaded PCL/SF/HA particle (50 mg) were kept in a centrifuge tube then added 10 mL PBS as the release medium. For further study, the centrifuge tube was then kept in the incubator at 37 °C. At specific time points, 5 mL of solution was taken from the release medium and that volume was replaced with 5 mL PBS. Subsequently, the quantity of MH released at various time intervals up to 30 days was analysed using UV-visible spectrophotometer at 260 nm. Using the calibration curve of MH measured in the same condition, MH release percentage was determined and plotted as the curve versus time according to this equation: Release (%)=Released MHTotal loaded MH × 100 (%) 3.5. Cell Culture Human mesenchymal stem cells (MSCs) (Lonza, Singapore) were cultured in DMEM/F12 (1:1) imbued with FBS in a 75 cm2 flask. MSCs were incubated at 37 °C in 5% CO2 for 7 days and every alternate day changed the medium. The confluences of cells were detached by adding trypsin with EDTA. Separated cells remained centrifuged and counted by Trypan blue assay using the hemocytometer. The coverslips were kept in well plates with stainless steel ring in each well to prevent swelling of the samples. The nanoparticles were then sterilized under UV light for 3–4 h, then dried with ethanol and then rinsed with PBS to eliminate remaining solvents. They were subsequently dipped in medium overnight incubated at 37 °C before cell culture. Cells were seeded on PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH at a concentration of 8 × 103 cells per well. 3.6. Cell Proliferation Cell proliferation was measured by using the colorimetric MTS assay (3-(4,5-dimethyl thiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt) Cell Titer 96® Aqueous One solution(Promega, Madison, WI, USA). The principle behind the estimation is that the mitochondria of metabolically energetic cells secretes dehydrogenase enzymes is responsible for the reduction of yellow tetrazolium salt in MTS to form purple colour formazan crystals. The colour exposed absorbance at 492 nm and the quantity of formazan formed is directly proportional to the amount of cells. Later the desired seeding periods of 5, 10 and 15 days, removed media from well plates and the samples were washed with PBS to remove any unattached cells followed by development with MTS mixture for 3 h at 37 °C in 5% CO2 incubator. Later, aliquots into 96 well plate and the solutions were analysed in a micro plate reader at 490nm (FLUOstar OPTIMA, BMG Lab Technologies Microplate Readers company, Ortenberg, Germany). 3.7. Cell Morphology Cell morphology was examined after 15 days of MSCs culturing on to the samples using FESEM, removed the media and rinsed the substrate with PBS followed by 3% glutaraldehyde for 3 h to fix the cells. The samples were then washed with H2O and dried with ethanol in a series of concentration increases from 30%, 50%, 70%, 90% and 100% (v/v) sequentially for 15 min each. Lastly, the particles were dehydrated with hexamethyldisilazane and dried overnight. Then the samples were treated with platinum and cell morphology was analyzed using FESEM at an accelerating voltage of 10 kV. CMFDA fluorescent dye was used to observe the live cell morphology. 3.8. Alkaline Phosphatase Activity Alkaline phosphatase activity (ALP) was measured by using alkaline phosphate yellow liquid substrate. Principle behind this reaction, ALP catalyzes the hydrolysis of colourless organic phosphate ester substrate, P-nitro-phenyl phosphate (pNPP) to yellow product p-nitrophenol and phosphate. After 5, 10 and 15 days of cell culture washed the samples with PBS followed by addition of 400 mL pNPP and then incubated for 30 min until the solution colour turns yellow. Stopped the reaction in addition of sodium hydroxide after which the solution was aliquot in 96-well plate and read in micro plate reader at 405 nm. 3.9. Mineralization Alizarin red-S (ARS) dye binds specifically to calcium salts which are employed to measure and quantity of mineralization in differentiated osteoblasts. After 15 days of cell culture, washed the samples with PBS and fixed with chilled ethanol (70%) and washed with deionized water (DI) water and stained with ARS solution for 30 min. After three times rings with DI water, the samples were detected under the inverted optical microscope and images of the stain showing calcium deposition in Leica FW 4000 (version 1.0.2) (Leica Microsystems Imaging Solutions Ltd., Cambridge, UK). The stain was eluted with 10% cetyl pyridinium chloride for 60 min and the absorbance was measured at 540 nm in a spectrophotometer (Thermo Spectronic, Waltham, MA, USA). 3.10. Osteocalcin Expression The differentiation of MSCs into osteoblasts was analyzed by using immunofluorescent staining of both MSC marker protein CD90 and osteoblast marker protein osteocalcin (OCN). After 15 days of culture, the samples were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. Non-specific binding spots were blocked with 3% BSA for 90 min. The biocomposite particles were treated with osteocalcin in the dilution of 1:100 for 90 min at 25 °C. In additional Alexa Fluor 594 secondary antibody was added in the dilution of 1:250 for 60 min. Then the samples were rinsed thrice with PBS to eliminate the additional staining and incubated using DAPI (1:3000) for 30 min. Then the samples were detached and fixed above the glass slide using Vectashield mounting medium (Vector lab, EON biotech PTE. Ltd., Singapore) and observed using fluorescent microscope (Olympus FV1000- Olympus Singapore PTE. Ltd., Singapore). 3.11. Statistical Analysis Experiments were carried out in triplicates and the data shown were presented as mean ± standard deviation. Statistical differences were analyzed using student t-test and significance was considered at p ≤ 0.05. 4. Conclusions The optimal combination of structural, biological and chemical properties of a bone implant plays a potential role in cell attachment and formation of new bone in bone tissue regeneration therapy. Nano-biomolecules can stimulate good cell adhesion, production of ECM and mineralization. PCL/SF/HA/MH nanoparticles fabricated by electrospraying method and MSCs grown on them showed greater cell proliferation and increased ALP activity as compared to all other particles. Osteogenesis was also induced on these biocomposite particles with maximum osteocalcin expression and subsequent mineralization. The electrosprayed PCL/SF/HA/MH hybrid nanoparticles harness abundant potential for cell attachment, proliferation, differentiation and subsequent mineralization of MSCs for bone tissue engineering. Acknowledgments This work was supported by research grants from the Lee Kong Chian School of Medicine, Nanyang Technological University and Singapore Ministry of Education Academic Research Fund Tier 1 (awarded to Srinivasan Dinesh Kumar, who is the corresponding author of this paper). Author Contributions Chinnasamy Gandhimathi and Srinivasan Dinesh Kumar conceived and designed the experiments; Allister Yingwei Tham and Jayaraman Praveena performed the experiments, Chinnasamy Gandhimathi and Jayarama Reddy Venugopal analyzed the data; Srinivasan Dinesh Kumar and Seeram Ramakrishna contributed reagents/materials/analysis tools; Chinnasamy Gandhimathi and Allister Yingwei Tham wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Field emission scanning electron microscope (FESEM) images of the electrosprayed micro/nanoparticles (a) polycaprolactone (PCL); (b) Polycaprolactone/silk fibroin (PCL/SF); (c) Polycaprolactone/silk fibroin/hyaluronic acid (PCL/SF/HA) and (d) Polycaprolactone/silk fibroin/hyaluronic acid/minocycline hydrochloride (PCL/SF/HA/MH) (0.54 ± 0.12 to 3.2 ± 0.18 µm). (Scale bar: 10 µm). Figure 2 Water contact angle values of composite micro/nanoparticles. Figure 3 Fourier transforms infrared spectroscopy (FT-IR) study of (a) Polycaprolactone (PCL); (b) Polycaprolactone/silk fibroin (PCL/SF); (c) Polycaprolactone/silk fibroin/hyaluronic acid (PCL/SF/HA) and (d) Polycaprolactone/silk fibroin/hyaluronic acid/minocycline hydrochloride (PCL/SF/HA/MH) micro/nanoparticles. Figure 4 FESEM images show the degradation studies of biocomposite particles at various time points (Scale bar: 10 µm). Figure 5 Biocomposite particles dry weight changes after degradation at various time points. Figure 6 Drug release profile of minocycline hydrochloride in biocomposite nano/micro particles. Figure 7 Proliferation analysis of mesenchymal stem cells (MSCs) on biocomposite particles on day 5, 10 and 15. * p ≤ 0.05, ** p ≤ 0.001. Figure 8 FESEM images showing the cell-biomaterial interactions on (a) TCP, (b) PCL (c) PCL/SF (d) PCL/SF/HA (e) PCL/SF/HA/MH micro/nanoparticles (Scale bar: 10 µm). Figure 9 Cell morphology and 5-Chloromethylfluorescein diacetate CMFDA dye extrusion image to analyses the cell morphology on (a) TCP, (b) PCL (c) PCL/SF (d) PCL/SF/HA (e) PCL/SF/HA/MH micro/nanoparticles at 10× magnifications. The green color fluorescence indicate live cell morphology. Figure 10 Alkaline phosphatase activity showing the osteogenic differentiation of MSCs on TCP, PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH micro/nanoparticles on day 5, 10 and 15. * p ≤ 0.05, ** p ≤ 0.001. Figure 11 Quantification of mineral deposition in MSCs differentiated into osteogenesis. Alizarin Red-S staining on TCP, PCL, PCL/SF, PCL/SF/HA and PCL/SF/HA/MH micro/nanoparticles. * p ≤ 0.05, ** p ≤ 0.001. Figure 12 Optical microscopic images showing the secretion of minerals by osteogenic differentiation of MSCs. Alizarin red staining on day 15 in TCP (a), PCL (b), PCL/SF (c), PCL/SF/HA (d) and PCL/SF/HA/MH (e), (10× magnification). Figure 13 Confocal microscopy images to confirm the MSCs differentiation into osteogenesis. Nuclear staining DAPI (a–e) blue, MSCs specific marker protein CD90 (f–j) green, and osteoblasts specific marker protein osteocalcin (k–o) red. Merged images showing the dual expression of both CD90 and osteocalcin, characteristic of MSCs cells which have undergone osteogenic differentiation (p–t) on TCP (a,f,k,p), PCL (b,g,l,q), PCL/SF (c,h,m,r), PCL/SF/HA (d,i,n,s) and PCL/SF/HA/MH (e,j,o,t) with the nuclear staining by DAPI (a–e) at 20× magnification (Scale bar: 50 µm). ijms-17-01222-t001_Table 1Table 1 Particle size measurement of biocomposite nanoparticles. Composite Particles Particle Size (µm) Polycaprolactone (PCL) 3.2 ± 0.18 Polycaprolactone/silk fibroin (PCL/SF) 1.62 ± 0.59 Polycaprolactone/silk fibroin/hyaluronic acid (PLC/SF/HA) 0.9 ± 0.15 Polycaprolactone/silk fibroin/hyaluronic acid/minocycline hydrochloride (PCL/SF/HA/MH) 0.54 ± 0.12 ==== Refs References 1. Hench L.L. Thompson I. Twenty-first century challenges for biomaterials J. R. Soc. Interface 2010 7 S379 S391 10.1098/rsif.2010.0151.focus 20484227 2. Schubert T. Xhema D. Vériter S. Schubert M. Behets C. Delloye C. The enhanced performance of bone allografts using osteogenic-differentiated adipose-derived mesenchymal stem cells Biomaterials 2011 32 8880 8891 10.1016/j.biomaterials.2011.08.009 21872925 3. Behr B. Tang C. Germann G. Longaker M.T. Quarto N. Locally applied vascular endothelial growth factor A increases the osteogenic healing capacity of human adipose-derived stem cells by promoting osteogenic and endothelial differentiation Stem Cells 2011 29 286 296 10.1002/stem.581 21732486 4. Tien Y.C. Chih T.T. Lin J.H. Ju C.P. Lin S.D. Augmentation of tendon-bone healing by the use of calcium-phosphate cement J. Bone Jt. Surg. Br. 2004 86 1072 1076 10.1302/0301-620X.86B7.14578 5. Huangfu X. Zhao J. Tendon-bone healing enhancement using injectable tricalcium phosphate in a dog anterior cruciate ligament reconstruction model Arthroscopy 2007 23 455 462 10.1016/j.arthro.2006.12.031 17478274 6. Dvir T. Timko B.P. Kohane D.S. Langer R. Nanotechnological strategies for engineering complex tissues Nat. Nanotechnol. 2011 6 13 22 10.1038/nnano.2010.246 21151110 7. Wu C. Zhou Y. Fan W. Hypoxia-mimicking mesoporous bioactive glass scaffolds with controllable cobalt ion release for bone tissue engineering Biomaterials 2012 33 2076 2085 10.1016/j.biomaterials.2011.11.042 22177618 8. Erices A. Conget P. Minguell J.J. Mesenchymal progenitor cells in human umbilical cord blood Br. J. Haematol. 2000 109 235 242 10.1046/j.1365-2141.2000.01986.x 10848804 9. Gronthos S. Franklin D.M. Leddy H.A. Robey P.G. Storms R.W. Gimble J.M. Surface protein characterization of human adipose tissue-derived stromal cells J. Cell. Physiol. 2001 189 54 63 10.1002/jcp.1138 11573204 10. Marion N.W. Mao J.J. Mesenchymal stem cells and tissue engineering Methods Enzymol. 2006 420 339 361 17161705 11. Priscilla C. Aveline J. Bourseguin I. Rochefort Y. Mesenchymal stem cells and acquisition of a bone phenotype: An ion channel overview J. Dev. Biol. Tissue Eng. 2011 3 103 118 12. Lemmouchi Y. Schatch E. Preparation and in vitro evaluation of biodegradable poly (ε-caprolactone-co-d,l lactide) (X–Y) devises containing tryparocidal drugs J. Control. Release 1997 45 227 233 10.1016/S0168-3659(96)01569-6 13. Hardy J.G. Leal-Egaña A. Scheibel T.R. Engineered spider silk protein-based composites for drug delivery Macromol. Biosci. 2013 13 1431 1437 10.1002/mabi.201300233 23881554 14. Numata K. Kaplan D.L. Silk-based delivery systems of bioactive molecules Adv. Drug Deliv. Rev. 2010 62 1497 1508 10.1016/j.addr.2010.03.009 20298729 15. Zhang Y. Fan M.W. Ma Z. Wu C. Fang W. Liu G. Xiao Y. The effects of pore architecture in silk fibroin scaffolds on the growth and differentiation of mesenchymal stem cells expressing BMP7 Acta Biomater. 2010 6 3021 3028 10.1016/j.actbio.2010.02.030 20188872 16. Meinel L. Fajardo R. Hofmann S. Langer R. Chen J. Snyder B. Vunjak-Novakovic G. Kaplan D. Silk implants for the healing of critical size bone defects Bone 2005 37 688 698 10.1016/j.bone.2005.06.010 16140599 17. Meinel L. Hofmann S. Betz O. Fajardo R. Merkle H.P. Langer R. Evans C.H. Vunjak-Novakovic G. Kaplan D.L. Osteogenesis by human mesenchymal stem cells cultured on silk biomaterials: Comparison of adenovirus mediated gene transfer and protein delivery of BMP-2 Biomaterials 2006 27 4993 5002 10.1016/j.biomaterials.2006.05.021 16765437 18. Young D.S. Krause E.W. Balik C.M. Hyaluronic Acid-Based Nanofibers via Electrospinning Master’s Thesis North Carolina State University Raleigh, NC, USA 2006 19. Hosseini Ravandi S.A. Gandhimathi C. Valizadeh M. Ramakrishna S. Application of electrospun natural biopolymer nanofibers Curr. Nanosci. 2013 9 423 433 10.2174/1573413711309040002 20. Xiao C.D. Xiao P.Z. Jian Z. Hua-Qiong C. Jian T. Quan-Li L. Minocycline-released hydroxyapatite–gelatin nanocomposite and its cytocompatibility in vitro Biomed. Mater. 2011 6 5002 10.1088/1748-6041/6/2/025002 21. Teixeria A.I. Nealey P.P.F. Murphy C.J. Responses of human keratocytes to micro and nanostructured substrates J. Biomed. Mater. Res. A 2004 71 369 376 10.1002/jbm.a.30089 15470741 22. Zahedi P. Rezaeian I. Ranaei-Siadat S.O. Jafari S.H. Supaphol P. A review on wound dressings with an emphasis on electrospun nanofibrous polymeric bandages Polym. Adv. Technol. 2010 21 77 95 10.1002/pat.1625 23. Maquet V. Boccaccini A.R. Pravata L. Notingher I. Jerome R. Porous poly(ahydroxyacid)/Bioglass composite scaffolds for bone tissue engineering. I: Preparation and in vitro characterisation Biomaterials 2004 25 4185 4194 10.1016/j.biomaterials.2003.10.082 15046908 24. Yang Y.Y. Shi M. Goh S.H. Moochala S. Heller J. POE/PLGA composite microspheres: Formation and in vitro behaviour of double walled microspheres J. Control. Release 2003 88 201 213 10.1016/S0168-3659(02)00491-1 12628328 25. Loo S.C.J. Ooi C.P. Boey Y.C.F. Radiation effects on poly(lactide-co-glycolide) (PLGA) and poly(l -lactide) (PLLA) Polym. Degrad. Stab. 2004 83 259 265 10.1016/S0141-3910(03)00271-4 26. Hurrell S. Cameron R.E. Polyglycolide: Degradation and drug release. Part I: Changes morphology during degradation J. Mater. Sci. Mater. Med. 2001 12 811 816 10.1023/A:1017925019985 15348229 27. Drury J.L. Mooney D.J. Hydrogels for tissue engineering: Scaffold design variables and applications Biomaterials 2003 24 4337 4351 10.1016/S0142-9612(03)00340-5 12922147 28. Venugopal J.R. Low S. Choon A.T. Kumar A.B. Ramakrishna S. Nanobioengineered electrospun composite nanofibers and osteoblasts for bone regeneration Artif. Organs 2008 32 388 397 10.1111/j.1525-1594.2008.00557.x 18471168 29. Venugopal J. Rajeswari R. Shayanthi M. Bongso A. Giri Dev V.R. Deepika G. Choon A.T. Ramakrishna S. Electrosprayed hydroxyapatite on polymer nanofibers to differentiate mesenchymal stem cells to osteogenesis J. Biomater. Sci. Polym. Ed. 2013 24 170 184 10.1163/156856212X629845 22370175 30. Anderson H.C. Sipe J.B. Hessle L. Dhanyamraju R. Atti E. Camacho N.P. Impaired calcification around matrix vesicles of growth plate and bone in alkaline phosphatase-deficient mice Am. J. Pathol. 2004 164 841 847 10.1016/S0002-9440(10)63172-0 14982838 31. Liu F. Malaval L. Aubin J.E. Global amplification polymerase chain reaction reveals novel transitional stages during osteoprogenitor differentiation J. Cell Sci. 2003 116 1787 1796 10.1242/jcs.00376 12665559 32. Popat K.C. Chatvanichkul K.I. Barnes G.L. Latempa T.J. Grimes C.A. Desai T.A. Osteogenic differentiation of marrow stromal cells cultured on nanoporous alumina surfaces J. Biomed. Mater. Res. A 2007 80 955 964 10.1002/jbm.a.31028 17089417 33. Popat K.C. Daniels R.H. Dubrow R.S. Hardev V. Desai T.A. Nanostructured surfaces for bone biotemplating applications J. Orthop. Res. 2006 24 619 627 10.1002/jor.20105 16514643 34. Park J.K. Shim J.H. Kang K.S. Yeom J. Jung H.S. Kim J.Y. Solid freeform fabrication of tissue-engineering scaffolds with a poly(lactic-co-glycolic acid) grafted hyaluronic acid conjugate encapsulating an intact bone morphogenetic protein–2/poly(ethylene glycol) complex Adv. Funct. Mater. 2011 21 2906 2912 10.1002/adfm.201100612 35. Mizuno M. Kuboki Y. Osteoblast-related gene expression of bone marrow cells during the osteoblastic differentiation induced by type I collagen J. Biochem. 2001 129 133 138 10.1093/oxfordjournals.jbchem.a002824 11134967 36. Stein G.S. Lian J.B. van Wijnen A.J. Stein J.L. Montecino M. Javed A. Runx2 control of organization, assembly and activity of the regulatory machinery for skeletal gene expression Oncogene 2004 23 4315 4329 10.1038/sj.onc.1207676 15156188 37. Young M.F. Kerr J.M. Ibaraki K. Heegaard A.M. Robey P.G. Structure, expression, and regulation of the major noncollagenous matrix proteins of bone Clin. Orthop. Relat. Res. 1992 281 275 294 10.1097/00003086-199208000-00042 1499220

PMC005xxxxxx/PMC5000621.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081223ijms-17-01223ReviewIncretin-Based Therapies for Diabetic Complications: Basic Mechanisms and Clinical Evidence Kawanami Daiji 1*Matoba Keiichiro 1Sango Kazunori 2Utsunomiya Kazunori 1Cai Lu Academic EditorWang Yuehui Academic EditorZhang Zhiguo Academic Editor1 Division of Diabetes, Metabolism and Endocrinology, Department of Internal Medicine, Jikei University School of Medicine, 3-25-8 Nishi-shinbashi, Minato-ku, Tokyo 105-8461, Japan; [email protected] (K.M.); [email protected] (K.U.)2 Diabetic Neuropathy Project, Department of Sensory and Motor Systems, Tokyo Metropolitan Institute of Medical Science, 2-1-6 Kamikitazawa, Setagaya-ku, Tokyo 156-8506, Japan; [email protected]* Correspondence: [email protected]; Tel.: +81-3-3433-1111; Fax: +81-3-3578-975329 7 2016 8 2016 17 8 122324 6 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).An increase in the rates of morbidity and mortality associated with diabetic complications is a global concern. Glycemic control is important to prevent the development and progression of diabetic complications. Various classes of anti-diabetic agents are currently available, and their pleiotropic effects on diabetic complications have been investigated. Incretin-based therapies such as dipeptidyl peptidase (DPP)-4 inhibitors and glucagon-like peptide-1 receptor agonists (GLP-1RA) are now widely used in the treatment of patients with type 2 diabetes. A series of experimental studies showed that incretin-based therapies have beneficial effects on diabetic complications, independent of their glucose-lowering abilities, which are mediated by anti-inflammatory and anti-oxidative stress properties. Based on these findings, clinical studies to assess the effects of DPP-4 inhibitors and GLP-1RA on diabetic microvascular and macrovascular complications have been performed. Several but not all studies have provided evidence to support the beneficial effects of incretin-based therapies on diabetic complications in patients with type 2 diabetes. We herein discuss the experimental and clinical evidence of incretin-based therapy for diabetic complications. incretinDPP-4glucose-dependent insulinotropic polypeptide (GIP)GLP-1diabetesdiabetic complicationscardiovascular disease ==== Body 1. Introduction The number of patients with diabetes is increasing worldwide. Although remarkable advances have been made in developing novel agents against diabetes, therapies that directly target diabetic complications are still not available. Both experimental and clinical studies suggested that the inflammatory process plays an important role in the development of diabetic complications. Therefore, agents that can control not only hyperglycemia but also the inflammatory process may aid in the prevention and regression of diabetic complications. The potential beneficial effects of incretin-based therapies are increasingly recognized. DPP-4 inhibitors and GLP-1RA were originally developed to lower plasma glucose levels, but accumulating evidence shows that they have vascular protective effects as well, independent of their glucose-lowering abilities, in an incretin-dependent and incretin-independent manner. In this review article, we discuss our current understanding of incretin-based therapies against diabetic nephropathy, retinopathy, neuropathy, and macrovasculopathy. 2. Incretins as Therapeutic Targets of Diabetes Incretins are gut-derived members of the glucagon superfamily that are released from the small intestine in response to nutrient ingestion. Glucagon-like peptide (GLP)-1 and glucose-dependent insulinotropic polypeptide (GIP) are major physiological incretins. They exert biological effects through their specific receptors: GLP-1 receptor (GLP-1R) and GIP receptor (GIPR), which are G-coupled protein receptors [1]. The binding of incretins to the receptors on pancreatic β cells leads to activation of adenylate cyclase-mediated signaling cascades. Accordingly, an increase in the intracellular cyclic adenosine monophosphate (cAMP)-mediated activation of protein kinase A stimulates the exocytosis of insulin-containing granules [1,2]. GLP-1 and GIP account for 50% to 70% of postprandial glucose-dependent insulin secretion [3]. These proteins have opposing effects on glucagon secretion from pancreatic α cells. GIP has been shown to stimulate glucagon secretion through GIPR in pancreatic α cells in a cAMP-dependent manner [1,4]. Enhanced glucagon secretion by GIP has been confirmed in patients with type 2 diabetes [4]. In contrast, GLP-1 has been shown to suppress glucagon secretion when plasma glucose levels are above the fasting level [5], meaning that GLP-1 does not suppress the counter-regulatory responses of glucagon against hypoglycemia. The inhibitory effects of GLP-1 on glucagon secretion are thought to be mediated by somatostatin. It has been shown that GLP-1 stimulates pancreatic somatostatin secretion [6], and blockade of somatostatin abolishes the inhibitory effect of GLP-1 on glucagon secretion [7]. Although the precise mechanism remains unknown, this evidence supports the notion of somatostatin-dependent glucagon inhibition by GLP-1. GIP and GLP-1 have very short half-lives (approximately 1–2 min) because they are quickly degraded by DPP-4 (also known as CD26), which drastically reduces their activity [8]. DPP-4 is a widely expressed serine peptidase that exists in various cell types including vascular cells, renal cells, and T cells. DPP-4 inactivates peptides with an alanine, proline, or serine residue in the penultimate position from the N-terminus [9,10]. In addition to its membrane-bound form, DPP-4 also circulates as a soluble form in the plasma, which lacks the cytoplasmic and transmembrane domain with preserved catalytic activity [11]. DPP-4 functions as a binding protein and is highly accessible to peptide substrates circulating through the gut, liver, lung, and kidney [12,13]. At present, DPP-4 inhibitors and GLP-1RA are available as incretin-based therapies. GLP-1RA is resistant to DPP-4 degradation. DPP-4 inhibitors inhibit DPP-4-mediated GIP and GLP-1 inactivation, thereby elevating the GLP-1 and GIP levels, although the extent of elevation (picomolar) is small compared with pharmacological supplementation with GLP-1 analogs (nanomolar) [9]. Incretin dynamics are impaired in type 2 diabetes. Clinical studies have shown that the incretin effect in diabetic patients is reduced compared to that in healthy individuals, although it still remains unclear whether this is a cause or consequence of diabetes [14,15,16]. Furthermore, DPP-4 activity has been shown to be elevated in both type 1 and type 2 diabetic subjects [17,18,19]. GIP-based therapy for diabetes was abandoned because the insulinotropic effect of GIP is reduced and GIP-dependent postprandial glucagon production is increased in type 2 diabetes [20,21]. In contrast, the insulinotropic effect of GLP-1 has been shown to be preserved in type 2 diabetes [4,21,22]. 3. Organ Protective Effects of Incretins Hyperglycemia is known to activate metabolic pathways, including the diacylglycerol (DAG)-protein kinase C (PKC) pathway, advanced glycation end-products (AGE) pathway, polyol pathway, and hexosamine pathway [23]. Activation of these signaling pathways induces inflammation and oxidative stress, which has been implicated in the pathogenesis of diabetic complications. Incretin-based therapies in experimental studies have been shown to attenuate diabetic vascular complications by inhibiting metabolic pathways such as the PKC pathway [24] and AGE pathway [25,26]. Accumulating evidence show that GLP-1 induces anti-inflammatory effects by downregulating ROS production and NF-κB activation in vascular cells [27,28] and renal cells [26,29]. GLP-1 exerts these beneficial effects via GLP-1R. In addition to a GLP-1-dependent mechanism, DPP-4 inhibitors also exert organ protective effects in part through GLP-1-independent mechanisms. DPP-4 has exopeptidase activity that cleaves dipeptides from the amino terminus of polypeptides with a proline or alanine at the second position [30]. DPP-4 has multiple substrates other than GLP-1, including brain natriuretic peptide (BNP) [31], substance P [32], neuropeptide Y (NPY) [33], stromal-derived factor 1α (SDF-1α) [34], and high-mobility group protein B1 (HMGB1) [35]. These substrates have been implicated in regulating vascular function such as vascular tone regulation, inflammation, cell migration, and cell differentiation [36]. For example, SDF-1α is a chemokine that attracts stem cell such as hematopoietic stem cells (HSCs) and endothelial progenitor cells (EPCs) [11]. Interestingly, the DPP-4 inhibitor linagliptin has been shown to reduce infarct size after myocardial ischemia in rats by inhibiting SDF-1α degradation, thereby enhancing the recruitment of CXC chemokine receptor 4 (CXCR4), a specific receptor for SDF-1α-positive circulating progenitor cells [37]. Similar effects mediated by DPP-4’s substrates can be expected in other organs, particularly in the kidney, in which DPP-4 is expressed at the highest level per organ weight [12]. Furthermore, a study of the tissue distribution of linagliptin found that the accumulation was highest in the kidney [38]. These findings suggest that the kidney is the organ where DPP-4 interacts with DPP-4 inhibitor. Taken together, these findings suggest that the pleiotropic effects of incretin-based therapies are mediated by both incretin-dependent and incretin-independent mechanisms (Figure 1), which thereby exert beneficial effects on diabetic complications (Figure 2) (Table 1, Table 2 and Table 3). 4. Effects of Incretin-Based Therapies on Diabetic Nephropathy 4.1. Experimental Studies Accumulating evidence shows the beneficial effect of incretin-based therapies on diabetic nephropathy. In rats, GLP-1R mRNA expression in the glomeruli and proximal tubules has been reported [70]. Furthermore, a study utilizing in situ hybridization showed that GLP-1R is expressed in the glomerular capillary wall and vascular structure but not in tubules in mice [71]. In humans, while one study found that GLP-1R was expressed in glomeruli and tubules [72], another study found it to be expressed dominantly in the proximal tubules [73]. To date, the localization of GLP-1R still remains controversial. These inconsistent results are thought to be due to the limited specificity and sensitivity of antibodies against GLP-1R [36]. It has been shown that renal DPP-4 expression and activity are upregulated in response to a high-fat diet in rats [74]. In addition, GLP-1R expression in the glomeruli has been shown to be downregulated in diabetic rats [75]. These observations indicate the potential usefulness of incretin-based therapies against diabetic nephropathy. A number of studies have investigated the effects of DPP-4 inhibitors in experimental diabetic models. Mega et al. revealed that the administration of sitagliptin attenuated glomerular, tubulointerstitial, and vascular lesions, accompanied by reduced lipid peroxidation in type 2 diabetic Zucker diabetic fatty (ZDF) rats [39]. Sitagliptin also has been shown to attenuate glomerulosclerosis and tubulointerstitial fibrosis by decreasing the levels of inflammatory cytokines such as IL-1β and TNF-α as well as cellular apoptosis in the kidney of ZDF rats [40]. However, it is difficult to determine the renoprotective effects beyond glucose reduction because DPP-4 inhibitors can improve glycemic control in type 2 diabetic models. Interestingly, the administration of DPP-4 inhibitors has been shown to attenuate diabetic nephropathy in type 1 diabetic animal models independent of glucose-lowering. Kodera et al. reported that the DPP-4 inhibitor PKF275-055 attenuated urinary albumin excretion in STZ-diabetic rats, whereas glycemic control was not affected. A mechanistic analysis showed that PKF275-055 suppressed NF-κB activation, thereby inhibiting the expression of adhesion molecules and macrophage infiltration in the glomeruli [41]. The endothelial-mesenchymal transition (EndMT) plays an important role in the pathogenesis of diabetic kidney fibrosis. Kanasaki et al. showed that the DPP-4 inhibitor linagliptin ameliorates diabetic kidney fibrosis by EndMT, which is associated with the inhibition of DPP-4 protein expression by miR-29, the miR that negatively regulates 3′UTR of DPP-4 mRNA [42]. Of note, the attenuation of EndMT by linagliptin is thought to be drug-specific and not a class effect. Indeed, Shi et al. found that linagliptin but not sitagliptin inhibits TGF-β2-induced EndMT and DPP-4 3′UTR activity in human dermal microvascular endothelial cells [12]. The difference in the effects of these gliptins seems to be dependent on their ability to inhibit homo-dimer formation of DPP-4, which was observed only in linagliptin [12]. Mima et al. demonstrated that GLP-1R is downregulated by diabetes-induced PKCβ activation in glomerular endothelial cells. Mice overexpressing PKCβ2 in endothelial cells showed exaggerated albuminuria, and a mechanistic analysis revealed that PKCβ2 activation promotes ubiquitin-mediated GLP-1R degradation [76]. GLP-1RA has been shown to have direct renoprotective effects independent of the glucose-lowering ability in STZ-diabetic rats, a type 1 diabetic model [29,43]. The inhibition of oxidative stress and the inflammatory process are involved in the direct renoprotective effects of GLP-1RA. GLP-1-mediated PKA activation attenuates oxidative stress, as NAD(P)H oxidase is activated through PKA [77]. Furthermore, the GLP-1RA-mediated attenuation of albuminuria was associated with a reduction in the urinary 8-OHdG excretion, an oxidative stress marker [43]. It has been shown that GLP-1RA attenuates intercellular adhesion molecule (ICAM)-1 expression and macrophage infiltration in the kidney via the amelioration of oxidative stress and reduction of NF-κB expression [29]. Finally, GLP-1RA has been shown to inhibit AGE-mediated monocyte chemoattractant protein (MCP)-1 expression by inhibiting RAGE expression and subsequent ROS production in mesangial cells [26]. Consistent with these observations, GLP-1R-deficient C57/BL6 Akita mice showed increased albuminuria, mesangial expansion, and oxidative stress with decreased cAMP and PKA activity in the kidney [71]. Taken together, these findings show that GLP-1RA exerts a renoprotective effect independent of glucose-lowering by attenuating the inflammation induced by oxidative stress and NF-κB activation. A summary of incretin-based therapies on experimental models is shown in Table 1. 4.2. Clinical Studies Clinical studies have shown that DPP-4 inhibitors attenuate albuminuria in type 2 diabetic subjects. Hattori et al. found that administration of sitagliptin (50 mg/day) for 6 months resulted in a significant reduction in urinary albumin excretion in 36 patients with type 2 diabetes whose HbA1c levels were higher than 6.5% [55]. In their study, sitagliptin improved glycemic control and lowered both systolic and diastolic blood pressures. Interestingly, significant reductions in highly sensitive C-reactive protein and soluble vascular cell adhesion molecule (VCAM)-1 were also observed [55], suggesting that the albuminuria reduction by sitagliptin was dependent on the glucose-lowering effect, as well as the blood pressure reduction and anti-inflammatory effects. Accordingly, Mori et al. investigated the effect of sitagliptin on albuminuria in comparison with other anti-diabetic agents [56]. Eighty-five patients (HbA1c < 6.5%) were allocated to either the sitagliptin group or the other anti-diabetic agents group. Improvement of glycemic control was observed in both groups, but a significant reduction in the urinary albumin excretion was obtained only in the sitagliptin group [56]. The Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus-Thrombolysis in Myocardial Infarction (SAVOR-TIMI 53) trial showed improvement of albuminuria by saxagliptin in type 2 diabetic patients at risk for cardiovascular events [78]. The patients were stratified by the renal function as normal function (estimated glomerular filtration rate (eGFR) >50 mL/min/1.73 m2; n = 13,916), moderate renal impairment (eGFR 30–50 mL/min/1.73 m2; n = 2240), or severe renal impairment (eGFR <30 mL/min/1.73 m2; n = 336) and randomized to receive saxagliptin or placebo. After a two-year follow-up period, saxagliptin did not affect the risk of ischemic cardiovascular events, neither of which were affected by the renal function [78]. Importantly, saxagliptin reduced albuminuria, regardless of the baseline renal function. Given that the HbA1c reduction was significant in the saxagliptin group at two years (7.5% in saxagliptin vs. 7.8% in placebo, p < 0.01) [57], whether the saxagliptin-mediated reduction of albuminuria was due to a glucose-lowering or incretin-dependent mechanism remains unclear. Groop et al. showed a potential glucose-independent effect of linagliptin on albuminuria [58]. In their study, 217 type 2 diabetic patients with albuminuria under RAAS inhibitors were randomized to a placebo group or linagliptin group. Linagliptin treatment induced a significant reduction (32%) in the urinary albumin-creatinine ratio (ACR), and this finding was not associated with the magnitude of the control of the blood glucose and blood pressure [58]. Fujita et al. showed an important finding that supports the notion that DPP-4 inhibitors attenuate albuminuria beyond glucose-lowering [59]. They investigated the effect of the combination of DPP-4 inhibitors with ARB in type 2 diabetic patients with incipient nephropathy. The study consisted of three treatment periods: sitagliptin 50 mg/day for four weeks (first period), alogliptin 25 mg/day for four weeks (second period), and sitagliptin 50 mg/day for four weeks (third period) [59]. Intriguingly, switching from sitagliptin to alogliptin resulted in a decrease in the urinary levels of albumin and 8-hydroxy-2′-deoxyguanosine (8-OHdG), and an increase in the urinary cAMP levels and plasma levels of SDF-1α [59], suggesting that alogliptin can attenuate albuminuria by inhibiting oxidative stress through the reduction of the SDF-1α degradation by DPP-4. To further understand the potential renoprotective effects of linagliptin beyond its glucose-lowering abilities, the Efficacy, Safety & Modification of Albuminuria in Type 2 Diabetes Subjects with Renal Disease with LINAgliptin (MARLINA-T2D) study is currently in progress [79]. In this study, 350 inadequately controlled type 2 diabetic individuals with albuminuria were randomized to either the linagliptin group (5 mg/day) or placebo group in addition to receiving stable glucose-lowering background therapy for 24 weeks [79]. The results will provide novel evidence of the pleiotropic effects of linagliptin on diabetic nephropathy. The beneficial effects of GLP-1RA on albuminuria in type 2 diabetic patients have also been reported. For instance, long-term treatment (one-year) of liraglutide has been shown to reduce albuminuria as well as lower the blood glucose and blood pressure [60,80]. The renoprotective effects of GLP-1RA appear to be exerted via the inhibition of the fibrotic process in the kidney. Zhang et al. demonstrated that exenatide can reduce urinary TGF-β1 and type IV collagen excretion in type 2 diabetic patients [61]. In their study, 31 type 2 diabetic patients with microalbuminuria were allocated to either the exenatide (initiated with 5 µg twice daily the first four weeks then increased to 10 µg twice daily) group or glimepiride (1–4 mg/day) group. All of the subjects were under metformin treatment (1.0–1.5 g/day). After 16 weeks, exenatide but not glimepiride treatment significantly reduced the urinary excretion of albumin, TGF-β1, and type IV collagen, with no significant difference in the glycemic control between the groups [61]. Taken together, these findings suggest that both DPP-4 inhibitors and GLP-1RA could be attractive therapeutic options against diabetic nephropathy. However, large randomized clinical trials are required to conclude the usefulness of incretin-based therapies for diabetic nephropathy. A summary of clinical studies that evaluate the effect of incretin-based therapies is shown in Table 2. 5. Effects of Incretin-Based Agents on Diabetic Retinopathy 5.1. Experimental Studies Retinal endothelial cell dysfunction plays an important role in the development of diabetic retinopathy because it causes pericyte loss and increases vascular permeability and leukocyte adhesion, all of which are key features in diabetic retinopathy [44,81]. Blood-retinal barrier (BRB) breakdown is an early step of vascular permeability that can be induced by disruption of tight junctions (TJs) of endothelial cells [82]. DPP-4 inhibitors and GLP-1RA have been shown to exert beneficial effects on these changes. Increased DPP-4 activity in the retina has been reported in STZ-diabetic rats [44]. GLP-1R is reported to be expressed abundantly in the retina of humans and mice [47]. However, the regulation of GLP1-R under diabetic condition remains inconclusive. A study utilizing STZ-diabetic rats demonstrated that hyperglycemia downregulates GLP-1R expression in the retina [83], although Hernandez et al. found no significant differences in the GLP-1R expression levels in retinas derived from diabetic patients and db/db mice versus those from non-diabetic controls despite GLP-1 levels being lower in the retinas of those with diabetes [47]. Goncalves et al. demonstrated that sitagliptin inhibits BRB breakdown in both type 1 and type 2 diabetic models by preventing the changes in the endothelial subcellular distribution of the TJ proteins, inflammatory cytokines such as IL-1β, and cell death by apoptosis in diabetic retinas [44,45]. They also demonstrated that sitaglitpin prevented the diabetes-induced reduction in the adhesion ability of endothelial progenitor cells (EPCs) to the retinal vessels [44]. Similarly, vildagliptin has been shown to inhibit inflammation and thrombogenic reactions in the retina of Otsuka Long-Evans Tokushima Fatty rats (OLETF rats), a model of obese type 2 diabetes [46]. Both systemic and topical administration of GLP-1RA have been shown to inhibit retinal neurodegeneration such as glial activation and retinal apoptosis in db/db mice independently of the glucose-lowering effect [47]. From a mechanistic standpoint, GLP-1RA exerted these beneficial effects through a significant reduction of retinal glutamate and stimulation of prosurvival signaling pathways by increasing the pAKT/AKT ratio [47]. GLP-1RA has been reported to attenuate ischemia-reperfusion-induced BRB damage as well as inflammatory cytokine production by microglia activated by the inhibition of NF-κB activation [84]. 5.2. Clinical Studies Retinal hyperperfusion is an early hemodynamic change that occurs prior to clinical manifestations of diabetic retinopathy. A clinical trial to investigate the effect of saxagliptin on early retinal microvascular changes in patients with type 2 diabetes was performed [63]. In this study, 50 type 2 diabetic individuals without micro- or macro-vascular complications were randomized to the placebo group or saxagliptin (5 mg) group. After six weeks, the retinal arteriolar structure and retinal capillary flow (RCF) were assessed. Interestingly, administration of saxagliptin resulted in normalization of the RCF [63]. It has been shown that 10-month exenatide treatment resulted in transient worsening of diabetic retinopathy in 30% of diabetic subjects, which was associated with the rapid reduction in HbA1c levels [64,65]. However, a follow-up study revealed that sustained exenatide treatment improved or maintained stable diabetic retinopathy in 80% of patients who showed transient progression of diabetic retinopathy by exenatide [65]. This finding demonstrates the potential protective effect of GLP-1RA for diabetic retinopathy. However, given that significant improvement of glycemic control was obtained in both studies, it remains uncertain whether these observations are independent of or dependent on the glucose-lowering effect. Further study will be required to elucidate the direct beneficial effect of DPP-4 inhibitors on diabetic retinopathy. 6. Effects of Incretin-Based Agents on Diabetic Neuropathy 6.1. Experimental Studies GLP1-R is expressed in the nervous tissues, including sensory neurons and Schwann cells in dorsal root ganglia (DRG). Sango & Utsunomiya demonstrated that GLP-1R is expressed predominantly in large and small peptidergic DRG neurons rather than small non-peptidergic neurons [85]. However, the expression of GLP-1RA is not altered under diabetic conditions. GLP-1RA has been shown to exert beneficial effects on diabetic neuropathy in STZ-induced diabetic rats, independent of the glucose-lowering effect. Himeno et al. showed that four-week administration of GLP-1RA restored motor and sensory nerve conduction velocities (NCV) and hypoalgesia [50]. Furthermore, GLP-1RA has been shown to activate the ERK signaling pathway in peripheral neurons and/or Schwann cells derived from diabetic rats and mice, thereby protecting against the reduction of motor nerve conduction velocity (NCV) [51]. Tsukamoto et al. demonstrated that GLP-1RA restored the reduced neurite outgrowth and viability of adult rat DRG neurons caused by the absence of insulin in culture medium and suppressed the activity of RhoA, a small GTP-ase binding protein that is an inhibitory regulator for peripheral nerve regeneration, in PC12 cells. Furthermore, these effects were attenuated by the phosphatidylinositol-3′-phosphate kinase (PI3K) inhibitor LY294002, indicating that GLP-1RA enhances neurite outgrowth and neuronal survival through the activation of the PI3K signaling pathway, which negatively regulates RhoA activity [86]. DPP-4 inhibitors have been shown to inhibit diabetic neuropathy in both type 1 and type 2 diabetic rodent models. Jin et al. showed that vildagliptin protected STZ-induced diabetic rats from peripheral nerve degeneration by ameliorating decreases in the intraepidermal nerve fiber density [48]. Accordingly, Bianchi et al. investigated the protective and therapeutic effects of vildagliptin on diabetic neuropathy in STZ-induced diabetic rats. They observed that the vildagliptin analog PKF275-055 partially counteracted the reduction in the NCV but failed to improve the mechanical and thermal sensitivity of diabetic rats in prevention and protection experiments. However, they showed that PKF275-055 treatment restored mechanical sensitivity thresholds by 50% and progressively improved changes in the thermal responsiveness in therapeutic experiments [49]. Tsuboi et al. studied the effects of vildagliptin on diabetic neuropathy in more detail. They demonstrated that the administration of vildagliptin improved NCV in Goto-Kakizaki (GK) rats. Vildagliptin ameliorated delayed NCV and neuronal atrophy and reduced the expression of calcitonin-gene-related peptide (CGRP), a potent vasodilator of epineurial arterioles [87], as well as lowered the intraepidermal nerve fiber density in GK rats. Similarly, vildagliptin restored impaired NCV in STZ-induced diabetic mice [88]. From a mechanistic standpoint, vildagliptin corrected the impaired phosphorylation of cAMP response element binding protein (CREB), protein kinase B/Akt (PKB/Akt), S6-ribosomal protein (S6RP), and insulin receptor substrate (IRS) 2 in DRG of diabetic models, suggesting that vildagliptin restores the diabetes-induced impairment of GLP-1 and insulin signaling that play important roles in neurite growth and cell survival, thereby exerting protective effects against diabetic neuropathy [88]. Alogliptin has also been shown to improve the NCV in STZ-induced diabetic rats by improving CGRP-mediated vascular relaxation in epineurial arterioles [89]. 6.2. Clinical Studies Only one clinical study has investigated the effects of incretin-based therapy on diabetic neuropathy. Jaiswal et al. performed an open-label randomized study to evaluate the effects of GLP-1RA exenatide on diabetic peripheral neuropathy (DPN) as well as cardiovascular autonomic neuropathy (CAN) in subjects with type 2 diabetes [66]. In this study, 46 type 2 diabetic subjects with mild-moderate DPN were randomized to a twice daily exenatide group (n = 22) or daily insulin glargine (n = 24). After 18 months of follow-up, no significant differences were observed in the prevalence of confirmed clinical neuropathy, intra-epidermal nerve fiber density, and nerve conductions studies. Furthermore, there were no significant changes in the measures of CAN [66]. Glycemic control was similar in both groups [66]. Although GLP-1RA did not induce any marked beneficial effect in this study, further studies with different design will be required to conclude the clinical usefulness of incretin-based therapies. 7. Diabetic Macrovascular Complications 7.1. Experimental Studies A series of experimental studies demonstrated the anti-atherosclerotic effects of incretin-based therapies in non-diabetic and diabetic animal models [90]. A study in STZ-induced diabetic apoE-deficient mice demonstrated that administration of alogliptin reduced the build-up of atherosclerotic plaque by inhibiting toll-like receptor 4-mediated IL-6 and IL-1β upregulation [52]. Consistently, vildagliptin has been shown to suppress atherosclerotic lesions by inhibiting macrophage accumulation and foam cell formation in STZ-induced apoE-null mice as well as db/db mice. Interestingly, these beneficial effects of DPP-4 inhibitors were mediated, at least in part, through GLP-1 and GIP, because both incretin receptor blockers induced partial but not complete attenuation of vildaglitpin’s anti-atherosclerotic effects [91]. Indeed, native incretins (both GLP-1 and GIP) have been shown to attenuate atherosclerotic lesions and macrophage infiltration in the aortic wall in apoE knockout mice [92,93]. Vildagliptin likely inhibited the activation of monocytes rather than macrophages because the expression of GIPR and GLP-1R were dramatically suppressed by differentiation of monocytes into macrophages [91]. Furthermore, DPP-4 inhibitors have been shown to attenuate atherosclerotic lesion in diabetic models such as ZDF rats [53] and high-fat diet low density lipoprotein (LDL)-receptor-deficient mice [94] via the downregulation of oxidative stress, chemokine production, and monocyte recruitment by inhibiting Rac activation [53,94]. Similarly, GLP-1RA has been shown to attenuate atherosclerosis in diabetic animal models. Arakawa et al. found that exendin-4 reduced monocyte adhesion to the endothelium of aorta, thereby leading to the suppression of atherosclerotic lesions in apoE knockout mice [95]. Tang et al. demonstrated that sitagliptin as well as exenatide administration improved endothelial dysfunction in STZ-induced diabetic rats [54]. From a mechanistic standpoint, these drugs recovered the diabetes-induced impairment of vasorelaxation; increased the serum NO levels and reductions of serum endothelin-1 and inflammatory cytokine levels such as VCAM-1, tumor necrosis factor (TNF)-α, and IL-6 levels; and inhibited the ROS production in the aorta [54], suggesting that incretin-based therapies improve diabetes-induced endothelial dysfunction by inhibiting inflammation and oxidative stress. Indeed, GLP-1RA has been shown to inhibit high-glucose mediated inductions of NAD(P)H oxidases such as p47phox and gp91phox [27]. The cAMP/PKA-mediated inhibition of small GTPase-binding protein Rho and its effector Rho-kinase, important factors in the pathogenesis of diabetic complications [23], are involved in the beneficial effects of GLP-1 on oxidative stress in endothelia cells [27] and the aorta [54] under diabetic conditions. 7.2. Clinical Studies Three large randomized trials—SAVOR-TIMI 53 [57], Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care (EXAMINE) [67], and Trial Evaluating Cardiovascular Outcomes with Sitagliptin (TECOS) study [68]—did not show any significant reductions in the rates of cardiovascular events by DPP-4 inhibitors in patients with type 2 diabetes. In addition, a meta-analysis (total of 75 studies comprising 45,648 patients with type 2 diabetes) demonstrated no significant protective effect of incretin-based therapies against cardiovascular events [96]. An unexpected finding in SAVOR-TIMI53 was an increased incidence of heart failure hospitalization [57], but this observation was not observed in EXAMINE or TECOS. Data that support the anti-atherosclerotic effect of incretin-based therapies are emerging. Recent studies have shown that aloglitpin and sitagliptin prevent the progression of carotid atherosclerosis in patients with diabetes [97,98]. Furthermore, a meta-analysis demonstrated that GLP-1-based therapy has beneficial effects on atherosclerotic markers (brain naturetic peptide, high-sensitivity C-reactive protein, plasminogen activator inhibitor-1, total cholesterol, LDL cholesterol, and triglycerides) in patients with type 2 diabetes [99]. The evaluation of lixisenatide in acute coronary syndrome (ELIXA) investigated the effect of GLP-1RA lixisenatide on cardiovascular morbidity and mortality in type 2 diabetic patients with recent acute coronary syndrome, but no significant differences in the rates of cardiovascular events were noted [69]. Recently, the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial provided evidence that liraglutide reduces the rates of cardiovascular events in patients with type 2 diabetes and high cardiovascular risk [62]. In this study, a total of 9340 type 2 diabetic patients who had HbA1c levels of ≥7.0% were assigned to either a liraglutide (1.8 mg) or placebo group [62]. The primary composite outcome in the time-to-event analysis was the first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. After a median follow-up of 3.8 years, the primary composite outcome was significantly lower in the liraglutide group than in the placebo group (HR, 0.87: 95% CI: 0.78–0.97; p < 0.001 for noninferiority; p = 0.01 for superiority). The administration of liraglutide significantly reduced the rates of death from cardiovascular causes (HR, 0.78: 95% CI: 0.66–0.93; p = 0.007) and death from any cause (HR, 0.85: 95% CI: 0.74–0.97; p = 0.02) [62]. Of note, the rate of incidence of a composite outcome of renal or retinal microvascular events (nephropathy (defined as the new onset of macroalbuminuria or a doubling of the serum creatinine level and an eGFR of ≤45 mL/min/1.73 m2, the need for renal replacement therapy, or death from renal disease) and retinopathy (defined as the need for retinal photocoagulation or treatment with intra-vitreal agents, vitreous hemorrhaging, or diabetes-related blindness)) was significantly lower in the liraglutide group than in the placebo group (HR, 0.84: 95% CI 0.73 to 0.97; p = 0.02) [62]. The LEADER trial demonstrated for the first time the usefulness of GLP-1RA liraglutide in the reduction of the rate of cardiovascular events in patients with type 2 diabetes. These beneficial effects of liraglutide may have been separate from the glucose-lowering effect, given the observation of significant mean differences in the change from baseline to 36 months of cardiovascular risk factors (weight loss (−2.3 kg), systolic blood pressure (−1.2 mmHg), and diastolic blood pressure (−0.6 mmHg)) by liraglutide compared to placebo [62]. Further investigations will be required to elucidate the mechanisms by which liraglutide provides cardiovascular benefit in patients with type 2 diabetes. A summary of the trials that investigated the effects of incretin-based therapies is shown in Table 3. 8. Conclusions Incretin-based therapies are among the most important therapeutic options in diabetes and have revolutionized the treatment of diabetes. Both basic and clinical evidence show that incretin-based therapies have beneficial effects on diabetic complications. However, several points remain to be elucidated. From a basic mechanism perspective, the differential functions between GIP and GLP-1 in diabetic complications are unclear. DPP-4 inhibitors and GLP-1RA are similar in their pleiotropic effects, but they have different pharmacologic actions because the former increases both GIP and GLP-1 levels. Furthermore, DPP-4 inhibitors exert beneficial effects via the DPP-4 substrate. Clarifying the role of GIP in diabetic complications may address the utility of incretin-based drugs. In addition, further studies will need to examine whether or not a combination of DPP-4 inhibitors and GLP-1RA can exert more potent beneficial effects on diabetic complications than either drug alone. Addressing these issues will help researchers develop a novel therapeutic strategy against diabetic complications using incretin-based therapies. Acknowledgments This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (to Daiji Kawanami, Kazunori Sango and Kazunori Utsunomiya), Takeda Science Foundation (to Daiji Kawanami), Banyu Foundation International (to Daiji Kawanami), the Uehara Memorial Foundation (to Daiji Kawanami and Keiichiro Matoba), and the Nukada Institute for Medical and Biological Research (to Kazunori Sango). Author Contributions Daiji Kawanami planned the study, searched the literature, wrote the manuscript, and made the figures. Keiichiro Matoba wrote the manuscript and made the figures. Kazunori Sango wrote the manuscript. Kazunori Utsunomiya helped edit the manuscript and revise the manuscript for important intellectual content. All of the authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflicts of interest. Figure 1 Mechanisms of beneficial effects of increased-based therapy. Dipeptidyl peptidase (DPP)-4 inhibition increases active glucagon-like peptide-1 (GLP-1) levels and GLP-1 signaling through its receptor. DPP-4 inhibition also inhibits degradation of its substrates other than GLP-1 (e.g., stromal-derived factor 1α (SDF-1α)), thereby activating incretin-independent signaling. GLP-1 inhibits inflammation and oxidative stress by downregulating inflammatory cytokine production (e.g., IL-1β, TNF-α), NF-κB, Rho-kinase activation, and the glycation end-products (AGE) pathway. GLP-1 inhibits apoptosis by decreasing the ratio of BAX/Bcl-2, which are a pro-apoptotic protein and an anti-apoptotic protein. The beneficial effects are also exerted via glucose-lowering by GLP-1. Figure 2 The effects of incretin-based therapies on diabetic complications. DPP-4 inhibitors and GLP-1RA attenuate diabetic complications through various beneficial effects. Incretin-based therapies have been shown to attenuate inflammation and oxidative stress, thereby inhibiting the fibrotic response in the kidney. In addition to these anti-inflammatory effects, endothelial dysfunction has been shown to be improved by incretin-based therapies, leading to improved capillary flow and inhibited thrombogenic activity in the retina. The anti-atherogenic effects of incretin-based therapies are mediated by the downregulation of inflammation, oxidative stress, and macrophage activation. The neuroprotective effects of incretin-based therapies are exerted by stimulating neurite growth and cell survival via the activation of GLP-1- and insulin-dependent signaling pathways. Blue arrows: decrease; Red arrows: increase. ijms-17-01223-t001_Table 1Table 1 Summary of the effects of incretin-based therapies on experimental models. The beneficial effects of DPP-4 inhibitors and GLP-1 RA on diabetic microvascular and macrovascular complications have been reported. Complication Model Drug/Dose/Duration Major Effects Nephropathy ZDF rats [39] Sitagliptin,10 mg/kg, 6 weeks ↓Glomerular lesion ZDF rats [40] Sitagliptin,10 mg/kg, 6 weeks ↓Glomerulosclerosis ↓Tubulointerstitial fibrosis STZ-diabetic rats [41] PKF275-055, 3 mg/kg, 8 weeks ↓Inflammation STZ-diabetic mice [42] Linagliptin, 5 mg/kg, 4 weeks ↓Kidney fibrosis STZ-diabetic rats [29] Exendin-4, 10 mg/kg, 8 weeks ↓Inflammation STZ-diabetic rats [43] Liragltuide, 0.3 mg/kg, 8 weeks ↓Oxidative stress Retinopathy STZ-diabetic rats [44] Sitagliptin, 5 mg/kg, 2 weeks ↓Blood-retinal barrier breakdown ↓Inflammation ↓Neuronal apoptosis ZDF-rats [45] Sitaglitpin 10 mg/kg, 6 weeks ↓Inflammation ↓Retinal cell apoptosis OLETF rats [46] Vildagliptin 3 mg/kg, 10 weeks ↓Thrombogenic reactions db/db mice [47] Liraglutide 400 μg/kg, 15 days ↓Retinal neurodegenartion Neuropathy STZ-diabetic rats [48] Vildagliptin 0.3 or 10 mg/kg, 32 weeks ↓Peripheral nerve degeneration STZ-diabetic rats [49] PKF275-055 3 mg/kg, 4 or 5 weeks ↑NCV STZ-diabetic mice [50] Exendin-4 10 nmol/kg, 4 weeks ↑Neurite DRG outgrowth ↑MNCV, SNCV STZ-diabetic mice [51] Exenatide 0.3 pmoles/kg/min, 8 weeks (infusion) ↑MNCV Macrovasculopathy STZ-diabetic apoE-null mice [52] Alogliptin 15 mg/kg, 24 weeks ↓Atherosclerotic plaque ZDF rats [53] Sitaglitpin 10 mg/kg or Linaglitpin 3 mg/kg, 4 weeks ↑Vascular relaxation, ↓Oxidative stress STZ-diabetic rats [54] Sitagliptin 30 mg/kg or ↓Inflammation Exenatide 30 μg/kg/12h (infusion), 12 weeks ↑Endothelial function ZDF: zucker diabetes fatty; STZ: streptozotocin; OLETF: Otsuka Long-Evans Tokushima Fatty; MNCV: motor nerve conduction velocity; SNCV: sensory nerve conduction velocity; NCV: nerve conduction velocity; DRG: dorsal root ganglion; ↓: decrease; ↑: increase. ijms-17-01223-t002_Table 2Table 2 Summary of clinical studies that evaluate the effect of incretin-based therapies on diabetic microvascular complications in patients with type 2 diabetes (T2D). The renoprotective effects of incretin-based therapies have been reported. Further investigations into the usefulness of incretin-based therapies on retinopathy and neuropathy should be performed. Complication Drug Doses (Duration) Patients Endpoint Nephropathy Sitagliptin [55] 50 mg/day (6 months) T2D patients (n = 36) ↓Albuminuria Sitagliptin [56] 50 mg/day (6 months) T2D patients (n = 85) ↓Albuminuria Saxagliptin [57] 2.5 or 5 mg/day (2 years) T2D patients (n = 16,492) ↓Albuminuria Linagliptin [58] 5 mg/day (6 months) T2D patients (n = 217) ↓Albuminuria Alogliptin [59] 25 mg/day (4 weeks) (vs. Sitagliptin 50 mg/day) (cross over) T2D patients (n = 12) ↓Albuminuria Liraglutide [60] 0.6-1.8 mg/day (1 year) T2D patients (n = 84) ↓Albuminuria Exenatide [61] 10 μg twice daily (16 weeks) (5 μg twice daily (first 4 weeks) T2D patients (n = 31) ↓Albuminuria Liraglutide [62] 1.8 mg/day (3.8 years) T2D patients (n = 9340) ↓Composite outcome of renal and retinal microvascular events Retinopathy Saxagliptin [63] 5 mg/day (6 weeks) T2D patients (n = 50) Normalization of retinal capillary flow Exenatide [64] N/A (300 days) T2D patients (n = 165) Transient worsening of diabetic retinopathy (DR) Exenatide [65] N/A (430 days) T2D patients (n = 39) Improvement of DR Neuropathy Exenatide [66] 10 μg twice daily (18 months) (5 μg twice daily (first 4 weeks)) T2D patients (n = 46) No changes in confirmed clinical neuropathy, cardiovascular autonomic neuropathy N/A: Not available. ijms-17-01223-t003_Table 3Table 3 Clinical trials that investigated the effects of incretin-based therapies on the cardiovascular outcome in patients with T2D. All of the studies shown here were performed with T2D patients at high risk of cardiovascular disease. To date, Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) is the only study that showed superiority of incretin-based therapy against cardiovascular events compared to placebo. Trial Drug/Doses Patients Primary Composite Outcome Result (Risk of Cardiovascular Events) SAVOR-TIMI53 [57] (2.1 years) Saxagliptin 2.5 mg or 5 mg/day (on the basis of estimated glomerular filtration rate (eGFR) at baseline) T2D patients who had a history of, or were at risk for, cardiovascular events (n = 16,492) Cardiovascular death, myocardial infarction, or ischemic stroke (no change) EXAMINE [67] (1.5 years) Alogliptin 6.25 mg or 12.5 mg or 25 mg (same as above) T2D patients with either an acute myocardial infarction or unstable angina requiring hospitalization within the previous 15 to 90 days (n = 5380) Cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke TECOS [68] (3.0 years) Sitagliptin 50 mg or 100 mg/day (same as above) T2D patients who had a history of major coronary artery disease, ischemic cerebrovascular disease, or atherosclerotic peripheral arterial disease (n = 14,671) Cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina ELIXA [69] (2.1 years) Lixisenatide 20 μg/day T2D patients who had had a myocardial infarction or who had been hospitalized for unstable angina within the previous 180 days (n = 6068) Cardiovascular death, myocardial infarction, stroke, or hospitalization for unstable angina LEADER [62] (3.8 years) Liraglutide 1.8 mg/day T2D patients ≥50 years of age with at least one cardiovascular coexisting condition or ≥60 years of age with at least one cardiovascular risk factor (n = 9340) Cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke (decrease) ==== Refs References 1. Seino Y. Fukushima M. Yabe D. GIP and GLP-1, the two incretin hormones: Similarities and differences J. Diabetes Investig. 2010 1 8 23 10.1111/j.2040-1124.2010.00022.x 24843404 2. Joao A.L. Reis F. Fernandes R. The incretin system ABCs in obesity and diabetes-novel therapeutic strategies for weight loss and beyond Obes. Rev. Off. J. Int. Assoc. Study Obes. 2016 17 553 572 10.1111/obr.12421 27125902 3. Nauck M.A. Incretin-based therapies for type 2 diabetes mellitus: Properties, functions, and clinical implications Am. J. Med. 2011 124 Suppl. 1 S3 S18 10.1016/j.amjmed.2010.11.002 21194578 4. Chia C.W. Carlson O.D. Kim W. Shin Y.K. Charles C.P. Kim H.S. Melvin D.L. Egan J.M. Exogenous glucose-dependent insulinotropic polypeptide worsens post prandial hyperglycemia in type 2 diabetes Diabetes 2009 58 1342 1349 10.2337/db08-0958 19276444 5. Nauck M.A. Heimesaat M.M. Behle K. Holst J.J. Nauck M.S. Ritzel R. Hufner M. Schmiegel W.H. Effects of glucagon-like peptide 1 on counterregulatory hormone responses, cognitive functions, and insulin secretion during hyperinsulinemic, stepped hypoglycemic clamp experiments in healthy volunteers J. Clin. Endocrinol. Metab. 2002 87 1239 1246 10.1210/jcem.87.3.8355 11889194 6. Orskov C. Holst J.J. Nielsen O.V. Effect of truncated glucagon-like peptide-1 (proglucagon-(78-107) amide) on endocrine secretion from pig pancreas, antrum, and nonantral stomach Endocrinology 1988 123 2009 2013 10.1210/endo-123-4-2009 2901341 7. De Heer J. Rasmussen C. Coy D.H. Holst J.J. Glucagon-like peptide-1, but not glucose-dependent insulinotropic peptide, inhibits glucagon secretion via somatostatin (receptor subtype 2) in the perfused rat pancreas Diabetologia 2008 51 2263 2270 10.1007/s00125-008-1149-y 18795252 8. Seino Y. Yabe D. Glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1: Incretin actions beyond the pancreas J. Diabetes Investig. 2013 4 108 130 10.1111/jdi.12065 24843641 9. Waldrop G. Zhong J. Peters M. Rajagopalan S. Incretin-based therapy for diabetes: What a cardiologist needs to know J. Am. Coll. Cardiol. 2016 67 1488 1496 10.1016/j.jacc.2015.12.058 27012410 10. Mulvihill E.E. Drucker D.J. Pharmacology, physiology, and mechanisms of action of dipeptidyl peptidase-4 inhibitors Endocr. Rev. 2014 35 992 1019 10.1210/er.2014-1035 25216328 11. Zhong J. Rajagopalan S. Dipeptidyl peptidase-4 regulation of SDF-1/CXCR4 axis: Implications for cardiovascular disease Front. Immunol. 2015 6 477 10.3389/fimmu.2015.00477 26441982 12. Shi S. Kanasaki K. Koya D. Linagliptin but not Sitagliptin inhibited transforming growth factor-β2-induced endothelial DPP-4 activity and the endothelial-mesenchymal transition Biochem. Biophys. Res. Commun. 2016 471 184 190 10.1016/j.bbrc.2016.01.154 26826382 13. Deacon C.F. What do we know about the secretion and degradation of incretin hormones? Regul. Pept. 2005 128 117 124 10.1016/j.regpep.2004.06.007 15780431 14. Knop F.K. Vilsboll T. Hojberg P.V. Larsen S. Madsbad S. Volund A. Holst J.J. Krarup T. Reduced incretin effect in type 2 diabetes: Cause or consequence of the diabetic state? Diabetes 2007 56 1951 1959 10.2337/db07-0100 17513701 15. Knop F.K. Aaboe K. Vilsboll T. Volund A. Holst J.J. Krarup T. Madsbad S. Impaired incretin effect and fasting hyperglucagonaemia characterizing type 2 diabetic subjects are early signs of dysmetabolism in obesity Diabetes Obes. Metab. 2012 14 500 510 10.1111/j.1463-1326.2011.01549.x 22171657 16. Tura A. Muscelli E. Gastaldelli A. Ferrannini E. Mari A. Altered pattern of the incretin effect as assessed by modelling in individuals with glucose tolerance ranging from normal to diabetic Diabetologia 2014 57 1199 1203 10.1007/s00125-014-3219-7 24658843 17. Ryskjaer J. Deacon C.F. Carr R.D. Krarup T. Madsbad S. Holst J. Vilsboll T. Plasma dipeptidyl peptidase-IV activity in patients with type-2 diabetes mellitus correlates positively with HbAlc levels, but is not acutely affected by food intake Eur. J. Endocrinol. Eur. 2006 155 485 493 10.1530/eje.1.02221 16914604 18. Mannucci E. Pala L. Ciani S. Bardini G. Pezzatini A. Sposato I. Cremasco F. Ognibene A. Rotella C.M. Hyperglycaemia increases dipeptidyl peptidase IV activity in diabetes mellitus Diabetologia 2005 48 1168 1172 10.1007/s00125-005-1749-8 15864535 19. Osawa S. Kawamori D. Katakami N. Takahara M. Sakamoto F. Katsura T. Yasuda T. Kaneto H. Matsuhisa M. Matsuoka T.A. Significant elevation of serum dipeptidyl peptidase-4 activity in young-adult type 1 diabetes Diabetes Res. Clin. Pract. 2016 113 135 142 10.1016/j.diabres.2015.12.022 26827118 20. Nauck M.A. Heimesaat M.M. Orskov C. Holst J.J. Ebert R. Creutzfeldt W. Preserved incretin activity of glucagon-like peptide 1 (7-36 amide) but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus J. Clin. Investig. 1993 91 301 307 10.1172/JCI116186 8423228 21. Vilsboll T. Krarup T. Madsbad S. Holst J.J. Defective amplification of the late phase insulin response to glucose by GIP in obese Type II diabetic patients Diabetologia 2002 45 1111 1119 12189441 22. Nauck M.A. Homberger E. Siegel E.G. Allen R.C. Eaton R.P. Ebert R. Creutzfeldt W. Incretin effects of increasing glucose loads in man calculated from venous insulin and C-peptide responses J. Clin. Endocrinol. Metab. 1986 63 492 498 10.1210/jcem-63-2-492 3522621 23. Kawanami D. Matoba K. Utsunomiya K. Signaling pathways in diabetic nephropathy Histol. Histopathol. 2016 11777 24. Inoue T. Inoguchi T. Sonoda N. Hendarto H. Makimura H. Sasaki S. Yokomizo H. Fujimura Y. Miura D. Takayanagi R. GLP-1 analog liraglutide protects against cardiac steatosis, oxidative stress and apoptosis in streptozotocin-induced diabetic rats Atherosclerosis 2015 240 250 259 10.1016/j.atherosclerosis.2015.03.026 25818251 25. Ishibashi Y. Matsui T. Takeuchi M. Yamagishi S. Glucagon-like peptide-1 (GLP-1) inhibits advanced glycation end product (AGE)-induced up-regulation of VCAM-1 mRNA levels in endothelial cells by suppressing AGE receptor (RAGE) expression Biochem. Biophys. Res. Commun. 2010 391 1405 1408 10.1016/j.bbrc.2009.12.075 20026306 26. Ishibashi Y. Nishino Y. Matsui T. Takeuchi M. Yamagishi S. Glucagon-like peptide-1 suppresses advanced glycation end product-induced monocyte chemoattractant protein-1 expression in mesangial cells by reducing advanced glycation end product receptor level Metab. Clin. Exp. 2011 60 1271 1277 10.1016/j.metabol.2011.01.010 21388644 27. Wang D. Luo P. Wang Y. Li W. Wang C. Sun D. Zhang R. Su T. Ma X. Zeng C. Glucagon-like peptide-1 protects against cardiac microvascular injury in diabetes via a cAMP/PKA/Rho-dependent mechanism Diabetes 2013 62 1697 1708 10.2337/db12-1025 23364453 28. Lee Y.S. Park M.S. Choung J.S. Kim S.S. Oh H.H. Choi C.S. Ha S.Y. Kang Y. Kim Y. Jun H.S. Glucagon-like peptide-1 inhibits adipose tissue macrophage infiltration and inflammation in an obese mouse model of diabetes Diabetologia 2012 55 2456 2468 10.1007/s00125-012-2592-3 22722451 29. Kodera R. Shikata K. Kataoka H.U. Takatsuka T. Miyamoto S. Sasaki M. Kajitani N. Nishishita S. Sarai K. Hirota D. Glucagon-like peptide-1 receptor agonist ameliorates renal injury through its anti-inflammatory action without lowering blood glucose level in a rat model of type 1 diabetes Diabetologia 2011 54 965 978 10.1007/s00125-010-2028-x 21253697 30. Salles T.A. dos Santos L. Barauna V.G. Girardi A.C. Potential role of dipeptidyl peptidase IV in the pathophysiology of heart failure Int. J. Mol. Sci. 2015 16 4226 4249 10.3390/ijms16024226 25690036 31. Boerrigter G. Costello-Boerrigter L.C. Harty G.J. Lapp H. Burnett J.C. Jr. Des-serine-proline brain natriuretic peptide 3–32 in cardiorenal regulation Am. J. Physiol. Regul. Integ. Comp. Physiol. 2007 292 R897 R901 10.1152/ajpregu.00569.2006 17068158 32. Grouzmann E. Bigliardi P. Appenzeller M. Pannatier A. Buclin T. Substance P-induced skin inflammation is not modulated by a single dose of sitagliptin in human volunteers Biol. Chem. 2011 392 217 221 10.1515/bc.2011.003 21194357 33. Mentlein R. Dahms P. Grandt D. Kruger R. Proteolytic processing of neuropeptide Y and peptide YY by dipeptidyl peptidase IV Regul. Pept. 1993 49 133 144 10.1016/0167-0115(93)90435-B 7907802 34. Shioda T. Kato H. Ohnishi Y. Tashiro K. Ikegawa M. Nakayama E.E. Hu H. Kato A. Sakai Y. Liu H. Anti-HIV-1 and chemotactic activities of human stromal cell-derived factor 1α (SDF-1α) and SDF-1β are abolished by CD26/dipeptidyl peptidase IV-mediated cleavage Proc. Natl. Acad. Sci. USA 1998 95 6331 6336 10.1073/pnas.95.11.6331 9600965 35. Marchetti C. Di Carlo A. Facchiano F. Senatore C. De Cristofaro R. Luzi A. Federici M. Romani M. Napolitano M. Capogrossi M.C. High mobility group box 1 is a novel substrate of dipeptidyl peptidase-IV Diabetologia 2012 55 236 244 10.1007/s00125-011-2213-6 21656024 36. Makino Y. Fujita Y. Haneda M. Dipeptidyl peptidase-4 inhibitors in progressive kidney disease Curr. Opin. Nephrol. Hypertens. 2015 24 67 73 10.1097/MNH.0000000000000080 25415611 37. Hocher B. Sharkovska Y. Mark M. Klein T. Pfab T. The novel DPP-4 inhibitors linagliptin and BI 14361 reduce infarct size after myocardial ischemia/reperfusion in rats Int. J. Cardiol. 2013 167 87 93 10.1016/j.ijcard.2011.12.007 22217485 38. Fuchs H. Binder R. Greischel A. Tissue distribution of the novel DPP-4 inhibitor BI 1356 is dominated by saturable binding to its target in rats Biopharm. Drug Dispos. 2009 30 229 240 10.1002/bdd.662 19562682 39. Mega C. de Lemos E.T. Vala H. Fernandes R. Oliveira J. Mascarenhas-Melo F. Teixeira F. Reis F. Diabetic nephropathy amelioration by a low-dose sitagliptin in an animal model of type 2 diabetes (Zucker diabetic fatty rat) Exp. Diabetes Res. 2011 2011 162092 10.1155/2011/162092 22203828 40. Marques C. Mega C. Goncalves A. Rodrigues-Santos P. Teixeira-Lemos E. Teixeira F. Fontes-Ribeiro C. Reis F. Fernandes R. Sitagliptin prevents inflammation and apoptotic cell death in the kidney of type 2 diabetic animals Mediat. Inflamm. 2014 2014 538737 10.1155/2014/538737 24817793 41. Kodera R. Shikata K. Takatsuka T. Oda K. Miyamoto S. Kajitani N. Hirota D. Ono T. Usui H.K. Makino H. Dipeptidyl peptidase-4 inhibitor ameliorates early renal injury through its anti-inflammatory action in a rat model of type 1 diabetes Bioche. Biophys. Res. Commun. 2014 443 828 833 10.1016/j.bbrc.2013.12.049 24342619 42. Kanasaki K. Shi S. Kanasaki M. He J. Nagai T. Nakamura Y. Ishigaki Y. Kitada M. Srivastava S.P. Koya D. Linagliptin-mediated DPP-4 inhibition ameliorates kidney fibrosis in streptozotocin-induced diabetic mice by inhibiting endothelial-to-mesenchymal transition in a therapeutic regimen Diabetes 2014 63 2120 2131 10.2337/db13-1029 24574044 43. Hendarto H. Inoguchi T. Maeda Y. Ikeda N. Zheng J. Takei R. Yokomizo H. Hirata E. Sonoda N. Takayanagi R. GLP-1 analog liraglutide protects against oxidative stress and albuminuria in streptozotocin-induced diabetic rats via protein kinase A-mediated inhibition of renal NAD(P)H oxidases Metab. Clin. Exp. 2012 61 1422 1434 10.1016/j.metabol.2012.03.002 22554832 44. Goncalves A. Marques C. Leal E. Ribeiro C.F. Reis F. Ambrosio A.F. Fernandes R. Dipeptidyl peptidase-IV inhibition prevents blood-retinal barrier breakdown, inflammation and neuronal cell death in the retina of type 1 diabetic rats Biochim. Biophys. Acta 2014 1842 1454 1463 10.1016/j.bbadis.2014.04.013 24769045 45. Goncalves A. Leal E. Paiva A. Teixeira Lemos E. Teixeira F. Ribeiro C.F. Reis F. Ambrosio A.F. Fernandes R. Protective effects of the dipeptidyl peptidase IV inhibitor sitagliptin in the blood-retinal barrier in a type 2 diabetes animal model Diabetes Obes. Metab. 2012 14 454 463 10.1111/j.1463-1326.2011.01548.x 22151893 46. Maeda S. Yamagishi S. Matsui T. Nakashima S. Ojima A. Maeda S. Nishino Y. Ishibashi Y. Yoshida Y. Yamakawa R. Beneficial effects of vildagliptin on retinal injury in obese type 2 diabetic rats Ophthalmic Res. 2013 50 221 226 10.1159/000354116 24081217 47. Hernandez C. Bogdanov P. Corraliza L. Garcia-Ramirez M. Sola-Adell C. Arranz J.A. Arroba A.I. Valverde A.M. Simo R. Topical Administration of GLP-1 receptor agonists prevents retinal neurodegeneration in experimental diabetes Diabetes 2016 65 172 187 10.2337/db15-0443 26384381 48. Jin H.Y. Liu W.J. Park J.H. Baek H.S. Park T.S. Effect of dipeptidyl peptidase-IV (DPP-IV) inhibitor (Vildagliptin) on peripheral nerves in streptozotocin-induced diabetic rats Arch. Med. Res. 2009 40 536 544 10.1016/j.arcmed.2009.09.005 20082866 49. Bianchi R. Cervellini I. Porretta-Serapiglia C. Oggioni N. Burkey B. Ghezzi P. Cavaletti G. Lauria G. Beneficial effects of PKF275–055, a novel, selective, orally bioavailable, long-acting dipeptidyl peptidase IV inhibitor in streptozotocin-induced diabetic peripheral neuropathy J. Pharmacol. Exp. Ther. 2012 340 64 72 10.1124/jpet.111.181529 21984837 50. Himeno T. Kamiya H. Naruse K. Harada N. Ozaki N. Seino Y. Shibata T. Kondo M. Kato J. Okawa T. Beneficial effects of exendin-4 on experimental polyneuropathy in diabetic mice Diabetes 2011 60 2397 2406 10.2337/db10-1462 21810596 51. Jolivalt C.G. Fineman M. Deacon C.F. Carr R.D. Calcutt N.A. GLP-1 signals via ERK in peripheral nerve and prevents nerve dysfunction in diabetic mice Diabetes Obes. Metab. 2011 13 990 1000 10.1111/j.1463-1326.2011.01431.x 21635674 52. Ta N.N. Schuyler C.A. Li Y. Lopes-Virella M.F. Huang Y. DPP-4 (CD26) inhibitor alogliptin inhibits atherosclerosis in diabetic apolipoprotein E-deficient mice J. Cardiovasc. Pharmacol. 2011 58 157 166 10.1097/FJC.0b013e31821e5626 21558879 53. Takai S. Sakonjo H. Jin D. Significance of vascular dipeptidyl peptidase-4 inhibition on vascular protection in Zucker diabetic fatty rats J. Pharmacol. Sci. 2014 125 386 393 10.1254/jphs.14052FP 25030743 54. Tang S.T. Zhang Q. Tang H.Q. Wang C.J. Su H. Zhou Q. Wei W. Zhu H.Q. Wang Y. Effects of glucagon-like peptide-1 on advanced glycation endproduct-induced aortic endothelial dysfunction in streptozotocin-induced diabetic rats: Possible roles of Rho kinase- and AMP kinase-mediated nuclear factor κB signaling pathways Endocrine 2016 53 107 116 10.1007/s12020-015-0852-y 26758998 55. Hattori S. Sitagliptin reduces albuminuria in patients with type 2 diabetes Endocr. J. 2011 58 69 73 10.1507/endocrj.K10E-382 21206136 56. Mori H. Okada Y. Arao T. Tanaka Y. Sitagliptin improves albuminuria in patients with type 2 diabetes mellitus J. Diabetes Investig. 2014 5 313 319 10.1111/jdi.12142 24843780 57. Scirica B.M. Bhatt D.L. Braunwald E. Steg P.G. Davidson J. Hirshberg B. Ohman P. Frederich R. Wiviott S.D. Hoffman E.B. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus N. Engl. J. Med. 2013 369 1317 1326 10.1056/NEJMoa1307684 23992601 58. Groop P.H. Cooper M.E. Perkovic V. Emser A. Woerle H.J. von Eynatten M. Linagliptin lowers albuminuria on top of recommended standard treatment in patients with type 2 diabetes and renal dysfunction Diabetes Care 2013 36 3460 3468 10.2337/dc13-0323 24026560 59. Fujita H. Taniai H. Murayama H. Ohshiro H. Hayashi H. Sato S. Kikuchi N. Komatsu T. Komatsu K. Komatsu K. DPP-4 inhibition with alogliptin on top of angiotensin II type 1 receptor blockade ameliorates albuminuria via up-regulation of SDF-1α in type 2 diabetic patients with incipient nephropathy Endocr. J. 2014 61 159 166 10.1507/endocrj.EJ13-0305 24225429 60. Zavattaro M. Caputo M. Sama M.T. Mele C. Chasseur L. Marzullo P. Pagano L. Mauri M.G. Ponziani M.C. Aimaretti G. One-year treatment with liraglutide improved renal function in patients with type 2 diabetes: A pilot prospective study Endocrine 2015 50 620 626 10.1007/s12020-014-0519-0 25572181 61. Zhang H. Zhang X. Hu C. Lu W. Exenatide reduces urinary transforming growth factor-β1 and type IV collagen excretion in patients with type 2 diabetes and microalbuminuria Kidney Blood Press. Res. 2012 35 483 488 10.1159/000337929 22687869 62. Marso S.P. Daniels G.H. Brown-Frandsen K. Kristensen P. Mann J.F. Nauck M.A. Nissen S.E. Pocock S. Poulter N.R. Ravn L.S. Liraglutide and cardiovascular outcomes in type 2 diabetes N. Engl. J. Med. 2016 10.1056/NEJMoa1603827 27295427 63. Ott C. Raff U. Schmidt S. Kistner I. Friedrich S. Bramlage P. Harazny J.M. Schmieder R.E. Effects of saxagliptin on early microvascular changes in patients with type 2 diabetes Cardiovasc. Diabetol. 2014 13 19 10.1186/1475-2840-13-19 24423149 64. Varadhan L. Humphreys T. Hariman C. Walker A.B. Varughese G.I. GLP-1 agonist treatment: Implications for diabetic retinopathy screening Diabetes Res. Clin. Pract. 2011 94 e68 e71 10.1016/j.diabres.2011.08.017 21906831 65. Varadhan L. Humphreys T. Walker A.B. Varughese G.I. The impact of improved glycaemic control with GLP-1 receptor agonist therapy on diabetic retinopathy Diabetes Res. Clin. Pract. 2014 103 e37 e39 10.1016/j.diabres.2013.12.041 24456992 66. Jaiswal M. Martin C.L. Brown M.B. Callaghan B. Albers J.W. Feldman E.L. Pop-Busui R. Effects of exenatide on measures of diabetic neuropathy in subjects with type 2 diabetes: Results from an 18-month proof-of-concept open-label randomized study J. Diabetes Complic. 2015 29 1287 1294 10.1016/j.jdiacomp.2015.07.013 26264399 67. White W.B. Cannon C.P. Heller S.R. Nissen S.E. Bergenstal R.M. Bakris G.L. Perez A.T. Fleck P.R. Mehta C.R. Kupfer S. Alogliptin after acute coronary syndrome in patients with type 2 diabetes N. Engl. J. Med. 2013 369 1327 1335 10.1056/NEJMoa1305889 23992602 68. Green J.B. Bethel M.A. Armstrong P.W. Buse J.B. Engel S.S. Garg J. Josse R. Kaufman K.D. Koglin J. Korn S. Effect of sitagliptin on cardiovascular outcomes in type 2 diabetes N. Engl. J. Med. 2015 373 232 242 10.1056/NEJMoa1501352 26052984 69. Pfeffer M.A. Claggett B. Diaz R. Dickstein K. Gerstein H.C. Kober L.V. Lawson F.C. Ping L. Wei X. Lewis E.F. Lixisenatide in patients with type 2 diabetes and acute coronary syndrome N. Engl. J. Med. 2015 373 2247 2257 10.1056/NEJMoa1509225 26630143 70. Crajoinas R.O. Oricchio F.T. Pessoa T.D. Pacheco B.P. Lessa L.M. Malnic G. Girardi A.C. Mechanisms mediating the diuretic and natriuretic actions of the incretin hormone glucagon-like peptide-1 Am. J. Physiol. Ren. Physiol. 2011 301 F355 F363 10.1152/ajprenal.00729.2010 21593184 71. Fujita H. Morii T. Fujishima H. Sato T. Shimizu T. Hosoba M. Tsukiyama K. Narita T. Takahashi T. Drucker D.J. The protective roles of GLP-1R signaling in diabetic nephropathy: Possible mechanism and therapeutic potential Kidney Int. 2014 85 579 589 10.1038/ki.2013.427 24152968 72. Korner M. Stockli M. Waser B. Reubi J.C. GLP-1 receptor expression in human tumors and human normal tissues: Potential for in vivo targeting J. Nucl. Med. 2007 48 736 743 10.2967/jnumed.106.038679 17475961 73. Schlatter P. Beglinger C. Drewe J. Gutmann H. Glucagon-like peptide 1 receptor expression in primary porcine proximal tubular cells Regul. Pept. 2007 141 120 128 10.1016/j.regpep.2006.12.016 17276524 74. Yang J. Campitelli J. Hu G. Lin Y. Luo J. Xue C. Increase in DPP-IV in the intestine, liver and kidney of the rat treated with high fat diet and streptozotocin Life Sci. 2007 81 272 279 10.1016/j.lfs.2007.04.040 17583752 75. Liu W.J. Xie S.H. Liu Y.N. Kim W. Jin H.Y. Park S.K. Shao Y.M. Park T.S. Dipeptidyl peptidase IV inhibitor attenuates kidney injury in streptozotocin-induced diabetic rats J. Pharmacol. Exp. Ther. 2012 340 248 255 10.1124/jpet.111.186866 22025647 76. Mima A. Hiraoka-Yamomoto J. Li Q. Kitada M. Li C. Geraldes P. Matsumoto M. Mizutani K. Park K. Cahill C. Protective effects of GLP-1 on glomerular endothelium and its inhibition by PKCβ activation in diabetes Diabetes 2012 61 2967 2979 10.2337/db11-1824 22826029 77. Cuevas S. Villar V.A. Jose P.A. Armando I. Renal dopamine receptors, oxidative stress, and hypertension Int. J. Mol. Sci. 2013 14 17553 17572 10.3390/ijms140917553 23985827 78. Udell J.A. Bhatt D.L. Braunwald E. Cavender M.A. Mosenzon O. Steg P.G. Davidson J.A. Nicolau J.C. Corbalan R. Hirshberg E.B. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes and moderate or severe renal impairment: Observations from the SAVOR-TIMI 53 Trial Diabetes Care 2015 38 696 705 10.2337/dc14-1850 25552421 79. Groop P.H. Cooper M.E. Perkovic V. Sharma K. Schernthaner G. Haneda M. Hocher B. Gordat M. Cescutti J. Woerle H.J. Dipeptidyl peptidase-4 inhibition with linagliptin and effects on hyperglycaemia and albuminuria in patients with type 2 diabetes and renal dysfunction: Rationale and design of the MARLINA-T2D trial Diabetes Vasc. Dis. Res. 2015 12 455 462 10.1177/1479164115579002 26224765 80. Von Scholten B.J. Hansen T.W. Goetze J.P. Persson F. Rossing P. Glucagon-like peptide 1 receptor agonist (GLP-1 RA): Long-term effect on kidney function in patients with type 2 diabetes J. Diabetes Complicat. 2015 29 670 674 10.1016/j.jdiacomp.2015.04.004 25935863 81. Cai J. Boulton M. The pathogenesis of diabetic retinopathy: Old concepts and new questions Eye 2002 16 242 260 10.1038/sj.eye.6700133 12032713 82. Goncalves A. Ambrosio A.F. Fernandes R. Regulation of claudins in blood-tissue barriers under physiological and pathological states Tissue Barriers 2013 1 e24782 10.4161/tisb.24782 24665399 83. Kim D.I. Park M.J. Choi J.H. Lim S.K. Choi H.J. Park S.H. Hyperglycemia-induced GLP-1R downregulation causes RPE cell apoptosis Int. J. Biochem. Cell Biol. 2015 59 41 51 10.1016/j.biocel.2014.11.018 25483438 84. Goncalves A. Lin C.M. Muthusamy A. Fontes-Ribeiro C. Ambrosio A.F. Abcouwer S.F. Fernandes R. Antonetti D.A. Protective Effect of a GLP-1 Analog on ischemia-reperfusion induced blood-retinal barrier breakdown and inflammation Investig. Ophthalmol. Vis. Sci. 2016 57 2584 2592 10.1167/iovs.15-19006 27163772 85. Sango K. Utsunomiya K. Efficacy of glucagon-like peptide-1 mimetics for neural regeneration Neural Regen. Res. 2015 10 1723 1724 10.4103/1673-5374.169611 26807090 86. Tsukamoto M. Niimi N. Sango K. Takaku S. Kanazawa Y. Utsunomiya K. Neurotrophic and neuroprotective properties of exendin-4 in adult rat dorsal root ganglion neurons: Involvement of insulin and RhoA Histochem. Cell Biol. 2015 144 249 259 10.1007/s00418-015-1333-3 26026990 87. Yorek M.A. Coppey L.J. Gellett J.S. Davidson E.P. Sensory nerve innervation of epineurial arterioles of the sciatic nerve containing calcitonin gene-related peptide: Effect of streptozotocin-induced diabetes Exp. Diabesity Res. 2004 5 187 193 10.1080/15438600490486732 15512786 88. Tsuboi K. Mizukami H. Inaba W. Baba M. Yagihashi S. The dipeptidyl peptidase IV inhibitor vildagliptin suppresses development of neuropathy in diabetic rodents: Effects on peripheral sensory nerve function, structure and molecular changes J. Neurochem. 2016 136 10.1111/jnc.13439 26603140 89. Davidson E.P. Coppey L.J. Dake B. Yorek M.A. Treatment of streptozotocin-induced diabetic rats with alogliptin: Effect on vascular and neural complications Exp. Diabetes Res. 2011 2011 810469 10.1155/2011/810469 21822422 90. Hirano T. Mori Y. Anti-atherogenic and anti-inflammatory properties of glucagon-like peptide-1, glucose-dependent insulinotropic polypepide, and dipeptidyl peptidase-4 inhibitors in experimental animals J. Diabetes Investig. 2016 7 Suppl. 1 80 86 10.1111/jdi.12446 27186361 91. Terasaki M. Nagashima M. Nohtomi K. Kohashi K. Tomoyasu M. Sinmura K. Nogi Y. Katayama Y. Sato K. Itoh F. Preventive effect of dipeptidyl peptidase-4 inhibitor on atherosclerosis is mainly attributable to incretin’s actions in nondiabetic and diabetic apolipoprotein E-null mice PLoS ONE 2013 8 1223 10.1371/journal.pone.0070933 23967137 92. Nagashima M. Watanabe T. Terasaki M. Tomoyasu M. Nohtomi K. Kim-Kaneyama J. Miyazaki A. Hirano T. Native incretins prevent the development of atherosclerotic lesions in apolipoprotein E knockout mice Diabetologia 2011 54 2649 2659 10.1007/s00125-011-2241-2 21786155 93. Nogi Y. Nagashima M. Terasaki M. Nohtomi K. Watanabe T. Hirano T. Glucose-dependent insulinotropic polypeptide prevents the progression of macrophage-driven atherosclerosis in diabetic apolipoprotein E-null mice PLoS ONE 2012 7 1223 10.1371/journal.pone.0035683 22536426 94. Shah Z. Kampfrath T. Deiuliis J.A. Zhong J. Pineda C. Ying Z. Xu X. Lu B. Moffatt-Bruce S. Durairaj R. Long-term dipeptidyl-peptidase 4 inhibition reduces atherosclerosis and inflammation via effects on monocyte recruitment and chemotaxis Circulation 2011 124 2338 2349 10.1161/CIRCULATIONAHA.111.041418 22007077 95. Arakawa M. Mita T. Azuma K. Ebato C. Goto H. Nomiyama T. Fujitani Y. Hirose T. Kawamori R. Watada H. Inhibition of monocyte adhesion to endothelial cells and attenuation of atherosclerotic lesion by a glucagon-like peptide-1 receptor agonist, exendin-4 Diabetes 2010 59 1030 1037 10.2337/db09-1694 20068138 96. Kim J.Y. Yang S. Lee J.I. Chang M.J. Cardiovascular effect of incretin-based therapy in patients with type 2 diabetes mellitus: Systematic review and meta-analysis PLoS ONE 2016 11 1223 10.1371/journal.pone.0153502 27078018 97. Mita T. Katakami N. Yoshii H. Onuma T. Kaneto H. Osonoi T. Shiraiwa T. Kosugi K. Umayahara Y. Yamamoto T. Alogliptin, a Dipeptidyl Peptidase 4 Inhibitor, Prevents the Progression of Carotid Atherosclerosis in Patients With Type 2 Diabetes: The Study of Preventive Effects of Alogliptin on Diabetic Atherosclerosis (SPEAD-A) Diabetes Care 2016 39 139 148 10.2337/dc15-0781 26628419 98. Mita T. Katakami N. Shiraiwa T. Yoshii H. Onuma T. Kuribayashi N. Osonoi T. Kaneto H. Kosugi K. Umayahara Y. Sitagliptin attenuates the progression of carotid intima-media thickening in insulin-treated patients with type 2 diabetes: The Sitagliptin Preventive Study of Intima-Media Thickness Evaluation (SPIKE): A randomized controlled trial Diabetes Care 2016 39 455 464 10.2337/dc15-2145 26822324 99. Song X. Jia H. Jiang Y. Wang L. Zhang Y. Mu Y. Liu Y. Anti-atherosclerotic effects of the glucagon-like peptide-1 (GLP-1) based therapies in patients with type 2 Diabetes Mellitus: A meta-analysis Sci. Rep. 2015 5 10202 10.1038/srep10202 26111974

PMC005xxxxxx/PMC5000622.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081224ijms-17-01224ArticleGenetic Mapping of a Major Resistance Gene to Pea Aphid (Acyrthosipon pisum) in the Model Legume Medicago truncatula Kamphuis Lars G. 12*Guo Su-Min 13Gao Ling-Ling 1Singh Karam B. 12Maffei Massimo Academic Editor1 Commenwealth Scientific and Industrial Research Organisation, Agriculture and Food, 147 Underwood Avenue, Floreat, WA 6014, Australia; [email protected] (S.-M.G.); [email protected] (L.-L.G.); [email protected] (K.B.S.)2 University of Western Australia Insititute of Agriculture, 35 Stirling Highway, Crawley, WA 6009, Australia3 Boyce Thompson Institute for Plant Research, Ithaca, NY 14853, USA* Correspondence: [email protected]; Tel.: +61-8-9333-632029 7 2016 8 2016 17 8 122421 6 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Resistance to the Australian pea aphid (PA; Acyrthosiphon pisum) biotype in cultivar Jester of the model legume Medicago truncatula is mediated by a single dominant gene and is phloem-mediated. The genetic map position for this resistance gene, APR (Acyrthosiphon pisum resistance), is provided and shows that APR maps 39 centiMorgans (cM) distal of the A. kondoi resistance (AKR) locus, which mediates resistance to a closely related species of the same genus bluegreen aphid (A. kondoi). The APR region on chromosome 3 is dense in classical nucleotide binding site leucine-rich repeats (NLRs) and overlaps with the region harbouring the RAP1 gene which confers resistance to a European PA biotype in the accession Jemalong A17. Further screening of a core collection of M. truncatula accessions identified seven lines with strong resistance to PA. Allelism experiments showed that the single dominant resistance to PA in M. truncatula accessions SA10481 and SA1516 are allelic to SA10733, the donor of the APR locus in cultivar Jester. While it remains unclear whether there are multiple PA resistance genes in an R-gene cluster or the resistance loci identified in the other M. truncatula accessions are allelic to APR, the introgression of APR into current M. truncatula cultivars will provide more durable resistance to PA. barrel mediclegumesinsect resistanceresistance gene ==== Body 1. Introduction Sap-sucking insects such as aphids, psyllids, scales and whiteflies cause significant damage in agricultural crops throughout the world. Damage is caused by direct feeding from the phloem sap as well as vectoring viruses, with aphids transmitting over 50% of all plant viruses [1]. Sap-sucking insects have a close association with their host and feed from a single cell type, the phloem sieve element. Sap-sucking insects have developed the ability to disguise their presence and/or suppress plant defences, ultimately leading to the establishment of a successful feeding site [2,3]. In recent years an increased research focus on studying plant—sap-sucking insect interactions has occurred, resulting in the identification of several sap-sucking insect resistance loci [4,5] and an improved understanding of the molecular mechanisms of basal defense as well as gene mediated resistance to sap-sucking insects is emerging [5]. The evolutionary origins of recognition of attackers of plants mainly stems from studies involving plant pathogens rather than insects and is better known as the plants innate immune system [6]. Recognition of an attacker often occurs through resistance (R) gene products which recognize specific attacker-derived product(s) and upon recognition mount a defence response. While these R-genes mediate resistance to a variety of different pathogens and pests, their architecture is highly similar and includes one of the following conserved motifs: Nucleotide binding site, leucine-rich repeat (NLRs) or serine/threonine protein kinase domains. This would imply that basic modes of recognition and subsequent signalling pathways that trigger the defence response have been retained through plant evolution and diversification [7,8]. An important advance in understanding R-gene mediated resistance to sap-sucking insects came from the discovery of the major dominant resistance gene Mi1.2, which confers resistance to three sap-sucking insects, being potato aphid (Macrosiphum euphorbiae), whiteflies (Bemisia tabaci) biotypes B and Q and psyllids (Bactericerca cockerelli) as well as three species of root-knot nematodes (Meloidogyne spp.) [9,10,11]. The second major R-gene identified and cloned was the Vat gene conferring resistance to cotton-melon aphid (Aphis gossypii) [12]. Mi1.2 and Vat belong to the largest class of R-genes encoding proteins with NLR motifs of the subclass with coiled-coiled (CC) motifs. The silencing of the Resistance Gene Candidate 2 (RGC2) cluster of NLR encoding genes in lettuce (Lactuca sativa) led to the loss of resistance to the lettuce root aphid (Phemphigus bursarius) [13]. In the model legume Medicago truncatula, single dominant resistance genes to other aphid species including bluegreen aphid (BGA; Acyrthosiphon kondoi), spotted alfalfa aphid (Therioaphis trifolii) and pea aphid (PA; Acyrthosiphon pisum) map to regions dense in these NLR encoding genes [14,15,16,17]. For both Mi1.2 and Vat as well as the single dominant resistance genes identified in M. truncatula resistance to aphids is exerted in the phloem, which shows that plants are able to utilize their innate immune systems to defend against parasitism of the phloem. Over the last decade M. truncatula has emerged as an excellent model plant to study plant insect interactions [5,18], with major dominant resistance genes identified to bluegreen aphid [14], spotted alfalfa aphid [15] and pea aphid [17,19]. Furthermore, quantitative trait loci (QTLs) controlling different aspects of aphid resistance including antibiosis, antixenosis and tolerance to BGA, PA, spotted alfalfa aphid and cowpea aphid have been identified [20,21,22]. Resistance to BGA, PA and spotted alfalfa aphid has been introgressed into the M. truncatula variety Jemalong (A17) through recurrent backcrosses to create a new aphid-resistant cultivar Jester [19,23]. Resistance to these three aphid species in Jester has been dissected over the last decade and it was shown that in all cases it involves antibiosis and antixenosis, with resistance exerted at the phloem [14,15,24]. Resistance in M. truncatula to PA was of particular interest as PA has been chosen by the international aphid genome consortium (IAGC) as the model aphid and there is a reference genome sequence [25] and other genomic resources available [26] as well as a number of distinct PA biotypes [27]. In the case of the Medicago-PA interaction in Jester, it was unclear whether resistance to BGA and PA was conferred by the same single dominant resistance gene, AKR (Acyrthosiphon kondoi resistance). In 2009, Guo et al. demonstrated that resistance to the Australian PA biotype was introgressed into the Jester background from a different donor than the resistance to BGA, thus there were two distinct resistance genes for the Australian PA biotype and BGA, where the resistance locus to the Australian PA biotype was termed APR for Acyrthosiphon pisum resistance [19]. In M. truncatula resistance to an European pea aphid biotype (PS01) is distinct from resistance to the Australian biotype. Resistance to the European biotype was identified in M. truncatula accession A17 which is moderately resistant to the Australian biotype [17]. Like APR mediated resistance RAP1 resistance is also exerted through the phloem. The genetic map position of RAP1 is on linkage group 3 in a region harbouring both serine-threonine kinase and NLR proteins. RAP1 mediated resistance causes 100% mortality to the European clone PS01 and is therefore different from APR mediated resistance since the antibiotic effect of APR on the Australian PA biotype shows no mortality, but rather a reduced reproductive rate [17,24]. Here we present a genetic map position for the APR locus and demonstrate that APR and RAP1 map to the same region on chromosome 3. We also report on a screen of additional M. truncatula germplasm for PA resistance and elaborate on the hypotheses that APR and RAP1 are two distinct genes tightly linked to one another in an R-gene cluster, or are alternative alleles of the same locus. 2. Results 2.1. Resistance to Pea Aphid in the Cultivar Jester Is Controlled by a Single Dominant Gene Previous mapping data suggested that PA resistance in Jester was linked to that of bluegreen aphid resistance mediated by the AKR locus on chromosome 3 [24]. To identify the genetic location of the APR locus, two genetic mapping populations were developed between Jester and A20, a wide cross as well as Jester and A17, a narrow cross. Molecular markers developed by the M. truncatula community [28,29,30], were screened for polymorphisms between the parents for each population (Table S1). A total of 129 F2 individuals were genotyped with 15 molecular markers polymorphic between Jester and A20. This resulted in the construction of a genetic linkage map for chromosome 3 spanning 100.9 centiMorgans (cM) with an average interval size of 7.2 cM. Seed was collected for these 129 individuals and their F3 offspring (n = 12 per F3 family) was infested with PA to determine their PA resistance response and thus the F2 alleles for the APR locus. This determined that the PA resistance locus APR is located between markers h2_39a22a and h2_180m21a spanning a 12.1 cM interval (Figure 1). Jester and A17 are 89% identical in their genome organisation [19] with Jester mainly having a large insertion from different donors on chromosome 3. Therefore, the chance to identify recombinants in the APR region of interest from a cross derived between Jester and A17 is higher than that from a cross derived between Jester and A20; thus, 384 F2 individuals of the narrow cross derived between Jester and A17 were genotyped with eight polymorphic markers near the region of APR to identify individuals with recombination events around the APR locus. This identified a total of 26 individuals with recombination events in the APR region of interest and their F3 progeny (n = 12 per F3 family) were infested with PA to determine their resistance status. As shown in Figure 1 the region of interest for the APR locus in the Jester × A17 cross spans 13.4 cM between markers MTIC51 and h2_151m16a. This region spans a physical distance of 3972.4 Kb in the M. truncatula v4.0 genome assembly of accession A17, which harbours a cluster of classical nucleotide-binding site leucine-rich repeats (NLR) resistance genes, including the RAP1 resistance gene to the European PA clone LS01 [17], but not the region where the bluegreen aphid resistance gene AKR has been mapped [14]. 2.2. Screening of M. truncatula Accessions for Additional Sources of PA Resistance With both APR and RAP1 located in an NLR cluster on chromosome 3, we wanted to determine whether additional major PA resistance genes to the Australian PA biotype exist besides APR and perhaps with a more striking lethal resistance as conferred by RAP1 to the European PA biotype LS01. Therefore, additional lines of M. truncatula were screened for aphid performance and plant damage. Thirty-five accessions of the South Australian Research and Development Institute (SARDI) M. truncatula core collection, which represent the major clades in the phylogenetic tree of the SARDI core accessions [31] were selected to evaluate PA resistance performance. These included accessions A20, Cyprus and Borung, previously identified as being highly susceptible to PA, A17 which is moderately resistant, as well as Jester and Caliph which are highly resistant to PA [32]. Plant damage and aphid populations were monitored over a 28-day period. One of the typical aphid infestation phenotypes in M. truncatula following infestation with PA is necrotic flecks on local leaves [17,24]; however this was only observed in M. truncatula accessions Jester and A17. No lethal resistance to PA was observed and all accessions showed varying degrees of stunting and wilting, with damage symptoms appearing as yellowing patches or leaf chlorosis surrounding the aphid infestation sites within 9 days after infestation. Nine accessions including two resistant controls (Jester and Caliph) were resistant and survived PA infestation after 28 days post infestation (dpi) and went on to flower and set seed, with the exception of one individual of accession SA27063 (Table 1). The remaining 26 accessions succumbed to the PA infestation, with 15 accessions including susceptible controls (Borung and A20) with higher plant damage scores than the moderately resistant accession A17 (Table 1). In a subsequent experiment the nine resistant accessions and five highly susceptible accessions from the initial screen were infested to confirm their resistance response to PA infestation with A17 included as a moderately resistant control. Starting with the initial two adult apterous aphids, PA colony density on all susceptible accessions peaked around 12 dpi; thereafter, the plants succumbing to PA infestation by 15 dpi. PA population density on A17 plants, the moderately resistant accession, reached the peak around 15 dpi (Table S2), whereas aphid populations were the largest at 21 dpi on the resistant accessions and declined thereafter at 24 dpi (Table S2). Plant damage on resistant accessions SA1516, SA28645, SA10481, SA10733, Jester and SA11753 remained stable from 21 dpi onwards with an average score of 3.4 (Table S3). There were some notable differences in the population sizes of PA on the different resistant accessions with a notably lower population density on SA1516 and SA10481 compared to Jester. In a follow-up short-term infestation experiment the performance of PA nymphs over a four-day period was observed, and this reflected the plant damage and aphid densities seen in the long term experiments (Figure 2). The PA nymph population had a significantly lower mean relative growth rate (MRGR) on Jester, SA10733, SA1516 and SA10481 compared to the moderately resistant A17, which, in turn, had a significantly lower MRGR compared to the highly susceptible accessions A20 and Cyprus (Figure 2a) (Tukey Kramer HSD test; p < 0.05). No significant differences between the accessions were found for the survivorship of PA nymphs over this four-day period (Figure 2b) (Tukey Kramer HSD test; p < 0.05). 2.3. Resistance in M. truncatula Accessions SA10733 and SA10481 Is Controlled by Single Dominant Gene SA1516 and SA10481 had the lowest average plant damage scores, albeit similar resistance phenotype to Jester and SA10733, the donor of APR in cultivar Jester. Moreover, notably lower PA population densities on accessions SA1516 and SA10481 were observed in the long-term experiments. Therefore F2 populations were generated between the resistant accessions SA10733 and SA10481 and the highly susceptible accession A20 to determine the genetic control underlying the PA resistance in these accessions. Phenotyping of 264 and 355 F2 individuals of the SA10733 × A20 and SA10481 × A20 showed a Mendelian segregation ratio of 3:1 for PA resistance in both populations (Table 2). To determine whether the single dominant resistance in SA10481 was allelic to that of SA10733 and/or SA1516, crosses were generated and F2 individuals for three crosses evaluated for their resistance to PA. As shown in Table 3 no susceptible individuals were identified, for any of the 535 individuals assayed, whereas the susceptible controls and moderately resistant controls behaved as seen in previous experiments. Thus the single dominant resistance in SA1516 and SA10481 and SA10733 are either alleles of the same gene (e.g., APR) or genes in a tightly linked resistance gene cluster. 3. Discussion Previously, we have characterised PA resistance in the M. truncatula cultivar Jester, which also harbours resistance to bluegreen aphid [24]. The biology of the resistance to both aphid species in this cultivar shared similarities with resistances occurring at the phloem level and requires an intact plant and involves a combination of antibiosis, antixenosis and plant tolerance [14,24]. However, the donor for bluegreen aphid resistance (accession SA1499) was a different donor than that of PA resistance (accession SA10733), thus resistance to both aphids are controlled by distinct single dominant resistance genes with the PA resistance locus tentatively named APR [19]. Here we demonstrated that resistance to PA mapped 39 cM distal of the flanking markers for the bluegreen aphid resistance locus AKR (h2_6g9b and 004H01) on chromosome 3 in a region rich in classical NLR type of resistance gene (Figure 1). Moreover, the region that contains APR in the genetic background of Jester spans the same region as the region harbouring RAP1 to the European PA biotype LS01 in the genetic background of A17 [17]. This could mean that APR and RAP1 are either two different alleles of the same orthologous gene, or, alternatively, two different genes in a NLR cluster of resistance genes. Further fine-mapping will be achieved in future work by generating re-sequencing data for cultivar Jester to identify single nucleotide polymorphisms (SNPs) or insertions/deletions (indels) in the APR region with the 26 recombinant F3 families. This would narrow-down the region of interest further and allow a map-based cloning approach for the APR locus. Similarly, the use of the Medicago HapMap resources [33] that contains re-sequencing data for DZA315 would allow the identification of SNPs and indels to generate novel markers for further fine-mapping of the RAP1 locus. Screening of diverse M. truncatula accessions with eight different European biotypes has previously been conducted by Kanvil and colleagues [27] and showed a range of differences in performance of the different biotypes across 23 M. truncatula accessions. They demonstrated that aphid virulence and host resistance were strongly dependent on the genotype of both the aphid and the host where diverse host-specific PA performance and biotype specific resistance in M. truncatula were observed. In Australia, there is currently only one biotype present and in contrast to the study by Kanvil et al. [27], no lethal resistance to the Australian biotype was identified in M. truncatula germplasm. Despite this result, seven new accessions were identified as being resistant to PA at a similar level to SA10733 and Jester both harbouring the APR gene, with notably lower PA population densities on accessions SA1516 and SA10481 compared to current cultivar Jester (Figure 2, Table 1). To determine the genetic control of PA resistance in the resistant accessions, crosses were generated to the susceptible A20 and phenotyping of the F2 populations showed that resistance segregated in a Mendelian fashion for a single dominant gene (Table 2), raising the question whether the resistance identified in these accessions were allelic to APR, a gene somewhat linked to APR or an unlinked gene. Out of the 494 F2 individuals phenotyped none of them showed susceptibility, which suggests that the single dominant resistance in SA1516, SA10481 and SA10733 are either alleles of the same gene (e.g., APR) or genes in a tightly linked resistance gene cluster. The latter could be a valid hypothesis as the RAP1 gene is also located in the same region on chromosome 3, and this region contains a suite of NLR resistance genes. The RAP1 gene in M. truncatula provides race-specific resistance to pea aphid biotype PS01 but not to biotype LL01 [17]. Furthermore, it has been shown that different PA biotypes (both sexual and asexual clones) differ in their performance on a range of M. truncatula accessions, including Jester and A17 plants [27]. Another PA biotype, N116, was virulent on RAP1 genotypes like biotype LL01 as well as on a wide range of other cultivars and wild M. truncatula genotypes [27]. On the contrary, PS01 was avirulent on most of the M. truncatula accessions. The divergent performance of these PA biotypes allowed the determination of inheritance of aphid virulence, and it was demonstrated through a series of F1 progenies of clones N116 and PS01 that the RAP1 mediated resistance can be overcome by progeny from either selfing or reciprocal crosses [34]. This suggests that the annual sexual cycle in aphids can lead to the generation of novel genotypes, which might have increased or decreased virulence. In turn, M. truncatula has to adapt and develop new forms of resistance to PA. In other plant species, this adaptation to other forms of virulent pathogens/pests occurs according to the birth and death model of R genes where R-genes duplicate and diversify in gene clusters [35]. Further fine-mapping of the identified PA resistance loci would shed more light on whether this has occurred in M. truncatula in response to different PA biotypes. The identification of the APR resistance gene in M. truncatula cv. Jester is the fourth major aphid resistance gene in this genetic background (Figure 3), which also harbours resistance to bluegreen aphid conferred by genes AKR [14] and AIN [16] and spotted alfalfa aphid conferred by TTR [15]. Breeders introgressed resistance to bluegreen aphid and spotted alfalfa aphid into the genetic background of Jemalong A17 from various resistance sources [19,23]. Since the APR locus is located 10.5 cM distal of the flanking marker for TTR in the Jester x A20 population and thus somewhat linked to TTR, they coincidentally introduced resistance to PA as well (Figure 3). The wealth of M. truncatula genomic resources including a reference genome sequence for Jemalong A17 [36,37] and a genome sequence for the model aphid PA [25] makes the M. truncatula-PA system a great one to study plant-insect interactions and R gene specificity and evolution. Similarly, PA genomic datasets such as numerous Expressed Sequence Tags (EST) and transcriptome resources [38] and RNA interference methods to silence aphid genes [39,40] would complement the plant based studies and allow the identification of aphid effectors recognised by the resistance genes. The use of these resources and in addition to the advances in sequencing technologies and Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 should allow the development of new ways to identify essential PA genes to establish a feeding site and/or effectors recognized by the resistance locus and might lead to effective durable resistance to aphids. 4. Materials and Methods 4.1. Plants and Aphids Three genotypes of M. truncatula were mainly used being: Jester, A17 and A20. Genetic F2;3 mapping populations derived from crosses derived between both Jester and A20, and Jester and A17, were generated using a crossing procedure described by Thoquet et al [41] and used in this study for the genetic mapping and phenotyping for PA resistance. The M. truncatula core collection accessions were acquired from the South Australian Research and Development Institute (SARDI, Urrbrae, Australia). Accessions DZA315 and DZA045 were obtained from the Institut National de la Recherche Agronomique (INRA), Montpellier, France. Seeds were germinated and plants grown as described by Klingler et al. [16]. The aphid species used was PA collected in Western Australia and were reared on faba bean (Vicia faba), as described by Gao et al. [32]. 4.2. Plant Damage and PA Performance Tests To assess the performance of PA and plant feeding damage, two-week-old seedlings of M. truncatula lines A17, A20 and Jester as well as 129 F3 families (n = 12 per F3 family) of the Jester × A20 population and 26 F3 families (n = 12 per F3 family) of the Jester × A17 population were grown in separate 0.9 L pots and were infested with two apterous adult aphids. Similarly, the 35 accessions (Table 1) were screened for PA resistance in a glasshouse when two-week-old and infested with two apterous adult aphids. The screening of the 35 accessions was arranged in a randomized complete block design with three replicates per accession infested for 28 days. In all phenotyping experiments the aphids were allowed to develop, reproduce, and move freely among plants. Aphid population build-up and feeding damage on plants were assessed at a three-day interval from the third day up to 28 days post infestation using a scale from 1–5 and 0–5, respectively as described previously [20]. 4.3. Aphid Performance on Caged Leaves The survival and growth rate of PA were measured after four days on individual plants of each M. truncatula accession with ten replicates for each accession and the mean relative growth rate (MRGR) calculated as described by Gao et al. [32]. The proportion of aphids that survived and MRGR were compared using the Tukey-Kramer Honestly Significant Difference test with the JMP-IN 5.1 software (SAS Institute, Cary, NC, USA). 4.4. Genetic Mapping of PA Resistance in the Various Mapping Populations Genetic maps for the Jester × A20 and Jester × A17 mapping populations were generated using both microsatellite and gene-based markers generated by the Medicago research community. Previously we established linkage association with markers on linkage group 3 [24] and therefore markers were initially selected to be evenly distributed over linkage group 3 and were obtained from several published sources [28,29,30]. A total of 26 markers were characterised for the Jester × A20 (n = 129) and for Jester × A17 (n = 384) populations with the polymorphic markers for the respective populations listed in Table S1. Linkage group 3 was constructed for both mapping populations using a set of 15 and 8 markers for the Jester × A20 and the Jester × A17 population respectively, using Multipoint v1.2 (Institute of Evolution, Haifa University, Haifa, Israel) as described by Kamphuis et al. [42]. 4.5. Allelism Tests Pairwise crosses were made among SA10733, SA1516 and SA10481 to test the allelic status of the PA resistance in SA1516 and SA10481 as in Table 3. The seedlings of F2 from each cross with at least eight replicates of their respective parental genotypes and A20 were tested for PA resistance. Each three-to-four-week-old seedling was infested with two apterous adult PAs for 28 days. During this period, aphids were allowed to develop, reproduce and move freely. Aphid resistance were scored as either resistant or susceptible at 28 dpi. Susceptible plants die before 20 dpi and with overwhelming aphids around 12 dpi and then totally migrate to the other plants due to the death of the host plant; resistant plants are still surviving at 28 dpi and reasonably healthy. The appearance of parental lines and A20 was used as controls. Acknowledgments We would like to thank Elaine Smith and Jenny Reidy-Croft for technical assistance on the project. We would like to thank Jonathan Anderson, Andrew James and Silke Jacques for providing helpful comments on the manuscript. Su-Min Guo was supported by a CSIRO/China Scholarship Council fellowship. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1224/s1. Click here for additional data file. Author Contributions Lars G. Kamphuis, Ling-Ling Gao and Karam B. Singh conceived and designed the experiments; Lars G. Kamphuis and Su-Min Guo performed the experiments; Lars G. Kamphuis and Su-Min Guo analyzed the data; Lars G. Kamphuis and Karam B. Singh wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Genetic map position of the APR (Acyrthosiphon pisum resistance) locus conferring resistance to the Australian pea aphid biotype, covers the same region of interest as the region of interest for RAP1 conferring resistance to a European PA biotype in the genetic background A17. Figure 2 (a) Mean relative growth rate (MRGR) of pea aphid nymphs on nine Medicago truncatula accessions over four days. Values are mean and standard error of ten replicates. Accessions that do not share the same letters indicate significant differences in pea aphid MRGR from the other accessions by Tukey Kramer HSD test (p < 0.05); (b) Survivorship of pea aphid nymphs on nine M. truncatula accessions over four days. No significant differences were observed in survivorship by Tukey Kramer HSD test (p < 0.05). Figure 3 Overview of the major resistance genes identified in M. truncatula cv. Jester to three different aphid species. ijms-17-01224-t001_Table 1Table 1 Evaluation of 35 Medicago truncatula accessions from the South Australian Research and Development Institute (SARDI) core collection for resistance to an Australian biotype of pea aphid. Each value represents the mean and standard error (SE) of three biological replicates. For the aphid population build-up, the rating scale was as described by Gao et al. [32]. Accession Aphid Score 9 dpi (SE) Plant Score 15 dpi (SE) Plant Score 21 dpi (SE) Plant Survivorship 28 dpi Comment SA11753 2.5(0.3) 1.6(0.7) 3.0(0.0) 3/3 Resistant SA28645 2.5(0.9) 2.2(0.3) 3.0(0.1) 3/3 Resistant SA3047 1.8(0.6) 2.5(0.3) 3.1(0.1) 3/3 Resistant SA10481 2.8(0.4) 2.3(0.3) 3.3(0.2) 3/3 Resistant SA1516 1.7(0.2) 2.5(0.3) 3.6(0.5) 3/3 Resistant SA27192 1.7(0.2) 1.3(0.1) 3.6(0.6) 3/3 Resistant SA27063 2.3(0.3) 3.5(0.3) 3.6(0.4) 2/3 Resistant Caliph 1.8(0.3) 2.4(0.3) 3.8(0.4) 3/3 Resistant (control) Jester 1.5(0.3) 2.7(0.2) 3.9(0.2) 3/3 Resistant (control) SA25654 2.3(0.3) 2.0(0.6) 3.3(0.3) 0/3 Moderately susceptible SA18395 1.2(0.2) 2.1(0.6) 3.5(0.0) 0/3 Moderately susceptible SA8604 2.3(0.2) 2.0(0.6) 4.0(0.4) 0/3 Moderately susceptible SA9062 2.3(0.3) 2.3(0.3) 4.0(0.4) 0/3 Moderately susceptible SA30199 2.2(0.4) 2.3(0.2) 4.1(0.2) 0/3 Moderately susceptible SA3569 2.2(0.4) 2.2(0.3) 4.3(0.3) 0/3 Moderately susceptible SA10419 2.7(0.7) 4.3(0.6) 4.4(0.6) 0/3 Moderately susceptible DZA315 2.8(0.6) 4.0(0.1) 4.6(0.1) 0/3 Moderately susceptible SA17590 2.8(0.7) 3.0(0.6) 4.6(0.2) 0/3 Moderately susceptible A17 2.7(0.2) 3.0(0.5) 4.7(0.1) 0/3 Moderately susceptible (control) SA3919 1.7(0.4) 2.3(0.6) 4.7(0.2) 0/3 Susceptible SA24968 2.2(0.4) 3.3(0.7) 4.8(0.1) 0/3 Susceptible SA3054 2.8(0.4) 2.9(0.9) 4.8(0.1) 0/3 Susceptible SA8618 3.2(0.2) 3.3(0.3) 4.8(0.3) 0/3 Susceptible SA11734 2.5(0.3) 4.1(0.5) 4.9(0.1) 0/3 Susceptible SA9357 3.8(0.2) 4.3(0.2) 4.9(0.1) 0/3 Susceptible SA22323 3.3(0.3) 4.3(0.3) 5.0(0.0) 0/3 Susceptible SA7749 3.2(0.3) 4.6(0.1) 5.0(0.0) 0/3 Susceptible SA9710 2.5(0.3) 4.5(0.3) 5.0(0.0) 0/3 Susceptible SA9712 2.7(0.2) 4.4(0.2) 5.0(0.0) 0/3 Susceptible Cyprus 3.0(0.6) 4.5(0.3) 5.0(0.0) 0/3 Susceptible (control) Borung 3.3(0.2) 4.9(0.1) 5.0(0.0) 0/3 Susceptible (control) A20 3.7(0.2) 4.8(0.1) 5.0(0.0) 0/3 Susceptible (control) SA1499 3.3(0.2) 4.4(0.2) 5.0(0.0) 0/3 Susceptible DZA045 3.0(0.5) 4.8(0.1) 5.0(0.0) 0/3 Susceptible SA1489 3.7(0.2) 4.9(0.1) 5.0(0.0) 0/3 Susceptible ijms-17-01224-t002_Table 2Table 2 Segregation of resistance to pea aphid in resistant M. truncatula accession crossed with accession A20. Chi-square analysis for a single dominant Mendelian inheritance of resistance of the two F2 populations indicates single dominant, Mendelian inheritance of resistance to PA in both populations. Population Resistant: Susceptible Observed Expected χ2 p SA10733 × A20 200:64 198:66 0.081 0.776 SA10481 × A20 264:91 266:89 0.076 0.783 ijms-17-01224-t003_Table 3Table 3 Pairwise allelism test between resistant M. truncatula accessions. Chi-square analysis for two unlinked Mendelian dominant genes indicates the resistance genes are either allelic or tightly linked. Population Resistant: Susceptible Observed Expected χ2 p SA10733 × SA1516 144:0 135:9 9.6 0.0019 SA1516 × SA10481 100:0 93.75:6.25 6.667 0.010 SA10481 × SA10733 250:0 234.4:15.6 16.667 0.00005 ==== Refs References 1. Ng J.C.K. Perry K.L. Transmission of plant viruses by aphid vectors Mol. Plant Pathol. 2004 5 505 511 10.1111/j.1364-3703.2004.00240.x 20565624 2. Walling L.L. Avoiding effective defenses: Strategies employed by phloem-feeding insects Plant Physiol. 2008 146 859 866 10.1104/pp.107.113142 18316641 3. Will T. Furch A.C. Zimmermann M.R. How phloem-feeding insects face the challenge of phloem-located defenses Front. Plant Sci. 2013 4 336 10.3389/fpls.2013.00336 24009620 4. Kaloshian I. Gene-for-gene disease resistance: Bridging insect pest and pathogen defense J. Chem. Ecol. 2004 30 2419 2438 10.1007/s10886-004-7943-1 15724964 5. Kamphuis L.G. Zulak K. Gao L.-L. Anderson J.P. Singh K.B. Plant-aphid interactions with a focus on legumes Funct. Plant Biol. 2013 40 1271 1284 10.1071/FP13090 6. Dangl J.L. Horvath D.M. Staskawicz B.J. Pivoting the plant immune system from dissection to deployment Science 2013 341 746 751 10.1126/science.1236011 23950531 7. Jones J.D.G. Dangl J.L. The plant immune system Nature 2006 444 323 329 10.1038/nature05286 17108957 8. Eitas T.K. Dangl J.L. NB–LRR proteins: Pairs, pieces, perception, partners and pathways Curr. Opin. Plant Biol. 2010 13 472 477 10.1016/j.pbi.2010.04.007 20483655 9. Casteel C.L. Walling L.L. Paine T.D. Behavior and biology of the tomato psyllid, Bactericerca cockerelli , in response to the Mi-1.2 gene Entomol. Exp. Appl. 2006 121 67 72 10.1111/j.1570-8703.2006.00458.x 10. Nombela G. Williamson V.M. Muñiz M. The root-knot nematode resistance gene Mi-1.2 of tomato is responsible for resistance against the whitefly Bemisia tabaci Mol. Plant Microbe Interact. 2003 16 645 649 10.1094/MPMI.2003.16.7.645 12848430 11. Rossi M. Goggin F. Milligan S.B. Kaloshian I. Ullman D.E. Williamson V.M. The nematode resistance gene Mi of tomato confers resistance against the potato aphid Proc. Natl. Acad. Sci. USA 1998 95 9750 9754 10.1073/pnas.95.17.9750 9707547 12. Dogimont C. Chovelon V. Pauquet J. Boualem A. Bendahmane A. The Vat locus encodes for a CC–NBS–LRR protein that confers resistance to Aphis gossypii infestation and A. gossypii -mediated virus resistance Plant J. 2014 80 993 1004 10.1111/tpj.12690 25283874 13. Wroblewski T. Piskurewicz U. Tomczak A. Ochoa O. Michelmore R.W. Silencing of the major family of NBS–LRR encoding genes in lettuce results in the loss of multiple resistance specificities Plant J. 2007 51 803 818 10.1111/j.1365-313X.2007.03182.x 17587302 14. Klingler J. Creasy R. Gao L. Nair R.M. Calix A.S. Spafford Jacob H. Edwards O.R. Singh K.B. Aphid resistance in Medicago truncatula involves antixenosis and phloem-specific, inducible antibiosis, and maps to a single locus flanked by NBS–LRR resistance gene analogs Plant Physiol. 2005 137 1445 1455 10.1104/pp.104.051243 15778464 15. Klingler J.P. Edwards O.R. Singh K.B. Independent action and contrasting phenotypes of resistance genes against spotted alfalfa aphid and bluegreen aphid in Medicago truncatula New Phytol. 2007 173 630 640 10.1111/j.1469-8137.2006.01939.x 17244058 16. Klingler J.P. Nair R.M. Edwards O.R. Singh K.B. A single gene, AIN , in Medicago truncatula mediates a hypersensitive response to both bluegreen aphid and pea aphid, but confers resistance only to bluegreen aphid J. Exp. Bot. 2009 60 4115 4127 10.1093/jxb/erp244 19690018 17. Stewart S.A. Hodge S. Ismail N. Mansfield J.W. Feys B.J. Prosperi J.M. Huguet T. Ben C. Gentzbittel L. Powell G. The RAP1 gene confers effective, race-specific resistance to the pea aphid in Medicago truncatula independent of the hypersensitive reaction Mol. Plant Microbe Interact. 2009 22 1645 1655 10.1094/MPMI-22-12-1645 19888829 18. Edwards O.R. Singh K.B. Resistance to insect pests, what do legumes have to offer? Euphytica 2006 147 273 285 10.1007/s10681-006-3608-1 19. Guo S. Kamphuis L.G. Gao L.-L. Edwards O.R. Singh K.B. Two independent resistance genes in the Medicago truncatula cultivar Jester confer resistance to two different aphid species of the genus Acyrthosiphon Plant Signal. Behav. 2009 4 1 4 10.4161/psb.4.4.8190 19568332 20. Guo S. Kamphuis L.G. Gao L.-L. Klingler J.P. Lichtenzveig J. Edwards O.R. Singh K.B. Identification of distinct quantitative trait loci associated with defense against the closely related aphids Acyrthosiphon pisum and A. kondoi in Medicago truncatula J. Exp. Bot. 2012 63 3913 3922 10.1093/jxb/ers084 22442407 21. Kamphuis L.G. Gao L.-L. Singh K.B. Identification and characterization of resistance to cowpea aphid (Aphis craccivora Koch) in Medicago truncatula BMC Plant Biol. 2012 12 1224 10.1186/1471-2229-12-101 22759788 22. Kamphuis L.G. Lichtenzveig J. Peng K. Guo S. Klingler J.P. Siddique K. Gao L.L. Singh K.B. Physiology and genetics of resistance to spotted alfalfa aphid (Therioaphis trifolii ) in Medicago truncatula J. Exp. Bot. 2013 64 5157 5172 10.1093/jxb/ert305 24058162 23. Hill J.R. Jester Plant Var. J. 2000 13 40 24. Gao L.L. Klingler J.P. Anderson J.P. Edwards O.R. Singh K.B. Characterization of pea aphid resistance in Medicago truncatula Plant Physiol. 2008 146 996 1009 10.1104/pp.107.111971 18184733 25. IAGC Genome sequence of the pea aphid Acyrthosiphon pisum PLoS Biol. 2010 8 1224 10.1371/journal.pbio.1000313 26. Legeai F. Shigenobu S. Gauthier J.P. Colbourne J. Rispe C. Collin O. Richards S. Wilson A.C. Murphy T. Tagu D. AphidBase: A centralized bioinformatic resource for annotation of the pea aphid genome Insect Mol. Biol. 2010 2 5 12 10.1111/j.1365-2583.2009.00930.x 20482635 27. Kanvil S. Powell G. Turnbull C. Pea aphid biotype performance on diverse Medicago host genotypes indicates highly specific virulence and resistance functions Bull. Entomol. Res. 2014 104 689 701 10.1017/S0007485314000443 25375216 28. Choi H.K. Kim D. Uhm T. Limpens E. Lim H. Mun J.H. Kalo P. Penmetsa R.V. Seres A. Kulikova O. A sequence-based genetic map of Medicago truncatula and comparison of marker colinearity with M. sativa Genetics 2004 166 1463 1502 10.1534/genetics.166.3.1463 15082563 29. Gutierrez M.V. Vaz Patto M.C. Huguet T. Cubero J.I. Moreno M.T. Torres A.M. Cross-species amplification of Medicago truncatula microsatellites across three major pulse crops Theor. Appl. Genet. 2005 110 1210 1217 10.1007/s00122-005-1951-6 15806349 30. Mun J.H. Kim D.J. Choi H.K. Gish J. Debellé F. Mudge J. Denny R. Endré S. Saurat O. Dudez A.M. Distribution of microsatellites in the genome of Medicago truncatula : A resource of genetic markers that integrate genetic and physical maps Genetics 2006 172 2541 2555 10.1534/genetics.105.054791 16489220 31. Ellwood S.R. D’Souza N.K. Kamphuis L.G. Burgess T.I. Nair R.M. Oliver R.P. SSR analysis of the Medicago truncatula SARDI core collection reveals substantial diversity and unusual genotype dispersal throughout the Mediterranean basin Theor. Appl. Genet. 2006 112 977 983 10.1007/s00122-005-0202-1 16402186 32. Gao L.L. Horbury R. Nair R.M. Edwards O.R. Singh K.B. Characterization of resistance to multiple aphid species (Hemiptera: Aphididae) in Medicago truncatula Bull. Entomol. Res. 2007 97 41 48 10.1017/S0007485307004786 17298680 33. Kang Y. Sakiroglu M. Krom N. Stanton-Geddes J. Wang M. Lee Y.C. Young N.D. Udvardi M. Genome-wide association of drought-related and biomass traits with HapMap SNPs in Medicago truncatula Plant Cell Environ. 2015 38 1997 2011 10.1111/pce.12520 25707512 34. Kanvil S. Collins C.M. Powell G. Turnbull C.G.N. Cryptic virulence and avirulence alleles revealed by controlled sexual recombination in pea aphids Genetics 2015 199 581 593 10.1534/genetics.114.173088 25519896 35. Michelmore R.W. Meyers B.C. Clusters of resistance genes in plants evolve by divergent selection and a birth-and-death process Genome Res. 1998 8 1113 1130 9847076 36. Tang H. Krishnakumar V. Bidwell S. Rosen B. Chan A. Zhou S. Gentzbittel L. Childs K.L. Yandell M. Gundlach H. An improved genome release (version Mt4.0) for the model legume Medicago truncatula BMC Genom. 2014 15 1224 10.1186/1471-2164-15-312 24767513 37. Young N.D. Debellé F. Oldroyd G.E. Geurts R. Cannon S.B. Udvardi M.K. Benedito V.A. Mayer K.F. Gouzy J. Schoof H. The Medicago genome provides insight into the evolution of rhizobial symbioses Nature 2011 480 520 524 10.1038/nature10625 22089132 38. Gerardo N.M. Altincicek B. Anselme C. Atamian H. Barribeau S.M. de Vos M. Duncan E.I. Evans J.D. Gabaldon T. Ghanim M. Immunity and other defenses in pea aphids, acyrthosiphon pisum Genome Biol. 2010 11 R21 10.1186/gb-2010-11-2-r21 20178569 39. Mutti N.S. Louis J. Pappan L.K. Pappan K. Begum K. Chen M.-S. Park Y. Dittmer N. Marshall J. Reese J.C. A protein from the salivary glands of the pea aphid, acyrthosiphon pisum , is essential in feeding on a host plant Proc. Natl. Acad. Sci. USA 2008 105 9965 9969 10.1073/pnas.0708958105 18621720 40. Mutti N.S. Park Y. Reese J.C. Reeck G.R. RNAi knockdown of a salivary transcript leading to lethality in the pea aphid Acyrthosiphon pisum J. Insect Sci. 2006 6 38 10.1673/031.006.3801 20233093 41. Thoquet P. Ghérardi M. Journet E. Kereszt A. Ané J. Prosperi J.M. Huguet T. The molecular genetic linkage map of the model legume Medicago truncatula : An essential tool for comparative legume genomics and the isolation of agronomically important genes BMC Plant Biol. 2002 2 1224 10.1186/1471-2229-2-1 11825338 42. Kamphuis L.G. Lichtenzveig J. Oliver R.P. Ellwood S.R. Two alternative recessive quantitative trait loci influence resistance to spring black stem and leaf spot in Medicago truncatula BMC Plant Biol. 2008 8 1224 10.1186/1471-2229-8-30 18366746

PMC005xxxxxx/PMC5000623.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081225ijms-17-01225ReviewThe Metabolic Role of Gut Microbiota in the Development of Nonalcoholic Fatty Liver Disease and Cardiovascular Disease Sanduzzi Zamparelli Marco 1Compare Debora 1Coccoli Pietro 1Rocco Alba 1Nardone Olga Maria 1Marrone Giuseppe 2Gasbarrini Antonio 2Grieco Antonio 2Nardone Gerardo 2*Miele Luca 2*Haybaeck Johannes Academic Editor1 Department of Clinical Medicine and Surgery, Gastroenterology Unit, Federico II University of Naples, 80131 Napoli, Italy; [email protected] (M.S.Z.); [email protected] (D.C.); [email protected] (P.C.); [email protected] (A.R.); [email protected] (O.M.N.)2 Internal Medicine and Gastroenterology Area, Fondazione Policlinico Universitario A. Gemelli, Catholic University of Rome, 00168 Rome, Italy; [email protected] (G.M.); [email protected] (A.G.); [email protected] (A.G.)* Correspondence: [email protected] (L.M.); [email protected] (G.N.); Tel.: +39-06-3015-5451 (L.M. & G.N.)29 7 2016 8 2016 17 8 122507 6 2016 14 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The prevalence of metabolic disorders, such as type 2 diabetes (T2D), obesity, and non-alcoholic fatty liver disease (NAFLD), which are common risk factors for cardiovascular disease (CVD), has dramatically increased worldwide over the last decades. Although dietary habit is the main etiologic factor, there is an imperfect correlation between dietary habits and the development of metabolic disease. Recently, research has focused on the role of the microbiome in the development of these disorders. Indeed, gut microbiota is implicated in many metabolic functions and an altered gut microbiota is reported in metabolic disorders. Here we provide evidence linking gut microbiota and metabolic diseases, focusing on the pathogenetic mechanisms underlying this association. gut microbiotacardiovascular diseaseNAFLD ==== Body 1. Introduction According to the World Health Organization cardiovascular disease (CVD) remains the leading cause of death in Western countries, accounting for about 20 million deaths/year worldwide [1]. Nevertheless, the prevalence of metabolic diseases, such as type 2 diabetes (T2D), obesity, and non-alcoholic fatty liver disease (NAFLD), which are common risk factors for CVD, has dramatically increased over the last decades worldwide. Both genetic susceptibility and environmental factors contribute to cardio-metabolic pathogenesis. Unfortunately, despite the introduction of new extensive investigation on genetic determinants, such as large-scale genome-wide association study (GWAS), only a few cases can be assigned to genetic factors [2,3]. Thus, the environmental component seems to be primarily involved in CVD burden. The largest environmental exposure is food that we take into our intestine in kilogram quantities every day. However, there is an imperfect correlation between dietary habits and the development of metabolic disease. Diet, on the other hand, is known to deeply alter the microbiota composition in the gut. Therefore, recent evidence focused on the role of the gut microbiota in cardio-metabolic disorders and growing interest is aimed to modulate the gut microbiota as a therapeutic strategy. 2. Gut Microbiota The human gut harbors an enormously complex, dynamic, and vast microbial community that is composed mainly of bacteria, but also includes viruses, fungi, protozoa, and archaea [4]. The gut microbiota is estimated to consist of at least 1014 bacteria and archaea, including more than 1100 species and 150-fold more genes than our own host genomes [5]. The introduction of culture-independent methods to study the microbial community revealed the complexity of the gut microbiota. The human gut microbiota is dominated by bacteria belonging to four major phyla: Firmicutes, Bacteroidetes, Actinobacteria (representing >95% of the total microbiota), and Proteobacteria [6]. Interestingly, Actinobacteria and Proteobacteria are more abundant in childhood, while Firmicutes and Bacteroidetes are prevalent in adulthood. The fetal human gut was supposed to be sterile at birth; however, very recently, it has been reported that there is a passage of microbes between the mother and the fetus through the placenta. The most important colonization of the gastrointestinal tract starts suddenly after the birth and depends on delivery type, maternal flora, environment hygiene, and infant diet. During the first years of life, the gut microbiota composition is widely shifting and changeable, while it stabilizes when the infant reaches 1–3 years of age [7]. Microbial density increases from the proximal to the distal gut, and along the mucosal-lumen axis [8]. Similarly to bacterial density, microbial diversity also increases along the same axis. The composition of the gut microbiota is dynamically influenced by several host factors, including diet, lifestyle, antibiotics, and genetic background. In a C57B/L6J mouse model, the use of antibiotics, in early life, alters the host metabolism and adiposity by a modification of the microbiome [9]. Another study described that the maternal administration of antibiotics during the last six months before the pregnancy or the early infancy is related to reduced bacterial diversity of the feces of the neonate, decreased levels of Lactobacilli and Bifidobacteria, and increased risk of childhood obesity [10]. Diet seems to play a critical role, being linked to quantitative and qualitative modifications of the microbiota composition. A study comparing the gut microbiota of Italian children with those living in a rural African village showed that Bacteroidetes levels (mainly Prevotella) were higher in the stool of African children who consumed high amounts of plant polysaccharides while Enterobacteriaceae levels were higher in the stool of Italian children [11]. Recently, exercise has also been demonstrated to modulate gut microbiota composition, showing qualitative and quantitative modifications after running [12] in voluntary athletes. 3. Gut Microbiota and Energy Balance 3.1. Carbohydrate Metabolism 3.1.1. Animal Studies The gut microbes can benefit the host by extracting energies from otherwise non-digestible carbohydrates and plant polysaccharides via enzymes not encoded by humans [13]. Indeed, indigestible polysaccharides are fermented by colonic microbiota, leading to the generation of short-chain fatty acids (SCFAs) in the form of acetate (60%), propionate (25%), and butyrate (15%). SCFAs are readily absorbed in the colon, where butyrate is an important energy source for colonic epithelial cells, while acetate and propionate reach the liver and peripheral organs, where they are used as substrates for gluconeogenesis [14]. SCFAs can also regulate gene expression by acting as signaling molecules, by binding to G-protein-coupled receptors (GPCRs), such as GPR41 (free fatty acid receptor, FFAR3) and GPR43 (FFAR2) [15]. A recent study demonstrated that GPR43-deficient mice, even if fed with a low-fat diet, are obese, while mice overexpressing GPR43 in adipose tissue are lean, even if under a high-fat diet (HFD) [16]. SCFAs-dependent activation of GPR43 can modulate insulin signaling in the adipose tissue, thus preventing fat accumulation. Furthermore, SCFAs directly modulate intestinal gluconeogenesis. Propionate affected intestinal gluconeogenesis via the gut-brain neuronal circuit, involving GPR41 and butyrate through cyclic adenosine monophosphate (cAMP)-dependent pathway, independently of GPR43 [17]. As far as butyrate is concerned, it is able to regulate the appetite via the central nervous system, by stimulating the liberation of peptide YY (PYY) and the satietogenic hormones glucagon-like peptide 1 (GLP-1) from enteroendocrine L-cells in the distal small intestine and the colon [18]. PYY is an intestinal hormone [19] known for its ability to decrease intestinal transit rate and increase the harvest of energy from the diet, while GLP-1 improves adipocyte insulin sensitivity and remarkably reduces fat storage in adipose tissue [20]. Gut microbes can also control the metabolic activity of the host by affecting the composition and the abundance of certain bile acid species [21]. The cholic and chenodeoxycholic acids are produced in the liver from cholesterol, and are needed for the absorption of cholesterol, dietary fats, and fat-soluble vitamins. In the ileum microbes deconjugate these bile acids, which escape intestinal uptake, and are converted into secondary bile acids. Bile acids can also act as signaling molecules by binding cellular receptors, such as the bile-acid-synthesis controlling nuclear receptor farnesoid X receptor (FXR), GPCR, and the G protein-coupled bile acid receptor TGR5 [22]. While primary bile acids can impair glucose metabolism by binding FXR [23], secondary bile acids (deoxycholic and chenodeoxycholic acid) improve glucose homeostasis by binding TGR5 [24] (Figure 1). Furthermore, animal studies demonstrated that Akkermansia muciniphila concentrations are inversely correlated with the body weight and glucose tolerance [25]. Hansen et al. recently demonstrated that vancomycin administration in non-obese diabetic rats increased A. muciniphila levels and ameliorated glucose homeostasis [26]. 3.1.2. Human Studies In humans, a metagenome-wide association study demonstrated significant modifications of specific gut microbes and metabolic pathways in T2D patients [27]. The study was performed using stool samples of 344 Chinese patients and showed a diminution of butyrate-producing bacteria, such as Roseburia intestinalis and Faecalibacterium prausnitzii, and an abundance of Akkermansia muciniphila. Recently, an interesting study conducted in Europe on postmenopausal patients with normal, impaired, or diabetic glucose regulation, showed somewhat contradictory results in comparison to the Chinese study. Different techniques, but also ethnic and dietetic factors, can account for the discrepancy [28]. A previous, smaller study demonstrated higher levels of Lactobacillus spp. in T2D patients in comparison to healthy controls [29] and both Chinese and European studies showed enhanced concentration of Lactobacillus gasseri, Streptococcus mutans and certain Clostridiales, and lower levels of Roseburia intestinalis and Faecalibacterium prausnittzii, in the diabetic cohort. Moreover, certain antidiabetic medications can produce modification of the gut microbiota composition. A recent paper demonstrated that metformin administration resulted in increased levels of A. muciniphila, improvement of glucose tolerance, and reduced systemic inflammation [30,31]. In this light, the administration of prebiotics, such as oligofructose, increased A. muciniphila concentrations with various metabolically-beneficial outcomes [30]. In this way A. muciniphila reveals to be a promising candidate in the understanding of the complex role of intestinal microbes in metabolic disorders even if data are still somewhat controversial. In summary, evidence suggests that gut microbiota plays a critical role in the energy balance by fermentation of carbohydrates in SCFA. Gut microbiota composition appears to be involved in the pathogenesis of diabetes, but further interventional studies are needed to use it as a diagnostic and therapeutic tool. 3.2. Lipid Metabolism 3.2.1. Animal Studies It is supposed that HFD may affect epithelial integrity, leading to impaired intestinal permeability and, consequently, to metabolic endotoxemia and systemic inflammation [32]. Chronic exposition to low-dose lipopolysaccharide (LPS) in mice led to hepatic fat deposition, insulin resistance (IR), hyperlipidemia, adipose tissue macrophages infiltration, and obesity, similar as after feeding a HFD [33,34]. Furthermore, these effects were abolished when TLR4 (Toll-like receptor, directly involved in LPS binding) knockout mice were fed a HFD, or ob/ob mice and HFD-fed mice were treated with antibiotics or prebiotics [32]. Another complex diet-independent effect of microbiota on host metabolism is mediated by bacterial compounds, such as peptidoglycan and flagellin, that are able to activate inflammatory pathways [35]. For example, TLR5 is a pattern recognizer for flagellin, which can modulate intestinal homeostasis by activating various intracellular signaling pathways. TLR5−/− mice show a metabolic phenotype, exhibiting gut microbiome modifications, dyslipidemia, hypertension, IR, and obesity. Similarly to other studies, colonization of germ-free mice with TLR5−/− mice caecum microbes, led to a metabolic phenotype [36]. Studies in rodents genetically predisposed to obesity (ob/ob), first revealed an increase in the Firmicutes/Bacteroidetes ratio [37] and analogous differences were observed in the gut microbiota in human obesity [37,38]. Interestingly, it has been clearly shown that the obese phenotype is a transmissible trait, by transplanting the caecal microbiome from obese mice into germ-free mice who developed the same phenotype of the donor [39,40]. Similarly, when germ-free mice were colonized with a caecum-derived microbiome of conventional mice, the amount of the total body fat increased and insulin sensitivity decreased [41]. A recent paper showed that colonization of germ-free mice with fecal microbiota from twin pairs discordant for obesity resulted in an obese phenotype in mice receiving microbiota from the obese twin, while mice receiving microbiota from the lean twin showed less weight gain and adiposity [42]. In the same study, authors also showed that cohousing of ob/ob mice and lean mice led to the acquisition of the gut microbiota of the lean rodents by ob/ob mice but not to the acquisition by lean mice of the ob/ob mice microbiota. In addition, cohousing of ob/ob mice and lean mice fed with high vegetables and fruit, a low-fat diet (LFD), resulted in a greater acquisition of microbes of the lean mice by ob/ob rodents. A recent study showed that the increased levels of acetate in dysbiosis stimulated the parasympathetic nervous system which, in turn, enhanced the secretion of glucose-stimulated insulin and ghrelin, leading to hyperphagia and obesity [43]. Overall, these findings suggest an active role of gut microbiome in the development of obesity, and the major mechanism involved appears to be the more efficient energy harvest from food [44]. 3.2.2. Human Studies Several studies demonstrated that obese patients usually show chronic adipose tissue inflammation and that the obese-associated gut microbiome was related with a high production of pro-inflammatory cytokines [45]. In addition, several, but not all, human adult evidence showed a decrease of the diversity of the microbiota and an increased Firmicutes/Bacteroidetes ratio [46]. Weight loss and a LFD reversed the obese core microbiome versus the healthy microbiota, by increasing the relative proportion of Bacteroidetes spp. and the bacterial diversity [47]. Furthermore, after bariatric surgery, levels of Bacteroidetes and Prevotella negatively correlated with adiposity and energy harvest [48]. Recent data on the gut fungal species of obese humans showed a reduced family biodiversity in obese individuals but no modifications in the richness of the gut mycobiome between obese and healthy subjects [49]. In summary, animal and human evidence described an ‘obese core microbiome’ which seems to be involved in the pathogenesis of obesity. 3.3. Gut Microbiota and NAFLD The first evidence for a putative role of the gut microflora in NAFLD was suggested more than 20 years ago in patients with small intestinal bacterial overgrowth (SIBO) who were prone to develop NAFLD, and in a rat model with a blind intestinal loop in which hepatic injury was prevented by antibiotics [50]. An human study showed an increased gut permeability and prevalence of SIBO in NAFLD patients [51]. Recent studies focused on the increased endogenous ethanol production by the gut microbiota that is primary involved in NAFLD development [52,53]. Indeed, ob/ob mice showed an increase in breath ethanol amount, while normal ethanol levels were observed if mice were treated with neomycin [54]. Convincing evidence also demonstrated that the microbiota-dependent choline conversion into methylamines in strain 129S6 on a HFD decreases the bioavailability of choline mimicking the effects achieved by choline-deficient diets [55]. In this study, the authors observed that dietary choline depletion induced gut microbiota modification, and that Erysipelotrichi and Gammaproteobacteria levels were associated with changes in liver fat. A recent study demonstrated that differences in intestinal microbiota composition could explain the varying response to a HFD in mice [56]. In this study, two donor C57BL/6J mice were identified on the basis of their responses to a high-fat diet (HFD); although both groups showed similar body weight gain, one mouse, called the ‘responder’, displayed hyperglycemia and systemic inflammation, whereas the other, called a ‘non-responder’, was normoglycemic. When the microbiota of both groups were transplanted into germ-free mice which were then fed with the same HFD, only mice colonized with ‘responders’ microbiota showed hyperglycemia and fatty liver disease, in absence of systemic and hepatic inflammation. Moreover, an interventional study on ApoE−/− mice, a genetic model of dyslipidemia, intestinal inflammation, and steatohepatitis, showed that the administration of the probiotic VSL#3 improved insulin resistance, and reduced the aortic plaque extension, mesenteric adipose tissue inflammation, and steatohepatitis [57]. A human study demonstrated that the administration of Bifidobacterium longum with fructo-oligosaccharides, plus lifestyle modifications, were able to significantly reduce serum aspartate transaminase (AST) levels, tumor necrosis factor (TNF)-α, CRP (C-reactive protein), HOMA-IR (Homeostasis Assessment Model-Insulin Resistance), serum endotoxin, steatosis, and the non-alcoholic steatohepatitis (NASH) activity index when compared to lifestyle modification alone [58]. Dysbiosis-driven inflammatory response seems to play a major role in the pathogenesis of NAFLD. Intestinal microorganisms have highly conserved molecules, named “pathogen associated molecular patterns” (PAMPs) which are recognized specifically by pattern recognition receptors (PRRs), such as TLRs and nucleotide binding oligomerization domain-like receptors (NLRs). Several studies have shown that PRR stimulation can increase pro-inflammatory cytokines and chemokines by different intracellular signaling cascades, while commensal microbiota are able to counteract this inflammatory response by interfering in TLR-dependent nuclear factor kappa B (NF-κB) transcription [59,60,61]. The multimeric signaling platforms, called ‘inflammasomes’, appear to be primarily involved in the gut microbiota-driven liver steatosis and inflammation. The inflammasome, which is frequently identified by its first sensing molecules, such as NLRP6 and NLRP3 (NOD-like receptors, pyrin domain containing 6 and 3), have common pathways leading to interleukin (IL)-18 and IL-1 activation via caspase-1 activation. Inflammasome-deficient mice showed a reduction of Firmicutes and an increase of Bacteroidetes, associated with increased steatosis and inflammation via TLR4 and TLR9 activation, leading to enhanced hepatic TNF-transcription [62]. Co-housing of inflammasome-deficient mice with wild-type mice resulted in exacerbation of obesity and steatosis, concluding that NLRP6 and NLRP3 inflammasomes negatively regulate NAFLD/NASH progression. Furthermore, co-housing mice defective in TLR4 and TLR with inflammasome-deficient mice did not worsen NAFLD/NASH [63]. In summary, evidences suggest that dysbiosis may play a critical role in the development of NAFLD/NASH by metabolic and inflammatory pathways. 3.4. Cardiovascular Disease Gut microbiota composition is modified not only in diabetes, obesity, and NAFLD, with adverse cardiovascular outcomes, but also in hypertension. Indeed, animals with hypertension showed decreased gut bacterial richness and diversity with associated reduced levels of acetate and butyrate which negatively correlate with systemic inflammation [64]. Furthermore, several studies showed that the “metabolic endotoxemia” caused by the exposure to LPS promoted a systemic low-grade inflammation with an increased cardiovascular risk [65,66]. Indeed, the binding of TLR4 by LPS, and the activation of an immune response, triggered the release of proinflammatory molecules that promoted endothelial dysfunction, oxidation of low-density lipoproteins (LDLs), thrombogenesis, and the formation and rupture of the atherosclerotic plaque [66]. Population studies supported a relation between infection and CVD, in consideration of higher cardiovascular risk and blood pressure in patients affected by periodontal disease [67,68]. Mounting evidence in animals and in humans is accumulating showing that gut microbiota is associated with CVD [69]. One study failed to show any overall difference in fecal microbiota composition of patients with atherosclerosis, however, gut and atherosclerotic plaque microbiota showed several operational taxonomic units (OTUs) [70]. Indeed, atherosclerotic plaque was shown to harbor its own microbiota dominated by Proteobacteria and Collinsella [71]. Furthermore, a meta-analysis of clinical trials of antibiotics therapy in people with atherosclerosis failed to demonstrate their positive effect on cardiovascular mortality [72]. Nowadays, only one prospective randomized trial with azithromycin performed for secondary prophylaxis of coronary disease failed to reduce cardiovascular event rates [73]. Metabolomic analyses of plasma samples in humans identified novel metabolites and connecting pathways related with cardiovascular risk [74]. Major differences were found in choline, betaine, and trimethylamine N-oxide (TMAO), metabolites linked to phosphatidylcholine (PC) metabolism. Choline is known to be an essential nutrient [75] and its dietary deficiency can lead to muscle damage and liver steatosis. TMAO arises from the bacterial metabolism of choline via an intermediate trimethylamine (TMA), which reaches the liver and is converted into TMAO by oxidizing flavin monooxygenases 3 (FMO3) [76]. TMAO was undetectable in germ-free mice following a PC or carnitine challenge, but conventionalization of the rodents increased TMAO levels [77]. Similarly, when mice were treated with a cocktail of antibiotics, PC administration did not result in TMAO. Oral, but not parental, delivery of phosphatidylcholine was associated to higher levels of TMAO, suggesting that gut metabolic step is required for TMAO production. Interestingly, TMAO administration to apolipoprotein E-null mice was observed to produce macrophage foam cell formation in both the artery wall and the peritoneal cavity and to enhance aortic root atherosclerotic plaque development. In a large cohort study of over 4000 patients undergoing elective coronary angiography, TMAO levels were able to predict major adverse cardiac events independently from traditional CV risk factors [78]. Blood choline and carnitine levels were also predictors of major cardiac events with respect to stable coronary artery disease, but only when TMAO levels were concomitantly increased [76]. In summary, animal and human evidence suggest that the gut microbiota plays a contributory role in the development of CVD but need further studies to use the gut microbiota as a new target for prevention and treatment of CVD. 4. Concluding Remarks Over the last years, gut microbiota has emerged as a fascinating “new organ”, involved in many intestinal and extra-intestinal pathologies. New sequencing techniques led to the discovery of a huge complexity of the microbiome and to the identification of potential genes involved in gut microbiota-host interaction. Exciting new data suggest a clear contributory role for gut microbes in cardiometabolic disorders, such as diabetes, obesity, NAFLD/NASH, and atherosclerosis. However, more interest should be paid to the association with metabolomics, metagenomic, and functional studies, supplemented by prospective observational and interventional studies. The exploration of gut microbiota composition and function has now become a field of interest that offers a new frontier to discover the complex human host physiology and disease and providing new biomarkers and innovative therapeutic approaches. Acknowledgments Financial support: Grant PRIN 2010–2011 (Prot. N. 2010C4JJWB) (to A.Gr.), Catholic University Linea D1 (to A.Gr.) and Linea D3/2013 (to A.Ga. and A.Gr.). Author Contributions Marco Sanduzzi Zamparelli, Debora Compare, Pietro Coccoli, Alba Rocco, Olga Maria Nardone and Giuseppe Marrone revised and analized the literature and wrote the revied; Antonio Gasbarrini, Antonio Grieco, Gerardo Nardone and Luca Miele wrote the paper and revised the text. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Gut microbiota and energy balance. The gut microbes can benefit the host by extracting energies from otherwise non-digestible carbohydrates and plant polysaccharides via enzymes not encoded by humans. Short-chain fatty acids (SCFAs) modulate intestinal gluconeogenesis via the gut-brain neuronal circuit, involving GPR41 (free fatty acid receptor, FFAR3) and through the cyclic adenosine monophosphate (cAMP)-dependent pathway. Butyrate is able to regulate the appetite in the central nervous system by stimulating the liberation of peptide YY (PYY) and the satietogenic hormone glucagon-like peptide 1 (GLP-1) from enteroendocrine L-cells. PYY decreases the intestinal transit rate and increases the harvest of energy from the diet, while GLP-1 improves adipocyte insulin sensitivity and remarkably reduces fat storage in adipose tissue. Gut microbes can also control the metabolic activity of the host by affecting the composition and the abundance of certain bile acid species. In the ileum microbes deconjugate cholic and chenodeoxycholic acids, which escape intestinal uptake, and are converted into secondary bile acids. Bile acids can also act as signaling molecules by binding cellular receptors such as the bile-acid-synthesis controlling nuclear receptor farnesoid X receptor (FXR), G-protein-coupled receptors (GPCR), and TGR5. While primary bile acids can impair glucose metabolism by binding FXR, secondary bile acids, by binding TGR5, improve glucose homeostasis and increase energy expenditure in skeletal muscle. ==== Refs References 1. Go A.S. Mozaffarian D. Roger V.L. Benjamin E.J. Berry J.D. Blaha M.J. Dai S. Ford E.S. Fox C.S. Franco S. Heart disease and stroke statistics—2014 update: A report from the American Heart Association Circulation 2014 129 e28 e292 10.1161/01.cir.0000441139.02102.80 24352519 2. O’Donnell C.J. Nabel E.G. Genomics of cardiovascular disease N. Engl. J. Med. 2011 365 2098 2109 22129254 3. Ripatti S. Tikkanen E. Orho-Melander M. Havulinna A.S. Silander K. Sharma A. Guiducci C. Perola M. Jula A. Sinisalo J. A multilocus genetic risk score for coronary heart disease: Case-control and prospective cohort analyses Lancet 2010 376 1393 1400 10.1016/S0140-6736(10)61267-6 20971364 4. Ley R.E. Peterson D.A. Gordon J.I. Ecological and evolutionary forces shaping microbial diversity in the human intestine Cell 2006 124 837 848 10.1016/j.cell.2006.02.017 16497592 5. Neish A.S. Microbes in gastrointestinal health and disease Gastroenterology 2009 136 65 80 10.1053/j.gastro.2008.10.080 19026645 6. Tremaroli V. Bäckhed F. Functional interactions between the gut microbiota and host metabolism Nature 2012 489 242 249 10.1038/nature11552 22972297 7. Yatsunenko T. Rey F.E. Manary M.J. Trehan I. Dominguez-Bello M.G. Contreras M. Magris M. Hidalgo G. Baldassano R.N. Anokhin A.P. Human gut microbiome viewed across age and geography Nature 2012 486 222 227 22699611 8. Qin J. Li R. Raes J. Arumugam M. Burgdorf K.S. Manichanh C. Nielsen T. Pons N. Levenez F. Yamada T. A human gut microbial gene catalogue established by metagenomic sequencing Nature 2010 464 59 65 10.1038/nature08821 20203603 9. Cox L.M. Yamanishi S. Sohn J. Alekseyenko A.V. Leung J.M. Cho I. Kim S.G. Li H. Gao Z. Mahana D. Altering the intestinal microbiota during a critical developmental window has lasting metabolic consequences Cell 2014 158 705 721 10.1016/j.cell.2014.05.052 25126780 10. Bailey L.C. Forrest C.B. Zhang P. Richards T.M. Livshits A. DeRusso P.A. Association of antibiotics in infancy with early childhood obesity JAMA Pediatr. 2014 168 1063 1069 10.1001/jamapediatrics.2014.1539 25265089 11. De Filippo C. Cavalieri D. Di Paola M. Ramazzotti M. Poullet J.B. Massart S. Collini S. Pieraccini G. Lionetti P. Impact of diet in shaping gut microbiota revealed by a comparative study in children from Europe and rural Africa Proc. Natl. Acad. Sci. USA 2010 107 14691 14696 10.1073/pnas.1005963107 20679230 12. Matsumoto M. Inoue R. Tsukahara T. Ushida K. Chiji H. Matsubara N. Hara H. Voluntary running exercise alters microbiota composition and increases n-butyrate concentration in the rat cecum Biosci. Biotechnol. Biochem. 2008 72 572 576 10.1271/bbb.70474 18256465 13. Flint H.J. Bayer E.A. Rincon M.T. Lamed R. White B.A. Polysaccharide utilization by gut bacteria: Potential for new insights from genomic analysis Nat. Rev. Microbiol. 2008 6 121 131 10.1038/nrmicro1817 18180751 14. Bäckhed F. Ding H. Wang T. Hooper L.V. Koh G.Y. Nagy A. Semenkovich C.F. Gordon J.I. The gut microbiota as an environmental factor that regulates fat storage Proc. Natl. Acad. Sci. USA 2004 101 15718 15723 10.1073/pnas.0407076101 15505215 15. Brown A.J. Goldsworthy S.M. Barnes A.A. Eilert M.M. Tcheang L. Daniels D. Muir A.I. Wigglesworth M.J. Kinghorn I. Fraser N.J. The Orphan G protein-coupled receptors GPR41 and GPR43 are activated by propionate and other short chain carboxylic acids J. Biol. Chem. 2003 278 11312 11319 10.1074/jbc.M211609200 12496283 16. Kimura I. Ozawa K. Inoue D. Imamura T. Kimura K. Maeda T. Terasawa K. Kashihara D. Hirano K. Tani T. The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43 Nat. Commun. 2013 4 1829 10.1038/ncomms2852 23652017 17. De Vadder F. Kovatcheva-Datchary P. Goncalves D. Vinera J. Zitoun C. Duchampt A. Bäckhed F. Mithieux G. Microbiota-generated metabolites promote metabolic benefits via gut-brain neural circuits Cell 2014 156 84 96 10.1016/j.cell.2013.12.016 24412651 18. Tolhurst G. Heffron H. Lam Y.S. Parker H.E. Habib A.M. Diakogiannaki E. Cameron J. Grosse J. Reimann F. Gribble F.M. Short-chain fatty acids stimulate glucagon-like peptide-1 secretion via the G-protein-coupled receptor FFAR2 Diabetes 2012 61 364 371 10.2337/db11-1019 22190648 19. Samuel B.S. Shaito A. Motoike T. Rey F.E. Backhed F. Manchester J.K. Hammer R.E. Williams S.C. Crowley J. Yanagisawa M. Effects of the gut microbiota on host adiposity are modulated by the short-chain fatty-acid binding G protein-coupled receptor, GPR41 Proc. Natl. Acad. Sci. USA 2008 105 16767 16772 10.1073/pnas.0808567105 18931303 20. Yadav H. Lee J.H. Lloyd J. Walter P. Rane S.G. Beneficial metabolic effects of a probiotic via butyrate-induced GLP-1 hormone secretion J. Biol. Chem. 2013 288 25088 25097 10.1074/jbc.M113.452516 23836895 21. Swann J.R. Want E.J. Geier F.M. Spagou K. Wilson I.D. Sidaway J.E. Nicholson J.K. Holmes E. Systemic gut microbial modulation of bile acid metabolism in host tissue compartments Proc. Natl. Acad. Sci. USA 2011 108 Suppl. 1 4523 4530 10.1073/pnas.1006734107 20837534 22. Sinal C.J. Tohkin M. Miyata M. Ward J.M. Lambert G. Gonzalez F.J. Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis Cell 2000 102 731 744 10.1016/S0092-8674(00)00062-3 11030617 23. Prawitt J. Abdelkarim M. Stroeve J.H. Popescu I. Duez H. Velagapudi V.R. Dumont J. Bouchaert E. van Dijk T.H. Lucas A. Farnesoid X receptor deficiency improves glucose homeostasis in mouse models of obesity Diabetes 2011 60 1861 1871 10.2337/db11-0030 21593203 24. Thomas C. Gioiello A. Noriega L. Strehle A. Oury J. Rizzo G. Macchiarulo A. Yamamoto H. Mataki C. Pruzanski M. TGR5-mediated bile acid sensing controls glucose homeostasis Cell Metab. 2009 10 167 177 10.1016/j.cmet.2009.08.001 19723493 25. Everard A. Lazarevic V. Derrien M. Girard M. Muccioli G.G. Neyrinck A.M. Possemiers S. van Holle A. François P. de Vos W.M. Responses of gut microbiota and glucose and lipid metabolism to prebiotics in genetic obese and diet-induced leptin-resistant mice Diabetes 2011 60 2775 2786 10.2337/db11-0227 21933985 26. Hansen C.H. Krych L. Nielsen D.S. Vogensen F.K. Hansen L.H. Sørensen S.J. Buschard K. Hansen A.K. Early life treatment with vancomycin propagates Akkermansia muciniphila and reduces diabetes incidence in the NOD mouse Diabetologia 2012 55 2285 2294 10.1007/s00125-012-2564-7 22572803 27. Qin J. Li Y. Cai Z. Li S. Zhu J. Zhang F. Liang S. Zhang W. Guan Y. Shen D. A metagenome wide association study of gut microbiota in type 2 diabetes Nature 2012 490 55 60 10.1038/nature11450 23023125 28. Karlsson F.H. Tremaroli V. Nookaew I. Bergström G. Behre C.J. Fagerberg B. Nielsen J. Bäckhed F. Gut metagenome in European women with normal, impaired and diabetic glucose control Nature 2013 498 99 103 10.1038/nature12198 23719380 29. Larsen N. Vogensen F.K. van den Berg F.W. Nielsen D.S. Andreasen A.S. Pedersen B.K. Al-Soud W.A. Sørensen S.J. Hansen L.H. Jakobsen M. Gut microbiota in human adults with type 2 diabetes differs from non-diabetic adults PLoS ONE 2010 5 1225 10.1371/journal.pone.0009085 20140211 30. Shin N.R. Lee J.C. Lee H.Y. Kim M.S. Whon T.W. Lee M.S. Bae J.W. An increase in the Akkermansia spp. population induced by metformin treatment improves glucose homeostasis in diet-induced obese mice Gut 2014 63 727 735 10.1136/gutjnl-2012-303839 23804561 31. Everard A. Belzer C. Geurts L. Ouwerkerk J.P. Druart C. Bindels L.B. Guiot Y. Derrien M. Muccioli G.G. Delzenne M.M. Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced obesity Proc. Natl. Acad. Sci. USA 2013 110 9066 9071 10.1073/pnas.1219451110 23671105 32. Cani P.D. Bibiloni R. Knauf C. Waget A. Neyrinck A.M. Delzenne N.M. Burcelin R. Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice Diabetes 2008 57 1470 1481 10.2337/db07-1403 18305141 33. Cani P.D. Amar J. Iglesias M.A. Poggi M. Knauf C. Bastelica D. Neyrinck A.M. Fava F. Tuohy K.M. Chabo C. Metabolic endotoxemia initiates obesity and insulin resistance Diabetes 2007 56 1761 1772 10.2337/db06-1491 17456850 34. Erridge C. Attina T. Spickett C.M. Webb D.J. A high-fat meal induces low-grade endotoxemia: Evidence of a novel mechanism of postprandial inflammation Am. J. Clin. Nutr. 2007 86 1286 1292 17991637 35. Clarke T.B. Davis K.M. Lysenko E.S. Zhou A.Y. Yu Y. Weiser J.N. Recognition of peptidoglycan from the microbiota by Nod1 enhances systemic innate immunity Nat. Med. 2010 16 228 231 10.1038/nm.2087 20081863 36. Vijay-Kumar M. Aitken J.D. Carvalho F.A. Cullender T.C. Mwangi S. Srinivasan S. Sitaraman S.V. Knight R. Ley R.E. Gewirtz A.T. Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5 Science 2010 328 228 231 10.1126/science.1179721 20203013 37. Ley R.E. Bäckhed F. Turnbaugh P. Lozupone C.A. Knight R.D. Gordon J.I. Obesity alters gut microbial ecology Proc. Natl. Acad. Sci. USA 2005 102 11070 11075 10.1073/pnas.0504978102 16033867 38. Tilg H. Kaser A. Gut microbiome, obesity, and metabolic dysfunction J. Clin. Investig. 2011 121 2126 2132 10.1172/JCI58109 21633181 39. Bäckhed F. Manchester J.K. Semenkovich C.F. Gordon J.I. Mechanisms underlying the resistance to diet-induced obesity in germ-free mice Proc. Natl. Acad. Sci. USA 2007 104 979 984 10.1073/pnas.0605374104 17210919 40. Turnbaugh P.J. Ley R.E. Mahowald M.A. Magrini V. Mardis E.R. Gordon J.I. An obesity-associated gut microbiome with increased capacity for energy harvest Nature 2006 444 1027 1031 10.1038/nature05414 17183312 41. Sonnenburg J.L. Bäckhed F. Diet microbiota interactions as moderators of human metabolism Nature 2016 535 56 64 10.1038/nature18846 27383980 42. Ridaura V.K. Faith J.J. Rey F.E. Cheng J. Duncan A.E. Kau A.L. Griffin N.W. Lombard V. Henrissat B. Bain J.R. Gut microbiota from twins discordant for obesity modulate metabolism in mice Science 2013 341 1241214 10.1126/science.1241214 24009397 43. Perry R.J. Peng L. Barry N.A. Cline G.W. Zhang D. Cardone R.L. Petersen K.F. Kibbey R.G. Goodman A.L. Shulman G.I. Acetate mediates a microbiome-brain-β-cell axis to promote metabolic syndrome Nature 2016 534 213 217 10.1038/nature18309 27279214 44. Tremaroli V. Kovatcheva-Datchary P. Bäckhed F. A role for the gut microbiota in energy harvesting? Gut 2010 59 1589 1590 10.1136/gut.2010.223594 20940281 45. Gesta S. Blüher M. Yamamoto Y. Norris A.W. Berndt J. Kralisch S. Boucher J. Lewis C. Kahn C.R. Evidence for a role of developmental genes in the origin of obesity and body fat distribution Proc. Natl. Acad. Sci. USA 2006 103 6676 6681 10.1073/pnas.0601752103 16617105 46. Turnbaugh P.J. Hamady M. Yatsunenko T. Cantarel B.L. Duncan A. Ley R.E. Sogin M.L. Jones W.J. Roe B.A. Affourtit J.P. A core gut microbiome in obese and lean twins Nature 2009 457 480 484 10.1038/nature07540 19043404 47. Ley R.E. Turnbaugh P.J. Klein S. Gordon J.I. Microbial ecology: Human gut microbes associated with obesity Nature 2006 444 1022 1023 10.1038/4441022a 17183309 48. Furet J.P. Kong L.C. Tap J. Poitou C. Basdevant A. Bouillot J.L. Mariat D. Corthier G. Doré J. Henegar C. Differential adaptation of human gut microbiota to bariatric surgery-induced weight loss: Links with metabolic and low-grade inflammation markers Diabetes 2010 59 3049 3057 10.2337/db10-0253 20876719 49. Mar Rodríguez M. Pérez D. Javier Chaves F. Esteve E. Marin-Garcia P. Xifra G. Vendrell J. Jové M. Pamplona R. Ricart W. Obesity changes the human gut mycobiome Sci. Rep. 2015 5 14600 10.1038/srep14600 26455903 50. Lichtman S.N. Keku J. Schwab J.H. Sartor R.B. Hepatic injury associated with small bowel bacterial overgrowth in rats is prevented by metronidazole and tetracycline Gastroenterology 1991 100 513 519 10.1016/0016-5085(91)90224-9 1985047 51. Miele L. Valenza V. La Torre G. Montalto M. Cammarota G. Ricci R. Mascianà R. Forgione A. Gabrieli M.L. Perotti G. Increased intestinal permeability and tight junction alterations in nonalcoholic fatty liver disease Hepatology 2009 49 1877 1887 10.1002/hep.22848 19291785 52. Moschen A.R. Kaser S. Tilg H. Non-alcoholic steatohepatitis: A microbiota-driven disease Trends. Endocrinol. Metab. 2013 24 537 545 10.1016/j.tem.2013.05.009 23827477 53. Miele L. Marrone G. Lauritano C. Cefalo C. Gasbarrini A. Day C. Grieco A. Gut-liver axis and microbiota in NAFLD: Insight pathophysiology for novel therapeutic target Curr. Pharm. Des. 2013 19 5314 5324 10.2174/1381612811319290011 23432669 54. Cope K. Risby T. Diehl A.M. Increased gastrointestinal ethanol production in obese mice: Implications for fatty liver disease pathogenesis Gastroenterology 2000 119 1340 1347 10.1053/gast.2000.19267 11054393 55. Spencer M.D. Hamp T.J. Reid R.W. Fischer L.M. Zeisel S.H. Fodor A.A. Association between composition of the human gastrointestinal microbiome and development of fatty liver with choline deficiency Gastroenterology 2011 140 976 986 10.1053/j.gastro.2010.11.049 21129376 56. Le Roy T. Llopis M. Lepage P. Bruneau A. Rabot S. Bevilacqua C. Martin P. Philippe C. Walker F. Bado A. Intestinal microbiota determines development of non-alcoholic fatty liver disease in mice Gut 2013 62 1787 1794 10.1136/gutjnl-2012-303816 23197411 57. Mencarelli A. Cipriani S. Renga B. Bruno A. D’Amore C. Distrutti E. Fiorucci S. VSL#3 resets insulin signaling and protects against NASH and atherosclerosis in a model of genetic dyslipidemia and intestinal inflammation PLoS ONE 2012 7 1225 58. Malaguarnera M. Vacante M. Antic T. Giordano M. Chisari G. Acquaviva R. Mastrojeni S. Malaguarnera G. Mistretta A. Li Volti G. Bifidobacterium longum with fructo-oligosaccharides in patients with non alcoholic steatohepatitis Dig. Dis. Sci. 2012 57 545 553 10.1007/s10620-011-1887-4 21901256 59. Zhang G. Ghosh S. Toll-like receptor-mediated NF-κB activation: A phylogenetically conserved paradigm in innate immunity J. Clin. Investig. 2001 107 13 19 10.1172/JCI11837 11134172 60. Macpherson A.J. Harris N.L. Interactions between commensal intestinal bacteria and the immune system Nat. Rev. Immunol. 2004 4 478 485 10.1038/nri1373 15173836 61. Compare D. Coccoli P. Rocco A. Nardone O.M. de Maria S. Cartenì M. Nardone G. Gut—Liver axis: The impact of gut microbiota on non alcoholic fatty liver disease Nutr. Metab. Cardiovasc. Dis. 2012 22 471 476 10.1016/j.numecd.2012.02.007 22546554 62. Henao-Mejia J. Elinav E. Jin C. Hao L. Mehal W.Z. Strowig T. Thaiss C.A. Kau A.L. Eisenbarth S.C. Jurczak M.J. Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity Nature 2012 482 179 185 10.1038/nature10809 22297845 63. Elinav E. Thaiss C.A. Flavell R.A. Analysis of microbiota alterations in inflammasome-deficient mice Methods Mol. Biol. 2013 1040 185 194 23852605 64. Yang T. Santisteban M.M. Rodriguez V. Li E. Ahmari N. Carvajal J.M. Zadeh M. Gong M. Qi Y. Zubcevic J. Gut dysbiosis is linked to hypertension Hypertension 2015 65 1331 1340 10.1161/HYPERTENSIONAHA.115.05315 25870193 65. Arroyo-Espliguero R. Avanzas P. Jeffery S. Kaski J.C. CD14 and toll-like receptor 4: A link between infection and acute coronary events? Heart 2004 90 983 988 10.1136/hrt.2002.001297 15310678 66. Curtiss L.K. Tobias P.S. Emerging role of Toll-like receptors in atherosclerosis J. Lipid. Res. Suppl. 2009 165 S340 S345 10.1194/jlr.R800056-JLR200 18980945 67. Mattila K.J. Nieminen M.S. Valtonen V.V. Rasi V.P. Kesäniemi Y.A. Syrjälä S.L. Jungell P.S. Isoluoma M. Hietaniemi K. Jokinen M.J. Association between dental health and acute myocardial infarction BMJ 1989 298 779 781 10.1136/bmj.298.6676.779 2496855 68. Holmes E. Loo R.L. Stamler J. Bictash M. Yap I.K. Chan Q. Ebbels T. de Iorio M. Brown I.J. Veselkov K.A. Human metabolic phenotype diversity and its association with diet and blood pressure Nature 2008 453 396 400 10.1038/nature06882 18425110 69. Miele L. Giorgio V. Alberelli M.A. de Candia E. Gasbarrini A. Grieco A. Impact of Gut Microbiota on Obesity, Diabetes, and Cardiovascular Disease Risk Curr. Cardiol. Rep. 2015 17 120 10.1007/s11886-015-0671-z 26497040 70. Koren O. Spor A. Felin J. Fåk F. Stombaugh J. Tremaroli V. Behre C.J. Knight R. Fagerberg B. Ley R.E. Bäckhed F. Human oral, gut, and plaque microbiota in patients with atherosclerosis Proc. Natl. Acad. Sci. USA 2011 108 Suppl. 1 4592 4598 10.1073/pnas.1011383107 20937873 71. Karlsson F.H. Fåk F. Nookaew I. Tremaroli V. Fagerberg B. Petranovic D. Bäckhed F. Nielsen J. Symptomatic atherosclerosis is associated with an altered gut metagenome Nat. Commun. 2012 3 1245 10.1038/ncomms2266 23212374 72. Andraws R. Berger J.S. Brown D.L. Effects of antibiotic therapy on outcomes of patients with coronary artery disease: A meta-analysis of randomized controlled trials JAMA 2005 293 2641 2647 10.1001/jama.293.21.2641 15928286 73. Grayston J.T. Kronmal R.A. Jackson L.A. Parisi A.F. Muhlestein J.B. Cohen J.D. Rogers W.J. Crouse J.R. Borrowdale S.L. ACES Investigators. Azithromycin for the secondary prevention of coronary events N. Engl. J. Med. 2005 352 1637 1645 10.1056/NEJMoa043526 15843666 74. Wang Z. Klipfell E. Bennett B.J. Koeth R. Levison B.S. Dugar B. Feldstein A.E. Britt E.B. Fu X. Chung Y.M. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease Nature 2011 472 57 63 10.1038/nature09922 21475195 75. Zeisel S.H. Mar M.H. Howe J.C. Holden J.M. Concentrations of choline-containing compounds and betaine in common foods J. Nutr. 2003 133 2918 76. Lever M. Slow S. The clinical significance of betaine, an osmolyte with a key role in methyl group metabolism Clin. Biochem. 2010 43 732 744 10.1016/j.clinbiochem.2010.03.009 20346934 77. Tang W.H. Wang Z. Levison B.S. Koeth R.A. Britt E.B. Fu X. Wu Y. Hazen S.L. Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk N. Engl. J. Med. 2013 368 1575 1584 10.1056/NEJMoa1109400 23614584 78. Wang Z. Tang W.H. Buffa J.A. Fu X. Britt E.B. Koeth R.A. Levison B.S. Fan Y. Wu Y. Hazen S.L. Prognostic value of choline and betaine depends on intestinal microbiota-generated metabolite trimethylamine-N -oxide Eur. Heart J. 2014 35 904 910 10.1093/eurheartj/ehu002 24497336

PMC005xxxxxx/PMC5000624.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081226ijms-17-01226ArticlePeroxisome Proliferator-Activated Receptor γ Expression Is Inversely Associated with Macroscopic Vascular Invasion in Human Hepatocellular Carcinoma Hsu Hui-Tzu 12Sung Ming-Ta 3Lee Chih-Chun 14Kuo Yin-Ju 56Chi Chin-Wen 13Lee Hsin-Chen 1Hsia Cheng-Yuan 67*Haybaeck Johannes Academic Editor1 Department and Institute of Pharmacology, School of Medicine, National Yang-Ming University, Taipei 112, Taiwan; [email protected] (H.-T.H.); [email protected] (C.-C.L.); [email protected] (C.-W.C.); [email protected] (H.-C.L.)2 Program in Molecular Medicine, National Yang-Ming University and Academia Sinica, Taipei 112, Taiwan3 Department of Medical Research, Taipei Veterans General Hospital, Taipei 112, Taiwan; [email protected] Department of Surgery, Koo Foundation Sun Yat-Sen Cancer Center, Taipei 112, Taiwan5 Department of Pathology, Taipei Veterans General Hospital, Taipei 112, Taiwan; [email protected] Faculty of Medicine, School of Medicine, National Yang-Ming University, Taipei 112, Taiwan7 Department of Surgery, Taipei Veterans General Hospital, Taipei 112, Taiwan* Correspondence: [email protected]; Tel.: +886-2-2875-733529 7 2016 8 2016 17 8 122602 6 2016 21 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Peroxisome proliferator-activated receptor γ (PPARγ) is a ligand-activated nuclear receptor that regulates cellular lipid and glucose metabolism and also plays an inhibitory role in various cancers. However, the role of PPARγ in hepatocellular carcinoma (HCC) remains controversial. This study aimed to investigate the prognostic value of PPARγ in HCC and its role in inhibiting tumor progression, namely, HCC cell growth, migration, and angiogenesis. Immunohistochemical PPARγ staining was examined in 83 HCC specimens to investigate the clinicopathological correlations between PPARγ expression and various parameters. The functional role of PPARγ was determined via PPARγ overexpression and knockdown in HCC cells. Patients with low HCC tissue PPARγ expression were significantly younger (p = 0.006), and exhibited more tumor numbers (p = 0.038), more macroscopic vascular invasion (MVI) (p = 0.008), and more advanced TNM (size of primary tumor, number of regional lymph nodes, and distant metastasis) stages at diagnosis (p = 0.013) than patients with high HCC tissue PPARγ expression. PPARγ knockdown increased HCC cell growth, migration, and angiogenesis, while PPARγ overexpression reduced HCC cell growth, migration, and angiogenesis. These results suggest that low PPARγ expression is an independent predictor of more MVI in HCC patients. PPARγ contributes to the suppression of HCC cell growth, migration, and angiogenesis. Therefore, PPARγ may be a therapeutic target in HCC patients. PPARγmacroscopic vascular invasionhepatocellular carcinoma ==== Body 1. Introduction Hepatocellular carcinoma (HCC) is a major cause of cancer-related death worldwide [1], particularly in Asia and Africa. Hepatitis B virus (HBV) and hepatitis C virus (HCV) infection are two risk factors for HCC development [2]. Surgical resection is curative in HCC, but post-treatment recurrence and distant metastasis remain the major causes of death affected patients [3]. Therefore, understanding the molecular mechanism underlying HCC invasiveness is important for developing prognostic markers and new therapeutic targets for preventing tumor recurrence and improving survival rates. Peroxisome proliferator-activated receptor γ (PPARγ) is a ligand-activated nuclear hormone receptor that regulates insulin sensitivity, glucose metabolism, and inflammation in liver tissue, adipose tissue, and skeletal muscle tissue. Ligand binding to PPARγ triggers PPARγ and retinoid X receptor (RXR) heterodimerization, which may recruit co-activators or co-repressors to PPAR response elements (PPREs) within the promoters of PPARγ target genes and regulate their transcription [4]. PPARγ activity can be induced by natural and synthetic ligands. 15-Deoxy-∆12,14-prostaglandin J2 (15d-PGJ2) is a natural PPARγ ligand [5] and thiazolidinediones (TZDs), such as rosiglitazone, troglitazone, and pioglitazone are synthetic PPARγ ligands [6]. Accumulating evidence indicates that PPARγ plays a critical role in cancer cell growth [7], migration [8], invasion [9], and apoptosis [10]. PPARγ combined with its ligands to exert inhibitory effects on HCC cell growth, migration, and metastasis in vitro and in mouse models [11,12,13]. In addition, Krüppel-like factor 4 (KLF4), a tumor suppressor in HCC [14,15,16], has been reported to be up-regulated by the PPARγ agonist troglitazone and promotes cell cycle arrest in colorectal cancer cells [17]. These findings indicate that PPARγ collaborates with KLF4 to regulate HCC tumorigenesis and cancer progression. To date, only a few studies have described the changes in PPARγ expression in human HCC tissues. Schaefer et al. reported that high PPARγ protein expression was detected in 20 HCC tissues, but no expression was detected in non-tumorous livers [18]. Similarly, another group reported that the PPARγ mRNA expression was significantly increased in 16 HCC tissues compared with the non-tumorous livers [19]. However, a third group reported that PPARγ protein expression was decreased in HCC tissues compared with non-tumorous livers in 20 HCC patients [20]. These results suggest that not only are the findings regarding PPARγ expression in HCC controversial, but the clinicopathological significance of PPARγ in human HCC also still unclear. The relationship between PPARγ expression and patient survival after curative treatment also remains unclear. In this study, we examined the relationship between PPARγ protein expression and various clinicopathological variables such as age, tumor number, macroscopic vascular invasion (MVI), TNM (size of primary tumor, number of regional lymph nodes, and distant metastasis) stage, and survival rate in 83 HCC patients, who have underwent surgical resection. We also investigated the role of PPARγ in cell proliferation, migration, and angiogenesis via its overexpression and knockdown of PPARγ (peroxisome proliferator-activated receptor γ) in HCC cells. 2. Results 2.1. Peroxisome Proliferator-Activated Receptor γ (PPARγ) Protein Expression in Human Hepatocellular Carcinoma (HCC) Tissues and Associated Clinicopathological Characteristics To determine the clinicopathological significance of PPARγ protein expression in HCC, we examined PPARγ and downstream KLF4 expression in 83 HCC tissue samples via immunohistochemistry (IHC). IHC results were scored from 0 to 3 to indicate the percentages of cells with positive PPARγ staining (Figure 1A). Of the 83 HCC tissue samples, 30 (36.1%) exhibited positive PPARγ staining (score > 0), and were considered to have high PPARγ expression, and 53 (63.9%) exhibited negative PPARγ staining (score = 0), and were considered to have low PPARγ expression (Figure 1B). A similar trend was observed regarding KLF4 expression (Figure 1B). Moreover, IHC staining of KLF4 and PPARγ revealed a significantly positive correlation between the expression levels of the two proteins (r = 0.35, p = 0.01, Chi-square < 0.001) (Figure 1C). Various clinicopathological parameters, including patient age (p = 0.006), tumor number (p = 0.038), MVI (p = 0.008), and TNM stage (p = 0.013), exhibited significant associations with PPARγ expression (Table 1). Notably, MVI was independently associated with PPARγ expression based on the results of the multiple logistic regression analysis, indicating that low PPARγ expression is independently predictive of more MVI in HCC patients. In this study, patients with MVI exhibited significantly worse survival than patients without MVI (Figure S2). Subgroup analysis showed that patients with high PPARγ expression and no MVI exhibited superior disease-free survival (DFS) and overall survival (OS) rates (40.7% and 46.1%, respectively) than patients with low PPARγ expression and MVI (23.5% and 39.2%, respectively), although this difference was not statistically significant (p = 0.202 and p = 0.720, respectively) (Figure S3). These results suggest that low PPARγ expression is significantly correlated with poor clinicopathological findings in HCC patients. 2.2. PPARγ Suppresses HCC Cell Proliferation Given that PPARγ is significantly associated with important HCC diagnostic and clinicopathological variables, we characterized its function in HCC via in vitro assays. Endogenous PPARγ and E-cadherin expression levels were examined in various HCC cell lines, including PLC/PRF/5, SK-Hep1, and Mahlavu cells (Figure S1). Our results revealed that Mahlavu cells, which are poorly differentiated and highly migratory, exhibited both low PPARγ and E-cadherin expression, whereas PLC/PRF/5 cells, which are well-differentiated and less migratory, exhibited both high PPARγ and high E-cadherin expression. To simulate different clinical scenarios, we overexpressed PPARγ in Mahlavu cells via a retrovirus-mediated gene transfer. We also knock down PPARγ expression in PLC/PRF/5 cells via a lentivirus-mediated gene transfer. Moreover, STAT3 and cyclin D1 protein expression was analyzed because both proteins are downstream targets of PPARγ-mediated signaling and are essential for cell cycle progression [21,22]. We found that PPARγ-overexpressing cells (Mahlavu-PPARγ) exhibited decreased cell growth rates (Figure 2A) and reduced STAT3 and cyclin D1 expression compared with vector control cells (Mahlavu-ctr) (Figure 2B). In contrast, PPARγ knockdown cells (PLC/PRF/5-shPPARγ) exhibited increased cell growth rates (Figure 2C) and higher STAT3 and cyclin D1 expression compared with luciferase control cells (PLC/PRF/5-ctr) (Figure 2D). Taken together, these findings indicate that PPARγ suppresses HCC cell proliferation, and down-regulates STAT3 and cyclin D1 expression. 2.3. PPARγ Inhibits HCC Cell Migration We investigated the effect of PPARγ on HCC cell migration and found that PPARγ-overexpressing cells (Mahlavu-PPARγ) exhibited a 16% decrease in cell migration compared with control cells as determined via wound healing assay (Figure 3A). Conversely, PPARγ-knockdown cells (PLC/PRF/5-shPPARγ) exhibited significant four-fold increases in migration compared with control cells (Figure 3B). These results suggest that PPARγ inhibits HCC cell migration. 2.4. PPARγ Decreases HCC Cell Angiogenesis Table 1 shows that PPARγ expression was inversely associated with MVI in human HCC patients and that MVI is associated with angiogenesis. We examined the effects of PPARγ on HCC cell angiogenesis. The results showed that PPARγ-overexpressing cell (Mahlavu-PPARγ)-conditioned medium decreased the number of vessel joints in human umbilical vein endothelial cells (HUVECs) by 20% compared with the control cell-conditioned medium (Figure 4A). In contrast, the PPARγ-knockdown cell (PLC/PRF/5-shPPARγ)-conditioned medium increased the number of vessel joints by 40% compared with the control cell-conditioned medium (Figure 4B). In addition, the PPARγ knockdown cells (PLC/PRF/5-shPPARγ) exhibited increased vascular endothelial growth factor (VEGF) expression compared with control cells; however, no significant differences in VEGF expression were observed between PPARγ-overexpressing cells (Mahlavu-PPARγ) and control cells (Figure S4). These results suggest that PPARγ inhibits HCC cell angiogenesis. 3. Discussions This study was the first to analyze the clinicopathological significance of PPARγ in HCC using a relatively large sample size. We demonstrated that HCC patients with low PPARγ expression were significantly younger than 65 years old and exhibited more tumor numbers, more MVI, and more advanced TNM stages at diagnosis than HCC patients with high PPARγ expression. All of these clinicopathological factors are considered prognostic factors for patient survival. In particular, MVI has long been considered a major determining factor of TNM stage and patient survival and HCC patients exhibiting low PPARγ expression were independently predicted to have more MVI. In addition, we demonstrated that PPARγ inhibits HCC cell proliferation, migration, and angiogenesis. These results suggest that PPARγ functions as a tumor suppressor in HCC cells and may be a therapeutic target in HCC. We found that 53 of 83 HCC samples exhibited negative PPARγ staining (score = 0) and that only 30 samples exhibited positive PPARγ staining (score > 0), indicating that most HCC tissues exhibited low or no PPARγ protein expression (Figure 1B). Consistent with the results from the Human Protein Atlas (www.proteinatlas.org), nine of 11 HCC tissue samples exhibited non-detectable PPARγ staining. However, previous studies demonstrated that PPARγ protein and gene expression varies in HCC tissues compared with normal tissues across different assays [18,19,20]. A study of colorectal cancer cells demonstrated that PPARγ regulates KLF4 transcription by directly binding to the KLF4 promoter [17]. Reduced KLF4 expression was also observed in HCC [14]. Our analysis revealed that a significantly positive correlation exists between PPARγ and KLF4 expression in HCC tissues, suggesting that PPARγ collaborates with KLF4 to facilitate tumor suppression. HCC patients with MVI involving portal or hepatic veins exhibited an increased risk of tumor recurrence and worse prognoses after liver resection or transplantation than patients without extensive MVI [23]. MVI has also been suggested to be an independent predictor of recurrence after liver transplantation [24]. Consistent with these findings, our results also showed that HCC patients with MVI exhibited a significantly decreased in DFS rates compared with patients without MVI (Figure S2). HCC patients with high PPARγ expression and no MVI exhibited better DFS and OS than patients with low PPARγ expression and MVI (Figure S3). These results indicated that PPARγ expression is inversely associated with tumor development, namely, the number of tumor foci, the extent of MVI, and the TNM stage. Our findings indicate that low PPARγ expression is significantly associated with a poor prognosis in HCC patients. In this study, PPARγ expression was negatively associated with TNM stage and MVI in HCC tissues. To simulate this clinical scenario and determine the functional role of PPARγ in HCC, we used the poorly differentiated and relatively low PPARγ-expressing HCC cell line Mahlavu to overexpress PPARγ. We also used the well-differentiated and relatively high PPARγ-expressing HCC cell line PLC/PRF/5 to knock down PPARγ. We noted decreased growth and decreased STAT3 and cyclin D1 expression in PPARγ-overexpressing HCC cells (Figure 2A). Conversely, we noted increased growth and increased STAT3 and cyclin D1 expression in PPARγ knockdown HCC cells (Figure 2B). These results are consistent with those of previous studies involving PPARγ-deficient (PPARγ+/−) mice and different cell lines. Yu et al. demonstrated that PPARγ-deficient (PPARγ+/−) mice were susceptible to diethylnithrosamine-induced liver carcinogenesis compared with wild-type mice (PPARγ+/+), suggesting that PPARγ functions as a tumor suppressor in hepatocarcinogenesis [12]. In pancreatic cancer cells, PPARγ activation by PPARγ agonists suppressed STAT3 expression through transcriptional repression to inhibit cell growth [21]. In addition, PPARγ activation by a synthetic PPARγ agonist, pioglitazone, resulted in growth inhibition and decreased cyclin D1 expression in human HCC SMMC-7721 and HepG2 cells [22]. Together, these data suggest that PPARγ inhibits cell growth by down-regulating STAT3 and cyclin D1 expression in HCC cells. Moreover, we observed PPARγ inhibited HCC cell migration (Figure 3) and in vitro angiogenesis (Figure 4). Previous studies involving human HCC MHCC97L and BEL-7404 cells suggested that PPARγ overexpression suppressed cell migration and invasion by down-regulating matrix metalloproteinase (MMP) 9, MMP13, and heparanase (HPSE) expression, while up-regulating E-cadherin and tissue inhibitor of metalloproteinase (TIMP) 3 expression [11]. PPARγ overexpression decreased cell invasion by up-regulating plasminogen activator inhibitor-1 (PAI-1) expression in HepG2 cells [13]. Moreover, PPARγ ligands facilitated cell cycle arrest, apoptosis, and metastasis inhibition in HCC via multiple pathways [25]. In addition, PPARγ ligands have also been shown to exert anti-angiogenic effects and to decrease VEGF expression in different types of cancers, such as glioblastoma and Lewis lung carcinoma cells [26,27]. It has been reported that the human VEGF promoter contains a PPRE. Treatments with PPARγ ligands, such as rosiglitazone and 15-Deoxy-∆12,14-prostaglandin J2, repressed VEGF gene expression through direct binding to the VEGF PPRE promoter in human endometrial cells [28]. These results suggest that the absence of PPARγ and the loss of the repressor in the VEGF promoter may result in increased VEGF expression, indicating that PPARγ plays an inhibitory role in the HCC cell growth, migration, and angiogenesis; thus, PPARγ may be a therapeutic target for HCC treatment. In conclusion, we demonstrated for the first time that low PPARγ expression was significant associated with patient age, tumor number, MVI, and TNM stage in HCC patients. In vitro experiments showed that PPARγ suppresses tumor progression, namely, growth, migration, and angiogenesis in HCC cells. Cyclin D1 and STAT3 may be involved in PPARγ-mediated signaling pathways that inhibit HCC cell growth. Absence of PPARγ expression may lead to increased invasiveness in HCC. Our findings indicate that PPARγ expression may determine patient prognosis in HCC and that PPARγ may serve as a therapeutic target for HCC treatment. 4. Materials and Methods 4.1. Human Tissue Specimens and Patient Information Eighty-three patients who underwent curative liver resection for HCC in Taipei Veteran General Hospital were enrolled in the study. These patients ranged from 28 to 88 years (average of 61 ± 14 years). Sixty patients were male and 23 were female. A total of 83 paraffin-embedded HCC samples were obtained from the surgical tissue bank of Taipei Veterans General Hospital, Taiwan. Institutional review board (IRB) approval was obtained for this retrospective study (IRB No: 2013-02-031BC). HCC was morphologically classified, according to the World Health Organization guideline. Representative paraffin blocks were obtained from pathologically confirmed waxed-preserved HCC specimens. The paraffin blocks were then used to generate a tissue array and cut into 3-μm-thick sections for further investigation. 4.2. Immunohistochemical PPARγ and KLF4 protein expression was examined by IHC using a DAKO LSAB2 Kit (Agilent Technologies, Produktionsvej, Denmark). Tissue sections were microwaved in sodium citrate buffer (10 mM, pH 6), treated with 3.0% H2O2 for 10 min and soaked with blocking solution for 10 min. The tissue sections were incubated overnight with antibodies specific for PPARγ (Santa Cruz Biotechnology, Dallas, TX, USA) and KLF4 (Atlas, Stockholm, Sweden) at dilutions of 1:100 at room temperature in a moist chamber. Then, the tissue section slides were washed in PBS and incubated with a biotin-labeled secondary antibody for 10 min, before being incubated with a streptavidin horseradish peroxidase (HRP)-conjugated secondary antibody for 10 min. After the sections were incubated with a 3,3-diaminobenzidine tetrahydrochloride (DAB) substrate chromogen for 10 min, the Mayer’s hematoxylin counterstain was applied for 10 min (Muto Pure Chemicals, Tokyo, Japan). Finally, mounting solution (Kaiser’s glycerol gelatin, Merck, Kenilworth, NJ, USA) was added to the sections, which were covered with cover slides for histological examination. A pathologist blinded patient clinicopathological data perform examination. The staining results were graded on a scale of 0 to 3 and represented as percentages of positively stained cells. A score of 0 indicated negative staining (<10%), a score of 1 indicated weak staining (10% to 25% of cells stained positive), a score of 2 indicated moderate staining (25% to 50% of cells stained positive), and a score of 3 indicated strong staining (greater than 50% of cells stained positive). A score of 0 indicated low PPARγ expression, and scores ranging from 1 to 3 were indicated high PPARγ expression. 4.3. Cell Culture PLC/PRF/5, Mahlavu, and SK-Hep1 HCC cell lines (obtained from Cell Bank of Taipei Veterans General Hospital, Taipei, Taiwan) and HEK293T cells (ATCC, Manassas, VA, USA) were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 0.1 mM non-essential amino acids, 2 mM l-glutamine, and 1% penicillin/streptomycin in a humidified atmosphere containing with 5% CO2 at 37 °C. All cell culture reagents were obtained from Invitrogen (Carlsbad, CA, USA). 4.4. PPARγ Overexpression and Knockdown Retroviral expression vectors carrying PPARγ full-length cDNA or pBABE-puro-PPARγ were constructed. HEK293T cells were co-transfected with these retroviral expression vectors and packaging plasmids (pAmpho and pCMV-VSV-G) using TurboFECT (Thermo Scientific, Waltham, MA, USA). Supernatants were collected 72 h after transfection and filtered. Mahlavu cells were infected with the retroiviral expression vectors in the presence of 8 μg/mL polybrene (Sigma-Aldrich, St. Louis, MO, USA). The lentiviral vector pLVO.1-shPPARG, which was used for PPARγ knockdown, was purchased from the RNAi core of Academia Sinica (Taipei, Taiwan). The oligonucleotide targeting human PPARγ was 5′-GACAACAGACAAATCACCATT-3′. The lentiviruses were generated by co-transfecting HEK293T cells with the indicated lentiviral expression vectors (pLVO.1-puro and pLVO.1-shPPARG) and packaging plasmids (pCMVΔR8.91 and pCMV-VSV-G) using TurboFECT (Thermo Scientific). Supernatants containing the lentiviruses were collected 72 h after transfection and filtered. PLC/PRF/5 cells were infected with lentiviruses in the presence of 8 μg/mL polybrene (Sigma-Aldrich). Stable clones of PPARγ-overexpressing and PPARγ-knockdown HCC cells were selected by puromycin (Sigma-Aldrich). The packaging plasmids pAmpho, pCMV-VSV-G, and pCMVΔR8.91 were purchased from the RNAi core of Academia Sinica. 4.5. Cell Proliferation Measurement Sulforhodamine B (SRB) colorimetric analysis was used to measure cell proliferation. Cells were seeded at a density of 4 × 103 cells/well in 96-well plates. First, the cells were fixed in cold 10% trichloroacetic acid (Sigma-Aldrich) at 4 °C for 1 h. After being washed with water and air-dried the fixed cells were incubated with 1% SRB (Sigma-Aldrich), dissolved in 1% acetic acid for 30 min. Unincorporated dye was removed by five rinses with 1% acetic acid. The protein-bound dye was extracted with 10 mM Tris and then optical absorbance was measured at a wavelength of 510 nm was measured by a spectrophotometer. 4.6. Western Blot Analyses Total proteins were extracted from cells using RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, 0.25% sodium deoxycholate, 1% Triton X-100, 0.1% SDS), supplemented with a protease inhibitor cocktail (Calbiochem, San Diego, CA, USA). Proteins were separated via 10% SDS-PAGE gel and electrotransferred onto a polyvinylidene difluoride (PVDF) membrane. The PVDF membrane was blocked in 5% skimmed milk at room temperature for 1 h and probed with primary antibodies. PPARγ (Cell Signaling, Danvers, MA, USA), E-cadherin (Cell Signaling), signal transducer and activator transcription 3 (STAT3) (Cell Signaling), cyclin D1 (Millipore, Darmstadt, Germany), and β-actin (Sigma-Aldrich) antibodies were used to probe the proteins on the membrane at 4 °C overnight. After incubation with an HRP-conjugated secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), the probed proteins were detected by an enhanced chemiluminescence system (Thermo Scientific), according to the manufacturer’s instructions. 4.7. Wound Healing Assay Changes in cell migration ability were assessed via in vitro wound healing assay with a Culture-Insert (Ibidi, Am Klopferspitz, Germany), which is a special sticky and biocompatible surface, the bottom side of which works like glue and avoids leaking. Cells were seeded into each well of the Culture-Insert and then incubated overnight at 37 °C and 5% CO2. After the cells attached and achieved confluence, the Culture-Insert was gently removed to allow cell migration. We measured the migratory areas of Mahlavu cells and PLC/PRF/5 cells for 14 and 24 h, respectively, after removing the Culture-Insert. Four bright field images were obtained at 100× magnification at the indicated time points, and Image J software (National Institute of Health, Bethesda, MD, USA) was used to analyze the migratory areas. 4.8. Matrigel Tube Formation Assay HUVECs (human umbilical vein endothelial cells) were cultured in Medium 200 with Low Serum Growth Supplement (Gibco, Carlsbad, CA, USA) in a humidified atmosphere, with 5% CO2 at 37 °C. Briefly, 96-well culture plates were coated with 50 µL of Matrigel (BD Biosciences, San Jose, CA, USA) per well and the Matrigel was allowed to polymerize for 30 min. HCC cell-conditioned medium was harvested over 2 days, and HUVECs were separately suspended in Mahlavu-ctr, Mahlavu-PPARγ, PLC/PRF/5-shLuc, and PLC/PRF/5-shPPARγ cell-conditioned medium. HUVEC culture medium was subsequently added to the suspension at a 1:1 ratio. The HUVECs were then seeded onto the polymerized Matrigel-coated wells at a density of 104 cells/100 μL per well. Bright field images were obtained after the cells were incubated for 6 h in a 37 °C incubator. Tube-like structures were detected under an inverted light microscope at 200× to evaluate in vitro angiogenesis. The numbers of vessel joints in five fields were counted. 4.9. Statistical Analyses A Chi-square test was used for categorical variables. Continuous variables were expressed as the mean ± SEM (standard error of the mean), and the differences between groups were evaluated by Student’s t-test. Multivariate analysis was performed using the logistic regression model. Correlations between variables were calculated by Spearman’s co-efficient method. Survival was calculated using the Kaplan–Meier method, and the survival differences were assessed by the log-rank test. SPSS software version 19 (IBM Corporation, Armonk, NY, USA) was used to perform the statistical analyses. Differences were considered statistically significant at p < 0.05. Acknowledgments The authors thank Shih-Hwa Chiou for providing the lentiviral expression vector and packaging plasmids and thank Ling-Chen Tai for performing the statistical analyses. This work was supported by grants NSC 102-2320-B-075-003, MOST-103-2320-B-075-006, and MOST-104-2320-B-010-031 from the Ministry of Science and Technology, Taiwan; and DOH101-TD-C-111-007, DOH102-TD-C-111-007, and MOHW105-TDU-13-211-134003 from the Ministry of Health and Welfare, Taiwan. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1226/s1. Click here for additional data file. Author Contributions Hui-Tzu Hsu, Chin-Wen Chi, Hsin-Chen Lee and Cheng-Yuan Hsia conceived and designed the experiments. Hui-Tzu Hsu, Cheng-Yuan Hsia and Yin-Ju Kuo performed the experiments and analyzed the data. Hui-Tzu Hsu wrote the manuscript. Chin-Wen Chi, Hsin-Chen Lee, Cheng-Yuan Hsia, Ming-Ta Sung and Chih-Chun Lee revised the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Peroxisome proliferator-activated receptor γ (PPARγ) and Krüppel-like factor 4 (KLF4) protein expression in human hepatocellular carcinoma (HCC) tissues: (A) representative views indicated PPARγ expression scores ranging from 0 to 3, as determined by immunohistochemistry (IHC); (B) the case number bar chart; and (C) correlation plot were generated using PPARγ and KLF4 staining scores from 83 human HCC tissue samples. Scale bar represents 100 μm. Figure 2 Effects of PPARγ overexpression and knockdown on cell proliferation and PPARγ downstream target protein expression in Mahlavu and PLC/PRF/5 HCC cells, respectively. (A,C) The cell proliferation rates of Mahlavu-ctr, Mahlavu-PPARγ, PLC/PRF/5-shLuc, and PLC/PRF/5-shPPARγ cells were analyzed by SRB assay; (B,D) the expression of PPARγ downstream target proteins STAT3 and cyclin D1 was analyzed by Western blot and the quantification results are shown. * p < 0.05 indicates a significant difference from vector control cells at the same time point. Figure 3 Effects of PPARγ overexpression and knockdown on cell migration of Mahlavu cells and PLC/PRF/5 cells, respectively. (A) Cell migration abilities of Mahlavu-ctr and Mahlavu-PPARγ cells were analyzed over 14 h by wound healing assay; (B) cell migration abilities of PLC/PRF/5-shLuc and PLC/PRF/5-shPPARγ cells were assessed over 24 h by wound healing assay. Relative quantification data are expressed as the mean ± SEM (standard error of the mean) from three independent experiments. * p < 0.05 and ** p < 0.001 indicate significant differences compared with vector control cells. Scale bar represents 100 μm. Figure 4 Effects of PPARγ overexpression and knockdown on in vitro human umbilical vein endothelial cells (HUVEC) tube formation in Mahlavu and PLC/PRF/5 cells, respectively. Conditioned medium was harvested from: (A) Mahlavu-ctr and Mahlavu-PPARγ; and (B) PLC/PRF/5-shLuc and PLC/PRF/5-shPPARγ cell cultures. The conditioned medium was used for HUVEC cell tube formation, and photos were taken after 6 h of incubation. Quantification data are expressed as the mean ± SEM from three independent experiments. * p < 0.05 indicates significant differences compared with vector control cells. Scale bar represents 100 μm. ijms-17-01226-t001_Table 1Table 1 Multivariate analysis of PPARγ expression in relation to clinicopathological findings in HCC patients. Characteristic PPARγ Expression (Percentage) p Value Low (=0) n = 53 High (>0) n = 30 Age ≤65 39 13 0.006 * >65 14 17 Sex Male 39 21 0.726 Female 14 9 Tumor size <3 cm 8 9 0.106 >3 cm 45 21 HBsAg ∆ (−) 15 14 0.053 (+) 38 14 Anti-HCV ∆ (−) 44 22 0.148 (+) 7 8 Cell differentiation Well differentiated 3 3 0.63 Moderately differentiated 32 19 Poorly differentiated 18 8 Tumor number ∆ Single 33 25 0.038 * Multiple 18 4 Liver cirrhosis No 42 21 0.344 Yes 11 9 Chronic hepatitis ∆ No 6 7 0.227 Yes 41 23 Fibrosis ∆ No 29 17 0.293 Yes 12 12 MVI No 36 28 0.008 * Yes 17 2 Bile duct invasion No 49 29 0.649 Yes 4 1 AFP ∆ <20 19 16 0.139 >20 33 14 TNM stage I + II 30 25 0.013 * III 23 5 DFS Event/all 34/53 18/30 0.967 5-year survival 34.6% 38% OS Event/all 25/53 15/30 0.349 5-year survival 55.3% 46.6% PPARγ: Peroxisome proliferator-activated receptor γ; HCC: hepatocellular carcinoma; HBsAg: hepatitis B virus surface antigen, Anti-HCV: anti-hepatitis C virus, MVI: macroscopic vascular invasion, AFP: α-fetoprotein, DFS: disease-free survival, OS: overall survival. * p < 0.05. ∆ indicates a missing number. TNM: size of primary tumor, number of regional lymph nodes, and distant metastasis. ==== Refs References 1. Torre L.A. Bray F. Siegel R.L. Ferlay J. Lortet-Tieulent J. Jemal A. Global cancer statistics, 2012 CA Cancer J. Clin. 2015 65 87 108 10.3322/caac.21262 25651787 2. El-Serag H.B. Epidemiology of viral hepatitis and hepatocellular carcinoma Gastroenterology 2012 142 1264 1273 10.1053/j.gastro.2011.12.061 22537432 3. Tang Z.Y. Hepatocellular carcinoma—Cause, treatment and metastasis World J. Gastroenterol. 2001 7 445 454 10.3748/wjg.v7.i4.445 11819809 4. Desvergne B. Wahli W. Peroxisome proliferator-activated receptors: Nuclear control of metabolism Endocr. Rev. 1999 20 649 688 10.1210/er.20.5.649 10529898 5. Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. 15-Deoxy-∆12,14 -prostaglandin J2 is a ligand for the adipocyte determination factor PPARγ Cell 1995 83 803 812 10.1016/0092-8674(95)90193-0 8521497 6. Murphy G.J. Holder J.C. PPAR-γ agonists: Therapeutic role in diabetes, inflammation and cancer Trends Pharmacol. Sci. 2000 21 469 474 10.1016/S0165-6147(00)01559-5 11121836 7. Cao L.Q. Shao Z.L. Liang H.H. Zhang D.W. Yang X.W. Jiang X.F. Xue P. Activation of peroxisome proliferator-activated receptor-γ (PPARγ) inhibits hepatoma cell growth via downregulation of SEPT2 expression Cancer Lett. 2015 359 127 135 10.1016/j.canlet.2015.01.004 25592041 8. Lee H.J. Su Y. Yin P.H. Lee H.C. Chi C.W. PPARγ/PGC-1α pathway in E-cadherin expression and motility of HepG2 cells Anticancer Res. 2009 29 5057 5063 20044617 9. Gu J.J. Zhang J.H. Chen H.J. Wang S.S. MicroRNA-130b promotes cell proliferation and invasion by inhibiting peroxisome proliferator-activated receptor-γ in human glioma cells Int. J. Mol. Med. 2016 37 1587 1593 10.3892/ijmm.2016.2580 27122306 10. Ohta K. Endo T. Haraguchi K. Hershman J.M. Onaya T. Ligands for peroxisome proliferator-activated receptor γ inhibit growth and induce apoptosis of human papillary thyroid carcinoma cells J. Clin. Endocrinol. Metab. 2001 86 2170 2177 11344222 11. Shen B. Chu E.S. Zhao G. Man K. Wu C.W. Cheng J.T. Li G. Nie Y. Lo C.M. Teoh N. PPARγ inhibits hepatocellular carcinoma metastases in vitro and in mice Br. J. Cancer 2012 106 1486 1494 10.1038/bjc.2012.130 22472882 12. Yu J. Shen B. Chu E.S. Teoh N. Cheung K.F. Wu C.W. Wang S. Lam C.N. Feng H. Zhao J. Inhibitory role of peroxisome proliferator-activated receptor γ in hepatocarcinogenesis in mice and in vitro Hepatology 2010 51 2008 2019 10.1002/hep.23550 20512989 13. Pang X. Wei Y. Zhang Y. Zhang M. Lu Y. Shen P. Peroxisome proliferator-activated receptor-γ activation inhibits hepatocellular carcinoma cell invasion by upregulating plasminogen activator inhibitor-1 Cancer Sci. 2013 104 672 680 10.1111/cas.12143 23461356 14. Li Q. Gao Y. Jia Z. Mishra L. Guo K. Li Z. Le X. Wei D. Huang S. Xie K. Dysregulated kruppel-like factor 4 and vitamin D receptor signaling contribute to progression of hepatocellular carcinoma Gastroenterology 2012 143 799 810 10.1053/j.gastro.2012.05.043 22677193 15. Sung M.T. Hsu H.T. Lee C.C. Lee H.C. Kuo Y.J. Hua K. Hsia C.Y. Chi C.W. Kruppel-like factor 4 modulates the migration and invasion of hepatoma cells by suppressing TIMP-1 and TIMP-2 Oncol. Rep. 2015 34 439 446 25954999 16. Lin Z.S. Chu H.C. Yen Y.C. Lewis B.C. Chen Y.W. Kruppel-like factor 4, a tumor suppressor in hepatocellular carcinoma cells reverts epithelial mesenchymal transition by suppressing slug expression PLoS ONE 2012 7 1226 10.1371/journal.pone.0043593 22937066 17. Li S. Zhou Q. He H. Zhao Y. Liu Z. Peroxisome proliferator-activated receptor γ agonists induce cell cycle arrest through transcriptional regulation of kruppel-like factor 4 (KLF4) J. Biol. Chem. 2013 288 4076 4084 10.1074/jbc.M111.317487 23275339 18. Schaefer K.L. Wada K. Takahashi H. Matsuhashi N. Ohnishi S. Wolfe M.M. Turner J.R. Nakajima A. Borkan S.C. Saubermann L.J. Peroxisome proliferator-activated receptor γ inhibition prevents adhesion to the extracellular matrix and induces anoikis in hepatocellular carcinoma cells Cancer Res. 2005 65 2251 2259 10.1158/0008-5472.CAN-04-3037 15781638 19. Lin Y.M. Velmurugan B.K. Yeh Y.L. Tu C.C. Ho T.J. Lai T.Y. Tsai C.H. Tsai F.J. Tsai C.H. Huang C.Y. Activation of estrogen receptors with E2 downregulates peroxisome proliferator-activated receptor γ in hepatocellular carcinoma Oncol. Rep. 2013 30 3027 3031 24126791 20. Yu J. Qiao L. Zimmermann L. Ebert M.P. Zhang H. Lin W. Rocken C. Malfertheiner P. Farrell G.C. Troglitazone inhibits tumor growth in hepatocellular carcinoma in vitro andin vivo Hepatology 2006 43 134 143 10.1002/hep.20994 16374840 21. Vitale G. Zappavigna S. Marra M. Dicitore A. Meschini S. Condello M. Arancia G. Castiglioni S. Maroni P. Bendinelli P. The PPAR-γ agonist troglitazone antagonizes survival pathways induced by stat-3 in recombinant interferon-β treated pancreatic cancer cells Biotechnol. Adv. 2012 30 169 184 10.1016/j.biotechadv.2011.08.001 21871555 22. Yang Y. Zhao L.H. Huang B. Wang R.Y. Yuan S.X. Tao Q.F. Xu Y. Sun H.Y. Lin C. Zhou W.P. Pioglitazone, a PPARγ agonist, inhibits growth and invasion of human hepatocellular carcinoma via blockade of the rage signaling Mol. Carcinog. 2015 54 1584 1595 10.1002/mc.22231 25307746 23. Faber W. Stockmann M. Kruschke J.E. Denecke T. Bahra M. Seehofer D. Implication of microscopic and macroscopic vascular invasion for liver resection in patients with hepatocellular carcinoma Dig. Surg. 2014 31 204 209 10.1159/000365257 25196847 24. Roayaie S. Frischer J.S. Emre S.H. Fishbein T.M. Sheiner P.A. Sung M. Miller C.M. Schwartz M.E. Long-term results with multimodal adjuvant therapy and liver transplantation for the treatment of hepatocellular carcinomas larger than 5 centimeters Ann. Surg. 2002 235 533 539 10.1097/00000658-200204000-00012 11923610 25. Wu C.W. Farrell G.C. Yu J. Functional role of peroxisome-proliferator-activated receptor γ in hepatocellular carcinoma J. Gastroenterol. Hepatol. 2012 27 1665 1669 10.1111/j.1440-1746.2012.07213.x 22742931 26. Panigrahy D. Huang S. Kieran M.W. Kaipainen A. PPARγ as a therapeutic target for tumor angiogenesis and metastasis Cancer Biol. Ther. 2005 4 687 693 10.4161/cbt.4.7.2014 16082179 27. Panigrahy D. Singer S. Shen L.Q. Butterfield C.E. Freedman D.A. Chen E.J. Moses M.A. Kilroy S. Duensing S. Fletcher C. PPARγ ligands inhibit primary tumor growth and metastasis by inhibiting angiogenesis J. Clin. Investig. 2002 110 923 932 10.1172/JCI0215634 12370270 28. Peeters L.L. Vigne J.L. Tee M.K. Zhao D. Waite L.L. Taylor R.N. PPAR γ represses VEGF expression in human endometrial cells: Implications for uterine angiogenesis Angiogenesis 2005 8 373 379 10.1007/s10456-005-9027-4 16400517

PMC005xxxxxx/PMC5000625.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081227ijms-17-01227CommunicationAssociation between IRS1 Gene Polymorphism and Autism Spectrum Disorder: A Pilot Case-Control Study in Korean Males Park Hae Jeong 1Kim Su Kang 1Kang Won Sub 2Park Jin Kyung 2Kim Young Jong 2Nam Min 3Kim Jong Woo 2Chung Joo-Ho 1*Butler Merlin G. Academic EditorProkai-Tatrai Katalin Academic Editor1 Kohwang Medical Research Institute, School of Medicine, Kyung Hee University, Seoul 02447, Korea; [email protected] (H.J.P.); [email protected] (S.K.K.)2 Department of Neuropsychiatry, School of Medicine, Kyung Hee University, Seoul 02447, Korea; [email protected] (W.S.K.); [email protected] (J.K.P.); [email protected] (Y.J.K.); [email protected] (J.W.K.)3 Seoul Metropolitan Eunpyeong Hospital, Seoul 06801, Korea; [email protected]* Correspondence: [email protected]; Tel.: +82-2-961-028129 7 2016 8 2016 17 8 122711 5 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The insulin-like growth factor (IGF) pathway is thought to play an important role in brain development. Altered levels of IGFs and their signaling regulators have been shown in autism spectrum disorder (ASD) patients. In this study, we investigated whether coding region single-nucleotide polymorphisms (cSNPs) of the insulin receptor substrates (IRS1 and IRS2), key mediators of the IGF pathway, were associated with ASD in Korean males. Two cSNPs (rs1801123 of IRS1, and rs4773092 of IRS2) were genotyped using direct sequencing in 180 male ASD patients and 147 male control subjects. A significant association between rs1801123 of IRS1 and ASD was shown in additive (p = 0.022, odds ratio (OR) = 0.66, 95% confidence interval (CI) = 0.46–0.95) and dominant models (p = 0.013, OR = 0.57, 95% CI = 0.37–0.89). Allele frequency analysis also showed an association between rs1801123 and ASD (p = 0.022, OR = 0.66, 95% CI = 0.46–0.94). These results suggest that IRS1 may contribute to the susceptibility of ASD in Korean males. autism spectrum disorderinsulin receptor substratesingle nucleotide polymorphisminsulin-like growth factor ==== Body 1. Introduction Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social communication, impaired reciprocal social interaction, and repetitive patterns of behaviors or interests [1]. ASD has been reported to occur in approximately six out of 1000 births, affecting males and females in a ratio of 4:1 [2]. ASD has been found throughout the world across all racial, ethnic and social backgrounds. Although the cause of ASD remains elusive, accumulating evidence suggests that genetic factors play a prominent role in this disease. Heritability is estimated to be above 90% [3]. Inherited copy number variations (CNVs) and chromosomal abnormalities have shown to contribute to genetic vulnerability to ASD [4,5]. In addition, recent multiple genome-wide association studies suggest several common single-nucleotide polymorphisms (SNPs) as markers of ASD [6,7,8]. Previous studies have revealed the importance of the insulin-like growth factor (IGF) pathway in the development and maintenance of the central nervous system (CNS). During both development and adulthood, increased IGF1 and IGF2 levels were associated with increased neuronal complexity and impaired learning, suggested that IGFs support the activity-dependent neuronal plasticity underlying cognitive processes [9,10]. Moreover, recent studies have highlighted a significant role for IGF1 in complex social interactions [11,12]. It has been reported that levels of IGF1, IGF2, and IGF binding protein 3 (IGFBP3) were significantly increased in patients with ASD [13]. Furthermore, both cross-sectional and longitudinal studies have reported that a subset of ASD patients show age-dependent brain overgrowth [14,15]. Brain overgrowth at an early age could be caused by conditional overexpression of IGF1 in the brain, which is responsible for significant increases in brain volume during the embryonic and early postnatal period [16]. Indeed, the significant correlation between head circumference and IGF1 levels was also shown in ASD patients, but not in the controls [13]. On the other hand, other studies reported lower IGF1 levels in the cerebrospinal fluid (CSF) of autistic patients [17,18]. In addition, it was proposed that a reduced peripartum level of IGF1 due to genetic, epigenetic, or environmental factors may be a sentinel biomarker of increased probability of the later development of autism [19]. Although little is known about the biological function of the IGF pathway molecules in ASD, given the previous reports, we speculated that IGFs and their signaling regulator genes might be candidate genes involved in ASD. However, to our knowledge, there have not been any studies on the possible genetic association of IGFs or IGF signaling regulator genes with ASD. In this study, we focused on key molecules of the IGF pathway, insulin receptor substrate 1 (IRS1) and IRS2, which are the major cytosolic substrates of the IGF receptors and mediators for the downstream pathway processes [20,21]. Insulin has been regarded as primarily a metabolic signal, while IGFs has been implicated as an important mitogen and cell differentiation factor [22,23]. The IRS family contains several members (IRS1-6), of which IRS1 and IRS2 have been most widely studied. IRS1 and IRS2 regulate body weight control and glucose homeostasis [24]. They could also control body growth and peripheral insulin action. Thus, they have been suggested as markers of an active IGF pathway within tumors [25,26], although they are involved in insulin signaling. Indeed, polymorphisms of IRS1 and/or IRS2 have shown the significant associations with diabetes, glucose levels [27], and obesity [28], as well as with cancers, along with IGF signaling regulator genes [29,30,31]. Herein, we investigated the association of the coding region single-nucleotide polymorphisms (cSNP) of IRS1 and IRS2, active markers of the IGF pathway, with ASD in Korean males. 2. Results Two cSNPs of IRS1 and IRS2 were polymorphic, and the genotype distributions of the SNPs were in Hardy-Weinberg equilibrium (HWE) (p > 0.05; data not shown). We calculated the power of the sample size to verify our data using a genetic power calculator [32]. Considering a two-fold genotype relative risk, the sample powers of the SNPs were 0.900 (rs1801123, number of effective samples for 80% power = 148) and 0.967 (rs4773092, n = 110), respectively (α = 0.05). In addition, the sample powers of rs1801123 were 0.841 (n = 175) and 0.761 (n = 213) for a 1.9- and 1.8-fold relative risk, respectively. The sample powers of rs4773092 were 0.938 (n = 128) for a 1.9-fold relative risk, 0.889 (n = 154) for a 1.8-fold relative risk, and 0.889 (n = 189) for a 1.7-fold relative risk. Therefore, the results of our study had a significant power and sample size to detect the genotype relative risks up to 1.9-fold on rs1801123 and 1.8-fold on rs4773092. As shown in Table 1, rs1801123 of IRS1 was associated with ASD in additive (AG vs. GG vs. AA, p = 0.022, odds ratio (OR) = 0.66, 95% confidence interval (CI) = 0.46–0.95) and dominant models (p = 0.013, OR = 0.57, 95% CI = 0.37–0.89). The frequency of the genotypes containing the G allele (AG/GG, 36.7%) was decreased in the ASD patients compared to the control subjects (50.3%). In allele frequency analysis, we also found that rs1801123 was associated with ASD (p = 0.022, OR = 0.66, 95% CI = 0.46–0.94). The frequency of the G allele was lower in ASD patients (21.1%) than in control subjects (28.9%). This significance remained after the Bonferroni correction. Interestingly, when we analyzed the differences between patients with autistic disorder and healthy individuals, rs1801123 of IRS1 showed a statistically more significant association (Table 2). The association was revealed in the additive (p = 0.0037, OR = 0.56, 95% CI = 0.37–0.83) and dominant models (p = 0.0041, OR = 0.50, 95% CI = 0.31–0.81). Allele frequency analysis also revealed a stronger association between rs1801123 and autistic disorder (p = 0.004, OR = 0.56, 95% CI = 0.38–0.84). The frequencies of the AG/GG genotypes (33.6% and 50.3% in patients with autistic disorder and control subjects, respectively) and the G allele (18.6% and 28.9%) were more remarkably decreased in patients with autistic disorder compared to control subjects. 3. Discussion In our study, we found that rs1801123 of IRS1 was significantly associated with ASD in Korean males. The G allele of rs1801123 contributed to a decreased risk of ASD and, particularly, the contribution was potently shown in patients with autistic disorder. The IGF pathway plays an important role in regulating cell proliferation, differentiation and apoptosis, and, thus, IGFs and their signaling regulators have been studied in growth-, weight gain- , and obesity-related diseases [20,28,33,34,35]. IRS1 and IRS2 are key mediators of the IGF pathway [20,21]. Binding of IGFs to IGF receptors phosphorylates IRSs and triggers downstream cascades such as MAPK and PI3K/AKT signaling, which finally leads to cell proliferation and differentiation [20,21]. Thus, IRS1 and IRS2 together with IGFs and IGFRs have been involved in obesity, birth weight, diabetes mellitus, insulin sensitivity and cancer, showing the genetic associations of their polymorphisms [28,29,30,31,33,34]. In ASD patients, increased head growth, and particularly brain overgrowth in early life, has been reported with or without higher weights and body mass indexes (BMIs) Thus, several studies have suggested the involvement of growth-related hormones such as IGFs and their regulators, which lead to increased head growth and higher weights and BMIs, in the pathophysiology of autistic disorder/ASD [13,17,18]. Indeed, Mills et al. [13] reported increased levels of IGF1, IGF2, IGFBP3 and GHBP in the plasma of autistic disorder/ASD patients, and also showed a positive correlation between IGF1 level and head circumference in autistic disorder/ASD patients. On the other hands, IGFs and their regulators have been also reported to play a role in growth retardation. Indeed, transgenic mice lacking IRS1 showed prenatal and postnatal growth retardation [36,37]. In mice lacking IRS2, growth retardation was also observed, although it was minimal compared to mice lacking IRS1 [36]. Moreover, in the brain, IGFs and their regulators are essential factors for normal brain growth and development, as well as synaptogenesis and myelination [16,19,38]. They directly affect the rate that oligodendrocytes promote myelination, and thus factors which relatively reduce the production or availability of IGFs could retard normal nerve programming [19]. Indeed, in early laboratory embryos, the addition of IGF-receptor inhibitors blocked the normal formation of midbrain neurons [39]. IGF1 knockout mice had defective neurologic development [40]. Moreover, other studies on the relationship between IGF1 level and autistic patients reported lower IGF1 levels in the cerebrospinal fluid (CSF) of autistic patients, although the IGF1 levels of patients were compared to abnormal controls instead of normal controls due to ethical reasons that do not allow researchers to obtain CSF from normal control subjects [17,18]. Thus, relatively low activities of IGF pathway molecules such as IGF1, IRS1 and IRS2 may play a role as risk factors in the pathophysiology of autistic disorder/ASD, leading to the growth and development retardation. Hence, it is controversial which one among excessive activations and reduced availabilities of IGF-related factors is involved in the pathology of autistic disorder/ASD. In the present study, we found that rs1801123 of IRS1 was associated with ASD, and the association was shown more strongly in patients with autistic disorder. In particular, our results showed that the frequency of the minor G allele of rs1801123 was decreased in patients with autistic disorder/ASD; thus, the G allele of rs1801123 may contribute to a decreased risk of autistic disorder/ASD as a protective factor. In a previous study, carriers of the minor allele of rs1801123 (TG/GG) were reported to be associated with higher fasting plasma glucose and insulin levels [27]. Furthermore, the G allele of rs1801123 was associated with an increased risk of breast cancer in women carrying the BRCA1 mutation [30], although in a recent study of the same group using a large set of BRCA1 and BRCA2 mutation carriers, its lack of association was revealed [41]. Moreover, the G allele of rs1801123 was significantly associated with lymph node involvement in estrogen-receptor-positive primary invasive breast cancer patients, who were treated with surgery and tamoxifen [29]. These reports indicated that the minor allele of rs1801123 may be involved in the increased product and ability of the IGFs, along with the activation of the insulin-related signaling pathway. Therefore, the decreased frequency of the minor allele of rs1801123 in autistic disorder/ASD patients in our study may contribute to relatively low activation of the IGF pathway molecules. Taken together, we postulated that the activation of IGF pathway molecules may be reduced in autistic disorder/ASD patients, although it is controversial as mentioned above. Also, the G allele of rs1801123 may play a role as a protective factor against the decreased activity of the IGF pathway in autistic disorder/ASD. Further studies are needed to determine how rs1801123 and the IGF pathway affect the pathophysiology of autistic disorder/ASD. Our study is the first pilot to report an association of the IRS1 with autistic disorder/ASD. The limitation of our study is that only one SNP of each IRS1 and IRS2 was selected and analyzed. Replication studies are needed to determine the association between the IRS1 and autistic disorder/ASD, as well as the lack of association of IRS2, analyzing more polymorphisms in addition to rs1801123 and rs4773092. Moreover, as shown in our sample power analysis, our results have statistical confidence, only assuming a genotype relative risk up to 1.9-fold on rs1801123. Thus, the relatively small sample size limits the generalizability of the findings from the present study. Our findings are preliminary and need to be validated in further studies with larger sample sizes. Our work provides evidence that the IRS1 gene may play a role in the pathophysiology of ASD. 4. Experimental Section 4.1. Subjects One hundred eighty male ASD patients (mean age ± standard deviation (SD), 15.5 ± 4.8 years) and 147 healthy male individuals (39.9 ± 5.8 years) were enrolled in this study. ASD patients were diagnosed with ASDs by well-trained psychiatrists, child and adolescent specialists according to Diagnostic and Statistical Manual of Mental Disorders, 4th ed (DSM-IV) criteria [42], using available historical information from interviews and clinical records. Each ASD patient was also evaluated using the Childhood Autism Rating Scale (CARS) [43], one of the most widely used instrument to evaluate the developmental degree of autism, applying cut-off score of 30. The average of CARS score was 38.6 ± 5.8 (mean ± SD). The ASD group consisted of 137 patients with autistic disorder, 11 patients with Asperger’s disorder, and 32 patients with Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS). A summary of clinical characteristics of ASD patients is provided in Table 3. The healthy adult controls were recruited from subjects who visited the hospital for routine health checkups. Controls were investigated to determine whether they or their first-degree relatives had psychiatric disturbances or previous psychiatric treatment through personal interviews. Only unaffected subjects with no psychiatric disorder or family history were included in this study. All the ASD patients and control subjects were of Korean background. The present study was conducted in accordance with the guidelines of the Helsinki Declaration and was approved by the Ethics Review Committee of Medical Research Institute, Kyung Hee University Medical Center on 15 September 2004 (2004-09-15). Written informed consents were obtained from the parents or guardians of ASD patients and control subjects. 4.2. Single-Nucleotide Polymorphism (SNP) Selection and Genotyping Of SNPs in the IRS1 and IRS2, cSNPs were targeted and selected from the National Center for Biotechnology Information SNP database [44]. Some cSNPs alter a functionally important amino acid residue, and these are of interest for their potential links with phenotype. Other cSNPs may prove useful for their potential links to functional cSNPs via linkage disequilibrium mapping [45,46]. We selected common SNPs with a minor allele frequency of >0.1 in Chinese and Japanese populations, excluding SNPs without data on genotype frequency. Finally, we selected two cSNP (rs1801123 (Ala804Ala) of IRS1, and rs4773092 (Cys816Cys) of IRS2). Genomic DNA was extracted from the whole blood of each subject using the High Pure PCR Template Preparation kit (Roche, Mannheim, Germany) following the manufacturer’s protocol. SNP genotyping was conducted with direct sequencing using the following primers for each SNP: rs1801123 in IRS1 (sense, 5′-TCCTACTACTCATTGCCAAGATC-3′; antisense, 5′-CTATTGGTCTGAGCAGCTGTGT-3′), and rs4773092 in IRS2 (sense, 5′-ATGTGGTGCGGTTCCAAGCTGT-3′; antisense, 5′-GCCAAAGTCGATGTTGATGTACT-3′). The PCR products were sequenced using the ABI PRISM 3730XL analyzer (PE Applied Biosystems, Foster City, CA, USA), and sequence data were then analyzed using SeqManII software (DNASTAR Inc., Madison, WI, USA). 4.3. Statistical Analysis SNPStats [47] and SPSS 18.0 software (SPSS Inc., Chicago, IL, USA) were used to analyze the genetic data and the HWE. The association between SNP genotypes and ASD were estimated by computing the ORs and their 95% CIs with logistic regression analyses. In the logistic regression analysis for each SNP, the following models were used: codominant inheritance (that is, where the relative hazard differed between subjects with one minor allele and those with two minor alleles), dominant inheritance (subjects with one or two minor alleles had the same relative hazard for the disease), or recessive inheritance (subjects with two minor alleles were at increased risk of the disease). The chi-square test was used to compare allele frequencies between groups. To avoid chance findings due to multiple testing, a Bonferroni correction was applied by lowering the significance levels to p = 0.025 (p = 0.05/2) for two SNPs. Acknowledgments This work was supported by a grant from Kyung Hee University (Seoul, Korea) in 2012 (KHU-20121738). Author Contributions Joo-Ho Chung designed and directed the whole project. Won Sub Kang, Jin Kyung Park, Young Jong Kim, Min Nam and Jong Woo Kim collected the blood samples from ASD patients and control subjects. Hae Jeong Park and Su Kang Kim performed the experiments, collected the results, and analyzed the data. Joo-Ho Chung and Hae Jeong Park discussed and interpreted the data and results. Hae Jeong Park wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. ijms-17-01227-t001_Table 1Table 1 Multiple logistic regression analysis of IRS1 and IRS2 polymorphisms in autism spectrum disorder (ASD) patients and control subjects. SNP Model/Allele Genotype Control ASD OR (95% CI) p n (%) n (%) rs1801123 Additive AA 73 (49.7) 114 (63.3) 1 Ala804Ala AG 63 (42.9) 56 (31.1) IRS1 GG 11 (7.5) 10 (5.6) 0.66 (0.46–0.95) 0.022 Dominant AA 73 (49.7) 114 (63.3) 1 AG/GG 74 (50.3) 66 (36.7) 0.57 (0.37–0.89) 0.013 Recessive AA/AG 136 (92.5) 170 (94.4) 1 GG 11 (7.5) 10 (5.6) 0.73 (0.30–1.76) 0.48 Allele A 209 (71.1) 284 (78.9) 1 G 85 (28.9) 76 (21.1) 0.66 (0.46–0.94) 0.022 rs4773092 Additive AA 41 (27.9) 51 (28.3) 1 Cys816Cys AG 76 (51.7) 95 (52.8) IRS2 GG 30 (20.4) 34 (18.9) 0.96 (0.70–1.32) 0.8 Dominant AA 41 (27.9) 51 (28.3) 1 AG/GG 106 (72.1) 129 (71.7) 0.98 (0.60–1.59) 0.93 Recessive AA/AG 117 (79.6) 146 (81.1) 1 GG 30 (20.4) 34 (18.9) 0.91 (0.53–1.57) 0.73 Allele A 158 (53.7) 197 (54.7) 1 G 136 (46.3) 163 (45.3) 0.96 (0.70–1.31) 0.8 Bold characters represent statistically significant values (p < 0.025). ASD, autism spectrum disorder. 1—It is a statistical reference in our genetic analysis. ijms-17-01227-t002_Table 2Table 2 Multiple logistic regression analysis of IRS1 and IRS2 polymorphisms in patients with autistic disorder and control subjects. SNP Model/allele Genotype Control Autistic Disorder OR (95% CI) p n (%) n (%) rs1801123 Additive AA 73 (49.7) 91 (66.4) 1 Ala804Ala AG 63 (42.9) 41 (29.9) IRS1 GG 11 (7.5) 5 (3.6) 0.56 (0.37–0.83) 0.0037 Dominant AA 73 (49.7) 91 (66.4) 1 AG/GG 74 (50.3) 46 (33.6) 0.50 (0.31–0.81) 0.0041 Recessive AA/AG 136 (92.5) 132 (96.3) 1 GG 11 (7.5) 5 (3.6) 0.47 (0.16–1.38) 0.16 Allele A 209 (71.1) 223 (81.4) 1 G 85 (28.9) 51 (18.6) 0.56 (0.38–0.84) 0.004 rs4773092 Additive AA 41 (27.9) 36 (26.3) 1 Cys816Cys AG 76 (51.7) 70 (51.1) IRS2 GG 30 (20.4) 31 (22.6) 1.08 (0.77–1.51) 0.64 Dominant AA 41 (27.9) 36 (26.3) 1 0.76 AG/GG 106 (72.1) 101 (73.7) 1.09 (0.64–1.83) Recessive AA/AG 117 (79.6) 106 (77.4) 1 0.65 GG 30 (20.4) 31 (22.6) 1.14 (0.65–2.01) Allele A 158 (53.7) 142 (51.8) 1 G 136 (46.3) 132 (48.2) 0.96 (0.70–1.31) 0.8 Bold characters represent statistically significant values (p < 0.025). 1—It is a statistical reference in our genetic analysis. ijms-17-01227-t003_Table 3Table 3 Clinical characteristics of ASD patients and control subjects. Characteristics ASDs Control Total no. of subject 180 147 Age (mean ± SD, years) 15.5 ± 4.8 39.9 ± 5.8 CARS score 38.6 ± 5.8 Autistic disorder (n = 137) 41.1 ± 4.3 Asperger‘s disorder (n = 11) 30.7 ± 0.4 PDD-NOS (n = 32) 30.9 ± 1.0 ASD, autism spectrum disorder; CARS, Childhood Autism Rating Scale; PDD-NOS, Pervasive Developmental Disorder-Not Otherwise Specified. ==== Refs References 1. Tanguay P.E. Pervasive developmental disorders: A 10-year review J. Am. Acad. Child Adolesc. Psychiatry 2000 39 1079 1095 10986804 2. Nicholas J.S. Charles J.M. Carpenter L.A. King L.B. Jenner W. Spratt E.G. Prevalence and characteristics of children with autism-spectrum disorders Ann. Epidemiol. 2008 18 130 136 18083540 3. Burmeister M. McInnis M.G. Zollner S. Psychiatric genetics: Progress amid controversy Nat. Rev. Genet. 2008 9 527 540 18560438 4. Nowakowska B.A. de Leeuw N. Ruivenkamp C.A. Sikkema-Raddatz B. Crolla J.A. Thoelen R. Koopmans M. den Hollander N. van Haeringen A. van der Kevie-Kersemaekers A.M. Parental insertional balanced translocations are an important cause of apparently de novo CNVs in patients with developmental anomalies Eur. J. Hum. Genet. 2012 20 166 170 21915152 5. Carter M.T. Nikkel S.M. Fernandez B.A. Marshall C.R. Noor A. Lionel A.C. Prasad A. Pinto D. Joseph-George A.M. Noakes C. Hemizygous deletions on chromosome 1p21.3 involving the DPYD gene in individuals with autism spectrum disorder Clin. Genet. 2011 80 435 443 21114665 6. Anney R. Klei L. Pinto D. Regan R. Conroy J. Magalhaes T.R. Correia C. Abrahams B.S. Sykes N. Pagnamenta A.T. A genome-wide scan for common alleles affecting risk for autism Hum. Mol. Genet. 2010 19 4072 4082 20663923 7. Weiss L.A. Arking D.E. Gene Discovery Project of Johns Hopkins and the Autism Consortium. A genome-wide linkage and association scan reveals novel loci for autism Nature 2009 461 802 808 19812673 8. Wang K. Zhang H. Ma D. Bucan M. Glessner J.T. Abrahams B.S. Salyakina D. Imielinski M. Bradfield J.P. Sleiman P.M. Common genetic variants on 5p14.1 associate with autism spectrum disorders Nature 2009 459 528 533 19404256 9. Chen D.Y. Stern S.A. Garcia-Osta A. Saunier-Rebori B. Pollonini G. Bambah-Mukku D. Blitzer R.D. Alberini C.M. A critical role for IGF-II in memory consolidation and enhancement Nature 2011 469 491 497 21270887 10. O’Kusky J.R. Ye P. D’Ercole A.J. Insulin-like growth factor-I promotes neurogenesis and synaptogenesis in the hippocampal dentate gyrus during postnatal development J. Neurosci. 2000 20 8435 8442 11069951 11. Schaevitz L.R. Moriuchi J.M. Nag N. Mellot T.J. Berger-Sweeney J. Cognitive and social functions and growth factors in a mouse model of Rett syndrome Physiol. Behav. 2010 100 255 263 20045424 12. Moy S.S. Nadler J.J. Young N.B. Nonneman R.J. Grossman A.W. Murphy D.L. D’Ercole A.J. Crawley J.N. Magnuson T.R. Lauder J.M. Social approach in genetically engineered mouse lines relevant to autism Genes Brain Behav. 2009 8 129 142 19016890 13. Mills J.L. Hediger M.L. Molloy C.A. Chrousos G.P. Manning-Courtney P. Yu K.F. Brasington M. England L.J. Elevated levels of growth-related hormones in autism and autism spectrum disorder Clin. Endocrinol. 2007 67 230 237 14. Courchesne E. Carper R. Akshoomoff N. Evidence of brain overgrowth in the first year of life in autism JAMA 2003 290 337 344 12865374 15. Courchesne E. Karns C.M. Davis H.R. Ziccardi R. Carper R.A. Tigue Z.D. Chisum H.J. Moses P. Pierce K. Lord C. Unusual brain growth patterns in early life in patients with autistic disorder: An MRI study Neurology 2001 57 245 254 11468308 16. Popken G.J. Hodge R.D. Ye P. Zhang J. Ng W. O’Kusky J.R. D’Ercole A.J. In vivo effects of insulin-like growth factor-I (IGF-I) on prenatal and early postnatal development of the central nervous system Eur. J. Neurosci. 2004 19 2056 2068 15090033 17. Riikonen R. Neurotrophic factors in the pathogenesis of Rett syndrome J. Child Neurol. 2003 18 693 697 14649551 18. Vanhala R. Turpeinen U. Riikonen R. Low levels of insulin-like growth factor-I in cerebrospinal fluid in children with autism Dev. Med. Child Neurol. 2001 43 614 616 11570630 19. Steinman G. Mankuta D. Insulin-like growth factor and the etiology of autism Med. Hypotheses 2013 80 475 480 23375408 20. Laban C. Bustin S.A. Jenkins P.J. The GH-IGF-I axis and breast cancer Trends Endocrinol. Metab. 2003 14 28 34 12475609 21. Jones J.I. Clemmons D.R. Insulin-like growth factors and their binding proteins: Biological actions Endocr. Rev. 1995 16 3 34 7758431 22. White M.F. Insulin signaling in health and disease Science 2003 302 1710 1711 14657487 23. Burks D.J. White M.F. IRS proteins and β-cell function Diabetes 2001 50 S140 S145 11272176 24. Withers D.J. Burks D.J. Towery H.H. Altamuro S.L. Flint C.L. White M.F. IRS-2 coordinates IGF-1 receptor-mediated beta-cell development and peripheral insulin signalling Nat. Genet. 1999 23 32 40 10471495 25. Werner H. Le Roith D. The insulin-like growth factor-I receptor signaling pathways are important for tumorigenesis and inhibition of apoptosis Crit. Rev. Oncog. 1997 8 71 92 9516087 26. Van Obberghen E. Baron V. Scimeca J.C. Kaliman P. Insulin receptor: Receptor activation and signal transduction Adv. Second Messenger Phosphoprot. Res. 1993 28 195 201 27. Feng X. Tucker K.L. Parnell L.D. Shen J. Lee Y.C. Ordovas J.M. Ling W.H. Lai C.Q. Insulin receptor substrate 1 (IRS1) variants confer risk of diabetes in the Boston Puerto Rican Health Study Asia Pac. J. Clin. Nutr. 2013 22 150 159 23353623 28. Lautier C. El Mkadem S.A. Renard E. Brun J.F. Gris J.C. Bringer J. Grigorescu F. Complex haplotypes of IRS2 gene are associated with severe obesity and reveal heterogeneity in the effect of Gly1057Asp mutation Hum. Genet. 2003 113 34 43 12687350 29. Winder T. Giamas G. Wilson P.M. Zhang W. Yang D. Bohanes P. Ning Y. Gerger A. Stebbing J. Lenz H.J. Insulin-like growth factor receptor polymorphism defines clinical outcome in estrogen receptor-positive breast cancer patients treated with tamoxifen Pharmacogen. J. 2014 14 28 34 30. Neuhausen S.L. Brummel S. Ding Y.C. Singer C.F. Pfeiler G. Lynch H.T. Nathanson K.L. Rebbeck T.R. Garber J.E. Couch F. Genetic variation in insulin-like growth factor signaling genes and breast cancer risk among BRCA1 and BRCA2 carriers Breast Cancer Res. 2009 11 R76 19843326 31. Slattery M.L. Samowitz W. Curtin K. Ma K.N. Hoffman M. Caan B. Neuhausen S. Associations among IRS1, IRS2, IGF1, and IGFBP3 genetic polymorphisms and colorectal cancer Cancer Epidemiol. Biomark. Prev. 2004 13 1206 1214 32. Genetic Power Calculator Available online: http://pngu.mgh.harvard.edu/~purcell/gpc/cc2.html (accessed on 28 July 2016) 33. Rasmussen S.K. Urhammer S.A. Hansen T. Almind K. Moller A.M. Borch-Johnsen K. Pedersen O. Variability of the insulin receptor substrate-1, hepatocyte nuclear factor-1α (HNF-1α ), HNF-4α , and HNF-6 genes and size at birth in a population-based sample of young Danish subjects J. Clin. Endocrinol. Metab. 2000 85 2951 2953 10946909 34. Almind K. Frederiksen S.K. Bernal D. Hansen T. Ambye L. Urhammer S. Ekstrom C.T. Berglund L. Reneland R. Lithell H. Search for variants of the gene-promoter and the potential phosphotyrosine encoding sequence of the insulin receptor substrate-2 gene: Evaluation of their relation with alterations in insulin secretion and insulin sensitivity Diabetologia 1999 42 1244 1249 10525667 35. Almind K. Bjorbaek C. Vestergaard H. Hansen T. Echwald S. Pedersen O. Aminoacid polymorphisms of insulin receptor substrate-1 in non-insulin-dependent diabetes mellitus Lancet 1993 342 828 832 8104271 36. Withers D.J. Gutierrez J.S. Towery H. Burks D.J. Ren J.M. Previs S. Zhang Y. Bernal D. Pons S. Shulman G.I. Disruption of IRS-2 causes type 2 diabetes in mice Nature 1998 391 900 904 9495343 37. Tamemoto H. Kadowaki T. Tobe K. Yagi T. Sakura H. Hayakawa T. Terauchi Y. Ueki K. Kaburagi Y. Satoh S. Insulin resistance and growth retardation in mice lacking insulin receptor substrate-1 Nature 1994 372 182 186 7969452 38. Anlar B. Sullivan K.A. Feldman E.L. Insulin-like growth factor-I and central nervous system development Horm. Metab. Res. 1999 31 120 125 10226791 39. Onuma T.A. Ding Y. Abraham E. Zohar Y. Ando H. Duan C. Regulation of temporal and spatial organization of newborn GnRH neurons by IGF signaling in zebrafish J. Neurosci. 2011 31 11814 11824 21849542 40. Beck K.D. Powell-Braxton L. Widmer H.R. Valverde J. Hefti F. Igf1 gene disruption results in reduced brain size, CNS hypomyelination, and loss of hippocampal granule and striatal parvalbumin-containing neurons Neuron 1995 14 717 730 7718235 41. Ding Y.C. McGuffog L. Healey S. Friedman E. Laitman Y. Paluch-Shimon S. Kaufman B. Swe B. Liljegren A. Lindblom A. A nonsynonymous polymorphism in IRS1 modifies risk of developing breast and ovarian cancers in BRCA1 and ovarian cancer in BRCA2 mutation carriers Cancer Epidemiol. Biomark. Prev. 2012 21 1362 1370 42. Association Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders 4th ed. American Psychiatric Press Washington, DC, USA 1994 43. Schopler E. Reichier R.J. Renner B.R. Childhood Autism Rating Scale Western Psychological Services Los Angeles, CA, USA 1988 44. NCBI databases Available online: http://www.ncbi.nlm.nih.gov/SNP (accessed on 28 July 2016) 45. Collins F.S. Guyer M.S. Chakravarti A. Variations on a theme: Cataloging human DNA sequence variations Science 1997 278 1580 1581 9411782 46. Garg K. Green P. Nickerson D.A. Identification of candidate coding region single nucleotide polymorphisms in 165 human genes using assembled expressed sequence tags Genome Res. 1999 9 1087 1092 10568748 47. ICO Available online: http://bioinfo.iconcologia.net/index.php (accessed on 28 July 2016)

PMC005xxxxxx/PMC5000626.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081228ijms-17-01228ArticleUrinary Dopamine as a Potential Index of the Transport Activity of Multidrug and Toxin Extrusion in the Kidney Kajiwara Moto 1*Ban Tsuyoshi 2Matsubara Kazuo 2Nakanishi Yoichi 3Masuda Satohiro 1Cho William Chi-shing Academic Editor1 Department of Pharmacy, Kyushu University Hospital, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan; [email protected] Department of Clinical Pharmacology and Therapeutics, Kyoto University Hospital, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan; [email protected] (T.B.); [email protected] (K.M.)3 Research Institute for Diseases of the Chest, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan; [email protected]* Correspondence: [email protected]; Tel.: +81-92-642-592130 7 2016 8 2016 17 8 122831 5 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Dopamine is a cationic natriuretic catecholamine synthesized in proximal tubular cells (PTCs) of the kidney before secretion into the lumen, a key site of its action. However, the molecular mechanisms underlying dopamine secretion into the lumen remain unclear. Multidrug and toxin extrusion (MATE) is a H+/organic cation antiporter that is highly expressed in the brush border membrane of PTCs and mediates the efflux of organic cations, including metformin and cisplatin, from the epithelial cells into the urine. Therefore, we hypothesized that MATE mediates dopamine secretion, a cationic catecholamine, into the tubule lumen, thereby regulating natriuresis. Here, we show that [3H]dopamine uptake in human (h) MATE1-, hMATE-2K- and mouse (m) MATE-expressing cells exhibited saturable kinetics. Fluid retention and decreased urinary excretion of dopamine and Na+ were observed in Mate1-knockout mice compared to that in wild-type mice. Imatinib, a MATE inhibitor, inhibited [3H]dopamine uptake by hMATE1-, hMATE2-K- and mMATE1-expressing cells in a concentration-dependent manner. At clinically-relevant concentrations, imatinib inhibited [3H]dopamine uptake by hMATE1- and hMATE2-K-expressing cells. The urinary excretion of dopamine and Na+ decreased and fluid retention occurred in imatinib-treated mice. In conclusion, MATE transporters secrete renally-synthesized dopamine, and therefore, urinary dopamine has the potential to be an index of the MATE transporter activity. dopamineMATEnatriuresisimatinibfluid retention ==== Body 1. Introduction Dopamine is a cationic natriuretic catecholamine. Excretion of this hormone and urinary Na+ concentration are both increased by Na+ intake and acute saline infusion [1,2,3]. Dopamine receptors are classified into two groups, D1-like (D1 and D5) and D2-like (D2, D3 and D4), which are both expressed in the kidney [4]. After moderate Na+ loading, more than 50% of the incremental urinary Na+ excretion is attributable to the stimulation of D1-like receptors with renally-synthesized dopamine [1,5]. The subsequent increase in urinary Na+ excretion is accompanied by an increase in urine output [1,5]. A study in humans indicates that urinary dopamine is derived from the kidney since plasma dopamine concentration (0.43 ± 0.06 nM) and total plasma volume are insufficient to achieve the almost 1000-fold greater urinary dopamine concentration (0.63 ± 0.17 µM) [6]. Renal dopamine synthesis is restricted to proximal tubular cells (PTCs), which internalize the circulating and glomerular-filtered forms of l-dihydroxyphenylalanine (l-DOPA) via dimers of the 4F2 heavy chain/l-type amino acid transporter 2 and related to B°,+ amino acid transporter/B°,+-type amino acid transporter, respectively [7,8]. PTCs express aromatic amino acid decarboxylase (AADC) [9], which converts internalized l-DOPA to dopamine. Renally-synthesized dopamine is secreted into the tubular lumen and acts at dopamine receptors expressed in multiple nephron segments, thereby inhibiting the Na+ transport activity of various targets, including the Na+/H+ exchanger (NHE)-1, NHE3, Na+/P cotransporter IIa, Na+/HCO3− cotransporter, Cl−/HCO3− exchanger and Na+/K+ ATPase [10]. Although dopamine secretion from PTCs to the lumen is a key step in natriuresis, the molecular mechanisms underlying dopamine secretion remain unknown. In the current study, we focused on multidrug and toxin extrusion (MATE), which is also known as SLC47A, a candidate transporter that provides dopamine into the proximal tubular lumen (PTL). MATE is a H+/organic cation antiporter that is highly expressed in brush border membranes of PTCs and mediates the tubular secretion of organic cations by using a H+ gradient [11,12,13]. MATE1 and MATE2-K are expressed in the human kidney tissue, whereas Mate1 is expressed in mice [13,14]. Organic cations, such as tetraethylammonium, cimetidine, metformin, creatinine and varenicline, are typical substrates for MATE transporters [15,16]. Because they transport several clinically-important drugs, MATE1 and MATE2-K are included in the battery of the in vitro tests used in the process of new drug development, as recommended by the International Transporter Consortium, the European Medicines Agency and the U.S. Food and Drug Administration [17,18,19]. MATE transporters play a critical role in the excretion of metformin, a biguanide antidiabetic drug that is mainly excreted in the urine in a non-metabolized form. Tubular secretion plays a major role in this process since the renal metformin clearance is almost five-fold greater than its creatinine clearance is [20]. In Mate1-knockout mice, the area under the blood concentration-time curve of metformin at 60 min and renal secretory clearance of metformin were two-fold higher and 86% lower, respectively, than the respective values in wild-type mice were [14]. An in vitro uptake study showed that tyrosine kinase inhibitors blocked [14C]metformin uptake by human MATE transporters; imatinib was the most effective agent, which displayed the lowest half-maximal inhibitory concentration (IC50) of all of the drugs tested [21]. In this study, we carried out uptake experiments and acute saline infusion experiments in Mate1-knockout mice and MATE inhibitor (imatinib)-treated mice. The results indicate that MATE facilitated the transfer of dopamine into the PTL and promoted natriuresis, and therefore, urinary dopamine has potential usefulness as a noninvasive index of transport activity of MATE in the kidney. 2. Results 2.1. Dopamine Transport Is Mediated by Multidrug and Toxin Extrusion (MATE) To examine whether dopamine is a substrate of MATE transporters, we carried out uptake experiments, which are frequently used to evaluate MATE transport properties [21,22,23,24,25]. Significant uptake of [3H]dopamine by human (h) MATE1 (Figure 1a), hMATE2-K (Figure 1a) and mouse (m) MATE1 (Figure 1b) compared to that of the cells transfected with an empty vector was observed at each time point (p < 0.01). The transport characteristics of MATE transporters were examined at 1 min in subsequent experiments because of technical limitations and reproducibility. An “overshoot” was observed because the driving force (an outward H+ gradient for MATE) was depleted during the uptake experiments, and substrate backflow occurred [26,27]. To estimate the kinetic parameters for [3H]dopamine uptake by hMATE1, hMATE2-K and mMATE1, concentration-dependent uptake was examined, and the dopamine uptake by all three transporters exhibited saturable kinetics, following the Michaelis–Menten equation (Figure 1c,d,e). The apparent maximal uptake velocity (Vmax), Michaelis–Menten constant (Km) and Vmax/Km values are summarized in Table 1. The rank order of the [3H]dopamine transport activity (Vmax/Km) was mMATE1 > hMATE1, hMATE2-K. 2.2. Effects of Mate1 Knockout on Urinary Dopamine and Na+ Excretion in Mice Next, to clarify the MATE transporter-mediated renal tubular secretion of dopamine and consequent promotion of renal Na+ excretion in vivo, we carried out acute saline volume expansion experiments in wild-type and Mate1-knockout mice. This is because intravenous saline infusion is known to accelerate dopamine synthesis in the kidney and promotes urinary Na+ excretion [28]. The results revealed that urinary dopamine was barely detectable in Mate1-knockout mice (Figure 2a). The renal dopamine level was 1.5-fold higher in the Mate1-knockout mice than it was in their wild-type counterparts after acute saline volume expansion (Figure 2b). These results show that the urinary dopamine excretion was impaired by Mate1 knockout and explained why dopamine accumulates in the kidneys. Because renally-synthesized dopamine is a natriuretic catecholamine, we examined the effect of urinary dopamine depletion in Mate1-knockout mice. Volume expansion induced a 12.3-fold increase in urinary Na+ excretion in wild-type mice, whereas that in the Mate1-knockout group decreased to a 1.5-fold increase (Figure 2c). The urinary K+ excretion slightly increased by 1.7-fold in the wild-type mice (Figure 2d) compared to the change in Na+ excretion (Figure 2c). The changes in urinary Cl− excretion between the control and volume expansion groups were similar to that of urinary Na+ excretion. Specifically, there were 10.4- and 2.7-fold increases in urinary Na+ excretion in the wild-type and Mate1-knockout mice, respectively (Figure 2e). Furthermore, the urinary volume increased by 5.7-fold and 1.7-fold in the wild-type mice in Mate1-knockout mice, respectively (Figure 2f). Together, these results indicate that Mate1 knockout impairs natriuresis because excretion of dopamine into the tubular lumen is impaired. Considering that Mate1 knockout impairs natriuresis, we assessed whether it also caused fluid retention. We discovered that the ratio of total body water weight to total body weight of the Mate1-knockout mice was significantly higher than that of wild-type mice (p < 0.01; Figure 2g). This result indicated that fluid retention occurred in Mate1 knockout mice. The body weights were similar between wild-type and Mate1-knockout animals (Table 2). Furthermore, the blood Na+, K+ and glucose levels were weakly changed by Mate1 knockout, but these results were statistically significant (Table 2). To examine whether Mate1 knockout alters the dopamine receptor localization, we examined the expression of D1 and D5 (D1-like receptors) in mouse kidneys because D1-like receptors are responsible for over 50% of the dopamine-induced natriuresis [1,5]. Immunohistochemical analysis revealed that localization of both receptor subtypes was similar in the kidneys of the wild-type and Mate1-knockout mice (Figure 3a–d). We also examined the expression of the NHE3 transporter because it plays a dominant role in Na+ reabsorption [29], and we discovered that it was also similarly localized in the kidneys of wild-type and Mate1-knockout mice (Figure 3e,f). 2.3. Effects of Imatinib on Urinary Dopamine and Na+ Excretion in Mice Since imatinib inhibits MATE [21] and edema is a common side effect in patients treated with imatinib [30], we tested whether imatinib inhibits urinary dopamine excretion in mice. We discovered that the urinary dopamine excretion was significantly decreased in the imatinib-treated group than it was in the vehicle-treated group during the control period (i.e., at a moderate saline infusion rate; Figure 4a). The renal dopamine levels of the imatinib-treated group after acute saline volume expansion were similar to those of the vehicle-treated groups (Figure 4b). During the control period, imatinib administration decreased the urinary Na+ and Cl− excretion (Figure 4c,e); in addition to these ions, it also decreased K+ and urinary volume during the volume expansion (Figure 4c–f). In the vehicle-treated group, there was a significant increase in urinary Na+, K+ and Cl− excretion (3.5-, 1.8- and 3.0-fold; Figure 4c–e, respectively) during the volume expansion treatment. Furthermore, imatinib administration increased the ratio of total body water weight to total body weight of mice (Figure 4g) without altering their body weights (Figure 4h). The blood total CO2 level was lower in the imatinib- than it was in vehicle-treated mice (Table 3). These results indicate that imatinib impaired urinary dopamine excretion and natriuresis. 2.4. Imatinib Inhibits MATE-Dependent Uptake of Dopamine To determine whether imatinib affects MATE-mediated dopamine transport, we carried out [3H]dopamine uptake experiments in the presence of imatinib. The results revealed that dopamine transport activities of hMATE1, hMATE2-K and mMATE1 were inhibited by imatinib in a dose-dependent manner, and the calculated IC50 values were 1.1, 13.8 and 100.6 µM for hMATE1, hMATE2-K and mMATE1, respectively (Table 4). In the evaluation of ratio of total body water weight to total body weight, imatinib reached a concentration of 92.1 and 59.0 µM in the plasma and kidney, respectively. In the acute saline volume expansion experiments, plasma and kidney concentrations were 65.6 and 347.1 µM, respectively. We calculated the dopamine transport activity of mMATE1 relative to the renal imatinib concentration. There was a 37% reduction in mMATE1 activity at 59.0 µM imatinib and a 77% reduction at 347.1 µM. Taken together, these results support the hypothesis that imatinib inhibits natriuresis by disturbing the MATE-mediated dopamine secretion into the tubular lumen and consequently causes fluid retention. We also examined the effects of the tyrosine kinase inhibitors, dasatinib and nilotinib, on MATE-dependent dopamine uptake. We found that dasatinib inhibited MATE-dependent [3H]dopamine uptake with IC50 values of 7.1 µM for hMATE1, 4.1 µM for hMATE2-K and 106.2 µM for mMATE1 (Table 4). However, IC50 values could not be determined for nilotinib because of its poor solubility (Table 4). 3. Discussion A previous study designed to elucidate the mechanism of renal dopamine secretion used a porcine-derived renal epithelial cell line (LLC-PK1) [31]. This line possesses proximal tubule cell-like properties and releases dopamine synthesized from l-DOPA [31,32]. Because both cocaine (a non-selective competitive inhibitor of monoamine transporters) and GBR-12909 (a specific dopamine transporter inhibitor) failed to inhibit outward transfer of synthesized dopamine in LLC-PK1, it was postulated that monoamine transporters are not involved in the secretion of renally-synthesized dopamine into the tubular lumen [31]. The inside of the cells is generally negatively charged. Therefore, uptake of organic cations such as dopamine from the circulation into PTCs is driven by a downhill gradient through a potential-driven facilitated diffusion process. In contrast, the secretion of organic cations into the tubular lumen from PTCs goes against an uphill charge gradient [33]. Uphill transport requires a concomitant driving force, which is provided by an oppositely-directed H+ gradient in the case of MATE and by adenosine triphosphate (ATP) in the case of P-glycoprotein (P-gp). P-gp transports large and hydrophobic cationic drugs, including digoxin, anticancer agents, cyclosporine and tacrolimus [34]. An in vitro study showed that vesicular monoamine transporter (VMAT) 1 sequestered renally-synthesized dopamine, and exocytosis was involved in dopamine release from PTCs [35]. However, the effect of a VMAT inhibitor on dopamine release was mild [35]. A vesicle uptake study revealed that a H+/organic cation antiporter system is present in the apical membrane of LLC-PK1 cells [36]. Furthermore, urinary dopamine excretion is defective in Mate1-knockout mice (Figure 2a). Therefore, MATE, rather than P-gp, exocytosis or monoamine transporters, mediates dopamine secretion into the tubular lumen. In the human kidney, both hMATE1 and hMATE2-K appear important for dopamine secretion into the PTL, since the levels of their respective mRNAs were similar in this tissue (approximately 80 and 60 amol/µg total RNA, respectively) [13]. Furthermore, the [3H]dopamine transport activity of hMATE1 and hMATE2-K was similar (Table 1). Functionally null mutants with complete loss of hMATE1 and hMATE2-K transport activities have been identified [37]. Only heterozygous carriers have been identified, and the allelic frequency of the loss-of-function mutant is low, at 0.6% for hMATE1 and 1.7% for hMATE2-K [37]. Heterozygous MATE variants do not affect oral clearance of metformin [38]. Therefore, MATE transporters may play important physiological roles, and hMATE1 and hMATE2-K may be considered to compensate for each other. In PTCs, dopamine receptors are expressed in the brush border membranes and basolateral membranes [39]. However, the degree of contribution of each receptor expressed in brush border membrane or basolateral membrane to the Na+ excretion effect remains unclear. The urinary Na+ level in the volume expansion period was 38.9 and 9.8 µ equivalent (Eq)/h for the wild-type and Mate1-knockout mice, respectively (Figure 2c), while the urinary dopamine was barely detectable in Mate1-knockout mice (Figure 2a). Therefore, it is likely that the 29.1 µEq/h (75%) difference in urinary Na+ between the wild-type and Mate1-knockout mice was due to differences in dopamine secretion into the tubular lumen via apical dopamine receptors. The Km value of dopamine uptake by mMATE1 (0.53 ± 0.08 mM; Table 1) was very similar to the renal level in intact wild-type mice (approximately 100 ng/mg or 0.65 mM) [40]. Additionally, the renal dopamine level in wild-type mice after volume expansion treatment was approximately 150–200 µg/g in kidney or 0.98–1.3 mM (Figure 2b and Figure 4b). Therefore, mMATE1 transport velocity is not saturated at in vivo dopamine concentrations. Thus, volume expansion treatment increased mouse renal dopamine concentration, which subsequently led to the nonlinear region of the mMATE1 dopamine transport velocity. In the human kidney PTCs, various organic cations (e.g., tetraethylammonium, metformin, oxaliplatin, varenicline, cimetidine and creatinine) are taken up from the circulation by the membrane potential-dependent OCT2, a basolateral transporter, and secreted into the tubular lumen by MATE1 and MATE2-K, which are both apical transporters [16,41,42]. Dopamine is known as a substrate of OCT2 [42], and in the present study, we showed that it is also a substrate of MATE. This suggests a model in which OCT2 mediates dopamine uptake from the circulation into PTC membrane in a membrane potential-dependent manner, whereas MATE1 and MATE2-K mediate tubular secretion of the renally-synthesized dopamine into the lumen in a direction opposite to the H+ gradient. Hepatic metabolism is the major route of imatinib elimination, and the cytochrome P-450 (CYP) 3A isoenzyme subfamily breaks down imatinib via oxidative reactions [43]. Imatinib is also a substrate of P-gp and the breast cancer resistance protein (BCRP) [44]. Imatinib inhibits creatinine and metformin transport by the organic cation transporter (OCT) and MATE [21,45]. In the current study, we found that imatinib also inhibited MATE-mediated dopamine transport. Monitoring of the pharmacokinetics and pharmacodynamics of tyrosine kinase inhibitors is critical during the clinical administration of these drugs. The plasma trough concentration (Ctrough) of imatinib must be >1000 ng/mL (2 µM) to achieve clinical efficacy in patients with chronic myelocytic leukemia (CML) or gastrointestinal stromal tumors [46]. Because the clinical plasma levels of imatinib are similar to the IC50 values of hMATE1 and hMATE2-K (Table 4), it is reasonable to assume that imatinib will inhibit MATE-dependent dopamine secretion into the tubular lumen in patients. In the case of dasatinib, the Ctrough should not exceed 2.5 ng/mL (0.005 µM) in order to meet safety requirements when used to treat the chronic phase in patients with CML [46]. The recommended Ctrough of nilotinib in patients with CML is >761 ng/mL (1.4 µM) [46]. In contrast to imatinib, the recommended clinical plasma concentrations of dasatinib and nilotinib are much lower than the IC50 values of these drugs concerning hMATE1- and hMATE2-K-mediated dopamine uptake (Table 4). Thus, we can infer that patients taking imatinib experience edema more frequently than those taking dasatinib and nilotinib because of the differential IC50 values that these drugs have for MATE inhibition. Edema is the most common side effect associated with imatinib treatment (it is observed in 55.5% of patients), and the risk factors are a high dosage, or plasma concentration, or both [30,47]. Dietary restriction of salt improves edema symptoms in some patients [30]. Patients with edema usually continue to take imatinib without dose reduction because, in most cases, the accompanying edema is superficial and presents with a mild to modest severity. However, edema can contribute to poor patient adherence, which is a key factor for achieving a stable major molecular response [48,49]. Periorbital edema is the most frequent type of imatinib-induced water retention, and some patients require surgical intervention to recover visual field defects [47]. The etiology of imatinib-induced edema is still unknown, which thwarts the development of novel dosing or adjuvant strategies to ameliorate these effects. In conclusion, our results show that MATE mediated dopamine secretion and, consequently, promoted natriuresis (Figure 5a), and thus, its dysfunction is a risk for fluid retention (Figure 5b). Therefore, urinary dopamine has potential usefulness as a noninvasive index of the MATE transporter activity in the kidney. 4. Materials and Methods 4.1. Cell Culture HEK293 cells (American Type Culture Collection, Manassas, VA, USA; ATCC CRL-1573) were cultured in complete medium consisting of Dulbecco’s Modified Eagle’s Medium (DMEM) (Wako Pure Chemical Industries, Osaka, Japan) with 10% fetal bovine serum (FBS, Life Technologies Corporation, Carlsbad, CA, USA) in an atmosphere of 5% CO2/95% air at 37 °C. For transient expression, HEK293 cells were seeded on 24-well (2 × 105 cells/well) poly-d-lysine-coated plates (BD Biocoat, Franklin Lakes, NJ, USA) and transfected with hMATE1, hMATE2-K and mMATE1 cDNA containing plasmid vectors and empty vectors (pcDNA3.1(+) for hMATE1 and hMATE2-K, pFLAG for mMATE1) by using the LipofectAMINE 2000 Reagent (Life Technologies Corporation), according to the manufacturer’s instructions. The cells were used for uptake experiments at 48 h after transfection, while the empty vector-expressing cells were used as control. 4.2. Dopamine Uptake Experiments The MATE-expressing HEK293 cells were preincubated with 0.2 mL of incubation medium (145 mM NaCl, 3 mM KCl, 1 mM CaCl2, 0.5 mM MgCl2, 5 mM d-glucose, 5 mM HEPES, 1 mM ascorbic acid and 10 µM U0521) containing 30 mM ammonium chloride (pH 7.4) for 20 min at 37 °C to induce intracellular acidification, because MATE moves solutes against a H+ gradient [11]. After the preincubation medium was removed, 0.2 mL of incubation medium containing 82 nM (1.13 µ curie (Ci)/mL) of [3H]dopamine (NET131250UC, PerkinElmer, Waltham, MA, USA) were added. In the cis-inhibition experiments, various concentrations of imatinib, dasatinib and nilotinib were added in addition to [3H]dopamine. Following the indicated incubation times at 37 °C, the medium containing [3H]dopamine was aspirated, and the monolayers were rapidly rinsed twice with 1 mL of ice-cold incubation medium without 10 µM U0521. The cells were dissolved in 0.5 mL of 0.5 normality (N) sodium hydroxide (NaOH), and then, the radioactivity in 200 µL was determined using liquid scintillation counting following addition to 2 mL of Ultima Gold (PerkinElmer). The protein content of each NaOH solution was determined using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA) by using bovine γ-globulin as a standard. The Km and Vmax values were calculated from the saturation curve using the Michaelis–Menten equation, V = Vmax[S]/(Km + [S]), with Kaleidagraph Version 4.00 (Synergy Software, Reading, PA, USA) after the nonspecific uptake values were subtracted. The nonspecific uptake values were calculated from the linear model obtained from 1 and 5 mM [3H]dopamine uptake values in the presence of 5 mM tetraethylammonium. The IC50 values were calculated from the inhibition plots using the equation, V = V0/[1 + ([I]/IC50)n], with Kaleidagraph Version 4.00, where V and V0 are the uptake amounts of [3H]dopamine in the presence and absence of the inhibitor, respectively, [I] is the concentration of the inhibitor and n is the Hill coefficient. 4.3. Animals Mice aged 17 weeks or younger were used in all experiments. The male wild-type mice and Mate1-knockout mice were on C57BL/6NCrSlc and C57BL/6 genetic backgrounds, respectively [14]. All of the mice were treated according to the Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology in Japan. MedKyo12133, 26 March 2012, Kyoto University Animal Care and User Committee. 070661, 16 May 2008, Kyoto University Safety Committee for Recombinant DNA Experiments. 4.4. Immunohistochemical Analysis The mice were anesthetized, and their kidneys were perfused via the left ventricle, first with saline containing 25 U/mL of heparin, followed by 4% paraformaldehyde in phosphate-buffered saline (PBS). The fixed tissues were embedded in OCT compound (Sakura Finetechnical, Tokyo, Japan), frozen rapidly in liquid nitrogen; 6 µm-thick sections were cut and then incubated with 10 mM of citrate buffer pH 6 (D1 and D5) or 1 mM of EDTA-2Na buffer pH 9 (NHE3) at 95 °C for 40 min for antigen retrieval. After washing with PBS, the sections were incubated with 3% skim milk in PBS at room temperature for 15 min. Following another PBS wash, the sections were incubated at 4 °C overnight with specific antiserum specific for D1 (ab20066; Abcam, Cambridge, MA, USA) (1:4500 dilution), D5 (GTX77969; GeneTex, Irvine, CA, USA) (1:1600 dilution) or anti-NHE3 antibody (NHE31-A; ALPHA DIAGNOSTIC, San Antonio, TX, USA) (1:1000 dilution). Following three PBS washes (5 min each), the sections were incubated with hydrogen peroxide (0.3% in methanol) at room temperature for 30 min. After three additional PBS washes (5 min each), the sections were incubated with EnVision Horseradish Peroxidase (K4003; DAKO, Hamburg, Germany) at room temperature for 30 min. After washing with distilled water, the nuclei were stained with hematoxylin, and the images were captured using an Aperio (Leica, Wetzlar, Germany). 4.5. Determination of Fluid Content The imatinib- (200 mg/kg oral administration, for 14 days) and vehicle-treated (lactated Ringer’s solution) wild-type, intact wild-type and intact [17] Mate1-knockout mice were anesthetized with isoflurane. Blood was collected from the inferior vena cava, and the chemical parameters were measured using a CHEM8+ cartridge and the i-STAT analyzer (Abbott, Abbott Park, IL, USA). Then, the mouse carcasses were heated in an oven at 60 °C over 14 days or until there was no further change in weight loss on 2 consecutive days. The difference between the live mouse weight and dried carcass weight was used to determine the fluid content of each mouse [50]. The total body water weight to total body weight ratio was calculated as follows: (living mouse weight-dried mouse weight)/living mouse weight. The mouse body weight was measured using an ANDGX-2000 (A&D Company, Tokyo, Japan), and the body weight was measured just before oral administration of imatinib or the vehicle. 4.6. Acute Saline Volume Expansion Experiments For the saline infusion, a catheter was inserted into the femoral vein with polyethylene tubing (Intramedic PE-10; BD Biosciences, San Jose, CA, USA) in mice anesthetized with 50 mg/kg of pentobarbital. To collect the urine, the urinary bladder was catheterized with SP-31 tubing (Natsume Seisakusho, Tokyo, Japan). The saline was infused using an automatic infusion pump (Harvard Apparatus, Inc., Holliston, MA, USA) at 0.24 mL/h for 30 min in the stabilization period, 0.12 mL/h for 60 min in the control period and 3 mL/h for 30 min in the volume expansion period. Following the latter period, the infusion pump was stopped, and the mice were then left for 30 min before the kidney samples were collected. The urinary electrolyte levels of the mice were measured using an ion-selective electrode (cobas6000, Roche, Basel, Switzerland). For dopamine detection, 1 N HCl was added to the urinary collection tube, and 0.2 N HClO4 was added to kidney collection tube to prevent dopamine degradation. In the imatinib inhibition study, 500 mg/kg imatinib dissolved in lactated Ringer’s solution were administered orally to wild-type mice 3 h or immediately before saline infusion. 4.7. Sample Preparation The catecholamines in the urine reagent kit (CHROMSYSTEMS, Munich, Germany) were used for dopamine sample preparation according to the manufacturer’s instructions with some modifications. For urinary dopamine detection, the urine was first diluted to a total volume of 300 µL with 1% 6 N HCl, and then, 50 µL of the internal standard (DHBA) were then added, followed by 6 mL of neutralization buffer. We applied 600 µL elution buffer to the column in a final step and added formic acid to the collected eluate. Forty microliters of sample was injected into the LC-MS/MS system after being filtrated with a filter (06543-04, Nacalai Tesque, Kyoto, Japan). The kidney homogenate was diluted 5- or 6-fold with 1% 6 N HCl. After centrifugation, 300 µL of the supernatant were subjected to the same procedure used for the urine sample above. For imatinib detection, the mouse plasma and kidney homogenate samples were diluted 40× and 400× with saline, respectively, and 10 µL of roscovitine solution (1 µg/mL) were added to 100 µL of the diluted samples. Then, the samples were deproteinized with 200 µL of acetonitrile for 10 min with agitation. After centrifugation (14,680 rpm, 15 min) using the Centrifuge 5424 (Eppendorf, Hamburg, Germany), the supernatant was diluted with 200 µL of 0.2% formic acid, and 1 µL of the sample was injected into the LC-MS/MS system after filtration using a filter (06541-24, Nacalai Tesque, Kyoto, Japan). 4.8. Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) The dopamine and imatinib were analyzed by using liquid chromatography-tandem mass spectrometry (LC-MS/MS). A liquid chromatography system consisting of a Prominence series chromatograph (Shimadzu, Kyoto, Japan) coupled to an API4000 triple-quadrupole tandem mass spectrometer (AB SCIEX, Foster City, CA, USA) was used for dopamine detection, and an Eksigent ultra-performance liquid chromatography (AB SCIEX) coupled to a QTRAP4500 triple-quadrupole tandem mass spectrometer (AB SCIEX) was used for imatinib detection. An Inertsil ODS-3 (GL Sciences, Tokyo, Japan) was used for chromatographic separation. The mobile phase was composed of 0.1% formic acid and acetonitrile containing 0.1% formic acid. For dopamine detection, gradient elution was carried out at a flow rate of 0.25 mL/min. 3,4-dihydroxybenzylamine (DHBA) was used as an internal standard. The detection was carried out in the multiple reaction monitoring mode by monitoring ion transitions of m/z 154.1 → 91.2 for dopamine and m/z 140.1 → 77.1 for DHBA. For imatinib detection, gradient elution was carried out at a flow rate of 0.2 mL/min. Roscovitine was used as an internal standard. Detection was carried out in the multiple reaction-monitoring mode by monitoring ion transitions of m/z 494.171 → 393.900 for imatinib and m/z 355.047 → 233.100 for roscovitine. 4.9. Statistical Analysis All data are expressed as the mean ± standard error (S.E.) and were analyzed statistically using the unpaired t-test. Acknowledgments We would like to thank Masahiro Shimomura Ph.D. for developing the dopamine measuring LC-MS/MS system. Funding sources: Grants-in-aid for scientific research (25860114 and 15K18922 to Moto Kajiwara, 15H04666 to Satohiro Masuda) and Funding program for next generation world-leading researchers (LS073 to Satohiro Masuda); Japan Research Foundation for Clinical Pharmacology. Author Contributions Moto Kajiwara and Satohiro Masuda designed the study. Moto Kajiwara and Tsuyoshi Ban carried out the research. Moto Kajiwara analyzed the data. Moto Kajiwara and Satohiro Masuda wrote the paper. Yoichi Nakanishi and Kazuo Matsubara approved the paper. Conflicts of Interest The authors declare no conflict of interests. Figure 1 Characteristics of dopamine transport mediated by human multidrug and toxin extrusion (hMATE)1-, hMATE2-K- and mouse (m) MATE1-expressing cells (a) Time course of [3H]dopamine uptake by hMATE1- (Ο) and hMATE2-K- (●) expressing cells. pcDNA3.1(+) (Δ) represents cells transfected with empty vector. Each set of points represents uptake values at 0.5, 1, 2 and 5 min. The level of [3H]dopamine uptake at all time points in the presence of MATE transporters was significantly higher than that of the controls was, p < 0.01 (n = 3); (b) Time course of [3H]dopamine uptake by mMATE1- (Ο) expressing cells. pFLAG (●) is the empty vector. Each set of points represents uptake values at 0.5, 1, 2 and 5 min. [3H]dopamine uptake in the cells expressing MATE transporters was significantly higher than that of controls at all time points, p < 0.01 (n = 3); (c) Concentration-dependent uptake of [3H]dopamine by hMATE1-expressing cells (n = 3); (d) Concentration-dependent uptake of [3H]dopamine by hMATE2-K-expressing cells (n = 3); (e) Concentration-dependent uptake of [3H]dopamine by mMATE1-expressing cells (n = 3). For analyses in (c–e), dopamine concentrations were 8.2 × 10−5, 0.1, 0.5, 1, 2.5 and 5 mM. Figure 2 Effect of multidrug and toxin extrusion (Mate1) knockout on natriuresis resulting from renally-synthesized dopamine and fluid retention in mice. (a) Urinary dopamine level in wild-type (WT) and Mate1-knockout (KO) mice during acute saline infusion. ** p < 0.01 compared to WT mice (n = 3 for each group); (b) Renal dopamine level in WT and Mate1 KO mice after acute saline infusion. * p < 0.05, compared to WT mice (n = 3 for each group); (c) Urinary Na+ excretion level of WT (n = 3) and Mate1 KO mice (n = 4) during acute saline infusion. * p < 0.05 compared to WT mice; † p < 0.05 and †† p < 0.01 compared to control period; (d) Urinary K+ excretion level of WT (n = 3) and Mate1 KO mice (n = 4) during acute saline infusion. † p < 0.05 compared to the control period; (e) Urinary Cl− excretion level of WT (n = 3) and Mate1 KO mice (n = 4) during acute saline infusion. * p < 0.05 compared to WT mice; † p < 0.05 and †† p < 0.01 compared to the control period; (f) Urinary volume of WT (n = 3) and Mate1 KO mice (n = 4) during acute saline infusion. ** p < 0.01 compared to WT mice; † p < 0.05 and ††p < 0.01 compared to the control period; (g) Ratio of total body water weight to total body weight of intact WT (n = 7) (Ο) and Mate1 KO mice (●) (n = 8). Bars indicate the mean values. ** p < 0.01 compared to WT mice. Figure 3 Expression of D1-like receptors and Na+/H+ exchanger (NHE)3 in wild-type (WT) and Mate1-knockout (KO) mouse kidneys. The scale bar represents 100 µm. Immunohistochemistry of renal D1 receptor (a,b), D5 receptor (c,d), NHE3 (e,f) in WT (a,c,e) and Mate1-KO (b,d,f) mice, respectively. Figure 4 Effect of imatinib on natriuresis resulting from renally-synthesized dopamine and fluid retention in mice. (a) Urinary dopamine level of vehicle- and imatinib-treated WT mice during acute saline infusion. * p < 0.05 and † p < 0.05 compared to vehicle-treated WT mice and control period, respectively, n = 3 for each group; (b) Renal dopamine level of vehicle- and imatinib-treated WT mice after acute saline infusion, n = 3 for each group; (c) Urinary Na+ excretion level of vehicle- and imatinib-treated WT mice during acute saline infusion (n = 3 and 4 for vehicle and imatinib groups, respectively). * p < 0.05 and ** p < 0.01 compared to vehicle-treated WT mice, † p < 0.05 compared to control period; (d) Urinary K+ excretion levels of vehicle- and imatinib-treated WT mice during acute saline infusion (n = 3 and 4 for vehicle and imatinib groups, respectively). * p < 0.05 and † p < 0.05 compared to vehicle-treated WT mice and control period, respectively; (e) Urinary Cl− excretion level of vehicle- and imatinib-treated WT mice during acute saline infusion (n = 3 and 4 for vehicle and imatinib groups, respectively). ** p < 0.01, compared to vehicle-treated WT mice; † p < 0.05, compared to control period; (f) Urinary volume of vehicle- and imatinib-treated WT mice during acute saline infusion (n = 3 and 4 for vehicle and imatinib groups, respectively). * p < 0.05 and † p < 0.05 compared to vehicle-treated WT mice and control period, respectively; (g) Ratio of total body water weight to total body weight of vehicle- and imatinib-treated wild-type (WT) mice (Ο and ●, respectively, n = 3 per group). Bars indicate the mean values. * p < 0.05, compared to vehicle-treated WT mice; (h) Body weight change of vehicle- and imatinib-treated WT mice (Ο and ●, respectively, n = 3 per group). Figure 5 Mechanism of the secretion of dopamine synthesized in the kidney into proximal tubular cells. (a) MATE mediates dopamine secretion into the apical lumen and, consequently, promotes urinary Na+ excretion by acting on dopamine receptors that are expressed in multiple nephron sites; (b) MATE dysfunction inhibits dopamine secretion into the apical lumen and, consequently, triggers fluid retention by inhibiting urinary Na+ excretion. Related to B°,+ amino acid transporter/ B°,+-type amino acid transporter dimer, rBAT/B°,+AT; 4F2 heavy chain/l-type amino acid transporter 2 dimer, 4F2h/LAT2; multidrug and toxin extrusion, MATE; aromatic amino acid decarboxylase, AADC; l-dihydroxyphenylalanine, l-DOPA; D1 receptor and D5 receptor, D1-like receptor. ijms-17-01228-t001_Table 1Table 1 Kinetic parameters of [3H]dopamine uptake in HEK293 cells transiently expressing human multidrug and toxin extrusion (hMATE) 1, hMATE2-K and mouse (m) MATE1. Kinetic Parameters hMATE1 hMATE2-K mMATE1 Km (mM) 0.56 ± 0.18 * 2.48 ± 0.65 ‡ 0.53 ± 0.08 Vmax (nmol·mg·protein−1·min−1) 3.71 ± 0.15 * 7.69 ± 1.12 8.73 ± 0.08 †† Vmax/Km (µL·mg·protein−1·min−1) 7.70 ± 1.67 3.44 ± 0.78 ‡‡ 17.20 ± 2.72 † Data represent the mean ± standard error (S.E.) of three separate experiments. * p < 0.05, hMATE1 vs. hMATE2-K unpaired t-test. ‡ p < 0.05, ‡‡ p < 0.01, hMATE2-K vs. mMATE1 unpaired t-test. † p < 0.05, †† p < 0.01, hMATE1 vs. mMATE1 unpaired t-test. ijms-17-01228-t002_Table 2Table 2 Blood parameters and body weight of wild-type and Mate1-knockout mice. Blood Parameters and Body Weight Wild-Type Mice Mate1-Knockout Mice Na+ (mmol/L) 147.0 ± 0.2 149.1 ± 0.6 ** K+ (mmol/L) 4.0 ± 0.1 3.8 ± 0.0 * Cl− (mmol/L) 113.6 ± 0.4 114.3 ± 0.7 iCa (mmol/L) 1.2 ± 0.0 1.2 ± 0.0 tCO2 (mmol/L) 19.3 ± 0.5 19.6 ± 0.4 Glucose (mg/dL) 249.9 ± 10.0 182.4 ± 17.3 ** BUN (mg/dL) 25.1 ± 1.1 26.0 ± 2.6 Hct (%) 38.9 ± 0.5 37.6 ± 2.1 Hb (g/dL)(via Hct) 13.2 ± 0.2 12.8 ± 0.7 AnGap (mmol/L) 19.1 ± 0.6 19.9 ± 0.8 Body weight (g) 28.7 ± 0.4 29.1 ± 0.4 iCa, ionized calcium; tCO2, total carbon dioxide; BUN, blood urea nitrogen; Hct, hematocrit; Hb, hemoglobin; AnGap, anion gap. Values are the mean ± standard error (S.E.) for seven and eight wild-type and Mate1-knockout mice, respectively. * p < 0.05 and ** p < 0.01, significantly different from wild-type mice (unpaired t-test). ijms-17-01228-t003_Table 3Table 3 Blood parameters and body weight of vehicle- and imatinib-treated mice. Blood Parameters and Body Weight Vehicle-Treated Mice Imatinib-Treated Mice Na+ (mmol/L) 147 ± 1.5 145.7 ± 1.5 K+ (mmol/L) 4.9 ± 0.3 4.5 ± 0.5 Cl− (mmol/L) 115.0 ± 0.6 115.3 ± 1.2 iCa (mmol/L) 1.2 ± 0.0 1.3 ± 0.0 tCO2 (mmol/L) 25.0 ± 0.6 19.7 ± 1.5 * Glucose (mg/dL) 284.0 ± 62.6 229.0 ± 39.3 BUN (mg/dL) 26.3 ± 0.3 21.0 ± 2.1 Hct (%) 39.7 ± 0.3 38.3 ± 2.0 Hb (g/dL)(via Hct) 13.5 ± 0.1 13.0 ± 0.7 AnGap (mmol/L) 13.0 ± 0.6 16.0 ± 1.5 Body weight (g) 26.6 ± 0.3 26.7 ± 0.8 iCa, ionized calcium; tCO2, total carbon dioxide; BUN, blood urea nitrogen; Hct, hematocrit; Hb, hemoglobin; AnGap, anion gap. Each value represents the means ± S.E. for three (vehicle- and imatinib-treated) mice. * p < 0.05, significantly different from vehicle-treated mice (unpaired t-test). ijms-17-01228-t004_Table 4Table 4 Half-maximal inhibitory concentration (IC50) values of tyrosine kinase inhibitors on [3H]dopamine uptake mediated by hMATE1, hMATE2-K and mMATE1. Transporter Imatinib (µM) Dasatinib (µM) Nilotinib (µM) hMATE1 1.1 ± 0.1 7.1 ± 0.6 >100 hMATE2-K 13.8 ± 4.4 4.1 ± 1.0 >100 mMATE1 100.6 ± 13.9 106.2 ± 7.0 >100 Imatinib and dasatinib data represent the mean ± S.E. of three separate experiments. Imatinib concentrations were 0, 0.3, 0.6, 1, 3 and 10 µM for hMATE1; 0, 2, 3, 6, 10 and 250 µM for hMATE2-K; and 0, 30, 60, 100, 150 and 250 µM for mMATE1. Dasatinib concentrations were 0, 0.3, 1, 3, 10 and 30 µM for hMATE1 and hMATE2-K and 0, 30, 60, 100, 150 and 200 µM for mMATE1. ==== Refs References 1. Chen C.J. Lokhandwala M.F. Role of endogenous dopamine in the natriuretic response to various degrees of iso-osmotic volume expansion in rats Clin. Exp. Hypertens. A 1991 13 1117 1126 10.3109/10641969109042117 1760885 2. Hayashi M. Yamaji Y. Kitajima W. Saruta T. Effects of high salt intake on dopamine production in rat kidney Am. J. Physiol. 1991 260 E675 E679 2035623 3. Alexander R.W. Gill J.R. Jr. Yamabe H. Lovenberg W. Keiser H.R. Effects of dietary sodium and of acute saline infusion on the interrelationship between dopamine excretion and adrenergic activity in man J. Clin. Investig. 1974 54 194 200 10.1172/JCI107743 4600048 4. Wang X. Villar V.A. Armando I. Eisner G.M. Felder R.A. Jose P.A. Dopamine, kidney, and hypertension: Studies in dopamine receptor knockout mice Pediatr. Nephrol. 2008 23 2131 2146 10.1007/s00467-008-0901-3 18615257 5. Chen C.J. Lokhandwala M.F. An impairment of renal tubular DA-1 receptor function as the causative factor for diminished natriuresis to volume expansion in spontaneously hypertensive rats Clin. Exp. Hypertens. A 1992 14 615 628 10.3109/10641969209036211 1352742 6. Ball S.G. Oats N.S. Lee M.R. Urinary dopamine in man and rat: Effects of inorganic salts on dopamine excretion Clin. Sci. Mol. Med. 1978 55 167 173 10.1042/cs0550167 28196 7. Quinones H. Collazo R. Moe O.W. The dopamine precursor l -dihydroxyphenylalanine is transported by the amino acid transporters rBAT and LAT2 in renal cortex Am. J. Physiol. Ren. Physiol. 2004 287 F74 F80 10.1152/ajprenal.00237.2003 15180924 8. Pinto V. Pinho M.J. Soares-da-Silva P. Renal amino acid transport systems and essential hypertension FASEB J. 2013 27 2927 2938 10.1096/fj.12-224998 23616567 9. Hayashi M. Yamaji Y. Kitajima W. Saruta T. Aromatic l -amino acid decarboxylase activity along the rat nephron Am. J. Physiol. 1990 258 F28 F33 2301593 10. Jose P.A. Soares-da-Silva P. Eisner G.M. Felder R.A. Dopamine and G protein-coupled receptor kinase 4 in the kidney: Role in blood pressure regulation Biochim. Biophys. Acta 2010 1802 1259 1267 10.1016/j.bbadis.2010.02.004 20153824 11. Tsuda M. Terada T. Asaka J. Ueba M. Katsura T. Inui K. Oppositely directed H+ gradient functions as a driving force of rat H+ /organic cation antiporter MATE1 Am. J. Physiol. Ren. Physiol. 2007 292 F593 F598 10.1152/ajprenal.00312.2006 17047166 12. Otsuka M. Matsumoto T. Morimoto R. Arioka S. Omote H. Moriyama Y. A human transporter protein that mediates the final excretion step for toxic organic cations Proc. Natl. Acad. Sci. USA 2005 102 17923 17928 10.1073/pnas.0506483102 16330770 13. Masuda S. Terada T. Yonezawa A. Tanihara Y. Kishimoto K. Katsura T. Ogawa O. Inui K. Identification and functional characterization of a new human kidney-specific H+ /organic cation antiporter, kidney-specific multidrug and toxin extrusion 2 J. Am. Soc. Nephrol. 2006 17 2127 2135 10.1681/ASN.2006030205 16807400 14. Tsuda M. Terada T. Mizuno T. Katsura T. Shimakura J. Inui K. Targeted disruption of the multidrug and toxin extrusion 1 (Mate1 ) gene in mice reduces renal secretion of metformin Mol. Pharmacol. 2009 75 1280 1286 10.1124/mol.109.056242 19332510 15. Yonezawa A. Inui K. Importance of the multidrug and toxin extrusion MATE/SLC47A family to pharmacokinetics, pharmacodynamics/toxicodynamics and pharmacogenomics Br. J. Pharmacol. 2011 164 1817 1825 10.1111/j.1476-5381.2011.01394.x 21457222 16. Kajiwara M. Masuda S. Watanabe S. Terada T. Katsura T. Inui K. Renal tubular secretion of varenicline by multidrug and toxin extrusion (MATE) transporters Drug Metab. Pharmacokinet. 2012 27 563 569 10.2133/dmpk.DMPK-11-RG-156 22510868 17. EMA Guideline on the Investigation of Drug Interactions Availabler online: http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/07/WC500129606.pdf (accessed on 20 May 2016) 18. Hillgren K.M. Keppler D. Zur A.A. Giacomini K.M. Stieger B. Cass C.E. Zhang L. International Transporter Consortium Emerging transporters of clinical importance: An update from the International Transporter Consortium Clin. Pharmacol. Ther. 2013 94 52 63 10.1038/clpt.2013.74 23588305 19. FDA Guidance for Industry: Drug Interaction Studies—Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations (Draft Guidance) Availabler online: http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm292362.pdf (accessed on 20 May 2016) 20. Lipska K.J. Bailey C.J. Inzucchi S.E. Use of metformin in the setting of mild-to-moderate renal insufficiency Diabetes Care 2011 34 1431 1437 10.2337/dc10-2361 21617112 21. Minematsu T. Giacomini K.M. Interactions of tyrosine kinase inhibitors with organic cation transporters and multidrug and toxic compound extrusion proteins Mol. Cancer Ther. 2011 10 531 539 10.1158/1535-7163.MCT-10-0731 21252289 22. Asaka J. Terada T. Tsuda M. Katsura T. Inui K. Identification of essential histidine and cysteine residues of the H+ /organic cation antiporter multidrug and toxin extrusion (MATE) Mol. Pharmacol. 2007 71 1487 1493 10.1124/mol.106.032938 17327464 23. Kato K. Mori H. Kito T. Yokochi M. Ito S. Inoue K. Yonezawa A. Katsura T. Kumagai Y. Yuasa H. Investigation of endogenous compounds for assessing the drug interactions in the urinary excretion involving multidrug and toxin extrusion proteins Pharm. Res. 2014 31 136 147 10.1007/s11095-013-1144-y 23907530 24. Tanihara Y. Masuda S. Sato T. Katsura T. Ogawa O. Inui K. Substrate specificity of MATE1 and MATE2-K, human multidrug and toxin extrusions/H+ -organic cation antiporters Biochem. Pharmacol. 2007 74 359 371 10.1016/j.bcp.2007.04.010 17509534 25. Zhang X. Cherrington N.J. Wright S.H. Molecular identification and functional characterization of rabbit MATE1 and MATE2-K Am. J. Physiol. Ren. Physiol. 2007 293 F360 F370 10.1152/ajprenal.00102.2007 17442726 26. Chan B.S. Satriano J.A. Pucci M. Schuster V.L. Mechanism of prostaglandin E2 transport across the plasma membrane of HeLa cells and Xenopus oocytes expressing the prostaglandin transporter “PGT” J. Biol. Chem. 1998 273 6689 6697 10.1074/jbc.273.12.6689 9506966 27. Lu R. Chan B.S. Schuster V.L. Cloning of the human kidney PAH transporter: Narrow substrate specificity and regulation by protein kinase C Am. J. Physiol. 1999 276 F295 F303 9950961 28. Hegde S.S. Jadhav A.L. Lokhandwala M.F. Role of kidney dopamine in the natriuretic response to volume expansion in rats Hypertension 1989 13 828 834 10.1161/01.HYP.13.6.828 2661432 29. Alexander R.T. Dimke H. Cordat E. Proximal tubular NHEs: Sodium, protons and calcium? Am. J. Physiol. Ren. Physiol. 2013 305 F229 F236 10.1152/ajprenal.00065.2013 23761670 30. Guilhot F. Indications for imatinib mesylate therapy and clinical management Oncologist 2004 9 271 281 10.1634/theoncologist.9-3-271 15169982 31. Soares-Da-Silva P. Serrao M.P. Vieira-Coelho M.A. Apical and basolateral uptake and intracellular fate of dopamine precursor l -DOPA in LLC-PK1 cells Am. J. Physiol. 1998 274 F243 F251 9486218 32. Nielsen R. Birn H. Moestrup S.K. Nielsen M. Verroust P. Christensen E.I. Characterization of a kidney proximal tubule cell line, LLC-PK1 , expressing endocytotic active megalin J. Am. Soc. Nephrol. 1998 9 1767 1776 9773777 33. Pritchard J.B. Miller D.S. Renal secretion of organic anions and cations Kidney Int. 1996 49 1649 1654 10.1038/ki.1996.240 8743470 34. Inui K. Masuda S. Saito H. Cellular and molecular aspects of drug transport in the kidney Kidney Int. 2000 58 944 958 10.1046/j.1523-1755.2000.00251.x 10972658 35. Maurel A. Spreux-Varoquaux O. Amenta F. Tayebati S.K. Tomassoni D. Seguelas M.H. Parini A. Pizzinat N. Vesicular monoamine transporter 1 mediates dopamine secretion in rat proximal tubular cells Am. J. Physiol. Ren. Physiol. 2007 292 F1592 F1598 10.1152/ajprenal.00514.2006 17244889 36. Inui K. Saito H. Hori R. H+ -gradient-dependent active transport of tetraethylammonium cation in apical-membrane vesicles isolated from kidney epithelial cell line LLC-PK1 Biochem. J. 1985 227 199 203 10.1042/bj2270199 2986603 37. Kajiwara M. Terada T. Ogasawara K. Iwano J. Katsura T. Fukatsu A. Doi T. Inui K. Identification of multidrug and toxin extrusion (MATE1 and MATE2-K) variants with complete loss of transport activity J. Hum. Genet. 2009 54 40 46 10.1038/jhg.2008.1 19158817 38. Toyama K. Yonezawa A. Tsuda M. Masuda S. Yano I. Terada T. Osawa R. Katsura T. Hosokawa M. Fujimoto S. Heterozygous variants of multidrug and toxin extrusions (MATE1 and MATE2-K) have little influence on the disposition of metformin in diabetic patients Pharmacogenet. Genom. 2010 20 135 138 10.1097/FPC.0b013e328335639f 20016398 39. Felder C.C. McKelvey A.M. Gitler M.S. Eisner G.M. Jose P.A. Dopamine receptor subtypes in renal brush border and basolateral membranes Kidney Int. 1989 36 183 193 10.1038/ki.1989.178 2528652 40. Zhang M.Z. Yao B. Wang S. Fan X. Wu G. Yang H. Yin H. Yang S. Harris R.C. Intrarenal dopamine deficiency leads to hypertension and decreased longevity in mice J. Clin. Investig. 2011 121 2845 2854 10.1172/JCI57324 21701066 41. Feng B. Obach R.S. Burstein A.H. Clark D.J. de Morais S.M. Faessel H.M. Effect of human renal cationic transporter inhibition on the pharmacokinetics of varenicline, a new therapy for smoking cessation: An in vitro-in vivo study Clin. Pharmacol. Ther. 2008 83 567 576 10.1038/sj.clpt.6100405 17971819 42. Koepsell H. Lips K. Volk C. Polyspecific organic cation transporters: Structure, function, physiological roles, and biopharmaceutical implications Pharm. Res. 2007 24 1227 1251 10.1007/s11095-007-9254-z 17473959 43. Gschwind H.P. Pfaar U. Waldmeier F. Zollinger M. Sayer C. Zbinden P. Hayes M. Pokorny R. Seiberling M. Ben-Am M. Metabolism and disposition of imatinib mesylate in healthy volunteers Drug Metab. Dispos. 2005 33 1503 1512 10.1124/dmd.105.004283 16006570 44. Oostendorp R.L. Buckle T. Beijnen J.H. van Tellingen O. Schellens J.H. The effect of P-gp (Mdr1a/1b), BCRP (Bcrp1) and P-gp/BCRP inhibitors on the in vivo absorption, distribution, metabolism and excretion of imatinib Investig. New Drugs 2009 27 31 40 10.1007/s10637-008-9138-z 18449471 45. Tanihara Y. Masuda S. Katsura T. Inui K. Protective effect of concomitant administration of imatinib on cisplatin-induced nephrotoxicity focusing on renal organic cation transporter OCT2 Biochem. Pharmacol. 2009 78 1263 1271 10.1016/j.bcp.2009.06.014 19540211 46. Yu H. Steeghs N. Nijenhuis C.M. Schellens J.H. Beijnen J.H. Huitema A.D. Practical guidelines for therapeutic drug monitoring of anticancer tyrosine kinase inhibitors: Focus on the pharmacokinetic targets Clin. Pharmacokinet. 2014 53 305 325 10.1007/s40262-014-0137-2 24566736 47. McClelland C.M. Harocopos G.J. Custer P.L. Periorbital edema secondary to imatinib mesylate Clin. Ophthalmol. 2010 4 427 431 10.2147/OPTH.S8521 20505834 48. Marin D. Bazeos A. Mahon F.X. Eliasson L. Milojkovic D. Bua M. Apperley J.F. Szydlo R. Desai R. Kozlowski K. Adherence is the critical factor for achieving molecular responses in patients with chronic myeloid leukemia who achieve complete cytogenetic responses on imatinib J. Clin. Oncol. 2010 28 2381 2388 10.1200/JCO.2009.26.3087 20385986 49. Salaroli A. Serrao A. Alimena G. Breccia M. Edema in patients treated with imatinib J. Symptoms Signs 2013 2 78 83 50. Guan Y. Hao C. Cha D.R. Rao R. Lu W. Kohan D.E. Magnuson M.A. Redha R. Zhang Y. Breyer M.D. Thiazolidinediones expand body fluid volume through PPARγ stimulation of ENaC-mediated renal salt absorption Nat. Med. 2005 11 861 866 10.1038/nm1278 16007095

PMC005xxxxxx/PMC5000627.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081229ijms-17-01229ReviewThe Interactions of Aquaporins and Mineral Nutrients in Higher Plants Wang Min 1Ding Lei 2Gao Limin 1Li Yingrui 1Shen Qirong 1Guo Shiwei 1*Rouached Hatem Academic Editor1 Jiangsu Key Lab for Organic Waste Utilization, Nanjing Agricultural University, Nanjing 210095, China; [email protected] (M.W.); [email protected] (L.G.); [email protected] (Y.L.); [email protected] (Q.S.)2 Institut des Sciences de la Vie, Université Catholique de Louvain, Louvain-la-Neuve B-1348, Belgium; [email protected]* Correspondence: [email protected]; Tel.: +86-25-8439-639329 7 2016 8 2016 17 8 122916 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Aquaporins, major intrinsic proteins (MIPs) present in the plasma and intracellular membranes, facilitate the transport of small neutral molecules across cell membranes in higher plants. Recently, progress has been made in understanding the mechanisms of aquaporin subcellular localization, transport selectivity, and gating properties. Although the role of aquaporins in maintaining the plant water status has been addressed, the interactions between plant aquaporins and mineral nutrients remain largely unknown. This review highlights the roles of various aquaporin orthologues in mineral nutrient uptake and transport, as well as the regulatory effects of mineral nutrients on aquaporin expression and activity, and an integrated link between aquaporins and mineral nutrient metabolism was identified. aquaporinwater transportmembrane proteinmineral nutrient ==== Body 1. Introduction Aquaporins, small integral proteins that belong to the ancient family of major intrinsic proteins (MIPs), have been found in all kingdoms of life. In plants, aquaporins reside in the plasma membrane and tonoplast and play important roles in plant water relations by facilitating the transport of water across biological membranes and regulating osmotic potential and hydraulic conductivity [1,2]. The regulatory roles of aquaporins in cellular water transport have been reported in previous studies [3,4,5,6]. In general, the molecular mechanisms of water transport across plasma membranes regulated by aquaporins are mainly attributed to co-translational and post-translational modification, aquaporin gating, and tetramer assembly and cellular trafficking of plasma membrane intrinsic proteins [1,4]. Based on amino acid sequence similarities, aquaporins are classified into seven subfamilies. The plasma membrane intrinsic proteins (PIPs) and the tonoplast intrinsic proteins (TIPs) are the most abundant aquaporins in the plasma membrane and tonoplast, respectively [3,4]. The nodulin 26-like intrinsic proteins (NIPs), are located in the peribacteroid membrane of nitrogen-fixing symbiotic root nodules of leguminous plants and are also present in the plasma membrane of other species [7]. The small basic intrinsic proteins (SIPs) are small proteins mainly localized in the ER membrane [8], and the uncharacterized X intrinsic proteins (XIPs) are plasma membrane aquaporins that function in the transport of uncharged substrates [9]. The hybrid intrinsic proteins (HIPs) and the glycerol facilitator (GlpF)-like intrinsic proteins (GIPs) are present exclusively in moss [10]. The large number of plant aquaporins has been explained by their importance in regulating plant metabolic processes under various physiological states and environmental conditions [11]. For example, aquaporins are essential for plant defence responses against biotic and abiotic stresses, such as drought [12], salt stress [13,14], cold [15,16], nutrient deprivation [17], heavy metals [18,19], and pathogen infection [20,21]. Aquaporins play complex integrated roles in the response to different environmental stressors and are involved in plant growth and metabolic processes. PIPs and TIPs are involved in drought, salt, and cold stress through hydraulic conductivity and transpiration regulation, while TIPs and NIPs are involved in biotic stress, predominantly nutrient homeostasis between pathogens and host plants [22]. Mineral nutrients, which are usually present in the soil solution in organic and inorganic forms, are essential for plant growth and production. Physiological analysis indicated that ion uptake was regulated by transporters in the root plasma membranes, and there is a strong interaction between mineral nutrients and water status in which mineral nutrient uptake is accompanied by water absorption. In addition to facilitating water diffusion, a number of aquaporins have also been shown to transport other small neutral molecules, such as urea [23,24], ammonia (NH3) [25,26], carbon dioxide (CO2) [27,28,29], boric acid [17,30,31], silicic acid [32,33,34], lactic acid [35], hydrogen peroxide (H2O2) [9,36,37,38], and other molecules with physiological significance [39]. Aquaporin trafficking and their subcellular relocalization act as a critical point for regulating the internal redistribution of mineral nutrients by transporting them from the endoplasmic reticulum (ER) to the plasma membrane via the Golgi apparatus, as well as undergoing repeated cycles of endocytosis and recycling through the early endosome to the multivesicular body/prevacuolar compartments before eventually being targeted to the vacuole [4]. However, the molecular and cellular mechanisms underlying the interactions of aquaporin and mineral nutrients should be further investigated. In this review, the role of aquaporins in maintaining the plant water and mineral nutrient status is discussed, and the cellular aspects of plant aquaporin functions and regulation of mineral nutrients are also extensively reviewed. 2. Nitrogen (N) Nitrogen (N), one of the most important mineral nutrients in higher plants, is involved in plant metabolism as a constituent of amino acids, proteins, nucleic acids, lipids, chlorophyll, co-enzymes, phytohormones, and secondary metabolites [40,41]. The interaction between aquaporins and N assimilation was first identified following the observation that the expression of several aquaporin genes responded to different N sources, such as AtTIP2;1, which was up-regulated by N starvation or NH4+ supply [26], while both PtdPIP1;2 and PtdSIP1;2 were down-regulated under high N fertilization levels [42]. Aquaporins have been suggested to be involved in water transport in response to nitrogen availability [42,43,44]. In our previous study, a high N (mixture of NH4+ and NO3−) supply enhanced aquaporin (AQP) expression and decreased root aerenchyma and lignin, resulting in a high water absorption rate [44], which was consistent with results that show that high N supply increases root hydraulic conductivity and AQP expression in rice plants [43]. The possible mechanisms of aquaporin regulated hydraulic conductivity in response to N availability may be attributed to the changes in aquaporin abundance and activity [43]. Aquaporins play an important role in N absorption, mobilization, and detoxification, as well as other nitrogen metabolic processes in higher plants [23]. The PIP, NIP, and TIP subfamilies have been shown to transport N compounds, including ammonia and urea [23,24]. 2.1. Nitrate Nitrate is the major inorganic N source absorbed by upland plants, and the process of nitrate uptake and metabolism is tightly associated with water utilization, which is regulated by AQP. Nitrate was suggested to be a critical signalling factor for radial water fluxes in the roots [45,46,47], and the increased root hydraulic conductivity (Lpr) by nitrate was shown to correlate with up-regulation of aquaporin expression [46,47,48]. In maize, the expression of ZmPIP1;5b was strongly up-regulated by nitrate [49], and in tomato plants, several AQP genes were up-regulated by the nitrate supply [50], which can mediate and control the increased water influx into the root cells. Recently, Li et al. [51] showed that Lpr and PIP expression were controlled by both exogenous and internal nitrate concentrations in Arabidopsis, and Lpr and PIP expressions were higher under 5 mM NO3− than 0.5 mM NO3−. In nrt2.1 (the high-affinity NO3− transporter) mutant plants, NO3− content decreased in both the roots and shoots, which resulted in decreases in Lpr and PIP expression, indicating that the nitrate supply was positively correlated with enhanced root AQP activity and Lpr. However, the interactions between nitrate and aquaporins vary over time. In the short term, hours to days, nitrate induced aquaporin expression [50], while over multiple days, root morphology and proliferation were significantly altered by the nitrate supply, resulting in increases in root nitrate acquisition [52,53]. 2.2. Ammonia/Ammonium Root NH4+ uptake occurs mainly via ammonium transporters (AMT) in the plasma membrane, while NH3 has been proposed to enter the cells by free diffusion in higher plants. Transport of NH3/NH4+ and urea into the vacuole would allow for N storage and eliminate their toxicity to the plant [54], and when N was needed, the stored nitrogen could be remobilized by a passive, low-affinity transport pathway, which may involve the TIP proteins [23]. Indeed, several tonoplast intrinsic proteins (TIPs) have been shown to facilitate the NH3 transport, such as ZmTIP1;1 and ZmTIP1;2 [55]. TIPs from wheat (TaTIP2;1) and Arabidopsis thaliana (AtTIP2;1 and AtTIP2;3) not only function as water-conducting membrane pores but also facilitate the transport of NH3 across membranes and therefore mediate the remarkable loading and acid-trapping of the protonated form (NH4+) in the vacuole [26,56]. However, the importance of the channel pores in ammonia transport by TIP2;2 from wheat has been challenged by the finding that NH3 is not transported with water but through a separate pathway [25]. The crystal structure of an NH3 permeable aquaporin AtTIP2;1 demonstrated that an intriguing water-filled side pore, next to the substrate-binding histidine, is involved in deprotonating ammonium ions, thereby increasing the permeation of NH3 [57]. Additionally, there was a potential correlation between ammonium uptake and water absorption, which was regulated by AQP. In rice plants, ammonium could increase the expression of PIP and TIP genes in the roots and resulted in a higher water uptake rate compared with that of nitrate (Figure 1a). Under water stress, ammonium increased drought tolerance of rice plants by inducing aquaporin expression and/or activity, which corresponded with increased root water uptake ability [58]. However, in French bean plants with a ‘one shoot-two roots’ split root system, Guo et al. [59] demonstrated that the mRNA expression of PIP1 was higher in the roots supplied with nitrate than those supplied with ammonium (Figure 1b). Generally, rice prefers ammonium nutrition while beans prefer nitrate nutrition, demonstrating that AQP expression is upregulated under favoured nitrogen nutrition. These results suggested that ammonium and nitrate differentially regulated water uptake and AQP in different plant species. 2.3. Urea Urea is a major N fertilizer used in agricultural production and is also a naturally occurring and readily available N source in soil. Urea is an uncharged small solute and passes through plant membranes via AQP [49,61,62,63], and members of the PIP, NIP, and TIP subfamilies have been shown to facilitate urea crossing membranes [23,24,64]. NIPs and PIPs, localized to the plasma membrane, function in urea movement between the apoplast and the symplast of plant cells [33,65,66]. In comparison, TIPs were targeted mainly to the tonoplast or other endo-membranes and are involved in equilibrating urea concentrations between different cellular compartments [23]. In Arabidopsis, several native NIPs, such as AtNIP6;1 and AtNIP5;1, were shown to transport urea, and AtNIP6;1 was also predicted to conduct substantial amounts of ammonia [64], and AtNIP5;1 was identified to transport boron acid [67]. However, AtNIP5;1 was identified to facilitate urea uptake only under B deficiency, in both high and low urea concentrations [67]. In maize plants, ZmNIP2;1, ZmNIP2;4, and ZmTIP4;4 were found to be involved in urea transport and played critical roles in urea uptake and movement, and stabilized urea concentrations in the tonoplast [23,63]. CsNIP2;1, a plasma membrane transporter from Cucumis sativus, facilitates urea uptake and internal transport during N remobilization and N delivery in plants [62]. In maize roots, the ZmPIP1;5, an aquaporin transport for water and urea, diverges from other PIP membranes by urea transport activity [49], and the expression of ZmPIP1;5 is induced by nitrate and modulated during the day-night cycle. Vacuoles could be used for short-term urea storage to avoid toxicity in the cytoplasm; this process was regulated by TIPs, which contribute to urea remobilization from the vacuole and equilibration within the cell [24,68]. Under nitrogen deficient conditions, expression of ZmNIP2;1 and ZmNIP2;4 was not affected, whereas the expression of ZmTIP4;4 increased significantly in the roots and expanded leaves, suggesting that ZmTIP4;4-regulated urea transport was essential for unloading vacuolar urea across the tonoplast under N starvation conditions [63]. Moreover, AtTIP1;1, AtTIP1;2, AtTIP2;1, and AtTIP4;1, which are different from the high-affinity H+/urea symporter AtDUR3, provide a less concentration- and pH-dependent pathway for urea transport from the external growth medium into the cytosol or from the cytosol into the vacuole [23,61]. AtTIP1;3 and AtTIP5;1, the only highly expressed pollen-specific aquaporins, function in N remobilization via transport of mitochondrial urea to the cytoplasm [61,69]. Aquaporins are tightly linked with N metabolism in higher plants, and the PIP, NIP, and TIP subfamilies have been shown to transport NH3 and urea, and maintain the balance between the cytoplasm and vacuole. Understanding the principles of N compounds passing through the plasma membrane by aquaporins allow us to modulate the N uptake and utilization, and improve the nitrogen use efficiency in plants. Nitrate and ammonium were different in regulation of plant water uptake and AQP expression depending on the plant species. Nitrate is suggested to be a critical signalling factor to induce PIPs expression and increase root hydraulic conductivity in nitrate preferred plants, while ammonia increases PIPs and TIPs expression and water uptake in ammonia preferred plants. The regulation of AQP by different nitrogen forms provides an effective pathway to increased plant water stress and water use efficiency in plants. 3. Phosphorus (P) Phosphorus (P) is necessary for the synthesis of nucleic acids, which contains the genetic code of the plant for production of proteins and other compounds essential for plant structure, seed yield, and genetic transfer. A number of studies have indicated that the enhancement in plant growth with P fertilization is associated with an increased capacity of the plants to transport water [70,71]. The activity or density of aquaporins in the plasma membrane of root cells is diminished during nutrient stress, such as N- and P deprivation [48]. P is involved in root water uptake by altering aquaporin expression and/or activity [48,72]. P deficiency reduced aquaporin activity or abundance in the root plasma membrane [48] and was associated with a decrease in water uptake [72], which was attributed to phosphorylation of the plant aquaporins [73]. In Arabidopsis roots, the changes in the phosphorylation status of PIP aquaporins were positively correlated to changes in root hydraulic conductivity under NaCl, NO, and N and P starvation treatments [74]. Additionally, plants often exhibit disruption of water transport that is associated with enhanced ethylene production, which modulates root hydraulic conductivity by affecting the aquaporin activity under P deficient conditions [75,76]. As plant aquaporins are regulated by cytosolic pH and free Ca2+ activity [77], ethylene can elicit a rapid increase in cytosolic Ca2+ concentration by activating the Ca2+-permeable channels [78], as a result of inhibiting aquaporin activity. In sorghum plants, the root hydraulic conductivity of water-stressed plants with a sufficient P supply recovered faster than that of plants without a P supply, suggesting that sufficient P could increase AQP expression and/or activity after water recovery [79]. Arbuscular mycorrhizal (AM) fungi, which formed symbiotic associations with host plants, can uptake and deliver inorganic P to the host through hyphal networks. Under water stress, AM symbiosis can increase the tolerance of plants by regulating the AQP gene expression, osmotic adjustments, and plant growth [80,81]. The role of P on aquaporins is mainly focused on its phosphorylation functions by regulating aquaporin activity and abundance, and corresponds with regulated root hydraulic conductivity and water uptake. Further studies are needed to elucidate the specific functions of AQP genes regulated by AM symbiosis, in order to reveal the exact mechanism of AM symbiosis to deliver P and alter plant adaptation to environmental stressors. 4. Potassium (K) Potassium ion (K+), the most abundant cation in higher plants, functions in osmo-regulation, cation-anion balance, stomatal movement, photosynthesis, energy transfer, carbohydrate phloem transport, enzyme activation, and protein synthesis, as well as stress resistance [41]. As K+ is the major osmolyte, its uptake will be accompanied by water flux through the aquaporins, and there was a positive correlation between K absorption and water uptake [82]. It was suggested that aquaporins could function as turgor sensors to modulate the conductance of K+ channels [83]. Transcripts encoding aquaporins were strongly affected by K+ starvation, even without water stress [84]. In Arabidopsis, iterative group analysis (iGA) identified 12 aquaporin genes in the shoots and 15 genes in the roots that were significantly up-regulated after K+ resupply [85]. The trafficking and activity of plasma membrane aquaporin PIPs is regulated by the SNARE SYP121, a plasma membrane resident syntaxin involved in vesicle trafficking, signaling, and regulation of K+ channels [86,87]. SYP121 plays a role in the regulation and maintenance of membrane osmotic water permeability through a coordinated regulation of the plasma membrane density of both PIP and K+ channels in membrane delivery and recycling [86]. The aquaporins may participate in ion homeostasis at the whole plant level by regulating the ratio of apoplastic/symplastic water flow and thus directing solute flux through different plant tissues. In onion roots, water transport was sensitive to inhibitors of aquaporins and K+ channels, and the decrease in hydraulic conductivity after K+ channel inhibitor treatment indicates that K+ fluxes are involved in aquaporin activity in the plasma membrane [88]. In Arabidopsis roots, the expression of genes encoding water channels of the aquaporins PIP1b, PIP2b, and TIP, as well as the K+ transporter HAK5 were reduced after K+ channel inhibitor (CsCl) treatment [89], suggesting that K+ channel blockers could reduce the hydraulic conductivity of the plasma membrane by down-regulating or blocking water channels. It has been reported that aquaporins and K+ channels can function as plant osmo-regulators to maintain cytosolic osmolarity and increase the tolerance of the plant to drought or other stressors [90,91]. In rice, the expression of PIP and K+ channels responded similarly to K deficiency and water stress, in which expression of PIPs and K+ channel-encoding genes was induced by K+ starvation and down-regulated by water deficit during a short time, suggesting that aquaporins and K+ channels are functionally co-regulated during cell turgor regulation [90]. Loading K+ into the plant xylem could regulate the xylem hydraulic conductivity, which can help maintain cell turgor, stomatal aperture, and gas exchange rates, as a result of increasing drought tolerance [92,93]. During drought stress, plants modulate their water and ion uptake capacities by regulating aquaporins and K+ channels at the transcriptional level to respond to the water deficiency [90,94,95]. Aquaporins participate in whole plant ion homeostasis and act as turgor sensors to modulate the K+ channels. Aquaporins and K+ channels can function as plant osmo-regulators to maintain cytosolic osmolarity and increase tolerance to drought stress, which corresponds with rapid recovery of the shoot water status, cell turgidity, and thus plant growth. The coordination of aquaporins and K+ transport in plants during different stressors and physiological states, may be regulated by different signalling pathways. 5. Calcium (Ca) Calcium (Ca) is an essential macronutrient that functions in the cell wall and membranes, acting as a counter-cation for inorganic and organic anions in the vacuole, as well as a secondary messenger in cell signal transduction [41,96,97]. Generally, Ca2+ enters the root apoplast via the mass flow from the soil solution [41], suggesting that transpiration-regulated water flow may be involved in Ca2+ delivery and storage, which could be regulated by aquaporins [97]. Conversely, Ca2+ could affect AQP activity and/or expression, and aquaporin expression was suppressed by Ca2+ starvation [84]. It has been reported that the inhibition of maize root water transport by HgCl2 was detected only in the presence of Ca2+ in the nutrient solution, suggesting that Ca2+ is involved in regulating aquaporin activity [98]. Several studies demonstrated that the aquaporin functions could be regulated by Ca2+ [99,100] and triggered by environmental stressors [101]. Under water stress, the expression or activity of aquaporins was affected [94,98], and this process could be facilitated by excess Ca2+ [102]. Salt stress decreased water transport through the plasma membrane and the root cortical cells by reducing Hg-sensitive aquaporin activity, and the ameliorative effect of Ca2+ on salt stress was related to aquaporin function [72,103,104]. In pepper plants, cytosolic Ca2+ decreased after long-term exposure to salt stress with a corresponding overall inhibition of aquaporins [105]. Reversible phosphorylation, a potential mechanism for plant aquaporin regulation during development and in the response of plants to environmental stimuli [73,106,107], could be regulated by Ca2+, indicating a link between aquaporin regulation and Ca2+ signalling [105]. In guard cells, extracellular Ca2+ is involved in stomatal movement by acting as an elicitor (second messenger) or aquaporin blocker, which may initiate the signal cascade and lead to the post-transcriptional regulation of aquaporins or directly block aquaporins [108]. The aquaporin gate was regulated by cytosolic Ca2+ transport, especially the opening and closing of verapamil-sensitive Ca2+ channels [104]. In vitro phosphorylation of the aquaporin PM28A was directly dependent on submicromolar Ca2+ concentrations [99]. Ca2+ is involved in plasma membrane aquaporin regulation via a chain of processes within the cell, but its effects are not due to alteration of the stability of the plasma membrane [104]. Cytosolic Ca2+ transport and Ca2+ channels might directly regulate water flow by acting on aquaporins, which would affect nutrient movement through the plant. In the response of plants to environmental stimuli, the functions of aquaporin could be regulated by Ca2+ via reversible phosphorylation. However, the regulation mechanism of Ca2+ on aquaporins and its physiological role in whole plant conditions remains to be established. 6. Boron (B) Boron (B) is an essential micronutrient for plant growth and development, especially for the structure and function of the plant cell wall [41]. B deficiency and toxicity in plants results in a significant reduction in quality and yield of many crops worldwide [109,110]. Aquaporins have been shown to function in B transport in higher plants [111,112], and are required for normal plant growth under B deficiency and toxic conditions [17,113,114]. AtNIP5;1, a boric acid channel that belongs to the major intrinsic proteins (MIPs), is predominantly expressed in epidermal, cortical, and endodermal cells [17,30]. Under B deficiency, AtNIP5;1 expression was strongly up-regulated, which is critical for efficient B transport into the roots [17]. In nip5;1 mutants of Arabidopsis thaliana, both root and shoot growth were inhibited under B deficiency [17], indicating that NIP5;1 was essential for the overall B uptake that was required for plant growth and development under B limitation. AtNIP6;1, which is homologous to AtNIP5;1, was shown to facilitate the rapid penetration of boric acid across the membrane and normal distribution of boric acid in plant tissues, but it is completely impermeable to water [113]. Similarly, the water channel OsNIP3;1 was also found to be a B-inducible channel in rice involved in B uptake and distribution [115]. In barley, the tolerance to excessive soil B is controlled by downregulated expression of HvNIP2;1 to reduce B uptake and leaf blade B accumulation. Expression of Bot1, a BOR1 ortholog that provides B tolerance to barley, was induced to eliminate B from the roots and sensitive tissues [114]. HvNIP2;1 is essential for B toxicity tolerance in barley in combination with Bot1. AtTIP5;1 plays a critical role in the B transport pathway possibly via vacuolar compartmentation, and the overexpression of AtTIP5;1 may facilitate the elimination of B toxicity in plants [116]. OsPIP1;3, OsPIP2;4, OsPIP2;6, and OsPIP2;7, members of the major intrinsic proteins (MIPs) family, were involved in both influx and efflux of B transport, and their expressions were strongly upregulated under B toxicity [117,118]. Briefly, aquaporins are essential for reducing the accumulation of toxic boric acid levels in plant tissues [9]. Aquaporins were involved in B uptake and distribution, and PIP, NIP, TIP, and XIP subfamilies have been shown to transport boric acid. Under B deficiency, NIPs are essential for efficient B uptake and distribution that is required for plant growth and development. Whereas under B toxicity, NIPs, TIPs, and PIPs are involved in reducing the accumulation of toxic boric acid levels in plant tissues. Manipulation of these aquaporins could be highly useful in improving plant tolerance to B deficiency or toxicity. 7. Silicon (Si) Silicon (Si), the second most abundant element in the earth’s crust, is important for plant growth and development. Si is beneficial to the mechanical and physiological properties of plants and helps plants to overcome biotic and abiotic stress [34,119,120,121]. Under salt stress, Si can improve plant tolerance through enhancing root water uptake which contributes to the regulation of aquaporin activity and gene expression [122,123]. The Si uptake by plants in soil solution is through silicic acid [Si(OH)4], an uncharged molecule [34]. Silicic acid enters the plant roots mainly by water flow via the apoplastic and symplastic pathways, and the symplastic pathway involves the presence of water channels, mainly NIPs [124]. In rice, two Si transporters, Lsi1 and Lsi2, have been shown to be involved in Si uptake. Lsi1 is localized on the distal side of the plasma membrane of the exodermal and endodermal cells and functions as an influx transporter [33,125], whereas Lsi2 is located on the proximal side of the same root cells and functions as an efflux transporter [125,126]. The combination of Lsi1 and Lsi2 enables rice to efficiently transport silicic acid from the soil solution into the xylem of the roots. Lsi1 (also named OsNIP2;1) belongs to the NIP subfamily of aquaporins, and previous studies have shown that NIP proteins are permeable to a wide range of substrates, such as silicic acid [33], arsenite [125], boric acid [17], urea and formamide [64], glycerol [127], lactic acid [35], as well as selenite [128]. Lsi6 (OsNIP2;2), which is localized in the xylem parenchyma cells of leaf blades and sheaths, was also identified as responsible for Si xylem unloading [129]. In barley, a Lsi1 ortholog of HvLsi1 (HvNIP2;1), localized in the plasma membrane of epidermal cells and all cortical cells in roots, was identified as a Si influx transporter and shown to be involved in the radial transport of Si through the epidermal and cortical layers of the basal roots [32]. Si absorption is facilitated by NIPs, and two Si transporters have been identified that are involved in Si uptake, Lsi1 and Lsi2, which function as an influx and efflux transporter, respectively. Cooperation of Lsi1 and Lsi2 is required for the efficient transport of Si. The identification of Si transporters provides an insight into the Si uptake system in plants and a new approach for producing crops with high resistance to various biotic and abiotic stresses by genetic modification. To further elucidate the Si accumulation mechanism and understand the critical role of Si at the whole plant level, molecular and physiological characterization of Si transporters in different plant species is required in the future. 8. Conclusions and Future Perspectives A wide range of selectivity profiles and regulatory properties allow aquaporins to be involved in multiple functions in plant growth and development, such as water transport, and nitrogen, carbon, and micronutrient acquisition. Aquaporins, mainly PIPs, TIPs, and NIPs, have been shown to facilitate the transport of plant mineral nutrients across plasma membranes and cell organelles (Figure 2), such as ammonia, urea, boric acid, and silicic acid. Aquaporins are responsible for ensuring different mineral nutrient availability for the plant and play essential roles in mineral nutrient absorption, mobilization, detoxification, and homeostasis. There is a tight link between plant aquaporins and mineral nutrients. Aquaporin expression is regulated by mineral nutrient availability and plant species. Aquaporin function can be regulated by mineral nutrients in the response of plants to environmental stimuli, such as drought and salt stress, nutrient deficiency, and toxicity. Understanding the interactions between aquaporins and mineral nutrients allow us to modulate the water and mineral nutrient uptake and utilization, and improve the water and nutrient use efficiency in plants, as well as increase tolerance to biotic and abiotic stress. In the future, attention should be focused on: (1) The functions of aquaporins in the transport of other novel putative substrates, such as Mg, S, and other micronutrients, which await further investigation. (2) New aquaporin subclasses and unknown functions of aquaporins recently discovered in certain plant species should be deciphered. (3) Investigation of the interactions between water and mineral nutrient transport, as well as interactions between different mineral nutrients regulated by aquaporins will be required. (4) The role of aquaporins during biotic and abiotic stress, and the relevance of altered aquaporin expression for biotechnological improvement of plant tolerance must be explored. (5) Aquaporin functions need to be further investigated concerning whole plant physiology, which requires a better understanding of how the various aquaporin transport activities are coupled with plant mineral nutrient transport proteins. Acknowledgments This work was financially supported by the National Basic Research Program of China (2015CB150505 and 2013CB127403), the National Key R & D Program (2016YFD0200300), the National Natural Science Foundation of China (31272236 and 31401941), the Key Fund Project of Jiangsu Province (BK20150059), the Jiangsu Postdoctoral Science Foundation (1402148C), and the China Postdoctoral Science Foundation (2015M571768). Author Contributions Min Wang and Shiwei Guo wrote the paper; Lei Ding performed the experiments; Limin Gao and Yingrui Li analyzed the data; Qirong Shen and Shiwei Guo commented and improved the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Effect of different nitrogen sources (ammonium vs. nitrate) on aquaporin (AQP) expression in rice (a) and French bean plants (b). Rice plants were supplied with 2.5 mM ammonium [(NH4)2SO4] or nitrate [Ca(NO3)2] for two weeks. Root samples were collected for RNA isolation, and quantitative real-time PCR (q-RT-PCR) was performed to detect the relative expression of the plasma membrane intrinsic proteins (PIPs) and tonoplast intrinsic proteins (TIPs) [60]; (b) French bean plants were grown in a split-root system, in which half of the roots were supplied with 5 mM ammonium [(NH4)2SO4] or nitrate [Ca(NO3)2]. The PIP1 aquaporin expression in the roots was determined via Northern blot analysis until day 5 after the treatments [59]. Figure 2 The multiple cellular functions of plant aquaporins in mineral nutrition. The figure illustrates the variety of transporter functions of aquaporins in various subcellular compartments. The different subclasses of aquaporins are identified in different colours. The plasma membrane intrinsic proteins (PIPs) might be involved in the internal re-distribution of mineral nutrients by transporting them from the endoplasmic reticulum (ER) to the plasma membrane via the Golgi apparatus. Moreover, PIPs also undergo repeated cycles of endocytosis and recycling through the early endosome to the multivesicular body/prevacuolar compartments before eventually being targeted to the vacuole. PIPs primarily facilitated urea and boric acid transport, while tonoplast intrinsic proteins (TIPs) are principally involved in urea, NH3, and boric acid transport, and nodulin 26-like intrinsic proteins (NIPs) are involved in urea, boric acid, and silicic acid transport. ==== Refs References 1. Heinen R.B. Ye Q. Chaumont F. Role of aquaporins in leaf physiology J. Exp. Bot. 2009 60 2971 2985 10.1093/jxb/erp171 19542196 2. Maurel C. Aquaporins and water permeability of plant membranes Annu. Rev. Plant Physiol. Plant Mol. Biol. 1997 48 399 429 10.1146/annurev.arplant.48.1.399 15012269 3. Maurel C. Santoni V. Luu D.T. Wudick M.M. Verdoucq L. The cellular dynamics of plant aquaporin expression and functions Curr. Opin. Plant Biol. 2009 12 690 698 10.1016/j.pbi.2009.09.002 19783200 4. Maurel C. Verdoucq L. Luu D.T. Santoni V. Plant aquaporins: Membrane channels with multiple integrated functions Annu. Rev. Plant Biol. 2008 59 595 624 10.1146/annurev.arplant.59.032607.092734 18444909 5. Maurel C. Chrispeels M.J. Aquaporins. A molecular entry into plant water relations Plant Physiol. 2001 125 135 138 10.1104/pp.125.1.135 11154316 6. Bienert G.P. Chaumont F. Plant aquaporins: Roles in water homeostasis, nutrition, and signaling processes Transporters and Pumps in Plant Signaling Geisler M. Venema K. Springer Berlin Heidelberg Berlin & Heidelberg, Germany 2011 3 36 7. Wallace I.S. Choi W.G. Roberts D.M. The structure, function and regulation of the nodulin 26-like intrinsic protein family of plant aquaglyceroporins Biochim. Biophys. Acta 2006 1758 1165 1175 10.1016/j.bbamem.2006.03.024 16716251 8. Ishikawa F. Suga S. Uemura T. Sato M.H. Maeshima M. Novel type aquaporin SIPs are mainly localized to the ER membrane and show cell-specific expression in Arabidopsis thaliana FEBS Lett. 2005 579 5814 5820 10.1016/j.febslet.2005.09.076 16223486 9. Bienert G.P. Bienert M.D. Jahn T.P. Boutry M. Chaumont F. Solanaceae XIPs are plasma membrane aquaporins that facilitate the transport of many uncharged substrates Plant J. 2011 66 306 317 10.1111/j.1365-313X.2011.04496.x 21241387 10. Danielson J.A.H. Johanson U. Unexpected complexity of the aquaporin gene family in the moss Physcomitrella patens BMC Plant Biol. 2008 8 10.1186/1471-2229-8-45 18430224 11. Hachez C. Zelazny E. Chaumont F. Modulating the expression of aquaporin genes in planta: A key to understand their physiological functions? Biochim. Biophys. Acta 2006 1758 1142 1156 10.1016/j.bbamem.2006.02.017 16580626 12. Rizhsky L. Liang H.J. Shuman J. Shulaev V. Davletova S. Mittler R. When defense pathways collide. The response of Arabidopsis to a combination of drought and heat stress Plant Physiol. 2004 134 1683 1696 10.1104/pp.103.033431 15047901 13. Sade N. Gebretsadik M. Seligmann R. Schwartz A. Wallach R. Moshelion M. The role of tobacco Aquaporin1 in improving water use efficiency, hydraulic conductivity, and yield production under salt stress Plant Physiol. 2010 152 245 254 10.1104/pp.109.145854 19939947 14. Xu Y. Hu W. Liu J.H. Zhang J.B. Jia C.H. Miao H.X. Xu B.Y. Jin Z.Q. A banana aquaporin gene, MaPIP1;1 , is involved in tolerance to drought and salt stresses BMC Plant Biol. 2014 14 10.1186/1471-2229-14-59 24606771 15. Huang C. Zhou S.Y. Hu W. Deng X.M. Wei S.Y. Yang G.X. He G.Y. The wheat aquaporin gene TaAQP7 confers tolerance to cold stress in transgenic tobacco Z. Naturforsch. C. 2014 69 142 148 10.5560/znc.2013-0079 24873035 16. Ahamed A. Murai-Hatano M. Ishikawa-Sakurai J. Hayashi H. Kawamura Y. Uemura M. Cold stress-induced acclimation in rice is mediated by root-specific aquaporins Plant Cell Physiol. 2012 53 1445 1456 10.1093/pcp/pcs089 22711693 17. Takano J. Wada M. Ludewig U. Schaaf G. von Wiren N. Fujiwara T. The Arabidopsis major intrinsic protein NIP5;1 is essential for efficient boron uptake and plant development under boron limitation Plant Cell 2006 18 1498 1509 10.1105/tpc.106.041640 16679457 18. Zhao F.J. Ago Y. Mitani N. Li R.Y. Su Y.H. Yamaji N. McGrath S.P. Ma J.F. The role of the rice aquaporin Lsi1 in arsenite efflux from roots New Phytol. 2010 186 392 399 10.1111/j.1469-8137.2010.03192.x 20163552 19. Zhao F.J. Ma J.F. Meharg A.A. McGrath S.P. Arsenic uptake and metabolism in plants New Phytol. 2009 181 777 794 10.1111/j.1469-8137.2008.02716.x 19207683 20. Zou J.J. Rodriguez-Zas S. Aldea M. Li M. Zhu J. Gonzalez D.O. Vodkin L.O. DeLucia E. Clough S.J. Expression profiling soybean response to Pseudomonas syringae reveals new defense-related genes and rapid HR-specific downregulation of photosynthesis Mol. Plant Microbe Interact. 2005 18 1161 1174 10.1094/MPMI-18-1161 16353551 21. Kim M.J. Kim H.R. Paek K.H. Arabidopsis tonoplast proteins TIP1 and TIP2 interact with the cucumber mosaic virus 1a replication protein J. Gen. Virol. 2006 87 3425 3431 10.1099/vir.0.82252-0 17030879 22. Afzal Z. Howton T.C. Sun Y. Mukhtar M.S. The roles of aquaporins in plant stress responses J. Dev. Biol. 2016 4 9 10.3390/jdb4010009 23. Liu L.H. Ludewig U. Gassert B. Frommer W.B. von Wiren N. Urea transport by nitrogen-regulated tonoplast intrinsic proteins in Arabidopsis Plant Physiol. 2003 133 1220 1228 10.1104/pp.103.027409 14576283 24. Gerbeau P. Guclu J. Ripoche P. Maurel C. Aquaporin Nt-TIPa can account for the high permeability of tobacco cell vacuolar membrane to small neutral solutes Plant J. 1999 18 577 587 10.1046/j.1365-313x.1999.00481.x 10417709 25. Bertl A. Kaldenhoff R. Function of a separate NH3 -pore in Aquaporin TIP2;2 from wheat FEBS Lett. 2007 581 5413 5417 10.1016/j.febslet.2007.10.034 17967420 26. Loque D. Ludewig U. Yuan L.X. von Wiren N. Tonoplast intrinsic proteins AtTIP2;1 and AtTIP2;3 facilitate NH3 transport into the vacuole Plant Physiol. 2005 137 671 680 10.1104/pp.104.051268 15665250 27. Flexas J. Ribas-Carbó M. Hanson D.T. Bota J. Otto B. Cifre J. McDowell N. Medrano H. Kaldenhoff R. Tobacco aquaporin NtAQP1 is involved in mesophyll conductance to CO2 in vivo Plant J. 2006 48 427 439 10.1111/j.1365-313X.2006.02879.x 17010114 28. Uehlein N. Lovisolo C. Siefritz F. Kaldenhoff R. The tobacco aquaporin NtAQP1 is a membrane CO2 pore with physiological functions Nature 2003 425 734 737 10.1038/nature02027 14520414 29. Ding L. Gao L. Liu W. Wang M. Gu M. Ren B. Xu G. Shen Q. Guo S. Aquaporin plays an important role in mediating chloroplastic CO2 concentration under high-N supply in rice (Oryza sativa ) plants Physiol. Plant. 2016 156 215 226 10.1111/ppl.12387 26382720 30. Kato Y. Miwa K. Takano J. Wada M. Fujiwara T. Highly boron deficiency-tolerant plants generated by enhanced expression of NIP5;1, a boric acid channel Plant Cell Physiol. 2009 50 58 66 10.1093/pcp/pcn168 19017629 31. Durbak A.R. Phillips K.A. Pike S. O’Neill M. Mares J. Gallavotti A. Malcomber S.T. Gassmann W. McSteen P. Transport of boron by the tassel-less1 aquaporin is critical for vegetative and reproductive development in maize Plant Cell 2014 26 2978 2995 10.1105/tpc.114.125898 25035406 32. Chiba Y. Mitani N. Yamaji N. Ma J.F. HvLsi1 is a silicon influx transporter in barley Plant J. 2009 57 810 818 10.1111/j.1365-313X.2008.03728.x 18980663 33. Ma J.F. Tamai K. Yamaji N. Mitani N. Konishi S. Katsuhara M. Ishiguro M. Murata Y. Yano M. A silicon transporter in rice Nature 2006 440 688 691 10.1038/nature04590 16572174 34. Ma J.F. Yamaji N. Silicon uptake and accumulation in higher plants Trends Plant Sci. 2006 11 392 397 10.1016/j.tplants.2006.06.007 16839801 35. Choi W.G. Roberts D.M. Arabidopsis NIP2;1, a major intrinsic protein transporter of lactic acid induced by anoxic stress J. Biol. Chem. 2007 282 24209 24218 10.1074/jbc.M700982200 17584741 36. Bienert G.P. Moller A.L.B. Kristiansen K.A. Schulz A. Moller I.M. Schjoerring J.K. Jahn T.P. Specific aquaporins facilitate the diffusion of hydrogen peroxide across membranes J. Biol. Chem. 2007 282 1183 1192 10.1074/jbc.M603761200 17105724 37. Dynowski M. Schaaf G. Loque D. Moran O. Ludewig U. Plant plasma membrane water channels conduct the signalling molecule H2 O2 Biochem. J. 2008 414 53 61 10.1042/BJ20080287 18462192 38. Bienert G.P. Chaumont F. Aquaporin-facilitated transmembrane diffusion of hydrogen peroxide Biochim. Biophys. Acta 2014 1840 1596 1604 10.1016/j.bbagen.2013.09.017 24060746 39. Bienert G.P. Schussler M.D. Jahn T.P. Metalloids: Essential, beneficial or toxic? Major intrinsic proteins sort it out Trends Biochem. Sci. 2008 33 20 26 10.1016/j.tibs.2007.10.004 18068370 40. Kusano M. Fukushima A. Redestig H. Saito K. Metabolomic approaches toward understanding nitrogen metabolism in plants J. Exp. Bot. 2011 62 1439 1453 10.1093/jxb/erq417 21220784 41. Marschner H. Marschner P. Marschner’s Mineral Nutrition of Higher Plants 3rd ed. Elsevier London, UK Academic Press London, UK 2012 42. Hacke U.G. Plavcova L. Almeida-Rodriguez A. King-Jones S. Zhou W.C. Cooke J.E.K. Influence of nitrogen fertilization on xylem traits and aquaporin expression in stems of hybrid poplar Tree Physiol. 2010 30 1016 1025 10.1093/treephys/tpq058 20610665 43. Ishikawa-Sakurai J. Hayashi H. Murai-Hatano M. Nitrogen availability affects hydraulic conductivity of rice roots, possibly through changes in aquaporin gene expression Plant Soil 2014 379 289 300 10.1007/s11104-014-2070-4 44. Ren B.B. Wang M. Chen Y.P. Sun G.M. Li Y. Shen Q.R. Guo S.W. Water absorption is affected by the nitrogen supply to rice plants Plant Soil 2015 396 397 410 10.1007/s11104-015-2603-5 45. Wilkinson S. Bacon M.A. Davies W.J. Nitrate signalling to stomata and growing leaves: Interactions with soil drying, ABA, and xylem sap pH in maize J. Exp. Bot. 2007 58 1705 1716 10.1093/jxb/erm021 17374875 46. Gloser V. Zwieniecki M.A. Orians C.M. Holbrook N.M. Dynamic changes in root hydraulic properties in response to nitrate availability J. Exp. Bot. 2007 58 2409 2415 10.1093/jxb/erm118 17562690 47. Gorska A. Ye Q. Holbrook N.M. Zwieniecki M.A. Nitrate control of root hydraulic properties in plants: Translating local information to whole plant response Plant Physiol. 2008 148 1159 1167 10.1104/pp.108.122499 18753287 48. Clarkson D.T. Carvajal M. Henzler T. Waterhouse R.N. Smyth A.J. Cooke D.T. Steudle E. Root hydraulic conductance: Diurnal aquaporin expression and the effects of nutrient stress J. Exp. Bot. 2000 51 61 70 10.1093/jexbot/51.342.61 10938796 49. Gaspar M. Bousser A. Sissoeff I. Roche O. Hoarau J. Mahe A. Cloning and characterization of ZmPIP1-5b , an aquaporin transporting water and urea Plant Sci. 2003 165 21 31 10.1016/S0168-9452(03)00117-1 50. Wang Y.H. Garvin D.F. Kochian L.V. Nitrate-induced genes in tomato roots. Array analysis reveals novel genes that may play a role in nitrogen nutrition Plant Physiol. 2001 127 1323 1323 10.1104/pp.127.1.345 51. Li G. Tillard P. Gojon A. Maurel C. Dual regulation of root hydraulic conductivity and plasma membrane aquaporins by plant nitrate accumulation and high-affinity nitrate transporter NRT2.1 Plant Cell Physiol. 2016 10.1093/pcp/pcw022 26823528 52. Remans T. Nacry P. Pervent M. Filleur S. Diatloff E. Mounier E. Tillard P. Forde B.G. Gojon A. The Arabidopsis NRT1.1 transporter participates in the signaling pathway triggering root colonization of nitrate-rich patches Proc. Natl. Acad. Sci. USA 2006 103 19206 19211 10.1073/pnas.0605275103 17148611 53. Walch-Liu P. Ivanov I.I. Filleur S. Gan Y.B. Remans T. Forde B.G. Nitrogen regulation of root branching Ann. Bot. 2006 97 875 881 10.1093/aob/mcj601 16339770 54. Wang W.H. Kohler B. Cao F.Q. Liu L.H. Molecular and physiological aspects of urea transport in higher plants Plant Sci. 2008 175 467 477 10.1016/j.plantsci.2008.05.018 55. Barzana G. Aroca R. Bienert G.P. Chaumont F. Ruiz-Lozano J.M. New insights into the regulation of aquaporins by the arbuscular mycorrhizal symbiosis in maize plants under drought stress and possible implications for plant performance Mol. Plant Microbe Interact. 2014 27 349 363 10.1094/MPMI-09-13-0268-R 24593244 56. Jahn T.P. Moller A.L.B. Zeuthen T. Holm L.M. Klaerke D.A. Mohsin B. Kuhlbrandt W. Schjoerring J.K. Aquaporin homologues in plants and mammals transport ammonia FEBS Lett. 2004 574 31 36 10.1016/j.febslet.2004.08.004 15358535 57. Kirscht A. Kaptan S.S. Bienert G.P. Chaumont F. Nissen P. de Groot B.L. Kjellbom P. Gourdon P. Johanson U. Crystal structure of an ammonia-permeable aquaporin PLoS Biol. 2016 14 1229 10.1371/journal.pbio.1002411 27028365 58. Ding L. Gao C.M. Li Y.R. Li Y. Zhu Y.Y. Xu G.H. Shen Q.R. Kaldenhoff R. Kai L. Guo S.W. The enhanced drought tolerance of rice plants under ammonium is related to aquaporin (AQP) Plant Sci. 2015 234 14 21 10.1016/j.plantsci.2015.01.016 25804805 59. Guo S.W. Kaldenhoff R. Uehlein N. Sattelmacher B. Brueck H. Relationship between water and nitrogen uptake in nitrate- and ammonium-supplied Phaseolus vulgaris L. plants J. Plant Nutr. Soil Sci. 2007 170 73 80 10.1002/jpln.200625073 60. Ding L. Li Y.R. Wang Y. Gao L.M. Wang M. Chaumont F. Shen Q.R. Guo S.W. Root ABA accumulation enhances rice seedling drought tolerance under ammonium supply: Interaction with aquaporins Front. Plant Sci. 2016 in press 61. Soto G. Alleva K. Mazzella M.A. Amodeo G. Muschietti J.P. AtTIP1;3 and AtTIP5;1, the only highly expressed Arabidopsis pollen-specific aquaporins, transport water and urea FEBS Lett. 2008 582 4077 4082 10.1016/j.febslet.2008.11.002 19022253 62. Zhang L. Yan J. Vatamaniuk O.K. Du X. CsNIP2;1 is a plasma membrane transporter from Cucumis sativus that facilitates urea uptake when expressed in Saccharomyces cerevisiae and Arabidopsis thaliana Plant Cell Physiol. 2016 57 616 629 10.1093/pcp/pcw018 26858284 63. Gu R.L. Chen X.L. Zhou Y.L. Yuan L.X. Isolation and characterization of three maize aquaporin genes, ZmNIP2 ;1 , ZmNIP2 ;4 and ZmTIP4 ;4 involved in urea transport BMB Rep. 2012 45 96 101 10.5483/BMBRep.2012.45.2.96 22360887 64. Wallace I.S. Roberts D.M. Distinct transport selectivity of two structural subclasses of the nodulin-like intrinsic protein family of plant aquaglyceroporin channels Biochemistry 2005 44 16826 16834 10.1021/bi0511888 16363796 65. Chaumont F. Barrieu F. Jung R. Chrispeels M.J. Plasma membrane intrinsic proteins from maize cluster in two sequence subgroups with differential aquaporin activity Plant Physiol. 2000 122 1025 1034 10.1104/pp.122.4.1025 10759498 66. Gao Z.X. He X.L. Zhao B.C. Zhou C.J. Liang Y.Z. Ge R.C. Shen Y.Z. Huang Z.J. Overexpressing a putative aquaporin gene from wheat, TaNIP , enhances salt tolerance in transgenic Arabidopsis Plant Cell Physiol. 2010 51 767 775 10.1093/pcp/pcq036 20360019 67. Yang H.Y. Menz J. Haussermann I. Benz M. Fujiwara T. Ludewig U. High and low affinity urea root uptake: Involvement of NIP5;1 Plant Cell Physiol. 2015 56 1588 1597 10.1093/pcp/pcv067 25957355 68. Tyerman S.D. Bohnert H.J. Maurel C. Steudle E. Smith J.A.C. Plant aquaporins: Their molecular biology, biophysics and significance for plant water relations J. Exp. Bot. 1999 50 1055 1071 10.1093/jxb/50.Special_Issue.1055 69. Soto G. Fox R. Ayub N. Alleva K. Guaimas F. Erijman E.J. Mazzella A. Amodeo G. Muschietti J. TIP5;1 is an aquaporin specifically targeted to pollen mitochondria and is probably involved in nitrogen remobilization in Arabidopsis thaliana Plant J. 2010 64 1038 1047 10.1111/j.1365-313X.2010.04395.x 21143683 70. Cheeseman J.M. Lovelock C.E. Photosynthetic characteristics of dwarf and fringe Rhizophora mangle L. in a Belizean mangrove Plant Cell Environ. 2004 27 769 780 10.1111/j.1365-3040.2004.01181.x 71. Lovelock C.E. Ball M.C. Choat B. Engelbrecht B.M.J. Holbrook N.M. Feller I.C. Linking physiological processes with mangrove forest structure: Phosphorus deficiency limits canopy development, hydraulic conductivity and photosynthetic carbon gain in dwarf Rhizophora mangle Plant Cell Environ. 2006 29 793 802 10.1111/j.1365-3040.2005.01446.x 17087463 72. Carvajal M. Cooke D.T. Clarkson D.T. Responses of wheat plants to nutrient deprivation may involve the regulation of water-channel function Planta 1996 199 372 381 10.1007/BF00195729 73. Johansson I. Karlsson M. Shukla V.K. Chrispeels M.J. Larsson C. Kjellbom P. Water transport activity of the plasma membrane aquaporin PM28A is regulated by phosphorylation Plant Cell 1998 10 451 459 10.1105/tpc.10.3.451 9501117 74. Di Pietro M. Vialaret J. Li G.W. Hem S. Prado K. Rossignol M. Maurel C. Santoni V. Coordinated post-translational responses of aquaporins to abiotic and nutritional stimuli in Arabidopsis roots Mol. Cell. Proteom. 2013 12 3886 3897 10.1074/mcp.M113.028241 24056735 75. Li Y.S. Mao X.T. Tian Q.Y. Li L.H. Zhang W.H. Phosphorus deficiency-induced reduction in root hydraulic conductivity in Medicago falcata is associated with ethylene production Environ. Exp. Bot. 2009 67 172 177 10.1016/j.envexpbot.2009.05.013 76. Kamaluddin M. Zwiazek J.J. Ethylene enhances water transport in hypoxic aspen Plant Physiol. 2002 128 962 969 10.1104/pp.010791 11891251 77. Alleva K. Niemietz C.M. Maurel C. Parisi M. Tyerman S.D. Amodeo G. Plasma membrane of Beta vulgaris storage root shows high water channel activity regulated by cytoplasmic pH and a dual range of calcium concentrations J. Exp. Bot. 2006 57 609 621 10.1093/jxb/erj046 16397000 78. Zhao M.G. Tian Q.Y. Zhang W.H. Ethylene activates a plasma membrane Ca2+ -permeable channel in tobacco suspension cells New Phytol. 2007 174 507 515 10.1111/j.1469-8137.2007.02037.x 17447907 79. Shangguan Z.P. Lei T.W. Shao M.A. Xue Q.W. Effects of phosphorus nutrient on the hydraulic conductivity of sorghum (Sorghum vulgare Pers.) seedling roots under water deficiency J. Integr. Plant Biol. 2005 47 421 427 10.1111/j.1744-7909.2005.00069.x 80. Yooyongwech S. Samphumphuang T. Tisarum R. Theerawitaya C. Cha-um S. Arbuscular mycorrhizal fungi (AMF) improved water deficit tolerance in two different sweet potato genotypes involves osmotic adjustments via soluble sugar and free proline Sci. Hortic. Amst. 2016 198 107 117 10.1016/j.scienta.2015.11.002 81. He F. Zhang H.Q. Tang M. Aquaporin gene expression and physiological responses of Robinia pseudoacacia L. to the mycorrhizal fungus Rhizophagus irregularis and drought stress Mycorrhiza 2016 26 311 323 10.1007/s00572-015-0670-3 26590998 82. Guo S.W. Shen Q.R. Brueck H. Effects of local nitrogen supply on water uptake of bean plants in a split root system J. Integr. Plant Biol. 2007 49 472 480 10.1111/j.1744-7909.2007.00436.x 83. Hill A.E. Shachar Hill B. Shachar Hill Y. What are aquaporins for? J. Membr. Biol. 2004 197 1 32 10.1007/s00232-003-0639-6 15014915 84. Maathuis F.J.M. Filatov V. Herzyk P. Krijger G.C. Axelsen K.B. Chen S.X. Green B.J. Li Y. Madagan K.L. Sanchez-Fernandez R. Transcriptome analysis of root transporters reveals participation of multiple gene families in the response to cation stress Plant J. 2003 35 675 692 10.1046/j.1365-313X.2003.01839.x 12969422 85. Armengaud P. Breitling R. Amtmann A. The potassium-dependent transcriptome of Arabidopsis reveals a prominent role of jasmonic acid in nutrient signaling Plant Physiol. 2004 136 2556 2576 10.1104/pp.104.046482 15347784 86. Besserer A. Burnotte E. Bienert G.P. Chevalier A.S. Errachid A. Grefen C. Blatt M.R. Chaumont F. Selective regulation of maize plasma membrane aquaporin trafficking and activity by the SNARE SYP121 Plant Cell 2012 24 3463 3481 10.1105/tpc.112.101758 22942383 87. Hachez C. Laloux T. Reinhardt H. Cavez D. Degand H. Grefen C. de Rycke R. Inze D. Blatt M.R. Russinova E. Arabidopsis SNAREs SYP61 and SYP121 coordinate the trafficking of plasma membrane aquaporin PIP2;7 to modulate the cell membrane water permeability Plant Cell 2014 26 3132 3147 10.1105/tpc.114.127159 25082856 88. Tazawa M. Sutou E. Shibasaka M. Onion root water transport sensitive to water channel and K+ channel inhibitors Plant Cell Physiol. 2001 42 28 36 10.1093/pcp/pce004 11158441 89. Sahr T. Voigt G. Paretzke H.G. Schramel P. Ernst D. Caesium-affected gene expression in Arabidopsis thaliana New Phytol. 2005 165 747 754 10.1111/j.1469-8137.2004.01282.x 15720685 90. Liu H.Y. Sun W.N. Su W.A. Tang Z.C. Co-regulation of water channels and potassium channels in rice Physiol. Plant. 2006 128 58 69 10.1111/j.1399-3054.2006.00709.x 91. Wang M. Zheng Q.S. Shen Q.R. Guo S.W. The critical role of potassium in plant stress response Int. J. Mol. Sci. 2013 14 7370 7390 10.3390/ijms14047370 23549270 92. Oddo E. Inzerillo S. La Bella F. Grisafi F. Salleo S. Nardini A. Short-term effects of potassium fertilization on the hydraulic conductance of Laurus nobilis L. Tree Physiol. 2011 31 131 138 10.1093/treephys/tpq115 21367746 93. Zwieniecki M.A. Melcher P.J. Holbrook N.M. Hydrogel control of xylem hydraulic resistance in plants Science 2001 291 1059 1062 10.1126/science.1057175 11161220 94. Galmes J. Pou A. Alsina M.M. Tomas M. Medrano H. Flexas J. Aquaporin expression in response to different water stress intensities and recovery in Richter-110 (Vitis sp.): Relationship with ecophysiological status Planta 2007 226 671 681 10.1007/s00425-007-0515-1 17447082 95. Cuéllar T. Pascaud F. Verdeil J.L. Torregrosa L. Adam-Blondon A.F. Thibaud J.B. Sentenac H. Gaillard I. A grapevine Shaker inward K+ channel activated by the calcineurin B-like calcium sensor 1-protein kinase CIPK23 network is expressed in grape berries under drought stress conditions Plant J. 2010 61 58 69 10.1111/j.1365-313X.2009.04029.x 19781051 96. Bush D.S. Calcium regulation in plant-cells and its role in signaling Annu. Rev. Plant Physiol. Plant Mol. Biol. 1995 46 95 122 10.1146/annurev.pp.46.060195.000523 97. Gilliham M. Dayod M. Hocking B.J. Xu B. Conn S.J. Kaiser B.N. Leigh R.A. Tyerman S.D. Calcium delivery and storage in plant leaves: Exploring the link with water flow J. Exp. Bot. 2011 62 2233 2250 10.1093/jxb/err111 21511913 98. Ionenko I.F. Anisimov A.V. Karimova F.G. Water transport in maize roots under the influence of mercuric chloride and water stress: A role of water channels Biol. Plant. 2006 50 74 80 10.1007/s10535-005-0077-7 99. Johansson I. Larsson C. Ek B. Kjellbom P. The major integral proteins of spinach leaf plasma membranes are putative aquaporins and are phosphorylated in response to Ca2+ and apoplastic water potential Plant Cell 1996 8 1181 1191 10.1105/tpc.8.7.1181 8768376 100. Azad A.K. Sawa Y. Ishikawa T. Shibata H. Characterization of protein phosphatase 2A acting on phosphorylated plasma membrane aquaporin of tulip petals Biosci. Biotechnol. Biochem. 2004 68 1170 1174 10.1271/bbb.68.1170 15170131 101. Steudle E. Henzler T. Water channels in plants: Do basic concepts of water transport change? J. Exp. Bot. 1995 46 1067 1076 10.1093/jxb/46.9.1067 102. Wu Y. Liu X.F. Wang W.F. Zhang S.Q. Xu B.C. Calcium regulates the cell-to-cell water flow pathway in maize roots during variable water conditions Plant Physiol. Biochem. 2012 58 212 219 10.1016/j.plaphy.2012.07.005 22841977 103. Carvajal M. CerdÁ A. MartÍNez V. Does calcium ameliorate the negative effect of NaCl on melon root water transport by regulating aquaporin activity? New Phytol. 2000 145 439 447 10.1046/j.1469-8137.2000.00593.x 104. Cabanero F.J. Martinez-Ballesta M.C. Teruel J.A. Carvajal M. New evidence about the relationship between water channel activity and calcium in salinity-stressed pepper plants Plant Cell Physiol. 2006 47 224 233 10.1093/pcp/pci239 16352698 105. Martinez-Ballesta M.C. Cabanero F. Olmos E. Periago P.M. Maurel C. Carvajal M. Two different effects of calcium on aquaporins in salinity-stressed pepper plants Planta 2008 228 15 25 10.1007/s00425-008-0714-4 18317798 106. Luu D.T. Maurel C. Aquaporins in a challenging environment: Molecular gears for adjusting plant water status Plant Cell Environ. 2005 28 85 96 10.1111/j.1365-3040.2004.01295.x 107. Vera-Estrella R. Barkla B.J. Bohnert H.J. Pantoja O. Novel regulation of aquaporins during osmotic stress Plant Physiol. 2004 135 2318 2329 10.1104/pp.104.044891 15299122 108. Yang H.M. Zhang X.Y. Tang Q.L. Wang G.X. Extracellular calcium is involved in stomatal movement through the regulation of water channels in broad bean Plant Growth Regul. 2006 50 79 83 10.1007/s10725-006-9128-0 109. Shorrocks V.M. The occurrence and correction of boron deficiency Plant Soil 1997 193 121 148 10.1023/A:1004216126069 110. Blevins D.G. Lukaszewski K.M. Boron in plant structure and function Annu. Rev. Plant Physiol. Plant Mol. Biol. 1998 49 481 500 10.1146/annurev.arplant.49.1.481 15012243 111. Dordas C. Chrispeels M.J. Brown P.H. Permeability and channel-mediated transport of boric acid across membrane vesicles isolated from squash roots Plant Physiol. 2000 124 1349 1361 10.1104/pp.124.3.1349 11080310 112. Fitzpatrick K.L. Reid R.J. The involvement of aquaglyceroporins in transport of boron in barley roots Plant Cell Environ. 2009 32 1357 1365 10.1111/j.1365-3040.2009.02003.x 19552667 113. Tanaka M. Wallace I.S. Takano J. Roberts D.M. Fujiwara T. NIP6;1 is a boric acid channel for preferential transport of boron to growing shoot tissues in Arabidopsis Plant Cell 2008 20 2860 2875 10.1105/tpc.108.058628 18952773 114. Schnurbusch T. Hayes J. Hrmova M. Baumann U. Ramesh S.A. Tyerman S.D. Langridge P. Sutton T. Boron toxicity tolerance in barley through reduced expression of the multifunctional aquaporin HvNIP2;1 Plant Physiol. 2010 153 1706 1715 10.1104/pp.110.158832 20581256 115. Hanaoka H. Fujiwara T. Channel-mediated boron transport in rice Plant Cell Physiol. 2007 48 S227 116. Pang Y.Q. Li L.J. Ren F. Lu P.L. Wei P.C. Cai J.H. Xin L.G. Zhang J.A. Chen J. Wang X.C. Overexpression of the tonoplast aquaporin AtTIP5;1 conferred tolerance to boron toxicity in Arabidopsis J. Genet. Genom. 2010 37 389 397 10.1016/S1673-8527(09)60057-6 117. Mosa K.A. Kumar K. Chhikara S. Musante C. White J.C. Dhankher O.P. Enhanced boron tolerance in plants mediated by bidirectional transport through plasma membrane intrinsic proteins Sci. Rep. 2016 6 10.1038/srep21640 26902738 118. Kumar K. Mosa K.A. Chhikara S. Musante C. White J.C. Dhankher O.P. Two rice plasma membrane intrinsic proteins, OsPIP2;4 and OsPIP2;7, are involved in transport and providing tolerance to boron toxicity Planta 2014 239 187 198 10.1007/s00425-013-1969-y 24142111 119. Epstein E. Silicon Annu. Rev. Plant Physiol. Plant Mol. Biol. 1999 50 641 664 10.1146/annurev.arplant.50.1.641 15012222 120. Richmond K.E. Sussman M. Got silicon? The non-essential beneficial plant nutrient Curr. Opin. Plant Biol. 2003 6 268 272 10.1016/S1369-5266(03)00041-4 12753977 121. Ma J.F. Role of silicon in enhancing the resistance of plants to biotic and abiotic stresses Soil Sci. Plant Nutr. 2004 50 11 18 10.1080/00380768.2004.10408447 122. Zhu Y.X. Xu X.B. Hu Y.H. Han W.H. Yin J.L. Li H.L. Gong H.J. Silicon improves salt tolerance by increasing root water uptake in Cucumis sativus L. Plant Cell Rep. 2015 34 1629 1646 10.1007/s00299-015-1814-9 26021845 123. Liu P. Yin L.N. Wang S.W. Zhang M.J. Deng X.P. Zhang S.Q. Tanaka K. Enhanced root hydraulic conductance by aquaporin regulation accounts for silicon alleviated salt-induced osmotic stress in Sorghum bicolor L. Environ. Exp. Bot. 2015 111 42 51 10.1016/j.envexpbot.2014.10.006 124. Guerriero G. Hausman J.F. Legay S. Silicon and the plant extracellular matrix Front. Plant Sci. 2016 7 10.3389/fpls.2016.00463 27148294 125. Ma J.F. Yamaji N. Mitani N. Xu X.Y. Su Y.H. McGrath S.P. Zhao F.J. Transporters of arsenite in rice and their role in arsenic accumulation in rice grain Proc. Natl. Acad. Sci. USA 2008 105 9931 9935 10.1073/pnas.0802361105 18626020 126. Ma J.F. Yamaji N. Mitani N. Tamai K. Konishi S. Fujiwara T. Katsuhara M. Yano M. An efflux transporter of silicon in rice Nature 2007 448 209 212 10.1038/nature05964 17625566 127. Dean R.M. Rivers R.L. Zeidel M.L. Roberts D.M. Purification and functional reconstitution of soybean nodulin 26. An aquaporin with water and glycerol transport properties Biochemistry 1999 38 347 353 10.1021/bi982110c 9890916 128. Zhao X.Q. Mitani N. Yamaji N. Shen R.F. Ma J.F. Involvement of silicon influx transporter OsNIP2; 1 in selenite uptake in rice Plant Physiol. 2010 153 1871 1877 10.1104/pp.110.157867 20498338 129. Yamaji N. Mitatni N. Ma J.F. A transporter regulating silicon distribution in rice shoots Plant Cell 2008 20 1381 1389 10.1105/tpc.108.059311 18515498

PMC005xxxxxx/PMC5000628.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081230ijms-17-01230ArticleIs Upregulation of Aquaporin 4-M1 Isoform Responsible for the Loss of Typical Orthogonal Arrays of Particles in Astrocytomas? Fallier-Becker Petra 1Nieser Maike 1*Wenzel Ulrike 1Ritz Rainer 2Noell Susan 3Ishibashi Kenichi Academic Editor1 Institute of Pathology and Neuropathology, University Hospital of Tuebingen, Liebermeisterstr. 8, 72076 Tuebingen, Germany; [email protected] (P.F.-B.); [email protected] (U.W.)2 Department of Neurosurgery, Schwarzwald-Baar Klinikum, Klinikstr. 11, 78052 Villingen-Schwenningen, Germany; [email protected] Department of Neurosurgery, University Hospital of Tuebingen, Hoppe-Seyler-Str. 3, 72076 Tuebingen, Germany; [email protected]* Correspondence: [email protected]; Tel.: +49-7071-29-8441029 7 2016 8 2016 17 8 123013 6 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The astrocytic endfoot membranes of the healthy blood-brain barrier—contacting the capillary—are covered with a large number of the water channel aquaporin 4 (AQP4). They form orthogonal arrays of particles (OAPs), which consist of AQP4 isoform M1 and M23. Under pathologic conditions, AQP4 is distributed over the whole cell and no or only small OAPs are found. From cell culture experiments, it is known that cells transfected only with AQP4-M1 do not form OAPs or only small ones. We hypothesized that in astrocytomas the situation may be comparable to the in vitro experiments expecting an upregulation of AQP4-M1. Quantitative Real-time PCR (qRT-PCR) of different graded astrocytomas revealed an upregulation of both isoforms AQP4 M1 and M23 in all astrocytomas investigated. In freeze fracture replicas of low-grade malignancy astrocytomas, more OAPs than in high-grade malignancy astrocytomas were found. In vitro, cultured glioma cells did not express AQP4, whereas healthy astrocytes revealed a slight upregulation of both isoforms and only a few OAPs in freeze fracture analysis. Taken together, we found a correlation between the decrease of OAPs and increasing grade of malignancy of astrocytomas but this was not consistent with an upregulation of AQP4-M1 in relation to AQP4 M23. low-grade astrocytomahigh-grade astrocytomamalignancyaquaporin ==== Body 1. Introduction Aquaporin 4 (AQP4) is the main water channel in the mammalian brain. It is morphologically assembled in square arrays called orthogonal arrays of particles (OAPs) [1,2,3] localized in superficial and perivascular astrocytic end foot membranes at the blood-brain barrier (BBB). In parenchymal membranes where the astrocyte loses contact to the basal lamina, only a small number of OAPs is found, indicating the polarization of astrocytes [4]. This polarization disappears in astrocytomas. Here, AQP4 protein is strongly upregulated compared to healthy brain tissue as was shown by Saadoun et al. by immunohistological stainings [5]. However, Noell et al. [6] reported that freeze fracture replicas of glioblastomas revealed no or only small OAPs. Western blot analyses of AQP4 expressing cells and tissues revealed two bands for AQP4 that refer to two different AQP4 isoforms: AQP4-M23 and AQP4-M1 with M1 being 22 amino acids (AAs) longer than M23 [7,8]. It is known from in vitro transfection experiments that cells transfected with AQP4-M23-isoform form large lattices of OAPs whereas in M1-transfected cells no or only small OAPs were observed [9,10]. Consequently, we hypothesized that in astrocytomas, where no typical OAPs have been found in freeze fracture replicas, AQP4-M1 might be upregulated in relation to AQP4-M23. In this work we examined tissues from patients with low-grade to high-grade astrocytomas (World Health Organization (WHO) grade II to IV) asking the question whether or not the AQP4-M1 mRNA is upregulated compared to AQP4-M23 mRNA. Furthermore, we performed freeze fracture analysis of these glioma tissues in order to find out if there is a correlation between the number of OAPs and the grade of malignancy. In addition, cultured glioma cells were analyzed in the same way and compared to normal murine astrocytes in vitro. 2. Results 2.1. Determination of Orthogonal Arrays of Particles (OAPs) from Low-Grade to High-Grade Malignancies Using freeze fracture techniques, we analyzed replicas of human astrocytomas from low-grade to high-grade malignancy in order to find out whether or not there is a relationship between the number of typical OAPs and the grade of malignancy of these astrocytomas. Figure 1 shows the freeze fracture replicas of the examined astrocytoma tissues: low-grade astrocytomas (WHO grade II; Figure 1A) revealed membranes with OAPs. The astrocytoma WHO grade III (Figure 1B) showed a few OAPs. IDH1 (isocitrate dehydrogenase 1 (NADP(+))-negative (non-mutated) glioblastomas (Figure 1C) formed only few and sometimes misshapen OAPs or membranes lacking any OAPs. Quantification was performed by counting OAPs of all astrocytoma tissues per 4 µm2. The results are shown in Figure 2 indicating that astrocytomas WHO graded II harbor significantly more OAPs than astrocytomas WHO grade III (p = 0.028) and glioblastomas (WHO graded IV; p < 0.001). These results point out that there is a correlation between the formation of OAPs and the grade of malignancy in glial tumors of the human brain. 2.2. Determination of Aquaporin 4 (AQP4)-M1 and -M23 mRNA in Low-Grade and High-Grade Astrocytomas We know from cell culture experiments that cells transfected with AQP4-M23 isoform are forming large OAP lattices whereas M1-transfected cells show no or only small OAPs [6,7]. Consequently, we performed qRT-PCR to investigate the expression of the two AQP4 isoforms in our patients’ tumor samples. AQP4-M1 expression was upregulated in all astrocytomas, without any correlation to the grade of malignancy (Figure 3). The average fold change of AQP4-M1 was 12.41 compared to a normal brain. Surprisingly, the average fold change of AQP4-M23 was 17.93 (Figure 4), indicating that AQP4-M23 was even more upregulated than AQP4-M1. The M23/M1 ratios ranged between 0.81 and 5.58, with an average ratio of 1.64, indicating that most gliomas express the M23 isoform over 1.5-fold more than M1. When comparing the different astrocytoma grades, astrocytoma II and III depict ratios of 1.5 and 1.14, respectively. The average ratio for astrocytomas IV IDH1- is 1.94. It seems that AQP4-M23/M1 ratio is higher in the more malignant IDH1-negative glioblastomas than in the IDH1-positive astrocytomas. 2.3. Determination of AQP4-M1 and -M23 mRNA in Cultures of Normal Astrocytes and Glioma Cells Cell cultures of murine astrocytes from healthy brain and human glioma cells (TuGlio 25 and U373) were processed for qRT-PCR to compare their expression of AQP4 isoforms to the patients’ tumor tissue. Cultured mouse brain astrocytes revealed a slight upregulation of both AQP4-M1 (fold change 4.35) and AQP4-M23X (fold change 4.66, Figure 5) compared to a normal brain. Freeze fracture replicas of mouse astrocytes showed several small sized OAPs (Figure 6). In U373 and TuGlio25 tumor cells, however, both isoforms were expressed at a very low level (average AQP4-M1 crossing point (CP) values of 32.04 and 34.54 for TuGlio25 and U373, and average AQP4-M23 CP values of 35 and 32.89, respectively) indicating that AQP4 is not expressed in these human tumor cell cultures. Accordingly, no OAPs were formed. 3. Discussion In this work, we furnished proof that the number of OAPs in human astrocytomas decreased from low-grade to high-grade malignancy although the amount of AQP4 protein was increasing [11]. We hypothesized that there might be a connection to the in vitro transfection results of Furman et al. [9] who found no or only small OAPs in AQP4-M1 transfected cell cultures. From these findings, we concluded that astrocytomas revealing no typical OAPs should express more AQP4-M1 than AQP4-M23 compared to the normal brain. However, not only AQP4-M1 but alsoAQP4-M23 was upregulated, even to a slightly higher degree than AQP4-M1. Astrocytomas WHO graded II and III depict average M23/M1 ratios of 1.5 and 1.14, whereas the average M23/M1 ratio of glioblastomas is elevated to 1.94. In the healthy brain, however, the M23/M1 is at least 3 [9]. In addition, Jin et al. report that low M23/M1 ratios yield small OAPs, whereas high M23/M1 ratios lead to large OAPs [12]. Our results show that in low-grade and high-grade astrocytomas more AQP4-M23 isoform than AQP4-M1 was expressed despite the decreasing number of OAPs compared to a normal brain. Investigating AQP4-M1 and -M23 mRNA expression of cell cultures of primary tumor cells (TuGlio25) and a tumor cell line (U373) revealed very low AQP4 mRNA expression levels. These results are consistent with the immunohistological investigations of C6 rat glioma cell cultures, which yielded no positive staining for AQP4 [12]. In contrast, in cultures of healthy murine astrocytes, AQP4–M23 and –M1 mRNA were expressed. Here we found the M23/M1 ratio of 1.07 which is comparable to M23/M1 ratios of astrocytomas WHO grade III. In addition, the formation of OAPs in healthy astrocyte cultures was also comparable to the OAPs found in astrocytomas WHO grade II. Taken together we conclude that loss of OAPs in astrocytomas depends on other factors than the upregulation of AQP4-M1 alone. Former western blot investigations of aquaporin 4 in glioblastoma tissues [6] and rat brain gliomas [13] always yielded the same results: the AQP4-M23 bands were thicker than the M1 bands independent of the formation of OAPs in the appropriate membranes. It seems that the formation of OAPs underlies other mechanisms than change of the M23/M1 ratio. As reported previously, the microenvironment of brain tumors differs from that of normal brains [6]. Our previous findings revealing a correlation between the presence of agrin in the extracellular matrix and the formation of OAPs in glioblastoma and agrin knock-out mice [6,14] suggest that this is also the case in astrocytomas and might be the reason for the decreasing number of OAPs with increasing malignancy of the astrocytomas. In addition to agrin, dystroglycan is also fundamental in forming OAPs. It links agrin in the extracellular matrix via utrophin to alpha synthrophin and to AQP4 [15]. Although investigation of subependymomas reveals also no expression of agrin and dystroglycan and hence no formation of OAPs, these benign tumors are not comparable to the infiltrating astrocytomas [16]. In addition, the activity of matrixmetalloproteinases MMP2, 9, and 3 play a pivotal role in this scenario: if agrin is cleaved by MMP3 [17]—such as in glioblastoma—no typical OAPs are found and if MMP2 and 9 cleave dystroglycan [18]—such as in glioblastoma—no OAPs are formed as well. All these in vivo investigations suggest that the microenvironment in the brain is more important for the formation of OAPs than the M23/M1 relation (Table 1). Furthermore, in vitro experiments showed that murine astrocyte cultures form more OAPs when growing on agrin-coated surfaces or in agrin-containing culture medium [19,20]. It still remains unclear why AQP4 form OAPs and what their role might be. It has been suggested that AQP4 water channels might have a better adhesion to the neighbored membrane if assembled in square arrays [8,21]; others propose a more effective water exchange when AQP4-molecules are packed in OAPs [22]. The isoform AQP4-M23 is suggested to be the faster water channel because it works more effectively in OAPs [23]. 4. Materials and Methods 4.1. Tissue Tumor tissues from 22 patients with astrocytoma graded WHO II, III, and IV (Table 2) were analyzed using freeze fracture techniques and qRT-PCR. The ethics committee of the medical faculty of the Eberhard-Karls University of Tuebingen approved the project and patients’ consent procedures. The ethics committee waived the need for consent (project no. 663/2013BO2). All patients were treated in the Department of Neurosurgery Tuebingen between 2013 and 2014 (Table 2). 4.2. Cell Culture Human primary glioma cells were isolated enzymatically from a patient’s glioblastoma. Briefly, resected surgical material was transported under sterile conditions in a 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPEs) buffered medium (Thermo Fisher Scientific, Waltham, MA, USA) and cells were isolated using trypsin (Sigma, Taufkirchen, Germany). After centrifugation, culture medium was added to the cell pellet and cell suspension was pipetted on a cell strainer (70 µm) to get single cells that were seeded in culture flasks and cultured in a humidified incubator at 37 °C and 5% CO2 under standard conditions with Dulbecco's modified Eagle’s medium (DMEM) (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal calf serum (Thermo Fisher Scientific, Waltham, MA, USA), 1% penicillin and streptomycin (Sigma-Aldrich, St. Louis, MO, USA). Culture medium was changed every two days. Ulrike Naumann, Hertie Institute of Clinical Research, Tuebingen, Germany, kindly provided U373 cells. They were cultured as described above. Murine cortex astrocytes were isolated from 5 day old mice as described above and cultured under standard conditions [24,25]. When cells reached confluence they were processed for qRT-PCR and freeze fracture analysis. 4.3. Freeze Fracture Technique and Determination of OAP Densities Tissues and cells were processed for freeze fracture analysis as described before [6]. 4.4. RNA Extraction Cryo tumor specimens of the above-mentioned astrocytomas were gathered from the surgical pathology files of the Department of Neuropathology of the Institute of Pathology of Tuebingen. RNA of the samples and the cell cultures was extracted manually with the RNeasy Mini Kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions. RNA quantification was performed with the NanoDrop ND-2000 (Thermo Fisher Scientific) spectrophotometer. Human Brain, Cerebral Cortex Total RNA, and Mouse Brain Total RNA (Clontech, Mountain View, CA, USA) served as healthy controls for qRT-PCR. 4.5. cDNA Synthesis Intron-spanning oligonucleotides for human HPRT1 (housekeeping gene), AQP4 M1, and AQP4 M23, as well as murine Hprt1, Aqp4 M1, and Aqp4 M23X [26] were designed using the program Primer3Plus (http://primer3plus.com/cgi-bin/dev/primer3plus.cgi). In Table 3 and Table 4 the primer sequences and the product sizes are shown. Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) was used to check for the specificity of the primers. 4.6. Quantitative Real Time-PCR (qRT-PCR) Total RNA (1 μg) was reversely transcribed using the High-Capacity cDNA RT Kit (Thermo Fisher Scientific) in combination with RNA Inhibitor (Thermo Fisher Scientific). The qRT-PCR mix included 10 μL Real-Time SYBR Green PCR master mix, 1 μL (20 ng) diluted reverse transcription product, 2 μL each of Primer (Table 2 and Table 3) and 7 μL DNase/RNase free water. The PCR was carried out under following conditions: 95 °C for 15 min followed by 40 cycles of 94 °C for 15 s, 55 °C for 30 s and 70 °C for 30 s in a LightCycler 480 II (Roche, Basel, Switzerland). Normal brain controls as well as a negative water control were included in every run. All samples were run in triplicates. By running the PCR for 40 cycles, the maximal crossing point (CP) value that could be achieved by the analysis was 35. The CP value is the number of PCR cycles at which a constant fluorescence level is achieved. CP values between 15 and 25 indicate strong expression, 25–30 medium expression, and >30 weak expression. To assess the specificity of the amplified PCR product a melting curve analysis was carried out. 4.7. Data Analysis and Statistics The average raw CP values (median of triplicates) were imported into Microsoft Excel (Version, Microsoft Corp., Seattle, WA, USA). The AQP4-M1 and -M23 expression were analyzed using the comparative ΔCP method, where ΔCP = CP candidate target − CP reference RNA. HPRT1 was used as endogenous reference gene for the analysis. ΔΔCP = ΔCP sample − ΔCP normal brain control, relative expression = 2−ΔΔCP, fold change: if relative expression >1 then fold change ≙ relative expression, if relative expression <1 then fold change = −1/relative expression. M23/M1 ratios were determined by simply dividing the M23 fold changes through the M1 fold changes. For the evaluation of the OAP densities and the AQP4-M1 and -M23 fold changes of the astrocytomas two-tailed unpaired t-tests were performed using GraphPad software (Version 4, San Diego, CA, USA). The scatter plot for the OAPs was also achieved by using this software. 5. Conclusions Taken together we were able to demonstrate a significant correlation between the decreasing number of OAPs and the grade of malignancy in human astrocytomas. Furthermore, we found an upregulation of AQP4-M1 as well as an upregulation of -M23 mRNA in all astrocytomas but the M23/M1 ratio differed from 1.14 to 1.5 in low-grade astrocytomas to 1.94 in glioblastomas. These results demand further investigations to elucidate the question whether or not they will yield an impact on the development of new therapeutics. Acknowledgments We gratefully acknowledge Ria Knittel for preparing the freeze fracture replicas. We are also grateful for Yeliz Donat’s excellent technical assistance. Author Contributions Petra Fallier-Becker and Susan Noell conceived and designed the experiments. Petra Fallier-Becker and Maike Nieser performed the experiments. Petra Fallier-Becker, Susan Noell, Maike Nieser and Ulrike Wenzel analyzed the data. Petra Fallier-Becker, Susan Noell and Maike Nieser wrote the paper. Rainer Ritz provided tumor samples, supervised ongoing experiments and critically revised the manuscript for important intellectual content. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Freeze fracture replicas of low-grade to high-grade astrocytomas. (A) Astrocytoma World Health Organization (WHO) grade II showing orthogonal arrays of particles (OAPs) (encircled); (B) Astrocytoma WHO grade III showing less OAPs than in A (encircled); (C) IDH1 (isocitrate dehydrogenase 1 (NADP(+))-negative glioblastoma: No typical OAPs are found. Bar: 250 nm. Figure 2 Number of OAPs counted per 4 µm2 in astrocytoma WHO grade II, III and IV IDH1-negative glioblastoma. The number of OAPs in astrocytoma WHO grade II is significant higher than in astrocytoma WHO grade III (p = 0.028) and glioblastoma (p = 0.0006). * p ≤ 0.05, *** p ≤ 0.001. Figure 3 Aquaporin 4 (AQP4)-M1 fold changes of astrocytomas. The average fold change is 12.41. No significant differences between the WHO grade groups were detected. Figure 4 AQP4-M23 fold changes of astrocytomas. The average fold change is 17.93. No significant differences between the WHO grade groups were detected. Figure 5 AQP4 M1 and M23X fold changes of primary astrocytes. Both isoforms depict 4–5-fold expression compared to the normal mouse brain. Figure 6 Freeze fracture replica of a healthy murine astrocyte in culture showing OAPs encircled. ijms-17-01230-t001_Table 1Table 1 Comparison of typical features between healthy brain and glioblastoma. Protein Expression/Protein Complex Healthy Brain Glioblastoma Agrin + − Dystroglycan + − Active MMP3 − + Active MMP2/9 − + OAPs + − ijms-17-01230-t002_Table 2Table 2 World Health Organization (WHO) grading and IDH1 (isocitrate dehydrogenase 1 (NADP(+)) status of patients. Patient Number Astrocytoma WHO Grade IDH1 1 II IDH1 (R132H) mutation (IDH1+) 2 II IDH1 R132H) mutation (IDH1+) 3 II IDH1 (R132H) mutation (IDH1+) 4 II IDH1 (R132H) mutation (IDH1+) 5 II IDH1 (R132H) mutation (IDH1+) 6 II IDH1 (R132H) mutation (IDH1+) 7 III IDH1 (R132H) mutation (IDH1+) 8 III IDH1 (R132H) mutation (IDH1+) 9 III IDH1 (R132H) mutation (IDH1+) 10 III IDH1 (R132H) mutation (IDH1+) 11 III IDH1 (R132H) mutation (IDH1+) 12 IV IDH1 wildtype (IDH1−) 13 IV IDH1 wildtype (IDH1−) 14 IV IDH1 wildtype (IDH1−) 15 IV IDH1 wildtype (IDH1−) 16 IV IDH1 wildtype (IDH1−) 17 IV IDH1 wildtype (IDH1−) 18 IV IDH1 wildtype (IDH1−) 19 IV IDH1 wildtype (IDH1−) 20 IV IDH1 wildtype (IDH1−) 21 IV IDH1 wildtype (IDH1−) 22 IV IDH1 wildtype (IDH1−) ijms-17-01230-t003_Table 3Table 3 Primer (Homo sapiens). Primer Sequence Product Size HPRT1 Ex6 for HPRT1 Ex7 rev TGACACTGGCAAAACAATGC TTCGTGGGGTCCTTTTCACC 101 bp AQP4 M1 Ex1 for AQP4 M1 Ex2 rev GGGGAAGGCATGAGTGACAG AAAGCTTGAGTCCAGACCCC 110 bp AQP4 M23 Ex1 for AQP4 M23 Ex2 rev TCTCTTTTCAGTAAGTGTGGACCT CATGGCCAGAAATTCCGCTG 114 bp ijms-17-01230-t004_Table 4Table 4 Primer (Mus musculus). Primer Sequence Product Size HPRT Ex6 Mus for HPRT Ex7 Mus rev CAAACTTTGCTTTCCCTGGT GGCCTGTATCCAACACTTCG 91 bp AQP4 M1 Mus Ex1 for AQP4 M1 Mus Ex2 rev AGGGAAGGCATGAGTGACAG GACTCCTTTGAAAGCCACCA 96 bp AQP4 M23X Mus for AQP4 M23X Mus rev TATGGTTCACGGGTTTGGAT CCCTTTGTCACCTGCTCATT 139 bp ==== Refs References 1. Rash J.E. Yasumura T. Hudson C.S. Agre P. Nielsen S. Direct immunogold labeling of aquaporin-4 in square arrays of astrocyte and ependymocyte plasma membranes in rat brain and spinal cord Proc. Natl. Acad. Sci. USA 1998 95 11981 11986 10.1073/pnas.95.20.11981 9751776 2. Wolburg H. Wolburg-Buchholz K. Fallier-Becker P. Noell S. Mack A.F. Structure and functions of aquaporin-4-based orthogonal arrays of particles Int. Rev. Cell Mol. Biol. 2011 287 1 41 21414585 3. Nicchia G.P. Rossi A. Mola M.G. Pisani F. Stigliano C. Basco D. Mastrototaro M. Svelto M. Frigeri A. Higher order structure of aquaporin-4 Neuroscience 2010 168 903 914 10.1016/j.neuroscience.2010.02.008 20153404 4. Rohlmann A. Gocht A. Wolburg H. Reactive astrocytes in myelin-deficient rat optic-nerve reveal an altered distribution of orthogonal arrays of particles (OAP) Glia 1992 5 259 268 10.1002/glia.440050404 1534070 5. Saadoun S. Papadopoulos M.C. Davies D.C. Krishna S. Bell B.A. Aquaporin-4 expression is increased in oedematous human brain tumours J. Neurol. Neurosurg. Psychiatry 2002 72 262 265 10.1136/jnnp.72.2.262 11796780 6. Noell S. Wolburg-Buchholz K. Mack A.F. Ritz R. Tatagiba M. Beschorner R. Wolburg H. Fallier-Becker P. Dynamics of expression patterns of AQP4, dystroglycan, agrin and matrix metalloproteinases in human glioblastoma Cell Tissue Res. 2012 347 429 441 10.1007/s00441-011-1321-4 22307776 7. Neely J.D. Christensen B.M. Nielsen S. Agre P. Heterotetrameric composition of aquaporin-4 water channels Biochemistry 1999 38 11156 11163 10.1021/bi990941s 10460172 8. Smith A.J. Jin B.J. Ratelade J. Verkman A.S. Aggregation state determines the localization and function of M1- and M23-aquaporin-4 in astrocytes J. Cell Biol. 2014 204 559 573 10.1083/jcb.201308118 24515349 9. Furman C.S. Gorelick-Feldman D.A. Davidson K.G. Yasumura T. Neely J.D. Agre P. Rash J.E. Aquaporin-4 square array assembly: Opposing actions of M1 and M23 isoforms Proc. Natl. Acad. Sci. USA 2003 100 13609 13614 10.1073/pnas.2235843100 14597700 10. Fryer J.P. Lennon V.A. Pittock S.J. Jenkins S.M. Fallier-Becker P. Clardy S.L. Horta E. Jedynak E.A. Lucchinetti C.F. Shuster E.A. Aqp4 autoantibody assay performance in clinical laboratory service Neuroimmunol. Neuroinflamm. 2014 1 1e11 10.1212/NXI.0000000000000011 25340055 11. Warth A. Simon P. Capper D. Goeppert B. Tabatabai G. Herzog H. Dietz K. Stubenvoll F. Ajaaj R. Becker R. Expression pattern of the water channel aquaporin-4 in human gliomas is associated with blood-brain barrier disturbance but not with patient survival J. Neurosci. Res. 2007 85 1336 1346 10.1002/jnr.21224 17335082 12. Jin B.J. Rossi A. Verkman A.S. Model of aquaporin-4 supramolecular assembly in orthogonal arrays based on heterotetrameric association of M1-M23 isoforms Biophys. J. 2011 100 2936 2945 10.1016/j.bpj.2011.05.012 21689527 13. Noell S. Ritz R. Wolburg-Buchholz K. Wolburg H. Fallier-Becker P. An allograft glioma model reveals the dependence of aquaporin-4 expression on the brain microenvironment PLoS ONE 2012 7 1230 10.1371/journal.pone.0036555 22590566 14. Noell S. Fallier-Becker P. Deutsch U. Mack A.F. Wolburg H. Agrin defines polarized distribution of orthogonal arrays of particles in astrocytes Cell Tissue Res. 2009 337 185 195 10.1007/s00441-009-0812-z 19449033 15. Noell S. Wolburg-Buchholz K. Mack A.F. Beedle A.M. Satz J.S. Campbell K.P. Wolburg H. Fallier-Becker P. Evidence for a role of dystroglycan regulating the membrane architecture of astroglial endfeet Eur. J. Neurosci. 2011 33 2179 2186 10.1111/j.1460-9568.2011.07688.x 21501259 16. Noell S. Fallier-Becker P. Mack A.F. Hoffmeister M. Beschorner R. Ritz R. Water channels aquaporin 4 and -1 expression in subependymoma depends on the localization of the tumors PLoS ONE 2015 10 1230 10.1371/journal.pone.0131367 26115524 17. Sole S. Petegnief V. Gorina R. Chamorro A. Planas A.M. Activation of matrix metalloproteinase-3 and agrin cleavage in cerebral ischemia/reperfusion J. Neuropathol. Exp. Neurol. 2004 63 338 349 10.1093/jnen/63.4.338 15099024 18. Agrawal S. Anderson P. Durbeej M. van Rooijen N. Ivars F. Opdenakker G. Sorokin L.M. Dystroglycan is selectively cleaved at the parenchymal basement membrane at sites of leukocyte extravasation in experimental autoimmune encephalomyelitis J. Exp. Med. 2006 203 1007 1019 10.1084/jem.20051342 16585265 19. Noell S. Fallier-Becker P. Beyer C. Kroger S. Mack A.F. Wolburg H. Effects of agrin on the expression and distribution of the water channel protein aquaporin-4 and volume regulation in cultured astrocytes Eur. J. Neurosci. 2007 26 2109 2118 10.1111/j.1460-9568.2007.05850.x 17927773 20. Fallier-Becker P. Sperveslage J. Wolburg H. Noell S. The impact of agrin on the formation of orthogonal arrays of particles in cultured astrocytes from wild-type and agrin-null mice Brain Res. 2011 1367 2 12 10.1016/j.brainres.2010.09.092 20920487 21. Crane J.M. Tajima M. Verkman A.S. Live-cell imaging of aquaporin-4 diffusion and interactions in orthogonal arrays of particles Neuroscience 2010 168 892 902 10.1016/j.neuroscience.2009.08.034 19699275 22. Tani K. Mitsuma T. Hiroaki Y. Kamegawa A. Nishikawa K. Tanimura Y. Fujiyoshi Y. Mechanism of aquaporin-4’s fast and highly selective water conduction and proton exclusion J. Mol. Biol. 2009 389 694 706 10.1016/j.jmb.2009.04.049 19406128 23. Fenton R.A. Moeller H.B. Zelenina M. Snaebjornsson M.T. Holen T. MacAulay N. Differential water permeability and regulation of three aquaporin 4 isoforms Cell. Mol. Life Sci. 2010 67 829 840 10.1007/s00018-009-0218-9 20013023 24. Beyer C. Raab H. Nongenomic effects of oestrogen: Embryonic mouse midbrain neurones respond with a rapid release of calcium from intracellular stores Eur. J. Neurosci. 1998 10 255 262 10.1046/j.1460-9568.1998.00045.x 9753134 25. Ivanova T. Karolczak M. Beyer C. Estradiol stimulates GDNF expression in developing hypothalamic neurons Endocrinology 2002 143 3175 3178 10.1210/endo.143.8.8794 12130584 26. Zelenin S. Gunnarson E. Alikina T. Bondar A. Aperia A. Identification of a new form of AQP4 mRNA that is developmentally expressed in mouse brain Pediatr. Res. 2000 48 335 339 10.1203/00006450-200009000-00012 10960499

PMC005xxxxxx/PMC5000629.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081231ijms-17-01231CommunicationSMA Human iPSC-Derived Motor Neurons Show Perturbed Differentiation and Reduced miR-335-5p Expression Murdocca Michela 1Ciafrè Silvia Anna 1*Spitalieri Paola 1Talarico Rosa Valentina 1Sanchez Massimo 2Novelli Giuseppe 1Sangiuolo Federica 1*Pichler Martin Academic Editor1 Department of Biomedicine and Prevention, University of Rome Tor Vergata, Via Montpellier, 1, 00133 Rome, Italy; [email protected] (M.M.); [email protected] (P.S.); [email protected] (R.V.T.); [email protected] (G.N.)2 Department of Cell Biology and Neurosciences, Istituto Superiore di Sanità, 00161 Rome, Italy; [email protected]* Correspondence: [email protected] (S.A.C.); [email protected] (F.S.); Tel.: +39-06-7259-6059 (S.A.C.); +39-06-7259-6154 (F.S.)30 7 2016 8 2016 17 8 123124 6 2016 18 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Spinal Muscular Atrophy (SMA) is a neuromuscular disease caused by mutations in the Survival Motor Neuron 1 gene, resulting in very low levels of functional Survival of Motor Neuron (SMN) protein. SMA human induced Pluripotent Stem Cells (hiPSCs) represent a useful and valid model for the study of the disorder, as they provide in vitro the target cells. MicroRNAs (miRNAs) are often reported as playing a key role in regulating neuronal differentiation and fate specification. In this study SMA hiPSCs have been differentiated towards early motor neurons and their molecular and immunocytochemical profile were compared to those of wild type cells. Cell cycle proliferation was also evaluated by fluorescence-activated cell sorting (FACS). SMA hiPSCs showed an increased proliferation rate and also higher levels of stem cell markers. Moreover; when differentiated towards early motor neurons they expressed lower levels of NCAM and MN specific markers. The expression of miR-335-5p; already identified to control self-renewal or differentiation of mouse embryonic stem cells (mESCs); resulted to be reduced during the early steps of differentiation of SMA hiPSCs compared to wild type cells. These results suggest that we should speculate a role of this miRNA both in stemness characteristic and in differentiation efficiency of these cells. SMAearly motor neuronmiRNAhiPSCs ==== Body 1. Introduction Spinal Muscular Atrophy (SMA) is a neuromuscular disease, recessively inherited, representing the primary genetic reason of infantile mortality. SMA is caused by mutations in the survival motor neuron 1 (SMN1; MIM#600354) gene, resulting in very low levels of functional SMN protein [1,2]. The disease is characterized by the degeneration of lower α-motor neurons (MN), progressive muscle weakness and paralysis [3]. Although SMA is caused by reduced levels of a ubiquitous protein, the higher sensitivity of motor neurons to SMN deficiency still represents an unsolved paradox. Moreover, no effective treatment has been developed for this motor neuron disease. In recent years, a great interest has been growing for induced pluripotent stem cells (iPSCs) as a model for the study of genetic diseases, and in particular for those characterized by the inaccessibility of the disease target cells in patients. SMA, affecting MNs, is one of these. Human iPSCs from SMA patients have already been described [4,5], closely resembling an in vitro model for the pathology. Ebert and collaborators showed that hiPSCs from type I SMA patients are able to differentiate into motor neurons lacking SMN1 expression, and undergo a selective death over time. Recent findings have placed microRNAs (miRNAs) in the midst of gene regulatory networks involved in neural induction, neuronal differentiation and fate specification [6]. We have recently shown that miR-335-5p, already identified to control self-renewal or differentiation of mESCs [7], is differentially expressed in SMN∆7 SMA versus wild type (WT) neural progenitor cells derived from E13.5 mice spinal cords, and probably correlated to the increased proliferation activity observed in SMA cells [8]. In this study, we analyze the molecular and phenotypic characteristics, in terms of gene expression and cell cycle proliferation, of SMA and wild type hiPSCs during their differentiation towards early motor neurons, and we confirm the under expression of miR-335-5p in SMA cells, associated to a reduced expression of early MN markers. 2. Results As we previously described, SMA neural progenitor cells obtained from spinal cords of SMA embryo mice showed an increased cell proliferation rate compared to wild type ones. Thus, we decided to characterize hiPSCs by evaluating the progression along the cell cycle of SMA and WT derived hiPSCs after 15, 60 and 180 min of BrdU incubation (Figure 1A). Analysis has been performed by flow cytometry in BrdU labelled cells. Although the results show an elevated proliferation proficiency of both SMA- and WT-derived hiPSCs, the former are characterized by a significantly increased number of cells entered in S phase (67.91% ± 0.66% vs. 56.38% ± 1.68% at 180 min), combined with significantly reduced number of cells in G2/M phase (9.57% ± 0.08% vs. 21.60% ± 0.73%), most likely due to a faster exit from mitosis (Figure 1B). In parallel, we analyzed by RT-qPCR the expression of three transcription regulators, NANOG, OCT4 and SOX2, essential for maintaining self-renewal of stem cells. As shown in Figure 1C, the SMA hiPSCs express significantly higher levels of all transcripts compared to wild type hiPSCs (Figure 1C). These data suggest a potential correlation between the observed increase of proliferation rate and the higher expression of “stemness” transcription regulators in SMA hiPSCs. Thus, to better clarify if these aspects could have any consequences on hiPSCs differentiation capacity, cells were induced to embryoid body (EB) formation and specifically committed to the ectodermal lineage. While the expression of stem cell markers (OCT4, NANOG and SOX2) equally decreased in both genotypes (data not shown), RT-qPCR analysis (Figure 2A) showed a statistically significant increase of the ectodermal marker “Neural Cell Adhesion Molecule” (NCAM) expression in WT hiPSCs, strongly evident at 22 days after EB adhesion (*** p < 0.001). Conversely, only a slight and not significant increase was observed in SMA hiPSCs over time. As the Sonic hedgehog (Shh)-induced transcriptional pathway is critical for the proper induction of early MN differentiation [9], we measured the expression levels of Shh-related MN markers at the same time points during EB differentiation. The expression levels of Lhx3, Isl1 and HB9 in SMA-hiPSC derived MNs strongly increase during the differentiation process until day 22 (p < 0.05) (Figure 2B). The same pattern was observed in wild type cells but the expression resulted to be more strongly incremented, especially following 22 days of differentiation (p < 0.01) (Figure 2B). In particular HB9 marker, involved in motor neuron specification and maturation [10], showed an evident boost of expression in WT cells (p < 0.001), though unparalleled in SMA cells (Figure 2B). Immunocytochemical analysis performed on hiPSC-derived MNs showed a strong positivity to both LIM3 and TUJ1 markers, also in this case the percentage of LIM3 positive cells resulted to be lower in SMA hiPSC-induced MNs supporting the molecular data (Figure 2C,D). Taken together, these results suggest that SMA hiPSCs demonstrate that they differentiate less efficiently into ectodermal derived cells, and specifically into the target cells of the disease. To further corroborate these data and considering our previous results [8], we also evaluated the expression of miR-335-5p in hiPSCs and followed it up along the in vitro differentiation steps from hiPSCs to EBs, at the same time points where we observed the differential expression of early MN markers. RT-qPCR revealed that, while miR-335-5p expression does not differ between SMA and WT hiPSCs, a significant difference is evident at both 14 and 22 days of differentiation. In particular, miR-335-5p results downregulated more than two-fold in SMA samples compared to WT ones after 14 days of differentiation, and even more strongly reduced after 22 days (Figure 3). This suggests a possible correlation between miR-335-5p expression and the ability of hiPSCs to enter the early stages of differentiation towards motor neurons. 3. Discussion One of the most important aspects of hiPSC technology is the possibility of modelling human diseases using patient-derived reprogrammed cells. The recapitulation of the pathological phenotypes using disease-specific hiPSCs that can be differentiated in vitro, sets the basis for studying the aetiopathology of diseases, especially those in which the target cells are really inaccessible [11,12]. In a previous work we described a perturbed pattern of microRNA expression when comparing neural progenitor cells derived from the spinal cords of E13.5 SMA mice to WT. In particular, miR-335-5p, associated with self-renewal, resulted as underexpressed in the SMA model of murine neural precursors. Our present data allow us to take a further step towards the characterization of some molecular events depicting SMA phenotype during the very early stages of differentiation. In fact, not only are we describing human cells, derived from SMA patients, as opposed to the previously studied murine cells, but we are studying them from the undifferentiated state along the early stages of differentiation. We believe that the time points modelled in vitro in our experiments may be important for neuronal differentiation. Our results about the reduced expression of miR-335-5p in SMA cells during differentiation are novel and important, as they indicate a possible role for this miRNA in SMA disease in humans, specifically in cells which are committed towards motor neurons. In fact, human iPSCs before differentiation do not show any differences in miR-335-5p expression, which become evident during EBs differentiation. These observations deserve further deepening in order to unravel if miR-335-5p reduction not only marks the differentiation of SMA neural precursors, but most importantly if it is causally related to their pathological phenotype. 4. Materials and Methods 4.1. Ethics Statement This study was conducted according to the principles expressed in the Declaration of Helsinki, and approved by the institutional review board of the Bioethical Committee of Fondazione PTV, Tor Vergata Hospital (prot. 0027655/2013). All patients provided written informed consent for the collection of samples and subsequent analysis. 4.2. Cell Culture and Reprogramming to hiPSCs Chorionic villus sampling (CVSs) were obtained from SMA I high-risk pregnancies or from wild type at the 11th week of gestation, following standard biopsy procedures. Cell culture and hiPSCs reprogramming protocol were previously published [11]. Two independent SMA and wild type hiPS cell preparations were analyzed. 4.3. Cytofluorimetry Flow cytometry analysis on hiPSCs (WT and SMA) were performed as already described [8]. 4.4. Expression Analyses For gene expression analyses, total RNA from iPS cells and embryoid bodies (EB) was extracted with TRIzol Reagent (Invitrogen; Life Technologies Corporation, Carlsbad, CA, USA), reverse transcribed with the High-Capacity cDNA Archive kit (Life Technologies Corporation) and used in RT-qPCR, by employing either SYBR Green or TaqMan chemistry (Life Technologies Corporation) and specific primers. The comparative ΔΔCt method was used to quantify relative gene expression levels. All primer sequences for molecular analyses are reported in Table 1. For miRNA expression, total RNA extracted as reported above, was reverse transcribed using the miR-335 and snRNA U6 Taqman assays (Life Technologies Corporation). The quantitative stem-loop real time polymerase chain reaction (qPCR) was performed according to conditions suggested by Life Technologies. 4.5. hiPSCs Differentiation and Immunocytochemical Analysis HiPSCs were differentiated inMNs according to the protocol described by Amoroso et al. [13]. At day 20, EBs were dissociated with PAPAIN (Worthington, Biochemical Corporation, Lakewood, NJ, USA) and plated onto poly-lysine/laminin-coated 8-well chamber slides (BD Biosciences, Two Oak Park, Bedford, MA, USA). Immunocytochemistry was carried out for the detection of specific neural markers: β-III-tubulin (TUJ1; Abcam, Cambridge, UK; 1:500), and LIM3 (Millipore Corporation, Billerica, MA, USA; 1:250). The cell nucleus was labeled with 4,6-diamidino-2-phenylindole (DAPI; Sigma Aldrich, St. Louis, MO, USA) and examined under a fluorescence microscope. Images were acquired using a Zeiss (Thornwood, NY, USA) Axioplan 2 microscope. All values provided in the text and figures are means of three independent experiments ± standard deviations (SD). Mean values were compared using the two-tailed Student t-test, for independent samples. p-Value was considered significant *** p < 0.001, ** p < 0.01, * p < 0.05. FACS Statistical analysis was performed according to paired Student’s t-test by using GraphPad Prism Software version 5.03 (GraphPad Software Inc., La Jolla, CA, USA). Acknowledgments Work granted by Agenzia Spaziale Italiana (ASI) Progetto CoReA project n. 2013-084-R.0. The authors thank Graziano Bonelli for expert technical help. Author Contributions Michela Murdocca, Silvia Anna Ciafrè and Federica Sangiuolo conceived and designed the experiments; Michela Murdocca, Rosa Valentina Talarico, Paola Spitalieri and Massimo Sanchez performed the experiments; Federica Sangiuolo and Silvia Anna Ciafrè analyzed the data; Giuseppe Novelli, Federica Sangiuolo and Silvia Anna Ciafrè wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Spinal Muscular Atrophy (SMA) human induced Pluripotent Stem Cells (hiPSCs) show an increased proliferation rate and expression levels of stem cell markers compared to wild type (WT) ones. (A) Flow cytometry analysis of bromodeoxyuridine (BrdU)-positive cells of SMA- and WT-derived hiPSCs incubated with BrdU for 15, 60 and 180 min. Representative dot plots of cell cycle reporting the BrdU-positive cells are shown. Three regions have been drawn in each dot plot to identify G0/G1, S and G2/M subpopulations; (B) The column bar graph reports the percent of cells in G0/G1, S and G2/M phases of SMA- and WT-derived hiPSCs after 15, 60 and 180 min of BrdU incubation. The results (mean ± standard deviation) are representative of three independent experiments (** p < 0.01, *** p < 0.001); (C) Real time-qPCR analysis of OCT4, NANOG and SOX2 in SMA and WT-hiPSCs using the expression of WT sample as the reference. The data were normalized to 5S ribosomal RNA expression. Data are representative of three independent replicates; values represent mean ± SD; *** p < 0.001, ** p < 0.01, * p < 0.05. Figure 2 Differential expression of Neural Cell Adhesion Molecule (NCAM) and MN-specific transcription factors along differentiation of SMA and WT hiPSCs. (A) RT-qPCR analysis of NCAM expression in WT- and SMA-derived α-motor neurons (MNs) after 14 and 22 days using hiPSCs as a reference. The data are normalized to 5S ribosomal RNA and the expression in hiPSCs was set as =1 in each genotype. Data are representative of three independent replicates; values represent mean ± SD; when comparing WT versus SMA at each time point, *** p < 0.001, ** p < 0.01; (B) The induction of MN differentiation results in transcriptional boost of Isl1, Lhx3 and HB9 in WT cells, while it is lower in SMA ones. The data are normalized to 5S ribosomal RNA and the expression levels in hiPSCs were used as a reference in each genotype. Data are representative of three independent replicates; values represent mean ± SD; (C,D) Representative immunofluorescence images of in WT- and SMA-derived MNs after 22 days of differentiation, expressing β-III tubulin (TUJ1, green) and LIM3 (red). DAPI nuclear staining is in blue. Scale bars, 50 µm. Figure 3 Early MN derived from SMA hiPSCs express reduced levels of miR-335-5p. RT-qPCR of miR-335-5p both in hiPSCs and in early MNs (EB day 14 and EB day 22). The data were normalized to the expression of snRNA U6. Data are representative of three independent replicates; values represent mean ± SD; *** p < 0.001, ** p < 0.01. ijms-17-01231-t001_Table 1Table 1 Oligonucleotide sequence. Primers Forward (5′-3′) Reverse (5′-3′) 5S TCGTCTGATCTCGGAAGCTAAGCA AAAGCCTACAGCACCCGGTATT OCT4 AACCTGGAGTTTGTGCCAGGGTTT TGAACTTCACCTTCCCTCCAACCA SOX2 AGAAGAGGAGAGAGAAAGAAAGGGAGAGA GAGAGAGGCAAACTGGAATCAGGATCAAA NANOG CCTGAAGACGTGTGAAGATGAG GCTGATTAGGCTCCAACCATAC NCAM ATGGAAACTCTATTAAAGTGAACCTG TAGACCTCATACTCAGCATTCCAGT Isl1 GAATGGCATGCGGCATGTTTGA CGCATTTGATCCCGTACAACCTGA Lhx3 TCGGACAAGGACAGCGTTCAG TTTCCGCCAAGGAAGGCTCATCG HB9 CACCGAGACCCAGGTGAAGATTT CCCTTCTGTTTCTCCGCTTCCT OCT4: octamer-binding transcription factor 4; NCAM: Neural Cell Adhesion Molecule; Lhx3: LIM Homeobox 3. ==== Refs References 1. Lefebvre S. Bürglen L. Reboullet S. Clermont O. Burlet P. Viollet L. Benichou B. Cruaud C. Millasseau P. Zeviani M. Identification and characterization of a spinal muscular atrophy determining gene Cell 1995 80 155 165 10.1016/0092-8674(95)90460-3 7813012 2. Coovert D.D. Le T.T. McAndrew P.E. Strasswimmer J. Crawford T.O. Mendell J.R. Coulson S.E. Androphy E.J. Prior T.W. Burghes A.H. The survival motor neuron protein in spinal muscular atrophy Hum. Mol. Genet. 1997 6 1205 1214 10.1093/hmg/6.8.1205 9259265 3. Crawford T.O. Pardo C.A. The neurobiology of childhood spinal muscular atrophy Neurobiol. Dis. 1996 3 97 110 10.1006/nbdi.1996.0010 9173917 4. Ebert A.D. Svendsen C.N. Stem cell model of spinal muscular atrophy Arch. Neurol. 2010 67 665 669 10.1001/archneurol.2010.89 20558385 5. Frattini E. Ruggieri M. Salani S. Faravelli I. Zanetta C. Nizzardo M. Simone C. Magri F. Corti S. Pluripotent stem cell-based models of spinal muscular atrophy Mol. Cell. Neurosci. 2015 64 44 50 10.1016/j.mcn.2014.12.005 25511182 6. Stappert L. Roese-Koerner B. Brustle O. The role of microRNAs in human stem cells, neuronal differentiation and subtype specification Cell Tissue Res. 2015 359 47 64 10.1007/s00441-014-1981-y 25172833 7. Schoeftner S. Scarola M. Comisso E. Schneider C. Benetti R. An OCT4-pRb axis, controlled by miR-335, integrates stem cell self-renewal and cell cycle control Stem Cells 2013 31 717 728 10.1002/stem.1315 23307555 8. Luchetti A. Ciafrè S.A. Murdocca M. Malgieri A. Masotti A. Sanchez M. Farace M.G. Novelli G. Sangiuolo F. A perturbed microRNA expression pattern characterizes embryonic neural stem cells derived from a severe mouse model of Spinal Muscular Atrophy (SMA) Int. J. Mol. Sci. 2015 16 18312 18327 10.3390/ijms160818312 26258776 9. Briscoe J. Ericson J. Specification of neuronal fates in the ventral neural tube Curr. Opin. Neurobiol. 2001 11 43 49 10.1016/S0959-4388(00)00172-0 11179871 10. Arber S. Han B. Mendelsohn M. Smith M. Jessell T.M. Sockanathan S. Requirement for the homeobox gene Hb9 in the consolidation of motor neuron identity Neuron 1999 23 659 674 10.1016/S0896-6273(01)80026-X 10482234 11. Spitalieri P. Talarico R.V. Botta A. Murdocca M. D’Apice M.R. Orlandi A. Giardina E. Santoro M. Brancati F. Novelli G. Generation of human induced pluripotent stem cells from extraembryonic tissues of fetuses affected by monogenic diseases Cell. Reprogram. 2015 17 275 287 10.1089/cell.2015.0003 26474030 12. Qin Y. Gao W.Q. Concise Review: Patient-derived stem cell research for monogenic disorders Stem Cells 2016 34 44 54 10.1002/stem.2112 26227066 13. Amoroso M.W. Croft G.F. Williams D.J. O’Keeffe S. Carrasco M.A. Davis A.R. Roybon L. Oakley D.H. Maniatis T. Henderson C.E. Accelerated high-yield generation of limb-innervating motor neurons from human stem cells J. Neurosci. 2013 33 574 586 10.1523/JNEUROSCI.0906-12.2013 23303937

PMC005xxxxxx/PMC5000630.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081232ijms-17-01232ReviewCirculating Organ-Specific MicroRNAs Serve as Biomarkers in Organ-Specific Diseases: Implications for Organ Allo- and Xeno-Transplantation Zhou Ming 12Hara Hidetaka 3Dai Yifan 4Mou Lisha 1*Cooper David K. C. 3Wu Changyou 2Cai Zhiming 1*Masuda Satohiro Academic EditorCho William Chi-shing Academic Editor1 Shenzhen Xenotransplantation Medical Engineering Research and Development Center, Shenzhen Second People’s Hospital, First Affiliated Hospital of Shenzhen University, Shenzhen 518039, China; [email protected] Institute of Immunology, Zhongshan School of Medicine, Guangdong Provincial Key Laboratory of Organ Donation and Transplant Immunology, Sun Yat-sen University, Guangzhou 510275, China; [email protected] Thomas E. Starzl Transplantation Institute, University of Pittsburgh, Pittsburgh, PA 15261, USA; [email protected] (H.H.); [email protected] (D.K.C.C.)4 Jiangsu Key Laboratory of Xenotransplantation, Nanjing Medical University, Nanjing 210029, China; [email protected]* Correspondence: [email protected] (L.M.); [email protected] (Z.C.); Tel./Fax: +86-755-8355-7380 (L.M. & Z.C.)01 8 2016 8 2016 17 8 123226 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Different cell types possess different miRNA expression profiles, and cell/tissue/organ-specific miRNAs (or profiles) indicate different diseases. Circulating miRNA is either actively secreted by living cells or passively released during cell death. Circulating cell/tissue/organ-specific miRNA may serve as a non-invasive biomarker for allo- or xeno-transplantation to monitor organ survival and immune rejection. In this review, we summarize the proof of concept that circulating organ-specific miRNAs serve as non-invasive biomarkers for a wide spectrum of clinical organ-specific manifestations such as liver-related disease, heart-related disease, kidney-related disease, and lung-related disease. Furthermore, we summarize how circulating organ-specific miRNAs may have advantages over conventional methods for monitoring immune rejection in organ transplantation. Finally, we discuss the implications and challenges of applying miRNA to monitor organ survival and immune rejection in allo- or xeno-transplantation. allotransplantationbiomarkerimmune rejectionmiRNAcirculatingxenotransplantation ==== Body 1. Introduction 1.1. Current Status of Biomarkers in the Detection of Graft Rejection Organ failure is the leading cause of human death [1]. Replacing a non-functioning organ with a new one is the most effective strategy to cure terminal organ failure and maintain patients’ lives. However, immune rejection still remains a major obstacle for long-term survival of allo- or xeno-transplanted organs (reviewed in [2]). Real-time monitoring of immune rejection is critical for organ allo (xeno)-transplantation. However, existing methods are limited. Currently, detection of acute rejection is largely based on clinical data such as a patient’s symptoms and physical signs, but also on laboratory data such as biochemical and immunological assays and tissue biopsies [3], some of which are not ideal for clinical application. A patient’s symptoms and physical signs are often of value only at a ‘late-stage’ of rejection. Tissue biopsies are the ‘gold standard’, but are invasive and expensive with some limitation of diagnostic accuracy [3,4]. Biochemical and immunological assays are either of low sensitivity or low specificity [5,6]. For example, biochemical biomarkers of serum creatinine and urine albumin are classical indicators of kidney injury following renal allo (xeno)-transplantation, but are of relatively low sensitivity (reviewed in [7]). Furthermore, there are no reliable methods or biomarkers to monitor chronic rejection or acute-on-chronic rejection. Apart from immune rejection, there are special concerns in organ allo (xeno)-transplantation, such as recurrence of primary disease, thrombosis formation, and hemolysis, all of which need to be properly monitored post-transplantation. Therefore, novel non-invasive biomarkers that might specifically monitor immune rejection and/or rejection-associated complications are urgently needed. Circulating miRNAs may serve as non-invasive, specific, sensitive, and low-cost biomarkers in organ allo (xeno)-transplantation. 1.2. Potential of Circulating miRNAs as Biomarkers of Graft Rejection MicroRNA (miRNA) is a small non-coding RNA containing approximately 22 nucleotides; it is found in different species and functions in the RNA silencing and post-transcriptional regulation of 30% of the gene expression in humans [8,9], including proliferation, DNA repair, differentiation, metabolism, and apoptosis [10,11]. Different cells/tissues/organs possess different miRNA expression profiles [12,13]. Some miRNAs are specific or abundant in certain organs, e.g., miR-122 is liver-specific [13]. MiRNA expression profiles, including cell/tissue/organ-specific miRNAs, indicate different diseases, such as cancers [14]. Profiles of miRNA in the body fluids, also called circulating miRNA or cell-free miRNA, serve as non-invasive biomarkers for physiological and pathological changes, such as cancers [15], organ injury [16], diabetes [15], and even pregnancy [17,18]. Cell death, such as apoptosis or necrosis, is the final result of immune rejection. Theoretically, intracellular miRNAs are passively released from rejected cells and become part of the circulating miRNAs (reviewed in [19]) (also shown in Figure 1). In fact, this is the scientific basis for most of the rejection-related biomarkers, including proteins, DNA, and RNA, which follow cell death once the allo (xeno)-grafts are attacked by the host immune system. Both proteins and DNA are proved to be useful biomarkers for acute immune rejection [4,20,21,22,23,24]. As described above, there is much more information on circulating miRNA than on circulating DNA, and detection of miRNA is more sensitive and quantifiable than that of proteins, indicating miRNA may be more powerful in the diagnosing and prognosis of organ survival and immune rejection. The levels of circulating miRNAs are quite stable in healthy people, but may become deregulated once cell death occurs [15]. As mentioned above, circulating miRNAs are tissue/organ-specific and may be disease- and stage-specific, resistant to adverse conditions, easily detectable and quantifiable, and readily accessible using non-invasive methods, all of which confers on them promise for the detection of rejection. Organ/tissue-specific/enriched miRNAs may be of more value than some ubiquitously-expressed miRNAs in organ allo- or xeno-transplantation, as they are specific indicators for the state of the transplanted organ. Circulating organ/tissue-specific/enriched miRNAs may indicate direct organ/tissue injury or cell death, and circulating immune-associated miRNAs may serve as sensors of the immune state. A combination of organ/tissue-specific/enriched miRNA and immune-associated miRNAs might be valuable to distinguish between different diseases, such as lung injury caused by cytomegalovirus (CMV) infection or immune rejection. Therefore, this paper provides background information on circulating miRNAs as a proof of concept that circulating organ-specific miRNAs serve as non-invasive biomarkers for a wide spectrum of clinical organ-specific manifestations. We subsequently review the potential role of circulating miRNAs in organ allo (xeno)-transplantation, and finally discuss the challenges that remain if these biomarkers are to prove valuable in organ allo (xeno)-transplantation. 2. Background 2.1. Circulating miRNA: Sources and Functions miRNA biogenesis and mechanisms of action have been reviewed in detail [25,26,27]. Other than intracellular miRNA, mature miRNA can be found in body fluids, including plasma, serum, urine, tears, breast milk, amniotic fluid, bronchial lavage, pleural fluid, cerebrospinal fluid, and saliva [15,28,29,30,31] (reviewed in [32]). Circulating miRNA may be actively secreted from living cells, but may also be passively released by dying cells. Circulating miRNA may be found in microvesicles, in apoptotic bodies, in/on high density lipoprotein (HDL) particles, or may bind to argonaute (AGO) proteins (reviewed in [32,33,34]). Mostly, miRNAs are byproducts of cellular activities and cell death, except certain miRNA species might also function in cell–cell communication (reviewed in [32]). The major form of circulating miRNA is AGO-binding miRNA, which accounts for 90%–99% of the total circulating miRNA [35,36]. AGO proteins, as essential catalytic components of the RNA-induced silencing complex (RISC), play a central role in RNA-mediated regulation of gene expression, during which all mature miRNAs become associated with one of the four AGO proteins (mainly AGO2) [8,37]. The AGO protein provides stability to the miRNA. The function of circulating AGO-binding miRNAs is unknown. The miRNAs in microvesicles enveloped by a phospholipid bilayer is a mixed population of exosomes and shedding vesicles (reviewed in [32]). The phospholipid bilayer is impermeable to RNases and renders stability to miRNAs [38]. Although the underlying mechanisms of sorting and secretion have not yet been fully explained, the vesicular miRNAs are suggested to be released through a ceramide-dependent secretory pathway [39,40,41]. Although accounting for only a minority of the total circulating miRNA [42,43,44], miRNAs in microvesicles play an important role in cell-to-cell or organ-to-organ communication [42,45,46], including in immune regulation (reviewed in [47,48]) and in cancer metastasis (reviewed in [49,50]). Some miRNAs are packaged into or bind to HDL particles regulated by neutral sphingomyelinase, which also transports endogenous miRNAs and delivers them to recipient cells with functional targeting capabilities [51]. MiRNAs can also be entrapped into apoptotic bodies during apoptosis and accompanied by various cellular organelles [52]. Sources of circulating miRNA are shown in Figure 1 (details shown in Table 1). The level of circulating miRNA is quite stable in healthy people, but deregulated under certain conditions, such as physiological changes, inflammation, and cell death [15]. Hence, circulating miRNAs are promising biomarkers for specific diseases. In addition, circulating miRNAs resist harsh conditions, including RNase digestion, freeze-thawing, boiling, and extreme pH conditions, rendering them extremely promising markers for the non-invasive detection of various diseases [15,28]. 2.2. Circulating Liver-Specific miRNAs Serve as Novel Biomarkers in Liver Disease The liver is one of the most vital organs in humans and other animals, with a wide range of functions, including detoxification of various metabolites, protein synthesis, and the production of bile [55]. In the liver, the most abundant miRNA (>70%) is miR-122, which is also the hepatocyte-specific miRNA [12,56], acting as a novel regulator of diverse aspects of hepatic function, including metabolism [57], response to stress [58], and maintenance of hepatic phenotype [59]. Once the microenvironment changes, levels of circulating miR-122 are deregulated, for example, in certain pathological conditions such as drug- or alcohol-induced cytotoxicity, viral infections, and hepatic disease progression (reviewed in [60]). In recent years, a large number of reports have been published indicating that circulating miR-122 can be applied as a non-invasive biomarker for a wide spectrum of clinical liver-specific diseases, including hepatic virus infections [61,62,63,64,65,66,67,68], liver injury [69,70,71,72,73], hepatic cirrhosis [74,75], and hepatocellular carcinoma [65,76,77] (reviewed in [60]). By reviewing the published papers, we have summarized the information, and it suggests that circulating miR-122 is the most frequent biomarker for liver disease (Figure 2). Other circulating hepatic-abundant miRNAs could also serve as biomarkers for liver disease, such as miR-22 [71,78], miR-125b [64,65,68,79], miR-99a [65,68,80] and miR-192 [72,74,76,77] (Figure 2). It is worth noting that a panel of circulating miRNAs may enhance specificity and sensitivity [76,80], in which both liver-specific miRNAs and liver-abundant miRNAs might indicate a specific type of liver disease or dysfunction. 2.3. Circulating Cardiac-Specific miRNAs Serve as Novel Biomarkers in Heart Disease The heart is a muscular organ that pumps blood through the blood vessels of the circulatory system [81]. Unlike liver-specific miR-122, cardiac-specific miRNA (miR-208) is not a heart-enriched miRNA (e.g., miR-1, miR-133a, miR-499, and miR-296). Once the heart is injured, muscle-enriched miRNAs and cardiac-specific miR-208 are released into body fluids, and can serve as novel biomarkers for heart disease [82,83,84,85,86,87,88,89,90,91] (also reviewed in [92,93,94,95]). In fact, the boundary is not clear between “specific” and “enriched/abundant” miRNAs, terms which sometimes replace each other in different published papers [96]. The heart is rich in blood vessels, such as arteries, veins, and capillaries. Hence, enriched/specific miRNAs in vascular endothelial cells, such as miR-126, may also be indicators of cardiovascular diseases. For example, circulating miR-126 is significantly down-regulated in patients suffering symptomatic atherosclerosis [97], acute myocardial infarction [98], and heart failure [99]. 2.4. Circulating Kidney-Specific miRNAs Serve as Novel Biomarkers in Renal Disease Unlike the liver and heart, the kidney possesses more diverse cell types and structures which carry out complicated functions, including regulation of blood-electrolyte balance, maintenance of blood acid–base balance, removal of excess water-soluble wastes from the blood, and regulation of blood pressure [100]. A set of kidney-specific miRNAs is present, of which expression levels favor understanding and diagnosis for renal diseases [101]. These specific miRNAs include miR-192, miR-194, miR-204, miR-215, and miR-216, which have low expression in the liver, lung and heart. As a proof of concept, different renal cell types possess different profiles of miRNA expression [12,13], which may be indicators of different diseases in different segments of the kidney. For example, miR-192 and miR-194 are significantly more abundant in the cortex, while miR-30c and miR-200c are highly expressed in the medulla [102,103]. Surprisingly, few circulating kidney-specific miRNAs have been reported as non-invasive biomarkers for renal disease in humans, possibly because most research focused on excavating biomarkers for the more accessible sample of urine. Indeed, circulating miRNAs have rarely been studied in the diagnosis of renal diseases. We could find only one paper which reported that circulating levels of miR-16 and miR-320 are down-regulated while miR-210 is up-regulated in patients with acute kidney injury [104]. However, in rat models, plasma kidney-specific/enriched miRNAs (miR-10a, miR-192, and miR-194) have been reported to be potential biomarkers for renal ischemia-reperfusion injury [105]. Urine, which is more accessible than serum/plasma, can provide a non-invasive sample to test for renal disease. Urinary miRNAs are filtered or excreted from the kidney and/or urinary tract [28] (reviewed in [106]). Once the filtering function of the kidney is injured, miRNAs from the plasma may become urinary miRNAs, which can serve as potential biomarkers of renal disease. These miRNAs can be considered similar to certain urinary proteins which are normally absent from the urine and, when present, indicate kidney damage. Urinary miR-200a levels are useful for the diagnosis of renal tubular dysfunction (in the Dahl salt-sensitive rat with high salt administration) [107]. A high urinary level of miR-494 precedes an increase in serum creatinine in patients with kidney injury, indicating that urinary miR-494 can serve as an early and non-invasive indicator of acute renal injury [108]. Urinary miRNAs excreted from the kidney and/or urinary tract are also good sensors of disease of the urinary system. For example, a panel of miRNAs (miR-126, miR-152, miR-182) is significantly increased in the urine of patients with urothelial bladder cancer [109], which suggests that urine miRNAs may serve as markers of bladder cancer. 2.5. Circulating Lung-Specific miRNAs Serve as Novel Biomarkers in Lung Disease The lung is the primary organ of the respiratory system in mammals, functioning in the process of gas exchange. By using semi-quantitative real-time RT-PCR, miR-92, miR-26a, miR-200c, miR-16, let-7b, miR-125a, and miR-125b have been found to be the most highly expressed miRNAs in human airway tissues, having levels >70-fold higher than the average miRNA and contributing 55.5% of the total mRNA detected in airway biopsies [110]. Using microarrays, miR-195 and miR-200c were found to be expressed specifically in the rat lung [111], although miR-195 is only moderately expressed in the human lung [110]. In patients suffering from idiopathic pulmonary fibrosis, miR-200c was significantly increased in sera compared to healthy controls [112]. The down-regulation of both circulating miR-195 and miR-21 predicted poor differentiation of non-small cell lung cancer [113]. We could not identify other reports, probably because relatively few studies have been carried out in this field. 3. The Potential Role of Circulating miRNAs to Detect Graft Rejection 3.1. Circulating Organ-Specific miRNAs in Organ Allotransplantation In organ allotransplantation, the grafts may be damaged by the host immune system. Levels of certain circulating miRNAs from both the transplanted organ graft and from cells of the host’s immune system may change significantly. Indeed, there are already reports focusing on rejection of organ allografts [16,114,115,116,117,118]. Organ-specific/enriched miRNAs normally serve as biomarkers of direct injury or ischemia-reperfusion injury. For example, hepatocyte-abundant miRNAs (miR-122, miR-148a) and cholangiocyte-abundant miRNAs (miR-30e, miR-222, and miR-296) may be elevated in the solution in which the organ has been stored and transported, which may predict the quality of a donor liver [114,115]. Serum levels of hepatocyte-derived miRNAs (miR-122, miR-148a, miR-194) were elevated in patients with liver injury and positively-correlated with aminotransferase levels in patients with liver transplants [16]. Kidney-derived miRNAs (miR-21, miR-20a, miR-146a, miR-199a-3p, miR-214, miR-192, miR-187, miR-805, and miR-194) may reveal a signature of kidney damage following ischemia-reperfusion injury, which could be used as a biomarker of renal injury [116,117]. Circulating pancreas/islet-specific miR-375 could be a reliable biomarker to detect graft damage in clinical islet transplantation, and can be compared with C-peptide and pro-insulin levels [118]. Circulating organ/tissue-specific/enriched miRNAs which can potentially be applied to monitor typical allograft rejections are shown in Table 2. 3.2. Circulating Immune-Associated miRNAs in Organ Allotransplantation In contrast, circulating immune-associated miRNAs normally serve as biomarkers for immune activation and immune rejection. In patients with allotransplanted hearts, circulating miR-10a (down-regulated), miR-31, miR-92a, and miR-155 (all up-regulated) strongly discriminate between patients undergoing allograft rejection and those not undergoing rejection [120]. Notably, these miRNAs (miR-10a, miR-31, miR-92a, and miR-155) were associated with an inflammatory response [121,122,123,124]. Decreased levels of miR-210 were observed in the urine of patients undergoing acute T cell-mediated renal allograft rejection, and increased in response to corticosteroid therapy [125]. Eight miRNAs in peripheral blood mononuclear cells have been identified as sensors of operational tolerance in kidney transplant recipients (up-regulation: miR-450b-5p, miR142-3p, miR-876-3p, and miR-106b; down-regulation: miR-508-3p, miR-148b, miR-324-5p, and miR-98) [126]. In particular, up-regulation of miR-142-3p is correlated with operational tolerance of a B lymphocyte subset, which would be useful to identify immune tolerance and optimize individualized treatment by immunosuppressive drugs. 3.3. Circulating Organ-Specific miRNAs in Organ Xenotransplantation Organ allotransplantation is effective in the treatment of terminal organ failure. However, >90% patients remain on the waiting list due to the shortage of deceased human donors [127]. Pigs are regarded as the most promising alternatives as sources of organs and tissues [128,129]. Progress has been made by gene-editing, which dramatically reduces immune rejection [130,131], and may reduce the potential risk of the presence of porcine endogenous retrovirus (PERV) in the pig organ [132,133,134]. However, xenotransplantation is in urgent need of novel biomarkers to monitor graft survival or immune rejection. Circulating organ-specific/enriched miRNAs may provide methods to monitor xenograft survival, and immune-associated miRNAs may serve to monitor immune rejection and immune tolerance (by non-invasive procedures). In a pig model of acetaminophen-induced acute liver failure, pig (Sus scrofa)-derived miRNAs including ssc-miR-122 (liver-specific), ssc-miR-192 (kidney-specific), and ssc-miR-124-1 (brain-enriched) were associated with clinical evidence of liver, kidney and brain injury, respectively [135]. However, there are no further reports addressing this issue, probably due to limited pre-clinical experience. Lack of complete genome information further impeded profiling miRNA expression. As described from differences in miRNAs between humans and mice [13], most miRNAs are conserved in sequence and abundance between species. Indeed, there are already reports of pig miRNA sequencing [136,137,138,139,140,141,142] providing the expression profiles of different organs, tissues, and cells. Certain miRNAs may be quite different in sequence and expressed differentially between species [136,143,144,145] (sometimes called xeno-miRNAs), which may be useful to increase specificity. For example, the pig-species miRNA ssc-miR-199b* possesses an internal 2-nt mismatch compared with its counterpart from human and mouse. It is moderately expressed in liver, heart, and lung [146], and may be discriminated by Taqman probe in qPCR. ssc-miR-199b* may potentially serve as a biomarker for the fate of a xenograft. For a xeno-miRNA to serve as a biomarker in xenotransplantation, it would be best for it to be abundant, differentiable, and easily quantifiable. Substantial work needs to be carried out to address this issue. Circulating organ/tissue-specific/enriched miRNAs that might be applied to monitor typical xenograft rejection are shown in Table 3. 4. Discussion—Challenges and Solutions There remain too many gaps in our knowledge of miRNAs, including regulation of miRNA production, specific targets, and mechanisms of active secretion [148]. For example, intracellular levels of miRNAs and/or levels of secretion may fluctuate at different stages of rejection, which may complicate interpretation of the data. For example, circulating liver-specific miRNA-122 is dramatically up-regulated during liver injury [69,70,71,72,73], but significantly down-regulated during late-stage hepatic cirrhosis [75] and hepatocellular carcinoma [76,77], due to intracellular down-regulation of miRNA production. Therefore, levels of certain circulating organ-specific miRNAs may be significantly up-regulated during acute rejection, but significantly down-regulated during chronic or late rejection. Changes in circulating miRNAs cannot simply be explained by rejection increasing their levels. Substantial and critical studies need to be carried out to identify whether circulating miRNAs can be used as reliable biomarkers for diagnosis, prognosis, and response to therapy. Another challenge we face is the problem of specificity. (1) One circulating miRNA may not be sufficiently specific to target a certain organ, tissue or cell. For example, miR-375 is pancreas/islet-specific and reported to be a reliable biomarker to detect islet graft injury [118], but we note it is also expressed in the airways [110] and thus damage to an airway may interfere with the diagnosis of islet rejection; (2) Another problem is that immune-associated miRNAs may be elevated during opportunistic infection, inflammation, and cancer, which may also confuse interpretation of the results; (3) What may provide an additional problem is that most published researchers have designed their experiments to compare a disease group with a healthy group, rather than included alternative disease groups. A biomarker may powerfully discriminate between a disease group and a healthy group but may also prove to be a biomarker for another disease. Only a subset of reported blood-based miRNA biomarkers has specificity for a particular disease [149]. Therefore, a panel of miRNAs may be more specific to determine allo- or xeno-graft survival or rejection. Some researchers applying high through-put sequencing have already shown a panel of miRNAs with enhanced specificity in diagnosis [76,80]. In addition, if it is possible to identify xeno-miRNAs in pig-to-human xenotransplantation models, this may enhance specificity. Moreover, the sensitivity of circulating miRNAs should also be taken into consideration for different types of organ/tissue transplantation. A normal cell-turnover rate is mass-dependent [21]. Therefore, the level of a circulating miRNA may also be mass-dependent even during rejection. As reviewed above, we believe circulating organ-specific/enriched miRNAs are easily detectable during solid organ transplantations such as liver, heart, kidney, and lung. However, other transplants of small-mass tissues, such as corneal, islet, skin, and cardiac valves, may exhibit limited release of tissue-specific/enriched miRNAs during rejection. As we learnt from clinical islet allotransplantation, circulating islet-specific miR-375 is dramatically elevated during rejection [118], indicating tissue-specific/enriched miRNAs derived from small-mass tissues could be potentially detected. In addition, circulating miRNA can be found in different body fluids [15,28,29,30,31] (reviewed in [32]), thus proper choice of sampling may further enhance sensitivity. A technical challenge is that there is currently no standardized method for measuring miRNAs. Different technologies have been applied to measure circulating miRNAs, such as high through-put sequencing, microarrays, PCR arrays, qPCR, and droplet digital PCR, which make the data difficult to compare. A high through-put, precise, and reproducible detection method is urgently needed, which may lead to increased progress in this field. In addition, the heterogeneity of circulating miRNA (in microvesicles, apoptotic bodies, HDL particles, or bound to AGO protein) requires further standardization of protocols for sample processing to make sure the data between different studies can be directly compared. Although thousands of reports have demonstrated the potential of measuring circulating miRNA, there still is a huge gap between fundamental research and clinical application, because no standard is available by which clinicians can judge whether a graft is rejected or not. Apart from unsatisfactory specificity and sensitivity as described above, clinical variance or biases such as donor/recipient heterogeneity, immunosuppressive treatment, and potential contamination may further complicate the diagnosis. For example, the mass of the organ/tissue is variable resulting from donor heterogeneity, and levels of circulating tissue-specific/enriched miRNAs may also be variable. Therefore, it will be hard to establish a normal or healthy threshold for circulating miRNAs. As we learnt from cell-free DNA in the diagnosis of rejection [21], levels and features of circulating organ/tissue-specific/enriched miRNAs should be identified first in different recipients without evidence of graft rejection. Ideally, there should be a representative mathematical model or regression curve for recipients without graft rejection, which will then allow precise diagnosis of rejection. In this case, rejection judged by individual dynamics of circulating miRNAs may be more applicable than the fold-change normally reported in most fundamental studies. This issue needs further study. Nevertheless, circulating organ-specific miRNAs have the potential to provide biomarkers for various diseases and as non-invasive indicators of organ/tissue allograft and/or xenograft rejection. Acknowledgments Our research was supported in part by ‘Three Outstanding Projects’ of Shenzhen, the Project of Shenzhen Engineering Center (GCZX2015043017281705), Clinical Doctor-Basic Scientist Combination Foundation of Shenzhen Second People’s Hospital and Key Laboratory Project of Shenzhen Second People’s Hospital, and Shenzhen Foundation of Science and Technology (grant nos. GJHZ20140414170821192, JCYJ20140414170821337 and JCYJ20160229204849975). Author Contributions Lisha Mou and Zhiming Cai conceived and designed the study. Ming Zhou wrote the paper. Hidetaka Hara, Yifan Dai, David K.C. Cooper and Changyou Wu reviewed and edited the manuscript. All authors read and approved the manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations AGO argonaute protein HDL high density lipoprotein hsa Homo sapiens miRNA(miR-) MicroRNA PCR polymerase chain reaction qPCR quantitative PCR RNA ribonucleic acid RISC RNA-induced silencing complex ssc Sus scrofa CMV Cytomegalovirus Figure 1 A schematic model of sources of circulating miRNAs. Circulating miRNAs can be actively secreted from living cells, mainly in the form of microvesicles and AGO-binding miRNA derived from the exosome pathway and transmembrane transporter, respectively. They can also be passively released from dying cells in the form of necrosis lysate or apoptotic bodies. All the cell-free miRNAs finally diffuse into body fluids, such as the blood. Solid and broken green arrows between vascular endothelial cells indicate large-scale and micro-scale release of circulating miRNA, respectively. All the source materials were obtained from a web-accessible software plugin of PowerPoint: Science Slide 5. Figure 2 Circulating liver-specific/enriched miRNAs serve as biomarkers for different liver diseases. Expression profiles of liver miRNAs were obtained from a web-accessible database (http://www.mirz.unibas.ch/), of which miRNA expression was determined by small RNA library sequencing [13]. The frequencies of circulating miRNAs as biomarkers were determined from 65 published papers. ijms-17-01232-t001_Table 1Table 1 Types, sources, functions, content, and size of different types of circulating miRNA. Types Sources Functions Content Size Citation AGO-binding mainly necrotic cells a byproducts 90%–99% unknown [35,36] exosomes living cells cell-to-cell communication minority 30–100 nm [39,40,41,42,43,44,45,46] shedding vesicles living cells cell-to-cell communication minority 0.1–1 μm [32,53,54] HDL particles living cells cell-to-cell communication minority 8–12 nm [51] apoptotic bodies apoptotic cells byproducts minority b 1–4 μm [52] a Circulating AGO-binding miRNAs may also be actively secreted from living cells; b The content may increase to some extent under disease conditions. ijms-17-01232-t002_Table 2Table 2 Organ/tissue-specific/enriched miRNAs in organs/tissue of humans (Homo sapiens). Organ/Tissue Specific/Enriched miRNAs a Citation Liver miR-122, miR-125b, miR-16, miR-99a [12,56] Heart miR-1, miR-126, miR-133a, miR-208, miR-296, miR-499 [82,83,84,85,86,87,88,89,90,91] Kidney miR-192, miR-194, miR-204, miR-215, miR-216 [101] Lung let-7b, miR-125a, miR-125b, miR-16, miR-195, miR-200c, miR-26a, miR-92 [110,111] Pancreas/Islet miR-375 [13,119] a All the organ/tissue-specific miRNAs are underlined and in bold. ijms-17-01232-t003_Table 3Table 3 Organ/tissue-specific/enriched miRNAs in typical organs/tissue of pigs (Sus scrofa). Organ/Tissue Specific/Enriched miRNAs a Xeno-miRNAs Citation Liver miR-122, miR-153-3p, miR-194 miR-199b* [136,137,146] Heart miR-1, miR-133, miR-208, miR-499 miR-199b* [136,146] Kidney miR-125b, miR-192, miR-200a, miR-23b unknown [139,147] Lung let-7i, miR-143-3p, miR-145, miR-320 miR-199b* [138,139,146] a All the organ/tissue-specific miRNAs are underlined and in bold. ==== Refs References 1. Buchman T.G. Multiple organ failure Curr. Opin. Gen. Surg. 1993 26 31 7583986 2. Cooper D.K. Ekser B. Tector A.J. Immunobiological barriers to xenotransplantation Int. J. Surg. 2015 23 211 216 10.1016/j.ijsu.2015.06.068 26159291 3. Williams W.W. Taheri D. Tolkoff-Rubin N. Colvin R.B. Clinical role of the renal transplant biopsy Nat. Rev. Nephrol. 2012 8 110 121 10.1038/nrneph.2011.213 22231130 4. Snyder T.M. Khush K.K. Valantine H.A. Quake S.R. Universal noninvasive detection of solid organ transplant rejection Proc. Natl. Acad. Sci. USA 2011 108 6229 6234 10.1073/pnas.1013924108 21444804 5. Sood S. Testro A.G. Immune monitoring post liver transplant World J. Transplant. 2014 4 30 39 10.5500/wjt.v4.i1.30 24669365 6. Townamchai N. Eiam-Ong S. Biomarkers in kidney transplantation: From bench to bedside World J. Nephrol. 2015 4 487 491 26558185 7. Ronco C. Acute kidney injury: From clinical to molecular diagnosis Crit. Care 2016 20 201 10.1186/s13054-016-1373-7 27384344 8. Bartel D.P. MicroRNAs: Genomics, biogenesis, mechanism, and function Cell 2004 116 281 297 10.1016/S0092-8674(04)00045-5 14744438 9. Lewis B.P. Burge C.B. Bartel D.P. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets Cell 2005 120 15 20 10.1016/j.cell.2004.12.035 15652477 10. Croce C.M. Calin G.A. MiRNAs, cancer, and stem cell division Cell 2005 122 6 7 10.1016/j.cell.2005.06.036 16009126 11. Ambros V. The functions of animal microRNAs Nature 2004 431 350 355 10.1038/nature02871 15372042 12. Lagos-Quintana M. Rauhut R. Yalcin A. Meyer J. Lendeckel W. Tuschl T. Identification of tissue-specific microRNAs from mouse Curr. Biol. 2002 12 735 739 10.1016/S0960-9822(02)00809-6 12007417 13. Landgraf P. Rusu M. Sheridan R. Sewer A. Iovino N. Aravin A. Pfeffer S. Rice A. Kamphorst A.O. Landthaler M. A mammalian microRNA expression atlas based on small RNA library sequencing Cell 2007 129 1401 1414 10.1016/j.cell.2007.04.040 17604727 14. Lu J. Getz G. Miska E.A. Alvarez-Saavedra E. Lamb J. Peck D. Sweet-Cordero A. Ebert B.L. Mak R.H. Ferrando A.A. MicroRNA expression profiles classify human cancers Nature 2005 435 834 838 10.1038/nature03702 15944708 15. Chen X. Ba Y. Ma L. Cai X. Yin Y. Wang K. Guo J. Zhang Y. Chen J. Guo X. Characterization of microRNA s in serum: A novel class of biomarkers for diagnosis of cancer and other diseases Cell Res. 2008 18 997 1006 10.1038/cr.2008.282 18766170 16. Farid W.R. Pan Q. van der Meer A.J. de Ruiter P.E. Ramakrishnaiah V. de Jonge J. Kwekkeboom J. Janssen H.L. Metselaar H.J. Tilanus H.W. Hepatocyte-derived microRNAs as serum biomarkers of hepatic injury and rejection after liver transplantation Liver Transplant. 2012 18 290 297 10.1002/lt.22438 21932376 17. Miura K. Morisaki S. Abe S. Higashijima A. Hasegawa Y. Miura S. Tateishi S. Mishima H. Yoshiura K. Masuzaki H. Circulating levels of maternal plasma cell-free pregnancy-associated placenta-specific microRNAs are associated with placental weight Placenta 2014 35 848 851 10.1016/j.placenta.2014.06.002 25015175 18. Pan M. Ge Q. Li H. Yang Q. Lu J. Zhang D. Lu Z. Sequencing the miRNAs in maternal plasma from women before and after parturition J. Nanosci. Nanotechnol. 2012 12 4035 4043 10.1166/jnn.2012.6196 22852344 19. Schwarzenbach H. Nishida N. Calin G.A. Pantel K. Clinical relevance of circulating cell-free microRNA s in cancer Nat. Rev. Clin. Oncol. 2014 11 145 156 10.1038/nrclinonc.2014.5 24492836 20. Sigdel T.K. Vitalone M.J. Tran T.Q. Dai H. Hsieh S.C. Salvatierra O. Sarwal M.M. A rapid noninvasive assay for the detection of renal transplant injury Transplantation 2013 96 97 101 10.1097/TP.0b013e318295ee5a 23756769 21. De Vlaminck I. Martin L. Kertesz M. Patel K. Kowarsky M. Strehl C. Cohen G. Luikart H. Neff N.F. Okamoto J. Noninvasive monitoring of infection and rejection after lung transplantation Proc. Natl. Acad. Sci. USA 2015 112 13336 13341 10.1073/pnas.1517494112 26460048 22. Lo Y.M. Tein M.S. Pang C.C. Yeung C.K. Tong K.L. Hjelm N.M. Presence of donor-specific DNA in plasma of kidney and liver-transplant recipients Lancet 1998 351 1329 1330 10.1016/S0140-6736(05)79055-3 9643800 23. Adams D.H. Mainolfi E. Elias E. Neuberger J.M. Rothlein R. Detection of circulating intercellular adhesion molecule-1 after liver transplantation—Evidence of local release within the liver during graft rejection Transplantation 1993 55 83 87 10.1097/00007890-199301000-00016 8093567 24. Jin Z.K. Tian P.X. Wang X.Z. Xue W.J. Ding X.M. Zheng J. Ding C.G. Mao T.C. Duan W.L. Xi M. Kidney injury molecule-1 and osteopontin: New markers for prediction of early kidney transplant rejection Mol. Immunol. 2013 54 457 464 10.1016/j.molimm.2013.01.013 23435265 25. Pritchard C.C. Cheng H.H. Tewari M. MicroRNA profiling: Approaches and considerations Nat. Rev. Genet. 2012 13 358 369 10.1038/nrg3198 22510765 26. Mendell J.T. Olson E.N. MicroRNA s in stress signaling and human disease Cell 2012 148 1172 1187 10.1016/j.cell.2012.02.005 22424228 27. Fleissner F. Goerzig Y. Haverich A. Thum T. Microvesicles as novel biomarkers and therapeutic targets in transplantation medicine Am. J. Transplant. 2012 12 289 297 10.1111/j.1600-6143.2011.03790.x 22082333 28. Weber J.A. Baxter D.H. Zhang S. Huang D.Y. Huang K.H. Lee M.J. Galas D.J. Wang K. The microRNA spectrum in 12 body fluids Clin. Chem. 2010 56 1733 1741 10.1373/clinchem.2010.147405 20847327 29. Chim S.S. Shing T.K. Hung E.C. Leung T.Y. Lau T.K. Chiu R.W. Lo Y.M. Detection and characterization of placental microRNAs in maternal plasma Clin. Chem. 2008 54 482 490 10.1373/clinchem.2007.097972 18218722 30. Lawrie C.H. Gal S. Dunlop H.M. Pushkaran B. Liggins A.P. Pulford K. Banham A.H. Pezzella F. Boultwood J. Wainscoat J.S. Detection of elevated levels of tumour-associated microRNA s in serum of patients with diffuse large B-cell lymphoma Br. J. Haematol. 2008 141 672 675 10.1111/j.1365-2141.2008.07077.x 18318758 31. Mitchell P.S. Parkin R.K. Kroh E.M. Fritz B.R. Wyman S.K. Pogosova-Agadjanyan E.L. Peterson A. Noteboom J. O’Briant K.C. Allen A. Circulating microRNAs as stable blood-based markers for cancer detection Proc. Natl. Acad. Sci. USA 2008 105 10513 10518 10.1073/pnas.0804549105 18663219 32. Turchinovich A. Weiz L. Burwinkel B. Extracellular mirnas: The mystery of their origin and function Trends Biochem. Sci. 2012 37 460 465 10.1016/j.tibs.2012.08.003 22944280 33. Ajit S.K. Circulating microRNAs as biomarkers, therapeutic targets, and signaling molecules Sensors 2012 12 3359 3369 10.3390/s120303359 22737013 34. Grasedieck S. Sorrentino A. Langer C. Buske C. Dohner H. Mertens D. Kuchenbauer F. Circulating microRNAs in hematological diseases: Principles, challenges, and perspectives Blood 2013 121 4977 4984 10.1182/blood-2013-01-480079 23550041 35. Arroyo J.D. Chevillet J.R. Kroh E.M. Ruf I.K. Pritchard C.C. Gibson D.F. Mitchell P.S. Bennett C.F. Pogosova-Agadjanyan E.L. Stirewalt D.L. Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma Proc. Natl. Acad. Sci. USA 2011 108 5003 5008 10.1073/pnas.1019055108 21383194 36. Turchinovich A. Weiz L. Langheinz A. Burwinkel B. Characterization of extracellular circulating microRNA Nucleic Acids Res. 2011 39 7223 7233 10.1093/nar/gkr254 21609964 37. Winter J. Jung S. Keller S. Gregory R.I. Diederichs S. Many roads to maturity: MicroRNA biogenesis pathways and their regulation Nat. Cell Biol. 2009 11 228 234 10.1038/ncb0309-228 19255566 38. Pegtel D.M. Cosmopoulos K. Thorley-Lawson D.A. van Eijndhoven M.A. Hopmans E.S. Lindenberg J.L. de Gruijl T.D. Wurdinger T. Middeldorp J.M. Functional delivery of viral miRNAs via exosomes Proc. Natl. Acad. Sci. USA 2010 107 6328 6333 10.1073/pnas.0914843107 20304794 39. Kosaka N. Iguchi H. Yoshioka Y. Takeshita F. Matsuki Y. Ochiya T. Secretory mechanisms and intercellular transfer of microRNAs in living cells J. Biol. Chem. 2010 285 17442 17452 10.1074/jbc.M110.107821 20353945 40. Mittelbrunn M. Gutierrez-Vazquez C. Villarroya-Beltri C. Gonzalez S. Sanchez-Cabo F. Gonzalez M.A. Bernad A. Sanchez-Madrid F. Unidirectional transfer of MicroRNA-loaded exosomes from T cells to antigen-presenting cells Nat. Commun. 2011 2 282 10.1038/ncomms1285 21505438 41. Trajkovic K. Hsu C. Chiantia S. Rajendran L. Wenzel D. Wieland F. Schwille P. Brugger B. Simons M. Ceramide triggers budding of exosome vesicles into multivesicular endosomes Science 2008 319 1244 1247 10.1126/science.1153124 18309083 42. Valadi H. Ekstrom K. Bossios A. Sjostrand M. Lee J.J. Lotvall J.O. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells Nat. Cell Biol. 2007 9 654 659 10.1038/ncb1596 17486113 43. Hunter M.P. Ismail N. Zhang X. Aguda B.D. Lee E.J. Yu L. Xiao T. Schafer J. Lee M.L. Schmittgen T.D. Detection of microRNA expression in human peripheral blood microvesicles PLoS ONE 2008 3 1232 10.1371/journal.pone.0003694 19002258 44. Turchinovich A. Samatov T.R. Tonevitsky A.G. Burwinkel B. Circulating mirnas: Cell-cell communication function? Front. Genet. 2013 4 119 10.3389/fgene.2013.00119 23825476 45. Zhang Y. Liu D. Chen X. Li J. Li L. Bian Z. Sun F. Lu J. Yin Y. Cai X. Secreted monocytic mir-150 enhances targeted endothelial cell migration Mol. Cell 2010 39 133 144 10.1016/j.molcel.2010.06.010 20603081 46. Collino F. Deregibus M.C. Bruno S. Sterpone L. Aghemo G. Viltono L. Tetta C. Camussi G. Microvesicles derived from adult human bone marrow and tissue specific mesenchymal stem cells shuttle selected pattern of miRNAs PLoS ONE 2009 5 1232 10.1371/journal.pone.0011803 20668554 47. Robbins P.D. Dorronsoro A. Booker C.N. Regulation of chronic inflammatory and immune processes by extracellular vesicles J. Clin. Investig. 2016 126 1173 1180 10.1172/JCI81131 27035808 48. De Candia P. De Rosa V. Casiraghi M. Matarese G. Extracellular RNAs: A secret arm of immune system regulation J. Biol. Chem. 2016 291 7221 7228 10.1074/jbc.R115.708842 26887954 49. Mathivanan S. Hong J. Simpson R.J. Exosomes: Extracellular organelles important in intercellular communication J. Proteom. 2010 73 1907 1920 10.1016/j.jprot.2010.06.006 20601276 50. Clotilde T. Matias O. Elodie S. Membrane vesicles as conveyors of immune responses Nat. Rev. Immunol. 2009 9 581 593 19498381 51. Vickers K.C. Palmisano B.T. Shoucri B.M. Shamburek R.D. Remaley A.T. microRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins Nat. Cell Biol. 2011 13 423 433 10.1038/ncb2210 21423178 52. Zernecke A. Bidzhekov K. Noels H. Shagdarsuren E. Gan L. Denecke B. Hristov M. Koppel T. Jahantigh M.N. Lutgens E. Delivery of microRNA -126 by apoptotic bodies induces CXCL12-dependent vascular protection Sci. Signal. 2009 2 ra81 10.1126/scisignal.2000610 19996457 53. Cocucci E. Racchetti G. Meldolesi J. Shedding microvesicles: Artefacts no more Trends Cell Biol. 2009 19 43 51 10.1016/j.tcb.2008.11.003 19144520 54. Muralidharan-Chari V. Clancy J.W. Sedgwick A. D’Souza-Schorey C. Microvesicles: Mediators of extracellular communication during cancer progression J. Cell Sci. 2010 123 1603 1611 10.1242/jcs.064386 20445011 55. Iveson-Iveson J. Anatomy and physiology: The accessory organs of digestion Nurs. Mirror 1979 148 24 25 254931 56. Chang J. Nicolas E. Marks D. Sander C. Lerro A. Buendia M.A. Xu C. Mason W.S. Moloshok T. Bort R. MiR-122, a mammalian liver-specific microRNA, is processed from HCR mRNA and may downregulate the high affinity cationic amino acid transporter CAT-1 RNA Biol. 2004 1 106 113 10.4161/rna.1.2.1066 17179747 57. Esau C. Davis S. Murray S.F. Yu X.X. Pandey S.K. Pear M. Watts L. Booten S.L. Graham M. McKay R. MiR-122 regulation of lipid metabolism revealed by in vivo antisense targeting Cell Metab. 2006 3 87 98 10.1016/j.cmet.2006.01.005 16459310 58. Marsit C.J. Eddy K. Kelsey K.T. microRNA responses to cellular stress Cancer Res. 2006 66 10843 10848 10.1158/0008-5472.CAN-06-1894 17108120 59. Deng X.G. Qiu R.L. Wu Y.H. Li Z.X. Xie P. Zhang J. Zhou J.J. Zeng L.X. Tang J. Maharjan A. Overexpression of miR-122 promotes the hepatic differentiation and maturation of mouse ESCs through a miR-122/FOXA1/HNF4A-positive feedback loop Liver Int. 2014 34 281 295 10.1111/liv.12239 23834235 60. Enache L.S. Enache E.L. Ramiere C. Diaz O. Bancu L. Sin A. Andre P. Circulating RNA molecules as biomarkers in liver disease Int. J. Mol. Sci. 2014 15 17644 17666 10.3390/ijms151017644 25272224 61. Wang J.H. Jiang D. Rao H.Y. Zhao J.M. Wang Y. Wei L. Absolute quantification of serum microRNA -122 and its correlation with liver inflammation grade and serum alanine aminotransferase in chronic hepatitis C patients Int. J. Infect. Dis. 2015 30 52 56 10.1016/j.ijid.2014.09.020 25461662 62. Cermelli S. Ruggieri A. Marrero J.A. Ioannou G.N. Beretta L. Circulating microRNA s in patients with chronic hepatitis C and non-alcoholic fatty liver disease PLoS ONE 2011 6 1232 10.1371/journal.pone.0023937 21886843 63. Zhang S. Ouyang X. Jiang X. Gu D. Lin Y. Kong S.K. Xie W. Dysregulated serum microRNA expression profile and potential biomarkers in hepatitis C virus-infected patients Int. J. Med. Sci. 2015 12 590 598 10.7150/ijms.11525 26283876 64. Bala S. Tilahun Y. Taha O. Alao H. Kodys K. Catalano D. Szabo G. Increased microRNA -155 expression in the serum and peripheral monocytes in chronic HCV infection J. Transl. Med. 2012 10 151 10.1186/1479-5876-10-151 22846613 65. Li L.M. Hu Z.B. Zhou Z.X. Chen X. Liu F.Y. Zhang J.F. Shen H.B. Zhang C.Y. Zen K. Serum microRNA profiles serve as novel biomarkers for HBV infection and diagnosis of HBV-positive hepatocarcinoma Cancer Res. 2010 70 9798 9807 10.1158/0008-5472.CAN-10-1001 21098710 66. Bihrer V. Friedrich-Rust M. Kronenberger B. Forestier N. Haupenthal J. Shi Y. Peveling-Oberhag J. Radeke H.H. Sarrazin C. Herrmann E. Serum miR-122 as a biomarker of necroinflammation in patients with chronic hepatitis C virus infection Am. J. Gastroenterol. 2011 106 1663 1669 10.1038/ajg.2011.161 21606975 67. Zhang H. Li Q.Y. Guo Z.Z. Guan Y. Du J. Lu Y.Y. Hu Y.Y. Liu P. Huang S. Su S.B. Serum levels of microRNAs can specifically predict liver injury of chronic hepatitis B World J. Gastroenterol. 2012 18 5188 5196 23066312 68. Winther T.N. Bang-Berthelsen C.H. Heiberg I.L. Pociot F. Hogh B. Differential plasma microRNA profiles in HBEAG positive and hbeag negative children with chronic hepatitis B PLoS ONE 2013 8 1232 10.1371/journal.pone.0058236 23469271 69. Xu J. Wu C. Che X. Wang L. Yu D. Zhang T. Huang L. Li H. Tan W. Wang C. Circulating MicroRNAs, miR-21, miR-122, and miR-223, in patients with hepatocellular carcinoma or chronic hepatitis Mol. Carcinog. 2011 50 136 142 10.1002/mc.20712 21229610 70. Qi P. Cheng S.Q. Wang H. Li N. Chen Y.F. Gao C.F. Serum microRNA s as biomarkers for hepatocellular carcinoma in chinese patients with chronic hepatitis B virus infection PLoS ONE 2011 6 1232 10.1371/journal.pone.0028486 22174818 71. Anadol E. Schierwagen R. Elfimova N. Tack K. Schwarze-Zander C. Eischeid H. Noetel A. Boesecke C. Jansen C. Dold L. Circulating microRNA S as a marker for liver injury in human immunodeficiency virus patients Hepatology 2015 61 46 55 10.1002/hep.27369 25125218 72. Wang K. Zhang S. Marzolf B. Troisch P. Brightman A. Hu Z. Hood L.E. Galas D.J. Circulating microRNAs, potential biomarkers for drug-induced liver injury Proc. Natl. Acad. Sci. USA 2009 106 4402 4407 10.1073/pnas.0813371106 19246379 73. Starkey Lewis P.J. Dear J. Platt V. Simpson K.J. Craig D.G. Antoine D.J. French N.S. Dhaun N. Webb D.J. Costello E.M. Circulating microRNAs as potential markers of human drug-induced liver injury Hepatology 2011 54 1767 1776 10.1002/hep.24538 22045675 74. Koberle V. Waidmann O. Kronenberger B. Andrei A. Susser S. Fuller C. Perner D. Zeuzem S. Sarrazin C. Piiper A. Serum MicroRNA-122 kinetics in patients with chronic hepatitis C virus infection during antiviral therapy J. Viral Hepat. 2013 20 530 535 10.1111/jvh.12075 23808991 75. Trebicka J. Anadol E. Elfimova N. Strack I. Roggendorf M. Viazov S. Wedemeyer I. Drebber U. Rockstroh J. Sauerbruch T. Hepatic and serum levels of miR-122 after chronic HCV-induced fibrosis J. Hepatol. 2013 58 234 239 10.1016/j.jhep.2012.10.015 23085648 76. Tan Y. Ge G. Pan T. Wen D. Chen L. Yu X. Zhou X. Gan J. A serum microRNA panel as potential biomarkers for hepatocellular carcinoma related with hepatitis B virus PLoS ONE 2014 9 1232 10.1371/journal.pone.0107986 25238238 77. Zhou J. Yu L. Gao X. Hu J. Wang J. Dai Z. Wang J.F. Zhang Z. Lu S. Huang X. Plasma microRNA panel to diagnose hepatitis B virus-related hepatocellular carcinoma J. Clin. Oncol. 2011 29 4781 4788 10.1200/JCO.2011.38.2697 22105822 78. Arataki K. Hayes C.N. Akamatsu S. Akiyama R. Abe H. Tsuge M. Miki D. Ochi H. Hiraga N. Imamura M. Circulating microRNA-22 correlates with microRNA-122 and represents viral replication and liver injury in patients with chronic hepatitis B J. Med.Virol. 2013 85 789 798 10.1002/jmv.23540 23508904 79. Giray B.G. Emekdas G. Tezcan S. Ulger M. Serin M.S. Sezgin O. Altintas E. Tiftik E.N. Profiles of serum microRNAs; miR-125b-5p and miR223–3p serve as novel biomarkers for hbv-positive hepatocellular carcinoma Mol. Biol. Rep. 2014 41 4513 4519 10.1007/s11033-014-3322-3 24595450 80. Tan Y. Ge G. Pan T. Wen D. Gan J. A pilot study of serum microRNAs panel as potential biomarkers for diagnosis of nonalcoholic fatty liver disease PLoS ONE 2014 9 1232 10.1371/journal.pone.0105192 25141008 81. Kent A.F. Researches on the structure and function of the mammalian heart J. Physiol. 1893 14 233 254 10.1113/jphysiol.1893.sp000451 82. Pleister A. Selemon H. Elton S.M. Elton T.S. Circulating miRNAs: Novel biomarkers of acute coronary syndrome? Biomark. Med. 2013 7 287 305 10.2217/bmm.13.8 23547823 83. Wang G.K. Zhu J.Q. Zhang J.T. Li Q. Li Y. He J. Qin Y.W. Jing Q. Circulating microRNA: A novel potential biomarker for early diagnosis of acute myocardial infarction in humans Eur. Heart J. 2010 31 659 666 10.1093/eurheartj/ehq013 20159880 84. Dimmeler S. Zeiher A.M. Circulating microRNAs: Novel biomarkers for cardiovascular diseases? Eur. Heart J. 2010 31 2705 2707 10.1093/eurheartj/ehq221 20605798 85. D’Alessandra Y. Devanna P. Limana F. Straino S. Di Carlo A. Brambilla P.G. Rubino M. Carena M.C. Spazzafumo L. De Simone M. Circulating microRNAs are new and sensitive biomarkers of myocardial infarction Eur. Heart J. 2010 31 2765 2773 10.1093/eurheartj/ehq167 20534597 86. Corsten M.F. Dennert R. Jochems S. Kuznetsova T. Devaux Y. Hofstra L. Wagner D.R. Staessen J.A. Heymans S. Schroen B. Circulating microRNA-208b and microRNA-499 reflect myocardial damage in cardiovascular disease Circ. Cardiovasc. Genet. 2010 3 499 506 10.1161/CIRCGENETICS.110.957415 20921333 87. Cheng Y. Tan N. Yang J. Liu X. Cao X. He P. Dong X. Qin S. Zhang C. A translational study of circulating cell-free microRNA-1 in acute myocardial infarction Clin. Sci. 2010 119 87 95 10.1042/CS20090645 20218970 88. Ai J. Zhang R. Li Y. Pu J. Lu Y. Jiao J. Li K. Yu B. Li Z. Wang R. Circulating microRNA-1 as a potential novel biomarker for acute myocardial infarction Biochem. Biophys. Res. Commun. 2010 391 73 77 10.1016/j.bbrc.2009.11.005 19896465 89. Adachi T. Nakanishi M. Otsuka Y. Nishimura K. Hirokawa G. Goto Y. Nonogi H. Iwai N. Plasma microRNA 499 as a biomarker of acute myocardial infarction Clin. Chem. 2010 56 1183 1185 10.1373/clinchem.2010.144121 20395621 90. Lippi G. Mattiuzzi C. Cervellin G. Circulating microRNAs (miRs) for diagnosing acute myocardial infarction: Meta-analysis of available studies Int. J. Cardiol. 2013 167 277 278 10.1016/j.ijcard.2012.09.152 23079084 91. Olivieri F. Antonicelli R. Lorenzi M. D’Alessandra Y. Lazzarini R. Santini G. Spazzafumo L. Lisa R. La Sala L. Galeazzi R. Diagnostic potential of circulating miR-122-499–5p in elderly patients with acute non ST-elevation myocardial infarction Int. J. Cardiol. 2013 167 531 536 10.1016/j.ijcard.2012.01.075 22330002 92. Li C. Pei F. Zhu X. Duan D.D. Zeng C. Circulating microRNAs as novel and sensitive biomarkers of acute myocardial infarction Clin. Biochem. 2012 45 727 732 10.1016/j.clinbiochem.2012.04.013 22713968 93. Fichtlscherer S. Zeiher A.M. Dimmeler S. Circulating microRNAs: Biomarkers or mediators of cardiovascular diseases? Arterioscler. Thromb. Vasc. Biol. 2011 31 2383 2390 10.1161/ATVBAHA.111.226696 22011751 94. Ji X. Takahashi R. Hiura Y. Hirokawa G. Fukushima Y. Iwai N. Plasma miR-122-208 as a biomarker of myocardial injury Clin. Chem. 2009 55 1944 1949 10.1373/clinchem.2009.125310 19696117 95. Zampetaki A. Willeit P. Drozdov I. Kiechl S. Mayr M. Profiling of circulating microRNAs: From single biomarkers to re-wired networks Cardiovasc. Res. 2012 93 555 562 10.1093/cvr/cvr266 22028337 96. Nassirpour R. Ramaiah S.K. Whiteley L.O. Nephron segment specific microRNA biomarkers of pre-clinical drug-induced renal toxicity: Opportunities and challenges Toxicol. Appl. Pharmacol. 2016 in press 10.1016/j.taap.2016.01.021 26827821 97. Willeit P. Zampetaki A. Dudek K. Kaudewitz D. King A. Kirkby N.S. Crosby-Nwaobi R. Prokopi M. Drozdov I. Langley S.R. Circulating microRNAs as novel biomarkers for platelet activation Circ. Res. 2013 112 595 600 10.1161/CIRCRESAHA.111.300539 23283721 98. Long G. Wang F. Duan Q. Chen F. Yang S. Gong W. Wang Y. Chen C. Wang D.W. Human circulating microRNA-1 and microRNA-126 as potential novel indicators for acute myocardial infarction Int. J. Biol. Sci. 2012 8 811 818 10.7150/ijbs.4439 22719221 99. Fukushima Y. Nakanishi M. Nonogi H. Goto Y. Iwai N. Assessment of plasma miRNAs in congestive heart failure Circ. J. 2011 75 336 340 10.1253/circj.CJ-10-0457 21157109 100. Wallace M.A. Anatomy and physiology of the kidney AORN J. 1998 68 799 820 10.1016/S0001-2092(06)62377-6 101. Sun Y. Koo S. White N. Peralta E. Esau C. Dean N.M. Perera R.J. Development of a micro-array to detect human and mouse microRNAs and characterization of expression in human organs Nucleic Acids Res. 2004 32 e188 10.1093/nar/gnh186 15616155 102. Liang M. Liu Y. Mladinov D. Cowley A.W. Jr. Trivedi H. Fang Y. Xu X. Ding X. Tian Z. microRNA: A new frontier in kidney and blood pressure research Am. J. Physiol. Ren. Physiol. 2009 297 F553 F558 10.1152/ajprenal.00045.2009 19339633 103. Tian Z. Greene A.S. Pietrusz J.L. Matus I.R. Liang M. MicroRNA-target pairs in the rat kidney identified by microRNA microarray, proteomic, and bioinformatic analysis Genome Res. 2008 18 404 411 10.1101/gr.6587008 18230805 104. Lorenzen J.M. Kielstein J.T. Hafer C. Gupta S.K. Kumpers P. Faulhaber-Walter R. Haller H. Fliser D. Thum T. Circulating miR-122-210 predicts survival in critically ill patients with acute kidney injury Clin. J. Am. Soc. Nephrol. CJASN 2011 6 1540 1546 10.2215/CJN.00430111 21700819 105. Wang J.F. Zha Y.F. Li H.W. Wang F. Bian Q. Lai X.L. Yu G. Screening plasma miRNAs as biomarkers for renal ischemia-reperfusion injury in rats Med. Sci. Monit. Int. Med. J. Exp. Clin. Res. 2014 20 283 289 106. Schena F.P. Serino G. Sallustio F. MicroRNAs in kidney diseases: New promising biomarkers for diagnosis and monitoring Nephrol. Dial. Transplant. 2014 29 755 763 10.1093/ndt/gft223 23787546 107. Kito N. Endo K. Ikesue M. Weng H. Iwai N. MiRNA profiles of tubular cells: Diagnosis of kidney injury BioMed Res. Int. 2015 2015 465479 10.1155/2015/465479 26106607 108. Lan Y.F. Chen H.H. Lai P.F. Cheng C.F. Huang Y.T. Lee Y.C. Chen T.W. Lin H. MicroRNA-494 reduces ATF3 expression and promotes AKI J. Am. Soc. Nephrol. JASN 2012 23 2012 2023 10.1681/ASN.2012050438 23160513 109. Hanke M. Hoefig K. Merz H. Feller A.C. Kausch I. Jocham D. Warnecke J.M. Sczakiel G. A robust methodology to study urine microRNA as tumor marker: MicroRNA-126 and microRNA-182 are related to urinary bladder cancer Urol. Oncol. 2010 28 655 661 10.1016/j.urolonc.2009.01.027 19375957 110. Williams A.E. Larner-Svensson H. Perry M.M. Campbell G.A. Herrick S.E. Adcock I.M. Erjefalt J.S. Chung K.F. Lindsay M.A. microRNA expression profiling in mild asthmatic human airways and effect of corticosteroid therapy PLoS ONE 2009 4 1232 10.1371/journal.pone.0005889 19521514 111. Wang Y. Weng T. Gou D. Chen Z. Chintagari N.R. Liu L. Identification of rat lung-specific microRNAs by micoRNA microarray: Valuable discoveries for the facilitation of lung research BMC Genom. 2007 8 1232 10.1186/1471-2164-8-29 17250765 112. Yang G. Yang L. Wang W. Wang J. Wang J. Xu Z. Discovery and validation of extracellular/circulating microRNAs during idiopathic pulmonary fibrosis disease progression Gene 2015 562 138 144 10.1016/j.gene.2015.02.065 25725128 113. Zhao Q. Cao J. Wu Y.C. Liu X. Han J. Huang X.C. Jiang L.H. Hou X.X. Mao W.M. Ling Z.Q. Circulating miRNAs is a potential marker for Gefitinib sensitivity and correlation with EGFR mutational status in human lung cancers Am. J. Cancer Res. 2015 5 1692 1705 26175938 114. Verhoeven C.J. Farid W.R.R. Ruiter P.E.E.D. Jonge J.D. Kwekkeboom J. Metselaar H.J. Tilanus H.W. Laan L.J.W.V.D. Kazemier G. MicroRNAs in preservation solution are more predictive of graft quality than their expression in liver tissue Liver Transplant. 2012 18 S116 115. Verhoeven C.J. Farid W.R. de Ruiter P.E. Hansen B.E. Roest H.P. de Jonge J. Kwekkeboom J. Metselaar H.J. Tilanus H.W. Kazemier G. MicroRNA profiles in graft preservation solution are predictive of ischemic-type biliary lesions after liver transplantation J. Hepatol. 2013 59 1231 1238 10.1016/j.jhep.2013.07.034 23928409 116. Shapiro M.D. Bagley J. Latz J. Godwin J.G. Ge X. Tullius S.G. Iacomini J. MicroRNA expression data reveals a signature of kidney damage following ischemia reperfusion injury PLoS ONE 2011 6 1232 10.1371/journal.pone.0023011 21887224 117. Godwin J.G. Ge X. Stephan K. Jurisch A. Tullius S.G. Iacomini J. Identification of a microRNA signature of renal ischemia reperfusion injury Proc. Natl. Acad. Sci. USA 2010 107 14339 14344 10.1073/pnas.0912701107 20651252 118. Kanak M.A. Takita M. Shahbazov R. Lawrence M.C. Chung W.Y. Dennison A.R. Levy M.F. Naziruddin B. Evaluation of microRNA375 as a novel biomarker for graft damage in clinical islet transplantation Transplantation 2015 99 1568 1573 10.1097/TP.0000000000000625 25769078 119. Poy M.N. Eliasson L. Krutzfeldt J. Kuwajima S. Ma X. Macdonald P.E. Pfeffer S. Tuschl T. Rajewsky N. Rorsman P. A pancreatic islet-specific microRNA regulates insulin secretion Nature 2004 432 226 230 10.1038/nature03076 15538371 120. Duong Van Huyen J.P. Tible M. Gay A. Guillemain R. Aubert O. Varnous S. Iserin F. Rouvier P. Francois A. Vernerey D. MicroRNAs as non-invasive biomarkers of heart transplant rejection Eur. Heart J. 2014 35 3194 3202 10.1093/eurheartj/ehu346 25176944 121. Fang Y. Shi C. Manduchi E. Civelek M. Davies P.F. MicroRNA-10a regulation of proinflammatory phenotype in athero-susceptible endothelium in vivo and in vitro Proc. Natl. Acad. Sci. USA 2010 107 13450 13455 10.1073/pnas.1002120107 20624982 122. Lind E.F. Ohashi P.S. Mir-155, a central modulator of t-cell responses Eur. J. Immunol. 2014 44 11 15 10.1002/eji.201343962 24571026 123. Suarez Y. Wang C. Manes T.D. Pober J.S. Cutting edge: Tnf-induced microRNAs regulate tnf-induced expression of e-selectin and intercellular adhesion molecule-1 on human endothelial cells: Feedback control of inflammation J. Immunol. 2010 184 21 25 10.4049/jimmunol.0902369 19949084 124. Zhang L. Zhou M. Wang Y. Huang W. Qin G. Weintraub N.L. Tang Y. MiR-122-92a inhibits vascular smooth muscle cell apoptosis: Role of the mkk4-jnk pathway Apoptosis Int. J. Programm. Cell Death 2014 19 975 983 10.1007/s10495-014-0987-y 24705900 125. Lorenzen J.M. Volkmann I. Fiedler J. Schmidt M. Scheffner I. Haller H. Gwinner W. Thum T. Urinary miR-210 as a mediator of acute t-cell mediated rejection in renal allograft recipients Am. J. Transplant. 2011 11 2221 2227 10.1111/j.1600-6143.2011.03679.x 21812927 126. Danger R. Pallier A. Giral M. Martinez-Llordella M. Lozano J.J. Degauque N. Sanchez-Fueyo A. Soulillou J.P. Brouard S. Upregulation of miR-122-142–3p in peripheral blood mononuclear cells of operationally tolerant patients with a renal transplant J. Am. Soc. Nephrol. JASN 2012 23 597 606 10.1681/ASN.2011060543 22282590 127. Lopez-Fraga M. Dominguez-Gil B. Capron A.M. Van Assche K. Martin D. Cozzi E. Delmonico F.L. A needed convention against trafficking in human organs Lancet 2014 383 2187 2189 10.1016/S0140-6736(14)60835-7 24976316 128. Reardon S. New life for pig-to-human transplants Nature 2015 527 152 154 10.1038/527152a 26560282 129. Ekser B. Ezzelarab M. Hara H. van der Windt D.J. Wijkstrom M. Bottino R. Trucco M. Cooper D.K. Clinical xenotransplantation: The next medical revolution? Lancet 2012 379 672 683 10.1016/S0140-6736(11)61091-X 22019026 130. Dai Y. Vaught T.D. Boone J. Chen S.H. Phelps C.J. Ball S. Monahan J.A. Jobst P.M. McCreath K.J. Lamborn A.E. Targeted disruption of the α1,3-galactosyltransferase gene in cloned pigs Nat. Biotechnol. 2002 20 251 255 10.1038/nbt0302-251 11875425 131. Phelps C.J. Koike C. Vaught T.D. Boone J. Wells K.D. Chen S.H. Ball S. Specht S.M. Polejaeva I.A. Monahan J.A. Production of alpha 1,3-galactosyltransferase-deficient pigs Science 2003 299 411 414 10.1126/science.1078942 12493821 132. Yang L. Guell M. Niu D. George H. Lesha E. Grishin D. Aach J. Shrock E. Xu W. Poci J. Genome-wide inactivation of porcine endogenous retroviruses (PERVS) Science 2015 350 1101 1104 10.1126/science.aad1191 26456528 133. Salomon D.R. A crispr way to block pervs—Engineering organs for transplantation N. Engl. J. Med. 2016 374 1089 1091 10.1056/NEJMcibr1515623 26981939 134. Jarchum I. Getting rid of pervs Nat. Biotechnol. 2016 34 46 10.1038/nbt.3448 26744982 135. Baker L.A. Lee K.C. Palacios Jimenez C. Alibhai H. Chang Y.M. Leckie P.J. Mookerjee R.P. Davies N.A. Andreola F. Jalan R. Circulating microRNAs reveal time course of organ injury in a porcine model of acetaminophen-induced acute liver failure PLoS ONE 2015 10 1232 10.1371/journal.pone.0128076 26018205 136. Reddy A.M. Zheng Y. Jagadeeswaran G. Macmil S.L. Graham W.B. Roe B.A. Desilva U. Zhang W. Sunkar R. Cloning, characterization and expression analysis of porcine microRNAs BMC Genom. 2009 10 1232 10.1186/1471-2164-10-65 19196471 137. Li Y. Li X. Sun W.K. Cheng C. Chen Y.H. Zeng K. Chen X. Gu Y. Gao R. Liu R. Comparison of liver microRNA transcriptomes of tibetan and yorkshire pigs by deep sequencing Gene 2016 577 244 250 10.1016/j.gene.2015.12.003 26656174 138. Li J. Chen Z. Zhao J. Fang L. Fang R. Xiao J. Chen X. Zhou A. Zhang Y. Ren L. Difference in microRNA expression and editing profile of lung tissues from different pig breeds related to immune responses to hp-prrsv Sci. Rep. 2015 5 9549 10.1038/srep09549 25856272 139. Xie S.S. Li X.Y. Liu T. Cao J.H. Zhong Q. Zhao S.H. Discovery of porcine microRNAs in multiple tissues by a solexa deep sequencing approach PLoS ONE 2011 6 1232 10.1371/journal.pone.0016235 21283541 140. Cho I.S. Kim J. Seo H.Y. Lim do H. Hong J.S. Park Y.H. Park D.C. Hong K.C. Whang K.Y. Lee Y.S. Cloning and characterization of microRNAs from porcine skeletal muscle and adipose tissue Mol. Biol. Rep. 2010 37 3567 3574 10.1007/s11033-010-0005-6 20180025 141. Mai M. Jin L. Tian S. Liu R. Huang W. Tang Q. Ma J. Jiang A. Wang X. Hu Y. Deciphering the microRNA transcriptome of skeletal muscle during porcine development PeerJ 2016 4 e1504 10.7717/peerj.1504 26793416 142. Mentzel C.M. Anthon C. Jacobsen M.J. Karlskov-Mortensen P. Bruun C.S. Jorgensen C.B. Gorodkin J. Cirera S. Fredholm M. Gender and obesity specific microRNA expression in adipose tissue from lean and obese pigs PLoS ONE 2015 10 1232 10.1371/journal.pone.0131650 26222688 143. Glazov E.A. Cottee P.A. Barris W.C. Moore R.J. Dalrymple B.P. Tizard M.L. A microRNA catalog of the developing chicken embryo identified by a deep sequencing approach Genome Res. 2008 18 957 964 10.1101/gr.074740.107 18469162 144. Berezikov E. Thuemmler F. van Laake L.W. Kondova I. Bontrop R. Cuppen E. Plasterk R.H. Diversity of microRNAs in human and chimpanzee brain Nat. Genet. 2006 38 1375 1377 10.1038/ng1914 17072315 145. Burnside J. Ouyang M. Anderson A. Bernberg E. Lu C. Meyers B.C. Green P.J. Markis M. Isaacs G. Huang E. Deep sequencing of chicken microRNAs BMC Genom. 2008 9 1232 10.1186/1471-2164-9-185 18430245 146. Kim J. Cho I.S. Hong J.S. Choi Y.K. Kim H. Lee Y.S. Identification and characterization of new microRNAs from pig Mamm. Genome 2008 19 570 580 10.1007/s00335-008-9111-3 18548309 147. Timoneda O. Balcells I. Nunez J.I. Egea R. Vera G. Castello A. Tomas A. Sanchez A. MiRNA expression profile analysis in kidney of different porcine breeds PLoS ONE 2013 8 1232 10.1371/journal.pone.0055402 23372853 148. Mas V.R. Dumur C.I. Scian M.J. Gehrau R.C. Maluf D.G. MicroRNAs as biomarkers in solid organ transplantation Am. J. Transplant. 2013 13 11 19 10.1111/j.1600-6143.2012.04313.x 23136949 149. Haider B.A. Baras A.S. McCall M.N. Hertel J.A. Cornish T.C. Halushka M.K. A critical evaluation of microRNA biomarkers in non-neoplastic disease PLoS ONE 2014 9 1232 10.1371/journal.pone.0089565 24586876

PMC005xxxxxx/PMC5000631.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081233ijms-17-01233CorrectionCorrection: Yan Chen, et al. Dual Agent Loaded PLGA Nanoparticles Enhanced Antitumor Activity in a Multidrug-Resistant Breast Tumor Eenograft Model. Int. J. Mol. Sci. 2014, 15, 2761–2772. Chen Yan 1†Zheng Xue-Lian 2†Fang Dai-Long 1Yang Yang 1Zhang Jin-Kun 1Li Hui-Li 1Xu Bei 1Lei Yi 1Ren Ke 3Song Xiang-Rong 1*1 State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China; [email protected] (Y.C.); [email protected] (D.-L.F.); [email protected] (Y.Y.); [email protected] (J.-K.Z.); [email protected] (H.-L.L.); [email protected] (B.X.); [email protected] (Y.L.)2 Key Laboratory of Obstetric & Gynecologic and Pediatric Diseases and Birth Defects of Ministry of Education, West China Second University Hospital, Sichuan University, Chengdu 610041, Sichuan, China; [email protected] Department of Pharmaceutical Sciences, University of Nebraska Medical Center, Omaha, NE 68198, USA; [email protected]* Correspondence: [email protected]; Tel./Fax: +86-28-8550-3817† These authors contributed equally to the work. 29 7 2016 8 2016 17 8 123327 7 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). ==== Body The authors wish to make a change to their published paper [1]. The title should read: “Dual Agent Loaded PLGA Nanoparticles Enhanced Antitumor Activity in a Multidrug-Resistant Breast Tumor Xenograft Model”. The authors apologize for any inconvenience the change may cause. The change does not affect the scientific results. The manuscript will be updated and the original will remain online on the article webpage. ==== Refs Reference 1. Chen Y. Zheng X.-L. Fang D.-L. Yang Y. Zhang J.-K. Li H.-L. Xu B. Lei Y. Ren K. Song X.-R. Dual Agent Loaded PLGA Nanoparticles Enhanced Antitumor Activity in a Multidrug-Resistant Breast Tumor Eenograft Model Int. J. Mol. Sci. 2014 15 2761 2772 10.3390/ijms15022761 24552875

PMC005xxxxxx/PMC5000632.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081234ijms-17-01234ReviewMolecular Mechanisms of Cutaneous Inflammatory Disorder: Atopic Dermatitis Kim Jung Eun 1Kim Jong Sic 2Cho Dae Ho 3Park Hyun Jeong 2*Jackson Chris Academic Editor1 Department of Dermatology, St. Paul’s Hospital, College of Medicine, The Catholic University of Korea, Seoul 02559, Korea; [email protected] Department of Dermatology, Yeouido St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 62 Yeouido-dong, Yeongdeungpo-gu, Seoul 07345, Korea; [email protected] Department of Life Science, SookmyungWomen’s University, Seoul 140-742, Korea; [email protected]* Correspondence: [email protected]; Tel.: +82-2-3779-1230; Fax: +82-2-783-760430 7 2016 8 2016 17 8 123431 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Atopic dermatitis (AD) is a multifactorial inflammatory skin disease resulting from interactions between genetic susceptibility and environmental factors. The pathogenesis of AD is poorly understood, and the treatment of recalcitrant AD is still challenging. There is accumulating evidence for new gene polymorphisms related to the epidermal barrier function and innate and adaptive immunity in patients with AD. Newly-found T cells and dendritic cell subsets, cytokines, chemokines and signaling pathways have extended our understanding of the molecular pathomechanism underlying AD. Genetic changes caused by environmental factors have been shown to contribute to the pathogenesis of AD. We herein present a review of the genetics, epigenetics, barrier dysfunction and immunological abnormalities in AD with a focus on updated molecular biology. atopic dermatitisgeneticsepigenomicsbarrierimmunologic abnormalities ==== Body 1. Introduction Atopic dermatitis (AD) is a common chronic inflammatory skin disease. The prevalence of AD in children is about 10%–20%, while the prevalence in adults is approximately 1%–3% worldwide, depending on the different ethnic populations [1]. It is known that most of AD-related genes do not follow Mendelian law, but are highly heritable. Thus, patients with familial history of AD have a higher risk of developing AD [2]. The prevalence of AD is higher in developed countries, such as those in Western Europe, and much lower in the countries specialized in agriculture, including China and Eastern Europe, rural Africa and Central Asia. This trend is consistent with the hygiene hypothesis [3]. Additionally, AD patients have various triggering factors and disease courses, which emphasize the influence of inter-individual differences. About 70% of patients with AD show elevated serum IgE levels with allergic sensitization and are categorized as having extrinsic AD, while other patients showing AD lesions with normal serum IgE levels are categorized as having intrinsic AD, although both subtypes share common clinical features [4]. Lesional distribution patterns vary depending on the patient’s age and disease activity. Whereas food allergens are the main triggering factors in pediatric AD, inhalant allergens are the main cause of AD exacerbation in adults [5]. The symptoms subside in many pediatric patients as they grow. However, some patients show persistent disease courses and tend to have concomitant allergic diseases, such as allergic rhinitis and asthma [6]. Generally diagnosis is made based on relevant clinical history and symptoms of the patients. The key clinical features of AD are pruritus and chronically-relapsing eczematous dermatitis that has a typical morphology and distribution distinct to individual age. These features can be used to distinguish AD from other clinical conditions, such as psoriasis and seborrheic dermatitis. Among various diagnostic criteria, Hanifin–Rajka’s criteria have been widely used [7]. The treatment strategy of AD mainly depends on the disease severity. At any stage, moisturizer should be properly used, and during flare-ups, topical and/or systemic immunomodulators can be used to control the conditions with different disease severity. The objective SCORing AD (SCORAD) index is widely used to assess AD severity [8]. Mild AD (SCORAD < 15) can be controlled by using moisturizer and topical anti-inflammatory agents, including topical steroids and calcineurin inhibitors [9]. The main treatment option of moderate (15 ≤ SCORAD < 40) to severe AD (SCORAD ≥ 40) includes systemic immunosuppressants, such as steroids, cyclosporine, azathioprine, methotrexate, interferon-gamma (INF-γ), intravenous immunoglobulin, allergen-specific immunotherapy and phototherapy. The known indication for the above therapies is the AD patients with sleep disorders, emotional stress or SCORAD index >40. Antihistamines and antimicrobial drugs can be used if needed. Although biologics have been tried to treat AD, the efficacy is limited when compared to that of psoriasis. Adjunctive treatment, such as evening primrose, probiotics and alternative medicines, can be tried in refractory AD patients [9]. However, the ultimate outcome of current treatment modalities is often not satisfactory in severe AD patients and has significant side effects. There is accumulating evidence that the heterogeneity of AD may result from the complex interactions between genetic susceptibility and environment, resulting in decreased skin barrier function, defects in innate immunity and aberrant immune responses to allergens and pathogens. This information offers the potential for individually-tailored therapeutic approaches. Genome-wide association studies (GWAS) and Immunochip analyses have identified several gene polymorphisms, susceptibility loci for AD and genetic changes caused by environmental factors that may be involved in the pathogenesis of AD. Newly-found T cells and dendritic cell (DC) subsets, cytokines, chemokines and signaling pathways have extended our understanding of the molecular pathomechanism of AD and modified the conventional concept of T helper type 1 (Th1)/T helper type 2 (Th2) imbalance paradigms [10,11,12,13]. Recent advances in the understanding of the pathomechanism of AD regarding the barrier dysfunction and immune dysregulation have led to the development of new therapeutic drugs of AD, and the efficacy and safety of these drugs are currently under the investigation. This review summarized the updated pathogenesis of AD with regard to genetics, epigenetics, epidermal barrier disruption and immunological dysregulation. 2. Genetics Many barrier and immune molecules related to the pathogenesis of AD showed potential genetic polymorphisms that may serve as protective or risk factors for AD (Table 1). Genes associated with epidermal barrier constituents and enzymes that maintain homeostasis have been linked to AD. Filaggrin gene (FLG) mutations are found in 10%–50% of patients with AD, but also in 9% of the normal population [14,15,16]. Loss-of-function mutations in FLG are well-known predisposing factors for AD and account for 40% of the total mutations in these patients. Whereas the frequency of null mutations of FLG is reportedly about 50% in European patients with moderate to severe AD, this frequency is much lower (<27%) in Asian patients with AD [17,18,19]. Only about 15% of patients with mild to moderate AD show the null mutation of FLG, even in European populations [17]. Therefore, FLG deficiency alone seems insufficient to explain epidermal barrier dysfunction in these patients. In addition to various mutations, copy number variation within FLG also increases the risk of AD [20,21], and the levels of filaggrin degradation products in the stratum corneum are correlated with the FLG genotype or copy number and AD severity [21,22]. Hornerin and FLG2 are fused S-100 proteins that are functionally related to FLG. Among patients with FLG2 mutations, African-Americans tend to develop more persistent AD [23]. A single-nucleotide polymorphism (SNP) in epidermal differentiation complex (EDC) on chromosome 1 and variants of C11orf30 and LRRC32 on chromosome 11 were found to be associated with AD [24]. Small proline-rich protein 3 (SPRR3) is a cornified envelope (CE) precursor protein. Several mutations related to SPRR3 overexpression in patients with AD are associated with decreased production of lipid levels and a thinner CE [25,26]. Mattrin encoded by transmembrane protein 79 (TMEM79) regulates lamellar body secretion. The missense mutation rs6684514 of TMEM79 reportedly impairs lamellar body secretion in patients with AD [27,28,29]. Serine protease inhibitor Kazal-type 5 (SPINK5) encodes the protease inhibitor lymphoepithelial Kazal-type-related inhibitor (LEKTI), which counteracts the activity of epidermal proteases, such as kallikrein-5 in the epidermis. Loss-of-function mutations in SPINK5 are known to cause Netherton syndrome, which has AD manifestations. The rs2303070 T allele of SPINK5 is a risk factor for AD in Taiwanese populations [30]. The E420K LEKTI variant is associated with an increased risk of developing AD through increased TSLP expression and barrier permeability by enhancement of epidermal protease activity and profilaggrin proteolysis [31]. The claudins are key adhesion molecules comprising tight junctions. The claudin-1 (CLDN1) gene haplotype-tagging SNP reportedly has associations with AD in North American populations [32]. The rs9290929 polymorphism, located in CLDN1, reduced the expression of CLDN1 and enhanced the production of IgE after mold exposure [33]. Several gene polymorphisms related to innate and adaptive immunity have been found in patients with AD. These include mutations in pathogen-associated molecular patterns, such as toll-like receptor (TLR) and nucleotide-binding oligomerization domain receptors (NOD) and antimicrobial peptides (AMPs), TSLP and the receptor for TSLP (TSLPR), IL-1 family cytokines and receptor genes, vitamin D pathway genes, the nerve growth factor pathway, Th2 and other cytokine genes and the genes encoding the high affinity IgE receptor (FcεRI, FCER1A) [24,34,35,36,37,38,39]. About 12% of patients with AD have the (TLR)2 R753Q mutation, and this mutation is associated with the severe AD phenotype and concomitant atopic diseases in certain populations [40,41,42,43]. TLR2 A-16934T, TLR4 D299G and A-896G mutations are also associated with severe AD [43,44]. The TLR9 promoter polymorphism, C-1237T, has been reported in patients with intrinsic AD [45]. SNPs in NOD-like receptor 1 genes related to caspase recruitment domain (CARD)4, CARD12, CARD15, NACHT, LRR and PYD domain-containing protein (NALP)1, NALP12 and NOD1 have been associated with AD [7,46]. Several SNPs of the human β-defensin (hBD) 1 gene were found to be linked to severe AD with allergic sensitization [47]. TSLP plays a crucial role in DC-driven Th2 responses. A recent study showed that patients with AD with a certain TSLP polymorphism showed eczema herpeticum [48,49]. IL-1 family cytokines play important roles in innate immune responses in patients with AD. While some variants of the IL-18 gene and the receptors (IL18RA) are associated with AD [35,50], rs1946518 and rs187238 polymorphisms in the IL-18 gene were shown to be the protective factors against the development of AD [51]. Genetic variants in IL-12 and IL-12RB, IFN-γ genes (IFNG) and IFNGR1 leading to partial IFNGR1 deficiency are related to AD in patients susceptible to eczema herpeticum [52,53]. Whereas CYP27A1 variants in the vitamin D pathway genes showed a protective effect, other variants are associated with severe AD with eosinophilia and high IgE levels [35,54,55,56]. Several distinct polymorphisms of IL-4, IL-13 and IL-31 and their receptors were found to influence AD predisposition [34]. Genetic differences in the genes encoding IL-4 and IL-13 were suggested to alter transcriptional activity. The signal transducer and activator of transcription 6 (STAT6) is a key transcription factor in responses mediated by IL-4 and IL-13. STAT6 variants were found to be associated with AD [57,58]. Neonates with the rs324011 polymorphism in the STAT6 had a lower risk of AD as they showed a reduced number of regulatory T cells (Tregs) and an increased Th1 response at birth [59]. A common haplotype encoding IL-31 was shown to be a risk factor for intrinsic AD. The rs7977932 G allele of IL31 variants was shown to be a risk factor for AD in Taiwanese populations [30]. The haplotype AAA or GAA of IL-31 was correlated with increased serum levels of IL-31 and severe pruritus in certain populations [60]. The AA genotype of IL-17A was found to be a predisposing factor of severe AD with concomitant asthma [61]. Other cytokine variants were also identified in patients with AD, including IL-2, IL-5, IL-6, IL-7, IL-9 and IL-10 [50,53,62,63,64,65,66,67,68]. Polymorphisms of regulated on activation, normal T cell expressed and secreted (RANTES) and eotaxin were associated with allergen sensitization [69,70,71]. A haplotype variant of the histamine 4 receptor (H4R) and a copy number variation were found to be associated with AD [72,73]. An SNP of FCER1A was reported to be associated with AD in Asians with an elevated serum IgE level [74,75,76]. The T allele of brain-derived neurotrophic factor gene polymorphism in C270T is associated with intrinsic AD and male sex. Serum brain-derived neurotrophic factor levels were reported to be correlated with the severity of intrinsic AD [77]. Barrier strengthening therapy to improve the barrier defect can be achieved by the proper use of moisturizer. Biologics treatment targeting Th2 immunity could be the best therapeutic option in the future. Duplimumab, anti-IL-4Ra monoclonal antibodies (mAb), has shown promising therapeutic responses in phase III clinical trials [78]. Anti-IL-13 mAb (lebrikizumab and tralokinumab) is currently in phase II clinical trials for AD [79]. Anti-IL-22 mAb (ILV-094), anti-IL-31 (BMS-981164) and anti-IL-31R (CIM331) mAb, anti-TSLP (AMG 157) and anti-TSLPR (MK-8226) mAb have been developed and are currently in phase II, I, II, I and I clinical trials, respectively [80]. OC000459 and other several small molecules that antagonize the chemoattractant receptor-homologous molecule expressed on Th2 cells have been in a phase II clinical trial of AD patients [80]. 3. Epigenetics The modern lifestyle and other environmental factors, such as air pollutants and tobacco smoke, have been suggested to be responsible for the high prevalence of AD since the advent of industrialization. There is accumulating evidence that epigenetic changes in response to these environmental factors contribute to the pathogenesis of AD. These epigenetic mechanisms include DNA methylation, histone modification and microRNA (miR) responsible for barrier function and immunological regulation [81]. Prenatal tobacco smoke exposure is correlated with high-level miR-223 expression and DNA methylation of the FOXP3 locus in cord blood, which are associated with lower Treg numbers. Infants with lower Treg numbers in cord blood at birth had a higher risk of AD during the first three years of life [82,83]. The potential risks of AD are determined by both the amount of exposure to or composition of the pollutants and the genetic susceptibility of the host [84]. A Taiwanese birth cohort revealed that a gene polymorphism related to a deficiency of the antioxidant enzyme glutathione-S-transferase may explain the individual differences in susceptibility to AD after prenatal tobacco smoke exposure [85]. DNA demethylation of a specific regulatory region of the TSLP gene was significantly associated with TSLP overexpression in lesional skin of patients with AD [81]. Methylation of the TSLP 5′-CpG island was significantly linked with prenatal smoke exposure. The lower degree of such methylation in cord blood leads to TSLP overexpression and subsequent development of AD [86]. DNA methylation of genes related to FcεRI and IgE production may modify allergic sensitization in certain patients with AD. In AD-affected patients with high IgE levels, the levels of DNA cytosine methyltransferase 1 transcripts were significantly decreased in peripheral blood mononuclear cells [87]. Overexpression of FcεRI on monocytes and DCs in patients with AD was shown to be due to the demethylation of specific regions within the FCER1G locus [88]. Some miRs that are upregulated or downregulated in AD-induced lesions have been identified and involved in the pathogenesis of AD. Exposure to relevant allergens could induce miR-155 expression in AD lesions. Increased miR-155 downregulates cytotoxic T lymphocyte-associated antigen, a negative regulator of T cell function, which in turn stimulates T cell proliferation and leads to a sustained inflammatory state [89]. miR-155 also positively modulates the differentiation and function of T helper type 17 (Th17) cells and is correlated with AD severity [90]. Increased miR-146a expression has been reported in the lesional skin of patients with AD. miR-146a can alleviate AD inflammation by inhibiting nuclear factor κ B-mediated proinflammatory cytokines and chemokines [91]. The forced expression of miR-143 reversed IL-13-induced inhibition of epidermal differentiation by blocking IL-13Rα1. Thus, miR-143 may be a potential therapeutic target in AD [92]. However, evidence regarding epigenetics responsible for AD is still currently limited, and further studies are needed to clarify the gene–environmental interactions and potential therapeutic targets. 4. Barrier Dysfunction Impaired epidermal barrier function in AD is characterized by abnormalities in skin microenvironment, gene functioning epidermal structural proteins, such as filaggrin and claudin, and lipid synthesis. The disrupted barrier causes an increased trans-epidermal water loss and enables the capture of more allergens, thus promoting allergic sensitization and initiation or exacerbation of AD inflammation. Impaired barrier function causes increased IL-1 release from keratinocytes, which activates the vascular endothelium to induce adhesion molecule expression and promotes cutaneous inflammation [12,93]. Epicutaneous sensitization to allergens causes heightened allergic immune responses and serves as a predisposing factor for a more severe allergic march. The use of soap and detergent raises skin pH in AD patients, which induces the imbalance between serine proteases and protease inhibitors. The activities of endogenous and exogenous proteases from house dust mites or S. aureus are increased in AD lesions. Lack of endogenous protease inhibitor activity accelerates barrier permeability and inflammation. Increases in pH and in serine protease activity result in increased microbial colonization and accelerated degradation of the ceramide synthesis enzymes [94]. In a LEKTI-knockout mouse model, increased kallikrein-5 stimulated proteinase-associated receptor-2 (PAR2), which activates nuclear factor κ B-induced overexpression of thymic stromal lymphopoietin (TSLP) and induces pruritus [95]. TSLP also suppresses the expression of EDC proteins, such as filaggrin, by activating STAT3 and extracellular signal-regulated kinase (ERK) signaling in keratinocytes [96]. Phosphodiesterase 4 (PDE4), which catalyzes the conversion of cyclic adenosine 3′,5′-monophosphate (cAMP) to 5′-AMP, contributes to the pathogenesis of AD via PAR2. Anti-PDE4 agents could theoretically inhibit PAR2 and leukotriene B4 production mediated by increased cAMP levels, thus relieving pruritus in AD [97]. However, the oral anti-PDE4 apremilast and topical anti-PDE4 crisaborole showed limited efficacy in treating AD [98,99]. Filaggrin, a key protein in the skin barrier, is involved in cornification and hydration. FLG deficiency is known to increase [13] and impair skin integrity, hydration, protease activity and AMP function [12]. Reduced levels of hornerin and filaggrin-2 expression were suggested to be related to abnormal cornified envelope (CE) formation in AD skin [100]. In addition to filaggrin, loricrin is a component of CE, and decreased loricrin levels are observed in patients with AD. Filaggrin expression is restored after topical anti-inflammatory treatment with calcineurin inhibitors or corticosteroids [12,101]. A recent study demonstrated that JTC801, a new synthetic compound, increased filaggrin expression in human keratinocytes in vitro and decreased the development of AD-like lesions in mice in vivo [102]. The importance of barrier strengthening therapy is supported by the outcome that a topical recombinant filaggrin delivery through cell penetrating peptide could restore the AD-like inflammation in filaggrin knockout mice [103]. Defects of tight junction proteins, such as claudin, could have a permissive effect on the entry of irritants, allergens or pathogens into the epidermis [33,104]. The claudin-1 level showed an inverse correlation with a high serum IgE level and eosinophilia. Claudin-1 expression was significantly suppressed by IL-4, IL-13 and IL-31 in a human skin equivalent [105]. While claudin-1 knockout is lethal, low claudin-1-expressing conditioned mice exhibited AD-like dermatitis and an increased recruitment of neutrophil and macrophage in the skin. TLR2 activation increased the expression of tight junction proteins, including claudin-1, in human keratinocytes. TLR2-deficient mice showed slow and incomplete barrier recovery by suppressing the tight junction proteins [106]. Patients with AD show impaired skin integrity and subclinical inflammation even in uninvolved skin. Consistent with these features, markedly reduced levels of filaggrin, filaggrin-2 and claudin-1 expression were observed not only in lesional AD skin, but also in non-lesional skin in patients with AD [107]. Lipids in the stratum corneum comprise ceramides, free fatty acids (FFAs) and cholesterol. An overall reduction in lipid levels, especially in the ceramide content, and a reduced ceramide chain length are observed in patients with AD and are associated with the severity of AD [108]. A decrease in the FFA chain length and an increase in the proportion of unsaturated FFAs have been found in patients with AD [109]. A decreased lipid content leads to a less compact lipid organization and defective skin barrier function [109]. Th2-dominant cytokine profiles in AD further contribute to the decrease in ceramide and long-chain FFA levels [28,110]. 5. Immunological Abnormalities 5.1. Innate Immunity Patients with AD develop recurrent skin infections. Early studies suggested that suppressed levels of AMPs, such as hBD-2, hBD-3 and cathelicidin, in AD-affected skin compared to the skin of patients with psoriasis or healthy subjects, are responsible for this susceptibility to infection [111]. IL-4, IL-13, IL-10 and IL-33 could suppress the expression of hBD-2 and hBD-3 [112], contributing to superinfection [110]. In contrast, recent data showed that the levels of AMPs in AD lesions are increased as much as those in healthy subjects, but are still insufficient to defend against S. aureus infection, possibly because of the huge amount of S. aureus colonization or functional defects in the AMPs [113]. Members of the S100 protein family (S100A7, S100A8 and S100A9) function as AMPs, as well as damage-associated molecular pattern molecules, which have proinflammatory activities. The levels of S100 proteins are increased in patients with acute and chronic AD, and their proinflammatory properties may result in defects of epidermal differentiation and cutaneous inflammation [113,114]. TLRs play important roles in innate immunity by recognizing PAMP and antimicrobial defenses. The activation of TLRs induces the expression of antimicrobial effector molecules and the release of various proinflammatory and immunomodulatory cytokines, which lead to the activation of adaptive immune responses. In AD, enhanced Th2 cytokines were found to downregulate the expression of TLRs, which renders AD skin more susceptible to skin infections. Several TLR polymorphism have been reported to be associated with AD. Monocytes from patients with AD with the TLR2 R753Q mutation show enhanced IL-6 and IL-12 production and downregulated CD36 expression. These abnormalities cause impaired TLR2/TLR6 heterodimer-CD36 complex internalization, leading to increased susceptibility to Staphylococcus aureus (S. aureus) infection [115]. Vitamin D contributes to cathelicidin and lipid synthesis, and the efficacy of vitamin D supplementation has been suggested in small studies [116,117]. Several SNP polymorphism in vitamin D could affect TLR activity by inhibiting the cathelicidin expression [55,56]. The main sources of TSLP are epidermal keratinocytes, and the TSLP level is increased in the epidermis of AD lesions. TSLP is involved in the activation of Langerhans cells (LCs) and DCs to induce Th2 immune responses. TSLP showed a positive correlation with IL-31 and IL-33 and has been suggested as new biomarker in AD [118]. 5.2. Adaptive Immunity 5.2.1. Th1/Th2 Imbalance Skin-homing memory T cells are key players of immune dysregulation in the pathogenesis of AD. Traditionally, the main pathogenesis of AD has been interpreted as immune dysregulation with predominant Th2 cytokines, such as IL-4, IL-5 and IL-13. As new T cell subsets were identified, the view of AD was modified as a Th2/Th22 polarized environment. Acute lesions of AD are driven by Th2 and Th22 responses, while chronic lesions are driven by a Th1 response. A recent study confirmed AD to be a result of Th2-skewed immune response, specifically the ratio of Th1:Th2 of chronic AD was 0.09. Consistently, most of the CD3+ T cells in biopsy specimens from chronic AD lesions were comprised of Th2 (64.6%), followed by Th17 (30.4%), Th22 (3.3%) and Th1 cells (4.8%) [119]. The Th2 cytokines IL-4 and IL-13 have a permissive effect on microbial invasion and epidermal barrier disruption by inhibiting AMP production [120], reducing lipid production in the stratum corneum and inducing spongiosis [80]. Th2 cytokines downregulate the expression of the major EDC genes, including FLG, LOR and involucrin [121,122,123,124], independent of the FLG genotype [125], and suppress keratinocyte differentiation via STAT3. Topical treatment with the Janus kinase (JAK) inhibitor downregulated STAT3 activation and restored skin barrier function by inducing terminal differentiation in AD animal models [126]. Topical and oral JAK inhibitors, such as tofacitinib, baricitinib and PF-04965842, are in phase II trials for AD [79]. IL-4 and IL-13 stimulate keratinocytes to express TSLP, which serves as a link between barrier defects and Th2 polarization. IL-4-overexpressing transgenic mice develop AD-like lesions. IL-5 attracts eosinophils into chronic AD lesions. Th2 cytokines IL-4 promote immunoglobulin switching in B cells, resulting in IgE synthesis; induce the expression of adhesion molecules; and recruit various immune cells into skin. With regard to biological agents targeting the Th2 response, several clinical trials have confirmed the therapeutic efficacy and safety of the anti-IL-4 receptor antibody duplimumab [127]. Levels of the Th1 chemokine CCL20 are increased in patients with chronic AD [71]. Among the Th1 cytokines, IL-1α, IL-2 and TGF-β were found to be decreased in patients with AD [71], whereas IFN-γ, IL-12 and granulocyte monocyte-colony stimulating factor were elevated in patients with chronic AD. IFN-γ is known to activate keratinocytes and induce their apoptosis. Granulocyte monocyte-colony-stimulating factor is known to prolong the survival of monocytes and induce persistent inflammation. IL-11 and TGF-β1 are associated with tissue remodeling in chronic AD [128]. Treg cells are known to suppress both Th1 and Th2 immune responses and to be deficient in AD lesions. 5.2.2. Th17 Cells Th17 cells contribute to the onset of acute AD, although their role in AD is relatively small in psoriasis [12]. Th17 cells produce IL-17 and IL-22, which induce the production of S100 proteins, the AMPs in keratinocytes and various proinflammatory cytokines. IL-17 may have a role in the differentiation of Th2 cells [80]. IL-17 level decreases gradually in chronic AD lesions as Th2 cytokines inhibit IL-17 production. The relative absence of IL-17 in AD lesions may be related to reduced AMP levels and may explain the increased susceptibility to skin infection in patients with AD [129]. IL-17 is responsible for eosinophil and neutrophil-mediated inflammation [130]. The number of Th17 cells and IL-17 expressed in AD lesions and serum is correlated with disease severity [12,131], and this correlation is more prominent in intrinsic AD. A recent study comparing the transcriptome between intrinsic and extrinsic AD indicated that patients with intrinsic AD showed greater Th17 and Th22 immune activation than did those with extrinsic AD [132]. Compared to European and American AD patients, Asian patients presented more epidermal hyperplasia, parakeratosis and stronger Th17 and Th2 activation resembling psoriasis, even in patients with extrinsic AD [133]. However, another study in the European American population found that a number of Th17 cells and IL-17 expression level were reduced in the AD patients with severe symptoms, whereas those of Th2 and Th22 cell subsets were correlated with disease severity [134]. This discrepancy may result from the heterogeneous characteristics of the patients, such as phenotype, disease duration and ethnic differences. 5.2.3. Th22 T Cells The original concept of AD cannot account for the epidermal hyperplasia in patients with chronic AD and has some limitations. Th22 cells produce IL-22, which is responsible for skin barrier impairment and epidermal hyperplasia. Filaggrin expression can be modulated in AD by both the Th2 and Th22 cytokine milieu. Similar to Th2 cytokines, IL-22 compromises the epidermal barrier by suppressing major terminal differentiation proteins [135]. IL-22 increases S100A7, S100A8 and S100A9 gene expression, thus inhibiting epidermal differentiation by enhancing IL-6 secretion, and exerts a proinflammatory effect in AD lesions [114,121]. IL-22 and IL-17 synergistically increase the levels of S100 proteins and AMPs in the epidermis [12,136]. IL-22 secretion can be induced immediately in response to staphylococcal exotoxins and house dust mites and can potentially amplify chronic skin inflammation in patients with AD [137,138]. IL-22 can also stimulate CCL17 production from human keratinocytes and promote the migration of T cells into the skin [138]. Upregulated IL-22 in chronic AD induces matrix metalloproteinase-3, a marker of remodeling; stromelysin-1; platelet-derived growth factor A; and CXCL5 (chemokine, CXC motif, ligand 5) [139]. IL-22 promotes the migration of keratinocytes through matrix metalloproteinases-1 and -3 and is involved in epidermal hyperplasia [136]. IL-22 binds to its receptors in the form of an IL-22R1/IL-10R2 complex and activates the JAK-STAT signaling pathway via STAT3 activation [140,141]. IL-22 expression can be enhanced by the activated Notch signaling pathway without activation of STAT3 [142]. The role of Th22 cells varies depending on the patient’s age and disease severity. Infants with AD show only a Th2/Th1 cell imbalance, whereas adults with AD exhibit Th22/Tc22 cell subsets [143]. The numbers of Th22 cells are correlated with disease activity. Moreover, the skin-homing T cell population of patients with severe AD mostly comprises circulating Th2 and Th22 cells, rather than Th17 cells in European Americans [134]. The role of anti-IL-22 monoclonal antibody is currently being investigated in a phase II trial [80]. 5.2.4. IL-18 IL-18 is a Th1-like cytokine, which is produced by various cells. Keratinocytes and mast cells produce IL-18 in response to exposure to allergens or pathogens, such as house dust mites and S. aureus [144,145]. Cytotoxic T cells release perforin and granzyme B after the recognition of viral infection in keratinocytes, then subsequently activate pro-IL-18 [146]. Inflammatory dendritic epidermal cells (IDECs) and monocyte-derived DCs also release IL-18 [147]. IL-18 stimulates mast cells to release chymase, which in turn cleaves and activates pro-IL-18. IL-18 stimulates either Th2 or Th1 cytokines based on IL-12. In acute AD lesions, IL-18 stimulates basophils, mast cells and CD4+ T cells to produce Th2 cytokines without IL-12. In chronic AD lesions, IL-18 stimulates Th1 cells to produce IFN-γ with IL-12 [148]. IL-18 contributes to CD4+ T cell- and natural killer T cell-dependent IgE production [146,149]. Corticotrophin-releasing hormone, a skin hypothalamic-pituitary axis molecule, decreases IL-18 expression in monocyte-derived DCs independent of corticotropin-releasing hormone receptors in patients with AD [150]. This suggests that stress may aggravate AD symptoms by modulating the expression of IL-18. IL-18 was shown to be a key player in the development of spontaneous AD-like lesions in a mouse model under pathogen-free conditions [151]. Recently, IL-18 levels in the maternal serum and cord blood were found to be a predisposing factor of childhood AD [152]. Elevated levels of serum IL-18 and IL-18 receptors have been suggested as biomarkers of AD severity [146,153], but additional research is required to confirm this suggestion [154,155]. 5.2.5. IL-31 IL-31, a new Th2 cytokine, is a major pruritogenic inflammatory substance in AD [156]. IL-31 protein and mRNA levels are elevated in AD lesions. Serum levels of IL-31 are proportional to disease severity in patients with AD [157]. IL-31 was found to inhibit epidermal terminal differentiation and enhance proinflammatory cytokine secretion [158,159]. IL-31 can also compromise epidermal barrier function by affecting epidermal terminal differentiation and lipid constituents, and it may inhibit terminal differentiation by downregulating filaggrin and loricrin expression [80,159] IL-31 also decreases ceramide production, increases long-chain FFAs, but decreases ester-linked ω-hydroxy ceramides in stratum corneum lipids [159]. IL-31 functions after binding to a heterodimeric receptor consisting of receptors for IL-31 (IL-31RA) and the oncostatin M receptor β complex. IL-31RA is found on keratinocytes, macrophages, eosinophils and nerve fibers in AD and in the neurons of dorsal root ganglia in healthy subjects [160,161]. IL-31/IL-31RA complexes activate signal transduction cascades, such as the Janus kinase-STAT (JAK-STAT), mitogen-activated protein kinase and phosphatidyl-inositol 3-kinase pathways [160]. The IL-31/IL-31RA complex stimulates the growth and branching of the nerve in sensory neurons via the STAT3 pathway and not via transient receptor potential cation channel vanilloid subtype 1 channels (TRPV1) [162]. A recent study demonstrated the safety and efficacy of a humanized anti-human IL-31RA monoclonal antibody in relieving pruritus in AD [163]. This suggests that the pruritogenic effect of IL-31 is mediated by IL-31RA. Activation of IL-31RA in keratinocytes induces calcium influx and produces more β-endorphin in STAT3-dependent pathways [164]. Meanwhile, a recent study suggested that the pruritogenic effect of IL-31 was mediated indirectly via keratinocytes and secondary pruritogenic substances, rather than through its receptors on cutaneous nerves [165]. Low doses of IL-31 promote the antimicrobial barrier, and thus, complete inhibition of IL-31 signaling may be undesirable [166]. 5.2.6. B cells Although T cells are key players in the pathogenesis of AD, B cells are also found in the dermis of AD lesions and play significant roles. B cells can present antigens to CD4+ T cells and activate T cells. B cells interact with T cells via MHC class II molecule/T cell receptor and co-stimulatory molecules, as well as their ligands. Co-stimulatory molecules include CD40 and CD80/CD86 on B cells and their ligands, CD40L and CD28 on T cells. The expression of costimulatory molecule CD86 was reported to be increased on B cells in AD [167]. Th2 cytokine IL-4 promotes immunoglobulin switching in B cells, resulting in IgE synthesis, which then induces the expression of adhesion molecules and recruits various immune cells into skin. B cells also produce chemokines CCL17, CCL22 and IL-16, attracting T cells to AD lesions. The levels of IgE are significantly increased in AD patients along with allergic sensitization. IgE contributes to IgE-mediated inflammation by stimulating FcεRI-expressing cells, such as mast cells and basophils. Additionally, IgE also contributes to autoallergic inflammation in a certain subset of AD patients. Serum IgE autoantibodies to epidermal self-antigens, such as Homo sapiens antigens 1, had a 38.2% prevalence in AD patients of autoreactivity, while patients in the control group showed a 6.4% prevalence [168]. However, the role of IgE in pathogenesis of AD is not major, and previous attempts to inhibit IgE with omalizumab, respectively, showed heterogeneously different therapeutic efficacy in AD patients [169]. Although B cell-targeting therapy with rituximab (anti-CD20 antibody) showed some improvement in AD patients, it was conducted only in a case series [170]. 5.2.7. Dendritic Cells (DCs) DCs play a key role in antigen uptake and presentation and induce a Th2-predominant response. Several types of DCs have been shown to express FcεRI in patients with AD [13]. These DCs are easily stimulated and activated by allergens, pathogens, irritants and scratching. Whereas LCs are the predominant DC type in non-lesional AD skin, inflammatory subtype DCs are increased in AD lesions [171]. IDECs have an increased capacity for antigen presentation and T cell activation. Topical anti-inflammatory treatment showed a reduced number of inflammatory DC subtypes in AD lesions [172]. On the other hand, the low number of plasmacytoid DCs and reduced amount of type 1 IFN in the lesional epidermis of patients with AD have been suggested to be responsible for the high susceptibility to viral skin infections in these patients [173]. TSLP is abundantly expressed in the epidermis of AD lesions. TSLP induces DCs to express OX40 ligand, which is essential for the differentiation of Th2 cells in the lymph nodes [174]. Activated DCs migrate to the lymph nodes and prime naive T cell differentiation into Th2 cells in the presence of IL-4 [174]. Expression of the IFN-γ receptor on epidermal DCs and their response to IFN-γ are attenuated in patients with AD [175]. DCs produce IL-25, which induces Th2 cell differentiation and suppresses filaggrin expression in vitro [12]. Activated epidermal LC and dermal DCs release CCL17 and CCL22, respectively, which in turn recruit and expand Th2 cells [176]. Chemokines (i.e., CCL17, CCL18, CCL19, CXCL9, CXCL10 and CXCL11) expressed by DCs induce further attraction and activation of various inflammatory cells [121]. Epidermal LCs and dermal DCs also induce the differentiation of Th22 cells, and IL-6 and TNF-α released from DCs have been suggested as contributing factors [177]. Plasmacytoid DCs also stimulate naive T cells to differentiate into Th22 cells in chronic AD lesions. IDECs release the proinflammatory cytokines IL-12 and IL-18, which promote Th1 responses in chronic AD. H4R is expressed by LCs, IDECs, plasmacytoid DCs and 6-sulfoLacnac (slan)-expressing DCs in patients with AD [178,179]. Histamine released from mast cells is relatively rich in AD lesions, subsequently leading to the activation of H4R-expressing DCs, and contributes to inflammatory processes. 5.2.8. Chemokines Thymus- and activation-regulated chemokine, also known as CCL17, is a key chemokine expressed on vascular endothelium and is involved in the homing of chemokine C-C motif receptor 4 (CCR4)-expressing T cells to the skin. CCL17 promotes the Th2 response via CCR4. The number of CCR4-expressing lymphocytes in serum is correlated with AD severity, the serum IgE level and the blood eosinophil count [180,181]. Furthermore, increased CCL17 levels in cord blood are associated with the development of infantile AD [182]. A recent meta-analysis indicated that serum CCL17 is the most reliable biomarker identified to date [154]. Similar to CCL17, macrophage-derived chemokine, also known as CCL22, is a chemoattractant for CCR4-expressing T cells. The Th2 chemokines CCL17 and CCL22 are mainly produced by LCs [183]. The levels of CCL17 and CCL22 may reflect the degree of skin barrier impairment. CCL17 and CCL22 induce the skin homing of T cells into AD lesions, and H4R antagonist inhibits CCL17 and CCL22 chemokine production by LCs in patients with AD [184]. Cutaneous T cell-attracting chemokine, also known as CCL27, is produced by keratinocytes and attracts CCR10-expressing Th22 cells into the skin. CCR6 is expressed in Th17 cells and facilitates their migration to the skin depending on the ligand CCL20 [185]. IFN-γ, monocyte chemotactic protein-4, eotaxin and RANTES secreted from keratinocytes after stimulation with Th1 cytokines facilitate the migration of macrophages, eosinophils and Th1 cells into chronic AD lesions [10]. CCL26 (eotaxin-3), which is essential to eosinophil recruitment into lesional epidermis, is enhanced by IL-4 and IL-13 [186]. IL-4 also stimulates dermal fibroblasts to express CCL11 (eotaxin-1) in AD lesions [187]. CCL11 was also detected on the lymphocytes, macrophages and eosinophils in AD lesions [188]. RANTES/CCL5 is a potent chemoattractant for eosinophils. CCL11, CCL5 and CCL26 bind to CCR3 expressed on eosinophils and activate these cells [189]. Fractalkine (CX3CL1) is expressed on the vascular endothelium and traffics leukocytes into AD lesions. Serum CX3CL1 levels were correlated with disease severity in pediatric AD patients [190,191]. 5.2.9. Mast Cells The role of mast cells in the pathogenesis of AD is not completely understood. Increased numbers of mast cells are seen in AD lesions, especially in the chronic state. After exposure to allergens, FcεRI and IgE binding induce mast cell degranulation and cause acute symptoms by releasing preformed mediators, such as histamine, heparin, serotonin, prostaglandins, leukotrienes, major basic protein and platelet-activating factor. Patients with distinct STAT3 mutations present enhanced STAT3 signaling in mast cells and accelerated degranulation [192]. Histamine and tryptase released by mast cells evoke scratching behavior and secondary barrier disruption. Mast cells are involved in the pathogenesis of AD not only by releasing inflammatory mediators, but also by directly regulating the recruitment and action of various inflammatory cells. Mast cells regulate the differentiation, activation and migration of T cells by inducing the expression of chemokines and adhesion molecules on endothelial cells [193]. Prostaglandin D2, a mast cell mediator, downregulates IL-12 production by DCs, which leads to a Th2-polarized immune response [194]. Mast cells are among the sources of IL-4 and IL-13 production [195]. Overexpression of TSLP in AD lesions can further activate mast cells to generate more Th2 cytokines [196]. Ligands for CD40 on mast cells interact with CD40 on the surface of B cells and stimulate B cell development and IgE synthesis in the presence of IL-4 [197]. Histamine and TNF are known to facilitate the migration of DCs into lymph nodes [198,199]. Chemokines derived from mast cells attract eosinophils and type 2 innate lymphoid cells [10,200]. IL-33 can induce mast cells to produce proinflammatory cytokines and chemokines [201]. Mast cells not only serve as key inflammatory mediators, but also play a protective role in the development of AD. A recent study indicated that mast cell-knockout mice showed incomplete epidermal differentiation with decreased EDC gene expression and easily developed severe AD-like skin inflammation [202]. 5.2.10. Eosinophils and Basophils Eosinophils are increased in number in both the serum and lesions of patients with AD, and they contribute to the pathogenesis of AD. Tissue eosinophilia is reportedly correlated with the severity of AD [203]. The Th2 cytokines IL-4, IL-5 and IL-13 play important roles in the development, survival, recruitment and function of eosinophils. Various inflammatory mediators are released during degranulation of eosinophils, including eosinophil cationic protein, eosinophil-derived neurotoxin and major basic protein. Eosinophil cationic protein and eosinophil-derived neurotoxin exhibit RNase activity and neurotoxicity. Eosinophil-derived neurotoxin induces maturation and activation of DCs by the TLR2-MyD88 signaling pathway and increases Th2 responses [204]. Major basic protein can downregulate the integrity of lipid bilayers [205]. Eosinophils constitutively express IL-31RA and release various proinflammatory cytokines and chemokines, such as IL-1β, IL-6, IL-31, CXCL1, CXCL8, CCL2, CCL18 and CCL26, in response to IL-31 [206,207,208]. IL-31 can prolong the survival of eosinophils by activating ERK signaling [207]. Similar to mast cells, basophils express FcεRI and degranulate after binding of IgE to FcεRI. Basophils play a role in producing Th2 cytokines in response to IL-18, as well as in promoting type 2 innate lymphoid cell (ILC2) recruitment and proliferation by releasing IL-4 in mouse AD models [209]. Basophils express ST2, a receptor for IL-33, and are activated by IL-33 [201]. Basophils express PAMPs, such as NOD2 and TLR2, and play a role in innate immune defense. S. aureus exacerbates AD symptoms by binding to NOD2 and TLR2 and by activating basophils and eosinophils in AD mouse models [210]. Recently, NOD2 expression by basophils in AD patients was found to be downregulated, which may explain the ineffective host defense to S. aureus in AD [211]. 5.2.11. Innate Lymphoid Cells ILCs are effector cells of innate immunity that are derived from a common lymphoid progenitor, but they do not express cell lineage markers for myeloid and DCs. ILC2s have been detected in the skin, peripheral blood, gastrointestinal tract and airways. ILCs can be divided into three groups based on predominant cytokine type. ILC1s produce Th1 cytokines, including IFN-γ; ILC2s produce Th2 cytokines, including IL-5 and IL-13; and ILC3s produce Th17 cytokines, such as IL-17 and IL-22 [212]. The transcription factor RAR-related orphan receptor alpha and GATA-3 induce differentiation of ILC2s in the presence of IL-25 and IL-33 [213]. Aberrant ILC2 function may contribute to allergic inflammation, such as AD [214,215]. ILC2s express CLA and CCR4, infiltrate the skin after allergen exposure and produce Th2 cytokines. Filaggrin deficiency is associated with increased ILC2 infiltration into the skin in both mice and patients with AD [216]. In response to barrier disruption, keratinocytes produce IL-25, IL-33 and TSLP. ILC2s express IL-25R, ST2 and TSLPR and are activated by IL-25, IL-33 and TSLP to produce IL-5 and IL-13. E-cadherin, the adhesion molecule between keratinocytes, is known to suppress the activation of skin ILC2s, possibly via ligation through killer cell lectin-like receptor G1 on human ILC2s. E-cadherin expression in keratinocytes is reduced due to loss of the skin barrier in AD lesions [217]. Subsequently, ILC2 may contribute to the production of high levels of the Th2 cytokines IL-5 and IL-13 in the absence of the inhibitory E-cadherin signal [213,218]. ILC2 cells produce IL-9, which attracts mast cells into the skin and augments their activation [214]. ILC2s are also activated and migrate in response to prostaglandin D2 [219]. Moreover, depletion of ILC2s blocked the skin homing of Th2 cells by inhibiting the production of the Th2 chemokine CCL17 in DCs in mouse models of AD [214,215,220]. ILC2s stimulated with IL-2 alone were sufficient to drive Th2 responses and AD-like inflammation without the influence of adaptive immunity in mouse models [221,222]. 5.3. Signal Pathways 5.3.1. GATA-3 GATA-3 is a key regulator of CD4+ T cell development, homeostasis, activation and proliferation. GATA-3 is known as a transcription factor that drives the differentiation of Th2 cells, stimulates the secretion of Th2 cytokines from Th2 cells and inhibits the development of B cells. GATA-3 plays a critical role in the differentiation of Th2 cells and is involved in the Th2 cytokine-mediated signaling network in AD. Th2 differentiation could be induced by both STAT6-dependent and STAT6-independent pathways. Acting in the downstream pathway of STAT6, GATA-3 activates IL-4, IL-5 and IL-13, but inhibits the expression of IFN-γ via STAT6-dependent pathways. GATA3 activates the common cellular signaling pathway including c-Jun N-terminal kinases, protein kinase C, JAK-STAT6 and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) in T cells [223]. GATA-3 has a positive-feedback loop that further activates GATA-3 and reinforces Th2 differentiation. In AD lesions, GATA-3 signaling is inhibited by T-bet, which is the Th1 transcription factor. By downregulating STAT-4, GATA-3 enabled Th2-predominant cytokine profiles to be maintained. GATA3 is regulated by diverse upstream signals, including IL-4R/STAT6, IL-2 receptor/STAT5, p38 MAP kinase, T cell receptor and Notch. Notch signaling or IL-2 receptor signaling could induce Th2 differentiation through GATA-3 in the STAT6-independent pathway [224]. Furthermore, AD patients showed a weak T cell receptor signaling, which enhances Th2 immunity and IgE production in order to compensate a weak T cell receptor signaling [225]. 4-hydroxy-3-methoxycinnamaldehyde downregulated T cell proliferation and differentiation into Th1 and Th2 cells by inhibiting T-bet and GATA3, respectively, and could ameliorate the symptoms of AD in mice models [226]. Recently, SB011, a topical DNAzyme that cleaves GATA-3, has been developed and is in a phase II clinical trial [79]. GATA-3 is also involved in Th9 differentiation, but inhibits Th1 and Th17 differentiation. GATA-3 regulates the function of Tregs by inducing forkhead box P3 (Foxp3) expression. As mentioned above, GATA-3 plays important roles in the generation and function of ILCs [223]. A recent GWAS of severe AD found new loci including candidate genes that may contribute to the defects in GATA-3 and STAT6 [59,227,228]. B cell activation is initiated by the engagement of B cell receptor and co-receptor, which leads to the change in the gene expression, as well as common cellular signaling, such as protein kinase C, NF-κB and c-fos. Information regarding unique abnormalities of B cell signaling in the AD pathogenesis is not well known. In order to establish B cell generation and maturation, CD40-CD40L interaction between B cells and T cells is essential, and B cell activation is parallel to T cell activation in a thymus-dependent manner [229]. 5.3.2. Notch Signaling The lesional epidermis of patients with AD exhibits markedly suppressed expression of Notch and its receptors, and reduced Notch expression in the lesions was shown to be normalized after successful treatment [230]. Impaired Notch signaling adversely influences epidermal terminal differentiation by affecting filaggrin, involucrin and transglutaminase-3 activity, resulting in incomplete CE and barrier formation. Suppressed Notch signaling downregulates the expression of aquaporin 3 and claudin-1, thus resulting in barrier dysfunction and increased transepidermal water loss [230,231]. Notch inhibits TLR-activated innate immunity by a negative feedback mechanism [232]. Thus, deficiency in Notch signaling may result in persistent activation of macrophages and DCs as observed in AD. The IFNG has recently been found as a target of Notch1 [233]. Compromised Notch signaling may thus explain the increased susceptibility to viral skin infection in AD through diminished IFN-γ production. Notch1 is an inhibitor of a transcription factor called activator protein-1, which is upregulated in the AD-affected epidermis and promotes Th2 cytokine production. The absence of Notch-mediated downregulation of activator protein-1 results in upregulation of the levels of IL-31 and may aggravate IL-31-mediated pruritus in AD [231]. Notch deficiency induces keratinocytes to secrete TSLP, which has recently been shown to stimulate cutaneous sensory neurons to promote itch [234]. Notch-knockout mice exhibit AD-like skin inflammation. In the presence of TGF-β, Notch1 regulates Foxp3 expression, which is important in the differentiation of Tregs. 6. Conclusions Patients with AD have impaired skin integrity and show elevated susceptibility to allergens and pathogens, which activate innate and adaptive immune responses. AD is characterized as a Th2/Th22-predominant inflammatory disease, but Th1 and Th17 responses modulate the development and progression of AD. Although T cells play key roles in the inflammation seen in AD, keratinocytes, DCs, B cells, mast cells, eosinophils, basophils and ILC2s act together through various cytokines and chemokines (Figure 1). Therapeutics include: duplimumab, anti-IL-4Ra monoclonal antibodies (mAb); lebrikizumab, anti-IL-13 mAb; ILV-094, anti-IL-22 mAb; BMS-981164, anti-IL-31 mAb; CIM331, anti-IL-31R mAb; AMG 157, anti-TSLP mAb; MK-8226 anti-TSLPR mAb; ustekinumab, anti-IL-12/IL-23 mAb; secukinumab, anti-IL-17 mAb; OC000459, chemoattractant receptor-homologous molecule expressed on Th2 cell antagonist. IL-4, -5, -13 and -31 drive Th2 inflammatory responses, but also inhibit epidermal terminal differentiation and lipid barrier formation and, thus, disrupt barrier functions. This is consistent with the “outside-in” and “inside-out” hypotheses [235]. Complex interactions between genes and environmental factors have been revealed, and further studies will reveal how environmental changes can regulate the development and triggering of AD. Current therapeutic options are often unsatisfactory to the refractory AD patients. Taken together, these data extend our knowledge of newly-found genes and signaling molecules as possible therapeutic targets for AD. Acknowledgments This work was supported by the Basic Science Research program and Creative Materials Discovery Program through the National Research Foundation of Korea (NRF), which is funded by the Ministry of Education, Science and Technology and the Ministry of Science, ICT and Future Planning (2015R1C1A2A01055073, 2016M3D1A1021387). We also thank Yujin Jung for help in preparation of the revised paper. Author Contributions Jung Eun Kim wrote and revised the manuscript. Jong Sic Kim assisted with the writing and revision of the manuscript. Dae Ho Cho and Hyun Jeong Park designed the review and assisted with the writing and revision of the manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations AD Atopic dermatitis AMPs Antimicrobial peptides CARD Caspase recruitment domain CCL C-C motif chemokine ligand CCR C-C motif chemokine receptor CLDN1 Claudin-1 CE Cornified envelope CXCL C-X-C motif chemokine ligand DCs Dendritic cells EDC Epidermal differentiation complex ERK Extracellular signal-regulated kinase FcεRI High affinity IgE receptor FFA Free fatty acid FLG Filaggrin gene Foxp3 Forkhead box P3 GATA-3 GATA-binding protein 3 GWAS Genome-wide association study hBD Human β-defensin HRH4 Histamine receptor H4 IDECs Inflammatory dendritic epidermal cells IFNG IFN-γ gene ILCs Innate lymphoid cells IL18RA P Interleukin 18 receptor accessory protein JAK Janus kinase JNK c-Jun N-terminal kinase LCs Langerhans cells LEKTI Lymphoepithelial Kazal-type trypsin inhibitor mAb Monoclonal antibodies miR MicroRNA MMP Matrix metalloproteinase NALP NACHT, LRR and PYD domain-containing protein NF-κB Nuclear factor κ-light-chain-enhancer of activated B cells NOD Nucleotide-binding oligomerization domain receptors PAR2 Protease-activated type 2 receptor PDE4 Phosphodiesterase 4 RANTES Regulated on activation, normal T cell expressed and secreted SC Stratum corneum SNP Single nucleotide polymorphism SP Serine protease STAT Signal transducer and activator of transcription Th T helper TLR Toll-like receptor TMEM79 Transmembrane protein 79 Treg Regulatory T cell TSLP Thymic stromal lymphopoietin SPINK5 Serine protease inhibitor Kazal-type 5 SPRR3 Small proline-rich protein 3 Figure 1 Barrier and immune network in the pathogenesis of atopic dermatitis. Red words: drug name; Dotted line (chronic stage): disease progression; Line: secretion. ijms-17-01234-t001_Table 1Table 1 Susceptibility genes for skin barrier and immunity in atopic dermatitis. Gene Locus Alleles or Mutation SNP Population Disease Severity Reference Epidermal Differentiation Complex FLG 1q21.3 Loss-of-function Copy number variation R501X, 2282del4 European, Chinese, Singaporean Increases the risk of AD correlated with severity [14,16,18,19,20,21,22] FLG2 1q21.3 Premature stop codon rs12568784, Q2053del224 rs16833974 African American More persistent AD [23] SPINK5 5q31 Loss-of-function rs2303070 T Taiwanese Increases the risk of AD [30] E420K Italian Increases the risk of AD [31] SPRR3 1q21.3 Copy number variation rs28989168 German Increases the risk of AD [25] TMEM79 Missense mutation rs6684514 Ireland Increases the risk of AD [27] claudin-1 3q28 Haplotype-tagging rs893051 African American Increases the risk of AD [32] AG or GG genotype rs9290929 Korean Mold infection [33] Innate Immunity TLR2 4 Missense mutation R753Q German, Italian Severe AD [40] A 16934T German, Japanese Severe AD [41,42] TLR4 9 N/A D299G Italian Increased in AD [43] 896A/G N/A Ukrainian Increased viral respiratory infections [44] TLR9 3 TT C-1237T German Intrinsic AD [45] NOD1 N/A rs2907748 rs2907749 German Allergen sensitization [46] hBD 1 8 haplotype CT rs5743399 Korean allergen sensitization [47] N/A rs5743409 Korean AD [47] TSLP 5q22 C/T rs1898671 European American Eczema herpeticum [48] Adaptive Immunity IL-18 11q22 G-allele rs1946518 rs187238 Chinese Low risk of AD [51] IL18RAP 2q12 N/A rs6419573 Japanese Increase the risk of AD [35] IL-12 chr3 IVS-798A/T, haplotype TA rs582504, rs582054, rs2243151 Korean Increase the risk of AD [53] IL-12RB chr5 chr1 TT AA rs438421, rs2066446 Korean Allergen sensitization [53] IFNG/IFNGR1 12/6q23-24 Loss-of-function V14M and Y397C African American Eczema herpeticum [52] IL-4 5q31-33 T allele 590 C/T of IL-4 promoter Egyptian Increase the risk of AD [58] IL-4Rα 16 Gain of function I50V, Q576R Egyptian Increase the risk of AD [58] IL-13 5q31.1 N/A rs12188917 Association with asthma [34] STAT6 12 Minor allele homozygotes rs324011 German Low risk of AD [59] IL-31 12 Haplotype AAA and GAA 1066, −2057, and ivs2 + 12 polish High IL-31 serum level severe pruritus [60] IL-17A AA genotype 152 G/A polish Severe AD in coexistence of asthma [61] FCER1A 1 N/A promoter Japanese Allergen sensitization [74] Chemokines N/A RANTES 17.35 28G N/A German Allergen sensitization [69] 403A overexpression N/A Japanese, German Allergen sensitization [70] Vitamin D Pathway Cyp24a1 20q54 Major C allele rs2248359 German Severe AD [55] VDR 20q13 AT rs7975232 Chinese Severe, eosinophilia and high IgE levels [54] Nerve Growth Factor Pathway BDNF 11 T C270T Chinese Intrinsic AD and male sex [77] N/A: not available. ==== Refs References 1. Odhiambo J.A. Williams H.C. Clayton T.O. Robertson C.F. Asher M.I. Global variations in prevalence of eczema symptoms in children from isaac phase three J. Allergy Clin. Immunol. 2009 124 1251 1258 10.1016/j.jaci.2009.10.009 20004783 2. Barnes K.C. An update on the genetics of atopic dermatitis: Scratching the surface in 2009 J. Allergy Clin. Immunol. 2010 125 16 29 10.1016/j.jaci.2009.11.008 20109730 3. Tay Y.K. Kong K.H. Khoo L. Goh C.L. Giam Y.C. The prevalence and descriptive epidemiology of atopic dermatitis in singapore school children Br. J. Dermatol. 2002 146 101 106 10.1046/j.1365-2133.2002.04566.x 11841373 4. Tokura Y. Extrinsic and intrinsic types of atopic dermatitis J. Dermatol. Sci. 2010 58 1 7 10.1016/j.jdermsci.2010.02.008 20207111 5. Katsarou A. Armenaka M. Atopic dermatitis in older patients: Particular points J. Eur. Acad. Dermatol. Venereol. 2011 25 12 18 10.1111/j.1468-3083.2010.03737.x 20569298 6. Bantz S.K. Zhu Z. Zheng T. The atopic march: Progression from atopic dermatitis to allergic rhinitis and asthma J. Clin. Cell. Immunol. 2014 5 10.4172/2155-9899.1000202 7. Andersen R.M. Thyssen J.P. Maibach H.I. Qualitative vs. Quantitative atopic dermatitis criteria—In historical and present perspectives J. Eur. Acad. Dermatol. Venereol. 2016 30 604 618 10.1111/jdv.13442 26538253 8. Oranje A.P. Glazenburg E.J. Wolkerstorfer A. de Waard-van der Spek F.B. Practical issues on interpretation of scoring atopic dermatitis: The scorad index, objective scorad and the three-item severity score Br. J. Dermatol. 2007 157 645 648 10.1111/j.1365-2133.2007.08112.x 17714568 9. Kim J.E. Kim H.J. Lew B.L. Lee K.H. Hong S.P. Jang Y.H. Park K.Y. Seo S.J. Bae J.M. Choi E.H. Consensus guidelines for the treatment of atopic dermatitis in korea (Part II): Systemic treatment Ann. Dermatol. 2015 27 578 592 10.5021/ad.2015.27.5.578 26512172 10. Mu Z. Zhao Y. Liu X. Chang C. Zhang J. Molecular biology of atopic dermatitis Clin. Rev. Allergy Immunol. 2014 47 193 218 10.1007/s12016-014-8415-1 24715253 11. Ghosh D. Ding L. Sivaprasad U. Geh E. Biagini Myers J. Bernstein J.A. Khurana Hershey G.K. Mersha T.B. Multiple transcriptome data analysis reveals biologically relevant atopic dermatitis signature genes and pathways PLoS ONE 2015 10 1234 10.1371/journal.pone.0144316 26717000 12. Czarnowicki T. Krueger J.G. Guttman-Yassky E. Skin barrier and immune dysregulation in atopic dermatitis: An evolving story with important clinical implications J. Allergy Clin. Immunol. Pract. 2014 2 371 379 10.1016/j.jaip.2014.03.006 25017523 13. Boguniewicz M. Leung D.Y. Atopic dermatitis: A disease of altered skin barrier and immune dysregulation Immunol. Rev. 2011 242 233 246 10.1111/j.1600-065X.2011.01027.x 21682749 14. Palmer C.N. Irvine A.D. Terron-Kwiatkowski A. Zhao Y. Liao H. Lee S.P. Goudie D.R. Sandilands A. Campbell L.E. Smith F.J. Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis Nat. Genet. 2006 38 441 446 10.1038/ng1767 16550169 15. Leung D.Y. Our evolving understanding of the functional role of filaggrin in atopic dermatitis J. Allergy Clin. Immunol. 2009 124 494 495 10.1016/j.jaci.2009.07.041 19733296 16. Henderson J. Northstone K. Lee S.P. Liao H. Zhao Y. Pembrey M. Mukhopadhyay S. Smith G.D. Palmer C.N. McLean W.H. The burden of disease associated with filaggrin mutations: A population-based, longitudinal birth cohort study J. Allergy Clin. Immunol. 2008 121 872 877 10.1016/j.jaci.2008.01.026 18325573 17. Brown S.J. McLean W.H. Eczema genetics: Current state of knowledge and future goals J. Investig. Dermatol. 2009 129 543 552 10.1038/jid.2008.413 19209157 18. Li M. Liu Q. Liu J. Cheng R. Zhang H. Xue H. Bao Y. Yao Z. Mutations analysis in filaggrin gene in Northern China patients with atopic dermatitis J. Eur. Acad. Dermatol. Venereol. 2013 27 169 174 10.1111/j.1468-3083.2011.04435.x 22220561 19. Chen H. Common J.E. Haines R.L. Balakrishnan A. Brown S.J. Goh C.S. Cordell H.J. Sandilands A. Campbell L.E. Kroboth K. Wide spectrum of filaggrin-null mutations in atopic dermatitis highlights differences between Singaporean Chinese and European populations Br. J. Dermatol. 2011 165 106 114 10.1111/j.1365-2133.2011.10331.x 21428977 20. Li K. Seok J. Park K.Y. Yoon Y. Kim K.H. Seo S.J. Copy-number variation of the filaggrin in Korean patients with atopic dermatitis: What really matters, “number” or “variation”? Br. J. Dermatol. 2016 174 1098 1100 10.1111/bjd.14287 26554544 21. Brown S.J. Kroboth K. Sandilands A. Campbell L.E. Pohler E. Kezic S. Cordell H.J. McLean W.H. Irvine A.D. Intragenic copy number variation within filaggrin contributes to the risk of atopic dermatitis with a dose-dependent effect J. Investig. Dermatol. 2012 132 98 104 10.1038/jid.2011.342 22071473 22. Kezic S. O’Regan G.M. Yau N. Sandilands A. Chen H. Campbell L.E. Kroboth K. Watson R. Rowland M. McLean W.H. Levels of filaggrin degradation products are influenced by both filaggrin genotype and atopic dermatitis severity Allergy 2011 66 934 940 10.1111/j.1398-9995.2010.02540.x 21261659 23. Margolis D.J. Gupta J. Apter A.J. Ganguly T. Hoffstad O. Papadopoulos M. Rebbeck T.R. Mitra N. Filaggrin-2 variation is associated with more persistent atopic dermatitis in African American subjects J. Allergy Clin. Immunol. 2014 133 784 789 10.1016/j.jaci.2013.09.015 24184149 24. Weidinger S. Willis-Owen S.A. Kamatani Y. Baurecht H. Morar N. Liang L. Edser P. Street T. Rodriguez E. O’Regan G.M. A genome-wide association study of atopic dermatitis identifies loci with overlapping effects on asthma and psoriasis Hum. Mol. Genet. 2013 22 4841 4856 10.1093/hmg/ddt317 23886662 25. Marenholz I. Rivera V.A. Esparza-Gordillo J. Bauerfeind A. Lee-Kirsch M.A. Ciechanowicz A. Kurek M. Piskackova T. Macek M. Lee Y.A. Association screening in the epidermal differentiation complex (EDC) identifies an SPRR3 repeat number variant as a risk factor for eczema J. Investig. Dermatol. 2011 131 1644 1649 10.1038/jid.2011.90 21490620 26. Kelsell D.P. Byrne C. Snping at the epidermal barrier J. Investig. Dermatol. 2011 131 1593 1595 10.1038/jid.2011.92 21753762 27. Saunders S.P. Goh C.S. Brown S.J. Palmer C.N. Porter R.M. Cole C. Campbell L.E. Gierlinski M. Barton G.J. Schneider G. Tmem79/matt is the matted mouse gene and is a predisposing gene for atopic dermatitis in human subjects J. Allergy Clin. Immunol. 2013 132 1121 1129 10.1016/j.jaci.2013.08.046 24084074 28. Elias P.M. Wakefield J.S. Mechanisms of abnormal lamellar body secretion and the dysfunctional skin barrier in patients with atopic dermatitis J. Allergy Clin. Immunol. 2014 134 781 791 10.1016/j.jaci.2014.05.048 25131691 29. Sasaki T. Shiohama A. Kubo A. Kawasaki H. Ishida-Yamamoto A. Yamada T. Hachiya T. Shimizu A. Okano H. Kudoh J. A homozygous nonsense mutation in the gene for TMEM79, a component for the lamellar granule secretory system, produces spontaneous eczema in an experimental model of atopic dermatitis J. Allergy Clin. Immunol. 2013 132 1111 1120 10.1016/j.jaci.2013.08.027 24060273 30. Lan C.C. Tu H.P. Wu C.S. Ko Y.C. Yu H.S. Lu Y.W. Li W.C. Chen Y.C. Chen G.S. Distinct spink5 and IL-31 polymorphisms are associated with atopic eczema and non-atopic hand dermatitis in taiwanese nursing population Exp. Dermatol. 2011 20 975 979 10.1111/j.1600-0625.2011.01374.x 22017185 31. Fortugno P. Furio L. Teson M. Berretti M. El Hachem M. Zambruno G. Hovnanian A. D’Alessio M. The 420K LEKTI variant alters LEKTI proteolytic activation and results in protease deregulation: Implications for atopic dermatitis Hum. Mol. Genet. 2012 21 4187 4200 10.1093/hmg/dds243 22730493 32. De Benedetto A. Rafaels N.M. McGirt L.Y. Ivanov A.I. Georas S.N. Cheadle C. Berger A.E. Zhang K. Vidyasagar S. Yoshida T. Tight junction defects in patients with atopic dermatitis J. Allergy Clin. Immunol. 2011 127 773 786 10.1016/j.jaci.2010.10.018 21163515 33. Yu H.S. Kang M.J. Kwon J.W. Lee S.Y. Lee E. Yang S.I. Jung Y.H. Hong K. Kim Y.J. Lee S.H. Claudin-1 polymorphism modifies the effect of mold exposure on the development of atopic dermatitis and production of IgE J. Allergy Clin. Immunol. 2015 135 827 830 10.1016/j.jaci.2014.10.040 25512082 34. Paternoster L. Standl M. Chen C.M. Ramasamy A. Bonnelykke K. Duijts L. Ferreira M.A. Alves A.C. Thyssen J.P. Albrecht E. Meta-analysis of genome-wide association studies identifies three new risk loci for atopic dermatitis Nat. Genet. 2012 44 187 192 10.1038/ng.1017 22197932 35. Hirota T. Takahashi A. Kubo M. Tsunoda T. Tomita K. Sakashita M. Yamada T. Fujieda S. Tanaka S. Doi S. Genome-wide association study identifies eight new susceptibility loci for atopic dermatitis in the japanese population Nat. Genet. 2012 44 1222 1226 10.1038/ng.2438 23042114 36. Suarez-Farinas M. Ungar B. Correa da Rosa J. Ewald D.A. Rozenblit M. Gonzalez J. Xu H. Zheng X. Peng X. Estrada Y.D. RNA sequencing atopic dermatitis transcriptome profiling provides insights into novel disease mechanisms with potential therapeutic implications J. Allergy Clin. Immunol. 2015 135 1218 1227 10.1016/j.jaci.2015.03.003 25840722 37. Esaki H. Ewald D.A. Ungar B. Rozenblit M. Zheng X. Xu H. Estrada Y.D. Peng X. Mitsui H. Litman T. Identification of novel immune and barrier genes in atopic dermatitis by means of laser capture microdissection J. Allergy Clin. Immunol. 2015 135 153 163 10.1016/j.jaci.2014.10.037 25567045 38. Tamari M. Hirota T. Genome-wide association studies of atopic dermatitis J. Dermatol. 2014 41 213 220 10.1111/1346-8138.12321 24628071 39. Paternoster L. Standl M. Waage J. Baurecht H. Hotze M. Strachan D.P. Curtin J.A. Bønnelykke K. Tian C. Takahashi A. Multi-ancestry genome-wide association study of 21,000 cases and 95,000 controls identifies new risk loci for atopic dermatitis Nat. Genet. 2015 47 1449 1456 10.1038/ng.3424 26482879 40. Ahmad-Nejad P. Mrabet-Dahbi S. Breuer K. Klotz M. Werfel T. Herz U. Heeg K. Neumaier M. Renz H. The toll-like receptor 2 R753Q polymorphism defines a subgroup of patients with atopic dermatitis having severe phenotype J. Allergy Clin. Immunol. 2004 113 565 567 10.1016/j.jaci.2003.12.583 15007364 41. Oh D.Y. Schumann R.R. Hamann L. Neumann K. Worm M. Heine G. Association of the toll-like receptor 2 A-16934T promoter polymorphism with severe atopic dermatitis Allergy 2009 64 1608 1615 10.1111/j.1398-9995.2009.02066.x 19627277 42. Potaczek D.P. Nastalek M. Okumura K. Wojas-Pelc A. Undas A. Nishiyama C. An association of TLR2–16934A>T polymorphism and severity/phenotype of atopic dermatitis J. Eur. Acad. Dermatol. Venereol. 2011 25 715 721 10.1111/j.1468-3083.2010.03812.x 21134221 43. Salpietro C. Rigoli L. Miraglia Del Giudice M. Cuppari C. Di Bella C. Salpietro A. Maiello N. La Rosa M. Marseglia G.L. Leonardi S. TLR2 and TLR4 gene polymorphisms and atopic dermatitis in Italian children: A multicenter study Int. J. Immunopathol. Pharmacol. 2011 24 33 40 22032785 44. Levchenko L. Izmailova O.V. Shlykova O.A. Kaidashev I.P. Polymorphism 896A/G of TLR4 gene rather than 1196C/T and 2258G/A of TLR2 gene determines severe and complicated course of atopic dermatitis in children Tsitol. Genet. 2013 47 46 53 23821954 45. Novak N. Yu C.F. Bussmann C. Maintz L. Peng W.M. Hart J. Hagemann T. Diaz-Lacava A. Baurecht H.J. Klopp N. Putative association of a TLR9 promoter polymorphism with atopic eczema Allergy 2007 62 766 772 10.1111/j.1398-9995.2007.01358.x 17573724 46. Macaluso F. Nothnagel M. Parwez Q. Petrasch-Parwez E. Bechara F.G. Epplen J.T. Hoffjan S. Polymorphisms in NACHT-LRR (NLR ) genes in atopic dermatitis Exp. Dermatol. 2007 16 692 698 10.1111/j.1600-0625.2007.00589.x 17620097 47. Kim E. Lee J.E. Namkung J.H. Kim P.S. Kim S. Shin E.S. Cho E.Y. Yang J.M. Single nucleotide polymorphisms and the haplotype in the DEFB1 gene are associated with atopic dermatitis in a Korean population J. Dermatol. Sci. 2009 54 25 30 10.1016/j.jdermsci.2008.12.005 19135873 48. Gao P.S. Rafaels N.M. Mu D. Hand T. Murray T. Boguniewicz M. Hata T. Schneider L. Hanifin J.M. Gallo R.L. Genetic variants in thymic stromal lymphopoietin are associated with atopic dermatitis and eczema herpeticum J. Allergy Clin. Immunol. 2010 125 1403 1407 10.1016/j.jaci.2010.03.016 20466416 49. Margolis D.J. Kim B. Apter A.J. Gupta J. Hoffstad O. Papadopoulos M. Mitra N. Thymic stromal lymphopoietin variation, filaggrin loss of function, and the persistence of atopic dermatitis JAMA Dermatol. 2014 150 254 259 10.1001/jamadermatol.2013.7954 24401911 50. Novak N. Kruse S. Potreck J. Maintz L. Jenneck C. Weidinger S. Fimmers R. Bieber T. Single nucleotide polymorphisms of the IL18 gene are associated with atopic eczema J. Allergy Clin. Immunol. 2005 115 828 833 10.1016/j.jaci.2005.01.030 15806006 51. Cheng D. Hao Y. Zhou W. Ma Y. The relationship between interleukin-18 polymorphisms and allergic disease: A meta-analysis BioMed Res. Int. 2014 2014 290687 10.1155/2014/290687 24995282 52. Gao L. Bin L. Rafaels N.M. Huang L. Potee J. Ruczinski I. Beaty T.H. Paller A.S. Schneider L.C. Gallo R. Targeted deep sequencing identifies rare loss-of-function variants in IFNGR1 for risk of atopic dermatitis complicated by eczema herpeticum J. Allergy Clin. Immunol. 2015 136 1591 1600 10.1016/j.jaci.2015.06.047 26343451 53. Namkung J.H. Lee J.E. Kim E. Kim S. Kim S. Shin E.S. Cho E.Y. Yang J.M. Association of single nucleotide polymorphisms in the IL-12 (IL-12a and b) and IL-12 receptor (IL-12rβ1 and β2 ) genes and gene-gene interactions with atopic dermatitis in koreans J. Dermatol. Sci. 2010 57 199 206 10.1016/j.jdermsci.2009.12.003 20060272 54. Wang S.S. Hon K.L. Kong A.P. Tang M.F. Sy H.Y. Chan J.C. Leung T.F. Eczema phenotypes are associated with multiple vitamin d pathway genes in chinese children Allergy 2014 69 118 124 10.1111/all.12337 24730053 55. Hallau J. Hamann L. Schumann R.R. Worm M. Heine G. A promoter polymorphism of the vitamin D metabolism gene CYP24A1 is associated with severe atopic dermatitis in adults Acta Derm. Venereol. 2016 96 169 172 10.2340/00015555-2226 26315479 56. Heine G. Hoefer N. Franke A. Nothling U. Schumann R.R. Hamann L. Worm M. Association of vitamin d receptor gene polymorphisms with severe atopic dermatitis in adults Br. J. Dermatol. 2013 168 855 858 10.1111/bjd.12077 23034014 57. Tamura K. Arakawa H. Suzuki M. Kobayashi Y. Mochizuki H. Kato M. Tokuyama K. Morikawa A. Novel dinucleotide repeat polymorphism in the first exon of the STAT-6 gene is associated with allergic diseases Clin. Exp. Allergy 2001 31 1509 1514 10.1046/j.1365-2222.2001.01191.x 11678849 58. Hussein Y.M. Shalaby S.M. Nassar A. Alzahrani S.S. Alharbi A.S. Nouh M. Association between genes encoding components of the IL-4/IL-4 receptor pathway and dermatitis in children Gene 2014 545 276 281 10.1016/j.gene.2014.04.024 24742632 59. Casaca V.I. Illi S. Klucker E. Ballenberger N. Schedel M. von Mutius E. Kabesch M. Schaub B. STAT6 polymorphisms are associated with neonatal regulatory T cells and cytokines and atopic diseases at 3 years Allergy 2013 68 1249 1258 10.1111/all.12220 24053457 60. Sokolowska-Wojdylo M. Glen J. Zablotna M. Rebala K. Trzeciak M. Sikorska M. Ruckemann-Dziurdzinska K. Nedoszytko B. Florek A. Nowicki R. The frequencies of haplotypes defined by three polymorphisms of the IL-31 gene: −1066, −2057, and IVS2+12 in Polish patients with atopic dermatitis Int. J. Dermatol. 2015 54 62 67 10.1111/ijd.12666 25534405 61. Narbutt J. Wojtczak M. Zalinska A. Salinski A. Przybylowska-Sygut K. Kuna P. Majak P. Sysa-Jedrzejowska A. Lesiak A. The A/A genotype of an interleukin-17A polymorphism predisposes to increased severity of atopic dermatitis and coexistence with asthma Clin. Exp. Dermatol. 2015 40 11 16 10.1111/ced.12438 25223230 62. Hong C.H. Yu H.S. Ko Y.C. Chang W.C. Chuang H.Y. Chen G.S. Lee C.H. Functional regulation of interleukin-31 production by its genetic polymorphism in patients with extrinsic atopic dermatitis Acta Derm. Venereol. 2012 92 430 432 10.2340/00015555-1203 21952721 63. Arshad S.H. Karmaus W. Kurukulaaratchy R. Sadeghnejad A. Huebner M. Ewart S. Polymorphisms in the interleukin 13 and GATA binding protein 3 genes and the development of eczema during childhood Br. J. Dermatol. 2008 158 1315 1322 10.1111/j.1365-2133.2008.08565.x 18410415 64. Hussein P.Y. Zahran F. Ashour Wahba A. Ahmad A.S. Ibrahiem M.M. Shalaby S.M. El Tarhouny S.A. El Sherbiny H.M. Bakr N. Interleukin 10 receptor α subunit (IL-10RA ) gene polymorphism and IL-10 serum levels in Egyptian atopic patients J. Investig. Allergol. Clin. Immunol. 2010 20 20 26 20232770 65. Namkung J.H. Lee J.E. Kim E. Park G.T. Yang H.S. Jang H.Y. Shin E.S. Cho E.Y. Yang J.M. An association between IL-9 and IL-9 receptor gene polymorphisms and atopic dermatitis in a Korean population J. Dermatol. Sci. 2011 62 16 21 10.1016/j.jdermsci.2011.01.007 21371865 66. Gharagozlou M. Farhadi E. Khaledi M. Behniafard N. Sotoudeh S. Salari R. Darabi B. Fathi S.M. Mahmoudi M. Aghamohammadi A. Association between the interleukin 6 genotype at position −174 and atopic dermatitis J. Investig. Allergol. Clin. Immunol. 2013 23 89 93 23654074 67. Namkung J.H. Lee J.E. Kim E. Cho H.J. Kim S. Shin E.S. Cho E.Y. Yang J.M. IL-5 and IL-5 receptor α polymorphisms are associated with atopic dermatitis in Koreans Allergy 2007 62 934 942 10.1111/j.1398-9995.2007.01445.x 17620072 68. Hoffjan S. Beygo J. Akkad D.A. Parwez Q. Petrasch-Parwez E. Epplen J.T. Analysis of variation in the IL7RA and IL2RA genes in atopic dermatitis J. Dermatol. Sci. 2009 55 138 140 10.1016/j.jdermsci.2009.05.001 19523791 69. Nickel R.G. Casolaro V. Wahn U. Beyer K. Barnes K.C. Plunkett B.S. Freidhoff L.R. Sengler C. Plitt J.R. Schleimer R.P. Atopic dermatitis is associated with a functional mutation in the promoter of the C-C chemokine rantes J. Immunol. 2000 164 1612 1616 10.4049/jimmunol.164.3.1612 10640782 70. Tanaka K. Roberts M.H. Yamamoto N. Sugiura H. Uehara M. Hopkin J.M. Upregulating promoter polymorphisms of rantes relate to atopic dermatitis Int. J. Immunogenet. 2006 33 423 428 10.1111/j.1744-313X.2006.00635.x 17117952 71. Yamanaka K. Mizutani H. The role of cytokines/chemokines in the pathogenesis of atopic dermatitis Curr. Probl. Dermatol. 2011 41 80 92 21576949 72. Yu B. Shao Y. Zhang J. Dong X.L. Liu W.L. Yang H. Liu L. Li M.H. Yue C.F. Fang Z.Y. Polymorphisms in human histamine receptor H4 gene are associated with atopic dermatitis Br. J. Dermatol. 2010 162 1038 1043 10.1111/j.1365-2133.2010.09675.x 20199554 73. Chen B. Ye T. Shao Y. Zhang J. Zhong Q. Hu X. Zhang W. Yu B. Association between copy-number variations of the human histamine H4 receptor gene and atopic dermatitis in a Chinese population Clin. Exp. Dermatol. 2013 38 295 300 10.1111/ced.12117 23517361 74. Niwa Y. Potaczek D.P. Kanada S. Takagi A. Shimokawa N. Ito T. Mitsuishi K. Okubo Y. Tajima M. Hobo A. Fcepsilonrialpha gene (FCER1A ) promoter polymorphisms and total serum IgE levels in Japanese atopic dermatitis patients Int. J. Immunogenet. 2010 37 139 141 10.1111/j.1744-313X.2010.00901.x 20141544 75. Park K.Y. Park M.K. Kim E.J. Lee M.K. Seo S.J. Fcepsilonri gene promoter polymorphisms and total IgE levels in susceptibility to atopic dermatitis in korea J. Korean Med. Sci. 2011 26 870 874 10.3346/jkms.2011.26.7.870 21738338 76. Zhou J. Zhou Y. Lin L.H. Wang J. Peng X. Li J. Li L. Association of polymorphisms in the promoter region of FCER1A gene with atopic dermatitis, chronic uticaria, asthma, and serum immunoglobulin E levels in a Han Chinese population Hum. Immunol. 2012 73 301 305 10.1016/j.humimm.2011.12.001 22222815 77. Ma L. Gao X.H. Zhao L.P. Di Z.H. McHepange U.O. Zhang L. Chen H.D. Wei H.C. Brain-derived neurotrophic factor gene polymorphisms and serum levels in chinese atopic dermatitis patients J. Eur. Acad. Dermatol. Venereol. 2009 23 1277 1281 10.1111/j.1468-3083.2009.03308.x 19522715 78. Blakely K. Gooderham M. Papp K. Dupilumab, a monoclonal antibody for atopic dermatitis: A review of current literature Skin Ther. Lett. 2016 21 1 5 79. Wang D. Beck L.A. Immunologic targets in atopic dermatitis and emerging therapies: An update Am. J. Clin. Dermatol. 2016 1 19 10.1007/s40257-016-0205-5 26518880 80. Noda S. Krueger J.G. Guttman-Yassky E. The translational revolution and use of biologics in patients with inflammatory skin diseases J. Allergy Clin. Immunol. 2015 135 324 336 10.1016/j.jaci.2014.11.015 25541257 81. Liang Y. Chang C. Lu Q. The genetics and epigenetics of atopic dermatitis-filaggrin and other polymorphisms Clin. Rev. Allergy Immunol. 2015 10.1007/s12016-015-8508-5 26385242 82. Herberth G. Bauer M. Gasch M. Hinz D. Roder S. Olek S. Kohajda T. Rolle-Kampczyk U. von Bergen M. Sack U. Maternal and cord blood mir-223 expression associates with prenatal tobacco smoke exposure and low regulatory T-cell numbers J. Allergy Clin. Immunol. 2014 133 543 550 10.1016/j.jaci.2013.06.036 23978443 83. Hinz D. Bauer M. Roder S. Olek S. Huehn J. Sack U. Borte M. Simon J.C. Lehmann I. Herberth G. Cord blood tregs with stable FOXP3 expression are influenced by prenatal environment and associated with atopic dermatitis at the age of one year Allergy 2012 67 380 389 10.1111/j.1398-9995.2011.02767.x 22187950 84. Ahn K. The role of air pollutants in atopic dermatitis J. Allergy Clin. Immunol. 2014 134 993 999 10.1016/j.jaci.2014.09.023 25439225 85. Wang I.J. Guo Y.L. Lin T.J. Chen P.C. Wu Y.N. GSTM1, GSTP1, prenatal smoke exposure, and atopic dermatitis Ann. Allergy Asthma Immunol. 2010 105 124 129 10.1016/j.anai.2010.04.017 20674822 86. Wang I.J. Chen S.L. Lu T.P. Chuang E.Y. Chen P.C. Prenatal smoke exposure, DNA methylation, and childhood atopic dermatitis Clin. Exp. Allergy 2013 43 535 543 10.1111/cea.12108 23600544 87. Nakamura T. Sekigawa I. Ogasawara H. Mitsuishi K. Hira K. Ikeda S. Ogawa H. Expression of DNMT-1 in patients with atopic dermatitis Arch. Dermatol. Res. 2006 298 253 256 10.1007/s00403-006-0682-0 16897079 88. Liang Y. Wang P. Zhao M. Liang G. Yin H. Zhang G. Wen H. Lu Q. Demethylation of the FCER1G promoter leads to fcepsilonri overexpression on monocytes of patients with atopic dermatitis Allergy 2012 67 424 430 10.1111/j.1398-9995.2011.02760.x 22150093 89. Sonkoly E. Janson P. Majuri M.L. Savinko T. Fyhrquist N. Eidsmo L. Xu N. Meisgen F. Wei T. Bradley M. miR-155 is overexpressed in patients with atopic dermatitis and modulates T-cell proliferative responses by targeting cytotoxic T lymphocyte-associated antigen 4 J. Allergy Clin. Immunol. 2010 126 581 589 10.1016/j.jaci.2010.05.045 20673989 90. Ma L. Xue H.B. Wang F. Shu C.M. Zhang J.H. MicroRNA-155 may be involved in the pathogenesis of atopic dermatitis by modulating the differentiation and function of T Helper type 17 (Th17) cells Clin. Exp. Immunol. 2015 181 142 149 10.1111/cei.12624 25761610 91. Rebane A. Runnel T. Aab A. Maslovskaja J. Ruckert B. Zimmermann M. Plaas M. Karner J. Treis A. Pihlap M. MicroRNA-146A alleviates chronic skin inflammation in atopic dermatitis through suppression of innate immune responses in keratinocytes J. Allergy Clin. Immunol. 2014 134 836 847 10.1016/j.jaci.2014.05.022 24996260 92. Zeng Y.P. Nguyen G.H. Jin H.Z. MicroRNA-143 inhibits IL-13-induced dysregulation of the epidermal barrier-related proteins in skin keratinocytes via targeting to IL-13rα1 Mol. Cell. Biochem. 2016 416 63 70 10.1007/s11010-016-2696-z 27048505 93. Dinarello C.A. Immunological and inflammatory functions of the interleukin-1 family Annu. Rev. Immunol. 2009 27 519 550 10.1146/annurev.immunol.021908.132612 19302047 94. Hachem J.P. Wagberg F. Schmuth M. Crumrine D. Lissens W. Jayakumar A. Houben E. Mauro T.M. Leonardsson G. Brattsand M. Serine protease activity and residual lekti expression determine phenotype in netherton syndrome J. Investig. Dermatol. 2006 126 1609 1621 10.1038/sj.jid.5700288 16601670 95. Briot A. Deraison C. Lacroix M. Bonnart C. Robin A. Besson C. Dubus P. Hovnanian A. Kallikrein 5 induces atopic dermatitis-like lesions through PAR2-mediated thymic stromal lymphopoietin expression in netherton syndrome J. Exp. Med. 2009 206 1135 1147 10.1084/jem.20082242 19414552 96. Kim J.H. Bae H.C. Ko N.Y. Lee S.H. Jeong S.H. Lee H. Ryu W.I. Kye Y.C. Son S.W. Thymic stromal lymphopoietin downregulates filaggrin expression by signal transducer and activator of transcription 3 (STAT3) and extracellular signal-regulated kinase (ERK) phosphorylation in keratinocytes J. Allergy Clin. Immunol. 2015 136 205 208 10.1016/j.jaci.2015.04.026 26026341 97. Andoh T. Kuraishi Y. Antipruritic mechanisms of topical e6005, a phosphodiesterase 4 inhibitor: Inhibition of responses to proteinase-activated receptor 2 stimulation mediated by increase in intracellular cyclic amp J. Dermatol. Sci. 2014 76 206 213 10.1016/j.jdermsci.2014.10.005 25458869 98. Volf E.M. Au S.C. Dumont N. Scheinman P. Gottlieb A.B. A phase 2, open-label, investigator-initiated study to evaluate the safety and efficacy of apremilast in subjects with recalcitrant allergic contact or atopic dermatitis J. Drugs Dermatol. 2012 11 341 346 22395585 99. Samrao A. Berry T.M. Goreshi R. Simpson E.L. A pilot study of an oral phosphodiesterase inhibitor (apremilast) for atopic dermatitis in adults Arch. Dermatol. 2012 148 890 897 10.1001/archdermatol.2012.812 22508772 100. Pellerin L. Henry J. Hsu C.Y. Balica S. Jean-Decoster C. Mechin M.C. Hansmann B. Rodriguez E. Weindinger S. Schmitt A.M. Defects of filaggrin-like proteins in both lesional and nonlesional atopic skin J. Allergy Clin. Immunol. 2013 131 1094 1102 10.1016/j.jaci.2012.12.1566 23403047 101. Jensen J.M. Pfeiffer S. Witt M. Brautigam M. Neumann C. Weichenthal M. Schwarz T. Folster-Holst R. Proksch E. Different effects of pimecrolimus and betamethasone on the skin barrier in patients with atopic dermatitis J. Allergy Clin. Immunol. 2009 123 1124 1133 10.1016/j.jaci.2009.03.032 19410693 102. Otsuka A. Doi H. Egawa G. Maekawa A. Fujita T. Nakamizo S. Nakashima C. Nakajima S. Watanabe T. Miyachi Y. Possible new therapeutic strategy to regulate atopic dermatitis through upregulating filaggrin expression J. Allergy Clin. Immunol. 2014 133 139 146 10.1016/j.jaci.2013.07.027 24055295 103. Stout T.E. McFarland T. Mitchell J.C. Appukuttan B. Stout J.T. Recombinant filaggrin is internalized and processed to correct filaggrin deficiency J. Investig. Dermatol. 2014 134 423 429 10.1038/jid.2013.284 23792461 104. De Benedetto A. Slifka M.K. Rafaels N.M. Kuo I.H. Georas S.N. Boguniewicz M. Hata T. Schneider L.C. Hanifin J.M. Gallo R.L. Reductions in claudin-1 may enhance susceptibility to herpes simplex virus 1 infections in atopic dermatitis J. Allergy Clin. Immunol. 2011 128 242 246 10.1016/j.jaci.2011.02.014 21489616 105. Gruber R. Bornchen C. Rose K. Daubmann A. Volksdorf T. Wladykowski E. Vidal Y.S.S. Peters E.M. Danso M. Bouwstra J.A. Diverse regulation of claudin-1 and claudin-4 in atopic dermatitis Am. J. Pathol. 2015 185 2777 2789 10.1016/j.ajpath.2015.06.021 26319240 106. Kuo I.H. Carpenter-Mendini A. Yoshida T. McGirt L.Y. Ivanov A.I. Barnes K.C. Gallo R.L. Borkowski A.W. Yamasaki K. Leung D.Y. Activation of epidermal Toll-Like receptor 2 enhances tight junction function: Implications for atopic dermatitis and skin barrier repair J. Investig. Dermatol. 2013 133 988 998 10.1038/jid.2012.437 23223142 107. Suarez-Farinas M. Tintle S.J. Shemer A. Chiricozzi A. Nograles K. Cardinale I. Duan S. Bowcock A.M. Krueger J.G. Guttman-Yassky E. Nonlesional atopic dermatitis skin is characterized by broad terminal differentiation defects and variable immune abnormalities J. Allergy Clin. Immunol. 2011 127 954 964 10.1016/j.jaci.2010.12.1124 21388663 108. Janssens M. van Smeden J. Gooris G.S. Bras W. Portale G. Caspers P.J. Vreeken R.J. Hankemeier T. Kezic S. Wolterbeek R. Increase in short-chain ceramides correlates with an altered lipid organization and decreased barrier function in atopic eczema patients J. Lipid Res. 2012 53 2755 2766 10.1194/jlr.P030338 23024286 109. Van Smeden J. Janssens M. Kaye E.C. Caspers P.J. Lavrijsen A.P. Vreeken R.J. Bouwstra J.A. The importance of free fatty acid chain length for the skin barrier function in atopic eczema patients Exp. Dermatol. 2014 23 45 52 10.1111/exd.12293 24299153 110. Sawada E. Yoshida N. Sugiura A. Imokawa G. Th1 cytokines accentuate but Th2 cytokines attenuate ceramide production in the stratum corneum of human epidermal equivalents: An implication for the disrupted barrier mechanism in atopic dermatitis J. Dermatol. Sci. 2012 68 25 35 10.1016/j.jdermsci.2012.07.004 22884781 111. Ong P.Y. Ohtake T. Brandt C. Strickland I. Boguniewicz M. Ganz T. Gallo R.L. Leung D.Y. Endogenous antimicrobial peptides and skin infections in atopic dermatitis N. Engl. J. Med. 2002 347 1151 1160 10.1056/NEJMoa021481 12374875 112. Howell M.D. Novak N. Bieber T. Pastore S. Girolomoni G. Boguniewicz M. Streib J. Wong C. Gallo R.L. Leung D.Y. Interleukin-10 downregulates anti-microbial peptide expression in atopic dermatitis J. Investig. Dermatol. 2005 125 738 745 10.1111/j.0022-202X.2005.23776.x 16185274 113. Schittek B. The antimicrobial skin barrier in patients with atopic dermatitis Curr. Probl. Dermatol. 2011 41 54 67 21576947 114. Son E.D. Kim H.J. Kim K.H. Bin B.H. Bae I.H. Lim K.M. Yu S.J. Cho E.G. Lee T.R. S100a7 (psoriasin) inhibits human epidermal differentiation by enhanced IL-6 secretion through IκB/NF-κB signaling Exp. Dermatol. 2016 25 636 641 10.1111/exd.13023 27060579 115. Niebuhr M. Langnickel J. Sigel S. Werfel T. Dysregulation of CD36 upon TLR-2 stimulation in monocytes from patients with atopic dermatitis and the TLR2 R753Q polymorphism Exp. Dermatol. 2010 19 e296 e298 10.1111/j.1600-0625.2009.00989.x 19849713 116. Camargo C.A. Jr. Ganmaa D. Sidbury R. Erdenedelger K. Radnaakhand N. Khandsuren B. Randomized trial of vitamin d supplementation for winter-related atopic dermatitis in children J. Allergy Clin. Immunol. 2014 134 831 835 10.1016/j.jaci.2014.08.002 25282565 117. Hata T.R. Audish D. Kotol P. Coda A. Kabigting F. Miller J. Alexandrescu D. Boguniewicz M. Taylor P. Aertker L. A randomized controlled double-blind investigation of the effects of vitamin D dietary supplementation in subjects with atopic dermatitis J. Eur. Acad. Dermatol. Venereol. 2014 28 781 789 10.1111/jdv.12176 23638978 118. Nygaard U. Hvid M. Johansen C. Buchner M. Folster-Holst R. Deleuran M. Vestergaard C. TSLP, IL-31, IL-33 and SST2 are new biomarkers in endophenotypic profiling of adult and childhood atopic dermatitis J. Eur. Acad. Dermatol. Venereol. 2016 10.1111/jdv.13679 27152943 119. Moy A.P. Murali M. Kroshinsky D. Duncan L.M. Nazarian R.M. Immunologic overlap of helper T-cell subtypes 17 and 22 in erythrodermic psoriasis and atopic dermatitis JAMA Dermatol. 2015 151 753 760 10.1001/jamadermatol.2015.2 25760640 120. Howell M.D. Fairchild H.R. Kim B.E. Bin L. Boguniewicz M. Redzic J.S. Hansen K.C. Leung D.Y. Th2 cytokines act on S100/A11 to downregulate keratinocyte differentiation J. Investig. Dermatol. 2008 128 2248 2258 10.1038/jid.2008.74 18385759 121. Gittler J.K. Shemer A. Suarez-Farinas M. Fuentes-Duculan J. Gulewicz K.J. Wang C.Q. Mitsui H. Cardinale I. de Guzman Strong C. Krueger J.G. Progressive activation of Th2/Th22 cytokines and selective epidermal proteins characterizes acute and chronic atopic dermatitis J. Allergy Clin. Immunol. 2012 130 1344 1354 10.1016/j.jaci.2012.07.012 22951056 122. Sehra S. Yao Y. Howell M.D. Nguyen E.T. Kansas G.S. Leung D.Y. Travers J.B. Kaplan M.H. IL-4 regulates skin homeostasis and the predisposition toward allergic skin inflammation J. Immunol. 2010 184 3186 3190 10.4049/jimmunol.0901860 20147633 123. Kim B.E. Leung D.Y. Boguniewicz M. Howell M.D. Loricrin and involucrin expression is down-regulated by Th2 cytokines through STAT-6 Clin. Immunol. 2008 126 332 337 10.1016/j.clim.2007.11.006 18166499 124. Howell M.D. Kim B.E. Gao P. Grant A.V. Boguniewicz M. Debenedetto A. Schneider L. Beck L.A. Barnes K.C. Leung D.Y. Cytokine modulation of atopic dermatitis filaggrin skin expression J. Allergy Clin. Immunol. 2007 120 150 155 10.1016/j.jaci.2007.04.031 17512043 125. McAleer M.A. Irvine A.D. The multifunctional role of filaggrin in allergic skin disease J. Allergy Clin. Immunol. 2013 131 280 291 10.1016/j.jaci.2012.12.668 23374260 126. Amano W. Nakajima S. Kunugi H. Numata Y. Kitoh A. Egawa G. Dainichi T. Honda T. Otsuka A. Kimoto Y. The janus kinase inhibitor JTE-052 improves skin barrier function through suppressing signal transducer and activator of transcription 3 signaling J. Allergy Clin. Immunol. 2015 136 667 677 10.1016/j.jaci.2015.03.051 26115905 127. Thaci D. Simpson E.L. Beck L.A. Bieber T. Blauvelt A. Papp K. Soong W. Worm M. Szepietowski J.C. Sofen H. Efficacy and safety of dupilumab in adults with moderate-to-severe atopic dermatitis inadequately controlled by topical treatments: A randomised, placebo-controlled, dose-ranging phase 2B trial Lancet 2016 387 40 52 10.1016/S0140-6736(15)00388-8 26454361 128. Toda M. Leung D.Y. Molet S. Boguniewicz M. Taha R. Christodoulopoulos P. Fukuda T. Elias J.A. Hamid Q.A. Polarized in vivo expression of IL-11 and IL-17 between acute and chronic skin lesions J. Allergy Clin. Immunol. 2003 111 875 881 10.1067/mai.2003.1414 12704372 129. Nograles K.E. Zaba L.C. Shemer A. Fuentes-Duculan J. Cardinale I. Kikuchi T. Ramon M. Bergman R. Krueger J.G. Guttman-Yassky E. IL-22-producing “T22” T cells account for upregulated IL-22 in atopic dermatitis despite reduced IL-17-producing Th17 T cells J. Allergy Clin. Immunol. 2009 123 1244 1252 10.1016/j.jaci.2009.03.041 19439349 130. Park H. Li Z. Yang X.O. Chang S.H. Nurieva R. Wang Y.H. Wang Y. Hood L. Zhu Z. Tian Q. A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17 Nat. Immunol. 2005 6 1133 1141 10.1038/ni1261 16200068 131. Jin S. Park C.O. Shin J.U. Noh J.Y. Lee Y.S. Lee N.R. Kim H.R. Noh S. Lee Y. Lee J.H. Damp molecules S100A9 and S100A8 activated by IL-17A and house-dust mites are increased in atopic dermatitis Exp. Dermatol. 2014 23 938 941 10.1111/exd.12563 25308296 132. Suarez-Farinas M. Dhingra N. Gittler J. Shemer A. Cardinale I. de Guzman Strong C. Krueger J.G. Guttman-Yassky E. Intrinsic atopic dermatitis shows similar Th2 and higher Th17 immune activation compared with extrinsic atopic dermatitis J. Allergy Clin. Immunol. 2013 132 361 370 10.1016/j.jaci.2013.04.046 23777851 133. Noda S. Suarez-Farinas M. Ungar B. Kim S.J. de Guzman Strong C. Xu H. Peng X. Estrada Y.D. Nakajima S. Honda T. The asian atopic dermatitis phenotype combines features of atopic dermatitis and psoriasis with increased Th17 polarization J. Allergy Clin. Immunol. 2015 136 1254 1264 10.1016/j.jaci.2015.08.015 26428954 134. Czarnowicki T. Gonzalez J. Shemer A. Malajian D. Xu H. Zheng X. Khattri S. Gilleaudeau P. Sullivan-Whalen M. Suarez-Farinas M. Severe atopic dermatitis is characterized by selective expansion of circulating TH2/TC2 and TH22/TC22, but not TH17/TC17, cells within the skin-homing T-cell population J. Allergy Clin. Immunol. 2015 136 104 115 10.1016/j.jaci.2015.01.020 25748064 135. Gutowska-Owsiak D. Schaupp A.L. Salimi M. Taylor S. Ogg G.S. Interleukin-22 downregulates filaggrin expression and affects expression of profilaggrin processing enzymes Br. J. Dermatol. 2011 165 492 498 10.1111/j.1365-2133.2011.10400.x 21564072 136. Wolk K. Witte E. Wallace E. Docke W.D. Kunz S. Asadullah K. Volk H.D. Sterry W. Sabat R. IL-22 regulates the expression of genes responsible for antimicrobial defense, cellular differentiation, and mobility in keratinocytes: A potential role in psoriasis Eur. J. Immunol. 2006 36 1309 1323 10.1002/eji.200535503 16619290 137. Niebuhr M. Mainardy J. Heratizadeh A. Satzger I. Werfel T. Staphylococcal exotoxins induce interleukin 22 in human Th22 cells Int. Arch. Allergy Immunol. 2014 165 35 39 10.1159/000367923 25301279 138. Jang M. Kim H. Kim Y. Choi J. Jeon J. Hwang Y. Kang J.S. Lee W.J. The crucial role of IL-22 and its receptor on tarc production and T cell migration by HDM extract Exp. Dermatol. 2016 25 598 603 10.1111/exd.12988 26914146 139. Boniface K. Bernard F.X. Garcia M. Gurney A.L. Lecron J.C. Morel F. IL-22 inhibits epidermal differentiation and induces proinflammatory gene expression and migration of human keratinocytes J. Immunol. 2005 174 3695 3702 10.4049/jimmunol.174.6.3695 15749908 140. Wu P.W. Li J. Kodangattil S.R. Luxenberg D.P. Bennett F. Martino M. Collins M. Dunussi-Joannopoulos K. Gill D.S. Wolfman N.M. IL-22R, IL-10R2, and IL-22BP binding sites are topologically juxtaposed on adjacent and overlapping surfaces of IL-22 J. Mol. Biol. 2008 382 1168 1183 10.1016/j.jmb.2008.07.046 18675824 141. Witte E. Witte K. Warszawska K. Sabat R. Wolk K. Interleukin-22: A cytokine produced by T, NK and NKT cell subsets, with importance in the innate immune defense and tissue protection Cytokine Growth Factor Rev. 2010 21 365 379 10.1016/j.cytogfr.2010.08.002 20870448 142. Alam M.S. Maekawa Y. Kitamura A. Tanigaki K. Yoshimoto T. Kishihara K. Yasutomo K. Notch signaling drives IL-22 secretion in CD4+ T cells by stimulating the aryl hydrocarbon receptor Proc. Natl. Acad. Sci. USA 2010 107 5943 5948 10.1073/pnas.0911755107 20231432 143. Czarnowicki T. Esaki H. Gonzalez J. Malajian D. Shemer A. Noda S. Talasila S. Berry A. Gray J. Becker L. Early pediatric atopic dermatitis shows only a cutaneous lymphocyte antigen (CLA)(+) TH2/TH1 cell imbalance, whereas adults acquire CLA+ TH22/TC22 cell subsets J. Allergy Clin. Immunol. 2015 136 941 951 10.1016/j.jaci.2015.05.049 26242300 144. Inoue Y. Aihara M. Kirino M. Harada I. Komori-Yamaguchi J. Yamaguchi Y. Nagashima Y. Ikezawa Z. Interleukin-18 is elevated in the horny layer in patients with atopic dermatitis and is associated with staphylococcus aureus colonization Br. J. Dermatol. 2011 164 560 567 10.1111/j.1365-2133.2010.10145.x 21087228 145. Kim B.S. Choi J.K. Jung H.J. Park K.H. Jang Y.H. Lee W.J. Lee S.J. Kim S.H. Kang H.Y. Kim J.M. Effects of topical application of a recombinant staphylococcal enterotoxin a on dncb and dust mite extract-induced atopic dermatitis-like lesions in a murine model Eur. J. Dermatol. 2014 24 186 193 24721827 146. Lee J.H. Cho D.H. Park H.J. IL-18 and cutaneous inflammatory diseases Int. J. Mol. Sci. 2015 16 29357 29369 10.3390/ijms161226172 26690141 147. Novak N. Valenta R. Bohle B. Laffer S. Haberstok J. Kraft S. Bieber T. Fcepsilonri engagement of langerhans cell-like dendritic cells and inflammatory dendritic epidermal cell-like dendritic cells induces chemotactic signals and different T-cell phenotypes in vitro J. Allergy Clin. Immunol. 2004 113 949 957 10.1016/j.jaci.2004.02.005 15131579 148. Yoshimoto T. Tsutsui H. Tominaga K. Hoshino K. Okamura H. Akira S. Paul W.E. Nakanishi K. IL-18, although antiallergic when administered with IL-12, stimulates IL-4 and histamine release by basophils Proc. Natl. Acad. Sci. USA 1999 96 13962 13966 10.1073/pnas.96.24.13962 10570181 149. Hoshino T. Yagita H. Ortaldo J.R. Wiltrout R.H. Young H.A. In vivo administration of IL-18 can induce IgE production through Th2 cytokine induction and up-regulation of CD40 ligand (CD154) expression on CD4+ T cells Eur. J. Immunol. 2000 30 1998 2006 10.1002/1521-4141(200007)30:7<1998::AID-IMMU1998>3.0.CO;2-U 10940889 150. Lee H.J. Kwon Y.S. Park C.O. Oh S.H. Lee J.H. Wu W.H. Chang N.S. Lee M.G. Lee K.H. Corticotropin-releasing factor decreases IL-18 in the monocyte-derived dendritic cell Exp. Dermatol. 2009 18 199 204 10.1111/j.1600-0625.2008.00781.x 18803662 151. Konishi H. Tsutsui H. Murakami T. Yumikura-Futatsugi S. Yamanaka K. Tanaka M. Iwakura Y. Suzuki N. Takeda K. Akira S. IL-18 contributes to the spontaneous development of atopic dermatitis-like inflammatory skin lesion independently of IgE/STAT6 under specific pathogen-free conditions Proc. Natl. Acad. Sci. USA 2002 99 11340 11345 10.1073/pnas.152337799 12151598 152. Jin J.J. Zou Y.X. Zeng S.W. Risk factors for and expression of immune and inflammatory factors in atopic dermatitis in chinese population: A birth cohort study Mol. Cell. Probes 2016 30 168 173 10.1016/j.mcp.2016.03.006 27040441 153. Tanaka T. Tsutsui H. Yoshimoto T. Kotani M. Matsumoto M. Fujita A. Wang W. Higa S. Koshimoto T. Nakanishi K. Interleukin-18 is elevated in the sera from patients with atopic dermatitis and from atopic dermatitis model mice, NC/NGA Int. Arch. Allergy Immunol. 2001 125 236 240 10.1159/000053821 11490156 154. Thijs J. Krastev T. Weidinger S. Buckens C.F. de Bruin-Weller M. Bruijnzeel-Koomen C. Flohr C. Hijnen D. Biomarkers for atopic dermatitis: A systematic review and meta-analysis Curr. Opin. Allergy Clin. Immunol. 2015 15 453 460 10.1097/ACI.0000000000000198 26226355 155. Zedan K. Rasheed Z. Farouk Y. Alzolibani A.A. Bin Saif G. Ismail H.A. Al Robaee A.A. Immunoglobulin E, interleukin-18 and interleukin-12 in patients with atopic dermatitis: Correlation with disease activity J. Clin. Diagn. Res. 2015 9 Wc01 Wc05 10.7860/JCDR/2015/12261.5742 26023628 156. Sonkoly E. Muller A. Lauerma A.I. Pivarcsi A. Soto H. Kemeny L. Alenius H. Dieu-Nosjean M.C. Meller S. Rieker J. IL-31: A new link between t cells and pruritus in atopic skin inflammation J. Allergy Clin. Immunol. 2006 117 411 417 10.1016/j.jaci.2005.10.033 16461142 157. Ezzat M.H. Hasan Z.E. Shaheen K.Y. Serum measurement of interleukin-31 (IL-31) in paediatric atopic dermatitis: Elevated levels correlate with severity scoring J. Eur. Acad. Dermatol. Venereol. 2011 25 334 339 10.1111/j.1468-3083.2010.03794.x 21294778 158. Cornelissen C. Marquardt Y. Czaja K. Wenzel J. Frank J. Luscher-Firzlaff J. Luscher B. Baron J.M. IL-31 regulates differentiation and filaggrin expression in human organotypic skin models J. Allergy Clin. Immunol. 2012 129 426 433 10.1016/j.jaci.2011.10.042 22177328 159. Danso M.O. van Drongelen V. Mulder A. van Esch J. Scott H. van Smeden J. El Ghalbzouri A. Bouwstra J.A. TNF-α and Th2 cytokines induce atopic dermatitis-like features on epidermal differentiation proteins and stratum corneum lipids in human skin equivalents J. Investig. Dermatol. 2014 134 1941 1950 24518171 160. Zhang Q. Putheti P. Zhou Q. Liu Q. Gao W. Structures and biological functions of IL-31 and IL-31 receptors Cytokine Growth Factor Rev. 2008 19 347 356 10.1016/j.cytogfr.2008.08.003 18926762 161. Kato A. Fujii E. Watanabe T. Takashima Y. Matsushita H. Furuhashi T. Morita A. Distribution of IL-31 and its receptor expressing cells in skin of atopic dermatitis J. Dermatol. Sci. 2014 74 229 235 10.1016/j.jdermsci.2014.02.009 24667097 162. Feld M. Garcia R. Buddenkotte J. Katayama S. Lewis K. Muirhead G. Hevezi P. Plesser K. Schrumpf H. Krjutskov K. The pruritus- and TH2-associated cytokine IL-31 promotes growth of sensory nerves J. Allergy Clin. Immunol. 2016 10.1016/j.jaci.2016.02.020 27212086 163. Nemoto O. Furue M. Nakagawa H. Shiramoto M. Hanada R. Matsuki S. Imayama S. Kato M. Hasebe I. Taira K. The first trial of CIM331, a humanized antihuman interleukin-31 receptor a antibody, in healthy volunteers and patients with atopic dermatitis to evaluate safety, tolerability and pharmacokinetics of a single dose in a randomized, double-blind, placebo-controlled study Br. J. Dermatol. 2016 174 296 304 26409172 164. Lee C.H. Hong C.H. Yu W.T. Chuang H.Y. Huang S.K. Chen G.S. Yoshioka T. Sakata M. Liao W.T. Ko Y.C. Mechanistic correlations between two itch biomarkers, cytokine interleukin-31 and neuropeptide beta-endorphin, via STAT3/calcium axis in atopic dermatitis Br. J. Dermatol. 2012 167 794 803 10.1111/j.1365-2133.2012.11047.x 22578170 165. Hawro T. Saluja R. Weller K. Altrichter S. Metz M. Maurer M. Interleukin-31 does not induce immediate itch in atopic dermatitis patients and healthy controls after skin challenge Allergy 2014 69 113 117 10.1111/all.12316 24251414 166. Hanel K.H. Pfaff C.M. Cornelissen C. Control of the physical and antimicrobial skin barrier by an IL-31-IL-1 signaling network J. Immunol. 2016 196 3233 3244 10.4049/jimmunol.1402943 26944931 167. Jirapongsananuruk O. Hofer M.F. Trumble A.E. Norris D.A. Leung D.Y. Enhanced expression of B7.2 (CD86) in patients with atopic dermatitis: A potential role in the modulation of IgE synthesis J. Immunol. 1998 160 4622 4627 9574570 168. Navarrete-Dechent C. Perez-Mateluna G. Silva-Valenzuela S. Vera-Kellet C. Borzutzky A. Humoral and cellular autoreactivity to epidermal proteins in atopic dermatitis Arch. Immunol. Ther. Exp. 2016 10.1007/s00005-016-0400-3 27147107 169. Heil P.M. Maurer D. Klein B. Hultsch T. Stingl G. Omalizumab therapy in atopic dermatitis: Depletion of IgE does not improve the clinical course—A randomized, placebo-controlled and double blind pilot study J. Dtsch. Dermatol. Ges. 2010 8 990 998 10.1111/j.1610-0387.2010.07497.x 20678148 170. Simon D. Hosli S. Kostylina G. Yawalkar N. Simon H.U. Anti-CD20 (rituximab) treatment improves atopic eczema J. Allergy Clin. Immunol. 2008 121 122 128 10.1016/j.jaci.2007.11.016 18206507 171. Wollenberg A. Wen S. Bieber T. Phenotyping of epidermal dendritic cells: Clinical applications of a flow cytometric micromethod Cytometry 1999 37 147 155 10.1002/(SICI)1097-0320(19991001)37:2<147::AID-CYTO8>3.0.CO;2-Y 10486527 172. Wollenberg A. Sharma S. von Bubnoff D. Geiger E. Haberstok J. Bieber T. Topical tacrolimus (FK506) leads to profound phenotypic and functional alterations of epidermal antigen-presenting dendritic cells in atopic dermatitis J. Allergy Clin. Immunol. 2001 107 519 525 10.1067/mai.2001.112942 11240954 173. Wollenberg A. Wagner M. Gunther S. Towarowski A. Tuma E. Moderer M. Rothenfusser S. Wetzel S. Endres S. Hartmann G. Plasmacytoid dendritic cells: A new cutaneous dendritic cell subset with distinct role in inflammatory skin diseases J. Investig. Dermatol. 2002 119 1096 1102 10.1046/j.1523-1747.2002.19515.x 12445198 174. Ito T. Wang Y.H. Duramad O. Hori T. Delespesse G.J. Watanabe N. Qin F.X. Yao Z. Cao W. Liu Y.J. TSLP-activated dendritic cells induce an inflammatory T helper type 2 cell response through OX40 ligand J. Exp. Med. 2005 202 1213 1223 10.1084/jem.20051135 16275760 175. Gros E. Petzold S. Maintz L. Bieber T. Novak N. Reduced ifn-gamma receptor expression and attenuated IFN-γ response by dendritic cells in patients with atopic dermatitis J. Allergy Clin. Immunol. 2011 128 1015 1021 10.1016/j.jaci.2011.05.043 21774972 176. Ziegler S.F. Thymic stromal lymphopoietin and allergic disease J. Allergy Clin. Immunol. 2012 130 845 852 10.1016/j.jaci.2012.07.010 22939755 177. Fujita H. Nograles K.E. Kikuchi T. Gonzalez J. Carucci J.A. Krueger J.G. Human langerhans cells induce distinct IL-22-producing CD4+ T cells lacking IL-17 production Proc. Natl. Acad. Sci. USA 2009 106 21795 21800 10.1073/pnas.0911472106 19996179 178. Dijkstra D. Stark H. Chazot P.L. Shenton F.C. Leurs R. Werfel T. Gutzmer R. Human inflammatory dendritic epidermal cells express a functional histamine H4 receptor J. Investig. Dermatol. 2008 128 1696 1703 10.1038/sj.jid.5701250 18239617 179. Gschwandtner M. Schakel K. Werfel T. Gutzmer R. Histamine H4 receptor activation on human slan-dendritic cells down-regulates their pro-inflammatory capacity Immunology 2011 132 49 56 10.1111/j.1365-2567.2010.03336.x 20722760 180. Murray C. Ahrens K. Devalaraja M. Dymond M. Fagura M. Hargreaves A. Holt A. Peers I. Price S. Reens J. Use of a canine model of atopic dermatitis to investigate the efficacy of a CCR4 antagonist in allergen-induced skin inflammation in a randomized study J. Investig. Dermatol. 2016 136 665 671 10.1016/j.jid.2015.11.001 26876716 181. Morita E. Takahashi H. Niihara H. Dekio I. Sumikawa Y. Murakami Y. Matsunaka H. Stratum corneum tarc level is a new indicator of lesional skin inflammation in atopic dermatitis Allergy 2010 65 1166 1172 10.1111/j.1398-9995.2010.02361.x 20374230 182. Miyahara H. Okazaki N. Nagakura T. Korematsu S. Izumi T. Elevated umbilical cord serum TARC/CCL17 levels predict the development of atopic dermatitis in infancy Clin. Exp. Allergy 2011 41 186 191 10.1111/j.1365-2222.2010.03634.x 21054588 183. Onoue A. Kabashima K. Kobayashi M. Mori T. Tokura Y. Induction of eosinophIL- and Th2-attracting epidermal chemokines and cutaneous late-phase reaction in tape-stripped skin Exp. Dermatol. 2009 18 1036 1043 10.1111/j.1600-0625.2009.00899.x 19558500 184. Miyano K. Matsushita S. Tsuchida T. Nakamura K. Inhibitory effect of a histamine 4 receptor antagonist on CCL17 and CCL22 production by monocyte-derived langerhans cells in patients with atopic dermatitis J. Dermatol. 2016 10.1111/1346-8138.13294 26892271 185. Kleinewietfeld M. Puentes F. Borsellino G. Battistini L. Rotzschke O. Falk K. CCR6 expression defines regulatory effector/memory-like cells within the CD25+CD4+ T-cell subset Blood 2005 105 2877 2886 10.1182/blood-2004-07-2505 15613550 186. Esche C. de Benedetto A. Beck L.A. Keratinocytes in atopic dermatitis: Inflammatory signals Curr. Allergy Asthma Rep. 2004 4 276 284 10.1007/s11882-004-0071-8 15175141 187. Gahr N. Folster-Holst R. Weichenthal M. Christophers E. Schroder J.M. Bartels J. Dermal fibroblasts from acute inflamed atopic dermatitis lesions display increased eotaxin/CCL11 responsiveness to interleukin-4 stimulation Br. J. Dermatol. 2011 164 586 592 10.1111/j.1365-2133.2010.10112.x 21039413 188. Yawalkar N. Uguccioni M. Scharer J. Braunwalder J. Karlen S. Dewald B. Braathen L.R. Baggiolini M. Enhanced expression of eotaxin and CCR3 in atopic dermatitis J. Investig. Dermatol. 1999 113 43 48 10.1046/j.1523-1747.1999.00619.x 10417617 189. Park C.W. Lee B.H. Han H.J. Lee C.H. Ahn H.K. Tacrolimus decreases the expression of eotaxin, CCR3, rantes and interleukin-5 in atopic dermatitis Br. J. Dermatol. 2005 152 1173 1181 10.1111/j.1365-2133.2005.06474.x 15948978 190. Chong S. Lan H. Zeng K. Zhao X. Serum fractalkine (CX3CL1) concentration correlates with clinical severity in pediatric atopic dermatitis patients Ann. Clin. Lab. Sci. 2016 46 168 173 27098623 191. Echigo T. Hasegawa M. Shimada Y. Takehara K. Sato S. Expression of fractalkine and its receptor, CX3CR1, in atopic dermatitis: Possible contribution to skin inflammation J. Allergy Clin. Immunol. 2004 113 940 948 10.1016/j.jaci.2004.02.030 15131578 192. Siegel A.M. Stone K.D. Cruse G. Lawrence M.G. Olivera A. Jung M.Y. Barber J.S. Freeman A.F. Holland S.M. O’Brien M. Diminished allergic disease in patients with STAT3 mutations reveals a role for STAT3 signaling in mast cell degranulation J. Allergy Clin. Immunol. 2013 132 1388 1396 10.1016/j.jaci.2013.08.045 24184145 193. Mekori Y.A. Metcalfe D.D. Mast cell-T cell interactions J. Allergy Clin. Immunol. 1999 104 517 523 10.1016/S0091-6749(99)70316-7 10482820 194. Theiner G. Gessner A. Lutz M.B. The mast cell mediator PGD2 suppresses IL-12 release by dendritic cells leading to Th2 polarized immune responses in vivo Immunobiology 2006 211 463 472 10.1016/j.imbio.2006.05.020 16920486 195. Horsmanheimo L. Harvima I.T. Jarvikallio A. Harvima R.J. Naukkarinen A. Horsmanheimo M. Mast cells are one major source of interleukin-4 in atopic dermatitis Br. J. Dermatol. 1994 131 348 353 10.1111/j.1365-2133.1994.tb08522.x 7918008 196. Nagarkar D.R. Poposki J.A. Comeau M.R. Biyasheva A. Avila P.C. Schleimer R.P. Kato A. Airway epithelial cells activate Th2 cytokine production in mast cells through IL-1 and thymic stromal lymphopoietin J. Allergy Clin. Immunol. 2012 130 225 232 10.1016/j.jaci.2012.04.019 22633328 197. Gauchat J.F. Henchoz S. Mazzei G. Aubry J.P. Brunner T. Blasey H. Life P. Talabot D. Flores-Romo L. Thompson J. Induction of human IgE synthesis in B cells by mast cells and basophils Nature 1993 365 340 343 10.1038/365340a0 7690905 198. Jawdat D.M. Albert E.J. Rowden G. Haidl I.D. Marshall J.S. IgE-mediated mast cell activation induces langerhans cell migration in vivo J. Immunol. 2004 173 5275 5282 10.4049/jimmunol.173.8.5275 15470073 199. Suto H. Nakae S. Kakurai M. Sedgwick J.D. Tsai M. Galli S.J. Mast cell-associated tnf promotes dendritic cell migration J. Immunol. 2006 176 4102 4112 10.4049/jimmunol.176.7.4102 16547246 200. Modena B.D. Dazy K. White A.A. Emerging concepts: Mast cell involvement in allergic diseases Transl. Res. 2016 174 98 121 10.1016/j.trsl.2016.02.011 26976119 201. Saluja R. Ketelaar M.E. Hawro T. Church M.K. Maurer M. Nawijn M.C. The role of the IL-33/IL-1RL1 axis in mast cell and basophil activation in allergic disorders Mol. Immunol. 2015 63 80 85 10.1016/j.molimm.2014.06.018 25017307 202. Sehra S. Serezani A.P. Ocana J.A. Travers J.B. Kaplan M.H. Mast cells regulate epidermal barrier function and the development of allergic skin inflammation J. Investig. Dermatol. 2016 136 1429 1437 10.1016/j.jid.2016.03.019 27021404 203. Simon D. Braathen L.R. Simon H.U. Eosinophils and atopic dermatitis Allergy 2004 59 561 570 10.1111/j.1398-9995.2004.00476.x 15147438 204. Yang D. Chen Q. Su S.B. Zhang P. Kurosaka K. Caspi R.R. Michalek S.M. Rosenberg H.F. Zhang N. Oppenheim J.J. EosinophIL-derived neurotoxin acts as an alarmin to activate the TLR2-MyD88 signal pathway in dendritic cells and enhances Th2 immune responses J. Exp. Med. 2008 205 79 90 10.1084/jem.20062027 18195069 205. Abu-Ghazaleh R.I. Gleich G.J. Prendergast F.G. Interaction of eosinophil granule major basic protein with synthetic lipid bilayers: A mechanism for toxicity J. Membr. Biol. 1992 128 153 164 10.1007/BF00231888 1501242 206. Kagi M.K. Joller-Jemelka H. Wuthrich B. Correlation of eosinophils, eosinophil cationic protein and soluble interleukin-2 receptor with the clinical activity of atopic dermatitis Dermatology 1992 185 88 92 10.1159/000247419 1421636 207. Cheung P.F. Wong C.K. Ho A.W. Hu S. Chen D.P. Lam C.W. Activation of human eosinophils and epidermal keratinocytes by Th2 cytokine IL-31: Implication for the immunopathogenesis of atopic dermatitis Int. Immunol. 2010 22 453 467 10.1093/intimm/dxq027 20410259 208. Kunsleben N. Rudrich U. Gehring M. Novak N. Kapp A. Raap U. IL-31 induces chemotaxis, calcium mobilization, release of reactive oxygen species, and CCL26 in eosinophils, which are capable to release IL-31 J. Investig. Dermatol. 2015 135 1908 1911 10.1038/jid.2015.106 25789701 209. Kim B.S. Wang K. Siracusa M.C. Saenz S.A. Brestoff J.R. Monticelli L.A. Noti M. Tait Wojno E.D. Fung T.C. Kubo M. Basophils promote innate lymphoid cell responses in inflamed skin J. Immunol. 2014 193 3717 3725 10.4049/jimmunol.1401307 25156365 210. Jiao D. Wong C.K. Qiu H.N. Dong J. Cai Z. Chu M. Hon K.L. Tsang M.S. Lam C.W. NOD2 and TLR2 ligands trigger the activation of basophils and eosinophils by interacting with dermal fibroblasts in atopic dermatitis-like skin inflammation Cell. Mol. Immunol. 2015 10.1038/cmi.2015.77 26388234 211. Wong C.K. Chu I.M. Hon K.L. Tsang M.S. Lam C.W. Aberrant expression of bacterial pattern recognition receptor NOD2 of basophils and microbicidal peptides in atopic dermatitis Molecules 2016 21 417 10.3390/molecules21040471 27070566 212. Spits H. Cupedo T. Innate lymphoid cells: Emerging insights in development, lineage relationships, and function Annu. Rev. Immunol. 2012 30 647 675 10.1146/annurev-immunol-020711-075053 22224763 213. Salimi M. Barlow J.L. Saunders S.P. Xue L. Gutowska-Owsiak D. Wang X. Huang L.C. Johnson D. Scanlon S.T. McKenzie A.N. A role for IL-25 and IL-33-driven type-2 innate lymphoid cells in atopic dermatitis J. Exp. Med. 2013 210 2939 2950 10.1084/jem.20130351 24323357 214. Roediger B. Kyle R. Le Gros G. Weninger W. Dermal group 2 innate lymphoid cells in atopic dermatitis and allergy Curr. Opin. Immunol. 2014 31 108 114 10.1016/j.coi.2014.10.008 25459002 215. Doherty T.A. Broide D.H. Group 2 innate lymphoid cells: New players in human allergic diseases J. Investig. Allergol. Clin. Immunol. 2015 25 1 11 25898689 216. Saunders S.P. Moran T. Floudas A. Wurlod F. Kaszlikowska A. Salimi M. Quinn E.M. Oliphant C.J. Nunez G. McManus R. Spontaneous atopic dermatitis is mediated by innate immunity, with the secondary lung inflammation of the atopic march requiring adaptive immunity J. Allergy Clin. Immunol. 2016 137 482 491 10.1016/j.jaci.2015.06.045 26299987 217. Kim B.S. Innate lymphoid cells in the skin J. Investig. Dermatol. 2015 135 673 678 10.1038/jid.2014.401 25339380 218. Savinko T. Matikainen S. Saarialho-Kere U. Lehto M. Wang G. Lehtimaki S. Karisola P. Reunala T. Wolff H. Lauerma A. IL-33 and ST2 in atopic dermatitis: Expression profiles and modulation by triggering factors J. Investig. Dermatol. 2012 132 1392 1400 10.1038/jid.2011.446 22277940 219. Xue L. Salimi M. Panse I. Mjosberg J.M. McKenzie A.N. Spits H. Klenerman P. Ogg G. Prostaglandin D2 activates group 2 innate lymphoid cells through chemoattractant receptor-homologous molecule expressed on Th2 cells J. Allergy Clin. Immunol. 2014 133 1184 1194 10.1016/j.jaci.2013.10.056 24388011 220. Halim T.Y. Hwang Y.Y. Scanlon S.T. Zaghouani H. Garbi N. Fallon P.G. McKenzie A.N. Group 2 innate lymphoid cells license dendritic cells to potentiate memory Th2 cell responses Nat. Immunol. 2016 17 57 64 10.1038/ni.3294 26523868 221. Roediger B. Kyle R. Yip K.H. Sumaria N. Guy T.V. Kim B.S. Mitchell A.J. Tay S.S. Jain R. Forbes-Blom E. Cutaneous immunosurveillance and regulation of inflammation by group 2 innate lymphoid cells Nat. Immunol. 2013 14 564 573 10.1038/ni.2584 23603794 222. Kim B.S. Siracusa M.C. Saenz S.A. Noti M. Monticelli L.A. Sonnenberg G.F. Hepworth M.R. van Voorhees A.S. Comeau M.R. Artis D. Tslp elicits IL-33-independent innate lymphoid cell responses to promote skin inflammation Sci. Transl. Med. 2013 5 10.1126/scitranslmed.3005374 23363980 223. Wan Y.Y. Gata3: A master of many trades in immune regulation Trends Immunol. 2014 35 233 242 10.1016/j.it.2014.04.002 24786134 224. Ho I.C. Tai T.S. Pai S.Y. Gata3 and the T-cell lineage: Essential functions before and after T-helper-2-cell differentiation Nat. Rev. Immunol. 2009 9 125 135 10.1038/nri2476 19151747 225. Datta S. Milner J.D. Altered T-cell receptor signaling in the pathogenesis of allergic disease J. Allergy Clin. Immunol. 2011 127 351 354 10.1016/j.jaci.2010.11.033 21281865 226. Lee H.S. Choi E.J. Choi H. Lee K.S. Kim H.R. Na B.R. Kwon M.S. Jeong G.S. Choi H.G. Choi E.Y. Oral administration of 4-hydroxy-3-methoxycinnamaldehyde attenuates atopic dermatitis by inhibiting T cell and keratinocyte activation PLoS ONE 2015 10 1234 10.1371/journal.pone.0144521 26656486 227. Kim K.W. Myers R.A. Lee J.H. Igartua C. Lee K.E. Kim Y.H. Kim E.J. Yoon D. Lee J.S. Hirota T. Genome-wide association study of recalcitrant atopic dermatitis in Korean children J. Allergy Clin. Immunol. 2015 136 678 684 10.1016/j.jaci.2015.03.030 25935106 228. Hussein Y.M. Alzahrani S.S. Alharthi A.A. Alhazmi A.S. Ghonaim M.M. Alghamdy A.A. El Askary A. Gene polymorphism of interleukin-4, interleukin-4 receptor and STAT6 in children with atopic dermatitis in Taif, Saudi Arabia Immunol. Investig. 2016 45 223 234 10.3109/08820139.2015.1135943 27018548 229. Chen Y.Q. Shi H.Z. CD28/CTLA-4–CD80/CD86 and ICOS–B7RP-1 costimulatory pathway in bronchial asthma Allergy 2006 61 15 26 10.1111/j.1398-9995.2006.01008.x 16364152 230. Melnik B.C. Does therapeutic intervention in atopic dermatitis normalize epidermal notch deficiency? Exp. Dermatol. 2014 23 696 700 10.1111/exd.12460 24889007 231. Melnik B.C. The potential role of impaired notch signalling in atopic dermatitis Acta Derm. Venereol. 2015 95 5 11 10.2340/00015555-1898 24853951 232. Zhang Q. Wang C. Liu Z. Liu X. Han C. Cao X. Li N. Notch signal suppresses toll-like receptor-triggered inflammatory responses in macrophages by inhibiting extracellular signal-regulated kinase 1/2-mediated nuclear factor κB activation J. Biol. Chem. 2012 287 6208 6217 10.1074/jbc.M111.310375 22205705 233. Bailis W. Yashiro-Ohtani Y. Fang T.C. Hatton R.D. Weaver C.T. Artis D. Pear W.S. Notch simultaneously orchestrates multiple helper t cell programs independently of cytokine signals Immunity 2013 39 148 159 10.1016/j.immuni.2013.07.006 23890069 234. Wilson S.R. The L. Batia L.M. Beattie K. Katibah G.E. McClain S.P. Pellegrino M. Estandian D.M. Bautista D.M. The epithelial cell-derived atopic dermatitis cytokine TSLP activates neurons to induce itch Cell 2013 155 285 295 10.1016/j.cell.2013.08.057 24094650 235. Elias P.M. Hatano Y. Williams M.L. Basis for the barrier abnormality in atopic dermatitis: Outside-inside-outside pathogenic mechanisms J. Allergy Clin. Immunol. 2008 121 1337 1343 10.1016/j.jaci.2008.01.022 18329087

PMC005xxxxxx/PMC5000633.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081235ijms-17-01235ArticleBiological Effect of a Hybrid Anticancer Agent Based on Kinase and Histone Deacetylase Inhibitors on Triple-Negative (MDA-MB231) Breast Cancer Cells Librizzi Mariangela 1Spencer John 2Luparello Claudio 1*Matsuzawa Atsushi Academic EditorMalemud Charles J. Academic Editor1 Dipartimento di Scienze e Tecnologie Biologiche, Chimiche, Farmaceutiche (STEBICEF), Edificio 16, Università di Palermo, Viale delle Scienze, Palermo 90128, Italy; [email protected] Department of Chemistry, School of Life Sciences, University of Sussex, Falmer, Brighton BN1 9QJ, UK; [email protected]* Correspondence: [email protected]; Tel.: +39-091-2389-740530 7 2016 8 2016 17 8 123518 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).We examined the effects of the histone deacetylase inhibitor (HDACi) suberoylanilide hydroxamic acid (SAHA) combined with the vascular endothelial growth factor receptor-1/2 inhibitor (3Z)-5-hydroxy-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-2-one on MDA-MB-231 breast cancer cells (triple-negative) in the form of both a cocktail of the separate compounds and a chemically synthesized hybrid (N-hydroxy-N'-[(3Z)-2-oxo-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-5-yl]octanediamide). Comparative flow cytometric and Western blot analyses were performed on cocktail- and hybrid-treated cells to evaluate cell cycle distribution, autophagy/apoptosis modulation, and mitochondrial metabolic state in order to understand the cellular basis of the cytotoxic effect. Cell cycle analysis showed a perturbation of the rate of progression through the cycle, with aspects of redistribution of cells over different cycle phases for the two treatments. In addition, the results suggest that the two distinct classes of compounds under investigation could induce cell death by different preferential pathways, i.e., autophagy inhibition (the cocktail) or apoptosis promotion (the hybrid), thus confirming the enhanced potential of the hybrid approach vs. the combination approach in finely tuning the biological activities of target cells and also showing the hybrid compound as an additional promising drug-like molecule for the prevention or therapy of “aggressive” breast carcinoma. breast cancerMDA-MB231 cellshistone deacetylase inhibitorvascular endothelial growth factor receptor-2 inhibitorcytotoxicitycell cycleapoptosisautophagymitochondrial metabolism ==== Body 1. Introduction It is generally acknowledged that several signaling pathways are involved in the aggressiveness and metastatic potential of malignant tumors. Therefore, the multifactorial nature of cancer illustrates the need for multifunctional therapeutic tools, such as employing the use of more than one compound to modulate different pathways. Combination therapy has been implemented by combining compounds in a “cocktail” of two or more unmodified molecules in a single solution, whereas the production of hybrid compounds is another emerging strategy which has gained popularity in the last decade [1,2]. In a previous publication, we [3] reported the synthesis of a hybrid drug (N-hydroxy-N'-[(3Z)-2-oxo-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-5-yl]octanediamide) based on the merging of fragments of the histone deacetylase inhibitor (HDACi) suberoylanilide hydroxamic acid (SAHA) and the vascular endothelial growth factor-1 and -2 receptor inhibitor (VEGFR1/2i) (3Z)-5-hydroxy-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-2-one (Figure 1), which was effective in reducing the viability of MDA-MB-231 cells, in an in vitro model system for triple-negative breast cancer (TNBC). It is known that, due to a lack of expression of estrogen, progesterone and epidermal growth factor receptor by TNBC cells, this neoplastic cytotype is extremely “aggressive”; moreover, it is endowed with a higher malignant potential than other breast tumor subtypes [4]. Further, the limitation of treatment options has prompted the development of novel drugs or analogues of pre-existing drugs in the attempt to counteract TNBC cell growth. These compounds, on the other hand, necessitate a thorough biological evaluation. Experimental evidence has shown that the two parental molecules tested by us [3] are active in restraining MDA-MB-231 cell survival and growth. Compound 1 was found to affect the cell cycle and promote apoptosis [5], induce polyploidy-dependent senescence [6], and inhibit epidermal growth factor receptor (EGFR) expression, thereby disrupting the associated downstream signaling [7]. The combination of 1 with other drugs in hybrid anti-cancer molecules has proven to be effective, e.g., as reported by Mendoza-Sanchez et al. [8] who developed a bifunctional anti-proliferative 1/estrogen receptor modulator ICI-164,384 compound active on both TNBC (MDA-MB-231) and estrogen receptor–positive (MCF-7) cells. VEGF is a renowned angiogenic growth factor that is recognized by a family of receptor tyrosine kinases, the VEGFRs, thereby promoting endothelial cell proliferation and migration [9]. In addition, it is known that VEGFRs are able to control the biological behavior of other non-endothelial cytotypes, including TNBC cells such as MDA-MB-231, thereby showing the potential to directly monitor tumor cell survival and the development of breast cancer [10]. In particular, the inhibition of VEGFR2 phosphorylation activity, which is directed to switch on the intracellular signalization of phosphatidylInositol 3-kinase (PI3K), AKT and signal transducer and activator of transcription 3 (STAT3), by the natural products xanthatin (a sesquiterpene lactone) and rhamnazin (an O-methylated flavonol), has been shown to significantly inhibit MDA-MB-231 cell growth both in culture and in nude mice [11,12]. In our previous work [3], both the 1/2 cocktail and compound 3 were shown to be cytotoxic on TNBC cells with an approximate 1:3 ratio of their half maximal inhibitory concentration (IC50) at 72 h. On the other hand, although comparable in their final effects on MDA-MB-231 cell survival, the different dose-response curves for the two treatments suggested a likely distinction, at least partially, of their biochemical and molecular mechanisms of action. Since characterization of the biological properties of new drugs is a key point for the evaluation of drugs’ potency within the composite intracellular microenvironment, the present work was aimed at acquiring the initial experimental data on the functional properties of the compounds under study when administered to MDA-MB-231 cell cultures through a panel of flow cytometric and immunoblot assays. 2. Results In order to obtain comparative information on biological parameters of the cytotoxic action of the 1/2 cocktail and 3 on triple-negative MDA-MB-231 cells, cell cycle state, apoptosis induction markers (phosphatydilserine externalization and caspase-8 activation), mitochondrial metabolism and cell redox state markers (mitochondrial transmembrane potential (MMP) and reactive oxygen species (ROS) production), and autophagy markers (acidic vesicular organelle (AVO) and beclin-1 accumulation) were investigated. First, MDA-MB-231 cells were examined for distribution of cell cycle phases, and the results obtained are shown in Figure 2. Exposure to 3 induced a more prominent increase of the G0/G1 phase fraction than that recorded for the 1/2 cocktail (control vs. 1/2 cocktail = 55.69% vs. 64.05%; control vs. 3 = 53.68% vs. 78.66%), indicative of a more pronouncedly restrained progression via the S phase due to the conceivable activation of the corresponding checkpoint. In both experimental conditions, a similar marked decrease of the S phase fraction (control vs. 1/2 cocktail = 34.11% vs. 10.66%; control vs. 3 = 36.82% vs. 8%) was observed, a result that appears noteworthy since in breast cancer this fraction is regarded as prognostic. Moreover, an accumulation of cells in the G2/M phase (control vs. 1/2 cocktail = 10.2% vs. 25.29%; control vs. 3 = 9.5% vs. 13.34%), more conspicuous for the cocktail treatment and indicative of the inhibition of cell division, was also recorded. Literature reports indicate that drug-induced G2/M arrest of MDA-MB-231 cells is consistently associated with apoptosis promotion (e.g., [13]); on the other hand, an increase of the sub-G0/G1 cell fraction, consistent with the occurrence of apoptosis-triggered fragmentation of DNA, was observed at the left of the G0/G1 peak in both treated conditions. To assess if the cytotoxicity of the 1/2 cocktail and 3 were to be ascribed, at least in part, to the onset of programmed cell death, control and exposed cells were submitted to flow cytometric evaluation of apoptosis and mitochondrial metabolism markers. The panel in Figure 3 shows that, compared to controls, exposure to the drugs was associated with an increase of annexin V+/propidium iodide- (Figure 3A) and activated caspase-8+ (Figure 3B) apoptotic cells. In particular, 3 appeared to be more effective than the 1/2 cocktail in promoting phosphatydilserine externalization (3 vs. 1/2 cocktail vs. control = 63.85% vs. 8.21% vs. 0.03%), whereas the extent of the enzyme activation between the two experimental conditions was more comparable (3 vs. 1/2 cocktail vs. control = 24.66% vs. 20.83% vs. 1.34%). Variations of MMP after cell exposure to the drugs were detected using the JC1 probe. As shown in Figure 4, flow cytometry analysis suggests a loss of MMP in treated cells, in particular 3-exposed cells to a higher extent, with the percentage of low red-emitting cells (bottom quadrants) being about 61% and 75% after 72 h of exposure to the 1/2 cocktail and 3, respectively, vs. approximately 35% of control cells. The ability of the drugs to affect the mitochondrial metabolism was also checked by assaying ROS production using a commercial kit, which differentiates between total ROS and superoxide ions. As shown in Figure 5, the dissipation of MMP was mirrored by an enhanced production of ROS (3 vs. 1/2 cocktail vs. control = 9.66% vs. 8.41% vs. 0.94%) including a more moderate increase of the superoxide anion (3 vs. 1/2 cocktail vs. control = 2.75% vs. 3.43% vs. 0.6%). It has been reported that MDA-MB-231 cells have a constitutively high autophagy rate [14], thereby providing cells with energy and basic elements to counterbalance the metabolic stress associated with hypoxia and nutrient shortage and fast proliferation. It is also acknowledged that the inhibition of autophagy sensitizes MDA-MB-231 tumor cells to the lethal effect of chemical and physical agents (e.g., [15,16]). Therefore, in a last set of analyses, we checked whether 1/2 cocktail and 3 might induce a modification of the amount of autolysosomes, also known as AVOs, a hallmark of autophagy, through acridine orange staining. Interestingly, Figure 6 shows that 1/2 cocktail–treated cells underwent a consistent reduction of AVO accumulation, whereas their amount in 3-treated cells was comparable to that of the control (3 vs. 1/2 cocktail vs. control = 96.89% vs. 79.05% vs. 98.23%). The flow cytometric result was confirmed by a reduction in the amount of beclin-1, an essential mediator involved in autophagy machinery, in MDA-MB-231 cells exposed to the 1/2 cocktail, as visualized by Western blot (Figure 7). This revealed autophagy inhibition as a possible further aspect, alternative to apoptotic induction, the level of which was lower than in 3-treated cells, involved in the cytotoxicity exerted by the drug cocktail on the TNBC cell line. 3. Discussion In this paper, we have examined the cytotoxic effect of an HDACi (1) and a kinase inhibitor (2) on MDA-MB-231 cells, derived from a pleural effusion of a TNBC of the basal subtype and endowed with an “aggressive” phenotype in vivo [17]. We compared their biological activity in the form of both a 1:1 cocktail of the separate compounds and a chemically synthesized hybrid, at a concentration equal to their IC50 at 72 h, as already reported [3]. Several previous studies have confirmed the synergistic effects of drugs used in combination as potential anticancer agents. For example, Zhang et al. [18] have shown that a combination of Vorinostat (SAHA, 1) and the antiangiogenic kinase inhibitor Sorafenib, acting on RAF kinase and the VEGFR-2/PDGFR-β pathway, exhibited in vivo synergism on death induction in different tumor cytotypes. Since their findings suggested the activation of the extrinsic apoptotic pathway, in order to assess the biological aspects of the cytotoxicity of the compounds under study, it prompted us to examine whether this could also be the mode of action of the 1/2 cocktail and/or 3. Preliminarily, cell cycle analysis showed a perturbation of the rate of progression through the cycle, with aspects of redistribution of cells over different cycle phases for the two treatments, and a significant increase of the sub-G0/G1 cell population, indicative of DNA fragmentation which might occur in apoptotic cells, after exposure to both the cocktail and the hybrid. The annexin-V assay confirmed that apoptosis was promoted by exposure to the drugs, more prominently in the case of hybrid 3, and the observed caspase-8 activation was suggestive of the occurrence of receptor-mediated death signaling [19]. Interestingly, both treatments were effective in inducing the dissipation of MMP, also in this case more prominently in the presence of 3, and augmenting the levels of ROS, which are classically considered as crucial events in the intrinsic mitochondria-mediated pathway of apoptosis [20]. It is currently acknowledged that caspase-8 may stimulate the mitochondrial pathway of apoptosis via the proteolytic maturation of BH3 interacting-domain death agonist (BID) protein, thereby promoting the permeabilization of the outer mitochondrial membrane and the release of cytochrome c [21,22]. On the other hand, the co-existence of both apoptotic pathways in MDA-MB-231 after treatment with a plant metabolite has been reported [23]. Thus, whether exposure especially to 3 induces apoptosis in MDA-MB-231 cells by converging the death receptor-mediated extrinsic and the mitochondrial intrinsic pathways, although plausible, remains to be determined through further assays. Nevertheless, our cumulative results imply a higher efficacy of 3 in triggering programmed cell death on TNBC cells. Autophagy is a cellular homeostatic function in which cells autodigest cytoplasmic substrates for removal or turnover after sequestration in multi-membrane-bound structures, the autophagic vacuole, and subsequent fusion with lysosomes generating AVOs or autolysosomes [24]. It plays a complex and highly controversial role in breast cancer: on one hand, autophagy can result in cell death, thereby acting as a tumor-suppressor mechanism, but on the other hand, it can exert a cell-protective role via intracellular recycling, providing energy and basic elements which allow tumor cell survival in stress conditions of oxygen and nutrient shortage and rapid proliferative rate [25]. The latter pro-survival role appears to be the effect of the constitutively-elevated autophagy rate of MDA-MB-231 cells [14]. In fact, various publications have demonstrated that inhibition of autophagy in this cell line leads to sensitization to the lethal effect of chemicals and apoptosis activation [14,26,27]. In this paper we have checked whether the drugs under study might modulate the autophagic rate by two different approaches, i.e., the flow cytometric evaluation of AVO accumulation in conjunction with the immunorevelation of the protein marker beclin-1 [28], which is a key component of autophagy machinery belonging to the signal-initiating class III phosphatidylinositol-3 kinase complex [29]. Interestingly, down-regulation of autophagy appeared to be most prominently triggered by the 1/2 cocktail, thus suggesting that the two forms of the examined compounds could induce cell death by different preferential pathways, i.e., autophagy inhibition (1/2 cocktail) or apoptosis promotion (3). It is noteworthy that a preliminary evaluation of the individual IC50 of either 1 or 2 after 72 h of exposure revealed a different potency of the two compounds on MDA-MB-231 cells (approximately 1 μM for 1 [30] and 100 μM for 2), thereby indicating that equipotent co-treatment could have been performed with a 1:100 1/2 cocktail. Surprisingly, this formulation was not able to induce more than about 20% cell loss at the highest concentrations tested (Luparello, unpublished data) and therefore an equimolar mixture of the drugs (10 μM for both), i.e., the same used in our previous publication [3], was used in the present study. Thus, it can be hypothesized that excess 1 might be responsible for the more marked G2/M arrest and involvement of apoptosis induced by treatment with 1:1 1/2 cocktail, in light of previous studies on the effects of SAHA on MDA-MB-231 cell cycle distribution [31] and viability reduction due to activation of programmed cell death [32,33]. In particular, consistent with our findings from these studies, both apoptotic pathways appeared to be triggered, since the studies [32] demonstrated the activation of the intrinsic apoptotic pathway and the occurrence of caspase-3 cleavage, whereas others [33] reported the upregulation of genes associated with the extrinsic apoptotic pathway, such as TNF-related apoptosis-inducing ligand (TRAIL) and caspase-8 among the others. In conclusion, our results further substantiate the impressive potential of the hybrid vs. combination approach in finely tuning the biological activities of target cells and also strongly encourage the further optimization of 3-like molecules to develop a promising additional prevention and/or treatment agent effective against “aggressive” breast carcinomas, such as TNBC. 4. Materials and Methods 4.1. Cell Culture and Drug Treatments MDA-MB-231 breast cancer cells, were available in laboratory stocks and already used for study in our lab (e.g., [14]), were maintained in a RPMI 1640 medium (Sigma, St. Louis, MO, USA) supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% antibiotics (v/v) (100 U/mL penicillin, 100 μg/mL streptomycin and 2.5 mg/L amphotericin B (Invitrogen, Carlsbad, CA, USA)) in a humidified atmosphere at 37 °C in 5% CO2. Cells were detached from flasks with 0.05% trypsin-EDTA, counted, and plated at the required density for treatment once 60%–80% confluency was attained. The HDACi 1 was purchased from Selleck Chemicals, compounds 2 and 3 were synthesized as reported [34]. In light of the values of IC50 at 72 h already reported for 1:1 1/2 cocktail and for 3 [3], the biological assays were performed in the presence of either 10 μM of the 1/2 cocktail or 29 μM of 3. Control cells were exposed to dimethyl sulphoxide, (Santa Cruz Biotechnology, Dallas, TX, USA) at the same concentrations. 4.2. Flow Cytometry Flow cytometric assays were performed according to Librizzi et al. [14]. MMP was verified by use of a fluorescent dye JC1 (Molecular Probes, Eugene, OR, USA), which is selectively taken up into mitochondria, leading to a fluorescence emission shift from green (~529 nm) to red (~590 nm) for intact MMP, although in the case of mitochondrial depolarization a decrease in the red/green fluorescence intensity ratio is observed. The ionophore valinomycin (Sigma), which induces mitochondrial gradient dissipation, was co-incubated at 1 μM concentration with JC1 for positive control. Data were represented as dot plots using Flowing Software v.2.5.1. (Mr. Perttu Terho, Turku Centre for Biotechnology, Turku, Finland), which discriminate, in the bottom quadrants, the amount of cells undergoing a loss of MMP. Quantification of AVOs was obtained via acridine orange staining (Sigma, final concentration of 100 μg/mL) for 20 min (in the dark) prior to analysis. Data are displayed as dot-plots using the Flowing software v.2.5.1., which is able to discriminate cells with increased AVO accumulation in the top quadrants. Cell cycle distribution was checked using propidium iodide stain following pre-incubation with Triton X-100 and RNase A (Sigma), and analyzed with Weasel v.3.0.1. software (Murray Jeffs, Walter & Eliza Hall Institute of Medical Research, Parkville, Australia). Activation of caspase-8 was assessed with a Vybrant FAM caspase-8 assay kit (Molecular Probes) following the manufacturer’s instructions. Data are represented as dot-plots with Flowing software v.2.5.1., which discriminates early and late apoptotic cells with increased enzyme activation in the bottom and top right quadrants, respectively. All the preparations assayed contained both attached and floating cells, and all the analyses were carried out in a FACSCanto apparatus (BD Biosciences, Franklin Lakes, NJ, USA). 4.3. Western Blot Electrophoretic analyses and immunoblots were carried out according to Librizzi et al. [35]. Acknowledgments The work was supported in part by University of Palermo (FFR 2013) for Claudio Luparello. Author Contributions Mariangela Librizzi performed the cell treatments and the flow cytometric and Western blot assays. John Spencer and Claudio Luparello supervised the work. Conflicts of Interest The authors declare no conflict of interest. Abbreviations HDACi histone deacetylase inhibitor SAHA suberoylanilide hydroxamic acid VEGFR1/2i vascular endothelial growth factor-1 and -2 receptor inhibitor TNBC triple-negative breast cancer EGFR epidermal growth factor receptor VEGFR vascular endothelial growth factor receptor PI3K phosphatidylInositol 3-kinase STAT3 signal transducer and activator of transcription 3 IC50 half maximal inhibitory concentration MMP mitochondrial transmembrane potential ROS reactive oxygen species AVO acidic vesicular organelle PDGFR platelet-derived growth factor receptors BID BH3 interacting-domain death agonist TRAIL TNF-related apoptosis-inducing ligand FBS fetal bovine serum EDTA ethylenediaminetetraacetic acid SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis Figure 1 Compounds used in the present study. (A) SAHA (1), (B) (3Z)-5-hydroxy-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-2-one (2); (C) N-hydroxy-N'-[(3Z)-2-oxo-3-(1H-pyrrol-2-ylmethylidene)-2,3-dihydro-1H-indol-5-yl]octanediamide (3). SAHA, suberoylanilide hydroxamic acid. Figure 2 Effect of the 1/2 cocktail and 3 on the MDA-MB-231 cell cycle. DNA profiles of MDA-MB-231 cells following 72 h of culture under control conditions (red line in A,B) and in the presence of either 10 μM 1/2 cocktail (green line in A) or 29 μM 3 (green line in B). Cell distribution in the different cycle phases is reported in the Table (annex). Figure 3 Flow cytometric analysis of control (A,D), 1/2 cocktail–treated (B,E) and 3-treated MDA-MB-231 cells (C,F) stained with annexin V-FITC and propidium iodide for phosphatydilserine externalization (A–C), and with FAM-LETD-FMK caspase-8 reagent for caspase-8 activation (D–F). The percentage in the quadrants of plots A, B and C relates to late apoptotic/necrotic annexin V+/propidium iodide+ cells (top right quadrant) and early apoptotic annexin V+/propidium iodide− cells (bottom right quadrants). The percentage indicated in the quadrants of plots D, E and F refers to late apoptotic/necrotic cells (top right quadrants) and early apoptotic cells (bottom right quadrants) with activated caspase-8. Figure 4 Flow cytometric analysis of untreated (A), valinomycin-treated (positive control) (B), 1/2 cocktail–treated (C) and 3-treated (D) MDA-MB-231 cells stained with JC1 after 72 h of exposure for mitochondrial transmembrane potential (MMP) evaluation. The percentage in the bottom quadrants in each frame relates to low red-emitting cells that underwent MMP dissipation. Figure 5 Flow cytometric analysis of untreated (A), 1/2 cocktail–treated (B) and 3-treated (C) MDA-MB-231 cells stained with two-color reactive oxygen species (ROS) detection reagent after 72 h of exposure for MMP evaluation. The percentage indicated in the bottom right quadrants refers to total ROS overproducing cells, whereas that in the top left quadrants to superoxide anion–overproducing cells. Figure 6 Flow cytometric analysis of untreated (A), 1/2 cocktail–treated (B) and 3-treated (C) MDA-MB-231 cells stained with acridine orange after 72 h of exposure for evaluation of acidic vesicular organelle (AVO) accumulation. The percentage in the top quadrants relates to AVO-positive cells. Figure 7 Western blot analysis of beclin-1. The image shows a prototypical example of a Western blot of total cell lysates after exposure of MDA-MB-231 cells to a 1/2 cocktail and 3 and analyzed with an antibody raised against beclin-1. Treatment with cocktail led to a reduction of the intensity of the beclin-1 band. Actin was immunostained as a loading control. The molecular weight marker is indicated on the right (Cruz Marker, Santa Cruz Biotechnology, Dallas, TX, USA). ==== Refs References 1. Fortin S. Bérubé G. Advances in the development of hybrid anticancer drugs Expert Opin. Drug Discov. 2013 8 1029 1047 10.1517/17460441.2013.798296 23646979 2. Riviere M.K. le Tourneau C. Paoletti X. Dubois F. Zohar S. Designs of drug-combination phase I trials in oncology: A systematic review of the literature Ann. Oncol. 2015 26 669 674 10.1093/annonc/mdu516 25403591 3. Patel H. Chuckowree I. Coxhead P. Guille M. Wang M. Zuckermann A. Williams R.S.B. Librizzi M. Paranal R.N. Bradner J.E. Synthesis of Hybrid Anticancer Agents Based on Kinase and Histone Deacetylase Inhibitors MedChemComm 2014 5 1829 1833 10.1039/C4MD00211C 4. De Laurentiis M.D. Cianniello D. Caputo R. Stanzione B. Arpino G. Cinieri S. Lorusso V. de Placido S. Treatment of triple-negative breast cancer (TNBC): Current options and future perspectives Cancer Treat. Rev. 2010 36 Suppl. 3 580 586 10.1016/S0305-7372(10)70025-6 5. Huang L. Pardee A.B. Suberoylanilide hydroxamic acid as a potential therapeutic agent for human breast cancer treatment Mol. Med. 2000 6 849 866 11126200 6. Xu W.S. Perez G. Ngo L. Gui C.Y. Marks P.A. Induction of polyploidy by histone deacetylase inhibitor: a pathway for antitumor effects Cancer Res. 2005 65 7832 7839 16140952 7. Zhou Q. Shaw P.G. Davidson N.E. Inhibition of histone deacetylase suppresses EGF signaling pathways by destabilizing EGFR mRNA in ER-negative human breast cancer cells Breast Cancer Res. Treat. 2009 117 443 451 10.1007/s10549-008-0148-5 18683042 8. Mendoza-Sanchez R. Cotnoir-White D. Kulpa J. Jutras I. Pottel J. Moitessier N. Mader S. Gleason J.L. Design, synthesis and evaluation of antiestrogen and histone deacetylase inhibitor molecular hybrids Bioorg. Med. Chem. 2015 23 7597 7606 10.1016/j.bmc.2015.11.005 26613635 9. Shibuya M. Vascular endothelial growth factor (VEGF) and its receptor (VEGFR) signaling in angiogenesis: A crucial target for anti- and pro-angiogenic therapies Genes Cancer 2011 2 1097 1105 10.1177/1947601911423031 22866201 10. Ning Q. Liu C. Hou L. Meng M. Zhang X. Luo M. Shao S. Zuo X. Zhao X. Vascular endothelial growth factor receptor-1 activation promotes migration and invasion of breast cancer cells through epithelial-mesenchymal transition PLoS ONE 2013 8 1235 10.1371/journal.pone.0065217 23776453 11. Yu Y. Yu J. Pei C.G. Li Y.Y. Tu P. Gao G.P. Shao Y. Xanthatin, a novel potent inhibitor of VEGFR2 signaling, inhibits angiogenesis and tumor growth in breast cancer cells Int. J. Clin. Exp. Pathol. 2015 8 10355 10364 26617743 12. Yu Y. Cai W. Pei C.G. Shao Y. Rhamnazin, a novel inhibitor of VEGFR2 signaling with potent antiangiogenic activity and antitumor efficacy Biochem. Biophys. Res. Commun. 2015 458 913 919 10.1016/j.bbrc.2015.02.059 25704088 13. Cyr L. Langler R. Lavigne C. Cell cycle arrest and apoptosis responses of human breast epithelial cells to the synthetic organosulfur compound p- methoxyphenyl p -toluenesulfonate Anticancer Res. 2008 28 2753 2763 19035306 14. Librizzi M. Longo A. Chiarelli R. Amin J. Spencer J. Luparello C. Cytotoxic effects of Jay Amin hydroxamic acid (JAHA), a ferrocene-based class I histone deacetylase inhibitor, on triple-negative MDA-MB231 breast cancer cells Chem. Res. Toxicol. 2012 25 2608 2616 10.1021/tx300376h 23094795 15. Apel A. Herr I. Schwarz H. Rodemann H.P. Mayer A. Blocked autophagy sensitizes resistant carcinoma cells to radiation therapy Cancer Res. 2008 68 1485 1494 10.1158/0008-5472.CAN-07-0562 18316613 16. Kanematsu S. Uehara N. Miki H. Yoshizawa K. Kawanaka A. Yuri T. Tsubura A. Autophagy inhibition enhances sulforaphane-induced apoptosis in human breast cancer cells Anticancer Res. 2010 30 3381 3390 20944112 17. Holliday D.L. Speirs V. Choosing the right line for breast cancer research Breast Cancer Res. 2011 13 215 10.1186/bcr2889 21884641 18. Zhang G. Park M.A. Mitchell C. Hamed H. Rahmani M. Martin A.P. Curiel D.T. Yacoub A. Graf M. Lee R. Vorinostat and sorafenib synergistically kill tumor cells via FLIP suppression and CD95 activation Clin. Cancer Res. 2008 14 5385 5399 10.1158/1078-0432.CCR-08-0469 18765530 19. Kruidering M. Evan G. Caspase-8 in apoptosis: The beginning of “the end”? IUBMB Life 2000 50 85 90 10.1080/713803693 11185963 20. Wang L.H. Li H.H. Li M. Wang S. Jiang X.R. Li Y. Ping G.F. Cao Q. Liu X. Fang W.H. SL4, a chalcone-based compound, induces apoptosis in human cancer cells by activation of the ROS/MAPK signalling pathway Cell Prolif. 2015 48 718 728 10.1111/cpr.12226 26500049 21. Kroemer G. Galluzzi L. Brenner C. Mitochondrial membrane permeabilization in cell death Physiol. Rev. 2007 87 99 163 10.1152/physrev.00013.2006 17237344 22. Cullen S.P. Martin S.J. Caspase activation pathways: Some recent progress Cell Death Differ. 2009 16 935 938 10.1038/cdd.2009.59 19528949 23. Lee E. Lee C.G. Yim J.H. Lee H.K. Pyo S. Ramalin-mediated apoptosis is enhanced by autophagy inhibition in human breast cancer cells Phytother. Res. 2016 30 426 438 10.1002/ptr.5544 26676298 24. Fitzwalter B.E. Thorburn A. Recent insights into cell death and autophagy FEBS J. 2015 282 4279 4288 10.1111/febs.13515 26367268 25. Arroyo D.S. Gaviglio E.A. Peralta Ramos J.M. Bussi C. Rodriguez-Galan M.C. Iribarren P. Autophagy in inflammation, infection, neurodegeneration and cancer Int. Immunopharmacol. 2014 18 55 65 10.1016/j.intimp.2013.11.001 24262302 26. Rai G. Mishra S. Suman S. Shukla Y. Resveratrol improves the anticancer effects of doxorubicin in vitro and in vivo models: A mechanistic insight Phytomedicine 2016 23 233 242 10.1016/j.phymed.2015.12.020 26969377 27. Jung C.H. Kim H. Ahn J. Jung S.K. Um M.Y. Son K.H. Kim T.W. Ha T.Y. Anthricin isolated from Anthriscus sylvestris (L.) Hoffm. inhibits the growth of breast cancer cells by inhibiting Akt/mTOR signaling, and its apoptotic effects are enhanced by autophagy inhibition Evid. Based Complement. Altern. Med. 2013 2013 385219 10.1155/2013/385219 23818925 28. Klionsky D.J. Guidelines for the use and interpretation of assays for monitoring autophagy 3rd ed Autophagy 2016 12 1 222 10.1080/15548627.2015.1100356 26799652 29. Toton E. Lisiak N. Sawicka P. Rybczynska M. Beclin-1 and its role as a target for anticancer therapy J. Physiol. Pharmacol. 2014 65 459 467 25179078 30. Joung K.E. Min K.N. An J.Y. Kim D.K. Kong G. Sheen Y.Y. Potent in vivo anti-breast cancer activity of IN-2001, a novel inhibitor of histone deacetylase, in MMTV/c-Neu mice Cancer Res. 2006 66 5394 5402 10.1158/0008-5472.CAN-05-3835 16707467 31. Li Z.H. Zhang X.B. Han X.Q. Feng C.R. Wang F.S. Wang P.G. Shen J. Shi Y.K. Antitumor effects of a novel histone deacetylase inhibitor NK-HDAC-1 on breast cancer Oncol. Rep. 2013 30 499 505 23624828 32. Uehara N. Kanematsu S. Miki H. Yoshizawa K. Tsubura A. Requirement of p38 MAPK for a cell-death pathway triggered by vorinostat in MDA-MB-231 human breast cancer cells Cancer Lett. 2012 315 112 121 10.1016/j.canlet.2011.07.032 22093617 33. Bellarosa D. Bressan A. Bigioni M. Parlani M. Maggi C.A. Binaschi M. SAHA/Vorinostat induces the expression of the CD137 receptor/ligand system and enhances apoptosis mediated by soluble CD137 receptor in a human breast cancer cell line Int. J. Oncol. 2012 41 1486 1494 10.3892/ijo.2012.1551 22797667 34. Spencer J. Novel Hybrid Compounds Patent WO2012025726A1 1 5 2012 35. Librizzi M. Tobiasch E. Luparello C. The conditioned medium from osteo-differentiating human mesenchymal stem cells affects the viability of triple negative MDA-MB231 breast cancer cells Cell Biochem. Funct. 2016 34 7 15 10.1002/cbf.3157 26628086

PMC005xxxxxx/PMC5000634.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081236ijms-17-01236ReviewIs the Mouse a Good Model of Human PPARγ-Related Metabolic Diseases? Pap Attila 1Cuaranta-Monroy Ixchelt 1Peloquin Matthew 2Nagy Laszlo 123*Desvergne Béatrice Academic Editor1 Department of Biochemistry and Molecular Biology, Faculty of Medicine, University of Debrecen, Debrecen H-4012, Hungary; [email protected] (A.P.); [email protected] (I.C.-M.)2 Sanford Burnham Prebys Medical Discovery Institute at Lake Nona, Orlando, FL 32877, USA; [email protected] MTA-DE “Lendulet” Immunogenomics Research Group, University of Debrecen, Debrecen H-4012, Hungary* Correspondence: [email protected]; Tel.: +1-407-745-215030 7 2016 8 2016 17 8 123622 6 2016 21 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).With the increasing number of patients affected with metabolic diseases such as type 2 diabetes, obesity, atherosclerosis and insulin resistance, academic researchers and pharmaceutical companies are eager to better understand metabolic syndrome and develop new drugs for its treatment. Many studies have focused on the nuclear receptor peroxisome proliferator-activated receptor gamma (PPARγ), which plays a crucial role in adipogenesis and lipid metabolism. These studies have been able to connect this transcription factor to several human metabolic diseases. Due to obvious limitations concerning experimentation in humans, animal models—mainly mouse models—have been generated to investigate the role of PPARγ in different tissues. This review focuses on the metabolic features of human and mouse PPARγ-related diseases and the utility of the mouse as a model. PPARγ expressionhuman mutationsmouse modelsmetabolic syndromelipodystrophyligand activation ==== Body 1. Introduction Peroxisome proliferator-activated receptors (PPARs) are ligand-inducible transcription factors of the nuclear receptor superfamily [1]. There are three PPARs in mammals: PPARα, PPARβ/δ and PPARγ. The PPARs forms obligatory heterodimers with retinoid X receptors (RXRs) and bind to PPAR-responsive elements (PPRE), which regulate the expression of different genes involved in adipogenesis, lipid metabolism and inflammation. PPARs have modular structures containing a N-terminal A/B region with a transactivation domain (AF1), a DNA binding domain (DBD) with two zinc-finger motifs and a C-terminal ligand-binding domain (LBD) with the ligand-dependent transactivation function (AF2) [2,3]. PPARγ was first identified in 1992 in Xenopus [4] and then in 1993 in mice [5]. PPARγ is highly expressed in white adipose tissue (WAT) and brown adipose tissue (BAT), where it plays a critical role in adipogenesis, lipid metabolism and insulin sensitivity. PPARγ is expressed at much lower levels in other metabolic tissues, such as liver and muscle but shows a relatively high expression in placenta, where it is a regulator of vascularization [6]. PPARγ is a modulator of lipid metabolism and inflammatory function in macrophages and dendritic cells [7,8,9]. Comparing the expression of PPARγ in human and mouse tissues, both show a very similar expression pattern, suggesting conserved function of PPARγ across species (Figure 1) [10]. PPARγ has two isoforms: PPARγ1 and PPARγ2. While PPARγ1 is expressed in many tissues, PPARγ2 is adipose tissue-specific under normal physiological conditions, however it is also expressed in macrophages [11]. These two isoforms differ at the N-terminal end of the protein, where PPARγ2 contains an additional 28 amino acids in humans and 30 amino acids in mouse that are absent in PPARγ1. The amino acid sequence of human and mouse PPARγ is highly conserved, with only nine amino acids are differing in the PPARγ1 (Figure 2). This suggests a very similar way of folding and DNA binding with RXR [12]. PPARγ can be modulated by posttranslational modifications such as ubiquitination, acetylation, phosphorylation and sumoylation. These modifications confer cell and tissue specificity [13,14]. Within the last two decades, PPARγ became a focus of attention as a transcription factor implicated in metabolic syndrome. Metabolic syndrome is a concerning public health issue worldwide, which is characterized by a cluster of different symptoms, including obesity, insulin resistance, hyperglycemia, hypertension, hypertriglyceridemia and decreased serum HDL cholesterol levels [15]. All of the aforementioned symptoms contribute to cardiovascular disease, the leading cause of death throughout the world. Researchers go to great lengths attempting to understand the human physiology and uncover those genetic, physiological and environmental changes, which contribute to impair metabolic processes. Plenty of studies have demonstrated the central role of PPARγ in metabolic diseases [16]. Thiazolidinediones (TZDs) are synthetic ligands and potent activators of PPARγ; they have been amply used in treating type 2 diabetes (T2D) in the past. Rosiglitazone and pioglitazone have been withdrawn from the US and European market due to critical cardiovascular diseases and bladder cancer as side effects [14,17,18]. Insulin resistance is a major player in the pathogenesis of metabolic syndrome. Furthermore, PPARγ agonists have been reported to modulate insulin sensitivity and glucose metabolism. Thus, it is of general research interest the finding of new PPARγ–modulators that could improve insulin sensitivity with less important side effects. However, clinical investigations involving human subjects have ethical and methodological limitations, creating a need for a physiologically relevant model organism. This need was addressed by using mice as a model organism for metabolic syndrome. Conversely, using model systems to investigate a biological process always raise the question: how good is the model that we use? In this review, we center our attention on the new developments of the field to answer this question. 2. Human Aspects of PPARγ in Metabolic Syndrome 2.1. PPARg Polymorphisms Related to Metabolic Traits without Lipodystrophy There are several mutations described in the PPARγ gene that affects metabolism traits in humans. These mutations have been classified previously as followed: common polymorphisms (Pro12Ala, His477His), dominant-negative (Val290Met, Cys162Tyr), haploinsufficient mutations (Arg425Cys, Phe388Leu), gain of function mutations (Pro115Gln) and promoter variants (P2 C-689T, P4 A-14G) [19]. In 1997, the most studied and well-characterized polymorphism in epidemiologic studies, Pro12Ala (rs1801282), was first described (Table 1). Susceptibility to T2D with the Pro12 allele and resistance with Ala12 allele was described later [20,21]. It has been hypothesized that the increased insulin clearance and sensitivity in Ala12 allele are due to improved lipolysis [22]. Furthermore, the association of LPL activity in vitro and in vivo has been reported [20,23]. In a large meta-analysis, the Pro12Ala SNP has been found to increase body mass index (BMI) [24]. The association of cardiovascular disease (CVD) and the Pro12Ala polymorphism has been widely studied. However, the results are contradictory. Ridker et al. found a protective role of the Ala allele for myocardial infarction risk [25]. However, in 2004 Tobin and colleagues did not find the same protective effect [26]. A more recent meta-analysis found an increased risk of CVD of the Ala allele in Caucasians patients but not in an Asian population [27]. Importantly, in genome-wide association studies (GWAS), the Pro12Ala variant was among those found in type 2 monogenic diabetes [28]. Moreover, this allele also interacts with BMI in regard to increasing insulin resistance [29]. Interestingly, PPARγ rs1801282 polymorphism has also been studied in several populations, and it exhibits population-based susceptibility to different metabolic traits. In overweight Brazilian pubertal sample, this polymorphism showed higher risks of altered insulin levels [30]. Meanwhile, Pro12Ala could predict BMI, overweight, and total cholesterol in females but not in male Taiwanese patients [31]. In children diagnosed with T2D the Pro12Ala polymorphism of PPARγ was significantly associated with obesity and T2D [32]. However, in a Japanese cohort, this polymorphism was not associated with BMI, and visceral and subcutaneous fat accumulation assessed by computed tomography [33]. Furthermore, Pro12Ala allele is a strong predictor for T2D susceptibility in Asian Indian Sikhs and Chinese population [34,35]. In Russian populations, this variant is associated with insulin sensitivity in type 2 diabetic and normoglycemic subjects [36]. PPARγ Pro12Ala polymorphism is associated with insulin sensitivity and BMI in patients with polycystic ovary syndrome (PCOS) [37]. Importantly, lifestyle interventions appeared to be allele-dependent. The association of Pro12 PPARγ carriers with T2D and low physical activity has been described [38]. PPARγ Pro12Ala variant improves glucose homeostasis as a result of regular exercising with a GWAS approach according to results from the HERITAGE Family Study [39]. Interaction of PPARγ Pro12Ala with dietary fat influences plasma lipids in subjects who are at risk for cardiometabolic diseases [40]. Furthermore, the PPARγ SNP rs1175544 influences the weight loss in a longitudinal study with short-term calorie restriction [37]. Importantly, Pro12Ala variant did not affect the response of pioglitazone treatment in patients with T2D (Figure 3) [41]. However, in a genome-wide study using ChIP-seq, RNA-seq and Gro-seq, Soccio et al. showed that PPARγ binding and the response to rosiglitazone depends on SNPs in human and mouse subcutaneous fat tissues and cell lines respectively [42]. The association between five PPARγ promoter variants and T2D has been described in T2D postmenopausal women [53]. In addition, different PPARγ polymorphisms (rs2972164, rs11128598, rs17793951, rs1151996, rs1175541, and rs3856806), contributed to the deterioration of β-cell function in Mexican Americans population with T2D risk [44]. In a large cohort of T2D cases and controls from multiple studies and ethnic groups, Majithia et al. in 2014 described unidentified PPARγ variants. Nine of these 49 variants have reduced activity in adipocyte differentiation and were associated with a higher risk of T2D [54] (Table 1). Hypercholesterolemia and hypertriglyceridemia have also been associated with PPARγ polymorphisms in a large meta-analysis in 2012 by Asselbergs et al. [55]. Moreover, PPARα V162 allele increases total cholesterol and LDL-cholesterol levels. This effect was reduced by carrying the PPARγ T161 allele in patients with non-diabetic coronary heart disease (CHD) [45]. Another polymorphism associated with CHD is C161T in patients with T2D. The phenotype of this SNP was weakening with the presence of P12P homozygote genotype [46]. Moreover, in an Italian cohort, the 93695C > T PPARγ promoter polymorphism was found to have a protective role in acute coronary syndrome [56]. Furthermore, the C1431T PPARγ polymorphism was associated not only with altered plasma lipids during fasting but also with higher risk of an angiography defined CVD [47]. However, in a large meta-analysis there was no statistically significant difference of serum lipids levels in an Asian population carrying this SNP [57]. Epigenetic changes of the PPARγ gene locus have also been found in metabolic syndrome-related diseases. Recently Kokosar et al. investigate methylation and gene expression in adipose tissue in women with PCOS. Methylation and gene expression of PPARγ was inversely correlated in this study [58]. Furthermore, Nilsson et al. described differential DNA methylation in 15,627 sites, representing 7046 genes including PPARγ in adipose tissue from patients with T2D compared to control subjects [59]. PPARγ loss of function mutations have been reported in colorectal cancers. Not surprisingly in 2010 a novel germline mutation in this gene (S289C) was found in a patient with dyslipidemia, obesity, and hypertension not associated with T2D and a large intestine polyp that progressed to adenocarcinoma [48]. 2.2. PPARg Mutations Associated with Lipodystrophy Lipodystrophy is a syndrome characterized by adipose tissue deficiency; this results in ectopic lipid accumulation in organs and causes non-alcoholic fatty liver disease (NAFLD), reduced blood leptin insulin resistance and T2D [60,61]. Human lipodystrophies are genetic or acquired and may be partial or generalized. Familial partial lipodystrophies (FPLD) are diseases relating to abnormal adipose tissue topography and reduction in total fat mass. The FPLDs have been subclassified into three groups: FPLD1, FPLD2 or FPLD3. A set of mutations in PPARγ gene is associated with FPLD3 (Table 1). Patients with dominant-negative mutations in a single allele of PPARγ have partial lipodystrophy and insulin resistance. The FPLD3 clinical presentation is characterized by a deficiency of limb and gluteal fat, meanwhile abdominal and facial fat is usually preserved [62]. The presentation is usually in adulthood, but insulin resistance and lipodystrophy have been described in prepubertal children as well [63,64,65]. In the patients carrying PPARγ F388L mutant, the transcriptional levels of PPARγ were threefold lower than in the wild type in luciferase assay [64]. Two heterozygous mutations (P467L and V290M) were reported in the PPARγ ligand-binding domain and the clinical presentation in three patients was severe insulin resistance, liver steatosis, T2D and hypertension at an early age (Figure 3). Later, patients carrying these mutations were found to have partial lipodystrophy as assessed by a complete evaluation of body composition and fat distribution. There have been approximately 60 patients in the world identified with FPLD3. The most recent reported mutations in the PPARγ gene that has been found in patients with FPLD3 are summarized below. The PPARγ mutation D424N is located in the ligand-binding domain, and the patients carrying this mutation exhibited a loss of function; which is partially restored by adding the PPARγ agonist rosiglitazone during in vitro analysis using luciferase assays [66]. PPARγ H449L mutation was associated with hypertriglyceridemia, insulin resistance, and NAFLD in four patients related with variable severity in the clinical features. Three subjects presented diabetes or impaired glucose tolerance. Pioglitazone therapy in these three patients resulted in a modest improvement in their metabolic control and consistent menstrual cycles in the two female subjects [49]. Novel mutations in PPARγ (R165T and L339X) linked to FPLD3 are associated with a defective transrepression of cellular RAS leading to cellular dysfunction, contributing to the specific FPLD3-linked severe hypertension [50]. Recently, a heterozygous PPARγ mutation c.1040A > C was identified in all five patients of a family. The resulting amino acid substitution is predicted to disrupt critical molecular interactions at the ligand-binding domain [51]. All pathogenic mutations described until 2014 were heterozygous and located in the DNA- or ligand-binding domains of the PPARγ protein. Most of them show dominant negative activity [43,67]. Recently, Dyment et al. described a biallelic mutation at PPARγ that causes a congenital generalized lipodystrophy (E138V and R164W). A female patient presented a particular phenotype since birth: clear general absence of adipose tissue, later during childhood developed hypertriglyceridemia, pancreatitis, refractory diabetes, irregular menses and renal failure [52]. These new mutations open the possibility of analyzing PPARγ sequence in patients with congenital generalized lipodystrophy (CGL) when no mutation in well-established CGL causing genes could be found. Further studies investigating PPARγ binding and general gene expression are needed in patients with partial lipodystrophies and human common polymorphisms. Also iPS technology should be used to the generation of patient-specific cell lines and the differentiation of such cells to adipocytes and other cell types should allow disease-in-a-dish type experiments and molecular dissection of the mutant receptor to cellular processes. 3. Mouse Models for Study the Role of PPARγ in Metabolic Diseases 3.1. PPARγ Full Body Knockout Mice The first attempts to generate whole body PPARγ knockout (KO) mice showed that loss of PPARγ caused impaired terminal differentiation of the trophoblast and placental vascularization resulting in utero lethality of null embryos tetraploid-rescued PPARγ-null mice survived and showed lack of adipose tissues, which established the essential role of PPARγ in adipogenesis (Figure 4) [68]. The solution for generating full body PPARγ null mice was the Mox2-Cre-floxed PPARγ (MORE-PG) KO, in which Cre recombinase is expressed only in epiblast-derived tissues and preserves PPARγ expression in the trophoblast but only 10% reach adulthood [69]. The characteristics of these mice are: lipodystrophy, organomegaly, decreased leptin and adiponectin in plasma, insulin resistance, elevated free fatty acids (FFAs) and hypotension (Table 2). Moreover, these mice show sex-dependent response to rosiglitazone, which induced regrowth of specific fat depots and improved insulin sensitivity in female, but not in male mice. In contrast, rosiglitazone improved glucose homeostasis with further increase in insulin production but not insulin sensitivity in male mice. Due to the high rate of mortality of this full PPARγ deletion, a tamoxifen inducible whole body PPARγ KO system has been used, and together with the MORE-PG mice, they showed a different gene expression of clock genes in relevant metabolic tissues than controls [70]. Expounding on these systems, Sox2Cre is another type of recombination technology for generating epiblast-specific conditional KO mice [71]. Sox2Cre-floxed PPARγ KO mice escape from embryonic lethality due to normal placental angiogenesis. Several diseases affect these full-body PPARγ deficient mice, therefore only some of them reach maturity [6]. The full body ablation of PPARγ using epiblast-specific KO mice gives an excellent opportunity to investigate the physiological effects of global PPARγ deletion in adult mice. The major limitation of the approach is that the pathologies affect multiple organs and therefore cell autonomous and primary effects are difficult to identify and dissect. A possible solution could be the development and more systematic usage of total body inducible and cell type specific inducible KO models in which the recombination can be induced at will in different developmental or diseases states. 3.2. Heterozygous PPARγ Mice Mice heterozygous for PPARγ showed increased insulin sensitivity instead of the expected insulin resistance. These mice showed decreased triglyceride content in metabolic relevant organs due to elevated leptin expression and induction of fatty acid metabolism [72]. Heterozygous PPARγ mice are resistant to high fat diet (HFD) induced obesity and under these conditions, remained more sensitive to insulin than their WT counterparts [73]. This effect may be caused by the release of some genes that are repressed by PPARγ in adipose tissue. Although they use different mechanisms, activation and partial loss of PPARγ both increase insulin sensitivity [74]. Deletion of one PPARγ allele not only affected lipid storage, but mainly in fasting conditions, also reduced the expression of genes involved in glucose uptake and utilization, fatty acid synthesis, lipolysis and glycolysis. These deregulations led to reduce circulating adiponectin levels in the WAT. Expression of metabolic genes decreased in WAT, but was not affected in liver and skeletal muscle. In addition, there was a decrease in the metabolic rate and physical activity of the PPARγ+/− mice, which was abolished by thiazolidinedione treatment, thereby linking regulation of the metabolic rate and physical activity to PPARγ [75]. 3.3. Hypomorph Mouse Model Targeting the exon B of adipose tissue specific PPARγ2 isoform generated the PPARγ KO in WAT (Figure 4). These mice also displayed decreased levels of PPARγ1 [76]. The homozygous (PPARγhyp/hyp) mice are born normally, indicating that the PPARγ2 isoform may not be required for placental development. However, these animals present growth retardation, severe lipodystrophy and about a 40%–50% mortality rate before the age of five weeks. Neonatal PPARγhyp/hyp mice have insulin resistance, hyperinsulinemia, hyperglycemia and fatty liver, which resembles to human CGL. In contrast, adult mice overcome the fatty liver and hyperlipidemia. However, the skeletal muscle and the heart accumulated more lipids and it was associated with glucose intolerance. The PPARγ agonist, rosiglitazone, reversed glucose intolerance, but not the insulin resistance in homozygote mice. Adipogenic markers and PPARγ target genes were reduced. The mild insulin resistance was explained by an up-regulation of β-oxidation in muscle. However, lipid metabolism and β-oxidation genes in the liver remained unchanged. This model demonstrates the compensatory mechanisms in the absence of WAT [76]. 3.4. Ablation of PPARγ2 Isoform Two PPARγ2 KO mouse models were generated [77]. PPARγ2 KO mice generated by Zhang et al. are viable, but have lipodystrophy and reduced leptin and adiponectin plasma levels. The PPARγ2 KO mice have insulin resistance in male but not in female mice. Surprisingly, the insulin resistance, hypertriglyceridemia and liver steatosis in these males could be reversed by PPARγ agonist treatment, demonstrating that PPARγ2 is not essential for TZDs action on insulin sensitivity [78]. Medina-Gomez et al. generated another PPARγ2 KO mouse model [79] that despite normal adipose tissue development, exhibit insulin resistance under chow diet, suggesting that PPARγ2 could modulate insulin sensitivity [79]. In both models in vitro adipocyte differentiation from precursors is impaired. This suggests a compensating mechanism, which protect in vivo adipogenesis. PPARγ2 deletion in the leptin deficient ob/ob background resulted in decreased fat mass, dyslipidemia, β-cell failure and insulin resistance. PPARγ2 isoform prevents lipotoxicity by promoting adipose tissue proliferation and decreasing ectopic lipid deposition in peripheral organs [80]. 3.5. PPARγ Mutant Mice Modeling the human PPARγ dominant negative mutations is important due to its impact in human metabolic diseases. Therefore, researchers generated mouse lines carrying similar dominant negative mutations in the PPARγ gene. Tsai et al. [81] generated a mouse model containing the P465L amino acid substitution in PPARγ (Figure 4), which is the equivalent with human (P467L) mutation [43]. The homozygous P465L PPARγ mutation is lethal, but the heterozygous animals display hypertension and altered adipose tissue distribution similarly to human phenotypes. In contrast with the severe insulin resistance in P467L PPARγ patients, the P465L PPARγ mutant mice have normal insulin sensitivity. However, P465L mutation shows more similarity to humans on obese ob/ob backgrounds [82]. Another dominant negative PPARγ (L466A) mouse model shows lipodystrophy, increased FFA levels, liver steatosis, hypertension and develops mild insulin resistance, when fed with high-fat diet (Figure 4) [83]. Moreover, mice harboring dominant negative mutations of PPARγ show altered adipose tissue localization and distribution revealing a role for PPARγ controlling the fat distribution in the body [81]. One of the models was the knockin of alanine at position 112 (S112A), which blocks the serine phosphorylation results in a constitutively active PPARγ, with elevated serum adiponectin and reduced FFA levels on high-fat diet. This result suggests that modulation of PPARγ phosphorylation may serve as pharmacological target for insulin sensitization [84]. Importantly, the well-known PPARγ2 P12A mutation in human populations was also generated in mice as a P12A knockin model. Homozygous Ala/Ala mice are viable, however they have lean phenotype, improved insulin sensitivity and plasma lipid profiles on chow diet. Heikkinen et al. demonstrates that P12A variant of PPARγ2 is an important modulator in metabolic control, but the effects depend on the metabolic context and gene–environment interactions [85]. Inducible PPARγ knockin mouse model was also developed in which the endogenous PPARγ gene was substituted with recombinant inducible PPARγldi allele. The PPARγldi/+ mouse show reduced fat mass and insulin sensitivity giving a unique model of human conditional lipodystrophy [86]. 3.6. Tissue Specific Ablation of PPARγ 3.6.1. Adipose-Specific PPARγ Knockout Imai et al. selectively deleted PPARγ in adipocytes of adult mice using the tamoxifen-dependent Cre-ERt2 recombination system (Figure 4). The mature PPARγ-null white and brown adipocytes die within a few days, demonstrating that PPARγ is essential for the in vivo survival of mature adipocytes. After some days without tamoxifen fat depots are replaced with newly formed PPARγ-positive adipocytes [87]. Two similar adipose-specific PPARγ KO mice have been published both KO mice use floxed PPARγ knock-in mice crossed with transgenic aP2-Cre mice model. He and colleagues report that WAT and BAT decreased in young mice, has decreased leptin and adiponectin plasma levels, increased circulating FFAs and triglycerides; therefore, developing liver steatosis. However, adipose-specific PPARγ KO mice have insulin resistance in adipose tissue and liver, but not in skeletal muscle when challenged with high-fat diet. Administration of TZDs to these mice improves insulin sensitivity in skeletal muscle and liver, but not in adipose tissue [88]. Jones et al. have published the other adipose-specific PPARγ KO mouse line. These animals exhibited impairment in brown and white adipogenesis and physiology. When fed with high-fat diet, these mice showed decreased weight gain despite hyperphagia, increased triglyceride levels, liver steatosis, reduced adiponectin and leptin levels and did not develop glucose intolerance or insulin resistance [89]. Characterization of in vivo glucose dynamics pointed to improved hepatic glucose metabolism as the basis for preventing high-fat diet-induced insulin resistance [89]. The differences between these rather similar models might be caused by the different expression of aP2-Cre. The aP2 promoter is a direct PPARγ target, such that PPARγ inactivation during differentiation will reduce the levels of Cre resulting different PPARγ inactivation and potentially distinct phenotypes. PPARγ activation by TZDs increases the uptake of fatty acids and the containing capacity of adipocytes. Selective activation of PPARγ in adipocytes can cause whole body insulin sensitization in mice without an increase of body weight [90]. Models of adipose tissue-specific impairment of PPARγ function demonstrate that PPARγ activity is necessary for normal adipose tissue development and maintenance. Recently, Jonker and colleagues identified the fibroblast growth factor 1 (FGF1) as a critical transducer in the process of adipose tissue sensing nutrients and it stay under the regulation of PPARγ via the promoter of FGF1 gene. Interestingly, FGF1 KO have no significant phenotype under standard laboratory care, these mice develop severe diabetic phenotype and impaired adipose tissue expansion with multiple pathologies when challenged with a HFD. The phenotype of FGF1 KO mouse establishes the PPARγ–FGF1 axis as critical for maintaining metabolic homeostasis and insulin sensitization [91]. An in vivo conditional PPARγ KO adipocyte specific (Adipotrack marked cells) was described recently, and this model has been used to elucidate different cell progenitor depots and its importance in adipocyte differentiation within developmental and adult stages [92,93]. 3.6.2. Muscle-Specific Ablation of PPARγ Skeletal muscle is one of the main insulin responsive tissues in the body. Although PPARγ is expressed to a much smaller extent in muscle than in adipose tissue (Figure 1), it is able to induce the expression of genes that regulate glucose uptake. Two independent groups examined mice with targeted PPARγ deletion in skeletal muscle using the creatinine kinase promoter driven Cre-loxP recombination system [94,95]. In the first study, Hevener et al. used older mice and showed that lack of PPARγ in skeletal muscle resulted in adiposity, severe insulin resistance, and being unable to respond to TZD treatment [94]. In another study, Norris et al. used younger mice with targeted deletion of PPARγ in muscle resulting in obese mice with no insulin resistance and remained responsive to TZD treatment [95]. The role of PPARγ in increasing lipid oxidation in muscle has been published [96]. Findings in muscle-specific PPARγ KO mice suggested that PPARγ in muscle can regulate whole-body lipid metabolism and insulin sensitivity, however TZDs have indirect and age dependent effects on skeletal muscle [14,77,95]. 3.6.3. Liver-Specific Disruption of PPARγ PPARγ is expressed most highly in adipose tissue, but is also detectable in many other tissues such as liver (Figure 1), where PPARγ expression is increased in several mouse models of liver steatosis [97,98]. Liver PPARγ disruption has been developed using Cre recombination system driven by liver-specific albumin promoter [98]. Gavrilova et al. deleted PPARγ in the liver of A-ZIP/F-1 lipoatrophic mice. Lack of PPARγ in this lipoatrophic background protected the mice to develop fatty liver by reducing liver triglyceride and increasing serum FFA levels, but these mice have muscle insulin resistance. Liver-specific ablation of PPARγ in mice leads to increased adiposity and insulin resistance, but these mice respond to TZD treatment [98]. However, liver-specific PPARγ disruption on a lipoatrophic background results mice becoming resistant to TZD treatment, indicating that in the absence of WAT, the liver takes over the role of regulating lipid and glucose homeostasis [14,98]. Matsusue et al. also showed that disruption of liver PPARγ in leptin deficient ob/ob mice results decreased hepatic triglyceride accumulation, but elevated serum lipid levels and insulin resistance [99]. These reinforce the pivotal role of PPARγ in the liver regulating lipid homeostasis and protecting other organs from lipotoxicity and insulin resistance [77]. 3.6.4. PPARγ Ablation in Pancreatic Beta Cells Pancreatic β-cells also express PPARγ (Figure 1) [100,101], where activation of PPARγ regulates the expression of genes involved in glucose-stimulated insulin secretion and TZDs can enhance the insulin secretion and insulin sensitivity in mice and human [102]. Surprisingly, deletion of PPARγ in mouse β-cells caused altered islet mass and morphology, but do not affect the whole body glucose homoeostasis. These mice showed weakened TZD response [103]. Another study used a pancreatic-specific PPARγ KO model generated by crossing mice with floxed PPARγ to mice with pdx-1 promoter driven Cre recombinase and showed that loss of PPARγ in the whole pancreas results normal size of β-cell islets, but hyperglycemia and impaired insulin secretion [104]. Vivas et al. shown that ob/ob mice with genetic ablation of PPARγ2, known as POKO mice failed to enlarged its β-cell mass. They identified genes that regulate β-cells proliferation and survival and identified some PPARγ dependent pathways (cholesterol biosynthesis, apoptosis through TGF-β signaling), which are differentially regulated in POKO mice [105]. However, Welters et al. detected minimal changes in gene expression of important β-cells genes in tamoxifen-inducible β-cell-specific PPARγ KO mice, which could be modified with HFD or rosiglitazone treatment. There were no significant differences in body weight, plasma insulin, glucagon and glucose levels when the mice are kept on normal diet. Based on this study PPARγ seems to be not directly essential for normal β-cell function [106]. 3.6.5. Disruption of PPARγ in Macrophages and Dendritic cells PPARγ has an important role in many immune cell types [107]. Many studies have been focused on macrophages and dendritic cells. The PPARγ expression level in these cell types is low, but they have an important role in regulating expression levels of genes involved in lipid homeostasis and immune function [108,109]. Macrophages accumulated in adipose tissue in obese state are able to induce inflammation and affect glucose homeostasis [110]. Generation of inducible macrophage-specific PPARγ KO mouse revealed the importance of this receptor in the macrophages regulation of cholesterol efflux [111]. Macrophage-directed PPARγ KO mice are more predisposed to obesity and insulin resistance after challenged with HFD, however these mice do not have liver steatosis [112]. Macrophage PPARγ has been claimed to be required for normal skeletal muscle and liver insulin sensitivity and for the maturation of anti-inflammatory M2 type macrophages [113]. More recently it has been shown that STAT6 acts as a facilitating factor for PPARγ by promoting DNA binding and increasing number of genes connected to lipid metabolism and inflammatory response in macrophages and dendritic cells [114]. This work also established that M2 activation at the initial transcriptional response could take place without PPARγ. In addition PPARγ is a potent regulator of various processes in dendritic cells; however in vivo model system for DC-specific PPARγ ablation was unavailable for a long time. A recent study has shown that the dendritic cell-specific CD11c-Cre PPARγfl/fl conditional KO mice have spontaneous lung inflammation and emphysema. Using genome wide microarray analysis, they identified potential PPARγ regulated genes in emphysema [115]. Schneider and colleagues found that PPARγ is required for alveolar macrophage differentiation, however absence of PPARγ did not affect the development and recruitment of macrophages and dendritic cells in other tissues such as liver, brain, heart, kidneys, lamina propria and WAT. GM-CSF induces PPARγ expression in fetal monocytes and plays an important role in alveolar macrophage development. Transcriptome analysis of alveolar macrophage precursors from newborn mice showed that PPAR confers a unique alveolar macrophage signature and identity [116]. 4. Testing of Novel PPARγ Modulators in Mice Novel compounds have been developed in the last few years, which might be potential modulators of PPARγ. One of them is Z-551 that has both PPARα agonistic and PPARγ antagonistic activities. The effects of Z-551 were examined in wild type mice on HFD and it could suppress body weight gain, ameliorated insulin resistance and abnormal lipid metabolism, significantly reducing the plasma levels of glucose, FFAs, insulin and leptin [117]. Another potent modulator is a new thiazolidinedione, GQ-177, which has shown a therapeutic potential on diet-induced obesity and atherosclerosis. This molecule was identified as a partial and selective PPARγ agonist, which improved insulin sensitivity and lipid profile without affecting body weight, fat accumulation or bone density in LDLr-/- mice fed with high-fat diet [118]. 5. Comparison of Human and Mouse Findings Several years ago Heikkinen et al. [19] already summarized the role of PPARγ in human and mouse physiology listing the different human mutations found and mouse models generated until 2007. They highlighted the complex function of PPARγ in cell differentiation, inflammation, glucose and lipid homeostasis pointing ahead its role in metabolic diseases. In this current review, we provide an update and focused on human and mouse experiments regarding metabolic syndrome, summarizing the earlier findings and more recent studies as well. As PPARγ effects occur in a tissue specific manner and the different PPARγ full agonists have severe side effects, also suggested the need of tissue selective PPARγ modulation. PPARγ allelic variants are the most common cause of metabolic traits related to the PPARγ gene. However, knockin mice (P465L) carrying a similar human mutation P467L have normal insulin sensitivity in contrast with the severe insulin resistance in these patients. This clearly indicates a difference in the response of the genetic variants between human and mouse. The Pro12Ala mutation of PPARγ2 is a risk factor of weight gains in human obese patients; in contrast, the Ala12 allele improves insulin sensitivity and has protective effect against obesity and type 2 diabetes mellitus in lean patients [20,119]. The Pro12Ala knockin mice show similar phenotypes, the Ala/Ala homozygous animals are leaner and more insulin sensitive than Pro/Pro mice on normal chow, however they put on weight and lost insulin sensitivity on HFD. It was suggested that the Pro12Ala variant is a diet-dependent metabolic sensor with the ability to modify the PPARγ2 efficacy [85]. To our current knowledge this polymorphism in human and in mouse behaves similarly. Although PPARγ binding is conserved in mouse and human orthologous regions, there is a marked difference in reclusion of this transcription factor due to the motif turnover in different species [120]. It is also important to point out that the PPARγ binding retention during mammalian evolution from mouse to human is C/EBPa interdependent [121]. Differences in PPARγ binding due to SNPs in mouse and human adipose tissue have been also reported [42]. Adding complexity to the system, polymorphisms in PPARγ cofactors can also affect insulin and glucose metabolism like PPARGC1A cofactor mutation Gly482Ser. These genes also should be considered T2D risk factors [122]. Phosphorylation of PPARγ at Ser273 by cyclin-dependent kinase 5 (Cdk5) can affect the expression of distinct PPARγ target genes increasing insulin resistance in mouse models. In obesity a variety of cytokines such as TNFα secreted by adipose tissue can induce the Cdk5 dependent PPARγ phosphorylation. Mutation of Ser273 to alanine and RSG could effectively block the Cdk5-mediated phosphorylation of PPARγ [123]. Interestingly, adipose tissue specific Cdk5 KO mice have increased PPARγ phosphorylation and insulin resistance due to ERK dependent phosphorylation [124]. Again underscoring the complexity and redundancy affecting PPARγ activation in vivo. Mouse full body PPARγ KO MORE-PG does not mimic the human lipodystrophy findings (Table 2). Although, crossing these mice with the obese ob/ob model may provide some insight into the human lipodystrophy [98]. On the other hand, Gray et al. hypothesize that manipulations of PPARγ gene in mice generates very similar defects than in humans but these can only be seen when these mice are challenged with a HFD, exposure to low temperature, during exercise or food deprivation [82,125] arguing that external conditions are critical in the development of metabolic phenotypes and diseases in the presence of a particular genetic disposition. Another important recent development is the discovery of PPARγ biallelic mutations in a human patient. These mutations are rare; therefore, fibroblasts and/or tissues derived from these patients are usually limited. Better full body PPARγ KO mice models are needed to elucidate PPARγ related CGL (Table 1 and Table 3). In summary it is likely that a better and more strategic integration of mouse and human phenotypes and data will be possible by direct comparison of disease-in-a-dish type experiments on human derived iPS cell lines and mouse cellular models. As far as animal models are concerned more “dynamic” models allowing recombination of PPARγ in a temporal and tissue specific manner can accelerate the rate of discovery. CRISPR/Cas9 technology can be used to knock out or knock in genes in whole body animal models and in human cell lines; allowing researchers to study at molecular and physiological level the effect of this gene disruption. This can also lead to identification of the tissue and cell type specific roles including tissue-specific gene expression. The extensive usage of genomic and epigenomic approaches are also going to help dissect the gene expression networks coordinated by the receptors. The ultimate goal of therapy should be to develop tissue selective PPARγ modulators to avoid side effects. For this to happen, mouse and human need to go forward hand in hand in an even more intertwined manner. 6. Conclusions Mice have many genes in common with humans (99% of human genes are conserved in mouse genome) and show many similarities in organ physiology, metabolic processes and pathogenesis of different diseases. Mice are excellent model organisms for other reasons as well. First of all, they are small in size and have a short generation time, which makes breeding and housing relatively simple and cost-effective. Second, since the mouse genome is known, the use of genetically modified mouse models in research and preclinical studies has increased. Furthermore, the mouse is the only mammalian model in which it is technically possible to replace a particular mouse gene with its human counterpart. These so-called “humanized” mouse models are able to produce the human version of the protein of interest, or it can be created to carry a mutated version of the human gene, which is known to be associated with a human disease. Comparing the findings from human and mouse PPARγ related metabolic diseases, we can conclude that mice models can generally be used to investigate and more deeply understand the processes of human diseases. However, it is important to know that despite genetic and physiological similarities, mice have a lot of specific features, which make it difficult to extrapolate mouse results to human. Moreover, there are different conditions in mice, such as genetic background, gender, age, diet and environmental conditions, which could further modify the results. In the last decade, genome-wide studies have changed the epidemiological and functional research of PPAR variants in both human and mouse. However, more systemically used epigenetic and transcriptomic analyses are necessary in the different PPARγ mouse mutants for elucidating PPARγ and its cistrome’s role in metabolic syndrome. The future of PPARγ research relies on using humanized mouse models coupled with human iPS cells derived tissues and genome-wide studies for not only clarify the molecular mechanism of PPARγ in its target genes that have an impact on metabolic syndrome conditions, but also to find suitable PPARγ modulators for human insulin resistance and diabetes treatment. Acknowledgments We thank Gerardo Alvarado Contreras and Erika Sari for artwork and the members of the Nagy laboratory for the helpful discussions, especially Tamas Varga and Gergely Nagy. L.N. is supported by grants from the Hungarian Scientific Research Fund (OTKA K100196, K111941 and K116855). Author Contributions Attila Pap: contributed to concept generation, data interpretation, drafting, critical revision and approval of the manuscript; Ixchelt Cuaranta-Monroy: contributed to concept generation, data interpretation, drafting, critical revision and approval of the manuscript; Matthew Peloquin: grammatical and critical revision and approval of the manuscript; and Laszlo Nagy: contributed to concept generation, data interpretation, grammatical and critical revision and approval of the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 PPARγ tissue distribution in human and mouse. We re-analyzed the expression data set GDS596 (human) and GDS592 (mouse) from Su et al. [10] available on NCBI GEO database (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE1133). The expression values were normalized to median and the tissues were selected based on the levels of PPARγ expression. Metabolic tissues are highlighted in blue. WAT = white adipose tissue, BAT = brown adipose tissue, NA = data not available. Figure 2 Comparison of PPARγ protein homology between human and mouse. We used Ensembl database to obtain the protein sequences and then we compared the human and mouse PPARγ sequences with Clustal 2.1 multiple sequence alignment program. For protein modeling of PPARγ Phyre2 web portal was used and for visualization Geneious 9.1.4 software was applied. Figure 3 Human PPARγ mutations. Mutations on the PPARγ2 are marked with a black frame. A/B region = N terminal region with activation function 1; DBD = DNA binding domain; LBD = ligand binding domain; AF2 = activation function 2. Figure 4 Gene editing strategies applied to the mouse PPARγ allele. The different targeting approaches that have been described in the literature are summarized in this figure. ijms-17-01236-t001_Table 1Table 1 Summary of human PPARγ polymorphism associated to metabolic syndrome conditions. Polymorphism Metabolic Traits Involved References Pro12Ala T2D [20,32] Monogenic diabetes [21] Higher BMI [24,32] Altered insulin levels [30] Insulin sensitivity [36] BMI and insulin sensitivity in PCOS [37] P467L V290M Insulin resistance, liver steatosis, T2D and hypertension [43] Promoter variants polymorphism rs29722164 rs11128598 rs17793951 rs1151996 rs1175541 rs3856806 Deterioration of B-cell function [44] V162 Increase total cholesterol and LDL-cholesterol levels [45] C161T CHD in patients with T2D [46] C1431T Altered fasting serum lipids and risk factor for CHD [47] S289C Dyslipidemia, obesity and hypertension [48] H449L Hypertriglyceridemia, insulin resistance and hepatic steatosis, FPLD3 [49] R165T L339X FPLD3 and severe hypertension [50] c.1040A > C FPLD3, Diabetes Mellitus, hypertension and dyslipidemia [51] Biallelic mutation E138V and R164W CGL, hypertriglyceridemia, diabetes mellitus, pancreatitis and renal failure [52] T2D = type 2 diabetes mellitus; BMI = body mass index; PCOS = polycystic ovarian syndrome; LDL = low- density lipoprotein; CHD = coronary heart disease; FPLD3 = familiar partial lipodystrophy 3; CGL = congenital general lipodystrophy. ijms-17-01236-t002_Table 2Table 2 Comparison of the metabolic features of PPARγ whole body and tissue-specific KO mice. Features Mouse Models MORE- PG KO [69] HET-PPARγ [72,75] HYPO- PPARγ [76] PPARγ2 KO [78] Adipo PPARγ KO [88,89] Sc.M. PPARγ KO [94,95] Liver PPARγ KO [98] β-cell PPARγ KO [103] MΦ PPARγ KO [112] Obesity No ↓ ↓ No ↓ (HFD) ↑ (HFD) No No ↑ (HFD) Insuline resistance Yes IS Yes Yes (male) unclear Yes Yes No Yes Glucose tolerance ↓ (male) ND ↓ ↓ ND ↓ ND NC ↓ (HFD) Type 2 diabetes Yes (male) No ND ND Yes Yes ND No ND Lipodystrophy Yes No Yes Yes Yes No No ND ND Liver steatosis No No No No Yes ND No ND No Hypertension hypoten. ND ND ND ND ND ND ND ND Organomegaly Yes No No No ND Yes No ND No Food intake NC ↓ NC NC ↑ (HFD) ↓ NC ND ND Triglicerides ↑ ↓ ↓ NC ↑ ↑ ↑ * ND NC Free fatty acids ↑ ↓ ↑ (fed) ND ↑ ↑ NC ND ND Cholesterol ND ND ND ND ND ND NC ND LDL ↓ Glucose ↑ ND ↑ (fed) NC NC ↑ ↑ * NC ↑ (HFD) Insulin ↑ ↓ ↑ ND ↑ ↑ ↑ * NC ↑ (HFD) Leptin ↓ ↑ ↓ ↓ ↓ ↑ ↑ * ND ↑ Adiponectin ↓ ↑ ↓ ↓ ↓ ND ↓ * ND ↓ TZD effectiveness ND Yes ND Yes partial partial Yes Yes Yes HET = heterozygous; HYPO = hypomorph; Adipo = adipocyte; Sc. M. = skeletal muscle; MΦ = macrophage; HFD = on high fat diet; IS = insulin sensitivity; hypoten = hypotension; male = just in male mice; fed = in fed state; * = only in 40 weeks old mice; NC = not changed; ND = not determined. ijms-17-01236-t003_Table 3Table 3 Comparison of the metabolic features between human and mouse PPARγ mutants. Features Human Mutants Mouse Mutants P12A Mutant [20] P467L Mutant [43,65] F388L Mutant [64] Biallelic E138V R164W [52] P12A Mutant [85] P465L Mutant [81] Obesity Yes No No No No No Insuline resistance Yes Yes Yes Yes IS No Glucose tolerance ND ↓ ND ND ↑ ↑ Type 2 diabetes Yes Yes Yes Yes No No Lipodystrophy No No FPLD3 CGL No redistr. Liver steatosis ND ND No ND ND ND Hypertension ND Yes Yes No ND Yes Organomegaly No ND No Yes No ND Food intake ND ND ND ND NC NC Triglicerides ↑ ↑ ↑ ↑ ↓ NC Free fatty acids ND ND ND ND NC NC Cholesterol ↑ HDL ↓ HDL ↓ NC ↓ NC Glucose ↑ ND ↑ ↑ NC NC Insulin ND ↑ ↑ ↑ NC ↑ (HFD) Leptin ND ND ND ↓ NC ND Adiponectin ND ND ND ↓ NC ND TZD effectiveness ND ND partial ND partial ND FPLD3 = familiar partial lipodystrophy 3; CGL = congenital generalized lipodystrophy; HFD = on high fat diet; HDL = high-density lipoprotein; IS = insulin sensitivity; NC = not changed; redistr. = redistribution of adipose tissue; ND = not determined. ==== Refs References 1. Chawla A. Repa J.J. Evans R.M. Mangelsdorf D.J. Nuclear receptors and lipid physiology: Opening the x-files Science 2001 294 1866 1870 10.1126/science.294.5548.1866 11729302 2. Tontonoz P. Spiegelman B.M. Fat and beyond: The diverse biology of ppargamma Annu. Rev. Biochem. 2008 77 289 312 10.1146/annurev.biochem.77.061307.091829 18518822 3. Poulsen L. Siersbaek M. Mandrup S. Ppars: Fatty acid sensors controlling metabolism Semin. Cell Dev. Biol. 2012 23 631 639 10.1016/j.semcdb.2012.01.003 22273692 4. Dreyer C. Krey G. Keller H. Givel F. Helftenbein G. Wahli W. Control of the peroxisomal beta-oxidation pathway by a novel family of nuclear hormone receptors Cell 1992 68 879 887 10.1016/0092-8674(92)90031-7 1312391 5. Zhu Y. Alvares K. Huang Q. Rao M.S. Reddy J.K. Cloning of a new member of the peroxisome proliferator-activated receptor gene family from mouse liver J. Biol. Chem. 1993 268 26817 26820 8262913 6. Nadra K. Quignodon L. Sardella C. Joye E. Mucciolo A. Chrast R. Desvergne B. Ppargamma in placental angiogenesis Endocrinology 2010 151 4969 4981 10.1210/en.2010-0131 20810566 7. Tontonoz P. Nagy L. Alvarez J.G. Thomazy V.A. Evans R.M. Ppargamma promotes monocyte/macrophage differentiation and uptake of oxidized ldl Cell 1998 93 241 252 10.1016/S0092-8674(00)81575-5 9568716 8. Nagy L. Tontonoz P. Alvarez J.G. Chen H. Evans R.M. Oxidized ldl regulates macrophage gene expression through ligand activation of ppargamma Cell 1998 93 229 240 10.1016/S0092-8674(00)81574-3 9568715 9. Szatmari I. Torocsik D. Agostini M. Nagy T. Gurnell M. Barta E. Chatterjee K. Nagy L. Ppargamma regulates the function of human dendritic cells primarily by altering lipid metabolism Blood 2007 110 3271 3280 10.1182/blood-2007-06-096222 17664351 10. Su A.I. Wiltshire T. Batalov S. Lapp H. Ching K.A. Block D. Zhang J. Soden R. Hayakawa M. Kreiman G. A gene atlas of the mouse and human protein-encoding transcriptomes Proc. Natl. Acad. Sci. USA 2004 101 6062 6067 10.1073/pnas.0400782101 15075390 11. Temelkova-Kurktschiev T. Hanefeld M. Chinetti G. Zawadzki C. Haulon S. Kubaszek A. Koehler C. Leonhardt W. Staels B. Laakso M. Ala12ala genotype of the peroxisome proliferator-activated receptor gamma2 protects against atherosclerosis J. Clin. Endocrinol. Metab. 2004 89 4238 4242 10.1210/jc.2003-032120 15356014 12. Chandra V. Huang P. Hamuro Y. Raghuram S. Wang Y. Burris T.P. Rastinejad F. Structure of the intact ppar-gamma-rxr- nuclear receptor complex on DNA Nature 2008 456 350 356 10.1038/nature07413 19043829 13. Van Beekum O. Fleskens V. Kalkhoven E. Posttranslational modifications of ppar-gamma: Fine-tuning the metabolic master regulator Obesity 2009 17 213 219 10.1038/oby.2008.473 19169221 14. Ahmadian M. Suh J.M. Hah N. Liddle C. Atkins A.R. Downes M. Evans R.M. Ppargamma signaling and metabolism: The good, the bad and the future Nat. Med. 2013 19 557 566 10.1038/nm.3159 23652116 15. Grundy S.M. Metabolic syndrome update Trends Cardiovasc. Med. 2016 26 364 373 10.1016/j.tcm.2015.10.004 26654259 16. Vacca M. Degirolamo C. Mariani-Costantini R. Palasciano G. Moschetta A. Lipid-sensing nuclear receptors in the pathophysiology and treatment of the metabolic syndrome WIREs Syst. Biol. Med. 2011 3 562 587 10.1002/wsbm.137 21755605 17. Kung J. Henry R.R. Thiazolidinedione safety Expert. Opin. Drug Saf. 2012 11 565 579 10.1517/14740338.2012.691963 22616948 18. Nissen S.E. Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes N. Engl. J. Med. 2007 356 2457 2471 10.1056/NEJMoa072761 17517853 19. Heikkinen S. Auwerx J. Argmann C.A. Ppargamma in human and mouse physiology Biochim. Biophys. Acta 2007 1771 999 1013 10.1016/j.bbalip.2007.03.006 17475546 20. Deeb S.S. Fajas L. Nemoto M. Pihlajamaki J. Mykkanen L. Kuusisto J. Laakso M. Fujimoto W. Auwerx J. A pro12ala substitution in ppargamma2 associated with decreased receptor activity, lower body mass index and improved insulin sensitivity Nat. Genet. 1998 20 284 287 10.1038/3099 9806549 21. Gouda H.N. Sagoo G.S. Harding A.H. Yates J. Sandhu M.S. Higgins J.P. The association between the peroxisome proliferator-activated receptor-gamma2 (pparg2) pro12ala gene variant and type 2 diabetes mellitus: A huge review and meta-analysis Am. J. Epidemiol. 2010 171 645 655 10.1093/aje/kwp450 20179158 22. Tonjes A. Stumvoll M. The role of the pro12ala polymorphism in peroxisome proliferator-activated receptor gamma in diabetes risk Curr. Opin. Clin. Nutr. Metab. Care 2007 10 410 414 10.1097/MCO.0b013e3281e389d9 17563457 23. Schneider J. Kreuzer J. Hamann A. Nawroth P.P. Dugi K.A. The proline 12 alanine substitution in the peroxisome proliferator-activated receptor-gamma2 gene is associated with lower lipoprotein lipase activity in vivo Diabetes 2002 51 867 870 10.2337/diabetes.51.3.867 11872694 24. Galbete C. Toledo E. Martinez-Gonzalez M.A. Martinez J.A. Guillen-Grima F. Marti A. Pro12ala variant of the pparg2 gene increases body mass index: An updated meta-analysis encompassing 49,092 subjects Obesity 2013 21 1486 1495 10.1002/oby.20150 23666678 25. Ridker P.M. Cook N.R. Cheng S. Erlich H.A. Lindpaintner K. Plutzky J. Zee R.Y. Alanine for proline substitution in the peroxisome proliferator-activated receptor gamma-2 (pparg2) gene and the risk of incident myocardial infarction Arterioscler. Thromb. Vasc. Biol. 2003 23 859 863 10.1161/01.ATV.0000068680.19521.34 12663371 26. Tobin M.D. Braund P.S. Burton P.R. Thompson J.R. Steeds R. Channer K. Cheng S. Lindpaintner K. Samani N.J. Genotypes and haplotypes predisposing to myocardial infarction: A multilocus case-control study Eur. Heart J. 2004 25 459 467 10.1016/j.ehj.2003.11.014 15039125 27. Wu Z. Lou Y. Jin W. Liu Y. Lu L. Lu G. The pro12ala polymorphism in the peroxisome proliferator-activated receptor gamma-2 gene (ppargamma2) is associated with increased risk of coronary artery disease: A meta-analysis PLoS ONE 2012 7 1236 10.1371/journal.pone.0053105 28. Yang Y. Chan L. Monogenic diabetes: What it teaches us on the common forms of type 1 and type 2 diabetes Endocr. Rev. 2016 37 190 222 10.1210/er.2015-1116 27035557 29. Manning A.K. LaValley M. Liu C.T. Rice K. An P. Liu Y. Miljkovic I. Rasmussen-Torvik L. Harris T.B. Province M.A. Meta-analysis of gene-environment interaction: Joint estimation of snp and snp x environment regression coefficients Genet. Epidemiol. 2011 35 11 18 10.1002/gepi.20546 21181894 30. Queiroz E.M. Candido A.P. Castro I.M. Bastos A.Q. Machado-Coelho G.L. Freitas R.N. Igf2, lepr, pomc, pparg, and ppargc1 gene variants are associated with obesity-related risk phenotypes in brazilian children and adolescents Braz. J. Med. Biol. Res. 2015 48 595 602 10.1590/1414-431X20154155 25923461 31. Hsiao T.J. Lin E. The pro12ala polymorphism in the peroxisome proliferator-activated receptor gamma (pparg) gene in relation to obesity and metabolic phenotypes in a taiwanese population Endocrine 2015 48 786 793 10.1007/s12020-014-0407-7 25182148 32. Dubinina I.A. Chistiakov D.A. Eremina I.A. Brovkin A.N. Zilberman L.I. Nikitin A.G. Kuraeva T.L. Nosikov V.V. Peterkova V.A. Dedov II. Studying progression from glucose intolerance to type 2 diabetes in obese children Diabetes Metab. Syndr. 2014 8 133 137 10.1016/j.dsx.2014.07.002 25127329 33. Hotta K. Kitamoto A. Kitamoto T. Mizusawa S. Teranishi H. So R. Matsuo T. Nakata Y. Hyogo H. Ochi H. Association between type 2 diabetes genetic susceptibility loci and visceral and subcutaneous fat area as determined by computed tomography J. Hum. Genet. 2012 57 305 310 10.1038/jhg.2012.21 22377712 34. Sanghera D.K. Demirci F.Y. Been L. Ortega L. Ralhan S. Wander G.S. Mehra N.K. Singh J. Aston C.E. Mulvihill J.J. Pparg and adipoq gene polymorphisms increase type 2 diabetes mellitus risk in asian indian sikhs: Pro12ala still remains as the strongest predictor Metabolism 2010 59 492 501 10.1016/j.metabol.2009.07.043 19846176 35. Hu C. Zhang R. Wang C. Wang J. Ma X. Lu J. Qin W. Hou X. Wang C. Bao Y. Pparg, kcnj11, cdkal1, cdkn2a-cdkn2b, ide-kif11-hhex, igf2bp2 and slc30a8 are associated with type 2 diabetes in a chinese population PLoS ONE 2009 4 1236 10.1371/journal.pone.0007643 19862325 36. Chistiakov D.A. Potapov V.A. Khodirev D.S. Shamkhalova M.S. Shestakova M.V. Nosikov V.V. The ppargamma pro12ala variant is associated with insulin sensitivity in russian normoglycaemic and type 2 diabetic subjects Diab. Vasc. Dis. Res. 2010 7 56 62 10.1177/1479164109347689 20368233 37. Matsuo T. Nakata Y. Katayama Y. Iemitsu M. Maeda S. Okura T. Kim M.K. Ohkubo H. Hotta K. Tanaka K. Pparg genotype accounts for part of individual variation in body weight reduction in response to calorie restriction Obesity 2009 17 1924 1931 10.1038/oby.2009.199 19543210 38. Nelson T.L. Fingerlin T.E. Moss L.K. Barmada M.M. Ferrell R.E. Norris J.M. Association of the peroxisome proliferator-activated receptor gamma gene with type 2 diabetes mellitus varies by physical activity among non-hispanic whites from colorado Metabolism 2007 56 388 393 10.1016/j.metabol.2006.10.022 17292728 39. Ruchat S.M. Rankinen T. Weisnagel S.J. Rice T. Rao D.C. Bergman R.N. Bouchard C. Perusse L. Improvements in glucose homeostasis in response to regular exercise are influenced by the pparg pro12ala variant: Results from the heritage family study Diabetologia 2010 53 679 689 10.1007/s00125-009-1630-2 20043145 40. Alsaleh A. O'Dell S.D. Frost G.S. Griffin B.A. Lovegrove J.A. Jebb S.A. Sanders T.A. investigators R.S. Interaction of pparg pro12ala with dietary fat influences plasma lipids in subjects at cardiometabolic risk J. Lipid. Res. 2011 52 2298 2303 10.1194/jlr.P019281 21949049 41. Glamoclija U. Jevric-Causevic A. Genetic polymorphisms in diabetes: Influence on therapy with oral antidiabetics Acta Pharm. 2010 60 387 406 10.2478/v10007-010-0040-9 21169132 42. Soccio R.E. Chen E.R. Rajapurkar S.R. Safabakhsh P. Marinis J.M. Dispirito J.R. Emmett M.J. Briggs E.R. Fang B. Everett L.J. Genetic variation determines ppargamma function and anti-diabetic drug response in vivo Cell 2015 162 33 44 10.1016/j.cell.2015.06.025 26140591 43. Barroso I. Gurnell M. Crowley V.E. Agostini M. Schwabe J.W. Soos M.A. Maslen G.L. Williams T.D. Lewis H. Schafer A.J. Dominant negative mutations in human ppargamma associated with severe insulin resistance, diabetes mellitus and hypertension Nature 1999 402 880 883 10622252 44. Black M.H. Wu J. Takayanagi M. Wang N. Taylor K.D. Haritunians T. Trigo E. Lawrence J.M. Watanabe R.M. Buchanan T.A. Variation in pparg is associated with longitudinal change in insulin resistance in mexican americans at risk for type 2 diabetes J. Clin. Endocrinol. Metab. 2015 100 1187 1195 10.1210/jc.2014-3246 25584717 45. Yilmaz-Aydogan H. Kurnaz O. Kucukhuseyin O. Akadam-Teker B. Kurt O. Eronat A.P. Tekeli A. Bugra Z. Ozturk O. Different effects of ppara, pparg and apoe snps on serum lipids in patients with coronary heart disease based on the presence of diabetes Gene 2013 523 20 26 10.1016/j.gene.2013.03.136 23583468 46. Yilmaz-Aydogan H. Kurnaz O. Kurt O. Akadam-Teker B. Kucukhuseyin O. Tekeli A. Isbir T. Effects of the pparg p12a and c161t gene variants on serum lipids in coronary heart disease patients with and without type 2 diabetes Mol. Cell. Biochem. 2011 358 355 363 10.1007/s11010-011-0987-y 21833536 47. Oladi M. Nohtani M. Avan A. Mirhafez S.R. Tajbakhsh A. Ghasemi F. Asadi A. Elahdadi Salmani M. Mohammadi A. Hoseinzadeh L. Impact of the c1431t polymorphism of the peroxisome proliferator activated receptor-gamma (ppar-gamma) gene on fasted serum lipid levels in patients with coronary artery disease Ann. Nutr. Metab. 2015 66 149 154 10.1159/000381358 25896411 48. Capaccio D. Ciccodicola A. Sabatino L. Casamassimi A. Pancione M. Fucci A. Febbraro A. Merlino A. Graziano G. Colantuoni V. A novel germline mutation in peroxisome proliferator-activated receptor gamma gene associated with large intestine polyp formation and dyslipidemia Biochim. Biophys. Acta 2010 1802 572 581 10.1016/j.bbadis.2010.01.012 20123124 49. Demir T. Onay H. Savage D.B. Temeloglu E. Uzum A.K. Kadioglu P. Altay C. Ozen S. Demir L. Cavdar U. Familial partial lipodystrophy linked to a novel peroxisome proliferator activator receptor -gamma (pparg) mutation, h449l: A comparison of people with this mutation and those with classic codon 482 lamin a/c (lmna) mutations Diabet. Med. 2016 10.1111/dme.13061 26756202 50. Auclair M. Vigouroux C. Boccara F. Capel E. Vigeral C. Guerci B. Lascols O. Capeau J. Caron-Debarle M. Peroxisome proliferator-activated receptor-gamma mutations responsible for lipodystrophy with severe hypertension activate the cellular renin-angiotensin system Arterioscler. Thromb. Vasc. Biol. 2013 33 829 838 10.1161/ATVBAHA.112.300962 23393388 51. Miehle K. Porrmann J. Mitter D. Stumvoll M. Glaser C. Fasshauer M. Hoffmann K. Novel peroxisome proliferator-activated receptor gamma mutation in a family with familial partial lipodystrophy type 3 Clin. Endocrinol. 2016 84 141 148 10.1111/cen.12837 26119484 52. Dyment D.A. Gibson W.T. Huang L. Bassyouni H. Hegele R.A. Innes A.M. Biallelic mutations at pparg cause a congenital, generalized lipodystrophy similar to the berardinelli-seip syndrome Eur. J. Med. Genet. 2014 57 524 526 10.1016/j.ejmg.2014.06.006 24980513 53. Chan K.H. Niu T. Ma Y. You N.C. Song Y. Sobel E.M. Hsu Y.H. Balasubramanian R. Qiao Y. Tinker L. Common genetic variants in peroxisome proliferator-activated receptor-gamma (pparg) and type 2 diabetes risk among women's health initiative postmenopausal women J. Clin. Endocrinol. Metab. 2013 98 E600 E604 10.1210/jc.2012-3644 23386649 54. Majithia A.R. Flannick J. Shahinian P. Guo M. Bray M.A. Fontanillas P. Gabriel S.B. Go T.D.C. Project N.J.F.A.S. Consortium S.T.D. Rare variants in pparg with decreased activity in adipocyte differentiation are associated with increased risk of type 2 diabetes Proc. Natl. Acad. Sci. USA 2014 111 13127 13132 10.1073/pnas.1410428111 25157153 55. Asselbergs F.W. Guo Y. van Iperen E.P. Sivapalaratnam S. Tragante V. Lanktree M.B. Lange L.A. Almoguera B. Appelman Y.E. Barnard J. Large-scale gene-centric meta-analysis across 32 studies identifies multiple lipid loci Am. J. Hum. Genet. 2012 91 823 838 10.1016/j.ajhg.2012.08.032 23063622 56. Evangelisti L. Attanasio M. Lucarini L. Sofi F. Marcucci R. Giglioli C. Valente S. Gensini G. Abbate R. Pepe G. Ppargamma promoter polymorphisms and acute coronary syndrome Atherosclerosis 2009 205 186 191 10.1016/j.atherosclerosis.2008.11.009 19117570 57. Li Q. Chen R. Bie L. Zhao D. Huang C. Hong J. Association of the variants in the pparg gene and serum lipid levels: A meta-analysis of 74 studies J. Cell. Mol. Med. 2015 19 198 209 10.1111/jcmm.12417 25265984 58. Kokosar M. Benrick A. Perfilyev A. Fornes R. Nilsson E. Maliqueo M. Behre C.J. Sazonova A. Ohlsson C. Ling C. Epigenetic and transcriptional alterations in human adipose tissue of polycystic ovary syndrome Sci. Rep. 2016 6 22883 10.1038/srep22883 26975253 59. Nilsson E. Jansson P.A. Perfilyev A. Volkov P. Pedersen M. Svensson M.K. Poulsen P. Ribel-Madsen R. Pedersen N.L. Almgren P. Altered DNA methylation and differential expression of genes influencing metabolism and inflammation in adipose tissue from subjects with type 2 diabetes Diabetes 2014 63 2962 2976 10.2337/db13-1459 24812430 60. Garg A. Lipodystrophies: Genetic and acquired body fat disorders J. Clin. Endocrinol. Metab. 2011 96 3313 3325 10.1210/jc.2011-1159 21865368 61. Garg A. Acquired and inherited lipodystrophies N. Engl. J. Med. 2004 350 1220 1234 10.1056/NEJMra025261 15028826 62. Semple R.K. Chatterjee V.K. O’Rahilly S. Ppar gamma and human metabolic disease J. Clin. Investig. 2006 116 581 589 10.1172/JCI28003 16511590 63. Agarwal A.K. Garg A. A novel heterozygous mutation in peroxisome proliferator-activated receptor-gamma gene in a patient with familial partial lipodystrophy J. Clin. Endocrinol. Metab. 2002 87 408 411 11788685 64. Hegele R.A. Cao H. Frankowski C. Mathews S.T. Leff T. Pparg f388l, a transactivation-deficient mutant, in familial partial lipodystrophy Diabetes 2002 51 3586 3590 10.2337/diabetes.51.12.3586 12453919 65. Savage D.B. Tan G.D. Acerini C.L. Jebb S.A. Agostini M. Gurnell M. Williams R.L. Umpleby A.M. Thomas E.L. Bell J.D. Human metabolic syndrome resulting from dominant-negative mutations in the nuclear receptor peroxisome proliferator-activated receptor-gamma Diabetes 2003 52 910 917 10.2337/diabetes.52.4.910 12663460 66. Ludtke A. Buettner J. Schmidt H.H. Worman H.J. New pparg mutation leads to lipodystrophy and loss of protein function that is partially restored by a synthetic ligand J. Med. Genet. 2007 44 e88 10.1136/jmg.2007.050567 17766367 67. Agostini M. Schoenmakers E. Mitchell C. Szatmari I. Savage D. Smith A. Rajanayagam O. Semple R. Luan J. Bath L. Non-DNA binding, dominant-negative, human ppargamma mutations cause lipodystrophic insulin resistance Cell Metab. 2006 4 303 311 10.1016/j.cmet.2006.09.003 17011503 68. Barak Y. Nelson M.C. Ong E.S. Jones Y.Z. Ruiz-Lozano P. Chien K.R. Koder A. Evans R.M. Ppar gamma is required for placental, cardiac and adipose tissue development Mol. Cell 1999 4 585 595 10.1016/S1097-2765(00)80209-9 10549290 69. Duan S.Z. Ivashchenko C.Y. Whitesall S.E. D’Alecy L.G. Duquaine D.C. Brosius F.C. 3rd Gonzalez F.J. Vinson C. Pierre M.A. Milstone D.S. Hypotension, lipodystrophy, and insulin resistance in generalized ppargamma-deficient mice rescued from embryonic lethality J. Clin. Investig. 2007 117 812 822 10.1172/JCI28859 17304352 70. Yang G. Jia Z. Aoyagi T. McClain D. Mortensen R.M. Yang T. Systemic ppargamma deletion impairs circadian rhythms of behavior and metabolism PLoS ONE 2012 7 1236 10.1371/journal.pone.0038117 71. Hayashi S. Lewis P. Pevny L. McMahon A.P. Efficient gene modulation in mouse epiblast using a sox2cre transgenic mouse strain Mech. Dev. 2002 119 S97 S101 10.1016/S0925-4773(03)00099-6 14516668 72. Yamauchi T. Kamon J. Waki H. Murakami K. Motojima K. Komeda K. Ide T. Kubota N. Terauchi Y. Tobe K. The mechanisms by which both heterozygous peroxisome proliferator-activated receptor gamma (ppargamma) deficiency and ppargamma agonist improve insulin resistance J. Biol. Chem. 2001 276 41245 41254 10.1074/jbc.M103241200 11533050 73. Kubota N. Terauchi Y. Miki H. Tamemoto H. Yamauchi T. Komeda K. Satoh S. Nakano R. Ishii C. Sugiyama T. Ppar gamma mediates high-fat diet-induced adipocyte hypertrophy and insulin resistance Mol. Cell 1999 4 597 609 10.1016/S1097-2765(00)80210-5 10549291 74. Miles P.D. Barak Y. He W. Evans R.M. Olefsky J.M. Improved insulin-sensitivity in mice heterozygous for ppar-gamma deficiency J. Clin. Investig. 2000 105 287 292 10.1172/JCI8538 10675354 75. Anghel S.I. Bedu E. Vivier C.D. Descombes P. Desvergne B. Wahli W. Adipose tissue integrity as a prerequisite for systemic energy balance: A critical role for peroxisome proliferator-activated receptor gamma J. Biol. Chem. 2007 282 29946 29957 10.1074/jbc.M702490200 17699161 76. Koutnikova H. Cock T.A. Watanabe M. Houten S.M. Champy M.F. Dierich A. Auwerx J. Compensation by the muscle limits the metabolic consequences of lipodystrophy in ppar gamma hypomorphic mice Proc. Natl. Acad. Sci. USA 2003 100 14457 14462 10.1073/pnas.2336090100 14603033 77. Gray S.L. Dalla Nora E. Vidal-Puig A.J. Mouse models of ppar-gamma deficiency: Dissecting ppar-gamma's role in metabolic homoeostasis Biochem. Soc. Trans. 2005 33 1053 1058 10.1042/BST0331053 16246044 78. Zhang J. Fu M. Cui T. Xiong C. Xu K. Zhong W. Xiao Y. Floyd D. Liang J. Li E. Selective disruption of ppargamma 2 impairs the development of adipose tissue and insulin sensitivity Proc. Natl. Acad. Sci. USA 2004 101 10703 10708 10.1073/pnas.0403652101 15249658 79. Medina-Gomez G. Virtue S. Lelliott C. Boiani R. Campbell M. Christodoulides C. Perrin C. Jimenez-Linan M. Blount M. Dixon J. The link between nutritional status and insulin sensitivity is dependent on the adipocyte-specific peroxisome proliferator-activated receptor-gamma2 isoform Diabetes 2005 54 1706 1716 10.2337/diabetes.54.6.1706 15919792 80. Medina-Gomez G. Gray S.L. Yetukuri L. Shimomura K. Virtue S. Campbell M. Curtis R.K. Jimenez-Linan M. Blount M. Yeo G.S. Ppar gamma 2 prevents lipotoxicity by controlling adipose tissue expandability and peripheral lipid metabolism PLoS Genet. 2007 3 1236 10.1371/journal.pgen.0030064 17465682 81. Tsai Y.S. Kim H.J. Takahashi N. Kim H.S. Hagaman J.R. Kim J.K. Maeda N. Hypertension and abnormal fat distribution but not insulin resistance in mice with p465l ppargamma J. Clin. Investig. 2004 114 240 249 10.1172/JCI200420964 15254591 82. Gray S.L. Nora E.D. Grosse J. Manieri M. Stoeger T. Medina-Gomez G. Burling K. Wattler S. Russ A. Yeo G.S. Leptin deficiency unmasks the deleterious effects of impaired peroxisome proliferator-activated receptor gamma function (p465l ppargamma) in mice Diabetes 2006 55 2669 2677 10.2337/db06-0389 17003330 83. Freedman B.D. Lee E.J. Park Y. Jameson J.L. A dominant negative peroxisome proliferator-activated receptor-gamma knock-in mouse exhibits features of the metabolic syndrome J. Biol. Chem. 2005 280 17118 17125 10.1074/jbc.M407539200 15716267 84. Rangwala S.M. Rhoades B. Shapiro J.S. Rich A.S. Kim J.K. Shulman G.I. Kaestner K.H. Lazar M.A. Genetic modulation of ppargamma phosphorylation regulates insulin sensitivity Dev. Cell 2003 5 657 663 10.1016/S1534-5807(03)00274-0 14536066 85. Heikkinen S. Argmann C. Feige J.N. Koutnikova H. Champy M.F. Dali-Youcef N. Schadt E.E. Laakso M. Auwerx J. The pro12ala ppargamma2 variant determines metabolism at the gene-environment interface Cell Metab. 2009 9 88 98 10.1016/j.cmet.2008.11.007 19117549 86. Kim S. Huang L.W. Snow K.J. Ablamunits V. Hasham M.G. Young T.H. Paulk A.C. Richardson J.E. Affourtit J.P. Shalom-Barak T. A mouse model of conditional lipodystrophy Proc. Natl. Acad. Sci. USA 2007 104 16627 16632 10.1073/pnas.0707797104 17921248 87. Imai T. Takakuwa R. Marchand S. Dentz E. Bornert J.M. Messaddeq N. Wendling O. Mark M. Desvergne B. Wahli W. Peroxisome proliferator-activated receptor gamma is required in mature white and brown adipocytes for their survival in the mouse Proc. Natl. Acad. Sci. USA 2004 101 4543 4547 10.1073/pnas.0400356101 15070754 88. He W. Barak Y. Hevener A. Olson P. Liao D. Le J. Nelson M. Ong E. Olefsky J.M. Evans R.M. Adipose-specific peroxisome proliferator-activated receptor gamma knockout causes insulin resistance in fat and liver but not in muscle Proc. Natl. Acad. Sci. USA 2003 100 15712 15717 10.1073/pnas.2536828100 14660788 89. Jones J.R. Barrick C. Kim K.A. Lindner J. Blondeau B. Fujimoto Y. Shiota M. Kesterson R.A. Kahn B.B. Magnuson M.A. Deletion of ppargamma in adipose tissues of mice protects against high fat diet-induced obesity and insulin resistance Proc. Natl. Acad. Sci. USA 2005 102 6207 6212 10.1073/pnas.0306743102 15833818 90. Sugii S. Olson P. Sears D.D. Saberi M. Atkins A.R. Barish G.D. Hong S.H. Castro G.L. Yin Y.Q. Nelson M.C. Ppargamma activation in adipocytes is sufficient for systemic insulin sensitization Proc. Natl. Acad. Sci. USA 2009 106 22504 22509 10.1073/pnas.0912487106 20018750 91. Jonker J.W. Suh J.M. Atkins A.R. Ahmadian M. Li P. Whyte J. He M. Juguilon H. Yin Y.Q. Phillips C.T. A ppargamma-fgf1 axis is required for adaptive adipose remodelling and metabolic homeostasis Nature 2012 485 391 394 10.1038/nature10998 22522926 92. Jiang Y. Berry D.C. Tang W. Graff J.M. Independent stem cell lineages regulate adipose organogenesis and adipose homeostasis Cell Rep. 2014 9 1007 1022 10.1016/j.celrep.2014.09.049 25437556 93. Berry D.C. Jiang Y. Graff J.M. Emerging roles of adipose progenitor cells in tissue development, homeostasis, expansion and thermogenesis Trend. Endocrinol. Metab. 2016 10.1016/j.tem.2016.05.001 27262681 94. Hevener A.L. He W. Barak Y. Le J. Bandyopadhyay G. Olson P. Wilkes J. Evans R.M. Olefsky J. Muscle-specific pparg deletion causes insulin resistance Nat. Med. 2003 9 1491 1497 10.1038/nm956 14625542 95. Norris A.W. Chen L. Fisher S.J. Szanto I. Ristow M. Jozsi A.C. Hirshman M.F. Rosen E.D. Goodyear L.J. Gonzalez F.J. Muscle-specific ppargamma-deficient mice develop increased adiposity and insulin resistance but respond to thiazolidinediones J. Clin. Investig. 2003 112 608 618 10.1172/JCI17305 12925701 96. Cha B.S. Ciaraldi T.P. Park K.S. Carter L. Mudaliar S.R. Henry R.R. Impaired fatty acid metabolism in type 2 diabetic skeletal muscle cells is reversed by ppargamma agonists Am. J. Physiol. Endocrinol. Metabol. 2005 289 E151 E159 10.1152/ajpendo.00141.2004 15727952 97. Vidal-Puig A. Jimenez-Linan M. Lowell B.B. Hamann A. Hu E. Spiegelman B. Flier J.S. Moller D.E. Regulation of ppar gamma gene expression by nutrition and obesity in rodents J. Clin. Investig. 1996 97 2553 2561 10.1172/JCI118703 8647948 98. Gavrilova O. Haluzik M. Matsusue K. Cutson J.J. Johnson L. Dietz K.R. Nicol C.J. Vinson C. Gonzalez F.J. Reitman M.L. Liver peroxisome proliferator-activated receptor gamma contributes to hepatic steatosis, triglyceride clearance, and regulation of body fat mass J. Biol. Chem. 2003 278 34268 34276 10.1074/jbc.M300043200 12805374 99. Matsusue K. Haluzik M. Lambert G. Yim S.H. Gavrilova O. Ward J.M. Brewer B. Jr. Reitman M.L. Gonzalez F.J. Liver-specific disruption of ppargamma in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes J. Clin. Investig. 2003 111 737 747 10.1172/JCI200317223 12618528 100. Dubois M. Pattou F. Kerr-Conte J. Gmyr V. Vandewalle B. Desreumaux P. Auwerx J. Schoonjans K. Lefebvre J. Expression of peroxisome proliferator-activated receptor gamma (ppargamma) in normal human pancreatic islet cells Diabetologia 2000 43 1165 1169 10.1007/s001250051508 11043863 101. Welters H.J. McBain S.C. Tadayyon M. Scarpello J.H. Smith S.A. Morgan N.G. Expression and functional activity of ppargamma in pancreatic beta cells Brit. J. Pharmacol. 2004 142 1162 1170 10.1038/sj.bjp.0705844 15237101 102. Kim H.I. Kim J.W. Kim S.H. Cha J.Y. Kim K.S. Ahn Y.H. Identification and functional characterization of the peroxisomal proliferator response element in rat glut2 promoter Diabetes 2000 49 1517 1524 10.2337/diabetes.49.9.1517 10969836 103. Rosen E.D. Kulkarni R.N. Sarraf P. Ozcan U. Okada T. Hsu C.H. Eisenman D. Magnuson M.A. Gonzalez F.J. Kahn C.R. Targeted elimination of peroxisome proliferator-activated receptor gamma in beta cells leads to abnormalities in islet mass without compromising glucose homeostasis Mol. Cell. Biol. 2003 23 7222 7229 10.1128/MCB.23.20.7222-7229.2003 14517292 104. Gupta D. Jetton T.L. Mortensen R.M. Duan S.Z. Peshavaria M. Leahy J.L. In vivo and in vitro studies of a functional peroxisome proliferator-activated receptor gamma response element in the mouse pdx-1 promoter J. Biol. Chem. 2008 283 32462 32470 10.1074/jbc.M801813200 18718916 105. Vivas Y. Martinez-Garcia C. Izquierdo A. Garcia-Garcia F. Callejas S. Velasco I. Campbell M. Ros M. Dopazo A. Dopazo J. Early peroxisome proliferator-activated receptor gamma regulated genes involved in expansion of pancreatic beta cell mass BMC Med. Genom. 2011 4 1236 10.1186/1755-8794-4-86 22208362 106. Welters H.J. El Ouaamari A. Kawamori D. Meyer J. Hu J. Smith D.M. Kulkarni R.N. Rosiglitazone promotes ppargamma-dependent and -independent alterations in gene expression in mouse islets Endocrinology 2012 153 4593 4599 10.1210/en.2012-1243 22807489 107. Szeles L. Torocsik D. Nagy L. Ppargamma in immunity and inflammation: Cell types and diseases Biochim. Biophys. Acta 2007 1771 1014 1030 10.1016/j.bbalip.2007.02.005 17418635 108. Szatmari I. Rajnavolgyi E. Nagy L. Ppargamma, a lipid-activated transcription factor as a regulator of dendritic cell function Ann. N. Y. Acad. Sci. 2006 1088 207 218 10.1196/annals.1366.013 17192567 109. Nagy L. Szanto A. Szatmari I. Szeles L. Nuclear hormone receptors enable macrophages and dendritic cells to sense their lipid environment and shape their immune response Physiol. Rev. 2012 92 739 789 10.1152/physrev.00004.2011 22535896 110. Weisberg S.P. McCann D. Desai M. Rosenbaum M. Leibel R.L. Ferrante A.W. Jr. Obesity is associated with macrophage accumulation in adipose tissue J. Clin. Investig. 2003 112 1796 1808 10.1172/JCI200319246 14679176 111. Akiyama T.E. Sakai S. Lambert G. Nicol C.J. Matsusue K. Pimprale S. Lee Y.H. Ricote M. Glass C.K. Brewer H.B. Jr. Conditional disruption of the peroxisome proliferator-activated receptor gamma gene in mice results in lowered expression of abca1, abcg1, and apoe in macrophages and reduced cholesterol efflux Mol. Cell. Biol. 2002 22 2607 2619 10.1128/MCB.22.8.2607-2619.2002 11909955 112. Odegaard J.I. Ricardo-Gonzalez R.R. Goforth M.H. Morel C.R. Subramanian V. Mukundan L. Red Eagle A. Vats D. Brombacher F. Ferrante A.W. Macrophage-specific ppargamma controls alternative activation and improves insulin resistance Nature 2007 447 1116 1120 10.1038/nature05894 17515919 113. Hevener A.L. Olefsky J.M. Reichart D. Nguyen M.T. Bandyopadyhay G. Leung H.Y. Watt M.J. Benner C. Febbraio M.A. Nguyen A.K. Macrophage ppar gamma is required for normal skeletal muscle and hepatic insulin sensitivity and full antidiabetic effects of thiazolidinediones J. Clin. Investig. 2007 117 1658 1669 10.1172/JCI31561 17525798 114. Szanto A. Balint B.L. Nagy Z.S. Barta E. Dezso B. Pap A. Szeles L. Poliska S. Oros M. Evans R.M. Stat6 transcription factor is a facilitator of the nuclear receptor ppargamma-regulated gene expression in macrophages and dendritic cells Immunity 2010 33 699 712 10.1016/j.immuni.2010.11.009 21093321 115. Shan M. You R. Yuan X. Frazier M.V. Porter P. Seryshev A. Hong J.S. Song L.Z. Zhang Y. Hilsenbeck S. Agonistic induction of ppargamma reverses cigarette smoke-induced emphysema J. Clin. Investig. 2014 124 1371 1381 10.1172/JCI70587 24569375 116. Duan S.Z. Usher M.G. Foley E.L.t. Milstone D.S. Brosius F.C. Mortensen R.M. Sex dimorphic actions of rosiglitazone in generalised peroxisome proliferator-activated receptor-gamma (ppar-gamma)-deficient mice Diabetologia 2010 53 1493 1505 10.1007/s00125-010-1748-2 20401461 117. Shiomi Y. Yamauchi T. Iwabu M. Okada-Iwabu M. Nakayama R. Orikawa Y. Yoshioka Y. Tanaka K. Ueki K. Kadowaki T. A novel peroxisome proliferator-activated receptor (ppar)alpha agonist and ppargamma antagonist, z-551, ameliorates high-fat diet-induced obesity and metabolic disorders in mice J. Biol. Chem. 2015 290 14567 14581 10.1074/jbc.M114.622191 25907553 118. Silva J.C. Cesar F.A. de Oliveira E.M. Turato W.M. Tripodi G.L. Castilho G. Machado-Lima A. de Las Heras B. Bosca L. Rabello M.M. New ppargamma partial agonist improves obesity-induced metabolic alterations and atherosclerosis in LDLr(-/-) mice Pharmacol. Res. 2016 104 49 60 10.1016/j.phrs.2015.12.010 26706782 119. Ryan K.K. Li B. Grayson B.E. Matter E.K. Woods S.C. Seeley R.J. A role for central nervous system ppar-gamma in the regulation of energy balance Nat. Med. 2011 17 623 626 10.1038/nm.2349 21532595 120. Mikkelsen T.S. Xu Z. Zhang X. Wang L. Gimble J.M. Lander E.S. Rosen E.D. Comparative epigenomic analysis of murine and human adipogenesis Cell 2010 143 156 169 10.1016/j.cell.2010.09.006 20887899 121. Schmidt S.F. Jorgensen M. Chen Y. Nielsen R. Sandelin A. Mandrup S. Cross species comparison of c/ebpalpha and ppargamma profiles in mouse and human adipocytes reveals interdependent retention of binding sites BMC Genom. 2011 12 1236 10.1186/1471-2164-12-152 21410980 122. Ruchat S.M. Weisnagel S.J. Vohl M.C. Rankinen T. Bouchard C. Perusse L. Evidence for interaction between pparg pro12ala and ppargc1a gly482ser polymorphisms in determining type 2 diabetes intermediate phenotypes in overweight subjects Exp. Clin. Endocrinol. Diabetes 2009 117 455 459 10.1055/s-0029-1216352 19536736 123. Choi J.H. Banks A.S. Estall J.L. Kajimura S. Bostrom P. Laznik D. Ruas J.L. Chalmers M.J. Kamenecka T.M. Bluher M. Anti-diabetic drugs inhibit obesity-linked phosphorylation of ppargamma by cdk5 Nature 2010 466 451 456 10.1038/nature09291 20651683 124. Banks A.S. McAllister F.E. Camporez J.P. Zushin P.J. Jurczak M.J. Laznik-Bogoslavski D. Shulman G.I. Gygi S.P. Spiegelman B.M. An erk/cdk5 axis controls the diabetogenic actions of ppargamma Nature 2015 517 391 395 10.1038/nature13887 25409143 125. Asterholm I.W. Halberg N. Scherer P.E. Mouse models of lipodystrophy key reagents for the understanding of the metabolic syndrome Drug Discov. Today Dis. Model. 2007 4 17 24 10.1016/j.ddmod.2007.10.003 18193096

PMC005xxxxxx/PMC5000635.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081237ijms-17-01237ArticleVEGF and FGF2 Improve Revascularization, Survival, and Oocyte Quality of Cryopreserved, Subcutaneously-Transplanted Mouse Ovarian Tissues Li Sheng-Hsiang 12†Hwu Yuh-Ming 1234†Lu Chung-Hao 3Chang Hsiao-Ho 1Hsieh Cheng-En 3Lee Robert Kuo-Kuang 135*Muraca Maurizio Academic Editor1 Department of Medical Research, Mackay Memorial Hospital, Tamsui District, New Taipei City 251, Taiwan; [email protected] (S.-H.L.); [email protected] (Y.-M.H.); [email protected] (H.-H.C.)2 Mackay Junior College of Medicine, Nursing, and Management, Beitou District, Taipei City 112, Taiwan3 Department of Obstetrics and Gynecology, Mackay Memorial Hospital, Taipei City 104, Taiwan; [email protected] (C.-H.L.); [email protected] (C.-E.H.)4 Mackay Medical College, Sanzhi District, New Taipei City 252, Taiwan5 Department of Obstetrics and Gynecology, Taipei Medical University, Taipei City 110, Taiwan* Correspondence: [email protected]; Tel.: +886-2-2543-3535† These authors contributed equally to this work. 30 7 2016 8 2016 17 8 123708 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).This study was conducted to investigate the effect of the vascular endothelial growth factor (VEGF) and fibroblast growth factor 2 (FGF2) on revascularization, survival, and oocyte quality of cryopreserved, subcutaneously-transplanted mouse ovarian tissue. Autologous subcutaneous transplantation of vitrified-thawed mouse ovarian tissues treated with (experimental group) or without (control group) VEGF and FGF2 was performed. After transplantation to the inguinal region for two or three weeks, graft survival, angiogenesis, follicle development, and oocyte quality were examined after gonadotropin administration. VEGF coupled with FGF2 (VEGF/FGF2) promoted revascularization and significantly increased the survival rate of subcutaneously-transplanted cryopreserved ovarian tissues compared with untreated controls. The two growth factors did not show long-term effects on the ovarian grafts. In contrast to the untreated ovarian grafts, active folliculogenesis was revealed as the number of follicles at various stages and of mature oocytes in antral follicles after gonadotropin administration were remarkably higher in the VEGF/FGF2-treated groups. Although the fertilization rate was similar between the VEGF/FGF2 and control groups, the oocyte quality was much better in the VEGF/FGF2-treated grafts as demonstrated by the higher ratio of blastocyst development. Introducing angiogenic factors, such as VEGF and FGF2, may be a promising strategy to improve revascularization, survival, and oocyte quality of cryopreserved, subcutaneously-transplanted mouse ovarian tissue. angiogenic factorovarian cryopreservationovarian transplantationoocyte qualityfertility reservation ==== Body 1. Introduction Cryopreservation of ovarian tissue is rapidly evolving and is a promising clinical technique for preserving gonadal reproductive function. It avoids the necessity for ovarian stimulation and stores a substantial amount of ovarian tissue [1,2]. Cryopreservation of ovarian tissue is feasible to restore fertility for young women requiring urgent treatment for cancer and is even the only option to preserve fertility in children with cancer [3]. While cryopreservation and transplantation of the entire ovary takes advantage of immediate restoration of the blood supply by vascular anastomosis [4,5,6], storage of ovarian tissue slices is more practical [2,7]. Previous studies have demonstrated human live birth after orthotopic transplantation of cryopreserved ovarian tissues [8,9,10]. In mammals, several cases of successful reproduction from heterotopic transplantation into the kidney capsule of cryopreserved ovarian tissues also were reported [11,12,13]. Despite the successes in orthotopic and renal capsular transplantation, patients must endure the risk of operation, and ovum retrieval in these locations is very difficult. In contrast to the aforementioned transplantation, subcutaneous transplantation, such as in the forearm or abdominal wall, is of practical value because it is easily accessible for ovum retrieval [2,14,15] and follicle development is relatively easy to monitor [14,15]. Lee et al. reported the birth of a monkey following in vitro fertilization (IVF) of an oocyte obtained from subcutaneously-transplanted fresh ovarian tissue [14]. Oktay et al. reported that a four-cell human embryo was produced from subcutaneously transplanted cryopreserved ovarian tissue, but no conception occurred [16]; however, spontaneous conceptions and live birth following the subcutaneous transplantation of frozen banked ovarian tissue from a Hodgkin lymphoma survivor were also reported [17]. In mice, subcutaneous grafted ovaries demonstrated poorer graft survival and lower numbers of retrieved oocytes compared with those grafted under the renal capsule or in the bursal cavity [18]. Nevertheless, blastocyst development and live pups from subcutaneous transplantation of cryopreserved ovaries have been reported [19]. Despite some cases of successful reproduction, the technique of subcutaneous transplantation remains premature and needs more investigations. Follicle loss in the cryopreserved grafts was very common. The key factor is ischemic injury during the early revascularization period of ovarian transplantation [3]. The kidney capsule with ample capillaries provides an ideal grafting site for rapid revascularization. Several previous reports demonstrated that the ovary and renal capsules are rich in angiogenic factors, such as vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF), which may enhance angiogenesis and promote graft survival [20,21,22,23]. Here, we encapsulated VEGF and FGF2 with vitrified-thawed mouse ovarian tissues and attempted to induce the production of blood vessels in subcutaneously-transplanted cryopreserved mouse ovarian tissue. We evaluated the effect of the two angiogenic factors on revascularization and survival of the ovarian grafts, and subsequent oocyte maturation, fertilization, and embryo development. 2. Results We removed ovarian tissues after subcutaneous transplantation of the VEGF/FGF2-treated ovarian tissues for one, two, or three weeks. Following the experimental scheme in Figure A1, we assessed angiogenesis, survival rate, and oocyte quality of the grafted ovarian tissues. 2.1. VEGF Coupled with FGF2 Improved Angiogenesis and Survival of Cryopreserved, Subcutaneously-Transplanted Mouse Ovarian Tissues After cryopreserved ovarian tissues were transplanted subcutaneously for two or three weeks, the grafted tissues were removed and their appearance evaluated. The VEGF/FGF2-treated ovarian tissue demonstrated a substantially larger size compared with the untreated controls. More blood vessels, even larger vessels, were readily visualized in VEGF/FGF2-treated ovarian tissues, while only microvessels were observed in the control grafts (Figure 1A). Tissue fibrosis occurred in approximately 34% and 50% of control implants as well as 6% and 18% of VEGF/FGF2-treated grafts two and three weeks after transplantation (Table S1). Nevertheless, the tissue survival rate was significantly enhanced under the treatment of VEGF and FGF2. In addition, the grafted tissues two weeks after transplantation seemed to have better survival rates than those at three weeks and no statistical difference was observed (Figure 1B). Histological analysis revealed that the density of small (Figure 1C) and large (Figure 1D) blood vessels was markedly detected in the VEGF/FGF2-treated ovarian tissues, while nearly only small vessels were found in the untreated controls (Figure 1E,F). In addition, more follicles were obviously detected in the VEGF/FGF2-treated ovarian tissues (Figure 1G) compared with the untreated controls (Figure 1H). 2.2. Levels of Angiogenic Cytokines in the Ovarian Grafts We examined the relative levels of angiogenic cytokines, including TNF-α, IGF-1, VEGF, IL-6, FGF2, IFNγ, EGF, and leptin, in the ovarian grafts 1, 2, or 3 weeks after transplantation. All proteins demonstrated higher levels in VEGF/FGF2-treated grafts compared with untreated controls one week after transplantation (Figure 2), though there was no statistical differences between the groups (Figure 2A); however, protein levels were similar two or three weeks after transplantation (Figure 2B,C). 2.3. VEGF and FGF2 Improved Oocyte Quality of the Transplanted Ovarian Tissues Untreated control had a significantly increased number of preantral follicles and an increased trend of antral follicles, despite there was no significant difference, compared with VEGF/FGF2-treated ovarian grafts three weeks after transplantation; however, when different doses of gonadotropins were administered to assess folliculogenesis and oocyte quality of the ovarian grafts, VEGF/FGF2-treated ovarian grafts demonstrated relatively active folliculogenesis compared with the untreated grafts in the administration of various doses of gonadotropins. As the doses of gonadotropins increased, the numbers of primordial, primary, preantral, and antral follicles significantly increased or had an increased trend in VEGF/FGF2-treated groups (Table 1). The status of oocytes in the antral follicles of the ovarian grafts is shown in Table 2 and Table S2. The total number of oocytes retrieved from VEGF/FGF2-treated ovarian grafts was higher than that of controls, 191 vs. 162, respectively. The number of matured oocytes was significantly higher in the VEGF/FGF2-treated grafts than that in the controls (Figure 3A). As metaphase II (MII) oocytes were subjected to IVF, the fertilization rate was comparable between the two groups (Figure 3B). However, the zygotes developed to the blastocyst stage were significantly enhanced in the VEGF/FGF2-treated grafts (Figure 3C). 3. Discussion We examined the effect of exogenous angiogenic factors, i.e., VEGF and FGF2, on revascularization, survival, and oocyte quality of cryopreserved, subcutaneously-transplanted murine ovarian tissue. The results revealed that VEGF and FGF2 induced angiogenesis and enhanced the survival of the ovarian grafts and, thus, the oocyte quality was also improved. Cryopreserved ovarian tissues are subjected to freezing and thawing injury, while grafting to a subcutaneous site causes ischemic injury and also potentially exposes the tissue to altered temperature than normal ovaries [16,24]. Presumably, these injuries impair the quality of the oocyte and subsequent embryogenesis. Due to the lack of vessels under the skin, ischemia occurring during the critical time for revascularization is the primary injury to subcutaneously transplanted ovarian tissue, leading to tissue fibrosis or severe follicle loss [3]. Heterotopic transplantation of the ovarian tissue to the kidney capsules of nude mice led to better follicle development [25,26]. Renal capsules express rich angiogenic factors, such as VEGF and FGF2 and, thus, may enhance graft survival and revascularization [21,22]. To reduce the ischemic injury during subcutaneous transplantation, we encapsulated ovarian tissues with VEGF, FGF2, and basement membrane extract (BME) to create an environment for angiogenesis. Being a mixture of extracellular matrix, BME can form an interface between epithelial, muscle, and stromal cells. It may provide a suitable microenvironment for angiogenesis. VEGF and FGF2, being potent angiogenic factors, are demonstrated to promote angiogenesis and graft survival [27,28,29]. Therefore, coupling of BME, VEGF, and FGF2 may promote revascularization of the ovarian implants during the critical time of transplantation. This could be the reason for the higher survival rate in this study. In general, oxygen, nutrient, and gonadotropin supplements are essential for folliculogenesis of subcutaneously transplanted ovarian tissue. In the present study, we demonstrated that VEGF and FGF2 promoted revascularization of the ovarian graft; therefore, the survival rate of the graft was increased, and the size of the graft generally was larger than the untreated control graft, suggesting that ischemic injury can be improved to some extent and, thus, oxygen and nutrients could reach the graft. However, the ovarian response to gonadotropins seemed to be unexpected. Higher doses of gonadotropins were used because the general dose (e.g., 5 IU pregnant mare’s serum gonadotropin (PMSG) and 10 IU human chorionic gonadotropin (hCG)) routinely used in the normal mice could not well stimulate folliculogenesis of the ovarian grafts. The situation was also observed previously [30]. This phenomenon may reflect a poor ovarian reserve that affects the response to gonadotropin stimulation. While the two growth factors induced angiogenesis in ovarian grafts, and some larger vessels and more microvessels were detected, the coverage and the deep vessel may remain insufficient and impact the ovarian reserve. Another clue for poor ovarian reserve is the ratio of matured oocytes. Only an average of approximately 2.7 MII oocytes per grafted ovarian tissue can be retrieved from the VEGF/FGF2-treated graft. Consequently, poor ovarian response may be due to limited vascularization of the grafts in the subcutaneous space and poor ovarian reserve after freezing, thawing, and transplantation. Despite untreated control with a relatively higher numbers of preantral and antral follicles without treatment with gonadotropins, the quality of oocytes seemingly damaged since a relatively low number of oocytes can be developed into the blastocyst stage [31]. Even after gonadotropin treatment, the quality of oocytes collected from the untreated group still is poorer as demonstrated by the very lower rate of blastocyst formation. In fact, the oocytes retrieved from subcutaneous transplanted ovaries showed a relatively lower fertilization rate compared with those normally-ovulated oocytes (Table S4). The increases in the numbers of primordial and primary follicles were found by increasing doses of gonadotropin; however, these increases are not associated with gonadotropin treatment since follicles at these stages do not respond to FSH stimuli in nature. As taking the 150 IU group, with a larger sample size, for consideration, there are no differences in the number of primordial follicles, while the number of primary follicles is just reaching statistical significance (p = 0.0466) between the two groups. In addition, the numbers of primordial and primary follicles have no differences between the two groups without gonadotropins treatment (Table 1). Therefore, whether the numbers of primordial and primary follicles are increased in VEGF/FGF2-treated groups or increasing the sample size may change the differences needs further evaluation. Counting sections of the grafted ovary found that the numbers of antral follicles in VEGF/FGF2-treated ovarian grafts were approximately three-fold higher than those in untreated controls. However, when oocytes were directly retrieved from antral follicles of the grafted ovaries, the differences of the retrieved oocytes between control and VEGF/FGF2 treatment may be reduced. The ovarian grafts were buried under the subcutaneous site of the inguinal region where the surface of the grafts were covered with a layer of tough connective tissues, it must be removed first during the oocyte retrieval, and only the larger antral follicles were collected. Nevertheless, many oocytes need to be picked up from a small grafted ovary; the scene is muddy and in a mess. Thus, some oocytes are possibly missing. In addition, we also need to consider the standard deviation found in different batches of experiment. This may explain why the differences of the retrieved oocytes between control and VEGF/FGF2 treatment were too small. Despite the poor response of the ovarian grafts, more available MII oocytes were retrieved from the VEGF/FGF2-treated grafts compared with the control grafts. Ischemia may largely impair the ovarian reserve of the control grafts, leading to the only approximately 20% rate of MII oocyte recovery [31] as shown in our control group; however, oocyte quality is much better in VEGF/FGF2-treated ovarian than in the control grafts. More blastocyst embryos could be obtained under the treatment of VEGF and FGF2. Blastocyst has been reported to have better pregnancy and live birth rates in assisted reproductive technology [32,33], suggesting that taking the advantage of potent angiogenesis factors is beneficial for fertility preservation. Protein levels of VEGF, FGF2, and some cytokines were relatively higher, but there was no statistical difference between the VEGF/FGF2-treated and untreated ovarian grafts one week after transplantation. This cannot be explained by inflammation after the graft operation. While inflammation can induce the expression of cytokines and growth factors, the operative procedure was the same between the VEGF/FGF2-treated and untreated control grafts. Nevertheless, the promotion of revascularization of the ovarian graft by VEGF/FGF2 must include biological processes such as wound healing and tissue remodeling. Thus, VEGF and FGF2 may induce the elevation of other angiogenic-related cytokines. The elevation disappeared two and three weeks after transplantation, indicating that the grafted ovary may recover to a stable status at least two weeks after transplantation as demonstrated by the similar levels of growth factors and cytokines between the two groups. The underlying causes regarding the increased levels of angiogenic cytokines were declined by longer cultivation time in bodies after grafting require further investigation. The effect of growth factors on the survival and fertility potential of the ovarian graft is controversial. In monkeys, VEGF does not improve the viability of subcutaneously transplanted ovarian grafts [34]. Gao et al. used high concentrations of FGF2 (75–150 µg/mL) to treat the transplanted fresh ovarian tissue and demonstrated that FGF2 significantly improved primordial follicle survival and angiogenesis one week after transplantation, and apoptosis of follicles and stromal cells was significantly decreased [35]. However, they showed that lower dose of FGF2 (25 and 50 µg/mL) had no effect. In our study, combined VEGF (3 µg/mL) and FGF2 (9 µg/mL) showed beneficial effects on cryopreserved, subcutaneously transplanted ovarian tissues with a relatively low concentration of FGF2, even at eight- to 17-fold lower concentrations than used in their study. Recently, the authors mixed VEGF and FGF2 using a six- and three-fold higher concentration, respectively, compared with ours to treat subcutaneously transplanted cryopreserved ovarian tissues and demonstrated that the two growth factors can promote angiogenesis and enhance graft survival [36]. Kang et al. reported that culture of vitrified-thawed human ovarian tissues supplemented with a relatively low concentration of FGF2 (150 ng/mL) alone for two days and then subcutaneous xenografting to severe combined immune deficiency mice for one week significantly increased microvessel density and the number of follicles, while apoptosis significantly decreased. Their results showed that these effects were ascribed to FGF2 alone but not VEGF (25–100 ng/mL) [37]. A recent study reported that preinjection of VEGF into the subcutaneous site of Nu mice and then grafting cryopreserved ovaries increased the number of follicles at each stage [38]. These studies and ours all showed similar effects on cryopreserved, subcutaneously autotransplanted ovarian grafts with increases in blood vessels and survival rate. Unfortunately, the competence of oocytes retrieved from the ovarian grafts has not been assessed in the studies mentioned above. The novelty of this study is that we demonstrated that the quality of oocytes retrieved from subcutaneously transplanted mouse ovarian tissue has fertility potential. Nevertheless, whether these effects on the ovarian grafts are attributed to FGF2 alone or in combination with VEGF and what is the optimal concentration still needs to be determined. In conclusion, VEGF and FGF2 promoted angiogenesis and significantly increased the survival rate of subcutaneously transplanted cryopreserved ovarian tissues. The number of various stages of follicles was prominently increased in the VEGF/FGF2-treated ovarian grafts after gonadotropin administration. In addition, oocyte quality was much better in the VEGF/FGF2-treated ovarian grafts. While cryopreservation of ovarian tissue followed by autotransplantation still is under investigation, combined VEGF and FGF2 might be the promising remedy to preserve fertility for children and young women with cancer. 4. Materials and Methods 4.1. Animals Specific pathogen-free outbred CD-1 mice (BioLASCO Taiwan, Taipei, Taiwan) were bred in the Animal Center at the Department of Medical Research, Mackay Memorial Hospital following institutional guidelines for the care and use of experimental animals. The animal use protocol was reviewed and approved by the hospital’s Institutional Animal Care and Use Committee (approval number, MMH-A-S-97042). All efforts were made to minimize suffering. Animals were housed under controlled lighting (14 h light, 10 h dark) at 21–22 °C and were provided with water and chow ad libitum. Sexually mature (6–8-week old) female mice were used for the main study, while male mice (14-week old) were used for IVF. 4.2. Vitrification of Ovarian Tissues The mice were injected intraperitoneally with 15 µL/g body weight of avertin solution, which was made by dissolving 1 g of 2,2,2-tribromoethanol (Sigma-Aldrich Chemical Co., St. Louis, MO, USA) in 1 mL of tertiary-amyl alcohol (J.T. Baker, Phillipsburg, NJ, USA) for mouse anesthesia. Both ovaries were removed, freed of fat, and cut into three equal pieces in M2 medium (Sigma-Aldrich) under a dissection microscope. Next, the six pieces of ovarian tissues were randomly mixed and immersed in MEMα medium (Life Technologies, Grand Island, NY, USA) supplemented with 12% fetal bovine serum (Sigma-Aldrich), 100 µg/mL streptomycin, and 100 µg/mL penicillin, pH 7.4, for 5 min. Afterward, the tissues were incubated in the same medium containing 2.0 mol/L dimethyl sulfoxide (DMSO) and 0.1 mol/L sucrose for 5 min, and then transferred into another fresh medium containing 2.0 mol/L DMSO, 0.2 mol/L sucrose, and 2.0 mol/L propanediol for 5 min. The tissues were vitrified by directly dipping in liquid nitrogen, transferred into the precooled cryovial, and stored in the liquid nitrogen tank. 4.3. Treatment with VEGF and FGF2, and Autologous Subcutaneous Transplantation of the Vitrified-Thawed Ovarian Tissue A week later, cryovials were removed from liquid nitrogen, held in the air at room temperature for 20 s, and then the cryopreserved tissues were thawed quickly by immersing in 37 °C prewarmed phosphate-buffered saline (PBS). Once thawed, the tissues were consecutively diluted in MEMα medium containing 0.5, 0.25, and 0.125 mol/L sucrose, respectively, for 5 min each. After briefly rinsing in MEMα medium three times, the tissues were cultured in M2 medium in a humidified 5% CO2 atmosphere at 37 °C for 15 min. Simultaneously, VEGF and FGF2 were mixed with BME and heparin solution to prepare a final concentration of 3 and 9 µg/mL, respectively, on ice to prevent gel formation according to the instructions in the Directed In Vivo Angiogenesis Assay starter kit (Trevigen, Gaithersburg, MD, USA). Subsequently, the ovarian pieces wiped with a tissue paper were soaked in 10 µL premix reagent, with or without VEGF/FGF2, at 37 °C on a rack provided by the kit to form a gel-like substance, and then transplanted in the inguinal region of the mouse. After the mouse was anesthetized with avertin, an approximately 0.3 cm incision was made in the inguinal region, followed by blunt dissection of a 1-cm subcutaneous pocket using a pair of fine curved watchmaker’s forceps. Three ovarian pieces treated with VEGF, FGF2, and BME were inserted into one side of the subcutaneous space, and the other three pieces mixed with BME, but without VEGF and FGF2, were inserted into the opposite side of the same mouse. Finally, the skin incision was closed with a single nylon suture. 4.4. Graft Retrieval and Histological Analysis of Follicles Ovarian grafts were retrieved two and three weeks after transplantation for analysis of morphology and graft survival. The survival rate was defined as the percentage of ovarian grafts still present at the retrieval time relative to the total ovarian grafts. The untreated control group or the VEGF/FGF2-treated group was counted independently. Only those grafts where all survived from the control and VEGF/FGF2-treated groups in the same mouse were used for the following assay. The retrieved grafts, which had been grafted for two weeks, were fixed in neutral formalin, embedded in paraffin, and cut into 5-μm sections for morphology observation or immunohistochemical staining. Mice grafted with ovarian tissue three weeks after transplantation were treated with gonadotropins. Three dosages of pregnant mare’s serum gonadotropin (PMSG, Sigma-Aldrich), i.e., 50, 150, and 250 international units (IU), were injected intraperitoneally, and an equivalent dose of human chorionic gonadotropin (hCG, China Chemical and Pharmaceutical, Hsinchu, Taiwan) was administered 48 h later. The mice were killed by cervical dislocation 16 h after hCG administration. The ovarian grafts were retrieved, fixed, and embedded in paraffin as aforementioned. Ten serial sections were stained with hematoxylin and eosin (H&E). Follicles were counted via light microscopy at three randomly selected fields in 10 serial sections. Follicles were classified as described previously [39]. 4.5. Immunohistochemical Staining of Vessels We stained for von Willebrand factor (vWF), one of the markers for the identification of endothelial cells of blood vessels, to recognize the blood vessel. After the ovarian sections on slides were deparaffinized and hydrated, tissues were demasked with antigen retrieval AR-10 solution (BioGenex, San Ramon, CA, USA); the slides were treated with 3% (v/v) H2O2 in PBS for 15 min, blocked with 10% (v/v) normal goat serum in PBS (blocking solution) for 1 h at room temperature, and then incubated with rabbit anti-vWF antiserum (ab6994; Abcam, Cambridge, UK) or the control antiserum diluted 1:1000 in blocking solution at 4 °C for 16 h. After washing, the slides were treated with biotin-conjugated goat anti-rabbit IgG (3 μg/mL; Zymed Laboratories, South San Francisco, CA, USA) for 1 h at room temperature. After washing, slides were incubated with 1 μg/mL of horseradish peroxidase (HRP)-conjugated streptavidin (Zymed Laboratories) for 40 min at room temperature. The slides were immunostained by 3-amino-9-ethylcarbazole staining (Zymed Laboratories), counterstained with hematoxylin (Vector Laboratories, Burlingame, CA, USA), and photographed using an Olympus BX 40 microscope (Olympus, Tokyo, Japan) equipped with an Olympus DP-70 digital camera. 4.6. Enzyme-Linked Immunosorbent Assay (ELISA) Mouse angiogenesis ELISA strips for profiling eight cytokines (EA-1021; Signosis, Santa Clara, CA, USA), including VEGF and FGF2, were applied to determine the relative levels of angiogenic factors in the grafted ovarian tissues. Ovarian grafts were removed one, two, or three weeks after transplantation and were homogenized in PBS containing the protease inhibitor cocktail to prepare the total protein extract. The protein concentration was determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL, USA). The protein extract (30 µg) was subjected to analysis according to the manufacturer’s instructions. 4.7. Gonadotropin Administration and Oocyte Retrieval We superovulated the transplanted mice three weeks after grafting. Approximately 150 IU of PMSG was injected intraperitoneally, and an equivalent dose of hCG was given 48 h later. The transplanted ovarian tissues were removed carefully 16 h after hCG administration from the inguinal areas, and oocyte-granulosa cell complexes from the antral follicles were isolated using 30-gauge needles (Becton Dickinson, Bedford, MA, USA). After washing with M16 medium, the maturation status of the oocytes was scored as a germinal vesicle (GV) being present as an enlarged nucleus, a metaphase I (MI) referring to the breakdown of the GV, and a metaphase II (MII) when the first polar body had been extruded. 4.8. In Vitro Fertilization (IVF) and Embryo Culture Mature MII oocytes retrieved from the transplanted ovarian tissues were subjected to IVF following the method reported previously [40]. The zygotes were cultured in M16 medium in a humidified 5% CO2 atmosphere at 37 °C for four days. 4.9. Statistical Analysis Data are presented as the mean ± SD. The differences were analyzed by a paired Student’s t-test using GraphPad Prism 5 (San Diego, CA, USA). p < 0.05 was considered significant. Acknowledgments This work was supported by the grants from the National Science Council (NSC 98-2314-B-195-009-MY3, Robert Kuo-Kuang Lee) and the Mackay Memorial Hospital (MMH 9921, 10013, and 10105, Robert Kuo-Kuang Lee), Taipei, Taiwan. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1237/s1. Click here for additional data file. Author Contributions Sheng-Hsiang Li, Yuh-Ming Hwu, and Robert Kuo-Kuang Lee conceived and designed the experiments; Chung-Hao Lu and Hsiao-Ho Chang performed the experiments; Sheng-Hsiang Li, Yuh-Ming Hwu, and Cheng-En Hsieh analyzed the data; Sheng-Hsiang Li contributed to manuscript preparation and editing; Yuh-Ming Hwu and Robert Kuo-Kuang Lee provided professional consultation. Conflicts of Interest The authors declare no conflict of interest. Appendix Figure A1 Experimental scheme of this study. The cryopreserved mouse ovarian tissues were autologously transplanted to the subcutaneous site of the inguinal region. The VEGF/FGF2-treated ovarian tissues were implanted into one side of the subcutaneous space, and the ones without treatment of VEGF/FGF2 were inserted into the opposite side of the same mouse. For the assay purpose, ovarian grafts were removed one, two, or three weeks after transplantation as shown on the diagram. Figure 1 Morphology, survival rate, and blood vessels of the cryopreserved ovarian tissue after transplantation. (A) The morphologies of the grafted ovarian tissue. Left: VEGF/FGF2-treated tissue; right: untreated control tissue. An arrow shows the larger blood vessel; (B) The survival rate of the grafted ovarian tissue two and three weeks after transplantation. Survival rate is defined in the text. ** p < 0.01 compared with relative control group; (C–F) Immunohistochemical staining of vessels. The von Willebrand factor (vWF) protein, a marker of the endothelial cells of blood vessels, was stained red-brown (where arrow pointed); (G,H) Representative photos of H&E-stained ovarian sections show the morphology of ovarian grafts two weeks after transplantation. Scale bars: 100 µm. Figure 2 Analysis of protein levels of angiogenic cytokines in the grafted ovarian tissues treated with or without VEGF and FGF2. The grafted ovarian tissues one, two, and three weeks (A–C, respectively) after transplantation were retrieved. The relative amount of angiogenic cytokines determined by ELISA was represented as bar charts. Figure 3 The effect of VEGF and FGF2 on the rate of metaphase II (MII) oocyte, fertilization, and blastocyst development. Mice (n = 24) grafted with ovarian tissue three weeks after transplantation were treated with 150 IU gonadotropins, and MII oocytes were collected from the antral follicles for in vitro fertilization (IVF). Fertilization oocytes were cultured to the blastocyst stage. (A) The rate of MII oocytes; (B) fertilization rate; and (C) the rate of blastocyst formation. These results were statistically analyzed using the relative percentage data shown in Table S3. Values are the mean ± SD of six independent experiments (n = 4 each). * p < 0.05 compared with the control group. ijms-17-01237-t001_Table 1Table 1 Comparison of the average number of follicles in subcutaneously transplanted cryopreserved ovarian tissue, treated with or without vascular endothelial growth factor (VEGF) and fibroblast growth factor 2 (FGF2), after gonadotropin administration. Gonadotropins Untreated Control VEGF/FGF2 Treatment IU PMF PF PAF AF PMF PF PAF AF 0 22.50 ± 10.65 17.33 ± 6.12 12.00 ± 2.97 ** 12.00 ± 4.69 21.17 ± 21.08 13.67 ± 11.06 6.33 ± 3.88 6.33 ± 6.71 50 22.14 ± 14.37 12.29 ± 10.72 5.86 ± 7.31 6.71 ± 5.71 38.43 ± 17.60 * 27.71 ± 13.78 * 8.14 ± 7.38 13.14 ± 10.56 150 29.36 ± 13.50 13.91 ± 10.89 8.64 ± 5.92 6.73 ± 6.07 36.64 ± 22.35 24.09 ± 13.74 * 11.73 ± 6.48 17.91 ± 12.19 * 250 15.75 ± 12.09 9.75 ± 8.66 4.25 ± 3.86 4.50 ± 5.26 52.75 ± 13.84 ** 39.25 ± 15.11 * 18.00 ± 11.80 18.00 ± 16.75 Data represent the mean ± SD of the number of follicles at each stages (for 0, 50, 150, and 250 IU, n = 6, 7, 11, and 4, respectively). * p < 0.05, ** p < 0.01 compared with the untreated control. IU, international unit; PMF, primordial follicle; PF, primary follicle; PAF, preantral follicle; AF, antral follicle. ijms-17-01237-t002_Table 2Table 2 Outcomes of oocyte maturation in antral follicles collected from cryopreserved, subcutaneously-transplanted ovarian tissue, treated with or without VEGF and FGF2, after gonadotropin administration. Oocyte Status No. Oocytes—Control No. Oocytes—VEGF/FGF2 GV 111 (69%) 98 (51%) MI 18 (11%) 28 (14%) MII 33 (20%) 65 (35%) Total No. oocytes 162 191 Three weeks after transplantation, mice (n = 24) grafted with ovarian tissue were treated with 150 IU gonadotropins, and various stages of oocytes were collected from the antral follicles of the ovarian grafts. GV, germinal vesicle; MI, metaphase I; MII, metaphase II. ==== Refs References 1. Practice Committee of the American Society for Reproductive Medicine Practive Committee of the Society for Assisted Reproductive Technology Ovarian tissue and oocyte cryopreservation Fertil. Steril. 2006 86 S142 S147 17055811 2. Poirot C.J. Martelli H. Genestie C. Golmard J.L. Valteau-Couanet D. Helardot P. Pacquement H. Sauvat F. Tabone M.D. Philippe-Chomette P. Feasibility of ovarian tissue cryopreservation for prepubertal females with cancer Pediatr. Blood Cancer 2007 49 74 78 10.1002/pbc.21027 16977608 3. Demeestere I. Simon P. Emiliani S. Delbaere A. Englert Y. Orthotopic and heterotopic ovarian tissue transplantation Hum. Reprod. Update 2009 15 649 665 10.1093/humupd/dmp021 19474206 4. Jadoul P. Donnez J. Dolmans M.M. Squifflet J. Lengele B. Martinez-Madrid B. Laparoscopic ovariectomy for whole human ovary cryopreservation: Technical aspects Fertil. Steril. 2007 87 971 975 10.1016/j.fertnstert.2006.10.012 17173904 5. Martinez-Madrid B. Camboni A. Dolmans M.M. Nottola S. van Langendonckt A. Donnez J. Apoptosis and ultrastructural assessment after cryopreservation of whole human ovaries with their vascular pedicle Fertil. Steril. 2007 87 1153 1165 10.1016/j.fertnstert.2006.11.019 17239875 6. Martinez-Madrid B. Donnez J. Cryopreservation of intact human ovary with its vascular pedicle—Or cryopreservation of hemiovaries? Hum. Reprod. 2007 22 1795 1796 10.1093/humrep/dem047 17400594 7. Shamonki M.I. Oktay K. Oocyte and ovarian tissue cryopreservation: Indications, techniques, and applications Semin. Reprod. Med. 2005 23 266 276 10.1055/s-2005-872455 16059833 8. Donnez J. Dolmans M.M. Demylle D. Jadoul P. Pirard C. Squifflet J. Martinez-Madrid B. van Langendonckt A. Livebirth after orthotopic transplantation of cryopreserved ovarian tissue Lancet 2004 364 1405 1410 10.1016/S0140-6736(04)17222-X 15488215 9. Donnez J. Squifflet J. Jadoul P. Demylle D. Cheron A.C. van Langendonckt A. Dolmans M.M. Pregnancy and live birth after autotransplantation of frozen-thawed ovarian tissue in a patient with metastatic disease undergoing chemotherapy and hematopoietic stem cell transplantation Fertil. Steril. 2011 95 10.1016/j.fertnstert.2010.11.041 21145049 10. Ernst E. Bergholdt S. Jorgensen J.S. Andersen C.Y. The first woman to give birth to two children following transplantation of frozen/thawed ovarian tissue Hum. Reprod. 2010 25 1280 1281 10.1093/humrep/deq033 20172869 11. Chen S.U. Chien C.L. Wu M.Y. Chen T.H. Lai S.M. Lin C.W. Yang Y.S. Novel direct cover vitrification for cryopreservation of ovarian tissues increases follicle viability and pregnancy capability in mice Hum. Reprod. 2006 21 2794 2800 10.1093/humrep/del210 16982660 12. Snow M. Cox S.L. Jenkin G. Trounson A. Shaw J. Generation of live young from xenografted mouse ovaries Science 2002 297 10.1126/science.1073693 12351780 13. Waterhouse T. Cox S.L. Snow M. Jenkin G. Shaw J. Offspring produced from heterotopic ovarian allografts in male and female recipient mice Reproduction 2004 127 689 694 10.1530/rep.1.00081 15175505 14. Lee D.M. Yeoman R.R. Battaglia D.E. Stouffer R.L. Zelinski-Wooten M.B. Fanton J.W. Wolf D.P. Live birth after ovarian tissue transplant Nature 2004 428 137 138 10.1038/428137a 15014485 15. Oktay K. Economos K. Kan M. Rucinski J. Veeck L. Rosenwaks Z. Endocrine function and oocyte retrieval after autologous transplantation of ovarian cortical strips to the forearm JAMA 2001 286 1490 1493 10.1001/jama.286.12.1490 11572742 16. Oktay K. Buyuk E. Veeck L. Zaninovic N. Xu K. Takeuchi T. Opsahl M. Rosenwaks Z. Embryo development after heterotopic transplantation of cryopreserved ovarian tissue Lancet 2004 363 837 840 10.1016/S0140-6736(04)15728-0 15031026 17. Oktay K. Turkcuoglu I. Rodriguez-Wallberg K.A. Four spontaneous pregnancies and three live births following subcutaneous transplantation of frozen banked ovarian tissue: What is the explanation? Fertil. Steril. 2011 95 10.1016/j.fertnstert.2010.07.1072 20801441 18. Yang H.Y. Cox S.L. Jenkin G. Findlay J. Trounson A. Shaw J. Graft site and gonadotrophin stimulation influences the number and quality of oocytes from murine ovarian tissue grafts Reproduction 2006 131 851 859 10.1530/rep.1.00916 16672350 19. Li F. Tao Y. Zhang Y. Li Y. Fang F. Liu Y. Cao H. Zhang X. Zhou S. Follicle growth and oocyte development after ovary transplantation into back muscle of immune-intact adult castrated male mice Reproduction 2010 140 465 476 10.1530/REP-10-0076 20530631 20. Barboni B. Turriani M. Galeati G. Spinaci M. Bacci M.L. Forni M. Mattioli M. Vascular endothelial growth factor production in growing pig antral follicles Biol. Reprod. 2000 63 858 864 10.1095/biolreprod63.3.858 10952932 21. Jakeman L.B. Winer J. Bennett G.L. Altar C.A. Ferrara N. Binding sites for vascular endothelial growth factor are localized on endothelial cells in adult rat tissues J. Clin. Investig. 1992 89 244 253 10.1172/JCI115568 1729274 22. Koos R.D. Potential relevance of angiogenic factors to ovarian physiology Semin. Reprod. Endocrinol. 1989 7 29 40 10.1055/s-2007-1021379 23. Neufeld G. Ferrara N. Schweigerer L. Mitchell R. Gospodarowicz D. Bovine granulosa cells produce basic fibroblast growth factor Endocrinology 1987 121 597 603 10.1210/endo-121-2-597 3036478 24. Wolner-Hanssen P. Hagglund L. Ploman F. Ramirez A. Manthorpe R. Thuring A. Autotransplantation of cryopreserved ovarian tissue to the right forearm 4(1/2) years after autologous stem cell transplantation Acta Obstet. Gynecol. Scand. 2005 84 695 698 10.1111/j.0001-6349.2005.00654.x 15954881 25. Oktay K. Newton H. Gosden R.G. Transplantation of cryopreserved human ovarian tissue results in follicle growth initiation in SCID mice Fertil. Steril. 2000 73 599 603 10.1016/S0015-0282(99)00548-8 10689020 26. Oktay K. Newton H. Mullan J. Gosden R.G. Development of human primordial follicles to antral stages in SCID/hpg mice stimulated with follicle stimulating hormone Hum. Reprod. 1998 13 1133 1138 10.1093/humrep/13.5.1133 9647533 27. De Laporte L. des Rieux A. Tuinstra H.M. Zelivyanskaya M.L. de Clerck N.M. Postnov A.A. Preat V. Shea L.D. Vascular endothelial growth factor and fibroblast growth factor 2 delivery from spinal cord bridges to enhance angiogenesis following injury J. Biomed. Mater. Res. Part A 2011 98 372 382 10.1002/jbm.a.33112 21630429 28. Sun X.T. Ding Y.T. Yan X.G. Wu L.Y. Li Q. Cheng N. Qiu Y.D. Zhang M.Y. Angiogenic synergistic effect of basic fibroblast growth factor and vascular endothelial growth factor in an in vitro quantitative microcarrier-based three-dimensional fibrin angiogenesis system World J. Gastroenterol. 2004 10 2524 2528 10.3748/wjg.v10.i17.2524 15300897 29. Yoshiji H. Kuriyama S. Yoshii J. Ikenaka Y. Noguchi R. Hicklin D.J. Huber J. Nakatani T. Tsujinoue H. Yanase K. Synergistic effect of basic fibroblast growth factor and vascular endothelial growth factor in murine hepatocellular carcinoma Hepatology 2002 35 834 842 10.1053/jhep.2002.32541 11915029 30. Oktay K. Aydin B.A. Karlikaya G. A technique for laparoscopic transplantation of frozen-banked ovarian tissue Fertil. Steril. 2001 75 1212 1216 10.1016/S0015-0282(01)01776-9 11384651 31. Lee R.K. Ho H.Y. Yu S.L. Lu C.H. Blastocyst development after cryopreservation and subcutaneous transplantation of mouse ovarian tissue J. Assist. Reprod. Genet. 2005 22 95 101 10.1007/s10815-005-1499-z 15844735 32. Devine K. Connell M.T. Richter K.S. Ramirez C.I. Levens E.D. DeCherney A.H. Stillman R.J. Widra E.A. Single vitrified blastocyst transfer maximizes liveborn children per embryo while minimizing preterm birth Fertil. Steril. 2015 103 1454 1460 10.1016/j.fertnstert.2015.02.032 25813283 33. Frattarelli J.L. Leondires M.P. McKeeby J.L. Miller B.T. Segars J.H. Blastocyst transfer decreases multiple pregnancy rates in in vitro fertilization cycles: A randomized controlled trial Fertil. Steril. 2003 79 228 230 10.1016/S0015-0282(02)04558-2 12524098 34. Schnorr J. Oehninger S. Toner J. Hsiu J. Lanzendorf S. Williams R. Hodgen G. Functional studies of subcutaneous ovarian transplants in non-human primates: Steroidogenesis, endometrial development, ovulation, menstrual patterns and gamete morphology Hum. Reprod. 2002 17 612 619 10.1093/humrep/17.3.612 11870112 35. Gao J.M. Yan J. Li R. Li M. Yan L.Y. Wang T.R. Zhao H.C. Zhao Y. Yu Y. Qiao J. Improvement in the quality of heterotopic allotransplanted mouse ovarian tissues with basic fibroblast growth factor and fibrin hydrogel Hum. Reprod. 2013 28 2784 2793 10.1093/humrep/det296 23892320 36. Gao J. Huang Y. Li M. Zhao H. Zhao Y. Li R. Yan J. Yu Y. Qiao J. Effect of local basic fibroblast growth factor and vascular endothelial growth factor on subcutaneously allotransplanted ovarian tissue in ovariectomized mice PLoS ONE 2015 10 1237 10.1371/journal.pone.0134035 26208097 37. Kang B.J. Wang Y. Zhang L. Xiao Z. Li S.W. bFGF and VEGF improve the quality of vitrified-thawed human ovarian tissues after xenotransplantation to SCID mice J. Assist. Reprod. Genet. 2016 33 281 289 10.1007/s10815-015-0628-6 26712576 38. Cha S.K. Shin D.H. Kim B.Y. Yoon S.Y. Yoon T.K. Lee W.S. Chung H.M. Lee D.R. Effect of human endothelial progenitor cell (EPC)- or mouse vascular endothelial growth factor-derived vessel formation on the survival of vitrified/warmed mouse ovarian grafts Reprod. Sci. 2014 21 859 868 10.1177/1933719113518983 24401473 39. Pedersen T. Peters H. Proposal for a classification of oocytes and follicles in the mouse ovary J. Reprod. Fertil. 1968 17 555 557 10.1530/jrf.0.0170555 5715685 40. Summers M.C. McGinnis L.K. Lawitts J.A. Raffin M. Biggers J.D. IVF of mouse ova in a simplex optimized medium supplemented with amino acids Hum. Reprod. 2000 15 1791 1801 10.1093/humrep/15.8.1791 10920105

PMC005xxxxxx/PMC5000636.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081238ijms-17-01238CommunicationNavicula sp. Sulfated Polysaccharide Gels Induced by Fe(III): Rheology and Microstructure Fimbres-Olivarría Diana 1López-Elías José Antonio 1Carvajal-Millán Elizabeth 2*Márquez-Escalante Jorge Alberto 2Martínez-Córdova Luis Rafael 1Miranda-Baeza Anselmo 3Enríquez-Ocaña Fernando 1Valdéz-Holguín José Eduardo 1Brown-Bojórquez Francisco 4Audic Jean-Luc Academic Editor1 DICTUS, Department of Scientific and Technological Investigations, University of Sonora, Hermosillo, Sonora 83000, Mexico; [email protected] (D.F.-O.); [email protected] (J.A.L.-E.); [email protected] (L.R.M.-C.); [email protected] (F.E.-O.); [email protected] (J.E.V.-H.)2 CIAD, A.C., Research Center for Food and Development, Hermosillo, Sonora 83000, Mexico; [email protected] UES, State University of Sonora, Navojoa, Sonora 85875, Mexico; [email protected] Department of Polymers and Materials, University of Sonora, Hermosillo, Sonora 83000, Mexico; [email protected]* Correspondence: [email protected]; Tel.: +52-662-289-2400; Fax: +52-662-280-042130 7 2016 8 2016 17 8 123816 6 2016 19 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).A sulfated polysaccharide extracted from Navicula sp. presented a yield of 4.4 (% w/w dry biomass basis). Analysis of the polysaccharide using gas chromatography showed that this polysaccharide contained glucose (29%), galactose (21%), rhamnose (10%), xylose (5%) and mannose (4%). This polysaccharide presented an average molecular weight of 107 kDa. Scanning electron microscopy (SEM) micrographs showed that the lyophilized Navicula sp. polysaccharide is an amorphous solid with particles of irregular shapes and sharp angles. The polysaccharide at 1% (w/v) solution in water formed gels in the presence of 0.4% (w/v) FeCl3, showing elastic and viscous moduli of 1 and 0.7 Pa, respectively. SEM analysis performed on the lyophilized gel showed a compact pore structure, with a pore size of approximately 150 nm. Very few studies on the gelation of sulfated polysaccharides using trivalent ions exist in the literature, and, to the best of our knowledge, this study is the first to describe the gelation of sulfated polysaccharides extracted from Navicula sp. Navicula sp.sulfated polysaccharidegelationtrivalent ions ==== Body 1. Introduction For several years, marine microalgae have been of great interest because they contain a great variety of bioactive compounds with biotechnological potential, especially in the biomedical, pharmaceutical, nutraceutical, and cosmetic areas. Among the wide variety of microalgae used for biotechnological purposes are the diatoms, whose principal purpose is the production of biodiesel due to their high lipid content [1]. Some diatoms are benthic microalgae; they produce mucilage that binds them to their substrate. This mucilage is a matrix with a high content of extracellular polymeric substances, including polysaccharides [2]. The marine microalgae of the Navicula genus are benthic diatoms, and several bioactive compounds of commercial interest can be obtained from them, including polysaccharides [3,4,5]. Several studies have proven that microalgae polysaccharides have great potential as antiviral, antibacterial, and antioxidant compounds, among other uses. Despite some research on their applications appearing already, information on sulfated polysaccharides from species of the genus Navicula is still scarce. Currently, no reports exist on the gelation behavior of sulfated polysaccharides from this genus. However, there is some evidence that sulfate polysaccharides can form gels in the presence of trivalent ions, as shown for λ-carrageenan from seaweeds [6]. The aim of this study was to investigate the gelation of a sulfated polysaccharide from Navicula sp. in the presence of trivalent iron ions and to study the rheological and microstructural characteristics of the gel formed. 2. Results and Discussions 2.1. Polysaccharide Characteristics The polysaccharide yield was 4.4 (% w/w dry biomass basis), nearest to the values reported in the diatom Gomphonema olivaceum (3% w/w) [7] and in the benthic seaweed Sargassum qingdaoense (7.2% w/w) [8], but lower than the values reported for the planktonic specie Spirulina platensis (13.6% w/w) [9]. These differences could be due to the extraction methods used and/or the type of species investigated. The extracted polysaccharide consisted of a white-colored powder with fine and granulated parts. Scanning electron microscopy (SEM) can be a useful tool to analyze the surface morphology of polysaccharide powder. The SEM micrographs showed that the lyophilized Navicula sp. polysaccharide is an amorphous solid (Figure 1). The observed particles were mostly aggregates of irregular shapes with sharp angles similar to those reported for other sulfated polysaccharides [10]. The main sugars present in the polysaccharide were glucose, galactose, rhamnose, xylose and mannose (Table 1), glucose being the most abundant, with ca. 30% of the polysaccharide dry weight. Staats et al. [11] found that extracellular polysaccharides from Navicula salinarum were mainly composed of glucose, galactose, mannose, rhamnose and xylose, with galactose concentrations similar to those found in the present study. On the other hand, Lee et al. [3] reported the presence of fucose, xylose, galactose, mannose and rhamnose in Navicula directa extracts, but at higher concentrations (% w/w dry weight basis) than those found in this study. A small amount of protein (0.48% w/w) was also detected in the polysaccharide from Navicula sp. (Table 1). However, a higher content of protein has been reported for polysaccharides from Chlorella pyrenoidosa at different ethanol concentrations (0.75%–11.21% w/w) [12]. The sulfate content found in the polysaccharide from Navicula sp. in this study (0.33%) (Table 1) was in the range reported for a sulfated galactan from the red algae Ahnfeltia tobuchiensis (0.2%–0.3% w/w) [13] but lower than that in other reports for Navicula species (8% and 11% w/w) [3,11]. However, it is well known that the sulfate content in microalgae is highly variable and can range from 0 to approximately 90% [14]. In the present study, the molecular weight (Mw) for the Navicula sp. sulfated polysaccharide was 107 kDa, lower than the reported value in another study with Navicula directa (222 kDa) [3]. However, it should be mentioned that the information and characterization of sulfated polysaccharides of the genus Navicula are still emerging. It should also be noted that the characteristics of microalgae and their biological compounds depend heavily on the culture conditions used and, to an even greater extent, on the species [15,16]. The Fourier transform infrared (FT-IR) spectrum of the sulfated polysaccharide extract showed five distinct bands at wave numbers ranging from 3405–821 cm−1 (Figure 2). The bands were assigned to particular functional groups according to previously published literature [17,18]. The spectrum of this polysaccharide showed the typical infrared footprint of carbohydrates. The band in the region of 3405 cm−1 corresponds to the stretching vibration characteristic of OH groups; a similar band around this wavenumber was observed for sulfated polysaccharides from green and brown seaweeds [19,20]. The band related to amides associated with the protein was detected at 1656 cm−1; this band has also been detected in other microalgae [18]. The most important band was found at 1137 cm−1, assigned to C–O–C bending; similar bands were reported for sulfated polysaccharides from brown and red seaweeds [21,22]. The band corresponding to the S=O vibration (1244 cm−1) possesses a low intensity; this result could be due to the low sulfate content detected on the sample (0.33% w/w) as reported in Table 1. Some studies have reported the presence of this band in sulfated polysaccharides extracted from the diatom Navicula directa [3] and from the three major groups of seaweeds (green, brown and red algae) [19,20,21,23]. Finally, the band at 821 cm−1 was attributed to C–O–S stretching vibrations; sulfated polysaccharides extracted from some green and brown seaweed species also showed a band specific to the C–O–S group [19,20,21,23,24] around a similar wavelength as in our study. 2.2. Sulfated Polysaccharide Gelation Previous experiments were carried out in the present research in order to evaluate the gelation ability of the sulfated polysaccharide from Navicula sp. in the presence of mono and divalent cations (KCl and CaCl2, respectively). However, for those cations no gelation was observed. When a 0.4% (w/v) FeCl3 solution was dropped into a 1% (w/v) sulfated polysaccharide aqueous solution, a yellow-orange-colored gel-like substance precipitated as previously reported for λ-carrageenan [6]. The gel-like material was formed after 60 s of FeCl3 addition. It has been suggested that trivalent iron metals promote appropriate ionic interactions between sulfated polysaccharide chains, causing their union and subsequent gelation. However, the gelling mechanism of sulfated polysaccharides in the presence of trivalent ions is currently unknown [6,25]. The precipitated coagulum formed in the present study was recovered for further rheological and microstructural characterization. The FeCl3-induced gelation of the sulfated polysaccharide from Navicula sp. was rheologically investigated by small amplitude oscillatory shear. Figure 3 shows the changes in the elastic (G’) and viscous (G”) moduli of 1% (w/v) polysaccharide/FeCl3 from 5 to 70 °C. The sample showed G’ and G” values of 1.0 and 0.7 Pa, respectively, from 20 to 40 °C, indicating a gelation behavior. Higher G’ and G” values were found in λ-carrageenan/FeCl3 gels (G’ = 1200 Pa, G” = 150 Pa) [6] which could be related to a higher sulfate content reported for that sample [26]. In the present study, the polysaccharide/FeCl3 gel was thermally stable from 20 to 40 °C, as there was no crossover between G’ and G” in this temperature region. The temperature at which this polysaccharide gel was thermally stable could allow its use in biomedical applications, where the implementation of organic material that supports the body temperature is needed. In a study by Vorvolakos et al. [27], it was observed that hyaluronic acid could form gels in the presence of trivalent cations, a behavior similar to the polysaccharide in our study. The hyaluronic acid gel can be utilized in laparoscopic surgeries to avoid adhesions [28]; because its gelling characteristics were similar to those in our study, Navicula sp. sulfated polysaccharide evaluation in that application could be of keen interest. Mechanical spectra (Figure 4) of the gel were recorded at 25 °C, being typical of a solid-like material with a linear G’ independent of frequency and a G” much smaller than G’ in the frequency interval from 0.1 up to 1.0 Hz [29]. At higher frequency values (from 1.0 up to 10 Hz), G’ and G” enter into the non-linear range as a result of excessive oscillation frequency exposure, corresponding to a weak gel-like behavior [30]. The tangent delta values (tan δ = G”/G’) of the gel as a function of frequency sweep are also presented in Figure 4. Under the experimental conditions used in the present study, the tan δ values registered varied from 0.46 to 0.12 when the frequency changed from 0.1 to 10.0 Hz. These tan δ values are typical of so-called weak gels [29]. When subjected to the strain sweep test, this polysaccharide gel showed a linear behavior from 1.5% to 10.0% strain (Figure 5). The elastic character of this gel could be attributed to the temporary association of sulfated polysaccharide chains during short oscillation periods. It has been suggested that trivalent ions could be more suitable than monovalent ions for balancing the three negative sulfate charges, per disaccharide repeat unit, of polysaccharides such as λ-carrageenan [6]. In Figure 6 the sulfated polysaccharide solution before (a) and after (b) FeCl3 addition is observed. The yellow gel-like substance was lyophilized (Figure 6c) and analyzed by SEM (Figure 6d). SEM micrographs of the lyophilized polysaccharide gel present a compact pore structure, with an irregular pore size of approximately 150 nm. The gel formed with this trivalent metal consisted of fine-stranded networks with strand thickness on the nm scale. In could be assumed that the SEM preparation method does not affect the sizes of the domains of the network structure. Nevertheless, it is important to note that lyophilized gel does not allow visualizing the original wet-polymeric network but it can be useful to investigate the dried microstructure of the polysaccharide gels. 3. Materials and Methods 3.1. Materials The microalgae Navicula sp. was obtained and cultured as previously reported [31]. All chemical reagents were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO, USA). 3.2. Methods 3.2.1. Extraction of Polysaccharide At the end of the microalgal culture, the full biomass was harvested by gravity sedimentation method [32] and lyophilized using a Freezone 6 freeze dry system (Labconco, Kansas, MO, USA). Once lyophilized, soluble sulfated polysaccharides were obtained by suspending the lyophilized total biomass in distilled water for 1 h at 30 °C, the suspended biomass was then centrifuged for 15 min at 20,000× g. Finally, the supernatant was separated and precipitated overnight under cold conditions with 96% (v/v) ethanol to allow for the precipitation of sulfated polysaccharides from Navicula sp. [11]. Precipitate was recovered and dried by solvent exchange (96% (v/v) ethanol and pure acetone) and the polysaccharide from Navicula sp. was obtained as reported for other marine sulfated polysaccharides [8,9]. 3.2.2. Chemical Analysis The sulfate content of the extracted polysaccharide was determined after hydrolysis with 1 N HCl at 100 °C for 1 h following the sodium-rhodizonate method proposed by Terho and Hartiala [33]. Na2SO4 was utilized as a standard. The protein content was analyzed using the Dumas method (Leco FP-528 nitrogen analyzer, St. Joseph, MI, USA) [34]. The monosaccharide content was analyzed by gas chromatography (Agilent HP 6890 GC Series, Santa Clara, CA, USA) [35]. Briefly, the polysaccharide sample was hydrolyzed with 3 N H2SO4 (98% v/v) at 100 °C, and inositol was added as the internal standard. The external standards were glucose, mannose, galactose, xylose and rhamnose (1 mg/mL, w/v), which were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO, USA). Sugars were reduced to alditols with sodium borohydride, acetylated with acetic anhydride in the presence of methyl imidazole, and finally extracted with chloroform. After extraction, the alditol-acetates were injected (5 µL) in a DB 225 type column (50% cyanopropylphenyl-dimethylpolysiloxane, 30 m × 0.32 mm ID, 0.15 μm). The gas chromatography conditions were as follows: injection temperature 220 °C, detector temperature 260 °C, and oven temperature programmed to 205 °C at 10 °C/min. Nitrogen was used as the carrier gas and maintained at 1.0 mL/min. 3.2.3. Fourier Transform Infrared (FT-IR) Spectroscopy The polysaccharide powder and the lyophilized trivalent gels were pressed into KBr pellets. A blank KBr disk was used as background. FT-IR spectrum was recorded on a Nicolet FT-IR spectrophotometer (Nicolet Instruments Corp., Madison, WI, USA). The FT-IR spectrum was measured in absorbance mode from 4000–400 cm−1. 3.2.4. Molecular Weight Determination The molecular characteristics based on the absolute weight-average molecular weight (MW) of polysaccharide was analyzed by high-performance size-exclusion chromatography (HPSEC) attached to a multiangle laser-light scattering (MALLS) and refractive index (RI) detector (mini-Dawn®, Wyatt, Milford, MA, USA). The polysaccharide extract (1 mg/mL w/v) was dissolved in 100 mM NaNO3, filtered through a 0.2 µm membrane, and injected at 25 °C. The RI increment (dn/dc) utilized for the polysaccharide extract was 0.147 mL/g. 3.2.5. Rheological Measurements The gelation of the polysaccharide extract was carried out with the following reaction mixture: 1% w/v of polysaccharide solution with 0.4% w/v FeCl3 in water. For rheological tests, the sulfated polysaccharide gel formation was followed using a strain controller rheometer (Discovery HR-2 rheometer; TA Instruments, New Castle, DE, USA) along with a parallel plate geometry with a plate diameter of 40 mm. A temperature ramp was carried out from 5 to 70 °C at a frequency of 1 Hz and 2% strain. Frequency sweep test was performed from 0.1 to 10 Hz at 2% strain and 25 °C. Strain sweep experiment was done from 0.4 to 10% strain at a 1 Hz frequency and 25 °C. All measurements were performed in duplicate. 3.2.6. Scanning Electron Microscopy Imaging The polysaccharide powder and the lyophilized gel were all analyzed by field emission scanning electron microscopy (SEM) (JEOL 5410LV, JEOL, Peabody, MA, USA) using a voltage of 10 kV and ×100, ×200 or ×5000 magnifications. SEM images were obtained in secondary and backscattered electrons imaging modes. 4. Conclusions The present study demonstrated that the sulfated polysaccharide from Navicula sp. can form gels in the presence of trivalent iron cations and showed the basic viscoelastic and microstructural characteristics of this material. This finding has the potential to expand the utility of sulfated polysaccharides from microalgae in different biotechnological applications and provides a basis for further structural analysis and evaluation of the bioactivities of this sulfated polysaccharide and its trivalent gel. Acknowledgments This research was supported by “Fondo de Infraestructura de CONACYT, Mexico (Grant 226082 to E. Carvajal-Millan)”. The authors are pleased to acknowledge Alma C. Campa-Mada and Karla Martínez-Robinson (CIAD, Mexico) for technical assistance. Author Contributions Diana Fimbres-Olivarría performed the experiments, analyzed the data and wrote the paper draft. José Antonio López-Elías and Elizabeth Carvajal-Millán conceived and designed the experiments and edited the paper. Jorge Alberto Márquez-Escalante performed the rheological experiments. Anselmo Miranda-Baeza, Luis Rafael Martínez-Córdova, Fernando Enríquez-Ocaña and José Eduardo Valdéz-Holguín analyzed the data and collaborated to edit the paper. Francisco Brown-Bojórquez performed the SEM analysis. Conflicts of Interest The authors declare no conflicts of interest. Figure 1 Scanning electron microscopy (SEM) micrographs of lyophilized polysaccharide extracted from Navicula sp. at ×100 (a) and ×200 (b). Figure 2 Fourier transform infrared (FT-IR) spectrum of sulfated polysaccharide from Navicula sp. The arrows indicate the principal absorption bands. Figure 3 Temperature ramp for sulfated polysaccharide at 1% (w/v) in the presence of trivalent ions of FeCl3 at 0.4% (w/v) at 1 Hz and 2% strain. G’ (●), G” (○). Figure 4 Mechanical spectra of sulfated polysaccharide gel at 1% (w/v) induced by FeCl3 at 0.4% (w/v). Measurements at 2% strain and 25 °C. G’ (●), G” (○), tan δ (×). Figure 5 Strain sweep of sulfated polysaccharide gel at 1% (w/v) induced by FeCl3 at 0.4% (w/v). Measurements at 1 Hz and 25 °C. G’ (●), G” (○). Figure 6 Sulfated polysaccharide from Navicula sp. before (a) and after (b) the addition of FeCl3; lyophilized gel (c); SEM micrograph of the lyophilized gel (magnification ×5000, scale bar 25 µm) (d). ijms-17-01238-t001_Table 1Table 1 Composition of sulfated polysaccharides from Navicula sp. Compounds % w/w Dry Weight Basis Glucose 29.23 ± 2.04 Galactose 21.37 ± 2.27 Rhamnose 10.67 ± 2.66 Xylose 5.18 ± 1.09 Mannose 4.43 ± 0.79 Protein 0.480 ± 0.001 Sulfate 0.330 ± 0.004 All results were obtained from duplicates. ==== Refs References 1. Markou G. Angelidaki I. Georgakakis D. Microalgal carbohydrates: An overview of the factors influencing carbohydrates production, and of main bioconversion technologies for production of biofuels Appl. Microbiol. Biotechnol. 2012 96 631 645 10.1007/s00253-012-4398-0 22996277 2. Raposo M.F.D.J. de Morais R.M.S.C. de Morais B.A.M.M. Bioactivity and applications of sulphated polysaccharides from marine microalgae Mar. Drugs 2013 11 233 252 10.3390/md11010233 23344113 3. Lee J.B. Hayashi K. Hirata M. Kuroda E. Suzuki E. Kubo Y. Hayashi T. Antiviral sulfated polysaccharide from Navicula directa , a diatom collected from deep-sea water in Toyama Bay Biol. Pharm. Bull. 2006 29 2135 2139 10.1248/bpb.29.2135 17015966 4. Jiao G. Yu G. Zhang J. Ewart H.S. Chemical structures and bioactivities of sulfated polysaccharides from marine algae Mar. Drugs 2011 9 196 223 10.3390/md9020196 21566795 5. Wang W. Wang S.X. Guan H.S. The antiviral activities and mechanisms of marine polysaccharides: An overview Mar. Drugs 2012 10 2795 2816 10.3390/md10122795 23235364 6. Running C.A. Falshaw R. Janaswamy S. Trivalent iron induced gelation in λ-carrageenan Carbohydr. Polym. 2012 87 2735 2739 10.1016/j.carbpol.2011.11.018 22408280 7. Huntsman S.A. Sloneker J.H. An exocellular polysaccharide from the diatom Gomphonema olivaceum J. Phycol. 1971 7 261 264 8. Song L. Chen X. Liu X. Zhang F. Hu L. Yue Y. Kecheng L. Li P. Characterization and Comparison of the Structural Features, Immune-Modulatory and Anti-Avian Influenza Virus Activities Conferred by Three Algal Sulfated Polysaccharides Mar. Drugs 2015 14 4 10.3390/md14010004 26729137 9. Kurd F. Samavati V. Water soluble polysaccharides from Spirulina platensis : Extraction and in vitro anti-cancer activity Int. J. Biol. Macromol. 2015 74 498 506 10.1016/j.ijbiomac.2015.01.005 25583023 10. Kravchenko A.O. Anastyuk S.D. Sokolova E.V. Isakov V.V. Glazunov V.P. Helbert W. Yermak I.M. Structural analysis and cytokine-induced activity of gelling sulfated polysaccharide from the cystocarpic plants of Ahnfeltiopsis flabelliformis Carbohydr. Polym. 2016 151 523 534 10.1016/j.carbpol.2016.05.086 27474596 11. Staats N. de Winder B. Stal L. Mur L. Isolation and characterization of extracellular polysaccharides from the epipelic diatoms Cylindrotheca closterium and Navicula salinarum Eur. J. Phycol. 1999 34 161 169 10.1080/09670269910001736212 12. Chen Y.X. Liu X.Y. Xiao Z. Huang Y.F. Liu B. Antioxidant activities of polysaccharides obtained from Chlorella pyrenoidosa via different ethanol concentrations Int. J. Biol. Macromol. 2016 91 505 509 10.1016/j.ijbiomac.2016.05.086 27235730 13. Truus K. Tuvikene R. Vaher M. Kailas T. Toomik P. Pehk T. Structural and compositional characteristics of gelling galactan from the red alga Ahnfeltia tobuchiensis (Ahnfeltiales, the Sea of Japan) Carbohydr. Polym. 2006 63 130 135 10.1016/j.carbpol.2005.08.029 14. Pérez Loyola M. Popowski Casaña G. Pérez Castillo G. Caracterización biológica y química del biogel del balneario San Diego de los Baños Rev. Cubana Plant Med. (Online) 2003 8 Available online: http://scielo.sld.cu/scielo.php?script=sci_arttext&pid=S1028-47962003000300011&lng=es&nrm=isoISSN 1028-4796 (accessed on 22 May 2016) 15. Dautania G.K. Singh G.P. Role of Light and Dark Cycle and Different Temperatures in the Regulation of Growth and Protein Expression in Oscillatoria agardhii Strain Braz. Arch. Biol. Technol. 2014 57 933 940 16. De Morais M.G. Vaz B.D.S. de Morais E.G. Costa J.A.V. Biologically Active Metabolites Synthesized by Microalgae BioMed Res. Int. 2015 2015 1 15 17. Socrates G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts 3rd ed. John Wiley & Sons Chichester, UK 2001 327 18. Stehfest K. Toepel J. Wilhelm C. The application of micro-FTIR spectroscopy to analyze nutrient stress-related changes in biomass composition of phytoplankton algae Plant Physiol. Biochem. 2005 43 717 726 16122937 19. Shao P. Chen M. Pei Y. Sun P. In intro antioxidant activities of different sulfated polysaccharides from chlorophyta seaweeds Ulva fasciata Int. J. Biol. Macromol. 2013 59 295 300 23643973 20. Saboural P. Chaubet F. Rouzet F. Al-Shoukr F. Azzouna R.B. Bouchemal N. Picton L. Louedec L. Maire M. Rolland L. Purification of a low molecular weight fucoidan for spect molecular imaging of myocardial infarction Mar. Drugs 2014 12 4851 4867 25251032 21. Morimoto M. Takatori M. Hayashi T. Mori D. Takashima O. Yoshida S. Sato K. Kawamoto H. Tamura J. Izawa H. Depolymerization of sulfated polysaccharides under hydrothermal conditions Carbohydr. Res. 2014 384 56 60 24361592 22. Abdala-Díaz R.T. Chabrillón M. Cabello-Pasini A. López-Soler B. Figueroa F.L. Efecto de los polisacáridos de Porphyridium cruentum sobre la actividad de la línea celular de macrófagos murinos RAW 264.7 Cienc. Mar. 2010 36 345 353 23. Souza B.W. Cerqueira M.A. Bourbon A.I. Pinheiro A.C. Martins J.T. Teixeira J.A. Vicente A.A. Chemical characterization and antioxidant activity of sulfated polysaccharide from the red seaweed Gracilaria birdiae Food Hydrocoll. 2012 27 287 292 24. Shao P. Chen X. Sun P. In vitro antioxidant and antitumor activities of different sulfated polysaccharides isolated from three algae Int. J. Biol. Macromol. 2013 62 155 161 23994786 25. Le Nguyen Q.T. Okajima M. Mitsumata T. Kan K. Tran H.T. Kaneko T. Trivalent metal-mediated gelation of novel supergiant sulfated polysaccharides extracted from Aphanothece stagnina Colloid Polym. Sci. 2012 290 163 172 26. Falshaw R. Furneaux R.H. Structural analysis of carrageenans from the tetrasporic stages of the red algae, Gigartina lanceata and Gigartina chapmanii (Gigartinaceae, Rhodophyta) Carbohydr. Res. 1998 307 325 331 27. Vorvolakos K. Isayeva I.S. Do Luu H.M. Patwardhan D.V. Pollack S.K. Ionically cross-linked hyaluronic acid: Wetting, lubrication, and viscoelasticity of a modified adhesion barrier gel Med. Devices (Auckl.) 2011 4 1 10 22915924 28. Farell-Rivas J. Cruz Santiago C.A. Mata Quintero C.J. Cuevas Osorio V.J. Luna Martínez J. Aplicación de carboximetilcelulosa/ácido hialurónico líquido sobre mallas intraperitoneales para disminuir la formación de adherencias. Estudio experimental Cir. Gen. 2014 36 3 8 29. Clark A.H. Gels and gelling Physical Chemistry of Foods 1st ed. Schwartzberg H.G. Hartel R.W. Marcel Dekker New York, NY, USA 1992 263 306 30. Steffe J.F. Rheological Methods in Food Process Engineering 2nd ed. Freeman Press East Lansing, MI, USA 1996 418 31. Fimbres-Olivarría D. López-Elías J.A. Martínez-Córdova L.R. Carvajal-Millán E. Enríquez-Ocaña F. Valdéz-Holguín E. Miranda-Baeza A. Growth and Biochemical Composition of Navicula sp. Cultivated at Two Light Intensities and Three Wavelengths Isr. J. Aquac. Bamidgeh 2015 67 1 7 32. Shelef G. Sukenik A. Green M. Microalgae harvesting and processing: A literature review (No. SERI/STR-231-2396) Tech. Res. Dev. Found. Ltd. Haifa (Isr.) 1984 1 66 33. Terho T.T. Hartiala K. Method for Determination of the Sulfate Content of Glycosaminoglycans Anal. Biochem. 1971 41 471 476 4253906 34. AOAC Association of official analytical chemists Official Methods of Analysis of AOAC International AOAC International Arlington, VA, USA 1995 35. Rouau X. Surget A. A rapid semi-automated method for the determination of total and water-extractable pentosans in wheat flours Carbohydr. Polym. 1994 24 123 132

PMC005xxxxxx/PMC5000637.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081239ijms-17-01239ArticleA Next-Generation Sequencing Approach to Identify Gene Mutations in Early- and Late-Onset Hypertrophic Cardiomyopathy Patients of an Italian Cohort Rubattu Speranza 12*†Bozzao Cristina 1†Pennacchini Ermelinda 1‡Pagannone Erika 1Musumeci Beatrice Maria 1Piane Maria 1Germani Aldo 1Savio Camilla 1Francia Pietro 1Volpe Massimo 12Autore Camillo 1*†Chessa Luciana 1†Cho William Chi-shing Academic Editor1 Department of Clinical and Molecular Medicine, School of Medicine and Psychology, University Sapienza of Rome, 00185 Rome, Italy; [email protected] (C.B.); [email protected] (E.Pe.); [email protected] (E.Pa.); [email protected] (B.M.M.); [email protected] (M.P.); [email protected] (A.G.); [email protected] (C.S.); [email protected] (P.F.); [email protected] (M.V.); [email protected] (L.C.)2 Department of Angiocardioneurology, IRCCS Neuromed, 86077 Pozzilli, Italy* Correspondence: [email protected] (S.R.); [email protected] (C.A.); Tel.: +39-06-3377-5979 (S.R. & C.A.); Fax: +39-06-3377-5061 (S.R. & C.A.)† These authors contributed equally to this work. ‡ Present address: Universitatsklinik fur Kardiologie, Inselspital, Freiburgstrasse 4, 3010 Bern, Switzerland. 30 7 2016 8 2016 17 8 123916 6 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Sequencing of sarcomere protein genes in patients fulfilling the clinical diagnostic criteria for hypertrophic cardiomyopathy (HCM) identifies a disease-causing mutation in 35% to 60% of cases. Age at diagnosis and family history may increase the yield of mutations screening. In order to assess whether Next-Generation Sequencing (NGS) may fulfil the molecular diagnostic needs in HCM, we included 17 HCM-related genes in a sequencing panel run on PGM IonTorrent. We selected 70 HCM patients, 35 with early (≤25 years) and 35 with late (≥65 years) diagnosis of disease onset. All samples had a 98.6% average of target regions, with coverage higher than 20× (mean coverage 620×). We identified 41 different mutations (seven of them novel) in nine genes: MYBPC3 (17/41 = 41%); MYH7 (10/41 = 24%); TNNT2, CAV3 and MYH6 (3/41 = 7.5% each); TNNI3 (2/41 = 5%); GLA, MYL2, and MYL3 (1/41=2.5% each). Mutation detection rate was 30/35 (85.7%) in early-onset and 8/35 (22.9%) in late-onset HCM patients, respectively (p < 0.0001). The overall detection rate for patients with positive family history was 84%, and 90.5% in patients with early disease onset. In our study NGS revealed higher mutations yield in patients with early onset and with a family history of HCM. Appropriate patient selection can increase the yield of genetic testing and make diagnostic testing cost-effective. geneticsgene variantshypertrophic cardiomyopathynext-generation sequencing ==== Body 1. Introduction Hypertrophic cardiomyopathy (HCM) is a common genetic cardiac disease that affects one out of 500 individuals from the general population [1]. It is a clinically variable and genetically heterogeneous disease. In fact, more than 20 genes were related with HCM and a total number of about 1400 distinct mutations were identified in affected patients [2]. The most frequently encountered mutations fall within myosin heavy chain 7 (MYH7) and myosin binding protein C (MBPC3) [3,4]. Sequencing of sarcomere protein genes in patients fulfilling clinical diagnostic criteria identifies a disease-causing mutation in only 35% to 60% of cases [5,6,7,8]. Identification of an HCM-causing mutation is an important step in the disease’s clinical management, not only to better support the clinical diagnosis in the proband but also to either exclude or confirm the presence of disease-causing mutations in other family members. Considering the extreme genetic heterogeneity of the disease and the cost of genetic testing, several attempts were made to identify the clinical predictors of an underlying mutation [9,10,11]. In a large study of HCM patients genotyped for mutations in nine genes, the presence of a set of five clinical markers, including age at diagnosis <45 years, accounted for an 80% likelihood of positive genetic testing [11]. In addition, more reliable, precise, and possibly not time-consuming molecular diagnostic approaches are needed. In this regard, Next-Generation Sequencing (NGS), which has already been applied for the diagnosis of hereditary cardiovascular conditions as well as of other diseases [12,13,14,15,16], may represent a suitable tool. Targeted gene panels were shown to generate results with analytical quality identical to Sanger sequencing, and to have the advantage of being faster and cheaper with better coverage and sensitivity than that used in more expanded analyses. The purpose of the present study was to analyse the yield of NGS applied to the genetic screening of a well-phenotyped Italian HCM cohort, composed of patients with both early- and late-onset diagnosis, also including patients with positive family history, and to explore the ability of NGS to accomplish the molecular diagnostic needs in clinical practice. 2. Results 2.1. Description of Study Population The clinical characteristics of patients enrolled in the study are shown in Table 1A. The patients were divided into two subgroups of 35 patients each, depending on the age at HCM diagnosis: the early-onset (EO) group with a mean age at diagnosis of 18.6 ± 8.5 years and the late-onset (LO) group with a mean age at diagnosis of 70.4 ± 4.8 years. The number of patients with a positive family history for HCM was significantly higher in the EO group (p = 0.0001) (Table 1B). Thirty-four patients were women and 36 were men. The sex distribution of patients was different in the two subgroups, with more males in the EO group (p = 0.0001). The left atrium size was significantly different in the two groups (p = 0.0001), with LO patients more frequently exhibiting left atrial enlargement. The obstructive form of HCM was less frequently observed in the EO as compared to the LO group (p = 0.03). Evolution of the disease towards end stage (left ventricular ejection fraction <50%) was observed only in the EO group. None of the other clinical features considered in the study was significantly different between the two groups. 2.2. Sequencing The coding region of each of the 17 HCM phenotype causative genes included in the HCM panel was sequenced on Personal Genome Machine (PGM) IonTorrent sequencer. The 17 genes included in the HCM panel used for this analysis are shown in Table 2. Sequencing produced an average of 240,000 reads per patients; the mean read length was 130 bp; the average read depth per sample was 620× with a mean percentage of reads on target of 93.77%; the mean percentage of regions of interest (ROI) covered at least by 20× was 98.6%, and that covered at least by 100× was 94.7%. Details of the sequencing metrics for each patient are reported in Table 3. Two hundred eighty-two variants were identified within the 17 genes analysed: two were ins/del, 175 were intronic, 37 missense, 59 synonimous, five splicing, and four stop mutations. After filtration, 41 variants with a possible clinical effect were selected and confirmed by Sanger sequencing (data not shown). These variants were located in nine of the 17 genes: MYBPC3 (17/41 = 41%); MYH7 (10/41 = 24%); troponin T2 (TNNT2), caveolin 3 (CAV3), and myosin heavy chain 6 (MYH6) (3/41 = 7.5% each); troponin I 3 (TNNI3) (2/41 = 4.8%); and galactosidase alpha (GLA), myosin light chain 2 (MYL2), and myosin light chain 3 (MYL3) (1/41 = 2.5% each). Thirty-four were known variants, whereas seven were novel. Out of the seven new missense mutations, four had uncertain significance, two were likely pathogenic, and one was likely benign. Considering the 34 known variants, 15 were known to have pathogenic effect, six were likely pathogenic, one was likely benign, and 12 were known registered variants but with unknown clinical significance (Table 4). Mutations in sarcomeric genes accounted for 90% of all identified mutations, with MYBPC3 and MYH7 alone accounting for 65% of all mutations. Considering only mutations in MYBPC3, eight missense mutations and nine truncating mutations were identified (Table 4). 2.3. Group Comparison after Sequencing The mutation detection rate was 85.7% (30/35) in the EO group and 22.9% (8/35) in the LO group. The number of patients in which the molecular screening allowed the identification of at least one mutation was significantly different in the two groups of patients with different age at diagnosis (p < 0.0001). The overall detection rate, regardless of the age of onset, was 54.3% (38/70). The NGS analysis confirmed the known mutational status of the 22 controls (seven positive and 15 negative) included in this study. Mutations identified in each patient are listed in Table 5. Considering only patients with positive family history, the detection rate was 88% (22/25), ranging from 75% (3/4) in the LO group to 90.5% (19/21) in the EO group. Considering sporadic cases only, the overall detection rate was 35.5%, with a significant difference between EO (11/14, 78.6%) and LO (5/31, 16%), p < 0.0002. In the EO group, patients EO13 and EO33 carried three different mutations in MYBPC3. One of them was clinically characterized by an unfavourable course with evolution to end stage disease. Four patients carried two different mutations: EO23 carried two mutations in MYH7, whereas EO6, EO11, and EO21 carried two mutations in two different genes (Table 5). In the LO group, only two patients, LO8 and LO17, harboured two mutations in different genes (Table 5). The distribution of the identified gene mutations was similar between the two groups with the exceptions of MYH6 and TNNT2. In fact, mutations in MYH6 were identified in the LO group only, whereas mutations in TNNT2 were identified in the EO group only. 3. Discussion This report describes the results of a genetic screening obtained through NGS approach in an Italian population of unrelated and clinically well characterized HCM cases, divided into two groups according to age at diagnosis. Our population included a good percentage of patients with a family history of HCM. As expected, the prevalence of familial forms was higher in the EO group, whereas the prevalence of sporadic forms was higher in the LO group. The key finding of our investigation was the higher yield of mutation detection rate in the EO group and in patients with a family history of disease, with 90.5% of cases carrying an identified mutation. The overall yield of genetic testing was close to 50%, and, as previously reported in the literature [4,7,8,9,11], mutations in MYBPC3 and MYH7 accounted for about 65% of all variants. Other mutations were found in six additional sarcomeric genes (TNNT2, CAV3, MYH6, TNNI3, MYL2, and MYL3) and in one non-sarcomeric gene (GLA). Approximately a quarter of all variants were novel, most of them belonging to MYH7. The pathogenicity of novel mutations was verified through appropriate software for analysis. HCM is a disease characterized by a relevant heterogeneity of both morphological and clinical features. For this reason, despite the growing knowledge on its genetic basis, the establishment of a more precise genotype–phenotype correlation has been difficult to achieve. The main original aspect of our investigation was to test through NGS a wide range of HCM-causing genes (14 sarcomeric and three non-sarcomeric) while comparing the extreme ages of disease onset and evaluating the impact of familial occurrence of the disease even in patients with late diagnosis. Due to the small sample size of the population, our study could not address the issue of a relationship between genetic variants and phenotypic characteristics of different HCM onset patients. Notably, the presence of double and triple mutations was detected mostly among younger patients, and one of them showed a more severe form of the disease. The different rate of pathogenic mutations found in HCM patients with early and late onset of the disease was consistent with the literature [17,18,19], confirming that some mutations can be found mainly in young HCM patients (TNNT2) whereas other mutations are detected exclusively in the elderly (MYH6) [17,18,19]. In our study, a majority of patients with young age at diagnosis had a positive genetic testing (80% of cases), four-fold higher than that of the elderly and sporadic HCM cases. These data, together with previous observations, reinforce the concept that age at HCM diagnosis is a powerful predictor of positive genetic testing [11,17,18,19]. We also support the notion that family history of HCM has a key role in appropriately addressing the genetic test. In fact, among HCM patients with a late diagnosis, those with a family history of the disease had a higher rate of mutation detection (75%). We used an expanded panel of 17 genes in the attempt to improve the mutation detection rate. With this approach we mostly confirmed the type of mutations and the mutation distribution already described in the literature for HCM. In particular, the most frequent sarcomeric gene mutations, namely those in MYBPC3 and MYH7, accounted for the majority of the positive findings. Moreover, six of the seven novel mutations identified in our patients were in the main sarcomeric genes (three in MYH7, one in MYH6, one in MYBPC3, and one in TNNI3). In this regard, the limitations of using a wide diagnostic panel for HCM genetic testing have been recently highlighted in one of the largest clinical genetic studies ever reported for HCM [20]. Consistently, a panel designed only for the main HCM genes (n = 9), was able to successfully screen a large cohort of HCM patients [21]. Our findings support the choice of a limited, well-selected panel of HCM genes as the best tool for diagnostic purposes. 4. Materials and Methods 4.1. Patient Selection Seventy patients with clinical diagnosis of HCM were included in the study. We selected 35 patients with early diagnosis of the disease (≤25 years, EO-early onset) and 35 patients with a late diagnosis (≥65 years, LO-late onset). All patients underwent a cardiologic evaluation as well as genetic counselling. Clinical data for each patient included a detailed personal and family history and a thorough scrutiny of the age at which HCM was first diagnosed. Both electrocardiographic and echocardiographic examinations were performed at the time of inclusion into the study. The echocardiographic parameters included both structural measurements and resting LV outflow tract gradients derived from the continuous-wave Doppler velocities. The clinical diagnosis of HCM was based on the echocardiographic demonstration of a hypertrophied and not dilated left ventricle (wall thickness >15 mm in adults, or the equivalent wall thickness relative to body surface area in children) in the absence of another cardiac or systemic disease that could produce comparable left ventricular hypertrophy [22,23]. The mutational status for MYH7, MYBPC3, TNNI3, TNNT2, TPM1, and MYL2 genes was already known in 22/70 patients (8 EO and 14 LO patients). All coding exons (±20 bp) of the six genes were previously analysed by Sanger sequencing. The 22 samples were included in our study as positive and negative controls for the six genes also present in our NGS panel. The seven positive controls carried mutations in MYBPC3 (EO7, EO29, EO33, EO35), MYH7 (LO13), TNNI3 (EO30), and MYL2 (EO20). The 15 negative controls for the six genes were: EO10, EO15, LO5, LO6, LO9, LO12, LO19, LO21, LO22, LO25, LO27, LO28, LO29, LO32, and LO33. This study was conducted in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards (The approval identification number: 42 of 28 September 2007). A signed informed consent for blood sampling was obtained from all patients included in the study. 4.2. DNA Extraction and Quantification Genomic DNA was extracted from peripheral whole blood using a commercially available kit (Invitrogen, Milan, Italy), and then quantified using Qubitds DNA HS Assay Kit on Qubit 2.0 Fluorometer (Invitrogen). 4.3. Sequencing Seventeen genes known to be causative of HCM phenotype were selected for targeted sequencing (Table 2). A custom panel for coding DNA (+/−25 bp of intronic flanking regions) analysis of selected genes was designed online using Ion AmpliSeq Designer 2.0.3 (https://www.ampliseq.com/browse.action) [24]. The final custom panel was composed of 358 amplicons divided into two primer pools for a total of 61.89 kb of DNA. The panel covered 96.47% of regions of interest (ROI). Libraries were prepared using Ion AmpliSeq Library Kit v2.0 (Life Technologies, Carlsbad, CA, USA), according to the manufacturer’s instructions. One of 16 barcodes of the Ion Xpress Barcode Adapters1-16 Kit (Thermo Fisher Scientific Life Sciences Solutions, Carlsbad, CA, USA) was added to each sample. Libraries were quantified with Qubit dsDNA HS Assay Kit on Qubit 2.0 Fluorometer (Molecular Probes, Eugene, OR, USA) and equimolar amounts of each library were used to prepare template for clonal amplification. Emulsion PCR with Ion PGM Template OT2 200 Kit (Life Technologies, Carlsbad, CA, USA) was performed on OneTouch2 Systems (Life Technologies, Carlsbad, CA, USA). Templates were enriched using Ion OneTouch ES (Life Technologies, Carlsbad, CA, USA) and prepared for 316v2 chip loading (Life Technologies, Carlsbad, CA, USA). Groups from 12 to 16 sample libraries were sequenced on each chip. Sequencing runs were performed on Ion Torrent Personal Genome Machine (PGM, Life Technologies) using Ion PGM Sequencing 200 Kit v2, according to the manufacturer’s instructions. 4.4. Alignment Data analysis was performed using the Torrent Suite Software v.4.0.2. (Life Technologies, Carlsbad, CA, USA). Reads were aligned to human reference genome hg19 from UCSC Genome Browser [25] and to a designed bed file from Ion AmpliSeq Designer results. Alignments were visually verified with Integrative Genomics Viewer IGV v.2.3, Broad Institute [26]. 4.5. Coverage Analysis The average read depth and the percentage of reads that mapped on ROI out of the total number of reads (reads on target) was calculated using Coverage Analysis plug-in (Life Technologies, Carlsbad, CA, USA). For each sample the percentage of ROI covered by at least 100× and 20× using amplicon coverage matrix file was calculated. 4.6. Variant Analysis Variant calling was performed with Variant Caller plug-in configured with germ line-low stringency parameters. Variants were annotated using Ion Reporter 4.0 software (Carlsbad, CA, USA) [27]. Common single nucleotide variants (minor allele frequency MAF>5%, source 1000 Genomes), exonic synonymous variants, and intronic variants were removed from the analysis, while exonic non-synonymous, splice-site, and loss-of-function variants were analysed. The novel variants were analysed by means of three types of prediction software (SIFT, POLYPHEN, and PROVEAN) and classified based on the concordance of the prediction between the three types: “likely pathogenic,” “likely benign” (3/3 concordance), or “uncertain significance” (2/3 concordance). 4.7. Variant Validation The identified variants were validated by Sanger sequencing using standard protocols. Specific primers were designed for the analysis. Polymerase Chain Reaction (PCR) products were directly sequenced by using the BigDye Terminator v3.1 Cycle Sequencing Kit (Life Technologies Corporation, Carlsbad, CA, USA). Sample analysis was performed on an ABI PRISM 3130xl Genetic Analyser (Applied Biosystems, Carlsbad, CA, USA). 4.8. Statistical Analysis Statistical analysis was performed with SPSS statistical software (SPSS Inc., Chicago, IL, USA, version 17.0). Continuous variables are expressed as mean±SD. Comparisons between the two groups were performed using a Student’s t-test. The association between the mutational status and the clinical features of the two patient groups was evaluated using Chi-square and Fisher’s exact tests. A p value was considered statistically significant when <0.05. 5. Conclusions In summary, through NGS, we were able to detect pathogenic mutations responsible for HCM, particularly in patients with early onset of the disease and in those with a family history of HCM. Our findings document the suitability of a novel molecular diagnostic strategy for clinical purposes and the important role of appropriate patient selection in making genetic molecular testing more cost-effective. Acknowledgments This work was supported by a 5% grant (Ricerca Corrente) from the Italian Ministry of Health to Massimo Volpe and Speranza Rubattu. The funding sources had no involvement in the study design, in the collection, analyses, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. Author Contributions Speranza Rubattu and Camillo Autore conceived and designed the study. Beatrice Maria Musumeci, Erika Pagannone, Ermelinda Pennacchini, and Pietro Francia collected the study population. Cristina Bozzao, Maria Piane, Camilla Savio, and Aldo Germani performed the genetic analyses. Speranza Rubattu and Camillo Autore drafted and Luciana Chessa and Massimo Volpe finalized the manuscript. All authors closely interpreted all the results, reviewed, and approved the final version of the paper. Conflicts of Interest The authors declare no conflict of interest. Table 1 (A) Clinical characteristics of HCM patients with early or late onset of disease; (B) Familial vs. sporadic HCM. ijms-17-01239-t001a_Table 1(A) Variables Early-Onset n = 36 Late-Onset n = 35 p Age at diagnosis (years) 18.6 ± 8.5 70.4 ± 4.8 0.0001 Male 27 (77.1) 9 (25.7) 0.0001 LV obstruction 14 (40) 24 (68.6) 0.03 Family history of HCM 21 (60) 4 (11.4) 0.0001 NYHA functional class I 24 (68.6) 4 (11.4) 0.0001 II 9 (25.7) 25 (71.4) III 2 (5.7) 6 (17.1) Unexplained syncope 5 (14.3) 6 (17.1) 1 Non sustained ventricular tachycardia 6 (24) 5 (22.7) 1 Left atrial dimension (mm) 39.3 ± 6.2 45 ± 4.5 0.0001 Maximal LV wall thickness (mm) 21.4 ± 6.2 18.7 ± 2.6 0.02 Late gadolinium enhancement 24/29 (82.8) 9/19 (47.4) 0.01 Atrial fibrillation 11 (31.4) 10 (28.6) 1 End stage disease 4 (11.4) 0 (0) 0.11 Myectomy 2 (5.7) 0 (0) 0.49 ICD implantation 12 (34.3) 2 (5.7) 0.006 Death 0 (0) 1 (2.9) 1 ijms-17-01239-t001b_Table 1(B) Patients All n = 70 EO n = 35 LO n = 35 p Familial HCM 25 (36) 21 (60) 4 (14.4) 0.0001 Sporadic HCM 45 (64) 14 (40) 31 (88.6) 0.0001 In (A): Continuous variables are expressed as mean ± SD. Qualitative variable are expressed as n (%). HCM: hypertrophic cardiomyopathy; NYHA: New York Functional Class; LV: left ventricular; ICD: implantable cardioverter defibrillator; In (B): Variable are expressed as n (%); EO: early-onset; LO: late-onset. ijms-17-01239-t002_Table 2Table 2 Metrics of the 17 genes included into the HCM panel. #No. Gene Name Ref Seq NCBI Genomic Location (hg19) Description Amplicons Coverage (%) Target (bp) Missed (bp) 1 MYBPC3 NM_000256 chr11:47352958-47374253 myosin binding protein C, cardiac 53 100 5458 105 2 MYH7 NM_000257 chr14:23881948-23904870 myosin, heavy chain 7, cardiac muscle, β 67 98 7746 231 3 TPM1 NM_001018005 chr15:63334838-63364111 tropomyosin 1 α chain isoform 7 23 99.91 2245 2 4 TNNT2 NM_001001430 chr1:201328143-201346805 troponin T type 2, cardiac isoform 1 20 100 2357 0 5 TNNI3 NM_000363 chr19:55663137-55669100 troponin I, cardiac 10 99.9 989 1 6 MYL2 NM_000432 chr12:111348626-111358404 slow cardiac myosin regulatory light chain 2 9 84.8 858 46 7 MYL3 NM_000258 chr3:46899357-46904973 slow skeletal ventricular myosin alkali light 9 94,6 894 136 8 ACTC1 NM_005159 chr15:35080297-35087927 cardiac muscle α actin 1 proprotein 13 100 1440 0 9 LAMP2 NM_002294 chrX:119560004-119603204 lysosomal-associated membrane protein 2 isoform 21 100 2077 0 10 PRKAG2 NM_016203 chr7:151253203-151574316 AMP-activated protein kinase γ 2 subunit 26 84.3 2713 426 11 GLA NM_000169 chrX:100652779-100663001 α-galactosidase A precursor 14 100 1647 0 12 MYH6 NM_002471 chr14:23851199-23877482 myosin heavy chain 6 66 94.52 7707 422 13 TNNC1 NM_003280 chr3:52485108-52488057 troponin C, slow 8 98.2 792 14 14 CSRP3 NM_003476 chr11:19203578-19223589 cysteine and glycine-rich protein 3 8 100 840 0 15 PLN NM_002667 chr6:118869442-118881586 phospholamban 2 100 210 0 16 TCAP NM_003673 chr17:37821599-37822806 telethonin 5 100 606 0 17 CAV3 NM_033337 chr3:8775486-8788451 Homo sapiens caveolin 3 (CAV3), transcript variant 1, mRNA. 4 100 558 0 Gene symbols: TPM1: tropomyosin 1; ACTC1: actin, α, cardiac muscle 1; LAMP2: lysosomal associated membrane protein 2; PRKAG2: protein kinase AMP-activated non-catalytic subunit γ 2; TNNC1: troponin C 1; CSRP3: cystein and glycine-rich protein 3; PLN: phospholamban; TCAP: telethonin. ijms-17-01239-t003_Table 3Table 3 Patient sequencing metrics. Patients Mapped Reads Reads on Target (%) Uniformity (%) ROI MEAN COVERAGE ROI ≥ 20× (%) n of Amplicons < 20× ROI ≥ 100× (%) n of Amplicons < 100× EO1 178,727 92.13 93.95 459.94 98.60 5 94.97 18 EO2 178,731 90.57 94.83 452.19 98.88 4 95.53 16 EO3 72,440 91.78 93.75 185.71 96.93 11 83.52 59 EO4 247,711 90.70 93.90 627.61 99.44 2 96.09 14 EO5 111,232 91.03 93.57 282.82 97.21 10 91.62 30 EO6 280,419 93.08 94.15 729.08 99.16 3 96.09 14 EO7 623,594 92.53 93.81 1611.77 99.44 2 98.32 6 EO8 561,715 97.46 92.18 1529.12 99.44 2 97.49 9 EO9 77,846 93.36 93.83 203.00 96.93 11 86.87 47 EO10 381,796 96.33 93.71 1027.32 99.44 2 97.21 10 EO11 311,658 93.28 93.70 812.08 99.44 2 96.09 14 EO12 239,783 93.00 94.15 622.93 98.88 4 95.53 16 EO13 276,453 93.48 94.44 721.90 99.44 2 96.09 14 EO14 215,672 93.30 94.53 562.09 99.44 2 95.81 15 EO15 465,323 94.73 93.01 1231.34 99.44 2 96.65 12 EO16 465,619 97.25 92.65 1264.84 99.44 2 96.93 11 EO17 441,220 95.42 92.96 1176.05 99.72 1 97.49 9 EO18 192,373 98.07 91.50 526.97 98.60 5 94.13 21 EO19 313,968 95.80 93.72 840.19 99.16 3 96.65 12 EO20 192,211 95.35 94.02 517.24 98.60 5 95.81 15 EO21 196,251 95.05 94.02 521.07 98.88 4 95.81 15 EO22 303,435 96.01 93.55 813.79 98.88 4 96.65 12 EO23 322,467 94.14 92.17 847.94 98.88 4 95.81 15 EO24 253,552 95.97 91.35 679.71 98.88 4 95.53 16 EO25 188,696 95.33 93.96 502.45 98.60 5 95.53 16 EO26 182,956 94.99 92.88 485.47 98.88 4 94.13 21 EO27 191,880 94.62 93.58 507.12 98.88 4 94.97 18 EO28 228,313 92.89 93.22 592.43 98.88 4 94.69 19 EO29 199,442 98.07 92.15 546.33 98.32 6 94.97 18 EO30 190,915 97.24 92.66 518.58 98.04 7 94.69 19 EO31 161,793 95.49 92.19 431.54 97.77 8 93.30 24 EO32 245,414 89.57 93.45 613.99 98.32 6 95.81 15 EO33 205,079 95.46 85.54 546.83 96.65 12 89.39 38 EO34 210,900 97.24 93.66 572.83 98.60 5 95.25 17 EO35 147,306 97.24 92.62 402.14 98.32 6 94.13 21 LO1 178,290 90.77 93.68 321.94 97.77 8 91.90 29 LO2 205,008 93.84 93.97 537.36 99.16 3 96.09 14 LO3 159,828 93.15 93.80 502.12 98.88 4 95.53 16 LO4 193,973 93.90 94.09 1062.97 99.44 2 96.37 13 LO5 191,160 93.72 93.21 1097.36 99.16 3 96.65 12 LO6 177,316 94.10 93.42 931.78 99.44 2 96.37 13 LO7 238,812 94.23 93.77 593.70 97.77 8 92.18 28 LO8 158,483 93.67 93.01 708.10 99.44 2 97.49 9 LO9 213,370 93.89 94.34 861.05 99.44 2 97.21 10 LO10 190,285 94.47 93.70 415.86 98.32 6 94.41 20 LO11 182,160 93.99 94.02 505.35 98.32 6 94.97 18 LO12 213,052 92.51 94.69 532.23 98.88 4 94.97 18 LO13 249,591 93.48 93.92 304.45 96.93 11 90.78 33 LO14 195,422 94.82 94.22 378.77 98.04 7 92.74 26 LO15 201,815 95.14 93.81 717.45 98.60 5 95.53 16 LO16 400,156 98.18 90.79 500.41 98.60 5 94.41 20 LO17 274,868 95.75 91.07 423.42 98.04 7 92.74 26 LO18 158,695 92.07 94.09 457.87 98.32 6 94.97 18 LO19 195,752 89.66 93.66 206.68 96.65 12 86.03 50 LO20 83,846 88.25 93.28 490.24 98.32 6 94.13 21 LO21 179,015 91.57 94.00 408.12 98.60 5 94.69 19 LO22 170,180 89.07 93.89 735.17 98.60 5 94.97 18 LO23 161,290 90.82 93.26 680.84 99.16 3 96.09 14 LO24 281,769 92.16 93.07 536.34 98.60 5 95.53 16 LO25 145,219 93.37 93.76 517.60 99.16 3 95.81 15 LO26 390,025 97.57 92.65 508.78 98.88 4 95.53 16 LO27 124,848 87.30 93.29 651.69 99.16 3 95.25 17 LO28 214,031 89.02 93.84 550.53 99.16 3 96.37 13 LO29 203,428 88.93 94.12 479.56 98.88 4 95.53 16 LO30 127,496 90.40 93.91 452.05 98.32 6 94.69 19 LO31 25,524 95.49 93.62 409.17 98.04 7 93.02 25 LO32 329,472 93.56 94.58 559.60 99.16 3 96.09 14 LO33 265,155 95.60 93.94 414.68 98.32 6 94.13 21 LO34 21,736 97.78 89.25 628.59 98.88 4 95.81 15 LO35 352,805 94.55 92.69 466.05 97.77 8 94.69 19 ijms-17-01239-t004_Table 4Table 4 Mutations detected per gene. Gene ID Chrom Position Exon DNA Change Protein Change Mutation Type dbSNP Prev. Rep. GMAF SIFT POLYPHEN PROVEAN (cutoff = −2.5) Clinical Significance CAV3 chr3 8787313 2 c.216C>G Cys72Trp MISSENSE rs116840776 yes 0.00100 (G) deleterious 0 probably damaging 0.999 deleterious −6.167 known/uncertain significance chr3 8787330 2 c.233C>T Thr78Met MISSENSE rs72546668 yes 0.00200 (T) tolerated 0.05 possibly damaging 0.537 neutral −0.833 known/uncertain significance chr3 8787497 2 c.400G>T Ala134Ser MISSENSE deleterious 0.01 benign 0.07 neutral 0.862 new/uncertain significance GLA chrX 100653420 6 c.937G>T Asp313Tyr MISSENSE rs28935490 yes 0.0021 (A) deleterious 0 probably damaging 0.952 deleterious −3.183 known/uncertain significance MYBPC3 chr11 47371426 5 c.553A>T Lys185Ter STOP rs375607980 yes known/pathogenic chr11 47371414 5 c.565G>A Val189Ile MISSENSE rs11570052 yes 0.00200 (T) tolerated 0.44 benign 0.132 Neutral −0.418 known/likely benign chr11 47365154 13 c.1112C>G Pro371Arg MISSENSE rs397515887 yes 0.00020 (A) deleterious 0 probably damaging 0.994 deleterious −8.043 known/uncertain significance chr11 47365147 13 c.1120C>T Gln374Ter STOP rs730880635 yes known/pathogenic chr11 47364429 15 c.1409G>A Arg470Gln MISSENSE yes deleterious 0.01 probably damaging 0.982 deleterious −3.094 known/uncertain significance chr11 47364270 16 c.1483C>T Arg495Trp MISSENSE rs397515905 yes deleterious 0 probably damaging 0.999 deleterious −5.228 known/uncertain significance chr11 47364162 16 c.1591G>C Gly531Arg MISSENSE rs397515912 yes 0.00020 (G) deleterious 0 probably damaging 0.996 deleterious −7.038 known/likely pathogenic chr11 47364129 16 c.1624G>C Glu542Gln MISSENSE/SPLICING rs121909374 yes 0.00008 (G) known/pathogenic chr11 47360071 22 c.2308G>A Asp770Asn MISSENSE/SPLICING rs36211723 yes known/pathogenic chr11 47359347 23 c.2309-2A>G SPLICING rs111729952 yes known/pathogenic chr11 47359115 24 c.2429G>A Arg810His MISSENSE rs375675796 yes 0.00008 (T) deleterious 0 probably damaging 1 deleterious −4.564 known/likely pathogenic chr11 47359085 24 c.2459G>A Arg820Gln MISSENSE rs2856655 yes deleterious 0 probably damaging 0.98 deleterious −2.925 known/likely pathogenic chr11 47356592 26 c.2905+1G>A SPLICING rs397515991 yes known/pathogenic chr11 47355264 28 c.3034C>T Gln1012Ter STOP rs730880586 yes known/pathogenic chr11 47354882 29 c.3192_3193insC Lys1065Glnfs INS rs397516007 yes known/pathogenic chr11 47353801 32 c.3636T>G Ile1212Met MISSENSE deleterious 0 probably damaging 0.918 deleterious −2.498 new/likely pathogenic chr11 47353662 32 c.3775C>T Gln1259Ter STOP rs730880605 yes known/pathogenic MYH7 chr14 23900850 8 c.676G>A Ala226Thr MISSENSE deleterious 0 probably damaging 0.985 neutral −1.757 new/uncertain significance chr14 23896866 16 c.1816G>A Val606Met MISSENSE rs121913627 yes known/pathogenic chr14 23896042 18 c.1988G>A Arg663His MISSENSE rs371898076 yes 0.00008 (T) known/pathogenic chr14 23895189 19 c.2146G>C Gly716Arg MISSENSE rs121913638 yes deleterious 0.01 probably damaging 0.995 deleterious −3.728 known/likely pathogenic chr14 23895179 19 c.2156G>A Arg719Gln MISSENSE rs121913641 yes known/pathogenic chr14 23894116 22 c.2543_2545 delAAG Lys847del DEL yes known/pathogenic chr14 23893234 23 c.2804A>T Glu935Val MISSENSE rs730880761 yes known/pathogenic chr14 23891501 25 c.3133C>T Arg1045Cys MISSENSE rs45611033 yes 0.00020 (A) deleterious 0.03 benign 0.203 deleterious −6.180 known/uncertain significance chr14 23889413 27 c.3367G>C Glu1123Gln MISSENSE deleterious 0.01 probably damaging 0.968 neutral −2.389 new/uncertain significance chr14 23887615 30 c.3973G>A Ala1325Thr MISSENSE/SPLICING deleterious 0.02 possibly damaging 0.751 neutral −2.329 new/uncertain significance TNNT2 chr1 201334751 9 c.281G>C Arg94Thr MISSENSE rs397516452 yes deleterious 0 possibly damaging 0.573 deleterious −5.588 known/uncertain significance chr1 201330414 14 c.794A>T Lys265Ile MISSENSE rs397516482 yes deleterious 0 probably damaging 0.958 deleterious −6.86 known/uncertain significance chr1 201328373 16 c.853C>T Arg285Cys MISSENSE rs121964857 yes tolerated 0.06 probably damaging 0.978 neutral −2.09 known/likely pathogenic MYH6 chr14 23873951 7 c.611G>A Arg204His MISSENSE rs200623022 yes tolerated 0.05 possibly damaging 0.807 neutral −1.327 known/uncertain significance chr14 23865497 20 c.2425C>T Arg809Cys MISSENSE deleterious 0 probably damaging 0.963 deleterious −5.294 new/likely pathogenic chr14 23853697 36 c.5519A>G Lys1840Arg MISSENSE rs373629059 tolerated 0.13 probably damaging 0.999 neutral −1.731 known/uncertain significance MYL2 chr12 111350901 6 c.401A>C Glu134Ala MISSENSE rs143139258 yes deleterious 0.01 possibly damaging 0.755 Deleterious −5.696 known/likely pathogenic MYL3 chr3 46902303 3 c.170C>A Ala57Asp MISSENSE rs139794067 yes deleterious 0 probably damaging 0.996 deleterious −5.236 known/uncertain significance TNNI3 chr19 55665561 6 c.385C>G Thr128Ser MISSENSE tolerated 0.186 benign 0.000 neutral 0.61 new/likely benign chr19 55665516 6 c.431T>A Leu144Gln MISSENSE rs121917760 yes known/pathogenic Prev. Rep.: previously reported; GMAF: Global minor allele frequency; Software prediction programs used for sequence variant interpretation: SIFT: Evolutionary conservation; POLYPHEN: Protein structure/function and evolutionary conservation; PROVEAN: Alignment and measurement of similarity between variant sequence and protein sequence homolog. ijms-17-01239-t005_Table 5Table 5 Mutations detected per patient. Early-Onset Patient ID Familiarity Gene ID Exon DNA Change Protein Change Mutation Type Clinical Significance dbSNP Previously Reported Coverage Allele Coverage EO1 yes MYBPC3 5 c.553A>T Lys185Ter STOP known/pathogenic rs375607980 yes 384 202 EO2 yes MYH7 19 c.2156G>A Arg719Gln MISSENSE known/pathogenic rs121913641 yes 399 204 EO3 CAV3 2 c.233C>T Thr78Met MISSENSE known/uncertain significance rs72546668 yes 124 57 EO4 MYBPC3 23 c.2309-2A>G SPLICING known/pathogenic rs111729952 yes 400 186 EO5 MYH7 16 c.1816G>A Val606Met MISSENSE known/pathogenic rs121913627 yes 383 204 EO6 yes MYH7 8 c.676G>A Ala226Thr MISSENSE new/uncertain significance 399 208 GLA 6 c.937G>T Asp313Tyr MISSENSE known/uncertain significance rs28935490 yes 399 183 EO7 yes MYBPC3 28 c.3034C>T Gln1012Ter STOP known/pathogenic rs730880586 yes 397 194 EO8 yes MYBPC3 23 c.2309-2A>G SPLICING known/pathogenic rs111729952 yes 398 204 EO9 yes MYH7 19 c.2146G>C Gly716Arg MISSENSE known/likely pathogenic rs121913638 yes 354 169 EO11 yes MYBPC3 16 c.1483C>T Arg495Trp MISSENSE known/uncertain significance rs397515905 yes 400 259 CAV3 2 c.216C>G Cys72Trp MISSENSE known/uncertain significance rs116840776 yes 400 182 EO12 yes MYBPC3 23 c.2309-2A>G SPLICING known/pathogenic rs111729952 yes 399 185 EO13 MYBPC3 32 c.3636T>G Ile1212Met MISSENSE new/likely pathogenic 400 201 23 c.2309-2A>G SPLICING known/pathogenic rs111729952 yes 399 187 16 c.1591G>C Gly531Arg MISSENSE known/likely pathogenic rs397515912 yes 400 184 EO14 CAV3 2 c.233C>T Thr78Met MISSENSE known/uncertain significance rs72546668 yes 399 214 EO17 MYBPC3 22 c.2308G>A Asp770Asn MISSENSE/SPLICING known/pathogenic rs36211723 yes 399 195 EO18 yes MYBPC3 13 c.1120C>G Tyr374Ter STOP known/pathogenic rs730880635 yes 400 225 EO19 yes MYBPC3 32 c.3775C>T Gln1259Ter STOP known/pathogenic rs730880605 yes 398 204 EO20 yes MYL2 6 c.401A>C Glu134Ala MISSENSE known/likely pathogenic rs143139258 yes 398 191 EO21 yes MYBPC3 5 c.565G>A Val189Ile MISSENSE known/uncertain significance rs11570052 yes 309 253 MYH7 22 c.2543_2545 delAAG Lys847del DELETION known/pathogenic yes 391 194 EO22 CAV3 2 c.400G>T Ala134Ser MISSENSE new/uncertain significance 330 168 EO23 MYH7 30 c.3973G>A Ala1325Thr MISSENSE/SPLICING new/uncertain significance 400 176 23 c.2804A>T Glu935Val MISSENSE known/pathogenic rs730880761 yes 400 206 EO25 yes MYBPC3 15 c.1409G>A Arg470Gln MISSENSE known/uncertain significance yes 293 130 EO26 yes TNNT2 16 c.853C>T Arg285Cys MISSENSE known/likely pathogenic rs121964857 yes 323 167 EO27 yes TNNT2 14 c.794A>T Lys265Ile MISSENSE known/uncertain significance rs397516482 yes 395 193 EO29 yes MYBPC3 24 c.2429G>A Arg810His MISSENSE known/likely pathogenic rs375675796 yes 400 148 EO30 yes TNNI3 6 c.431T>A Leu144Gln MISSENSE known/pathogenic rs121917760 yes 398 227 EO31 MYBPC3 5 c.565G>A Val189Ile MISSENSE known/likely benign rs11570052 yes 312 152 EO32 MYH7 18 c.1988G>A Arg663His MISSENSE known/pathogenic rs371898076 yes 400 211 EO33 MYBPC3 29 c.3193_3194 insC Lys1065Glnfs INSERTION known/pathogenic rs397516007 yes 398 194 16 c.1591G>C Gly531Arg MISSENSE known/likely pathogenic rs397515912 yes 400 212 13 c.1112C>G Pro371Arg MISSENSE known/uncertain significance rs397515887 yes 235 87 EO34 yes TNNT2 9 c.281G>C Arg94Thr MISSENSE known/uncertain significance rs397516452 yes 400 196 EO35 MYBPC3 26 c.2905+1G>A ex26 SPLICING known/pathogenic rs397515991 Yes 296 139 Late-Onset Patient ID Familiarity Gene ID Exon DNA Change Protein Change Mutation Type Clinical Significance dbSNP Previously Reported Coverage Allele Coverage LO1 Yes MYH7 25 c.3133C>T Arg1045Cys MISSENSE known/uncertain significance rs45611033 yes 213 113 LO4 MYH7 27 c.3367G>C Glu1123Gln MISSENSE new/uncertain significance 400 220 LO6 MYH6 20 c.2425C>T Arg809Cys MISSENSE new/ likely pathogenic 299 139 LO8 Yes MYBPC3 16 c.1624G>C Glu542Gln MISSENSE/SPLICING known/pathogenic rs121909374 yes 353 188 TNNI3 6 c.385C>G Thr128Ser MISSENSE new/likely benign 400 204 LO13 Yes MYH7 16 c.1816G>A Val606Met MISSENSE known/pathogenic rs121913627 yes 383 204 LO14 MYH6 36 c.5519A>G Lys1840Arg MISSENSE known/uncertain significance rs373629059 yes 399 196 LO16 MYBPC3 24 c.2459G>A Arg820Gln MISSENSE known/likely pathogenic rs2856655 yes 400 213 LO17 MYH6 7 c.611G>A Arg204His MISSENSE known/uncertain significance rs200623022 yes 398 201 MYL3 3 c.170C>A Ala57Asp MISSENSE known/uncertain significance rs139794067 yes 398 168 dbSNP: database single nucleotide polymorphisms (www.ncbi.nlm.nih.gov/SNP). ==== Refs References 1. Maron B.J. Gardin J.M. Flack J.M. Gidding S.S. Kurosaki T.T. Bild D.E. Prevalence of hypertrophic cardiomyopathy in a general population of young adults. Echocardiographic analysis of 4111 subjects in the cardia study. Coronary artery risk development in (young) adults Circulation 1995 92 785 789 10.1161/01.CIR.92.4.785 7641357 2. Ho C.Y. Charron P. Richiard P. Girolami F. van Spaendonck-Zwarts K.Y. Pinto Y. Genetic advances in sarcomeric cardiomyopathies: State of the art Cardiovasc. Res. 2015 105 397 408 10.1093/cvr/cvv025 25634555 3. Maron B.J. Maron M.S. Semsarian C. Genetics of hypertrophic cardiomyopathy after 20 years: Clinical perspectives J. Am. Coll. Cardiol. 2012 60 705 715 10.1016/j.jacc.2012.02.068 22796258 4. Richard P. Charron P. Carrier L. Ledeuil C. Cheav T. Pichereau C. Benaiche A. Isnard R. Dubourg O. Burban M. Hypertrophic cardiomyopathy: Distribution of disease genes, spectrum of mutations, and implications for a molecular diagnostic strategy Circulation 2003 107 2227 2232 10.1161/01.CIR.0000066323.15244.54 12707239 5. Olivotto I. Girolami F. Ackerman M.J. Nistri S. Bos J.M. Zachara E. Ommen S.R. Theis J.L. Vaubel R.A. Re F. Myofilament protein gene mutation screening and outcome of patients with hypertrophic cardiomyopathy Mayo Clin. Proc. 2008 83 630 638 10.1016/S0025-6196(11)60890-2 18533079 6. Andersen P.S. Havndrup O. Hougs L. Sørensen K.M. Jensen M. Larsen L.A. Hedley P. Thomsen A.R. Moolman-Smook J. Christiansen M. Diagnostic yield, interpretation, and clinical utility of mutation screening of sarcomere encoding genes in Danish hypertrophic cardiomyopathy patients and relatives Hum. Mutat. 2009 30 363 370 10.1002/humu.20862 19035361 7. Brito D. Miltenberger-Miltenyi G. Vale Pereira S. Silva D. Diogo A.N. Madeira H. Sarcomeric hypertrophic cardiomyopathy: Genetic profile in a Portuguese population Rev. Port. Cardiol. 2012 31 577 587 22857948 8. Lopes L.R. Zekavati A. Syrris P. Hubank M. Giambartolomei C. Dalageorgou C. Jenkins S. McKenna W. Plagnol V. Elliott P.M. Genetic complexity in hypertrophic cardiomyopathy revealed by high-throughput sequencing J. Med. Genet. 2013 50 228 239 10.1136/jmedgenet-2012-101270 23396983 9. Ingles J. Sarina T. Yeates L. Hunt L. Macciocca I. McCormack L. Winship I. McGaughran J. Atherton J. Semsarian C. Clinical predictors of genetic testing outcomes in hypertrophic cardiomyopathy Genet. Med. 2013 15 972 977 10.1038/gim.2013.44 23598715 10. Gruner C. Ivanov J. Care M. Williams L. Moravsky G. Yang H. Laczay B. Siminovitch K. Woo A. Rakowski H. The Toronto HCM Genotype score for prediction of a positive genotype in hypertrophic cardiomyopathy Circ. Cardiovasc. Genet. 2013 6 19 26 10.1161/CIRCGENETICS.112.963363 23239831 11. Bos J.M. Will M.L. Gersh B.J. Kruisselbrink T.M. Ommen S.R. Ackerman M.J. Characterization of a phenotype-based genetic test prediction score for unrelated patients with hypertrophic cardiomyopathy Mayo Clin. Proc. 2014 89 727 737 10.1016/j.mayocp.2014.01.025 24793961 12. Voelkerding K.V. Dames S.A. Durtschi J.D. Next-generation sequencing: From basic research to diagnostics Clin. Chem. 2009 55 641 658 10.1373/clinchem.2008.112789 19246620 13. Millat G. Chanavat V. Rousson R. Evaluation of a new NGS method based on a custom ampliseq library and ion torrent PGM sequencing for the fast detection of genetic variation in cardiomyopathies Clin. Chim. Acta 2014 433 266 271 10.1016/j.cca.2014.03.032 24721642 14. Sikkema-Raddatz B. Johansson L.F. de Boer E.N. Almomani R. Boven L.G. van den Berg M.P. van Spaendonck-Zwarts K.Y. van Tintelen J.P. Sijmons R.H. Jongbloed J.D. Targeted next-generation sequencing can replace sanger sequencing in clinical diagnostics Hum. Mutat. 2013 34 1035 1042 10.1002/humu.22332 23568810 15. Meder B. Haas J. Keller A. Heid C. Just S. Borries A. Boisguerin V. Scharfenberger-Schmeer M. Stahler P. Beier M. Targeted next generation sequencing for the molecular genetic diagnostics of cardiomyopathies Circ. Cardiovasc. Genet. 2011 4 110 122 10.1161/CIRCGENETICS.110.958322 21252143 16. Mogensen J. van Tintelen J.P. Fokstuen S. Elliott P. van Langen I.M. Meder B. Richard P. Syrris P. Caforio A.L.P. Adler Y. The current role of next-generation DNA sequencing in routine care of patients with hereditary cardiovascular conditions: A viewpoint paper of the European Society of Cardiology working group on myocardial and pericardial diseases and members of the European Society of Human Genetics Eur. Heart J. 2015 36 1367 1370 25845928 17. Niimura H. Patton K.K. McKenna W.J. Soults J. Maron B.J. Seidman J.G. Seidman C.E. Sarcomere protein gene mutations in hypertrophic cardiomyopathy in the elderly Circulation 2002 105 446 451 10.1161/hc0402.102990 11815426 18. Binder J. Ommen S.R. Gersh B.J. van Driest S.L. Tajik A.J. Nishimura R.A. Ackerman M.J. Echocardiography-guided genetic testing in hypertrophic cardiomyopathy: Septal morphological features predict the presence of myofilament mutations Mayo Clin. Proc. 2006 81 459 467 10.4065/81.4.459 16610565 19. Anan R. Niimura H. Takenaka T. Hamasaki S. Tei C. Mutations in the genes for sarcomeric proteins in Japanese patients with onset sporadic hypertrophic cardiomyopathy after age of 40 years Am. J. Cardiol. 2007 99 1750 1754 10.1016/j.amjcard.2007.01.066 17560888 20. Alfares A.A. Kelly M.A. McDermott G. Funke B.H. Lebo M.S. Baxter S.B. Shen J. McLaughlin H.M. Clark E.H. Babb L.J. Results of clinical genetic testing of 2,912 probands with hypertrophic cardiomyopathy: Expanded panels offer limited additional sensitivity Genet. Med. 2015 17 880 888 10.1038/gim.2014.205 25611685 21. Gómez J. Reguero J.R. Morís C. Martín M. Alvarez V. Alonso B. Iglesias S. Coto E. Mutation analysis of the main hypertrophic cardiomyopathy genes using multiplex amplification and semiconductor next-generation sequencing Circ. J. 2014 78 2963 2971 10.1253/circj.CJ-14-0628 25342278 22. Spirito P. Autore C. Rapezzi C. Bernabò P. Badagliacca R. Maron M.S. Bongioanni S. Coccolo F. Mark Estes N.A. Barillà C.S. Syncope and risk of sudden death in hypertrophic cardiomyopathy Circulation 2009 119 1703 1710 10.1161/CIRCULATIONAHA.108.798314 19307481 23. Elliott P.M. Anastasakis A. Borger M.A. Borggrefe M. Cecchi F. Charron P. Hagege A.A. Lafont A. Limongelli G. Mahrholdt H. 2014 ESC guidelines on diagnosis and management of hypertrophic cardiomyopathy: The task force for the diagnosis and management of hypertrophic cardiomyopathy of the european society of cardiology (ESC) Eur. Heart J. 2014 35 2733 2779 10.1016/j.rec.2014.12.001 25173338 24. Ion AmpliSeqDesigner Available online: http://www.ampliseq.com/ (accessed on 3 July 2013) 25. UCSC Genome Browser Available online: http://genome.ucsc.edu/ (accessed on 24 January 2014) 26. Broad Institute Available online: http://www.broadinstitute.org/igv/home (accessed on 24 January 2014) 27. Ion Reporter Software Available online: https://ionreporter.lifetechnologies.com/ir/ (accessed on 24 January 2014)

PMC005xxxxxx/PMC5000638.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081240ijms-17-01240ArticleCoordination Environment of Cu(II) Ions Bound to N-Terminal Peptide Fragments of Angiogenin Protein Magrì Antonio 1Munzone Alessia 2Peana Massimiliano 3Medici Serenella 3Zoroddu Maria Antonietta 3Hansson Orjan 4Satriano Cristina 2*Rizzarelli Enrico 12La Mendola Diego 5*Maki Masatoshi Academic Editor1 Institute of Biostructures and Bioimages, National Council of Research ( CNR), Via P. Gaifami 18, 95126 Catania, Italy; [email protected] (A.M.); [email protected] (E.R.)2 Department of Chemical Sciences, University of Catania, Viale A. Doria 6, 95125 Catania, Italy; [email protected] Department of Chemistry and Pharmacy, University of Sassari, Via Vienna 2, 07100 Sassari, Italy; [email protected] (M.P.); [email protected] (S.M.); [email protected] (M.A.Z.)4 Department of Chemistry and Molecular Biology, University of Gothenburg, Medicinaregatan 9C, 41390 Göteborg, Sweden; [email protected] Department of Pharmacy, University of Pisa, Via Bonanno Pisano 6, 56126 Pisa, Italy* Correspondence: [email protected] (C.S.); [email protected] (D.L.M.); Tel.: +39-095-7385136 (C.S.); +39-050-2219533 (D.L.M.)01 8 2016 8 2016 17 8 124001 5 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Angiogenin (Ang) is a potent angiogenic factor, strongly overexpressed in patients affected by different types of cancers. The specific Ang cellular receptors have not been identified, but it is known that Ang–actin interaction induces changes both in the cell cytoskeleton and in the extracellular matrix. Most in vitro studies use the recombinant form (r-Ang) instead of the form that is normally present in vivo (“wild-type”, wt-Ang). The first residue of r-Ang is a methionine, with a free amino group, whereas wt-Ang has a glutamic acid, whose amino group spontaneously cyclizes in the pyro-glutamate form. The Ang biological activity is influenced by copper ions. To elucidate the role of such a free amino group on the protein–copper binding, we scrutinized the copper(II) complexes with the peptide fragments Ang(1–17) and AcAng(1–17), which encompass the sequence 1–17 of angiogenin (QDNSRYTHFLTQHYDAK-NH2), with free amino and acetylated N-terminus, respectively. Potentiometric, ultraviolet-visible (UV-vis), nuclear magnetic resonance (NMR) and circular dichroism (CD) studies demonstrate that the two peptides show a different metal coordination environment. Confocal microscopy imaging of neuroblastoma cells with the actin staining supports the spectroscopic results, with the finding of different responses in the cytoskeleton organization upon the interaction, in the presence or not of copper ions, with the free amino and the acetylated N-terminus peptides. copperangiogenesisrecombinant proteinpeptidomimeticconfocal microscopyactinstability constantsNMRneuroblastoma cells ==== Body 1. Introduction Human Angiogenin (Ang) is a 14 kDa protein belonging to the ribonucleases family, with a RNase catalytic activity about 106 lower than pancreatic RNase [1]. Ang is a physiological constituent of human plasma (in a concentration range of 250–360 µg/L) but is over-expressed in patients affected by different types of cancers. Indeed, Ang has been discovered and isolated for the first time from a colon adenocarcinoma cell line [2,3]. The protein influences nearly all steps of tumorigenesis, including cell proliferation, migration and metastatic progression [3,4,5]. Ang is among the most potent angiogenic factors known thus far, as determined in various models, both in vitro and in vivo [6,7]. Moreover, Ang regulates the expression of other angiogenic factors, such as the vascular endothelial growth factor (VEGF), the epidermal growth factor (EGF), the acidic fibroblast growth factor (aFGF) and the basic fibroblast growth factor (bFGF) [8,9,10,11,12]. The angiogenic activity of this protein occurs through a series of events, involving: (i) the Ang ribonucleolytic activity; (ii) the stimulation of basement membrane degradation; (iii) the stimulation of signal transduction; and (iv) the nuclear translocation [13,14]. From a molecular point of view, three distinct regions of the protein are necessary: (a) the catalytic site for RNase activity that involve the residues His-13, Lys-40 and His-114; (b) the putative cell binding region, encompassing the residues 60–68 (KNGNPHREN sequence); and (c) the nuclear translocation residues 31–35 (RRRGL sequence) [5,13,14]. The angiogenin functions are also exerted extracellularly, where the protein activates signal-related kinase1/2 (ERK1/2) in human umbilical vein endothelial cells (HUVECs) or stress-associated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) in human umbilical artery smooth muscle cells (HuASMCs) [15,16]. The binding of Ang to a putative 170 kDa receptor has been found in HUVEC [17], but a possible receptor role has been suggested also for other proteins, including actinin [18], follistatin [19] fibulin 1 [20] and actin [21,22,23]. In particular, the formation of a high-affinity complex between Ang and actin has been reported, although the actin-binding site on Ang is still unknown [22,23]. The interaction of actin with Ang induces changes in the cell cytoskeleton, remodeling of the extracellular matrix (ECM) and the degradation of basement membrane, thus promoting cell invasion into the perivascular tissue [22,23]. A possible activity of Ang towards actin aggregation has been indicated and related to the tendency of the protein to form multivalent intermolecular interactions. Interestingly, the Ang residues involved in such processes are localized in the N- and/or C-terminal domains of the protein [22]. It has to be noted that Ang’s functional role is not limited to the angiogenesis stimulation, since the protein is widely expressed in all mammalian organs and tissues [5,12]. For instance, the protein exists at high concentrations in motoneurons, where it is involved in the onset of amyotrophic lateral sclerosis (ALS) [24,25]. In addition, Ang is down-regulated in patients affected by Alzheimer’s diseases [26] and in α-synuclein mouse model of Parkinson’s disease [27]. Copper is also an angiogenic factor in vivo [28,29,30] and plays a role in the progression of different cancers as wells as in the onset/progression of neurodegenerative diseases [31,32,33,34]. Moreover, during angiogenesis process, copper translocates from intracellular to extracellular space [35]. In this pathway, still unknown, it is very likely that copper acts as signaling factor, by means of the binding with extracellular proteins involved in angiogenesis, such as Ang. Furthermore, the binding affinity between human angiogenin and endothelial cells is largely increased in the presence of copper ions [36,37]. Different peptide fragments encompassing either the putative binding site or the N-terminal domain of angiogenin have been demonstrated to be able to bind copper ions tightly [38,39]. Previous studies relied on the hypothesis that the angiogenic activity of copper and Ang occurred through different and independent biological pathways [40]. Recently, a strong correlation between the protein and the metal ion has been demonstrated: Cu2+ increases the expression of Ang and modulates its intracellular localization in HUVEC, affecting angiogenic activity of protein and ERK activation signaling [41]. This contrast is likely related to the use of different forms of the Ang protein: the recombinant (r-Ang), mostly used in many literature reports, containing an extra methionine as first residue, and the native form, present in vivo (“wild-type”-angiogenin, wt-Ang), which lacks the free amino terminal group, owing to the spontaneous cyclization of the first glutamine group residue to pyroglutamate [1]. Noteworthy, the two proteins, r-Ang and wt-Ang, bind copper ions differently: in the recombinant form, the anchoring site of metal ion is the terminal amino group, whereas the native wild-type protein binds Cu2+ through His-114 and His-13 [42]. Such amino acids constitute two out of the three catalytic residues of the protein, and the addition of copper ions at physiological pH influences much more the RNase activity and the capillary-like tubes formation in wt-Ang than in r-Ang [42]. A comprehensive characterization of the Ang–copper(II) complex species is therefore a valuable support in the improved understanding of potential mutual biological influences. In this work, we report on the copper(II) complexes formed with the peptide fragment encompassing the sequence 1–17 (QDNSRYTHFLTQHYDAK-NH2; named Ang(1–17)) and its acetylated form (Ac-QDNSRYTHFLTQHYDAK-NH2; named AcAng(1–17)). A comprehensive chemical characterization by means of potentiometry, nuclear magnetic resonance (NMR), ultraviolet visible (UV-vis) and circular dichroism (CD) spectroscopies is presented to depict the copper coordination environment in the N-terminal domain of the protein. The obtained data demonstrate the role of N-terminal free amino group in the metal binding. Moreover, the effect of Ang(1–17) and AcAng(1–17) peptides on the actin aggregation has been tested on human neuroblastoma cells, in the presence or not of copper ions, to exploit their potential use as mimicking system of whole r-Ang and wt-Ang proteins, respectively. 2. Results 2.1. Conformational Features of Ang(1–17) and AcAng(1–17) Peptides The five and seven protonation constant values, respectively, of Ang(1–17) and AcAng(1–17) peptides, as determined by potentiometric titrations, are reported in Table 1. In the investigated pH range, Ang(1–17) and AcAng(1–17) have a total of eight and seven protonation centres, respectively. This difference is related to the amino group in the N-terminal amino acid residue, free in Ang(1–17) and blocked by acetylation in AcAng(1–17). However, due to precipitation phenomena observed at pH = 9 during the titrations of Ang(1–17), the three protonation constant values of Tyr and Lys side chains for this ligand were not determined. For both peptides, the two lowest pKs refer to the carboxylic group of the two aspartic residues. Since their deprotonation occurred with overlapping titration curves, the first and second protonation constants have not been assigned to the specific residue, Asp-2 or Asp-15, respectively [43]. The next two deprotonation steps involve the imidazole nitrogen atoms of the two His residues; also in this case hence each protonation constant has been considered as a macroconstant value, due to the partially overlap of the titration curves,. Until the deprotonation reaction step of histidine (pH < 7), the behavior of the two peptides is similar. At higher pH values, the deprotonation processes of Tyr-6, Tyr-14 and Lys-17, measured only for AcAng(1–17), resulted in the assignment of three macroconstant values, in good agreement with those observed for analogous peptide fragments [39,44]. The far-UV CD spectra of Ang(1–17) and AcAng(1–17) obtained in the pH range of 4–10 are shown in Figure 1. At acidic pH, for both peptides, the presence of a band with a minimum at 198 nm and a maximum at 228 nm suggest a prevalent random coil conformation. Ang(1–17) displays a slight increase of the signal at 198 nm and a decrease of that at 228 nm at increasing pH values. Above pH 7, the histidine residues are deprotonated and the spectra show a decrease of the minimum and the appearance of two new bands displaying a maximum at 232 nm and a minimum at 242 nm. These effects are more evident at pH 10 where the 198 nm band is shifted at 201 nm and a new broad band is observed at 248 nm, assigned to the deprotonated phenolate group of tyrosine residues [44]. All these features point to conformational changes of Ang(1–17), likely due to the electrostatic charges, and regulated by the pH change through different protonation states of the peptide. As to AcAng(1–17), the curve changes are less prominent in comparison to Ang(1–17). The ligands have been also characterized by means of one dimensional (1D) 1H, two dimensional (2D) 1H-13C Heteronuclear Single-Quantum Correlation (HSQC), 2D 1H-1H Total Correlation Spectroscopy (TOCSY) and 2D 1H-1H Rotating-frame Overhauser Spectroscopy (ROESY) NMR measurements at pH values of 5.5 and 7. All signals have been assigned (Tables S1–S8 in Supplementary Materials); the range of ppm chemical shift related to peptide backbone N–H protons (7.79 < HN < 8.42) is narrower for both Ang(1–17) and AcAng(1–17) in comparison to that reported for the whole protein (7.74 < HN < 8.62) [42]. However, such a shift is larger than that assigned to completely unfolded protein (8.12 < HN < 8.42) [45], suggesting a partial folding of both peptides. In Figure S1, the comparison of the spread HN chemical shift (ppm) values with respect to a random coil conformation are reported for the fragment peptides Ang(1–17) and AcAng(1–17), respectively. 2.2. Far ultraviolet-Circular Dichroism (Far-UV CD) Study of Copper Complexes with Ang(1–17) and AcAng(1–17) Peptides The secondary structure of the peptides, correlated to the far-UV CD spectra features, visibly changes upon the addition of copper ions (Figure 2). Specifically, for Ang(1–17), one equivalent of copper(II) induces a decrease for the signal at 198 nm in the pH ranges of 4–5 and 7–9, whereas a opposite trend is observed at pH 6 and 10, accompanied also by a red-shift at pH = 6 and a blue-shift at pH = 10, respectively (Figure 2a). Moreover, at the highest pH values of 9 and 10, two new maxima and one minimum are clearly visible, at 214, 245 and 224 nm, respectively. The CD difference spectra (inset in Figure 2a) do not indicate an increase of specific conformational structure at the different pH values, but the presence of a turn structure can be figured out at the highest pH values of 9 and 10, as reported for other linear peptides [46]. The enhancement of the peptide turn conformation, induced by the pH increase, can be interpreted as due to the involvement of backbone amide nitrogen atoms in Cu2+ coordination (see Section 2.3). The far-UV CD spectra of Cu-peptide system are governed by the amide chromophore, but may contain a contribution of aromatic side chains, in particular the band at 224 and 245 nm might be related to the tyrosine phenolic group deprotonation [47]. On the other hand, by addition of one equivalent of copper to AcAng(1–17), a general decrease and slight shifts of the bands at 198 and 228 nm in the whole pH range of 4–9 are observed (Figure 2b). The difference spectra indicate, already at acidic pH, the formation of turn structures that increase at basic pH. The intensity of signals is much lower than that observed for Cu2+–Ang(1–17). Indeed, in the acetylated peptide, the imidazole nitrogen atoms of His-8 and His-13 are the potential metal anchoring sites; hence, the formation of a macrochelate, resulting in a turn structure, might include primarily the central portion of the peptide (residues 8–13). Otherwise, in Ang(1–17), the free amino terminal amino group is a further potential copper anchoring site; the stronger dichroic effect observed suggests the formation of a greater macrochelate involving amino group of first residue and imidazole nitrogen of His-8 and/or His-13. 2.3. Speciation and Characterization of Copper Complexes with Ang(1–17) and AcAng(1–17) Peptides The stability constant values are reported in Table 2 and the corresponding distribution diagrams in Figure 3, respectively. The distribution diagram in Figure 3a shows that [CuLH] is the first copper(II) complex species formed by Ang(1–17). The logK value determined for this species (logK = logβ111 − logβ011 = 6.05) suggests the involvement of two nitrogen atoms and a 2N2O coordination mode, in good agreement with data reported for analogous peptide sequences [48]. Different isomers involved as copper(II) anchoring sites, either the N-terminal amino group and one imidazole nitrogen or two imidazole nitrogens are likely. UV-vis and CD parameters can discriminate the actual copper(II) coordination environment (Table 3). The UV-vis parameters of [CuLH] species (λmax = 628 nm ε = 90 M−1·cm−1, see Table 3) rule out the formation of a macrochelate, involving two imidazole nitrogen atoms and one carboxylate, which would exhibit the absorption at higher wavelength [49]. Our data are indeed very similar to those reported for a peptide binding Cu2+ by means of the terminal amino group, the deprotonated amide nitrogen atom and the oxygen of a carboxylate group of contiguous aspartic residue [50]. Accordingly, the CD spectrum carried out at pH = 5.5 shows the presence of a band at 299 nm ascribable to a N-amide →Cu2+ charge transfer, confirming the presence of a deprotonated amide nitrogen in metal coordination environment. In addition, a band at 265 nm is found, assigned to either imidazole or amino groups. Hence, the [CuLH] species has to be considered instead as a [CuLH−1(H)2] species, in which one imidazole nitrogen is still protonated and the actual chromophore is Cu(NH2, N−, OCOO−, Owater). The next species formed, [CuL], is predominant at physiological pH. The stepwise constant logK110 (logK110 = logβ110 − logβ111 = 6.35) is compatible with a deprotonation of another amide nitrogen atom. The 25 nm blue shift in the UV-vis λmax absorption confirms the involvement of a further nitrogen atom in the metal coordination environment. In Cu2+ 3N1O coordination mode two or more isomers may exist; among them the most likely are the following: (i)-form where a second deprotonated amide atom is coordinated to Cu2+ (NH2, 2N−, OCOO−); and (ii)-form, with a deprotonated imidazole ring (Nim) in the equatorial plane (NH2, N−, NIm, OCOO−). The presence in the CD spectra recorded at pH = 7 of a dichroic band centered at 336 nm (diagnostic of N-imidazole →Cu2+ charge transfer) point to the predominant (ii)-form. Indeed, the UV-vis parameters (λmax = 605 nm ε = 95 M−1·cm−1) confirm this coordination mode, as found in literature for complex species formed by peptides with a similar primary sequence [49]. Increasing the pH, [CuLH−1] complex species is formed; this species displays its maximum percentage of formation at pH = 8.5. The logK value (logK = logβ11-1 − logβ110 = 8.14) indicates deprotonation and a further coordination of an amide nitrogen atom. This hypothesis is confirmed by UV-vis spectrum where a 35 nm blue shift of the d-d band is observed (Table 3), and by the intensity increase of the dichroic band relative to N-amide →Cu2+ charge transfer, at 322 nm. The next deprotonation species formed is [CuLH−2], at pH ~9, closely followed, at pH ~9.5, by the formation of [CuLH−3]. The UV-vis spectra recorded at in the pH range 9–10 result as the superposition of three complex species, namely [CuLH−1], [CuLH−3], and [CuLH−3]. However, based on the potentiometric data, the deprotonation of a third amide nitrogen atom in the [CuLH−2] species can be assumed, with the formation of a complex where the metal ion is bound to four nitrogen atoms [48,49,50]. As side comment, it is interesting to note that Cu(II) coordination to Ang(1–17) prevented the precipitation of peptide at basic pH values, as instead observed for its apo-form. Above pH = 10, the Tyr and Lys deprotonation occurs, with the formation of [CuLH−3], [CuLH−4] and [CuLH−5] species. Spectroscopic data do not evidence any change, indicating that Tyr and Lys residue are not involved in metal ion binding [48]. As to AcAng(1–17), Figure 3b shows that the peptide forms the [CuLH4] species at pH = 4, which reaches its maximum percentage of formation (around 25%) at pH = 5. The calculated stability constant (logK = logβ114 − logβ014 = 3.99) is higher respect to the typical stability constant of a copper complex in which the metal ion is coordinated to only one imidazole nitrogen of a histidine [51]. Accordingly, this complex species involves a carboxylate group of one aspartate residue, whereas the other aspartate is deprotonated but not coordinated to the metal. Therefore, four isomers with [CuLH4] stoichiometry can exist, formed by His-8, His13, Asp-2 and Asp-15. The smaller macrochelate formed by His13 and Asp-15 is the most thermodynamically favored isomer form. Increasing the pH, [CuLH3] is formed and its stability constant (logK = logβ131 − logβ031 = 5.95) is indicative of a macrochelate formation with coordination to Cu2+ of two imidazole rings and one COO− group of aspartic. This stability constant value is in a good agreement with the value reported for the corresponding complex species formed by other similar peptide fragments [39]. The UV-vis parameters of [CuLH3] species (λmax = 650 nm, ε = 60 M−1·cm−1, see Table 3) indicate a metal coordination to two imidazole nitrogen atoms, one carboxylate group and a water molecule. It is to note that for the corresponding species formed by the unacetylated peptide, λmax value is at the lower wavelength of 628 nm. The CD spectra at pH 5.9 do not show any evidence of the amide band while a band at 331 nm is a fingerprint of imidazole coordination. The next deprotonation step involves one amide backbone nitrogen atom, but the [CuLH2] is only a minor species; for this reason, no spectroscopic data for this complex species have been obtained. At physiological pH, the main species for the Cu-AcAng(1–17) system is [CuLH]. The calculated pK value (pK(2/1) = 5.79; see Table 2) indicates the deprotonation of a further amide nitrogen. This finding suggests the simultaneous presence of the imidazole rings and amide nitrogen atoms bound to the copper(II), as confirmed by the CD spectrum recorded at pH = 7.4, showing the presence of two bands centered around 360 nm and 320 nm, respectively. Furthermore, UV-vis parameters (λ = 580 nm and ε = 105 M−1·cm−1) indicate the involvement of four nitrogen atoms in the equatorial plane of the metal. Increasing the pH, the [CuL] species is formed. In this case, the pK determined for this deprotonation step is 7.38 and is reported in Table 2 as pK(1/0). The coordination of a third nitrogen atom deprotonated is confirmed by the blue-shift of 40 nm observed for the λmax (λ = 540 nm and ε = 125 M−1·cm−1). Furthermore, the increase of the ε is a clear indication that the coordination to the Cu(II) of this third amide nitrogen atom deprotonated with a release of an imidazole ring, created a more distorted copper coordination polyhedron with a stronger ligand field. The other complex species formed at basic pH is assigned to the deprotonation of the side chains of the tyrosine and lysine residues that do not influence the complexation of Cu(II), as shown by the unchanged CD ad UV-vis spectral parameters. 2.4. Nuclear Magnetic Resonance (NMR) Study of Peptide-Copper Complexes Peptides were scrutinised by NMR to determine the specific anchoring site and to discriminate among the different possible copper complexes isomers formed by Ang(1–17) and AcAng(1–17). Selective paramagnetic line broadening and signal disappearance in the 1D 1H, 2D 1H-13C HSQC and 2D 1H-1H TOCSY NMR spectra were recorded at pH 5.5 and 7, and at different copper to ligand molar ratios. Depending on the distance of the paramagnetic center, the observation of proton signals of the residues close to the metal site may be precluded, due to the broadening of NMR signals [52]. Only sub-stoichiometric amounts of Cu2+ ion solution were added to both the peptide solutions, in order to follow the relaxation effect of any nucleus that is in closer vicinity to the paramagnetic center. The line-broadening effects were taken into account in order to localize the metal binding sites along the sequence of the peptides. At pH 5.5, by addition of Cu(II), a gradual but specific decrease and disappearance of proton signal intensity from H8, H13 and D15 residues and, to a lesser extent, of protons from F9 and Y14 residues, are detected in the 1H NMR spectra of AcAng(1–17). The superposition of 1H aromatic and aliphatic region for AcAng(1–17) is reported in Figure S2. Figure 4 shows the comparison of 2D 1H-13C HSQC NMR spectra, in the aromatic and aliphatic region, of AcAng(1–17) free (red) with Cu(II):AcAng(1–17) complexes at 0.02/1 molar ratio (blue) at pH 5.5. The disappearance of signals from Hε1 and Hδ2 of H8 and H13 residues together with Asp-15 residue is visible, supporting the concomitant participation of both histidine residues together with aspartate residue in the coordination to Cu2+ ions. Moreover, the attenuation of signals intensities owing to residues close to the paramagnetic binding site is detected for the residues F9, L10, Q12 and Y14. The direct involvment of D15 residue in the Cu2+ coordination, rather than D2, is also supported by the disappearance in the 1H-1H TOCSY spectrum of its Hβ-HN and Hα-HN correlations (Figure S3), confirming the representation of [CuLH3] coordination mode above described. Figure 5 shows the superposition of 1H aromatic and aliphatic region for AcAng(1–17) by increasing sub-stoichiometric metal to ligand molar ratio, from 1:0 to 1:0.1, obtained at pH 7. From the spectra, the gradual but specific decrease in intensity involving mainly proton signals from H8, H13 and D15 and, to a lesser extent, from F9 and Y14 residues is evident. The involvement of H8, H13 and D15 residues in metal binding is further proven by the disappearance or broadening of their protons in the 1H-13C HSQC spectra (Figure S4). Additional data to support the involvement of these residues in the coordination are provided by the 1H-1H TOCSY, where the spin system HN from H8, H13 and D15 in the AcAng(1–17):Cu(II) 1:0.05 system disappears (Figure 6). A relevant line broadening is also detected for F9, T11, Q12 and Y14, indicating that the residues mainly involved in the coordination are those between H8 and H13 (Figure S5). The NMR study of Ang(1–17) has been carried out by using the same increasing sub-stoichiometric copper to ligand molar ratios, at both pH of 5.5 and 7. In Figure 7, the 2D 1H-13C HSQC NMR spectra of Ang(1–17) at pH 5.5, in the aromatic and aliphatic region, for the free (in red) and the copper complex systems at 0.02/1 and 0.05/1 molar ratios (in blue) are reported. The most evident difference with respect to AcAng(1–17) is the strong decrease of Q1, D2, N3 and Y6 signals, other than Hβ of D15, Hε1 and Hδ2 of H8 and H13 residues. The 1H-1H TOCSY spectra of Ang(1–17), both free and Cu(II)-complexed at pH 5.5, are reported in Figure 8. The proton spin systems of D15 residue totally disappear, whereas those of D2 residue are still present, though decreased in intensity. The titration of Ang(1–17) with increasing amount of Cu(II), in the molar ratio range from 1:0 to 1:0.05, has been followed at pH 7 and the corresponding 1H spectra are reported in Figure S6. From the comparison of 1H-13C HSQC NMR spectra in the aromatic and aliphatic region of Ang(1–17) free (red) and of Cu(II):Ang system (blue) (Figure S7) and from the selection of 1H-1H TOCSY spectrum of the same systems (Figure S8), it is evident that the signals from H8 and H13 entirely disappear. Moreover, the H-C correlations from Asp residues, D2 and D15, almost disappeared in the HSQC spectrum, are instead still present in the TOCSY spectrum, suggesting that their involvement in the coordination to Cu(II) ion is reduced, mainly for D15 residue. Line broadening can be clearly noticed for R5, F9 and Y14 residues. The residues mainly affected by the interaction of Cu(II) ion with Ang(1–17) are evidenced in Figure S9. Conversely to what has been determined with AcAng(1–17), the D15 residue in Ang(1–17) shows a limited participation in complex formation, as demonstrated by its almost unaffected resonances in the TOCSY spectra obtained at pH 7 (Figure S7). In summary, the NMR data at pH 7 indicate the involvement of both N-terminal residue (Q1, Asp2) together with imidazole, as deduced from potentiometry and UV-vis CD results. 2.5. Neuroblastoma Cells Experiments The results of thetetrazolium dye 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (Figure 9) evidence that all the tested treatments are non-cytotoxic. No significant differences in cell viability are found, except for the incubation with 10 µM of the complex Ang(1–17):Cu(II), where a small increase in neuroblastoma cell viability is measured compared to the control. Preliminary cellular experiments of confocal imaging were performed to scrutinize qualitatively the effects of both peptides and the proteins in terms of actin organization, when supplemented to the cells together with the metal ions. The distribution of actin filaments in the neuroblastoma cells treated with Ang(1–17) and AcAng(1–17), both apo- and complexed-forms with Cu(II) ions, in comparison with the whole Ang protein, both wt-Ang and r-Ang, is shown in Figure 10. The common patterns for actin microfilament organization in the cell cytoplasm, i.e., peripheral actin bands, actin bundles (stress fibers), and diffuse actin networks, are visible for all the used cell treatments. The actin pattern is brightest for the cells treated with r-Ang (Figure 10f), while it visibly changes in cells treated with r-Ang:Cu(II) complex (Figure 10g). On the contrary, both the treatments with wt-Ang (Figure 10h) and wt-Ang:Cu(II) complex (Figure 10i) induce comparable actin patterns, with an evident polygonal arrangement of actin filaments. It is noteworthy that the cells treated with Ang(1–17):Cu(II) complex (Figure 10b) display a similar actin pattern than those treated with the free-amino r-Ang complexed to copper. On the other hand, for the acetylated peptide, similarities are found between the cells treated with the apo-form (Figure 10c) and the N terminal amino-blocked wt-Ang (Figure 10h). The quantitative analysis of fluorescence for the actin staining is shown in Figure 11. The cells treated with the metal alone at the concentration used with the proteins (100 nM) do not significantly differ with respect to the control untreated cells. On the other hand, the cells supplemented with 10 µM copper (i.e., the concentration used for the peptides) visibly increase the actin fluorescence intensity. The overall trend observed, for both the proteins and the peptides, is a decrease of intensity for cell treatment with their copper complexes in comparison to the treatment with the free ligands. It is noteworthy that such intensity decrease exhibit similarities in wt-Ang and Ang(1–17) (i.e., huge decrease in intensity) and r-Ang and AcAng(1–17) (i.e., small decrease in intensity), which are in contrast to what was expected on the basis of the peptidomimetic design strategy, namely AcAng(1–17) mimicking wt-Ang and Ang(1–17) mimicking the r-Ang. This finding suggest that other factors are likely to be involved in the peptide-metal/cell membrane interaction which cannot simply be ruled out by this experiments. Specifically, the involvement of other protein domains and/or the formation of various Ang–Ang aggregates driven by the presence of the free amino group can be invoked. Once again, these considerations stress the relevant issue of the use of the “proper” protein form, wt-Ang, to scrutinize the pathways involving the protein in angiogenesis processes, especially in the presence of the metal ions. 3. Discussion The protein angiogenin (Ang) is a potent angiogenic factor whose activity is influenced by copper ions. Many literature reports on the protein activity have been obtained using the recombinant form (r-Ang), which contains an extra methionine as first residue. In contrast, the protein effectively present in human plasma (wt-Ang), exhibit the amino terminal group blocked, owing to the spontaneously cyclization of the first residue glutamine to pyroglutamate. To focus on the copper binding events within the N-terminal domain of the protein, we synthesized two peptides encompassing the residues 1–17 of the protein, Ang(1–17), with the amino free, and AcAng(1–17), the analogous form with the N-terminal amino group acetylated. The chemical characterization by means of potentiometry, NMR, UV-vis and CD spectroscopies of the copper(II) complexes formed with Ang(1–17) and AcAng(1–17) demonstrate that the copper coordination environment in the N-terminal domain of the protein is strongly influenced by the free amino group. Specifically, for Ang(1–17), mimicking the recombinant protein form (r-Ang), the predominant copper complex species at physiological pH involves: (i) the amino group; (ii) the deprotonated amide of Asp-2; and (iii) one of imidazole of His-8 or His-13. On the contrary, the acetylated peptide AcAng(1–17), which has a blocked amino group in the N-terminus as the wild type protein, wt-Ang, binds copper ions through: (i) the two imidazole groups; and (ii) the deprotonated amide nitrogen atoms nearby the histidine residues. Taking into account the biological role of the protein in physiological as well as pathological conditions involving the angiogenic processes, we tested the activity of the two peptides in terms of cell viability and staining of actin, which is one of the potential target receptors driving the Ang–cell membrane interaction. Cellular experiments in neuroblastoma cell line SH-SY5Y cells demonstrate that, at the used experimental conditions, both the peptides and their correspondent copper complexes are not-cytotoxic. Rather, a small increase of cell viability, which can be explained as proliferative activity [53], is found for the cells treated with free amino peptide Ang(1–17) complexed with Cu(II) ions. The actin patterns, followed by confocal microscopy, of the cells treated with the peptides and the proteins, in the apo- or copper-complexed form, evidence strong differences between acetylated and free-amino peptides. Indeed, notwithstanding a general trend of decreased fluorescence intensity for the copper-complexed forms with respect to the free peptide ligand, the AcAng(1–17):Cu(II) complex exhibits a huge decrease of the emission of the stained actin, whereas the effect is smoothed for Ang(1–17):Cu(II) species. This observation is explained in terms of the different metal coordination modes and binding affinities likely determining different aggregation morphologies that might result in significantly different biological effects [54]. As to the actin staining results for the cells treated with the whole proteins, the recombinant form is more active (i.e., strong fluorescence emission) than the wild-type angiogenin. Analogously to the peptides, the respective copper complexes exhibit a decrease of fluorescence intensity, this effect being larger for r-Ang than wt-Ang. Since one possible process related to the actin activation mechanism is the occurrence of both intramolecular Ang–Ang and intermolecular Ang–actin interactions [22], the presence of free amino groups might actually affect such a pathway. Again, the strong effect displayed by copper addition in the case of r-Ang can be correlated to the metal-chelating capability of such free amino groups, as demonstrated for short peptide sequences as well as oligopeptides [54,55]. For example, in the case of r-Ang, the formation of dimeric protein structures with bridges through the amino groups could be prompted by the presence of the copper [37]. In conclusion, this work provides experimental proof to support the significance of using the actual angiogenin present in the human plasma, the wild type form (wt-Ang), to scrutinize the protein activity, especially in the presence of the metal ions. 4. Materials and Methods 4.1. Chemicals The peptide Ang(1–17) and Ac(ang(1–17) were supplied by Caslo Aps, Lyngby, Denmark. All other chemicals, of the highest available grade, were purchased from Sigma-Aldrich (Munich, Germany) and used without further purification. 4.2. Potentiometric Titrations Potentiometric titrations were performed with a home-assembled fully automated apparatus sets (Metrohm E654 pH-meter, combined micro pH glass electrode, Orion 9103SC, Hamilton digital dispenser, Model 665) controlled by the appropriate software set up in our laboratory. The titration cell (2.5 mL) was thermostated at 298.0 ± 0.2 K, and all solutions were kept under an atmosphere of argon, which was bubbled through a solution having the same ionic strength and temperature as the measuring cell. KOH solutions (0.1 M) were added through a Hamilton buret equipped with 1 cm3 syringe. The ionic strength of all solutions was adjusted to 0.10 M (KNO3). In order to determine the stability constants, solutions of the ligands (protonation constants) or the ligands with Cu2+ (copper complex constants) were titrated using 0.1 M sodium hydroxide. Ligand concentration ranged from 1.4 to 2.0 × 10−3 for the protonation and complexation experiments, respectively. A minimum of three independent runs were performed to determine the protonation constants, while four independent experiments were run for the copper(II) complexation constants. Metal to ligand ratios of 1:1 were employed. The initial pH was always adjusted to 2.4. To avoid systematic errors and verify reproducibility, the electromotive force (EMF) values of each experiment were taken at different time intervals. To obtain protonation and complexation constants, the potentiometric data were refined using Hyperquad [56], which minimizes the error square sum of the measured electrode potentials through a nonlinear iterative refinement of the sum of the squared residuals, U, and also allows for the simultaneous refinement of data from different titrations: U = Σ(Eexp − Ecalc)2 where Eexp and Ecalc are the experimental and calculated electrode potentials, respectively. Errors in stability constant values are reported as three times standard deviations. The formation reaction equilibria of ligands with protons and copper(II) ions are given in Equation (1): pCu + qH + rL ⇆ CupHqLr(1) where L are the peptides under study. The stability constant βpqr is defined in Equation (2): βpqr = [CupHqLr]/[Cu]p · [H]q · [L]r(2) The species distribution as a function of the pH was obtained using the computer program Hyss [57]. 4.3. Ultraviolet-Visible (UV-vis) Measurements UV-vis spectra were recorded at 25 °C using an Agilent 8453 or a Varian Cary 500 spectrophotometer (Agilent Technologies, Santa Clara, CA, USA). The concentrations of the peptides and copper(II) used to record absorption spectra were the same as those for the potentiometric titrations. Combined spectroscopic and potentiometric metal-complex titrations were performed into a 3 mL quartz cuvette with a 1 cm path length to obtain the spectrum in the Visible region at each pH value simultaneously. These experiments were replicated at least three times for each copper-peptide system. Spectroscopic data were processed by means of HYPERQUAD program [56]. 4.4. Circular Dichroism (CD) Measurements CD spectra were obtained at 25 °C under a constant flow of nitrogen on a Jasco model 810 spectropolarimeter (Jasco, Easton, MD, USA) at a scan rate of 50 nm·min−1 and a resolution of 0.1 nm, the path length being 1 cm, in the 280–800 nm range. The spectra were recorded as an average of either 3 or 5 scans. Calibration of the instrument was performed with a 0.06% aqueous solution of ammonium camphorsulfonate. The CD spectra of the copper(II) complexes on varying the solution pH were obtained in both the 190–250 and 250–800 nm wavelength regions. All the solutions were freshly prepared using double distilled water. The copper(II) ion and peptide concentrations used for the acquisition of the CD spectra in the Visible region were identical to those used in the potentiometric titrations. The results are reported as ε (molar adsorption coefficient) and Δε (molar dichroic coefficient) in M−1·cm−1. 4.5. Nuclear Magnetic Resonance (NMR) Spectroscopy NMR experiments were carried out on a Bruker AscendTM 400 MHz spectrometer (Bruker, Billerica, MA, USA) equipped with a 5 mm automated tuning and matching broad band probe (BBFO) with z-gradients, as previously described [58,59]. NMR measurements were performed by using a concentration of the peptides of 2 mM, in 90/10 (v/v) H2O/D2O, at 298 K. HSQC (2D 1H-13C heteronuclear correlation spectra) were acquired using a phase-sensitive sequence utilizing Echo-Antiecho-TPPI gradient selection with a heteronuclear coupling constant JXH = 145 Hz, and shaped pulses for all 180° pulses on f2 channel with decoupling during acquisition. Sensitivity improvement and gradients in back-inept were also used. In all of the experiments, relaxation delays of 2 s and 90° pulses of about 10 µs were used. The solvent suppression in the 1D 1H, 2D 1H-1H TOCSY and 2D 1H-1H ROESY experiments was performed by using excitation sculpting with gradients. The spin-lock mixing time of TOCSY experiments was obtained with MLEV17. 1H-1H TOCSY spectra were carried out using 60 ms as mixing times. The signals of both free and metal-bound peptides at different pH values and different metal to ligand molar ratios have been assigned by using a combination of 1D, 2D TOCSY, HSQC and ROESY experiments. All NMR results were processed by using TopSpin (Bruker Instruments) software and analyzed using Sparky 3.11 and MestRe Nova 6.0.2 (Mestrelab Research S.L.) programs (Santiago de Compostela, Spain). 4.6. Cellular Experiments Human neuroblastoma SH-SY5Y cells were grown in DMEM-F12 (1:1) medium, supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 2 mM l-glutamine, and maintained in a humidified incubator at 37 °C in 5% CO2 atmosphere. 4.6.1. Cell Viability Assay In order to evaluate the proliferative and pro-survival effects of Ang(1–17) and AcAng(1–17) peptides, the colorimetric MTT assay was performed. The assay is based on the reduction, worked by cellular oxidoreductase enzymes that reflect the number of viable cells present, of the tetrazolium dye MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide to its insoluble formazan, which has a purple color. Cells were plated in a 48-well plate and grown in DMEM-F12 up to 70%–80% confluency, followed by treatment for 24 h with each of the two peptides at the concentration of 5 × 10−6, 1 × 10−5 and 2 × 10−5 M, as well as the correspondent complexes with CuSO4 at equimolar concentrations. MTT (1 mg/mL) in PBS was then added to wells and incubated at 37 °C for 2 h to allow for complete cleavage of the tetrazolium salt by metabolically active cells. Next, MTT was removed and 220 µL of dimethyl sulfoxide (DMSO) was added. One hundred microliters of each well was transferred to a 96-well plate followed by colorimetric analysis using a multilabel plate reader at 560 nm of wavelength. Absorbance values plotted are the mean from triplicate experiments. Statistical relevance was calculated by One-way Analysis of variance (ANOVA) test performed with Origin 8.3 software (Northampton, MA, USA). 4.6.2. Confocal Microscopy Imaging SH-SY5Y cells in culture at 80% of confluence were split on glass bottom Petri dishes (WillCo Wells, glass diameter of 22 mm). The day after cells were treated for 1 h with Ang(1–17), AcAng(1–17), wt-Ang, r-Ang, CuSO4, the peptide–copper and the protein–copper complexes at 1:1 mole ratio. The used concentrations were of 1 × 10−7 M for the peptides and 1 × 10−5 M for the proteins. After the incubation time, cells were washed with phosphate buffer saline solution (10 mM PBS, 37 °C, pH = 7.4), fixed with high purity 4% formaldehyde in PBS (pH = 7.3) and stained with the nuclear dye DAPI (ThermoFisher). Afterwards, cells were permeabilized with 0.5% Triton X-100 and stained with a high-affinity F-actin probe, conjugated to green-fluorescent Alexa Fluor® 488 dye (ActinGreen™ 488 ReadyProbes® Reagent, TermoFisher (Thermo Fisher Scientific, Waltham, MA, USA). Confocal imaging was performed with an Olympus FV1000 confocal laser scanning microscope (LSM, Olympus, Shinjuku, Japan), equipped with diode UV (405 nm, 50 mW), multiline Argon (457, 488, 515 nm, total 30 mW), HeNe(G) (543 nm, 1 mW) and HeNe(R) (633 nm, 1 mW) lasers. An oil immersion objective (60xO PLAPO) and spectral filtering system were used. The detector gain was fixed at a constant value and images were taken, in sequential mode, for all the samples at random locations throughout the area of the well. Quantitative analysis of fluorescence was performed using the ImageJ software (1.50i version, NIH), in terms of integrated density ID = N·[M − B], where N is the number of pixels in the selection, M is the average gray value of the pixels and B is the most common pixel value [60]. Acknowledgments Maria Antonietta Zoroddu, Diego La Mendola and Enrico Rizzarelli thank Italian Ministry of University and Research (MIUR) for partial support (PRIN 2010M2JARJ; Cristina Satriano, Diego La Mendola and Enrico Rizzarelli thank the University Consortium for Research in the Chemistry of Metal ions in Biological Systems (CIRCMSB); Diego La Mendola thanks University of Pisa (PRA_2015_0015 “Liquidi Ionici funzionalizzati per applicazione nella chimica fine”. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1240/s1. Click here for additional data file. Author Contributions Diego La Mendola, Cristina Satriano and Enrico Rizzarelli conceived and designed the experiments; Antonio Magrì, Alessia Munzone, Cristina Satriano, Massimiliano Peana, Serenella Medici performed the experiments; Cristina Satriano, Massimiliano Peana, Maria Antonietta Zoroddu, Orjan Hansson, Enrico Rizzarelli and Diego La Mendola analyzed the data; Diego La Mendola and Enrico Rizzarelli contributed reagents/materials/analysis tools; Diego La Mendola, Cristina Satriano, Antonio Magrì, Maria Antonietta Zoroddu and Enrico Rizzarelli wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Far ultraviolet-circular dichroism (Far-UV CD) spectra of: (a) Ang(1–17); and (b) AcAng(1–17). Figure 2 Far-UV CD spectra of copper(II) complexes with ligands (L): (a) Ang(1–17); and (b) AcAng(1–17), [L] = 5 × 10−5 M; metal to ligand molar ratio of 1:1. Insets: the spectra of copper complexes subtracted by the spectra of the free ligands. Figure 3 Species distribution of copper(II) complexes with: (a) Ang(1–17); and (b) AcAng(1–17). [L] = 1 × 10−3 M; metal to ligand molar ratio of 1:1. Figure 4 Aromatic and aliphatic region of 1H-13C HSQC spectrum of AcAng(1–17) free (red) and AcAng(1–17):Cu(II), 1:0.02 mole ratio (blue), at pH 5.5. Disappearing peaks and those with major broadening have been labelled. Figure 5 Superposition of 1H aromatic (a) and aliphatic (b) region for AcAng(1–17) peptide by increasing substechiometric metal to ligand molar ratio: 0, 1:0.002, 1:0.005, 1:0.01, 1:0.02, 1: 0.05, and 1:0.1 at pH 7. Insets: aromatic and aliphatic protons with total disappearance or broadening of the signals. Figure 6 Comparison in the 1H-1H TOCSY spectrum, aromatic region, of AcAng(1–17) free (red) and AcAng(1–17):Cu(II) 1:0.05 system (blue) at pH 7. Figure 7 1H-13C HSQC NMR spectra, in the aromatic and aliphatic region, of Ang(1–17) free (red) and Cu(II):Ang(1–17) system at 1 to 0.02 and 1 to 0.05 (blue) molar ratio at pH 5.5. Figure 8 NMR 1H-1H TOCSY spectrum of Ang(1–17) free (red) and Cu(II):Ang(1–17) system at 1 to 0.02 and 1 to 0.05 (blue) molar ratio at pH 5.5. Figure 9 Results of MTT assay (expressed in terms of intensity normalized to the control) for SH-SY5Y cells incubated 1 h (37 °C, 5% CO2) with CuSO4, Ang(1–17), Ang(1–17):CuSO4 (1:1), AcAng(1–17) and AcAng(1–17):CuSO4 (1:1) in the culture medium. Data are average of three different experiments (error bar = standard deviation). Peptide concentrations: 5 × 10−6, 1 × 10−5, and 2 × 10−5 M. (* p < 0.05 of significance with respect to control, One-way Analysis of variance (ANOVA) test). The red dot line is to guide the eye at the control level. Figure 10 Merged confocal images recorded in the blue (DAPI staining of nuclei, λex/λem = 405/425–475 nm) and in the green (actin staining, λex/λem = 488/500–530 nm) channel for SH-SY5Y cells incubated 1 h (37 °C, 5% CO2) with: (a) 10 µM Ang(1–17); (b) Ang(1–17):CuSO4 (1:1, 10 µM); (c) AcAng(1–17); (d) AcAng(1–17):CuSO4 (1:1, 10 µM); (e) 10 µM CuSO4; (f) 100 nM r-Ang; (g) r-Ang:CuSO4 (1:1, 100 nM); (h) wt-Ang; (i) wt-Ang:CuSO4 (1:1, 100 nM); (j) 100 nM CuSO4; and (k) control. Scale bar = 30 µm. Figure 11 Quantitative analysis of emission recorded for actin staining (λex/λem = 488/500–530 nm) from the confocal microscopy analyses. The fluorescence intensities were calculated in terms of integrated density (ID) values. Mean ID (with the standard deviation) from five randomly chosen fields are shown. Data are normalized with respect to the control. ijms-17-01240-t001_Table 1Table 1 Protonation constant (logβqr) and pK values of (Ang(1–17) and AcAng(1–17) (T = 298 K and I = 0.1 M KNO3). a Speciesqr logβqr b logβqr b L = Ang(1–17) L = AcAng(1–17) HL 7.18 (2) 10.89 (1) H2L 13.60 (2) 20.90 (1) H3L 19.45 (2) 29.87 (3) H4L 23.28 (3) 36.39 (5) H5L 26.55 (2) 42.34 (4) H6L - 46.24 (5) H7L - 49.63 (4) pK COO− 3.27 3.39 pK COO− 3.83 3.90 pK His 5.84 5.95 pK His 6.42 6.52 pK NH2 7.18 - pK Tyr or Lys - 8.97 pK Tyr or Lys - 10.01 pK Tyr or Lys - 10.89 a Standard deviations (3σ values) are given in parentheses; [L] = 1 × 10−3 M; b qH + rL = HqLr; βqr = [HqLr]/[H]q[L]r. ijms-17-01240-t002_Table 2Table 2 Stability constants (logβpqr) and pK values of copper(II) complexes with Ang(1–17) and AcAng(1–17); (T = 298 K, I = 0.1 M KNO3). a Species (pqr) logβpqr b L = Cu–Ang(1–17) pK(n/m) Species (pqr) logβpqr b L = Cu–AcAng(1–17) pK(n/m) CuLH 13.23 (1) - CuLH4 40.38 (3) - CuL 6.80 (2) (1/0) = 6.43 CuLH3 35.82 (1) (4/3) = 4.56 CuLH−1 −1.34 (2) (0/−1) = 8.14 CuLH2 29.52 (3) (3/2) = 6.30 CuLH−2 −10.39 (2) (−1/−2) = 9.04 CuLH 23.73 (1) (2/1) = 5.79 CuLH−3 −19.85 (2) (−2/−3) = 9.46 CuL 16.35 (2) (1/0) = 7.38 CuLH−4 −30.12 (2) (−3/−4) = 10.25 CuLH−1 7.66 (2) (−1/0) = 8.69 CuLH−5 −40.04 (2) (−4/−5) = 9.94 CuLH−2 −2.16 (2) (−2/−1) = 9.82 CuLH−3 −12.36 (2) (−3/−2) = 10.20 a Standard deviations (3σ values) are given in parentheses. Charges are omitted for clarity; pK(n/m) values reflect the pK value of copper(II) complexes; [L] = 1 × 10−3 M; molar ratio 1:1; b pCu + qH + rL = CupHqLr; βbqr = [CupHqLr]/[Cu]p[H]q[L]r. ijms-17-01240-t003_Table 3Table 3 Spectroscopic parameters of Copper(II) complexes. L pH Species (CuLH) UV-vis λ (nm) (ε, M−1·cm−1) a CD λ (nm) (Δε, M−1·cm−1) Ang(1–17) 5.5 CuLH 628 (90) 265 (1.171); 298 (−1.224); 620 (−0.223) 7 CuL 605 (95) 264 (3.28); 299 (−1.318); 336 (0.456); 597 (−0.456) 8.5 CuLH−1 574 (106) 265 (3.847); 297 (0.305) 322 (1.051); 574 (−0.704) 9 CuLH−1 CuLH−2 CuLH-3 - 263 (3.805); 315 (1.409); 563 (−0.875) 10 CuLH−3 CuLH−3 CuLH−4 - 263 (3.225); 309 (1.675); 363 (−0.221); 555 (−1.023) 11 CuLH−5 525 (130) 262 (3.051); 308 (1.454); 352 (−0.314); 555 (−1.056) AcAng(1–17) 5 Cu (40%), CuLH4 (20%), CuLH3 (40%) - 619 (0.163) 5.9 CuLH3 650 (60) 258 (2.518); 331 (0.304); 598 (−0.236) 6.5 CuLH3; CuLH2; CuLH - 256 (7.449); 330 (0.602); 526 (0.203); 594 (−0.403) 7.5 CuLH 585 (105) 262 (8.525); 323 (1.031); 365 (−0.127); 502 (−0.745); 678 (0.621) 8.3 CuL 540 (125) 263 (7.999); 318 (1.143); 359 (-0.524); 500 (−1.127); 648 (0.989) 9.5 CuLH-1; 540 (135) 264 (7.812); 310 (1.360); 352 (−0.791); 501 (−1.216); 641 (1.276) 10.2 CuLH-2 540 (128) 264 (7.622); 312 (1.252); 351 (−0.789); 502 (−1.192); 648 (1.262) 11 CuLH−2; CuLH−3 540 (135) 264 (7.115); 312 (1.181); 350 (−0.825); 501 (−1.161); 641 (1.269) a Ref. [42]. ==== Refs References 1. Strydom D.J. Fett J.W. Lobb R.R. Alderman E.M. Bethune J.L. Riordan J.F. Vallee B.L. Amino acid sequence of human tumor derived angiogenin Biochemistry 1985 24 5486 5494 10.1021/bi00341a031 2866794 2. Fett J.W. Strydom D.J. Lobb R.R. Alderman E.M. Bethune J.L. Riordan J.F. Vallee B.L. Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells Biochemistry 1985 24 5480 5486 10.1021/bi00341a030 4074709 3. Strydom D.J. The angiogenesis Cell. Mol. Life Sci. 1998 54 811 824 10.1007/s000180050210 9760990 4. Adams S.A. Subramanian V. The angiogenins: An emerging family of ribonuclease related proteins with diverse cellular functions Angiogenesis 1999 3 189 199 10.1023/A:1009015512200 14517418 5. Tello-Montoliu A. Patel J.V. Lip G.Y. Angiogenin: A review of the pathophysiology and potential clinical applications Thromb. Haemost. 2006 4 1864 1874 10.1111/j.1538-7836.2006.01995.x 16961595 6. Badet J. Soncin F. N’Guyen T. Barritault D. In vivo and in vitro studies of angiogenin a potent angiogenic factor Blood Coagul. Fibrinolysis 1990 1 721 724 1723310 7. Sheng J. Xu Z. Three decades of research on angiogenin: A review and perspective Acta Biochim. Biophys. Sin. 2016 48 399 410 10.1093/abbs/gmv131 26705141 8. Hu G. Xu C. Riordan J.F. Human angiogenin is rapidly translocated to the nucleus of human umbilical vein endothelial cells and binds to DNA J. Cell. Biochem. 2000 76 452 462 10.1002/(SICI)1097-4644(20000301)76:3<452::AID-JCB12>3.0.CO;2-Z 10649442 9. Xu Z.P. Tsuji T. Riordan J.F. Hu G.F. The nuclear function of angiogenin in endothelial cells is related to rRNA production Biochem. Biophys. Res. Commun. 2002 294 287 292 10.1016/S0006-291X(02)00479-5 12051708 10. Xu Z.P. Tsuji T. Riordan J.F. Hu G.F. Identification and characterization of an angiogenin-binding DNA sequence that stimulates luciferase reporter gene expression Biochemistry 2003 42 121 128 10.1021/bi020465x 12515546 11. Kishimoto K. Liu S. Tsuji T. Olson A.K. Hu G.F. Emerging role of angiogenin in stress response and cell survival under adverse conditions Oncogene 2005 24 445 456 10.1038/sj.onc.1208223 15558023 12. Li S. Hu G.-F. Angiogenin-mediated rRNA transcription in cancer and neurodegeneration Int. J. Biochem. Mol. Biol. 2010 1 26 35 20827423 13. Vallee B.L. Riordan J.F. Organogenesis and angiogenin Cell. Mol. Life Sci. 1997 53 803 815 10.1007/s000180050101 9413551 14. Gao X. Xu Z. Mechanism of action of angiogenin Acta Biochim. Biophys. Sin. 2008 40 619 624 10.1111/j.1745-7270.2008.00442.x 18604453 15. Liu S. Yu D. Xu Z.P. Riordan J.F. Hu G.F. Angiogenin activates Erk1/2 in human umbilical vein endothelial cells Biochem. Biophys. Res. Commun. 2001 287 305 310 10.1006/bbrc.2001.5568 11549292 16. Xu Z. Monti D.M. Hu G. Angiogenin activates human umbilical artery smooth muscle cells Biochem. Biophys. Res. Commun. 2001 285 909 914 10.1006/bbrc.2001.5255 11467837 17. Hu G.F. Riordan J.F. Vallee B.L. A putative angiogenin receptor in angiogenin-responsive human endothelial cells Proc. Natl. Acad. Sci. USA 1997 94 2204 2209 10.1073/pnas.94.6.2204 9122172 18. Hu H. Gao X. Sun Y. Zhou J. Yang M. Xu Z. α-Actinin-2, a cytoskeletal protein, binds to angiogenin Biochem. Biophys. Res. Commun. 2005 329 661 667 10.1016/j.bbrc.2005.01.158 15737636 19. Gao X. Hu H. Zhu J. Xu Z. Identification and characterization of follistatin as a novel angiogenin-binding protein FEBS Lett. 2007 581 5505 5510 10.1016/j.febslet.2007.10.059 17991437 20. Zhang H. Gao X. Weng C. Xu Z. Interaction between angiogenin and fibulin 1: Evidence and implication Acta Biochim. Biophys. Sin. 2008 40 375 380 10.1111/j.1745-7270.2008.00420.x 18465022 21. Hu G.F. Strydom D.J. Fett J.W. Riordan J.F. Vallee B.L. Actin is a binding protein for angiogenin Proc. Natl. Acad. Sci. USA 1993 90 1217 1221 10.1073/pnas.90.4.1217 7679494 22. Pyatibratov M.G. Tolkatchev D. Plamondon J. Xu P. Ni F. Kostyukova A.S. Binding of human angiogenin inhibits actin polymerization Arch. Biochem. Biophys. 2010 495 74 81 10.1016/j.abb.2009.12.024 20045391 23. Pyatibratov M.G. Kostyukova A.S. New insigthts into the role of angiogenin in actin polymerization Int. Rev. Cell Mol. Biol. 2012 295 175 198 22449490 24. Greenway M.J. Andersen P.M. Russ C. Ennis S. Cashman S. Donaghy C. Patterson V. Swingler R. Kieran D. Prehn J. ANG mutations segregate with familial and “sporadic” amyotrophic lateral sclerosis Nat. Genet. 2006 38 411 413 10.1038/ng1742 16501576 25. Li S. Sheng J. Hu J.K. Yu W. Kishikawa H. Hu M.G. Shima K. Wu D. Xu Z. Xin W. Ribonuclease 4 protects neuron degeneration by promoting angiogenesis, neurogenesis, and neuronal survival under stress Angiogenesis 2013 16 387 404 10.1007/s10456-012-9322-9 23143660 26. Kim Y.N. Kim H. Decreased serum angiogenin level in Alzheimer’s disease Prog. Neuropsychopharmacol. Biol. Psychiatry 2012 38 116 120 10.1016/j.pnpbp.2012.02.010 22449478 27. Steidinger U. Slone S.R. Ding H. Standaert D.G. Yacoubian T.A. Angiogenin in Parkinson disease models: Role of Akt phosphorylation and evaluation of AAV-mediated angiogenin expression in MPTP treated mice PLoS ONE 2013 8 1240 10.1371/journal.pone.0056092 23409128 28. McAuslan B.R. Reilly W.G. Hannan G.N. Gole G.A. Angiogenic factors and their assay: Activity of formyl methionyl leucyl phenylalanine, adenosine diphosphate, heparin, copper, and bovine endothelium stimulating factor Microvasc. Res. 1983 26 323 338 10.1016/0026-2862(83)90080-8 6197617 29. Kaplan J.H. Lutsenko S. Copper transport in mammalian cells: Special care for a metal with special needs J. Biol. Chem. 2009 284 25461 25465 10.1074/jbc.R109.031286 19602511 30. He W. James Kang Y. Ischemia-induced copper loss and suppression of angiogenesis in the pathogenesis of myocardial infarction Cardiovasc. Toxicol. 2013 13 1 8 10.1007/s12012-012-9174-y 22644803 31. Finney L. Vogt S. Fukai T. Glesne D. Copper and angiogenesis: Unravelling a relationship key to cancer progression Clin. Exp. Pharmacol. Physiol. 2009 36 88 94 10.1111/j.1440-1681.2008.04969.x 18505439 32. Davies K.M. Mercer J.F. Chen N. Double K.L. Copper dyshomeostasis in Parkinson’s disease: Implications for pathogenesis and indications for novel therapeutics Clin. Sci. 2016 130 565 574 10.1042/CS20150153 26957644 33. Lovejoy D.B. Guillemin G.J. The potential for transition metal-mediated neurodegeneration in amyotrophic lateral sclerosis Front. Aging Neurosci. 2014 6 173 10.3389/fnagi.2014.00173 25100994 34. Ayton S. Lei P. Bush A.I. Metallostasis in Alzheimer’s disease Free Radic. Biol. Med. 2013 62 76 89 10.1016/j.freeradbiomed.2012.10.558 23142767 35. Finney L. Mandava S. Ursos L. Zhang W. Rodi D. Vogt S. Legnini D. Maser J. Ikpatt F. Olopade O.I. X-ray fluorescence microscopy reveals large-scale relocalization and extracellular translocation of cellular copper during angiogenesis Proc. Natl. Acad. Sci. USA 2007 104 2247 2252 10.1073/pnas.0607238104 17283338 36. Badet J. Soncin F. Guitton J.D. Lamare O. Cartwright T. Barritault D. Specific binding of angiogenin to calf pulmonary artery endothelial cells Proc. Natl. Acad. Sci. USA 1989 86 8427 8431 10.1073/pnas.86.21.8427 2813401 37. Soncin F. Guitton J.D. Cartwrigh T. Badet J. Interaction of human angiogenin with copper modulates angiogenin binding to endothelial cells Biochem. Biophys. Res. Commun. 1997 236 604 610 10.1006/bbrc.1997.7018 9245697 38. La Mendola D. Magrì A. Vagliasindi L.I. Hansson Ö. Bonomo R.P. Rizzarelli E. Copper(II) complex formation with a linear peptide encompassing the putative cell binding site of angiogenin Dalton Trans. 2010 10678 10684 10.1039/c0dt00732c 20941439 39. La Mendola D. Farkas D. Bellia F. Magrì A. Travaglia A. Hansson Ö. Rizzarelli E. Probing the copper(II) binding features of angiogenin. Similarities and differences between a N-terminus peptide fragment and the recombinant human protein Inorg. Chem. 2012 51 128 141 10.1021/ic201300e 22148849 40. Hu G.F. Copper stimulates proliferation of human endothelial cells under culture J. Cell. Biochem. 1998 69 326 335 10.1002/(SICI)1097-4644(19980601)69:3<326::AID-JCB10>3.0.CO;2-A 9581871 41. Giacomelli C. Trincavelli M.L. Satriano C. Hansson Ö. La Mendola D. Rizzarelli E. Martini C. Copper (II) ions modulate Angiogenin activity in human endothelial cells Int. J. Biochem. Cell Biol. 2015 60 185 196 10.1016/j.biocel.2015.01.005 25596488 42. La Mendola D. Arnesano F. Hansson Ö. Giacomelli C. Calò V. Mangini V. Magrì A. Bellia F. Trincavelli M.L. Martini C. Copper binding to naturally occurring, lactam form of angiogenin differs from that to recombinant protein, affecting their activity Metallomics 2016 8 118 124 10.1039/C5MT00216H 26594037 43. Kallay C. Varnagy K. Micera G. Sanna D. Sovago I.J. Copper(II) complexes of oligopeptides containing aspartyl and glutamyl residues. Potentiometric and spectroscopic studies Inorg. Biochem. 2005 99 1514 1525 10.1016/j.jinorgbio.2005.04.009 15927267 44. La Mendola D. Bonomo R.P. Caminati S. Di Natale G. Emmi S. Hansson Ö. Maccarrone G. Pappalardo G. Pietropaolo A. Rizzarelli E. Copper(II) complexes with an avian prion N-terminal region and their potential SOD-like activity J. Inorg. Biochem. 2009 103 195 204 10.1016/j.jinorgbio.2008.10.002 19019452 45. Jiskoot W. Crommelin D.J.A. Methods for Structural Analysis of Protein Pharmaceuticals AAPS Press New York, NY, USA 2005 46. La Mendola D. Magrì A. Hansson Ö. Bonomo R.P. Rizzarelli E. Copper(II) complexes with peptide fragments encompassing the sequence 122–130 of human doppel protein J. Inorg. Biochem. 2009 103 758 765 10.1016/j.jinorgbio.2009.01.017 19237200 47. Pietropaolo A. Muccioli L. Zannoni C. La Mendola D. Maccarrone G. Pappalardo G. Rizzarelli E. Unveiling the role of histidine and tyrosine residues on the conformation of the avian prion hexarepeat domain J. Phys. Chem. B 2008 112 5182 5188 10.1021/jp710702q 18386869 48. Kowalik-Jankowska T. Ruta-Dolejsz M. Wiśniewska K. Łankiewicz L. Kozłowski H. Copper(II) complexation by human and mouse fragments (11–16) of β-amyloid peptide J. Chem. Soc. Dalton Trans. 2000 4511 4519 10.1039/b006125p 49. Kowalik-Jankowska T. Ruta M. Wiśniewska K. Lankiewicz L. Coordination abilities of the 1–16 and 1–28 fragments of β-amyloid peptide towards copper(II) ions: A combined potentiometric and spectroscopic study J. Inorg. Biochem. 2003 95 270 282 10.1016/S0162-0134(03)00128-4 12818797 50. Kowalik-Jankowska T. Rajewska A. Wiśniewska K. Grzonka Z. Jezierska J. Coordination abilities of N-terminal fragments of α-synuclein towards copper(II) ions: A combined potentiometric and spectroscopic study J. Inorg. Biochem. 2005 99 2282 2291 10.1016/j.jinorgbio.2005.08.008 16203037 51. La Mendola D. Magrì A. Campagna T. Campitiello M.A. Raiola L. Isernia C. Hansson O. Bonomo R.P. Rizzarelli E. A doppel α-helix peptide fragment mimics the copper(II) interactions with the whole protein Chemistry 2010 16 6212 6223 10.1002/chem.200902405 20411530 52. Bertini I. Luchinat C. Parigi G. Pierattelli R. NMR spectroscopy of paramagnetic metalloproteins Chembiochem 2005 6 1536 1549 10.1002/cbic.200500124 16094696 53. Travaglia A. Arena G. Fattorusso R. Isernia C. La Mendola D. Malgieri G. Nicoletti V.G. Rizzarelli E. The inorganic perspective of nerve growth factor: Interactions of Cu2+ and Zn2+ with the N-terminus fragment of nerve growth factor encompassing the recognition domain of the TrkA receptor Chemistry 2011 17 3726 3738 10.1002/chem.201002294 21394800 54. Pietropaolo A. Satriano C. Strano G. La Mendola D. Rizzarelli E. Different zinc(II) complex species and binding modes at Aβ N-terminus drive distinct long range cross-talks in the Aβ monomers J. Inorg. Biochem. 2015 153 367 376 10.1016/j.jinorgbio.2015.08.013 26298865 55. Amorini A.M. Bellia F. Di Pietro V. Giardina B. La Mendola D. Lazzarino G. Sortino S. Tavazzi B. Rizzarelli E. Vecchio G. Synthesis and antioxidant activity of new homocarnosine β-cyclodextrin conjugates Eur. J. Med. Chem. 2007 42 910 920 10.1016/j.ejmech.2006.12.036 17336428 56. Gans P. Sabatini A. Vacca A. Investigation of equilibria in solution. Determination of equilibrium constants with the HYPERQUAD suite of programs Talanta 1996 43 1739 1753 10.1016/0039-9140(96)01958-3 18966661 57. Alderighi L. Gans P. Ienco A. Peters D. Sabatini A. Vacca A. Hyperquad simulation and speciation (HySS): A utility program for the investigation of equilibria involving soluble and partially soluble species Coord. Chem. Rev. 1999 184 311 318 10.1016/S0010-8545(98)00260-4 58. Remelli M. Peana M. Medici S. Ostrowska M. Gumienna-Kontecka E. Zoroddu M.A. Manganism and Parkinson’s disease: Mn(II) and Zn(II) interaction with a 30-amino acid fragment Dalton Trans. 2016 45 5151 5161 10.1039/C6DT00184J 26885710 59. Zoroddu M.A. Peana M. Medici S. Anedda R. An NMR study on nickel binding sites in Cap43 protein fragments Dalton Trans. 2009 28 5523 5534 10.1039/b903305j 19587996 60. Satriano C. Carnazza S. Licciardello A. Guglielmino S. Marletta G. Cell adhesion and spreading on polymer surfaces micropatterned by ion beams J. Vac. Sci. Technol. A 2003 21 1145 1151 10.1116/1.1575217

PMC005xxxxxx/PMC5000639.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081241ijms-17-01241ArticleEvaluation of Pulmonary Toxicity of Zinc Oxide Nanoparticles Following Inhalation and Intratracheal Instillation Morimoto Yasuo 1*Izumi Hiroto 1Yoshiura Yukiko 1Tomonaga Taisuke 1Oyabu Takako 2Myojo Toshihiko 2Kawai Kazuaki 3Yatera Kazuhiro 4Shimada Manabu 5Kubo Masaru 5Yamamoto Kazuhiro 6Kitajima Shinichi 7Kuroda Etsushi 8Kawaguchi Kenji 6Sasaki Takeshi 6Sivakov Vladimir Academic Editor1 Department of Occupational Pneumology, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan; [email protected] (H.I.); [email protected] (Y.Y.); [email protected] (T.T.)2 Department of Environmental Health Engineering, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan; [email protected] (T.O.); [email protected] (T.M.)3 Department of Environmental Oncology, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan; [email protected] Department of Respiratory Medicine, School of Medicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahata-nishi-ku, Kitakyushu, Fukuoka 807-8555, Japan; [email protected] Department of Chemical Engineering, Hiroshima University, Higashi-Hiroshima 739-8528, Japan; [email protected] (M.S.); [email protected] (M.K.)6 National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan; [email protected] (K.Y.); [email protected] (K.K.); [email protected] (T.S.)7 National Sanatorium Hoshizuka Keiaien, 4204 Hoshizuka-cho, Kanoya, Kagoshima 893-8502, Japan; [email protected] Laboratory of Vaccine Science, WPI Immunology Frontier Research Center, 6F IFReC Research Building, Osaka University, 3-1 Yamada-oka, Suita, Osaka 565-0871, Japan; [email protected]* Correspondence: [email protected]; Tel.: +81-93-691-713601 8 2016 8 2016 17 8 124114 6 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).We conducted inhalation and intratracheal instillation studies of zinc oxide (ZnO) nanoparticles in order to examine their pulmonary toxicity. F344 rats were received intratracheal instillation at 0.2 or 1 mg of ZnO nanoparticles with a primary diameter of 35 nm that were well-dispersed in distilled water. Cell analysis and chemokines in bronchoalveolar lavage fluid (BALF) were analyzed at three days, one week, one month, three months, and six months after the instillation. As the inhalation study, rats were exposed to a concentration of inhaled ZnO nanoparticles (2 and 10 mg/m3) for four weeks (6 h/day, 5 days/week). The same endpoints as in the intratracheal instillation study were analyzed at three days, one month, and three months after the end of the exposure. In the intratracheal instillation study, both the 0.2 and the 1.0 mg ZnO groups had a transient increase in the total cell and neutrophil count in the BALF and in the expression of cytokine-induced neutrophil chemoattractant (CINC)-1, CINC-2, chemokine for neutrophil, and heme oxygenase-1 (HO-1), an oxidative stress marker, in the BALF. In the inhalation study, transient increases in total cell and neutrophil count, CINC-1,-2 and HO-1 in the BALF were observed in the high concentration groups. Neither of the studies of ZnO nanoparticles showed persistent inflammation in the rat lung, suggesting that well-dispersed ZnO nanoparticles have low toxicity. zinc oxidenanoparticleCINCneutrophilintratracheal instillationlunginhalation ==== Body 1. Introduction Various applications of nanomaterials, including metal oxide nanoparticles, have been enabled by new characteristics that have resulted from the progress of nanotechnology. Zinc oxide (ZnO) nanoparticles are white powders that are widely used in cosmetics, paint pigment, rubber additives, pharmaceutical products, and electronic materials. Many in vitro studies [1,2,3] and in vivo studies [4,5,6] have concluded that ZnO nanoparticles have a strong potential of toxicity, but these results are insufficient and controversial because the endpoints of toxicity in many such studies reflect acute responses, such as cytotoxicity and acute inflammation. In lung disorders caused by inhaled particle, phagocytosis of inhaled particle induces infiltration of neutrophils and alveolar macrophages, and sustained or progressive inflammation is likely to cause lung injury and lead to irreversible chronic lesions, such as fibrosis and tumors [7,8,9]. Persistent inflammation, reported in animal exposure models using asbestos and silica, is important in the pathology of the formation of irreversible chronic lesions [7,9]. Most of the reports on ZnO nanoparticles show acute pulmonary inflammation in vivo and cytotoxicity in vitro, suggesting that ZnO nanoparticles may have harmful effects on humans [1,2,3,4,6,7]. However, it is also important to examine chronic responses, such as persistent inflammation. There are reports [10,11] that exposure to crystalline silica, a material known to have high toxicity, induced the onset of pulmonary inflammation after a certain observation time and more severe inflammation in the chronic phase. Considering the pulmonary toxicity of nanomaterials, it is important to evaluate the endpoints, such as inflammation and fibrosis, not only in the acute but also in the chronic phase. Therefore, we performed intratracheal instillation and inhalation studies of ZnO nanoparticles with more than three months of observation periods and examined pulmonary inflammation and fibrosis as the endpoints of toxicity in order to examine the pulmonary toxicity of ZnO nanoparticles. 2. Results 2.1. Intratracheal Instillation Study 2.1.1. Cell Analysis in Bronchoalveolar Lavage Fluid (BALF) Figure 1 shows the cellular analysis of the BALF following the intratracheal instillation of ZnO nanoparticles. The total cell count and neutrophil counts in the BALF were significantly and dose-dependently higher in the 1 mg group from three days post exposure compared with the negative control. The peak level of these counts was at three days, and they returned to nearly the level of the negative control according to a time course. The macrophage counts in the BALF were also higher in the ZnO groups than in the negative control at three days post exposure, although not dose-dependently. This response was also transient. A transient increase in the released Lactate Dehydrogenase (LDH) activity was observed in the 0.2 and 1 mg groups. This LDH activity was high, but decreased thereafter to nearly the same level as the negative control groups after one month. 2.1.2. Cytokine-Induced Neutrophil Chemoattractant (CINC) Concentration in BALF Figure 1E,F show the concentrations of CINC-1 and CINC-2 in the BALF following the intratracheal instillation of ZnO nanoparticles. The concentrations of CINC-1 were dose-dependently high in both the 0.2 and the 1 mg groups at three days post exposure, but at one month and three months post exposure, the value of CINC-1 in the ZnO groups was lower than that in negative control group. The concentrations of CINC-2 in the 0.2 and 1 mg groups were transiently higher than in the negative control group at three days post exposure, as like the concentration of CINC-1, and at one month and three months post exposure, the value of CINC-2 in the ZnO groups was lower than that in the negative control group. 2.1.3. Heme Oxigenase-1 (HO-1) Concentration in BALF Figure 1G shows the concentration of HO-1 in the BALF following the intratracheal instillation of ZnO nanoparticles. The concentration increased at three days post exposure, but there was no difference compared to the negative control group during the observation period after three days. 2.1.4. Histopathological Changes in the Lungs The lung specimens on day three showed infiltration of macrophages and inflammatory cells in the alveoli around the terminal bronchioles (Table 1). More inflammatory cell infiltration was observed in the lungs of the ZnO 1 mg instillation rats (Figure 2A) than in those of the 0.2 mg installation rats. Particle-laden macrophages were observed among the inflammatory cells, but the inflammation diminished one month after installation (Figure 2B). Minimal fibrosis occurred after inflammation, which disappeared three months after instillation. Some particle-laden macrophages were distributed around the alveolar ducts and the surrounding alveolar spaces. 2.1.5. Morphological Features of Alveolar Macrophages by TEM Figure 3A–D shows TEM images of the alveolar space near the inflammation in the high dose ZnO instillation group lung tissue at three days after exposure. Accumulation of alveolar macrophages and neutrophil cells can be seen (Figure 3A). TEM images of the inflammation are shown in Figure 3C, and neutrophil cells can be observed in the alveolar space. Figure 3D is a magnified image of the boxed area in Figure 3C. Black particles formed aggregates in the cell organelles, as indicated by the arrow. The shape of these nanoparticles indicate that they are ZnO nanoparticles, and that instilled ZnO nanoparticles reached the alveolar space. Figure 3B also shows an accumulation of alveolar macrophages, and many vacuoles can be observed. In our previous studies on the intratracheal instillation of NiO, TiO2, and CeO2 nanoparticles into rat lung, uptake of nanoparticles into alveolar macrophages was observed, but no ZnO nanoparticles were observed there in the present study. We speculate that ZnO nanoparticles were dissolved in the alveolar macrophages. 3. Inhalation Study 3.1. Cell Analysis in BALF Figure 4A–C shows the cellular analysis of the BALF following the inhalation of ZnO nanoparticles. The total cell, neutrophil and macrophage counts were higher in the 1 mg groups than in the negative control groups at three days. The value decreased to nearly the negative control level at 1–3 months. The pattern of LDH activity (Figure 4D) was the same as in the cellular analysis. There was a significant increase in LDH activity in the high concentration group at three days, but no significant increase in LDH activity was observed in the ZnO compared to the negative control groups in the other time courses. 3.2. CINC Concentration in BALF Figure 4E,F show the concentrations of CINC-1 and CINC-2 in the BALF following the inhalation of ZnO nanoparticles. Both values in the high concentration groups were significantly elevated at three days post exposure, but the values in the ZnO groups were lower than that in negative control group after one month. 3.3. HO-1 Concentration in BALF Figure 4G shows the concentration of HO-1 in the BALF following the inhalation of ZnO nanoparticles. The concentration of HO-1 in the high concentration groups was higher than that in the negative control group at three days post exposure. There were no significant differences in the concentration of HO-1 between the ZnO and the negative control groups in any other time course. 3.4. Histopathological Changes in the Lungs Mild inflammation was induced in small areas of the lungs in the high dose inhalation mice after three days of inhalation (Figure 5A) (Table 2), but there was no significant inflammation after one or three months (Figure 5B,C) (Table 2), nor in any period in the low dose groups. Foamy macrophages and particle-laden macrophages were observed in the alveoli, and some macrophages fused and formed multinucleated cells. 3.5. Morphological Features of Alveolar Macrophages by TEM TEM images of alveolar macrophages in the high dose ZnO inhalation group lung tissue at three days after exposure are shown in Figure 6A,B. Many vacuoles were observed in the alveolar macrophages. No ZnO nanoparticles were seen in the alveolar macrophages, the same as in the TEM observation in the intratracheal instillation study. Accumulation of alveolar macrophages (Figure 6C) and neutrophil cells (Figure 6D) was observed in the alveolar space. 4. Discussion In the present study, exposure to ZnO nanoparticles following intratracheal instillation and inhalation transiently induced neutrophil inflammation in the rat lung in the acute phase. Many in vivo studies [4,5,6] have shown pulmonary inflammation in animal models. Ho et al. [5] reported that inhalation of not only nanoscale, but also submicron, ZnO induced acute inflammation in the rat lung, and showed that both mass and surface area were affected by the influx of neutrophils in the lung. In vitro studies [1,2,3] have also shown that ZnO induced high cytotoxicity. Lu et al. reported [3] that, among PM and metal oxide nanoparticles, the highest lactate dehydrogenase level was caused by nano-ZnO particles in the A549 cell line (human alveolar adenocarcinoma cell line). The acute inflammatory level in the present study was approximately two times higher than that by nickel oxide and cerium oxide nanoparticles in our previous studies. We speculate that Zn ions, dissolved by ZnO nanoparticles, affected these high inflammatory responses. Kondura et al. [12] reported that the pulmonary clearance of ZnO nanoparticles in the lung following intratracheal instillation was biphasic, and that both rapid initial and slower terminal half times of ZnO nanoparticles were less than two days. Adamcakova-Dodd [13] showed that 100% of ZnO nanoparticles dissolved within the first 24 h of mixing in an artificial interstitial fluid (pH 4.5). In addition, copper oxide nanoparticles, which are considered to have high solubility, were reported to induce inflammation in the lung through dissolution [14]. These pulmonary responses were based on acute responses, and if the inflammogenic potential of nanoparticles is considered to lead to fibrosis and carcinoma in the lung, sustained inflammation is an important endpoint to speculate the harmful effect of nanoparticles. Even nanoparticles with low toxicity, such as titanium dioxide nanoparticles [15,16,17] and fullerene, induced transient inflammation in rat lung following intratracheal instillation, but not after inhalation. On the other hand, chemicals with high toxicity, such as asbestos and crystalline silica, induced persistent or progressive inflammation mainly by neutrophils, causing irreversible chronic lesions, such as fibrosis and tumors [7,9,18,19]. If the initial lung burden of ZnO following inhalation is calculated by the Multi-Path Particle Model (MPPD model) [20], the initial lung burden in the low and high concentrations following four weeks of inhalation was 0.269 and 1.302 mg/rat (data: low concentration, count median diameter (CMD) 0.126 µm (geometric standard deviation (GSD) 1.77) 2.1 mg/m3; high concentration, CMD 0.148 µm (GSD 1.79) 10.4 mg/m3), respectively. We think that the initial lung burden in the low and high concentrations approximately correspond to the low and high doses of injected ZnO nanoparticles in the intratracheal instillation study. Compared with either concentrations of ZnO nanoparticles in the inhalation study, inflammatory responses, such as cell analysis, chemokines, and oxidative stress in BALF in both the doses in the intratracheal instillation study were the same, or higher, qualitative level. The bolus effect may have resulted in the values of the data in the intratracheal instillation being higher than those in the inhalation. These tendencies of difference between intratracheal instillation and inhalation studies were also observed in exposure to nickel oxide, titanium dioxide, and multi-wall carbon nanotube (MWCNT) [16,21,22,23,24]. The observation period is important when examining the sustainability of inflammation in an animal model, and we arranged for an observation period of at least three months in the present study. Acute responses were not observed just after the end of exposure to the chemical, and the onset and peak of inflammation were observed after a certain period of observation in animal models. Intratracheal exposure of nickel oxide nanoparticles induced pulmonary inflammation in rats, and the peak of inflammation was at three months post exposure [23]. Sellamuthu et al. [11] reported that the number of neutrophils and the concentration of MCP-1 in the BALF were at the maximum at 16 weeks following inhalation of crystalline silica. Langley et al. [10] performed a six-week inhalation study of silica with 27 weeks of post exposure, and the counts of neutrophils and lymphocytes in the BALF was high at 10 weeks post exposure, although not at four days, and the LDH and protein concentrations in the BALF were significantly higher at 10 and 17 weeks, but not at four days. Pan et al. [25] performed pulmonary protein profiles in response to ZnO nanoparticles at 24 h and 28 days post exposure following intratracheal instillation, and found that detoxification pathways were activated at the 28-day time-point after exposure, suggesting that insufficient recovery response may develop into irreversible lesions. As we saw no chronic responses through at least three months of observation following both approaches, we assessed that the inflammation induced by ZnO nanoparticles following both approaches was transient. Both pathological features and cell analysis in BALF showed the same transient responses, and both signs were in accordance with each other. We also examined the concentration of CINC-1 and CINC-2, representative chemokines for neutrophils, in the BALF exposed to ZnO nanoparticles. Exposure to ZnO nanoparticles following intratracheal instillation and inhalation-induced transient elevation of CINC-1 and CINC-2 accompanied by neutrophil influx. In our previous studies [16,23], the results of neutrophil concentration in BALF showed that the inhalation exposure of NiO and CeO2 upregulated the concentration of CINC-1 and CINC-2, but TiO2 did not, and the intratracheal instillation of NiO, CeO2, and TiO2 induced persistent and transient concentration of CINC-1 and CINC-2 in BALF, respectively. These patterns of inflammation by these metal oxide nanoparticles were accompanied by changes in the concentration of CINC-1 and CINC-2. The transient responses in CINC-1 and CINC-2 expression accompanied by neutrophil inflammation in the present study may correspond to previous studies. HO-1 is known to be one of the representative biomarkers that affect oxidative stress. Li et al. [26] reported that in a dithiothreitol (DTT) assay a quantitative measure of in vitro ROS formation correlated with HO-1 expression in the Abeison murine leukemia virus-induced tumor (RAW264.7) cell line exposed to ultrafine particulate pollutants, in the murine macrophage cell line, human bronchial epithelial cell (BEAS-2B cell) line, and in the human bronchial epithelial cell line. In the present study, both exposures of ZnO nanoparticles induced transient elevation of HO-1 concentration in BALF. Like the CINC family, the chemicals with high toxicity induced persistent elevation of HO-1 expression, while the chemicals with low toxicity induced transient, or no, elevation. Considering the expression pattern of HO-1, we speculate that ZnO nanoparticles may have a low inflammatory potential. However, if there is strong oxidative stress in the acute phase of ZnO nanoparticle exposure, there may be a potential for genetic disorders, such as driver gene mutation. As pathological features, only mild and transient hyperplasia was observed, and oxidative DNA injury was not observed (data not shown), suggesting that oxidative stress from ZnO nanoparticles may not be strong, nor would there be induction of genetic disorder. We look forward to future research on the relationship between ZnO nanoparticles and driver gene mutation. 5. Methods and Materials 5.1. Sample Preparation of ZnO Nanoparticle Suspensions Commercial ZnO nanoparticle dispersion (Sigma-Aldrich Co. LLC., Tokyo, Japan, 51 wt % ZnO) with a water dispersion medium was employed as a source material. The source dispersion contained 2 wt % 3-aminopropyltriethoxysilane as a dispersing agent according to the company’s datasheet. Since no information about the purity was given by the company, we asked Sumika Chemical Analysis Service (Tokyo, Japan) for a purity analysis, and 99.94 wt % purity was reported. The source dispersion was diluted to 10 mg/mL with deionized endotoxin-free water, and was well homogenized by 2 h ultrasonic homogenizing (Branson 5510J-MT, Yamato Scientific Co., Ltd. Tokyo, Japan, 42 kHz 180 W). The prepared dispersion showed a simple secondary particle diameter distribution around the primary particle diameter without agglomeration, as shown in Figure 7A. The average for 500 particles of 35 nm given in Table 3 corresponded to the company’s datasheet (<35 nm). The average secondary particle diameter measured by dynamic light scattering (DLS) for nine samples was 33 nm, which was approximately the same as the primary size, as listed in Table 3, meaning that the ZnO nanoparticles were well dispersed in the suspension. Though the high concentration dispersion with 10 mg/mL was stable for more than a week, the lower concentration (less than 2 mg/mL) sometimes showed agglomeration (larger than 3 µm diameter) within a week. Therefore, 10 mg/mL dispersions were prepared weekly and diluted to the actual experimental conditions (0.6–5 mg/mL) by 20 min ultrasonic homogenizing just before the experiments. The ZnO nanoparticle suspensions were observed by a transmission electron microscope (TEM, EM922, Carl Zeiss, Jena, Germany). The accelerating voltage was 160 kV. The TEM specimens were prepared on TEM grids with carbon support films by dropping suspensions and then drying them. TEM images of the ZnO nanoparticle suspensions are shown in Figure 7B. The primary particle size of the ZnO was between 15 and 50 nm. This size distribution was in good agreement with the DLS measurement shown in Figure 7A. Most of the ZnO particles were mono-dispersed, however some made up aggregates from a few primary particles, which were less than 100 nm in size. A high-resolution TEM image of the ZnO primary particles is shown in Figure 7C, and a magnified image of C is shown in Figure 7D. The ZnO particles had a clear crystalline form by the high-resolution TEM image, and it was clarified that no damage was caused by the preparation processes. 5.2. Animals Male Fischer 344 rats (from 9–11 weeks old) were purchased from Charles River Laboratories Japan, Inc. (Yokohama, Kanagawa, Japan). All animals were acclimated in the Laboratory Animal Research Center of the University of Occupational and Environmental Health for at least one week prior to use. All experimental procedures were conducted in accordance with the guidelines described in the Japanese Guide for the Care and Use of Laboratory Animals as approved by the Animal Care and Use Committee, University of Occupational and Environmental Health, Japan (AE12-004, AE12-005). 5.3. Intratracheal Instillation of ZnO Nanoparticles The ZnO nanoparticles were suspended with 0.4 mL distilled water. Rats (12 weeks old) were exposed to 0.2 mg/rat (0.8 mg/kg) or 1 mg/rat (4 mg/kg) of ZnO nanoparticles intratracheally. The negative control groups received distilled water. Low dose (0.2 mg/rat) and high dose (1 mg/rat) were the dose of minimum level which nanomaterials with high toxicity and low toxicity induced minimum persistent inflammation in rat lung following intratracheal instillation [8,27]. Animals were dissected at three days, one week, one month, three months, and six months after the instillation. 5.4. Inhalation of ZnO Nanoparticles ZnO aerosol particles were supplied for the inhalation test at two target concentrations (10 and 2 mg/m3). The ZnO nanoparticle suspensions were sprayed with a pressurized nebulizer and dried to disperse the particles in the air flow. They were then delivered into a whole body exposure chamber attached to the rat cages. The setup used here has been described in more detail in our previous papers [28,29]. ZnO suspensions at concentrations of 3–5 and 0.6–0.8 mg/mL were used for the high- and low-dose chambers, respectively. Each of the suspensions was sprayed with the nebulizer at a rate of 0.8 mL/min, using the flow of compressed air at 40 L/min. The droplets generated from the spraying were mixed with 15 L/min of air containing bipolar airborne ions to reduce the electrical charge and, thus the electrically-enhanced loss of droplets. The droplets were successively passed through a heated (150 °C) tube to remove water from them. Clean air was added to the resulting aerosol flow to set the total airflow rate to 100 L/min. This aerosol flow was admitted into the exposure chamber for 6 h per day. The inhalation test period was four weeks. A particle size spectrometer (model 1000XP WPS, MSP Corp., Shoreview, MN, USA) was used to measure the aerosol particle size distribution in the exposure chambers twice an hour. A small amount of the aerosol was sampled periodically outside of the chamber. The particles in the aerosol were deposited onto a Cu TEM grid and subjected to TEM observation. Particles were also collected on a fibrous filter and weighed to determine the mass concentration of the aerosols in the chambers, which took place 3–5 times per day. After the inhalation test period, the rats were dissected after three days, one month, and three months of recovery. The particle size distributions of the ZnO aerosols in both the high- and low-dose chambers were sufficiently stable for 6 h on every day of the test period The geometric mean diameter of the aerosol particles averaged for the test period was 148 ± 14 nm (n = 480) for the high-dose chamber, and 126 ± 11 nm (n = 480) for the low-dose chamber. The average mass concentrations in the test period were 10.4 ± 1.39 mg/m3 (n = 73) and 2.11 ± 0.45 mg/m3 (n = 70) for the high- and low-dose chambers, respectively. Figure 8A–C show typical TEM images of the ZnO aerosol particles sampled from the high dose chamber. The particles were aggregates, and their sizes were mostly in the range of 50 and 300 nm, with a peak at around 150 nm. This was consistent with the result obtained with the particle size spectrometer. Crystal lattices can clearly be seen in a high resolution TEM image (Figure 8C), indicating that the aerosol generation process did not cause any damage to the ZnO particles. 5.5. Animals after the Inhalation and Intratracheal Instillation Studies There were 10 rats, classified into two subgroups of five animals each, in the negative control, low-dose, and high-dose groups in each time course for BALF and lung tissue analysis. In the first subgroups (five animals in each dose group), the lungs were divided into right and left lungs. Histopathological evaluation was performed with the left lung inflated and fixed by 10% formaldehyde. In the second subgroups (five animals in each dose group), the lungs were inflated with physiological saline with 20 mL water under a pressure of 20 cm, and recovered fluid was collected from whole lung divided two to three times. Between 15 and 18 mL of the recovered fluid was collected in collection tubes by free fall. Analysis of cytokine was performed with BALF. 5.6. Analysis of Inflammatory Cells in BALF From 10–13 mL of recovered BALF was centrifuged at 400× g at 4 °C for 15 min. The supernatant was transferred to a new tube and used for measuring the cytokines in the BALF. The pellets were washed by suspension with polymorphonuclear leukocyte (PMN) Buffer (137.9 mM NaCl, 2.7 mM KCl, 8.2 mM Na2HPO4, 1.5 mM KH2PO4, 5.6 mM C6H12O6) and centrifuged at 400× g at 4 °C for 15 min. After the supernatant was removed, the pellets were resuspended with 1 mL of PMN Buffer. The total cell numbers in the BALF was counted by Celltac (Nihon Kohden, Tokyo, Japan), and the cells were splashed on a slide glass using cytospin. After the cells were fixed and stained with Diff-Quik (System Corporation, Hyogo, Japan), the number of neutrophils and alveolar macrophages was counted by microscopic observation. 5.7. Chemokines, LDH, and HO-1 in BALF The concentrations of rat chemoattractant (CINC)-1 and rat CINC-2α/β in the BALF supernatant were measured by ELISA kits #RCN100 and #RCN200 (R and D Systems, Minneapolis, MN, USA), respectively. The concentrations of rat HO-1 in BALF supernatant were measured by an ELISA kit, ADI-EKS-810A (Enzo Life Sciences, Farmingdale, NY, USA), and the activity of released LDH in BALF supernatant was measured by a Cytotoxicity Detection KitPLUS(LDH) (Roche Diagnostics GmbH, Mannheim, Germany). The LDH activity in BALF supernatant was determined in an enzymatic test. All procedures were performed according to the manufacturer’s instructions. 5.8. Histopathology The lung tissue, which was inflated and fixed with 10% formaldehyde under a pressure of 25 cm water, was dehydrated and embedded in paraffin, and 5 µm-thick sections were cut from the lobe, then stained with hematoxylin and eosin. 6. TEM Experimental Methods Lung tissues were observed by TEM after the inhalation and intratracheal instillation studies. The TEM specimen preparation method is described below. The lung tissues were fixed by a perfusion system of a 4% paraformaldehyde solution, and then were post-fixed in a 1% osmium tetroxide solution. They were dehydrated in ethanol subsequently, followed by embedding in epoxy resin. Ultrathin sections were cut by using a diamond knife using microtomy. The specimens were stained with a 2% uranyl acetate solution, and then a mixed solution of 0.3% lead nitrate and 0.3% lead acetate. All of them were prepared at room temperature. Conventional TEM observation was performed with an H-7600 (Hitachi High-Technologies Corp., Tokyo, Japan). The accelerating voltage was 80 kV. Statistical Analysis Analysis of Mann-Whitney test were applied where appropriate to determine individual differences using a computer statistical package (SPSS, SPSS Inc., Chicago, IL, USA). 7. Conclusions We conducted inhalation and intratracheal instillation of ZnO nanoparticles in order to examine their toxicity. In the intratracheal instillation study, F344 rats were exposed to 0.2 or 1 mg of ZnO nanoparticles. In the inhalation study, rats inhaled ZnO nanoparticles at a maximum concentration of 10 mg/m3 for four weeks. The intratracheal instillation and the inhalation of a high dose of ZnO nanoparticles caused a transient increase in neutrophil influx in the lung and a transient increase in concentration of CINC-1, CINC-2, and HO-1 in BALF in the acute phase. These parameters returned to control level in the chronic phase, and reversible inflammation of neutrophils in the lung was observed by both approaches. The transient inflammation in the lung exposed to ZnO nanoparticles suggests that ZnO nanoparticles may have a low toxic potential. Acknowledgments This work was supported by “Development of Innovative Methodology for Safety Assessment of Industrial Nanomaterials” by the Ministry of Economy, Trade and Industry (METI) of Tokyo, Japan. Author Contributions Yasuo Morimoto, Hiroto Izumi, Toshihiko Myojo and Kazuaki Kawai conceived and designed the experiments; Yukiko Yoshiura, Taisuke Tomonaga and Takako Oyabu performed the animal experiments; Manabu Shimada and Masaru Kubo monitored nano-aerosol; Kazuhiro Yamamoto performed electric microscopy; Kazuhiro Yatera and Etsushi Kuroda discussed the results; Shinichi Kitajima performed microscopy; Kenji Kawaguchi and Takeshi Sasaki analyzed characteristics of materials the data; Yasuo Morimoto wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Cell number and cytokine level in bronchoalveolar lavage fluid (BALF) following intratracheal instillation of ZnO nanoparticles. (A) Total cell count in BALF; (B) neutrophil count in BALF; (C) macrophage count of in BALF; (D) lactate dehydrogenase (LDH) activity in BALF; (E) concentration of chemoattractant (CINC)-1 in BALF; (F) concentration of CINC-2 in BALF; and (G) concentration of heme oxigenase-1 (HO-1) in BALF. Intratracheal instillation of ZnO nanoparticles induced transient influx of inflammatory cells and expression of CINC-1, CINC-2 and HO-1 in BALF. * indicates p < 0.05 compared to negative control. ** indicates p < 0.01 compared to negative control. Figure 2 Histological changes in lungs of 1.0 mg-administered group (40×, inset 200×). (A) three days post exposure; (B) one month post exposure; and (C) three months post exposure. Bronchopneumonia was observed three days after intratracheal instillation of ZnO nanoparticles. Figure 3 Lung tissue TEM images in the high dose exposed group at three days following intratracheal instillation. (A) Accumulation of alveolar macrophages and neutrophil cells in alveolar space; (B) accumulation of alveolar macrophages with vacuoles; (C) neutrophil cell and inflammatory cells; and (D) magnified image of boxed area in (C). Arrow: black particles formed aggregates in the cell organelles. Figure 4 Cell number and cytokine level in BALF following intratracheal instillation of ZnO nanoparticles. (A) Total cell count in BALF; (B) neutrophil count in BALF; (C) macrophage count of in BALF; (D) LDH activity in BALF; (E) concentration of CINC-1 in BALF; (F) concentration of CINC-2 in BALF; (G) concentration of HO-1 in BALF. Inhaled ZnO nanoparticles at high concentration transiently induced the influx of inflammatory cells such as neutrophils and expression of CINC-1, CINC-2, and HO-1 in BALF. * indicates p < 0.05 compared to negative control. ** indicates p < 0.01 compared to negative control. Figure 5 Histological changes in lungs of high dose-inhalation group (40×, inset 200×). (A) three days post exposure; (B) one month post exposure; and (C) three months post exposure. Inflammation of three days after inhalation exposure is milder than that of three days after instillation exposure. Figure 6 Lung tissue TEM images in the high concentration group at three days following inhalation. (A,B) Alveolar macrophages with vacuoles in alveolar space; (C) Accumulation of alveolar macrophages; (D) Neutrophil cells in alveolar space. Figure 7 Zinc oxide (ZnO) nanoparticles suspended in distilled water. (A) Size distribution of particles was determined by dynamic light scattering technique; (B) Low magnification image of ZnO nanoparticles by transmission electron microscopy; (C) High magnification TEM image of ZnO nanoparticles; (D) Magnified image of (C). The crystalline lattice can be clearly observed. Figure 8 Inhaled ZnO nanoparticles in exposure chambers by transmission electron microscopy (A,B); (C) High magnification TEM image of ZnO nanoparticles. ijms-17-01241-t001_Table 1Table 1 Pathological features in the rat lung following intratracheal instillation of ZnO nanoparticles. Time 3 Days (n = 5) 1 Week (n = 5) 1 Month (n = 5) 3 Months (n = 5) 6 Months (n = 5) Pathological Feature Negative Control ZnO 0.2 mg ZnO 1.0 mg Negative Control ZnO 0.2 mg ZnO 1.0 mg Negative Control ZnO 0.2 mg ZnO 1.0 mg Negative Control ZnO 0.2 mg ZnO 1.0 mg Negative Control ZnO 0.2 mg ZnO 1.0 mg Macrophage infiltration in alveolar space − ++ ++ − + + − ± ± − − ~ ± − ~ ± − − ~ ± − ~ ± Inflammatory cell infiltration in alveolar space − ++ +++ − + + − − − ~ ± − − − − − − Infiltration in interstitial area − + ++ − ± ± − − − ~ ± − − − − − − Hyperplasia of bronchiolar epithelial cell − + + − − ~ ± − ~ ± − − ~ ± − ~ ± − − − ~ ± − − − ~ ± Hyperplasia of alveolar epithelial cell − ++ ++ ~ +++ − ± ± − − − − − − − − − Fibrosis − ± ~ + ± ~ + − ± ± − − − ~ ± − − − ~ ± − − − ~ ± Tumor − − − − − − − − − − − − − − − Grade of changes: −, none; ±, minimum; +, mild; ++, moderate; +++, remarked. ijms-17-01241-t002_Table 2Table 2 Pathological features in the rat lung following inhalation of ZnO nanoparticles. Time 3 Days (n = 5) 1 Month (n = 5) 3 Months (n = 5) Pathological Feature Negative Control ZnO Low ZnO High Negative Control ZnO Low ZnO High Negative Control ZnO Low ZnO High Macrophage infiltration in alveolar space − + ++ − ± + − ± ± Inflammatory cell infiltration in alveolar space − − − ~ ± − − − − − − Infiltration in interstitial area − − − ~ ± − − − − − − Hyperplasia of bronchiolar epithelial cell − − − ~ ± − − − − − − Hyperplasia of alveolar epithelial cell − − − − − − − − − Fibrosis − − − − − − − − − tumor − − − − − − − − − Grade of changes: −, none; ±, minimum; +, mild; ++, moderate. ijms-17-01241-t003_Table 3Table 3 Physicochemical properties of zinc oxide (ZnO) nanoparticles used in the experiment. Nanomaterials ZnO Nanoparticle Manufacturer Sigma-Aldrich Co. LLC. Chemical formula ZnO Primary diameter 35 nm Specific surface area 31 m2/g Shape Polyhedral roughly round Secondary diameter (DLS) 33 nm Purity 99.94 wt % Bulk density 5.6 g/cm3 Solubility high ==== Refs References 1. Kim I.S. Baek M. Choi S.J. Comparative cytotoxicity of Al2 O3 , CeO2 , TiO2 and ZnO nanoparticles to human lung cells J. Nanosci. Nanotechnol. 2010 10 3453 3458 10.1166/jnn.2010.2340 20358977 2. Kim Y.H. Fazlollahi F. Kennedy I.M. Yakobi N.R. Hamm-Alvarez S.F.H. Borok Z. Kim K.J. Crandall E.D. Alveolar epithelial cell injury due to zinc oxide nanoparticle exposure Am. J. Respir. Crit. Care Med. 2010 182 1398 1409 10.1164/rccm.201002-0185OC 20639441 3. Lu S. Zhang W. Zhang R. Liu P. Wang Q. Shang Y. Wu M. Donaldson K. Wang Q. Comparison of cellular toxicity caused by ambient ultrafine particles and engineered metal oxide nanoparticles Part Fibre Toxicol. 2015 12 5 10.1186/s12989-015-0082-8 25888760 4. Fukui H. Horie M. Endoh S. Kato H. Fujita K. Nishio K. Komaba L.K. Maru J. Miyauhi A. Nakamura A. Association of zinc ion release and oxidative stress induced by intratracheal instillation of ZnO nanoparticles to rat lung Chem. Biol. Interact. 2012 198 29 37 10.1016/j.cbi.2012.04.007 22640810 5. Ho M. Wu K.Y. Chein H.M. Chen L.C. Cheng T.J. Pulmonary toxicity of inhaled nanoscale and fine zinc oxide particles: Mass and surface area as an exposure metric Inhal. Toxicol. 2011 23 947 956 10.3109/08958378.2011.629235 22122307 6. Liu H. Yang D. Yang H. Zhang H. Zhang W. Fang Y. Lin Z. Tian L. Yan J. Xi Z. Comparative study of respiratory tract immune toxicity induced by three sterilization nanoparticles: Silver, zinc oxide and titanium dioxide J. Hazard Mater. 2013 248–249 478 486 10.1016/j.jhazmat.2013.01.046 23419906 7. Borm P.J. Driscoll K. Particles, inflammation and respiratory tract carcinogenesis Toxicol Lett. 1996 88 109 113 8920724 8. Nishi K. Morimoto Y. Ogami A. Murakami M. Myojo T. Oyabu T. Kadoya C. Yamamoto M. Todoroki M. Hirohashi M. Expression of cytokine-induced neutrophil chemoattractant in rat lungs by intratracheal instillation of nickel oxide nanoparticles Inhal. Toxicol. 2009 21 1030 1039 10.1080/08958370802716722 19772482 9. Shacter E. Weitzman S.A. Chronic inflammation and cancer Oncology 2002 16 217 232 11866137 10. Langley R.J. Kalra R. Mishra N.C. Hahn F.F. Razani-Boroujerdi S. Singh S. Benson J.M. Pefia-Philippides J.C. Barr E.B. Sopori M.L. A biphasic response to silica. I. Immunostimulation is restricted to the early stage of silicosis in Lewis rats Am. J. Respir. Cell Mol. Biol. 2004 30 823 829 10.1165/rcmb.2003-0284OC 14742293 11. Sellamuthu R. Umbright C. Roberts J.R. Chapman R. Young S.H. Richardson D. Leonard H. McKinney W. Chen B. Frazer D. Blood gene expression profiling detects silica exposure and toxicity Toxicol. Sci. 2011 122 253 264 10.1093/toxsci/kfr125 21602193 12. Konduru N.V. Murdaugh K.M. Sotiriou G.A. Donaghey T.C. Demokritou P. Brain J.D. Molina R.M. Bioavailability, distribution and clearance of tracheally-instilled and gavages uncoated or silica-coated zinc oxide nanoparticles Part Fibre Toxicol. 2014 11 44 10.1186/s12989-014-0044-6 25183210 13. Adamcakova-Dodd A. Stebounova LV. Kim J.S. Vorrink S.U. Ault A.P. O’Shaughnessy P.T. Grassian V.H. Thorne P.S. Toxicity assessment of zinc oxide nanoparticles using sub-acute and sub-chronic murine inhalation models Part Fibre Toxicol. 2014 11 15 10.1186/1743-8977-11-15 24684892 14. Ahamed M. Akhtar M.J. Alhadiag H.A. Alrokayan S.A. Assessment of the lung toxicity of copper oxide nanoparticles: Current status Nanomedicine 2015 10 2365 2377 10.2217/nnm.15.72 26251192 15. Morimoto Y. Hirohashi M. Ogami A. Oyabu T. Myojo T. Nishi K. Kadoya C. Todoroki M. Yamamoto M. Murakami M. Inflammogenic effect of well-characterized fullerenes in inhalation and intratracheal instillation studies Part Fibre Toxicol. 2010 7 4 10.1186/1743-8977-7-4 20226088 16. Morimoto Y. Izumi H. Yoshiura Y. Tomonaga T. Lee B.W. Okada T. Oyabu T. Myojo T. Yatera K. Shimada M. Comparison of pulmonary inflammatory responses following intratracheal instillation and inhalation nanoparticles Nanotoxicology 2016 10 607 618 10.3109/17435390.2015.1104740 26558952 17. Morimoto Y. Izumi H. Yoshiura Y. Fujishima K. Yatera K. Yamamoto K. Usefulness of intratracheal instillation studies for estimating nanoparticle-induced pulmonary toxicity Int. J. Mol. Sci. 2016 17 165 10.3390/ijms17020165 26828483 18. Fubini B. Hubbard A. Reactive oxygen species (ROS) and reactive nitrogen species (RNS) generation by silica in inflammation and fibrosis Free Radic. Biol. Med. 2003 34 1507 1516 10.1016/S0891-5849(03)00149-7 12788471 19. Schins R.P. Mechanisms of genotoxicity of particles and fibers Inhal. Toxicol. 2002 14 57 78 10.1080/089583701753338631 12122560 20. Kuempel E.D. Tran C.L. Castranova V. Bailer A.J. Lung dosimetry and risk assessment of nanoparticles: Evaluating and extending current models in rats and humans Inhal. Toxicol. 2006 18 717 724 10.1080/08958370600747887 16774860 21. Baisch B.L. Corson N.M. Wade-Mercer P. Gelein R. Kennell A.J. Obertorster G. Elder A. Equivalent titanium dioxide nanoparticle deposition by intratacheal instillation and whole body inhalation: The effect of dose rate on acute respiratory tract inflammation Part Fibre Toxicol. 2014 11 5 10.1186/1743-8977-11-5 24456852 22. Morimoto Y. Hirohashi M. Ogami A. Oyabu T. Myojo T. Todoroki M. Yamamoto M. Hashiba M. Mizuguchi Y. Lee B.W. Pulmonary toxicity of well-dispersed multi-wall carbon nanotubes following inhalation and intratracheal instillation Nanotoxicology 2012 6 587 599 10.3109/17435390.2011.594912 21714591 23. Morimoto Y. Izumi H. Yoshiura Y. Tomonaga T. Oyabu T. Myojo T. Kawai K. Yatera K. Shimada M. Kubo M. Pulmonary toxicity of well-dispersed cerium oxide nanoparticles following intratracheal instillation and inhalation J. Nanopart. Res. 2015 17 442 10.1007/s11051-015-3249-1 26594128 24. Silva R.M. Doudrick K. Franzi L.M. TeeSy C. Anderson D.S. Wu Z. Mitra S. Vu V. Dutrow G. Evans J.E. Instillation versus inhalation of multiwalled carbon nanotubes: Exposure-related health effects, clearance, and the role of particle characteristics ACS Nano 2014 8 8911 8931 10.1021/nn503887r 25144856 25. Pan C.H. Chuang K.J. Chen J.K. Hsiao T.C. Lai C.H. Jones T.P. Berube K.A. Hong G.B. Ho K.F. Chuang H.C. Characterization of pulmonary protein profiles in response to zinc oxide nanoparticles in mice: A 24-hour and 28-day follow-up study Int. J. Nanomed. 2015 10 4705 4716 26. Li N. Sioutas C. Cho A. Schmitz D. Misra C. Sempf J. Wang M. Oberley T. Froines J. Nel A. Unitrafine particulate pollutants induce oxidative stress and mitochondrial damage Environ. Health Perspect. 2003 111 455 460 10.1289/ehp.6000 12676598 27. Oyabu T. Morimoto Y. Hirohashi M. Horie M. Kambara T. Lee B.W. Hashiba M. Mizuguchi Y. Myojo T. Kuroda E. Dose-dependent pulmonary response of well-dispersed titanium dioxide nanoparticles following intratracheal instillation J. Nanopart. Res. 2013 15 1600 10.1007/s11051-013-1600-y 28. Kubo M. Nakaoka A. Morimoto K. Shimada M. Horie M. Morimoto Y. Sasaki T. Aerosol generation by a spray-drying technique under coulomb explosion and rapid evaporation for the preparation of aerosol particles for inhalation tests Aerosol Sci. Technol. 2014 48 698 705 10.1080/02786826.2014.918930 29. Shimada M. Wang W.N. Okuyama K. Myojo T. Oyabu T. Morimoto Y. Tanaka I. Endoh S. Uchida K. Ehara K. Development and evaluation of an aerosol generation and supplying system for inhalation experiments of manufactured nanoparticles Environ. Sci. Technol. 2009 43 5529 5534 10.1021/es9008773 19708392

PMC005xxxxxx/PMC5000640.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081242ijms-17-01242ReviewThe Impact of Anti-Epileptic Drugs on Growth and Bone Metabolism Fan Hueng-Chuen 12Lee Herng-Shen 3Chang Kai-Ping 4Lee Yi-Yen 56Lai Hsin-Chuan 12Hung Pi-Lien 7Lee Hsiu-Fen 8Chi Ching-Shiang 12*De Berardis Domenico Academic EditorKunz Wolfram S. Academic Editor1 Department of Pediatrics, Tungs’ Taichung Metroharbor Hospital, Wuchi, 435 Taichung, Taiwan; [email protected] (H.-C.F.); [email protected] (H.-C.L.)2 Department of Nursing, Jen-Teh Junior College of Medicine, Nursing and Management, 356 Miaoli, Taiwan3 Department of Pathology and Laboratory Medicine, Kaohsiung Veterans General Hospital, 813 Kaohsiung, Taiwan; [email protected] Department of Pediatrics, Taipei Veterans General Hospital, 112 Taipei, Taiwan; [email protected] Division of Pediatric Neurosurgery, Neurological Institute, Taipei Veterans General Hospital, 112 Taipei, Taiwan; [email protected] Faculty of Medicine, National Yang-Ming University, 112 Taipei, Taiwan7 Department of Pediatrics, Kaohsiung Chang Gung Medical Center, 833 Kaohsiung, Taiwan; [email protected] Department of Pediatrics, Taichung Veterans General Hospital, 407 Taichung, Taiwan; [email protected]* Corresponding: [email protected]; Tel.: +886-4-2658-1919 (ext. 4116); Fax: +886-4-2658-115501 8 2016 8 2016 17 8 124213 5 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Epilepsy is a common neurological disorder worldwide and anti-epileptic drugs (AEDs) are always the first choice for treatment. However, more than 50% of patients with epilepsy who take AEDs have reported bone abnormalities. Cytochrome P450 (CYP450) isoenzymes are induced by AEDs, especially the classical AEDs, such as benzodiazepines (BZDs), carbamazepine (CBZ), phenytoin (PT), phenobarbital (PB), and valproic acid (VPA). The induction of CYP450 isoenzymes may cause vitamin D deficiency, hypocalcemia, increased fracture risks, and altered bone turnover, leading to impaired bone mineral density (BMD). Newer AEDs, such as levetiracetam (LEV), oxcarbazepine (OXC), lamotrigine (LTG), topiramate (TPM), gabapentin (GP), and vigabatrin (VB) have broader spectra, and are safer and better tolerated than the classical AEDs. The effects of AEDs on bone health are controversial. This review focuses on the impact of AEDs on growth and bone metabolism and emphasizes the need for caution and timely withdrawal of these medications to avoid serious disabilities. epilepsybone metabolismanti-epileptic drugs (AEDs)classical anti-epileptic drugs (AEDs)newer anti-epileptic drugs (AEDs)cytochrome P450 (CYP450)bone mineral density (BMD) ==== Body 1. Introduction Epilepsy, a common neurological disorder, affects about 50 million people around the world. The prevalence of epilepsy is approximately 6.8 per 1000 in the US [1], 5.5 per 1000 in Europe, 1.5 to 14 per 1000 in Asia [2], and 3.3 per 1000 in Taiwan [3]. Although there are many alternative treatment choices for epilepsy, including vagus nerve stimulation (VNS), surgery, and a ketogenic diet, anti-epileptic drugs (AEDs) are always the first choice because numerous patients with epilepsy were seizure-free while taking an AED [4]. However, AEDs should be used carefully because of drug-drug interactions and potential side effects, such as dizziness, drowsiness, mental slowing, skin rashes, hepatotoxicity, movement and behavioral disorders, and metabolic disturbances, such as weight gain, metabolic acidosis, and nephrolithiasis [5]. Although few unusual adverse effects such as rickets, osteomalacia, and abnormal dentition were previously identified in patients taking the AEDs [6,7], more than 50% of patients with epilepsy who take AEDs are reported to have bone abnormalities [8,9,10], and several case-control studies have traced a link between long-term AED use, bone diseases [11,12,13], and the increase of fracture risks [12,14,15]. The newer AEDs, including levetiracetam (LEV), oxcarbazepine (OXC), lamotrigine (LTG), topiramate (TPM), gabapentin (GP), and vigabatrin (VB) are also effective in the treatment of various seizures, in addition to being safer and better-tolerated than the classical AEDs. However, studies regarding the effects of the newer AEDs on bone health and growth are limited. Symptoms of patients with AED-associated bone diseases include bone pain, muscle weakness, and fractures, with minimal or no trauma. These symptoms do not appear until the first fracture occurs [16]. Their biochemical studies may show abnormal serum levels of vitamin D metabolites, phosphorous, Ca2+, and alkaline phosphatase. Routine X-rays can identify bone fractures, but cannot detect these bone diseases if the reduction of bone mass density (BMD) is less than 30% [17]. BMD represents a complex and dynamic balance between the actions of osteoclasts, which are responsible for bone resorption, and the actions of osteoblasts, which are responsible for the bone-formation. The values of BMD in twins and siblings with epilepsy receiving AEDs treatment were significantly lower than that without treatment [18]. Dual energy X-ray absorptiometry (DXA) is an X-ray technique to measure the levels of BMD [17]. One-third to two-thirds of epileptic patients with AEDs showed abnormal BMD values by using DXA [9,19], but the safety of the ionizing radiation exposure is a large concern. Although it is clear that AEDs affect bone metabolism and increase fractures are not clear, the metabolism of drugs may play an important role in the development of these adverse effects. The metabolism of drugs can be divided into two phases. CYP 450 is responsible for the phase I metabolism, including activation, metabolism, and clearance of medications. Several medications cause unwanted side effects and decreased or no therapeutic effects because these medications, including AEDs administered parenterally or non-parenterally, can induce or suppress CYP450, leading to unanticipated drug-drug interactions [20]. Reports showed that enzyme-inducing AEDs (EIAEDs) could induce CYP450 to accelerate the degradation of vitamin D, contributing to hypocalcemia [21,22,23], reduced BMD, and a higher risk of fractures [24,25]. Studies showed that valproic acid (VPA), one of the non-enzyme-inducing AEDs (NEIAEDs), was associated with low bone mass [13,26,27]. The glucuronidation is responsible for the phase II metabolism. Organisms utilize glucuronidation to detoxify environmental toxins and carcinogens and participate in essential biochemical processes. UDP-glucuronosyltransferases (UGTs), which are the most important enzymes in the glucuronidation, comprise a superfamily of key proteins, UGT1 and UGT2. Each of the proteins UGT1 and UGT2 has at least eight isoenzymes [28]. UGTs facilitate the glucuronic acid group of uridine diphosphoglucuronic acid (UDPGlcA) transferring to several structurally diverse chemicals, such as AEDs, to increase the polarity and enhance their chemicals excretion in the urine and bile [29]. No definitive guidelines for evaluation of the effect of AEDs on bone metabolism are available. The diagnosis and the treatments of epilepsy are commonly initiated in childhood and adolescence, which are a critical period of growth in life. Therefore, it is worth conducting a short review to discuss the impact of classical and newer AEDs and how the metabolites of these AEDs affect bone health. The results of this review may allow for patients with AED-associated skeletal bone diseases to be recognized earlier and appropriate therapy to be implemented without delay. 1.1. Benzodiazepines (BZDs) BZDs, such as diazepam, lorazepam, midazolam, and clonazepam, are widely prescribed. Minimal toxicity and rapid onset of action make BZDs among the top 100 most commonly prescribed medications [30]. One of the main effects of BZDs is the enhancement of the neurotransmitter gamma-aminobutyric acid (GABA) and GABA receptor-mediated chloride conductance, contributing to the effects of sedation, hypnosis, anxiolysis, anti-seizure, and muscle relaxation [31,32]. Metabolism of BZDs includes liver microsomal oxidation, hydroxylation, glucuronidation, acetylation, etc. [33]. CYP3A4, CYP3A5, CYP2C19, and others are associated with the hydroxylation of BZD [34,35,36]. Some hydroxylated metabolites of BZDs still have pharmacological activities. UGTs are responsible for the process of glucuronidation of BZD [36,37]. Midazolam, S-oxazepam, and R-oxazepam undergoes glucuronidation by UGT1A4 [36,37,38], UGT2B15 [37], and UGT2B7 and UGT1A9 [37], respectively. Clonazepam undergoes acetylation by NAT2 [39,40] (Figure 1). BZD metabolites are mainly eliminated through renal excretion. A retrospective investigation concluded that the use of diazepam, lorazepam, and clonazepam [41,42] might induce a substantial number of fractures and consequential costs. Temazepam, a metabolite of diazepam via CYP3A4, was found to increase the risk of fractures [43]. There was only one case report regarding the use of oxazepam and recurrent mandibular luxation [44]. BZDs have also been reported to disturb bone metabolism, including a reduction of BMD and 25-hydroxy vitamin D (25OHD), and an increase in the serum alkaline phosphatase (ALP) levels. The levels of total calcium, phosphorus, magnesium, and parathyroid hormone (PTH) were unaffected by BZDs [24,35] although some other results are controversial [45,46,47]. Interestingly, a report showed that midazolam could exert negative effects on cell viability and osteogenic differentiation of cultured human bone marrow stem cells, suggesting a detrimental effect of the use of midazolam on bone formation and growth [48] (Table 1). 1.2. Carbamazepine (CBZ) CBZ, an iminodibenzyl derivative, is extensively bio-transformed in the liver and approximately 5% of CBZ is eliminated through renal excretion [49]. CBZ 10.11-epoxide (CBZ-E), which possesses anti-convulsant properties, is generated through the action of CYP3A4, CYP3A5, and CYP2C8 [50,51]. CBZ diol is generated via the action of epoxide hydrolase 1 (EPXH1) (Figure 2). Although glucuronidation is not important in the metabolism of CBZ, UGT2B7 may be involved in the metabolism of CBZ and CBZ-E [52,53]. Other metabolites of CBZ include 2-OH CBZ and 3-OH CBZ. The former is generated through the actions of multiple CYPs and the latter is produced by the actions of CYP2B6 and CYP3A4 [50]. 2-OH CBZ is oxidized by CYP3A4 to produce an iminoquinone intermediate [50], whereas 3-OH-CBZ is oxidized by CYP3A4 to generate CBZ o-quinone [50]. 3-OH CBZ may generate radicals through the action of myeloperoxidase (MPO) [50]. CBZ stabilizes voltage-gated sodium channels (VGSCs), minimizes VGSCs in the rest status subsequently to be excited, and reduce polysynaptic responses to block post-tetanic potentiation. These actions make CBZ a widely used AED for partial and secondary generalized seizures [54]. Additionally, CBZ’s structure is similar to that of the tricyclic anti-depressants and a function of CBZ is a GABA receptor agonist. These may partially explain the effects of CBZ on bipolar disorder and the treatment of pain in trigeminal neuralgia [50]. CBZ may cause several adverse effects, including sedation, ataxia, dizziness, nausea, vomiting, constipation, diarrhea, interference with the metabolism of lipids and sex hormones, hyponatremia, weight-gain, anemia, agranulocytosis, toxic epidermal necrolysis (TEN), Stevens Johnson syndrome (SJS), and drug reactions with eosinophilia and systemic symptoms (DRESS) [55,56,57]. Erythromycin, clarithromycin, and triacetyloleandomycin are the most potent CYP3A4 inhibitors and are best avoided in CBZ-treated patients. Azithromycin does not interact with CYP3A4 and, therefore, does not affect CBZ concentrations. CBZ was reported to cause spinal bifida in 1% of neonates whose mothers had an exposure history in pregnancy [58]. Moreover, long-term use of CBZ may increase the risks of fracture and bone loss, induce a status of decreased bone and mineral metabolism, increase bone turnover, and decrease BMD [19,42]. CBZ may induce CYP450 to decrease the levels of vitamin D. A study of previously drug-naive Koreans with CBZ revealed a significant decrease in BMD [59]. On the contrary, high levels of bone formation markers have been detected in patients treated with CBZ, despite normal levels of vitamin D [60]. Pack et al. [61] found that serum calcium and estrogen levels were lower in epileptic women in premenopausal status taking CBZ. However, there was no connection between bone turnover marker or calciotropic hormone levels and BMD change in these women, suggesting it was estrogen rather than vitamin D that led to bone loss in epileptic women in premenopausal status [61]. Whether CBZ affects bone through the induction of CYP450 and/or its metabolites remains unknown (Table 1). 1.3. Phenytoin (PT) PT (5,5-Diphenyl-Imidazolidine-2,4-Dione) is available in oral and intravenous formulations. The bioavailability of oral PT is 70%–90% and the t1/2 of PT is 12–36 h. The peak blood level of PT is 3–12 h [62]. Adverse effects, such as nausea, vomiting, gingival hyperplasia, burning sensation at the local injection site, nystagmus, ataxia, hypotension, bradyarrhythmias, cardiac arrest, SJS, TEN, and birth defects, may occur [62]. Dissolving PT in a base solution with pH of 12 that contains sodium hydroxide, ethanol, and propylene glycol can improve the aqueous solubility. However, the cardiovascular toxicity of intravenous phenytoin infusion may contribute to the strong effects of propylene glycol on the vagal nerve [62]. Additionally, the high pH is responsible for propylene glycol’s veno-irritant properties [62]. The therapeutic range of PT is narrow and the clearance of PT is variable between individuals. Additionally, co-ingestion of PT with an antacid mixture of magnesium trisilicate and aluminum hydroxide reduces serum PT concentrations [63]. Moreover, some medications, such as Cisplatin and other anti-neoplastic drugs may affect serum PT concentrations [64]. All of them suggest that it is necessary to do therapeutic drug monitoring when using PT. PT inhibits GABA and glutamate transport [65], reduces calcium influx into neurons to decrease the release of neurotransmitters [66], and reduces synaptic post-tetanic potentiation, and excitatory synaptic transmission to stop the cortical abnormal current propagation [67]. Moreover, PT can bind to and stabilize the inactive VGSCs [68]. VGSCs are highly conserved and responsible for the upstroke of the action potentials in neurons involving the propagation of the electrical impulse in the CNS, PNS, and cardiovascular and skeletal muscle tissue. After binding, PT prevents further generation of action potentials, which initiate seizures [68]. These mechanisms may significantly prevent generalized tonic-clonic seizures, complex partial seizures, and status epilepticus, but not absence seizures. PT is well-absorbed orally, and up to 90% of PT is biotransformed to HPPH, 5-(4′-hydroxyphenyl)-5-phenylhydantoin and hydroxyphenytoin [69], which are inactive metabolites and are excreted into urine after glucuronidation [70]. HPPH proceeds to form phenytoin-arene oxide (PAO), which may be the reason why epileptic patients develop hepatotoxicity, hypersensitivity, TEN, SJS, and idiosyncratic toxicity after taking PT [71]. PAO is metabolized to phenytoin dihydrodiol (PDH) via CYP1A2, CYP2C19, CYP2E1, CYP2A6, CYP2D6, CYP2C8, CYP2C9, CYP3A4, and epoxide hydrolase (EPHX1) [69,72]. Phenytoin catechol (PC) is a downstream metabolite of PDH [69]. Hydroxyphenytoin is turned into PC through the actions of CYP2C19, CYP3A4, CYP3A5, CYP3A7, and CYP2C9 [69,72,73]. PC is spontaneously and reversibly oxidized to form a phenytoin quinone by NAD(P)H dehydrogenase, quinone 1 (NQO1). PC is converted to phenytoin methylcatechol (PMC) through the action of Catechol-O-methyl transferase (COMT) [69]. Hydroxyphenytoin is glucuronidated by UGT1A1, UGT1A4, UGT1A6, and UGT1A9 [74]. PT can induce CYP3A, CYP2C, and UGTs [75] (Figure 3). Fetal hydantoin syndrome is characterized by learning disabilities, low IQ scores, growth retardation, microcephaly, and facial dysmorphologies [76], suggesting a significant influence on bone growth. PT might induce a substantial number of fractures and consequential costs [42] in vivo and in vitro [77,78]. PT may also induce the expression of CYP450, which increases the degradation of bioavailable vitamin D, decreases absorption of calcium in the gut, decreases serum levels of calcium and phosphate, and increases PTH. These effects may then lead to increased bone turnover, reduced BMD, and increased susceptibility to fractures [79,80,81]. Among phenytoin’s metabolites, only HPPH was found to affect bone in vitro [82]. Therefore, the bone condition of patients taking PT should be monitored regularly (Table 1). 1.4. Phenobarbital (PB) PB (5-ethyl-5-phenyl-1,3-diazinane-2,4,6-trione) was the most commonly-used AED in the world [83,84]. PB is available in oral and intravenous formulations. Its pharmacokinetics are linear and protein binding is 55%. The bioavailability of oral PB is more than 95% and the peak blood level of PB is 0.5–4 h. The t1/2 of PB is 2–7 days [85]. Discontinuing PB should be done with caution because a case report showed an increase of seizure frequency in patients tapering the doses of PB while stabilized on another AED [86]. Twenty-five percent of PB is cleared by renal excretion in unchanged form [87]. After administration, PB was detected in hepatic tissue and the portal vein, vena cava, and aorta [88], suggesting that the liver is the main organ for the metabolism of PB. The metabolites of PB include free PB and two inactive metabolites. p-hydroxy PB (6%–40% of the dose) is created by CYP2C9, CYP2C19, and CYP2E1 through the process of aromatic hydroxylation and 9-d-glucopyranosyl-PB by glucuronidation (25% of the dose). The enzymes involved in this N-glucosidation have not yet been identified; however, UGT 2B has been proposed as the enzyme responsible for this process [89]. These processes are complicated and exhibit a large inter-individual variability [90]. Orphan nuclear receptors, including pregnane X receptor (PXR) and constitutive androstane receptors (CAR), are activated by PB to upregulate CYP 450 gene expression [91], causing increased clearance and decreased serum concentrations of drugs, including AEDs (e.g., CBZ, PT, VPA, LTG, TPM), and lipid-soluble drugs (e.g., oral contraceptives, warfarin, corticosteroids, sex hormones, vitamin D) [92]. VPA may change serum levels or prolong the t1/2 of PB by affecting the metabolism of PB [92,93], leading to variable dose requirements for PB. Therefore, therapeutic drug monitoring of PB levels is needed when PB is used in combination with other drugs. PB enhances GABA and GABAA receptor-associated inhibition [94] and facilitates Cl− conductance by extending the time of channel opening [95]. These effects lead to an increased Cl− influx to hyperpolarize the postsynaptic neurons and block the propagation of aberrant epileptic currency. PB may directly activate the GABAA receptor [96]. The actions of PB through these effects may reduce anxiety, promote sleep, induce general anesthesia, and act as an effective control of generalized and partial tonic–clonic seizures [97,98]. PB was the World Health Organization’s first-line AED in developing countries because of its low cost and effectiveness in the treatment of seizures. However, the use of PB has decreased even though there is no obvious connection between the use of PB and the development of behavioral problems [99]. Side effects, such as sedation, hypnosis, dizziness, nystagmus, ataxia, excitement, confusion, and paradoxical hyperactivity may occur. Contraindications for PB use include acute intermittent porphyria, hypersensitivity to PB, a prior history of dependence on PB, and hyperkinesia in children [62]. In vivo studies showed that long-term use of PB might diminish the t1/2 of the plasma vitamin D3 and enhance excretion in the bile [100]. Long-term use of PB may increase the risks of fracture [42] and bone loss [19]. Liver microsomes dissected from animals with PB treatment enabled vitamin D3, 25-hydroxycholecalciferol, and 1,25-dihydroxycholecalciferol to turn into inactive products [100], causing rickets, osteomalacia, and hypocalcemia. Therefore, vitamin D supplementation should be considered for patients receiving long-term PB therapy. p-hydroxy PB and 9-d-glucopyranosyl-PB have not been reported to be associated with bone diseases (Table 1). 1.5. Valproic Acid (VPA) Valproic acid (VPA, 2-propylpentanoic acid), a branched-chain fatty acid, is originally extracted from Valeriana officinalis. VPA is commonly used in people with epilepsy because it is effective and can be administered orally, intravenously, or rectally. The oral bioavailability of VPA is more than 80%. Clinically, it is puzzling that the doses of VPA in the treatment of patients with epilepsy are variable and the toxicities of the drug are poorly correlated with VPA serum concentrations [101]. Studies showed that VPA has a very high protein binding (≥90%) in the plasma and few unchanged VPA (<3%) appears in the urine [102], suggesting a very complicated biotransformation of VPA in humans (Figure 4). First, 30%–50% of VPA may be metabolized via glucuronization by UGTs, including UGT1A3, UGT1A4, UGT1A6, UGT1A8, UGT1A9, UGT1A10, UGT2B7, and UGT2B15. End products are mostly excreted in the bile and urine. However, VPA can directly inhibit the activity of UGT1A4, and UGT2B7 [103,104]. Second, 30% of VPA metabolism occurs via β-oxidation in the mitochondria. VPA as a medium chain fatty acid is able to enter the mitochondrial matrix and is turned into valproyl-CoA (VPA-CoA) by medium-chain acyl-CoA synthase (EC 6.2.1.2) [105]. VPA-CoA is converted into VPA-dephospho-CoA and 2-propyl-valproyl-CoA (2-ene-VPA-CoA) by the phosphatase 2-methyl-branched chain acyl-CoA dehydrogenase (2MBCAD) and Isovaleryl-CoA dehydrogenase (IVD), respectively [106,107]. 3-hydroxyl-valproyl-VPA (3-OH-VPA-CoA) is generated from 2-ene-VPA-CoA through 2-enoyl-CoA hydratase (EH). 3-OH-VPA-CoA is converted into 3-keto-valproyl-CoA (3-oxo-VPA-CoA) or propionyl-CoA (C3-CoA) and pentanoyl-CoA (C5-CoA) by the action of 2-methyl-3-hydroxybutyryl-CoA dehydrogenase (MHBD) [108,109] or hydroxyacyl-CoA dehydrogenase (HADH) [105,108]. 3-oxo-VPA CoA is metabolized by the glutathione (GSH) into thiols [110]. 4-ene-VPA CoA, which is generated by the metabolism of VPA through 4-ene-VPA-CoA ester, is converted into 2,4-diene-VPA-CoA ester through 2MBCAD [110,111]. 2,4-diene-VPA-CoA and 4-ene-VPA-CoA are turned into thiols by GSH [110]. Third, 10% of VPA is biotransformed through CYP450-mediated oxidation. CYP2A6 is partially connected to the generation of 3-OH-VPA [112]. CYP2A6, CYP2C9, and CYP2B6 are involved in the VPA metabolism to generate 4-ene-VPA, 4-OH-VPA, and 5-OH-VPA [113]. Interestingly, VPA can inhibit CYP2C9, CYP2C19, and CYP3A4, but not CYP1A2, CYP2D6, or CYP2E1 [103,104]. VPA may undergo β-oxidation or glucuronidation when the doses are below or in therapeutic range [114]. This may explain why different doses of VPA cause distinct responses. VPA affects the GABAergic system, inhibits α-ketoglutarate dehydrogenase (αKGD), GABA transaminase (GABA-T), and succinate semialdehyde dehydrogenase (SSD), and enhances glutamate decarboxylase (GAD) to elevate GABA levels in plasma and in several brain regions. Consequently, VPA may affect cerebral metabolism, activate GABA receptors to block sodium channels, and modulate calcium and potassium conductance and dopaminergic and serotoninergic transmission [115,116]. These mechanisms make VPA a multi-functional medication for absence, partial, and tonic-clonic seizures, bipolar disorder, depression, migraine, personality disorders or mental retardation, dementia and cognitive problems, and a potential chemotherapeutic agent [116]. Moreover, VPA can inhibit histone deacetylase (HDAC), which is a crucial factor in the pathogenesis of cancer and transcriptional regulation [117,118]. VPA is currently under investigation to be an adjunctive therapeutic option in neurodegenerative diseases, HIV, and cancers. Nausea, vomiting, abdominal cramps, diarrhea, weight gain, impaired coagulation, and neutropenia are the most common side effects of VPA. Hepatotoxicity, pancreatitis, teratogenicity, and endocrine disturbance, such as menstrual abnormalities, increased total testosterone levels, teratogenicity, obesity, and polycystic ovary syndrome (PCOS) may be associated with VPA. Hepatotoxicity is one of the most serious complications in the use of VPA. Although mitochondrial dysfunction and abnormal fatty acid metabolism have been proposed for the causes of VPA associated hepatotoxicity [119], the exact mechanisms are still unknown. Fetal valproate syndrome is characterized by orofacial clefts, congenital heart disease, neural tube defects, limb defects, genitourinary defects, and craniosynostosis. VPA may affect limb and organ morphogenesis, suggesting a significant effect on bone growth and metabolism [120]. Long-term use of VPA may increase the risks of bone loss [19]. In vitro studies, our study [121] and others [122,123,124,125] showed that VPA may directly affect bone growth. VPA may have neuroprotective and anti-tumor activities through the modulation of epigenetic mechanisms [126,127,128]. VPA within therapeutic concentrations effectively inhibits histone deacetylases (HDACs). HDACs are enzymes crucial for the control of histone acetylation status and for the epigenetic regulation of gene activation involved in the modulation of cell growth, differentiation, and apoptosis [129,130]. VPA may cause short stature by directly inhibiting cell growth and proliferation through activation of apoptosis by hyperacetylation of histone tails and chromatin. In addition, serious side effects, teratogenesis, liver toxicity, and associated bone diseases have prompted the search for a newer generation of AEDs to provide better efficacy and fewer side effects (Table 1). 2. New Generation AEDs 2.1. Levetiracetam (LEV) LEV ((S)-α-ethyl-2-oxo-1-pyrrolidine acetamide) was discovered through screening for effective AEDs in audiogenic seizure mice [131]. The chemical structure of LEV is the α-ethyl analog of piracetam and is unrelated to other AEDs [132]. LEV is a safe and well-tolerated new AED and no significant drug interactions were noted between LEV and concomitant medications because of lower protein binding and no involvement of hepatic CYP isozymes [131,132]. LEV is rapidly absorbed in the digestive tract and mainly excreted in urine. Approximately 1/3 of an administered dose of LEV was metabolized and 2/3 was excreted in urine in unchanged form [133]. The major pathway involves hydrolysis through the type B esterases primarily in the liver and blood [134] to generate (2S)-2-(2-oxopyrrolidin-1-y butanoic acid and two minor metabolites without significant pharmacological activities [135]. Pharmacologically, LEV effectively reduces partial seizures, intractable partial seizures, and patients with other medical conditions by several proposed mechanisms, including: (1) targeting synaptic vesicle protein 2A (SV2A), which is associated with vesicle neurotransmitter exocytosis; (2) negative modulation of neuron-associated GABA- and glycine-gated currents; (3) inhibiting voltage-gated calcium channels or reducing voltage-operated potassium currents [136,137,138,139]. Low-dose LEV was found to impair longitudinal skeletal growth and increase the risk of fractures in immature rats [140]. LEV was found to affect serum estradiol levels, suggesting that young and female individuals might be at risk of fractures with long-term use of LEV [141]. However, other reports [142,143] did not observe this effect. No reports are available regarding hydrolytic metabolites of LEV on bone diseases (Table 1). 2.2. Oxcarbazepine (OXC) OXC (10,11-dihydro-10-oxo-5H-dibenz(b,f)azepine-5-carboxamide) has been designed via structural variation of CBZ [144]. After oral administration, metabolites of OXC in urine included MHD and two diastereoisomeric O-glucuronides (79%), unchanged OXC, OXC’s sulfate and glucuronide conjugates (13%), the cis- and trans-isomers of 10,11-dihydro-10,11-dihydroxy-carbamazepine (approximately 4%), and a phenolic derivative of MHD (less than 1%) [145]. Orally-administered OXC is rapidly metabolized to form the 10,11-dihydro-10-hydroxy-carbazepine (monohydroxy derivative, MHD) through cytosolic arylketone reductases. MHD is dissolved in water and a biologically active metabolite [144]. Therefore, MHD is as potent an anti-epileptic drug as OXC. MHD has two enantiomers: S enantiomers of MHD [(S)-MHD] (accounts for 80%) and R enantiomers of MHD [(R)-MHD] (accounts for 20%) [146]. The antiepileptic efficacy and tolerability of (R)-MHD and (S)-MHD is similar [147]. OXC, like CBZ, specifically inhibits voltage-dependent sodium [148], potassium [149], and calcium channels [150]. Although the efficacy of these two medications is similar, the safety of OXC is superior. Therefore, the FDA approved OXC as adjunctive therapy or monotherapy for children and adults with partial seizures. Hyponatremia is the main adverse effect of OXC [151]. Decreased BMD, altered levels of 25OHD [152,153], and bone turnover biomarkers such as PTH and bALP [152] were reported in patients with long-term OXC use [153,154]. However, our previous study [121] and others [155] did not discover any significant hypocalcemia or growth retardation in pediatric patients receiving OXC, and OXC did not significantly impair the proliferation of growth plate chondrocytes in an in vitro experiment [121]. Our recent study showed when patients with epilepsy took OXC and/or VPA for one year, their growth velocity was significantly decreased through affected bone resorption [156]. The use of VPA and/or OXC therapy affecting bone metabolism deserves further investigation (Table 1). 2.3. Lamotrigine (LTG) LTG (6-(2,3-dichlorophenyl)-1,2,4-triazine-3,5-diamine) is rapidly and completely absorbed after oral administration. The oral bioavailability of LTG is 98%. The blood level of LTG is 1.4 to 4.8 h [157]. Metabolite identification studies demonstrate that N-2 glucuronide, N-5 glucuronide, N-2 methyl, and N-2 oxide are the main metabolites of LTG [158]. Most of these metabolites are non-active. LTG is eliminated via glucuronidation—mainly through UGT1A4, UGT2B7, and UGT1A1 [159]. LTG generally does not interfere with drug metabolizing enzymes. More frequent dosing and higher doses may be needed when co-administered with AEDs, such as PB, PT, CBZ, and OXC because these AEDs may enhance LTG clearance and decrease its plasma concentration by activating the glucuronidation pathway. On the contrary, co-administration of VPA may raise LTG plasma concentration as much as two-fold by inhibiting LTG clearance. Therefore, the recommended maintenance dose of LTG should be two-fold lower if LTG is co-administered with VPA. However, newer AEDs rarely affect LTG clearance [103]. LTG acts on pre-synaptical voltage-sensitive sodium channels. LTG blocks N- and P/Q/R-type calcium channels. These blocking effects and others may stabilize neuronal membrane potential [160]. LTG can abolish the repetitive firing in mouse spinal cord neurons in vitro. For these mechanisms, LTG is effective as a monotherapy or polytherapy for primary or secondarily generalized clonic-tonic seizures and simple or complex partial seizures. Additionally, LTG can be used as an adjuvant therapy in typical or atypical absence seizures, infantile spasms, juvenile myoclonic epilepsy, Lennox-Gastaut syndrome (LGS), and myoclonic seizures [161]. The antiepileptic effect of LTG is similar to that of PT and CBZ, but LTG are multi-functional when compared with these two drugs. LTG may inhibit the release of glutamate in the ventral part of the striatum and limbic areas, leading to the mood stabilization effect [161]. Headache, dizziness, sedation, nausea, insomnia, diplopia, and ataxia are common problems in patients taking this medication. The incidence of rash in the use of LTG is approximately 0.1% in all cases. The rash can vary from transient mild rash to fatal SJS. Children generally tend to have skin rashes more than adults. Adverse effects of LTG on bone, including bone loss [19], disturbed growth in children, impaired BMD, and elevated bone turnover markers have been reported [162] while our [121] and other [163,164,165] results were contradictory (Table 1). 2.4. Topiramate (TPM) TPM (2,3:4,5-Bis-O-(1-methylethylidene)-β-d-fructopyranose sulfamate) is a derivative of monosaccharide d-fructose. TPM is rapidly and completely absorbed after oral administration and concomitant food intake does not affect the metabolism of TPM. The peak blood level of TPM is 1.4–4.3 h [166]. The protein binding of TPM in humans range 3%–4% [166]. An estimated 85% of an administered dose of TPM was predominantly excreted in urine as unchanged form [167]. The t1/2 of TPM is 19–25 h and is decreased by co-administration of EIAEDs such as CBZ and PT [166,167,168]. The remainder (15%) is metabolized through hydrolysis, hydroxylation, and glucuronidation. Six metabolites of TPM were detected in human urine without significant clinical activity [168]. TPM can partially inhibit CYP2C19 [169]. Pharmacologically, TPM affects GABAergic activity, inhibits voltage-sensitive sodium channels, calcium channels, and kainite/α-amino-3-hydroxy-5-methyllisoxazole-4-proprionic acid (AMPA)-type glutamate receptors, and blocks kinases to activate these channels [170]. All of these mechanisms not only make TPM approved as adjunctive therapy for adults and pediatric patients ages 2–16 years with primary generalized clonic-tonic seizures, partial seizures or LGS [171], but also contribute to a wide spectrum, including prophylaxis of migraines, alleviation of neuropathic pain, alcoholism, obesity, depression, bipolar disorder, and post-traumatic stress disorder [172]. Somnolence, nystagmus, hypo- or anhydrosis, paresthesia, poor concentration and word finding, weight loss, and decreased appetite, were the common complaints when using TPM. TPM may cause metabolic acidosis and nephrolithiasis through the inhibition of the carbonic anhydrase. The acid-base imbalance may accelerate osteopathy [173,174]. PTH secretion may be reduced after exposure to TPM, disrupting the balance between calcium resorption, the synthesis of 1,25(OH)2D, and the activities of osteoclasts [175]. In addition, TPM is a carbonic anhydrase inhibitor that may inhibit PTH-induced bone resorption, resulting in hypocalcemia. However, patients receiving TPM in our study did not show significant hypocalcemia or growth retardation [121]. More human studies may clarify these conflicting results (Table 1). 2.5. Gabapentin (GP) GP (1-(aminomethyl) cyclohexane acetic acid), structurally-related to GABA, was originally developed to treat spasticity [176]. GP is absorbed in the gastrointestinal tract. The GP concentrations in CSF and brain are 20% and 80% of the concentrations in plasma, respectively [177,178]. GB can bind to voltage-dependent calcium channels containing the α2δ subunit to attenuate their activities [179,180]. GB does not bind to GABAA or GABAB receptors, nor does it disturb GABA uptake or metabolism, but can increase the concentration of GABA to reduce firing inputs [181] and enhance GABA responses in neuronal tissues [182]. For its high lipid solubility and structural uniqueness, GP can freely cross the blood-brain barrier, promptly elevate brain GABA, and presumably offer partial protection against further seizures within hours of the first dose [183]. GB inhibits neuronal calcium influx to reduce the release of mono-amine neurotransmitters, including glutamate, noradrenaline, and serotonin [184], causing decreased AMPA receptor activation in the brain. GB can bind to presynaptic NMDA receptors with inhibitory effects [185]. Due to these mechanisms that neither induce nor suppress hepatic microsomal enzymes [186], low level of protein binding [187], and renal excretion with an unchanged GB form in urine [178], GB is less likely to interact with other AEDs and is approved in persons over three years of age as an adjunctive AED for partial seizures with or without secondary generalization. In addition, GB can inhibit the descending noradrenergic system, leading to anti-hyperalgesic and anti-allodynic effects [188]. GB is effective in the treatment of a variety of pains including headaches, inflammatory pain, central pain, diabetic neuropathy, post-herpetic neuralgia, HIV-related neuropathy, trigeminal neuralgia, malignant pain, and postoperative pain management [176]. Sexual dysfunction, weight gain, dizziness, somnolene, and fatigue, but no serious idiosyncratic reactions or toxicities, have been reported [189,190]. Long-term GP therapy may increase the risks of fracture [42] and bone loss [19], suggesting that GP may have adverse effects on bone health (Table 1). 2.6. Vigabatrin (VB) VB (4-amino-5-hexenoic acid) is a GABA-aminotransferase inhibitor to antagonize the GABA degradation in synapses [191]. VB is rapidly absorbed in small intestines and widely distributed throughout the body [192]. However, hepatic dysfunction has no impact on VB dosing because VB is predominantly excreted unchanged in the urine [192]. Lower doses are necessary in patients with renal dysfunction (creatinine clearance less than 80 mL/min). Younger subjects may demand a higher dose because their clearance is higher [104]. VB mainly relies on renal elimination and it does not need binding plasma proteins [192] or metabolism [193]. When patients with epilepsy were co-treated VB with other AEDs, VB might cause a significant increase in plasma clearance of CBZ [194] and decrease in the serum PT concentration [195]. VB is effective in the treatment of pediatric patients with infantile spasms, infantile spasms secondary to tuberous sclerosis, refractory complex partial seizures, and adult patients with LGS [196]. Patients treated with VB frequently complain of headache, ataxia, dizziness, tremors, fatigue, hyperactivity, and weight gain. Patients with myoclonic seizures should not use VB as it may aggravate this sort of seizure. Patients receiving VB should routinely undergo ophthalmological examinations because VB may damage the visual field. There was a study that enrolled patients with epilepsy receiving AEDs [197] and the study could not make a conclusion regarding the negative effects of VB on human bone metabolism because of limited subjects; however, immature rats treated with VB were found to have decreased body mass gain and inhibited compact bone growth [198]. Therefore, VB should be used cautiously in children, and the bone condition of pediatric patients should be closely monitored (Table 1). 3. Conclusions AEDs are widely used for seizure treatment. However, abnormalities in bone and mineral metabolism have been frequently reported in individuals receiving EIAEDs because EIAEDs may cause hypocalcemia through triggering the catabolism of vitamin D. In vitro studies demonstrated that PB induces cultured human hepatocytes to increase the mRNA of CYP2C9, CYP2C19 [197], CYP2B6, and 3A4 [199]. Another in vitro study showed that CYP1A2, CYP2B6, and CYP3A4 were significantly induced by OXC and CBZ in HepaRG cells and human hepatocytes [200]. However, a systemic review analyzed 13 observational studies representing 68,973 patients with epilepsy. In all EIAED users, five studies illustrated decreased BMD; five studies demonstrated irrelevance to BMD; two studies reported increased incidence of fracture, and one study showed nothing to do with the incidence of fracture [201]. This finding led to no conclusion regarding the relationship between EIAEDs and bone metabolism. Additionally, it was reported that vitamin D deficiency was parallel to the low BMD in epilepsy patients on AEDs [19]. Numerous studies have shown that serum 25-hydroxyvitamin D levels are not significantly different between groups of subjects treated with either EIAEDs or NEIAEDs [60,202]. Moreover, calcium and vitamin D supplementation did not influence the prevalence of fractures in a retrospective study enrolling over 3000 patients with AEDs [203]. Taken together, these results raise public concerns on the bone growth or other medical conditions of children with epilepsy taking AEDs. So far, several newer-generation AEDs, including fosphenytoin, zonisamide, lacosamide, perampanel, eslicarbazepine, felbamate, ezogabine/retigabine, stiripentol, tiagabine, and rufinamide, have been designed and marketed [204]. Most of them have broader spectrums, fewer drug interactions, better tolerance, and minimal side effects, including bone diseases [205]. Timely withdrawal of AEDs and proper use of a new medication may avoid serious disabilities in users. In addition, supplementation of calcium and vitamin D are still recommended to epileptic patients on AEDs even though the effects of supplementation on AED-related osteopathy are controversial [206]. Acknowledgments Hueng-Chuen Fan and Ching-Shiang Chi express their gratitude to the Tungs’ Metro Harbor Hospital for the grants TTMHH-105C-0028. Author contributions Hueng-Chuen Fan, Herng-Shen Lee and Ching-Shiang Chi were involved in conception and literature review and drafting the manuscript; Hueng-Chuen Fan and Ching-Shiang Chi revised the manuscript critically for important intellectual content; Kai-Ping Chang, Yi-Yen Lee, Hsin-Chuan Lai and Pi-Lien Hung and Hsiu-Fen Lee provided critical questions and suggestions to the manuscripts; Ching-Shiang Chi conceptualized the review, supervised all aspects of the study, critically reviewed and revised the manuscript, and approved the final manuscript as submitted. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Pathways of the benzodiazepines (BZD) biotransformation. CYP: cytochrome P450; UGT: Uridine 5′-diphospho-glucuronosyltransferase. Figure 2 Pathways of the benzodiazepines (CBZ) biotransformation; CBZ-E: CBZ 10.11-epoxide; MPO: myeloperoxidase; EPXH1: epoxide hydrolase 1. Figure 3 Pathways of the PT biotransformation. HPPH: hydroxyphenytoin, 5-(4′-hydroxyphenyl)-5-phenylhydantoin; PAO: phenytoin-arene oxide; PDH: phenytoin dihydrodiol; PC: phenytoin catechol; NQO1: NAD(P)H dehydrogenase, quinone 1; PMC: phenytoin methylcatechol; COMT: Catechol-O-methyltransferase. Figure 4 Pathways of the VPA biotransformation. VPA-CoA: valproyl-CoA; EC 6.2.1.2: medium-chain acyl-CoA synthase; 2-ene-VPA-CoA: 2-propyl-valproyl-CoA; 2MBCAD: 2-methyl-branched chain acyl-CoA dehydrogenase; IVD: Isovaleryl-CoA dehydrogenase; 3-OH-VPA-CoA: 3-hydroxyl-valproyl-VPA; EH: 2-enoyl-CoA hydratase; 3-oxo-VPA-CoA: 3-keto-valproyl-CoA; HADH: hydroxyacyl-CoA dehydrogenase; MHBD: 2-methyl-3-hydroxybutyryl-CoA dehydrogenase; C3-CoA: propionyl-CoA; and C5-CoA: pentanoyl-CoA. ijms-17-01242-t001_Table 1Table 1 Review of literature regarding each anti-epileptic drug (AED) on the bone metabolism. Literature was classified into in vitro, in vivo, pediatric, adult, and animal group according to the study design. Abbreviation: BZD: benzodiazepines; CBZ: carbamazepine; PT: phenytoin; PB: phenobarbital; VPA: valproic acid; LEV: levetiracetam; OXC: oxcarbazepine; LTG: lamotrigine; TPM: topiramate; GP: gabapentin; VB: vigabatrin. Drug Study Design In Vitro In Vivo Pediatric Adult Animal BZD 48 24, 41, 42, 43, 44, 46, 47, 49 21, 44 24, 41, 42, 43, 46, 47, 79 CBZ 77 19, 42, 58, 29, 60, 61, 79, 81, 164 58, 60, 194 42, 59, 60, 61, 79, 81, 164, 194 PT 65, 66, 67, 77, 78, 79, 80, 82 42, 79, 80, 81, 86, 164 19, 42, 79, 80, 81, 86, 164, 195 PB 100 19, 42, 45, 79, 81, 100 100 19, 42, 45, 79, 81, 100 100 VPA 121, 125 19, 121, 122, 123, 124, 164 121, 122, 124 123, 164 LEV 140, 142, 143 142 142, 143 140 OXC 121 152, 153, 154, 156 153, 154, 155, 156 152, 155 LTG 121 19, 121, 162, 163, 164, 165 121, 162, 164, 165 19, 163, 164 TPM 121, 173 121, 174, 175 121, 174 121 GBP 19, 42 19, 42 VGB 198 194, 195 194, 195 194, 195 ==== Refs References 1. Kim H. Thurman D.J. Durgin T. Faught E. Helmers S. Estimating Epilepsy Incidence and Prevalence in the US Pediatric Population Using Nationwide Health Insurance Claims Data J. Child Neurol. 2015 10.1177/0883073815620676 26719495 2. Maguire M. Marson A.G. Ramaratnam S. Epilepsy (generalised) BMJ Clin. Evid. 2012 2 1201 3. Chiang K.L. Cheng C.Y. Prevalence and neuro-psychiatric comorbidities of pediatric epilepsy in Taiwan: A national population-based study Epilepsy Res. 2014 108 1451 1460 10.1016/j.eplepsyres.2014.07.004 25107685 4. Kwan P. Brodie M.J. Early identification of refractory epilepsy N. Engl. J. Med. 2000 342 314 319 10.1056/NEJM200002033420503 10660394 5. Walia K.S. Khan E.A. Ko D.H. Raza S.S. Khan Y.N. Side effects of antiepileptics—A review Pain Pract. 2004 4 194 203 10.1111/j.1533-2500.2004.04304.x 17173601 6. Robinson P.B. Harris M. Harvey W. Abnormal skeletal and dental growth in epileptic children Br. Dent. J. 1983 154 9 13 10.1038/sj.bdj.4804977 6600617 7. Morijiri Y. Sato T. Factors causing rickets in institutionalised handicapped children on anticonvulsant therapy Arch. Dis. Child. 1981 56 446 449 10.1136/adc.56.6.446 6973324 8. Ali I.I. Schuh L. Barkley G.L. Gates J.R. Antiepileptic drugs and reduced bone mineral density Epilepsy Behav. 2004 5 296 300 10.1016/j.yebeh.2004.02.005 15145297 9. Farhat G. Yamout B. Mikati M.A. Demirjian S. Sawaya R. El-Hajj F.G. Effect of antiepileptic drugs on bone density in ambulatory patients Neurology 2002 58 1348 1353 10.1212/WNL.58.9.1348 12011279 10. Lazzari A.A. Dussault P.M. Thakore-James M. Gagnon D. Baker E. Davis S.A. Houranieh A.M. Prevention of bone loss and vertebral fractures in patients with chronic epilepsy—Antiepileptic drug and osteoporosis prevention trial Epilepsia 2013 54 1997 2004 10.1111/epi.12351 24010637 11. Tsiropoulos I. Andersen M. Nymark T. Lauritsen J. Gaist D. Hallas J. Exposure to antiepileptic drugs and the risk of hip fracture: A case-control study Epilepsia 2008 49 2092 2099 10.1111/j.1528-1167.2008.01640.x 18479399 12. Carbone L.D. Johnson K.C. Robbins J. Larson J.C. Curb J.D. Watson K. Gass M. LaCroix A.Z. Antiepileptic drug use, falls, fractures, and BMD in postmenopausal women: Findings from the women’s health initiative (WHI) J. Bone Miner. Res. 2010 25 873 881 10.1359/jbmr.091027 19839772 13. Souverein P.C. Webb D.J. Weil J.G. van Staa T.P. Egberts A.C. Use of antiepileptic drugs and risk of fractures: Case-control study among patients with epilepsy Neurology 2006 66 1318 1324 10.1212/01.wnl.0000210503.89488.88 16682661 14. Shen C. Chen F. Zhang Y. Guo Y. Ding M. Association between use of antiepileptic drugs and fracture risk: A systematic review and meta-analysis Bone 2014 64 246 253 10.1016/j.bone.2014.04.018 24780876 15. Souverein P.C. Webb D.J. Petri H. Weil J. Van Staa T.P. Egberts T. Incidence of fractures among epilepsy patients: A population-based retrospective cohort study in the General Practice Research Database Epilepsia 2005 46 304 310 10.1111/j.0013-9580.2005.23804.x 15679512 16. Cock H.R. Bone health in epilepsy Epilepsy 2003 391 400 17. Abes M. Sarihan H. Madenci E. Evaluation of bone mineral density with dual X-ray absorptiometry for osteoporosis in children with bladder augmentation J. Pediatr. Surg. 2003 38 230 232 10.1053/jpsu.2003.50050 12596110 18. Petty S.J. O’Brien T.J. Wark J.D. Anti-epileptic medication and bone health Osteoporos. Int. 2007 18 129 142 10.1007/s00198-006-0185-z 17091219 19. Andress D.L. Ozuna J. Tirschwell D. Grande L. Johnson M. Jacobson A.F. Spain W. Antiepileptic drug-induced bone loss in young male patients who have seizures Arch. Neurol. 2002 59 781 786 10.1001/archneur.59.5.781 12020260 20. Meier C. Kraenzlin M.E. Antiepileptics and bone health Ther. Adv. Musculoskelet. Dis. 2011 3 235 243 10.1177/1759720X11410769 22870482 21. Gough H. Goggin T. Bissessar A. Baker M. Crowley M. Callaghan N. A comparative study of the relative influence of different anticonvulsant drugs, UV exposure and diet on vitamin D and calcium metabolism in out-patients with epilepsy QJM 1986 59 569 577 3092278 22. Pack A.M. Gidal B. Vazquez B. Bone disease associated with antiepileptic drugs Clevel. Clin. J. Med. 2004 71 S42 S48 10.3949/ccjm.71.Suppl_2.S42 23. Pack A.M. The Association Between Antiepileptic Drugs and Bone Disease Epilepsy Curr. 2003 3 91 95 10.1046/j.1535-7597.2003.03306.x 15309069 24. Vestergaard P. Rejnmark L. Mosekilde L. Fracture risk associated with use of antiepileptic drugs Epilepsia 2004 45 1330 1337 10.1111/j.0013-9580.2004.18804.x 15509233 25. Sheth R.D. Harden C.L. Screening for bone health in epilepsy Epilepsia 2007 48 39 41 10.1111/j.1528-1167.2007.01401.x 18047601 26. Sheth R.D. Metabolic concerns associated with antiepileptic medications Neurology 2004 63 S24 S29 10.1212/WNL.63.10_suppl_4.S24 15557547 27. Samaniego E.A. Sheth R.D. Bone consequences of epilepsy and antiepileptic medications Semin. Pediatr. Neurol. 2007 14 196 200 10.1016/j.spen.2007.08.006 18070676 28. Mackenzie P.I. Owens I.S. Burchell B. Bock K.W. Bairoch A. Belanger A. Gigleux S.F. Green M. Hum D.W. Iyanagi T. The UDP glycosyltransferase gene superfamily: Recommended nomenclature update based on evolutionary divergence Pharmacogenetics 1997 7 255 269 10.1097/00008571-199708000-00001 9295054 29. Guillemette C. Pharmacogenomics of human UDP-glucuronosyltransferase enzymes Pharmacogenom. J. 2003 3 136 158 10.1038/sj.tpj.6500171 12815363 30. American Druggist Top 200 Drugs of 1995 Hearst Corp New York, NY, USA 1996 18 26 31. Gavish M. Snyder S.H. Benzodiazepine recognition sites on GABA receptors Nature 1980 287 651 652 10.1038/287651a0 6253806 32. Olsen R.W. GABA-benzodiazepine-barbiturate receptor interactions J. Neurochem. 1981 37 1 13 10.1111/j.1471-4159.1981.tb05284.x 6265597 33. Olkkola K.T. Ahonen J. Midazolam and other benzodiazepines Handb. Exp. Pharmacol. 2008 182 335 360 18175099 34. Picotte J.J. Rosenthal D.M. Rhode J.M. Cruzan M.B. Plastic responses to temporal variation in moisture availability: Consequences for water use efficiency and plant performance Oecologia 2007 153 821 832 10.1007/s00442-007-0794-z 17636336 35. Riss J. Cloyd J. Gates J. Collins S. Benzodiazepines in epilepsy: Pharmacology and pharmacokinetics Acta Neurol. Scand. 2008 118 69 86 10.1111/j.1600-0404.2008.01004.x 18384456 36. Mandrioli R. Mercolini L. Raggi M.A. Benzodiazepine metabolism: An analytical perspective Curr. Drug Metab. 2008 9 827 844 10.2174/138920008786049258 18855614 37. Court M.H. Duan S.X. Guillemette C. Journault K. Krishnaswamy S. Von Moltke L.L. Greenblatt D.J. Stereoselective conjugation of oxazepam by human UDP-glucuronosyltransferases (UGTs): S-oxazepam is glucuronidated by UGT2B15, while R-oxazepam is glucuronidated by UGT2B7 and UGT1A9 Drug Metab. Dispos. 2002 30 1257 1265 10.1124/dmd.30.11.1257 12386133 38. Klieber S. Hugla S. Ngo R. Arabeyre-Fabre C. Meunier V. Sadoun F. Fedeli O. Rival M. Bourrie M. Guillou F. Contribution of the N -glucuronidation pathway to the overall in vitro metabolic clearance of midazolam in humans Drug Metab. Dispos. 2008 36 851 862 10.1124/dmd.107.019539 18256203 39. Miller M.E. Garland W.A. Min B.H. Ludwick B.T. Ballard R.H. Levy R.H. Clonazepam acetylation in fast and slow acetylators Clin. Pharmacol. Ther. 1981 30 343 347 10.1038/clpt.1981.170 7273597 40. Olivera M. Martinez C. Gervasini G. Carrillo J.A. Ramos S. Benitez J. García-Martin E. Agúndez J.A. Effect of common NAT2 variant alleles in the acetylation of the major clonazepam metabolite, 7-aminoclonazepam Drug Metab. Lett. 2007 1 3 5 10.2174/187231207779814283 19356010 41. Finkle W.D. Der J.S. Greenland S. Adams J.L. Ridgeway G. Blaschke T. Wang Z. Dell R.M. VanRiper K.B. Risk of fractures requiring hospitalization after an initial prescription for zolpidem, alprazolam, lorazepam, or diazepam in older adults J. Am. Geriatr. Soc. 2011 59 1883 1890 10.1111/j.1532-5415.2011.03591.x 22091502 42. Jette N. Lix L.M. Metge C.J. Prior H.J. McChesney J. Leslie W.D. Association of antiepileptic drugs with nontraumatic fractures: A population-based analysis Arch. Neurol. 2011 68 107 112 10.1001/archneurol.2010.341 21220681 43. Carbone L. Chin A.S. Lee T.A. Burns S.P. Svircev J.N. Hoenig H. Akhigbe T. Thomas F. Bailey L. Weaver F. The association of anticonvulsant use with fractures in spinal cord injury Am. J. Phys. Med. Rehabil. 2013 92 1037 1046 10.1097/PHM.0000000000000014 24252933 44. Willemsen M.A. van der Wal K.G. Medication-induced mandibular luxation in a seven-year-old patient Tijdschr. Psychiatr. 2008 50 61 64 18188830 45. Young R.E. Ramsay L.E. Murray T.S. Barbiturates and serum calcium in the elderly Postgrad. Med. J. 1977 53 212 215 10.1136/pgmj.53.618.212 323838 46. Ensrud K.E. Walczak T.S. Blackwell T.L. Ensrud E.R. Barrett-Connor E. Orwoll E.S. Antiepileptic drug use and rates of hip bone loss in older men: A prospective study Neurology 2008 71 723 730 10.1212/01.wnl.0000324919.86696.a9 18765648 47. Kulak C.A. Borba V.Z. Bilezikian J.P. Silvado C.E. Paola L. Boguszewski C.L. Bone mineral density and serum levels of 25 OH vitamin D in chronic users of antiepileptic drugs Arq. Neuropsiquiatr. 2004 62 940 948 10.1590/S0004-282X2004000600003 15608949 48. Zhang T. Shao H. Xu K.Q. Kuang L.T. Chen R.F. Xiu H.H. Midazolam suppresses osteogenic differentiation of human bone marrow-derived mesenchymal stem cells Eur. Rev. Med. Pharmacol. Sci. 2014 18 1411 1418 24867522 49. Kim K.A. Oh S.O. Park P.W. Park J.Y. Effect of probenecid on the pharmacokinetics of carbamazepine in healthy subjects Eur. J. Clin. Pharmacol. 2005 61 275 280 10.1007/s00228-005-0940-7 15915352 50. Pearce R.E. Lu W. Wang Y. Uetrecht J.P. Correia M.A. Leeder J.S. Pathways of carbamazepine bioactivation in vitro . III. The role of human cytochrome P450 enzymes in the formation of 2,3-dihydroxycarbamazepine Drug Metab. Dispos. 2008 36 1637 1649 10.1124/dmd.107.019562 18463198 51. Kerr B.M. Thummel K.E. Wurden C.J. Klein S.M. Kroetz D.L. Gonzalez F.J. Levy R.H. Human liver carbamazepine metabolism. Role of CYP3A4 and CYP2C8 in 10,11-epoxide formation Biochem. Pharmacol. 1994 47 1969 1979 10.1016/0006-2952(94)90071-X 8010982 52. Hara Y. Nakajima M. Miyamoto K. Yokoi T. Morphine glucuronosyltransferase activity in human liver microsomes is inhibited by a variety of drugs that are co-administered with morphine Drug Metab. Pharmacokinet. 2007 22 103 112 10.2133/dmpk.22.103 17495417 53. Staines A.G. Coughtrie M.W. Burchell B. N -Glucuronidation of carbamazepine in human tissues is mediated by UGT2B7 J. Pharmacol. Exp. Ther. 2004 311 1131 1137 10.1124/jpet.104.073114 15292462 54. Mantegazza M. Curia G. Biagini G. Ragsdale D.S. Avoli M. Voltage-gated sodium channels as therapeutic targets in epilepsy and other neurological disorders Lancet Neurol. 2010 9 413 424 10.1016/S1474-4422(10)70059-4 20298965 55. Hung S.-I. Chung W.-H. Liu Z.-S. Chen C.-H. Hsih M.-S. Hui R.-C. Chu C.-Y. Chen Y.-T. Common risk allele in aromatic antiepileptic-drug induced Stevens-Johnson syndrome and toxic epidermal necrolysis in Han Chinese Pharmacogenomics 2010 11 349 356 10.2217/pgs.09.162 20235791 56. Albani F. Riva R. Baruzzi A. Carbamazepine clinical pharmacology: A review Pharmacopsychiatry 1995 28 235 244 10.1055/s-2007-979609 8773289 57. Ganeva M. Gancheva T. Lazarova R. Troeva J. Baldaranov I. Vassilev I. Hristakieva E. Tzaneva V. Carbamazepine-induced drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome: Report of four cases and brief review Int. J. Dermatol. 2008 47 853 860 10.1111/j.1365-4632.2008.03637.x 18717872 58. Oakeshott P. Hunt G.M. Carbamazepine and spina bifida BMJ 1991 303 651 10.1136/bmj.303.6803.651-b 1932927 59. Kim S.H. Lee J.W. Choi K.G. Chung H.W. Lee H.W. A 6-month longitudinal study of bone mineral density with antiepileptic drug monotherapy Epilepsy Behav. 2007 10 291 295 10.1016/j.yebeh.2006.11.007 17224308 60. Verrotti A. Greco R. Latini G. Morgese G. Chiarelli F. Increased bone turnover in prepubertal, pubertal, and postpubertal patients receiving carbamazepine Epilepsia 2002 43 1488 1492 10.1046/j.1528-1157.2002.13002.x 12460249 61. Pack A.M. Morrell M.J. McMahon D.J. Shane E. Normal vitamin D and low free estradiol levels in women on enzyme-inducing antiepileptic drugs Epilepsy Behav. 2011 21 453 458 10.1016/j.yebeh.2011.05.001 21704565 62. Gallop K. Review article: Phenytoin use and efficacy in the ED Emerg. Med. Australas. 2010 22 108 118 10.1111/j.1742-6723.2010.01269.x 20534046 63. Kulshrestha V.K. Thomas M. Wadsworth J. Richens A. Interaction between phenytoin and antacids Br. J. Clin. Pharmacol. 1978 6 177 179 10.1111/j.1365-2125.1978.tb00848.x 678395 64. Vecht C.J. Wagner G.L. Wilms E.B. Interactions between antiepileptic and chemotherapeutic drugs Lancet Neurol. 2003 2 404 409 10.1016/S1474-4422(03)00435-6 12849118 65. Wong P.T. Teo W.L. The effect of phenytoin on glutamate and GABA transport Neurochem. Res. 1986 11 1379 1382 10.1007/BF00966131 2946971 66. Pincus J.H. Lee S. Diphenylhydantoin and calcium. Relation to norepinephrine release from brain slices Arch. Neurol. 1973 29 239 244 10.1001/archneur.1973.00490280051007 4728183 67. Griffith W.H. Taylor L. Phenytoin reduces excitatory synaptic transmission and post-tetanic potentiation in the in vitro hippocampus J. Pharmacol. Exp. Ther. 1988 246 851 858 2843632 68. Lipkind G.M. Fozzard H.A. Molecular model of anticonvulsant drug binding to the voltage-gated sodium channel inner pore Mol. Pharmacol. 2010 78 631 638 10.1124/mol.110.064683 20643904 69. Leeder J.S. Mechanisms of idiosyncratic hypersensitivity reactions to antiepileptic drugs Epilepsia 1998 39 S8 S16 10.1111/j.1528-1157.1998.tb01679.x 9798756 70. Nakajima M. Yamanaka H. Fujiwara R. Katoh M. Yokoi T. Stereoselective glucuronidation of 5-(4′-hydroxyphenyl)-5-phenylhydantoin by human UDP-glucuronosyltransferase (UGT) 1A1, UGT1A9, and UGT2B15: Effects of UGT-UGT interactions Drug Metab. Dispos. 2007 35 1679 1686 10.1124/dmd.107.015909 17576806 71. Claesen M. Moustafa M.A. Adline J. Vandervorst D. Poupaert J.H. Evidence for an arene oxide-NIH shift pathway in the metabolic conversion of phenytoin to 5-(4-hydroxyphenyl)-5-phenylhydantoin in the rat and in man Drug Metab. Dispos. 1982 10 667 671 6130920 72. Komatsu T. Yamazaki H. Asahi S. Gillam E.M. Guengerich F.P. Nakajima M. Yokoi T. Formation of a dihydroxy metabolite of phenytoin in human liver microsomes/cytosol: Roles of cytochromes P450 2C9, 2C19, and 3A4 Drug Metab. Dispos. 2000 28 1361 1368 11038165 73. Cuttle L. Munns A.J. Hogg N.A. Scott J.R. Hooper W.D. Dickinson R.G. Gillam E.M. Phenytoin metabolism by human cytochrome P450: Involvement of P450 3A and 2C forms in secondary metabolism and drug-protein adduct formation Drug Metab. Dispos. 2000 28 945 950 10901705 74. Yamanaka H. Nakajima M. Hara Y. Katoh M. Tachibana O. Yamashita J. Yokoi T. Urinary excretion of phenytoin metabolites, 5-(4′-hydroxyphenyl)-5-phenylhydantoin and its O -glucuronide in humans and analysis of genetic polymorphisms of UDP-glucuronosyltransferases Drug Metab. Pharmacokinet. 2005 20 135 143 10.2133/dmpk.20.135 15855726 75. Anderson G.D. A mechanistic approach to antiepileptic drug interactions Ann. Pharmacother. 1998 32 554 563 10.1345/aph.17332 9606477 76. Dessens A.B. Cohen-Kettenis P.T. Mellenbergh G.J. Koppe J.G. van De Poll N.E. Boer K. Association of prenatal phenobarbital and phenytoin exposure with small head size at birth and with learning problems Acta Paediatr. 2000 89 533 541 10.1111/j.1651-2227.2000.tb00333.x 10852187 77. Feldkamp J. Becker A. Witte O.W. Scharff D. Scherbaum W.A. Long-term anticonvulsant therapy leads to low bone mineral density—Evidence for direct drug effects of phenytoin and carbamazepine on human osteoblast-like cells Exp. Clin. Endocrinol. Diabetes 2000 108 37 43 10.1055/s-0032-1329213 10768830 78. Ikedo D. Ohishi K. Yamauchi N. Kataoka M. Kido J. Nagata T. Stimulatory effects of phenytoin on osteoblastic differentiation of fetal rat calvaria cells in culture Bone 1999 25 653 660 10.1016/S8756-3282(99)00222-7 10593409 79. Kinjo M. Setoguchi S. Schneeweiss S. Solomon D.H. Bone mineral density in subjects using central nervous system-active medications Am. J. Med. 2005 118 1414.e7 1414.e12 10.1016/j.amjmed.2005.07.033 16378792 80. Lau K.H. Nakade O. Barr B. Taylor A.K. Houchin K. Baylink D.J. Phenytoin increases markers of osteogenesis for the human species in vitro and in vivo J. Clin. Endocrinol. Metab. 1995 80 2347 2353 7629228 81. Ensrud K.E. Walczak T.S. Blackwell T. Ensrud E.R. Bowman P.J. Stone K.L. Antiepileptic drug use increases rates of bone loss in older women: A prospective study Neurology 2004 62 2051 2057 10.1212/01.WNL.0000125185.74276.D2 15184613 82. Takahashi A. Onodera K. Shinoda H. Mayanagi H. Phenytoin and its metabolite, 5-(4-hydroxyphenyl)-5-phenylhydantoin, show bone resorption in cultured neonatal mouse calvaria Jpn. J. Pharmacol. 2000 82 82 84 10.1254/jjp.82.82 10874595 83. Preux P.M. Tiemagni F. Fodzo L. Kandem P. Ngouafong P. Ndonko F. Macharia W. Dongmo L. Dumas M. Antiepileptic therapies in the Mifi Province in Cameroon Epilepsia 2000 41 432 439 10.1111/j.1528-1157.2000.tb00185.x 10756409 84. Galimberti C.A. Magri F. Magnani B. Arbasino C. Cravello L. Marchioni E. Tartara A. Antiepileptic drug use and epileptic seizures in elderly nursing home residents: A survey in the province of Pavia, Northern Italy Epilepsy Res. 2006 68 1 8 10.1016/j.eplepsyres.2005.09.031 16330188 85. Wilensky A.J. Friel P.N. Levy R.H. Comfort C.P. Kaluzny S.P. Kinetics of phenobarbital in normal subjects and epileptic patients Eur. J. Clin. Pharmacol. 1982 23 87 92 10.1007/BF01061382 7128675 86. Theodore W.H. Porter R.J. Raubertas R.F. Seizures during barbiturate withdrawal: Relation to blood level Ann. Neurol. 1987 22 644 647 10.1002/ana.410220515 3426170 87. Bernus I. Dickinson R.G. Hooper W.D. Eadie M.J. Urinary excretion of phenobarbitone and its metabolites in chronically treated patients Eur. J. Clin. Pharmacol. 1994 46 473 475 10.1007/BF00191914 7957546 88. Levin S.S. Vars H.M. Schleyer H. Cooper D.Y. The metabolism and excretion of enzyme-inducing doses of phenobarbital by rats with bile fistulas Xenobiotica 1986 16 213 224 10.3109/00498258609043524 3705618 89. Toide K. Terauchi Y. Fujii T. Yamazaki H. Kamataki T. Uridine diphosphate sugar-selective conjugation of an aldose reductase inhibitor (AS-3201) by UDP-glucuronosyltransferase 2B subfamily in human liver microsomes Biochem. Pharmacol. 2004 67 1269 1278 10.1016/j.bcp.2003.11.010 15013842 90. Kwan P. Brodie M.J. Phenobarbital for the treatment of epilepsy in the 21st century: A critical review Epilepsia 2004 45 1141 1149 10.1111/j.0013-9580.2004.12704.x 15329080 91. Zelko I. Negishi M. Phenobarbital-elicited activation of nuclear receptor CAR in induction of cytochrome P450 genes Biochem. Biophys. Res. Commun. 2000 277 1 6 10.1006/bbrc.2000.3557 11027630 92. Patsalos P.N. Froscher W. Pisani F. van Rijn C.M. The importance of drug interactions in epilepsy therapy Epilepsia 2002 43 365 385 10.1046/j.1528-1157.2002.13001.x 11952767 93. Bruni J. Wilder B.J. Perchalski R.J. Hammond E.J. Villarreal H.J. Valproic acid and plasma levels of phenobarbital Neurology 1980 30 94 97 10.1212/WNL.30.1.94 6985722 94. MacDonald R.L. Rogers C.J. Twyman R.E. Barbiturate regulation of kinetic properties of the GABAA receptor channel of mouse spinal neurones in culture J. Physiol. 1989 417 483 500 10.1113/jphysiol.1989.sp017814 2482885 95. Twyman R.E. Rogers C.J. Macdonald R.L. Differential regulation of γ-aminobutyric acid receptor channels by diazepam and phenobarbital Ann. Neurol. 1989 25 213 220 10.1002/ana.410250302 2471436 96. Rho J.M. Donevan S.D. Rogawski M.A. Direct activation of GABAA receptors by barbiturates in cultured rat hippocampal neurons J. Physiol. 1996 497 509 522 10.1113/jphysiol.1996.sp021784 8961191 97. Nimaga K. Desplats D. Doumbo O. Farnarier G. Treatment with phenobarbital and monitoring of epileptic patients in rural Mali Bull. World Health Organ. 2002 80 532 537 12163916 98. Yokoro C.M. Pesquero S.M. Turchetti-Maia R.M. Francischi J.N. Tatsuo M.A. Acute phenobarbital administration induces hyperalgesia: Pharmacological evidence for the involvement of supraspinal GABA-A receptors Braz. J. Med. Biol. Res. 2001 34 397 405 10.1590/S0100-879X2001000300015 11262592 99. Nicholas J.M. Ridsdale L. Richardson M.P. Ashworth M. Gulliford M.C. Trends in antiepileptic drug utilisation in UK primary care 1993–2008: Cohort study using the General Practice Research Database Seizure 2012 21 466 470 10.1016/j.seizure.2012.04.014 22608976 100. Hahn T.J. Birge S.J. Scharp C.R. Avioli L.V. Phenobarbital-induced alterations in vitamin D metabolism J. Clin. Investig. 1972 51 741 748 10.1172/JCI106868 4335442 101. Serrano B.B. Garcia Sanchez M.J. Otero M.J. Buelga D.S. Serrano J. Dominguez-Gil A. Valproate population pharmacokinetics in children J. Clin. Pharm. Ther. 1999 24 73 80 10.1046/j.1365-2710.1999.00202.x 10319910 102. Leppik I.E. Birnbaum A.K. Epilepsy in the elderly Ann. N. Y. Acad. Sci. 2010 1184 208 224 10.1111/j.1749-6632.2009.05113.x 20146700 103. Patsalos P.N. Perucca E. Clinically important drug interactions in epilepsy: Interactions between antiepileptic drugs and other drugs Lancet Neurol. 2003 2 473 481 10.1016/S1474-4422(03)00483-6 12878435 104. Perucca E. Pharmacokinetic variability of new antiepileptic drugs at different ages Ther. Drug Monit. 2005 27 714 717 10.1097/01.ftd.0000179847.37965.4e 16404801 105. Silva M.F. Ruiter J.P. Overmars H. Bootsma A.H. van Gennip A.H. Jakobs C. Duran M. de Almeida I.T. Wanders R.J.A. Complete beta-oxidation of valproate: Cleavage of 3-oxovalproyl-CoA by a mitochondrial 3-oxoacyl-CoA thiolase Biochem. J. 2002 362 755 760 10.1042/bj3620755 11879205 106. Luis P.B. Ruiter J.P. Ijlst L. de Tavares A.I. Duran M. Mohsen A.W. Vockley J. Wanders R.J. Silva M.F. Role of isovaleryl-CoA dehydrogenase and short branched-chain acyl-CoA dehydrogenase in the metabolism of valproic acid: Implications for the branched-chain amino acid oxidation pathway Drug Metab. Dispos. 2011 39 1155 1160 10.1124/dmd.110.037606 21430231 107. Ito M. Ikeda Y. Arnez J.G. Finocchiaro G. Tanaka K. The enzymatic basis for the metabolism and inhibitory effects of valproic acid: Dehydrogenation of valproyl-CoA by 2-methyl-branched-chain acyl-CoA dehydrogenase Biochim. Biophys. Acta 1990 1034 213 218 10.1016/0304-4165(90)90079-C 2112956 108. Li J. Norwood D.L. Mao L.F. Schulz H. Mitochondrial metabolism of valproic acid Biochemistry 1991 30 388 394 1988037 109. Luis P.B. Ruiter J.P. Ofman R. Ijlst L. Moedas M. Diogo L. Garcia P. de Almeida I.T. Duran M. Wanders R.J. Valproic acid utilizes the isoleucine breakdown pathway for its complete β-oxidation Biochem. Pharmacol. 2011 82 1740 1746 10.1016/j.bcp.2011.07.103 21843514 110. Kassahun K. Hu P. Grillo M.P. Davis M.R. Jin L. Baillie T.A. Metabolic activation of unsaturated derivatives of valproic acid. Identification of novel glutathione adducts formed through coenzyme A-dependent and -independent processes Chem. Biol. Interact. 1994 90 253 275 10.1016/0009-2797(94)90014-0 8168173 111. Kassahun K. Farrell K. Abbott F. Identification and characterization of the glutathione and N -acetylcysteine conjugates of (E )-2-propyl-2,4-pentadienoic acid, a toxic metabolite of valproic acid, in rats and humans Drug Metab. Dispos. 1991 19 525 535 1676665 112. Kiang T.K. Ho P.C. Anari M.R. Tong V. Abbott F.S. Chang T.K. Contribution of CYP2C9, CYP2A6, and CYP2B6 to valproic acid metabolism in hepatic microsomes from individuals with the CYP2C9*1/*1 genotype Toxicol. Sci. 2006 94 261 271 10.1093/toxsci/kfl096 16945988 113. Ho P.C. Abbott F.S. Zanger U.M. Chang T.K. Influence of CYP2C9 genotypes on the formation of a hepatotoxic metabolite of valproic acid in human liver microsomes Pharmacogenom. J. 2003 3 335 342 10.1038/sj.tpj.6500210 14597963 114. Diaz F.J. Eap C.B. Ansermot N. Crettol S. Spina E. de Leon J. Can valproic acid be an inducer of clozapine metabolism? Pharmacopsychiatry 2014 47 89 96 10.1055/s-0034-1371866 24764199 115. Ximenes J.C. de Oliveira G.D. Siqueira R.M. Neves K.R. Santos C.G. Correia A.O. Félix F.H. Leal L.K. de Castro Brito G.A. da Graça Naffah-Mazzacorati M. Valproic acid: An anticonvulsant drug with potent antinociceptive and anti-inflammatory properties Naunyn Schmiedebergs Arch. Pharmacol. 2013 386 575 587 10.1007/s00210-013-0853-4 23584602 116. Monti B. Polazzi E. Contestabile A. Biochemical, molecular and epigenetic mechanisms of valproic acid neuroprotection Curr. Mol. Pharmacol. 2009 2 95 109 10.2174/1874467210902010095 20021450 117. Tan J. Cang S. Ma Y. Petrillo R.L. Liu D. Novel histone deacetylase inhibitors in clinical trials as anti-cancer agents J. Hematol. Oncol. 2010 10.1186/1756-8722-3-5 20132536 118. Terbach N. Williams R.S. Structure-function studies for the panacea, valproic acid Biochem. Soc. Trans. 2009 37 1126 1132 10.1042/BST0371126 19754465 119. Begriche K. Igoudjil A. Pessayre D. Fromenty B. Mitochondrial dysfunction in NASH: Causes, consequences and possible means to prevent it Mitochondrion 2006 6 1 28 10.1016/j.mito.2005.10.004 16406828 120. Kini U. Adab N. Vinten J. Fryer A. Clayton-Smith J. Dysmorphic features: An important clue to the diagnosis and severity of fetal anticonvulsant syndromes Arch. Dis. Child. Fetal Neonatal. Ed. 2006 91 F90 F95 10.1136/adc.2004.067421 16239295 121. Lee H.S. Wang S.Y. Salter D.M. Wang C.C. Chen S.J. Fan H.C. The impact of the use of antiepileptic drugs on the growth of children BMC Pediatr. 2013 10.1186/1471-2431-13-211 24354857 122. Tsukahara H. Kimura K. Todoroki Y. Ohshima Y. Hiraoka M. Shigematsu Y. Tsukahara Y. Miura M. Mayumi M. Bone mineral status in ambulatory pediatric patients on long-term anti-epileptic drug therapy Pediatr. Int. 2002 44 247 253 10.1046/j.1442-200X.2002.01561.x 11982890 123. Sato Y. Kondo I. Ishida S. Motooka H. Takayama K. Tomita Y. Maeda H. Satoh K. Decreased bone mass and increased bone turnover with valproate therapy in adults with epilepsy Neurology 2001 57 445 449 10.1212/WNL.57.3.445 11502911 124. Oner N. Kaya M. Karasalihoglu S. Karaca H. Celtik C. Tutunculer F. Bone mineral metabolism changes in epileptic children receiving valproic acid J. Paediatr. Child Health 2004 40 470 473 10.1111/j.1440-1754.2004.00431.x 15265190 125. Wu S. Legido A. De L.F. Effects of valproic acid on longitudinal bone growth J. Child Neurol. 2004 19 26 30 15032379 126. Kawagoe R. Kawagoe H. Sano K. Valproic acid induces apoptosis in human leukemia cells by stimulating both caspase-dependent and -independent apoptotic signaling pathways Leuk. Res. 2002 26 495 502 10.1016/S0145-2126(01)00151-5 11916526 127. Phillips A. Bullock T. Plant N. Sodium valproate induces apoptosis in the rat hepatoma cell line, FaO Toxicology 2003 192 219 227 10.1016/S0300-483X(03)00331-7 14580788 128. Tang R. Faussat A.M. Majdak P. Perrot J.Y. Chaoui D. Legrand O. Marie J.P. Valproic acid inhibits proliferation and induces apoptosis in acute myeloid leukemia cells expressing P-gp and MRP1 Leukemia 2004 18 1246 1251 10.1038/sj.leu.2403390 15116123 129. Blaheta R.A. Cinatl J. Jr. Anti-tumor mechanisms of valproate: A novel role for an old drug Med. Res. Rev. 2002 22 492 511 10.1002/med.10017 12210556 130. Cui S.S. Yang C.P. Bowen R.C. Bai O. Li X.M. Jiang W. Zhang X. Valproic acid enhances axonal regeneration and recovery of motor function after sciatic nerve axotomy in adult rats Brain Res. 2003 975 229 236 10.1016/S0006-8993(03)02699-4 12763612 131. Rogawski M.A. Brivaracetam: A rational drug discovery success story Br. J. Pharmacol. 2008 154 1555 1557 10.1038/bjp.2008.221 18552880 132. Haria M. Balfour J.A. Levetiracetam CNS Drugs 1997 7 159 164 10.2165/00023210-199707020-00006 23338133 133. Patsalos P.N. Clinical pharmacokinetics of levetiracetam Clin. Pharmacokinet. 2004 43 707 724 10.2165/00003088-200443110-00002 15301575 134. Coupez R. Nicolas J.M. Browne T.R. Levetiracetam—A new antiepileptic agent: Lack of in vitro and in vivo pharmacokinetic interaction with valproic acid Epilepsia 2003 44 171 178 10.1046/j.1528-1157.2003.25302.x 12558570 135. Benedetti M.S. Coupez R. Whomsley R. Nicolas J.M. Collart P. Baltes E. Comparative pharmacokinetics and metabolism of levetiracetam—A new anti-epileptic agent, in mouse, rat, rabbit and dog Xenobiotica 2004 34 281 300 10.1080/0049825042000196749 15204700 136. Lukyanetz E.A. Shkryl V.M. Kostyuk P.G. Selective blockade of N-type calcium channels by levetiracetam Epilepsia 2002 43 9 18 10.1046/j.1528-1157.2002.24501.x 11879381 137. Rigo J.M. Hans G. Nguyen L. Rocher V. Belachew S. Malgrange B. Leprince P. Moonen G. Selak I. Matagne A. The anti-epileptic drug levetiracetam reverses the inhibition by negative allosteric modulators of neuronal Br. J. Pharmacol. 2002 136 659 672 10.1038/sj.bjp.0704766 12086975 138. Madeja M. Margineanu D.G. Gorji A. Siep E. Boerrigter P. Klitgaard H. Speckmann E.J. Reduction of voltage-operated potassium currents by levetiracetam: A novel antiepileptic mechanism of action? Neuropharmacology 2003 45 661 671 10.1016/S0028-3908(03)00248-X 12941379 139. Gillard M. Chatelain P. Fuks B. Binding characteristics of levetiracetam to synaptic vesicle protein 2A (SV2A) in human brain and in CHO cells expressing the human recombinant protein Eur. J. Pharmacol. 2006 536 102 108 10.1016/j.ejphar.2006.02.022 16556440 140. Nissen-Meyer L.S. Svalheim S. Tauboll E. Reppe S. Lekva T. Solberg L.B. Melhus G. Reinholt F.P. Gjerstad L. Jemtland R. Levetiracetam, phenytoin, and valproate act differently on rat bone mass, structure, and metabolism Epilepsia 2007 48 1850 1860 10.1111/j.1528-1167.2007.01176.x 17634065 141. Svalheim S. Tauboll E. Surdova K. Ormel L. Dahl E. Aleksandersen M. McNeilly A. Gjerstad L. Ropstad E. Long-term levetiracetam treatment affects reproductive endocrine function in female Wistar rats Seizure 2008 17 203 209 10.1016/j.seizure.2007.11.018 18155931 142. Briggs D.E. French J.A. Levetiracetam safety profiles and tolerability in epilepsy patients Expert Opin. Drug Saf. 2004 3 415 424 10.1517/14740338.3.5.415 15335297 143. Di B.C. Mari F. Fattouch J. Egeo G. Vaudano A.E. Manfredi M. Prencipe M. Giallonardo A.T. Use of levetiracetam in treating epilepsy associated with other medical conditions Acta Neurol. Scand. 2006 113 82 86 16411967 144. Ambrosio A.F. Soares-Da-Silva P. Carvalho C.M. Carvalho A.P. Mechanisms of action of carbamazepine and its derivatives, oxcarbazepine, BIA 2-093, and BIA 2-024 Neurochem. Res. 2002 27 121 130 10.1023/A:1014814924965 11926264 145. Schutz H. Feldmann K.F. Faigle J.W. Kriemler H.P. Winkler T. The metabolism of 14C-oxcarbazepine in man Xenobiotica 1986 16 769 778 10.3109/00498258609043567 3765657 146. Flesch G. Francotte E. Hell F. Degen P.H. Determination of the R-(−) and S-(+) enantiomers of the monohydroxylated metabolite of oxcarbazepine in human plasma by enantioselective high-performance liquid chromatography J. Chromatogr. 1992 581 147 151 10.1016/0378-4347(92)80459-4 1429999 147. Schmutz M. Brugger F. Gentsch C. McLean M.J. Olpe H.R. Oxcarbazepine: Preclinical anticonvulsant profile and putative mechanisms of action Epilepsia 1994 35 S47 S50 10.1111/j.1528-1157.1994.tb05967.x 8039471 148. Kalis M.M. Huff N.A. Oxcarbazepine—An antiepileptic agent Clin. Ther. 2001 23 680 700 10.1016/S0149-2918(01)80019-9 11394728 149. Herranz J.L. Argumosa A. Characteristics and indications of oxcarbazepine Rev. Neurol. 2002 35 S101 S109 12373662 150. Stefani A. Pisani A. De Murtas M. Mercuri N.B. Marciani M.G. Calabresi P. Action of GP 47779, the active metabolite of oxcarbazepine, on the corticostriatal system. II. Modulation of high-voltage-activated calcium currents Epilepsia 1995 36 997 1002 10.1111/j.1528-1157.1995.tb00958.x 7555964 151. Elger C.E. Bauer J. New antiepileptic drugs in epileptology Neuropsychobiology 1998 38 145 148 10.1159/000026530 9778602 152. Mintzer S. Boppana P. Toguri J. DeSantis A. Vitamin D levels and bone turnover in epilepsy patients taking carbamazepine or oxcarbazepine Epilepsia 2006 47 510 515 10.1111/j.1528-1167.2006.00460.x 16529614 153. Cansu A. Yesilkaya E. Serdaroglu A. Hirfanoglu T.L. Camurdan O. Gulbahar O. Gücüyener K. Cinaz P. Evaluation of bone turnover in epileptic children using oxcarbazepine Pediatr. Neurol. 2008 39 266 271 10.1016/j.pediatrneurol.2008.07.001 18805365 154. Babayigit A. Dirik E. Bober E. Cakmakci H. Adverse effects of antiepileptic drugs on bone mineral density Pediatr. Neurol. 2006 35 177 181 10.1016/j.pediatrneurol.2006.03.004 16939856 155. Koo D.L. Hwang K.J. Han S.W. Kim J.Y. Joo E.Y. Shin W.C. Lee H.W. Seo D.W. Hong S.B. Effect of oxcarbazepine on bone mineral density and biochemical markers of bone metabolism in patients with epilepsy Epilepsy Res. 2014 108 442 447 10.1016/j.eplepsyres.2013.09.009 24507862 156. Lin C.M. Fan H.C. Chao T.Y. Chu D.M. Lai C.C. Wang C.C. Chen S.J. Potential effects of valproate and oxcarbazepine on growth velocity and bone metabolism in epileptic children—A medical center experience BMC Pediatr. 2016 10.1186/s12887-016-0597-7 27142370 157. Rambeck B. Wolf P. Lamotrigine clinical pharmacokinetics Clin. Pharmacokinet. 1993 25 433 443 10.2165/00003088-199325060-00003 8119045 158. Doig M.V. Clare R.A. Use of thermospray liquid chromatography-mass spectrometry to aid in the identification of urinary metabolites of a novel antiepileptic drug, Lamotrigine J. Chromatogr. 1991 554 181 189 10.1016/S0021-9673(01)88448-X 1795036 159. Werz M.A. Pharmacotherapeutics of epilepsy: Use of lamotrigine and expectations for lamotrigine extended release Ther. Clin. Risk Manag. 2008 4 1035 1046 19209284 160. Zamponi G.W. Targeting voltage-gated calcium channels in neurological and psychiatric diseases Nat. Rev. Drug Discov. 2016 15 19 34 10.1038/nrd.2015.5 26542451 161. Armijo J.A. Bravo J. Cuadrado A. Herranz J.L. Lamotrigine serum concentration-to-dose ratio: Influence of age and concomitant antiepileptic drugs and dosage implications Ther. Drug Monit. 1999 21 182 190 10.1097/00007691-199904000-00008 10217338 162. Guo C.Y. Ronen G.M. Atkinson S.A. Long-term valproate and lamotrigine treatment may be a marker for reduced growth and bone mass in children with epilepsy Epilepsia 2001 42 1141 1147 10.1046/j.1528-1157.2001.416800.x 11580761 163. Pack A. Bone health in people with epilepsy: Is it impaired and what are the risk factors? Seizure 2008 17 181 186 10.1016/j.seizure.2007.11.020 18187347 164. Pack A.M. Morrell M.J. Randall A. McMahon D.J. Shane E. Bone health in young women with epilepsy after one year of antiepileptic drug monotherapy Neurology 2008 70 1586 1593 10.1212/01.wnl.0000310981.44676.de 18443309 165. Sheth R.D. Hermann B.P. Bone mineral density with lamotrigine monotherapy for epilepsy Pediatr. Neurol. 2007 37 250 254 10.1016/j.pediatrneurol.2007.05.018 17903668 166. Doose D.R. Walker S.A. Gisclon L.G. Nayak R.K. Single-dose pharmacokinetics and effect of food on the bioavailability of topiramate, a novel antiepileptic drug J. Clin. Pharmacol. 1996 36 884 891 10.1002/j.1552-4604.1996.tb04754.x 8930774 167. Garnett W.R. Clinical pharmacology of topiramate: A review Epilepsia 2000 41 S61 S65 10.1111/j.1528-1157.2000.tb02174.x 10768303 168. Johannessen S.I. Pharmacokinetics and interaction profile of topiramate: Review and comparison with other newer antiepileptic drugs Epilepsia 1997 38 S18 S23 10.1111/j.1528-1157.1997.tb04512.x 9092953 169. Sachdeo R.C. Sachdeo S.K. Levy R.H. Streeter A.J. Bishop F.E. Kunze K.L. Mather G.G. Roskos L.K. Shen D.D. Thummel K.E. Topiramate and phenytoin pharmacokinetics during repetitive monotherapy and combination therapy to epileptic patients Epilepsia 2002 43 691 696 10.1046/j.1528-1157.2002.41701.x 12102670 170. Angehagen M. Ronnback L. Hansson E. Ben-Menachem E. Topiramate reduces AMPA-induced Ca(2+ ) transients and inhibits GluR1 subunit phosphorylation in astrocytes from primary cultures J. Neurochem. 2005 94 1124 1130 10.1111/j.1471-4159.2005.03259.x 16092949 171. Lyseng-Williamson K.A. Yang L.P. Topiramate: A review of its use in the treatment of epilepsy Drugs 2007 67 2231 2256 10.2165/00003495-200767150-00008 17927286 172. Perucca E. A pharmacological and clinical review on topiramate—A new antiepileptic drug Pharmacol. Res. 1997 35 241 256 10.1006/phrs.1997.0124 9264038 173. Sayer J.A. Pearce S.H. Diagnosis and clinical biochemistry of inherited tubulopathies Ann. Clin. Biochem. 2001 38 459 470 10.1177/000456320103800503 11587124 174. Glauser T.A. Preliminary observations on topiramate in pediatric epilepsies Epilepsia 1997 38 S37 S41 10.1111/j.1528-1157.1997.tb04517.x 9092958 175. Fraser W.D. Hyperparathyroidism Lancet 2009 374 145 158 10.1016/S0140-6736(09)60507-9 19595349 176. Rose M.A. Kam P.C. Gabapentin: Pharmacology and its use in pain management Anaesthesia 2002 57 451 462 10.1046/j.0003-2409.2001.02399.x 11966555 177. Vollmer K.O. von Hodenberg A. Kolle E.U. Pharmacokinetics and metabolism of gabapentin in rat, dog and man Arzneimittelforschung 1986 36 830 839 3730018 178. Ojemann L.M. Friel P.N. Ojemann G.A. Gabapentin concenrations in human brain (abstract) Epilepsia 1988 29 694 179. Striano P. Striano S. Gabapentin: A Ca2+ channel α 2-δ ligand far beyond epilepsy therapy Drugs Today 2008 44 353 368 10.1358/dot.2008.44.5.1186403 18548137 180. Bryans J.S. Davies N. Gee N.S. Dissanayake V.U. Ratcliffe G.S. Horwell D.C. Kneen C.O. Morrell A.I. Oles R.J. O’Toole J.C. Identification of novel ligands for the gabapentin binding site on the alpha2delta subunit of a calcium channel and their evaluation as anticonvulsant agents J. Med. Chem. 1998 41 1838 1845 10.1021/jm970649n 9599234 181. Loscher W. Honack D. Taylor C.P. Gabapentin increases aminooxyacetic acid-induced GABA accumulation in several regions of rat brain Neurosci. Lett. 1991 128 150 154 10.1016/0304-3940(91)90249-S 1945036 182. Hill D.R. Suman-Chauhan N. Woodruff G.N. Localization of [3 H] gabapentin to a novel site in rat brain: Autoradiographic studies Eur. J. Pharmacol. 1993 244 303 309 10.1016/0922-4106(93)90156-4 8384571 183. Petroff O.A. Hyder F. Rothman D.L. Mattson R.H. Effects of gabapentin on brain GABA, homocarnosine, and pyrrolidinone in epilepsy patients Epilepsia 2000 41 675 680 10.1111/j.1528-1157.2000.tb00227.x 10840398 184. Schlicker E. Reimann W. Gothert M. Gabapentin decreases monoamine release without affecting acetylcholine release in the brain Arzneimittelforschung 1985 35 1347 1349 4084337 185. Hara K. Sata T. Inhibitory effect of gabapentin on N -methyl-d -aspartate receptors expressed in Xenopus oocytes Acta Anaesthesiol. Scand. 2007 51 122 128 10.1111/j.1399-6576.2006.01183.x 17073851 186. LaRoche S.M. Helmers S.L. The new antiepileptic drugs: Scientific review JAMA 2004 291 605 614 10.1001/jama.291.5.605 14762040 187. Goa K.L. Sorkin E.M. Gabapentin. A review of its pharmacological properties and clinical potential in epilepsy Drugs 1993 46 409 427 10.2165/00003495-199346030-00007 7693432 188. Tanabe M. Takasu K. Kasuya N. Shimizu S. Honda M. Ono H. Role of descending noradrenergic system and spinal alpha2-adrenergic receptors in the effects of gabapentin on thermal and mechanical nociception after partial nerve injury in the mouse Br. J. Pharmacol. 2005 144 703 714 10.1038/sj.bjp.0706109 15678083 189. Dalal A. Zhou L. Gabapentin and sexual dysfunction: Report of two cases Neurologist 2008 14 50 51 10.1097/NRL.0b013e3181454823 18195659 190. DeToledo J.C. Toledo C. DeCerce J. Ramsay R.E. Changes in body weight with chronic, high-dose gabapentin therapy Ther. Drug Monit. 1997 19 394 396 10.1097/00007691-199708000-00006 9263379 191. Hadjiloizou S.M. Bourgeois B.F. Antiepileptic drug treatment in children Expert Rev. Neurother. 2007 7 179 193 10.1586/14737175.7.2.179 17286551 192. Johannessen S.I. Tomson T. Pharmacokinetic variability of newer antiepileptic drugs: When is monitoring needed? Clin. Pharmacokinet. 2006 45 1061 1075 10.2165/00003088-200645110-00002 17048972 193. Benedetti M.S. Enzyme induction and inhibition by new antiepileptic drugs: A review of human studies Fundam. Clin. Pharmacol. 2000 14 301 319 10.1111/j.1472-8206.2000.tb00411.x 11030437 194. Sanchez-Alcaraz A. Quintana M.B. Lopez E. Rodriguez I. Llopis P. Effect of vigabatrin on the pharmacokinetics of carbamazepine J. Clin. Pharm. Ther. 2002 27 427 430 10.1046/j.1365-2710.2002.00441.x 12472982 195. Rimmer E.M. Richens A. Interaction between vigabatrin and phenytoin Br. J. Clin. Pharmacol. 1989 27 27S 33S 10.1111/j.1365-2125.1989.tb03458.x 2757906 196. Chong D.J. Bazil C.W. Update on anticonvulsant drugs Curr. Neurol. Neurosci. Rep. 2010 10 308 318 10.1007/s11910-010-0120-8 20490745 197. Stephen L.J. McLellan A.R. Harrison J.H. Shapiro D. Dominiczak M.H. Sills G.J. Brodie M.J. Bone density and antiepileptic drugs: A case-controlled study Seizure 1999 8 339 342 10.1053/seiz.1999.0301 10512774 198. Nowinska B. Folwarczna J. Dusilo A. Pytlik M. Sliwinski L. Cegiela U. Kaczmarczyk-Sedlak I. Pietryka W. Hanke T. Trzeciak H.I. Effects of vigabatrin on the skeletal system of young rats Acta Pol. Pharm. 2012 69 327 334 22568048 199. Martin H. Sarsat J.P. de Waziers I. Housset C. Balladur P. Beaune P. Albaladejo V. Lerche-Langrand C. Induction of cytochrome P450 2B6 and 3A4 expression by phenobarbital and cyclophosphamide in cultured human liver slices Pharm. Res. 2003 20 557 568 10.1023/A:1023234429596 12739762 200. Sugiyama I. Murayama N. Kuroki A. Kota J. Iwano S. Yamazaki H. Hirota T. Evaluation of cytochrome P450 inductions by anti-epileptic drug oxcarbazepine, 10-hydroxyoxcarbazepine, and carbamazepine using human hepatocytes and HepaRG cells Xenobiotica 2016 46 765 774 10.3109/00498254.2015.1118774 26711482 201. Fraser L.A. Burneo J.G. Fraser J.A. Enzyme-inducing antiepileptic drugs and fractures in people with epilepsy: A systematic review Epilepsy Res. 2015 116 59 66 10.1016/j.eplepsyres.2015.07.003 26354168 202. Pack A.M. Morrell M.J. Marcus R. Holloway L. Flaster E. Done S. Randall A. Seale C. Shane E. Bone mass and turnover in women with epilepsy on antiepileptic drug monotherapy Ann. Neurol. 2005 57 252 257 10.1002/ana.20378 15668966 203. Espinosa P.S. Perez D.L. Abner E. Ryan M. Association of antiepileptic drugs, vitamin D, and calcium supplementation with bone fracture occurrence in epilepsy patients Clin. Neurol. Neurosurg. 2011 113 548 551 10.1016/j.clineuro.2011.03.011 21507568 204. Abraham S. Shaju M. Innovations in epilepsy management—An overview J. Pharm. Pharm. Sci. 2013 16 564 576 10.18433/J32K5B 24210064 205. Asconape J.J. Epilepsy: New drug targets and neurostimulation Neurol. Clin. 2013 31 785 798 10.1016/j.ncl.2013.04.001 23896505 206. Miziak B. Blaszczyk B. Chroscinska-Krawczyk M. Danilkiewicz G. Jagiello-Wojtowicz E. Czuczwar S.J. The problem of osteoporosis in epileptic patients taking antiepileptic drugs Expert Opin. Drug Saf. 2014 13 935 946 10.1517/14740338.2014.919255 24821596

PMC005xxxxxx/PMC5000641.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081243ijms-17-01243ArticleFemtosecond Laser Patterning of the Biopolymer Chitosan for Biofilm Formation Estevam-Alves Regina 12Ferreira Paulo Henrique Dias 3Coatrini Andrey C. 12Oliveira Osvaldo N. Jr.1Fontana Carla Raquel 4*Mendonca Cleber Renato 1*Sashiwa Hitoshi Academic Editor1 São Carlos Institute of Physics, University of São Paulo (USP), São Carlos 13566-590, SP, Brazil; [email protected] (R.E.-A.); [email protected] (A.C.C.); [email protected] (O.N.O.J.)2 Department of Materials Engineering, School of Engineering of São Carlos (USP), São Carlos 13563-120, SP, Brazil3 Physics Department, Federal University of São Carlos (UFSCAR), São Carlos 13565-905, SP, Brazil; [email protected] Faculdade de Ciencias Farmaceuticas, UNESP—Univ. Estadual Paulista, Campus Araraquara, Departamento de Analises Clinicas, Araraquara 14800-903, SP, Brazil* Correspondence: [email protected] (C.R.F.); [email protected] (C.R.M.); Tel.: +55-16-3301-5727 (C.R.F.)19 8 2016 8 2016 17 8 124316 6 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Controlling microbial growth is crucial for many biomedical, pharmaceutical and food industry applications. In this paper, we used a femtosecond laser to microstructure the surface of chitosan, a biocompatible polymer that has been explored for applications ranging from antimicrobial action to drug delivery. The influence of energy density on the features produced on chitosan was investigated by optical and atomic force microscopies. An increase in the hydrophilic character of the chitosan surface was attained upon laser micromachining. Patterned chitosan films were used to observe Staphylococcus aureus (ATCC 25923) biofilm formation, revealing an increase in the biofilm formation in the structured regions. Our results indicate that fs-laser micromachining is an attractive option to pattern biocompatible surfaces, and to investigate basic aspects of the relationship between surface topography and bacterial adhesion. fs-laser micromachiningmicropatterningchitosanbacterial growth ==== Body 1. Introduction Nano- and micro-patterned surfaces have attracted increasing attention in recent years among the scientific communities of physics, chemistry, medicine and biology [1,2,3,4]. In particular, it has been shown that the surface chemistry and topography (surface structure) of biomaterial substrates has a strong effect not only on cell morphology [5], but it can also have an influence on regulating cell behavior, such as adhesion, migration, orientation, guidance, differentiation, proliferation, gene expression and protein synthesis [6]. Structured surfaces are now used in biosensors, in biochips for diagnostics and cell microarrays, in drug delivery and in prostheses for medical implants. Understanding cell-substrate interactions is crucial to further develop these technologies efficiently [7]. Surface adhesion, for instance, is known as a survival mechanism for bacteria [8]. Therefore, the understanding of the bacterial-surface interaction mechanism is essential for the design of biomaterial surfaces with improved properties, to either allow or prevent bacteria attachment. The interactions between bacterial cells and different material surfaces such as glass, ceramics, metals and polymers have been studied to minimize or prevent bacterial colonization [9,10,11,12]. Cell adhesion is governed by several factors, including the physicochemical properties of the cell and of the substrate, and the conditions under which attachment takes place [13]. Bacterial adhesion, in particular, is less understood because of the diversity and complexity of both the bacterial cells and the surfaces [11]. Basically, two types of strategies to control cell adhesion have been used: (i) chemical or physical modification of substrates and (ii) coating the surface with biocompatible and/or bioactive agents favorable for cell attachment [11]. Surfaces have been modified by coating with nanostructured films or by changes in topography [14,15], in some cases with decreased bio-adhesion by tailored micropatterns [16], and in others to allow control of the cell growth direction [17]. Another approach to achieve controlled surface textures is femtosecond laser micromachining [18,19], which is advantageous in comparison to similar methods because it involves a direct, single-step and maskless procedure [20,21,22,23]. Although most of the papers using surface modification by laser technologies are focused on studying cell behavior, there are only a few studies aimed at bacterial growth. In this paper we study the fs-laser microstructuring on films of the biopolymer chitosan, a linear cationic polysaccharide [(1→4)-2-amino-2-deoxy-β-d-glucopyranose] obtained from the deacetylation of chitin [(1→4)-2-acetamido-2-deoxy-β-d-glucopyranose] encountered in crustaceans. Chitosan is biocompatible and biodegradable, and is explored as an antimicrobial agent in blood coagulation, taste sensors, bone regeneration, controlled drug delivery and conductive membranes [24,25,26,27,28,29,30,31]. We studied the influence of energy density on chitosan micromachining, the features of which were characterized by optical and atomic force microscopy. We were able to determine the energy density threshold for biopolymer removal, distinguishing it from the energy density range that leads to changes in the material. The wetting properties of micromachined surfaces were characterized by water contact angle measurements, which revealed an increase in the hydrophilic character of the chitosan surface. Once the proper fs-laser micromachining conditions were determined, we patterned chitosan films that were used as templates to observe Staphylococcus aureus biofilm formation. The results obtained indicate that micropatterned surfaces increase the hydrophilic character of the chitosan sample, which helps biofilm formation. Overall, we provide evidence that fs-laser micromachining is adequate to pattern biocompatible surfaces, with which one can also investigate the dependence of surface topography on bacterial adhesion. 2. Results and Discussion The influence of energy density on fs-laser micromachining features was analyzed by optical and atomic force microscopy. Figure 1a shows an optical microscopy image of microstructures produced in chitosan at a translation speed of 50 μm/s. Line patterns (500 μm long separated by 10 μm) were directly micromachined on the film surface using different energy densities (from 62 to 341 mJ/cm2). Figure 1b,c show two three-dimensional atomic force microscopy (AFM) morphological images of the micromachined region at energy densities of 149 and 341 mJ/cm2, respectively. When energy densities above 101 mJ/cm2 are used, the removal of the chitosan from the sample is observed, as illustrated in the AFM images. For energy densities below 62 mJ/cm2, no removal of material was observed from the sample surface (data not shown), indicating that the energy density threshold for chitosan ablation is at about 62 mJ/cm2 [32,33]. Therefore, the micromachined lines observed in Figure 1a for an energy density of 62 mJ/cm2 are only related to material changes and not removal. The width of the lines fabricated on the chitosan film increases with the energy density, as illustrated in Figure 2. As can be seen, the obtained groove width ranges from (0.54 ± 0.07) μm up to (1.33 ± 0.06) µm, representing an increase of approximately 2.5 when the energy density is changed from about 60 to 300 mJ/cm2. The use of biomaterials for applications in tissue engineering and implants depends on their surface characteristics, mainly the degree of hydrophobicity or hydrophilicity that has been reported to be an important factor influencing cell adhesion [34,35]. Microstructured chitosan surfaces revealed interesting wetting characteristics, as can be noted in Figure 3, for images and contact angles obtained with the sessile drop method on micromachined surfaces (grooves) composed of 3-mm-long lines separated by 10 μm (total area of 9.0 mm2). The contact angles were (77.0 ± 0.5)° for the non-microstructured surface (reference) and in the order of 40° for the structured surfaces made with 487 mJ/cm2 energy density. Also, the contact angle on the structured surface is anisotropic, for it depends on the observation direction of the produced pattern (lines). As illustrated in Figure 3b,c, the contact angle was (42.0 ± 0.5)° and (39.0 ± 0.5)° when observed perpendicularly and parallel to the microstructured lines, respectively. The dashed lines in Figure 3b,c represent the direction along which the images for the contact angle measurements were obtained. Thus, surface roughness caused by the fs-laser microstructuring increases the hydrophilic character of the original chitosan sample, a behavior that can be explained by the Wenzel model [36], which assumes a homogeneous contact between the water droplet and the polymer surface. Wetting characteristics can affect microbial interactions and, hence, bio-adhesion control. To demonstrate the potential of the micromachined surfaces to allow for the adhesion of bacteria, we used a 162 mJ/cm2 energy density at a 50 μm/s scan speed to fabricate micropatterned surfaces on chitosan films (thickness of approximately 3 μm). These parameters were found to be adequate for micropatterning according to Figure 2. Each microenvironment was composed by 500-μm-long lines separated by distinct spacing, varying from 4 to 12 μm. The total area of each microenvironment was 500 × 500 μm2. For comparison, the same microenvironments were fabricated on 430-μm-thick poly(methyl methacrylate) (PMMA) films. Figure 4 shows optical microscopy images of bacteria grown on micropatterned surfaces (lines spaced by 10 μm). Similar results were observed for the other spacing used. For the control PMMA sample, no difference was observed on the Staphylococcus aureus grown on the unpatterned (Figure 4a) and micropatterned surface (Figure 4b), i.e., the density of bacteria between the microfabricated lines is the same as the one observed on the unpatterned surface. For the chitosan substrate, however, the images in Figure 4c (unpatterned surface) and Figure 4d (10 μm patterned surface) reveal an increased bacteria density, seen as white dots (focus) and cloudy spots (out-of-focus), on the structured sample. According to [37], the effect of topography on cell adhesion is associated with the creation of a stable hydrophobic state, in which air pockets at the microscopic features inhibit the interaction of microbes with the surface. As displayed in the results of Figure 3, the microstructured pattern on chitosan films creates a more hydrophilic surface, which helps the interaction with bacteria cells, enabling adhesion and, therefore, biofilm formation. The overall smaller bacterial density on the chitosan substrate in comparison to that on PMMA samples is explained by the well-known antimicrobial action of chitosan [29,38]. For PMMA, given the high bacterial density achieved, the effect of the surface microstructuring is not evidenced. Biofilm formation initiates with microbial adhesion to biotic or abiotic surfaces. Growth of adherent organisms results in cluster formation referred to as complex aggregates or microcolonies. Figure 5 shows SEM images of biofilm formation on PMMA and chitosan surfaces patterned with lines separated by 8 and 12 μm. The PMMA microstructured surfaces (Figure 5a,b) exhibit dense biofilm after five days, as compared to chitosan (Figure 5c,d), which is related to the antimicrobial characteristic of chitosan. Also, the results show a higher density of bacteria in the grooves microstructured with 8 μm (Figure 5a,c) in comparison to the ones at 12 μm (Figure 5b,d), respectively. Such results indicate that the closer the microstructured lines are, the greater the bacterial interaction, because nucleation and colonization are favored by niches created by fs-laser micromachining. Owing to the antimicrobial chitosan activity, the chitosan samples (Figure 5c,d) were shown to resist biofilm formation, inhibiting bacterial attachment. However, even with the antimicrobial effect of chitosan, a behavior similar to the one displayed for PMMA samples was observed; as lines became closer, a denser biofilm was obtained on the chitosan microstructured surface, indicating an increase in bio-adhesion. Therefore, surface microstructuring can contribute to nucleation and guide the formation of three-dimensional mature biofilms. For separation between lines in the microstructured surfaces larger than 12 μm, no difference was observed on the biofilm as compared to the unpatterned surface (data not shown). In summary, there is a tradeoff between the natural antibacterial action from chitosan and the increasing adhesion of the microstructured surfaces, which promotes biofilm formation. Furthermore, micropatterning only affects bacterial growth if the grooves created by fs-laser micromachining are smaller than a given size, since 12 μm microstructured surfaces showed no difference compared to the smooth surface. 3. Materials and Methods Chitosan was purchased from Galena Chemistry & Pharmaceutical (Campinas, Brazil) and used as received. It was dissolved in acetic acid (4 mg/mL) and deposited onto glass substrates by spin coating to form ca. 100-nm-thick films. For the bacterial growth experiments, cast chitosan films with thickness up to approximately 3 μm were prepared with (4% w/v) solutions being spread onto glass substrates (1.5 mm diameter). The chitosan films were micromachined using an extended-cavity Ti:Sapphire laser oscillator, centered at 800 nm and operating at a repetition rate of 5.2 MHz, which produces pulses with energy up to 100 nJ and duration of 50 fs. The beam was focused on the sample surface by a 40× (0.67-NA) microscope objective. Further details about the micromachining system can be found elsewhere [39]. The sample was moved at a constant speed with respect to the laser beam using a computer-controlled xyz stage. Sample morphology was analyzed using an Atomic Force Microscope (AFM) from Nanosurf (Nanosurfe EasyScan 2 FlexAFM, Liestal, Switzerland) in the tapping mode, with images collected with high resolution (512 lines/scan) at a scan rate of 0.5 Hz. The samples were also examined with scanning electron microscopy (SEM), using a HITACHI TM 3000 microscope operating at 15 kV, and by optical microscopy using a LSM 700 from Zeiss (Jena, Germany). The wetting properties of the samples were studied by measuring the static contact angle (CA) for water, using a goniometer coupled to a horizontal microscope (KSV CAM 200, KSV, Helsinki, Finland). The CA measurements were performed at 25 °C and relative air humidity around 40%. Water droplets with a volume of about 3 μL (radius of ca. 1 mm) were used for all measurements. For the biological studies we used the American Type Culture Collection (Manassas, VA, USA) reference strain of Staphylococcus aureus (ATCC 25923), whose culture was maintained by weekly subculture in plates composed of Trypticase soy agar (Becton, Dickinson, and Co., Sparks, MD, USA). Biofilm development was performed by using the approach described in Reference [40]; 600 μL of the inoculum (~108 cells/mL—estimated by spectrophotometry) were carefully pipetted to cover the samples (3-μm thick microstructures chitosan substrates) placed in each well of 6-well plates. The plates were then incubated aerobically at 35 °C for five days. After an initial incubation period of 48 h, the liquid medium was carefully aspirated from each well and the biofilms were replenished with fresh broth. Then, fresh Tryptic Soy Broth (TSB) was added daily into each well, very slowly to avoid disrupting the biofilm. The chitosan films coated with biofilms were analyzed with SEM, for which the samples were thoroughly washed with sterile 0.1 M phosphate buffer (pH 7.4). The washing procedure was repeated three times and 1mL of 3% glutaraldehyde and 2% paraformaldehyde in 0.1 M potassium phosphate buffer, pH 7.4, was added. Then, three washes with pure buffer solution were performed. Dehydration was carried out with increasing concentrations of ethanol (50%, 60%, 70%, 80%, 90% and 100%). Following dehydration, the samples were dried in a desiccator with silica for 72 h and then analyzed in the microscope. 4. Conclusions This paper focused on the fs-laser microfabrication of a chitosan surface and its characterization, aiming at the production of engineered surface topographies to control bio-adhesion. Micropatterned chitosan surfaces are more hydrophilic than a smooth chitosan film, which helps the formation of Staphylococcus aureus biofilms. Our results also indicate a balance between increased adhesion on microstructured surfaces and the natural antibacterial action of chitosan. The results of this study demonstrated that fs-laser micromachining is an interesting option to pattern bio-surfaces to study bacterial growth and development, which can be used to elucidate the connection between surface topography and bacterial adhesion. The applicability of the methodology can even be extended if modified chitosans are employed, e.g., with antibiotic-loaded chitosan gels [41]. Acknowledgments This work was supported by FAPESP (2011/12399-0, 2013/14262-7, 2016/05345-4), CNPq, CAPES (Brazil), Programa de Apoio ao Desenvolvimento Científico da Faculdade de Ciências Farmacêuticas da UNESP-PADC, in addition to EU FP7 PIRSES Project (2013/612267). The authors are also thankful to André Romero and Debora T. Balogh for technical support. Author Contributions Several people contributed to the work described in this paper. Regina Estevam-Alves, Paulo Henrique Dias Ferreira and Cleber Renato Mendonca conceived the basic idea for this work and designed the experiments. Regina Estevam-Alves and Paulo Henrique Dias Ferreira carried out the experiments. Osvaldo N. Oliveira Jr. and Andrey C. Coatrini were responsible for the samples’ preparation. Carla Raquel Fontana carried out the bacteria culture and corresponding microscopy analysis. All authors took part in the writing process of the manuscript. Cleber Renato Mendonca supervised the research and edited the final manuscript. Conflicts of Interest Authors declare no conflict of interest. Figure 1 (a) Transmission optical microcopy of lines micromachined in chitosan film at a translation speed of 50 μm/s and distinct energy densities. Three-dimensional AFM images of micromachined chitosan surface with energy densities of (b) 149 and (c) 341 mJ/cm2. Figure 2 Width of micromachined lines as a function of energy density, for a scan speed of 50 μm/s. The solid line is only drawn to guide the eye. Figure 3 Water contact angle measured for: (a) reference sample (θ = (77.0 ± 0.5)°); (b) perpendicular to the microstrucutred lines (θ = (42.0 ± 0.5)°); (c) parallel to the microstructured lines (θ = (39.0 ± 0.5)°). The dashed lines represent the direction along which the contact angles were obtained. The left column is only an illustration of the contact angle anisotropy observed in the right column. Figure 4 Optical microscope images of microenvironments with biofilm formation on the unpatterned (a) and micropatterned (b) PMMA surfaces; and unpatterned (c) and micropatterned (d) chitosan surfaces. Figure 5 SEM images of microenvironments with biofilm formation on the 8 μm (a) and 12 μm (b) micropatterned PMMA surface; and 8 μm (c) and 12 μm (d) micropatterned chitosan surface. ==== Refs References 1. Caruso F. Nanoengineering of particle surfaces Adv. Mater. 2001 13 11 12 10.1002/1521-4095(200101)13:1<11::AID-ADMA11>3.0.CO;2-N 2. Cheng J.-Y. Wei C.-W. Hsu K.-H. Young T.-H. Direct-write laser micromachining and universal surface modification of PMMA for device development Sens. Actuators B Chem. 2004 99 186 196 10.1016/j.snb.2003.10.022 3. Katsikogianni M.G. Missirlis Y.F. Bacterial adhesion onto materials with specific surface chemistries under flow conditions J. Mater. Sci. Mater. Med. 2010 21 963 968 10.1007/s10856-009-3975-y 20044774 4. Bazaka K. Jacob M.V. Crawford R.J. Ivanova E.P. Plasma-assisted surface modification of organic biopolymers to prevent bacterial attachment Acta Biomater. 2011 7 2015 2028 10.1016/j.actbio.2010.12.024 21194574 5. Liu Y. Sun S. Singha S. Cho M.R. Gordon R.J. 3D femtosecond laser patterning of collagen for directed cell attachment Biomaterials 2005 26 4597 4605 10.1016/j.biomaterials.2004.11.033 15722129 6. Murugan R. Molnar P. Rao K.P. Hickman J.J. Biomaterial surface patterning of self-assembled monolayers for controlling neuronal cell behaviour Int. J. Biomed. Eng. Technol. 2009 2 104 134 10.1504/IJBET.2009.022911 20174479 7. Anselme K. Davidson P. Popa A.M. Giazzon M. Liley M. Ploux L. The interaction of cells and bacteria with surfaces structured at the nanometre scale Acta Biomater. 2010 6 3824 3846 10.1016/j.actbio.2010.04.001 20371386 8. Busscher H.J. van der Mei H.C. How do bacteria know they are on a surface and regulate their response to an adhering state? PLoS Pathog. 2012 8 1243 10.1371/journal.ppat.1002440 22291589 9. Mitik-Dineva N. Wang J. Truong V. Stoddart P. Malherbe F. Crawford R. Ivanova E. Escherichia coli , Pseudomonas aeruginosa , and Staphylococcus aureus Attachment Patterns on Glass Surfaces with Nanoscale Roughness Curr. Microbiol. 2009 58 268 273 10.1007/s00284-008-9320-8 19020934 10. Truong V. Rundell S. Lapovok R. Estrin Y. Wang J. Berndt C. Barnes D. Fluke C. Crawford R. Ivanova E. Effect of ultrafine-grained titanium surfaces on adhesion of bacteria Appl. Microbiol. Biotechnol. 2009 83 925 937 10.1007/s00253-009-1944-5 19296098 11. Ivanova E.P. Mitik-Dineva N. Wang J. Pham D.K. Wright J.P. Nicolau D.V. Mocanasu R.C. Crawford R.J. Staleya guttiformis attachment on poly(tert-butylmethacrylate) polymeric surfaces Micron 2008 39 1197 1204 10.1016/j.micron.2008.04.009 18539467 12. Puckett S.D. Taylor E. Raimondo T. Webster T.J. The relationship between the nanostructure of titanium surfaces and bacterial attachment Biomaterials 2010 31 706 713 10.1016/j.biomaterials.2009.09.081 19879645 13. Crawford R.J. Webb H.K. Truong V.K. Hasan J. Ivanova E.P. Surface topographical factors influencing bacterial attachment Adv. Colloid Interface Sci. 2012 179 142 149 10.1016/j.cis.2012.06.015 22841530 14. Cooke M.J. Phillips S.R. Shah D.S.H. Athey D. Lakey J.H. Przyborski S.A. Enhanced cell attachment using a novel cell culture surface presenting functional domains from extracellular matrix proteins Cytotechnology 2008 56 71 79 10.1007/s10616-007-9119-7 19002844 15. Alves N.M. Shi J. Oramas E. Santos J.L. Tomas H. Mano J.F. Bioinspired superhydrophobic poly(l -lactic acid) surfaces control bone marrow derived cells adhesion and proliferation J. Biomed. Mater. Res. A 2009 91 480 488 10.1002/jbm.a.32210 18985759 16. Kim E. Kinney W.H. Ovrutsky A.R. Vo D. Bai X. Honda J.R. Marx G. Peck E. Lindberg L. Falkinham J.O. III A surface with a biomimetic micropattern reduces colonization of Mycobacterium abscessus FEMS Microbiol. Lett. 2014 360 17 22 10.1111/1574-6968.12587 25155501 17. Langheinrich D. Yslas E. Broglia A. Rivarola V. Acevedo D. Lasagni A. Control of cell growth direction by direct fabrication of periodic micro- and submicrometer arrays on polymers J. Polym. Sci. B Polym. Phys. 2012 50 415 422 10.1002/polb.23017 18. Nayak B.K. Kolasinski K.W. Formation of nano-textured conical microstructures in titanium metal surface by femtosecond laser irradiation Appl. Phys. 2008 90 399 402 19. Ortiz R. Moreno-Flores S. Quintana I. Vivanco M. Sarasua J.R. Toca-Herrera J.L. Ultra-fast laser microprocessing of medical polymers for cell engineering applications Mater. Sci. Eng. C 2014 37 241 250 10.1016/j.msec.2013.12.039 24582245 20. Davis K.M. Miura K. Sugimoto N. Hirao K. Writing waveguides in glass with a femtosecond laser Opt. Lett. 1996 21 1729 1731 10.1364/OL.21.001729 19881782 21. Eldada L. Shacklette L.W. Advances in polymer integrated optics IEEE J. Sel. Top. Quantum Electron. 2000 6 54 68 10.1109/2944.826873 22. Cumpston B.H. Ananthavel S.P. Barlow S. Dyer D.L. Ehrlich J.E. Erskine L.L. Heikal A.A. Kuebler S.M. Lee I.Y.S. McCord-Maughon D. Two-photon polymerization initiators for three-dimensional optical data storage and microfabrication Nature 1999 398 51 54 23. Maruo S. Nakamura O. Kawata S. Three-dimensional microfabrication with two-photon-absorbed photopolymerization Opt. Lett. 1997 22 132 134 10.1364/OL.22.000132 18183126 24. He Q. Gong K. Ao Q. Ma T. Yan Y. Gong Y. Zhang X. Positive charge of chitosan retards blood coagulation on chitosan films J. Biomater. Appl. 2011 27 1032 1045 10.1177/0885328211432487 22207609 25. Dos Santos D.S. Jr. Riul A. Jr. Malmegrim R.R. Fonseca F.J. Oliveira O.N. Jr. Mattoso L.H.C. A layer-by-layer film of chitosan in a taste sensor application Macromol. Biosci. 2003 3 591 595 10.1002/mabi.200350027 26. Bojar W. Kucharska M. Ciach T. Koperski L. Jastrzębski Z. Szałwiński M. Bone regeneration potential of the new chitosan-based alloplastic biomaterial J. Biomater. Appl. 2014 28 1060 1068 10.1177/0885328213493682 23796631 27. Bhattarai N. Gunn J. Zhang M. Chitosan-based hydrogels for controlled, localized drug delivery Adv. Drug Deliv. Rev. 2010 62 83 99 10.1016/j.addr.2009.07.019 19799949 28. Xiong Y. Wang H. Wu C. Wang R. Preparation and characterization of conductive chitosan-ionic liquid composite membranes Polym. Adv. Technol. 2012 23 1429 1434 10.1002/pat.2061 29. Rabea E.I. Badawy M.E.-T. Stevens C.V. Smagghe G. Steurbaut W. Chitosan as antimicrobial agent: Applications and mode of action Biomacromolecules 2003 4 1457 1465 10.1021/bm034130m 14606868 30. Hu Q. Li B. Wang M. Jiacong Shen J. Preparation and characterization of biodegradable chitosan/hydroxyapatite nanocomposite rods via in situ hybridization: A potential material as internal fixation of bone fracture Biomaterials 2004 25 779 785 10.1016/S0142-9612(03)00582-9 14609666 31. Liang D. Lu Z. Yang H. Gao J. Chen R. Novel Asymmetric Wettable AgNPs/Chitosan Wound Dressing: In Vitro and In Vivo Evaluation ACS Appl. Mater. Interface 2016 10.1021/acsami.5b11160 26800283 32. Estevam-Alves R. Ferreira P.H.D. Almeida G.F.B. Sousa W.S. Mendonça C.R. Microfabrication of electroluminescent polymer for devices construction Appl. Surf. Sci. 2014 314 633 637 10.1016/j.apsusc.2014.07.074 33. Correa D.S. Cardoso M.R. Tribuzi V. Misoguti L. Mendonca C.R. Femtosecond Laser in Polymeric Materials: Microfabrication of Doped Structures and Micromachining IEEE J. Sel. Top. Quantum Electron. 2012 18 176 186 10.1109/JSTQE.2011.2106764 34. Dulinska-Molak I. Lekka M. Kurzydłowski K.J. Surface properties of polyurethane composites for biomedical applications Appl. Surf. Sci. 2013 270 553 560 10.1016/j.apsusc.2013.01.085 35. Tang Z. Akiyama Y. Okano T. Temperature-responsive polymer modified surface for cellsheet engineering Polymers 2012 4 1478 1498 10.3390/polym4031478 36. Wenzel R.N. Resistance of solid surfaces to wetting by water Ind. Eng. Chem. 1936 28 988 994 10.1021/ie50320a024 37. Decker J.T. Kirschner C.M. Long C.J. Finlay J.A. Callow M.E. Callow J.A. Brennan A.B. Engineered Antifouling Microtopographies: An Energetic Model That Predicts Cell Attachment Langmuir 2013 29 13023 13030 10.1021/la402952u 24044383 38. Regiel-Futyra A. Kus-Liśkiewicz M. Sebastian V. Irusta S. Arruebo A. Stochel G. Kyzioł A. Development of Noncytotoxic Chitosan–Gold Nanocomposites as Efficient Antibacterial Materials ACS Appl. Mater. Interface 2015 7 1087 1099 10.1021/am508094e 25522372 39. Ferreira P.H.D. Stefanutti R. Pavinatto F.J. Mendonça C.R. Femtosecond laser fabrication of waveguides in DR13-doped PMMA Opt. Commun. 2014 318 53 56 40. Fontana C.R. Abernethy A.D. Som S. Ruggiero K. Doucette S. Marcantonio R.C. Boussios C.I. Kent R. Goodson J.M. Tanner A.C.R. Soukos N.S. The antibacterial effect of photodynamic therapy in dental plaque-derived biofilms J. Periodontal Res. 2009 4 751 759 10.1111/j.1600-0765.2008.01187.x 19602126 41. Wu F. Meng G. He J. Wu Y. Wu F. Gu Z. Antibiotic-Loaded Chitosan Hydrogel with Superior Dual Functions: Antibacterial Efficacy and Osteoblastic Cell Responses ACS Appl. Mater. Interface 2014 6 10005 10013 10.1021/am502537k 24938653

PMC005xxxxxx/PMC5000642.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081244ijms-17-01244ArticleSalvianolic Acid A, as a Novel ETA Receptor Antagonist, Shows Inhibitory Effects on Tumor in Vitro Zhang Qiao 1Wang Shifeng 1Yu Yangyang 1Sun Shengnan 2Zhang Yuxin 1Zhang Yanling 1Yang Wei 3Li Shiyou 4*Qiao Yanjiang 1*Zhang Ge Academic EditorLu Aiping Academic EditorZhu Hailong Academic Editor1 School of Chinese Materia Medica, Beijing University of Chinese Medicine, No. 6 Wangjingzhonghuan South Road, Chaoyang District, Beijing 100102, China; [email protected] (Q.Z.); [email protected] (S.W.); [email protected] (Y.Y.); [email protected] (Yu.Z.); [email protected] (Ya.Z.)2 Pharmacogenetics, HD Biosciences, Co., Ltd., 590 Ruiqing Road, Zhangjiang Hi-Tech Park East Campus, Pudong New Area, Shanghai 201201, China; [email protected] Technical Department, ACEA Biosciences Inc., No. 5 Sandunxiyuan Road, Hangzhou 310030, China; [email protected] Beijing Institute of Genomics, Chinese Academy of Sciences, No. 1 Beichen West Road, Chaoyang District, Beijing 100101, China* Correspondence: [email protected] (S.L.); [email protected] (Y.Q.); Tel.: +86-10-8049-7628 (S.L.); +86-10-8473-8661 (Y.Q.)02 8 2016 8 2016 17 8 124430 5 2016 15 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Endothelin-1 (ET-1) autocrine and paracrine signaling modulate cell proliferation of tumor cells by activating its receptors, endothelin A receptor (ETAR) and endothelin B receptor (ETBR). Dysregulation of ETAR activation promotes tumor development and progression. The potential of ETAR antagonists and the dual-ETAR and ETBR antagonists as therapeutic approaches are under preclinical and clinical studies. Salvianolic acid A (Sal A) is a hydrophilic polyphenolic derivative isolated from Salvia miltiorrhiza Bunge (Danshen), which has been reported as an anti-cancer and cardio-protective herbal medicine. In this study, we demonstrate that Sal A inhibits ETAR activation induced by ET-1 in both recombinant and endogenous ETAR expression cell lines. The IC50 values were determined as 5.7 µM in the HEK293/ETAR cell line and 3.14 µM in HeLa cells, respectively. Furthermore, our results showed that Sal A suppressed cell proliferation and extended the doubling times of multiple cancer cells, including HeLa, DU145, H1975, and A549 cell lines. In addition, Sal A inhibited proliferation of DU145 cell lines stimulated by exogenous ET-1 treatment. Moreover, the cytotoxicity and cardio-toxicity of Sal A were assessed in human umbilical vein endothelial cells (HUVEC) and Human-induced pluripotent stem cell-derived cardiomyocytes (hiPS-CMs), which proved that Sal A demonstrates no cytotoxicity or cardiotoxicity. Collectively, our findings indicate that Sal A is a novel anti-cancer candidate through targeting ETAR. Salvianolic acid Aendothelin A receptoranticancerantagonistcardiotoxicity ==== Body 1. Introduction Endothelin-1 (ET-1) is a small vasoactive peptide overexpressed in plasma and tissue samples from patients with various solid cancers [1]. ET-1 effects are mediated by two distinct G protein-coupled receptors (GPCR), endothelin A receptor (ETAR) and endothelin B receptor (ETBR). Endothelin receptors in cancer cells can be activated through either autocrine production of ligand or production of ligand from stromal cells that may be expressed physiologically, or in response to cancer cells in a paracrine loop [2,3]. In vitro, ET-1 production has been detected in a number of human cancer cell lines, from such colorectal, stomach, breast, and prostate [4]. Furthermore, exogenous ET-1 added to ovarian and prostate cancer cells stimulates proliferation [5,6]. Early studies in various tumor cells revealed that spontaneous growth was significantly inhibited by ETAR antagonists, such as Atrasentan (ABT-627) and Zibotentan (ZD4054), demonstrating that endogenous ET-1 acts as an autocrine modulator of cell proliferation through ETAR [7,8,9]. ET-1 provided a growth stimulus to colorectal cancer cells; the signal was also propagated via the transactivation of the epidermal growth factor receptor (EGFR) [10,11,12]. Moreover, preclinical studies using combined treatment of Zibotentan and the EGFR inhibitor gefitinib suppressed proliferation, invasion, and vascular endothelial growth factor (VEGF) production in ovarian cancer cells [13]. Therefore, the discovery of novel ETAR antagonists would be helpful for anti-tumor procession. Natural ingredients derived from traditional Chinese medicines have shown obvious advantages in the therapy of certain diseases, such as multi-drug resistant cancer [14,15,16]. Thus, looking for candidates from traditional Chinese medicine provides a new path for the development of new drugs. Salvia polyphenols (Salvianolic acid A, B, C, D) are major active ingredients in Salvia miltiorrhizae Bunge (also termed as Danshen in China). Salvianolic acid A (Sal A) possesses multiple pharmacological activities, such as antiplatelet, anti-thrombosis, improvement of microcirculation, anti-inflammation, and antioxidant [17,18,19]. Furthermore, in recent years, it has been recognized that Sal A exerted effects on drug-resistant breast cancer cells [20,21]. Nonetheless, the potential target for Sal A on suppressing tumor cell proliferation remains to be illustrated. Here we identified Sal A as a potential antagonist of ETAR via calcium mobilization assay. The anti-proliferative effect was evaluated by multiple cell lines with or without exogenous ET-1 by cell viability and real-time cell analysis assay. Furthermore, Sal A exhibited neither remarkable cytotoxicity in human umbilical vein endothelial cells (HUVEC) nor cardiotoxicity in human-induced pluripotent stem cell-derived cardiomyocytes (hiPS-CMs). 2. Results 2.1. HEK293/Endothelin A Receptor (ETAR) Cell Line Validation The Gαq pathway is involved in ETAR activation, which increases intracellular Ca2+. Based on this characteristic, a cell-based calcium mobilization assay has been used to study the function of GPCR and Ca2+-permeable ion channels by measuring the changes of intracellular free Ca2+ levels. In our previous study, recombinant ETAR in HEK293 cells was developed according to standard procedures. The utility of the cell line was validated using ETAR agonist (ET-1) and antagonist (BQ-123). The EC50 value of ETAR agonist endothelin-1 (ET-1) was determined as 4.78 nM (shown in Figure 1), and the IC50 value of ETAR antagonist BQ-123 stimulated with 20 nM ET-1 (EC90 value) was determined as 0.1 nM, which were consistent with previously reported data [22]. 2.2. ETAR Antagonist Primary Screening Salvianolic acid A (Sal A), Salvianolic acid B (Sal B), Salvianolic C (Sal C), Salvianolic acid D (Sal D), and Salvianolic acid (Sal); these five major active compounds of Salvia miltiorrhizae Bunge (also termed as Danshen in China) were tested in the primary screening, and their chemical formulas are shown in Figure 2A. These compounds were evaluated for inhibitory effect in HEK293/ETAR cells at various concentrations of 30, 10, and 3 µM using the cell-based calcium mobilization assay. BQ-123 (1 nM) was set as positive antagonist control and 0.25% Dimethyl sulfoxide (DMSO) was defined as vehicle control. The results suggested that Sal A should be an antagonist of ETAR, and the inhibitory effect of Sal A was concentration-dependent in the HEK293/ETAR cell line. In parallel, no obvious inhibitory effect was observed for the other tested compounds (Figure 2B). 2.3. Salvianolic Acid A (Sal A) Dose-Dependently Blocked Exogenous ETAR without Inducing Cytotoxicity in the HEK293/ETAR Cell Line To acquire the antagonist dose response, we tested the inhibitory effects of Sal A at different concentrations in the HEK293/ETAR cell line with 20 nM ET-1 stimulating. A dose-dependent trend of Sal A was observed with an IC50 value of 5.7 µM (Figure 3A). To omit a potential false positive, the cytotoxicity was tested in the HEK293/ETAR cell line by ATP CellTiter-Glo assay. In brief, Sal A was incubated in HEK293/ETAR cells for 30 min before luminescence signal measurement. Sal A did not show obvious cytotoxicity in the HEK293/ETAR cell line within 45.7 nM to 100 µM, compared with vehicle control (Figure 3B). The selectivity of Sal A was further assessed by calcium influx assay on 5 GPCRs. HEK293 cell lines stably expressing human ETA receptors, ETB receptors, adenosine A1 receptor (A1), angiotensin II type 1 receptor (AT1), and proteinase-activated receptor 1 (PAR1) were used. All cell lines were developed by standard procedures. Sal A was tested at a final concentration of 10 µM when cells were challenged with their selective agonists for antagonist identification. The selective antagonists of specific cell lines were used as positive control (Table S1). The HEK293 cell line was set up as the naïve group, and cells were challenged with 10 µM ATP as an agonist. The results showed that Sal A only interacted with the ETA receptor in the screening panel, suggesting that Sal A is a selective ETAR antagonist (Figure 3C). 2.4. Sal A Inhibited Endogenously Expressed ETAR in HeLa Cells Since endogenous ETAR and ETBR were previously detected in HeLa cell lines [23], we tested whether Sal A inhibited endogenous ETAR and ETBR. Stimulation of HeLa cells with ET-1 led to a concentration-dependent, time-resolved impedance response as measured by the xCelligence system [24]. In our study, after 24 h incubation with complete medium, the medium was changed to hanks balanced salt solution (HBSS) with Ca2+ and Mg2+. Then, various concentrations of Sal A, 10 µM Bosentan (positive control), and 0.1% DMSO (negative control) were performed in parallel. A rapid impedance response was observed as cell index (CI) decreased and recovered within 15 min after initial treatment. Subsequently, 10 nM ET-1 was added. Cell index was sharply elevated by ET-1, and the response was completely blocked by Bosentan. Sal A also declined ET-1-induced changes of CI and showed a dose-dependent trend (Figure 4A). The concentration–response curve was calculated based on (MaxCI − MinCI) during 29.7 to 30.4 h, and the IC50 value of Sal A was 3.14 µM (Figure 4B). Furthermore, the effects of Bosentan and Sal A in HeLa cells were investigated by Ca2+ Influx assay. ET-1 dose-dependently induced calcium mobilization from 500 nM to 6.4 pM with an EC50 value of ET-1 of 39 nM (Figure 4C). This response was fully blocked by Bosentan and the IC50 was 4.3 µM (Figure 4D). In contrast, Sal A exhibited partial antagonism of endogenous ETAR, with IC50 of 5.19 µM (Figure 4E). These responses were further confirmed by analysis of respond-span, in which Bosentan completely and Sal A partially suppressed (p < 0.01) ET-1 activation (Figure 4F). These results suggested that Sal A could block both recombinant and endogenous ETAR. 2.5. Sal A Suppressed Proliferations in Multiple Cancer Cell Lines As indicated in Table 1, anti-proliferative effects were tested in five cell lines—including human ovarian carcinoma cells (SKOV3), human cervical cancer cells (HeLa), human non-small cell lung cancers with T790 mutation (NCI-H1975), human prostate carcinoma cells (DU145), human lung adenocarcinoma cells (A549)—with Sal A treatment at various concentrations of 100, 25, 6.25, and 1.56 µM. After 72 h treatment, cell viability was determined by CellTiter-Glo kit. With the increase of Sal A concentration, cell growth was suppressed in all cell lines. However, different IC50 values were observed in each cell line. 2.6. Sal A Extended Cell Lins Proliferation Cycle by Real-Time Cell Analysis (RTCA) To characterize the mechanism of Sal A, we assessed the effect of Sal A and Bosentan on DU145, A549, H1975, HeLa cell lines by using the xCELLigence system. After cell seeding onto 96 well E-plate for 16 h, cells were cultured in reduced serum medium (2% FBS) in the presence of 20, 5, 1.2 µM Sal A, and 10 µM Bosentan, and 0.1% DMSO as negative control. Relative impedance signal level (represented as “cell index” in manufacturer’s software) was monitored continuously for 130 h. The xCELLigence system were set to record the cell index every 70 min. All cell lines reach a stationary phase of growth in 80 to 130 h (Figure 5). With the treatment of 10 µM Bosentan, the time of reaching stationary phase was extended in four cell lines. Similarly, 5 and 1.2 µM Sal A extended the proliferation time. Meanwhile, cell indexes of four cell lines induced by 20 µM Sal A were detected to decrease, in which the concentration was lower than that we tested in the CellTiter-Glo assay. To compare the inhibitory effect in quantification, we calculated the doubling time (DT) index by using Formula 1. Based on the exponent proliferation curve, we estimated the doubling time by using cell index from 16 h to the time point of maximum cell index. Time indicates the time from starting to the time of maximum cell index. The parameter estimation was by Prism 6 (GraphPad Software, La Jolla, CA, USA), initial values were 1. The results were output, including the value of doubling time (DT) and 95% confidence interval of DT. (1) Cell Index=A×2Time/DT The results are shown in Table 2. Ten µM Bosentan extended the doubling time of four cell lines significantly. The extension of doubling time was observed with the treatment of 5 and 1.2 µM Sal A. The doubling times of HeLa, DU145, and H1975 were significantly more than negative control, but A549 cell lines showed no influence of Sal A on doubling time. Interestingly, Sal A induced a cell index decrease after 60 h in A549 cell lines. The influence of Sal A in A549 cell lines should be further investigated. 2.7. Sal A Inhibited the Proliferation Induced by Exogenous ET-1 in DU145 Cells The ability of Sal A—an ETA receptor antagonist—to inhibit ET-1 growth stimulation was tested in DU145 cells. Growth effects were determined independently in both cell lines in the presence of ET-1 alone, ET-1, ETA and ETB receptor antagonists (Sal A, Bosentan), and all antagonists alone (48 h). As shown in Figure 6, exogenous 10 nM ET-1 induced an increase in proliferation of approximately 30%. Meanwhile, ET-1-stimulated growth was significantly blocked by Bosentan (ETAR and ETBR antagonist) and Sal A. Incubation with the compounds alone did not significantly affect cell growth. 2.8. Effect of Sal A on Cell Cytotoxicity in HUVEC We investigated the effect of Sal A on cell cytotoxicity in HUVEC cells. As shown in Figure 7, released lactate dehydrogenase (LDH) cytotoxicity was evaluated in this experiment. HUVEC cells were treated with 10 and 2 µM of Sal A for 24 and 48 h. LDH cytotoxicity was measured using the Pierce LDH Cytotoxicity Assay Kit (Thermo Scientific, Hudson, NH, USA). The results indicated that there are no significant differences in extracellular LDH between test and control groups in both 24 and 48 h. 2.9. Effect of Sal A on Cardio-Toxicity in hiPS-CMs The xCelligence RTCA cardio system has been a useful tool used to monitor the cardio-toxicity effect on hiPS-CMs in the short and long term [25,26]. In this study, the effect of Sal A on available commercial hiPS-CMs was evaluated. Cells were treated with compounds post cell seeding for 24 h. As shown in Figure 8A, cell index remained stable after treatment with Sal A. Meanwhile, Adriamycin (5 µM) as the positive control decreased the cell index, indicating that Sal A induced no cell death. At the same time, we monitored the effect of Sal A on hiPS-CMs contractility activities in real-time. Before compound treatment, hiPS-CMs contractility was detected every minute for at least 10 min to confirm the stability of the hiPS-CMs. In Figure 8B–D, 10 µM Sal A had no inhibitory effect on hiPS-CMs beating rates and amplitude in the long term. Simultaneously, amiodarone in 1 µM was treated as positive control for inducing cardio-toxicity in hiPS-CMs. As shown in Figure 8B, the amiodarone (positive control) decreased beating rate within 10 h, and the recovery of hiPS-CMs contractility activities were observed after 10 h. The decreasing contractility activities indicated cardio-toxicity [27]. Compared with amiodarone, the contractility activities were not influenced by 10 µM Sal A. Above all, these results suggested that 10 µM Sal A was free of cardio-toxicity and cell toxicity on hiPS-CMs with 24 h. 3. Discussion Salvianolic acid derivatives are the major water-soluble components of Dashen, which is the dried roots and rhizomes of Salvia miltiorrhiza Bunge, a type of well-known traditional Chinese herbal medicine. Among various Salvianolic acids, Salvianolic acid A has drawn considerable research attention for its diverse potent bioactivities, including the suppressive effect of transgelin 2 and the inhibitory effect of matrix metalloproteinase-9 [20,28]. In our study, we demonstrate for the first time that Sal A is a selective ETAR antagonist in both exogenous and endogenous cell lines and reveal the inhibitory effect of Sal A on the proliferation of multiple tumor cell lines. Moreover, we traced the non-cytotoxicity and non-cardio-toxicity of Sal A in HUVECs and hiPS-CMs. In the ETA receptor-overexpressed HEK293 cell line, we observed the dose-dependent inhibitory effect of Sal A. The results of the selective assay suggested that Sal A possessed a selective inhibitory effect on ETA receptors. The similar inhibitory effect was observed after treatment of serial concentration of Sal A in HeLa cell line. By using RTCA, we determined that Sal A inhibited the Cell Index increase induced by ET-1 partially, and the CI increase was completely suppressed by Bosentan. In the Ca2+ influx assay (Figure 4C–F), the dose–response span of Sal A was less than Bosentan. The results demonstrated that Sal A inhibits ETAR activation and suggested Sal A as a functional antagonist of ETAR. The potential binding mechanism of Sal A to the endothelin receptor will be evaluated in further research. The ET-1–ETAR axis has been widely involved in the cancer process. ETAR antagonists, such as Zibotentan and Atrasentan, were investigated in a series of cancer therapy clinical trials [29,30]. Thus, several cancer cell lines, including HeLa, SKOV3, DU145, H1975, and A549—all of which expressed ETAR [31,32,33,34]—were applied to investigations of the anti-proliferative effects of Sal A. By CellTiter-Glo kit and RTCA assays, Sal A showed anti-proliferative effects in HeLa, DU145, H1975, and A549 cell lines with IC50 values under 20 µM. During the continuous monitor proliferation assay, both Sal A and Bosentan delayed the increase of cell index. However, IC50 values between ETAR antagonist and anti-tumor assays showed a remarkable difference (approximately 3 µM for antagonism assay and up to 15 µM for the anti-proliferation assay). A possible explanation for this variation may lie in the differences in detection approaches and cell lines. As shown in Figure 5D, both Sal A and Bosentan induced a cell index decrease after 60 h in A549 cell lines. In the study of Bi, L. [35], Sal A induced A549 cell lines apoptosis by up-regulating the phosphatase and tensin homolog (PTEN) protein level and down-regulating Akt phosphorylation. The doubling time index was used in our study for the comparison of the effect of Sal A in four cell lines. In the study of Masarik, M. [36], the slope of the cell index increase curve was used to describe the proliferation phenomena in the RTCA assay. Based on a similar principle, we employed the simple doubling time formula to estimate the doubling time. In brief, we estimated the parameter of Formula 1 by using the cell index from the time of adding compounds to the time point of maximum cell index. Furthermore, we demonstrated that Sal A suppresses the proliferation induced by exogenous ET-1 in DU145 cells (shown in Figure 6). Ping Sun [32] reported that exogenous ET-1 reduces the apoptosis induced by paclitaxel in DU145 cells. Generally, the majority of cancer cell types—for example, prostate, ovarian, and lung cancers—show a reduction of ET-1-stimulated growth in response to ETAR antagonism [12,37,38]. These findings suggested that Sal A may play a role in cancer therapy through inhibition of ETAR activation. Advances in cancer therapy have resulted in significant improvement in long-term survival for many types of cancer, but have also resulted in untoward side effects associated with treatment. One such complication that has become increasingly recognized is the development of cardiomyopathy and heart failure [39,40]. Although Sal A is reported to attenuate H9C2 cell apoptosis [41], the potential toxic risk of Sal A on HUVEC and hiPS-CMs at bioactive doses still remains to be explored. LDH release assay showed that Sal A did not show cytotoxicity to HUVECs at a concentration of 10 µM during a 48 h period. Compared to cytotoxicity assays on H9C2 [42], the RTCA cardio system provides a sensitive tool to predict potential cardiac side effects. Additionally, the sensitive label-free assay made it possible to detect the regular beating pattern of cardiomyocytes under normal physiological conditions [42,43,44]. As mentioned [26], the loss of cell index reflected the loss of mitochondria and cell viability induced by Adriamycin. Amiodarone lengthens the cardiac action potential to induce acute cardiac myocyte dysfunction [45]. Cytotoxicity of hiPS-CMs was clearly shown to be induced by 5 µM Adriamycin with a cell index decrease, but not by 10 µM Sal A. Treatment with 10 µM Sal A did not induce modification of amplitude nor beat rate compared with vehicle control. These observations suggest that Sal A is above the cardio-toxic threshold and provide a cardio-friendly anti-cancer leading compound. 4. Materials and Methods 4.1. Reagents and Materials Dulbecco’s modified eagles medium (DMEM) and fetal bovine serum (FBS) for cell culture were purchased from Gibco BRL (Grand Island, NY, USA). Salvianolic acid A, Salvianolic acid B, Salvianolic acid C, Salvianolic acid D, and Salvianolic acid were purchased from National Institutes for Food and Drug Control (Beijing, China) with purities greater than 98%. Hygromycin B, ET-1, Angiotensin II, Adenosine, TRAP-6, probenecid, and acid red 1 (purity ≥ 98%) were purchased from Sigma-Aldrich Chemicals (St. Louis, MO, USA). Bosentan hydrate was purchased from Yuanye (Shanghai, China). Fluo-4 AM was purchased from Molecular Probes (Grand Island, NY, USA). Matrigel was purchased from Becton Dickinson (New York, NY, USA). All the chemicals were dissolved in DMSO if not otherwise stated. 4.2. Compund Preparation Salvianolates were dissolved in DMSO at 36 mM. For primary screening, compounds were diluted to 12 mM in DMSO. ET-1 and BQ-123 were stored at a concentration of 50 µM and 1 mM, respectively. Initial serial dilutions were made in DMSO with compound concentrations at 400× final for concentration-response determinations. For the compound plates, 2 µL of the 400× DMSO solution was added to 160 µL HBSS (5× solution, DMSO concentration 1.25%). The final DMSO concentration in each well was controlled as 0.25% for all the tested compounds. 4.3. Cell Culture HEK293/ETAR cells, provided by the Beijing Institute of Genomics in the Chinese Academy of Science (Beijing, China), were well-characterized cell lines expressing the ETA receptor. All HEK293 cell lines used in this study were routinely maintained in DMEM containing 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified atmosphere of 5% CO2 at 37 °C. HEK293/ETAR cells were incubated with complete culture medium along with 50 µg/mL hygromycin B. HeLa, A549, SKOV3, DU145, and H1975 cell lines were maintained in DMEM containing 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin. The hiPS-CMs cells were obtained from CELLAPYBIO (Cat# CA2001106, Beijing, China), which are a well-validated cell line [46], and CMs were thawed from cryopreserved vials into CMs plating medium following recommended procedures. The cells were incubated in DMEM with 10% FBS at 37 °C, 5% CO2. Cell culture media were refreshed every two days. HUVEC cells, a gift from Zengchangqing Lab, were maintained in EBM-2 medium with SingleQuot growth kit (Lonza, Allendale, NJ, USA). 4.4. Antagonist Screening Assay 4.4.1. Ca2+ Influx Assay HEK293/ETAR cells were seeded at a density of 3 × 104 per well into 96-well clear-bottom black plates coated with matrigel and incubated in 5% CO2 at 37 °C overnight. On the day of assay, the growth medium was replaced by 80 µL loading buffer containing a final concentration of 4 µM Ca2+-sensitive dye Fluo-4 AM and 2 mM acid red 1 in HBSS. The plate was then incubated at 37 °C in the dark for 30 min before calcium signal read out. For antagonist study, 80 µL loading buffer was added into each well and 20 µL HBSS containing tested compound was added 10 min prior to calcium-flux measurement. Cells were transferred to a Flexstation II (Molecular Devices, Sunnyvale, CA, USA) for experimentation. Basal fluorescence was recorded for 16 s before agonist application. The integrated Flexstation II fluidics system added 25 µL compound (5× solution) from the agonist compound plate to the assay plate containing 100 µL loading buffer solution. Relative fluorescence units (RFU) were read by FlexStation II at 37 °C with an excitation wavelength of 485 nm and an emission wavelength of 525 nm. The fluorescence intensity was read every 1.52 s for 80 s. RFU indicated the peak of calcium response. The inhibition (%) were calculated as follows, (2) Inhibition (%)=(1−RFUcompound−RFUnegetiveRFUpositive−RFUnegetive)×100% 4.4.2. Real-Time Cell Analysis Assay HeLa cell lines were seeded at a density of 1 × 104 per well into a 96-well E-plate and the impedance was monitored by xCELLgence system in 5% CO2 at 37 °C overnight. On the day of the assay, the growth medium was replaced by 80 µL HBSS. After 30 min, the 20 µL compound (5× solution) were added into the E-plate. Then, the 25 µL ET-1 (5× solution) were added into the E-plate when the cell indexes were steady. The cell index was read every 20 s for 1 h. 4.5. Compound Cytotoxicity Assay 4.5.1. Cell Viability Assay HEK293/ETAR cells were seeded at 3 × 104 per well into 96-well clear-bottom black plates and incubated in 5% CO2 at 37 °C overnight. Different concentrations of the compound were added into the 96-well plates and incubated for 2 h. Luminescence was read by Envision 2100 multilabel reader to detect cells’ viability following incubation with CellTiter-Glo reagent for 10 min. 4.5.2. Lactate Dehydrogenase Leakage Assay HUVEC cells (1.5 × 104 cells/well) were seeded in a 96-well plate. The cultured cells were treated with various concentrations of compounds or vehicle and were incubated for another 24 and 48 h. The culture medium was aspirated and centrifuged at 1000× g for 10 min to obtain a cell-free supernatant. LDH activity was examined using a commercially available kit (Thermo Fisher Scientific, Pittsburgh, PA, USA) following the manufacturer’s instructions. These measurements were performed with VersaMax (Molecular Devices, Sunnyvale, CA, USA). The results are given as fractions of LDH release compared to the positive controls, which consisted of Lysis Buffer. The results presented represent mean values from triplicate measurements. 4.5.3. Cardio-Toxicity Assay The hiPS-CM were seeded at 1.7 × 104 in 96-well E-plate. Procedures have been described in [25]. In brief, compound treatment was initiated 48–72 h after cell seeding, and the E-plate was monitoring every 15 min on the RTCA Cardio Instrument incubator. The concentration of DMSO were control in 0.1%. 4.6. Cell Proliferation Assay 4.6.1. Cell Viability Assay Cell lines were seeded at 5 × 103 per well into 96-well plates and incubated in 5% CO2 at 37 °C overnight. On the day of assay, the cell culture medium was replaced with 108 µL complete culture medium within 10% FBS. The 12 µL of different concentrations of Salvianolic A (10× solution) were added into the 96-well plates and incubated for 72 h. The concentration of DMSO was controlled at 0.1%. Luminescence was read by Envision 2100 multilabel reader to detect cells’ viability following incubation with CellTiter-Glo reagent for 10 min. 4.6.2. Real-Time Cell Analysis Assay Cell lines were seeded at 5 × 103 per well into 96-well E-plates and incubated in 5% CO2 at 37 °C overnight. Then, the cell culture medium was replaced with 108 µL reduced serum medium within 2% FBS. The 12 µL of different concentrations of compounds (10× solution) were added into the 96-well E-plates and incubated for 138 h. The cell index was read every 70 min. 4.7. Data Analysis and Statistics Data were analyzed using xCELLigence Cardio Software (Roche, Basel, Switzerland) and further analyzed with GraphPad Prism 6. Data were presented as mean ± SD. Statistical significance of differences was estimated by one-way ANOVA. p < 0.05 (marked with an asterisk) was considered significant. 5. Conclusions In conclusion, we found that Sal A is a novel ETA receptor antagonist, and further observed anti-tumor effect with or without exogenous endothelin-1. Our data strongly suggest that Sal A is a potential candidate for the development of a novel anti-cancer drug. Acknowledgments This work supported by a grant from the National Natural Science Foundation of China (No. 81430094). We thank Lan Feng at Beijing lab for Cardiovascular Precision Medicine, Capital Medical University for technical expertise on hiPSC and ACEA Co., Ltd. (Hangzhou, China) for technical assistance of RTCA cardio system. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1244/s1. Click here for additional data file. Author Contributions Shiyou Li and Yanjiang Qiao conceived designed the project. Qiao Zhang designed and performed most experiments and wrote the paper. Yangyang Yu performed the in vitro experiments obtaining CMs. Yanling Zhang and Yuxin Zhang helped to prepare the compounds. Wei Yang provided xCELLigence Cardio equipment and technical support. Shifeng Wang and Shengnan Sun read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 HEK293/endothelin A receptor (ETAR) cell line validation. (A) Agonist concentration–response curve of endothelin-1 (ET-1). The EC50 value of ET-1 was 4.78 nM; (B) Antagonist concentration–response curve of BQ-123 (evoked by 20 nM ET-1). The IC50 value of ETAR antagonist BQ-123 was 0.1 nM. Data were expressed as the fold change of max relative fluorescence units (RFU) and min RFU. Data were from three independent experiments. Figure 2 The primary evaluation of five compounds in HEK293/ETAR cell line. (A) Chemical formulas of Salvianolic acid A, Salvianolic acid B, Salvianolic acid C, Salvianolic acid D, Salvianolic acid; (B) Inhibitory responses of the five tested compounds in HEK293/ETAR cell line. The cells were pre-incubated with tested compounds for 10 min and then stimulated by 20 nM ET-1. Data (mean ± SD) were expressed as inhibition (%) from two independent experiments. Figure 3 ETAR antagonist verification of Sal A. (A) Concentration-response curve in HEK293/ETAR cell line. The IC50 value of Sal A was 5.7 µM (evoked by 20 nM ET-1); (B) Cytotoxicity evaluation of Sal A in HEK293/ETAR cells. The cell viability was determined after 30 min compound treatment by using ATP CellTiter-Glo kit; (C) Selectivity assay of Sal A in five recombinant HEK293 cell lines. Purple dot-line: 50% inhibition. Data were expressed as means ± SD of three independent experiments. Statistical analysis by using one-way ANOVA with Tukey test. ns: no significance, *** p < 0.01. A1: adenosine A1 receptor; AT1: angiotensin II type 1 receptor; ETA: endothelin A; ETB: endothelin B; PAR1: proteinase-activated receptor 1. Figure 4 Effect of Sal A on (A,B) ET-1 stimulated cell index and (C–F) calcium influx in HeLa cells. (A) Time-dependent cell index (CI) curves are shown, dose concentration of Sal A and 10 µM Bosentan (positive control) were added at 29.4 h. After 15 min, 10 nM ET-1 was added. Black arrow: compound addition; (B) Concentration–response curves of Sal A calculated by (MaxCI − MinCI) during 29.7 to 30.4 h. The IC50 value of Sal A was 3.14 µM; (C) Concentration–response curve of ET-1 agonist in HeLa cells. The EC50 value was 39 nM; (D) Concentration–response curves of Bosentan in HeLa cells. The IC50 value was 4.3 nM; (E) Concentration–response curves of Sal A in HeLa cells. The IC50 value was 5.19 µM; (F) The comparison with response-span of three compounds. Data (C–E) were expressed as the fold change of max RFU and min RFU. Data are expressed as means ± SD of three independent experiments. Statistical analysis was conducted by using one-way ANOVA with Tukey test. ** p < 0.05. Figure 5 Real-time monitoring of cell proliferation using the xCELLigence system. (A) DU145; (B) HeLa; (C) H1975; and (D) A549 cell lines were seeded with 6000 cells per well onto E-plate. After 16 h, cells were cultured in reduced serum medium (2% FBS) in the presence of 20, 5, 1.2 µM Sal A, 10 µM Bosentan, and 0.1% DMSO as negative control. CI was monitored continuously for 130 h. The xCELLigence system was set to record the cell index every 70 min. Data were expressed as mean and SD in two independent experiments. Figure 6 Growth response of prostate cancer cell lines to ET-1 (10 nM) and Sal A (10 µM, ETA receptor antagonist) and Bosentan (10 µM, ETA, and ETB receptor antagonist). After 48 h incubation with compounds, cell viability (%) was compared with controls (set at 100%). Bosentan and Sal A significantly reduced growth in the presence of exogenous ET-1 (vs. ET-1 group). Without the presence of exogenous ET-1, both Bosentan and Sal A showed no inhibitory effect in DU145 (vs. Control). Data were expressed as mean and SD of three independent experiments. Statistical analysis of original RLU measurement was conducted using one-way ANOVA with Tukey test. *** p < 0.001. Figure 7 Determination of lactate dehydrogenase (LDH) cytotoxicity of Sal A in HUVEC cells. HUVEC cells were plated in a 96-well plate in maintaining medium. Different concentrations of Sal A were added to the culture media and incubated for 24 and 48 h at 37 °C, 5% CO2. Meanwhile, 1X Lysis Buffer was used as positive control. LDH cytotoxicity was measured using the Pierce LDH Cytotoxicity Assay Kit (Thermo Scientific). Data were expressed as mean ± SD, n = 3. Statistical analysis of original optical density value measurement was conducted using a one-way ANOVA and a with Tukey test. *** p < 0.001. ns: no significant difference. Figure 8 Typical contraction profiles of human-induced pluripotent stem cell-derived cardiomyocytes (hiPS-CMs) to compounds treatment. (A) Cell index fluctuation after Adriamycin (5 µM, cytotoxicity positive control), Amiodarone (1 µM, cardio-toxicity control), and Sal A (10 µM); (B) Temporal profiling of hiPS-CMs beating pattern. Beating pattern data were expressed in 0, 1, 6, 12, 18, 24 h; (C) Temporal profiling of hiPS-CMs beating rate. Data were normalized by the point before compounds were added; (D) Temporal profiling of hiPS-CMs amplitude. Data were normalized to the signal acquired prior to compound treatment. Data were analyzed using xCELLigence Cardio Software (ACEA, Hangzhou, China) and expressed as mean ± SD of three independent experiments. ijms-17-01244-t001_Table 1Table 1 Anti-proliferative effects and IC50 values of salvianolic acid A (Sal A). Cell Lines Inhibition (%) IC50 Values (µM) 100 µM 25 µM 6.25 µM 1.56 µM SKOV3 99.13(0.19) 30.87(3.18) 1.215(1.48) 6.001(0.21) 30.84 HeLa 96.30(0.49) 75.65(3.81) 8.751(1.25) 3.419(0.15) 15.85 H1975 96.38(0.19) 73.65(1.84) 35.09(5.75) 11.64(1.48) 10.19 DU145 96.48(0.04) 77.25(2.18) 32.94(2.26) 18.15(4.08) 9.512 A549 99.13(0.11) 58.43(2.39) 43.40(1.55) 38.35(2.56) 6.461 The anti-proliferative effect of Sal A on SKOV3, HeLa, H1975, DU145, and A549 cell lines. Cells were treated with Sal A at doses of 100, 25, 6.25, and 1.56 µM. After 72 h, cell viability was determined by CellTiter-Glo kit (Promega Co., Madison, WI, USA), and the inhibitions (%) were calculated (vs. Naïve group). The concentration of DMSO was controlled at 0.1%. Data are expressed as mean (SD) from three independent experiments. ijms-17-01244-t002_Table 2Table 2 The effect of Sal A on doubling time in different cell lines (h). Cell Lines Sal A Sal A Bosentan 0.1% DMSO 5 µM 1.2 µM 10 µM HeLa 33.5 ± 1.04 ** 23.7 ± 0.48 ** 26.3 ± 0.62 ** 21.1 ± 0.34 DU145 41.0 ± 1.35 ** 35.8 ± 0.37 ** 29.6 ± 0.26 ** 25.6 ± 0.19 H1975 94.1 ± 2.43 ** 49.6 ± 1.61 ** 29.4 ± 0.38 * 27.1 ± 0.63 A549 28.7 ± 1.10 28.9 ± 1.08 40.9 ± 2.33 ** 29.7 ± 0.99 The effect of Sal A on doubling time in HeLa, H1975, DU145, A549 cells line. Cells were treated with Sal A at a series of concentrations. The doubling time was calculated during 16 to 80 h. The concentration of DMSO was controlled at 0.1%. Data are expressed as mean ± 95% confidence interval. Statistical analysis was conducted using one-way ANOVA with Tukey test. * p < 0.1; ** p < 0.05. ==== Refs References 1. Grant K. Loizidou M. Taylor I. Endothelin-1: A multifunctional molecule in cancer Br. J. Cancer 2003 88 163 166 10.1038/sj.bjc.6700750 12610497 2. Rosano L. Spinella F. Bagnato A. Endothelin 1 in cancer: Biological implications and therapeutic opportunities Nat. Rev. Cancer 2013 13 637 651 10.1038/nrc3546 23884378 3. Shichiri M. Hirata Y. Nakajima T. Ando K. Imai T. Yanagisawa M. Masaki T. Marumo F. Endothelin-1 is an autocrine/paracrine growth factor for human cancer cell lines J. Clin. Investig. 1991 87 1867 1871 10.1172/JCI115210 2022753 4. Nelson J. Bagnato A. Battistini B. Nisen P. The endothelin axis: emerging role in cancer Nat. Rev. Cancer 2003 3 110 116 10.1038/nrc990 12563310 5. Jin J.K. Dayyani F. Gallick G.E. Steps in prostate cancer progression that lead to bone metastasis Int. J. Cancer 2011 128 2545 2561 10.1002/ijc.26024 21365645 6. Nie S. Zhou J. Bai F. Jiang B. Chen J. Zhou J. Role of endothelin A receptor in colon cancer metastasis: In vitro and in vivo evidence Mol. Carcinog. 2014 53 E85 E91 10.1002/mc.22036 23818293 7. Bagnato A. Cirilli A. Salani D. Simeone P. Muller A. Nicotra M.R. Natali P.G. Venuti A. Growth inhibition of cervix carcinoma cells in vivo by endothelin A receptor blockade Cancer Res. 2002 62 6381 6384 12438219 8. Banerjee S. Hussain M. Wang Z. Saliganan A. Che M. Bonfil D. Cher M. Sarkar F.H. In vitro and in vivo molecular evidence for better therapeutic efficacy of ABT-627 and taxotere combination in prostate cancer Cancer Res. 2007 67 3818 3826 10.1158/0008-5472.CAN-06-3879 17440096 9. Haque S.U. Dashwood M.R. Heetun M. Shiwen X. Farooqui N. Ramesh B. Welch H. Savage F.J. Ogunbiyi O. Abraham D.J. Efficacy of the specific endothelin a receptor antagonist zibotentan (ZD4054) in colorectal cancer: A preclinical study Mol. Cancer Ther. 2013 12 1556 1567 10.1158/1535-7163.MCT-12-0975 23723122 10. Daub H. Weiss F.U. Wallasch C. Ullrich A. Role of transactivation of the EGF receptor in signalling by G-protein-coupled receptors Nature 1996 379 557 560 10.1038/379557a0 8596637 11. Grant K. Knowles J. Dawas K. Burnstock G. Taylor I. Loizidou M. Mechanisms of endothelin 1-stimulated proliferation in colorectal cancer cell lines Br. J. Surg. 2007 94 106 112 10.1002/bjs.5536 17078114 12. Akhavan A. McHugh K.H. Guruli G. Bies R.R. Zamboni W.C. Strychor S.A. Nelson J.B. Pflug B.R. Endothelin receptor A blockade enhances taxane effects in prostate cancer Neoplasia 2006 8 725 732 10.1593/neo.06388 16984730 13. Rosano L. Di Castro V. Spinella F. Tortora G. Nicotra M.R. Natali P.G. Bagnato A. Combined targeting of endothelin A receptor and epidermal growth factor receptor in ovarian cancer shows enhanced antitumor activity Cancer Res. 2007 67 6351 6359 10.1158/0008-5472.CAN-07-0883 17616694 14. Liu T. Liu X. Li W. Tetrandrine, a Chinese plant-derived alkaloid, is a potential candidate for cancer chemotherapy Oncotarget 2016 10.18632/oncotarget.8315 27027348 15. Apaya M.K. Chang M.T. Shyur L.F. Phytomedicine polypharmacology: Cancer therapy through modulating the tumor microenvironment and oxylipin dynamics Pharmacol. Ther. 2016 162 58 68 10.1016/j.pharmthera.2016.03.001 26969215 16. Moselhy J. Srinivasan S. Ankem M.K. Damodaran C. Natural products that target cancer stem cells Anticancer Res. 2015 35 5773 5788 26503998 17. Fan H.Y. Fu F.H. Yang M.Y. Xu H. Zhang A.H. Liu K. Antiplatelet and antithrombotic activities of salvianolic acid A Thromb. Res. 2010 126 e17 e22 10.1016/j.thromres.2010.04.006 20451955 18. Liu G.T. Zhang T.M. Wang B.E. Wang Y.W. Protective action of seven natural phenolic compounds against peroxidative damage to biomembranes Biochem. Pharmacol. 1992 43 147 152 10.1016/0006-2952(92)90271-J 1739402 19. Oh K.S. Oh B.K. Mun J. Seo H.W. Lee B.H. Salvianolic acid A suppress lipopolysaccharide-induced NF-κB signaling pathway by targeting IKKβ Int. Immunopharmacol. 2011 11 1901 1906 10.1016/j.intimp.2011.07.022 21839184 20. Zheng X. Chen S. Yang Q. Cai J. Zhang W. You H. Xing J. Dong Y. Salvianolic acid A reverses the paclitaxel resistance and inhibits the migration and invasion abilities of human breast cancer cells by inactivating transgelin 2 Cancer Biol. Ther. 2015 16 1407 1414 10.1080/15384047.2015.1070990 26176734 21. Cai J. Chen S. Zhang W. Zheng X. Hu S. Pang C. Lu J. Xing J. Dong Y. Salvianolic acid A reverses paclitaxel resistance in human breast cancer MCF-7 cells via targeting the expression of transgelin 2 and attenuating PI3 K/Akt pathway Phytomed. Int. J. Phytother. Phytopharm. 2014 21 1725 1732 10.1016/j.phymed.2014.08.007 25442283 22. Ihara M. Ishikawa K. Fukuroda T. Saeki T. Funabashi K. Fukami T. Suda H. Yano M. In vitro biological profile of a highly potent novel endothelin (ET) antagonist BQ-123 selective for the ETA receptor J. Cardiovasc. Pharmacol. 1992 20 S11 S14 10.1097/00005344-199204002-00005 1282942 23. Drimal J. Drimal J. Jr. Drimal D. Enhanced endothelin ET(B) receptor down-regulation in human tumor cells Eur. J. Pharmacol. 2000 396 19 22 10.1016/S0014-2999(00)00198-9 10822048 24. Ke N. Nguyen K. Irelan J. Abassi Y.A. Multidimensional GPCR profiling and screening using impedance-based label-free and real-time assay Methods Mol. Biol. 2015 1272 215 226 25563187 25. Yu Y. Sun S. Wang S. Zhang Q. Li M. Lan F. Li S. Liu C. Liensinine- and neferine-induced cardiotoxicity in primary neonatal rat cardiomyocytes and human-induced pluripotent stem cell-derived cardiomyocytes Int. J. Mol. Sci. 2016 17 186 10.3390/ijms17020186 26840304 26. Clements M. Millar V. Williams A.S. Kalinka S. Bridging functional and structural cardiotoxicity assays using human embryonic stem cell-derived cardiomyocytes for a more comprehensive risk assessment Toxicol. Sci. 2015 148 241 260 10.1093/toxsci/kfv180 26259608 27. Nguemo F. Saric T. Pfannkuche K. Watzele M. Reppel M. Hescheler J. In vitro model for assessing arrhythmogenic properties of drugs based on high-resolution impedance measurements Cell. Physiol. Biochem. Int. J. Exp. Cell. Physiol. Biochem. Pharmacol. 2012 29 819 832 10.1159/000188069 22613982 28. Zhang T. Xu J. Li D. Chen J. Shen X. Xu F. Teng F. Deng Y. Ma H. Zhang L. Salvianolic acid A, a matrix metalloproteinase-9 inhibitor of Salvia miltiorrhiza, attenuates aortic aneurysm formation in apolipoprotein E-deficient mice Phytomed. Int. J. Phytother. Phytopharm. 2014 21 1137 1145 10.1016/j.phymed.2014.05.003 24916705 29. Chouaid C. Nathan F. Pemberton K. Morris T. A phase II, randomized, multicenter study to assess the efficacy, safety, and tolerability of zibotentan (ZD4054) in combination with pemetrexed in patients with advanced non-small cell lung cancer Cancer Chemother. Pharmacol. 2011 67 1203 1208 10.1007/s00280-010-1538-z 21153822 30. Nelson J.B. Love W. Chin J.L. Saad F. Schulman C.C. Sleep D.J. Qian J. Steinberg J. Carducci M. Atrasentan Phase 3 Study Guides Phase 3, randomized, controlled trial of atrasentan in patients with nonmetastatic, hormone-refractory prostate cancer Cancer 2008 113 2478 2487 10.1002/cncr.23864 18785254 31. Chauhan V. Breznan D. Thomson E. Karthikeyan S. Vincent R. Effects of ambient air particles on the endothelin system in human pulmonary epithelial cells (A549) Cell Biol. Toxicol. 2005 21 191 205 10.1007/s10565-005-0162-x 16323056 32. Sun P. Xiong H. Kim T.H. Ren B. Zhang Z. Positive inter-regulation between β-catenin/T cell factor-4 signaling and endothelin-1 signaling potentiates proliferation and survival of prostate cancer cells Mol. Pharmacol. 2006 69 520 531 10.1124/mol.105.019620 16291872 33. Von Brandenstein M.G. Ngum Abety A. Depping R. Roth T. Koehler M. Dienes H.P. Fries J.W. A p38-p65 transcription complex induced by endothelin-1 mediates signal transduction in cancer cells Biochim. Biophys. Acta 2008 1783 1613 1622 10.1016/j.bbamcr.2008.04.003 18457675 34. Tocci P. Caprara V. Cianfrocca R. Sestito R. Di Castro V. Bagnato A. Rosano L. Endothelin-1/endothelin A receptor axis activates RhoA GTPase in epithelial ovarian cancer Life Sci. 2016 10.1016/j.lfs.2016.01.008 26776837 35. Bi L. Chen J. Yuan X. Jiang Z. Chen W. Salvianolic acid A positively regulates PTEN protein level and inhibits growth of A549 lung cancer cells Biomed. Rep. 2013 1 213 217 24648921 36. Masarik M. Gumulec J. Hlavna M. Sztalmachova M. Babula P. Raudenska M. Pavkova-Goldbergova M. Cernei N. Sochor J. Zitka O. Monitoring of the prostate tumour cells redox state and real-time proliferation by novel biophysical techniques and fluorescent staining Integr. Biol. Quant. Biosci. Nano Macro 2012 4 672 684 10.1039/c2ib00157h 22592803 37. Kuzumaki N. Suzuki A. Narita M. Hosoya T. Nagasawa A. Imai S. Yamamizu K. Morita H. Suzuki T. Okada Y. Multiple analyses of G-protein coupled receptor (GPCR) expression in the development of gefitinib-resistance in transforming non-small-cell lung cancer PLoS ONE 2012 7 1244 10.1371/journal.pone.0044368 23144692 38. Rosano L. Cianfrocca R. Spinella F. di Castro V. Natali P.G. Bagnato A. Combination therapy of zibotentan with cisplatinum and paclitaxel is an effective regimen for epithelial ovarian cancer Can. J. Physiol. Pharmacol. 2010 88 676 681 10.1139/Y10-053 20628434 39. Bloom M.W. Hamo C.E. Cardinale D. Ky B. Nohria A. Baer L. Skopicki H. Lenihan D.J. Gheorghiade M. Lyon A.R. Cancer therapy-related cardiac dysfunction and heart failure: Part 1: Definitions, pathophysiology, risk factors, and imaging Circ. Heart Fail. 2016 9 e002661 10.1161/CIRCHEARTFAILURE.115.002661 26747861 40. Hamo C.E. Bloom M.W. Cardinale D. Ky B. Nohria A. Baer L. Skopicki H. Lenihan D.J. Gheorghiade M. Lyon A.R. Cancer therapy-related cardiac dysfunction and heart failure: Part 2: Prevention, treatment, guidelines, and future directions Circul. Heart Fail. 2016 9 e002843 10.1161/CIRCHEARTFAILURE.115.002843 26839395 41. Fan H. Yang L. Fu F. Xu H. Meng Q. Zhu H. Teng L. Yang M. Zhang L. Zhang Z. Cardioprotective effects of salvianolic Acid a on myocardial ischemia-reperfusion injury in vivo and in vitro Evid. Based Complement. Altern. Med. eCAM 2012 2012 508938 10.1155/2012/508938 21789047 42. Choi E.H. Chang H.J. Cho J.Y. Chun H.S. Cytoprotective effect of anthocyanins against doxorubicin-induced toxicity in H9c2 cardiomyocytes in relation to their antioxidant activities Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 2007 45 1873 1881 10.1016/j.fct.2007.04.003 17524543 43. Eldridge S. Guo L. Mussio J. Furniss M. Hamre J. 3rd Davis M. Examining the protective role of ErbB2 modulation in human-induced pluripotent stem cell-derived cardiomyocytes Toxicol. Sci. 2014 141 547 559 10.1093/toxsci/kfu150 25055963 44. Guo L. Eldridge S. Furniss M. Mussio J. Davis M. Use of human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) to monitor compound effects on cardiac myocyte signaling pathways Curr. Protoc. Chem. Biol. 2015 7 141 185 26331525 45. Guo L. Coyle L. Abrams R.M. Kemper R. Chiao E.T. Kolaja K.L. Refining the human iPSC-cardiomyocyte arrhythmic risk assessment model Toxicol. Sci. 2013 136 581 594 10.1093/toxsci/kft205 24052561 46. Liang P. Lan F. Lee A.S. Gong T. Sanchez-Freire V. Wang Y. Diecke S. Sallam K. Knowles J.W. Wang P.J. Drug screening using a library of human induced pluripotent stem cell-derived cardiomyocytes reveals disease-specific patterns of cardiotoxicity Circulation 2013 127 1677 1691 10.1161/CIRCULATIONAHA.113.001883 23519760

PMC005xxxxxx/PMC5000643.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081245ijms-17-01245ArticleImpact of HIV Infection and Anti-Retroviral Therapy on the Immune Profile of and Microbial Translocation in HIV-Infected Children in Vietnam Bi Xiuqiong 1†Ishizaki Azumi 1†Nguyen Lam Van 2Matsuda Kazunori 3Pham Hung Viet 2Phan Chung Thi Thu 2Ogata Kiyohito 3Giang Thuy Thi Thanh 2Phung Thuy Thi Bich 2Nguyen Tuyen Thi 2Tokoro Masaharu 4Pham An Nhat 2Khu Dung Thi Khanh 2Ichimura Hiroshi 1*Wigdahl Brian Academic Editor1 Department of Viral Infection and International Health, Graduate School of Medical Sciences, Kanazawa University, Kanazawa 920-8640, Japan; [email protected] (X.B.); [email protected] (A.I.)2 National Hospital of Pediatrics, Hanoi 100-000, Vietnam; [email protected] (L.V.N.); [email protected] (H.V.P.); [email protected] (C.T.T.P.); [email protected] (T.T.T.G.); [email protected] (T.T.B.P.); [email protected] (T.T.N.); [email protected] (A.N.P.); [email protected] (D.T.K.K.)3 Yakult Central Institute, Tokyo 186-8650, Japan; [email protected] (K.M.); [email protected] (K.O.)4 Department of Parasitology, Graduate School of Medical Sciences, Kanazawa University, Kanazawa 920-8640, Japan; [email protected]* Correspondence: [email protected]; Tel.: +81-76-265-2228; Fax: +81-76-234-4237† These authors contributed equally to this work. 02 8 2016 8 2016 17 8 124503 7 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).CD4+ T-lymphocyte destruction, microbial translocation, and systemic immune activation are the main mechanisms of the pathogenesis of human immunodeficiency virus type 1 (HIV) infection. To investigate the impact of HIV infection and antiretroviral therapy (ART) on the immune profile of and microbial translocation in HIV-infected children, 60 HIV vertically infected children (31 without ART: HIV(+) and 29 with ART: ART(+)) and 20 HIV-uninfected children (HIV(−)) aged 2–12 years were recruited in Vietnam, and their blood samples were immunologically and bacteriologically analyzed. Among the HIV(+) children, the total CD4+-cell and their subset (type 1 helper T-cell (Th1)/Th2/Th17) counts were inversely correlated with age (all p < 0.05), whereas regulatory T-cell (Treg) counts and CD4/CD8 ratios had become lower, and the CD38+HLA (human leukocyte antigen)-DR+CD8+- (activated CD8+) cell percentage and plasma soluble CD14 (sCD14, a monocyte activation marker) levels had become higher than those of HIV(−) children by the age of 2 years; the CD4/CD8 ratio was inversely correlated with the plasma HIV RNA load and CD8+-cell activation status. Among the ART(+) children, the total CD4+-cell and Th2/Th17/Treg-subset counts and the CD4/CD8 ratio gradually increased, with estimated ART periods of normalization being 4.8–8.3 years, whereas Th1 counts and the CD8+-cell activation status normalized within 1 year of ART initiation. sCD14 levels remained high even after ART initiation. The detection frequency of bacterial 16S/23S ribosomal DNA/RNA in blood did not differ between HIV-infected and -uninfected children. Thus, in children, HIV infection caused a rapid decrease in Treg counts and the early activation of CD8+ cells and monocytes, and ART induced rapid Th1 recovery and early CD8+-cell activation normalization but had little effect on monocyte activation. The CD4/CD8 ratio could therefore be an additional marker for ART monitoring. HIV-infected childrenintestinal microbial translocationimmune activation16S/23S ribosomal DNA ==== Body 1. Introduction Mucosa-associated lymphoid tissues, such as the gut-associated lymphoid tissue (GALT), harbor approximately 40%–60% of lymphocytes in the human body. GALT is the largest replication site and reservoir of human immunodeficiency virus type 1 (HIV) among HIV-infected individuals. In the early stages of HIV infection, the remarkable destruction of CD4+ T (CD4+) cells, particularly the type 17 helper T-cell (Th17) subset, occurs in GALT [1,2,3], resulting in the decline in the immune and mechanical barrier functions of the gut mucosa. Subsequently, the translocation of microbial products, such as lipopolysaccharides from gram-negative bacteria and bacterial DNA, from the gastrointestinal tract to systemic circulation occurs, which induces systemic immune activation, disrupts immune balance, and causes further loss of CD4+ cells [3,4,5,6]. CD4+ cells in the peripheral blood recover 1–3 years after antiretroviral therapy (ART) initiation, whereas the recovery of CD4+ cells in GALT is much slower. The barrier function of the gut mucosa therefore remains impaired, and the immune activation continues for many years after ART initiation [3,7,8]. Children who have contracted HIV from their mothers (vertical infection) progress to acquired immunodeficiency syndrome (AIDS) more rapidly than HIV-infected adults, showing a more rapid decline in CD4+-cell counts and high plasma HIV viral load (VL). Without ART, more than 50% of these children die by the age of 2 years [9,10]. In children, there is an abundance of lymphoid aggregates in the gut submucosa and memory CD4+CCR5+ T cells in gut epithelial cells [11]. Because these cells are highly susceptible to HIV infection even without activation, they are likely to be the prime site of HIV infection and replication [9,11]. Furthermore, a relatively low HIV-specific Th1 response coinciding with a low Th1/Th2 ratio and an expanded population of regulatory T cells (Tregs) are observed among HIV-infected children, particularly among infants [9,12,13]. HIV-infected children also exhibit a decrease in CD4+-cell counts, CD8+-cell activation, and intestinal microbial translocation [14,15], and ART effectively restores CD4+-cell populations and suppresses CD8+-cell activation [14,16,17], as is the case among HIV-infected adults. However, the impact of HIV infection and ART on different CD4+-cell subsets (Th1, Th2, Th17, and Treg) in children remains poorly understood. Since 2008, we have followed up with HIV-infected children in Hanoi, Vietnam to find an efficient and cost-effective method and immunological markers for monitoring ART in resource-limited settings [18]. Based on the results of that field study, we conducted this cross-sectional study to clarify the impact of HIV infection and ART on the immune profile and microbial translocation status of children aged over 2 years who have an immune status considered to be relatively mature and stable [9,19,20]. 2. Results 2.1. Characteristics of the Subjects This study included 60 HIV vertically infected children (31 without ART (HIV(+)) and 29 with ART (ART(+)) and 20 HIV-uninfected healthy children (HIV(−)). The characteristics of the study subjects are shown in Table 1. The ART(+) children had received ART for a median period of 3.5 years, and 22 (75.9%) had an undetectable plasma VL (< 220 copies/mL). More ART(+) (5/29, 17.2%) than HIV(+) (1/31, 3.2%) children were at WHO clinical stage 2, although the difference was not statistically significant (p = 0.098). The HIV(+) and ART(+) children were 2 years older than the HIV(−) children (HIV(+) vs. HIV(−): median age 6.2 vs. 4.1 years, respectively, p = 0.034; ART(+) vs. HIV(−): median age 6.1 vs. 4.1 years, respectively, p = 0.009), although height and body weight did not significantly differ among the three groups (all p > 0.1). 2.2. Immune Status of HIV-Infected Children The immune statuses of the children in the three groups are shown in Table 1. Compared with the HIV(−) children, the HIV(+) children had significantly lower total CD4+-cell (p = 0.003) and Th1/Th2/Th17/Treg-subset counts (p = 0.003/0.016/<0.001/<0.001, respectively), a lower percentage of CD4+ cells in lymphocytes (p = 0.001), a higher percentage of CD8+ cells in lymphocytes (p < 0.001), a lower CD4/CD8 ratio (p < 0.001), and a higher percentage of CD38+HLA (human leukocyte antigen)-DR+CD8+ cells in CD8+ cells (activated CD8+ cells, p < 0.001), although CD8+-cell counts and the CD4+-cell activation status (percentage of CD38+HLA-DR+CD4+ cells in CD4+ cells) did not significantly differ between the two groups (p = 0.239 and 0.969, respectively). The HIV(+) children had higher plasma soluble CD14 (sCD14; a monocyte activation marker) levels than the HIV(−) children (p = 0.009). The ART(+) children had significantly lower total CD4+-cell (p = 0.018) and Th2/Th17/Treg-subset counts (p = 0.009/<0.001/0.004), a lower CD4/CD8 ratio (p = 0.001), and higher sCD14 levels (p < 0.001) than the HIV(−) children. However, the CD4+-cell percentage in lymphocytes (p = 0.404), Th1 count (p = 0.611), and activated CD8+-cell percentage (p = 0.329) did not significantly differ between the two groups. The ART(+) children had significantly higher Th1//Th17/Treg-subset counts (p = 0.002/0.016/<0.001), higher sCD14s levels (p < 0.001), and lower activated CD8+ percentages (p < 0.001) than the HIV(+) children. However, the total CD4+-cell (p = 0.429) and Th2-subset counts (p = 0.970) and the CD4/CD8 ratio (p = 0.181) did not significantly differ between the two groups. 2.3. Impact of HIV Infection on Immune Profile Among the HIV(+) children, the total CD4+-cell counts (Figure 1A, p = 0.007), Th1/Th2/Th17-subset counts (Figure 1B–D, p = 0.040/0.012/0.024, respectively), CD4+-cell percentage in lymphocytes (Figure 1F, p = 0.033), and CD8+-cell counts (Figure 1G, p = 0.060) were inversely correlated with age (that is, nearly equal to their HIV-infection period). On the other hand, Treg counts (Figure 1E) and the CD4/CD8 ratio (Figure 1J) were lower and the CD38+HLA-DR+CD8+- (activated CD8+) cell percentage (Figure 1I) and plasma sCD14 levels (Figure 1K) were higher than those of the HIV(−) children by the age of 2 years. The CD38+HLA-DR+CD4+- (activated CD4+) cell percentage was not significantly correlated with age (Figure 1H). The activated CD8+-cell percentage (Figure 1L, p < 0.001) and sCD14 levels (Figure 1M, p = 0.030) were positively correlated with the plasma VL. The CD4/CD8 ratio was inversely correlated with the plasma VL (Figure 1N, p = 0.045) and CD8 activation status (Figure 1O, p = 0.003). 2.4. Impact of ART on the HIV-Induced Immune Profile Among the ART(+) children, the total CD4+-cell counts (Figure 2A, p = 0.001), Th2/Th17/Treg-subset counts (Figure 2C–E, p = 0.001/0.001/0.030, respectively), CD4+-cell percentage in lymphocytes (Figure 2F, p = 0.001), and CD4/CD8 ratio (Figure 2J, p = 0.001) gradually, but significantly, increased during ART, whereas the Th1 counts had increased to the range of those of the HIV(–) children by around 1 year of ART initiation (Figure 2B). The CD8+-cell counts were not significantly correlated with ART duration. None of these markers were significantly correlated with age (data not shown). Linear regression analyses were conducted to estimate the ART duration required for the immunological markers to reach the median values of the HIV(−) children. The estimations were as follows: 4.8 years (95% confidence interval (CI): 2.7–7.0 years) of ART for the total CD4+-cell counts, 3.6 years (95% CI: 1.4–5.7) for the CD4+-cell percentage in lymphocytes, 5.6 years (95% CI: 3.4–7.9) for the Th2 counts, 8.2 years (95% CI: 6.2–10.2) for the Th17 counts, 8.3 years (95% CI: 3.4–12.2) for the Treg counts, and 6.6 years (95% CI: 4.5–8.7) for the CD4/CD8 ratio. During the first year of ART initiation, the percentage of activated CD8+ cells rapidly decreased to the upper range of that among the HIV(–) children and then continued to slowly, but significantly, decrease (Figure 2I, p = 0.010). In contrast, the percentage of activated CD4+ cells marginally decreased with ART duration (Figure 2H, p = 0.053), but the percentage of activated CD4+ cells even among the HIV(+) children did not significantly differ from that among the HIV(−) children (5.6% vs. 6.4%, p = 0.969, Table 1). The plasma sCD14 levels showed no significant changes during almost 6 years of ART (Figure 2K). The CD4/CD8 ratio was inversely correlated with the percentage of activated CD8+ cells (p = 0.003, Figure 2L). 2.5. Physiological Change in Immunological Markers with Age among the HIV(−) Children Among the HIV(−) children, only the percentage of activated CD8+ cells significantly decreased with age (p = 0.023, Figure 3I), and plasma sCD14 levels marginally decreased with time by 4 years of age and remained stable thereafter (p = 0.066, Figure 3K). The other immunological markers did not show significant relationships with age (Figure 3A–H,J,L). 2.6. Microbial Translocation Status The impact of HIV infection and ART on the microbial translocation status of the children was determined by the detection of 12 bacterial 16S/23S ribosomal RNA genes (rDNA) in plasma and 16S/23S ribosomal RNA molecules (rRNA) in whole blood using quantitative PCR (qPCR) and reverse transcription (RT)-qPCR, respectively. Bacterial 16S/23S rDNA from Staphylococcus, Streptococcus, and/or Pseudomonas species was detected in 25.8% (8/31) of the HIV(+) children compared with 15.0% (3/20) of the HIV(−) children (Table 2). The detection frequency of Staphylococcus rDNA was significantly higher among the HIV(+) children than among the ART(+) children (22.6% vs. 0%, respectively, p = 0.011), whereas there was no significant difference between the HIV(+) and HIV(−) children in terms of the frequency of each bacterial rDNA (p = 0.169) and the number of bacterial DNA copies (data not shown). No target bacterial 16S/23S rRNA was detected in children, regardless of the HIV status (data not shown). 3. Discussion In the current study, we investigated the impact of HIV infection and ART on the immune profile of children aged over 2 years. Among the HIV(+) children, the total counts of CD4+ cells and their subsets (Th1/Th2/Th17) were inversely correlated with age, whereas Treg counts, CD38+HLA (human leukocyte antigen)-DR+CD8+- (activated CD8+) cell percentage, and plasma sCD14 levels (activated monocyte level) were not. By the age of 2 years, Treg counts had become lower and the activated CD8+ cell and monocyte levels had become higher than those of the HIV(−) children. In addition, activated CD8+ cell and monocyte levels were positively correlated with VL, and the CD8+-cell activation status was inversely, although marginally, correlated with Treg counts, (p = 0.0575). These findings indicate that vertical HIV infection induces a rapid decrease in the Treg subset and the early activation of CD8+ cells and monocytes. These data suggest that the early decline in Tregs contributes to CD8+-cell activation in the early phase of HIV infection in children, as previously reported in adults [21]. The early activation of CD8+ cells after HIV infection [15,16] and a correlation between CD8+ cell activation and VL [22] have been previously reported, whereas a rapid decrease in Treg subset has not been previously reported in children, although it has been reported in adults with primary HIV infection [23]. Among the ART(+) children, Th2/Th17/Treg-subset counts were positively correlated with ART duration and were estimated to reach the levels of HIV(−) children after 5.6, 8.2, and 8.3 years of ART initiation, respectively, compared with an estimate of only 1–2 years for Th1 counts. The activated CD8+ cell percentage decreased to the level of the HIV(−) children during the first year of ART initiation, whereas the plasma sCD14 level (activated monocytes level) was maintained at a level higher than that of the HIV(−) children even after VL was controlled and CD8+-cell activation was normalized. These findings indicate that in children, ART induced a more rapid Th1 recovery than the Th2/Th17/Treg subsets and a rapid normalization of CD8+-cell activation but had little effect on monocyte activation. Although previous studies have reported the early normalization of CD8+-cell activation after ART initiation [16,24], the recovery profile of CD4+-cell subsets after ART initiation, particularly the rapid recovery of Th1 and the slower recovery of Th2/Th17/Treg, has not been reported before in children. Incidentally, the rapid recovery of Th1 may play a role in the control of HIV infection together with ART. Similar to our findings, a previous study has reported that the plasma sCD14 level is not normalized even after 2 years of ART initiation in children, although it is significantly reduced after ART initiation [25]. Because the sCD14 level has been proposed as an independent predictor of non-AIDS-defining morbidity events even during suppressive ART in HIV-infected adults [26,27], further longitudinal studies may be needed to elucidate the implication of the high monocyte activation status in HIV-infected children under ART. Among the HIV(+) children, the CD4/CD8 ratio was inversely correlated with the plasma VL and the activated CD8+-cell percentage. Among the ART(+) children, the CD4/CD8 ratio was positively correlated with ART duration, as were the total CD4+-cell and Th2/Th17/Treg-subset counts, and inversely correlated with the percentage of activated CD8+ cells. Recent studies have suggested that the CD4/CD8 ratio is a marker of T-cell activation, senescence, and activation/exhaustion in treated HIV-infected children and young adults, and that it could be independently associated with the risk of non-AIDS-related morbidity and mortality [28,29]. Given the cost effectiveness of measuring CD4+ and CD8+ cell levels in clinical laboratories, the CD4/CD8 ratio could be used as an additional immunological marker to monitor ART outcomes, particularly in resource-limited settings. Intestinal microbial translocation plays a role in the pathogenesis of HIV infection [6]. Furthermore, immune activation is driven by intestinal microbial translocation in HIV-infected children both before and after ART [22,30]. However, in this study, none of the targeted bacterial 16S/23S rRNA fragments were detected in whole blood samples of the 80 study participants, indicating that live bacteria (bacteremia) are rarely detected in HIV-infected and -uninfected children. In contrast, bacterial rDNA was detected in 8 HIV(+) children and 3 HIV(−) children, and the detection frequency of staphylococcal rDNA was significantly higher among the HIV(+) children than among the ART(+) children (22.6% vs. 0%, respectively). However, there was no significant association of the copy number of bacterial 16S/23S rDNA with CD8+-cell or monocyte activation and no significant difference in the frequency of bacterial rDNA between the HIV(+) and HIV(−) children. These data suggest that ART decreases the occurrence of intestinal microbial translocation in HIV-infected children, although the possibility of contamination with Staphylococcus epidermidis cannot be excluded. Thus, further studies are needed to elucidate the reasons for the discrepancies between our data and previous findings [22,30] in relation to microbial translocation in HIV-infected children. In this study, we recruited children who were over 2 years of age because their immune systems are considered to be relatively mature and stable [9,19,20]. We showed that all targeted immunological markers, except CD8+-cell activation, did not significantly change with age among the HIV(–) children (Figure 3). These data demonstrate that the results presented here on the impact of HIV infection and ART on the immune profile of children were not influenced by growth-related physiological changes. There are some limitations to this study. First, this was a cross-sectional study rather than a longitudinal study; Second, the number of study subjects was relatively small, which limited our findings in relation to microbial translocation; Third, the HIV(+) and ART(+) children were 2 years older than the HIV(−) children; Finally, only cell-surface markers were used to identify the CD4+-subsets, particularly Treg. Therefore, these findings should be confirmed by longitudinal studies following HIV-infected children before and during ART with age-matched controls. To our knowledge, this is the first study to investigate the impact of HIV infection and ART on the immune profile of CD4+-cell subsets (Th1/Th2/Th17/Tregs) concurrently with the activation of T cells and monocytes in children. We found that HIV infection induced a more rapid decline in the Treg subset than in the Th1/Th2/Th17 subsets and the early activation of CD8+ cells and monocytes. We also found that ART induced a more rapid recovery of the Th1 subset than the Th2/Th17/Treg subsets and the early normalization of CD8+-cell activation but that it had little effect on the activation of monocytes in children. Finally, we have provided evidence that the CD4/CD8 ratio is an additional marker for ART outcomes. 4. Materials and Methods 4.1. Subjects and Study Design Sixty Vietnamese children vertically infected with HIV were recruited in May 2012. These children were assigned to one of two groups: the HIV(+) group and the ART(+) group. The HIV(+) group comprised 31 children with HIV who did not receive ART; the female/male ratio was 14/17, and the median age was 6.2 years (2.0–11.0 years). The ART(+) group comprised 29 children with HIV who had been treated with ART; the female/male ratio was 12/17, and the median age was 6.1 years (3.6–8.6 years). The study inclusion criteria for these HIV-1-infected groups were as follows: the children (1) had been followed at the National Hospital of Pediatrics (NHP) in Hanoi, Vietnam; and (2) were more than 2 years of age because at this point the immune system is relatively mature and stable [9,19,20]. The exclusion criteria were as follows: children who (1) had progressed to AIDS; (2) had received any treatment within the prior 8 weeks that might influence the immune system; and (3) had symptoms of gastrointestinal infections at the time of recruitment. The children in the ART(+) group resided at an orphanage center near Hanoi; the children in the HIV(+) group were followed at the outpatient department of NHP. A third control group, the HIV(−) group, was also included and comprised 20 healthy Vietnamese children without HIV infections. The female/male ratio was 8/12, and the median age was 4.1 years (2.0–8.3 years). The children in this group resided at an orphanage center near Hanoi. The characteristics of the study subjects are shown in Table 1. At recruitment, the median ART duration in the ART(+) group was 3.5 years (0.8–5.8) years. Of the 29 children, 8 received zidovudine (AZT)/lamivudine (3TC)/nevirapine (NVP), 7 received stavudine (d4T)/3TC/NVP, 6 received AZT/3TC/efavirenz (EFV), 4 received d4T/3TC/EFV, 2 received AZT/3TC/lopinavir boosted with ritonavir (LPV/r), 1 received abacavir (ABC)/3TC/LPV/r, and 1 received ABC/didanosine/LPV/r. The protocol of this cross-sectional study was approved by the Ethics Committee of Kanazawa University in Japan and the Ethics Committee of NHP in Vietnam. All of the HIV-infected children who met the inclusion criteria and did not correspond to the exclusion criteria, and all of the HIV-uninfected children who resided in the orphanage and were eligible for the criteria, were invited to join this study. The family or guardian of each subject was informed, and only those who voluntarily consented to participate were recruited. Written consent was obtained from all participants. 4.2. Plasma HIV VL and sCD14 Concentration Plasma HIV VL was measured using a Cobas Taqman HIV-1 Test Kit version 1.0 (Roche Molecular Systems, Inc., Branchburg, NJ, USA), following the manufacturer’s instructions (detection limit: 40 copies/mL). Plasma samples were diluted to 1:5.5 for measurements, which resulted in a final detection limit of 220 copies/mL. Plasma sCD14 concentration was measured with a Human sCD14 Immunoassay Kit (R&D Systems, Minneapolis, MN, USA), following the manufacturer’s instructions. 4.3. Immunological Analysis Immune activation was evaluated based on the percentage of CD4+ and CD8+ lymphocytes expressing CD38 and major histocompatibility complex class II (HLA-DR) molecules [31]. The different CD4+-cell subsets were identified using cell-surface markers: CXCR3+CCR6−CD4+ (“Th1”), CXCR3−CCR6−CD4+ (“Th2”), CXCR3−CCR6+CD4+ (“Th17”) [32], and CD25highCD4+ (“Treg”) [33]. Blood samples were processed for cell staining within 6 h after collection. Whole blood samples (50 µL) were stained for 15 min at 4 °C with a combination of four monoclonal antibodies: anti-CD4 PerCP and CD8 PE (BD Biosciences, San Jose, CA, USA), CD38 FITC (Miltenyi Biotec, Auburn, AL, USA), and HLA-DR PE-Cy7 (Biolegend, San Diego, CA, USA). Alternatively, samples were stained with a combination of four anti-CD4 monoclonal antibodies: anti-CD4 PerCP and CD25 PE (BD Biosciences), and CXCR3 FITC and CCR6 PE-Cy7 (Biolegend). After red blood cells were lysed with a lysing buffer (BD Biosciences), the remaining cells were washed once with 1 mL of phosphate buffered saline (PBS), kept at 4 °C in PBS with 1% paraformaldehyde and 0.5% bovine serum albumin, and analyzed within 48 h after staining at Kanazawa University with a JSAN flow cytometer (Bay Bioscience, Kobe, Japan). The data were analyzed with Flowjo V.7.5.5 (FLOWJO, OR, USA). 4.4. Detection of Bacterial Ribosomal RNA Genes (rDNA) in Plasma Isolated plasma (100 µL) was used for DNA extraction with the SMI TEST EX R&D (Medical and Biological Laboratories Co., Ltd., Aichi, Japan), following the manufacturer’s instructions. qPCR was then performed with a TaKaRa Taq kit (TaKaRa Bio Inc., Shiga, Japan). The reaction mixture (10 μL) contained 5 μL of template DNA and 0.2 μM of each specific primer set, except for the primer targeting g-Bfra-F2/g-Bfra-R (0.4 μM). The primers were designed to target bacterial 16S or 23S rRNA genes (Table S1). The amplification program comprised one cycle at 94 °C for 5 min and 45 cycles at 94 °C for 20 s, 55 °C or 60 °C for 20 s, and 72 °C for 50 s. A standard curve was generated with qPCR data and cycle threshold (Ct) values. The target gene copy number in the plasma samples was determined in a sample of the extracted DNA (1/20 of the DNA extracted from 100 µL of plasma). This DNA was subjected to qPCR, and the Ct value was applied to the standard curve to obtain the corresponding bacterial rDNA copy number/µL of plasma. With this procedure, the lower detection limit for the targeted gene was 2 copies/µL of plasma. 4.5. Detection of Bacterial rRNA in Blood Peripheral blood (1 mL) was added to two volumes of the RNA Protect bacterial reagent (QIAGEN GmbH, Hilden, Germany). After centrifugation of the mixture at 14,000× g for 10 min, the supernatant was discarded and the pellet was stored at −80 °C until further use. Total RNA extraction was followed by RT-qPCR as previously described [34]. Each RNA sample was diluted; then, diluted samples (corresponding to amounts of 1/200 and 1/20 of the extracted RNA from 1 mL of blood) were subjected to RT-qPCR with specific primer sets that targeted bacterial 16S/23S rRNA [34,35] to investigate if live bacteria existed in the blood stream of the study subjects (Table S1). With this procedure, the lower detection limit for the targeted bacteria was 4 cells/mL of blood. 4.6. Statistical Analysis Statistical analyses were performed with the SPSS programs (IBMSPSS statistics 19, IBM Corporation, NY, USA). The Mann–Whitney U test was used to compare the markers among the three groups (HIV(−), HIV(+),and ART(+)). Fisher’s exact test or the chi square test was used to compare the detection frequency of bacterial rDNA in plasma, sex distribution, and the WHO clinical stage across the groups. Spearman’s rank correlation was used to analyze the correlation among the biological markers in each group. p-values < 0.05 were considered statistically significant. 5. Conclusions In children, HIV infection caused a rapid decrease in Treg counts and the early activation of CD8+ cells and monocytes, and ART induced rapid Th1 recovery and early CD8+-cell activation normalization but had little effect on monocyte activation. The CD4/CD8 ratio could be an additional marker for ART monitoring. Acknowledgments This study was supported in part by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) in Japan (the Program of Founding Research Centers for Emerging and Reemerging Infectious Disease) and the 2012 Kanazawa University President Strategic Research fund. No potential conflict of interests relevant to this article exists. The authors are grateful to the children who participated in this study for their invaluable support throughout the use of their samples, and to the staff of the Departments of Infectious Diseases and Molecular Laboratory of NHP who made enormous contributions to this work. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1245/s1. Click here for additional data file. Author Contributions Azumi Ishizaki, Xiuqiong Bi, and Hiroshi Ichimura conceived and designed the study; Lam Van Nguyen, Hung Viet Pham, Chung Thi Thu Phan, Azumi Ishizaki, Thuy Thi Thanh Giang, Thuy Thi Bich Phung, and Tuyen Thi Nguyen collected the blood samples and clinical data; An Nhat Pham and Dung Thi Khanh Khu coordinated the groups of patients; Azumi Ishizaki, Kazunori Matsuda, Hung Viet Pham, and Chung Thi Thu Phan processed the samples; Xiuqiong Bi, Hung Viet Pham, Kazunori Matsuda, and Kiyohito Ogata performed the experiments; Azumi Ishizaki, Xiuqiong Bi, Masaharu Tokoro, and Hiroshi Ichimura analyzed the data; An Nhat Pham, Dung Thi Khanh Khu, and Hiroshi Ichimura supplied the laboratory materials; Xiuqiong Bi, Azumi Ishizaki, and Hiroshi Ichimura wrote the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Correlation between immunological markers, age, and/or plasma viral load among HIV(+) children (A–O). This bivariate correlation was estimated on the basis of Spearman’s rank correlation analysis. Regression lines are shown only for significantly correlated bivariates. Th: helper T-cell, Treg: regulatory T-cell, VL: viral load. * Median and interquartile range of the HIV(−) group. Figure 2 Correlations between ART duration and immunological markers among ART(+) children (A–L). * Median and interquartile range (IQR) of the HIV(−) group; # Median and IQR of the HIV(+) group. Figure 3 Correlation between immunological markers and age among HIV(−) children (A–L). This bivariate correlation was estimated on the basis of Spearman’s rank correlation analysis. Regression lines are shown only for significantly correlated bivariates. ijms-17-01245-t001_Table 1Table 1 Characteristics and immune status of each group. Items HIV(+) (n = 31) HIV(−) (n = 20) ART(+) (n = 29) p Values HIV(+) vs. HIV(−) ART(+) vs. HIV(−) HIV(+) vs. ART(+) Age (years) * 6.2 (2.0–11.0) 4.1 (2.0–8.3) 6.1 (3.6–8.6) 0.034 0.009 0.584 Gender, female (n)/male (n) * 14/17 8/12 12/17 0.718 0.920 0.764 Height (cm) * 109.0 (77.0–129.5) 110.0 (80.0–130.0) 110.0 (90.0–130.0) 0.771 0.418 0.539 Body weight (kg) * 17.5 (10.0–27.0) 16.0 (9.0–35.0) 19.8 (12.0–32.8) 0.395 0.140 0.192 WHO clinical stage: 2 (n)/1 (n) 1/30 5/24 - - 0.098 ART duration (years) * 3.5 (0.8–5.8) - - - Viral load (log10 copies/mL) * 5.0 (3.2–6.5) ** - - < 0.001 % CD4+ * 22.1 (3.6–44.5) 29.5 (19.6–54.9) 28.8 (12.4–47.1) 0.001 0.404 0.010 CD4+-cell counts (cells/μL) * 698 (97–1784) 1050 (693–2688) 894 (244–1711) 0.003 0.018 0.429 Th1 counts (cells/μL) * 80 (25–227) 136 (74–220) 147 (49–211) 0.003 0.611 0.002 Th2 counts (cells/μL) * 537 (63–1375) 822 (413–2196) 553 (119–1369) 0.016 0.009 0.970 Th17 counts (cells/μL) * 45 (6–116) 109 (51–192) 58 (23–144) <0.001 <0.001 0.016 Treg counts (cells/μL) * 14 (0–133) 48 (16–94) 30 (9–71) <0.001 0.004 <0.001 %CD38+HLA-DR+/CD4 * 5.6 (2.2–17.3) 6.4 (2.7–27.6) 4.3 (2.0–15.6) 0.969 0.036 0.003 % CD8+ * 43.4 (29.5–61.4) 31.4 (23.5–43.3) 44.7 (31.1–61.4) <0.001 <0.001 0.464 CD8+-cell counts (cells/μL) * 1368 (470–3127) 1101 (634–2874) 1212 (769–2064) 0.239 0.290 0.631 %CD38+HLA-DR+/CD8 * 27.5 (12.2–53.3) 12.9 (5.8–38.6) 10.2 (5.0–27.7) <0.001 0.329 <0.001 CD4/CD8 * 0.50 (0.06–1.19) 1.03 (0.45–2.34) 0.66 (0.20–1.42) <0.001 0.001 0.181 sCD14 (ng/mL) * 1637 (1049–3003) 1413 (944–2580) 1964 (1281–3169) 0.009 <0.001 <0.001 HIV(+): Children infected with HIV and without ART; ART(+): Children infected with HIV and on ART; HIV(−): Children not infected with HIV. P values are from the Man-Whitney U test, except the p values for sex and WHO clinical stage comparison, which are from the chi square test or Fisher’s exact test. * Median (range); ** 22 ART(+) children with undetectable viral load. ijms-17-01245-t002_Table 2Table 2 Bacterial 16S/23S ribosomal RNA gene (rDNA) detection in plasma. Target Bacteria HIV(+) (n = 31) HIV(−) (n = 20) ART(+) (n = 29) p Values HIV(+) vs. HIV(−) ART(+) vs. HIV(−) HIV(+) vs. ART(+) C. coccoides group 0 0 0 - - - C. leptum subgroup 0 0 0 - - - B. fragilis group 0 0 0 - - - Bifidobacterium 0 0 0 - - - Atopobium cluster 0 0 0 - - - Prevotella 0 0 0 - - - Enterobacteriaceae 0 0 0 - - - Streptococcus 1 (3.2%) 0 0 1 - 1 Enterococcus 0 0 0 - - - Staphylococcus 7 (22.6%) 1 (5.0%) 0 0.13 0.41 0.011 Pseudomonas 1 (3.2%) 2 (10.0%) 0 0.55 0.16 1 L. casei subgroup 0 0 0 - - - HIV(+): Children infected with HIV and without ART; ART(+): Children infected with HIV and on ART; HIV(−): Children not infected with HIV; C. coccoide: Clostridium coccoide; C. leptum; Clostridium leptum; B. fragilis: Bacteroides fragilis. p values: Fisher’s exact probability test. ==== Refs References 1. Mehandru S. Poles M.A. Tenner-Racz K. Horowitz A. Hurley A. Hogan C. Boden D. Racz P. Markowitz M. Primary HIV-1 infection is associated with preferential depletion of CD4+ T lymphocytes from effector sites in the gastrointestinal tract J. Exp. Med. 2004 200 761 770 10.1084/jem.20041196 15365095 2. Brenchley J.M. Paiardini M. Knox K.S. Asher A.I. Cervasi B. Asher T.E. Scheinberg P. Price D.A. Hage C.A. Kholi L.M. Differential Th17 CD4 T-cell depletion in pathogenic and nonpathogenic lentiviral infections Blood 2008 112 2826 2835 10.1182/blood-2008-05-159301 18664624 3. Hunt P.W. Th17, gut, and HIV: Therapeutic implications Curr. Opin. HIV AIDS 2010 5 189 193 10.1097/COH.0b013e32833647d9 20543599 4. Gordon S.N. Cervasi B. Odorizzi P. Silverman R. Aberra F. Ginsberg G. Estes J.D. Paiardini M. Frank I. Silvestri G. Disruption of intestinal CD4+ T cell homeostasis is a key marker of systemic CD4+ T cell activation in HIV-infected individuals J. Immunol. 2010 185 5169 5179 10.4049/jimmunol.1001801 20889546 5. Estes J.D. Harris L.D. Klatt N.R. Tabb B. Pittaluga S. Paiardini M. Barclay G.R. Smedley J. Pung R. Oliveira K.M. Damaged intestinal epithelial integrity linked to microbial translocation in pathogenic simian immunodeficiency virus infections PLoS Pathog. 2010 6 1245 10.1371/journal.ppat.1001052 20808901 6. Brenchley J.M. Price D.A. Schacker T.W. Asher T.E. Silvestri G. Rao S. Kazzaz Z. Bornstein E. Lambotte O. Altmann D. Microbial translocation is a cause of systemic immune activation in chronic HIV infection Nat. Med. 2006 12 1365 1371 10.1038/nm1511 17115046 7. Mehandru S. Poles M.A. Tenner-Racz K. Jean-Pierre P. Manuelli V. Lopez P. Shet A. Low A. Mohri H. Boden D. Lack of mucosal immune reconstitution during prolonged treatment of acute and early HIV-1 infection PLoS Med. 2006 3 1245 8. Bi X. Suzuki Y. Gatanaga H. Oka S. High frequency and proliferation of CD4+ FOXP3+ Treg in HIV-1-infected patients with low CD4 counts Eur. J. Immunol. 2009 39 301 309 10.1002/eji.200838667 19089812 9. Tobin N.H. Aldrovandi G.M. Immunology of pediatric HIV infection Immunol. Rev. 2013 254 143 169 10.1111/imr.12074 23772619 10. Prendergast A.J. Klenerman P. Goulder P.J. The impact of differential antiviral immunity in children and adults Nat. Rev. Immunol. 2012 12 636 648 10.1038/nri3277 22918466 11. Bunders M.J. van der Loos C.M. Klarenbeek P.L. van Hamme J.L. Boer K. Wilde J.C. de Vries N. van Lier R.A. Kootstra N. Pals S.T. Memory CD4+ CCR5+ T cells are abundantly present in the gut of newborn infants to facilitate mother-to-child transmission of HIV-1 Blood 2012 120 4383 4390 10.1182/blood-2012-06-437566 23033270 12. Muenchhoff M. Prendergast A.J. Goulder P.J. Immunity to HIV in early life Front. Immunol. 2014 5 391 10.3389/fimmu.2014.00391 25161656 13. Freguja R. Gianesin K. Mosconi I. Zanchetta M. Carmona F. Rampon O. Giaquinto C. de Rossi A. Regulatory T cells and chronic immune activation in human immunodeficiency virus 1 (HIV-1)-infected children Clin. Exp. Immunol. 2011 164 373 380 10.1111/j.1365-2249.2011.04383.x 21438872 14. Patel K. Hernan M.A. Williams P.L. Seeger J.D. McIntosh K. Dyke R.B. Seage G.R. 3rd Long-term effects of highly active antiretroviral therapy on CD4+ cell evolution among children and adolescents infected with HIV: 5 years and counting Clin. Infect. Dis. 2008 46 1751 1760 10.1086/587900 18426371 15. Ananworanich J. Apornpong T. Kosalaraksa P. Jaimulwong T. Hansudewechakul R. Pancharoen C. Bunupuradah T. Chandara M. Puthanakit T. Ngampiyasakul C. Characteristics of lymphocyte subsets in HIV-infected, long-term nonprogressor, and healthy Asian children through 12 years of age J. Allergy Clin. Immunol. 2010 126 1294 1301 10.1016/j.jaci.2010.09.038 21134574 16. Rosenblatt H.M. Stanley K.E. Song L.Y. Johnson G.M. Wiznia A.A. Nachman S.A. Krogstad P.A. Immunological response to highly active antiretroviral therapy in children with clinically stable HIV-1 infection J. Infect. Dis. 2005 192 445 455 10.1086/431597 15995958 17. Lederman M.M. Calabrese L. Funderburg N.T. Clagett B. Medvik K. Bonilla H. Gripshover B. Salata R.A. Taege A. Lisgaris M. Immunologic failure despite suppressive antiretroviral therapy is related to activation and turnover of memory CD4 cells J. Infect. Dis. 2011 204 1217 1226 10.1093/infdis/jir507 21917895 18. Pham H.V. Ishizaki A. Nguyen L.V. Phan C.T. Phung T.T. Takemoto K. Pham A.N. Bi X. Khu D.T. Ichimura H. Two-year outcome of first-line antiretroviral therapy among HIV-1 vertically-infected children in Hanoi, Vietnam Int. J. STD AIDS 2015 26 821 830 10.1177/0956462414556328 25332224 19. Shearer W.T. Rosenblatt H.M. Gelman R.S. Oyomopito R. Plaeger S. Stiehm E.R. Wara D.W. Douglas S.D. Luzuriaga K. McFarland E.J. Lymphocyte subsets in healthy children from birth through 18 years of age: The pediatric AIDS clinical trials group p1009 study J. Allergy Clin. Immunol. 2003 112 973 980 10.1016/j.jaci.2003.07.003 14610491 20. Schatorje E.J. Gemen E.F. Driessen G.J. Leuvenink J. van Hout R.W. de Vries E. Paediatric reference values for the peripheral T cell compartment Scand. J. Immunol. 2012 75 436 444 10.1111/j.1365-3083.2012.02671.x 22420532 21. Tenorio A.R. Martinson J. Pollard D. Baum L. Landay A. The relationship of T-regulatory cell subsets to disease stage, immune activation, and pathogen-specific immunity in HIV infection J. Acquir. Immune Defic. Syndr. 2008 48 577 580 10.1097/QAI.0b013e31817bbea5 18645514 22. Pilakka-Kanthikeel S. Huang S. Fenton T. Borkowsky W. Cunningham C.K. Pahwa S. Increased gut microbial translocation in HIV-infected children persists in virologic responders and virologic failures after antiretroviral therapy Pediatr. Infect. Dis. J. 2012 31 583 591 10.1097/INF.0b013e31824da0f5 22333700 23. Simonetta F. Lecuroux C. Girault I. Goujard C. Sinet M. Lambotte O. Venet A. Bourgeois C. Early and long-lasting alteration of effector CD45RA− Foxp3high regulatory T-cell homeostasis during HIV infection J. Infect. Dis. 2012 205 1510 1519 10.1093/infdis/jis235 22457280 24. Rainwater-Lovett K. Nkamba H. Mubiana-Mbewe M. Moore C.B. Margolick J. Moss W.J. Changes in cellular immune activation and memory T-cell subsets in HIV-infected Zambian children receiving HAART J. Acquir. Immune Defic. Syndr. 2014 67 455 462 10.1097/QAI.0000000000000342 25226208 25. Kourtis A.P. Ibegbu C.C. Wiener J. King C.C. Tegha G. Kamwendo D. Kumwenda J. Kaur S.P. Flax V. Ellington S. Role of intestinal mucosal integrity in HIV transmission to infants through breast-feeding: The BAN Study J. Infect. Dis. 2013 208 653 661 10.1093/infdis/jit221 23687226 26. Sandler N.G. Wand H. Roque A. Law M. Nason M.C. Nixon D.E. Pedersen C. Ruxrungtham K. Lewin S.R. Emery S. Plasma levels of soluble CD14 independently predict mortality in HIV infection J. Infect. Dis. 2011 203 780 790 10.1093/infdis/jiq118 21252259 27. Tenorio A.R. Zheng Y. Bosch R.J. Krishnan S. Rodriguez B. Hunt P.W. Plants J. Seth A. Wilson C.C. Deeks S.G. Soluble markers of inflammation and coagulation but not T-cell activation predict non-AIDS-defining morbid events during suppressive antiretroviral treatment J. Infect. Dis. 2014 210 1248 1259 10.1093/infdis/jiu254 24795473 28. Sainz T. Serrano-Villar S. Diaz L. Gonzalez Tome M.I. Gurbindo M.D. de Jose M.I. Mellado M.J. Ramos J.T. Zamora J. Moreno S. The CD4/CD8 ratio as a marker T-cell activation, senescence and activation/exhaustion in treated HIV-infected children and young adults AIDS 2013 27 1513 1516 10.1097/QAD.0b013e32835faa72 23435292 29. Serrano-Villar S. Perez-Elias M.J. Dronda F. Casado J.L. Moreno A. Royuela A. Perez-Molina J.A. Sainz T. Navas E. Hermida J.M. Increased risk of serious non-AIDS-related events in HIV-infected subjects on antiretroviral therapy associated with a low CD4/CD8 ratio PLoS ONE 2014 9 1245 10.1371/journal.pone.0085798 24497929 30. Pilakka-Kanthikeel S. Kris A. Selvaraj A. Swaminathan S. Pahwa S. Immune activation is associated with increased gut microbial translocation in treatment-naive, HIV-infected children in a resource-limited setting J. Acquir. Immune Defic. Syndr. 2014 66 16 24 10.1097/QAI.0000000000000096 24378729 31. Hunt P.W. Martin J.N. Sinclair E. Bredt B. Hagos E. Lampiris H. Deeks S.G. T cell activation is associated with lower CD4+ T cell gains in human immunodeficiency virus-infected patients with sustained viral suppression during antiretroviral therapy J. Infect. Dis. 2003 187 1534 1543 10.1086/374786 12721933 32. Maecker H.T. McCoy J.P. Nussenblatt R. Standardizing immunophenotyping for the human immunology project Nat. Rev. Immunol. 2012 12 191 200 10.1038/nri3229 22343568 33. Kinter A. McNally J. Riggin L. Jackson R. Roby G. Fauci A.S. Suppression of HIV-specific T cell activity by lymph node CD25+ regulatory T cells from HIV-infected individuals Proc. Natl. Acad. Sci. USA 2007 104 3390 3395 10.1073/pnas.0611423104 17360656 34. Sakaguchi S. Saito M. Tsuji H. Asahara T. Takata O. Fujimura J. Nagata S. Nomoto K. Shimizu T. Bacterial rRNA-targeted reverse transcription-PCR used to identify pathogens responsible for fever with neutropenia J. Clin. Microbiol. 2010 48 1624 1628 10.1128/JCM.01724-09 20351213 35. Matsuda K. Tsuji H. Asahara T. Matsumoto K. Takada T. Nomoto K. Establishment of an analytical system for the human fecal microbiota, based on reverse transcription-quantitative PCR targeting of multicopy rRNA molecules Appl. Environ. Microbiol. 2009 75 1961 1969 10.1128/AEM.01843-08 19201979

PMC005xxxxxx/PMC5000644.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081246ijms-17-01246ArticleUpregulation of Human ST8Sia VI (α2,8-Sialyltransferase) Gene Expression by Physcion in SK-N-BE(2)-C Human Neuroblastoma Cells Yoon Hyun-Kyoung 1†An Hyun-Kyu 1†Ko Min Jung 2Kim Kyoung-Sook 1Mun Seo-Won 1Kim Dong-Hyun 1Kim Cheol Min 3Kim Cheorl-Ho 4Choi Young Whan 2*Lee Young-Choon 1*Yang Li Academic Editor1 Department of Medicinal Biotechnology, College of Health Sciences, Dong-A university, Busan 49315, Korea; [email protected] (H.-K.Y.); [email protected] (H.-K.A.); [email protected] (K.-S.K.); [email protected] (S.-W.M.); [email protected] (D.-H.K.)2 Department of Horticultural Bioscience, Pusan National University, Miryang 50463, Korea; [email protected] Research Center for Anti-Aging Technology Development, Pusan National University, Busan 46241, Korea; [email protected] Molecular and Cellular Glycobiology Unit, Department of Biological Sciences, SungKyunKwan University, Kyunggi-Do 16419, Korea; [email protected]* Correspondence: [email protected] (Y.W.C.); [email protected] (Y.-C.L.); Tel.: +82-55-350-5522 (Y.W.C.); +82-51-200-7591 (Y.-C.L.); Fax: +82-55-350-5529 (Y.W.C.); +82-51-200-6536 (Y.-C.L.)† These authors contributed equally to this work. 02 8 2016 8 2016 17 8 124603 6 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).In this research, we firstly demonstrated that physcion, an anthraquinone derivative, specifically increased the expression of the human α2,8-sialyltransferase (hST8Sia VI) gene in SK-N-BE(2)-C human neuroblastoma cells. To establish the mechanism responsible for the up-regulation of hST8Sia VI gene expression in physcion-treated SK-N-BE(2)-C cells, the putative promoter region of the hST8Sia VI gene was functionally characterized. Promoter analysis with serially truncated fragments of the 5′-flanking region showed that the region between −320 and −240 is crucial for physcion-induced transcription of hST8Sia VI in SK-N-BE(2)-C cells. Putative binding sites for transcription factors Pax-5 and NF-Y are located at this region. The Pax-5 binding site at −262 to −256 was essential for the expression of the hST8Sia VI gene by physcion in SK-N-BE(2)-C cells. Moreover, the transcription of hST8Sia VI induced by physcion in SK-N-BE(2)-C cells was inhibited by extracellular signal-regulated protein kinase (ERK) inhibitor U0126 and p38 mitogen-activated protein kinase (MAPK) inhibitor SB203580, but not c-Jun N-terminal kinase (JNK) inhibitor SP600125. These results suggest that physcion upregulates hST8Sia VI gene expression via ERK and p38 MAPK pathways in SK-N-BE(2)-C cells. physcionhST8Sia VISK-N-BE(2)-Ctranscription factor Pax-5signal pathway ==== Body 1. Introduction Sialic acid (NeuAc) residues play crucial roles in diverse cellular events, including intercellular adhesion, cell differentiation, microbial attachment, and malignant transformation [1]. The sialylated glycans are formed by a family of sialyltransferases that catalyze the terminal addition of sialic acid to glycan chains [2]. Mammalian sialyltransferases characterized so far exhibit distinct specificities for acceptor substrate for glycoproteins and glycolipids and their gene expressions show marked tissue- and cell type-specific patterns, which is closely connected with cell type-specific glycan structures [2,3,4,5,6]. Because cell type-specific gene expression is generally known to be regulated on the transcriptional level, dissecting the molecular mechanisms for cell type-specific expression of sialyltransferase genes may be essential for understanding cell type-specific sialylation of glycoproteins and glycosphingolipids. Recently we studied the transcriptional regulation mechanism of human sialyltransferase genes that are specifically expressed at high levels in cell lines treated with natural compounds [7,8,9,10,11]. In the current study, we investigated the effect of physcion isolated from the Polygonum multiflorum roots on the human sialyltransferase gene expression in SK-N-BE(2)-C human neuroblastoma cells. Physcion is an anthraquinone derivative isolated from rhubarb, a Chinese herb medicine [12] and the marine-derived fungus Microsporum sp. [13]. Previous studies have demonstrated that physcion has various pharmacological activities such as hepatoprotective [14], anti-microbial [15], and anti-inflammatory effects [16]. In addition, recent studies have also demonstrated that physcion has anti-cancer effects through proliferation inhibition and apoptosis induction in cervical [13], breast [17], and colorectal cancer cells [18]. It was also recently reported that physcion has an anti-metastatic effect against human colorectal cancer SW620 cells by repressing the transcription factor SOX2 [19]. However, the effects of physcion on the expression of various genes, including sialyltransferase genes, and its underlying molecular mechanisms have never been investigated. Therefore, this work was undertaken to assess the effects of physcion on the gene expression of human sialyltransferases and the underlying mechanisms. In the present study, we discovered for the first time the specific increase of the human α2,8-sialyltransferase (hST8Sia VI) gene expression by physcion in human neuroblastoma SK-N-BE(2)-C cells and the molecular basis of physcion-induced hST8Sia VI gene expression was also investigated. 2. Results 2.1. Isolation and Identification of Physcion After extraction of the dried roots of P. multiflorum with ethyl acetate and evaporation of the solvent, the remaining residue was chromatographed over silica gel. Among the compounds isolated by an extensive separation process, a compound displaying induction activity for hST8Sia VI gene expression in SK-N-BE(2)-C cells was selected as the active compound. This compound was clarified to be physcion (Figure 1) by gas chromatography-mass spectrometry (GC-MS) and 1H- and 13C-nuclear magnetic resonance (NMR). GC-MS analysis showed that the isolated physcion had a purity of more than 97%. 2.2. Effect of Physcion on Gene Expression of hST8Sia VI and Cell Proliferation To investigate the effect of physcion on the human sialyltransferase gene expression in SK-N-BE(2)-C cells, mRNA levels of nineteen human sialyltransferases were checked by RT-PCR using total RNAs obtained from cells after 24 h treatment with 40 μM physcion. The result showed that with the exception of hST8Sia VI, the expression levels of eighteen human sialyltransferase genes were not affected by physcion treatment when compared to the control without physcion (Figure S1). Next, we treated the cells for varying periods of time with varying doses of physcion, and checked the transcript level of hST8Sia VI. As shown in Figure 2, hST8Sia VI mRNA levels were significantly increased in SK-N-BE(2)-C cells treated with 40 μM physcion for 24 h. This result clearly revealed that the gene expression of hST8Sia VI was induced by physcion. Next, we checked the cytotoxicity of physcion in SK-N-BE(2)-C cells using MTT assay. As shown in Figure 3A, 24 h incubation with 20 μM physcion did not alter the viability of the cells, whereas treatment with 40 μM physcion reduced the cell viability by about 20% when compared to the control, indicating that physcion has a slight cytotoxic effect at 40 μM. To assess the effect of physcion on apoptosis in SK-N-BE(2)-C cells, the samples obtained after treatment with various concentrations of physcion for 24 h were subjected to SDS-PAGE. Subsequently, caspase-3 activation and poly (ADP-ribose) polymerase (PARP) cleavage were investigated by immunoblot analysis with the corresponding antibodies. As shown in Figure 3B, the distinct increase of caspase-3 and PARP cleavages, typical apoptosis markers, was not observed after treatment with 50 μM physcion for 24 h, suggesting that physcion might not induce apoptosis at the highest tested concentration (50 μM) in SK-N-BE(2)-C cells. 2.3. Isolation and Sequence Analysis of the 5′-Flanking Region of the hST8Sia VI Gene Using the NCBI database, sequence comparison of the cDNA and the genomic DNA of the hST8Sia VI gene showed that the predicted transcription start site was located at 136 bp upstream of the ATG start codon. Based on this result, a 2660 bp region upstream of the transcription start site for the hST8Sia VI gene was obtained by LA-PCR and sequenced. By using a transcription factor binding site-prediction program, putative transcription factor binding sites (TFBS) within this region were screened and 52 TFBS were predicted (similarity margin ≥95%). 2.4. Promoter Analysis of the 5′-Flanking Region of the hST8Sia VI Gene in Physcion-Induced SK-N-BE(2)-C Cells To assess whether or not the 5′-flanking region of the hST8Sia VI gene includes a physcion-responsive promoter, six kinds of luciferase reporter plasmids (pGL3-2660, pGL3-1982, pGL3-1503, pGL3-968, pGL3-574, and pGL3-240) were constructed by inserting truncated fragments of the 5′-flanking region of the hST8Sia VI gene into the pGL3-Basic vector for analysis of luciferase activity. To identify physcion-responsive promoter activity, we transfected pGL3-Basic plasmid as a negative control along with the six reporter plasmids into physcion-untreated SK-N-BE(2)-C cells and checked the hST8Sia VI promoter activity by physcion induction. As shown in Figure 4A, all deletion plasmid constructs containing 5′-flanking regions between −2660 and −574 exhibited about two-fold higher promoter activity in physcion-treated cells than in physcion-untreated cells. However, when the 5′-flanking region was deleted up to −240, promoter activity by physcion induction was remarkably decreased. These results clearly indicate that the region between nucleotides −574 and −240 was crucial for the transcriptional activity of the hST8Sia VI gene in physcion-stimulated SK-N-BE(2)-C cells and that physcion-responsive elements exists within nucleotide −574 to −240 in the hST8Sia VI promoter. Based on this result, to identify the minimal physcion-responsive region controlling the maximal promoter activity of the hST8Sia VI gene in SK-N-BE(2)-C cells, we constructed two different kinds of reporter plasmids (pGL3-1982/-320 to pGL3-1982/-1066 and pGL3-1503/-320 to pGL3-1503/-1066) containing progressive deletions from the 3′ end of the hST8Sia VI gene promoter and transfected them into physcion-untreated SK-N-BE(2)-C cells, then checked for promoter activity by physcion stimulation. As shown in Figure 4B, deletion in the region from +1 to −320 remarkably reduced the promoter activity in SK-N-BE(2)-C cells with or without physcion induction. Collectively, these results apparently suggest that the physcion-responsive core promoter region for the transcription of the hST8Sia VI gene in SK-N-BE(2)-C cells is located between −320 and −240. 2.5. Identification of Physcion-Responsive Element in the Functional −320/−240 Region of hST8Sia VI Promoter To identify physcion-responsive elements in the nucleotide −320 to −240 region of the hST8Sia VI gene, we searched for TFBS within the nucleotide −320 to −240 region using bioinformatics analysis. Sequence analysis using the PROMO program [20,21] showed potential NF-Y (−289 to −282) and Pax-5 (−262 to −256) binding sites in this region (Figure 5A). To define whether these binding sites play a critical role in hST8sia VI expression by physcion induction in SK-N-BE(2)-C cells, two mutant plasmids (pGL3-Pax-5 mut and pGL3-NF-Y mut) were constructed (Figure 5B) and transfected into physcion-uninduced SK-N-BE(2)-C cells. Afterward, luciferase assays by physcion induction were performed. As shown in Figure 5B, the transcriptional activity of pGL3-Pax-5mut was markedly decreased to more than five-fold that of pGL3-574, whereas the activity of pGL3-NF-Ymut was slightly reduced. This result indicates that the Pax-5 binding site located at position −262 to −256 is crucial for the physcion-induced expression of the hST8Sia VI gene, and suggests that Pax-5 binding to this site is essential for the induction of hST8Sia VI gene expression by physcion stimulation. Next, we further evaluated whether Pax-5 can bind to this binding site in the hST8SIa VI promoter using chromatin immunoprecipitation (ChIP) assay, which examined the in vivo association of Pax-5 with the promoter region between positions −262 and −256 of the hST8STSia VI gene in physcion-treated SK-N-BE(2)-C cells. The specific PCR products obtained with input DNA are shown in Figure 5C. Although Pax-5-specific amplification by PCR was observed in physcion-untreated SK-N-BE(2)-C cells, physcion treatment resulted in a significant increase in Pax-5 binding in SK-N-BE(2)-C cells. On the other hand, a control assay using IgG antibody did not generate the PCR product in the absence or presence of physcion. These results indicate that the hST8Sia VI gene expression in physcion-induced SK-N-BE(2)-C cells is upregulated by the direct binding of Pax-5 to its site on the hST8Sia VI promoter region. 2.6. Transcriptional Activation of hST8Sia VI via ERK and p-38 MAPK Pathways in Physcion-Stimulated SK-N-BE(2)-C Cells To uncover the mechanism underlying the transcriptional activation of hST8Sia VI gene by physcion induction, we examined the MAPK signal pathway in response to physcion stimulation in SK-N-BE(2)-C cells. As shown in Figure 6A, the expression of phosphorylated ERK and p38 MAPK were clearly enhanced in a time-dependent fashion in physcion-induced SK-N-BE(2)-C cells, whereas JNK phosphorylation was slightly increased by physcion induction. This result suggests that ERK and p38 MAPK, including JNK, are activated by physcion induction in SK-N-BE(2)-C cells. To further assess whether physcion-stimulated transcriptional activity of a pGL3-574 promoter is induced via ERK and p38 MAPK signal pathways, we checked the promoter activity of pGL3-574 by physcion induction in SK-N-BE(2)-C cells pre-treated with MAPK inhibitors. As shown in Figure 6B, the promoter activity of pGL3-574 in SK-N-BE(2)-C cells induced by physcion was significantly decreased by ERK inhibitor U0126 and p38 MAPK inhibitor SB203580, but not by JNK inhibitor SP600125. Taken together, these results suggest that transcriptional activation of hST8Sia VI in physcion-stimulated SK-N-BE(2)-C cells is regulated through ERK and p38 MAPK pathways. 3. Discussion It was previously reported that hST8Sia VI acting on Neu5Acα2-3Galβ1-3GalNAcα1-O-Ser/Thr or Neu5Acα2-6GalNAcα1-O-Ser/Thr in O-glycans was highly expressed in MCF-7 breast cancer cell line, whereas its expression was very low in MDA-MB-231 and T47-D breast cancer cell lines and in HT-29 and Caco-2 colon cancer cell lines [22]. However, nothing is known about the transcriptional regulation hST8Sia VI gene. This study presents the first report on the transcriptional regulation hST8Sia VI gene. Here, we have demonstrated for the first time that hST8Sia VI gene expression was upregulated in SK-N-BE(2)-C cells stimulated with physcion and that the Pax-5 binding site at position −262 to −256 plays a pivotal role in the transcriptional activation of hST8Sia VI by physcion induction. Physcion, an anthraquinone derivative, has been reported to have a variety of pharmacological effects such as hepatoprotective [14], anti-microbial [15], anti-inflammatory [16], anti-cancer [13,17,18], and anti-metastatic activities [19]. However, the effect of physcion on gene expression, including sialyltransferase, has not been studied before. This study also represents the first report on the effect of physcion on gene expression and the underlying mechanism. Our results indicated that physcion treatment in SK-N-BE(2)-C cells enhanced hST8Sia VI mRNA levels. In this study, we isolated the 2.6 kb 5′-flanking region upstream of the transcription start site of the hST8Sia VI gene. While a number of putative regulatory cis-acting elements were found in this region, typical TATA and CCAAT boxes were not existed. Moreover, we demonstrated that a functional promoter in response to physcion was located at the 5′-flanking region (pGL3-2660) of the hST8Sia VI gene in SK-N-BE(2)-C cells. Our present result clarified that the region between −574 and −240 is the core promoter essential for transcriptional activation of hST8Sia VI gene in physcion-induced SK-N-BE(2)-C cells, as evidenced by deletion analysis. This region contains NF-Y (−289 to −282) and Pax-5 (−262 to −256) binding sites. By site-directed mutagenesis experiments and ChIP assay, we also verified that the Pax-5 binding site in this region plays a pivotal role in the transcriptional activation of the hST8Sia VI gene in physcion-stimulated SK-N-BE(2)-C cells. Pax-5 is well known as the B cell-specific transcription factor that plays a crucial role in the development and malignancy of B cells [23,24]. Pax5 regulates B lymphopoiesis through the activation of B cell-specific gene expression and concurrent repression of the expression of non-B cell genes [23]. It was documented that Pax-5 is mainly expressed at the midbrain-hindbrain boundary during mouse development, suggesting that Pax-5 is associated with the development of the central nervous system [25]. Pax-5 expression has also been reported in various cancer cells, including medulloblastoma [26], lymphoma [27], small-cell lung carcinoma [28], and neuroblastoma [29,30]. Previous studies have shown that SK-N-BE(2)-C cells expressed lower levels of Pax-5 [29], and downregulation of Pax-5 in SK-N-BE(2)-C cells resulted in significant reduction in the cell proliferation rate [30]. It was recently reported that stimulation of mouse primary B cells by lipopolysaccharide (LPS)-induced T-independent (TI) or IL4/IL5/CD40L-induced T-dependent (TD) pathway resulted in remarkable elevation of the mouse ST8Sia VI gene expression in mature and plasma cell populations as compared with ST8Sia II and ST8Sia IV [31]. This finding suggests transcriptional upregulation of ST8Sia VI during the differentiation of plasma cells by TI and TD stimulation. Considering that Pax-5 plays a role in mature B-cell differentiation [23], this finding also suggests the possibility that Pax-5 may be related to the transcriptional upregulation of ST8Sia VI, although it remains to be elucidated whether a Pax-5 binding site exists in the core region of the mouse ST8Sia VI promoter. Our data show that ERK, JNK and p38 MAPK activation were induced by physcion in SK-N-BE(2)-C cells, although physcion-mediated MAPK signaling pathways are not known at present. The present result also indicates that transcriptional activation of hST8Sia VI is mediated through ERK and p38 MAPK pathways induced by physcion in SK-N-BE(2)-C cells, as demonstrated by luciferase assay using cells treated with chemical inhibitors that selectively or specifically block these pathways. 4. Experimental Section 4.1. Extraction, Isolation and Structure Determination of Physcion Extraction and isolation of physcion from the dried root of Polygonum multiflorum were carried out by the same procedures as those described previously [32]. The structure of the purified physcion by 1H and 13C NMR spectroscopy with a Varian Spectrometer (Palo Alto, CA, USA) at 600 MHz with CDCl3 was determined. The gas chromatography-mass spectrometry (GC-MS) profiles of the purified physcion were recorded as described previously [32]. 1H NMR and 13C NMR results were compared with the data presented in the references [33,34]. 4.2. Cell Cultures Human neuroblastoma SK-N-BE(2)-C cells were obtained from American Type Culture Collection (Manassas, VA, USA) and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 100 μg/mL of streptomycin, 100 U/mL of penicillin, and 10% fetal bovine serum (WelGENE Co., Daegu, Korea) at 37 °C in a 5% CO2 incubator. 4.3. Cell Viability Assay Cell viability assay was performed as described previously [7,8,9,10,11]. The amount of formazan produced was quantified by measuring the absorbance at 490 nm with an ELISA plate reader (Bio-Rad, Hercules, CA, USA). 4.4. Reverse Transcription-Polymerase Chain Reaction (RT-PCR) Total RNA isolation and first-strand cDNA synthesis were performed as described previously [35]. PCR amplification was performed with a PC-818A Program Temp Control System (Astec, Fukuoka, Japan), with 1 cycle for 5 min at 95 °C and 30 cycles consisting of denaturation at 95 °C for 40 s, annealing at 60 °C for 40 s, and extension for 45 s at 72 °C, followed by incubation at 72 °C for 5 min. PCR products were analyzed by 1% agarose gel electrophoresis. Quantitation of the intensity of the amplified bands was performed using a Scion Image Instrument (Scion Corp.; Frederick, MD, USA). 4.5. Cloning of the 5′-Flanking Region of the hST8Sia VI Gene and Bioinformatics Analysis Using the sequence information for hST8Sia VI cDNA (GenBank accession number BC137102) and Homo sapiens chromosome 10, GRCh38.p2 Primary Assembly (GenBank accession number NC_000010 GPC_000001302) from the National Center for Biotechnology Information (NCBI), a 2660 bp fragment from the 5′-flanking sequence of the hST8Sia VI gene was amplified by long and accurate PCR (LA-PCR) with LA-Taq polymerase (Takara Bio, Shiga, Japan). LA-PCR was performed with the sense primer P-2660S containing a KpnI site and the antisense primer P-2660A containing a XhoI site (Table 1). Human genomic DNA as a template was isolated from SK-N-BE(2)-C cells. The LA-PCR condition was 1 cycle for 1 min at 95 °C and then 30 cycles of 98 °C for 20 s and 68 °C for 3 min, with a final extension for 10 min at 72 °C. The amplified PCR products were subcloned into pGEM-T Easy vector (Promega, Madison, WI, USA) to produce pGST8Sia VI. PCR products were sequenced in both directions by cloning convenient restriction fragments into the pUC118/9 vector. Putative binding sites for the transcription factors of the 5′-flanking region were analyzed using ALGGEN-PROMO search algorithm (http://alggen.lsi.upc.es/) based on TRANSFAC version 8.3 [20,21] with a maximum matrix dissimilarity rate of 5. 4.6. Construction of Luciferase Reporter Plasmids and Mutagenesis To determine the physcion-activated minimal promoter region of the hST8Sia VI gene, three kinds of luciferase reporter plasmids (pGL3-240 to pGL3-2660, pGL3-1982/-320 to pGL3-1982/-1066, and pGL3-1503/-320 to pGL3-1503/-1066), were constructed by LA-PCR with sense and antisense primers containing KpnI and XhoI sites, respectively (Table 1), using the pGST8Sia VI described above as the template. The PCR fragments were subcloned into pGEM-T Easy vector and sequenced. Each fragment produced by digestion with KpnI and XhoI was introduced into the corresponding sites of the pGL3-Basic vector used as a negative control. Mutations with base substitution and deletion at the Pax-5 and NF-Y binding sites, respectively, were generated using a QuikChange® II XL site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA) according to the manufacturer’s protocol using oligonucleotide primers (Table 1). The presence of mutation was verified by DNA sequencing. 4.7. Transfection and Luciferase Assay Transient transfection and luciferase assay were carried out as described previously [7,8,9], 0.5 μg of luciferase reporter plasmid and 50 ng of pRL-TK as the control Renilla luciferase vector (PRomega; Madison, WI, USA) were co-transfected into cells using 1 μL Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After 4 h culture, the transfection medium was replaced by normal medium without physcion and incubated for 7 h. Then, the medium was changed to medium containing 40 μM physcion (Sigma, St. Louis, MO, USA) and incubated for 24 h. Cells were harvested and assayed using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) and a GloMax™ 20/20 luminometer (Promega). 4.8. Western Blot Analysis Western blot analysis was conducted as described previously [35]. Thirty-five microgram samples of total cell lysates prepared using RIPA buffer were subjected to SDS-PAGE and transferred to PVDF membranes. The blotted membranes were incubated with primary and secondary antibodies. Blots were detected using the ECL chemiluminescence system (GE Healthcare, Piscataway, NJ, USA). The following primary antibodies were used: p-ERK, ERK (Santa Cruz, CA, USA), p-p38, p38 (Cell Signaling Technology, Beverly, MA, USA) and GAPDH (Millipore, Billerica, MA, USA). Horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from Enzo Life Science (Farmingdale, NY, USA). 4.9. Chromatin Immunoprecipitation (ChIP) Assay ChIP assay was performed using the ChIP kit (Upstate Biotechonology, Lake Placid, NY, USA) according to the manufacture’s protocol. Formaldehyde cross-linking of cells and shearing of chromatin were carried out as described previously [9,35]. Immunoprecipitation was conducted with Pax-5 (A-11, Santa Cruz, CA, USA) and IgG antibodies (Sigma; St. Louis, MO, USA). PCR analysis was conducted with primers flanking the Pax-5 binding site on the hST8Sia VI promoter (Table 1) and the purified ChIP DNA or input DNA. The PCR condition was 1 cycle at 94 °C for 3 min, followed by 32 cycles with a denaturing step at 94 °C for 20 s, an annealing step of 30 s at specific annealing temperatures for 59 °C, an elongation step at 72 °C for 30 s and a final extension step at 72 °C for 2 min. 5. Conclusions In the present study, we have demonstrated for the first time that hST8Sia VI gene expression was upregulated in SK-N-BE(2)-C cells stimulated with physcion and that the region between −320 and −240, which contains putative binding sites for Pax-5 and NF-Y in the hST8Sia VI promoter, acts as the major promoter for transcriptional activation of hST8Sia VI in physcion-induced SK-N-BE(2)-C cells. Furthermore, the Pax-5 binding site at position −262 to −256 plays a crucial role in the transcriptional activation of hST8Sia VI by physcion induction, as demonstrated by mutagenesis and ChIP assay. Our data suggest that physcion-induced transcriptional activation of hST8Sia VI is mediated through ERK and p38 MAPK pathways in SK-N-BE(2)-C cells. Acknowledgments This work was supported by the R&D program of MOTIE/KIAT (N0000697, Establishment of Infra Structure for Anti-aging Industry Support). Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1246/s1. Click here for additional data file. Author Contributions Conceived and designed the experiments: Hyun-Kyoung Yoon, Hyun-Kyu An, Kyoung-Sook Kim, Young Whan Choi, and Young-Choon Lee; Performed the experiments: Hyun-Kyoung Yoon, Hyun-Kyu An, Kyoung-Sook Kim, and Min Jung Ko; Analyzed the data: Hyun-Kyoung Yoon, Hyun-Kyu An, Kyoung-Sook Kim, Dong-Hyun Kim, Young Whan Choi, and Young-Choon Lee; Contributed reagents/materials/analysis tools: Seo-Won Mun, Dong-Hyun Kim, Young Whan Choi, Cheol Min Kim, and Cheorl-Ho Kim; Wrote the paper: Hyun-Kyoung Yoon, Hyun-Kyu An, Young Whan Choi, and Young-Choon Lee. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Purification chart (A); chemical structure (B); and key heteronuclear multiple bond correlation (HMBC) (C) of physcion from the ethyl acetate (EtOAc) extract of Polygonum multiflorum by silica gel column chromatography. CRPE is code name indicating ethyl acetate extract from Polygonum multiflorum. Physcion (CRPE5) and emodin (CRPE19) were finally isolated from CRPE4-6 and CRPE19-21 fractions, respectively. Figure 2 Effect of physcion on hST8Sia VI mRNA level in SK-N-BE(2)-C cells. After 24 h treatment with various concentration of physcion in the panel up, total RNA from SK-N-BE(2)-C cells was isolated. RT-PCR was used to detect hST8Sia VI mRNA level. In the panel below, the levels of each RT-PCR product were analyzed by densitometry and normalized by β-actin mRNA level as a control. Data reveal the relative values ± SD of three independent experiments. Figure 3 Effect of physcion on the viability and apoptosis of SK-N-BE(2)-C cells. (A) The cytotoxic effects of physcion on SK-N-BE(2)-C cells were examined using MTT assay. After 24 h incubation at various concentration of physcion, and cell viability was analyzed at 540 nm using an enzyme-linked immunosorbent assay (ELISA) plate reader (Bio-Rad, Hercules, CA, USA). The results represent the means ± SEM of three independent experiments. * p < 0.05 (compared with control); *** p < 0.001; (B) PARP-1/2 and caspase-3 were checked by Western blotting analysis. Figure 4 Deletion analysis of hST8Sia VI promoter in SK-N-BE(2)-C cells with physcion treatment. Reporter plasmids contain progressive deletions from the 5′ end (A) and the 3′ end (B) of the hST8Sia VI gene promoter. After co-transfection of each construct into SK-N-BE(2)-C cells with pRL-TK, relative firefly luciferase (Luc) activity was assessed. The pRL-TK-derived Renilla luciferase activity was used to normalize all firefly luciferase activity. Data are the means ± SD of triplicate measurements and are representative of three independent experiments. Figure 5 Promoter mutation assay for the transcription factor-binding sites in the nucleotide −320 to −240 region of the hST8Sia VI gene in physcion-induced SK-N-BE(2)-C cells. (A) Nucleotide sequences of the promoter region from −320 to −240 are shown; (B) After co-transfection of each construct into SK-N-BE(2)-C cells with pRL-TK, relative firefly luciferase activity was assessed. The Renilla luciferase activity derived from pRL-TK was used to normalize all firefly luciferase activity. Data are the means ± SD of triplicate measurements and are representative of three independent experiments. The circle and square indicate NF-Y and Pax-5 binding sites, respectively. The white and black colors indicate wild-type and mutant forms of these sites, respectively; (C) ChIP assay was conducted with primers amplifying the −379 and −131 region of the hST8Sia VI promoter and DNA isolated from chromatin immunoprecipitated with either anti-Pax-5 antibody or control IgG from SK-N-BE(2)-C cells treated for 24 h with or without 40 μM physcion. Figure 6 Physcion induced transcriptional activation of hST8Sia VI through ERK/p-38 MAPK signal pathways in SK-N-BE(2)-C cells. For representative immunoblot analysis of p-ERK, ERK, p-JNK, JNK, p-p38 and p38 protein expression levels, cells were treated with 40 μM physcion for different lengths of time (A). pGL3-574 or pGL3-Basic was co-transfected into SK-N-BE(2)-C cells with pRL-TK. The transfected cells were incubated for 24 h in the presence of 40 μM physcion with different inhibitors (10 μM U0126, 10 μM SP600125, and 20 μM SB203580); The Renilla luciferase activity derived from pRL-TK was used to normalize all firefly luciferase activity. Data are the means ± SD of triplicate measurements and are representative of three independent experiments (B). +, treatment; -, untreatment. ijms-17-01246-t001_Table 1Table 1 Primer sequences used for reverse transcription-polymerase chain reaction (RT-PCR) and deletion constructs, site-directed mutagenesis and chromatin immunoprecipitation (ChIP). Primer Sequence Strand Purpose hST8Sia VI 5′-CCCTATTTCTGGAGGACATTGCAACCTA-3′ Sense RT-PCR hST8Sia VI 5′-GTTGGAGGATCTGGCTGTATTCTTTG-3′ Antisense RT-PCR β-actin 5′-CAAGAGATGGCCACGGCTGCT-3′ Sense RT-PCR β-actin 5′-TCCTTCTGCATCCTGTCGGCA-3′ Antisense RT-PCR GAPDH 5′-AGCCTCAAGATCATCAGCAATGTCCT-3′ Sense RT-PCR GAPDH 5′-AAATTCGTTGTCATACCAGGAAATGAG-3′ Antisense RT-PCR P-2660 5′-ATGGTACCCTTCTGCTGTTGCCTTGAGCCCAGC-3′ Sense Deletion P-2660 5′-ATCTCGAGACAGCGTTCACAGGCGGCAGCGAG-3′ Antisense Deletion P-1982 5′-ATGGTACCGGCTGTCTGGCCTGGTTGCTCCCA-3′ Sense Deletion P-1503 5′-ATGGTACCAAGGATACCATAGGCTGGGTGACCG-3′ Sense Deletion P-968 5′-ATGGTACCAGGCTGCCTTGTGGGGCCTGGTATA-3′ Sense Deletion P-574 5′-ATGGTACCGCCCCTCATACCAGTTCGCTGTCCC-3′ Sense Deletion P-240 5′-ATGGTACCCGCGCGGCGGCGGCGGCAGCAGC-3′ Sense Deletion P-320A 5′-ATCTCGAGTTCTGCGCCCTCGCCTCGTCCCGA-3′ Antisense Deletion P-630A 5′-ATCTCGAGCCTGGAGACCCGTTTAGCCCCTG-3′ Antisense Deletion P-1066A 5′-ATCTCGAGGGGTGGACCTCATGGACCTCCTC-3′ Antisense Deletion Pax-5 Mut 5′-GGAGTTGAGCTTCCGCATTCCAACCTTCAGGTGACC-3′ Sense Mutagenesis Pax-5 Mut 5′-GGTCACCTGAAGGTTGGAATGCGGAAGCTCAACTCC-3′ Antisense Mutagenesis NF-Y Mut 5′-GCAGAAAACTTGGAGCAATCAGCACGGAGTTGAGC-3′ Sense Mutagenesis NF-Y Mut 5′-GCTCAACTCCGTGCTGATTGCTCCAAGTTTTCTGC-3′ Antisense Mutagenesis hST8Sia VI 5′-AGGCAGAGTTGTGGTGTGGC-3′ Sense ChIP hST8Sia VI 5′-TGGCAGATGACGATTCGCCGA-3′ Antisense ChIP Primers P-2660S to P-240S were used for construction of the deletion mutants. KpnI and XhoI sites in sense and antisense primers, respectively, are underlined. The mutated nucleotides in the oligonucleotides for mutation are in boldface and italic type. ==== Refs References 1. Schauer A. Sialic acids as regulators of molecular and cellular interactions Curr. Opin. Struct. Biol. 2009 19 507 514 10.1016/j.sbi.2009.06.003 19699080 2. Harduin-Lepers A. Mollicone R. Delannoy P. Oriol R. The animal sialyltransferases and sialyltransferase-related genes: A phylogenetic approach Glycobiology 2005 15 805 817 10.1093/glycob/cwi063 15843597 3. Paulson J.C. Weistein J. Schauer A. Tissue-specific expression of sialyltransferases J. Biol. Chem. 1989 264 10931 10934 2738054 4. Paulson J.C. Colley K.J. Glycosyltransferases: Structure, localization and control of cell type-specific glycosylation J. Biol. Chem. 1989 264 17617 17618 5. Harduin-Lepers A. Vallejo-Ruiz V. Krzewinski-Recchi M.A. Samyn-Petit B. Julien S. Delannoy P. The human sialyltransferase family Biochimie 2001 83 727 737 10.1016/S0300-9084(01)01301-3 11530204 6. Takashima S. Characterization of mouse sialyltransferase genes: Their evolution and diversity Biosci. Biotechnol. Biochem. 2008 72 1155 1167 10.1271/bbb.80025 18460788 7. Kwon H.Y. Kang N.Y. Dae H.M. Kim K.S. Kim C.H. Do S.I. Lee Y.C. Valproic acid-mediated transcriptional regulation of human GM3 synthase (hST3Gal V) in SK-N-BE(2)-C human neuroblastoma cells Acta Pharmacol. Sin. 2008 29 999 1005 10.1111/j.1745-7254.2008.00847.x 18718167 8. Kwon H.Y. Dae H.M. Song N.R. Kim K.S. Kim C.H. Lee Y.C. Valproic acid induces transcriptional activation of human GD3 synthase (hST8Sia I) in SK-N-BE(2)-C human neuroblastoma cells Mol. Cells 2009 27 113 118 10.1007/s10059-009-0012-4 19214441 9. Baik J.S. Kim K.S. Moon H.I. An H.K. Park S.J. Kim C.H. Lee Y.C. Cordycepin-mediated transcriptional regulation of human GD3 synthase (hST8Sia I) in human neuroblastoma SK-N-BE(2)-C cells Acta Biochim. Biophys. Sin. 2014 46 65 71 10.1093/abbs/gmt122 24225218 10. Kwon H.Y. Kim S.J. Kim C.H. Son S.W. Kim K.S. Lee J.H. Do S.I. Lee Y.C. Triptolide downregulates human GD3 synthase (hST8Sia I) gene expression in SK-MEL-2 human melanoma cells Exp. Mol. Med. 2010 42 849 855 10.3858/emm.2010.42.12.088 21072003 11. Song N.R. Kim S.J. Kwon H.Y. Son S.W. Kim K.S. An H.B. Lee Y.C. Transcriptional activation of human GM3 synthase (hST3Gal V) gene by valproic acid in ARPE-19 human retinal pigment epithelial cells BMB Rep. 2011 44 405 409 10.5483/BMBRep.2011.44.6.405 21699754 12. Huang Q. Lu G. Shen H.M. Chung M.C. Ong C.N. Anti-cancer properties of anthraquinones from rhubarb Med. Res. Rev. 2007 27 609 630 10.1002/med.20094 17022020 13. Wijesekara I. Zhang C. Ta Q.V. Vo T.S. Li Y.X. Kim S.K. Physcion from marine-derived fungus Microsporum sp. Induces apoptosis in human cervical carcinoma HeLa cells Microbiol. Res. 2014 169 255 261 10.1016/j.micres.2013.09.001 24071573 14. Zhao Y.L. Wang J.B. Zhou G.D. Shan L.M. Xiao X.H. Investigations of free anthraquinones from rhubarb against alpha-naphthylisothiocyanate-induced cholestatic liver injury in rats Basic Clin. Pharmacol. Toxicol. 2009 104 463 469 10.1111/j.1742-7843.2009.00389.x 19389047 15. Tamokou Jde D. Tala M.F. Wabo H.K. Kuiate J.R. Tane P. Antimicrobial activities of methanol extract and compounds from stem bark of Vismia rubescens J. Ethnopharmacol. 2009 124 571 575 10.1016/j.jep.2009.04.062 19464353 16. Ghosh S. Das Sarma M. Patra A. Hazra B. Anti-inflammatory and anticancer compounds isolated from Ventilago madraspatana Gaertn., Rubia cordifolia Linn. and Lantana camara Linn. J. Pharm. Pharmacol. 2010 62 1158 1166 10.1111/j.2042-7158.2010.01151.x 20796195 17. Hong J.Y. Chung H.J. Bae S.Y. Trung T.N. Bae K. Lee S.K. Induction of cell cycle arrest and apoptosis by physcion, an anthraquinone isolated from rhubarb (rhizomes of Rheum tanguticum ), in MDA-MB-231 human breast cancer cells J. Cancer Prev. 2014 19 273 278 10.15430/JCP.2014.19.4.273 25574462 18. Chen X. Gao H. Han Y. Ye J. Xie J. Wang C. Physcion induces mitochondria-driven apoptosis in colorectal cancer cells via downregulating EMMPRIN Eur. J. Pharmacol. 2015 764 124 133 10.1016/j.ejphar.2015.07.008 26144377 19. Han Y.T. Chen X.H. Gao H. Ye J.L. Wang C.B. Physcion inhibits the metastatic potential of human colorectal cancer SW620 cells in vitro by suppressing the transcription factor SOX2 Acta Pharmacol. Sin. 2016 37 264 275 10.1038/aps.2015.115 26707141 20. Messeguer X. Escudero R. Farre D. Nunez O. Martinez J. Alba M.M. PROMO: Detection of known transcription regulatory elements using species-tailored searches Bioinformatics 2002 18 333 334 10.1093/bioinformatics/18.2.333 11847087 21. Farré D. Roset R. Huerta M. Adsuara J.E. Roselló L. Albà M.M. Messeguer X. Identification of patterns in biological sequences at the ALGGEN server: PROMO and MALGEN Nucleic Acids Res. 2003 31 3651 3653 10.1093/nar/gkg605 12824386 22. Teintenier-Lelievre M. Julien S. Juliant S. Guerardel Y. Duonor-Cerutti M. Delannoy P. Harduin-Lepers A. Molecular cloning and expression of a human hST8Sia VI (α2,8-sialyltransferase) responsible for the synthesis of the diSia motif on O-glycosylproteins Biochem. J. 2005 393 665 674 10.1042/BJ20051120 16120058 23. Holmes M.L. Pridans C. Nutt S.L. The regulation of the B-cell gene expression programme by Pax5 Immunol. Cell Biol. 2008 86 47 53 10.1038/sj.icb.7100134 17998914 24. O’Brien P. Morin P. Jr. Ouellette R.J. Robichaud G.A. The Pax-5 gene: A pluripotent regulator of B-cell differentiation and cancer disease Cancer Res. 2011 71 7345 7350 10.1158/0008-5472.CAN-11-1874 22127921 25. Asano M. Gruss P. Pax-5 is expressed at the midbrain-hindbrain boundary during mouse development Mech. Dev. 1992 39 29 39 10.1016/0925-4773(92)90023-D 1283313 26. Kozmik Z. Sure U. Ruedi D. Busslinger M. Aguzzi A. Deregulated expression of PAX5 in medulloblastoma Proc. Natl. Acad. Sci. USA 1995 92 5709 5713 10.1073/pnas.92.12.5709 7777574 27. Balasenthil S. Gururaj A.E. TalUKder A.H. Bagheri-Yarmand R. Arrington T. Haas B.J. Braisted J.C. Kim I. Lee N.H. Kumar R. Identification of Pax5 as a target of MTA1 in B-cell lymphomas Cancer Res. 2007 67 7132 7138 10.1158/0008-5472.CAN-07-0750 17671180 28. Kanteti R. Nallasura V. Loganathan S. Tretiakova M. Kroll T. Krishnaswamy S. Faoro L. Cagle P. HUSAin A.N. Vokes E.E. PAX5 is expressed in small-cell lung cancer and positively regulates c-Met transcription Lab. Investig. 2009 89 301 314 10.1038/labinvest.2008.168 19139719 29. Lagergren A. Manetopoulos C. Axelson H. Sigvardsson M. Neuroblastoma and pre-B lymphoma cells share expression of key transcription factors but display tissue restricted target gene expression BMC Cancer 2004 4 1246 10.1186/1471-2407-4-80 15544702 30. Baumann Kubetzko F.B. Di Paolo C. Maag C. Meier R. Schafer B.W. Betts D.R. Stahel R.A. Himmelmann A. The PAX5 oncogene is expressed in N-type neuroblastoma cells and increases tumorigenicity of a S-type cell line Carcinogenesis 2004 25 1839 1846 10.1093/carcin/bgh190 15155532 31. Wang S.H. Tsai C.M. Lin K.I. Khoo K.H. Advanced mass spectRometry and chemical analyses reveal the presence of terminal disialyl motif on mouse B-cell glycoproteins Glycobiology 2013 23 677 689 10.1093/glycob/cwt008 23363740 32. Park S.J. Jin M.L. An H.K. Kim K.S. Ko M.J. Kim C.M. Choi Y.W. Lee Y.C. Emodin induces neurite outgrowth through PI3K/Akt/GSK-3β-mediated signaling pathways in Neuro2a cells Neurosci. Lett. 2015 588 101 107 10.1016/j.neulet.2015.01.001 25562207 33. Chu X. Sun A. Liu R. Preparative isolation and purification of five compounds from the Chinese medicinal herb Polygonum cuspidatum Sieb. et Zucc by high-speed counter-current chromatography J. Chromatogr. A 2005 1097 33 39 10.1016/j.chroma.2005.08.008 16298183 34. Liu R. Li A. Sun A. Preparative isolation and purification of hydroxyanthraquinones and cinnamic acid from the Chinese medicinal herb Rheum officinale Baill. by high-speed counter-current chromatography J. Chromatogr. A 2004 1052 217 221 10.1016/j.chroma.2004.08.101 15527141 35. Yoon H.K. Lee J.W. Kim K.S. Mun S.W. Kim D.H. Kim H.J. Kim C.H. Lee Y.C. Serum deprivation-induced human GM3 synthase (hST3Gal V) gene expression is mediated by Runx2 in human osteoblastic MG-63 cells Int. J. Mol. Sci. 2015 17 35 10.3390/ijms17010035 26729095

PMC005xxxxxx/PMC5000645.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081247ijms-17-01247ArticleConformation-Independent QSPR Approach for the Soil Sorption Coefficient of Heterogeneous Compounds Aranda José F. 1Garro Martinez Juan C. 2Castro Eduardo A. 1Duchowicz Pablo R. 1*Bacchus Marie-Christine Academic Editor1 Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas (INIFTA), CONICET, UNLP, Diag. 113 y 64, Sucursal 4, C.C. 16, La Plata 1900, Argentina; [email protected] (J.F.A.); [email protected] (E.A.C.)2 Instituto Multidisciplinario de Investigaciones Biológicas IMIBIO-SL (CCT San Luis), Departamento de Química, Universidad Nacional de San Luis, Chacabuco 917, San Luis 5700, Argentina; [email protected]* Correspondence: [email protected]; Tel.: +54-221-425-7430; Fax: +54-221-425-464203 8 2016 8 2016 17 8 124730 4 2016 22 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).We predict the soil sorption coefficient for a heterogeneous set of 643 organic non-ionic compounds by means of Quantitative Structure-Property Relationships (QSPR). A conformation-independent representation of the chemical structure is established. The 17,538 molecular descriptors derived with PaDEL and EPI Suite softwares are simultaneously analyzed through linear regressions obtained with the Replacement Method variable subset selection technique. The best predictive three-descriptors QSPR is developed on a reduced training set of 93 chemicals, having an acceptable predictive capability on 550 test set compounds. We also establish a model with a single optimal descriptor derived from CORAL freeware. The present approach compares fairly well with a previously reported one that uses Dragon descriptors. Quantitative Structure-Property RelationshipsReplacement Methodsoil sorption coefficientPharmaceutical Data Exploration Laboratory softwareEstimation Program Interface Suite softwareCorrelation and Logic software ==== Body 1. Introduction The soil sorption coefficient (Koc) describes the biodegradation and pollution impact of organic pesticides [1] when these compounds interact with the organic matter of soils and sediments either on surface, ground or drinking water [2]. The reliable estimation of the Koc parameter is very important in agriculture, as its experimental measurement is difficult, expensive and time-consuming. Predicting the soil sorption coefficient for a wide number of chemical structures is very convenient in risk assessment [3]. In the realm of the Quantitative Structure-Property Relationships (QSPR) theory [4,5,6], an experimental property of a chemical compound, i.e., Koc, can be predicted through the knowledge of its chemical structure. The structure is quantified by means of a set of suitable molecular descriptors, in other words, numerical quantities carrying specific information on the constitutional, topological, geometrical, hydrophobic, and/or electronic aspects [7,8,9]. Therefore, a set of descriptors is then statistically correlated with the experimental property, resulting in a mathematical model that can be used with find out useful parallelisms. It is known that many published QSPR models that predict the soil sorption coefficient involve the experimental octanol/water partition coefficient (Kow) or the water solubility (Sw) [10], while other QSPR are based on theoretical molecular descriptors [11,12,13]. However, usually, little work is done to examine the model’s predictivity (validation) and the chemical domain of application over a wide range of compounds, especially for new chemicals [14,15,16]. A previous QSPR study of Gramatica et al. [14] on a highly heterogeneous set of 643 organic non-ionic compounds predicts the soil sorption coefficient expressed in logarithmic units (logKoc). The training set with 93 compounds used in such work is peculiar, because it is much smaller than the test set of 550 compounds (1:6 ratio). The best Dragon molecular descriptors are selected through the Genetic Algorithms (GA) technique based on Multivariable Linear Regression analysis (MLR), leading to a four-dimensional QSPR having a predictivity of 78% on the test set. The best predicted data are obtained by consensus modeling from ten different models in the GA model population. In this work, we report new alternative QSPR models for the soil sorption coefficient in the same molecular set studied by Gramatica et al. [14], using an approach that does not consider the conformational representation of the chemical structure by only relying on the constitutional and topological aspects of the molecules [15]. As is known, every model that includes three-dimensional descriptors usually involves high computational costs and long times during the calculation of molecular geometry optimization. Therefore, the conformation-independent QSPR approach can be considered as a very useful methodology. In addition, we also explore the performance of QSPR models based on optimal descriptors [16]. Within this technique, the calculated optimal descriptor depends both on the molecular structure and the property under analysis (Koc), but does not explicitly depend on the 3D-molecular geometry. We have shown the importance of optimal descriptors in previous QSPR studies [17,18,19,20,21]. 2. Results and Discussion We begin our QSPR analysis by exploring the performance of molecular descriptors calculated with the PaDEL freeware. The most representative structural features of the training set of 93 heterogeneous compounds are searched through the RM technique. In this way, the best MLR models based on 1–6 molecular descriptors are found in a pool having 17,536 variables. In order to remove the ‘collinear’ (identical) descriptors, the linearly-dependent pairs are identified within RM, and only one variable from each pair is kept for further analysis. This process leads to a set containing 3491 linearly-independent descriptors. We follow the common practice of keeping the model’s dimension (d) as small as possible. The best MLR models are listed in Table 1, while a brief description of the descriptors meanings is provided in Table S1. It is appreciated from Table 1 that the RMStrain parameter continues improving beyond four descriptors, but RMStest does not significantly improve. According to this, we choose a structure-property relationship having four descriptors with an acceptable predictive power on the test set: (1) logKoc=0.18SP3+0.30CrippenLogP−0.090gmax+0.16XLogP+1.18 Ntrain=93, Rtrain2=0.87, RMStrain=0.45 Rijmax2=0.58, o2.5=0, Rloo2=0.85, RMSLOO=0.47, RMSrand=1.02 Ntest=550, Rtest2=0.81, RMStest=0.53 In this equation N is the number of compounds; Rijmax denotes the maximum correlation coefficient between descriptor pairs; o2.5 indicates the number of outlier compounds in the training set having a residual (difference between experimental and calculated activity) greater than 2.5-times RMStrain. The conformation-independent descriptors appearing in Equation (1) belong to four different classes [9]: (i) a PaDEL Chi Path Descriptor: SP3, simple path of order 3; (ii) a Crippen descriptor: CrippenLogP, Crippen’s LogP; (iii) an electrotopological state atom type descriptor: gmax, the maximum E-state; and (iv) the XLogP descriptor. A plot for the predicted logKoc as a function of the experimental values for the training and test sets is provided in Figure 1. The dispersion plot of residuals in Figure S1 tends to obey a random pattern around the zero line, suggesting that the assumption of the MLR technique is fulfilled. The correlation matrix for Equation (1) is given in Table S2, showing the absence of high correlations between descriptor pairs, while their numerical values are included in Table S3. Equation (1) has an acceptable predictive power on the external test set of 550 compounds, according to the Rtest2 and RMStest parameters. Such a model approves the internal validation process of Cross-Validation through the exclusion of one molecule at a time. The Y-Randomization technique demonstrates that Equation (1) has RMStrain<RMSrand and thus a valid structure-logKoc relationship is found. The external validation criteria recommended in [22] to assure predictive capability are also achieved and are summarized in Table S4. The statistical quality of Equation (1) is quite similar to various QSPR models reported previously by Gramatica et al. [14]. For instance, our QSPR with RMStrain=0.45 and RMStest=0.53 is better than the published four-topological descriptor model with RMStrain=0.52 and RMStest=0.56. Furthermore, Equation (1) is also comparable to the three-descriptor consensus model proposed in that paper (RMStrain=0.52 and RMStest=0.53), although such a model has as the disadvantage that it includes geometrical descriptors. In our approach, we do not consider the geometrical representation of the chemical structures, but consider their constitutional and topological aspects instead while achieving acceptable results. As a next step of this QSPR study, we include optimal molecular descriptor definitions in order to analyze the performance of such soil sorption-specific structural variables. The DCW optimal descriptor is optimized by increasing Rtrain2, until the model starts to lose predictive capability in the test set (measured by RMStest). The best structural representation for the 93 training compounds is hydrogen-filled graph, where the statistics for the stepwise evolution of the linear model is presented in Table 2. The first local descriptor selected is NNC (Nearest Neighboring Code), then the following ones are 0EC (Morgan Extended Connectivity of zero-th order) and NOSP (the presence of Nitrogen, Oxygen, Sulfur or Phosphorus) in that order. It is noted from Table 2 that the best quality optimal descriptor involves such three-variable types, and 64 active attributes are based on them (shown in Table S5). More complete details for the QSPR model are the following: (2) logKoc=0.073DCW+0.31 Ntrain=93, Rtrain2=0.87, RMStrain= 0.45 o2.5=1, Rloo2=0.86, RMSLOO=0.45, RMSrand=1.11 Ntest=550, Rtest2=0.76, RMStest=0.61 The parameters used for the DCW calculation are T=1 and Nepochs=7. Figures S2 and S3 demonstrate that the MLR technique is also satisfied for Equation (2). An example for the calculation of DCW for formaldehyde is provided in Table 3. Our results reveal that Equation (1) has a better performance on the test set than Equation (2). Both QSPRs are obtained through different approaches, i.e., by allowing or not the molecular descriptor representing the chemical structure to be dependent on the studied logKoc property. As a next step, we investigate what happens when the previous set of 3491 0D–2D descriptors from PaDEL is combined with the optimal DCW descriptor. The best 1–6 variable MLR models found in such pool of 3492 descriptors (Table S6) do not ameliorate the predictive power of our first model, as the training set statistics is better but not the one for the test set. In a new attempt to improve Equation (1), we consider the inclusion of EPI Suite predictions as semiempirical molecular descriptors, calculated through logKowEpi and logSwEpi predicted values. After searching the best MLR models in the set composed of 3493 independent descriptors from PaDEL and EPI Suite (refer to Table 4), the following structure-Koc relationship is achieved: (3) logKoc=0.60MLFER.E−0.36SubFP302+0.48logKowEpi+0.72 Ntrain=93, Rtrain2=0.87, RMStrain= 0.44 Rij max2=0.21, o2.5=0, Rloo2=0.86, RMSLOO=0.46, RMSrand=1.02 Ntest=550, Rtest2=0.84, RMStest=0.48 The performance of Equation (3) is better than Equation (1), and thus, we consider that this new QSPR model is the most suitable structure-soil sorption coefficient relationship for the 643 organic non-ionic compounds. Figure 2 and Figure S4 plot the predictions, while Tables S2 and S4 provide the correlation matrix and external validation criteria for Equation (3). The 2D molecular descriptors appearing in this last equation belong to three different classes: (i) a Molecular Linear Free Energy Relation (MLFER) descriptor: MLFER.E, measuring the excessive molar refraction; (ii) a substructure fingerprint: SubFP302, the presence of rotatable bonds; and (iii) an EPI Suite descriptor: logKowEpi. As the three descriptors take positive numerical values, Equation (3) indicates that a compound having higher values for both MLFER.E and logKowEpi descriptors together with a lower value for SubFP302 tend to have a higher predicted soil sorption coefficient. MLFER.E measures the excessive molar refraction: the molar refraction of the solute minus the molar refraction of an alkane of equivalent volume. This descriptor can be easily estimated from the knowledge of a compound’s refractive index, and suggests the propensity of the soil phase to interact with solute compounds having π- and σ-electron pairs. The SubFP302 descriptor has a clear interpretation as quantifies the presence (equal to one) or absence (equal to zero) of rotatable bonds in the chemical structure. This fingerprint identifies rotatable bonds that allow free rotation around themselves, that is to say, any single bond, not in a ring, bound to a non-terminal heavy atom. Finally, the logarithm of the octanol/water partition coefficient logKowEpi descriptor is a well-known physicochemical property that has been widely used in past QSPR studies for correlating the logKoc values. Therefore, hydrophobic compounds with high logKowEpi values tend to exhibit a higher retaining by the organic matter of soils and sediments. The analysis of the applicability domain of the new proposed QSPR reveals that 16 compounds out of the 550 included in the test set do not belong to the AD of the model, as hi>h*=0.13. The obtained leverage values are also provided in Table S7. We assume that this particular behavior is due to the complexity of the dataset, i.e., the great structural heterogeneity of the molecules considered in this study. Thus, the predicted logKoc values for all, with the exception to such 16 test set compounds, can be considered as reliable as they fall within the AD. As a final comparison, our best QSPR model with RMStrain=0.44 and RMStest=0.48 has a better performance on the heterogeneous compounds than the one provided by EPI Suite: RMStrain=0.47 and RMStest=0.56 (connectivity method) and RMStrain=0.48 and RMStest=0.56 (partition coefficient based method). This means that our developed QSPR model of Equation (3) represents an alternative/complementary tool to the EPI Suite program for predicting the studied property in present dataset of 643 organic non-ionic compounds. 3. Materials and Methods 3.1. Experimental Dataset The experimental soil sorption partition coefficient collected from [14] is quantified as the ratio between chemical concentration in soil and in water normalized to organic carbon. In the present dataset, logKoc ranges in the interval (−0.31, 6.02) in the training set (train) and (0, 6.33) in the test set (test); the complete list of 643 compounds studied here is included in Table S7 as Supplementary Material. The dataset is highly heterogeneous, and includes practically all of the principal functional groups present in pesticides and various organic pollutants. In addition and for comparison purposes, the calculated logarithm of the soil sorption partition coefficient is obtained through the Estimation Program Interface (EPI Suite) software from the KOCWIN module (logKocEpi) [23]. EPI Suite calculates logKocEpi via two different techniques: (a) based on the first order Molecular Connectivity Index (MCI); and (b) based on logKow (rather than MCI). In both cases, the program employs a series of group contribution factors. 3.2. Structural Representation and Molecular Descriptors Calculation The molecules are first drawn in mol format with ACDLabs ChemSketch freeware [24]. The set of conformation-independent molecular descriptors is computed using PaDEL Version 2.20 [25], because it has the advantage that it is a freely available and open source software. PaDEL currently calculates 1444 0D–2D descriptors and 12 fingerprint types (total 16,092 bits) [26]. Furthermore, semiempirical descriptors from EPI Suite are added, such as the calculated logarithm of the octanol/water partition coefficient from KOWWIN (logKowEpi) and the calculated logarithm of the water solubility from WATERNT (logSwEpi) [23]. Therefore, the total number of non-conformational descriptors explored in this work is 17,538. It is our intention to capture, with such a great number of descriptors, the most relevant structural characteristics affecting the studied property. 3.3. Model Development 3.3.1. Molecular Descriptors’ Selection in Multivariable Linear Regression (MLR) We employ the Replacement Method (RM) technique [27,28,29,30,31,32,33] in order to generate MLR models on the training set, by searching in a pool having D=17,538 descriptors for optimal subsets containing d descriptors (d is much lower than D), with smallest values for the standard deviation (Strain) or the root mean square deviation (RMStrain). Table S8 includes a list of mathematical equations involved in the present study. All of the MATLAB-programmed [34] algorithms used in our calculations are available upon request. 3.3.2. The Optimal Molecular Descriptors By means of the CORAL freeware (Correlation and Logic) [35] it is easy to define different optimal molecular descriptors. The Structural Representation (SR) used, i.e., graph or SMILES (Simplified Molecular Input Line Entry Specification), determines the Structural Attributes or local descriptors (SA) available for the QSPR. Therefore, it is necessary to decide which SA combination is the most appropriate, and this is done in a stepwise fashion, i.e., first search for the best single SA, then search for a second SA that combines the best with the previous one, and so on. The DCW descriptor is a linear combination of Correlation Weights (CW); refer to Table S8. The CW is calculated for each SA in the training set through the Monte Carlo (MC) simulation method. The DCW depends on the threshold (T) and the number of epochs (Nepochs): the appropriate selection of T and Nepochs avoids model over-fitting. The rare attributes are the ones that occur in less than T compounds, and in this work T is a positive integer analyzed in the range from 0-5. 3.3.3. Model Validation The linear regression models are theoretically validated through Leave-One-Out Cross-Validation (LOO) [22]. A more reliable validation is applied with an external test set of structures: the same training set-test set partition from [14] is used in present analysis, that is to say, 93 compounds in the training set and 550 compounds in the test set. We also scramble the experimental property values with Y-Randomization [36] and 10,000 cases, as a way of checking that the model is not a result of chance correlation when RMSrand is greater than RMStrain. 3.3.4. Applicability Domain A predictive QSPR model is only able to predict molecules falling within its Applicability Domain (AD) [37], so that the predicted property is not a result of substantial extrapolation (unreliable prediction). The AD definition is dependent on the model’s descriptors and the experimental property. Within the leverage approach [38], a test set compound must have a calculated leverage (hi) smaller than the warning leverage (h*). 4. Conclusions We have succeeded in establishing structure-property relationships for the soil sorption coefficient, a useful parameter related to sorption processes determining the environmental fate, distribution and persistence of chemicals. The chemical domain explored includes a heterogeneous set of 643 organic non-ionic compounds, having a Koc range of more than six log units. The QSPR models found on a training set composed of 93 compounds have an acceptable predictive performance on a test set including 550 compounds, and are able to fulfill other necessary mathematical conditions, such as Cross-Validation, Y-Randomization and Applicability Domain analysis. Our results compare favorably to previous reported ones from the literature, although the proposed models involve molecular descriptors calculated through freely available software like PaDEL, CORAL and EPI Suite. As we have developed a conformation-independent QSPR approach, the conformational representation of the chemical structures is avoided, and thus, no-experimental information on the X-ray crystal structure of compounds is required. Our research work continuously focuses on the use of new methods based on constitutional and topological approximations to QSPR studies, and thus, new results will be published shortly elsewhere. Acknowledgments We thank the financial support provided by the National Research Council of Argentina (CONICET) PIP11220100100151 project and to Ministerio de Ciencia, Tecnología e Innovación Productiva for the electronic library facilities. Juan C. Garro Martinez, Pablo R. Duchowicz, and Eduardo A. Castro are members of the scientific researcher career of CONICET. We thank the Prize Awarding Committee and the Editorial Board of International Journal of Molecular Sciences for considering one of our manuscripts in “IJMS 2015’s Best Paper Award”. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1247/s1. Click here for additional data file. Author Contributions The authors contributed equally to this work. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Predicted and experimental logKoc values according to QSPR based on Equation (1). Figure 2 Predicted and experimental logKoc values according to QSPR based on Equation (3). ijms-17-01247-t001_Table 1Table 1 The best linear QSPR models obtained from a pool of 3491 geometry independent descriptors obtained from PaDEL freeware; the selected model appears in bold. d Descriptors Rtrain2 Rtest2 RMStrain RMStest 1 CrippenLogP 0.72 0.68 0.65 0.67 2 CrippenLogP XLogP 0.80 0.76 0.55 0.59 3 CrippenLogP gmax TpiPC 0.84 0.79 0.49 0.56 4 SP3 CrippenLogP gmax XLogP 0.87 0.81 0.45 0.53 5 ALogp2 CrippenLogP maxHBint2 TpiPC XLogP 0.87 0.81 0.44 0.52 6 BCUTw-1l CrippenLogP gmax ETA_Epsilon_3 WPOL XLogP 0.89 0.81 0.41 0.53 ijms-17-01247-t002_Table 2Table 2 The stepwise search for finding the best structural attributes contributing the optimal descriptor; the selected result appears in bold. NNC, Nearest Neighboring Code; 0EC, Morgan Extended Connectivity of zero-th order; NOSP, the presence of Nitrogen, Oxygen, Sulfur or Phosphorus. Structural Attributes Rtrain2 Rtest2 RMStrain RMStest Nact NNC 0.84 0.73 0.49 0.64 50 NNC 0EC 0.86 0.75 0.46 0.62 70 NNC 0EC NOSP 0.87 0.76 0.45 0.61 64 ijms-17-01247-t003_Table 3Table 3 An example of the calculation of the optimal descriptor for formaldehyde by summing CW values: DCW=−0.64892. Structural Attribute CW EC0-O...1... 0.12508 EC0-C...3... 1.00094 EC0-H...1... −0.18254 EC0-H...1... −0.18254 NNC-O...101. 0.24867 NNC-C...303. −0.75284 NNC-H...101. −0.07978 NNC-H...101. −0.07978 NOSP01000000 −0.74613 ijms-17-01247-t004_Table 4Table 4 The best linear QSPR models obtained from a pool of 3493 geometry independent descriptors obtained from PaDEL and EPI Suite softwares; the selected model appears in bold. d Descriptors Rtrain2 Rtest2 RMStrain RMStest 1 logKowEpi 0.77 0.76 0.59 0.59 2 MLFER_E logKowEpi 0.86 0.83 0.46 0.50 3 MLFER_E SubFP302 logKowEpi 0.87 0.84 0.44 0.48 4 mindO MLFER_E KRFP1105 logKowEpi 0.88 0.84 0.42 0.48 5 MAXDP2 ZMIC1 TpiPC KRFP3788 logKowEpi 0.90 0.84 0.40 0.49 6 ATSC3c AATSC3c MATS4p MLFER_E AD2D393 logKowEpi 0.91 0.84 0.37 0.49 ==== Refs References 1. Sparks D.L. Environmental Soil Chemistry Academic Press Tokyo, Japan 2013 267 2. Jury W.A. Adsorption of organic chemicals onto soil Vadose Zone Modeling of Organic Pollutants Henn S.C. Melancon S.M. Lewis Publisher New York, NY, USA 1986 177 189 3. Gawlik B.M. Sotiriou N. Feicht E.A. Schulte-Hostede S. Kettrup A. Alternatives for the determination of the soil adsorption coefficient, KOC, of non-ionic organic compounds—A review Chemosphere 1997 34 2525 2551 10.1016/S0045-6535(97)00098-2 4. Hansch C. Leo A. Exploring QSAR. Fundamentals and Applications in Chemistry and Biology American Chemical Society Washington, DC, USA 1995 5. Kubinyi H. QSAR: Hansch Analysis and Related Approaches Wiley-Interscience New York, NY, USA 2008 6. Puzyn T. Leszczynski J. Cronin M.T.D. Recent Advances in QSAR Studies: Methods and Applications Springer Science & Business Media B.V. Houten, The Netherlands 2010 7. Katritzky A.R. Goordeva E.V. Traditional topological indices vs. Electronic, geometrical, and combined molecular descriptors in QSAR/QSPR research J. Chem. Inf. Comput. Sci. 1993 33 835 857 10.1021/ci00016a005 8113335 8. Diudea M.V.E. QSPR/QSAR Studies by Molecular Descriptors Nova Science Publishers New York, NY, USA 2001 9. Todeschini R. Consonni V. Molecular Descriptors for Chemoinformatics (Methods and Principles in Medicinal Chemistry) Wiley-VCH Weinheim, Germany 2009 10. Sabljic A. Gusten H. Verhaar H. Hermens J. QSAR modeling of soil sorption. Improvements and systematics of log Koc vs. Log kow correlations Chemosphere 1995 31 4489 4514 10.1016/0045-6535(95)00327-5 11. Duchowicz P.R. González M.P. Helguera A.M. Cordeiro M.N.D.S. Castro E.A. Application of the replacement method as novel variable selection in QSPR. 2. Soil sorption coefficients Chemom. Intell. Lab. Syst. 2007 88 197 203 10.1016/j.chemolab.2007.05.001 12. Goudarzi N. Goodarzi M. Araujo M.C.U. Galvão R.K.H. QSPR modeling of soil sorption coefficients (Koc ) of pesticides using SPA-ANN and SPA-MLR J. Agric. Food Chem. 2009 57 7153 7158 10.1021/jf9008839 19722589 13. Shao Y. Liu J. Wanga M. Shi L. Yao X. Gramatica P. Integrated QSPR models to predict the soil sorption coefficient for a large diverse set of compounds by using different modeling methods Atmos. Environ. 2014 88 212 218 10.1016/j.atmosenv.2013.12.018 14. Gramatica P. Giani E. Papa E. Statistical external validation and consensus modeling: A QSPR case study for Koc prediction J. Mol. Graph. Model. 2007 25 755 766 10.1016/j.jmgm.2006.06.005 16890002 15. Duchowicz P.R. Comelli N.C. Ortiz E.V. Castro E.A. QSAR study for carcinogenicity in a large set of organic compounds Curr. Drug Saf. 2012 7 282 288 10.2174/157488612804096623 23062240 16. Toropov A.A. Toropova A.P. Benfenati E. Gini G. OCWLGI descriptors: Theory and praxis Curr. Comput. Aided Drug Des. 2013 9 226 232 10.2174/1573409911309020007 23700994 17. Ibezim E. Duchowicz P.R. Ortiz E.V. Castro E.A. QSAR on aryl-piperazine derivatives with activity on malaria Chemom. Intell. Lab. Syst. 2012 110 81 88 10.1016/j.chemolab.2011.10.002 18. Mullen L.M.A. Duchowicz P.R. Castro E.A. QSAR treatment on a new class of triphenylmethyl-containing compounds as potent anticancer agents Chemom. Intell. Lab. Syst. 2011 107 269 275 10.1016/j.chemolab.2011.04.011 19. Toropov A.A. Leszczynska D. Leszczynski J. Predicting water solubility and octanol water partition coefficient for carbon nanotubes based on the chiral vector Comput. Biol. Chem. 2007 31 127 128 10.1016/j.compbiolchem.2007.02.002 17363328 20. Toropov A.A. Toropova A.P. Benfenati E. Gini G. Puzyn T. Leszczynska D. Leszczynski J. Novel application of the CORAL software to model cytotoxicity of metal oxide nanoparticles to bacteria Escherichia coli Chemosphere 2012 89 1098 1102 10.1016/j.chemosphere.2012.05.077 22704203 21. Toropova A.P. Toropov A.A. Martyanov S.E. Benfenati E. Gini G. Leszczynska D. Leszczynski J. CORAL: QSAR modeling of toxicity of organic chemicals towards Daphnia magna Chemom. Intell. Lab. Syst. 2012 110 177 181 10.1016/j.chemolab.2011.10.005 22. Golbraikh A. Tropsha A. Beware of q2 ! J. Mol. Graph. Model. 2002 20 269 276 10.1016/S1093-3263(01)00123-1 11858635 23. US EPA Available online: https://www.epa.gov/tsca-screening-tools/epi-suitetm-estimation-program-interface (accessed on 29 July 2016) 24. ACD/ChemSketch, 2016 Available online: http://www.acdlabs.com (accessed on 29 July 2016) 25. PaDEL, 2016 Available online: http://www.yapcwsoft.com/ (accessed on 29 July 2016) 26. Yap C.W. PaDEL-descriptor: An open source software to calculate molecular descriptors and fingerprints J. Comput. Chem. 2011 32 1466 1474 10.1002/jcc.21707 21425294 27. Duchowicz P.R. Castro E.A. Fernández F.M. Alternative algorithm for the search of an optimal set of descriptors in QSAR-QSPR studies MATCH Commun. Math. Comput. Chem. 2006 55 179 192 28. Duchowicz P.R. Castro E.A. Fernández F.M. González M. A new search algorithm of QSPR/QSAR theories: Normal boiling points of some organic molecules Chem. Phys. Lett. 2005 412 376 380 10.1016/j.cplett.2005.07.016 29. Duchowicz P.R. Talevi A. Bruno-Blanch L.E. Castro E.A. New QSPR study for the prediction of aqueous solubility of drug-like compounds Bioorg. Med. Chem. Lett. 2008 16 7944 7955 10.1016/j.bmc.2008.07.067 18701302 30. Goodarzi M. Duchowicz P.R. Wu C.H. Fernández F.M. Castro E.A. New hybrid genetic based support vector regression as QSAR approach for analyzing flavonoids-GABA(A) complexes J. Chem. Inf. Model. 2009 49 1475 1485 10.1021/ci900075f 19492793 31. Pomilio A.B. Giraudo M.A. Duchowicz P.R. Castro E.A. QSPR analyses for aminograms in food: Citrus juices and concentrates Food. Chem. 2010 123 917 927 10.1016/j.foodchem.2010.04.082 32. Talevi A. Goodarzi M. Ortiz E.V. Duchowicz P.R. Bellera C.L. Pesce G. Castro E.A. Bruno-Blanch L.E. Prediction of drug intestinal absorption by new linear and non-linear QSPR Eur. J. Med. Chem. 2011 46 218 228 10.1016/j.ejmech.2010.11.005 21112128 33. Pasquale G. Romanelli G.P. Autino J.C. García J. Ortiz E.V. Duchowicz P.R. Quantitative structure-activity relationships on chalcone derivatives: Mosquito larvicidal studies J. Agric. Food. Chem. 2012 60 692 697 10.1021/jf203374r 22217234 34. Matlab 7.0 Available online: http://www.mathworks.com (accessed on 29 July 2016) 35. Coral 1.5 Available online: http://www.insilico.eu/coral (accessed on 29 July 2016) 36. Wold S. Eriksson L. Statistical validation of qsar results Chemometrics Methods in Molecular Design van de Waterbeemd H. VCH Weinheim, Germany 1995 309 318 37. Gramatica P. Principles of qsar models validation: Internal and external QSAR Comb. Sci. 2007 26 694 701 10.1002/qsar.200610151 38. Eriksson L. Jaworska J. Worth A.P. Cronin M.T. McDowell R.M. Gramatica P. Methods for reliability and uncertainty assessment and for applicability evaluations of classification- and regression-based QSARS Environ. Health Perspect. 2003 111 1361 1375 10.1289/ehp.5758 12896860

PMC005xxxxxx/PMC5000646.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081248ijms-17-01248ArticleAssociation of Serum Uric Acid Concentration with Diabetic Retinopathy and Albuminuria in Taiwanese Patients with Type 2 Diabetes Mellitus Liang Ching-Chao 1Lin Pi-Chen 2Lee Mei-Yueh 234Chen Szu-Chia 345Shin Shyi-Jang 246Hsiao Pi-Jung 24Lin Kun-Der 247Hsu Wei-Hao 23*Cai Lu Academic Editor1 Department of Laboratory Technology, Kaohsiung Municipal CiJin Hospital, Kaohsiung 805, Taiwan; [email protected] Division of Endocrinology and Metabolism, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan; [email protected] (P.-C.L.); [email protected] (M.-Y.L.); [email protected] (S.-J.S.); [email protected] (P.-J.H.); [email protected] (K.-D.L.)3 Department of Internal Medicine, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung 812, Taiwan; [email protected] Graduate Institute of Clinical Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung 807, Taiwan5 Division of Nephrology, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung 807, Taiwan6 Center for Lipid and Glycomedicine Research, Kaohsiung Medical University, Kaohsiung 807, Taiwan7 Department of Internal Medicine, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung 801, Taiwan* Correspondence: [email protected]; Tel.: +886-7-8036783 (ext. 3440); Fax: +886-7-806334602 8 2016 8 2016 17 8 124826 5 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Patients with type 2 diabetes mellitus (DM) may experience chronic microvascular complications such as diabetic retinopathy (DR) and diabetic nephropathy (DN) during their lifetime. In clinical studies, serum uric acid concentration has been found to be associated with DR and DN. The goal of this study was to evaluate the relationship between the increases in serum uric acid level and the severity of DR and albuminuria in Taiwanese patients with type 2 DM. We recorded serum uric acid concentration, the severity of DR, and the severity of albuminuria by calculating urinary albumin-to-creatinine ratio (UACR) in 385 patients with type 2 DM. In multivariate logistic regression analysis, a high uric acid concentration was a risk factor for albuminuria (odds ratio (OR), 1.227; 95% confidence interval (CI) = 1.015–1.482; p = 0.034) and DR (OR, 1.264; 95% CI = 1.084–1.473; p = 0.003). We also demonstrated that there was a higher concentration of serum uric acid in the patients with more severe albuminuria and DR. In conclusion, an increased serum uric acid level was significantly correlated with the severity of albuminuria and DR in Taiwanese patients with type 2 DM. diabetes mellitusdiabetic retinopathyuric acidalbuminuriadiabetic nephropathy ==== Body 1. Introduction Diabetic retinopathy (DR) and diabetic nephropathy (DN) are two of the chronic microvascular complications in type 2 diabetes mellitus (DM). DR is a major cause of vision loss in adults [1], causing severe morbidity in patients with diabetes, resulting in public health and economic burdens. Prolonged exposure to the metabolic changes related to diabetes may damage the microvasculature of the retina, resulting in DR [2]. DN is a major cause of end stage renal disease in many countries [3,4], and it is a life-threatening condition. Identifying a clinical marker to detect the development and progression of diabetic microvascular complications is very important to allow for early management. A correlation between serum uric acid (SUA) level and the severity of DR has been reported in patients with type 2 diabetes [5]. SUA concentration has also been reported to be associated with DN and subclinical atherosclerosis [6,7]. In addition, DR and DN have been shown to be associated with SUA concentration. However, the association between DR and DN, and SUA level has yet to be investigated in Taiwanese patients with diabetes. An elevated uric acid level is a known major risk factor of diabetic microvascular diseases. Therefore, we performed this cross-sectional study to investigate the relationships between an increase in serum uric acid level, and the severity of DR and albuminuria in Taiwanese patients with type 2 DM. 2. Results Of the 385 patients included, 292 had a SUA level <7 mg/dL (low UA group), and 93 had a SUA level ≥7 mg/dL (high UA group). The clinical characteristics of the patients in the two groups are summarized in Table 1. The mean age of the high UA group was older than that in the low UA group (p = 0.013). The patients in the high UA group had higher systolic blood pressure (SBP), waist circumference (WC), and waist-to-hip circumference ratio (W-to-H) ratio than the patients in the low UA group (p = 0.015, <0.001, and 0.003, respectively). No significant differences were found between the two groups in term of coronary artery disease (CAD), cerebrovascular disease (CVD), duration of DM, diastolic blood pressure (DBP), hip circumference (HC), and body mass index (BMI). With regards to laboratory parameters, SUA, triglyceride, high-density lipoprotein (HDL) cholesterol, fasting plasma glucose, and estimated glomerular filtration rate (eGFR) were higher in the high UA group than in the low UA group (p < 0.001, 0.018, 0.001, 0.027, and <0.001, respectively). There were no significant differences in serum total cholesterol, low-density lipoprotein (LDL) cholesterol, and glycated hemoglobin (HbA1c) between the two groups. In univariate logistic regression analysis, we identified that the risk of albuminuria (urinary albumin-to-creatinine ratio (UACR) ≥30 mg/gm) (Table 2) was associated with high SBP (odds ratio (OR), 1.003; 95% confidence interval (CI) = 1.019–1.047; p < 0.001), high W-to-H ratio (OR, 2.154; 95% CI = 1.125–4.124; p = 0.021), high uric acid level (OR, 1.309; 95% CI = 1.156–1.483; p < 0.001), high HbA1c (OR, 1.129; 95% CI = 1.012–1.258; p = 0.029), and low eGFR (OR, 0.980; 95% CI = 0.973–0.987; p < 0.001), and that the risk of DR (Table 3) was associated with a long log duration of DM (OR, 5.295; 95% CI = 2.145–13.070; p < 0.001), high uric acid level (OR, 1.238; 95% CI = 1.086–1.411; p = 0.001), high fasting plasma glucose level (OR, 1.005; 95% CI = 1.001–1.008; p = 0.007), high HbA1c (OR, 1.172; 95% CI = 1.045–1.315; p = 0.007), and low eGFR (OR, 0.992; 95% CI = 0.984–0.999; p = 0.026). After multivariate adjustments, the risk factors for albuminuria were high SBP (OR, 1.023; 95% CI = 1.005–1.042; p = 0.015), high uric acid level (OR, 1.227; 95% CI = 1.015–1.482; p = 0.034), high HbA1c (OR, 1.183; 95% CI = 1.010–1.385; p = 0.037), and low eGFR (OR, 0.984; 95% CI = 0.972–0.997; p = 0.014), and the risk factors for DR were a long log duration of DM (OR, 6.133; 95% CI = 2.231–16.860; p < 0.001), and a high uric acid level (OR, 1.217; 95% CI = 1.013–1.461; p = 0.035). We then classified all of the patients into three groups: normalbuminuria (UACR < 30 mg/gm), microalbuminuria (UACR 30–299 mg/gm), and macroalbuminuria (UACR ≥ 300 mg/gm) according to urinary albumin excretion rate (Figure 1). The level of SUA was significantly higher in the macroalbuminuria group, compared with the normalbuminuria (6.9 ± 2.3 versus 5.6 ± 1.6, p < 0.001), and microalbuminuria (6.9 ± 2.3 versus 6.1 ± 1.7, p < 0.001) groups. In the microalbuminuria group, the level of SUA was significantly higher than that in the normalbuminuria group (6.1 ± 1.7 versus 5.6 ± 1.6, p < 0.001). We then divided all of the patients into three groups: no apparent DR (NDR), non-proliferative DR (NPDR), and proliferative DR (PDR) according to the severity of the DR (Figure 2). The level of SUA was significantly higher in the PDR group, compared with the NDR (7.4 ± 2.5 versus 5.7 ± 1.7, p < 0.001), and PDR (7.4 ± 2.5 versus 6.2 ± 1.8, p < 0.001) groups. 3. Discussion In this cross-sectional study, SUA was shown to be a risk factor for both albuminuria and DR in both univariate and multivariate logistic regression analyses. In addition, there was a trend that the more severe albuminuria or DR, the more elevated the level of SUA. These results are compatible with previous reports [5,6]. Our findings revealed that an increased level of SUA was associated with the severity of albuminuria and DR in Taiwanese patients with type 2 DM, and suggested that an elevated SUA level may reflect the severity of microvascular complications in patients with diabetes. SUA has been reported to be elevated with increasing in the severity of DR in patients with type 2 DM [5,8]. We also found a more elevated level of SUA in the patients with more severe DR. Krizova et al. reported that vitreous concentrations of uric acid were significantly higher in patients with diabetes than in nondiabetic controls [9,10]. An increased concentration in vascular endothelial vascular factor (VEGF) in the vitreous fluid has also been demonstrated in patients with DR [11]. In studies by Selim et al. and Funatsu et al., the aqueous and vitreous levels of VEGF were found to be significantly correlated with the severity of DR [12,13]. In another clinical study by Krizova et al., uric acid concentration in vitreous fluids was reported to be significantly associated with vitreous VEGF concentration in patients with DR [10], thus supporting the hypothesis that uric acid may be a contributory factor to the pathogenesis of DR. The SUA level has also been reported to be associated with the progression of albuminuria in patients with diabetes. Fukui et al. reported a significantly positive association between SUA concentration and the degree of urinary albumin excretion after adjusting for eGFR in men with type 2 DM [6]. Similarly, we also found a statistically significant relationship between SUA concentration and the severity of albuminuria classified into three grades. Two prospective cohort studies reported an association between SUA and the development or progression of DN in patients with type 1 DM and type 2 DM, respectively [14,15]. In addition, the level of SUA was independently associated with the progression of albuminuria to macroalbuminuria early in the course of type 1 DM [14]. A prospective study from Japan demonstrated a significant association between higher baseline SUA level and the subsequent risk of DN progression in patients with type 2 DM [15]. Several studies have reported that SUA may also play a role in diabetic peripheral neuropathy (DPN), a common microvascular complication in patients with diabetes. Higher levels of SUA in diabetic patients with peripheral neuropathy have also been reported [16,17]. Recently, a systemic review and meta-analysis reported an obvious increase in SUA levels in diabetic patients with DPN compared to those without DPN, and that hyperuricemia was significantly associated with an increased risk of DPN in patients with type 2 DM [18]. Hyperuricemia may be a marker and itself possibly be responsible for microvascular damage through inhibition of endothelial nitric oxide synthetase and activation of the renin-angiotensin system [19]. In addition, uric acid may cause microvascular disease independently of hypertension, possibly due to the direct effect of uric acid on endothelial cells and vascular smooth muscle cells [20]. In one review article, an elevated uric acid level was shown to be strongly associated with hypertension, kidney disease, metabolic syndrome, and carotid and coronary artery diseases [20], demonstrating an association between uric acid and microvascular and macrovascular diseases. A xanthine oxidase inhibitor can inhibit the activity of xanthine oxidase, the enzyme responsible for the conversion of hypoxanthine to xanthine to uric acid. Therefore, inhibition of xanthine oxidase can reduce the production of uric acid. Xanthine oxidase inhibitors include purine analogues (e.g., allopurinol), and non-purine analogues (e.g., febuxostat, and topiroxostat). In a randomized controlled trial enrolling patients with type 2 DM and nephropathy, Momeni et al. found that the 24-h urine protein level decreased after 4 months of allopurinol administration, probably due to a decreased level of serum uric acid [21]. However, in a systemic review including eight trials of patients with or without DM, or IgA nephropathy, meta-analysis of five trials showed that changes in proteinuria from baseline were similar between the allopurinol and control arms [22]. More recently, an ongoing clinical trial conducted by David Z.I. Cherney. investigated the effect of febuxostat on renal function including changes in glomerular filtration rate in adult patients with type 1 DM [23]. Another ongoing trial conducted by Sawako Kato et al. assesses the anti-albuminuric effect of topiroxostat in patients with hyperuricemia and DN [24]. However, the effect of lowering SUA level on DR has not previously been thoroughly discussed. Therefore, a randomized controlled trial focusing on the effect of uric acid-lowering agents on albuminuria and DR in patients with diabetes is needed to investigate the effect of decrease in SUA level. There are some limitations to this study. First, the cross-sectional design may have caused interference among the data; Second, we did not consider medications that may influence the concentration of SUA; Third, the number of enrolled subjects was small, and this may have influenced the power of the statistical analyses. Larger prospective trials are needed to assess the associations between uric acid, and DR and albuminuria; Finally, we did not evaluate diabetic neuropathy, and therefore our results cannot definitely conclude that SUA is associated with the severity of microvascular complications in patients with diabetes. 4. Materials and Methods 4.1. Subjects and Study Design This cross-sectional study included 385 patients with type 2 DM from Kaohsiung Municipal Ci-Jin Hospital, and Kaohsiung Medical University Hospital in Taiwan from 1 September 2014 to 29 February 2016. The inclusion criteria were patients attending the hospital clinic for treatment of type 2 DM, an age 18 years or older, and a duration of DM ≥1 year. The exclusion criteria were patients with type 1 DM, those who were pregnant, those treated for cancer in the last 5 years before study enrollment, those with blood disorders causing hemolysis (e.g., hemolytic anemia), kidney transplant recipients, and those with chronic glomerulonephritis. This study was conducted according to approved guidelines. The study protocol was approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20160048), and written informed consent was obtained from each participant. 4.2. Demographic and Clinical Data Demographic and medical data of the patients were collected by reviewing medical records, and included age, gender, SBP, DBP, WC, HC, W-to-H ratio, and BMI. Blood pressure was obtained from the upper limbs with the patient in a seated position using an automated device with the modified oscillometric pressure sensor method. The WC was measured at the proximate midpoint between the lower margin of the last palpable rib and the top iliac crest of the patients when standing. The HC was determined by measuring around the widest portion of the buttocks in a standing position. BMI was calculated as the ratio of weight in kilograms divided by the square of height in meters. Laboratory data were measured from fasting blood samples using an autoanalyzer (Roche Diagnostics GmbH, D-68298 Mannheim COBAS Integra 400, Mannheim, Germany). Serum creatinine was measured by using the compensated Jaffé method in a Roche/Integra 400 Analyzer. The abbreviated Modification of Diet in Renal Disease (MDRD) Study Group equation was used to calculate the eGFR: eGFR (mL/min/1.73 m2) = 186.3 × (serum creatinine−1.154) × (age−0.203) × 0.742 if female [25]. Urine albumin and creatinine were measured from a spot urine sample using an autoanalyzer (COBAS Integra 400 plus; Roche Diagnostics, Mannheim, Germany). Albuminuria was defined as UACR of ≥30 mg/gm. The ratio stood for urinary albumin excretion rate. The definitions of normalbuminuria, microalbuminuria, and macroalbuminuria were UACR < 30 mg/gm, UACR 30–299 mg/gm, and UACR ≥ 300 mg/gm, respectively. 4.3. Diabetes Retinopathy DR was evaluated by experienced ophthalmologists while the patients’ pupils were dilated. DR was classified as NDR, NPDR, and PDR. PDR was defined as the presence of the neovascularization of the retinal vessels and the complications of this neovascularization such as preretinal hemorrhage, vitreous hemorrhage, and traction retinal detachment. NPDR consisted of nerve fiber layer infarcts, intraretinal hemorrhage, hard exudates, and microvascular abnormalities such as microaneurysms in the retina without the presence of the neovascularization in the retina. Patients without these abnormalities in the retina were classified in the NDR group. 4.4. Statistical Analyses Data are expressed as mean ± standard deviation (SD). The Statistical Package for Social Science software (SPSS for Windows, version 19.0, International Business Machines Corporation (IBM), Armonk, NY, USA) was used to perform all statistical analyses. The Student’s t test was used for continuous variables, and the χ2 test was used for categorical variables. Binary logistic regression analysis was used to assess the influence of continuous and categorical variables on UACR, and DR to evaluate the risk factors for DN and DR. We also tried to run the log-binomial regression. Most of the coefficients from log-binomial regression were similar to those from logistic regression. However, some coefficients were undetectable due to stop of integration by Newton’s method. Therefore, we still run the logistic regression. A one-way analysis of variance (ANOVA) test was performed to compare the mean levels of SUA in two sets of three groups (albuminuria and DR, as shown in Figure 1 and Figure 2). A p value less than 0.05 was considered to be statistically significant. 5. Conclusions In conclusion, we found that an increased SUA level was significantly correlated with the severity of albuminuria and DR in Taiwanese patients with type 2 DM. Multivariate logistic regression analysis showed that SUA was a risk factor for both albuminuria and DR. Uric acid may play a role in the pathogenesis of diabetic microvascular diseases. Patients with type 2 diabetes mellitus often had coexisting microvascular complications when the diagnosis of diabetes mellitus was made. Therefore, identifying a clinical surrogate for the severity of diabetic microvascular complications is needed. Our findings suggested that in clinical practice, the SUA level obtained from diabetic patients may reflect the severity of the current microvascular complications in patients with type 2 DM. Regular measurements of SUA level as a potential marker for the severity of microvascular diseases may be beneficial for patients with diabetes. Acknowledgments The study was supported by a research grant from the Kaohsiung Medical University Hospital (Kaohsiung, Taiwan). The authors thank the help from the Statistical Analysis Laboratory, Department of Medical Research, Kaohsiung Medical University Hospital, Kaohsiung Medical University. Author Contributions Wei-Hao Hsu, and Mei-Yueh Lee conceived and designed the study. Ching-Chao Liang, and Pi-Chen Lin collected the clinical data. Kun-Der Lin interpreted the data. Szu-Chia Chen, Shyi-Jang Shin, and Pi-Jung Hsiao analyzed the data. Wei-Hao Hsu and Ching-Chao Liang wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Serum uric acid (SUA) concentrations in urinary albumin-to-creatinine ratio (UACR) < 30, UACR 30–299, and UACR ≥ 300 groups. UACR < 30 group, n = 232, SUA = 5.6 ± 1.6 mg/dL; UACR 30–299 group, n = 100, SUA = 6.1 ± 1.7 mg/dL; UACR ≥ 300 group, n = 53, SUA = 6.9 ± 2.3 mg/dL. * p < 0.001, UACR 30–299 group versus UACR < 30 group; UACR ≥ 300 group versus UACR 30–299 group. # p < 0.001, UACR ≥ 300 group versus UACR < 30 group. Figure 2 Serum uric acid (SUA) concentrations in no apparent diabetic retinopathy (NDR), non-proliferative DR (NPDR), and proliferative DR (PDR) groups. NDR group, n = 292, SUA = 5.7 ± 1.7 mg/dL; NPDR group, n = 73, SUA = 6.22 ± 1.8 mg/dL, PDR group, n = 20, SUA = 7.4 ± 2.5 mg/dL. * p < 0.001, PDR group versus NPDR group. # p < 0.001, PDR group versus NDR group. ijms-17-01248-t001_Table 1Table 1 Comparison of clinical characteristics between patients with serum uric acid (SUA) <7 and ≥7 mg/dL. Characteristics All Patients (n = 385) SUA < 7 mg/dL (n = 292) SUA ≥ 7 mg/dL (n = 93) Age (year) 64.6 ± 12.1 63.7 ± 11.6 67.3 ± 13.3 * Male gender (%) 49.6 43.5 68.8 CAD (%) 6.6 7.2 4.3 CVD (%) 3.8 2.7 7.2 Duration of DM (years) 9 (5–16) 9 (4–15) 11 (6–16) Systolic BP (mmHg) 138.6 ± 17.8 137.3 ± 17.1 142.7 ± 19.6 * Diastolic BP (mmHg) 76.1 ± 11.9 76.0 ± 11.5 76.1 ± 13.0 WC (cm) 92.6 ± 11.1 91.0 ± 10.4 97.5 ± 11.7 ** HC (cm) 100.0 ± 10.0 99.3 ± 10.2 102.4 ± 9.4 W-to-H ratio 0.93 ± 0.07 0.92 ± 0.07 0.95 ± 0.06 * BMI (kg/m2) 26.4 ± 4.6 26.1 ± 4.4 27.3 ± 5.2 Laboratory parameters Uric acid (mg/dL) 5.9 ± 1.8 5.1 ± 1.1 8.3 ± 1.3 ** Triglyceride (mg/dL) 121 (83.5–172) 117 (78–164) 144 (99.5–206.5) * Total cholesterol (mg/dL) 180.6 ± 47.7 179.9 ± 46.4 182.9 ± 51.6 HDL-cholesterol (mg/dL) 45.0 ± 12.9 46.4 ± 13.0 40.8 ± 11.6 * LDL-cholesterol (mg/dL) 101.7 ± 35.4 100.8 ± 33.2 104.5 ± 41.6 Fasting glucose (mg/dL) 154.8 ± 64.3 158.4 ± 67.3 143.4 ± 52.6 * HbA1c (g/dL) 7.6 ± 1.9 7.6 ± 1.9 7.5 ± 1.9 eGFR (mL/min/1.73 m2) 78.6 ± 33.0 84.5 ± 31.5 60.0 ± 30.7 ** Urinary albumin-to-creatinine ratio (UACR) (mg/gm) (p < 0.001) <30 60.3 66.4 40.9 30–300 26.0 22.9 35.5 ≥300 13.8 10.6 23.7 Diabetic retinopathy (DR) (p = 0.051) NDR 75.8 78.1 68.9 NPDR 19.0 18.2 21.5 PDR 5.2 3.8 9.7 * p < 0.05, ** p < 0.001 compared to patients with SUA <7 mg/dL. CAD, coronary artery disease; CVD, cerebrovascular disease; DM, type 2 diabetes mellitus; BP, blood pressure; WC, waist circumference; HC, hip circumference; W-to-H, waist-to-hip circumference ratio; BMI, body mass index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, glycated hemoglobin; eGFR, estimated glomerular filtration rate; NDR, no apparent DR; NPDR, non-proliferative DR. ijms-17-01248-t002_Table 2Table 2 Risk factors for urinary albumin-to-creatinine ratio (UACR) ≥ 30 mg/gm using binary logistic regression analysis. Parameters Univariate Multivariate OR (95% CI) p OR (95% CI) p Age (per 1 year) 1.009 (0.992–1.027) 0.278 0.987 (0.956–1.019) 0.424 Male gender (versus female) 1.486 (0.986–2.240) 0.059 0.910 (0.457–1.812) 0.788 CAD 0.703 (0.259–1.907) 0.489 - - CVD 1.918 (0.571–6.439) 0.292 - - Log duration of DM (per 1 year) 1.623 (0.839–3.138) 0.150 - - Systolic BP (per 1 mmHg) 1.033 (1.019–1.047) <0.001 1.023 (1.005–1.042) 0.015 Diastolic BP (per 1 mmHg) 1.013 (0.995–1.032) 0.156 - - WC (per 1 cm) 1.012 (0.987–1.039) 0.350 - - HC (per 1 cm) 0.988 (0.960–1.017) 0.404 - - W-to-H ratio (per 0.01) 2.154 (1.125–4.124) 0.021 1.816 (0.854–3.861) 0.121 BMI (per 1 kg/m2) 1.007 (0.959–1.058) 0.766 - - Laboratory parameters Uric acid (per 1 mg/dL) 1.309 (1.156–1.483) <0.001 1.227 (1.015–1.482) 0.034 Log Triglyceride (per 1 mg/dL) 1.339 (0.614–2.921) 0.463 - - Total cholesterol (per 1 mg/dL) 0.999 (0.994–1.003) 0.524 - - HDL-cholesterol (per 1 mg/dL) 0.986 (0.969–1.004) 0.120 - - LDL-cholesterol (per 1 mg/dL) 0.996 (0.990–1.002) 0.178 - - Fasting glucose (per 1 mg/dL) 1.002 (0.998–1.005) 0.348 - - HbA1c (per 1%) 1.129 (1.012–1.258) 0.029 1.183 (1.010–1.385) 0.037 eGFR (per 1 mL/min/1.73 m2) 0.980 (0.973–0.987) <0.001 0.984 (0.972–0.997) 0.014 Values express as odds ratios (OR) and 95% confidence interval (CI). Abbreviations are same as Table 1. ijms-17-01248-t003_Table 3Table 3 Risk factors for diabetic retinopathy using binary logistic regression analysis. Parameters Univariate Multivariate OR (95% CI) p OR (95% CI) p Age (per 1 year) 0.992 (0.973–1.011) 0.392 0.974 (0.948–1.001) 0.060 Male gender (versus female) 0.862 (0.541–1.373) 0.532 0.920 (0.495–1.709) 0.793 CAD 0.858 (0.275–2.679) 0.792 - - CVD 0.315 (0.04–2.506) 0.275 - - Log duration of DM (per 1 year) 5.295 (2.145–13.070) <0.001 6.133 (2.231–16.860) <0.001 Systolic BP (per 1 mmHg) 1.014 (1.000–1.028) 0.052 - - Diastolic BP (per 1 mmHg) 1.005 (0.984–1.026) 0.641 - - WC (per 1 cm) 0.975 (0.946–1.006) 0.109 - - HC (per 1 cm) 0.974 (0.941–1.008) 0.127 - - W-to-H ratio (per 0.01) 0.952 (0.478–1.894) 0.888 - - BMI (per 1 kg/m2) 0.961 (0.906–1.019) 0.185 - - Laboratory parameters Uric acid (per 1 mg/dL) 1.238 (1.086–1.411) 0.001 1.217 (1.013–1.461) 0.035 Log Triglyceride (per 1 mg/dL) 1.550 (0.644–3.732) 0.328 - - Total cholesterol (per 1 mg/dL) 1.005 (1.000–1.009) 0.057 - - HDL-cholesterol (per 1 mg/dL) 1.007 (0.988–1.027) 0.469 - - LDL-cholesterol (per 1 mg/dL) 1.001 (0.995–1.008) 0.701 - - Fasting glucose (per 1 mg/dL) 1.005 (1.001–1.008) 0.007 1.000 (0.995–1.005) 0.966 HbA1c (per 1%) 1.172 (1.045–1.315) 0.007 1.159 (0.963–1.395) 0.118 eGFR (per 1 mL/min/1.73 m2) 0.992 (0.984–0.999) 0.026 0.997 (0.986–1.008) 0.605 Values express as odds ratios (OR) and 95% confidence interval (CI). Abbreviations are same as Table 1. ==== Refs References 1. Fong D.S. Aiello L.P. Ferris F.L. 3rd Klein R. Diabetic retinopathy Diabetes Care 2004 27 2540 2553 10.2337/diacare.27.10.2540 15451934 2. Sivaprasad S. Gupta B. Crosby-Nwaobi R. Evans J. Prevalence of diabetic retinopathy in various ethnic groups: A worldwide perspective Surv. Ophthalmol. 2012 57 347 370 10.1016/j.survophthal.2012.01.004 22542913 3. Ritz E. Orth S.R. Nephropathy in patients with type 2 diabetes mellitus N. Engl. J. Med. 1999 341 1127 1133 10.1056/NEJM199910073411506 10511612 4. Hwang S.J. Tsai J.C. Chen H.C. Epidemiology, impact and preventive care of chronic kidney disease in taiwan Nephrology 2010 15 Suppl. 2 3 9 10.1111/j.1440-1797.2010.01304.x 20586940 5. Lee J.J. Yang I.H. Kuo H.K. Chung M.S. Chen Y.J. Chen C.H. Liu R.T. Serum uric acid concentration is associated with worsening in severity of diabetic retinopathy among type 2 diabetic patients in taiwan—A 3-year prospective study Diabetes Res. Clin. Pract. 2014 106 366 372 10.1016/j.diabres.2014.07.027 25108898 6. Fukui M. Tanaka M. Shiraishi E. Harusato I. Hosoda H. Asano M. Kadono M. Hasegawa G. Yoshikawa T. Nakamura N. Serum uric acid is associated with microalbuminuria and subclinical atherosclerosis in men with type 2 diabetes mellitus Metabolism 2008 57 625 629 10.1016/j.metabol.2007.12.005 18442624 7. Fu C.C. Wu D.A. Wang J.H. Yang W.C. Tseng C.H. Association of C-reactive protein and hyperuricemia with diabetic nephropathy in Chinese type 2 diabetic patients Acta Diabetol. 2009 46 127 134 10.1007/s00592-008-0069-0 18853099 8. Cai X.L. Wang F. Ji L.N. Risk factors of diabetic retinopathy in type 2 diabetic patients Chin. Med. J. 2006 119 822 826 16732984 9. Krizova L. Kalousova M. Kubena A. Benakova H. Zima T. Kovarik Z. Kalvoda J. Kalvodova B. Increased uric acid and glucose concentrations in vitreous and serum of patients with diabetic macular oedema Ophthalmic Res. 2011 46 73 79 10.1159/000322994 21242702 10. Krizova L. Kalousova M. Kubena A.A. Chrapek O. Chrapkova B. Sin M. Zima T. Correlation of vitreous vascular endothelial growth factor and uric acid concentration using optical coherence tomography in diabetic macular edema J. Ophthalmol. 2015 2015 10.1155/2015/478509 26682064 11. Aiello L.P. Avery R.L. Arrigg P.G. Keyt B.A. Jampel H.D. Shah S.T. Pasquale L.R. Thieme H. Iwamoto M.A. Park J.E. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders N. Engl. J. Med. 1994 331 1480 1487 10.1056/NEJM199412013312203 7526212 12. Selim K.M. Sahan D. Muhittin T. Osman C. Mustafa O. Increased levels of vascular endothelial growth factor in the aqueous humor of patients with diabetic retinopathy Indian J. Ophthalmol. 2010 58 375 379 10.4103/0301-4738.67042 20689190 13. Funatsu H. Yamashita H. Noma H. Mimura T. Nakamura S. Sakata K. Hori S. Aqueous humor levels of cytokines are related to vitreous levels and progression of diabetic retinopathy in diabetic patients Graefe’s Arch. Clin. Exp. 2005 243 3 8 10.1007/s00417-004-0950-7 15258777 14. Hovind P. Rossing P. Tarnow L. Johnson R.J. Parving H.H. Serum uric acid as a predictor for development of diabetic nephropathy in type 1 diabetes: An inception cohort study Diabetes 2009 58 1668 1671 10.2337/db09-0014 19411615 15. Hayashino Y. Okamura S. Tsujii S. Ishii H. Association of serum uric acid levels with the risk of development or progression of albuminuria among Japanese patients with type 2 diabetes: A prospective cohort study [Diabetes Distress And Care Registry at Tenri (DDCRT 10)] Acta Diabetol. 2016 10.1007/s00592-015-0825-x 26935413 16. Papanas N. Katsiki N. Papatheodorou K. Demetriou M. Papazoglou D. Gioka T. Maltezos E. Peripheral neuropathy is associated with increased serum levels of uric acid in type 2 diabetes mellitus Angiology 2011 62 291 295 10.1177/0003319710394164 21306998 17. Kiani J. Habibi Z. Tajziehchi A. Moghimbeigi A. Dehghan A. Azizkhani H. Association between serum uric acid level and diabetic peripheral neuropathy (a case control study) Casp. J. Intern. Med. 2014 5 17 21 18. Yu S. Chen Y. Hou X. Xu D. Che K. Li C. Yan S. Wang Y. Wang B. Serum uric acid levels and diabetic peripheral neuropathy in type 2 diabetes: A systematic review and meta-analysis Mol. Neurobiol. 2016 53 1045 1051 10.1007/s12035-014-9075-0 25579387 19. Mene P. Punzo G. Uric acid: Bystander or culprit in hypertension and progressive renal disease? J. Hypertens. 2008 26 2085 2092 10.1097/HJH.0b013e32830e4945 18854744 20. Oh J. Won H.Y. Kang S.M. Uric acid and cardiovascular risk N. Engl. J. Med. 2009 360 539 540 19186314 21. Momeni A. Shahidi S. Seirafian S. Taheri S. Kheiri S. Effect of allopurinol in decreasing proteinuria in type 2 diabetic patients Iran. J. Kidney Dis. 2010 4 128 132 20404423 22. Bose B. Badve S.V. Hiremath S.S. Boudville N. Brown F.G. Cass A. de Zoysa J.R. Fassett R.G. Faull R. Harris D.C. Effects of uric acid-lowering therapy on renal outcomes: A systematic review and meta-analysis Nephrol. Dial. Transplant. 2014 29 406 413 10.1093/ndt/gft378 24042021 23. David Z.I. Cherney. The Effect of Uric Acid Lowering in Type 1 Diabetes Available online: https://clinicaltrials.gov/ct2/show/NCT02344602 (accessed on 23 June 2016) 24. Kato S. Ando M. Mizukoshi T. Nagata T. Katsuno T. Kosugi T. Tsuboi N. Maruyama S. Randomized control trial for the assessment of the anti-albuminuric effects of topiroxostat in hyperuricemic patients with diabetic nephropathy (the ETUDE study) Nagoya J. Med.Sci. 2016 78 135 142 27303100 25. Levey A.S. Coresh J. Balk E. Kausz A.T. Levin A. Steffes M.W. Hogg R.J. Perrone R.D. Lau J. Eknoyan G. National kidney foundation practice guidelines for chronic kidney disease: Evaluation, classification, and stratification Ann. Intern. Med. 2003 139 137 147 10.7326/0003-4819-139-2-200307150-00013 12859163

PMC005xxxxxx/PMC5000647.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081249ijms-17-01249ReviewAquaporin-4 and Cerebrovascular Diseases Chu Heling 1†Huang Chuyi 2†Ding Hongyan 1Dong Jing 1Gao Zidan 1Yang Xiaobo 13Tang Yuping 1*Dong Qiang 1*Ishibashi Kenichi Academic Editor1 Department of Neurology, Huashan Hospital, State Key Laboratory of Medical Neurobiology, Fudan University, No. 12 Mid. Wulumuqi Road, Shanghai 200040, China; [email protected] (H.C.); [email protected] (H.D.); [email protected] (J.D.); [email protected] (Z.G.); [email protected] (X.Y.)2 Department of Neurology, Shanghai Jiaotong University Affiliated Sixth People’s Hospital, No. 600 Yishan Road, Shanghai 200030, China; [email protected] Department of Neurology, Jinshan Hospital, Fudan University, No. 1508 Longhang Road, Shanghai 201508, China* Correspondence: [email protected] (Y.T.); [email protected] (Q.D.); Tel.: +86-21-5288-9999 (Y.T.); +86-21-5288-7145 (Q.D.); Fax: +86-21-6248-1221 (Y.T.); +86-21-6248-1401 (Q.D.)† These authors contributed equally to this work. 11 8 2016 8 2016 17 8 124925 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Cerebrovascular diseases are conditions caused by problems with brain vasculature, which have a high morbidity and mortality. Aquaporin-4 (AQP4) is the most abundant water channel in the brain and crucial for the formation and resolution of brain edema. Considering brain edema is an important pathophysiological change after stoke, AQP4 is destined to have close relation with cerebrovascular diseases. However, this relation is not limited to brain edema due to other biological effects elicited by AQP4. Till now, multiple studies have investigated roles of AQP4 in cerebrovascular diseases. This review focuses on expression of AQP4 and the effects of AQP4 on brain edema and neural cells injuries in cerebrovascular diseases including cerebral ischemia, intracerebral hemorrhage and subarachnoid hemorrhage. In the current review, we pay more attention to the studies of recent years directly from cerebrovascular diseases animal models or patients, especially those using AQP4 gene knockout mice. This review also elucidates the potential of AQP4as an excellent therapeutic target. aquaporin-4cerebral ischemiaintracerebral hemorrhagesubarachnoid hemorrhage ==== Body 1. Introduction Aquaporins (AQPs) are a family of water channel proteins and famous for water transportation under physical and pathological conditions. Since the first water channel, termed AQP1, was discovered by Peter Agre in 1992 [1], at least 13 AQP members have been found in mammals [2]. A consensus motif is a common characteristic of all members of AQPs, which is essential for pore formation [3]. Apart from the pure water channel (AQP0, -1, -2, -4, -5, -6, and -8), there are a subset of AQPs that also transport glycerol called aquaglyceroporins (AQP3, -7, -9, and -10) [4]. In 1994, Agre’s group isolated the fourth mammalian member of the aquaporin water channel family (AQP4) by homology cloning, which regulated body water balance and mediated water flow within the central nervous system (CNS) as the osmoreceptor [5]. AQP4 is the most abundant water channel in CNS and predominantly expressed in astrocyte foot processes surrounding capillaries astrocyte processes which are comprised of the glial limiting membrane and in ependymal cells [6,7]. It is crucial for the formation and resolution of brain edema. Besides CNS water balance maintenance, by means of AQP4 knockout animals, several other biological effects of AQP4 have been demonstrated, including neural signal transduction regulation, synaptic plasticity, astrocyte migration, neurogenesis and neuroinflammation [8,9,10,11,12]. Cerebrovascular diseases are conditions caused by problems with brain vasculature, which mainly contain ischemic and hemorrhagic stroke. These diseases have a high morbidity and mortality throughout the world. Because brain edema is an important pathophysiological change after stoke, AQP4 is destined to have close relation with cerebrovascular diseases. However, this relation is not limited to brain edema. This review briefly introduces the structure and function of AQP4 and focuses on the effects of AQP4 on stroke. 2. Structure and Function of AQP4 2.1. Structure and Distribution of AQP4 AQP4 monomers consist of six helical, membrane-spanning domains and two highly conserved Asn-Pro-Ala (NPA) motifs that create a narrow aqueous pathway [13] (Figure 1A). Similar to other aquaporins, AQP4 monomers also assemble as tetramers. Importantly, AQP4 tetramers further cluster in the plasma membrane forming crystal-like supramo-lecular assemblies, termed orthogonal arrays of particles (OAPs). OAPs can be visualized in membranes by freeze-fracture electron microscopy whichare originally confirmed to be formed by AQP4 in AQP4-transfected Chinese hamster ovary cells [14,15]. AQP4 has two major isoforms: M1 and M23, which are transcribed from two different initiation sites on the same gene. M1 is a relatively long isoform with translation initiation at Met-1, while M23 is a shorter one with translation initiation at Met-23 [16]. Although AQP4 is the most abundant water channel in the brain, it is only detected in the plasma membrane of astrocytes and ependymal membranes since its discovery over two decades. Its location can be characterized as the cell surfaces of the blood–brain barrier (BBB) and cerebrospinal fluid (CSF)–brain barrier. Therefore, AQP4 is expressed in astrocyte foot processes surrounding capillaries, astrocyte processes which are comprised of the glial limiting membrane, ependymal cells and subependymal astrocytes [7,18] (Figure 1B). Besides, it was also found that AQP4 mRNA and protein are expressed by reactive microglial cells. However, this is still controversial because of a lack of support from further study [19,20]. The polarized distribution of AQP4 depends on some proteins also with polarized expressionin astrocytes. α-syntrophin, a member of the dystrophin associated protein complex, plays an important role in anchoring of AQP4 to astrocyte end-foot processes [21,22]. Besides, the matrix constituent agrin is also responsible for AQP4 polarization [21,23]. 2.2. Animal Models for Studying Function of AQP4 Currently, no effective and specific AQP4 inhibitors have been developed. AQP4 knockout mice play essential roles in exploring AQP4 function. Since all properties are similar to wild type mice except absence of AQP4, AQP4 knockout mice are excellent candidates for AQP4 study. Numerous reports have revealeddifferentAQP4 functions through comparing AQP4 deletion mice with wild type mice [8,9,10,11,12]. There are mainly three research groups that have reported AQP4 knockout lines. Verkman’s group, from San Francisco, USA, first generated AQP4 knockout lines in 1997, though they revealed many biological properties of AQP4 [24]. Afterwards, Hu’s group from Nanjing, China and the group of Ottersen and Nagelhus from Oslo, Norway also generated AQP4 knockout lines [25,26]. However, there are still some differences in the properties among the AQP4 null mice lines. For example, brain morphology and BBB integrity are not affected in AQP4 deletion mice from San Francisco and Oslo [20,26], which are distinct from those of Nanjing [27], suggesting tiny differences may be produced during the generating process. Besides AQP4 deletion, there is an alternative tool model; that is, mice lacking polarized AQP4 expression. It has been proven that α-syntrophin deficient mice lack polarized expression of AQP4, which are the most commonly used model with depolarized expression of AQP4 [21]. This model demonstrates the significance of polarized distribution of AQP4 in its biological properties. In α-syntrophin-null mice, development of brain edema of an experimental acute hyponatremia model was delayed and K+ clearance in epileptic seizures was prolonged, which is in accordance with AQP4 deletion mice [28,29]. Therefore, the AQP4 depolarized expression model can be considered as an alternative approach to test AQP4 function. 2.3. AQP4 and Brain Edema Brain edema is excess accumulation of fluid in the intracellular or extracellular spaces of the brain. Brian edema can result from several brain pathologies, including brain trauma, cerebrovascular diseases, brain tumors, brain inflammation and metabolic diseases [30]. Brain edema may aggravate the primary diseases. There are four types of brain edema: cytotoxic, vasogenic, osmotic and interstitial edema, with the former two being the most common and classic [31]. Cytotoxic edema is intracellular accumulation of water owing to energy failure resulting from impairment of the sodium and potassium pump in cell membrane. Astrocytes are the major cell type involved in cytotoxic edema [32]. The typical cytotoxic edema can be seen in early ischemia or hypoxia, cerebral malaria and hyponatremia. Vasogenic edema occurs due to disruption of BBB, which results in entry of intravascular proteins and fluid into extracellular space [33]. This kind of brain edema is usually found in brain tumors, focal inflammation, abscess and late ischemia. It has been revealed the existence of a brain-wide paravascular pathway for CSF and interstitial fluid exchange termed “glymphatic” system [34], which contributes greatly to the two types of brain edema. Polarized distribution of AQP4 plays an important role in this pathway [35]. By means of observation to appropriate brain edema models in AQP4 null mice versus wild types, it has been proven that AQP4 plays an important role in regulation of the two types of edema. Manley and colleagues established two models of cytotoxic edema, acute water intoxication and early cerebral ischemia, and demonstrated brain edema was significantly reduced in AQP4-deficient mice [36]. Besides, it was reported that AQP4-deficient mice had remarkably lower brain water accumulation in acute bacterial meningitis, another cytotoxic edema model [37]. In addition, mice lacking polarized AQP4 expression also had a reduction of water influx after early cerebral ischemia [28]. As to vasogenic edema, Papadopoulos et al. illustrated AQP4 knockout mice had elevated brain water in three vasogenic edema models: intracerebral fluid infusion, focal cortical freeze injury and brain tumor implantation [38]. Therefore, it can be concluded that AQP4 is involved in the formation of cytotoxic edema and elimination of vasogenic edema. Whereas, the brain edema formed by many CNS diseases is usually not a single type, and there are also some changes in brain edema types during the course. Therefore, the roles of AQP4 in those diseases are also complicated, and appropriate animal models are required for continuous and dynamic observation. 2.4. Other Function of AOP4 Besides regulation of brain edema, AQP4 can elicit multiple biological effects. Here, we briefly describe the important ones. Because AQP4 is selectively expressed in astrocytes, it is essential for astrocytes function. Verkman’s group demonstrated that cultured astrocytes derived from AQP4 null mice had markedly low migration efficiency towards the wound [10]. Moreover, it was verified using in vivo experiments, and it was found that astrocytes of wild type mice migrated faster than AQP4 knockout mice [39]. It was indicated that AQP4 promoted migration of astrocytes, which was beneficial for formation of glial scar. Meanwhile, AQP4 also modulates brain excitability in epilepsy [40]. Binder et al. demonstrated that the latency to generalized seizures was significantly lower in wild-type mice comparing with AQP4 deletion mice [41]. This group then explored the mechanism and the results showed AQP4 knockout mice increased seizure duration via slowing potassium kinetics [42]. α-syntrophin-null mice exhibited similar effects of prolonging potassium ions clearance [28]. However, the role of AQP4 in potassium ions buffering is controversial as Haj-Yasein et al. reported that AQP4 removal did not affect potassium ions recovery following synaptic activation [43]. Besides potassium ions, it was also reported AQP4 null mice had a reduction of astrocytic calcium ions spikes initiated by hypoosmotic stress, suggesting AQP4 is associated with calcium signal transduction [8]. Moreover, AQP4 has a potential to influence synaptic plasticity [44]. Skucas and colleagues demonstrated that absence of AQP4 selectively impaired neurotrophin-dependent synaptic plasticity [9]. Consistently, research from other groups also shows AQP4 is crucial for maintaining normal consolidation of long-term hippocampus-dependent memories by promoting incorporation of new neurons into spatial memory networks [45]. In addition, AQP4 is considered to promote neurogenesis. Hu’s group showed AQP4 deletion inhibited the proliferation, survival, migration and neuronal differentiation of neural stem cells derived from the subventricular zone by disrupting intracellular calcium ions dynamics [46]. Their following experiment using a depression model indicates AQP4 is required for the antidepressive action of fluoxetine through regulating adult hippocampal neurogenesis [11]. 3. Cerebral Ischemia 3.1. Expression of AQP4 The pathogenic process and pathophysiological mechanism of the corresponding clinical diseases are properly simulated by the current brain ischemia models. There are mainly focal ischemic models: middle cerebral arterial occlusion (MCAO), including permanent and transient MCAO and global ischemic models: two or four-vessel occlusion in animal experiments. The well-known cell culture model is oxygen and glucose deprivation (OGD) with or without reoxygenation. Studies on AQP4 expression after cerebral ischemia mainly focus on animal or cell models. It is well-known that AQP4 mRNA and protein are up-regulated at 30 min after permanent MCAO [47]. After transient MCAO of rat pup, AQP4 expression was increased on astrocytic end-feet in the border regions of injured tissues at 24 h, lasting at least 72 h and normalized at 28 days, which was in accord with brain edema showed by Magnetic resonance imaging (MRI) [48]. A continuous and dynamic observation was carried out in another research using a MCAO and reperfusion model, and the results demonstrated AQP4 expression was significantly increased on astrocyte end-feet both in the core and in the border of the lesion with two peaks: 1 h and 48 h [49]. However, a recent study of a global cerebral ischemic model showed no marked change of AQP4 within 48 h [50]. In astrocyte culture, Chikako et al. found expression of AQP4 was significantly decreased by OGD injury, but gradually recovered after reoxygenation with a significant up-regulation after 16 h [51]. Other research also showed up-regulation of AQP4 24 h after OGD/reoxygenation [52]. There are also studies from cerebral infarction patients. It was demonstrated AQP4 was only expressed on astrocytes and was highly localized on their end feet facing the outer surface of capillaries [53,54]. Stokum and coworkers’ work focused on AQP4 expression in different cerebral zones of both infarction patients and animal models. Their results revealed in cortex perivascular AQP4 was reduced with an unchanged AQP4 protein abundance, while an increase of perivascular and plasmalemmal AQP4 was observed in white matter with a 2.2- to 6.2-fold increase in AQP4 isoform abundance. Meanwhile, ischemic white matter swelled by approximately 40%, while cortex swelled by approximately 9% [55]. Accordingly, it can be concluded that AQP4 expression after cerebral ischemia tends to be up-regulated, although the specific change forms may be different due to different models. 3.2. Regulation of AQP4 after Cerebral Ischemia AQP4 expression is regulated by some signal transduction pathways in some in vivo and in vitro experiments. It was reported that AQP4 was down-regulated by activating protein kinase C (PKC) pathway by hydrogen sulfide or melatonin after MCAO, suggesting PKC pathway is essential for AQP4 down-regulation [56,57]. Mitogen-activated protein kinase (MAPK) pathways include three main members: extracellular signal-regulated kinase (ERK), C-Jun amino-terminal kinase (JNK) and p38-MAPK. There are two studies using astrocyte OGD/reoxygenation model, but the results are a little different. AQP4 was up-regulated by activation of ERK and p38-MAPK pathways [52], while JNK and p38-MAPK pathways were positive in another study [51]. The distinction may be owing to different experimental parameters (OGD 4 h vs. OGD 6 h) and the fact that cross talk exists generally among MAPK family. Therefore, MAPK pathways mainly have an important role in AQP4 up-regulation. 3.3. Effects of AQP4 on Ischemic Edema In the early stage of cerebral ischemia, the decline of cerebral blood flow causing hypoxia results in impairment of Na+/K+ ATPase. The energy failure leads to accumulation of intracellular sodium, which draws water into the cell inducing cytotoxic edema [58]. The development of ischemic cellular damage causes breakdown of BBB, giving rise to leakage of plasma proteins into extracellular space. The involved mechanisms are complex, including reverse pinocytosis, disputed Ca2+ signaling and action of other agents such as vascular endothelial growth factor (VEGF) and matrix metalloproteinases (MMPs) [17]. With the advance of BBB disruption, vasogenic edema occurs and even hemorrhagic conversion appears in some cases. The two types of cerebral edema coexist during the non-acute phase of cerebral ischemia [59] (Figure 2). The effects of AQP4 on brain edema after cerebral ischemia are mainly investigated by AQP4 inhibition models, including AQP4 knockout, AQP4 depolarized distribution and AQP4 gene silencing. Because cytotoxic edema is the predominant type in the early stage, AQP4 inhibits formation of edema based on the knowledge mentioned above. The first report by Manley et al. showed brain edema was decreased in AQP4-deficient mice at 24 h after permanent MCAO [36]. Then several studies reveal similar results in different ischemic models. AQP4 small interfering RNA (siRNA) relieved cellular edema at 6 h after MCAO [60]. Also, it was shown that AQP4 deletion reduced brain edema at 24 h after transient MCAO [61]. Yang and colleagues demonstrated that AQP4 knockdown by siRNA led to reduced brain edema accompanied by a higher apparent diffusion coefficient (ADC) value from 0 h to 12 h after hypoxia–ischemia established by suturing the bilateral carotid arteries in newborn piglets [62]. Moreover, in global cerebral ischemia models, AQP4 knockout alleviated brain water content at 24 h as well as astrocyte swelling in brain slice [63]. Since AQP4 plays dual roles in the two types of edema and the brain edema of cerebral ischemia during non-acute phase is mixed one, the effects of AQP4 during the phase are complex. It was reported brain edema was reduced in α-syntrophin deficient mice at 48 h and 72 h after transient MCAO [64]. Meanwhile, brain water content was increased in wild type mice compared with AQP4 deletion mice at 3 days and 5 days after severe global cerebral ischemia produced by transient four-vessel occlusion [65]. However, a recent study showed AQP4 deletion increase brain edema determined by MRI especially at 3 days and 7 days after transient MCAO [66]. Thus, the roles of AQP4 on brain edema in non-acute cerebral ischemia are complex and possibly related to models and detecting methods. In astrocytes culture, AQP4 siRNA protects against water influx in the formation of astrocyte swelling, while delays water clearance in the resolution of astrocyte swelling after OGD/reoxygenation [67]. In addition, it was reported thrombin preconditioning up-regulated AQP4 with predominant AQP4-M1 isoform at 24 h after MCAO, leading to reduction of brain edema [68]. It was suggested the ratio of AQP4-M23 and AQP4-M1 may be crucial for edema formation and elimination. 3.4. Effects of AQP4 on Ischemic BBB and Neural Cells Injury In addition to the influence of brain edema, AQP4 can elicit other biological effects as mentioned above. As a result, AQP4 may affect ischemic BBB and neural cells injury not only dependent on regulation of brain edema. Since AQP4 is highly concentrated in the important location of BBB, it is regarded to play important roles in maintaining BBB integrity in development and mature individuals [71,72]. However, in a global cerebral ischemic model, AQP4 deletion reduced BBB disruption measured by Evans blue dye extravasation [65]. Similar result was also found in a transient MCAO model [61]. It may result from inhibition of secondary injury to BBB by reduction of brain edema in AQP4-null mice, but dynamic and systematical observation to ischemic BBB injury remains to be carried out in the future. Moreover, it was demonstrated AQP4 knockout improved outcome and neurological function, reduced infarction volume, increased neuronal survival, and blocked apoptosis and inflammatory response after cerebral ischemia, which was consistent with brain edema reduction [61,62,63,65]. However, AQP4 deletion was reported to be beneficial at long term (14 days after MCAO) with neuronal survival improvement and neuroinflammation reduction without a direct effect on edema formation, suggesting a complex role of AQP4 in the ischemic pathophysiological cascades [66]. In an in vitro experiment, AQP4 siRNA also attenuated astrocytes injury induced by OGD/reoxygenation [52,73]. Nevertheless, one study revealed reverse results that AQP4 deletion aggravated inflammation and promoted neuronal loss at 24 and 72 h after MCAO [74]. Meanwhile, in chronic cerebral ischemia (35 days after MCAO), AQP4 knockout had more severe brain atrophy and more neuronal loss as well as impaired astrocyte proliferation and glial scar formation [75]. In summary, the effects of AQP4 on cerebral ischemia may be very complex and include several mechanisms. In the early stage, owing to the inhibition of relatively single cytotoxic edema, AQP4 should exhibit protective effects. As the disease development and long-term existence of mixed edema, effects of AQP4 are very complex, which are determined by the predominance of dual effects on the two types of edema and certain neuroprotective effects. As to chronic ischemia, AQP4 probably contributes to facilitative effects on neurorestoration because of its roles in astrocytes migration and neurogenesis promotion. 4. Intracerebral Hemorrhage (ICH) 4.1. Expression of AQP4 The research related to ICH is only limited to few animal model studies due to the absence of proper cell models. Animal models of ICH contain autologous blood injection, bacterial collagenase injection and spontaneous ICH models, and the former two are the most commonly used [76]. In collagenase models, bacterial collagenase disrupts the basal lamina of blood vessels and causes blood leakage leak into the surrounding tissue. Both have merits and demerits and they differ in ways that influence outcome. However, none of them mimic the pathophysiologic course of human spontaneous ICH, causing relative lag in experimental ICH research [77]. Several articles have revealed AQP4 expression is up-regulated from 3 h after ICH, reaches the peak at 2–5 day, and lasts for at least 14 days, which is not different between autologous blood [78,79,80,81,82,83] and collagenase models [84,85,86]. Meanwhile, it was also reported AQP4 polarity was disturbed in spite of AQP4 up-regulation [87]. Moreover, AQP4 is internalized following ICH and the lysosome is involved in degrading the internalized AQP4 [86]. No research has referred to AQP4 expression in ICH patients. Certain signal transduction pathways are crucial for regulation of AQP4 following ICH. It was reported that nuclear factor κB (NF-κB) participated in AQP4 up-regulation [80]. Furthermore, Chu et al. demonstrated that activation of JNK and ERK pathways, ERK pathway, and JNK and p38-MAPK was responsible for increase of AQP4 by VEGF, granulocyte-colony stimulating factor (G-CSF) and erythropoietin (EPO), respectively [88,89,90]. This indicates MAPK pathways also play important roles in AQP4 up-regulation. 4.2. Effects of AQP4 on Hemorrhagic Edema Brain edema following ICH remains complicated. Studies using animal models revealed autologous blood injected into the brain causes the activation of thrombin, plasminogen activator and urokinase. These substances activate inflammatory cells and disrupt BBB, leading to vasogenic edema. The above mechanism starts at several hours and peaks at several days after ICH. Subsequently, secondary cellular injury due to the substances from CNS cells disruption and red blood cells lysis leads to cytotoxic edema. Thus, these degradation products maintain mixed edema, which lasts about two to three weeks. As a result, ICH results in multiple forms of edema while the predominant type is probably vasogenic [69,70] (Figure 2). In general, brain edema and BBB disruption following ICH are more severe than cerebral ischemia, which usually greatly contribute to ICH-induced neurological deficits. Several studies have showed that the changing trend of AQP4 is parallel with brain edema after ICH [82,83]. However, the effects of AQP4 cannot be determined without further investigation. Tang and colleagues first observed the roles of AQP4 in ICH using AQP4 knockout mice. They focused on brain edema surrounding hematoma and used two methods to measure brain water content and brain specific gravity. The advantage of the latter is that it is possible to obtain reliable results in tissue samples as small as 10–30 mg. Both methods showed AQP4 deletion aggravated brain edema at 1, 3 and 7 days after ICH [78]. Similar results were obtained by Chu et al. [88,89]. Considering AQP4 contributes to clearance of vasogenic edema and this type is predominant in ICH, it is probable that AQP4 mainly acts on elimination of hemorrhagic edema. In addition, brain edema following ICH also has close relation with disruption of AQP4 polarized distribution [87]. As to patient studies, one study on polymorphism in AQP4 genes suggests AQP4 gene variant, single nucleotide polymorphism (SNP) rs1054827, is independently associated with brain edema after ICH [91]. 4.3. Effects of AQP4 on Hemorrhagic BBB and Neural Cells Injury Disruption of BBB is an important pathophysiological change after ICH and contributes to formation of vasogenic brain edema, giving rise to poor prognosis. It was found BBB function measured by Evans blue extravasation was worsened by AQP4 deletion at 1, 3 and 7 days after ICH compared with wild type mice [78,88,89]. As for the morphology of BBB, electron micrographs showed AQP4 deletion resulted in swelling and irregular capillary endothelial cells with opening of tight junction [78,90]. Meanwhile, AQP4 knockout reduced expression of tight junction proteins including occludin, zonula occluden-1 (ZO-1) and claudin-5 [90]. These results suggest AQP4 may have protective effects on BBB disruption after ICH both morphologically and functionally. The presence of AQP4 gene also improves neurological function, increases the survival rate and inhibits neuronal death and apoptosis after ICH [78,88,89,92]. Chu et al. first investigated the mechanisms involved in AQP4’s effects on apoptosis. In this work, AQP4 deletion increased apoptosis and the cell types involved were predominantly neurons and astrocytes. The apoptosis-related proteins including activated caspase-3 and caspase-8 were increased. Meanwhile, higher levels of tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) as well as their receptors were detected in AQP4 knockout mice. The inhibitors of the two cytokines alleviated cells apoptosis after ICH. It suggests AQP4 deletion increases apoptosis following ICH, and the underlying mechanism may be that cytokines, especially TNF-α and IL-1 β, initiate the apoptotic cascade and activates caspase-3 and caspase-8 [92]. Therefore, AQP4 may affect ICH and even other CNS diseases by edema independent of neuroinflammatory pathways. Furthermore, AQP4 can be located in downstream of certain drugs and proteins, thus mediates their effects on ICH. Although several articles reported some factors reduced brain edema and BBB disruption with a decrease of AQP4 [81,85], it is still hard to conclude the effects are associated with AQP4. Chu and colleagues tested the effects of VEGF, G-CSF and EPO on brain edema, BBB permeability and cells injury following ICH and examined whether they were AQP4 dependent using AQP4 deletion mice. They found these effects were associated with AQP4 [88,89,90]. Thissuggests AQP4 can mediate other neuroprotective factors’ effects as downstream pathways. 5. Subarachnoid Hemorrhage (SAH) SAH is a devastating subtype of stroke with high mortality, which is mostly followed by aneurysm rupture [93]. Early and delayed brain injuries are included in the pathophysiology. Early brain injury occurs immediately after SAH and lasts up to 72 h. Brain edema appears at early phase of SAH due to disruption of BBB via multiple mechanisms causing vasogenic edema, which is probably the major type [94]. Meanwhile, cytotoxic edema is detected also at early stage owing to ischemic insult [95]. Delayed brain injury mainly results in vasospasm and brain edema in this stage is similar to cerebral ischemia. Research on SAH patients shows AQP4 is up-regulated on the astrocytic processes with loss of polarization [96]. Till now, studies in SAH models mainly focus on effects of AQP4 on early brain injury. AQP4 is increased at 6 h after SAH and maintains high levels within 72 h [97,98,99,100,101]. Moreover, one study indicates an increase of AQP4 at 7 days after SAH, which is located in the phase of delayed brain injury [102]. Thus, AQP4 may be up-regulated early after SAH and last for a relatively long time. AQP4 may play a dual role in brain edema after SAH due to the mixed type of edema. Tait et al. demonstrated AQP4 knockout markedly reduced brain edema and BBB disruption at 6 h and 24 h after SAH with an increase of intracranial pressure and aggravation of neurological function [98]. In addition, it was shown that there was no improvement in neurological deficits and neuroinflammation at 7 days after SAH in AQP4 deletion mice compared with wide type control mice [103]. Therefore, AQP4 may play opposite roles in the early and delayed brain injury and further research using AQP4-null mice is urgently required. 6. Conclusions As the most abundant AQP in the CNS, the expression of AQP4 is increased in three kinds of cerebrovascular diseases, including cerebral ischemia, ICH and SAH. The direct effect on the related brain edema is the basis of the action of AQP4 for these diseases. However, vasogenic edema and cytotoxic edema may simultaneously appear in these diseases and the roles of AQP4 in the two types of edema are opposite. Meanwhile, AQP4 also shows other biological effects in addition to the effect on brain edema. Thus, the roles of AQP4 in these diseases are relatively complicated, which may be determined by the balance between its effect on the predominant brain edema type and other biological effects. Furthermore, it may also play different roles in alternative phase of the diseases. Thus, the effects of AQP4 on cerebrovascular diseases remain to be investigated, which may become the theoretical basis of AQP4 regulation treatment. AQP4 plays an important role in the formation and clearance of brain edema, and appropriate regulation of AQP4 may treat brain edema from perspectives of mechanism as a remedy of the disadvantages of the current common treatment, such as dehydrant or surgery. Although AQP4 gene knockout is an excellent tool to study AQP4, it cannot be used clinically. Furthermore, AQP4 gene silencing or over-expression is only locally applied, and invasive procedures are required. Therefore, the highly selective antagonist or agonist of AQP4 may be a better choice. It is well-known that neuromyelitisoptica (NMO) is characterized by autoantibodies directed against AQP4. However, purified NMO-IgG injected intravenously increased brain edema and infarct size at 24 h after MCAO [103]. Moreover, Igarashi et al. developed an AQP4 inhibitor TGN-020 and found intraperitoneal injection of TGN-020 reduced brain edema and infarct size at 24 h after MCAO. Nevertheless, no other research has duplicated their results [104]. Thus, a highly selective antagonist or agonist of AQP4 that can be used systemically remains to be further developed, and it promises to become a novel, effective measure of treating cerebrovascular diseases. Acknowledgments This research was supported by grants from the National Natural Science Foundation (No. 81500998; and 81571109)and National Basic Research Program (973 Program, No. 2012CB518602) of China. Author Contributions Heling Chu, Chuyi Huang, Yuping Tang and Qiang Dong were responsible for the overall structure and wrote the review. Hongyan Ding, Jing Dong, Zidan Gao and Xiaobo Yang contributed to the writing of the review. Conflicts of Interest The authors declare no conflict of interest. Figure 1 The structure and distribution of aquaporin-4 (AQP4). (A) AQP4 has six transmembrane domains (1–6) and five connecting loops (A–E). Loops B and E contain highly conserved “NPA” motifs (hemipores) that overlap midway creating a highly selective water pore; (B) AQP4 is polarized at the astrocyte processes facing cerebrospinal fluid (CSF)–brain and blood–brain barrier. Ependymal cells have basolateral expression of AQP4 [13,14,17]. Figure 2 Mechanisms of edema formation and the effects of AQP4 in cerebral ischemia and intracerebral hemorrhage (ICH). In cerebral ischemia, AQP4 promotes water entry into perivascular astrocyte end-feet resulting in cytotoxic edema. As further ischemic cellular damage evolves, the mechanism shifts into vasogenic edema. During ICH, release of multiple toxic substancescauses disruption of blood–brain barrier (BBB), which gives rise to vasogenic edema. AQP4 facilitates the reabsorption of edema fluid from the extracellular space. Then secondary cellular injury leads to energy failure causing cytotoxic edema. Thus, mixed brain edema exists in the two types of stroke but the predominant type is different [17,58,69,70]. ==== Refs References 1. Preston G.M. Carroll T.P. Guggino W.B. Agre P. Appearance of water channels in Xenopus oocytes expressing red-cell Chip28 protein Science 1992 256 385 387 10.1126/science.256.5055.385 1373524 2. Gonen T. Walz T. The structure of aquaporins Q. Rev. Biophys. 2006 39 361 396 10.1017/S0033583506004458 17156589 3. Xiao M. Hu G. Involvement of aquaporin 4 in astrocyte function and neuropsychiatric disorders CNS Neurosci. Ther. 2014 20 385 390 10.1111/cns.12267 24712483 4. Badaut J. Fukuda A.M. Jullienne A. Petry K.G. Aquaporin and brain diseases Biochim. Biophys. Acta 2014 1840 1554 1565 10.1016/j.bbagen.2013.10.032 24513456 5. Jung J.S. Bhat R.V. Preston G.M. Guggino W.B. Baraban J.M. Agre P. Molecular characterization of an aquaporin cDNA from brain: Candidate osmoreceptor and regulator of water balance Proc. Natl. Acad. Sci. USA 1994 91 13052 13056 10.1073/pnas.91.26.13052 7528931 6. Frigeri A. Gropper M.A. Umenishi F. Kawashima M. Brown D. Verkman A.S. Localization of MIWC and GLIP water channel homologs in neuromuscular, epithelial and glandular tissues J. Cell Sci. 1995 108 2993 3002 8537439 7. Nielsen S. Nagelhus E.A. Amiry-Moghaddam M. Bourque C. Agre P. Ottersen O.P. Specialized membrane domains for water transport in glial cells: High-resolution immunogoldcytochemistry of aquaporin-4 in rat brain J. Neurosci. 1997 17 171 180 8987746 8. Thrane A.S. Rappold P.M. Fujita T. Torres A. Bekar L.K. Takano T. Peng W. Wang F. Rangroo Thrane V. Enger R. Critical role of aquaporin-4 (AQP4) in astrocyte Ca2+ signaling events elicited by cerebral edema Proc. Natl. Acad. Sci. USA 2011 108 846 851 10.1073/pnas.1015217108 21187412 9. Skucas V.A. Mathews I.B. Yang J. Cheng Q. Treister A. Duffy A.M. Verkman A.S. Hempstead B.L. Wood M.A. Binder D.K. Impairment of select forms of spatial memory and neurotrophin-dependent synaptic plasticity by deletion of glial aquaporin-4 J. Neurosci. 2011 31 6392 6397 10.1523/JNEUROSCI.6249-10.2011 21525279 10. Saadoun S. Papadopoulos M.C. Watanabe H. Yan D. Manley G.T. Verkman A.S. Involvement of aquaporin-4 in astroglial cell migration and glial scar formation J. Cell Sci. 2005 118 5691 5698 10.1242/jcs.02680 16303850 11. Kong H. Sha L.L. Fan Y. Xiao M. Ding J.H. Wu J. Hu G. Requirement of AQP4 for antidepressiveefflciency of fluoxetine: Implication in adult hippocampal neurogenesis Neuropsychopharmacology 2009 34 1263 1276 10.1038/npp.2008.185 18923397 12. Li L. Zhang H. Varrin-Doyer M. Zamvil S.S. Verkman A.S. Proinflammatory role of aquaporin-4 in autoimmune neuroinflammation FASEB J. 2011 25 1556 1566 10.1096/fj.10-177279 21257712 13. Papadopoulos M.C. Verkman A.S. Aquaporin water channels in the nervous system Nat. Rev. Neurosci. 2013 14 265 277 10.1038/nrn3468 23481483 14. Nagelhus E.A. Ottersen O.P. Physiological roles of aquaporin-4 in brain Physiol. Rev. 2013 93 1543 1562 10.1152/physrev.00011.2013 24137016 15. Yang B. Brown D. Verkman A.S. The mercurial insensitive water channel (AQP-4) forms orthogonal arrays in stably transfected Chinese hamster ovary cells J. Biol. Chem. 1996 271 4577 4580 8617713 16. Papadopoulos M.C. Verkman A.S. Aquaporin 4 and neuromyelitisoptica Lancet Neurol. 2012 11 535 544 10.1016/S1474-4422(12)70133-3 22608667 17. Zador Z. Stiver S. Wang V. Manley G.T. Role of aquaporin-4 in cerebral edema and stroke Handb. Exp. Pharmacol. 2009 190 159 170 19096776 18. Rash J.E. Yasumura T. Hudson C.S. Agre P. Nielsen S. Direct immunogold labeling of aquaporin-4 in square arrays of astrocyte and ependymocyte plasma membranes in rat brain and spinal cord Proc. Natl. Acad. Sci. USA 1998 95 11981 11986 10.1073/pnas.95.20.11981 9751776 19. Tomás-Camardiel M. Venero J.L. de Pablos R.M. Rite I. Machado A. Cano J. In vivo expression of aquaporin-4 by reactive microglia J. Neurochem. 2004 91 891 899 10.1111/j.1471-4159.2004.02759.x 15525343 20. Eilert-Olsen M. Haj-Yasein N.N. Vindedal G.F. Enger R. Gundersen G.A. Hoddevik E.H. Petersen P.H. Haug F.M. Skare Ø. Adams M.E. Deletion of aquaporin-4 changes the perivascular glial protein scaffold without disrupting the brain endothelial barrier Glia 2012 60 432 440 10.1002/glia.22277 22131281 21. Neely J.D. Amiry-Moghaddam M. Ottersen O.P. Froehner S.C. Agre P. Adams M.E. Syntrophin-dependent expression and localization of Aquaporin-4 water channel protein Proc. Natl. Acad. Sci. USA 2001 98 14108 14113 10.1073/pnas.241508198 11717465 22. Enger R. Gundersen G.A. Haj-Yasein N.N. Eilert-Olsen M. Thoren A.E. Vindedal G.F. Petersen P.H. Skare Ø. Nedergaard M. Ottersen O.P. Molecular scaffolds underpinning macroglial polarization: an analysis of retinal Müller cells and brain astrocytes in mouse Glia 2012 60 2018 2026 10.1002/glia.22416 22987438 23. Noell S. Fallier-Becker P. Beyer C. Kröger S. Mack A.F. Wolburg H. Effects of agrin on the expression and distribution of the water channel protein aquaporin-4 and volume regulation in cultured astrocytes Eur. J. Neurosci. 2007 26 2109 2118 10.1111/j.1460-9568.2007.05850.x 17927773 24. Ma T. Yang B. Gillespie A. Carlson E.J. Epstein C.J. Verkman A.S. Generation and phenotype of a transgenic knockout mouse lacking the mercurial-insensitive water channel aquaporin-4 J. Clin. Investig. 1997 100 957 962 10.1172/JCI231 9276712 25. Fan Y. Zhang J. Sun X.L. Gao L. Zeng X.N. Ding J.H. Cao C. Niu L. Hu G. Sex- and region-specific alterations of basal amino acid and monoamine metabolism in the brain of aquaporin-4 knockout mice J. Neurosci. Res. 2005 82 458 464 10.1002/jnr.20664 16237719 26. Haj-Yasein N.N. Vindedal G.F. Eilert-Olsen M. Gundersen G.A. Skare Ø. Laake P. Klungland A. Thorén A.E. Burkhardt J.M. Ottersen O.P. Glial-conditional deletion of aquaporin-4 (AQP4) reduces blood-brain water uptake and confers barrier function on perivascular astrocyte endfeet Proc. Natl. Acad. Sci. USA 2011 108 17815 17820 10.1073/pnas.1110655108 21990350 27. Saadoun S. Tait M.J. Reza A. Davies D.C. Bell B.A. Verkman A.S. Papadopoulos M.C. AQP4 gene deletion in mice does not alter blood-brain barrier integrity or brain morphology Neuroscience 2009 161 764 772 10.1016/j.neuroscience.2009.03.069 19345723 28. Amiry-Moghaddam M. Williamson A. Palomba M. Eid T. de Lanerolle N.C. Nagelhus E.A. Adams M.E. Froehner S.C. Agre P. Ottersen O.P. Delayed K+ clearance associated with aquaporin-4 mislocalization: phenotypic defects in brains of α-syntrophin-null mice Proc. Natl. Acad. Sci. USA 2003 100 13615 13620 10.1073/pnas.2336064100 14597704 29. Amiry-Moghaddam M. Xue R. Haug F.M. Neely J.D. Bhardwaj A. Agre P. Adams M.E. Froehner S.C. Mori S. Ottersen O.P. α-Syntrophin deletion removes the perivascular but not endothelial pool of aquaporin-4 at the blood-brain barrier and delays the development of brain edema in an experimental model of acute hyponatremia FASEB J. 2004 18 542 544 10.1096/fj.03-0869fje 14734638 30. Raslan A. Bhardwaj A. Medical management of cerebral edema Neurosurg. Focus 2007 22 E12 10.3171/foc.2007.22.5.13 17613230 31. Qureshi A.I. Suarez J.I. Use of hypertonic saline solutions in treatment of cerebral edema and intracranial hypertension Crit. Care Med. 2000 28 3301 3313 10.1097/00003246-200009000-00032 11008996 32. Kimelberg H.K. Current concepts of brain edema. Review of laboratory investigations J. Neurosurg. 1995 83 1051 1059 10.3171/jns.1995.83.6.1051 7490620 33. Papadopoulos M.C. Binder D.K. Verkman A.S. Enhanced macromolecular diffusion in brain extracellular space in mouse models of vasogenic edema measured by cortical surface photobleaching FASEB J. 2005 19 425 427 10.1096/fj.04-2834fje 15596484 34. Iliff J.J. Wang M. Liao Y. Plogg B.A. Peng W. Gundersen G.A. Benveniste H. Vates G.E. Deane R. Goldman S.A. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β Sci. Transl. Med. 2012 4 147ra111 10.1126/scitranslmed.3003748 22896675 35. Thrane A.S. Rangroo Thrane V. Nedergaard M. Drowning stars: reassessing the role of astrocytes in brain edema Trends Neurosci. 2014 37 620 628 10.1016/j.tins.2014.08.010 25236348 36. Manley G.T. Fujimura M. Ma T. Noshita N. Filiz F. Bollen A.W. Chan P. Verkman A.S. Aquaporin-4 deletion in mice reduces brain edema after acute water intoxication and ischemicstroke Nat. Med. 2000 6 159 163 10.1038/72256 10655103 37. Papadoplos M.C. Verkman A.S. Aquaporin-4 gene disruption in mice reduces brain swelling and mortality in pneumococcal meningitis J. Biol. Chem. 2005 280 13906 13912 10.1074/jbc.M413627200 15695511 38. Papadopoulos M.C. Manley G.T. Krishna S. Verkman A.S. Aquaporin-4 facilitates reabsorption of excess fluid in vasogenic brain edema FASEB J. 2004 18 1291 1293 10.1096/fj.04-1723fje 15208268 39. Auguste K.I. Jin S. Uchida K. Yan D. Manley G.T. Papadopoulos M.C. Verkman A.S. Greatly impaired migration of implanted aquaporin-4-deficient astroglial cells in mouse brain toward a site of injury FASEB J. 2007 21 108 116 10.1096/fj.06-6848com 17135365 40. Binder D.K. Nagelhus E.A. Ottersen O.P. Aquaporin-4 and epilepsy Glia 2012 60 1203 1214 10.1002/glia.22317 22378467 41. Binder D.K. Oshio K. Ma T. Verkman A.S. Manley G.T. Increased seizure threshold in mice lacking aquaporin-4 water channels Neuroreport 2004 15 259 262 10.1097/00001756-200402090-00009 15076748 42. Binder D.K. Yao X. Zador Z. Sick T.J. Verkman A.S. Manley G.T. Increased seizure duration and slowed potassium kinetics in mice lacking aquaporin-4 water channels Glia 2006 53 631 636 10.1002/glia.20318 16470808 43. Haj-Yasein N.N. Bugge C.E. Jensen V. Østby I. Ottersen O.P. Hvalby Ø. Nagelhus E.A. Deletion of aquaporin-4 increases extracellular K(+) concentration during synaptic stimulation in mouse hippocampus Brain Struct. Funct. 2015 220 2469 2474 10.1007/s00429-014-0767-z 24744149 44. Scharfman H.E. Binder D.K. Aquaporin-4 water channels and synaptic plasticity in the hippocampus Neurochem. Int. 2013 63 702 711 10.1016/j.neuint.2013.05.003 23684954 45. Fan Y. Liu M. Wu X. Wang F. Ding J. Chen J. Hu G. Aquaporin-4 promotes memory consolidation in Morris water maze Brain Struct. Funct. 2013 218 39 50 10.1007/s00429-011-0373-2 22193336 46. Kong H. Fan Y. Xie J. Ding J. Sha L. Shi X. Sun X. Hu G. AQP4 knockout impairs proliferation, migration and neuronal differentiation of adult neural stem cells J. Cell Sci. 2008 121 4029 4036 10.1242/jcs.035758 19033383 47. Lu H. Hu H. He Z.P. Reperfusion of the rat brain tissues following acute ischemia: The correlation among diffusion-weighted imaging, histopathology, and aquaporin-4 expression Chin. Med. J. 2011 124 3148 3153 22040570 48. Badaut J. Ashwal S. Tone B. Regli L. Tian H.R. Obenaus A. Temporal and regional evolution of aquaporin-4 expression and magnetic resonance imaging in a rat pup model of neonatal stroke Pediatr. Res. 2007 62 248 254 10.1203/PDR.0b013e3180db291b 17622964 49. Ribeiro Mde C. Hirt L. Bogousslavsky J. Regli L. Badaut J. Time course of aquaporin expression after transient focal cerebral ischemia in mice J. Neurosci. Res. 2006 83 1231 1240 10.1002/jnr.20819 16511868 50. Akdemir G. Kaymaz F. Gursoy-ozdemir Y. Akalan N. Akdemir E.S. The time course changes in expression of aquaporin 4 and aquaporin 1 following global cerebral ischemic edema in rat Surg. Neurol. Int. 2016 7 4 10.4103/2152-7806.173316 26862443 51. Nito C. Kamada H. Endo H. Narasimhan P. Lee Y.S. Chan P.H. Involvement of mitogen-activated protein kinase pathways in expression of the water channel protein aquaporin-4 after ischemia in rat cortical astrocytes J. Neurotrauma 2012 29 2404 2412 10.1089/neu.2012.2430 22676888 52. Qi L.L. Fang S.H. Shi W.Z. Huang X.Q. Zhang X.Y. Lu Y.B. Zhang W.P. Wei E.Q. CysLT2 receptor-mediated AQP4 up-regulation is involved in ischemic-like injury through activation of ERK and p38 MAPK in rat astrocytes Life Sci. 2011 88 50 56 10.1016/j.lfs.2010.10.025 21055410 53. Aoki K. Uchihara T. Tsuchiya K. Nakamura A. Ikeda K. Wakayama Y. Enhanced expression of aquaporin 4 in human brain with infarction Acta Neuropathol. 2003 106 121 124 10.1007/s00401-003-0709-y 12715185 54. Satoh J. Tabunoki H. Yamamura T. Arima K. Konno H. Human astrocytes express aquaporin-1 and aquaporin-4 in vitro and in vivo Neuropathology 2007 27 245 256 10.1111/j.1440-1789.2007.00774.x 17645239 55. Stokum J.A. Mehta R.I. Ivanova S. Yu E. Gerzanich V. Simard J.M. Heterogeneity of aquaporin-4 localization and expression after focal cerebral ischemia underlies differences in white versus grey matter swelling Acta Neuropathol. Commun. 2015 3 61 10.1186/s40478-015-0239-6 26419740 56. Wei X. Zhang B. Cheng L. Chi M. Deng L. Pan H. Yao X. Wang G. Hydrogen sulfide induces neuroprotection against experimental stroke in rats by down-regulation of AQP4 via activating PKC Brain Res. 2015 1622 292 299 10.1016/j.brainres.2015.07.001 26168888 57. Bhattacharya P. Pandey A.K. Paul S. Patnaik R. Melatonin renders neuroprotection by protein kinase C mediated aquaporin-4 inhibition in animal model of focal cerebral ischemia Life Sci. 2014 100 97 109 10.1016/j.lfs.2014.01.085 24530291 58. Simard J.M. Kent T.A. Chen M. Tarasov K.V. Gerzanich V. Brain oedema in focal ischaemia: molecular pathophysiology and theoretical implications Lancet Neurol. 2007 6 258 268 10.1016/S1474-4422(07)70055-8 17303532 59. Wang W.W. Xie C.L. Zhou L.L. Wang G.S. The function of aquaporin4 in ischemic brain edema Clin. Neurol. Neurosurg. 2014 127 5 9 10.1016/j.clineuro.2014.09.012 25306413 60. He Z.P. Lu H. Aquaporin-4 gene silencing protects injured neurons after early cerebral infarction Neural Regen. Res. 2015 10 1082 1087 26330830 61. Yao X. Derugin N. Manley G.T. Verkman A.S. Reduced brain edema and infarct volume in aquaporin-4 deficient mice after transient focal cerebral ischemia Neurosci. Lett. 2015 584 368 372 10.1016/j.neulet.2014.10.040 25449874 62. Yang C. Liu Z. Li H. Zhai F. Liu J. Bian J. Aquaporin-4 knockdown ameliorates hypoxic-ischemic cerebral edema in newborn piglets IUBMB Life 2015 67 182 190 10.1002/iub.1356 25857369 63. Katada R. Akdemir G. Asavapanumas N. Ratelade J. Zhang H. Verkman A.S. Greatly improved survival and neuroprotection in aquaporin-4-knockout mice following global cerebral ischemia FASEB J. 2014 28 705 714 10.1096/fj.13-231274 24186965 64. Liu X. Zhang W. Alkayed N.J. Froehner S.C. Adams M.E. Amiry-Moghaddam M. Ottersen O.P. Hurn P.D. Bhardwaj A. Lack of sex-linked differences in cerebral edema and aquaporin-4 expression after experimental stroke J. Cereb. Blood Flow. Metab. 2008 28 1898 1906 10.1038/jcbfm.2008.83 18648381 65. Akdemir G. Ratelade J. Asavapanumas N. Verkman A.S. Neuroprotective effect of aquaporin-4 deficiency in a mouse model of severe global cerebral ischemia produced by transient 4-vessel occlusion Neurosci. Lett. 2014 574 70 75 10.1016/j.neulet.2014.03.073 24717641 66. Hirt L. Fukuda A.M. Ambadipudi K. Rashid F. Binder D. Verkman A. Ashwal S. Obenaus A. Badaut J. Improved long-term outcome after transient cerebral ischemia in aquaporin-4 knockout mice J. Cereb. Blood Flow. Metab. 2016 in press 10.1177/0271678X15623290 26767580 67. Fu X. Li Q. Feng Z. Mu D. The roles of aquaporin-4 in brain edema following neonatal hypoxia ischemia and reoxygenation in a cultured rat astrocyte model Glia 2007 55 935 941 10.1002/glia.20515 17437301 68. Hirt L. Ternon B. Price M. Mastour N. Brunet J.F. Badaut J. Protective role of early aquaporin 4 induction against postischemic edema formation J. Cereb. Blood Flow. Metab. 2009 29 423 433 10.1038/jcbfm.2008.133 18985050 69. Reggio E. Loretol C. Aquaporin and vascular diseases Curr. Neuropharmacol. 2010 8 105 111 21119881 70. Xi G.H. Keep R.F. Hoff J.T. Mechanisms of brain injury after intracerebral haemorrhage Lancet Neurol. 2006 5 53 63 10.1016/S1474-4422(05)70283-0 16361023 71. Nico B. Frigeri A. Nicchia G.P. Quondamatteo F. Herken R. Errede M. Ribatti D. Svelto M. Roncali L. Role of aquaporin-4 water channel in the development and integrity of the blood-brain barrier J. Cell Sci. 2001 114 1297 1307 11256996 72. Zhou J.P. Kong H. Hua X.D. Xiao M. Ding J. Hu G. Altered blood-brain barrier integrity in adult aquaporin-4 knockout mice Neuroreport 2008 19 1 5 10.1097/WNR.0b013e3282f2b4eb 18281883 73. Tang G. Liu Y. Zhang Z. Lu Y. Wang Y. Huang J. Li Y. Chen X. Gu X. Wang Y. Mesenchymal stem cells maintain blood-brain barrier integrity by inhibiting aquaporin-4 upregulation after cerebral ischemia Stem Cells 2014 32 3150 3162 10.1002/stem.1808 25100404 74. Shi W.Z. Zhao C.Z. Zhao B. Shi Q.J. Zhang L.H. Wang Y.F. Fang S.H. Lu Y.B. Zhang W.P. Wei E.Q. Aggravated inflammation and increased expression of cysteinyl leukotriene receptors in the brain after focal cerebral ischemia in AQP4-deficient mice Neurosci. Bull. 2012 28 680 692 10.1007/s12264-012-1281-z 23132680 75. Shi W.Z. Qi L.L. Fang S.H. Lu Y.B. Zhang W.P. Wei E.Q. Aggravated chronic brain injury after focal cerebral ischemia in aquaporin-4-deficient mice Neurosci. Lett. 2012 520 121 125 10.1016/j.neulet.2012.05.052 22634625 76. Alharbi B.M. Tso M.K. Macdonald R.L. Animal models of spontaneous intracerebral hemorrhage Neurol. Res. 2016 38 448 455 10.1080/01616412.2016.1144671 27098341 77. MacLellan C.L. Silasi G. Auriat A.M. Colbourne F. Rodent models of intracerebral hemorrhage Stroke 2010 41 S95 S98 10.1161/STROKEAHA.110.594457 20876518 78. Tang Y. Wu P. Su J. Xiang J. Cai D. Dong Q. Effects of Aquaporin-4 on edema formation following intracerebral hemorrhage Exp. Neurol. 2010 223 485 495 10.1016/j.expneurol.2010.01.015 20132816 79. Fang J. Li H. Li G. Wang L. Effect of hyperbaric oxygen preconditioning on peri-hemorrhagic focal edema and aquaporin-4 expression Exp. Ther. Med. 2015 10 699 704 10.3892/etm.2015.2539 26622378 80. Wang B.F. Cui Z.W. Zhong Z.H. Sun Y.H. Sun Q.F. Yang G.Y. Bian L.G. Curcumin attenuates brain edema in mice with intracerebral hemorrhage through inhibition of AQP4 and AQP9 expression Acta Pharmacol. Sin. 2015 36 939 948 10.1038/aps.2015.47 26119880 81. Sun Z. Zhao Z. Zhao S. Sheng Y. Zhao Z. Gao C. Li J. Liu X. Recombinant hirudin treatment modulates aquaporin-4 and aquaporin-9 expression after intracerebral hemorrhage in vivo Mol. Biol. Rep. 2009 36 1119 1127 10.1007/s11033-008-9287-3 18574711 82. Qing W.G. Dong Y.Q. Ping T.Q. Lai L.G. Fang L.D. Min H.W. Xia L. Heng P.Y. Brain edema after intracerebral hemorrhage in rats: The role of iron overload and aquaporin 4 J. Neurosurg. 2009 110 462 468 10.3171/2008.4.JNS17512 19025353 83. Wu H. Zhang Z. Li Y. Zhao R. Li H. Song Y. Qi J. Wang J. Time course of upregulation of inflammatory mediators in the hemorrhagic brain in rats: Correlation with brain edema Neurochem. Int. 2010 57 248 253 10.1016/j.neuint.2010.06.002 20541575 84. Yang Y. Zhang Y. Wang Z. Wang S. Gao M. Xu R. Liang C. Zhang H. Attenuation of acute phase injury in rat intracranial hemorrhage by cerebrolysin that inhibits brain edema and inflammatory response Neurochem. Res. 2016 41 748 757 10.1007/s11064-015-1745-4 26498936 85. Zhong Z. Wang B. Dai M. Sun Y. Sun Q. Yang G. Bian L. Carvacrol alleviates cerebral edema by modulating AQP4 expression after intracerebral hemorrhage in mice Neurosci. Lett. 2013 555 24 29 10.1016/j.neulet.2013.09.023 24051341 86. Xu J. Qiu G.P. Huang J. Zhang B. Sun S.Q. Gan S.W. Lu W.T. Wang K.J. Huang S.Q. Zhu S.J. Internalization of aquaporin-4 after collagenase-induced intracerebral hemorrhage Anat. Rec. 2015 298 554 561 10.1002/ar.23055 25257965 87. Qiu G.P. Xu J. Zhuo F. Sun S.Q. Liu H. Yang M. Huang J. Lu W.T. Huang S.Q. Loss of AQP4 polarized localization with loss of β-dystroglycan immunoreactivity may induce brain edema following intracerebral hemorrhage Neurosci. Lett. 2015 588 42 48 10.1016/j.neulet.2014.12.053 25545558 88. Chu H. Tang Y. Dong Q. Protection of vascular endothelial growth factor to brain edema following intracerebral hemorrhage and its involved mechanisms: effect of aquaporin-4 PLoS ONE 2013 8 1249 10.1371/journal.pone.0066051 23805198 89. Chu H. Tang Y. Dong Q. Protection of granulocyte-colony stimulating factor to hemorrhagic brain injuries and its involved mechanisms: effects of vascular endothelial growth factor and aquaporin-4 Neuroscience 2014 260 59 72 10.1016/j.neuroscience.2013.12.017 24355496 90. Chu H. Ding H. Tang Y. Dong Q. Erythropoietin protects against hemorrhagic blood-brain barrier disruption through the effects of aquaporin-4 Lab. Investig. 2014 94 1042 1053 10.1038/labinvest.2014.84 24978642 91. Appelboom G. Bruce S. Duren A. Piazza M. Monahan A. Christophe B. Zoller S. LoPresti M. Connolly E.S. Aquaporin-4 gene variant independently associated with oedema after intracerebral haemorrhage Neurol. Res. 2015 37 657 661 10.1179/1743132815Y.0000000047 26000774 92. Chu H. Xiang J. Wu P. Su J. Ding H. Tang Y. Dong Q. The role of aquaporin 4 in apoptosis after intracerebral hemorrhage J. Neuroinflamm. 2014 11 184 10.1186/s12974-014-0184-5 25359421 93. Connolly E.J. Rabinstein A.A. Carhuapoma J.R. Derdeyn C.P. Dion J. Higashida R.T. Hoh B.L. Kirkness C.J. Naidech A.M. Ogilvy C.S. Guidelines for the management of aneurismal subarachnoid hemorrhage: A guideline for healthcare professionals from the American Heart Association/American Stroke Association Stroke 2012 43 1711 1737 10.1161/STR.0b013e3182587839 22556195 94. Cahill J. Calvert J.W. Zhang J.H. Mechanisms of early brain injury after subarachnoid hemorrhage J. Cereb. Blood Flow. Metab. 2006 26 1341 1353 10.1038/sj.jcbfm.9600283 16482081 95. Busch E. Beaulieu C. de Crespigny A. Moseley M.E. Diffusion MR imaging during acute subarachnoid hemorrhage in rats Stroke 1998 29 2155 2161 10.1161/01.STR.29.10.2155 9756598 96. Badaut J. Brunet J.F. Grollimund L. Hamou M.F. Magistretti P.J. Villemure J.G. Regli L. Aquaporin 1 and aquaporin 4 expression in human brain after subarachnoid hemorrhage and in peritumoral tissue Acta. Neurochir. Suppl. 2003 86 495 498 14753493 97. Tait M.J. Saadoun S. Bell B.A. Verkman A.S. Papadopoulos M.C. Increased brain edema in aqp4-null mice in an experimental model of subarachnoid hemorrhage Neuroscience 2010 167 60 67 10.1016/j.neuroscience.2010.01.053 20132873 98. Cao S. Zhu P. Yu X. Chen J. Li J. Yan F. Wang L. Yu J. Chen G. Hydrogen sulfide attenuates brain edema in early brain injury after subarachnoid hemorrhage in rats: Possible involvement of MMP-9 induced blood-brain barrier disruption and AQP4 expression Neurosci. Lett. 2016 621 88 97 10.1016/j.neulet.2016.04.018 27080433 99. Wang Z. Meng C.J. Shen X.M. Shu Z. Ma C. Zhu G.Q. Liu H.X. He W.C. Sun X.B. Huo L. Zhang J. Chen G. Potential contribution of hypoxia-inducible factor-1α, aquaporin-4, and matrix metalloproteinase-9 to blood-brain barrier disruption and brain edema after experimental subarachnoid hemorrhage J. Mol. Neurosci. 2012 48 273 280 10.1007/s12031-012-9769-6 22528459 100. Shi S.S. Zhang H.B. Wang C.H. Yang W.Z. Liang R.S. Chen Y. Tu X.K. Propofol attenuates early brain injury after subarachnoid hemorrhage in rats J. Mol. Neurosci. 2015 57 538 545 10.1007/s12031-015-0634-2 26342279 101. Chen J.H. Yang L.K. Chen L. Wang Y.H. Wu Y. Jiang B.J. Zhu J. Li P.P. Atorvastatin ameliorates early brain injury after subarachnoid hemorrhage via inhibition of AQP4 expression in rabbits Int. J. Mol. Med. 2016 37 1059 1066 10.3892/ijmm.2016.2506 26935263 102. Luo C. Yao X. Li J. He B. Liu Q. Ren H. Liang F. Li M. Lin H. Peng J. Yuan T.F. Pei Z. Su H. Paravascular pathways contribute to vasculitis and neuroinflammation after subarachnoid hemorrhage independently of glymphatic control Cell Death Dis. 2016 7 e2160 10.1038/cddis.2016.63 27031957 103. Juenemann M. Braun T. Doenges S. Nedelmann M. Mueller C. Bachmann G. Singh P. Blaes F. Gerriets T. Tschernatsch M. Aquaporin-4 autoantibodies increase vasogenic edema formation and infarct size in a rat stroke model BMC Immunol. 2015 16 1249 10.1186/s12865-015-0087-y 25986484 104. Igarashi H. Huber V.J. Tsujita M. Nakada T. Pretreatment with a novel aquaporin 4 inhibitor, TGN-020, significantly reduces ischemic cerebral edema Neurol. Sci. 2011 32 113 116 10.1007/s10072-010-0431-1 20924629

PMC005xxxxxx/PMC5000648.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081250ijms-17-01250CommunicationA Novel Isothermal Assay of Borrelia burgdorferi by Recombinase Polymerase Amplification with Lateral Flow Detection Liu Wei 12Liu Hui-Xin 12Zhang Lin 12Hou Xue-Xia 12Wan Kang-Lin 12Hao Qin 12*Lin Li Academic Editor1 State Key Laboratory of Infectious Diseases Prevention and Control, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing 102206, China; [email protected] (W.L.); [email protected] (H.-X.L.); [email protected] (L.Z.); [email protected] (X.-X.H.); [email protected] (K.-L.W.)2 Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Hangzhou 310003, China* Correspondence: [email protected]; Tel.: +86-10-5890-077203 8 2016 8 2016 17 8 125003 5 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).A novel isothermal detection for recombinase polymerase amplification with lateral flow (LF-RPA) was established for Borrelia burgdorferi (B. burgdorferi) detection in this study. This assay with high sensitivity and specificity can get a visible result without any additional equipment in 30 min. We designed a pair of primers according to recA gene of B. burgdorferi strains and a methodology evaluation was performed. The results showed that the RPA assay based on the recA gene was successfully applied in B. burgdorferi detection, and its specific amplification was only achieved from the genomic DNA of B. burgdorferi. The detection limit of the new assay was about 25 copies of the B. burgdorferi genomic DNA. Twenty Lyme borreliosis patients’ serum samples were detected by LF-RPA assay, real-time qPCR and nested-PCR. Results showed the LF-RPA assay is more effective than nested-PCR for its shorter reaction time and considerably higher detection rate. This method is of great value in clinical rapid detection for Lyme borreliosis. Using the RPA assay might be a megatrend for DNA detection in clinics and endemic regions. Borrelia burgdorferiDNA amplificationRPArapid detection ==== Body 1. Introduction Lyme borreliosis (LB) is a zoonotic disease caused by tick-borne infection. The pathogen of Lyme borreliosis is Borrelia burgdorferi (B. burgdorferi) sensu lato which has complex genotypes around the world. Nowadays, the golden standard for laboratory diagnosis of Lyme borreliosis remains the culture of samples from patients. However, this method is not suitable for clinical diagnosis for its low detection rate [1,2]. In clinical practice, Lyme disease can be confirmed by the manifestation of erythema migrans (EM), a specific symptom for Lyme disease, together with some direct or indirect laboratory tests [3,4]. One of the main direct tests is to culture the pathogens from serum samples, plasma samples or skin biopsies. Another main direct test involves nucleotide amplification by using nested-PCR, real-time qPCR, or loop-mediated isothermal amplification (LAMP), etc. The indirect tests are serodiagnostic methods that are often performed by detecting the IgG and IgM antibodies of serum samples from hosts or patients using indirect immunofluorescence assay (IFA), enzyme immunoassay (EIA), enzyme-linked immunosorbent assay (ELISA), Western blot (WB), etc. [1]. With the development of laboratory tests for Lyme disease, many serodiagnostic methods have been developed, but there are still some aspects that need to be improved, such as the sensitivity in early infection, the reduction of cross-reaction with other bacteria, and the availability of a single standard method to confirm Lyme disease, etc. A novel isothermal detection of recombinase polymerase amplification (RPA) assay has been used for the diagnosis of many pathogens in recent years [5]. In this test, recombinase uvaX, DNA polymerase Bsu, single-strand binding protein (SSBP) gp32, and specific oligonucleotide primers were used, and the reaction started with magnesium acetate. The products could be detected between 30 and 42 °C with constant shaking for 20 min. It is much quicker and more convenient than PCR and many other tests. A series of RPA detection kits produced by the British company TwistDx can be used to detect both DNA and RNA targets. RPA is widely used in many fields [6,7,8,9] and many improved applications were shown in recent reports [10,11,12]. In this study, we used basic recombinase polymerase amplification (B-RPA) and recombinase polymerase amplification with lateral flow (LF-RPA) to detect a specific fragment of the recA gene of B. burgdorferi. The B-RPA reaction was used to determine the most appropriate primers for target fragment amplification. Then the primers would be used in the LF-RPA reaction with a few modifications. The products of LF-RPA were visualized by using Hybridetect 2T dipsticks after 5 min. This approach does not require complex equipment or procedures. 2. Results 2.1. Establishment of Recombinase Polymerase Amplification with Lateral Flow (LF-RPA) Assay We used LF-RPA assay to detect DNA samples extracted from 36 B. burgdorferi strains. The results showed that primers of the LF-RPA assay could be used to test DNA samples from different genotypes and different sources of B. burgdorferi strains in China (Figure 1). 2.2. Sensitivity and Specificity of Basic Recombinase Polymerase Amplification (B-RPA) and LF-RPA Assay The results of the B-RPA and LF-RPA assay using a dilution series of DNA ranging from 1 ng to 10 fg per reaction are shown in Figure 2. The results demonstrated that B-RPA and LF-RPA had the same detection threshold. DNA would be detected when its quantity in RPA reactions was as low as 50 fg. The lowest detection limit of LF-RPA was about 25 copies of the recA gene from B. burgdorferi genomic DNA. For specificity, DNA with no less than 1 pg of the non-Borrelia strains was detected by LF-RPA (Figure 3). No cross-amplification was observed in this study. 2.3. Evaluation of LF-RPA Assay for Clinical Samples Twenty samples from LB patients were detected by LF-RPA, real-time qPCR and nested-PCR (Table 1). In 20 samples, 17 samples were tested positive by WB, 14 samples were tested positive by real-time qPCR, 11 samples were tested positive by nested-PCR, and 18 samples were tested positive by LF-RPA. The LF-RPA assay got the highest positive rate in patient serum detection. Statistical analysis by the Chi-square test showed that LF-RPA was higher than nested-PCR (χ2 = 6.1442, p = 0.0132 < 0.05). No difference was observed between real-time qPCR and LF-RPA (p = 0.235). The high detection rate of the LF-RPA assay indicated that the LF-RPA assay is a very useful method in early Lyme borreliosis detection. 3. Discussion In China, the major genotypes of B. burgdorferi sensu lato were B. burgdorferi sensu stricto, Borrelia garinii, Borrelia afzelii and Borrelia valaisiana [13,14]. In this study, we chose 36 B. burgdorferi isolates of the major genotypes in endemic provinces of China for RPA detection. In our study, we found LF-RPA was a novel effective amplification method for B. burgdorferi detection. It has similar sensitivity to nested-PCR and real-time qPCR (Figure 4). In addition, it is much more convenient and faster than the other two methods. The LF-RPA reaction can be accomplished in 20 min at a temperature close to body temperature without any additional equipment (data is shown in Supplementary Materials Figure S1) [8,9,15,16]. In addition, the LF-RPA assay can tolerant some effects of the components in BSKII culture well (Supplementary Materials Figure S2). The detection limit of the LF-RPA assay for detecting the simulated serum samples was 104 copies, as low as the nested-PCR, and that of PCR was 105 copies (Supplementary Materials Figure S3). However, the detection limit of the LF-RPA assay for the purified genome DNA was 25 copies (Figure 2). The results demonstrated that the detection efficiency is different in different kinds of samples. The housekeeping gene recA was selected in RPA assays for its specificity and universality. This gene was always used in genome typing and real-time qPCR detection of B. burgdorferi [17,18,19]. We used the LF-RPA assay to detect the DNA of 36 B. burgdorferi strains, and nine non-Borrelia strains. Results showed all B. burgdorferi strains had a strong positive line on the detected strip. This implies that the LF-RPA assay is useful for B. burgdorferi detection. Besides, there was no cross-amplification between the non-Borrelia strains. There might be a cross-reaction in antibody detection from LB patients and other similar pathogen infections [20], but there was no cross-amplification in the LF-RPA test. Compared with real-time qPCR, LF-RPA showed similar detection efficiency (p > 0.05). However, its positive rate was higher than that of nested-PCR (p < 0.05). Compared with previous reports [1,2], the high positive rate of DNA detection in serum samples in this study might be related to the selection of the samples. All the samples were mainly collected from early LB patients with typical EM and tested by IFA and WB. The WB results also showed most samples were IgM-positive (Table 1). Pathogens usually enter the blood stream of the host in the initial infection, which initiates the host body’s immune system to produce antibodies to defend invasions. This study showed that serum DNA detection was helpful for diagnosing the early infection of B. burgdorferi, and can complement the antibody detection of B. burgdorferi. In conclusion, we have successfully applied a new method, LF-RPA, for B. burgdorferi detection. In addition, the RPA assay might also be applied in detection for tick samples in endemic fields. Experiments might be carried out in the future. 4. Materials and Methods 4.1. Strains and DNA Extraction A total of 36 B. burgdorferi strains, isolated from 13 provinces in China and classified to four genotypes [13], were provided by Department of Lyme disease, National Institute for Communicable Disease Control and Prevention, Chinese Center for Disease Control and Prevention (China CDC), except for B. burgdorferi sensu stricto B31 which was from the United States (Supplementary Materials Table S1). Genomic DNA of these strains was extracted by boiling 10 min and the quantity was determined by Nanodrop ND-1000 spectrophotometer (ThermoScientific, Waltham, MA, USA), and then stored at −20 °C until ready for use. Genomic DNA for sensitivity determination was extracted by Genomic DNA Purification kit (Promega, Madison, WI, USA). In addition, DNA of Ehrlichiae, Bartonella henselae, Anaplasma phagocytophilum, and Coxiella burnetii, were provided by Department of HGA (human granulocytic anaplasmosis), China CDC. DNA of Leptospira 56603, Leptospira 56602, Leptospira 56601, Leptospira 56613 was provided by Department of Leptospirosis, China CDC. Escherichia coli BL21 strain was brought from ComWin Biotech Company, Beijing, China. The DNA of Escherichia coli BL21 was obtained by the same method as B. burgdorferi strains. 4.2. Establishment of B-RPA and LF-RPA Assay A specific segment of recA sequence for B. burgdorferi up to 170 bp was selected to be a target. Suitable primers for RPA reaction were designed by using Primer 5.0 and parameters were set up according to the rules in the TwistDX instruction manual. The basic RPA reaction was initiated by using the TwistAmp Basic kit (TwistDx, Cambridge, UK), and the reaction system contained 4.8 μL primers (forward and reverse primer, 10 μM), 29.5 μL 1 × rehydration buffer, 12.2 μL ddH2O and 1 μL DNA, and then started the reaction by adding 2.5 μL magnesium acetate. Reactions were incubated for a typical 20 min at 37 °C in a Minitron shaker with constant shaking at 300 rpm [15,16]. The RPA production was purified by QIAquick PCR Purification kit (QIAGEN, Hilden, Germany) and analyzed on 3% agarose-gel electrophoresis subsequently. A series of RPA primers screened by the basic RPA reaction, and the most appropriate pair with a 170 bp amplicon was selected (Table 2). Primers of LF-RPA assay needed a hybridization probe and the reverse primer should be labeled with a biotin at the 5′ end (Table 2). TwistAmp nfo kit (TwistDX, Cambridge, UK) and Hybridetect 2T (Milenia Biotec GmbH, GieBen, Germany) dipsticks were used in this assay. The reaction system included 4.2 μL primers (10 μM), 29.5 μL 1× rehydration buffer, 0.6 μL probe, 13.2 μL mixture solution DNA with ddH2O and was also started by 2.5 μL magnesium acetate. Samples were incubated as previously described [15,16]. Then, the products were diluted at 1:20 with running buffer (Milenia Biotec GmbH, GieBen, Germany). Dipsticks were put into the diluted samples and the results can be read in 5 min. 4.3. Sensitivity and Specificity of LF-RPA Assay A gradient dilution of genomic DNA of B. burgdorferi to 100 pg, 10 pg, 1 pg, 100 fg, 50 fg, 10 fg per microliter was used in each reaction. In addition, the gradient copy number was diluted to 104, 103, 102, 50, 25, 10, 1 per microliter. Then 1 μL diluted DNA was added for each reaction, and then incubated at 37 °C for 20 min with 300 rpm constant shaking. These diluted DNA was used in B-RPA reaction and LF-RPA assay. In addition, we tested 36 strains of B. burdgorferi and nine non-Borrelia strains by using LF-RPA assay for specificity evaluation. The quantity of the non-Borrelia DNA used in the LF-RPA test was 1 ng per reaction. 4.4. Evaluating the Efficiency of LF-RPA Assay in Clinical Diagnosis Twenty patients’ serum samples of Lyme borreliosis used in this study were provided by Mudanjiang Linye Hospital, Heilongjiang Province. Clinic diagnosis of Lyme disease includes the following conditions: 1, history of tick-borne infection; 2, EM ≥ 5 cm; 3, IFA or ELISA test (positive, +); 4, Western blot (positive, +); 5, other symptoms of Lyme disease, such as carditis, neuroborreliosis, arthritis, etc. All participants met 1, 2, 3 or 1, 3, 4, 5 conditions can be confirmed with Lyme borreliosis [1,2]. DNA of these samples was extracted by using DNeasy Blood & Tissue kit (QIAGEN, Germany). All DNA specimens were detected by LF-RPA assay, real-time qPCR and nested-PCR. Primers and reaction conditions of real-time qPCR and nested-PCR were according to previous reports [20,21,22,23,24,25]. The conditions of nested-PCR: First round, P1: 5′-CGACCTTCTTCGCCTTAAAGC-3′, F1: 5′-TAAGCTGACTAATACTAATTACCC-3′; 35 cycles: at 94 °C for 45 s, 53 °C for 45 s, 72 °C for 45 s. And the second round: P2: 5′-TCCTAGGCATTCACCATA-3′, F2: 5′-GAGTTCGCGGGAGA-3′; 35 cycles at 94 °C for 45 s, 55 °C for 45 s, 72 °C for 45 s. The conditions of real-time qPCR: recA-rt-F: GTTCTGCAACATTAACACCTAAAGCTT; recA-rt-R: AGGTGGGATAGCTGCTTTTATTGAT; recA-rt-P: F-ACAGGATCAAGAGCATG-P; 40 cycles at 94 °C for 10 s, 54 °C for 30 s. Results of LF-RPA, real-time qPCR and nested-PCR were pairwise compared by chi-square test. In addition, B31 strains with different copy numbers were added into the healthy serum samples to preparing simulated serum samples. The DNA of the simulated serum samples was extracted by using DNeasy Blood & Tissue kit (QIAGEN, Hilden, Germany). The extracted DNA tested by PCR, nested-PCR and LF-RPA. The method of PCR was reference from previous report [13]. 4.5. Ethical Statement This study was approved by the Ethical Review Committee of the National Institute for Communicable Disease Control and Prevention (ICDC), Chinese Center for Disease Control and Prevention (China CDC). Patient serum samples were collected from endemic area of Lyme borreliosis. Written informed consent was obtained prior to the study. Acknowledgments We would like to thank the Department of Leptospirosis and Department of HGA (human granulocytic anaplasmosis) at the Chinese Center for Disease Control and Prevention for providing the DNA of non-Borrelia strains. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1250/s1. Click here for additional data file. Author Contributions Wei Liu participated in the study design, data collection, statistical analysis and the preparation of the manuscript. Hui-Xin Liu and Lin Zhang contributed to the study with data collection and critical review. Xue-Xia Hou prepared the DNA of the B. burgdorferi strains. Kang-Lin Wan and Qin Hao supervised this work. All authors read and approved the final version of the manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations LB Lyme borreliosis EM erythema migrans RPA recombinase polymerase amplification B-RPA basic recombinase polymerase amplification LF-RPA recombinase polymerase amplification with lateral flow LAMP loop-mediated isothermal amplification IFA indirect immunofluorescence assay EIA enzyme immunoassay ELISA immunosorbent assay WB western blot Figure 1 Results of recombinase polymerase amplification with lateral flow (LF-RPA) assay in 36 strains of Borrelia burgdorferi (B. burgdorferi) detection. N stands for negative control (water); Name of each strain was listed above the strips. Figure 2 Detection of a series of diluted B. burgdorferi DNA by basic recombinase polymerase amplification (B-RPA) and LF-RPA. (A) The B-RPA results shown by agarose gel electrophoresis. M stands for marker, N stands for negative control (water); (B) The LF-RPA results shown by Hybridetect 2T dipsticks. N stands for negative control (water); (C) Detection of B. burgdorferi DNA from 104 copies to one copy by LF-RPA assay. Figure 3 Result of LF-RPA assay in non-Borrelia strain detection. B31: positive control; 1, Ehrlichiae; 2, Bartonella henselae; 3, Anaplasma phagocytophilum; 4, Coxiella burnetii; 5, Leptospira 56603; 6, Escherichia coli BL21; 7, Leptospira 56602; 8, Leptospira 56601; 9, Leptospira 56613; 10, healthy blood DNA; N, negative control (water). Figure 4 Detection of a series of diluted B. burgdorferi DNA by real-time qPCR and nested-PCR. (A) The results of diluted DNA detected by real-time qPCR. Ct value for each concentration is shown in this figure (B); (C) results of diluted DNA detected by nested-PCR. M stands for marker. ijms-17-01250-t001_Table 1Table 1 Results of 20 Lyme borreliosis (LB) patients’ samples detected by real-time qPCR, nested-PCR and recombinase polymerase amplification with lateral flow (LF-RPA). Numbers of Samples Western Blot Real-Time qPCR (Ct Value ) Nested-PCR LF-RPA 1 +(IgM) 35.4 + + 2 +(IgM) 32.11 + + 3 +(IgG, IgM) 34.21 − + 4 +(IgM) − + + 5 +(IgG, IgM) 37.17 + + 6 +(IgG, IgM) − + + 7 +(IgM) − + + 8 +(IgM) − − + 9 +(IgG) − − − 10 +(IgG) − − − 11 +(IgM) 35.42 − + 12 +(IgG, IgM) 32.32 + + 13 − 34.63 + + 14 +(IgM) 33.36 − + 15 +(IgM) 37.14 − + 16 − 32.96 + + 17 +(IgM) 36.91 − + 18 +(IgM) 37.40 + + 19 +(IgM) 34.13 + + 20 − 35.30 − + ijms-17-01250-t002_Table 2Table 2 Oligonucleotides of basic recombinase polymerase amplification (B-RPA) and LF-RPA assay used in this study. Assays Primers Basic RPA reaction F: 5′-ATTGTATTAGATGAAGCTCTTGGCATTGGTGGA-3′ R: 5′-AATAGGATCGAGATCAAGTTCTGCTTCAATA-3′ Lateral flow RPA reaction LF-F: 5′-ATTGTATTAGATGAAGCTCTTGGCATTGGTGGA-3′ LF-R: 5′-biotin-TTGCATAAATAGGATCGAGATCAAGTTCTGC-3′ LF-P: 5′-(FAM)-ACTTTGATCCTTCAAGCGATTGCTGARGT-(dSpacer)-CAAAAAGAAGGAGGCAT-C3Spacer-3′ “R” degenerate bases; dSpacer is an exonuclease site; C3Spacers is a polymerase extension blocking site. ==== Refs References 1. Marques A.R. Laboratory diagnosis of lyme diseases: Advances and challenges Infect. Dis. Clin. N. Am. 2015 29 295 307 10.1016/j.idc.2015.02.005 25999225 2. Stanek G. Fingerle V. Hunfeld K.P. Jaulhac B. Kaiser R. Krause A. Kristoferitsch W. O’Connell S. Ornstein K. Strle F. Lyme borreliosis: Clinical case definitions for diagnosis and management in Europe Clin. Microbiol. Infect. 2011 17 69 79 10.1111/j.1469-0691.2010.03175.x 20132258 3. Stanek G. Strle F. Lyme borreliosis Lancet 2012 379 461 473 10.1016/S0140-6736(11)60103-7 21903253 4. Aguero-Rosenfeld M.E. Wang G. Schwartz I. Wormser G.P. Diagnosis of Lyme borreliosis Clin. Microbiol. Rev. 2005 18 484 509 10.1128/CMR.18.3.484-509.2005 16020686 5. James A. Macdonald J. Recombinase polymerase amplification: Emergence as a critical molecular technology for rapid, low-resource diagnostics Expert Rev. Mol. Diagn. 2015 15 1475 1489 10.1586/14737159.2015.1090877 26517245 6. Rosser A. Rollinson D. Forrest M. Webster B.L. Isothermal Recombinase Polymerase Amplification (RPA) of Schistosoma haematobium DNA and oligochromatographic lateral flow detection Parasites Vectors 2015 8 446 10.1186/s13071-015-1055-3 26338510 7. Nair G. Rebolledo M. White A.C. Crannell Z. Richards-Kortum R.R. Pinilla A.E. Ramírez J.D. López M.C. Castellanos-Gonzalez A. Detection of Entamoeba histolytica by Recombinase Polymerase Amplification Am. J. Trop. Med. Hyg. 2015 93 591 595 10.4269/ajtmh.15-0276 26123960 8. Wahed A.A.E. Patel P. Faye O. Thaloengsok S. Heidenreich D. Matangkasombut P. Manopwisedjaroen K. Sakuntabhai A. Sall A.A. Hufert F.T. Recombinase Polymerase Amplification assay for rapid diagnostics of dengue infection PLoS ONE 2015 10 10.1371/journal.pone.0129682 9. Liljander A. Yu M. O’Brien E. Heller M. Nepper J.F. Weibel D.B. Gluecks I. Younan M. Frey J. Falquet L. A field-applicable recombinase polymerase amplification assay for rapid detection of Mycoplasma capricolum subsp. capripneumoniae J. Clin. Microbiol. 2015 53 2810 2815 10.1128/JCM.00623-15 26085615 10. Kersting S. Rausch V. Bier F.F. von Nickisch-Rosenegk M. Multiplex isothermal solid-phase recombinase polymerase amplification for the specific and fast DNA-based detection of three bacterial pathogens Microchim. Acta 2014 181 1715 1723 10.1007/s00604-014-1198-5 25253912 11. Chao C.C. Belinskaya T. Zhang Z. Ching W.M. Development of Recombinase Polymerase Amplification assays for detection of Orientia tsutsugamushi or Rickettsia typhi PLoS Negl. Trop. Dis. 2015 9 1250 10.1371/journal.pntd.0003884 26161793 12. Crannell Z.A. Castellanos-Gonzalez A. Irani A. Rohrman B. White A.C. Richards-Kortum R. Nucleic acid test to diagnose cryptosporidiosis: Lab assessment in animal and patient specimens Anal. Chem. 2014 86 2565 2571 10.1021/ac403750z 24479858 13. Hao Q. Hou X. Geng Z. Wan K. Distribution of Borrelia burgdorferi sensu lato in China J. Clin. Microbiol. 2011 49 647 650 10.1128/JCM.00725-10 21106783 14. Chu C.Y. Jiang B.G. He J. Gao Y. Zhang P.H. Wu X.M. Zhang W.Y. Shi H. Gaowa H.S. Wang J.B. Genetic diversity of Borrelia burgdorferi sensu lato isolates from Northeastern China Vector Borne Zoonotic Dis. 2011 11 877 882 10.1089/vbz.2009.0195 21870355 15. Kersting S. Rausch V. Bier F.F. von Nickisch-Rosenegk M. Rapid detection of Plasmodium falciparum with isothermal recombinase polymerase amplification and lateral flow analysis Malar. J. 2014 13 99 10.1186/1475-2875-13-99 24629133 16. Castellanos-Gonzalez A. Saldarriaga O.A. Tartaglino L. Gacek R. Temple E. Sparks H. Melby P.C. Travi B.L. A novel molecular test to diagnose canine visceral leishmaniasis at the point of care Am. J. Trop. Med. Hyg. 2015 93 970 975 10.4269/ajtmh.15-0145 26240156 17. Dania R. Danièle P. Natacha S. Livey I. Matuschka F.R. Baranton G. Delineation of Borrelia burgdorferi sensu lato species by multilocus sequence analysis and confirmation of the delineation of Borrelia spielmanii sp. nov. Int. J. Syst. Evol. Microbiol. 2006 56 873 881 16585709 18. Momert S. Gutzmer R. Kapp A. Werfel T. Sensitive detection of Borrelia burgdorferi sensu lato DNA and differentiation of Borrelia species by LightCycler PCR J. Clin. Microbiol. 2001 39 2663 2667 10.1128/JCM.39.7.2663-2667.2001 11427590 19. Pietilä J. He Q. Oksi J. Viljanen M.K. Rapid differentiation of Borrelia garinii from Borrelia afzelii and Borrelia burgdorferi sensu stricto by LightCycler fluorescence melting curve analysis of a PCR product of the recA gene J. Clin. Microbiol. 2000 38 2756 2759 10878080 20. Magnarelli L.A. Anderson J.F. Johnson R.C. Cross-reactivity in serological tests for Lyme disease and other spirochetal infections J. Infect. Dis. 1987 156 183 188 10.1093/infdis/156.1.183 3298452 21. Zhang L. Zhu X. Hou X. Geng Z. Chen H. Hao Q. Test of 259 serums from patients with arthritis or neurological symptoms confirmed existence of Lyme disease in Hainan province, China Int. J. Clin. Exp. Med. 2015 8 9531 9536 26309619 22. Chu C.Y. Jiang B.G. Liu W. Zhao Q.M. Wu X.M. Zhang P.H. Zhan L. Yang H. Cao W.C. Presence of pathogenic Borrelia burgdorferi sensu lato in ticks and rodents in Zhejiang, south-east China J. Med. Microbiol. 2008 57 980 985 10.1099/jmm.0.47663-0 18628499 23. Postic D. Assous M.V. Grimont P.A. Baranton G. Diversity of Borrelia burgdorferi sensu lato evidenced by restriction fragment length polymorphism of rrf (5S)-rrl (23S) intergenic spacer amplicons Int. J. Syst. Bacteriol. 1994 44 743 752 10.1099/00207713-44-4-743 7981102 24. Schwartz J.J. Gazumyan A. Schwartz A.I. rRNA gene organization in the Lyme disease spirochete, Borrelia burgdorferi J. Bacteriol. 1992 174 3757 3765 1350586 25. Ge Z. Hou X. Zhang L. Hao Q. Evaluation of a new real-time PCR assay for detection of Borrelia burgdorferi in rodents Chin. J. Zoonoses 2015 31 812 816

PMC005xxxxxx/PMC5000649.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081251ijms-17-01251ArticleDrought-Induced Leaf Proteome Changes in Switchgrass Seedlings Ye Zhujia 1Sangireddy Sasikiran 1Okekeogbu Ikenna 1Zhou Suping 1*Yu Chih-Li 2Hui Dafeng 2Howe Kevin J. 3Fish Tara 3Thannhauser Theodore W. 3*Komatsu Setsuko Academic Editor1 Department of Agricultural Sciences, Tennessee State University, 3500 John Merritt Blvd, Nashville, TN 37209, USA; [email protected] (Z.Y.); [email protected] (S.S.); [email protected] (I.O.)2 Department of Biological Sciences, Tennessee State University, 3500 John Merritt Blvd, Nashville, TN 37209, USA; [email protected] (C.-L.Y.); [email protected] (D.H.)3 Functional & Comparative Proteomics Center, USDA-ARS, Cornell University, Ithaca, NY 14853, USA; [email protected] (K.J.H.); [email protected] (T.F.)* Correspondence: [email protected] (S.Z.); [email protected] (T.W.T.); Tel.: +1-615-963-2465 (S.Z.); +1-607-255-8808 (T.W.T.)02 8 2016 8 2016 17 8 125121 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Switchgrass (Panicum virgatum) is a perennial crop producing deep roots and thus highly tolerant to soil water deficit conditions. However, seedling establishment in the field is very susceptible to prolonged and periodic drought stress. In this study, a “sandwich” system simulating a gradual water deletion process was developed. Switchgrass seedlings were subjected to a 20-day gradual drought treatment process when soil water tension was increased to 0.05 MPa (moderate drought stress) and leaf physiological properties had expressed significant alteration. Drought-induced changes in leaf proteomes were identified using the isobaric tags for relative and absolute quantitation (iTRAQ) labeling method followed by nano-scale liquid chromatography mass spectrometry (nano-LC-MS/MS) analysis. Additionally, total leaf proteins were processed using a combinatorial library of peptide ligands to enrich for lower abundance proteins. Both total proteins and those enriched samples were analyzed to increase the coverage of the quantitative proteomics analysis. A total of 7006 leaf proteins were identified, and 257 (4% of the leaf proteome) expressed a significant difference (p < 0.05, fold change <0.6 or >1.7) from the non-treated control to drought-treated conditions. These proteins are involved in the regulation of transcription and translation, cell division, cell wall modification, phyto-hormone metabolism and signaling transduction pathways, and metabolic pathways of carbohydrates, amino acids, and fatty acids. A scheme of abscisic acid (ABA)-biosynthesis and ABA responsive signal transduction pathway was reconstructed using these drought-induced significant proteins, showing systemic regulation at protein level to deploy the respective mechanism. Results from this study, in addition to revealing molecular responses to drought stress, provide a large number of proteins (candidate genes) that can be employed to improve switchgrass seedling growth and establishment under soil drought conditions (Data are available via ProteomeXchange with identifier PXD004675). physiological propertiesisobaric tags for relative and absolute quantitation (iTRAQ)ProteoMinerfunctional pathwaysabscisic acid (ABA) signaling“Sandwich” plant growth system ==== Body 1. Introduction Switchgrass (Panicum virgatum), has been selected as a model herbaceous bioenergy species in the USA due to its high biomass yield, strong tolerance to drought and flooding conditions, relatively low herbicide and fertilizer input requirements, and widespread adaptability to temperate climate [1,2,3]. Recently, a shortage of fresh water and increasingly severe drought have become a significant challenge to crop production [4]. Based on data from the National Weather Service Centers for Environmental Prediction [5], soil moisture contents in the topsoil layer have declined over the past decade (2005–2015) in many regions of the USA, especially in central states. Drought tolerance is one of the most striking physiological properties of switchgrass. Mature plants have a very deep root system and a highly efficient C4 metabolic pathway [6]. However, switchgrass plants are slow to establish in the field, often requiring two to three growing seasons to develop deep root systems. During the early stages of growth when seedlings have a relatively shallow root distribution (0–15 cm) in the top soil, these plants are very susceptible to both periodic and long-term drought conditions [7]. A field trial shows that drought significantly affected seedling growth of switchgrass in the first year. Furthermore, biomass yield declined greatly after three consecutive years of drought [8]. Thus, developing switchgrass plants with strong drought tolerance during the early stages of growth is an effective strategy to ensure high biomass yields during subsequent years in the field. Plant growth depends on cell division, cell enlargement, and differentiation [9]. Under drought conditions, cell elongation and division are both suppressed by the reduced photosynthesis driven by diminished CO2 influx and limitation of carboxylation by abscisic acid (ABA)-dependent stomatal closure [9,10,11,12]. On the other hand, stomatal closing has been viewed as a drought tolerance mechanism to avoid excess water loss via transpiration. A set of physiological parameters related to drought tolerance has been identified including leaf relative water content (RWC), electrolyte leakage (EL), photosynthetic rate (Pn), stomatal conductance (gs), transpiration rate (Tr), intercellular CO2 concentration (Ci), and water use efficiency (WUE) [13]. Thus, whether plants can sustain active growth or just survive the water-deficient conditions depends on how efficiently they regulate these complex processes. Proteins are the primary molecules that carry out various biological functions in cells and in an entire organism [14]. Alterations in proteome composition provide the basis for a plant to perform different biological functions, including adapting to changing and/or suboptimal environmental conditions [15,16,17,18,19,20,21]. With the rapid development of proteomic technologies, two-dimensional liquid chromatography, in combination with multiplexed quantitative techniques such as isobaric tags for relative and absolute quantitation (iTRAQ), provides the ability to perform relative or absolute quantification of proteomes [22,23,24,25,26]. Quantitative proteomics using the shot-gun bottom-up approach has been used to evaluate drought-responsive proteins in important crop species, such as rice, maize, wheat, cotton, amaranth, alfalfa, sugar beets, and tomatoes [18,20,27,28,29,30,31,32,33,34,35,36,37]. Conclusively, these proteomics studies have significantly increased our understanding of molecular regulation at the translational and post-translational levels in plants. The separation and detection of all proteins contained in any given proteome remains a challenge because the analysis of low-abundance proteins is difficult in the presence of the highly abundant proteins. Characterization of the photosynthetically active leaf proteome is a very difficult task as the ribulose-1,5-biphosphate carboxylase/oxygenase (Rubisco) proteins would account for approximately 40% of total protein content [38]. An earlier study using immunoaffinity subtraction of Rubisco was able to increase the resolution of more protein species in leaf protein samples [39]. However, those antibodies are very expensive, which limits their usage in large quantitative proteomics experiments (unpublished data, Zhou, Tennessee State University, Nashville, TN, USA, 2016). The ProteoMiner protein depletion/enrichment technology, which employs a large, highly diverse bead-based library of combinatorial peptide ligands, has proven to be a powerful tool for uncovering low-abundance proteins. Using this approach, Fasoli et al. detected 79% more proteins from spinach leaves than could be detected without the depletion/enrichment process [40]. More importantly, the ProteoMiner protein enrichment method produces highly stable and reproducible results, which is extremely important in quantitative proteomics where two or more samples are analyzed in each treatment condition [41,42]. This study was carried out with a goal to understand the changes in leaf proteome in switchgrass under drought stress and to develop the association between the expression of these proteins and the physiological properties that give rise to drought tolerance. As described above, removal of highly abundant Rubisco protein is an effective strategy for increasing the overall number of identified proteins, thus the ProteoMiner depletion/enrichment procedure was performed to reduce the scale of dynamic range in protein abundance. By enabling the identification of low-abundance proteins and increasing the number of proteins quantified, this study provides an in-depth understanding of systemic changes in the drought-induced proteomes in switchgrass seedlings. 2. Results 2.1. Drought-Induced Physiological Properties of Switchgrass Sixteen days after the initiation of water withholding, young leaves on drought-treated plants started to show signs of wilting as the soil water tension of treated groups reached 0.05 MPa. Twenty days after water withholding, soil water tension increased to 0.08 ± 0.02 MPa (Table 1). At this time, the relative growth of drought-treated plants was reduced significantly (a 20% decrease), as well as the stomatal conductance and transpiration rate (p < 0.01), compared to the untreated control plants. The water use efficiency, which is defined as the ratio of the photosynthetic rate to the transpiration rate [43], showed an 7.1% increase in the drought-treated group, which was significantly higher than the untreated control plants (p < 0.01). Changes in these physiological properties showed that leaves and plants as a whole experienced a progressive drought-stress during the 20 days of withholding water. At this time-point (20 days after withholding water), the drought treatments were terminated and tissues were harvested for further analysis (Figure 1). 2.2. Effects of the ProteoMiner Enrichment Process on the Identification of Leaf Proteome The ProteoMiner enrichment method is used to increase the relative concentration of low-abundance proteins by depleting those high-abundance proteins in a protein sample, and thus to increase the depth of coverage of the leaf proteomes to be identified in a proteomics analysis. In plant leaf proteomes, more than 40% of the total leaf protein content consists of Rubisco [38]. One-dimensional gel electrophoresis showed that in the ProteoMiner-treated (PMT) samples, the band intensity of high abundance proteins (i.e., Rubisco protein) was reduced, whereas the intensity of several weaker protein bands was increased, compared to the counterparts of the Crude Leaf Extracts (CLE) protein samples (Figure S1). Analysis of the peptide numbers for the proteins detected in the PMT sample indicates that 1101 proteins were identified by a greater number of peptides after ProteoMiner enrichment—for instance, the number of peptides in an adenylate kinase protein (Pavir.Fa02159.1) was increased from 29 in CLE to 189 in PMT samples. On the other hand, 1876 proteins were identified by fewer peptides. For example, the number of peptides in Rubisco subunits including Pavir.Cb01593.1, Pavir.Cb01387.1, and Pavir.J32704.1 was decreased (Figure S2, Table S1-1,2). These results demonstrate that ProteoMiner did deplete the concentration of the highly abundant proteins while simultaneously enriching low-abundance proteins. 2.3. Identification of Quantified Proteins In this study, 7006 proteins were identified in the switchgrass leaf proteome with the assistance of the ProteoMiner enrichment method (Table 2, Tables S1-1–3 and S2). A total of 5493 proteins were identified in the CLE samples and 4839 unique proteins were identified in the PMT samples. Between the CLE and PMT samples, 3326 proteins overlapped. The use of a ProteoMiner enrichment step resulted in the identification of 1513 proteins that were not found in the CLE samples. It appears that the ProteoMiner enrichment is complementary to the analysis of the crude leaf protein extracts, and a combination of both approaches was shown to quantify more proteins than either individually. Among the total identified proteomes, 81.1% of them (5680/7006) contained at least two unique peptides (Table 2). Quantitative analysis revealed that 257 proteins, which was approximately 4% of the total quantified proteomes (257/7006), passed the threshold value of ±2σ (standard deviation), p < 0.05 (t-test and false discovery rate (FDR) corrections), and fold change <0.6 or >1.7. These proteins were considered significantly changed under the drought treatment conditions. Among the 257 drought-induced significant proteins, 55 proteins showed consistent changes in both the CLE and PMT protein samples, 150 proteins were found only in CLE samples, and 52 proteins were only identified in PMT samples (Tables S1-5 and S3). In addition, the false negative rate (β) was calculated as 0.02 by summing the probabilities that each of the proteins judged to be unchanged was in fact differentially expressed. This suggests that the power of the experiment was very high (p = 1 − β = 0.98). MapMan is a bioinformatics tool for developing the associations between gene (protein) expression and cellular processes, but this offline program only performs analysis of genomes contained in the MapMan Store. As the annotated switchgrass genome database is not listed, the program will not recognize the protein accession identity and therefore cannot map the protein expression data to biological functions. Instead, in this study, the Arabidopsis thaliana accessions annotated for those drought-induced switchgrass proteins were used when developing the functional pathways (Figure S3). Results showed that each functional group contained upregulated and downregulated proteins. A large number of the significantly changed proteins are associated with RNA transcription/processing, protein synthesis, and protein degradation pathways (Table 3). 2.4. Proteins in Regulation of Transcription and Translation For proteins involved in gene transcription, several members of the G2-like, myeloblastosis (MYB) and bZIP transcription factors (TFs) were identified. A MYB-related transcription factor TRY (Triptychon) (Pavir.Eb02165.1) and a G2-like transcription factor APL (altered phloem development) (Pavir.Fa01260.1) were significantly reduced under drought stress. The former TF did not pass the threshold as a significant protein in CLE, and the latter TF was identified only in the PMT samples. The GBF (G-box binding factor) (Pavir.Ea03718.1), a member of the bZIP TFs family involved in ABA and stress signaling [44], was significantly increased (>2-fold) (Table S1-4), and it was identified in both CLE and PMT samples. Proteins involved in protein synthesis and degradation were altered. The chloroplast-targeted FtsH protease (Pavir.J13145.1) was up-regulated at a higher than four-fold level. Moreover, the relative abundance of a senescence-specific Cys-protease protein (SAG) (Pavir.J08126.1) markedly declined in response to drought stress. Regarding to changes in protein synthesis, drought stress induced a plastid-specific 50S ribosomal protein (PSRP) (Pavir.Ea00033.1), which is an important member of the translation machinery in chloroplasts. However, the drought-induced significant change was found only in PMT samples, not in CLE samples (Table S1-4). 2.5. Cell Division and Cell Wall Modification Two proteins involved in the cell cycle and cell division were identified. Pavir.Gb00127.1, a regulator of chromosome condensation (RCC), was significantly decreased, whereas prohibitin (PHB) (Pavir.Aa01476.1) was significantly increased (Table S1-4). UDP-glucose 4-epimerase (UGE) (Pavir.J14539.1), with a proven function in cell wall carbohydrate metabolism [45], was upregulated more than six-fold. A cell-wall-modifying xyloglucan endotransglycosylase/hydrolase (XET) (Pavir.Fa01211.1) [46] was increased 3.39-fold. These two cell-wall-related proteins were not identified in PMT (Table S1-4). 2.6. Phyto-Hormone Metabolism and Signaling Transduction Pathways Four proteins involved in the metabolism of auxin and ethylene were all induced by drought stress (Pavir.Ga00273.1, Pavir.J01120.1, Pavir.J01160.1, and Pavir.Ia03739.1). Of the significantly changed proteins in the ABA-metabolic pathway, the upregulated proteins were classified as GRAM domain-containing proteins (Pavir.Ca02189.1 and Pavir.Cb00761.1), ABA-responsive elements-binding factor (ABF) (Pavir.J00256.1), and 9-cis-epoxycarotenoid dioxygenases (NCED) (Pavir.Ba03791.1) (Table S1-4). Six calcium-binding proteins that interact with the second messenger “Ca2+” to transduce stress signals into plant cells were identified, five of them markedly upregulated (Pavir.Ea00612.1, Pavir.Ca00053.1, Pavir.Eb03832.1, Pavir.Ib02894.1, and Pavir.J09383.1). These proteins were identified in both CLE and PMT, or in CLE but not in PMT. The one reduced (Pavir.Da01126.1) protein was identified in PMT but not in CLE (Table S1-4). 2.7. Stress-Responsive Proteins The drought treatments induced 31 abiotic/biotic stress responsive proteins. These stress proteins include six biotic stress responsive proteins (Pavir.Bb00478.1, Pavir.Fb02059.1, Pavir.Ga02124.1, Pavir.Ha00419.1, Pavir.J09667.1, and Pavir.J00406.1), five dehydrins (DHNs) (Pavir.Bb03589.1, Pavir.Ca01575.1, Pavir.Aa00887.1, Pavir.J04551.1, and Pavir.J13075.1), 13 heat shock proteins (HSP) (Pavir.Ea00289.1, Pavir.J35929.1, Pavir.J33423.1, Pavir.J24160.1, Pavir.Ia03665.1, Pavir.J40704.1, Pavir.Ib01136.1, Pavir.Ab00778.1, Pavir.J19824.1, Pavir.Fa01476.1, Pavir.Aa00282.1, Pavir.Hb01472.1, and Pavir.J21349.1), one cold stress-related protein (Pavir.J31919.1), and six other stress responsive proteins. Of them, the HSP20-like protein (Pavir.J21349.1) increased more than 9.73-fold (Table S1-4). 2.8. Carbohydrate Metabolism The relative abundance level of proteins in carbohydrate metabolic pathways, such as gluconeogenesis, starch metabolism, and the biosynthesis of raffinose family oligosaccharides (RFO), were altered in response to the drought treatments. The induced proteins include malate synthase (Pavir.Gb01372.1), β amylase protein (Pavir.J18576.1), and two galactinol synthase proteins (Pavir.J07018.1 and Pavir.J40731.1), but a starch synthase (Pavir.J06822.1) was repressed under the drought-treated conditions (Table S1-4). 2.9. Nitric Acid Metabolism Under moderate drought stress, three proteins involved in the biosynthesis of free amino acids were markedly upregulated, Δ1-pyrroline-5-carboxylate synthetase protein (P5CS) (Pavir.J02344.1), methionine-γ-lyase protein (MGL) (Pavir.Ib03758.1), and l-asparagine amidohydrolase (Pavir.Gb00328.1). An enzyme-catalyzing β-oxidation of fatty acids, 3-ketoacyl-CoA thiolase-2 (KAT2/PED1/PKT3) (Pavir.J16366.1) which is involved in ABA signaling, was also significant increased (Table S1-4). 3. Discussion Among all the drought tolerance mechanisms, an increased ABA content in leaves has been shown to play a key role in activating signaling pathways that control stomatal closure, thus reducing transpirational water loss [47,48]. During the 20 days of drought treatment period, the switchgrass leaves showed a gradual decline in stomatal conductance and transpiration rates, which are indications of a reduced stomatal aperture. This prediction of stomatal behavior is supported by the upregulation of PHB (Pavir.Aa01476.1), which regulates the level of nitric oxide accumulation that induces stomatal closure and thus enhances the adaptive plant responses against drought stress [49,50]. Changes in protein expression support the elevated biosynthesis of ABA and the induction of ABA-mediated signal transduction pathways during the drought treatment period (Figure 2). In the ABA biosynthesis pathway, 9-cis-epoxycarotenoid dioxygenase (NCED) catalyzes the step to convert 9-cis-xanthophylls to xanthoxin, which is the direct precursor of ABA [51]. The regulatory role of NCED in ABA biosynthesis in leaves under stress conditions has been clearly demonstrated in many studies, showing that the abundance of NCED proteins is directly correlated with ABA content [51,52,53,54,55]. The same metabolic changes may have occurred in switchgrass leaves where the significant increase (4.5-fold) of an NCED protein (Pavir.Ba03791.1) may result in an elevated ABA content in the drought-treated leaves (Table S1-4). In the ABA-dependent signaling pathway, bZIP transcription factors is one of the major families that have been described to be associated with plant responses to stress conditions [44]. In this study, two members of bZIP proteins, GBF (G-box binding factor) (Pavir.Ea03718.1) and ABF (ABA-responsive elements-binding factor) (Pavir.J00256.1), were significantly increased under drought stress. The overexpression of ABF can alter ABA sensitivity, dehydration tolerance, and the expression levels of ABA/stress-regulated genes [56]. Furthermore, the GBF and ABF protein, and an ABA-responsive GRAM domain-containing protein (Pavir.Cb00761.1) were identified as drought-induced proteins in both CLE and PMT samples, which validates the high confidence of these significant proteins. Taken together, we have shown for the first time that the ABA-dependent pathway are regulated at protein level, which in turn may have a significant role in activating the transcription of drought tolerance genes in switchgrass. Ribonuclease S1 (RNS1) (Pavir.Fa00890.1) plays a very important part in both wound- and ABA-responsive signaling pathways, and RNS1 itself is a target for post-transcriptional regulation by ABA [57]. The upregulated enzyme 3-ketoacyl-CoA thiolase-2 (KAT2/PED1/PKT3) (Pavir.J16366.1) has an important role in regulating reactive oxygen species (ROS) production in response to ABA [58]. In addition, three proteins annotated to the regulatory components of ABA receptor 3 (Pavir.Ab01039.1, Pavir.Ca00496.1, and Pavir.Cb01723.1) showed varied changes (0.78–1.48-fold), but none of them passed the threshold criteria for significantly changed proteins in this study. These results indicate a very dynamic adjustment system regulating the expression of proteins in ABA biosynthesis and signaling pathways, which in turn modulates the activation of drought tolerance mechanism in switchgrass leaves (Figure 2). Environmental stimuli usually require a second messenger, such as Ca2+, to transduce the signals into a plant cell. Under stress conditions, calcium-binding proteins (e.g., calmodulin or calmodulin-related protein) are induced in response to elevated levels of free Ca2+ in cells, and then they, in turn, activate signal transduction pathways with an impact on the activity of a variety of target enzymes [59,60,61,62,63,64,65,66,67,68]. The dynamic changes in the isoforms of these calcium-binding proteins quantified in this study represent the complex network of drought stress-induced signal transduction in switchgrass (Figure 2). The drought-induced metabolic rearrangement is one of the major components for plants to acquire tolerance to stress conditions. Soluble sugars can accumulate to function as osmolytes to maintain cell turgor and have the ability to protect membranes and proteins from stress damage [69,70,71]. In the drought-treated switchgrass leaves, the induced proteins include malate synthase (Pavir.Gb01372.1), which is a key enzyme in the glyoxylate cycle for the regeneration of glucose from organic acids (Table S1-4). Maruyama et al. detected an increased level of malate synthase transcripts in rice plants subjected to drought stress, and their data implied that regulation of the glyoxylate cycle may be involved in glucose accumulation in response to dehydration in rice [69]. In the starch metabolic pathway, two proteins showed a significant alteration under drought treatment condition: the downregulated starch synthase protein (Pavir.J06822.1), which participates in starch biosynthesis, and the upregulated β amylase protein (Pavir.J18576.1) involved in the hydrolysis of starch into sugars (Table S1-4). Starch is the main form of carbohydrate storage in most plants and can be rapidly mobilized into soluble sugars. Drought and salt stress generally lead to an active conversion of starch into soluble sugars in leaves [71,72,73]. Plants experiencing environmental stress like cold, heat, drought, or salinity accumulate raffinose family oligosaccharides (RFO) in leaves [71,73,74,75,76,77,78,79]. These sugars have been implicated in membrane protection and radical scavenging [80,81]. In this study, two galactinol synthase proteins (Pavir.J07018.1 and Pavir.J40731.1) were induced in drought-treated leaves (Table S1-4), and these enzymes catalyze formation of galactinol from myo-inositol and UDP-galactose in the biosynthesis of RFO [82]. In summary, the drought-induced proteome changes seem to favor accumulation of soluble sugars, which might serve a role in protecting against cellular dehydration under drought treatment conditions. Additionally, proteins associated with the biosynthesis of free amino acids were markedly upregulated in drought-treated leaves, which include ∆1-pyrroline-5-carboxylate synthetase protein (P5CS) (Pavir.J02344.1), the rate-limiting enzyme in proline biosynthesis, and methionine-γ-lyase protein (MGL) (Pavir.Ib03758.1), which is a precursor in isoleucine (Ile) biosynthesis (Table S1-4). Accumulation of proline (Pro) and branched-chain amino acids is commonly observed in plants subjected to osmotic stress [83,84]. Proline can serve as a free radical scavenger to overcome the oxidative stress by abiotic stress, and the accumulation of this amino acid enhances the ability of plants to grow in water-restricted or saline environments [85]. The accumulation of free isoleucine was induced in response to drought stress in A. thaliana [86]. The activation of these biosynthesis pathways leading to proline and isoleucine accumulation may also serve a critical role in amino acid homeostasis in drought-treated switchgrass leaves. In the drought-treated leaves, the downregulated expression of a regulator of chromosome condensation (RCC; Pavir.Gb00127.1) may have affected the cell division, since it can bind to chromatin and generate a Ras-related nuclear protein (RAN)-guanosine triphosphate (GTP)/RAN-guanosine diphosphate (GDP) (Ran-GTP/Ran-GDP) gradient across the nuclear envelope that is required both to drive nucleocytoplasmic transport and to regulate processes associated with progression of the cell cycle and mitosis [87,88]. This might be a mechanism underlying the smaller leaf areas on drought-treated plants. Additionally, the elevated level of xyloglucan endotransglycosylase (XET) (Pavir.Fa01211.1) may assist in the process of cell wall remodeling with an impact on strengthening the wall layers and protecting mesophyll cells against physiological dehydration stress [89]. 4. Materials and Methods As described in Figure 1, the experiment is comprised of four major steps: drought treatments, protein sample preparation, proteomics analysis, and functional pathway classification of the drought-induced leaf proteomes. 4.1. Construction of a “Sandwich” Drought Treatment System The “sandwich” treatment system was structured to simulate the process of a gradual decline in water content in the surface soil during drought under field conditions. It is comprised of double PVC pipes (an outer pipe and an inner pipe), a PVC sewer and drain coupling, and a PVC sewer and drain cap (Steinhouse Supply Company, Nashville, TN, USA). A fiberglass screen (New York Wire®, Grand Island, NY, USA) was placed inside the inner pipe, which assisted when pulling out the plants for checking root length. The “sandwich” treatment system is divided into three layers: garden soil (25 m), perlite (15 cm), and garden soil (30 cm). The garden soil and perlite were products of Scotts Miracle-Gro Company (Marysville, OH, USA). After withholding water, the top layer of soil becomes drier gradually as the middle perlite layer drains the water quickly and cuts off moisture movement upward, and the moist bottom soil layer serves to induce root growth downward. The water depletion process in the top layer would induce gradual drought stress on plants. A 200SS WATERMARK Soil Moisture Sensor (IRROMETER Company Inc., Riverside, CA, USA) was placed at the bottom of the top soil layer in each growth tube. 4.2. Preparation of Seedling Plants Switchgrass “Alamo” seeds were surface disinfected in 50% household bleach followed by three rinses in deionized water. Seeds were germinated in Magenta boxes partially filled with water and placed on an incubator shaker (50 rpm) at 25 °C for three days. Germinating seeds bearing 1 cm long radicals were transferred into seed cubes (Smithers-Oasis Company, Kent, OH, USA). These seedlings were watered every three days until they had grown to the three-leaf stage. At that stage, they were transplanted into the “sandwich” system and maintained in an open-roofed greenhouse at ambient temperature. The moisture content of the growing medium was maintained (soil water tension <0.01 MPa) until seedling roots reached the perlite layer. Each biological replicate contained 10 tubes each growing two plants, and four biological replicates were set up for drought-treated and non-treated control groups. A randomized block design was used in this study. 4.3. Drought Treatment and Physiological Measurements Two weeks after transplanting, the root length was evaluated every three days. Three samples from each replicate group were selected, randomly, in each inspection. Once the longest roots reached the perlite layer, drought treatment was initiated by withholding water to these test plants. The control groups received normal watering at the rate of 4 L of water every three days for each “sandwich” system. The drought treatment was initiated on 25 April and ended on 15 May 2013. Leaf photosynthetic rate, stomatal conductance, and transpiration rate were measured using a LI-COR 6400 Portable Photosynthesis System (Li-Cor Inc., Lincoln, NE, USA). Two fully expanded young leaves randomly selected from each plant were measured between 10:00 am and 3:00 pm. Light in the leaf chamber was set at 2000 µmol photons/m2/s. Water use efficiency (WUE) was calculated by WUE = leaf photosynthetic rate (Pn)/transpiration (Tr) [43]. Soil water tension was recorded daily. Plant height was collected before (Hb) and after (Ha) the drought treatment and was measured from the bottom of the tiller (start point) to the bottom of the latest node (end point). Relative plant height (Hr=Hb−Ha) was used to compare the difference of plant relative growth rate during the drought treatment. Fresh weight (Wf) of leaves were measured at harvest, and they were dried at 70 °C for three days until a constant dry weight (Wd). Relative water content was calculated by Wr=(Wf−Wd)/Wf. Data analysis was performed using PROC GLM procedure of SAS software (Version: 9.3. SAS Inc., Cary, NC, USA). The effect of drought treatments was analyzed using a randomized block design analysis of variance (ANOVA). When a significant effect of drought treatment was detected, least significant difference (LSD) was used for multiple comparisons. 4.4. Tissue Harvest and Preparation of Protein Samples Twenty days after water withholding, when the top-layer soil moisture declined to below 0.05 MPa and stomatal conductance, respiration, and water use efficiency of leaves showed a significant difference between drought-treated and non-treated control groups, plants were considered to have activated the drought-induced physiological process. The top three fully expanded leaves were cut into approximately 2 cm long pieces, wrapped with aluminum foil, frozen in liquid nitrogen, and stored at −80 °C until protein extraction. Frozen samples were ground into a fine powder under liquid nitrogen using a Retsch Mixer Mill MM 400 (Retsch GmbH, Haan, Germany). Protein extraction followed a previously described protocol [18]. Briefly, leaf tissue powder was washed sequentially in 10% trichloroacetic acid (TCA) in acetone, 80% methanol in 0.1 M ammonium acetate, and 80% acetone with centrifugation to pellet the powder after each step. Protein was then extracted in a phenol (pH 8.0) and dense sodium dodecyl sulfate (SDS) buffer (30% sucrose, 2% SDS, 5% β-mercaptoethanol (v/w) in 0.1 M Tris-HCl, pH 8.0). After incubation at 4 °C for 2 h, the mixture was centrifuged at 16,000× g at 4 °C for 20 min. Protein in the upper phenol phase was precipitated in 0.1 M ammonium acetate in methanol after incubation overnight at −20 °C. After washes in methanol and then acetone, the air-dried protein pellets were wetted in a buffer containing 500 mM triethylammonium bicarbonate (TEAB), 2 M urea, 0.1% sodium dodecyl sulfate (SDS), and a protease inhibitor cocktail for plant cell and tissue extracts (100× dilution in the extraction buffer) (Part #9599; Sigma, St. Louis, MO, USA). For enrichment of low-abundance proteins, the individual protein extracts were processed using a ProteoMiner Protein Enrichment kit (Bio-Rad, Hercules, CA, USA). One milliliter of each protein sample was added to the ProteoMiner columns. Proteins were bound to beads after shaking in the columns using a Mini LabRoller overnight at room temperature. Columns were then washed three times with a wash buffer (150 mM NaCl, 10 mM NaH2PO4, pH 7.4). Then, the columns were incubated at room temperature for 15 min in rehydrated elution reagent (8 M urea, 2% 3-((3-cholamidopropyl) dimethylammonium)-1-propanesulfonate (CHAPS) and 5% acetic acid) before eluting the proteins. Proteins were concentrated using 5 KDa Corning Spin-X UF centrifugal concentrator (Sigma, St. Louis, MO, USA). Protein concentration was determined using a Bradford Assay Kit (Bio-Rad). Protein quality was examined by separating 15 μg of proteins on 10%–20% precast Criterio TGX polyacrylamide gels (Bio-Rad). 4.5. Isobaric Tags for Relative and Absolute Quantification (iTRAQ) Labeling and Mass Spectrometry Analysis For iTRAQ labeling, protein samples containing 100 μg protein each were diluted using a buffer containing 500 mM TEAB, 0.1% SDS, and the same protease inhibitor as described above at the same concentration to reduce urea concentration to below 1 M. Then the protein sample was processed following the instructions of the 8-plex iTRAQ labeling kit [21]. Protein tryptic digestion was conducted using sequence grade modified trypsin (Promega, Madison, WI, USA) after incubation at 37 °C for 16 h. The control samples were labeled with tags 113, 115, 117, and 118 and the treated samples with 114, 116, 119, and 121. After combining all the labeled samples, unbound tags and SDS were removed through cation exchange cartridge (AB SCIEX). Salts and other impurities were removed using reverse-phase (RP) solid-phase extraction procedure involving 1-cm3, 50-mg Sep-Pak C18 cartridges following the manufacturer’s instructions (Waters; Milford, MA, USA). Peptides were eluted in 500 μL 50% (v/v) acetonitrile with 0.1% trifluoroacetic acid (TFA). Samples were dried at reduced pressure using a CentiVac Concentrator (labConco, Kansas City, MO, USA). The peptide samples were subjected to a first dimension of high-pH Ultra Performance Liquid. Chromatography (UPLC) separation using an Acquity UPLC System (Waters) coupled with a robotic fraction collector (Probot; Dionex, Sunnyvale, CA, USA) [21]. One hundred micrograms of the multiplexed sample were injected and fractionated into 48 fractions in a 96-well plate. The 48 fractions were concatenated to yield 22 samples as follows: samples 1–4 and 45–48 were combined to yield two 2nd dimension fractions (samples 1–4 not analyzed in 2nd dimension); then for the remaining samples (5–44), every 20th fraction was combined. For the low-pH second dimension, low-pH RP chromatography was employed. Dried samples were reconstituted with 15 μL of 2% acetonitrile with 0.5% formic acid. Nano-LC separations of tryptic peptides were performed as described previously. The eluent from the analytical column was delivered to the LTQ-Orbitrap Elite (Thermo-Fisher Scientific, Waltham, MA, USA) via a “Plug and Play” nano ion source (CorSolutions LLC, Ithaca, NY, USA). The mass spectrometer was externally calibrated across the m/z range from 375–1800 with Ultramark 1621 for the Fourier transform (FT) mass analyzer, and individual runs were internally calibrated with the background polysiloxane ion at m/z 445.1200025 as a lock mass [24,90,91]. The Orbitrap Elite was operated in the positive ion mode with nanosource voltage set at 1.7 kV and capillary temperature at 250 °C. A parallel data-dependent acquisition (DDA) mode was used to obtain one MS survey scan with the FT mass analyzer, followed by isolation and fragmentation of the 15 most abundant, multiply-charged precursor ions with a threshold ion count higher than 50,000 in both the LTQ mass analyzer and the high energy collisionally induced dissociation (HCD)-based FT mass analyzer at a resolution of 15,000 full width at half maximum (FWHM) and m/z 400. MS survey scans were acquired with resolution set at 60,000 across the survey scan range (m/z 375–1800). Dynamic exclusion was utilized with repeat count set to 1 with a 40 s repeat duration; exclusion list size was set to 500, 20 s exclusion duration, and low and high exclusion mass widths set to 1.5. Fragmentation parameters were set with isolation width at 1.5 m/z, normalized collision energy at 37%, and activation Q at 0.25. Activation time for HCD analysis was 0.1 min. All data were acquired using XCalibur 2.1 (Thermo-Fisher Scientific) [18,24]. Proteins were identified using the MS data to query the switchgrass annotated database (http://www.phytozome.net/) via Mascot v2.3.02 (Matrix Sciences, Boston, MA, USA). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [92] partner repository with the dataset identifier PXD004675 and 10.6019/PXD004675. 4.6. Protein Identification and Quantification, and Statistics Analysis For a protein to be included in the quantitative analysis, it was required that at least two unique peptides have to be identified in all eight biological samples. The intensities of reporter ions of constituent peptides were log2-transformed. Then, log2 fold values from all constituent peptides were subjected to t-test (general linear model procedure) followed by false discovery rate (FDR) corrections to test the statistical significance of the difference in normalized abundance of each protein between the drought-treated and control sample groups [21]. The log2 transformed abundance ratios were then fit to a normal distribution (p < 0.01) [93]. Two standard deviations (i.e., a 95% confidence level) of the log2 fold transformed protein abundance ratio (treated/control) were used as the cutoff for significantly changed proteins. The antilog conversion was used to represent the fold change of proteins. Statistical analyses were performed using SAS (version 9.3; SAS Institute, Cary, NC, USA) [18]. 4.7. Functional Pathway Analysis of Drought-Induced Proteins In the annotated switchgrass database (Panicum virgatum v1.1, Phytozome v11.0), each accession is associated with a unigene accession in Arabidopsis thaliana. The switchgrass annotated genome is not included in the database of the MapMan pathway tools. Therefore, in this study, the A. thaliana database in MapMan (MapMan, version 3.5.1R2, Max Planck Institute of Molecular Plant Physiology, Potsdam-Golm, Germanry) was used to develop the functional pathways [94]. Additional literature and database searches were conducted to develop the association between drought-induced proteins and drought tolerance, and highlight new discoveries using proteomics analysis. 4.8. Statistical Analysis All independent experiments were repeated four times. Experimental data were presented as means and standard deviations (SD). The SAS version 9.0 software (SAS Inc., Cary, NC, USA) was used to perform the analysis of variance (ANOVA) and least significant difference (LSD) tests for the physiological data, and t-tests and FDR tests in the analysis of quantitative proteomics data. 5. Conclusions This study has identified drought-induced changes in leaf proteomes that occurred when plants have shown significant physiological changes from drought-treated to non-treated control conditions. The identified proteins are involved in both ABA-dependent and ABA-independent signaling pathways, and diverting metabolic pathways toward increasing cellular concentrations of soluble sugars and stress-related amino acids (proline and isoleucine). The accumulation of a diverse species of stress proteins can be considered as the hallmark for switchgrass plants to acquire drought tolerance. Information provided in this paper advanced our understanding of molecular mechanisms underlying drought tolerance in C4 plants. Acknowledgments The authors wish to thank Sheng Zhang of the Proteomics and Mass Spectrometry Facility of the Cornell University Institute of Biotechnology for expert technical assistance and helpful discussion; Roger Sauve, Jason de Koff, Fur-Chi Chen, and George Smith at Tennessee State University for discussions in experimental design; and Mrs. Sarabjit Bhatti and Long Zhang for assisting in carrying out the study. This work was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Grant No. 2012-02466, project TENX-1507-SE, and the U.S. Department of Agriculture Agricultural Research Service, Grant No. 1907-21000-036/037-00D. ARS disclaimer: “Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.” Seeds of switchgrass “Alamo” were kindly provided by Jason de Koff, Tennessee State University, Nashville, TN, USA. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1251/s1. Click here for additional data file. Author Contributions Zhujia Ye designed and performed the experiments, and prepared the manuscript as part of her Ph.D. thesis. Kevin J. Howe and Tara Fish conducted mass-spectrometry analysis. Chih-Li Yu conducted photosynthesis measurements. Dafeng Hui designed the program for statistical analysis. Sasikiran Sangireddy and Ikenna Okekeogbu contributed equally and both conducted the drought treatment experiments. Suping Zhou and Theodore W. Thannhauser developed the experimental system and revised the paper. All authors read and agreed with the final manuscript. Suping Zhou and Theodore W. Thannhauser are the corresponding authors and are responsible for all contacts and correspondence. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Flow chart of the drought treatments and quantitative proteomics procedure. Plants were grown in a “sandwich” system (A); During the 20th day of the water withholding period, physiological data were recorded on both drought-treated (B); and well-watered control plants (C). Leaf protein samples were extracted followed by the ProteoMiner enrichment. Quantitative proteomics analysis was performed using the crude leaf protein extracts and the ProteoMiner-enriched samples. Functional pathways were developed using information on the drought-induced changes in the leaf proteomes, and the association between protein expression and physiological properties was developed focusing on drought stress tolerance. Figure 2 Schematic of the drought-induced signaling pathway based on proteome changes in switchgrass leaves. The biosynthesis of abscisic acid (ABA) was increased due to the elevated level of 9-cis-epoxycarotenoid dioxygenases (NCED) protein in drought-treated leaves. The elevated ABA level concurs with the induction of several ABA-responsive transcription factors, such as ABF2 (ABA-responsive elements-binding factor 2), GBF4 (G-box binding factor 4), GRAM, and ABA-responsive proteins including RNS (secreted ribonuclease) and KAT2 (3-ketoacyl-CoA thiolase-2). The ABA-independent signal transduction pathway appears to also play a role in drought-induced molecular regulation in switchgrass leaves. Several signal transduction processes may involve a second messenger (Ca2+). ijms-17-01251-t001_Table 1Table 1 Effects of drought treatments on physiological properties of switchgrass. Treatment Control Drought Soil Water Tension (MPa) 0.00 ± 0.00 A,† 0.08 ± 0.02 B,† Leaf Relative Water Content 77.35 ± 0.01 A 71.08 ± 0.02 B Plant Height (cm) 0 Day drought treatment 18.31 ± 6.18 A 19.08 ± 4.97 A 20 days drought treatment 43.26 ± 9.11 A 39.75 ± 8.49 B Relative plant height 24.96 ± 6.21 A 20.67 ± 6.22 B Photosynthesis Leaf photosynthetic rate (μmol CO2/m2/s) 22.96 ± 3.22 A 21.69 ± 7.17 A Stomatal conductance (mol H2O/m2/s) 0.138 ± 0.03 A 0.125 ± 0.05 B Transpiration rate (mmol H2O/m−2/s) 6.88 ± 1.11 A 6.09 ± 2.15 B Water use efficiency (μmol CO2/mmol H2O) 3.35 ± 0.20 A 3.59 ± 0.25 B Data for all the measurements except plant height were collected after 20 days of the water withholding treatments. Data are presented as means ± standard deviations (SD) of four independent replicates. Within columns, means followed by the same letter are not significantly different (p < 0.01). Leaf relative water content (Wr) was calculated using the following equation: Wr=(Wf−Wd)/Wf, where fresh weight (Wf) was taken immediately after harvest, and dry weight was measured after drying tissues at 70 °C for three days until a constant dry weight (Wd). Plant height was measured from the bottom of the tiller (start point) to the top of the latest node (end point). † Means within columns followed by the same letter are not different at the 1% level. ijms-17-01251-t002_Table 2Table 2 The number of proteins identified in the proteomes identified using the crude leaf protein extracts and ProteoMiner-enriched samples. Protein Classification CLE a PMT b The Number of Proteins from CLE and PMT Proteins identified with one or more peptides The total number of proteins 5493 4839 7006 The number of proteins overlapped in CLE and PMT 3326 The number of proteins identified in CLE 2167 - The number of protein identified in PMT - 1513 Quantified proteins with two or more peptides The total number of proteins 4746 4134 5680 The number of proteins overlapped in CLE and PMT 3200 The number of protein in CLE 1546 - The number of proteins in PMT - 934 Differentially expressed proteins (FDR < 0.01, fold change < 0.06 or > 1.7) The total number of proteins 205 107 257 The number of proteins in CLE and PMT 55 The number of proteins in CLE 150 - The number of proteins in PMT - 52 a The number of proteins identified in the crude leaf protein extracts; b The number of proteins identified in the ProteoMiner enriched samples; CLE: Crude Leaf Extracts; PMT: ProteoMiner-treated; FDR: false discovery rate. ijms-17-01251-t003_Table 3Table 3 The number of proteins identified in the crude leaf protein extracts and ProteoMiner-treated samples. Classification CLE a PMT b CLE and PMT c Molecular Function Abiotic/biotic stress 72 25 116 Cell division/cell cycle 11 7 42 Cell organization 26 11 47 Cell vesicle transport 21 6 31 Development 41 16 46 DNA repair 4 2 7 DNA synthesis 20 15 28 Functional enzyme 62 64 180 Metal binding 4 1 11 Phyto-hormone metabolism 21 11 36 Protein and amino acids activation 15 13 35 Protein degradation 88 39 172 Protein post-translation 27 12 41 Protein synthesis 61 54 209 Protein targeting 21 22 81 Redox balance 31 18 98 RNA transcription/processing 113 74 212 Signaling regulation 82 39 98 Transport 24 35 65 Cellular Metabolism Amino acid metabolism 39 38 91 C1-metabolism 4 5 16 Cell wall synthesis/modification 13 13 20 Fermentation 3 3 6 Glycolysis 9 12 41 Glyoxylate cycle 1 0 10 Lipid metabolism 22 30 51 Major CHO metabolism 11 10 35 Minor CHO metabolism 7 0 26 Mitochondrial electron transport/ATP synthesis 9 9 57 N-metabolism 2 2 7 Nucleotide metabolism 24 14 53 Oxidative pentose phosphate (OPP) pathway 7 3 12 Photosystem. Calvin cycle 4 6 36 Photosystem. Light reaction 15 10 82 Photorespiration 3 1 14 S-assimilation 2 2 5 Secondary metabolism 18 29 67 TCA cycle 8 10 52 Tetrapyrrole synthesis 13 9 20 Others and not assigned proteins 588 264 944 Total 1546 934 3200 a The number of proteins identified in the crude leaf protein extracts (CLE); b The number of proteins identified in the ProteoMiner-treated samples (PMT); c The number of proteins combining the proteomes identified in CLE and PMT. ==== Refs References 1. Wright L.I. Cushman J.H. Ehrenshaft A.R. McLaughlin S.B. McNabb W.A. Martin S.A. Ranney J.W. Tuskan A.G. Turhollow A.F. Biofuels Feedstock Development Program Annual Progress Report for 1992 ORNL-6781 Environmental Sciences Division Publication Washington, DC, USA 1993 2. Parrisha D.J. Fike J.H. The biology and agronomy of switchgrass for biofuels CRC. Crit. Rev. Plant Sci. 2005 24 423 459 10.1080/07352680500316433 3. Wright L.L. Historical Perspective on How and Why Switchgrass Was Selected as a “Model” High-Potential Energy Crop ORNL/TM-2007/109 Oak Ridge National Laboratory Oak Ridge, TN, USA 2007 4. Nezhadahmadi A. Prodhan Z.H. Faruq G. Drought tolerance in wheat Sci. World J. 2013 2013 610721 10.1155/2013/610721 24319376 5. National Weather Service Centers for Environmental Prediction Available online: http://www.cpc.ncep.noaa.gov/products/monitoring_and_data/topsoil.shtml (accessed on 1 January 2016) 6. Keyser P. Harper C. Bates G. Waller J. Doxon E. Native warm-season grasses for mid-south forage production UT Ext. 2011 1251 7. Sanderson M.A. Reed R.L. Switchgrass growth and development: Water, nitrogen, and plant density effects J. Range Manag. 2000 53 221 227 10.2307/4003287 8. Vogel K.P. Improved plant & production practices for grasslands & biomass crops in the mid-continental USA Proceedings of the DOE/USDA Biomass Feedstock Gate Review Meeting Washington, DC, USA 14–16 March 2005 9. Farooq M. Wahid A. Kobayashi N. Fujita D. Basra S.M.A. Plant drought stress: Effects, mechanisms and management Agron. Sustain. Dev. 2009 29 185 212 10.1051/agro:2008021 10. Nonami H. Plant water relations and control of cell elongation at low water potentials J. Plant Res. 1998 111 373 382 10.1007/BF02507801 11. Kaya M.D. Okçub G. Ataka M. Çıkılıc Y. Kolsarıcıa Ö. Seed treatments to overcome salt and drought stress during germination in sunflower (Helianthus annuus L.) Eur. J. Agron. 2006 24 291 295 10.1016/j.eja.2005.08.001 12. Hussain M. Malik M.A. Farooq M. Ashraf M.Y. Cheema M.A. Improving drought tolerance by exogenous application of glycinebetaine and salicylic acid in sunflower J. Agron. Crop Sci. 2008 194 193 199 10.1111/j.1439-037X.2008.00305.x 13. Liu Y. Zhang X. Tran H. Shan L. Kim J. Childs K. Ervin E.H. Frazier T. Zhao B. Assessment of drought tolerance of 49 switchgrass (Panicum virgatum ) genotypes using physiological and morphological parameters Biotechnol. Biofuels 2015 8 10.1186/s13068-015-0342-8 26396590 14. Alberts B. Johnson A. Lewis J. Molecular Biology of the Cell 4th ed. Garland Science New York, NY, USA 2002 15. Barkla B.J. Castellanos-Cervantes T. de León J.L.D. Matros A. Mock H.P. Perez-Alfocea F. Salekdeh G.H. Witzel K. Zörb C. Elucidation of salt stress defense and tolerance mechanisms of crop plants using proteomics—current achievements and perspectives Proteomics 2013 13 1885 1900 10.1002/pmic.201200399 23723162 16. Ghosh D. Xu J. Abiotic stress responses in plant roots: A proteomics perspective Front. Plant Sci. 2014 5 1 13 10.3389/fpls.2014.00006 24478786 17. Ngara R. Ndimba B.K. Understanding the complex nature of salinity and drought-stress response in cereals using proteomics technologies Proteomics 2014 14 611 621 10.1002/pmic.201300351 24339029 18. Okekeogbu I. Ye Z. Sangireddy S. Li H. Bhatti S. Hui D. Zhou S. Howe K. Fish T. Yang Y. Effect of aluminum treatment on proteomes of radicles of seeds derived from AL-treated tomato plants Proteomes 2014 2 169 190 10.3390/proteomes2020169 19. Zhou S. Sauve R. Fish T. Thannhauser T.W. Salt-induced and Salt-suppressed Proteins in Tomato Leaves J. Am. Soc. Hortic. Sci. 2009 134 289 294 20. Zhou S. Sauvé R. Liu Z. Reddy S. Bhatti S. Heat-induced proteome changes in tomato leaves J. Am. Soc. Hortic. Sci. 2011 136 219 226 21. Zhou S. Palmer M. Zhou J. Bhatti S. Howe K. Fish T. Thannhauser T.W. Differential root proteome expression in tomato genotypes with contrasting drought tolerance exposed to dehydration J. Am. Soc. Hort. Sci. 2013 138 131 141 22. Pottiez G. Wiederin J. Fox H.S. Ciborowski P. Comparison of 4-plex to 8-plex iTRAQ quantitative measurements of proteins in human plasma samples J. Proteome Res. 2012 11 3774 3781 10.1021/pr300414z 22594965 23. Chen X. Walker A.K. Strahler J.R. Simon E.S. Tomanicek-Volk S.L. Nelson B.B. Hurley M.C. Ernst S. Williams J. Andrews P.C. Organellar proteomics: Analysis of pancreatic zymogen granule membranes Mol. Cell. Proteom. 2006 5 306 312 10.1074/mcp.M500172-MCP200 16278343 24. Yang Q.S. Wu J.H. Li C.Y. Wei Y.R. Sheng O. Hu C.H. Kuang R.-B. Huang Y.-H. Peng X.-X. McCardle J. Quantitative proteomic analysis reveals that antioxidation mechanisms contribute to cold tolerance in plantain (Musa paradisiaca L.; ABB Group) seedlings Mol. Cell. Proteom. 2012 11 1853 1869 10.1074/mcp.M112.022079 22982374 25. Lan P. Li W.F. Wen T.N. Shiau J.Y. Wu Y.C. Lin W.D. Schmidt W. iTRAQ protein profile analysis of arabidopsis roots reveals new aspects critical for iron homeostasis Plant Physiol. 2011 155 821 834 10.1104/pp.110.169508 21173025 26. Redding A.M. Mukhopadhyay A. Joyner D.C. Hazen T.C. Keasling J.D. Study of nitrate stress in desulfovibrio vulgaris hildenborough using iTRAQ proteomics Brief. Funct. Genom. Proteom. 2006 5 133 143 10.1093/bfgp/ell025 16772278 27. Nveawiah-Yoho P. Zhou J. Palmer M. Sauve R. Zhou S. Howe K.J. Fish T. Thannhauser T.W. Identification of proteins for salt tolerance using a comparative proteomics analysis of tomato accessions with contrasting salt tolerance J. Am. Soc. Hort. Sci. 2013 138 382 394 28. Nveawiah-Yoho P. Mechanisms for Salt Tolerance and Susceptibility in Tomato Ph.D. Thesis Tennessee State University Nashville, TN, USA 2012 29. Ali G.M. Komatsu S. Proteomic analysis of rice leaf sheath during drought stress J. Proteom. Res. 2006 5 396 403 10.1021/pr050291g 16457606 30. Ke Y. Han G. He H. Li J. Differential regulation of proteins and phosphoproteins in rice under drought stress Biochem. Biophys. Res. Commun. 2009 379 133 138 10.1016/j.bbrc.2008.12.067 19103168 31. Salekdeh G. Siopongco J. Wade L. Ghareyazie B. Bennett J. Proteomic analysis of rice leaves during drought stress and recovery Proteomics 2002 2 1131 1145 10.1002/1615-9861(200209)2:9<1131::AID-PROT1131>3.0.CO;2-1 12362332 32. Salekdeh G. Siopongco J. Wade L. Ghareyazie B. Bennett J. A proteomic approach to analyzing drought- and salt-responsiveness in rice F. Crop. Res. 2002 76 199 219 10.1016/S0378-4290(02)00040-0 33. Shu L. Ding W. Wu J. Feng F. Luo L. Proteomic analysis of rice leaves shows the different regulations to osmotic stress and stress signals J. Integr. Plant Biol. 2010 52 981 995 10.1111/j.1744-7909.2010.00986.x 20977656 34. Xiong J. Fu B. Xu H. Li Y. Proteomic analysis of PEG-simulated drought stress-responsive proteins of rice leaves using a pyramiding rice line at the seedling stage Bot. Stud. 2010 51 137 145 35. De Vienne D. Leonardi A. Damerval C. Zivy M. Genetics of proteome variation for QTL characterization: Application to drought-stress responses in maize J. Exp. Bot. 1999 50 303 309 10.1093/jxb/50.332.303 36. Mohammadkhani N. Heidari R. Effects of drought stress on soluble proteins in two maize varieties Turkish J. Biol. 2008 32 23 30 37. Riccardi F. Gazeau P. de Vienne D. Zivy M. Protein changes in response to progressive water deficit in maize Plant Physiol. 1998 117 1253 1263 10.1104/pp.117.4.1253 9701581 38. McCabe M.S. Garratt L.C. Schepers F. Jordi W.J. Stoopen G.M. Davelaar E. van Rhijn J.H. Power J.B. Davey M.R. Effects of P(SAG12)-IPT gene expression on development and senescence in transgenic lettuce Plant Physiol. 2001 127 505 516 10.1104/pp.010244 11598225 39. Widjaja I. Naumann K. Roth U. Wolf N. Mackey D. Dangl J.L. Scheel D. Lee J. Combining subproteome enrichment and Rubisco depletion enables identification of low abundance proteins differentially regulated during plant defense Proteomics 2009 9 138 147 10.1002/pmic.200800293 19053141 40. Fasoli E. D’Amato A. Kravchuk A.V. Boschetti E. Bachi A. Righetti P.G. Popeye strikes again: The deep proteome of spinach leaves J. Proteom. 2011 74 127 136 10.1016/j.jprot.2010.10.009 21056706 41. Von Toerne C. Kahle M. Schäfer A. Ispiryan R. Blindert M. Hrabe De Angelis M. Neschen S. Ueffing M. Hauck S.M. Apoe, Mbl2, and Psp plasma protein levels correlate with diabetic phenotype in NZO mice-An optimized rapid workflow for SRM-based quantification J. Proteome Res. 2013 12 1331 1343 10.1021/pr3009836 23350727 42. Fonslow B.R. Carvalho P.C. Academia K. Freeby S. Xu T. Nakorchevsky A. Paulus A. Yates J.R. Improvements in proteomic metrics of low abundance proteins through proteome equalization using ProteoMiner prior to MudPIT J. Proteome Res. 2011 10 3690 3700 10.1021/pr200304u 21702434 43. Polley H.W. Implications of atmospheric and climatic change for crop yield and water use efficiency Crop Sci. 2002 42 131 140 10.2135/cropsci2002.0131 11756263 44. Finkelstein R.R. Lynch T.J. The Arabidopsis abscisic acid response gene ABI5 encodes a basic leucine zipper transcription factor Plant Cell 2000 12 599 609 10.1105/tpc.12.4.599 10760247 45. Rösti J. Barton C.J. Albrecht S. Dupree P. Pauly M. Findlay K. Roberts K. Seifert G.J. UDP-glucose 4-epimerase isoforms UGE2 and UGE4 cooperate in providing UDP-galactose for cell wall biosynthesis and growth of Arabidopsis thaliana Plant Cell 2007 19 1565 1579 10.1105/tpc.106.049619 17496119 46. Uozu S. Tanaka-Ueguchi M. Kitano H. Hattori K. Matsuoka M. Characterization of XET-related genes of rice Plant Physiol. 2000 122 853 859 10.1104/pp.122.3.853 10712549 47. Nambara E. Marion-Poll A. Abscisic acid biosynthesis and catabolism Annu. Rev. Plant Biol. 2005 56 165 185 10.1146/annurev.arplant.56.032604.144046 15862093 48. Leckie C.P. McAinsh M.R. Allen G.J. Sanders D. Hetherington A.M. Abscisic acid-induced stomatal closure mediated by cyclic ADP-ribose Proc. Natl. Acad. Sci. USA 1998 95 15837 15842 10.1073/pnas.95.26.15837 9861057 49. Wang Y. Ries A. Wu K. Yang A. Crawford N.M. The Arabidopsis prohibitin gene PHB3 functions in nitric oxide-mediated responses and in hydrogen peroxide-induced nitric oxide accumulation Plant Cell 2010 22 249 259 10.1105/tpc.109.072066 20068191 50. García-Mata C. Lamattina L. Nitric oxide induces stomatal closure and enhances the adaptive plant responses against drought stress Plant Physiol. 2001 126 1196 1204 10.1104/pp.126.3.1196 11457969 51. Lefebvre V. North H. Frey A. Sotta B. Seo M. Okamoto M. Nambara E. Marion-Poll A. Functional analysis of Arabidopsis NCED6 and NCED9 genes indicates that ABA synthesized in the endosperm is involved in the induction of seed dormancy Plant J. 2006 45 309 319 10.1111/j.1365-313X.2005.02622.x 16412079 52. Huo H. Dahal P. Kunusoth K. McCallum C.M. Bradford K.J. Expression of 9-cis -EPOXYCAROTENOID DIOXYGENASE4 is essential for thermoinhibition of lettuce seed germination but not for seed development or stress tolerance Plant Cell 2013 25 884 900 10.1105/tpc.112.108902 23503626 53. Tan B.C. Joseph L.M. Deng W.T. Liu L. Li Q.B. Cline K. McCarty D.R. Molecular characterization of the Arabidopsis 9-cis epoxycarotenoid dioxygenase gene family Plant J. 2003 35 44 56 10.1046/j.1365-313X.2003.01786.x 12834401 54. Seo M. Kanno Y. Frey A. North H.M. Marion-Poll A. Dissection of Arabidopsis NCED9 promoter regulatory regions reveals a role for ABA synthesized in embryos in the regulation of GA-dependent seed germination Plant Sci. 2016 246 91 97 10.1016/j.plantsci.2016.02.013 26993239 55. Iuchi S. Kobayashi M. Taji T. Naramoto M. Seki M. Kato T. Regulation of drought tolerance by gene manipulation of 9-cis -epoxycarotenoid dioxygenase, a key enzyme in abscisic acid biosynthesis in Arabidopsis Plant J. 2001 27 325 333 10.1046/j.1365-313x.2001.01096.x 11532178 56. Kim S. Kang J. Cho D.-I. Park J.H. Kim S.Y. ABF2, an ABRE-binding bZIP factor, is an essential component of glucose signaling and its overexpression affects multiple stress tolerance Plant J. 2004 40 75 87 10.1111/j.1365-313X.2004.02192.x 15361142 57. Hillwig M.S. LeBrasseur N.D. Green P.J. MacIntosh G.C. Impact of transcriptional, ABA-dependent, and ABA-independent pathways on wounding regulation of RNS1 expression Mol. Genet. Genom. 2008 280 249 261 10.1007/s00438-008-0360-3 18607631 58. Jiang T. Zhang X.F. Wang X.F. Zhang D.P. Arabidopsis 3-Ketoacyl-CoA Thiolase-2 (KAT2), an enzyme of fatty acid β-Oxidation, is involved in ABA signal transduction Plant Cell Physiol. 2011 52 528 538 10.1093/pcp/pcr008 21257607 59. Knight M.R. Campbell A.K. Smith S.M. Trewavas A.J. Transgenic plant aequorin reports the effects of touch and cold-shock and elicitors on cytoplasmic calcium Nature 1991 352 524 526 10.1038/352524a0 1865907 60. Knight M. Read N. Campbell A. Trewavas A. Imaging calcium dynamics in living plants using semi-synthetic recombinant aequorins J. Cell Biol. 1993 121 83 90 10.1083/jcb.121.1.83 8458875 61. Knight M.R. Smith S.M. Trewavas A.J. Wind-induced plant motion immediately increases cytosolic calcium Proc. Natl. Acad. Sci. USA 1992 89 4967 4971 10.1073/pnas.89.11.4967 11536497 62. Braam J. Regulated expression of the calmodulin-related TCH genes in cultured Arabidopsis cells: Induction by calcium and heat shock Proc. Natl. Acad. Sci. USA 1992 89 3213 3216 10.1073/pnas.89.8.3213 1373491 63. Haley A.N.N. Russell A.J. Wood N. Allan A.C. Knight M. Campbell A.K. Trewavas A.J. Ryan C.A. Lamb C.J. Jagendorf A.T. Effects of mechanical signaling on plant cell cytosolic calcium Proc. Natl. Acad. Sci. USA 1995 92 4124 4128 10.1073/pnas.92.10.4124 11536690 64. Klee C. Vanaman T. Calmodulin Adv. Protein Chem. 1982 35 213 322 6762067 65. Roberts D. Lukas T. Watterson D. Structure, function, and mechanisms of action of calmodulin CRC Crit. Rev. Plant Sci. 1986 4 311 339 10.1080/07352688609382230 66. Cohen P. Klee C. Calmodulin Elsevier Biomedical Press Amsterdam, The Netherlands 1988 67. Allan E. Hepler P. Calmodulin and calcium-binding proteins The Biochemistry of Plants Stumpf P. Lonn E. Academic Press New York, NY, USA 1989 455 484 68. Roberts D.M. Harmon A.C. Calcium modulated proteins targets of intracellular calcium signals in higher plants Annu. Rev. Plant Physiol. Plant Mol. Biol. 1992 43 375 414 10.1146/annurev.pp.43.060192.002111 69. Maruyama K. Urano K. Yoshiwara K. Morishita Y. Sakurai N. Suzuki H. Kojima M. Sakakibara H. Shibata D. Saito K. Integrated analysis of the effects of cold and dehydration on rice metabolites, phytohormones, and gene transcripts Plant Physiol. 2014 164 1759 1771 10.1104/pp.113.231720 24515831 70. Madden T. Bally M. Hope M. Cullis P. Schieren H. Janoff A. Protection of large unilamellar vesicles by trehalose during dehydration: Retention of vesicle contents Biochim. Biophys. Acta 1985 817 67 74 10.1016/0005-2736(85)90069-0 4005259 71. Kaplan F. Guy C.L. β-amylase induction and the protective role of maltose during temperature shock Plant Physiol. 2004 135 1674 1684 10.1104/pp.104.040808 15247404 72. Basu P.S. Ali M. Chaturvedi S.K. Osmotic adjustment increases water uptake, remobilization of assimilates and maintains photosynthesis in chickpea under drought Indian J. Exp. Biol. 2007 45 261 267 17373371 73. Kempa S. Krasensky J. Dal Santo S. Kopka J. Jonak C. A Central Role of Abscisic Acid in Stress-Regulated Carbohydrate Metabolism PLoS ONE 2008 3 1251 10.1371/journal.pone.0003935 19081841 74. Castonguay Y. Nadeau P. Enzymatic control of soluble carbohydrate accumulation in cold-acclimated crowns of alfalfa Crop Sci. 1998 38 1183 1189 10.2135/cropsci1998.0011183X003800050012x 75. Gilmour S. Sebolt A. Salazar M. Everard J. Thomashow M. Overexpression of the Arabidopsis CBF3 transcriptional activator mimics multiple biochemical changes associated with cold acclimation Plant Physiol. 2000 124 1854 1865 10.1104/pp.124.4.1854 11115899 76. Taji T. Ohsumi C. Iuchi S. Seki M. Kasuga M. Kobayashi M. Yamaguchi-Shinozaki K. Shinozaki K. Important roles of drought- and cold-inducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana Plant J. 2002 29 417 426 10.1046/j.0960-7412.2001.01227.x 11846875 77. Cook D. Fowler S. Fiehn O. Thomashow M.F. A prominent role for the CBF cold response pathway in conFigure uring the low-temperature metabolome of Arabidopsis Proc. Natl. Acad. Sci. USA 2004 101 15243 15248 15383661 78. Peters S. Mundree S. Thomson J. Farrant J. Keller F. Protection mechanisms in the resurrection plant Xerophyta viscosa (Baker): Both sucrose and raffinose family oligosaccharides (RFOs) accumulate in leaves in response to water deficit J. Exp. Bot. 2007 58 1947 1956 10.1093/jxb/erm056 17452754 79. Usadel B. Bläsing O.E. Gibon Y. Poree F. Höhne M. Günter M. Trethewey R. Kamlage B. Poorter H. Stitt M. Multilevel genomic analysis of the response of transcripts, enzyme activities and metabolites in Arabidopsis rosettes to a progressive decrease of temperature in the non-freezing range Plant Cell Environ. 2008 31 518 547 10.1111/j.1365-3040.2007.01763.x 18088337 80. Hincha D.K. Effects of calcium-induced aggregation on the physical stability of liposomes containing plant glycolipids Biochim. Biophys. Acta 2003 1611 180 186 10.1016/S0005-2736(03)00053-1 12659959 81. Nishizawa A. Yabuta Y. Shigeoka S. Galactinol and raffinose constitute a novel function to protect plants from oxidative damage Plant Physiol. 2008 147 1251 1263 10.1104/pp.108.122465 18502973 82. Peterbauer T. Richter A. Biochemistry and physiology of raffinose family oligosaccharides and galactosyl cyclitols in seeds Seed Sci. Res. 2001 11 185 197 83. Rhodes D. Handa S. Bressan R. Metabolic changes associated with adaptation of plant cells to water stress Plant Physiol. 1986 82 890 903 10.1104/pp.82.4.890 16665163 84. Girousse C. Bournoville R. Bonnemain J.L. Water deficit-induced changes in concentrations in proline and some other amino acids in the phloem sap of alfalfa Plant Physiol. 1996 111 109 113 12226278 85. Hong Z. Removal of feedback inhibition of delta 1-pyrroline-5-carboxylate synthetase results in increased proline accumulation and protection of plants from osmotic stress Plant Physiol. 2000 122 1129 1136 10.1104/pp.122.4.1129 10759508 86. Joshi V. Jander G. Arabidopsis methionine garmma-lyase is regulated according to isoleucine biosynthesis needs but Plays a subordinate Plant Physiol. 2009 151 367 378 10.1104/pp.109.138651 19571310 87. Bischoff F.R. Ponstingl H. Catalysis of guanine nucleotide exchange on Ran by the mitotic regulator RCC1 Nature 1991 354 80 82 10.1038/354080a0 1944575 88. Nemergut M.E. Lindsay M.E. Brownawell A.M. Macara I.G. Ran-binding protein 3 links Crm1 to the Ran guanine nucleotide exchange factor J. Biol. Chem. 2002 277 17385 17388 10.1074/jbc.C100620200 11932251 89. Cho S.K. Kim J.E. Park J.A. Eom T.J. Kim W.T. Constitutive expression of abiotic stress-inducible hot pepper CaXTH3, which encodes a xyloglucan endotransglucosylase/hydrolase homolog, improves drought and salt tolerance in transgenic Arabidopsis plants FEBS Lett. 2006 580 3136 3144 10.1016/j.febslet.2006.04.062 16684525 90. Thannhauser T. Shen M. Sherwood R. Howe K. Fish T. Yang Y. Chen W. Zhang S. A workflow for large-scale empirical identification of cell wall N-linked glycoproteins of tomato (Solanum lycopersicum ) Electrophoresis 2013 34 2417 2431 10.1002/elps.201200656 23580464 91. Yang Y. Qiang X. Owsiany K. Zhang S. Thannhauser T.W. Li L. Evaluation of different multidimensional LC-MS/MS pipelines for iTRAQ-based proteomic analysis of potato tubers in response to cold storage J. Proteome Res. 2011 10 4647 4660 10.1021/pr200455s 21842911 92. Vizcaíno J.A. Csordas A. del-Toro N. Dianes J.A. Griss J. Lavidas I. Mayer G. Perez-Riverol Y. Reisinger F. Ternent T. 2016 update of the PRIDE database and related tools Nucleic Acids Res. 2016 44 D447 D456 10.1093/nar/gkv1145 26527722 93. Zhou S. Okekeogbu I. Sangireddy S. Ye Z. Li H. Bhatti S. Hui D. Mcdonald D.W. Yang Y. Giri S. Proteome modification in tomato plants upon long-term aluminum treatment J. Proteome Res. 2016 15 1670 1684 10.1021/acs.jproteome.6b00128 27052409 94. MapMan, Version 3.5.1R2 2015 Available online: http://mapman.gabipd.org (accessed on 10 April 2016)

PMC005xxxxxx/PMC5000650.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081252ijms-17-01252ArticleEssential Oil of Cymbopogon nardus (L.) Rendle: A Strategy to Combat Fungal Infections Caused by Candida Species De Toledo Luciani Gaspar 1Ramos Matheus Aparecido Dos Santos 1Spósito Larissa 1Castilho Elza Maria 2Pavan Fernando Rogério 1Lopes Érica De Oliveira 1Zocolo Guilherme Julião 3Silva Francisca Aliny Nunes 3Soares Tigressa Helena 3dos Santos André Gonzaga 4Bauab Taís Maria 1*De Almeida Margarete Teresa Gottardo 2Battino Maurizio Academic Editor1 Department of Biological Sciences, School of Pharmaceutical Sciences, Universidade Estadual Paulista, Rodovia Araraquara-Jaú, km. 01, Araraquara, 14800-903 São Paulo, Brazil; [email protected] (L.G.D.T.); [email protected] (M.A.D.S.R.); [email protected] (L.S.); [email protected] (F.R.P.); [email protected] (É.D.O.L.)2 Department of Infectious Diseases, Faculty of Medicine of São José do Rio Preto, São José do Rio Preto, 15090-000 São Paulo, Brazil; [email protected] (E.M.C.); [email protected] (M.T.G.D.A.)3 Brazilian Agricultural Research Corporation, Embrapa Tropical Agroindustry, 60511-110 Fortaleza, Brazil; [email protected] (G.J.Z.); [email protected] (F.A.N.S.); [email protected] (T.H.S.)4 Department of Natural Active Principles and Toxicology, School of Pharmaceutical Sciences, Universidade Estadual Paulista, Araraquara, 14800-903 São Paulo, Brazil; [email protected]* Correspondence: [email protected]; Tel.: +55-16-3301-6955; Fax: +55-16-3301-694009 8 2016 8 2016 17 8 125230 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Background: The incidence of fungal infections, especially those caused by Candida yeasts, has increased over the last two decades. However, the indicated therapy for fungal control has limitations. Hence, medicinal plants have emerged as an alternative in the search for new antifungal agents as they present compounds, such as essential oils, with important biological effects. Published data demonstrate important pharmacological properties of the essential oil of Cymbopogon nardus (L.) Rendle; these include anti-tumor, anti-nociceptive, and antibacterial activities, and so an investigation of this compound against pathogenic fungi is interesting. Objective: The aim of this study was to evaluate the chemical composition and biological potential of essential oil (EO) obtained from the leaves of C. nardus focusing on its antifungal profile against Candida species. Methods: The EO was obtained by hydrodistillation and analyzed by gas chromatography-mass spectrometry (GC-MS). Testing of the antifungal potential against standard and clinical strains was performed by determining the minimal inhibitory concentration (MIC), time-kill, inhibition of Candida albicans hyphae growth, and inhibition of mature biofilms. Additionally, the cytotoxicity was investigated by the IC50 against HepG-2 (hepatic) and MRC-5 (fibroblast) cell lines. Results: According to the chemical analysis, the main compounds of the EO were the oxygen-containing monoterpenes: citronellal, geranial, geraniol, citronellol, and neral. The results showed important antifungal potential for all strains tested with MIC values ranging from 250 to 1000 μg/mL, except for two clinical isolates of C. tropicalis (MIC > 1000 μg/mL). The time-kill assay showed that the EO inhibited the growth of the yeast and inhibited hyphal formation of C. albicans strains at concentrations ranging from 15.8 to 1000 μg/mL. Inhibition of mature biofilms of strains of C. albicans, C. krusei and C. parapsilosis occurred at a concentration of 10× MIC. The values of the IC50 for the EO were 96.6 μg/mL (HepG-2) and 33.1 μg/mL (MRC-5). Conclusion: As a major virulence mechanism is attributed to these types of infections, the EO is a promising compound to inhibit Candida species, especially considering its action against biofilm. Cymbopogon nardusessential oilgas chromatographyCandidaantifungal activity ==== Body 1. Introduction Candida species have been a problem in human clinical practice due to the significant increase in cases of infection, especially in immunocompromised patients. The immune status of the host, the use of broad-spectrum antibiotics and corticosteroids, transplants, long-term intravascular and urethral catheters, and parenteral nutrition, are mentioned as risk factors in the development and increased incidence of fungal infections [1]. Candida species can develop on mucous membranes of the human body; this is associated with various types of diseases ranging from mucocutaneous overgrowth to disseminated infections [2]. Although the C. albicans is the prevalent species in candidemia, other species, such as C. krusei, C. glabrata, C. tropicalis, and C. parapsilosis, have been observed [3]. The ability of Candida species to cause disease is mainly related to mechanisms involving different virulence factors that include the morphological transition between yeast and hyphae, ability to defend themselves against the host immune system, adhesion, biofilm formation on host tissue or on medical devices, and production of harmful enzymes, such as hydrolytic proteases, phospholipases, and hemolysin [4]. Several antifungal agents have been indicated in the treatment of these infections, including those belonging to the polyenic, azole, and echinocandin classes; however, due to the indiscriminate use of these antimicrobial medications and physiological characteristics of the fungus, there has been a significant increase in resistance. Furthermore, the high toxicity, drug interactions, and insufficient bioavailability of active ingredients contribute to therapeutic failure [5]. Essential oils (EOs) from plants may be alternative bioactive compounds with antifungal properties because of the presence of secondary metabolites, such as tannins, terpenes, alkaloids, and flavonoids, etc. [6,7]. The genus Cymbopogon of the Poaceae family has been investigated for its pharmacological potential. Cymbopogon nardus (L.) Rendle, popularly known as citronella, is a grass cultivated in subtropical and tropical regions of Asia, Africa, and America, including Brazil [8]. The EO of the leaves of C. nardus is commonly used in perfumes, the production of cosmetics, and as an insect repellent. The major chemical constituents are geraniol, citral, citronellal, and citronellol [9]. Studies have demonstrated the antiviral [10], antibacterial [11], and antifungal activities [12] of this oil. The EO is a complex mixture of monoterpene and sesquiterpene hydrocarbons (10 and 15 carbon atoms, respectively), and their oxygenated derivatives such as alcohols, aldehydes, and ketones, phenylpropanoids, and other minor compounds [13]. EOs are also called volatile oils or ethereal oils, as they have a high degree of evaporation when exposed to air at room temperature; this feature confers the strong odor to plants, both to attract pollinators and to repel insects and herbivores [14]. EOs are important in several areas of science, especially in combatting pathogenic or opportunistic microorganisms [15,16]. The presence of terpenes, as one of the chemical compounds in EO, contributes to the complex constitution with the action against microorganisms being directly related to this characteristic [17]. The antimicrobial potential demonstrated by terpenes (e.g., monoterpenes) is attributed to their interference in the integrity and functioning of the cell membrane through induction of changes in membrane potential, loss of cytoplasmic material and inhibition of the respiratory chain. Thus, these characteristics of EO are relevant in the search of new antifungal agents [18]. Considering the fungal etiology of different diseases with great impact on public health, and reports on the use of plants of the Cymbopogon genus in medical literature, the aim of this study was to evaluate the chemical composition and biological potential of the EO of C. nardus. This study focuses on the exploration of the antifungal profile against Candida species in order to present this compound as a possible antifungal or adjuvant agent. 2. Results and Discussion 2.1. Chemical Composition of Essential Oil The qualitative and quantitative composition (GC-MS) of the EO is shown in Table 1. Oxygen-containing monoterpenes were the major constituents (90.61%), with citronellal (27.87%), geraniol (22.77%), geranial (14.54%), citronellol (11.85%), and neral (11.21%) as the main compounds. These monoterpenes are derived from geranyl diphosphate and are biosynthetically related [19]. The retention time (tR) of citronellal was 11.2 min by GC-FID analysis. The equation and R2 value obtained from the analytical curve for citronellal were y = 571,529.9016x − 102,555.3281 and 0.99976. The concentration of citronellal in the EO (GC-FID) was determined by means of the external standardization method as 282.5 mg/mL. Considering the relative density of commercial EO at 25 °C—0.897 g/mL (Sigma-Aldrich, St. Louis, MO, USA, 2016), the concentration of citronellal in the EO can be expressed as 31% (m/m). This value is consistent with the value obtained in the GC-MS analysis (28%). In a study performed by Wei and Wee [22] the concentration of citronellal, the major compound (29.6%), was similar to this work. Koba et al. [23] and Trindade et al. [24] found higher concentrations of citronellal at 35.5% and 37.75%, respectively. In Thailand, the concentrations were different [13]: geraniol (35.7%), trans-citral (22.7%), cis-citral (14.2%), geranyl acetate (5.8%), citronellal (5.8%), and citronellol (4.6%). In another recent study, the authors also obtained a different chemical composition of EO with the main compounds being geraniol (25.9%), citronellal (3.7%), and citronellol (3.1%) [25]. 2.2. Minimal Inhibitory Concentration and Minimal Fungicidal Concentration of Essential Oil of C. nardus The antifungal activity of the EO is shown in Table 2. The solvent and growth controls presented satisfactory results. Thus, the antifungal activity was attributed to essential oil. The results show that the EO had effective antifungal activity with a MIC range of 250–1000 μg/mL, including for isolates resistant to fluconazole and amphotericin-B. The lowest MIC value (250 μg/mL) of the EO was seen against C. krusei. Furthermore, the EO showed fungicidal activity against all fungi except two clinical isolates of C. tropicalis that were resistant to the EO with MIC > 1000 μg/mL. Unlike conventional antimicrobial drugs, the literature does not present a standard of MIC values (sensitive and resistant) for natural products against Candida species. A study performed by Webster et al. [26] that evaluated antifungal activity of 14 medicinal plant extracts, found MIC values equal to, or lower than, 1000 μg/mL. The authors believe that values equal to or lower than 1000 μg/mL confirm sensitivity. The values observed in this study were satisfactory, as the EO exhibited inhibitory action against 90% of the strains tested. Although some strains were inhibited with the highest concentration evaluated (1000 μg/mL), these data are relevant, since most of the strains are resistant to fluconazole (MIC > 64 μg/mL), the main drug used in the medical practice [27]. Interestingly, the MIC values (500 to 250 μg/mL) of the EO against the C. krusei ATCC clinical isolate are promising, due to the fact that this species presents intrinsic resistance to azoles [28]. The study carried out by Nakahara et al. [12] demonstrated that the EO inhibited filamentous fungus from the environment, however, the methodology used to determine the MIC was different from this study. Recent research by Trindade et al. [24] showed the antifungal activity of the EO against ATCC and clinical strains of C. albicans and C. tropicalisi, with MIC values ranging from 32 to 64 g/mL. The differences found are expected because factors, such as climate, region, and the time of harvest of C. nardus, in addition to the extraction method, can directly affect the characteristics and concentration of chemical compounds [29]. The assay used to determine the MFC showed that the fungicidal properties of EO against Candida species were capable of killing the fungal cells using the concentrations evaluated in this study. The antifungal activity of terpenoids, one of the major groups of volatile secondary metabolites, is known in the pharmaceutical field [17]. Thus, the antifungal activity of the EO in this present study may be related to the monoterpenes identified in the GC-MS assay. The anti-Candida potential of the terpenes, geraniol, and citronellol has been investigated previously, with effective inhibitory activity against C. albicans [18] and filamentous fungi of the Aspergillus species [30]. In addition, Mesa-Arango et al. [31] showed that oxygenated monoterpenes in the citral chemotype, such as geraniol, citral and citronellal, have antifungal activity against C. parapsilosis, C. krusei, Aspergillus flavus, and Aspergillus fumigatus. 2.3. Minimal Inhibitory Concentration and Minimal Fungicidal Concentration of Citronellal The MIC and MFC of citronellal are shown in Table 3. Citronellal showed antifungal activity against C. albicans ATCC, C. krusei (ATCC and clinical strain), and C. glabrata (ATCC and clinical strain). The species C. tropicalis, C. parapsilosis, C. orthopsilosis, and C. albicans clinical strains were resistant to citronellal, with MIC > 1000 μg/mL. Thus, in this present investigation, EO had better antifungal activity compared to citronellal, probably owing to synergisms among the chemical compounds present in the EO. 2.4. Inhibition on Candida albicans Hyphae Growth The results exhibited that the EO was able to inhibit the transition of C. albicans from yeast to the hyphal form. Microscopic observation of EO-treated fungal cells revealed an absence of filamentous cells in concentrations ranging from 1000 to 15 μg/mL (after 12 and 24 h) (Figure 1). Some therapeutic approaches are used to combat C. albicans, including blocking the transformation of yeast cells to filaments. This morphological change is considered to be a virulence factor, and, the biological mechanism has been explored using several active ingredients against this fungal species [32]. The ability of C. albicans to form hyphae is a risk factor in infections because hyphae play an important role in further tissue invasion due to their ability to adhere to host epithelial and endothelial cells [1]. Therefore, the results of this work are promising, since the EO was able to inhibit this morphological transition. Leite et al. [33] showed the action of citral, a mixture of two geometric isomers known as neral and geranial, that are found at high concentrations in the EO used in this study. These authors found that these components were able to inhibit pseudohyphae, chlamydospores, and blastoconidia at the concentration of 128 μg/mL over 48 h. The ability to EO to interfere in hyphal formation was demonstrated for different species of fungi. Chen et al., [8] for example, proved that the oil was able to promote deformities in the hyphal structure of the fungus Alternaria. 2.5. Time-Kill Assay The results showed that the EO inhibited the fungal growth of the different Candida species in a similar manner (Figure 2, Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7). The cell growth was constant until 24 h. After this, an exponential growth—CFU/mL—was noted which remained proportional for a long time. The strains CK-ATCC 6258, CK4, CG-ATCC 2001, GG3, CP1, CO-ATCC 96141, and CO1 presented a superior inhibitory behavior than amphotericin-B over 48 h. The current literature does not present data on time-kill assays evaluating this EO against Candida species. Thus, this study has an innovator character. Ahmad and Viljon [25] observed synergic activity of EOs from the genus Cymbopogon with silver ions against fungi. 2.6. Biofilm Biofilms produced by C. albicans, C. parapsilosis, and C. krusei were treated with EO (10× MIC). The EO showed expressive anti-biofilm activity at different concentrations (Table 4). The percentage of inhibition of biofilms by EO is demonstrated in Figure 8. The EO showed high inhibition of biofilms, mainly against C. albicans ATCC (97.7%), CA3 (82.0%), C. parapsilosis ATCC (93.6%), CP1 (86.2%), C. Krusei ATCC (65.0%), and CK4 (48.5%). The anti-biofilm potential of the EO was the main result of the present study. The elimination and control of fungal biofilm is very hard, due to several types of molecular, structural, and specifically physiological interactions [4]. Additionally, the mechanism of resistance presented by planktonic cells, and the quorum-sensing processes (signaling molecules), the production of specific enzymes and natural mutations can explain the increased resistance of biofilm to antimicrobial agents used in the clinical practice [1]. Biofilm is an important virulence factor and the treatment for its control is concentration dependent and can be high to inhibit the biofilm. The MIC to act against microbial biofilm ranges from 10–100 times higher than that necessary against the planktonic form [34]. The current results show that the EO was able to eliminate mature biofilm of C. albicans and C. parapsilosis at a concentration of 10× MIC. The scientific literature does not present any study of this EO against mature biofilm. Thus, these results are important and contribute to control strategies to eradicate mature biofilm. 2.7. Cytotoxic Evaluation The concentrations of the EO and citronellal that inhibited cell vitality by 50% (IC50) are shown in Table 5. Citronellal demonstrated a higher IC50 than the EO. The EO exhibited inhibitory effects against HepG-2 and MRC-5 with IC50 values of 96.6 μg/mL and 33.1 μg/mL, respectively. On comparing the results, the EO was less cytotoxic to HepG-2 than to MRC-5 cell lines. This can be explained by the metabolizer action of HepG-2. Although lower IC50 values were found, it is important to stress that amphotericin-B, the gold standard in the treatment of fungal diseases, has toxic effects, such as nephrotoxicity, and has an acute reaction after intravenous infusions [5]. In vivo tests must be performed because factors, such as immune response, metabolism, and the pharmacokinetics of the EO, are important. 3. Materials and Methods 3.1. Plant Material The leaves of C. nardus (L.) Rendle were collected in July 2013, in the morning, from the Garden of Toxic and Medicinal Plants: “Profa. Dra. Célia Cebrian de Araújo Reis” (Universidade Estadual Paulista, Araraquara, São Paulo, Brazil). A voucher specimen (HRCB-60752) was deposited in the Rioclarense Herbarium of the Institute of Biosciences (Universidade Estadual Paulista, Rio Claro, São Paulo, Brazil). 3.2. Extraction of the Essential Oil from the Leaves of C. nardus Fresh leaves of C. nardus (150 g) were submitted to hydrodistillation using a Clevenger-type apparatus attached to a round bottom flask (3 L) with 1500 mL of deionized water. The residual water in the EO was separated from the sample by freezing. The yield of the EO was 0, 7% (w/w). The EO was stored under refrigeration until chemical analysis and biological tests. 3.3. Citronellal The commercial (+/−)− citronellal standard (≥95% purity) used was purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). 3.4. Gas Chromatography Analysis of Essential Oil from the Leaves of C. nardus 3.4.1. Gas Chromatography-Mass Spectrometry Gas chromatography-mass spectrometry (GC-MS) analysis was performed using an Agilent® GC-7890B/MSD-5977A gas chromatograph (mass detector: electron impact ionization; mass quadrupole analyzer, (Agilent®, Santa Clara, CA, USA) fitted with a HP-5ms capillary column 5% diphenyl-polydimethylsiloxane (30 m × 0.25 mm, film thickness 0.25 μm—Agilent®). Helium was used as the carrier gas at a flow rate of 1.00 mL/min (8.2 psi) and linear velocity of 36.6 cm/s. Injector temperature: 250 °C; injection volume: 1 μL; splitting ratio: 1:100; oven temperature program: 60–246 °C (3 °C/min, 62 min); transfer line temperature: 280 °C; detector temperature: 150 °C; and ionization energy: 70 eV. EO was solubilized in hexane (chromatographic grade; Tedia®, Fairfield, OH, USA) 1:100 (v/v). The identification of the EO components was based on the comparison of acquired mass spectra (from chromatogram peaks) with reference spectra of the NIST mass-spectral library version 2.0, 2012 (243,893 compounds) and data from the literature. Furthermore, arithmetic retention indices [20] were calculated as described in [21] by linear interpolation relative to the retention times (tR) of a series of n-alkanes (C7–C30); the obtained values were compared with published retention index values [21]. Relative amounts of EO components were calculated based on the chromatogram peak area normalization method. 3.4.2. Gas Chromatography-Flame Ionization Detector Gas chromatography-flame ionization detector (GC-FID) analysis was performed using a Shimadzu® GC-2010 Plus gas chromatograph (flame ionization detector, Shimadzu®, Kyoto, Japan) fitted with a RTX-5MS capillary 5% diphenyl-polydimethylsiloxane column (30 m × 0.25 mm, film thickness 0.25 μm, Restek®, Bellefonte, PA, USA). Nitrogen was used as the carrier gas adjusted to a flow rate of 1.00 mL/min (8.2 psi) and linear velocity of 36.6 cm/s. Injector temperature: 250 °C; injection volume: 1 μL; splitting ratio: 1:30; oven temperature program: 70 °C–180 °C (4 °C/min) and 180 °C–250 °C (10 °C/min); total analysis time: 34.5 min; detector temperature: 280 °C. For the analytical curve, (+/−)− citronellal standard (Sigma-Aldrich®; ≥95% purity) solutions were prepared in hexane (chromatographic grade; Merck®): 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, and 5.0 mg/mL. EO was solubilized in hexane (chromatographic grade; Merck®) 1:100 (v/v). All analyses were performed in triplicate. The identification of citronellal in the EO was based on retention time (tR) and its quantification was achieved according to the external standard method using an analytical curve. 3.5. Antifungal Activity 3.6. Fungal Strains The strains—20 samples of Candida spp.—were obtained from the Laboratory of Microbiology, Department of Infectious Diseases, Medicine School in Sao José do Rio Preto (FAMERP), São Paulo, Brazil. These included three clinical isolates and one ATCC for each species: C. albicans (CA-ATCC 90028, CA2, CA3, CA4); C. krusei (CK-ATCC 6258, CK2, CK3, CK4); C. glabrata (CG-ATCC 2001, CG2, CG3, CG4); C. tropicalis (CT-ATCC 13803, CT2, CT3, CT4), parapsilosis complex—C. parapsilosis (CP-ATCC 22019, CP1), and C. orthopsilosis (CO-ATCC 96141, CO1). C. albicans ATCC 10231 was used to test for inhibition of hyphal growth. The clinical strains were donated to the Microbiology Laboratory of the Medicine School in Sao Jose do Rio Preto for the purposes of scientific research through a written consent of the donors. The use of these strains was approved by the Human Research Ethics Committee of FAMERP, project identification code 152/2006 (6 December2006), Medicine School in Sao José do Rio Preto (FAMERP). 3.7. Determination of Minimum Inhibitory Concentration The evaluation of the antifungal activity by determining the MIC was performed by the microplate dilution technique according to the protocol described in the M27-A3 document [35] with modifications. The concentration of the EO and citronellal was 7.8 to 1000 μg/mL. The EO was dissolved in 10% methanol and 2% Tween 80. A quantity of 0.1 mL was placed in a 96-well microtiter plate containing Roswell Park Memorial Institute (RPMI) 1640 medium. Each well was inoculated with 0.1 mL of a suspension containing 2.5 × 103 CFU/mL of yeast. Amphotericin B (AmB) (Sigma-Aldrich®) and fluconazole (FLU) (Sigma-Aldrich®) were used as the positive controls. Additional controls also included the culture medium, yeast growth, EO, and solvent. The microplates were incubated at 37 °C for 48 h. After incubation, 20 μL of an aqueous 2% solution of 2,3,5-triphenyltetrazolium chloride (TTC) was added, the plates were incubated at 37 °C for 2 h [32], and absorbance of the samples was measured by spectrophotometer (Biospectro, SP22, Curitiba, Brazil). All tests were performed in triplicate. According to obtained results of the EO MIC determination, the more sensitive strains (one ATCC and one clinical strain of each species) were selected to evaluate the antifungal activity of citronellal. 3.8. Determination of Minimum Fungicidal Concentration An aliquot from each well that showed antifungal activity was plated in Petri dishes containing Sabouraud Dextrose Agar (SDA)—DIFCO, to determine the minimum fungicidal concentration (MFC). The assays were carried out in triplicate. MFC was defined as the lowest concentration of the EO and citronellal that allowed no visible growth on the solid medium [32]. 3.9. Inhibition of C. albicans Hyphae Growth A microassay was developed to evaluate the inhibition effect on the growth of fungal strains. Growth of C. albicans (ATCC 10231) from a 48 h culture was transferred to a microplate with RPMI 1640 medium supplemented with fetal bovine serum (FBS) to obtain a final concentration of 2.5 × 103 yeast/mL. EO was added to the growth medium at concentrations ranging from 7.5 to 1000 μg/mL, and the cultures were incubated for 12 and 24 h at 37 °C. The hyphal formation of C. albicans was observed through an inverted light microscope. Amphotericin B (16 μg/mL) was used as a positive control [32]. 3.10. Time-Kill Assay The time-kill assay was performed according to Santos-Filho et al. [36], with modifications. This assay tested one ATCC strain and one clinical strain of each Candida species (CA ATCC 90028, CA3, CK ATCC 6258, CK4, CG ATCC 2001, CG3, CT ATCC 13803, CT3, CP ATCC 22019, CP1, CO ATCC, and CO1). In brief, Sabouraud Dextrose broth (SDB)-DIFCO, containing 2.5 × 103 CFU/mL of Candida spp. and 2× MIC of EO were incubated at 37 °C and aliquots of 100 μL were removed at different time intervals (0, 1, 2, 4, 8, 12, 24, 36 and 48 h). The aliquots were then diluted in a buffer solution of sterile PBS 1:100, twice. Each EO-cell suspension was spread onto Sabouraud plates and colonies were counted after 48 h incubation at 37 °C. Amphotericin B was used as a positive control. Negative controls were established with cell suspensions without the addition of EO. 3.11. Biofilm Assay The biofilm adhesion method was performed as described by Pitangui et al. [37], with modifications. The CA ATCC 90028, CA3, CK ATCC 6258, CK4, CP ATCC 22019, and CP1 strains were selected for the biofilm assay. Initially, 100 μL of inoculum (5.0 × 108 cells/mL), suspended in 0.9% saline solution, was added to the wells of microplates (96 wells) and incubated in a shaker at 80 rpm at 37 °C for 2 h. After the pre-adhesion period, the supernatant was removed and 100 μL of RPMI medium was added to each microplate well. Incubation continued at 37 °C for 48 h with the RPMI renewed after 24 h. The supernatant was then removed, and the wells were washed with 100 μL of 0.9% saline solution. Next, 100 μL of EO (10× MIC) were added to each microplate well. The microplates were incubated again for 24 h at 37 °C. Subsequently, the EO was removed and the wells were washed with sterile saline solution (to eliminate the drug carryover effect). Solvent, medium culture, and yeast growth were established as controls with the colorimetric indicator 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[carbonyl(phenylamino)]-2H-tetrazolium hydroxide (XTT®, Sigma-Aldrich®, Saint Louis, MO, USA). 3.12. Cytotoxic Activity 3.12.1. Cell Lines HepG-2 (hepatic) (ATCC®HB-8065™, Fiocruz, Rio de Janeiro, Brazil) and MRC-5 (fibroblast) (ATCC® CCl-171™, Fiocruz, Rio de Janeiro, Brazil) were used to determine cytotoxicity (IC50). The cells were maintained in flasks with a 12.50 cm2 surface area containing 10 mL of culture medium incubated at 37 °C in 5% carbon dioxide. The culture medium consisted of Dulbecco’s Modified Eagle Medium (DMEM, Vitrocell®, Campinas, São Paulo, Brazil) supplemented with 10% FBS, gentamicin sulfate (50 mg/L, Sigma-Aldrich®), and amphotericin B (2 mg/L, Sigma-Aldrich®). 3.12.2. Cytotoxic Assay The cytotoxic assay [38] consisted of collecting the cells using a solution of trypsin/ethylenediaminetetraacetic acid (EDTA, Vitrocell®), centrifuging the solution (2000 rpm for 5 min) and counting the number of cells in a Neubauer chamber followed by adjustment of the cell concentration to 7.5 × 104 cells/mL in DMEM. Then, 200 μL of this suspension was placed in each well of a 96-well microplate to obtain a concentration of 1.5 × 104 cells/well, and the microplates were then incubated at 37 °C in 5% carbon dioxide for 24 h to facilitate cell attachment to the plate. The serial dilutions of EO were prepared to obtain concentrations from 3.90 to 1000 μg/mL. These dilutions were added to the cells after the removal of the medium and the non-adherent cells. Then, the cells were incubated for an additional 24 h. The cytotoxicity of the compounds was determined by adding 30 μL of resazurin and reading on a microplate reader (BioTek®, Winoosky, VT, USA) after 6 h of incubation using a microplate and excitation emission filters with wavelengths of 530 and 590 nm, respectively. The IC50 was defined as the highest concentration of compound that allowed a viability of at least 50% of the cells. All experiments were performed in triplicate. A solution of 5% dimethyl sulfoxide (DMSO) was used as the control. 4. Conclusions According to the results of this study, it is possible to conclude that the EO from C. nardus is a promising source of active molecules with antifungal properties. The biological assays reported in this investigation show that the EO inhibits ATCC and clinical strains of Candida species, including those with resistance to drugs employed in medical practice. Additional to this simple inhibitory activity, the EO is able to inhibit and control the main virulence factors attributed to the Candida species used in this study, such as the formation and proliferation of hyphae of C. albicans and, more importantly, the eradication of mature biofilms. Moreover, the EO exhibits better antifungal action than citronellal, probably due to some synergistic effect among the EO components. Acknowledgments We thank grant#2013/19576-0 and grant#2016/16436-0, São Paulo Research Foundation (FAPESP) and “Programa de Apoio ao Desenvolvimento Científico (PADC)” of the School of Pharmaceutical Sciences of São Paulo State University, UNESP—Araraquara, São Paulo, Brazil by financial support. We thank Luciomar Gaspar de Toledo by the edition of the figures. Author Contributions Luciani Gaspar De Toledo, Matheus Aparecido dos Santos Ramos and Larissa Spósito designed all experiments of this work and the data analyses. Guilherme Julião Zocolo, Francisca Aliny Nunes Silva and Tigressa Helena Soares designed the chemical analysis. Fernando Rogério Pavan and Érica De Oliveira Lopes designed the cytotoxicity tests. André Gonzaga dos Santos supervised the obtainment of leaves from C. nardus and the extraction of essential oil. Margarete Teresa Gottardo De Almeida, Elza Maria Castilho, Taís Maria Bauab, André Gonzaga dos Santos and Guilherme Julião Zocolo supervised the design and data interpretation. The manuscript was written by Luciani Gaspar De Toledo, Matheus Aparecido dos Santos Ramos and Larissa Spósito and edited by Margarete Teresa Gottardo De Almeida. All authors discussed the results and commented on the manuscript. Conflicts of Interest The authors report no conflicts of interest in this work. Figure 1 Inhibitory effect of essential oil of C. nardus on the transition of C. albicans from yeast to the hyphal form (photomicrographs by inverted light microscopic under 400× magnification). Figure 2 Time-kill curves of C. albicans ATCC 90028 and CA3 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 3 Time-kill curves of C. krusei ATCC 6258 and CK4 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 4 Time-kill curves of C. glabrata ATCC 2001 and CG3 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 5 Time-kill curves of C. tropicalis ATCC 13803 and CT3 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 6 Time-kill curves of C. parapsilosis ATCC 22019 and CP1 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 7 Time-kill curves of C. orthopsilosis ATCC 96141 and CO1 following exposure to the essential oil of C. nardus and amphotericin-B. Control represents the untreated Candida cells. Figure 8 Percentage of inhibition of the essential oil of C. nardus against biofilms of Candida species. ijms-17-01252-t001_Table 1Table 1 Composition of essential oil from the leaves of C. nardus. Retention Time Compound Name AI 1 (Calculated) AI 2 (Literature) Concentration (%) 6.60 not identified – – 0.09 6.74 β-myrcene 992 988 0.09 7.08 n-octanal 1004 998 0.09 7.92 D-limonene 1029 1024 2.47 8.21 cis-ocimene 1037 1032 0.27 8.56 trans-ocimene 1048 1044 0.17 8.75 bergamal 1053 1051 0.37 9.40 not identified – – 0.17 10.40 linalool 1101 1095 0.53 10.57 α-pinene oxide 1105 1099 0.11 11.51 trans-rose oxide 1129 1122 0.14 12.16 neo-isopulegol 1145 1144 0.41 12.34 not identified – – 0.27 12.54 citronellal 1155 1148 27.87 12.95 not identified – – 0.25 13.69 not identified – – 0.33 14.16 cis-4-decenal 1195 1193 0.09 14.63 Decanal 1207 1201 0.46 15.60 β-citronellol 1230 1223 11.85 16.12 Neral 1242 1235 11.21 16.73 geraniol 1257 1249 22.77 17.38 geranial 1273 1264 14.54 20.82 citronellol acetate 1355 1350 0.31 22.08 geranyl acetate 1385 1379 0.26 23.45 β-cariophyllene 1419 1417 1.28 24.81 α-humulene 1453 1452 0.12 27.25 γ-cadinene 1514 1513 1.60 27.63 δ-cadinene 1524 1522 0.36 27.87 citronellyl butyrate 1530 1530 0.24 28.60 elemol 1550 1548 0.11 29.11 not identified – – 0.16 29.85 cariophyllene oxide 1582 1582 0.55 32.08 trans-cadinol 1642 1638 0.16 32.54 α-muurolol 1654 1644 0.30 Monoterpene hydrocarbons 3.00 Oxygen containing monoterpenes 90.61 Sesquiterpene hydrocarbons 3.36 Oxygen containing sesquiterpenes 1.12 Other compounds 0.64 Total identified 98.73 1 Arithmetic retention indices [20] relative to C7-C30 n-alkanes calculated [21]; 2 Arithmetic retention indices [20,21]. ijms-17-01252-t002_Table 2Table 2 Inimal inhibitory concentrations (MIC—μg/mL) and minimal fungicidal concentration (MFC—μg/mL) of the essential oil of C. nardus against Candida species. Candida Strains MIC EO * MFC EO * MIC AmB * MIC FLU * CA-ATCC 90028 1000 1000 1 1 CA2 1000 1000 4 16 CA3 1000 1000 1 >64 CA4 1000 1000 4 8 CK-ATCC 6258 250 500 8 >64 CK2 500 500 8 >64 CK3 500 500 8 >64 CK4 250 250 4 >64 CG-ATCC 2001 500 1000 1 >64 CG2 500 1000 4 >64 CG3 500 1000 2 >64 CG4 1000 1000 2 >64 CT-ATCC 13803 500 1000 8 >64 CT2 >1000 >1000 8 >64 CT3 1000 >1000 4 >64 CT4 >1000 >1000 4 >64 CP-ATCC 22019 500 1000 4 8 CP1 1000 1000 4 32 CO-ATCC 96141 500 1000 8 32 CO1 1000 1000 8 64 FLU: fluconazole; AmB: Amphotericin-B; * values in μg/mL. ijms-17-01252-t003_Table 3Table 3 Minimal inhibitory concentrations (MIC, μg/mL) and minimal fungicidal concentration (MFC, μg/mL) of citronellal against Candida species. Strains MIC Citronellal MFC Citronellal CA-ATCC 90028 1000 1000 CA3 >1000 >1000 CK-ATCC 6258 500 1000 CK4 500 500 CG-ATCC 2001 500 1000 CG3 500 >1000 CT-ATCC 13803 >1000 >1000 CT3 >1000 >1000 CP-ATCC 22019 >1000 >1000 CP1 >1000 >1000 ijms-17-01252-t004_Table 4Table 4 Anti-biofilm effect of the essential oil of C. nardus against C. albicans, C. krusei, and C. parapsilosis. Strains EO (mg/mL) CA-ATCC 90028 2.5 CA3 5 CK-ATCC 6258 2.5 CK4 2.5 CP-ATCC 22019 5 CP1 10 ijms-17-01252-t005_Table 5Table 5 Cytotoxic activity of the essential oil of C. nardus (EO) and citronellal. Cell Lines (EO) IC50 * Citronellal IC50 * (Control) IC50 a HepG-2 96.6 100.9 >1000 MRC-5 33.1 51 >1000 * Values in μg/mL; a Dimethyl sulfoxide. ==== Refs References 1. Mayer F.L. Wilson D. Hube B. Candida albicans pathogenicity mechanisms Virulence 2013 4 119 128 10.4161/viru.22913 23302789 2. Deorukhkar S.C. Saini S. Mathew S. Virulence factors contributing to pathogenicity of Candida tropicalis and its antifungal susceptibility profile Int. J. Microbiol. 2014 2014 1 6 10.1155/2014/456878 24803934 3. Silva S. Negri M. Henriques M. Oliveira R. Williams D.W. Azeredo J. Candida glabrata , Candida parapsilosis and Candida tropicalis : Biology, epidemiology, pathogenicity and antifungal resistance FEMS Microbiol. Rev. 2012 36 288 305 10.1111/j.1574-6976.2011.00278.x 21569057 4. Sardi J.D.C.O. Pitangui N.D.S. Rodríguez-Arellanes G. Taylor M.L. Fusco-Almeida A.M. Mendes-Giannini M.J.S. Highlights in pathogenic fungal biofilms Rev. Iberoam. Micol. 2014 31 22 29 10.1016/j.riam.2013.09.014 24252828 5. Calabrese E.C. Castellano S. Santoriello M. Sgherri C. Quartacci M.F. Calucci L. Warrilow A.G.S. Lamb D.C. Kelly S.L. Milite C. Antifungal activity of azole compounds CPA18 and CPA109 against azole-susceptible and -resistant strains of Candida albicans J. Antimicrob. Chemother. 2013 68 1111 1119 10.1093/jac/dks506 23292344 6. Costa G.M. Endo E.H. Cortez D.A.G. Ueda-nakamura T. Nakamura C.V. Filho B.P.D. Filho D. Effect of plant extracts on planktonic growth and biofilm of Staphylococcus aureus and Candida albicans Int. J. Curr. Microbiol. Appl. Sci. 2015 4 908 917 7. Piccinelli A.C. de Santana Aquino D.F. Morato P.N. Kuraoka-Oliveira Â.M. Strapasson R.L.B. dos Santos É.P. Stefanello M.É.A. Oliveira R.J. Kassuya C.A.L. Anti-inflammatory and antihyperalgesic activities of ethanolic extract and fruticulin a from salvia lachnostachys leaves in mice Evid. Based Complement. Altern. Med. 2014 2014 1 8 10.1155/2014/835914 25435893 8. Chen Q. Xu S. Wu T. Guo J. Sha S. Zheng X. Yu T. Effect of citronella essential oil on the inhibition of postharvest Alternaria alternata in cherry tomato J. Sci. Food Agric. 2014 94 2441 2447 10.1002/jsfa.6576 24425449 9. Chanthai S. Prachakoll S. Ruangviriyachai C. Luthria D.L. Influence of extraction methodologies on the analysis of five major volatile aromatic compounds of citronella grass Cymbopogon nardus and lemongrass (Cymbopogon citratus ) grown in Thailand J. AOAC Int. 2012 95 763 772 10.5740/jaoacint.11-335 22816268 10. Aini M.N.N. Said M.I. Nazlina I. Hanina M.N. Ahmad I.B. Screening for Antiviral activity of sweet lemon grass (Cymbopogon nardus (L.) rendle) fractions J. Biol. Sci. 2006 6 507 510 11. Innsan M.F.M.F. Shahril M.H. Samihah M.S. Asma O.S. Radzi S.M. Jalil A.K.A. Hanina M.N. Pharmacodynamic properties of essential oils from Cymbopogon species Afr. J. Pharm. Pharmacol. 2011 5 10.5897/AJPP11.693 12. Nakahara K. Alzoreky N.S. Yoshihashi T. Nguyen H.T.T. Trakoontivakorn G. Chemical composition and antifungal activity of essential oil from Cymbopogon nardus (Citronella Grass) Jpn. Agric. Res. Q. 2013 37 249 252 10.6090/jarq.37.249 13. Calo J.R. Crandall P.G. O’Bryan C.A. Ricke S.C. Essential oils as antimicrobials in food systems—A review Food Control 2015 54 111 119 10.1016/j.foodcont.2014.12.040 14. Bonifácio B.V. dos Santos Ramos M.A. da Silva P.B. Bauab T.M. Antimicrobial activity of natural products against Helicobacter pylori: A review Ann. Clin. Microbiol. Antimicrob. 2014 13 54 10.1186/s12941-014-0054-0 25406585 15. Freires I.D.A. Murata R.M. Furletti V.F. Sartoratto A. de Alencar S.M. Figueira G.M. de Oliveira Rodrigues J.A. Duarte M.C.T. Rosalen P.L. Coriandrum sativum L. (Coriander) essential oil: Antifungal activity and mode of action on Candida spp., and molecular targets affected in human whole-genome expression PLoS ONE 2014 9 1252 10.1371/journal.pone.0099086 24901768 16. Husain F.M. Ahmad I. Khan M.S. Al-Shabib N.A. Trigonella foenum-graceum (Seed) extract interferes with quorum sensing regulated traits and biofilm formation in the strains of Pseudomonas aeruginosa and Aeromonas hydrophila Evid. Based Complement. Altern. Med. 2015 2015 1 10 10.1155/2015/879540 26000026 17. Tolba H. Moghrani H. Benelmouffok A. Kellou D. Maachi R. Essential oil of algerian eucalyptus citriodora: Chemical composition, antifungal activity J. Med. Mycol. 2015 25 e128 e133 10.1016/j.mycmed.2015.10.009 26597375 18. Zore G.B. Thakre A.D. Jadhav S. Karuppayil S.M. Terpenoids inhibit Candida albicans growth by affecting membrane integrity and arrest of cell cycle Phytomedicine 2011 18 1181 1190 10.1016/j.phymed.2011.03.008 21596542 19. Dewick P.M. Medicinal Natural Products: A Biosynthetic Approach 3rd ed. John Wiley & Sons Chichester, UK 2009 20. Van Den Dool H. Kratz P.D. A generalization of the retention index system including linear temperature programmed gas–liquid partition chromatography J. Chromatogr. A 1963 11 463 471 10.1016/S0021-9673(01)80947-X 21. Adams R.P. Identification of Essential Oil Components by Gas Chromatography/Mass Spectrometry 4th ed. Allured Publishing Corporation Carol Stream, IL, USA 2007 22. Wei L.S. Wee W. Chemical composition and antimicrobial activity of Cymbopogon nardus citronella essential oil against systemic bacteria of aquatic animals Iran. J. Microbiol. 2013 5 147 152 23825733 23. Koba K. Sanda K. Guyon C. Raynaud C. Chaumont J.-P. Nicod L. In vitro cytotoxic activity of Cymbopogon citratus L. and Cymbopogon nardus L. essential oils from Togo Bangladesh J. Pharmacol. 2009 4 29 34 10.3329/bjp.v4i1.1040 24. Trindade L.A. de Araújo Oliveira J. de Castro R.D. de Oliveira Lima E. Inhibition of adherence of C. albicans to dental implants and cover screws by Cymbopogon nardus essential oil and citronellal Clin. Oral Investig. 2015 19 2223 2231 10.1007/s00784-015-1450-3 25804885 25. Ahmad A. Viljoen A. The in vitro antimicrobial activity of Cymbopogon essential oil (lemon grass) and its interaction with silver ions Phytomedicine 2015 22 657 665 10.1016/j.phymed.2015.04.002 26055131 26. Webster D. Taschereau P. Belland R.J. Sand C. Rennie R.P. Antifungal activity of medicinal plant extracts; preliminary screening studies J. Ethnopharmacol. 2008 115 140 146 10.1016/j.jep.2007.09.014 17996411 27. Li H. Zhang C. Chen Z. Shi W. Sun S. A promising approach of overcoming the intrinsic resistance of Candida krusei to fluconazole (FLC)—Combining tacrolimus with FLC FEMS Yeast Res. 2014 14 808 811 10.1111/1567-1364.12163 24830342 28. Dos Santos Ramos M.A. Calixto G.M.F. de Toledo L.G. Bonifácio B.V. dos Santos L.C. de Almeida M.T.G. Chorilli M. Bauab T.M. Liquid crystal precursor mucoadhesive system as a strategy to improve the prophylactic action of Syngonanthus nitens (Bong.) Ruhland against infection by Candida krusei Int. J. Nanomed. 2015 10 7455 10.2147/IJN.S92638 26719688 29. Blank A.F. Costa A.G. Arrigoni-Blank M.D.F. Cavalcanti S.C.H. Alves P.B. Innecco R. Ehlert P.A.D. de Sousa I.F. Influence of season, harvest time and drying on Java citronella (Cymbopogon winterianus Jowitt) volatile oil Rev. Bras. Farmacogn. 2007 17 557 564 10.1590/S0102-695X2007000400014 30. Mileva M. Krumova E. Miteva-Staleva J. Kostadinova N. Dobreva A. Galabov A.S. Chemical compounds, in vitro antioxidant and antifungal activities of some plant essential oils belonging to rosaceae family Proc. Bulg. Acad. Sci. 2014 67 1363 1368 31. Mesa-Arango A.C. Montiel-Ramos J. Zapata B. Durán C. Betancur-Galvis L. Stashenko E. Citral and carvone chemotypes from the essential oils of Colombian Lippia alba (Mill.) N.E. Brown: Composition, cytotoxicity and antifungal activity Mem. Inst. Oswaldo Cruz 2009 104 878 884 10.1590/S0074-02762009000600010 19876560 32. De Freitas Araújo M.G. Pacífico M. Vilegas W. Dos Santos L.C. Icely P.A. Miró M.S. Scarpa M.V.C. Bauab T.M. Sotomayor C.E. Evaluation of Syngonanthus nitens (Bong.) Ruhl. extract as antifungal and in treatment of vulvovaginal candidiasis Med. Mycol. 2013 51 673 682 10.3109/13693786.2013.795294 23758104 33. Leite M.C.A. de Brito Bezerra A.P. de Sousa J.P. Guerra F.Q.S. de Oliveira Lima E. Evaluation of Antifungal activity and mechanism of action of citral against Candida albicans Evid. Based Complement. Altern. Med. 2014 2014 1 9 10.1155/2014/378280 25250053 34. DiCicco M. Neethirajan S. Singh A. Weese J. Efficacy of clarithromycin on biofilm formation of methicillin-resistant Staphylococcus pseudintermedius BMC Vet. Res. 2012 8 1252 10.1186/1746-6148-8-225 23171620 35. Clinical and Laboratory Standards Institute Reference Methods for Broth Dilution Antifungal Susceptibility Tests for Yeasts: Approved Standards Clinical and Laboratory Standards Institute Wayne, IL, USA 2008 36. Santos-Filho N.A. Lorenzon E.N. Ramos M.A.S. Santos C.T. Piccoli J.P. Bauab T.M. Fusco-Almeida A.M. Cilli E.M. Synthesis and characterization of an antibacterial and non-toxic dimeric peptide derived from the C-terminal region of Bothropstoxin-I Toxicon 2015 103 160 168 10.1016/j.toxicon.2015.07.004 26160494 37. Pitangui N.S. Sardi J.C.O. Silva J.F. Benaducci T. da Silva R.A.M. Rodríguez-Arellanes G. Taylor M.L. Mendes-Giannini M.J.S. Fusco-Almeida A.M. Adhesion of Histoplasma capsulatum to pneumocytes and biofilm formation on an abiotic surface Biofouling 2012 28 711 718 10.1080/08927014.2012.703659 22784100 38. O’Brien J. Wilson I. Orton T. Pognan F. Investigation of the Alamar Blue (resazurin) fluorescent dye for the assessment of mammalian cell cytotoxicity Eur. J. Biochem. 2000 267 5421 5426 10.1046/j.1432-1327.2000.01606.x 10951200

PMC005xxxxxx/PMC5000651.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081253ijms-17-01253ArticleElectrochemical Determination of Food Preservative Nitrite with Gold Nanoparticles/p-Aminothiophenol-Modified Gold Electrode Üzer Ayşem 1Sağlam Şener 1Can Ziya 1Erçağ Erol 1Apak Reşat 12*Arráez-Román David Academic Editor1 Analytical Chemistry Division, Chemistry Department, Faculty of Engineering, Istanbul University, Avcilar, 34320 Istanbul, Turkey; [email protected] (A.Ü.); [email protected] (Ş.S.); [email protected] (Z.C.); [email protected] (E.E.)2 Turkish Academy of Sciences (TUBA) Piyade st. No: 27, 06690 Çankaya Ankara, Turkey* Correspondence: [email protected]; Tel.: +90-212-473-702802 8 2016 8 2016 17 8 125317 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Due to the negative impact of nitrate and nitrite on human health, their presence exceeding acceptable levels is not desired in foodstuffs. Thus, nitrite determination at low concentrations is a major challenge in electroanalytical chemistry, which can be achieved by fast, cheap, and safe electrochemical sensors. In this work, the working electrode (Au) was functionalized with p-aminothiophenol (p-ATP) and modified with gold nanoparticles (Au-NPs) to manufacture the final (Au/p-ATP-Aunano) electrode in a two-step procedure. In the first step, p-ATP was electropolymerized on the electrode surface to obtain a polyaminothiophenol (PATP) coating. In the second step, Au/p-ATP-Aunano working electrode was prepared by coating the surface with the use of HAuCl4 solution and cyclic voltammetry. Determination of aqueous nitrite samples was performed with the proposed electrode (Au/p-ATP-Aunano) using square wave voltammetry (SWV) in pH 4 buffer medium. Characteristic peak potential of nitrite samples was 0.76 V, and linear calibration curves of current intensity versus concentration was linear in the range of 0.5–50 mg·L−1 nitrite with a limit of detection (LOD) of 0.12 mg·L−1. Alternatively, nitrite in sausage samples could be colorimetrically determined with high sensitivity by means of p-ATP‒modified gold nanoparticles (AuNPs) and naphthylethylene diamine as coupling agents for azo-dye formation due to enhanced charge-transfer interactions with the AuNPs surface. The slopes of the calibration lines in pure NO2− solution and in sausage sample solution, to which different concentrations of NO2− standards were added, were not significantly different from each other, confirming the robustness and interference tolerance of the method. The proposed voltammetric sensing method was validated against the colorimetric nanosensing method in sausage samples. nitrite determinationgold electrodemodified electrodegold nanoparticlessquare wave voltammetry (SWV) ==== Body 1. Introduction Nitrite is an approved additive to meat products as an antimicrobial flavoring and coloring agent, in addition to its ability to retard lipid peroxidation [1]. Another interesting property of nitrite is its capability to retard the development of rancidity during storage, to reduce the thiobarbituric acid (TBA) response (used as a measure of biooxidative status), and the subsequent warmed-over flavors (WOF) developed upon heating of meat [2,3]. Unfortunately, nitrite is an essential precursor in the formation of nitrosamines as a cancer-suspect class of compounds [4]. In the last twenty years, many procedures have been designed to enable sensitive and selective estimation and monitoring of nitrite and nitrate [5], thereby focusing interest on related quantification techniques, such as chromatography [6], spectroscopy [7], and capillary electrophoresis [8]. The first, but still effective, spectrophotometric determination of NO2− involved the Griess method, which is more than a century old [9]. Spectrofluorimetry is also a sensitive technique for trace analysis, based on diazotization of nitrite with different reagents [10,11,12]. It should be added that these methods have certain drawbacks, such as toxicity of reagents, tedious extraction procedures, and inherent interference. Electrochemical methods are often favored over others due to their simplicity, rapidity, measurement precision in turbid solutions, and low cost [13,14]. In principle, electrochemical determination of nitrite may involve both oxidation and reduction, but in actual practice, oxidation is usually preferred over reduction because cathodic nitrite measurement is prone to interference from other readily reducible species such as nitrate ion and molecular oxygen [15]. Nitrite is electroactive at traditional electrodes made up of glassy carbon (GC), platinum, diamond, gold, copper, and transition metal oxides [14,16,17,18,19,20,21]. The electrochemical oxidation of nitrite on bare electrodes may be adversely affected by several species (such as nitrite oxidation products and intermediates) which can inhibit electrode processes via irreversible adsorption on the GCE surface and, thereby, decrease both sensitivity and accuracy [18,19,22,23]. In addition, the direct electro-reduction/oxidation of nitrite ions requires high overpotentials at bare electrode surfaces. As a remedy to overcome these limitations, various modified electrodes have been exploited for nitrite determination [24,25,26]. For the purpose of sensing nitrite, electrodes have been chemically modified with facile electron-transfer materials, such as metallophthalocyanines and metalloporphyrins [14,22,27,28,29,30,31,32], inorganic porous substances [33], and enzyme-based electrochemical transducers [34]. In recent years, an increasing variety of nano-materials has been employed for electrochemical studies, combining the advantages of electrochemistry and nanotechnology [35,36,37,38]. Generally compared to a bulk gold electrode, AuNPs-polymer nanocomposites forming an electrocatalytic layer on the surface of a modified electrode are expected to decrease the overpotential and enable faster electron-transfer kinetics for nitrite oxidation [39]. The conducting polymers used for electrode modification mainly comprise polyaniline, polypyrrole, and polythiophene. 4-aminothiophenol (p-ATP) is a focusing target in the manufacture of 2D or 3D nanoparticle assemblies by exploiting either covalent or electrostatic (ionic) interactions [40]. Since thiol and amine ends of p-ATP have different reactivities [41,42], the effective use of this molecular assembly may give rise to unique morphologies leading to multi-purpose chemical strategies. Moreover, the aromatic (conjugated π-electron system) ring of p-ATP intensifies the electrical coupling between NPs and the electrode. This work reports the fabrication, characterization, and analytical performance of a nitrite sensor involving the incorporation of AuNPs into a conductive polymer matrix, poly(4-aminothiophenol) (PATP), over the surface of a gold electrode. The proposed sensing procedure was tested for the tolerance and elimination of various interferences, and applied to nitrite determination in a real food sample, such as sausage. The nitrite findings obtained with this voltammetric sensor were compared to those found with an AuNPs-based colorimetric sensor also prepared in our laboratory. 2. Results 2.1. Fabrication of 4-ATP Polymer Film on an Au Electrode Electrochemical polymerization of an Au working electrode was achieved in two steps; the –SH groups of 4-ATP were arranged (via self-assembly) on the Au electrode surface, followed by electrochemical polymerization of the p-ATP monolayer from the –NH group end with the aid of the cyclic voltammetry (CV) technique [43]. The Au working electrode surface was coated with 10 mM p-ATP via CV within the potential range of 0–1.7 V at 20 mV·s−1 scanning speed for 20 cycles (Figure 1). As shown in Figure 1, the deposition amount of p-ATP on the Au electrode surface increased and, consequently, the current intensities of p-ATP (in solution) decreased with an increasing number of cycles (the optimal value of which was set at 20 cycles). As an alternative, this target could be achieved by immersing the working electrode in p-ATP solution (physical binding) for 12 h at room temperature. However, if the working electrode is coated electrochemically, it can be used repeatedly. A signal attenuation of about 2% was observed using the working electrode for 15 days, while 80% of the original signal remained after 30 days. At the second step, the Au working electrode surface was immersed in 5 mL of 0.5 M HClO4 solution and polymerized via cyclic voltammetry within the potential range of 0–0.8 V at 50 mV·s−1 scanning speed for 50 cycles (Figure 2). As a result, the Au working electrode surface was coated with polyaminothiophenol (PATP). The peak currents centering around 0.7 V were relatively stable for up to 50 voltammetric scans, beyond which a decreasing trend was observed in current. Since the peaks were recoverable in a fresh solution of perchloric acid and similar peaks were not observable with bare gold electrodes, the 0.7 V peak could be attributed to the presence of an electroactive monolayer of poly(4-aminothiophenol) on gold electrode. As the same electroactive film could not be formed in a solution of ATP and ATP→PATP conversion required the preliminary formation of a self-assembled layer on the flat surface of Au, the mentioned peaks were assumed to arise from the formation of polaronic and bipolaronic structures of PATP [44]. 2.2. Electrodeposition of Au-Nanoparticles on a PATP-Coated Au Electrode The peak manifesting itself at ≈−0.2 V in the first cycle (Figure 3) may be attributed to the reduction of gold nanoparticles on the polyaminothiophenol−coated electrode surface. As the number of cycles was increased, increasing amounts of metallic gold accumulated on the electrode, leaving a smaller number of free Au-NPs (and also diffused Au(III) ions in close proximity to the electrode), leading to a decrease in current intensities in the cathodic range. The current intensity tended to stabilize after 40 cycles, hinting to a saturation of Au-NPs accumulation on the electrode surface. After each regeneration step, material balance regarding gold accumulation on the electrode surface should be made by consideration of the values pertaining to initial and final cycles. An important advantage of this electrode design was time/labor saving by refraining from electrode recoating and cleansing prior to each measurement, as the once prepared Au/PATP-Aunano electrode could be used successively without tedious operations throughout the day. 2.3. CV Characterization of the Modified Electrodes The modified electrodes were characterized by a combination of CV scans, impedance spectral measurements, and scanning electron microscope (SEM) images. For the purpose of electrode characterization, CV was applied in monomer-free medium to bare Au, Au/PATP, and Au/PATP-Aunano electrodes with a scanning speed of 100 mV/s within the potential range of 0–0.8 V. As can be seen from Figure 4, anodic and cathodic currents could be obtained from both Au/PATP, and Au/PATP-Aunano electrodes, showing that both electrodes have electroactivity and that Aunano coating of the copolymer electrode did not cause a decrease in this activity. For the Au/PATP electrode, anodic and cathodic peaks emerged at 517.9 mV and 214.1 mV, respectively, whereas for the Au/PATP-Aunano electrode, the corresponding peaks shifted to slightly smaller potentials, appearing at 502.6 mV and 225.6 mV, respectively. These lower potential shifts and intensification of currents may be attributed to the increase in effective surface area of the modified electrode [23]. The bare Au electrode did not yield a noticeable peak under identical conditions. 2.4. SEM Images of an Au/PATP-Aunano Electrode The SEM image of the Au/PATP-Aunano electrode was recorded with the aid of a scanning electron microscopy (SEM; FEI Model Quanta 450 FEG, Hillsboro, OR, USA). The SEM images of the produced Au-NPs were shown in Figure 5. Au-NPs were homogeneously distributed on the p-ATP/Au surface with average diameters of about 75 nm; but in addition to well-distributed particles, large aggregated irregular particles with a size of about 300 nm were also observed. Hypothetically, Au-S covalent binding can contribute to Au-NPs immobilization on the electrode surface. This nanostructured film could significantly enhance the active surface area of Au electrode and might be very important to promote electron transfer. Consequently, the resulting modified electrode exhibited a good electrocatalytic capability towards the oxidation of nitrite. 2.5. Electrochemical Impedance Spectroscopy (EIS) Applied to Modified Electrodes Impedance measurements were made for the Au/PATP and Au/PATP-Aunano electrodes in monomer-free solution media via the potentiostat EIS method; the frequency range was 10 mHz–1 MHz, and points/decade 10 mV. Figure 6 shows the impedance spectra of Au/PATP- and Au/PATP-Aunano-modified electrodes. The high frequency region on the left hand side (beginning) of the spectrum identifies the electrolyte properties, while the mid-frequency region corresponds to the electrode/electrolyte interface process. The relaxation effect is represented by a semicircle whose intersections with the real axis are the electrolyte and the charge transfer resistances. Within the region of low frequency, impedance was controlled by counter-ion diffusion inside the electrode; the impedance response, a 45° straight line (Warburg impedance) [45]. The radius of the semicircle for Au/PATP was observed to be smaller than the corresponding value for the Au/PATP-Aunano electrode, indicating a lower charge-transfer resistance (RCT) and a higher electroactivity of the former. This may have resulted from the higher thickness of Au/PATP-Aunano electrode. It was reported in other sources that impedance curves depicted on the same Z’/Z” graph enabled the comparison of the electron-transfer capabilities of different (bare and modified) electrodes [46], where the monolayer coated on the modified electrode possibly showed a higher obstruction to electron transfer than the bare electrode [47]. In order to gain preliminary information about the charge storage capacity of the electrodes, lower frequency capacitance values (Csp) were calculated as suggested in the literature [48] and found as 1.81 × 10−4 and 1.68 × 10−4 farad/cm2 for Au/PATP and Au/PATP-Aunano, respectively, hinting to the better suitability of Au/PATP electrodes for capacitor application. It should be remembered that while impedance spectral curves enable a comparison of RCT values with slightly increased values of RCT upon Au-NPs modification of the electrode contrary to expectation, the advantages of the Au/PATP-Aunano electrode over Au/PATP have been demonstrated in this work as stability, sensitivity, reproducibility, and repeated use without a need for coating before each application. While studying the electrochemical characteristics of gold electrodes functionalized by carboxyl-terminated alkane thiol monolayers, Bradbury et al. [49] saw that the charge-transfer resistance of the ferri-/ferro-cyanide redox pair at equilibrium potential showed an exponential increase with increasing –CH2– units in the monolayer. These authors further argued that adsorption of the citrate-stabilized Au nanoparticles generated a local concentration polarization of the redox species at the interface, leading to an increase of RCT. Their preliminary investigations had shown that the apparent charge-transfer resistance in the presence of the Au nanoparticles (Rarray) was strongly dependent on the particle number density [49]. 2.6. Square Wave Voltammetric Response of the Recommended Sensor Electrode to NO2− Nitrite calibration was done with square wave voltammetry and the characteristic peak potential of NO2− appeared at 0.76 V. Square wave voltammograms of NO2− recorded within the range of 0.5 and 50 mg·L−1 (final concentrations) were given in Figure 7. Calibration of nitrite at 0.76 V potential gave a linear dependence of current intensity versus concentration: I0.76 V = 0.234 CNO2− + 2.537 (r = 0.999)(1) where I0.76 V is the peak current intensity (µA) at 0.76 V and CNO2− is the NO2− concentration (mg·L−1). The calibration curve for the analr Nyte is established as response (y) versus concentration (x). The minimal analyte response that can be detected (yLOD) corresponding to the concentration at the limit of detection (xLOD) is equal to the mean value of blank responses (ӯbl) exceeded by (ksbl), where sbl is the standard deviation of blank responses and k is a factor equal to 3 (by International Union of Pure and Applied Chemistry (IUPAC) recommendation) such that yLOD = ӯbl + ksbl = ӯbl + 3sbl. Thus, the calibration curve is intersected at yLOD response value, a vertical line is drawn from this intersection up to the horizontal axis, and the corresponding concentration (xLOD) is directly read on the intersection point of this perpendicular line with the horizontal axis [50]. In this work, LOD was assessed by calculating the peak current intensity at three standard deviations above the mean current intensity after n = 10 repetitive measurements of a reagent blank not containing any analyte (nitrite). The same procedure at ten standard deviations above the mean of the blank intensity was performed for the limit of quantification (LOQ) calculation. LOD and LOQ values of analytical results were found to be 0.12 and 0.40 mg·L−1, respectively. Thus, the developed sensor electrode was capable of detecting very low concentrations of nitrite, which may be important for food analysis. The analytical performance of the developed sensor was compared with those of other electrochemical methods utilizing nanoparticle-based sensor electrodes (Table 1). The reproducibility of the voltammetric sensor (Au/p-ATP-Aunano) for nitrite determination was investigated with intra- and inter-assay precision measurements. The intra-assay precision of the sensor was evaluated by determining 10 mg·L−1 nitrite with the same Au/p-ATP-Aunano electrode after five successive determinations. The inter-assay precision, or the fabrication reproducibility, was estimated by detecting the same amount of nitrite in duplicate with five sensor electrodes prepared in the same manner independently. The resulting intra- and inter-assay precisions in terms of percentage Relative Standard Deviation (RSD) values were 3.74% and 8.54%, respectively, indicating high reproducibility. In addition, there was no significant decrease in current response in the detection of 10 mg·L−1 nitrite during the first seven days. 2.7. Colorimetric Sensor Response to NO2− A colorimetric sensor was applied to the determination of nitrite yielding the calibration curve: A565 nm = 0.1331CNO2− − 0.0508 (r = 0.9992)(2) where CNO2− is the NO2− concentration (in mg·L−1) in the final solution and the molar absorptivity is ε = 6.52 × 103 L·mol−1·cm−1 with a limit of detection (LOD) = 0.23 mg·L−1 and a limit of quantification (LOQ) = 0.76 mg·L−1 (LOD = 3σbl/m and LOQ = 10σbl/m, where σbl denotes the standard deviation of a blank and m is the slope of the calibration curve for spectrophotometric nitrite determination). The spectra for nitrite determination at different concentrations are shown in Figure 8. 2.8. Real Sample Analysis Analysis of NO2− in sausage samples after preliminary extraction with the CUPRAC reagent: When nitrite amount of sausages samples was measured with the proposed method (SWV), an interference effect was observed owing to the reducing constituents and additives in these samples. Firstly the interference effect of ascorbic acid (AA), which is added to sausage samples as an antioxidant to protect from light and to stabilize the color, was tested and eliminated. For this purpose, 100 mg·L−1 ascorbic acid was measured with SWV and characteristic peak potential appeared at 0.97 V. This potential was close to the characteristic peak potential of NO2− (at 0.76 V). Therefore, ascorbic acid oxidase was used to inhibit the ascorbic acid-originated interference. The synthetic solutions of 20 mg·L−1 NO2−, 100 mg·L−1 ascorbic acid and the mixture of 20 mg·L−1 NO2− + 100 mg·L−1 ascorbic acid + 2 U·mL−1 ascorbic acid oxidase (in final solution) were separately measured with the proposed method. As shown in Figure 9, the interference of AA was eliminated with ascorbic acid oxidase (AAO) as the analyte-specific enzyme responsible for its oxidation, and the recovery was found to be 100.5% for NO2−. Although AA interference could be overcome with the AAO enzyme, other electro-active food additives of sausage still interfered with nitrite determination. In order to remove all reducing agents having possible interferent effects, a pre-extraction procedure associated with the modified CUPRAC method was applied. In this method, pre-prepared sausages were extracted with Cu(II)-neocuproine into dichloromethane, the reducing constituents oxidized with the CUPRAC reagent and the resulting yellow-orange colored organic phase discarded, and finally, the developed electrochemical method applied to NO2− estimation in sausage samples. Nitrite remained unaffected by the modified CUPRAC extraction. The standard potential for the nitrate-nitrous acid reduction reaction is given as [55]: (3) NO3−+3H++2e−↔HNO2+H2OEo=+0.94 V Therefore, nitrite—having such a high potential—behaves as a reducing agent only for strong oxidants (like permanganate, having a redox potential of +1.51 V in 1.0 N sulfuric acid medium) in acidic pH, for example: 5 NO2− + 2 MnO4− + 6 H+ → 5 NO3− + 2 Mn2+ + 3 H2O(4) On the other hand, Cu(II)-neocuproine complex has a redox potential of about +0.60 V at neutral pH. Therefore, thermodynamically speaking, Cu(II)-neocuproine cannot oxidize nitrite to nitrate, and is itself reduced to the highly-colored Cu(I)-neocuproine chelate at neutral pH. Tütem et al. [56] showed that Cu(II)-Nc does not oxidize nitrite (and is itself reduced to a colored product) at the working pH of the method. Table 2 shows the results obtained by simultaneously applying the electrochemical and colorimetric sensors to sausages samples, prepared as described in “Materials and Methods”. Statistical comparison between the results of the proposed voltammetric and reference colorimetric sensor procedures applied to sausage sample (for Brand “B”) was made on n = 5 repetitive determinations, essentially not showing significant differences between the results (Table 3). The confidence level used in method validation for sausage sample was 99% and 95%, respectively, for t- and F-tests. In the literature, noble metal nanoparticle-based electrochemical sensors have usually been tested in real matrices, such as sausage, milk, table salt, and tap water, while sausage is the most frequently utilized and most complex sample for nitrite. Nitrate and nitrite are used in dry-fermented sausage manufacture [57], for which comparisons are frequently made between samples manufactured with nitrite and those processed incorporating both nitrite and nitrate [58,59]. According to the Turkish Food Codex (TFC-Regulation) issued in the Republic of Turkey Official Newspaper, with Number 28693 and date 30 June 2013, nitrite is a permissible food additive that can be added to certain meat products, including sausage, while nitrite cannot be added to milk products. The nitrite contents of brand B and brand C sausages were below the permissible levels (60 ppm) with respect to the TFC-Regulation, while brand A exceeded this limit (Table 2). Most electrochemical sensors published previously have not considered the analytical problem of interference elimination while determining nitrite in sausage samples. In a study of Yang et al. [60], a phenoxazine dye was electropolymerized on GCE, and the manufactured amperometric sensor was used in the anodic oxidation of nitrite, but ascorbate was observed to seriously interfere (i.e., with a signal change >30% for one-fold molar ratio of ascorbate), and no measure could be taken to eliminate this interference. Saber-Tehrani et al. [40], using a Pt-NPs distributed poly(2-aminothiophenol)-modified electrode, reported that ascorbate, at a five-fold mass ratio to nitrite, caused a current attenuation exceeding 10%, while Wang et al. [61] noted a signal change >5% at 20-fold ascorbate with the use of a Au-NPs on choline chloride modified GCE. Cui et al. [24] used a chitosan-coated Prussian blue nanoparticle sensor electrode in nitrite determination, without mentioning how ascorbate interference was removed. In comparison to three of the methods listed in Table 1, the proposed method had a lower LOD of 2.6 µM. The electrode prepared by Miao et al. [52] used the corrosive and hazardous acid mixture called ‘piranha solution’ for cleansing, and electrode manufacture was cumbersome (requiring heat treatment at 95 °C for 20 min). Most of the listed methods did not report how interferences (arising from ascorbic acid and some amino acid residues on proteins) were eliminated. 3. Discussion This study provided a novel p-aminothiophenol (p-ATP)-modified and gold nanoparticles-derivatized gold electrode (Au/p-ATP-Aunano) for nitrite determination in food samples. The measurements were conducted in pH 4 buffer medium due to the decrease of current intensities below pKa of HNO2 (pKa between 3.2 and 3.4) [23] arising from the decomposition of free nitrous acid into N-oxides (2 HNO2 → NO + NO2 + H2O). Electrochemical approaches are favorable for nitrite determination owing to rapid response, simple operation, and capability of measurement in turbid samples. The developed sensor had good performance in food samples, and was sensitive and stable. The advantage of the co-deposition modification may find further utilization in the field of electrochemical sensing. Compared to the conventional GC electrode showing an oxidation peak at ca. +0.8 V for nitrite, this modified nano-electrode showed a lower potential (at ca. +0.76 V) with intensified current, possibly due to increased effective surface area favorably changing the diffusion regime through the nanoparticles dispersed in the electrode surface film [23]. The repeated use of the electrode without loss of reliability confirmed the absence of surface contamination by irreversible adsorption of nitrite oxidation products and other electrochemical byproducts, and this constitutes a distinct advantage over bare electrodes used for the same purpose. The interference-free utilization of the proposed voltammetric sensing method was also investigated by analysis of nitrite in sausage samples. The slopes of the calibration lines in pure NO2− solution and in the sausage sample solution to which different concentrations of NO2− standards were added were not significantly different from each other, confirming the robustness and interference tolerance of the method. As a major advantage over colorimetric methods employing the Griess reaction, nitrite can be easily determined with the recommended electrochemical method in admixture with nitrate, as the latter is electrochemically inactive. Generally, there is a lack of sufficient information regarding the interference effects of food additives on determination of nitrite, and the counter measures to compensate for possible interferences were not effectively discussed for various modified electrodes presented in literature. Especially, the adverse effect of ascorbic acid (causing a close potential peak overlap) is undeniable. The proposed method distinguishes itself from similar other methods reported in literature with its effective interference removal technique (i.e., removal of ascorbic acid with ascorbate oxidase, and removal of other chemical reductant interferents with Cu(II)-neocuproine extraction into dichloromethane) prior to electrochemical measurement. The sensitivity and selectivity of the proposed method is sufficient to assess the compatibility of meat products (such as sausage) to TFC-Regulation. 4. Materials and Methods 4.1. Chemicals, Solutions, and Instruments The alumina slurry used for electrode cleaning was from Baikowski International Corp (Charlotte, NC, USA) (0.05 µm, Baikalox 0.05CR). The supporting electrolyte for conductivity was 0.1 M pH:4 phosphate buffer solution (NaH2PO4–Na2HPO4) (Merck, Darmstadt, Germany). For electrode cleanliness, isopropyl alcohol (Sigma-Aldrich), acetone, and ethanol (both technical grade) were used. N-(1-naphtyl)-ethylenediamine dichloride (NED) (used in colorimetric sensor determination) was obtained from Fluka (St. Louis, MO, USA). The electrode coating material, 4-aminothiophenol (4-ATP), as well as the rest of the reagents, were supplied from E. Merck (Darmstadt, Germany). Gold(III) chloride solution (99.99% trace metals basis, 30% by wt. in dilute HCl) was used for deposition of gold nanoparticles to Au electrode coated with PATP copolymer, and the 0.04% (w/v) working solution was prepared from this stock solution. The final chemical form of Au(III) in acidic solution was HAuCl4. Preparation of Solutions in colorimetric sensor determination of NO2−: The working solutions of NO2− at 1–10 mg·L−1 were prepared from the corresponding aqueous stock solution at 500 mg·L−1. H3PO4 (0.1 M) and NED (20 mM) solutions were prepared in pure water and 4-ATP (10 mM) solution in absolute ethanol. Voltammetric experiments were performed with a Gamry Instruments model Reference 600 potentiostat/galvanostat/zero resistance ammeter (ZRA) interfaced to a PC computer and controlled Gamry Framework software (Warminster, PA, USA). Gold (BASi stationary voltammetry electrodes; ø 1.6 mm,) was used as the working electrode. A platinum (Pt) electrode and an Ag/AgCl, 3 M KCl electrode served as the auxiliary and reference electrodes, respectively. Optical absorption measurements were carried out in matched Hellma quartz cuvettes using a Varian CARY Bio 100 UV-VIS spectrophotometer (Agilent Technologies, Santa Clara, CA, USA). 4.2. Optimization of Voltammetric Method Electrode type, supporting electrolyte, scan rate, and concentration parameters were individually examined. A gold electrode was selected as the working electrode, by which a substantial signal could be produced (as current intensity) at sufficiently separated reduction potentials. A scan rate of 50 mV·s−1 was chosen, and 0.1 M phosphate buffer solution was preferred as the supporting electrolyte. The electrolyte pH was found to be optimal at pH 4 due to the stability problem of nitrite at more acidic pH. 4.3. Gold Electrode Pre-Treatment The Au electrode was polished with a suspension of alumina powder in the presence of pure water on a smooth polishing cloth [62] by circular movements for a few minutes, then washed with distilled water, and sonicated for 5 min in bi-distilled water. Sonication of the electrode was repeated for another 5 min in isopropyl alcohol–acetonitrile (1:1, v/v) mixture. The efficiency of this procedure for electrode cleanliness was confirmed from the absence of any baseline (blank) peak during CV scan. 4.4. Electrochemical Polymerization of PATP Film The monomer solution (10 mM 4-ATP) was prepared in ethanol. A three-electrode cell was used; an Au working electrode, a platinum wire counter electrode, and an Ag/AgCl reference electrode. Electrochemical polymerization of PATP film was achieved in two steps. In the first step -SH groups of 4-ATP were arranged by self-assembly on the Au electrode surface. Five milliliters of monomer: 4-aminothiophenol solution was taken into the working cell. Polymerization was performed via cyclic voltammetry within the potential range of 0–1.7 V at 20 mV·s−1 scanning speed for 20 cycles. The second step was carried out in a 0.5 M HClO4 solution. Electropolymerization monolayer ATP was formed by CV method from the –NH group end. Polymerization was performed via cyclic voltammetry within the potential range of 0 V to 0.8 V at 50 mV·s−1 scanning speed for 50 cycles. Finally, the modified Au electrode was rinsed with distilled water to remove any unbound material from the surface. 4.5. Electrodeposition of Au Nanoparticles on the Polymer Coated Electrode PATP/Au electrode was coated with Au nanoparticles using cyclic voltammetry with 0.04% (w/v) HAuCl4 (2.5 mL) + 0.1 M H2SO4 (2.5 mL) solutions via electrochemical deposition process. Coating was performed within the (−0.4 V to 0.4 V) range at 50 mV·s−1 scanning speed. The amount of deposited gold particles onto the PATP/Au surface was determined by checking cycle numbers, the optimal value of which was set at 40 cycles. The golden-colored electrode formed in this way was named as Au/PATP/Aunano 4.6. Electrochemical Determinaion of NO2− The working solutions of 2.5–250 mg·L−1 NO2− were prepared from the stock solution and transferred to the measurement cell; one milliliter of NO2− was added and the cell was filled up to 5 mL with pH:4 phosphate buffer solution (NO2− final concentration range was 0.5–50 mg·L−1). Square wave voltammetry (SWV) was performed in a potential range from 0.5–1.1 V, step size 2 mV and pulse size 25 mV, equilibrium time 15 s. The analytical experiment was carried out at a frequency of 50 Hz and the characteristic peak potential of NO2− was determined. Square wave voltammetry was performed as follows: the electrode was held at conditioning potential −0.5 V for 5 s, then equilibrated for 2 s; the analytical experiment was carried out at 25 Hz frequency from initial potential −0.5 V to end potential 1.2 V, where the step potential was 0.00195 V and amplitude 0.01245 V. 4.7. Preparation of Colorimetric Sensor for NO2− Determination Preparation of 4-ATP Functionalized AuNPs. Synthesis and modification of AuNPs was carried out by a previously published method in our laboratory [63]. The methods were originally developed for the determination of nitrite, derived from the hydrolytic cleavage of heterocyclic nitramine energetic materials (i.e., hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX). Fifty milliliters of 0.002% HAuCl4 was heated to boiling. To this solution, 0.5 mL of 1% trisodium citrate was added. The solution was heated until its color changed to the red-wine characteristic of the surface plasmons of gold, then removed from the hot plate, and cooled to room temperature. AuNPs were further functionalized in the same solution (i.e., without centrifugation) with 4-ATP. First, the pH of AuNPs was adjusted to 3.0 where the thiol group of 4-ATP showed maximum affinity toward the surface of AuNPs. This acidic AuNPs solution was mixed with 10 mM 4-ATP at 8:1 (v/v) ratio and then stirred at 60 °C for 3 h. The solution was then left to stabilize at room temperature for 48 h. After functionalization, 4-ATP−AuNP remained useful for quantitative analysis for at least two days without losing its maximal absorbance for nitrite. The scheme for the colorimetric sensor is summarized as follows: take 2 mL NO2− sample; add 1.0 mL of AuNP-4ATP + 0.2 mL of 0.1 M H3PO4 + 0.5 mL of 20.0 mM NED in this order; measure A565 nm against a reagent blank after 30 min of NED addition. 4.8. Application of Voltammetric and Colorimetric Sensors to Sausage Samples Three different brands of sausage samples were used for analysis. In accordance with literature [53], 10 g of sausage was shredded, 12.5 mL of saturated borax solution added, and then put into a 70 °C water bath for 15 min. At the end of this time, the proteins in the sausages were precipitated by adding 2.5 mL of 30% ZnSO4 solution. After that, the samples were put into a centrifuge tube and centrifuged for 10 min at 8000 rpm and filtered through Glass Fiber/PolyEthylene Terephthalate (GF/PET) 45/25 filter via a syringe, and finally diluted to 50 mL. Both voltammetric and colorimetric sensors were applied to the obtained samples. Voltammetric Sensing of NO2− in Sausage Samples: Since the direct determination of nitrite in the sausage samples with the developed voltammetric method was not possible due to the oxidation of other interfering constituents at close potentials, a preliminary extraction was performed, based on the CUPRAC method existing in literature [64]. Preliminary treatment and interference removal of sausage samples with an extractive CUPRAC method: 0.5 mL 10 mM CuCl2·2H2O + 1.2 mL 10 mM Nc + 2 mL pH:7 (NH4Ac) buffer + 4 mL sausage sample + 4 mL dicloromethane (after this extraction, 7.3 mL aqueous extract obtained). Voltammetric measurement on sausage samples after CUPRAC extraction: In the measurement cell; 0.5 mL sample + 0.5 mL water + 4 mL pH:4 buffer. Scan rate: 50 mV·s−1, readings were made with SWV within the 0.5–1.1 V potential range. Colorimetric Sensing of NO2− in Sausage Samples: The determination of nitrite in sausage samples with the use of the colorimetric sensor was performed using the method of standard additions without any need of pre-extraction into organic solvent with the CUPRAC reagent. The method of standard additions was applied to the extracted sausage samples as follows: Two milliliters of extracted sausage samples + nitrite addition within the linear calibration range; dilute with distilled water to 10 mL. As unspiked sample, 2 mL of sausage sample extract was also taken and diluted to 10 mL. The previously developed method for the determination of heterocyclic nitramine energetic materials, abbreviated as “4-ATP-AuNP+NED”, was modified for the determination of nitrite and standard-added samples as follows: Two milliliters of NO2− or standard-added samples + 1 mL AuNP-4ATP + 0.2 mL of 0.1 M H3PO4 + 0.5 mL of 20 mM NED, taken for colorimetric determination. Acknowledgments The authors thank TUBITAK (Turkish Scientific and Technical Research Council) for their support to the Research Project 112T792. Additionally, thanks are extended to the Scientific Research Projects Coordination Unit of Istanbul University (BAP), who supported this work with the project number: 57182. Author Contributions Ayşem Üzer and Erol Erçağ conceived and designed the experiments; Şener Sağlam and Ziya Can performed the experiments; Ayşem Üzer and Erol Erçağ contributed the reagents/materials/analysis tools; Reşat Apak and Ayşem Üzer recognized the analytical problem subject to this study, analyzed the data and wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Cyclic voltammograms of –SH groups array on the Au electrode surface (obtained with the use of 5 mL of 10 mM 4-aminothiophenol (p-ATP) solution); cycles in numerical order, from top to bottom. Figure 2 Cyclic voltammograms of electrochemical polymerization monolayer ATP; cycles in numerical order from top to bottom. Figure 3 Voltammograms regarding Au nanoparticles on a polyaminothiophenol (PATP) polymer-coated Au electrode; cycles in numerical order, from bottom to top. Figure 4 Cyclic voltammetry (CV) characterization of Au, Au/PATP, and Au/PATP-Aunano electrodes. Figure 5 The scanning electron microscope (SEM) images of Au colloidal particles ranging between 75 and 300 nm size (10,000-fold zoom). Figure 6 Impedance measurements on the Au/PATP and Au/PATP-Aunano electrodes using the potentiostat electrochemical impedance spectroscopy (EIS) method in monomer-free solution media. Figure 7 Square wave voltammograms of NO2− with Au/PATP-Aunano electrode within the concentrations of 0.5−50 mg·L−1 for NO2−. Figure 8 Spectra of NO2− with respect to the colorimetric sensing method within the initial concentration range of 1−10 mg·L−1. The color images of the test tubes containing from left to right, blank sample and colors obtained from samples of 1–10 mg·L−1 concentration range are shown in the inset figure. Figure 9 Square wave voltammograms of 20 mg·L−1 NO2−, 100 mg·L−1 ascorbic acid and the mixture of 20 mg·L−1 NO2− + 100 mg·L−1 ascorbic acid + 100 mg·L−1 ascorbic acid oxidase. ijms-17-01253-t001_Table 1Table 1 The comparison of analytical figures of merit with other electrochemical methods utilizing nanoparticle-based sensor electrodes. Electrode Material Method LOD µM Linear Range Reference EPPGE/SWCNT/Co and CoOx NP Chronoamperometry 5.6 32.3–189 µM [51] GE/AEBA/DPAN/PtNPs Amperometry 5 1–10 µM [52] ACNTs/thionin modified Differential Pulse Voltammetry (DPV) 1.12 3 × 10−6–5 × 10−4 mol·L−1 [53] GE/PATP-Ptnano Cyclic Voltammetry (CV) 1 3 µmol·L−1–1 mmol·L−1 [40] PGCE/nanometer-sized gold colloid/ethylenediamine Amperometry 45 1.3 × 10−4–4.4 × 10−2 mol·L−1 [54] GE/p-ATP-Aunano Square Wave Voltammetry (SWV) 2.6 0.5–50 mg·L−1 Üzer et al. (Proposed method) LOD: limit of detection; EPPGE: edge plane pyrolytic graphite electrode; GE: gold electrode; SWCNT: single-walled carbon nanotubes; CoOx: cobalt oxide; ACNTs: carbon nanotubes; AEBA: 4-(2-aminoethyl)benzenamine; DPAN: 5-[1,2]dithiolan-3-yl-pentanoic acid [2-(naphthalene-1-ylamino)-ethyl]amide; PtNPs: platinum nanoparticles; PGCE: pretreated glassy carbon electrode. ijms-17-01253-t002_Table 2Table 2 Voltammetric and colorimetric results of three different brand sausages samples (number of measurements for both methods: n = 5). Method Brand “A” Brand “B” Brand “C” Voltammetric 22.48 mg·L−1 5.83 mg·L−1 9.47 mg·L−1 Colorimetric 17.64 mg·L−1 5.18 mg·L−1 7.71 mg·L−1 ijms-17-01253-t003_Table 3Table 3 Statistical comparison of the proposed voltammetric sensor with a colorimetric sensor method applied to a sausage sample (number of measurements for both methods: n = 5). Method Mean Conc. (mg·L−1) SD (σ) S a,b t a,b ttable b F b Ftable b Voltammetric 5.83 0.522 - - - - - Colorimetric 5.18 0.276 0.418 2.402 3.355 0.278 6.39 a S2 = ((n1 − 1)s12 + (n2 − 1)s22)/(n1 + n2 − 2) and t = (ā1 − ā2)/(S(1/n1 + 1/n2)1/2), where S is the pooled standard deviation, s1 and s2 are the standard deviations of the two populations with sample sizes of n1 and n2, and sample means of ā1 and ā2, respectively (t has (n1 + n2 − 2) degrees of freedom); here, n1 = n2 = 5; b The statistical comparison is made on paired data produced with proposed and reference methods; the results are given only on the row of the reference method. Conc.: concentration, SD: standard deviation ==== Refs References 1. Alahakoon A.U. Jayasena D.D. Ramachandra S. Jo C. Alternatives to nitrite in processed meat: Up to date Trends Food Sci. Technol. 2015 45 37 49 10.1016/j.tifs.2015.05.008 2. Parthasarathy D.K. Bryan N.S. Sodium nitrite: The “cure” for nitric oxide insufficiency Meat Sci. 2012 92 274 279 10.1016/j.meatsci.2012.03.001 22464105 3. Pegg R.B. Shahidi F. Nitrite Curing of Meat Food & Nutrition Press Inc. Trumbull, CT, USA 2008 1 268 4. Jo C. Ahn H.J. Son J.H. Lee J.W. Byun M.W. Packaging and irradiation effect on lipid oxidation, color, residual nitrite content, and nitrosamine formation in cooked pork sausage Food Control 2003 14 7 12 10.1016/S0956-7135(02)00045-2 5. Moorcroft M.J. Davis J. Compton R.G. Detection and determination of nitrate and nitrite: A review Talanta 2001 54 785 803 10.1016/S0039-9140(01)00323-X 18968301 6. Helaleh M.I.H. Korenaga T. Ion chromatographic method for simultaneous determination of nitrate and nitrite in human saliva J. Chromatogr. B 2000 744 433 437 10.1016/S0378-4347(00)00264-4 7. Horikoshi N. Nagamoto S. Kitagawa T. Kohzuma T. Resonance raman spectroscopic studies of a copper-containing nitrite reductase from the Alcaligenes xylosoxidans NCIB 11015 J. Inorg. Biochem. 1997 67 46 10.1016/S0162-0134(97)89927-8 8. Nadzhafova O. Etienne M. Walcarius A. Direct electrochemistry of hemoglobin and glucose oxidase in electrodeposited sol–gel silica thin films on glassy carbon Electrochem. Commun. 2007 9 1189 1195 10.1016/j.elecom.2007.01.010 9. Griess P. Bemerkungen zu der Abhandlung der HH. Weselsky und Benedikt “Ueber einige Azoverbindungen” Ber. Dtsch. Chem. Ges. 1879 12 426 428 10.1002/cber.187901201117 10. Wang H. Yang W. Liang S.-C. Zhang Z.-M. Zhang H.-S. Spectrofluorimetric determination of nitrite with 5,6-diamino-1,3-naphthalene disulfonic acid Anal. Chim. Acta 2000 419 169 173 10.1016/S0003-2670(00)01000-X 11. Zhang X. Wang H. Liang S.-C. Zhang H.-S. Spectrofluorimetric determination of nitric oxide at trace levels with 5,6-diamino-1,3-naphthalene disulfonic acid Talanta 2002 56 499 504 10.1016/S0039-9140(01)00582-3 18968522 12. Dombrowski L.J. Pratt E.J. Fluorometric method for determining nanogram quantities of nitrite ion Anal. Chem. 1972 44 2268 2272 10.1021/ac60322a040 13. Camargo A. Aguirre M.J. Cheuquepán W. Chen Y.-Y. Ramírez G. Electrooxidation of Nitrite Mediated by Cu-x -Tetraaminophenylporphyrin (x = 2, 3, and 4) Glassy Carbon-Modified Electrodes: Effect of Substituent Position Electroanalysis 2008 20 2635 2641 10.1002/elan.200804369 14. Da Rocha J.R.C. Angnes L. Bertotti M. Araki K. Toma H.E. Amperometric detection of nitrite and nitrate at tetraruthenated porphyrin-modified electrodes in a continuous-flow assembly Anal. Chim. Acta 2002 452 23 28 10.1016/S0003-2670(01)01440-4 15. Lin C.-Y. Balamurugan A. Lai Y.-H. Ho K.-C. A novel poly(3,4-ethylenedioxythiophene)/iron phthalocyanine/multi-wall carbon nanotubes nanocomposite with high electrocatalytic activity for nitrite oxidation Talanta 2010 82 1905 1911 10.1016/j.talanta.2010.08.010 20875594 16. Armijo F. Goya M.C. Reina M. Canales M.J. Arévalo M.C. Aguirre M.J. Electrocatalytic oxidation of nitrite to nitrate mediated by Fe(III) poly-3-aminophenyl porphyrin grown on five different electrode surfaces J. Mol. Catal. A Chem. 2007 268 148 154 10.1016/j.molcata.2006.11.055 17. Da Silva S.M. Mazo L.H. Differential Pulse Voltammetric Determination of Nitrite with Gold Ultramicroelectrode Electroanalysis 1998 10 1200 1203 10.1002/(SICI)1521-4109(199811)10:17<1200::AID-ELAN1200>3.0.CO;2-5 18. Jiang Y.N. Luo H.Q. Li N.B. Determination of nitrite with a nano-gold modified glassy carbon electrode by cyclic voltammetry Int. J. Environ. Anal. Chem. 2007 87 295 306 10.1080/03067310601079582 19. Yang W. Bai Y. Li Y. Sun C. Amperometric nitrite sensor based on hemoglobin/colloidal gold nanoparticles immobilized on a glassy carbon electrode by a titania sol-gel film Anal. Bioanal. Chem. 2005 382 44 50 10.1007/s00216-005-3160-1 15900450 20. Milhano C. Pletcher D. The electrodeposition and electrocatalytic properties of copper–palladium alloys J. Electroanal. Chem. 2008 614 24 30 10.1016/j.jelechem.2007.11.001 21. Davis J. Moorcroft M.J. Wilkins S.J. Compton R.G. Cardosi M.F. Electrochemical detection of nitrate and nitrite at a copper modified electrode Analyst 2000 125 737 741 10.1039/a909762g 22. Agboola B.O. Ozoemena K.I. Nyokong T. Electrochemical properties of benzylmercapto and dodecylmercapto tetra substituted nickel phthalocyanine complexes: Electrocatalytic oxidation of nitrite Electrochim. Acta 2006 51 6470 6478 10.1016/j.electacta.2006.04.033 23. Kozub B.R. Rees N.V. Compton R.G. Electrochemical determination of nitrite at a bare glassy carbon electrode; why chemically modify electrodes? Sens. Actuators B 2010 143 539 546 10.1016/j.snb.2009.09.065 24. Cui L. Zhu J. Meng X. Yin H. Pan X. Ai S. Controlled chitosan coated Prussian blue nanoparticles with the mixture of graphene nanosheets and carbon nanoshperes as a redox mediator for the electrochemical oxidation of nitrite Sens. Actuators B 2012 161 641 647 10.1016/j.snb.2011.10.083 25. Li X.-R. Liu J. Kong F.-Y. Liu X.-C. Xu J.-J. Chen H.-Y. Potassium-doped graphene for simultaneous determination of nitrite and sulfite in polluted water Electrochem. Commun. 2012 20 109 112 10.1016/j.elecom.2012.04.014 26. Mani V. Periasamy A.P. Chen S.-M. Highly selective amperometric nitrite sensor based on chemically reduced graphene oxide modified electrode Electrochem. Commun. 2012 17 75 78 10.1016/j.elecom.2012.02.009 27. Wen Z.-H. Kang T.-F. Determination of nitrite using sensors based on nickel phthalocyanine polymer modified electrodes Talanta 2004 62 351 355 10.1016/j.talanta.2003.08.003 18969302 28. Caro C.A. Bedioui F. Zagal J.H. Electrocatalytic oxidation of nitrite on a vitreous carbon electrode modified with cobalt phthalocyanine Electrochim. Acta 2002 47 1489 1494 10.1016/S0013-4686(01)00875-1 29. Abbaspour A. Mehrgardi M.A. Electrocatalytic activity of Ce(III)–EDTA complex toward the oxidation of nitrite ion Talanta 2005 67 579 584 10.1016/j.talanta.2005.03.036 18970209 30. Pournaghi-Azar M.H. Dastangoo H. Electrocatalytic oxidation of nitrite at an aluminum electrode modified by a chemically deposited palladium pentacyanonitrosylferrate film J. Electroanal. Chem. 2004 567 211 218 10.1016/j.jelechem.2003.12.027 31. Winnischofer H. de Souza Lima S. Araki K. Toma H.E. Electrocatalytic activity of a new nanostructured polymeric tetraruthenated porphyrin film for nitrite detection Anal. Chim. Acta 2003 480 97 107 10.1016/S0003-2670(02)01594-5 32. Cardoso W.S. Gushikem Y. Electrocatalytic oxidation of nitrite on a carbon paste electrode modified with Co(II) porphyrin adsorbed on SiO2 /SnO2 /Phosphate prepared by the sol–gel method J. Electroanal. Chem. 2005 583 300 306 10.1016/j.jelechem.2005.06.015 33. Dai Z. Xu X. Ju H. Direct electrochemistry and electrocatalysis of myoglobin immobilized on a hexagonal mesoporous silica matrix Anal. Biochem. 2004 332 23 31 10.1016/j.ab.2004.03.067 15301945 34. Strehlitz B. Gründig B. Schumacher W. Kroneck P.M.H. Vorlop K.-D. Kotte H. A Nitrite Sensor Based on a Highly Sensitive Nitrite Reductase Mediator-Coupled Amperometric Detection Anal. Chem. 1996 68 807 816 10.1021/ac950692n 21619176 35. Golub E. Pelossof G. Freeman R. Zhang H. Willner I. Electrochemical, Photoelectrochemical, and Surface Plasmon Resonance Detection of Cocaine Using Supramolecular Aptamer Complexes and Metallic or Semiconductor Nanoparticles Anal. Chem. 2009 81 9291 9298 10.1021/ac901551q 19860374 36. Huang J. Yang G. Meng W. Wu L. Zhu A. Jiao X.A. An electrochemical impedimetric immunosensor for label-free detection of Campylobacter jejuni in diarrhea patients’ stool based on O-carboxymethylchitosan surface modified Fe3 O4 nanoparticles Biosens. Bioelectron. 2010 25 1204 1211 10.1016/j.bios.2009.10.036 19932018 37. Lin H. Ji X. Chen Q. Zhou Y. Banks C.E. Wu K. Mesoporous-TiO2 nanoparticles based carbon paste electrodes exhibit enhanced electrochemical sensitivity for phenols Electrochem. Commun. 2009 11 1990 1995 10.1016/j.elecom.2009.08.034 38. Tel-Vered R. Yildiz H.B. Willner I. Enhanced photocurrents generated by supramolecular relay/CdS-nanoparticle/electron-donor structures on gold electrodes Adv. Mater. 2009 21 716 720 10.1002/adma.200801612 39. Huang X. Li Y. Chen Y. Wang L. Electrochemical determination of nitrite and iodate by use of gold nanoparticles/poly(3-methylthiophene) composites coated glassy carbon electrode Sens. Actuators B 2008 134 780 786 10.1016/j.snb.2008.06.028 40. Saber-Tehrani M. Pourhabib A. Husain S.W. Arvand M. A Simple and Efficient Electrochemical Sensor for Nitrite Determination in Food Samples Based on Pt Nanoparticles Distributed Poly(2-aminothiophenol) Modified Electrode Food Anal. Methods 2013 6 1300 1307 10.1007/s12161-012-9543-y 41. Zhang L. Jia J. Zou X. Dong S. Simultaneous Determination of Dopamine and Ascorbic Acid at an In-Site Functionalized Self-Assembled Monolayer on Gold Electrode Electroanalysis 2004 16 1413 1418 10.1002/elan.200302965 42. Gopalan A.I. Lee K.-P. Manesh K.M. Santhosh P. Kim J.H. Kang J.S. Electrochemical determination of dopamine and ascorbic acid at a novel gold nanoparticles distributed poly(4-aminothiophenol) modified electrode Talanta 2007 71 1774 1781 10.1016/j.talanta.2006.08.026 19071522 43. Lukkari J. Kleemola K. Meretoja M. Ollonqvist T. Kankare J. Electrochemical Post-Self-Assembly Transformation of 4-Aminothiophenol Monolayers on Gold Electrodes Langmuir 1998 14 1705 1715 10.1021/la970931x 44. Gopalan A.I. Lee K.-P. Manesh K.M. Santhosh P. Kim J.H. Gold nanoparticles dispersed into poly(aminothiophenol) as a novel electrocatalyst—Fabrication of modified electrode and evaluation of electrocatalytic activities for dioxygen reduction J. Mol. Catal. A Chem. 2006 256 335 345 10.1016/j.molcata.2006.05.027 45. Arbizzani C. Catellani M. Mastragostino M. Mingazzini C. N - and P -doped Polydithieno[3,4-B:3′,4′-D] thiophene: A narrow band gap polymer for redox supercapacitors Electrochim. Acta 1995 40 1871 1876 10.1016/0013-4686(95)00096-W 46. Zhou M. Zhai Y. Dong S. Electrochemical Sensing and Biosensing Platform Based on Chemically Reduced Graphene Oxide Anal. Chem. 2009 81 5603 5613 10.1021/ac900136z 19522529 47. Gu H.-Y. Yu A.-M. Chen H.-Y. Direct electron transfer and characterization of hemoglobin immobilized on a Au colloid–cysteamine-modified gold electrode J. Electroanal. Chem. 2001 516 119 126 10.1016/S0022-0728(01)00669-6 48. Girija T.C. Sangaranarayanan M.V. Investigation of polyaniline-coated stainless steel electrodes for electrochemical supercapacitors Synth. Met. 2006 156 244 250 10.1016/j.synthmet.2005.12.006 49. Bradbury C.R. Zhao J. Fermin D.J. Distance-Independent Charge-Transfer Resistance at Gold Electrodes Modified by Thiol Monolayers and Metal Nanoparticles J. Phys. Chem. C 2008 112 10153 10160 10.1021/jp710165d 50. Analytical Methods Committee Recommendations for the definition, estimation and use of the detection limit Analyst 1987 112 199 204 51. Adekunlea A.S. Pillaya J. Ozoemena K.I. Probing the electrochemical behaviour of SWCNT-cobalt nanoparticles and their electrocatalytic activities towards the detection of nitrite at acidic and physiological pH conditions Electrochim. Acta 2010 55 4319 4327 10.1016/j.electacta.2009.02.102 52. Miao P. Shen M. Ning L.M. Chen G.F. Yin Y.M. Functionalization of platinum nanoparticles for electrochemical detection of nitrite Anal. Bioanal. Chem. 2011 399 2407 2411 10.1007/s00216-010-4642-3 21225247 53. Zhao K. Song H.Y. Zhuang S.Q. Dai L.M. He P.G. Fang Y.Z. Determination of nitrite with the electrocatalytic property to the oxidation of nitrite on thionine modified aligned carbon nanotubes Electrochem. Commun. 2007 9 65 70 10.1016/j.elecom.2006.07.001 54. Liu Y. Gu H.-Y. Amperometric detection of nitrite using a nanometer-sized gold colloid modified pretreated glassy carbon electrode Microchim. Acta 2008 162 101 106 10.1007/s00604-007-0946-9 55. Fanning J.C. The chemical reduction of nitrate in aqueous solution Coord. Chem. Rev. 2000 199 159 179 10.1016/S0010-8545(99)00143-5 56. Tütem E. Apak R. Baykut F. Spectrophotometric determination of trace amounts of copper(I) and reducing agents with neocuproine in the presence of copper(II) Analyst 1991 116 89 94 10.1039/AN9911600089 57. Marco A. Navarro J.L. Flores M. The influence of nitrite and nitrate on microbial, chemical and sensory parameters of slow dry fermented sausage Meat Sci. 2006 73 660 673 10.1016/j.meatsci.2006.03.011 22062567 58. Noel P. Briand E. Dumont J.P. Role of nitrite in flavor development in uncooked cured meat-products—Sensory assessment Meat Sci. 1990 28 1 8 10.1016/0309-1740(90)90015-X 22055434 59. Skjelvale R. Tjaberg T.B. Comparison of salami sausage produced with and without addition of sodium nitrite and sodium nitrate J. Food Sci. 1974 39 520 524 10.1111/j.1365-2621.1974.tb02938.x 60. Yang C. Lu Q. Hu S. A Novel Nitrite Amperometric Sensor and Its Application in Food Analysis Electroanalysis 2006 18 2188 2193 10.1002/elan.200603646 61. Wang P. Mai Z. Daia Z. Li Y. Zou X. Construction of Au nanoparticles on choline chloride modified glassy carbon electrode for sensitive detection of nitrite Biosens. Bioelectron. 2009 24 3242 3247 10.1016/j.bios.2009.04.006 19443208 62. Özkan S.A. Uslu B. Zuman P. Electrochemical reduction and oxidation of the antibiotic cefepime at a carbon electrode Anal. Chim. Acta 2002 457 265 274 10.1016/S0003-2670(02)00022-3 63. Üzer A. Can Z. Akın İ. Ercağ E. Apak R. 4-Aminothiophenol Functionalized Gold Nanoparticle-Based Colorimetric Sensor for the Determination of Nitramine Energetic Materials Anal. Chem. 2014 86 351 356 10.1021/ac4032725 24299426 64. Çelik S.E. Özyürek M. Güçlü K. Apak R. Solvent effects on the antioxidant capacity of lipophilic and hydrophilic antioxidants measured by CUPRAC, ABTS/persulphate and FRAP methods Talanta 2010 81 1300 1309 10.1016/j.talanta.2010.02.025 20441899

PMC005xxxxxx/PMC5000652.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081254ijms-17-01254ArticlePharmacological Activities of Ruthenium Complexes Related to Their NO Scavenging Properties Castellarin Anna 12Zorzet Sonia 2Bergamo Alberta 1Sava Gianni 12*Hadjikakou Sotiris K. Academic Editor1 Callerio Foundation Onlus, via A. Fleming 22-31, 34127 Trieste, Italy; [email protected] (A.C.); [email protected] (A.B.)2 Department of Life Sciences, University of Trieste, 34127 Trieste, Italy; [email protected]* Correspondence: [email protected]; Tel.: +39-40-558-863702 8 2016 8 2016 17 8 125429 6 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Angiogenesis is considered responsible for the growth of primary tumours and of their metastases. With the present study, the effects of three ruthenium compounds, potassiumchlorido (ethylendiamminotetraacetate)rutenate(III) (RuEDTA), sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2] (KP1339) and trans-imidazoledimethylsulphoxidetetrachloro-ruthenate (NAMI-A), are studied in vitro in models mimicking the angiogenic process. The ruthenium compounds reduced the production and the release of nitrosyls from either healthy macrophages and immortalized EA.hy926 endothelial cells. The effects of NAMI-A are qualitatively similar and sometimes quantitatively superior to those of RuEDTA and KP1339. NAMI-A reduces the production and release of nitric oxide (NO) by the EA.hy926 endothelial cells and correspondingly inhibits their invasive ability; it also strongly inhibits the angiogenesis in matrigel sponges implanted subcutaneously in healthy mice. Taken together, these data support the anti-angiogenic activity of the tested ruthenium compounds and they contribute to explain the selective activity of NAMI-A against solid tumour metastases, the tumour compartment on which angiogenesis is strongly involved. This anti-angiogenic effect may also contribute to the inhibition of the release of metastatic cells from the primary tumour. Investigations on the anti-angiogenic effects of NAMI-A at this level will increase knowledge of its pharmacological properties and it will give a further impulse to the development of this class of innovative metal-based drugs. rutheniumangiogenesisanticancernitric oxidecell cultures ==== Body 1. Introduction The role of the angiogenic processes for cancer growth and dissemination has been long demonstrated and angiogenesis identified as a target for the pharmacological control of tumour malignancy [1,2,3,4]. A significant example of anti-angiogenic drug is Avastin (bevacizumab), a monoclonal antibody that changed the treatment paradigm of tumours such as colorectal cancer [5]. The pharmacological control of tumour angiogenesis, besides the use of specific monoclonal antibodies directed to vascular endothelial growth factor (VEGF) (see the above reported Avastin), can be controlled with tyrosine kinase inhibitors (TKIs, for example Sunitinib [6]) that block the signalling pathway that activates the angiogenic processes, and/or by the use of chemicals that interrupt the signalling between cells (for example Lenalidomide, the levo enantiomer of the old drug Thalidomide [7]). Independently of the type of drug being used, and of its mechanism of action, the main pharmacological effect expected is the arrest of the formation of new vessels, induced by the tumour, and correspondingly the block to the growth of the tumour because of the inhibited arrival of nutrients to the cells of the tumour mass. Ruthenium-based drugs have also shown the ability to control the phenomena related to angiogenesis. In particular, imidazolium trans-imidazoledimethylsulphoxidetetrachloro-ruthenate (NAMI-A) inhibits the angiogenic process mimicked in the chick embryo chorioallantoic membrane (CAM) model [8] and in the pellets implanted in the rabbit cornea [9]. A more detailed study with the endothelial cell line ECV304 has clarified how NAMI-A was able to inhibit the growth of these cells through the inhibition of the receptor activated signal-extracellular regulated kinase (RAS-ERK) pathway at concentrations compatible with those reached in vivo in the metastatic sites [10,11,12]. The mechanism by which NAMI-A can control tumour angiogenesis has not been elucidated yet and support to this activity only comes from the observation that it is able to bind nitric oxide (NO), then reducing the activity of this gaseous transmitter that cancer cells use to modulate the angiogenesis on the endothelial cells [13,14,15]. The aim of the present study was therefore that of examining the pharmacological actions of NAMI-A, compared to those of two other ruthenium-based compounds (KP1339 and RuEDTA), in relation to their capacity to scavenge the nitric oxide. It is important to remember that the scavenging activity of NO of these compounds has already been studied through fourier transform infrared spectroscopy (FT-IR) and 1H-nuclear magnetic resonance (1H-NMR) spectroscopy techniques [9]. Here, the study will focus on the use of two in vitro cell models, the peritoneal murine macrophages (a cell population present in the tumour masses and to which the capacity to produce and release large quantities of NO is attributed [16]) and the human endothelial-like immortalised cell line EA.hy926 [17], and on the use of rat aorta rings cultured in vitro. Matrigel™ pellets implanted subcutaneously in mice will add further data from in vivo studies. 2. Results 2.1. Murine Peritoneal Macrophage Model Effects on Nitric Oxide (NO) Production Upon treatment with lipopolysaccharide (LPS), which is known to stimulate the inducible form of nitric oxide synthase (NOS), murine peritoneal macrophages are strongly activated to produce nitric oxide (NO) (Figure 1). The contemporary treatment with 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO) (used as the NO scavenger) significantly increased the extracellular NO2− concentration as detected by the Griess test. This effect is consequent to the mechanism by which PTIO exerts its NO scavenging activity; PTIO oxidises the excess of NO released in the extracellular medium, transforming it into NO2−, which is then detected by the Griess test as an increase of NO2− concentration (Figure 1). The treatment of murine peritoneal macrophages with 10−4 and 3 × 10−4 M NAMI-A significantly reduced the release of NO in the extracellular medium, that had been induced by the contemporary stimulation with LPS (Figure 2). The effect is comparable to that of equal concentrations of Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME), the well-known inhibitor of NOS. The treatment with potassiumchlorido (ethylendiamminotetraacetate)rutenate(III) (RuEDTA) is even more effective, showing a reduction of NO release of about 90% in comparison to untreated controls. The ruthenium compounds maintained their ability to lower the NO release also when the cells were treated before being activated with LPS (Figure 3). In this case, the reduction of NO release was quantitatively less relevant than that measured in the contemporary treatment reported in Figure 2, and it was similar for NAMI-A and RuEDTA (reduction of approximately 25%). The pre-treatment with sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2] (KP1339) was somewhat less effective. 2.2. EA.Hy926 Endothelial Cell Line Model 2.2.1. Cytotoxicity Test Before examining the activity of the ruthenium compounds on the NO production of the EA.hy926 endothelial cell line, we studied their effects on the cell viability after cell treatment for 48 or 72 h (Figure 4A,B, respectively). The same test was done with the NO scavenger PTIO, with the NOS inhibitor l-NAME, and with the NO donor S-nitroso-N-acetyl-dl-penicillamine (SNAP), using the same experimental conditions. Among the ruthenium compounds tested, KP1339 was the most cytotoxic, with IC50 values in the 10−5 M range after either 48 or 72 h of cell exposure (Table 1). NAMI-A and RuEDTA showed a quite similar activity on the endothelial cell viability being able to significantly reduce it only after treatment at the highest concentration tested (10−3 M). The NOS inhibitor, l-NAME, was virtually devoid of any cytotoxicity up to the maximum concentration used, i.e., 10−3 M. SNAP and PTIO significantly affected cell viability after treatments at the two highest concentrations tested (10−4 and 10−3 M). 2.2.2. Cell Invasion Test NAMI-A significantly reduced, in a concentration-dependent manner, the invasive ability of the EA.hy926 endothelial cells (Figure 5). RuEDTA was also effective, although to a lesser extent (−17% at 3 × 10−4 M compared to −45% of NAMI-A at equimolar concentrations), and its activity was similar to that of the NOS inhibitor l-NAME. The NO donor SNAP significantly decreased the invasion of the endothelial cells, although this effect seemed mostly related to its cytotoxicity, consequent to the consistent release of NO that it induced at the highest concentration tested. 2.2.3. NO Production and Release The basal release of NO by EA.hy926 cells is rather low; therefore an experimental condition of high level of exogenous NO was generated using the NO donor compound SNAP. To maximally limit the drawbacks of the production of NO at non-physiological levels, the time exposure of cells to SNAP was secured at 2 h. With these experimental conditions, SNAP significantly increased the NO levels starting from 10−5 and 10−4 M, respectively, in the extracellular medium (Figure 6A) and inside the treated cells (Figure 6B). From this preliminary experiment a concentration of 10−4 M SNAP was judged to be the most appropriate for the subsequent experiments. The treatment of the EA.hy926 endothelial cells with 10−4 M NAMI-A, RuEDTA and KP1339 for 2 h significantly counteracted the NO increase induced by the contemporary treatment with an equimolar concentration of SNAP in either the extracellular medium and intracellularly (Figure 7A,B). At the intracellular level, the NO scavenging ability of the ruthenium compounds was comparable, and even more pronounced than that of the standard NO scavenger PTIO. The increase of NO release in the extracellular medium upon PTIO treatment (Figure 7A) was the consequence of its mechanism of action, as already reported above. The NO scavenging ability of the ruthenium compounds during the contemporary exposure to the NO donor SNAP was further investigated to verify the relationship between the effect and the concentration tested (Figure 8A,B); NAMI-A and RuEDTA showed a concentration-dependent reduction of NO at either the extracellular and intracellular levels. The treatment with NAMI-A, RuEDTA, and KP1339, as well as with the positive control PTIO, for 2 h before the cell exposure to the NO donor SNAP, maintained their ability to reduce the NO levels inside the cells (Figure 9B). The effect was similar for the three ruthenium compounds, although quantitatively attenuated, in comparison to that observed with the contemporary treatment (approximately −25% vs. −90%; compare Figure 9B and Figure 7B). As expected, the pre-treatment with the ruthenium compounds, as well with PTIO, was completely ineffective to modulate the NO levels in the extra-cellular medium (Figure 9A). The NO scavenging ability of NAMI-A and RuEDTA at the intracellular level was measurable and it was relevant also when the treatment with the NO donor SNAP preceded the cell exposure to the ruthenium compounds (Figure 10). Once again the activity of the ruthenium compounds was quantitatively comparable (approximately −45% independently of the compound being tested). As expected, the detection of NO in the extracellular medium of control cells was under the threshold level (data not shown). 2.3. Anti-Angiogenic Activity of NAMI-A in the Model of Matrigel™ Pellets The morphological analysis of control Matrigel™ pellets showed a remarkable angiogenic activity, characterized by a diffuse network of blood vessels and an overt red colour (Figure 11A, left). In contrast, the pellets containing NAMI-A presented a pale coloration, indicative of a lesser presence of vessels (Figure 11A, right). This observation was confirmed by the quantitative analysis of the haemoglobin content in the pellets that was remarkably lower (−90%) in the treated group (Figure 11B). 3. Discussion A significant impulse to the study and development of anticancer complexes of ruthenium has been given by the pioneering works with the so-called “symmetrical bis-heterocycles ruthenium(III)” [18] and with the dimethylsulphoxide-containing ruthenium(II) and later ruthenium(III) compounds [19,20]. If the initial idea was to mimic cisplatin in either potency or mechanism of action, suddenly the “ruthenium-sulphoxides” showed a low level of cell cytotoxicity in vitro [21] and a relatively modest activity against the primary sites of growth of syngenic mouse tumours in vivo [22]. On the contrary, these metal-based compounds exhibited an innovative capacity to control the development of secondary tumours (metastases) with either mouse models of solid tumours [23,24] and leukaemias [25]. The progress of knowledge on the contribution of angiogenesis to the development of tumour metastases prompted us to verify whether the anti-metastatic properties of NAMI-A (the most studied ruthenium-based complex on mouse models of tumour metastases [22,23,24,25,26,27]) could be ascribed to the control of the angiogenic process. These studies showed NAMI-A capable of inhibiting angiogenesis in all the test systems employed, namely the CAM model [8], the rabbit cornea model [9] and the Matrigel™ sponges implanted subcutaneously in the mouse of the present study. NO is a key factor that stimulates the migration of endothelial cells during the angiogenesis process [13,28] and the interaction of ruthenium complexes with the nitrosyl ligand has been widely investigated [29,30,31]. All the ruthenium complexes tested in the present study have proven to control the production and release of NO by cells involved in the angiogenic processes facilitating the growth of tumours such as the macrophages and the endothelial cells (these latter here represented by the immortalized EA.hy926 cell line). It could be speculatively said that the anti-angiogenic activity of the complexes of the present study is due to their capacity to bind this small gaseous molecule as already demonstrated with the infrared spectroscopy [9]. In the case of the ruthenium complex NAMI-A, support to the role of NO scavenging for its innovative anti-tumour properties is given by the study of Das and Mondal [32] who have stressed the capacity of the adduct of NAMI-A with serum albumin to efficiently bind the nitrosyls, then immediately to undergo reduction to the more reactive ruthenium(II), a phenomenon that in the mind of these authors would explain the anti-metastatic properties of this drug. However, if the scavenging of extracellular NO can be easily understood, it is questionable that the reduction of the intracellular NO may depend on the same mechanism, given the demonstrated NAMI-A inability to penetrate into the cells [33]. Although capable of scavenging NO similarly to NAMI-A, RuEDTA and KP1339 do not share the same ability of NAMI-A to inhibit the endothelial cell invasion, suggesting that mechanisms others than NO scavenging are likely responsible of this activity, such as the reduced production/activation of MMPs caused by NAMI-A but not by the other ruthenium compounds ([8]; Callerio Foundation Onlus, data on file) It descends that the control of the production/release of NO is not sufficient to claim for anti-angiogenic properties for a given compound that correspondingly cannot be charged with capacities of controlling all the tumour cell activities related to angiogenesis. The anti-angiogenic effects of NAMI-A, here surrogated by the in vitro reduced invasive ability of EA.hy926 endothelial cells, are confirmed by the in vivo Matrigel™ pellets experiment, and further supported by a parallel study with aorta rings cultivated in vitro and exposed to increasing concentrations of the ruthenium-based drug. The results of this study can be summarized as follows: rat aorta rings obtained from healthy adult Wistar male rats, cultivated in vitro at 37 °C in Petri dishes over a Matrigel™ layer, allow the endothelial cells to grow forming elongated chains of cells invading the Matrigel™ structure. The addition of NAMI-A significantly reduces endothelial cell outgrowth from aorta rings in a dose-dependent manner and up to a complete suppression of the phenomenon at 3 × 10−4 M. The model was checked with SNAP (positive control that induces a pronounced increase of endothelial cell growth around the aorta ring) and with l-NAME (significant inhibition of the growth of endothelial cells around the aorta rings) (Callerio Foundation Onlus, data on file). This result confirms what already reported on the anti-angiogenic activity of NAMI-A adding a further model of angiogenesis in which the phenomenon is perturbed by concentrations of NAMI-A attainable with those obtained in vivo in the organs where tumour metastases are formed and grow [12]. The use and the results obtained with the reference standards PTIO, l-NAME and SNAP confirm the adequacy of the experimental protocols used to test the activity of NAMI-A and of the two other ruthenium complexes on the modulation of NO production. As expected, l-NAME reduces and SNAP increases the eNOS expression in the tested cells, as resulting from a series of Western blot analyses (Callerio Foundation Onlus, data on file). Therefore, the effects of NAMI-A and RuEDTA (marked reduction of eNOS expression) have pharmacological consistency. Similarly, the effect of l-NAME on the aorta rings cultivated in vitro (inhibition of the growth of endothelial structures), and those of SNAP (consistent promotion of endothelial growth), strengthen the meaning of the activity observed with NAMI-A on the same model and they provide for this drug evidence of the link between inhibition of eNOS, reduction of NO levels and inhibition of the angiogenesis. In conclusion, even examining the angiogenesis, NAMI-A proves to have unique qualities not shared with other ruthenium complexes, and is also very similar to the compound KP1339. The anti-angiogenic properties of NAMI-A are attributable in part to its capacity to remove the extracellular NO, even when the drug is bound to serum albumin (the greater amount of the drug in the blood after intravenous injection [34]), and in part to its ability to modify the activity of transcription factors responsible for the production of NO, as shown in cells cultured in vitro for 1 h in the presence of anti-metastatic concentrations of the drug [35]. Although these data are not meant to attribute the anti-metastatic properties of NAMI-A solely to one’s ability to exert anti-angiogenic activity, they unequivocally show the strong capacity of this drug to counteract the tumour cell angiogenesis. It is less clear why this property contributes to the reduction of tumour metastases, while it apparently does not apply on the primary site of growth of solid tumours, on which the angiogenesis is similarly important, and on which the effects of NAMI-A have always been quite moderate if not completely null. If it is not excluded that the anti-angiogenic effect of NAMI-A may contribute to the inhibition of the release of metastatic cells from the primary tumour, it is almost surprising that the reduction of angiogenesis at this level does not induce also suffering of tumour cells at this site. Clarifying this aspect will increase the knowledge on the pharmacological properties of NAMI-A and will give a further impulse to the development of this class of innovative metal-based drugs. 4. Materials and Methods 4.1. Compounds and Chemicals NAMI-A, imidazolium trans-imidazoledimethylsulfoxidetetrachlororuthenate(III), ImH [trans-RuCl4(DMSO)(Im)], was prepared according to the published procedure [36]. RuEDTA, potassiumchlorido(ethylendiamminotetraacetate)rutenate(III), K[Ru(HEDTA)Cl], was prepared according to the published procedure [37] and kindly provided by the group of Luigi Messori, University of Florence (Florence, Italy). KP1339, sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2], was kindly provided by Bernhard Keppler, University of Vienna (Vienna, Austria). 2-Phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO), S-nitroso-N-acetyl-dl-penicillamine (SNAP), Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME) and all other reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless otherwise indicated. The structures of the three ruthenium compounds, PTIO, SNAP, and l-NAME, are shown in Scheme 1. 4.2. Murine Peritoneal Macrophages Murine peritoneal macrophages were obtained according to the procedure described by Zhang et al. [38]. To this purpose, adult male Centro di Biotecnologie Avanzate (CBA) mice from an established colony of the animal house of the University of Trieste were used. Animal studies were carried out according to the guidelines in force in Italy (DDL 116 of 21/2/1992 and subsequent addenda) and in compliance with the Guide for the Care and Use of Laboratory Animals, DHHS pub. No. (NIH) 86–23, Bethesda, MD: NIH, 1985. Briefly, donor mice were injected in the peritoneal cavity with 1 mL of a 3% thioglycollate solution in sterile water. After four days, mice were euthanized with CO2, the abdomen of each mouse soaked with 70% alcohol and a small incision carried out along the midline with sterile scissors, then the abdominal skin was retracted to expose the intact peritoneal wall. Five millilitres of cold PBS (Phosphate Buffered Saline pH = 7.4) were injected into the peritoneum of each mouse, and then the fluid was aspirated from the peritoneum and dispensed into a 50 mL centrifuge tube on ice. The peritoneal exudate cells were centrifuged at 400× g for 5 min at 4 °C, and the supernatant discarded and the cell pellet re-suspended in Roswell Park Memorial Insitute culture medium 1640 (RPMI) medium containing 0.25 M Hepes (EuroClone, Devon, UK) and 105 cells/well seeded into a 96-well plate. After 2 h, the supernatant was discarded, cells were washed twice with PBS and macrophages maintained in RPMI culture medium supplemented with 10% foetal bovine serum (FBS, Invitrogen, Paisley, Scotland, UK), 2 mM l-glutamine (EuroClone), 100 IU/mL Penicillin, 100 μg/mL streptomycin (EuroClone), 0.5% gentamicin, 4 × 10−4 M sodium pyruvate (EuroClone), 0.25 M Hepes (EuroClone), and 1% non-essential amino acids (EuroClone). 4.3. EA.hy926 Cell Line The human endothelial-like immortalised cell line EA.hy926, derived from the fusion of human umbelical vein endothelial cells (HUVEC) with the lung carcinoma cell line A549 was kindly provided by Mauro Coluccia, University of Bari, Bari, Italy, and was maintained in Dulbecco’s modified Eagle’s medium (DMEM, EuroClone), supplemented with 10% heat-inactivated FBS, 2 mM l-glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin. The cell line was kept in an incubator with 5% CO2 and 100% relative humidity at 37 °C. Cells from a confluent monolayer were removed from flasks by a trypsin-EDTA solution. Cell viability was determined by the trypan blue dye exclusion test. For experimental purposes, cells were sown in flasks or in multi-well culture clusters. 4.4. Cytotoxicity Test Cells were seeded at 5000 per well on 96-well plates and allowed to grow until they reached a sub-confluence stage. Then, they were incubated for 48 and 72 h with 10−6–10−3 M solutions of NAMI-A, RuEDTA, KP1339, PTIO, SNAP, and l-NAME obtained by serial dilution of a stock solution (freshly prepared in sterile water at a concentration of 2 mM) with complete medium containing 5% FBS. Analysis of cell cytotoxicity by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was performed after 48 or 72 h of incubation. Briefly, MTT dissolved in PBS (5 mg/mL) was added (10 μL per 100 μL medium) to all wells, and the plates were then incubated at 37 °C with 5% CO2 and 100% relative humidity for 4 h. After this time, the medium was discarded and 200 μL of a 10% Igepal® solution in HCl 0.01 N were added to each well for 30 min at 37 °C. Absorbance units were measured at λ = 570 nm on an Automated Microplate Reader EL311s (BIO-TEK® Instruments, Winooski, VT, USA). IC50 values were calculated from dose–effect curves and are the mean ± standard deviation (S.D.) of at least three separate experiments. The fitting procedure applied is a nonlinear regression performed with GraphPad Prism version 6 for Mac OS X version 6.0b (GraphPad Software, San Diego, CA, USA). Experiments were conducted in quadruplicate and repeated three times. 4.5. Invasion Assay Invasion assay was performed using 8.0 μm pore size Transwell® inserts (Costar, Cambridge, MA, USA) coated with Matrigel™ (400 μg/mL, BD, Milano, Italy) at room temperature overnight. EA.hy926 cells were treated with NAMI-A, RuEDTA, SNAP, and l-NAME (10−5, 10−4, and 3 × 10−4 M) in serum-starved medium for 24 h at 37 °C, 5% CO2 before being seeded on inserts (105 cells/insert) in the same medium containing 0.1% bovine serum albumine (BSA). As invasion stimulus, the complete medium was applied in the plate wells, as negative control to detect the basal invasion rate a serum-free medium was used. After 24 h, cells that had remained on the upper side of the membrane were removed using cotton swabs, while the cells that invaded and were present in the lower surface of the inserts were fixed with methanol, stained with May-Grünwald-Giemsa and observed at light microscopy (400×) (Orthoplan, Leitz, Wetzlar, Germany). Cells that have invaded have been counted in 7 fields. The results are expressed as percentage of treated/untreated cells. 4.6. Nitric Oxide Measurements Murine peritoneal macrophages were seeded at 105/well into 96-well plates in complete medium and treated with the compounds and with 10 μg/mL LPS (Lipopolysaccharide), accordingly to literature data reporting that, in the presence of LPS, the production of nitric oxide was found to be induced in macrophages in a time- and dose-dependent manner [16]. The NO production was measured by the Griess test. EA.hy926 cells were seeded at 5000/well into 96-well plates in complete medium and treated with the compounds. The quantification of NO in the extracellular medium was measured by the Griess test while the fluorescent probe DAF-2 DA (Molecular Probes, Eugene, OR, USA) was used to detect NO at intracellular level. 4.7. Griess Test Nitric oxide (NO) was determined on supernatants of cell cultures with Griess reagent according to Stuher and Nathan [39]. Briefly, supernatants of cultures (85 μL) were put in microtitre 96-well plates and added with 5 μL nitrate reductase and 10 μL NADH for 20 min at room temperature (RT), and then with 100 μL Griess reagent (1% naphtylethylendiamine to 1% sulphanilamide, 1:1). After 10 min at RT, the absorbance units were measured at 540 nm. NaNO2 was used as standard. 4.8. DAF-2 DA Fluorescent Probe Diamminofluoresceins are usually used to detect NO in the intracellular milieu. DAF-2 DA has been widely applied to study NO in endothelial cells [40,41]. It enters by diffusion into cells where it is hydrolysed by cytosol esterases releasing the DAF-2 specie, which in the presence of NO and O2 is converted to the fluorescent derivative DAF-2T. The probe was added to the growth medium at 5 μM for 45 min at 37 °C and 5% CO2. The fluorescence (excitation at 492 nm and emission at 515 nm) was measured by a fluorimeter FluoroCount™ (Packard, Milano, Italy). 4.9. Matrigel™ Pellets Angiogenic Test The experiment was carried out at the Centro di Biotecnologie Avanzate (CBA) of the University of Genua, in collaboration with the group of Adriana Albini. C57/BL 6 N male mice were implanted subcutaneously (s.c.) with 600 μL of Matrigel™ added with VEGF (36 ng/pellet), heparin (12 U/pellet), TNF-α (0.72 ng/pellet) and PBS (for the controls) or NAMI-A to obtain a final concentration of 2.4 × 10−4 M. Four days after the implant mice were sacrificed with CO2, and the pellets extracted, photographed and processed for the determination of the haemoglobin content. Briefly, the pellets were put in eppendorf tubes with water and disaggregated mechanically with the aid of scissors. After centrifugation the supernatant was collected and the haemoglobin content determined with a kit purchased from Sigma and based on the method of Drabkin [42]. The procedure is a colorimetric cyanomethaemoglobin method where total haemoglobin at alkaline pH is rapidly converted to the cyanoderivative. The absorbance of the cyanoderivative is determined at 540 nm. The haemoglobin content was then normalized to the weight of the pellets. 4.10. Statistical Analysis Results obtained were processed using InstatGraph3 software (Version 3.0, GraphPad Software Inc., San diego, CA, USA) and presented as mean ± mean ± standard error medium (S.E.M.) The group means were compared using a Two-Way Analysis of Variance (ANOVA) followed by Tukey–Kramer post-test and considered significant when p < 0.05. Acknowledgments The work was funded by Callerio Foundation Onlus. The collaboration with the group of A. Albini at the University of Genua (Genua, Italy) is gratefully appreciated. Author Contributions Anna Castellarin and Gianni Sava conceived and designed the experiments; Anna Castellarin performed the experiments; Anna Castellarin and Alberta Bergamo analyzed the data; Gianni Sava and Sonia Zorzet contributed reagents/materials/analysis tools; and Alberta Bergamo and Gianni Sava wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figures, Scheme and Table Figure 1 Nitric oxide (NO) release in the extracellular medium by macrophages. Murine peritoneal macrophages were treated for 24 h with 10 μg/mL lipopolysaccharide (LPS) and with 3 × 10−4 M 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO), then NO release in the extracellular medium was measured by the Griess test. Data are the mean ± standard error medium (S.E.M.) of five samples per group. Unpaired t-test: *** p < 0.001 vs. no LPS no PTIO; °°° p < 0.001 vs. LPS. Figure 2 Effects of ruthenium compounds on the NO production by macrophages. Murine peritoneal macrophages were treated for 24 h with 10 μg/mL LPS and contemporary with 10−5, 10−4, and 3 × 10−4 M of Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME), imidazolium trans-imidazoledimethylsulfoxidetetrachloro-ruthenate (NAMI-A), and potassiumchlorido(ethylendiamminotetraacetate)rutenate(III) (RuEDTA). NO release in the extracellular medium was measured by the Griess test. Data are expressed as per cent of variation vs. the controls and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: * p < 0.05, *** p < 0.001 vs. Control. Figure 3 Effects of the pre-treatment with the ruthenium compounds on the NO production by macrophages. Murine peritoneal macrophages were treated with 10−4 M NAMI-A, RuEDTA, and sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2] (KP1339), then washed and further treated for 24 h with 10 μg/mL LPS. NO release in the extracellular medium was measured by the Griess test. Data are expressed as per cent of variation vs. the controls and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: *** p < 0.001 vs. Control; °°° p < 0.001 vs. KP1339. Figure 4 EA.hy926 cell viability after treatment with NAMI-A, RuEDTA, KP1339, PTIO, SNAP, and l-NAME. EA.hy926 cells were treated with compounds 10−6–10−3 M for: 48 h (A); or 72 h (B). Cell viability was measured with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test. Figure 5 Effects of NAMI-A, RuEDTA, l-NAME, and SNAP on the invasive ability of EA.hy926 cells. EA.hy926 cells were treated with compounds at 10−5, 10−4, and 3 × 10−4 M for 24 h before seeding them on Transwell® filters. Cell invasion was detected after further 24 h at 37 °C 5% CO2. Data are expressed as per cent of variation vs. the controls and are the mean ± S.E.M. of three samples per group. ANOVA and Tukey–Kramer: ** p < 0.01, *** p < 0.001 vs. Control. Figure 6 NO concentration at extracellular and intracellular level in EA.hy926 cells treated with SNAP. EA.hy926 cells were treated with SNAP 10−6–10−3 M for 2 h. Then, NO was quantified in the extracellular medium by Griess test (A) or at intracellular level by the fluorescent dye DAF-2 DA and expressed as Relative Fluorescence Units (RFU); (B) Data are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: *** p < 0.001 vs. Control. Figure 7 Effects of the contemporary treatment with NAMI-A, Ru-EDTA, KP1339, and PTIO on SNAP induced NO production/release by EA.hy926 cells. EA.hy926 cells were treated with 10−4 M SNAP ± an equimolar concentration of compounds for 2 h. Then, NO was measured in the extracellular medium by Griess test (A) or at intracellular level by the fluorescent dye DAF-2 DA; (B) Data are expressed as per cent of variation vs. SNAP treated cells and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: *** p < 0.001 vs. Control, °°° p < 0.001 vs. KP1339; ^^^ p < 0.001 vs. Ru-EDTA. Figure 8 Concentration-dependent effects of the contemporary treatment with NAMI-A, Ru-EDTA, and PTIO on SNAP induced NO production/release by EA.hy926 cells. EA.hy926 cells were treated with 3 × 10−4 M SNAP ± 10−5 and 10−4 M concentration of compounds for 2 h. Then, NO was measured in the extracellular medium by Griess test (A) or at intracellular level by the fluorescent dye DAF-2 DA; (B) Data are expressed as per cent of variation vs. SNAP treated cells and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: *** p < 0.001 vs. Control. Figure 9 Effects of the pre-treatment with NAMI-A, Ru-EDTA, KP1339, and PTIO on SNAP induced NO production/release by EA.hy926 cells. EA.hy926 cells were pre-treated with 10−4 M NAMI-A, Ru-EDTA, KP1339, and PTIO for 2 h, then, after washing, with an equimolar concentration of SNAP for further 2 h. NO was measured in the extracellular medium by Griess test (A) or at intracellular level by the fluorescent dye DAF-2 DA; (B) Data are expressed as per cent of variation vs. SNAP treated cells and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: ** p < 0.01 vs. Control. Figure 10 Effects of the post-treatment with NAMI-A, Ru-EDTA, and PTIO on SNAP induced NO production/release by EA.hy926 cells. EA.hy926 cells were pre-treated with 10−4 M SNAP for 2 h and then, after washing, with NAMI-A, Ru-EDTA, and PTIO for further 2 h. Then, NO was measured at intracellular level by the fluorescent dye DAF-2 DA. Data are expressed as per cent of variation vs. SNAP treated cells and are the mean ± S.E.M. of five samples per group. ANOVA and Tukey–Kramer: * p < 0.05, ** p < 0.01 vs. Control. Figure 11 Anti-angiogenic activity of NAMI-A in the model of Matrigel™ pellets. The Matrigel™ pellets containing vascular endothelial growth factor (VEGF), heparin, TNF-α and phosphate buffered saline (PBS) (controls), or 2 × 10−4 M NAMI-A (treated) were implanted subcutaneously (s.c.) in C57/BL 6N mice. After four days, the pellets were explanted, photographed (A) and the haemoglobin content measured with the method of Drabkin; (B) Data are the mean ± S.E.M. of three samples. Unpaired t-test: *** p < 0.001 vs. Controls. ijms-17-01254-sch001_Scheme 1Scheme 1 Schematic representation of the chemical structure of trans-imidazoledimethylsulphoxidetetrachloro-ruthenate (NAMI-A), sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2] (KP1339), potassiumchlorido(ethylendiamminotetraacetate)rutenate(III) (RuEDTA), and of 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO), S-nitroso-N-acetyl-dl-penicillamine (SNAP), and Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME). ijms-17-01254-t001_Table 1Table 1 IC50 values of imidazolium trans-imidazoledimethylsulfoxidetetrachloro-ruthenate (NAMI-A), potassiumchlorido(ethylendiamminotetraacetate)rutenate(III) (RuEDTA), sodium (bis-indazole)tetrachloro-ruthenate(III), Na[trans-RuCl4Ind2] (KP1339), 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide (PTIO), S-nitroso-N-acetyl-dl-penicillamine (SNAP), and Nω-nitro-l-arginine methyl ester hydrochloride (l-NAME) in EA.hy926 cells. Compound IC50 [μM] 48 h 72 h NAMI-A 960 (623–1481) 360 (150–885) RuEDTA >1000 880 (507–1534) KP1339 37 (20–68) 22 (12–38) PTIO 87 94 SNAP 500 (85–2893) 220 (85–599) l-NAME >1000 >1000 EA.hy926 cells were treated with NAMI-A, RuEDTA, KP1339, PTIO, SNAP, and l-NAME at 10−6–10−3 M for 48 or 72 h before measuring cell viability by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test. Values in parentheses represent the 95% Confidence Intervals. ==== Refs References 1. Hanahan D. Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis Cell 1996 86 353 364 10.1016/S0092-8674(00)80108-7 8756718 2. Ferrara N. Vascular endothelial growth factor Arterioscler. Thromb. Vasc. Biol. 2009 29 789 791 10.1161/ATVBAHA.108.179663 19164810 3. Giacomini A. Chiodelli P. Matarazzo S. Rusnati M. Presta M. Blocking the FGF/FGFR system as a “two-compartment” antiangiogenic/antitumor approach in cancer therapy Pharmacol. Res. 2016 107 172 185 10.1016/j.phrs.2016.03.024 27013279 4. Zhao Y. Adjei A.A. Targeting angiogenesis in cancer therapy: Moving beyond vascular endothelial growth factor Oncologist 2015 20 660 673 10.1634/theoncologist.2014-0465 26001391 5. Keating G.M. Bevacizumab: A review of its use in advanced cancer Drugs 2014 74 1891 1925 10.1007/s40265-014-0302-9 25315029 6. Rock E.P. Goodman V. Jiang J.X. Mahjoob K. Verbois S.L. Morse D. Dagher R. Justice R. Pazdur R. Food and Drug Administration drug approval summary: Sunitinib malate for the treatment of gastrointestinal stromal tumor and advanced renal cell carcinoma Oncologist 2007 12 107 113 10.1634/theoncologist.12-1-107 17227905 7. Sherbet G.V. Therapeutic potential of thalidomide and its analogues in the treatment of cancer Anticancer Res. 2015 35 5767 5772 26503997 8. Vacca A. Bruno M. Boccarelli A. Coluccia M. Ribatti D. Bergamo A. Garbisa S. Sartor L. Sava G. Inhibition of endothelial cell functions and of angiogenesis by the metastasis inhibitor NAMI-A Br. J. Cancer 2002 86 993 998 10.1038/sj.bjc.6600176 11953835 9. Morbidelli L. Donnini S. Filippi S. Messori L. Piccioli F. Orioli P. Sava G. Ziche M. Anti angiogenic properties of selected ruthenium(III) complexes that are nitric oxide scavengers Br. J. Cancer 2003 88 1484 1491 10.1038/sj.bjc.6600906 12778081 10. Sanna B. Dedidda M. Pintus G. Tadolini B. Posadino A.M. Bennardini F. Sava G. Ventura C. The anti-metastatic agent imidazolium trans- imidazoledimethylsulfoxide-tetrachlororuthenate induces endothelial cell apoptosis by inhibiting the mitogen-activated protein kinase/extracellular signal-regulated kinase signalling pathway Arch. Biochem. Biophys. 2002 403 209 218 10.1016/S0003-9861(02)00218-7 12139970 11. Pintus G. Tadolini B. Posadino A.M. Sanna B. Debidda M. Bennardini F. Sava G. Ventura C. Inhibition of the MEK/ERK signalling pathway by the novel antimetastatic agent NAMI-A down regulates c-myc gene expression and endothelial cell proliferation Eur. J. Biochem. 2002 269 5861 5870 10.1046/j.1432-1033.2002.03307.x 12444974 12. Cocchietto M. Sava G. Blood concentration and toxicity of the antimetastasis agent NAMI-A following repeated intravenous treatment in mice Pharmacol. Toxicol. 2000 87 193 197 10.1034/j.1600-0773.2000.d01-73.x 11129497 13. Ziche M. Morbidelli L. Masini B. Amerini S. Granger H.J. Maggi C.A. Geppetti P. Ledda F. Nitric oxide mediates angiogenesis in vivo and endothelial cell growth and migration in vitro promoted by substance P J. Clin. Investig. 1994 94 2036 2044 10.1172/JCI117557 7525653 14. Ziche M. Morbidelli L. Choudhuri R. Zhang H.T. Donnini S. Granger H.J. Bicknell R. Nitric oxide synthase lies downstream from vascular endothelial growth factor-induced but not basic fibroblast growth factor-induced angiogenesis J. Clin. Investig. 1997 99 2625 2634 10.1172/JCI119451 9169492 15. Parenti A. Morbidelli L. Cui X.L. Douglas J.G. Hood J. Granger H.J. Ledda F. Ziche M. Nitric oxide is an upstream signal for vascular endothelial growth factor-induced extracellular signal-regulated kinases1 /2 activation in postcapillary endothelium J. Biol. Chem. 1998 273 4220 4226 10.1074/jbc.273.7.4220 9461619 16. Krol W. Czuba Z.P. Threadgill M.D. Cunningham B.D.M. Pietsz G. Inhibition of nitric oxide (NO·) production in murine macrophages by flavones Biochem. Parmacol. 1995 50 1031 1035 10.1016/0006-2952(95)00237-T 17. Edgell C.J. McDonald C.C. Graham J.B. Permanent cell line expressing human factor VIII-related antigen established by hybridisation Proc. Natl. Acad. Sci. USA 1983 80 3734 3737 10.1073/pnas.80.12.3734 6407019 18. Keppler B.K. Henn M. Juhl U.M. Berger M.R. Niebl R. Wagner F.E. New ruthenium complexes for the treatment of cancer Progress in Clinical Biochemistry and Medicine Springer Berlin, Germany 1989 Volume 10 41 69 19. Sava G. Pacor S. Zorzet S. Alessio E. Mestroni G. Antitumour properties of dimethylsulphoxide ruthenium(II) complexes in the Lewis Lung Carcinoma system Pharmacol. Res. 1989 21 617 628 10.1016/1043-6618(89)90204-1 2594617 20. Sava G. Pacor S. Mestroni G. Alessio E. Effects of the Ru(III) complexes [mer-RuCl3 (DMSO)2Im][degrees] and Na[trans -RuCl4(DMSO)Im] on solid mouse tumors Anti-Cancer Drugs 1992 3 25 31 10.1097/00001813-199202000-00005 1623212 21. Sava G. Pacor S. Bergamo A. Cocchietto M. Mestroni G. Alessio E. Effects of ruthenium complexes on experimental tumours: Irrelevance of cytotoxicity for metastasis inhibition Chem. Biol. Interact. 1995 95 109 126 10.1016/0009-2797(94)03350-1 7697744 22. Sava G. Clerici K. Capozzi I. Cocchietto M. Gagliardi R. Alessio E. Mestroni G. Perbellini A. Reduction of lung metastasis by ImH[trans -RuCl4 (DMSO)Im]: Mechanism of the selective action investigated on mouse tumors Anti-Cancer Drugs 1999 10 129 138 10.1097/00001813-199901000-00016 10194556 23. Sava G. Capozzi I. Clerici K. Gagliardi R. Alessio E. Mestroni G. Pharmacological control of lung metastases of solid tumours by a novel ruthenium complex Clin. Exp. Metastasis 1998 16 371 379 10.1023/A:1006521715400 9626816 24. Sava G. Zorzet S. Turrin C. Vita F. Soranzo M.R. Zabucchi G. Cocchietto M. Bergamo A. DiGiovine S. Pezzoni G. Dual action of NAMI-A in inhibition of solid tumor metastasis: Selective targeting of metastatic cells and binding to collagen Clin. Cancer Res. 2003 9 1898 1905 12738748 25. Coluccia M. Sava G. Salerno G. Bergamo A. Pacor S. Mestroni G. Alessio E. Efficacy of 5-FU combined to Na[trans -RuCl4 (DMSO)Im], a novel selective antimetastatic agent, on the survival time of mice with p388 leukemia, P388/DDP subline and MCa mammary carcinoma Metal-Based Drugs 1995 2 195 199 10.1155/MBD.1995.195 18472766 26. Zorzet S. Bergamo A. Cocchietto M. Sorc A. Gava B. Alessio E. Iengo E. Sava G. Lack of in vitro cytotoxicity, associated to increased G2 -M cell fraction and inhibition of matrigel invasion, may predict in vivo-selective antimetastasis activity of ruthenium complexes J. Pharmacol. Exp. Ther. 2000 295 927 933 11082425 27. Gava B. Zorzet S. Spessotto P. Cocchietto M. Sava G. Inhibition of B16 melanoma metastases with the ruthenium complex imidazolium trans -imidazoledimethylsulfoxide-tetrachlororuthenate and down-regulation of tumor cell invasion J. Pharmacol. Exp. Ther. 2006 317 284 291 10.1124/jpet.105.095141 16368900 28. Goligorsky M.S. Budzikowski A.S. Tsukahara H. Noiri E. Co-operation between endothelin and nitric oxide in promoting endothelial cell migration and angiogenesis Clin. Exp. Pharmacol. Physiol. 1999 26 269 271 10.1046/j.1440-1681.1999.03029.x 10081626 29. Franke A. Oszajca M. Brindell M. Stochel G. van Eldik R. Metal-assisted activation of nitric oxide-mechanistic aspecte of complex nitrosylation processes Adv. Inorg. Chem. 2015 67 171 241 30. Oszajca M. Kulis E. Stochel G. Brindell M. Interaction of the NAMI-A complex with nitric oxide under physiological conditions New J. Chem. 2014 38 3386 3394 10.1039/C3NJ01631E 31. Bucinsky L. Büchel G.E. Ponec R. Rapta P. Breza M. Kozisek J. Gall M. Biskupic S. Fronc M. Schiessl K. On the electronic structure of mer ,trans -[RuCl3 (1H -indazole)2 (NO)], a hypothetical metabolite of the antitumor drug candidate KP1019: An experimental and DFT study Eur. J. Inorg. Chem. 2013 2013 2505 2519 10.1002/ejic.201201526 32. Das D. Mondal P. Quantum chemical studies on detail mechanism of nitrosylation of NAMI-A-HSA adduct J. Phys. Chem. 2015 119 10456 10465 10.1021/acs.jpcb.5b05071 26151453 33. Aitken J.B. Antony S. Weekley C.M. Lai B. Spiccia L. Harris H.H. Distinct cellular fates for KP1019 and NAMI-A determined by X-ray fluorescence imaging of single cells Metallomics 2012 4 1051 1056 10.1039/c2mt20072d 22907648 34. Rademaker-Lakhai J.M. van den Bongard D. Pluim D. Beijnen J.H. Schellens J.H. A Phase I and pharmacological study with imidazolium-trans -DMSO-imidazole-tetrachlororuthenate, a novel ruthenium anticancer agent Clin. Cancer Res. 2004 10 3717 3727 10.1158/1078-0432.CCR-03-0746 15173078 35. Bergamo A. Gerdol M. Lucafò M. Pelillo C. Battaglia M. Pallavicini A. Sava G. RNA-seq analysis of the whole transcriptome of MDA-MB-231 mammary carcinoma cells exposed to the antimetastatic drug NAMI-A Metallomics 2015 7 1439 1450 10.1039/C5MT00081E 26338673 36. Mestroni G. Alessio E. Sava G. New Salt of Anionic Complexes of Ru(III) as Antimetastatic and Antineoplastic Agents International Patent WO 98/00431 1 8 1998 37. Diamantis A.A. Dubrawsky J.V. Preparation and structure of ethylenediaminetetraacetate complexes of ruthenium(II) with dinitrogen, carbon monoxide, and other π-acceptor ligands Inorg. Chem. 1981 20 1142 1150 10.1021/ic50218a037 38. Zhang X. Goncalves R. Mosser D.M. The isolation and characterization of murine macrophages Curr. Protoc. Immunol. 2008 10.1002/0471142735.im1401s111 39. Stuehr D.J. Nathan C.F. Nitric oxide. A macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells J. Exp. Med. 1989 169 1543 1555 10.1084/jem.169.5.1543 2497225 40. Itoh Y. Ma F.H. Hoshi H. Oka M. Noda K. Ukai Y. Kojima H. Nagano T. Toda N. Determination and bioimaging method for nitric oxide in biological specimens by diaminofluorescein fluorimetry Anal. Biochem. 2000 287 203 209 11112265 41. Nakatsubo N. Kojima H. Kikuchi K. Nagoshi H. Hirata Y. Maeda D. Imai Y. Irimura T. Nagano T. Direct evidence of nitric oxide production from bovine aortic endothelial cells using new fluorescence indicators: Diaminofluoresceins FEBS Lett. 1998 427 263 266 10.1016/S0014-5793(98)00440-2 9607324 42. Drabkin D.L. Austin J.H. Spectrophotometric studies II. Preparations from washed blood cells; nitric oxide haemoglobin and sulfhemoglobin J. Biol. Chem. 1935 112 51 55

PMC005xxxxxx/PMC5000653.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081255ijms-17-01255ArticleSerum Interleukin-18, Fetuin-A, Soluble Intercellular Adhesion Molecule-1, and Endothelin-1 in Ankylosing Spondylitis, Psoriatic Arthritis, and SAPHO Syndrome Przepiera-Będzak Hanna 1*Fischer Katarzyna 2Brzosko Marek 1Moudgil Kamal D. Academic Editor1 Department of Rheumatology, Internal Medicine and Geriatrics, Pomeranian Medical University in Szczecin, Unii Lubelskiej 1, Szczecin 71-252, Poland; [email protected] Independent Laboratory of Rheumatic Diagnostics, Pomeranian Medical University in Szczecin, Unii Lubelskiej 1, Szczecin 71-252, Poland; [email protected]* Correspondence: [email protected]; Tel.: +48-914-253-321; Fax: +48-914-253-34403 8 2016 8 2016 17 8 125501 7 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).To examine serum interleukin 18 (IL-18), fetuin-A, soluble intercellular adhesion molecule-1 (sICAM-1), and endothelin-1 (ET-1) levels in ankylosing spondylitis (AS), psoriatic arthritis (PsA), and Synovitis Acne Pustulosis Hyperostosis Osteitis syndrome (SAPHO). We studied 81 AS, 76 PsA, and 34 SAPHO patients. We measured serum IL-18, fetuin-A, sICAM-1, ET-1, IL-6, IL-23, vascular endothelial growth factor (VEGF), and epidermal growth factor (EGF). IL-18 levels were higher in AS (p = 0.001), PsA (p = 0.0003), and SAPHO (p = 0.01) than in controls, and were positively correlated with CRP (p = 0.03), VEGF (p = 0.03), and total cholesterol (TC, p = 0.006) in AS and with IL-6 (p = 0.03) in PsA. Serum fetuin-A levels were lower in AS (p = 0.001) and PsA (p = 0.001) than in controls, and negatively correlated with C-reactive protein (CRP) in AS (p = 0.04) and SAPHO (p = 0.03). sICAM-1 positively correlated with CRP (p = 0.01), erythrocyte sedimentation rate (ESR, p = 0.01), and IL-6 (p = 0.008) in AS, and with IL-6 (p = 0.001) in SAPHO. Serum ET-1 levels were lower in AS (p = 0.0005) than in controls. ET-1 positively correlated with ESR (p = 0.04) and Disease Activity Score 28 (DAS28, p = 0.003) in PsA. In spondyloarthritis, markers of endothelial function correlated with disease activity and TC. interleukin 18fetuin-Asoluble intercellular adhesion molecule-1endothelin-1ankylosing spondylitispsoriatic arthritisSAPHO ==== Body 1. Introduction There is evidence of an increased risk of atherosclerosis and cardiovascular disease (CVD) in inflammatory rheumatic diseases [1,2,3,4,5,6,7,8,9]. Ankylosing spondylitis (AS), psoriatic arthritis (PsA), and SAPHO syndrome (SAPHO) are seronegative spondyloarthropathies (SpA) [10,11,12]. Inflammation in the course of arthritis, as well as traditional risk factors of atherosclerosis together with cytokines and adhesion molecules, influence endothelial activation and dysfunction. Interleukin 18 (IL-18), fetuin-A, soluble intercellular adhesion molecule-1 (sICAM-1), and endothelin-1 (ET-1) are a group of cytokines and adhesion molecules which activate and deregulate endothelial function, and could increase the risk of atherosclerosis [13,14,15,16,17,18,19,20]. Serum interleukin 6 (IL-6) and interleukin 23 (IL-23) are considered to be associated with the inflammatory process in spondyloarthritis [21,22]. Additionally, IL-6 stimulates the hepatic production of fibrinogen and is also involved in endothelial activation [22]. Angiogenesis, stimulated by vascular endothelial growth factor (VEGF) and epidermal growth factor (EGF), also plays a role in the pathogenesis of SpA [23,24,25,26]. The aim of the study was to examine serum IL-18, fetuin-A, sICAM-1, and ET-1 levels as markers of endothelial function in correlation with disease activity and the lipid profile in AS, PsA, and SAPHO. 2. Results The clinical, laboratory characteristics and treatment of the patients and controls are presented in Table 1. Serum IL-18 levels were higher in SpA, AS, PsA, and SAPHO patients than in controls (Table 2, Figure 1). Serum IL-18 levels were higher in male patients than in female patients with PsA (347.1 (269.8–429.6) vs. 269.6 (200.5–306.4) pg/mL, p = 0.003) and in controls (227.0 (206.9–273.1) vs. 190.0 (160.3–248.8) pg/mL, p = 0.05). No differences in serum IL-18 levels were found between the AS, PsA, and SAPHO groups (p > 0.05). SpA patients had an increased risk of increased serum IL-18 (Table 3) compared to controls. Serum fetuin-A levels were lower in SpA, AS, and PsA patients than in controls. (Table 2, Figure 2). In AS patients, serum fetuin-A levels were lower in males than in females (577.2 ± 138.1 vs. 673.7 ± 156.1 µg/mL, p = 0.009). No differences were found between female and male patients in terms of serum fetuin-A levels in the PsA, SAPHO, and control groups (all p > 0.05). No differences were observed in serum fetuin-A levels between the AS, PsA, and SAPHO groups (p > 0.05). SpA patients had an increased risk of decreased serum fetuin-A compared to controls (Table 3). No differences were found between SpA patients and controls in terms of sICAM-1 levels (Table 2). No differences were found between females and males in terms of sICAM-1 in SpA patients and in controls (all p > 0.05). No differences were found in serum sICAM-1 levels between the AS, PsA, and SAPHO groups (all p > 0.05). SpA patients, compared to controls, had an increased risk of increased serum sICAM-1 levels (Table 3). Serum ET-1 levels were lower in SpA and AS patients than in controls. (Table 2, Figure 3). Serum ET-1 levels were lower in AS than in SAPHO patients (p = 0.0003), and in PsA than in SAPHO patients (p = 0.05, Table 2). No differences were found between female and male patients in terms of serum ET-1 levels in SpA patients and in controls (all p > 0.05). SpA patients had an increased risk of decreased serum ET-1 levels compared to controls (Table 3). There was a positive correlation between serum IL-18 and sICAM-1 in SpA (r = 0.3; p = 0.00001), AS (r = 0.22; p = 0.04), and PsA (r = 0.43; p = 0.0001) patients. No correlations were found between IL-18 and fetuin-A or ET-1. In AS patients, sICAM-1 was positively correlated with ET-1. No correlations were found between sICAM-1 and fetuin-A in SpA. 2.1. Markers of Endothelial Function and Disease Activity in SpA There was a positive correlation between serum IL-18 and CRP (r = 0.24; p = 0.03) in AS patients. There was a positive correlation between serum IL-18 with IL-6 (r = 0.27; p = 0.03) and with BASDAI (r = 0.24; p = 0.03) in PsA patients. No correlations were found between serum IL-18 and disease activity in the SAPHO group. The assessment of the age–sex adjusted p-value by ANOVA confirmed the positive correlation between serum IL-18 and VEGF in SpA (r = 0.28; p = 0.004) and AS (r = 0.33; p = 0.03) patients. No correlations were found between serum IL-18 and VAS, ASDAS-ESR, DAS28, IL-23, EGF, or ERS (all p > 0.05) in SpA patients. There was a negative correlation between serum fetuin-A and CRP in AS (r = −0.23; p = 0.04) and SAPHO (r = −0.38; p = 0.03) patients. SAPHO patients with CRP ≥ 5 mg/L had lower fetuin-A compared to those with CRP < 5 mg/L (547.1 ± 143.3 vs. 690.0 ± 188.3 pg/mL, p = 0.02). There was a positive correlation between fetuin-A and VEGF in SpA (r = 0.32; p = 0.001) and AS (r = 0.4; p = 0.007) patients. The assessment of the age–sex adjusted p-value by ANOVA for SpA patients with CRP ≥ 5 mg/L confirmed the positive correlation between serum fetuin-A and CRP (r = 0.22; p = 0.002). Additionally, in SpA patients with VAS > 40, there was a positive correlation between serum fetuin-A and VAS (r = 0.21; p = 0.004). There was positive correlation between serum fetuin-A and IL-6 in SpA (r = 0.21; p = 0.01) and in AS (r = 0.29; p = 0.02) patients. There was a positive correlation between serum fetuin-A and IL-23 in SAPHO patients (r = 0.5; p = 0.01). No correlations were found between serum fetuin-A and BASDAI, ASDAS-ESR, DAS28, EGF, or ERS (all p > 0.05) in SpA patients. There was a positive correlation between serum sICAM-1 and IL-6 in AS (r = 0.34; p = 0.008) and SAPHO (r = 0.5; p = 0.001) patients. In AS patients, there was a positive correlation between serum sICAM-1 and CRP (r = 0.28; p = 0.01), ESR (r = 0.29; p = 0.01), and age (r = 0.2; p = 0.006). AS patients with CRP ≥ 5 mg/L, compared with AS patients with CRP < 5 mg/L, had higher serum sICAM-1 (238.9 ± 62.7 vs. 208.2 ± 40.4 ng/mL, p = 0.02). No correlations were found between sICAM-1 and VAS, BASDAI, ASDAS-ESR, DAS28, IL-23, VEGF, or EGF (all p > 0.05) in SpA patients. In PsA patients, serum ET-1 was positively correlated with ESR (r = 0.24; p = 0.04) and DAS28 (r = 0.7; p = 0.003). SpA patients with ESR ≥ 10 had higher levels of ET-1 compared to those with ESR < 10 (1.32 ± 0.6 vs. 1.11 ± 0.4 pg/mL, p = 0.02). There was a negative correlation between ET-1 and EGF in AS (r = −0.31; p = 0.04) and in SAPHO (r = −0.83; p = 0.0002) patients. The assessment of the age–sex adjusted p-value by ANOVA confirmed the positive correlation between serum ET-1 and IL-6 in SpA (r = 0.25; p = 0.001) and AS (r = 0.27; p = 0.03) patients. No correlations were found between ET-1 and VAS, BASDAI, ASDAS-ESR, IL-23, VEGF, or CRP (all p > 0.05) in SpA patients. SpA patients with a BASDAI score > 4 and those with a score ≤ 4 did not differ in terms of their levels of IL-18, fetuin-A, sICAM-1, and ET-1 (all p > 0.05) SpA patients with a VAS score > 40 and those with a score ≤ 40 did not differ in terms of their levels of IL-18, sICAM-1, and ET-1 (all p > 0.05). However, SpA patients with a VAS score > 40 had lower levels of fetuin-A (584.9 ± 144.4 vs. 664.9 ± 171.9 pg/mL, p = 0.005) compared to those with a VAS score ≤ 40. SAPHO patients with a VAS score > 40 had lower levels of fetuin-A (573.2 ± 124.0 vs. 800.8 ± 214.5 pg/mL, p = 0.005) compared to those with a VAS score ≤ 40. 2.2. Markers of Endothelial Function and HLA-B27 in SpA AS patients positive for HLA-B27 had higher serum sICAM-1 compared to those negative for HLA-B27 (228.7 ± 70.9 vs. 158.1 ± 34.3 ng/mL, p = 0.02) and ET-1 (1.07 ± 0.31 vs. 0.67 ± 0.22 pg/mL, p = 0.03). PsA patients positive for HLA-B27 had higher serum sICAM-1 compared to those negative for HLA-B27 (281.7 ± 74.8 vs. 212.4 ± 46.3 ng/mL, p = 0.04). No significant associations between HLA-B27 positivity and serum IL-18 and fetuin-A levels were found. 2.3. Markers of Endothelial Function and BMI and WHR in SpA SpA patients had a higher mean BMI than control subjects (p = 0.006). No differences in BMI were observed between AS, PsA, and SAPHO patients (p > 0.05, Table 1). There was a positive correlation between serum IL-18 and BMI in AS (r = 0.28; p = 0.02) and SAPHO (r = 0.53; p = 0.01) patients. AS patients with BMI > 25.0 had higher IL-18 levels compared to those with BMI ≤ 25.0 (271.9 (228.7–454.5) vs. 245.8 (187.6–323.5) pg/mL, p = 0.04). SAPHO patients with BMI > 25.0 had higher IL-18 levels compared to those with BMI ≤ 25.0 (295.29 (210.2–428.4) vs. 221.4 (155.5–240.8) pg/mL, p = 0.04). There was a positive correlation between IL-18 and WHR in AS patients (r = 0.26; p = 0.04). AS patients with an increased WHR (for female ≥ 0.8, for male ≥ 1.0) had higher IL-18 levels compared to those with a normal WHR (402.7 (267.8–486.1) vs. 245.8 (205.3–357.3) pg/mL, p = 0.02). There was a positive correlation between sICAM-1 and BMI in AS patients (r = 0.21; p = 0.01). AS patients with BMI > 30 had higher sICAM-1 compared to AS patients with BMI < 25 (256.5 ± 41.4 vs. 219.8 ng/mL, p = 0.01). AS patients with BMI > 25 had higher ET-1 compared to those with BMI ≤ 25 (1.22 ± 0.35 vs. 1.04 ± 0.44 pg/mL, p = 0.04). No correlations were found between serum fetuin-A or ET-1 with BMI or WHR in SpA patients (all p > 0.05). 2.4. Markers of Endothelial Function and Lipid Profile in SpA In AS patients, TC levels were lower than in controls (p = 0.001). In AS patients, LDL-C levels were lower than in controls (p = 0.02, Table 1). There was a positive correlation between serum TC and IL-18 (r = 0.36; p = 0.006) and fetuin-A (r = 0.38; p = 0.004) in AS patients. Additionally, in AS patients, there was a positive correlation between fetuin-A and LDL-C (r = 0.34; p = 0.01) and TG (r = 0.38; p = 0.004). No correlations were found between the lipid profile and sICAM-1 or ET-1 in SpA patients (all p > 0.05). 2.5. Markers of Endothelial Function, Treatment, and Comorbidities in SpA The results of the univariate and multivariate logistic regression analysis and step-wise analysis of serum IL-18, fetuin-A, sICAM-1, and ET-1 in terms of treatment with NSAIDs, NSAIDs with sulfasalazine, and NSAIDs with methotrexate showed no differences between different treatment groups in SpA patients (all p > 0.05). AS patients with AAU had higher levels of IL-18 compared to those without AAU (323.5 (228.7–454.5) vs. 251.9 (201.8–324.7) pg/mL, p = 0.03). The results of the univariate and multivariable logistic regression analysis and step-wise analysis of serum IL-18, fetuin-A, sICAM-1, and ET-1 levels in SpA patients who were smokers and non-smokers and with and without comorbidities such as IHD, hypertension, and diabetes showed no differences (all p > 0.05). 3. Discussion Patients with chronic inflammatory arthritis such as AS, PsA, and SAPHO are to prone to the early development and acceleration of atherosclerosis [1,2,3,4,5,6,7,8,9]. The pathogenesis of atherosclerosis is connected to dyslipidemia and markers of inflammation (e.g., CRP and IL-6), as well as other cytokines and adhesion molecules which affect endothelial cell activation [3,4,15]. It has been shown that in patients with AS, a high serum IL-6 level is associated with a higher risk of atherosclerosis [2,22]. Sparse data are available regarding serum IL-18, fetuin-A, sICAM-1, and ET-1 in SpA patients, as these markers are mainly considered in AS [27,28,29,30,31,32,33,34,35]. Expression of IL-18 has been found in atherosclerotic plaques, localized mainly in plaque macrophages [13]. This cytokine is associated with atherosclerosis, and is a strong predictor of cardiovascular death in stable and unstable angina [14]. In our study, serum IL-18 was higher in SpA patients than in control subjects; additionally, SpA patients had an increased risk of increased serum IL-18. The same was found by Sari et al. [30] in AS and by Hurdado-Nedelec et al. [29] in SAPHO and PsA. Rooney et al. [28] observed increased synovial tissue IL-18 in PsA, reactive arthritis, and seronegative patients, before treatment was started. These authors found that after treatment, tissue IL-18 expression was changed, and they concluded that IL-18 plays a role in the pathophysiology of inflammatory arthritis [28]. On the other hand, Surdacki et al. [31], in a group of AS patients smaller than that in our study, found no differences in serum IL-18 levels between AS patients and controls. We found no differences in serum IL-18 levels between the AS, PsA, and SAPHO groups. Additionally, serum IL-18 levels were higher in male than in female SpA patients, and correlated positively with markers of inflammation, such as IL-6 in PsA and CRP in AS. Serum IL-18 correlated positively with TC in AS; this is a well-known marker of the risk of cardiovascular disease. The same was shown by Sari et al. [30]. However, there was no correlation of IL-18 with LDL-C and HDL-C in SpA patients. The results of our study suggest that active inflammation in the course of SpA is connected with an increased IL-18 level, which stimulates endothelial dysfunction and may increase the risk of atherosclerosis in this group. Fetuin-A plays a role in physiological and pathological mineralization [20]. Fetuin-A has an established role in extracellular matrix mineralization. In vitro, fetuin-A protects smooth muscle cells from calcification [16]. Animal studies have suggested that fetuin-A may act as an ectopic calcification inhibitor [18]. The small number of studies concerning the role of fetuin-A in the pathogenesis of endothelial dysfunction in SpA is the reason that the problem is not well recognized. In our study, serum fetuin-A levels were lower in SpA patients than in controls. Additionally, compared to controls, SpA patients had an increased risk of decreased serum fetuin-A levels. These results are different from the results presented by Sari et al. [30], who found increased serum levels of fetuin-A in AS compared to controls. In another study of patients with AS, Tuylu et al. [32] found higher levels of fetuin-A in patients with syndesmophytes compared to AS patients without syndesmophytes and controls, but these levels were lower in patients without syndesmophytes compared to controls. There were some data showing that patients with AS had elevated levels of VEGF, which was associated with disease progression [25,26]. We found a positive correlation between fetuin-A and VEGF, which confirms the role of fetuin-A in disease progression caused by new bone formation. We can explain our results in two ways. On one hand, new bone formation—which is an intrinsic element of SpA—could reduce the concentration of serum fetuin-A in our patients. On the other hand, the decreased serum fetuin-A levels in SpA provide less of a protective effect on smooth muscle cell calcification, promoting endothelial cell dysfunction, and can increase the risk of atherosclerosis. We have also shown a negative correlation between serum fetuin-A and CRP. This confirms our hypothesis that increased disease activity is associated with greater new bone formation, and thereby results in decreased concentrations of serum fetuin-A. In our opinion, decreased fetuin-A in SpA patients could result in a decrease in its protective role in ectopic calcification, and in that way could increase the risk of endothelial dysfunction in SpA. In our study, SpA patients had higher levels of sICAM-1 than controls, but this was not statistically significant. We found a positive correlation between serum sICAM-1 and IL-6 and ESR in SpA. Some studies have shown high sICAM-1 levels in the serum of patients with AS [2,31]. Wendling et al. [34] found a positive correlation between sICAM-1 and ESR, CRP, and IL-6 in SpA patients. Therefore, we can conclude that, in SpA serum, sICAM-1 correlates with disease activity and may have an impact on endothelial dysfunction. ET-1 is a pro-inflammatory peptide that is mainly produced in endothelial cells, and contributes to several pathological events, including inflammation, fibrosis, and cardiac and vascular hypertrophy. The upregulation of ET-1 is regarded as an important pathogenic factor in the development of cardiovascular diseases [17]. In our study, serum ET-1 levels were lower in SpA patients than in controls. The same was found by Sari et al. [35] in AS patients. We could not find other studies that assessed ET-1 in SpA. We observed no influence of cigarette smoking and comorbidities such as IHD, hypertension, or diabetes on serum IL-18, fetuin-A, sICAM-1, and ET-1 levels in SpA patients. This confirms that SpA itself, not comorbidities, affected the expression of markers of endothelial dysfunction in patients with SpA in our study. In our study, the BMI positively correlated with serum IL-18 and sICAM-1 in AS and SAPHO patients. This suggests that overweight could be connected with IL-18 and sICAM-1 production in SpA patients. 4. Materials and Methods This study was approved by the Ethics Committee of the Pomeranian Medical University in Szczecin (KB-0012/106/10; 27SEP2010). Informed consent was obtained from all patients. We studied 191 SpA, caucasian patients: 81 had AS, 76 had PsA, and 34 had SAPHO. The controls were 30 healthy volunteers, matched to the patients according to age and sex. The diagnosis of AS was made according to modified New York criteria, the diagnosis of PsA according to the Caspar classification criteria, and the diagnosis of SAPHO according to the Kahn criteria [10,11,12]. The following data were recorded: age, sex, disease duration, peripheral joint involvement, acute anterior uveitis (AAU), inflammatory bowel disease (IBD), psoriasis, comorbidities (hypertension, diabetes, and ischemic heart disease), cigarette smoking, and current use of medication. Weight and height were measured to calculate the BMI (kg/m2). The waist and hip circumference were measured to calculate the waist/hip ratio (WHR). Skin changes were assessed according to the Psoriasis Area and Severity Index (PASI) [36]. The patient’s pain due to the disease was assessed by a visual analogue scale (VAS). We assessed the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI). We regarded patients as active if the BASDAI score was >4 [37]. In patients with axial disease, the Ankylosing Spondylitis Disease Activity Score (ASDAS) was assessed using erythrocyte sedimentation rate (ESR) (ASDAS-ESR), calculated using the online calculator available at the Assessment of SpondyloArthritis International Society website. Disease activity score calculation in patients with peripheral arthritis was performed using a free online Disease Activity Score 28 (DAS28) calculator. Serum was stored at −80 °C until analysis for IL-18, fetuin-A, sICAM-1, ET-1, IL-6, IL-23, VEGF, and EGF using a sensitive sandwich ELISA method: Human IL-18 Quantitative ELISA kit (MBL, Nagoya, Japan), Human Fetuin-A Immunoassay Quantikine® ELISA kit, Human sICAM-1 Immunoassay Quantikine® ELISA kit, Human ET-1 Immunoassay Quantikine® ELISA kit, Human IL-6 Immunoassay Quantikine® ELISA kit, Human IL-23 Immunoassay Quantikine® ELISA kit, Human VEGF Immunoassay Quantikine® ELISA kit, and Human EGF Immunoassay Quantikine® ELISA kit (R&D Systems, Minneapolis, MN, USA). All analyses and calibrations were performed in duplicate and read using a BioTek PowerWaveXS spectrophotometer (Winooski, VT, USA). Blood was taken after at least 8 h of fasting for the assessment of the erythrocyte sedimentation rate (ESR, mm/h, Westergren method), C-reactive protein (CRP, mg/dL) (turbimetric nephelometry, rate reaction), total cholesterol (TC, mmol/L), high-density lipoprotein cholesterol (HDL-C, mmol/L), low-density lipoprorotein cholesterol (LDL-C, mmol/L), and triglycerides (TG, mmol/L), measured according to standard procedures. Human leukocyte antigen B27 (HLA-B27) was determined using a BD Biosciences test (Becton, Dickinson and Company BD Biosciences, San Jose, CA, USA) based on flow cytometry and a BD FACSCanto II apparatus. Data distributions were assessed using the Kolmogorov–Smirnov test. Data are described as mean ± standard deviation and median (Q1, Q3). We used Spearman’s rank test to calculate correlations. The R values of correlations were determined and corresponding p-values <0.05 were considered significant. The groups were compared using Student’s t-test, the Mann–Whitney U test, or the Kruskal–Wallis test. To assess parameters associated with serum fetuin-A, sICAM-1, IL-18, and ET-1 levels, Pearson’s χ-squared test (χ2), logistic regression analysis, and step-wise analysis were performed. ANOVA was performed, controlling for age and sex. The level of significance was set at p < 0.05. The statistical analysis was performed using STATISTICA version 8.0, StatSoft Inc., Tulsa, OK, USA. 5. Conclusions In SpA patients, impaired endothelial function was found and correlated with disease activity and the lipid profile. The novelty of this work is a comprehensive assessment of a number of markers of endothelial function in combination with markers of disease activity, the lipid profile, comorbidities, and treatment in SpA. Acknowledgments This work was supported by a grant from the National Science Centre in Poland (DEC-2011/03/B/NZ5/04192). Author Contributions Hanna Przepiera-Będzak participated in the design and coordination of the study, performed the statistical analysis and drafted the manuscript. Katarzyna Fischer carried out the immunoassays. Marek Brzosko participated in the design and coordination of the study and helped to draft the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Serum concentrations of interleukin-18 (IL-18) in patients with spondyloarthritis (SpA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), SAPHO syndrome (SAPHO), and controls. Figure 2 Serum concentrations of fetuin-A in patients with spondyloarthritis (SpA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), SAPHO syndrome (SAPHO), and controls. Figure 3 Serum concentrations of endothelin-1 (ET-1) in patients with spondyloarthritis (SpA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), SAPHO syndrome (SAPHO), and controls. ijms-17-01255-t001_Table 1Table 1 Clinical, laboratory characteristics and treatment of the patients and controls. Assessed Parameter SpA Patients (n = 191) Mean ± SD Median (Q1, Q3) AS Patients (n = 81) Mean ± SD Median (Q1, Q3) PsA Patients (n = 76) Mean ± SD Median (Q1, Q3) SAPHO Syndrome Patients (n = 34) Mean ± SD Median (Q1, Q3) Controls (n = 30) Mean ± SD Median (Q1, Q3) Age (years) 48.3 ± 13.1 44.7 ± 13.2 50.8 ± 12.7 51.7 ± 12.0 43.5 ± 9.4 Sex F. 94, M. 97 F. 20, M. 61 F. 43, M. 33 F. 31, M. 3 F. 19, M. 11 Disease duration (years) 5.0 (2.0, 10.0) 10.0 (5.0, 15.0) 4.0 (2.0, 8.0) 2.0 (1.0, 5.0) 0 Peripheral arthritis, n (%) 97 (50.8%) 31 (38.3%) 61 (80.3%) 5 (14.7%) 0 AAU, n (%) 27 (14.1%) 27 (33.3%) 0 (0%) 0 (0%) 0 IBD, n (%) 8 (4.2%) 8 (9.9%) 0 0 0 VAS pain (mm) 51.4 ± 24.9 59.4 ± 26.8 44.9 ± 22.0 46.9 ± 20.7 0 BASDAI 4.2 ± 2.8 5.5 ± 2.6 2.8 ± 2.4 4.2 ± 2.5 0 ASDAS-ESR 2.4 (12.8, 3.2) 2.55 (2.0, 3.25) 1.8 (1.6, 2.2) - - DAS28 4.24 (3.76, 4.54) - 4.24 (3.76, 4.54) - - PASI 1.0 (0.0, 3.3) 0 1.0 (0.0, 3.3) 0 0 HLA-B27 number positive/number assessed (%) 91/151 (60.3%) 75/78 (96.1%) 13/51 (25.4%) 3/22 (13.6%) 0 CRP (mg/L) 5.5 (2.1, 12.3) 6.7 (3.5, 13.6) 4.1 (1.7, 10.2) 4.3 (1.0, 12.4) 0.0 ESR (mm/h) 14.0 (6.0, 25.0) 12.0 (5.0, 26.0) 13.0 (6.0, 22.0) 19.0 (6.0, 32.0) 9.0 (2.0, 16.0) IL-6 (pg/mL) 3.3 (1.5, 6.6) 4.06 (1.8, 7.92) 2.79 (1.47, 6.35) 2.5 (1.0, 6.61) 1.15 (0.6, 1.5) IL-23 (pg/mL) 0.3 (0.0, 2.0) 0.3 (0.0, 2.0) 0.0 (0.0, 1.4) 0.0 (0.0, 0.3) 0.0 (0.0, 0.0) VEGF (pg/mL) 347.4 (226.0, 562.2) 351.2 (223.2, 572.1) 343.5 (220.2, 666.4) 320.0 (240.0, 375.0) 270.0 (180.0, 445.0) EGF (pg/mL) 104.0 (68.0, 182.0) 96.0 (68.0, 182.0) 126.0 (66.0, 196.0) 93.0 (58.0, 168.0) 81. 0 (38.0, 134.0) Hypertension 47 (24.6%) 25 (30.9%) 13 (17.1%) 9 (26.5%) 0 Diabetes 6 (31.4%) 2 (2.5%) 1 (1.3%) 3 (8.8%) 0 IHD 19 (9.9%) 10 (12.3%) 7 (9.2%) 2 (5.9%) 0 Smoking 24 (12.6%) 12 (14.8%) 3 (3.7%) 9 (26.5%) 0 BMI 26.4 ± 4.1 26.0 ± 4.0 26.9 ± 4.2 26.8 ± 4.6 24.2 ± 3.4 WHR 0.88 ± 0.07 0.89 ± 0.08 0.88 ± 0.07 0.85 ± 0.06 - Total cholesterol (mmol/L) 209.8 ± 40.6 200.0 ± 38.2 217.4 ± 41.5 218.1 ± 40.8 229.0 ± 40.0 HDL cholesterol (mmol/L) 59.5 ± 17.4 58.8 ± 17.7 59.4 ± 18.3 61.6 ± 15.3 62.5 ± 10.1 LDL cholesterol (mmol/L) 127.7 ± 37.0 121.1 ± 34.5 132.1 ± 40.0 135.1 ± 35.8 139.2 ± 33.8 Triglyceride (mmol/L) 129.1 ± 69.3 115.2 ± 63.3 143.5 ± 71.3 133.9 ± 75.6 135.8 ± 60.3 NSAIDs only 46 (24.1%) 25 (30.9%) 16 (21.1%) 5 (14.7%) 0 (0%) NSADSs with sulfasalazine 90 (47.1%) 47 (58%) 31 (40.8%) 12 (35.3%) 0 (0%) NSAIDS with methotrexate 51 (26.7%) 9 (11.1%) 25 (32.9%) 17 (50.0%) 0 (0%) NSAIDS with methotrexate and cyclosporine 4 (2.1%) 0 (0%) 4 (5.3%) 0 (0%) 0 (0%) Hypertensive drugs 47 (24.6%) 25 (30.9%) 13 (17.1%) 9 (26.5%) 0 (0%) Data are presented as number (%), mean ± standard deviation, median (Q1, Q3). Abbreviations: AAU—Acute Anterior Uveitis; AS—ankylosing spondylitis; ASDAS-ESR—Ankylosing Spondylitis Disease Activity Score; BASDAI—Bath Ankylosing Spondylitis Disease Activity Index; CRP—C-reactive protein; DAS28—Disease Activity Score 28; EGF—epidermal growth factor; ESR—erythrocyte sedimentation rate; F—female; IL-6—interleukin 6; HDL: high-density lipoprotein; HLA-B27 - Human leukocyte antigen B27; IL-23—interleukin 23; IBD—inflammatory bowel disease; IHD—ischemic heart disease; LDL: low-density lipoprorotein; M—male; n—number of patients; NSAIDSs—nonsteroidal anti-inflammatory drugs; PASI—Psoriasis Area and Severity Index; PsA—psoriatic arthritis ; SD—standard deviation; SAPHO—Synovitis Acne Pustulosis Hyperostosis Osteitis syndrome; SpA—spondyloarthritis; VAS pain—visual analogue scale of patient’s pain; VEGF—vascular endothelial growth factor; WHR—waist/hip ratio. ijms-17-01255-t002_Table 2Table 2 Serum levels of markers of endothelial function in spondyloarthritis patients groups in comparison to controls. Assessed Parameter SpA Patients (n = 191) Mean ± SD Median (Q1, Q3) p-Value * p-Value AS Patients (n = 81) Mean ± SD Median (Q1, Q3) p-Value * p-Value PsA Patients (n = 76) Mean ± SD Median (Q1, Q3) p-Value * p-Value SAPHO Syndrome Patients (n = 34) Mean ± SD Median (Q1, Q3) p-Value * p-Value Controls (n = 30) Mean ± SD Median (Q1, Q3) IL-18 (pg/mL) 271.8 (207.3, 373.9) 0.0003 0.0116 259.7 (210.1, 372.3) 0.001 0.03 286.0 (219.6, 383.2) 0.0003 0.01 259.4 (193.3, 339.0) 0.01 0.01 198.9 (165.1, 271.5) Fetuin-A (µg/mL) 606.0 ± 155.3 0.001 0.0059 601.0 ± 147.8 0.001 0.1 599.5 ± 152.7 0.001 0.004 632.2 ± 179.0 0.08 0.1 709.1 ± 169.3 sICAM-1 (ng/mL) 234.4 ± 61.9 0.2 0.6710 228.8 ± 57.8 0.5 0.7 237.9 ± 63.4 0.2 0.6 239.7 ± 68.7 0.2 0.7 220.0 ± 69.9 ET-1 (pg/mL) 1.24 ± 0.53 0.03 0.0213 1.13 ± 0.38 0.0005 0.0003 1.25 ± 0.58 0.08 0.02 1.49 ± 0,64 0.87 0.82 1.47 ± 0.57 Data are presented as mean ± standard deviation (SD), median (Q1, Q3). The Student t-test was used for parameters with a normal distribution, Mann–Whitney test was used for parameters with a lack of normal distribution. * age–sex adjusted p-value for the defining groups of patients in ANOVA. AS—ankylosing spondylitis; ET-1—endothelin-1; IL-18—interleukin 18; PsA—psoriatic arthritis; SAPHO—Synovitis Acne Pustulosis Hyperostosis Osteitis syndrome; sICAM-1—soluble intercellular adhesion molecule-1; SpA—spondyloarthritis. ijms-17-01255-t003_Table 3Table 3 A logistic regression model of the odds ratio of the increased serum levels of interleukin-18, decreased serum levels of fetuin-A, increased serum levels of soluble intercellular adhesion molecule-1, and decreased serum levels of endothelin-1 in the spondyloarthritis group of patients compared to controls. Covariates IL-18 ≥ 227.45 pg/mL Fetuin-A ≤ 608.5 µg/mL sICAM-1 ≥ 172.95 ng/mL ET-1 ≤ 1.081 pg/mL OR (95% CI) p OR (95% CI) p OR (95% CI) p OR (95% CI) p SpA 3.22 (1.37–7.55) 0.007 4.14 (1.59–10.81) 0.004 2.7 (1.07–6.78) 0.03 6.73 (2.13–21.27) 0.001 AS 2.31 (0.84–6.32) 0.1 2.63 (0.87–7.92) 0.09 3.75 (1.03–13.74) 0.04 10.24 (2.89–36.23) <0.0001 PsA 4.07 (1.53–10.82) 0.005 5.5 (1.91–15.91) 0.002 3.08 (1.03–9.23) 0.04 7.01 (2.02–24.34) 0.002 SAPHO 3.08 (0.98–9.65) 0.05 3.89 (1.12–13.56) 0.03 1.87 (0.55–6.41) 0.3 3.07 (0.69–13.73) 0.142 AS—ankylosing spondylitis; ET-1—endothelin-1; IL-18—interleukin 18; OR—odds ratio; PsA—psoriatic arthritis; sICAM-1—soluble intercellular adhesion molecule-1; SAPHO—Synovitis Acne Pustulosis Hyperostosis Osteitis syndrome; SpA—spondyloarthritis. ==== Refs References 1. Peters M.J. van der Horst-Bruinsma I.E. Dijkmans B.A. Nurmohamed M.T. Cardiovascular risk profile of patients with spondylarthropathies, particularly ankylosing spondylitis and psoriatic arthritis Semin. Arthritis Rheum. 2004 34 585 592 10.1016/j.semarthrit.2004.07.010 15609262 2. Divecha H. Sattar N. Rumley A. Cherry L. Lowe G.D. Sturrock R. Cardiovascular risk parameters in men with ankylosing spondylitis in comparison with non-inflammatory control subjects: Relevance of systemic inflammation Clin. Sci. 2005 109 171 176 10.1042/CS20040326 15801904 3. Han C. Robinson D.W. Jr. Hackett M.V. Paramore L.C. Fraeman K.H. Bala M.V. Cardiovascular disease and risk factors in patients with rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis J. Rheumatol. 2006 33 2167 2672 16981296 4. Hahn B.H. Grossman J. Chen W. McMahon M. The pathogenesis of atherosclerosis in autoimmune rheumatic diseases: Roles of inflammation and dyslipidemia J. Autoimmun. 2007 28 69 75 10.1016/j.jaut.2007.02.004 17433865 5. Malesci D. Niglio A. Mennillo G.A. Buono R. Valentini G. La Montagna G. High prevalence of metabolic syndrome in patients with ankylosing spondylitis Clin. Rheumatol. 2007 26 710 714 10.1007/s10067-006-0380-5 16933103 6. Sidiropoulos P.I. Karvounaris S.A. Boumpas D.T. Metabolic syndrome in rheumatic diseases: Epidemiology, pathophysiology, and clinical implications Arthritis Res. Ther. 2008 10 207 10.1186/ar2397 18492218 7. Mok C.C. Ko G.T. Ho L.Y. Yu K.L. Chan P.T. To C.H. Prevalence of atherosclerotic risk factors and the metabolic syndrome in patients with chronic inflammatory arthritis Arthritis Care Res. (Hoboken) 2011 63 195 202 10.1002/acr.20363 20890981 8. Semb A.G. Kvien T.K. DeMicco D.A. Fayyad R. Wun C.C. LaRosa J. Betteridge J. Pedersen T.R. Holme I. Prediction of cardiovascular events in patients with ankylosing spondylitis and psoriatic arthritis: Role of lipoproteins in a high-risk population J. Rheumatol. 2012 39 1433 1440 10.3899/jrheum.111307 22660802 9. Papagoras C. Markatseli T.E. Saougou I. Alamanos Y. Zikou A.K. Voulgari P.V. Kiortsis D.N. Drosos A.A. Cardiovascular risk profile in patients with spondyloarthritis J. Bone Spine 2014 8 57 63 10.1016/j.jbspin.2013.03.019 23731637 10. Van Der Linden S. Valkenburg H.A. Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria Arthritis Rheum. 1984 27 361 368 10.1002/art.1780270401 6231933 11. Taylor W. Gladman D. Helliwell P. Marchesoni A. Mease P. Mielants H. CASPAR Study Group Classification criteria for psoriatic arthritis: Development of new criteria from a large international study Arthritis Rheum. 2006 54 2665 2673 10.1002/art.21972 16871531 12. Kahn M.F. Khan M.A. The SAPHO—Syndrome Baillieres Clin. Rheumatol. 1994 8 333 362 10.1016/S0950-3579(94)80022-7 8076391 13. Mallat Z. Corbaz A. Scoazec A. Besnard S. Lesèche G. Chvatchko Y. Tedqui A. Expression of interleukin-18 in human atherosclerotic plaques and relation to plaque instability Circulation 2001 104 1598 1603 10.1161/hc3901.096721 11581135 14. Blankenberg S. Luc G. Ducimetière P. Arveiler D. Ferrières J. Amouyel P. Evans A. Cambien F. Tiret L. Interleukin-18 and the risk of coronary heart disease in European men: The Prospective Epidemiological Study of Myocardial Infarction (PRIME) Circulation 2003 108 2453 2459 10.1161/01.CIR.0000099509.76044.A2 14581397 15. Maksimowicz-McKinnon K. Bhatt D.L. Calabrese L.H. Recent advances in vascular inflammation: C-reactive protein and other inflammatory biomarkers Curr. Opin. Rheumatol. 2004 16 18 24 10.1097/00002281-200401000-00005 14673384 16. Reynolds J.L. Skepper J.N. McNair R. Kasama T. Gupta K. Weissberg P.L. Jahnen-Dechent W. Shanahan C.M. Multifunctional roles for serum protein fetuin-a in inhibition of human vascular smooth muscle cell calcification J. Am. Soc. Nephrol. 2005 16 2920 2930 10.1681/ASN.2004100895 16093453 17. Bohm F. Pernow J. The importance of endothelin-1 for vascular dysfunction in cardiovascular disease Cardiovasc. Res. 2007 76 8 18 10.1016/j.cardiores.2007.06.004 17617392 18. Jahnen-Dechent W. Heiss A. Schäfer C. Ketteler M. Fetuin-A regulation of calcified matrix metabolism Circ. Res. 2011 108 1494 509 10.1161/CIRCRESAHA.110.234260 21659653 19. Badimon I. Romero J.C. Cubedo J. Circulating biomarkers Thromb. Res. 2012 130 S12 S15 10.1016/j.thromres.2012.08.262 23026650 20. Brylka L. Jahnen-Dechent W. The role of fetuin-A in physiological and pathological mineralization Calcif. Tissue Int. 2013 93 355 364 10.1007/s00223-012-9690-6 23277412 21. Przepiera-Będzak H. Fischer K. Brzosko M. Serum IL-6 and IL-23 levels and their correlation with angiogenic cytokines and disease activity in ankylosing spondylitis, psoriatic arthritis, and SAPHO syndrome Mediat. Inflamm. 2015 2015 785705 10.1155/2015/785705 26339141 22. Caliskan M. Erdogan D. Gullu H. Yilmaz S. Gursoy Y. Yildirir A. Yucel E. Muderrisoglu H. Impaired coronary microvascular and left ventricular diastolic functions in patients with ankylosing spondylitis Atherosclerosis 2008 196 306 312 10.1016/j.atherosclerosis.2006.11.003 17169363 23. Drouart M. Saas P. Billot M. Cedoz J.P. Tiberghien P. Wendling D. Toussirot É. High serum vascular endothelial growth factor correlates with disease activity of spondylarthropaties Clin. Exp. Immunol. 2003 132 158 162 10.1046/j.1365-2249.2003.02101.x 12653851 24. Przepiera-Będzak H. Fischer K. Brzosko M. Serum levels of angiogenic cytokines in psoriatic arthritis and SAPHO syndrome Polskie Arch. Med. Wewn. 2013 123 297 302 25. Poddubnyy D. Conrad K. Haibel H. Syrbe U. Appel H. Braun J. Rudwaleit M. Sieper J. Elevated serum level of the vascular endothelial growth factor predicts radiographic spinal progression in patients with axial spondyloarthritis Ann. Rheum. Dis. 2014 73 2137 2143 10.1136/annrheumdis-2013-203824 23956246 26. Przepiera-Będzak H. Fischer K. Brzosko M. Serum VEGF, EGF, basic FGF, and acidic FGF levels and their association with disease activity and extra‑articular symptoms in ankylosing spondylitis Polskie Arch. Med. Wewn. 2016 126 290 292 10.20452/pamw.3341 27025929 27. Sari I. Okan T. Akar S. Cece H. Altay C. Secil M. Onen F. Akkoc N. Impaired endothelial function in patients with ankylosing spondylitis Rheumatology (Oxford) 2006 45 283 286 10.1093/rheumatology/kei145 16204374 28. Rooney T. Murphy E. Benito M. Roux-Lombard P. FitzGerald O. Dayer J.M. Bresnihan B. Synovial tissue interleukin-18 expression and the response to treatment in patients with inflammatory arthritis Ann. Rheum. Dis. 2004 63 1393 1398 10.1136/ard.2003.016428 15479888 29. Hurtado-Nedelec M. Chollet-Martin S. Nicaise-Roland P. Grootenboer-Mignot S. Ruimy R. Meyer O. Hayem G. Characterization of the immune response in the synovitis, acne, pustulosis, hyperostosis, osteitis (SAPHO) syndrome Rheumatology 2008 47 1160 1167 10.1093/rheumatology/ken185 18559374 30. Sari I. Kebapcilar L. Taylan A. Bilgir O. Kozaci D.L. Yildiz Y. Yuksel A. Gunay N. Akkoc N. Fetuin-A and interleukin-18 levels in ankylosing spondylitis Int. J. Rheum. Dis. 2010 13 75 81 10.1111/j.1756-185X.2009.01448.x 20374388 31. Surdacki A. Sulicka J. Korkosz M. Mikolajczyk T. Telesinska-Jasiówka D. Klimek E. Kierzkowska I. Guzik T. Grodzicki T.K. Blood monocyte heterogeneity and markers of endothelial activation in ankylosing spondylitis J. Rheumatol. 2014 41 481 489 10.3899/jrheum.130803 24488416 32. Tuylu T. Sari I. Solmaz D. Kozaci D.L. Akar S. Gunay N. Onen F. Akkoc N. Fetuin-A is related to syndesmophytes in patients with ankylosing spondylitis: A case control study Clinics (Sao Paulo) 2014 69 688 693 10.6061/clinics/2014(10)07 25518021 33. Wang J. Sun C. Gerdes N. Liu C. Liao M. Liu J. Shi A. He A. Zhou Y. Sukhova G.K. Interleukin 18 function in atherosclerosis is mediated by the interleukin 18 receptor and the Na–Cl co-transporter Nat. Med. 2015 21 820 826 10.1038/nm.3890 26099046 34. Wendling D. Racadot E. Augé B. Toussirot E. Soluble intercellular adhesion molecule 1 in spondylarthropathies Clin. Rheumatol. 1998 17 202 204 10.1007/BF01451047 9694052 35. Sari I. Kebapcilar L. Alacacioglu A. Bilgir O. Yildiz Y. Taylan A. Yuksel A. Kozaci D.L. Increased levels of asymmetric dimethylarginine (ADMA) in patients with ankylosing spondylitis Intern. Med. 2009 48 1363 1368 10.2169/internalmedicine.48.2193 19687580 36. Ashcroft D.M. Li Wan Po A. Williams H.C. Griffiths C.E.M. Clinical measures of disease severity and outcome in psoriasis: A critical appraisal of their quality Dr. J. Dermatol. 1999 141 185 191 10.1046/j.1365-2133.1999.02963.x 37. Garrett S. Jenkinson T. Kennedy L.G. Whitelock H. Gaisford P. Calin A. A new approach to defining disease status in ankylosing spondylitis: The bath ankylosing spondylitis disease activity index J. Rheumatol. 1994 21 2286 2291 7699630

PMC005xxxxxx/PMC5000654.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081256ijms-17-01256ArticleSuppression of Lipid Accumulation by Indole-3-Carbinol Is Associated with Increased Expression of the Aryl Hydrocarbon Receptor and CYP1B1 Proteins in Adipocytes and with Decreased Adipocyte-Stimulated Endothelial Tube Formation Wang Mei-Lin 1Lin Shyh-Hsiang 1Hou Yuan-Yu 2Chen Yue-Hwa 13*Battino Maurizio Academic Editor1 School of Nutrition and Health Sciences, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan; [email protected] (M.-L.W.); [email protected] (S.-H.L.)2 Department of Food and Beverage Management, Mackay Medicine, Nursing and Management College, Taipei 112, Taiwan; [email protected] Cancer Research Center, Taipei Medical University Hospital, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan* Correspondence: [email protected]; Tel.: +886-2-2736-1661 (ext. 6550)03 8 2016 8 2016 17 8 125624 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).This study investigated the effects of indole-3-carbinol (I3C) on adipogenesis- and angiogenesis-associated factors in mature adipocytes. The cross-talk between mature adipocytes and endothelial cells (ECs) was also explored by cultivating ECs in a conditioned medium (CM) by using I3C-treated adipocytes. The results revealed that I3C significantly inhibited triglyceride accumulation in mature adipocytes in association with significantly increased expression of AhR and CYP1B1 proteins as well as slightly decreased nuclear factor erythroid-derived factor 2–related factor 2, hormone-sensitive lipase, and glycerol-3-phosphate dehydrogenase expression by mature adipocytes. Furthermore, I3C inhibited CM-stimulated endothelial tube formation, which was accompanied by the modulated secretion of angiogenic factors in adipocytes, including vascular endothelial growth factor, interleukin-6, matrix metalloproteinases, and nitric oxide. In conclusion, I3C reduced lipid droplet accumulation in adipocytes and suppressed adipocyte-stimulated angiogenesis in ECs, suggesting that I3C is a potential therapeutic agent for treating obesity and obesity-associated disorders. indole-3-carbinolobesityadipocytesadipogenesisangiogenesisaryl hydrocarbon receptor ==== Body 1. Introduction Obesity has been a major concern since the 20th century, and its prevalence is increasing in many countries. In addition to affecting the physical and psychological status of obese people, excess weight, mainly accumulated in the form of lipid in the adipocytes of white adipose tissue (WAT), considerably increases the risk of developing various chronic diseases, including diabetes, cardiovascular diseases, and cancers [1]. Moreover, in addition to mature adipocytes, WAT consists of fibroblastic preadipocytes, endothelial cells (ECs), and macrophages. The expansion of mature adipoctyes, the differentiation of preadiopocytes to adipocytes, the formation of new vessels of ECs, and the infiltration of macrophages accompany the progress of WAT expansion in obesity [2]. Mature adipocytes are not only a lipid storage site but also produce and secrete different adipokines and factors such as leptin, interleukin (IL)-6, and vascular endothelial growth factor (VEGF), which are closely associated with angiogenesis and other pathological conditions in obesity [2,3]. Therefore, the massive accumulation of lipids in mature adipocytes in obesity causes intimate interactions between mature adipocytes and adjacent cells, including vascular ECs in WAT, which may contribute to the pathological characteristics of obesity, including different metabolic disorders and cancers [4,5,6]. Indole-3-carbinol (I3C) is a bioactive indolic compound derived from glucosinolates in cruciferous vegetables, such as broccoli, cabbage, Brussels sprouts, kale, and cauliflower [7]. I3C possesses anticarcinogenic activities [7,8,9]. Recently, I3C has shown to exhibit antiobesity activity by reducing body weight and fat in animals fed a high-fat diet [10,11] and by inhibiting the differentiation of 3T3-L1 preadipocytes [12]. Being an activator of the aryl hydrocarbon receptor (AhR), a ligand-activated transcription factor crucial in adipogenesis and angiogenesis [13,14,15,16], we considered that I3C executes its activities through the AhR. Although I3C has been shown to inhibit the differentiation of preadipocytes, knowledge regarding its effects on lipid accumulation in mature adipocytes and on adipocyte-associated angiogenesis is limited. Because increased triglyceride (TG) accumulation in mature adipocytes is positively associated with obesity and associated metabolic disorders, compounds that reduce the TG content in adipocytes may have therapeutic roles in obesity and related pathological disorders. This study was aimed at examining the roles of I3C in the adipogenesis of mature adipocytes and in the cross-talk between mature adipocytes and ECs. Furthermore, the effects of I3C on factors associated with AhR-mediated pathways were determined. Results obtained in this study may facilitate elucidating the potential use of I3C in adjunctive treatment for obesity and associated disorders. 2. Results 2.1. Effects of I3C on Cell Viability and Lipid Accumulation in Mature Adipocytes At concentrations of 5–50 μM, I3C slightly, in a concentration-dependent manner, reduced (5%–22%) the viability of mature adipocytes after 24 h of treatment (Figure 1), and these concentrations were used for the following experiments. Results from the oil red O staining and analysis of intracellular TG content revealed that I3C concentration-dependently reduced lipid accumulation in the mature adipocytes (Figure 2A), and this reduction was associated with the increased release of glycerol by mature adipocytes (Figure 2B). 2.2. Effects of I3C on Adipocyte-Induced Tube Formation in ECs Expansion of adipose tissue is accompanied by increased endothelial angiogenesis; therefore, endothelial tube formation was used to explore the effects of I3C on interactions between mature adipocytes and ECs. The results revealed that the conditioned medium (CM) from adipocytes significantly stimulated the formation of tube-like endothelial structures; however, the presence of I3C reduced the interconnection networks among ECs (Figure 3), and the suppression was accompanied by the decreased production of proangiogenic factors, including VEGF (Figure 4A), IL-6 (Figure 4B), and to a lesser extent, NO (Figure 4C) and matrix metalloproteinases (MMPs) (Figure 4D,E), by the mature adipocytes. 2.3. Effects of I3C on Protein Expression by Mature Adipocytes To determine the effects of I3C on the expression of AhR-, adipogenesis-, and angiogenesis-associated proteins by mature adipocytes, a Western blot analysis was performed. Figure 5 shows that I3C significantly enhanced the expression of the AhR and the gene that it regulates, CYP1B1, in mature adipocytes, but only slightly enhanced ARNT expression at high concentrations. In contrast, the expression of nuclear factor erythroid-derived factor 2-related factor 2 (Nrf-2), HSL, GPDH, and VEGFR was slightly downregulated by I3C, especially at higher concentrations. The original blots for each protein are shown in the Figure S1. 3. Discussion In the present study, we demonstrated for the first time that a cruciferous vegetable bioactive component, I3C, at concentrations of 25–50 μM, significantly reduced TG accumulation in mature adipocytes, and this effect was associated with increased expression of the AhR and CYP1B1 proteins in adipocytes. In addition, I3C suppressed the production of proangiogenic mediators by mature adipocytes, leading to decreased endothelial tube formation stimulated by the adipocytes. Obesity is positively associated with various metabolic disorders and cancers, and the expansion of WAT accompanied by enlarged adipocytes is believed to be a key event in these pathological conditions involving the adipogenesis and angiogenesis of adipose tissues. Because the mature adipocyte is the major cell type in the WAT of obese people, the suppressive effects of I3C on TG accumulation and stimulated endothelial tube formation suggest that I3C has potential therapeutic effects on obesity, and possibly, on obesity-associated metabolic disorders. Several studies have indicated the protective effects of I3C on cancers and obesity; these effects are believed to be intimately associated with the AhR-mediated pathways. The liganded AhR dimerizes with the ARNT to modulate various pathways, including carcinogen metabolism and adipocyte differentiation. The AhR ligands, such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), β-naphthoflavone (BNF), and polychlorinated biphenyl, activate the transcription of xenobiotic metabolizing enzymes CYP1A and CYP1B [17]. On the other hand, the AhR negatively regulates adipocyte differentiation [13,18], and TCDD suppresses adipocyte differentiation [19,20], whereas the AhR antagonist α-naphthoflavone (ANF) increases lipid accumulation in mature adipocytes [21]. I3C is a naturally occurring AhR agonist that exhibits antiobesity activities, such as the reduction of body and WAT weights in high-fat-diet-induced obese mice and the inhibition of adipocyte differentiation by activating the silent mating type information regulation 2 homolog 1 and subsequently downregulating the expression of PPARγ2, C/EBPα, and aP2, factors crucial for differentiation [12,22]. In this study, we observed that I3C may act through activating lipolysis and/or inhibiting lipogenesis to reduce TG accumulation in mature adipocytes from the evidence that I3C increased the glycerol released into the medium and suppressed the expression of GPDH, a key enzyme in lipogenesis, at high concentrations. The generation of glycerol from TG hydrolysis acts through sequential actions of different lipases, including adipose TG lipase (ATGL), HSL, and monoglyceride lipase, and both HSL and ATGL are major enzymes involved in TG catabolism in adipose tissue [23,24]. I3C might not act chiefly through HSL activation to increase lipolysis in mature adipocytes, because the expression of the HSL protein was slightly reduced by I3C. However, the phosphorylation status of HSL was not determined, so the detailed mechanisms require investigated further. Not only by inhibiting adipocyte differentiation does AhR regulate lipid metabolism. AhR is a constitutive inhibitor of TG synthesis in mouse embryo fibroblasts (MEFs) [25]; in addition, a transient fatty liver was observed in AhR-null mice [26], suggesting that AhR acts as a suppressor of lipid accumulation. Because we observed that I3C enhanced the expression of the AhR protein in adipocytes, this may be also involved in the suppressed TG accumulation in adipocytes by I3C. Similarly, Tano et al. [27] indicated that BNF represses the expression of enzymes in the fatty acid synthesis pathway in primary hepatocytes, leading to a decrease in fatty acid production, and these effects are dependent upon AhR. TCDD increased the expression of lipolysis-associated factors in treated mice, and this increase is related to the TCDD-induced wasting syndrome [28,29]. On the other hand, we previously reported that the AhR antagonist ANF reduced the expression of the AhR in association with increased TG accumulation in adipocytes [21]. Therefore, we hypothesize that I3C reduces lipid accumulation in adipocytes by inducing AhR expression and then reducing lipogenesis and increasing lipolysis responses in 3T3-L1 adipoyctes. Nrf-2 is another transcription factor crucial in the expression of xenobiotic metabolism enzymes and in adipogenesis [30]. An intimate interaction between Nrf-2 and the AhR pathways exists. Miao et al. [31] reported that TCDD induces Nrf-2 expression by activating AhR-XRE binding in Hepa1c1c cells. Conversely, Nrf-2−/− MEFs had low levels of AhR expression, and the Nrf-2 activator upregulated the AhR pathways, subsequently inhibiting adipogenesis in Nrf-2+/+ MEFs. Furthermore, stable knockdown of Nrf-2 in 3T3-L1 cells inhibited enhanced adipogenesis caused by the deficiency of kelch-like ECH-associated protein 1 [32], indicating the suppressive role of Nrf-2 in adipogenesis [25]. However, not only I3C but also the AhR antagonist ANF suppressed Nrf-2 protein expression in mature adipocytes. These results suggest that Nrf-2 predominantly affects adipocyte differentiation; however, it either does not regulate or only slightly regulates AhR expression and angiogenesis in mature adipocytes. This mechanism requires further investigation. In addition to reducing lipid accumulation in mature adipocytes, I3C (5–50 μM) inhibited endothelial tube formation stimulated by the CM from mature adipocytes, and this suppression was associated with the decreased secretion of angiogenic factors, including VEGF, IL-6, NO, and MMPs, by mature adipocytes. Substantial tissue remodeling that occurs within adipose tissues during fat mass expansion is accompanied by angiogenesis [33,34]. WAT not only is a TG storage depot but also acts as an endocrine organ because of its abilities to produce and secrete various adipokines, growth factors, and inflammatory mediators, which may alter the functions of different cells. Higher serum levels of proinflammatory mediators, such as C-reactive protein, IL-6, and tumor necrosis factor α, as well as angiogenic factors, including VEGF, IGF-1, MMPs, and leptin, were observed in obese people compared with those in healthy people [15], and these factors are involved in the pathological changes of obesity. Mature human adipose tissue extract produces numerous angiogenic factors and induces endothelial tube formation [35]. The expression of VEGF positively correlates to the size of adipocytes [36], and is enhanced by IL-6 [37]. Because I3C was observed to reduce IL-6 expression in the WAT in high-fat-diet-induced obese mice [11] and in LPS-stimulated macrophages [38], we hypothesize that I3C (5–50 μM) suppresses adipocyte-induced angiogenesis both by reducing TG accumulation in mature adipocytes leading to reduce the secretion of angiogenic mediators, and by directly inhibiting IL-6 expression in mature adipocytes. Alternatively, I3C lowers leptin levels and increases serum adiponectin levels in obese animals [10]; these effects may contribute to the antiangiogenic effect of I3C observed in this study. Thus, I3C can not only diminish adipocyte differentiation but also eliminate angiogenesis, thereby inhibiting obesity. The estimated daily intake is at the equivalent of 6.4 mg of I3C in the UK, where the cruciferous vegetable tends to be a dietary staple [39]. Ideally, this dose would generate ca. 9 μM of plasma I3C concentration in a 70 kg subject on the basis that blood volume comprises 7% of the body weight without considering digestion and absorption. Similarly, the plasma I3C peaked concentration was 4.13 μg/mL (ca. 28 μM) after orally administering 250 mg/kg of I3C to mice [40]. Although the plasma I3C concentration from ordinary vegetable consumption is lower than the concentrations we used in this study, higher plasma concentrations may be possibly achieved by taking I3C dietary supplements. Alternatively, several acid-catalyzed compounds formed following oral consumption of I3C have been identified and may contribute to the protective roles of I3C [41,42,43]. Among these derivatives, 3,3′-diindolylmethane (DIM) has been reported to play a protective role in metabolic diseases, including reducing blood glucose and increasing antioxidative enzymes in diabetic mice [44], as well as in alleviating hepatic steatosis and inflammation in Nonalcoholic steatohepatitis (NASH) mouse models [45]. Furthermore, a significant amount of DIM could be detected in the plasma and various organs after oral administration of I3C [40]. In addition, different studies have indicated that orally giving 400 or 800 mg/day of I3C to human subjects for up to 4.8 years showed no adverse effects [46,47]. Finally, although the AhR has been associated with carcinogenesis and toxicity, most studies suggest that I3C acts as a chemopreventive agent. Because I3C decreases body weight, reduces adipocyte lipid accumulation, and because I3C derivatives possess protective effects in metabolic disorders associated with obesity, the potential use of I3C in treating obesity or obesity-associated diseases is plausible, although the effects of I3C in vivo require further investigation. 4. Materials and Methods 4.1. Chemicals and Biochemicals I3C, insulin, dexamethasone (Dex), 3-isobutyl-1-methyl-xanthine (IBMX), and dimethyl sulfoxide (DMSO) were purchased from Sigma Chemical (St. Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM), sodium bicarbonate, fetal bovine serum (FBS), calf serum, trypan blue, and trypsin were obtained from GIBCO BRL (Grand Island, NY, USA). I3C was dissolved in DMSO, and the final concentration of DMSO in culture media was 0.1% (v/v). 4.2. Cell Culture The murine preadipocyte cell line 3T3-L1, which is typically used as a model for studying adipocyte differentiation and biology, was purchased from the Bioresource Collection and Research Center (BCRC #60159; Hsinchu, Taiwan). The cells were grown in a monolayer in DMEM supplemented with 10% fetal bovine serum (FBS) at 37 °C in a 95% air and 5% CO2 environment. To induce adipocyte differentiation, 3T3-L1 preadipocytes were cultivated in a DMEM differentiation medium that contained 0.25 μM Dex, 10 μg/mL insulin, and 0.5 mM IBMX for 2 days and then maintained in insulin-containing DMEM for another 4 days to obtain round mature adipocytes. After the medium was replaced with DMEM containing 1% FBS, the mature adipocytes were treated with I3C for 24 h, and the cells and the medium were analyzed. To examine tube formation, a human endothelium-derived cell line with vascular EC characteristics, EA hy926, was used. EA hy926 cells were provided by Dr. Cora-Jean Edgell (University of North Carolina, Chapel Hill, NC, USA), who established and characterized the cells [48], which are now available from the American Type Culture Collection (ATCC® CRL-2922). The cells were maintained in DMEM supplemented with 10% FBS at 37 °C in a 95% air and 5% CO2 environment. 4.3. Cytotoxicity To determine the cytotoxic effects of I3C on mature adipocytes, cells were treated with different concentrations of I3C for 24 h, and cytotoxicity was evaluated by measuring the absorbance of the formazan product of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) produced by live cells by using a microplate reader at OD490 nm. 4.4. Lipid Accumulation and Glycerol Release of 3T3-L1 Adipocyte To explore the effects of I3C on lipid accumulation, intracellular TG accumulation in mature adipocytes was determined using oil red O staining and was quantified using a commercial Randox TRIGS (Cat TG213) assay kit (Randox Labs, Crumlin, UK). Glycerol released into a medium, regarded as a marker of lipolysis, was analyzed using a commercial Randox glycerol (Cat GY105) assay kit. 4.5. Tube Formation Assay To determine whether I3C affects the cross-talk between mature adipocytes and ECs, mature adipocytes were treated with I3C for 24 h, and the conditioned medium (CM) was retrieved and used for cultivating EA hy926 ECs, which were grown on BD Matrigel-coated plates for 24 h. The ECs were subsequently stained with calcein AM fluorescent dye (BD Biosciences, San Jose, CA, USA), and networks of vessel-like structures were observed and photographed using a fluorescent microscope. 4.6. Assays of Nitric Oxide, VEGF, IL-6, and Matrix Metalloproteinase Activities To determine the effects of I3C on the production of angiogenic factors by mature adipocytes, the levels of VEGF, and IL-6 in the CM of mature adipocytes were determined using mouse DuoSet VEGF, and IL-6 commercial enzyme-linked immunosorbent assay systems (R & D System, Minneapolis, MN, USA), respectively. The amount of nitric oxide (NO) in the CM was determined using the Griess reagent. The activities of matrix metalloproteinase (MMP)-2 and -9 in the CM were determined through gelatin zymography, in which a culture medium containing 5 μg of protein was separated on a 10% SDS-PAGE gel that contained 1 mg/mL gelatin and then stained with 0.5% Coomassie Blue R-250. Clear bands in the destained gel against a blue background indicated the presence of MMP-2 and -9 (92 kDa) and were quantitated using Image-Pro Plus software (Media Cybernetics, Silver Spring, MD, USA). 4.7. Western Blot Analysis Expression of proteins related to the AhR-mediated pathway, lipid metabolism, and angiogenesis in adipocytes was analyzed through Western blot. Following separation on a 10% SDS-PAGE gel, cellular proteins (15 μg), including AhR (1:1000, Santa Cruz Biotechnology, Santa Cruz, CA, USA), vascular endothelial growth factor receptor (VEGFR, 1:500, Santa Cruz Biotechnology), the AhR nuclear translocator (ARNT, 1:1000, Abcam Inc., Cambridge, MA, USA), CYP1B1 (1:1000, Abcam Inc.), Nrf2 (1:1000, Abcam Inc.), glycerol-3-phosphate dehydrogenase (GPDH, 1:1000, Abcam Inc.), hormone-sensitive lipase (HSL, 1:1000, Origene Technologies, Inc., Rockville, MD, USA) and β-actin (1:500, Novus Biologicals, Littleton, CO, USA), were electroblotted onto a polyvinylidene difluoride membrane and detected with specific protein monoclonal antibodies. Following the addition of peroxidase-conjugated immunoglobulin G (Millipore Corporation, Billerica, MA, USA) and detection with Amersham Enhanced Chemiluminescence™ western blotting detection reagents (GE Healthcare, Piscataway, NJ, USA), the specific proteins were quantitated using Image-Pro Plus software (Media Cybernetics, Silver Spring, MD, USA). 4.8. Statistical Analysis Values are expressed as the mean ± standard deviation (SD). One-way analysis of variance followed by Fisher’s least significant difference test were performed using SAS software version 9.1 (SAS Institute, Cary, NC, USA) to compare the differences between groups. Differences were considered statistically significant at p < 0.05. 5. Conclusions I3C significantly reduced lipid accumulation in mature adipocytes and suppressed adipocyte-stimulated tube formation in ECs, and these effects are associated with the decreased secretion of angiogenic factors by mature adipocytes, including VEGF, IL-6, and NO. In addition, I3C increased the expression of the AhR and CYP1B1 proteins in mature adipocytes. These results suggest that I3C can be potentially used for facilitating weight loss and alleviating obesity-associated disorders. Acknowledgments We thank Cora-Jean Edgell (University of North Carolina, Chapel Hill, NC, USA) for providing the EA hy926 endothelial cells. This work was supported by a grant (NSC 102-2320-B-038-029-MY2) from the Ministry of Science and Technology, Taipei, Taiwan. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1256/s1. Click here for additional data file. Author Contributions Mei-Lin Wang performed the experiments and analyzed the data; Shyh-Hsiang Lin helped to analyze the data and revised the manuscript; Yuan-Yu Hou contributed reagents/materials/analysis tools; Yue-Hwa Chen conceived and supervised the experiments and wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Effect of indole-3-carbinol (I3C) on cell viability in differentiated adipocytes. Cells were treated with various concentrations of I3C for 24 h, and cell viability was measured using an MTS assay kit. Values are the mean ± SD from three measurements. a–d, bars with different letters significantly differ from each other (p < 0.05). Figure 2 Effect of indole-3-carbinol (I3C) on lipid accumulation in differentiated adipocytes. Cells were treated with various concentrations of I3C for 24 h, and the cells were stained with oil red O. Intracellular oil red O and triglyceride (TG) (A) and extracellular glycerol (B) contents were quantified using the method described in the Materials and Methods. Values are the mean ± SD from three measurements. Bars with different letters (a–d) or different symbols (+, #, *) significantly differ from each other (p < 0.05). Figure 3 Effects of indole-3-carbinol (I3C) on tube formation in endothelial cells activated with the conditioned medium (CM) from mature adipocytes. Following 6 d differentiation, mature adipocytes were treated with I3C for 24 h, and the CM was collected and used to cultivate endothelial EA hy926 cells, which were grown on Matrigel-coated plates for 24 h. Formation of tube-like structures (as indicated in arrows) was observed and photographed under a microscope after staining with calcein AM fluorescent dye. Pictures are representative of three independent experiments. Scale Bars = 100 μm. Figure 4 Effects of indole-3-carbinol (I3C) on the angiogenic factors in the cultured medium from differentiated adipocytes. Following differentiation, adipocytes were treated with various concentrations of I3C in Dulbecco’s modified Eagle’s medium containing 1% fetal bovine serum for 24 h. The medium was retrieved for analyzing the vascular endothelial growth factor (VEGF) (A); interleukin (IL)-6 (B); nitric oxide (NO) (C); and matrix metalloproteinase (MMP) activities (D); and quantification (E) by using the methods described in the Materials and Methods. Values are the mean ± SD from three measurements, and bars with different letters (a–c) or different symbols (#,*) significantly differ from each other (p < 0.05). Figure 5 Effects of I3C on the aryl hydrocarbon receptor (AhR), AhR nuclear translocator (ARNT), CYP1B1, hormone-sensitive lipase (HSL), glycerol-3-phosphate dehydrogenase (GPDH), nuclear factor erythroid-derived factor 2-related factor 2 (Nrf-2), and vascular endothelial growth factor receptor (VEGFR) protein expression in differentiated adipocytes. Mature adipocytes were treated with various concentrations of I3C for 24 h, and proteins were retrieved and subsequently measured using a Western blot analysis. The expression of each protein was adjusted using β-actin, and values represented are a fold of the control. The values in parentheses represent standard deviations from three different measurements in this experiment. * indicates significant different from the negative control (p < 0.05). ==== Refs References 1. Rutkowski J.M. Davis K.E. Scherer P.E. Mechanisms of obesity and related pathologies: The macro- and microcirculation of adipose tissue FEBS J. 2009 276 5738 5746 10.1111/j.1742-4658.2009.07303.x 19754873 2. Henry S.L. Bensley J.G. Wood-Bradley R.J. Cullen-McEwen L.A. Bertram J.F. Armitage J.A. White adipocytes: More than just fat depots Int. J. Biochem. Cell Biol. 2012 44 435 440 10.1016/j.biocel.2011.12.011 22222895 3. Lafontan M. Adipose tissue and adipocyte dysregulation Diabetes Metab. 2014 40 16 28 10.1016/j.diabet.2013.08.002 24139247 4. Hosogai N. Fukuhara A. Oshima K. Miyata Y. Tanaka S. Segawa K. Furukawa S. Tochino Y. Komuro R. Matsuda M. Adipose tissue hypoxia in obesity and its impact on adipocytokine dysregulation Diabetes 2007 56 901 911 10.2337/db06-0911 17395738 5. Trayhurn P. Hypoxia and adipose tissue function and dysfunction in obesity Physiol. Rev. 2013 93 1 21 10.1152/physrev.00017.2012 23303904 6. Ye J. Adipose tissue vascularization: Its role in chronic inflammation Curr. Diabetes Rep. 2011 11 203 210 10.1007/s11892-011-0183-1 21327583 7. Fenwick G.R. Heaney R.K. Mullin W.J. Glucosinolates and their breakdown products in food and food plants Crit. Rev. Food Sci. Nutr. 1983 18 123 201 10.1080/10408398209527361 6337782 8. McDanell R. McLean A.E. Hanley A.B. Heaney R.K. Fenwick G.R. Chemical and biological properties of indole glucosinolates (glucobrassicins): A review Food Chem. Toxicol. 1988 26 59 70 10.1016/0278-6915(88)90042-7 3278958 9. Aggarwal B.B. Ichikawa H. Molecular targets and anticancer potential of indole-3-carbinol and its derivatives Cell Cycle 2005 4 1201 1215 10.4161/cc.4.9.1993 16082211 10. Chang H.P. Wang M.L. Chan M.H. Chiu Y.S. Chen Y.H. Antiobesity activities of indole-3-carbinol in high-fat-diet-induced obese mice Nutrition 2011 27 463 470 10.1016/j.nut.2010.09.006 21392705 11. Choi Y. Kim Y. Park S. Lee K.W. Park T. Indole-3-carbinol prevents diet-induced obesity through modulation of multiple genes related to adipogenesis, thermogenesis or inflammation in the visceral adipose tissue of mice J. Nutr. Biochem. 2012 23 1732 1739 10.1016/j.jnutbio.2011.12.005 22569347 12. Choi Y. Um S.J. Park T. Indole-3-carbinol directly targets SIRT1 to inhibit adipocyte differentiation Int. J. Obes. 2013 37 881 884 10.1038/ijo.2012.158 22986685 13. Alexander D.L. Ganem L.G. Fernandez-Salguero P. Gonzalez F. Jefcoate C.R. Aryl-hydrocarbon receptor is an inhibitory regulator of lipid synthesis and of commitment to adipogenesis J. Cell Sci. 1998 111 3311 3322 9788873 14. Shimba S. Todoroki K. Aoyagi T. Tezuka M. Depletion of arylhydrocarbon receptor during adipose differentiation in 3T3-L1 cells Biochem. Biophys. Res. Commun. 1998 249 131 137 10.1006/bbrc.1998.9100 9705844 15. Roman A.C. Carvajal-Gonzalez J.M. Rico-Leo E.M. Fernandez-Salguero P.M. Dioxin receptor deficiency impairs angiogenesis by a mechanism involving VEGF-A depletion in the endothelium and transforming growth factor-beta overexpression in the stroma J. Biol. Chem. 2009 284 25135 25148 10.1074/jbc.M109.013292 19617630 16. Shin J.H. Zhang L. Murillo-Sauca O. Kim J. Kohrt H.E. Bui J.D. Sunwoo J.B. Modulation of natural killer cell antitumor activity by the aryl hydrocarbon receptor Proc. Natl. Acad. Sci. USA 2013 110 12391 12396 10.1073/pnas.1302856110 23836658 17. Safe S. Lee S.O. Jin U.H. Role of the aryl hydrocarbon receptor in carcinogenesis and potential as a drug target Toxicol. Sci. 2013 135 1 16 10.1093/toxsci/kft128 23771949 18. Shimba S. Wada T. Tezuka M. Arylhydrocarbon receptor (AhR) is involved in negative regulation of adipose differentiation in 3T3-L1 cells: AhR inhibits adipose differentiation independently of dioxin J. Cell Sci. 2001 114 2809 2817 11683414 19. Brodie A.E. Azarenko V.A. Hu C.Y. 2,3,7,8-Tetrachlorodibenzo-ρ-dioxin (TCDD) inhibition of fat cell differentiation Toxicol. Lett. 1996 84 55 59 10.1016/0378-4274(95)03537-0 8597178 20. Phillips M. Enan E. Liu P.C. Matsumura F. Inhibition of 3T3-L1 adipose differentiation by 2,3,7,8-tetrachlorodibenzo-ρ-dioxin J. Cell Sci. 1995 108 395 402 7537747 21. Wang M.L. Lin S.H. Hou Y.Y. Chen Y.H. Alpha-naphthoflavone increases lipid accumulation in mature adipocytes and enhances adipocyte-stimulated endothelial tube formation Nutrients 2015 7 3166 3183 10.3390/nu7053166 25942489 22. Sul H.S. Smas C.M. Wang D. Chen L. Regulation of fat synthesis and adipose differentiation Prog. Nucleic Acid Res. Mol. Biol. 1998 60 317 345 9594578 23. Zimmermann R. Strauss J.G. Haemmerle G. Schoiswohl G. Birner-Gruenberger R. Riederer M. Lass A. Neuberger G. Eisenhaber F. Hermetter A. Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase Science 2004 306 1383 1386 10.1126/science.1100747 15550674 24. Schweiger M. Schreiber R. Haemmerle G. Lass A. Fledelius C. Jacobsen P. Tornqvist H. Zechner R. Zimmermann R. Adipose triglyceride lipase and hormone-sensitive lipase are the major enzymes in adipose tissue triacylglycerol catabolism J. Biol. Chem. 2006 281 40236 40241 10.1074/jbc.M608048200 17074755 25. Shin S. Wakabayashi N. Misra V. Biswal S. Lee G.H. Agoston E.S. Yamamoto M. Kensler T.W. NRF2 modulates aryl hydrocarbon receptor signaling: Influence on adipogenesis Mol. Cell. Biol. 2007 27 7188 7197 10.1128/MCB.00915-07 17709388 26. Schmidt J.V. Su G.H. Reddy J.K. Simon M.C. Bradfield C.A. Characterization of a murine Ahr null allele: Involvement of the Ah receptor in hepatic growth and development Proc. Natl. Acad. Sci. USA 1996 93 6731 6736 10.1073/pnas.93.13.6731 8692887 27. Tanos R. Murray I.A. Smith P.B. Patterson A. Perdew G.H. Role of the Ah receptor in homeostatic control of fatty acid synthesis in the liver Toxicol. Sci. 2012 129 372 379 10.1093/toxsci/kfs204 22696238 28. Houlahan K.E. Prokopec S.D. Sun R.X. Moffat I.D. Lindén J. Lensu S. Okey A.B. Pohjanvirta R. Boutros P.C. Transcriptional profiling of rat white adipose tissue response to 2,3,7,8-tetrachlorodibenzo-ρ-dioxin Toxicol. Appl. Pharmacol. 2015 288 223 231 10.1016/j.taap.2015.07.018 26232522 29. Lindén J. Lensu S. Pohjanvirta R. Effect of 2,3,7,8-tetrachlorodibenzo-ρ-dioxin (TCDD) on hormones of energy balance in a TCDD-sensitive and a TCDD-resistant rat strain Int. J. Mol. Sci. 2014 15 13938 13966 10.3390/ijms150813938 25119860 30. Schneider K.S. Chan J.Y. Emerging role of Nrf2 in adipocytes and adipose biology Adv. Nutr. 2013 4 62 66 10.3945/an.112.003103 23319124 31. Miao W. Hu L. Scrivens P.J. Batist G. Transcriptional regulation of NF-E2 p45-related factor (NRF2) expression by the aryl hydrocarbon receptor-xenobiotic response element signaling pathway: Direct cross-talk between phase I and II drug-metabolizing enzymes J. Biol. Chem. 2005 280 20340 20348 10.1074/jbc.M412081200 15790560 32. Hou Y. Xue P. Bai Y. Liu D. Woods C.G. Yarborough K. Fu J. Zhang Q. Sun G. Collins S. Nuclear factor erythroid-derived factor 2-related factor 2 regulates transcription of CCAAT/enhancer-binding protein beta during adipogenesis Free Radic. Biol. Med. 2012 52 462 472 10.1016/j.freeradbiomed.2011.10.453 22138520 33. Christiaens V. Lijnen H.R. Angiogenesis and development of adipose tissue Mol. Cell. Endocrinol. 2010 318 2 9 10.1016/j.mce.2009.08.006 19686803 34. Hausman G.J. Richardson R.L. Adipose tissue angiogenesis J. Anim. Sci. 2004 82 925 934 15032451 35. Sarkanen J.R. Kaila V. Mannerstrom B. Raty S. Kuokkanen H. Miettinen S. Ylikomi T. Human adipose tissue extract induces angiogenesis and adipogenesis in vitro Tissue Eng. Part A 2012 18 17 25 10.1089/ten.tea.2010.0712 21902602 36. Yamada T. Kawakami S. Nakanishi N. Fat depot-specific differences in angiogenic growth factor gene expression and its relation to adipocyte size in cattle J. Vet. Med. Sci. 2010 72 991 997 10.1292/jvms.10-0031 20234108 37. Rega G. Kaun C. Demyanets S. Pfaffenberger S. Rychli K. Hohensinner P.J. Kastl S.P. Speidl W.S. Weiss T.W. Breuss J.M. Vascular endothelial growth factor is induced by the inflammatory cytokines interleukin-6 and oncostatin m in human adipose tissue in vitro and in murine adipose tissue in vivo Arterioscler. Thromb. Vasc. Biol. 2007 27 1587 1595 10.1161/ATVBAHA.107.143081 17525365 38. Wang M.L. Shih C.K. Chang H.P. Chen Y.H. Antiangiogenic activity of indole-3-carbinol in endothelial cells stimulated with activated macrophages Food Chem. 2012 134 811 820 10.1016/j.foodchem.2012.02.185 23107695 39. Broadbent T.A. Broadbent H.S. 1. The chemistry and pharmacology of indole-3-carbinol (indole-3-methanol) and 3-(methoxymethyl)indole Curr. Med. Chem. 1998 5 337 352 9756978 40. Anderton M.J. Manson M.M. Verschoyle R.D. Gescher A. Lamb J.H. Farmer P.B. Steward W.P. Williams M.L. Pharmacokinetics and tissue disposition of indole-3-carbinol and its acid condensation products after oral administration to mice Clin. Cancer Res. 2004 10 5233 5241 10.1158/1078-0432.CCR-04-0163 15297427 41. Bjeldanes L.F. Kim J.Y. Grose K.R. Bartholomew J.C. Bradfield C.A. Aromatic hydrocarbon responsiveness-receptor agonists generated from indole-3-carbinol in vitro and in vivo: Comparisons with 2,3,7,8-tetrachlorodibenzo-ρ-dioxin Proc. Natl. Acad. Sci. USA 1991 88 9543 9547 10.1073/pnas.88.21.9543 1658785 42. Chen Y.H. Riby J. Srivastava P. Bartholomew J. Denison M. Bjeldanes L. Regulation of CYP1A1 by indolo[3,2-b]carbazole in murine hepatoma cells J. Biol. Chem. 1995 270 22548 22555 10.1074/jbc.270.38.22548 7673247 43. Chang X. Tou J.C. Hong C. Kim H.A. Riby J.E. Firestone G.L. Bjeldanes L.F. 3,3′-Diindolylmethane inhibits angiogenesis and the growth of transplantable human breast carcinoma in athymic mice Carcinogenesis 2005 26 771 778 10.1093/carcin/bgi018 15661811 44. Jayakumar P. Pugalendi K.V. Sankaran M. Attenuation of hyperglycemia-mediated oxidative stress by indole-3-carbinol and its metabolite 3,3′-diindolylmethane in C57BL/6J mice J. Physiol. Biochem. 2014 70 525 534 10.1007/s13105-014-0332-5 24715233 45. Liu Y. She W. Wang F. Li J. Wang J. Jiang W. 3,3′-Diindolylmethane alleviates steatosis and the progression of NASH partly through shifting the imbalance of Treg/Th17 cells to Treg dominance Int. Immunopharmacol. 2014 23 489 498 10.1016/j.intimp.2014.09.024 25281898 46. Bradlow H.L. Michnovicz J.J. Halper M. Miller D.G. Wong G.Y. Osborne M.P. Long-term responses of women to indole-3-carbinol or a high fiber diet Cancer Epidemiol. Biomark. Prev. 1994 3 591 595 47. Rosen C.A. Bryson P.C. Indole-3-carbinol for recurrent respiratory papillomatosis: Long-term results J. Voice 2004 18 248 253 10.1016/j.jvoice.2003.05.005 15193659 48. Edgell C.J. McDonald C.C. Graham J.B. Permanent cell line expressing human factor VIII-related antigen established by hybridization PNAS 1983 80 3734 3737 10.1073/pnas.80.12.3734 6407019

PMC005xxxxxx/PMC5000655.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081257ijms-17-01257ReviewDocosahexaenoic Acid Induces Oxidative DNA Damage and Apoptosis, and Enhances the Chemosensitivity of Cancer Cells Song Eun Ah Kim Hyeyoung *Sáez Guillermo T. Academic EditorDepartment of Food and Nutrition, Brain Korea 21 PLUS Project, College of Human Ecology, Yonsei University, Seoul 03722, Korea; [email protected]* Correspondence: [email protected]; Tel.: +82-2-2123-312503 8 2016 8 2016 17 8 125731 3 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The human diet contains low amounts of ω-3 polyunsaturated fatty acids (PUFAs) and high amounts of ω-6 PUFAs, which has been reported to contribute to the incidence of cancer. Epidemiological studies have shown that a high consumption of fish oil or ω-3 PUFAs reduced the risk of colon, pancreatic, and endometrial cancers. The ω-3 PUFA, docosahexaenoic acid (DHA), shows anticancer activity by inducing apoptosis of some human cancer cells without toxicity against normal cells. DHA induces oxidative stress and oxidative DNA adduct formation by depleting intracellular glutathione (GSH) and decreasing the mitochondrial function of cancer cells. Oxidative DNA damage and DNA strand breaks activate DNA damage responses to repair the damaged DNA. However, excessive DNA damage beyond the capacity of the DNA repair processes may initiate apoptotic signaling pathways and cell cycle arrest in cancer cells. DHA shows a variable inhibitory effect on cancer cell growth depending on the cells’ molecular properties and degree of malignancy. It has been shown to affect DNA repair processes including DNA-dependent protein kinases and mismatch repair in cancer cells. Moreover, DHA enhanced the efficacy of anticancer drugs by increasing drug uptake and suppressing survival pathways in cancer cells. In this review, DHA-induced oxidative DNA damage, apoptotic signaling, and enhancement of chemosensitivity in cancer cells will be discussed based on recent studies. docosahexaenoic acidoxidative DNA damageapoptosischemosensitivitycancer cells ==== Body 1. Introduction The human diet contains low amounts of ω-3 polyunsaturated fatty acids (PUFAs) and high amounts of ω-6 PUFAs, which might contribute to increased cancer incidence. In a previous study, breast cancer risk was positively associated with the ratio of dietary ω-6 to ω-3 PUFAs and inversely associated with the dietary intake of long-chain ω-3 PUFAs [1]. Studies conducted in humans have demonstrated that a high consumption of fish oil reduced the risk of cancer. In a Japanese cohort study, there was an inverse relationship between the risk of distal colon and pancreatic cancers and ω-3 PUFA consumption [2,3]. In a Scottish study, a significant reduction in colon cancer risk was associated with increased intake of total ω-3 PUFAs as well as eicosapentaenoic acid (EPA) or docosahexaenoic acid (DHA), consumed separately [4]. Chavarro et al. [5] analyzed the blood fatty acid levels of 476 men with prostate cancer during a 13-year follow-up and their matched controls. They found that whole blood levels of all long-chain ω-3 PUFAs were inversely related to overall prostate cancer risk. The blood levels of γ-linolenic and dihomo-γ-linolenic acids, fatty acids generated from the metabolism of linoleic acid, were directly associated with prostate cancer. In a human study evaluating the association between endometrial cancer risk and the intake of fatty acids and fish, the ratio of ω-6 to ω-3 PUFAs was inversely associated with the risk of endometrial cancer. Therefore, the dietary intake of EPA and DHA from foods and supplements may protect against the development of endometrial cancer [6]. Furthermore, regarding the relationship between the incidences of colorectal cancer (CRC) types such as proximal colon and distal colon cancer and ω-3 PUFA intake, high ω-3 PUFA was associated with a lower risk of proximal colon cancer and an unaltered or even increased risk of distal colon cancer [7,8]. Between 10% and 15% of CRCs display microsatellite instability (MSI) with predominance in the proximal colon [9,10,11], and the MSI was found to be induced by the loss of DNA mismatch repair (MMR) activity [12]. Song et al. [13] demonstrated that ω-3 PUFA intake inhibited inflammatory pathways associated with the development of tumors that arise from defective MMR. Therefore, ω-3 PUFA intake appears to be associated with a lower risk for MSI-high CRC but not microsatellite-stable (MSS) tumors. They suggested a potential role for ω-3 PUFAs in protecting against CRC through MMR. Natural, synthetic, and biological agents have been developed to reduce or delay the occurrence of malignancy [14]. Certain agents trigger DNA damage followed by cancer cell death, which is critical for the maintenance of proper physiological processes including tissue homeostasis and immune function regulation [15,16,17]. However, dysregulation of cell death is often observed in cancer cells [18,19]. Therefore, the induction of cancer cell death is pivotal in cancer treatment, which makes it an important strategy in cancer therapy. The ω-3 PUFAs play vital roles in the normal growth and development of various cells and tissues [20,21]. DHA is one of the longest and most unsaturated fatty acids found in biological systems, with 22 carbons and six double bonds. DHA has been shown to inhibit cancer cell proliferation and induce death of some cancer cells [22,23,24,25]. DHA induces oxidative stress, DNA adduct formation, and DNA damage in various cancer cells [26,27], showing selective cytotoxicity against various types of cancer but not normal cells [28,29]. DHA favorably modulates anticancer treatment through its incorporation into cellular membranes, induction of oxidative stress, interaction with cellular signaling mediators, including cyclooxygenase-2, nuclear factor-κB, mitogen-activated protein kinases, and peroxisome proliferator-activated receptor-γ (PPARγ), and ability to increase the sensitivity of anticancer drugs in in vitro studies [30,31]. For patients with advanced non-small cell lung cancer undergoing platinum-based chemotherapy (carboplatin with vinorelbine or gemcitabine), supplementation with fish oil (2.5 g EPA + DHA/day) increased the chemotherapy efficacy without affecting the toxicity profile compared with the standard of care [32]. Bougnoux et al. [33] studied the effect of daily supplementation with 1.8 g DHA on the efficiency of an anthracycline-based chemotherapy regimen in patients with breast cancer. DHA during chemotherapy was devoid of adverse side effects and improved the outcome of chemotherapy when it was highly incorporated in the body. They suggested that DHA has the potential to specifically chemosensitize tumors. A pilot phase II clinical trial investigating the treatment of patients with metastatic breast cancer with dietary DHA (1.8 g/day) and an anthracycline-based chemotherapy revealed an improved survival rate, especially in a patient subpopulation with a high incorporation of DHA in the plasma. DHA had a specific chemosensitizing effect on metastases that was not observed in non-tumor tissues [34]. Although DHA shows chemopreventive activity as mentioned above, its potential anticancer effects are suggested in this review based on its oxidative stress–induced DNA damage, apoptosis, and the enhancement of chemosensitivity in cancer cells. 2. DHA Induces Oxidative Stress and Oxidative DNA Damage to Cancer Cells High levels of reactive oxygen species (ROS) such as superoxide anions and hydroxyl radicals may induce oxidative stress in cancer cells. Superoxide anions react with nitric oxide (NO) to form peroxynitrite, a reactive nitrogen species (RNS). Both ROS and RNS induce DNA damage and DNA strand breaks. Merendino et al. [35] showed that DHA induced oxidative stress by depleting intracellular glutathione (GSH) in Paca-44 pancreatic cancer cells. They suggested that GSH depletion occurred via active GSH extrusion in cancer cells, since the inhibition of GSH efflux by treatment with the specific inhibitors of carrier-mediated GSH extrusion, cystathionine or methionine, completely reversed the apoptosis. Both EPA and DHA induced ROS accumulation and caspase-8–dependent apoptosis in breast cancer MCF-7 [36] and pancreatic cancer (MIA-PaCa-2 and Capan-2) cells in vitro. The growth of MIA-PaCa-2 human pancreatic cancer xenografts in athymic nude mice was suppressed by 5% fish oil supplementation, which induced oxidative stress and cell death [37]. Shin et al. [38] showed that DHA increased the cellular ROS levels and apoptosis of PC3 and DU145 prostate cancer cells expressing mutant p53. Pretreatment with the antioxidant N-acetyl-cysteine completely blocked the DHA-induced reduction in cell viability and reduced the elevated poly(ADP-ribose) polymerase cleavage caused by DHA. These findings suggest that DHA induces apoptosis by triggering intracellular ROS accumulation in these cells. Mitochondria are a major source of intracellular ROS in mammalian cells. DHA has been shown to induce intracellular ROS by promoting the generation of mitochondrial ROS in certain cancer cells [39,40,41]. DHA induced excessive mitochondrial ROS accumulation in PA-1 human ovarian cancer cells [39] and human papillomavirus (HPV)-infected HeLa and SiHa human cervical cancer cells [40,41]. In both studies, the DHA-induced mitochondrial ROS overproduction was accompanied by mitochondrial malfunction, evidenced by the loss of mitochondrial membrane potential following the addition of DHA. Since the oxygen consumption rate was decreased by DHA treatment, DHA might trigger excessive mitochondrial ROS generation by disrupting the mitochondrial electron transport chain from producing ROS in cancer cells. DHA is oxidized by ROS, resulting in the generation of electrophilic compounds that have the potential to form DNA adducts, thereby initiating apoptotic responses in cancer cells [42]. DHA increased the formation of acrolein-derived 1,N2-propanodeoxyguanosine (Acr-dG), a major DNA adduct formed from oxidized DHA, and induced apoptosis. Interestingly, it was only after the Acr-dG reached a certain threshold level, which was beyond the capacity of nucleotide excision repair (NER) and other DNA repair pathways, that the cells underwent cell cycle arrest and induced apoptotic signaling pathways. In colon cancer HT-29 cells, DHA caused DNA adduct formation and cell cycle arrest in the G1 phase [43]. Since cancer cells produce relatively large amounts of ROS, DHA can be easily oxidized to form DNA adducts and cause DNA damage, which may result in the apoptosis of cancer cells. 3. DHA Induces Apoptotic Signaling and Affects DNA Damage Response in Cancer Cells When DNA damage occurs from diverse factors including oxidative stress and exogenous sources, cellular processes lead to cell cycle arrest and DNA damage repair as a defense mechanism [44]. Excess DNA damage, beyond the capacity of DNA repair, results in apoptotic cell death. Defects in the apoptotic pathway are widely observed in cancer cells. The pro-apoptotic effect of DHA on cancer cells has been widely documented both in vitro and in vivo in various types of cancer cells such as colon HT-29 [45], gastric AGS [46], pancreatic Paca-44 [47], lung 549 [48], and colorectal stem-like [49] cancer cells. DHA induced apoptosis in the human colon adenocarcinoma cell line HCT116, which carries an activating mutation of the β-catenin gene (CTNNB1), and SW480 cells with wild-type CTNNB1 [50]. The proposed mechanisms of DHA did not involve the modification of the transcription of β-catenin, but induced ubiquitin-dependent proteasomal degradation of the protein was suggested [50]. In breast cancer lines MCF-7 and Hs578T, DHA pretreatment attenuated 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced cell migration as well as protein kinase C δ (PKCδ), Wnt-1, and β-catenin expression. A study suggested that the anti-metastatic potential of DHA is partly attributable to its suppression of TPA-activated PKCδ and Wnt-1 signaling [51]. Xue et al. [52] conducted an investigation in BABL/c mice bearing breast cancer tumors. A 5% fish oil–supplemented diet for a period of 30 days significantly reduced the growth of 4T1 mouse breast cancer cells by downregulating β-catenin in tumor tissues with a notable increase in apoptosis. Fluckiger et al. [53] showed that DHA triggered apoptosis in the HCT-116 and HCT-8 human colorectal cancer cell lines in an autocrine tumor necrosis factor (TNF)-α–dependent manner. They demonstrated that DHA stimulated nuclear accumulation of Forkhead box O3 which binds to the microRNA-21 promoter. Therefore, DHA induced the mRNA expression of TNF-α through post-transcriptional regulation by the downregulation of microRNA-21 expression. A recent study compared DHA-induced stress responses in two human colon cancer cell lines, SW620 and Caco-2 [54]. DHA inhibited the growth of SW620 cells at early time points while that of the Caco-2 cells was unaffected by the same treatment. Furthermore, oxidative stress was induced in both cell lines, although at different time points and to varying extents. Therefore, the anticancer activity of DHA may differ depending on the molecular properties of the cancer cells. Since DHA induces oxidative DNA damage in cancer cells, the DNA damage response may be affected by DHA in cancer cells. The p53 protein is a sensor at the center of the DNA damage response and is activated in response to multiple types of DNA damage. DHA was effective in the growth suppression of ovarian TOV-21G and breast MCF-10A cancer cells, which may be partly mediated by p53 activation [55,56]. Wan et al. [55] reported that EPA/DHA induced PPARγ and p53 overexpression in TOV-21G cells and the induction of p53 by EPA/DHA was abolished by the PPARγ antagonist GW9662. They found that the effect of DHA was more potent than that of EPA. The growth suppression of TOV-21G cells may be partly mediated by PPARγ and p53 activation during DHA treatment. Rescigno et al. [56] demonstrated that DHA differentially regulated the activation of extracellular signal-regulated kinase 1/2 (ERK1/2) and signal transducers and activators of transcription 3 (STAT3) pathways as well as cell cycle regulators such as p21 and p53 in breast cancer cell lines. DHA selectively arrested non-tumoral MCF-10A breast cells in the G0/G1 cell cycle phase by activation of p21 and p53. DHA induced cell death in highly transformed SK-BR-3 breast cells with the reduction of ERK1/2 and STAT3 phosphorylation, but only slightly affected the cell cycle in MCF-7 breast cancer cells with a transformation degree lower than that in the SK-BR-3 cells. These studies suggest DHA has a variable inhibitory effect on cancer cell growth that depends on the molecular properties and the degree of malignancy in each clinical case. Kato et al. [57] compared the effect of DHA on the growth of the human colon carcinoma COLO 205 cells carrying wild-type p53 and WiDr colon carcinoma cells containing mutated p53 (His237). DHA inhibited the growth of COLO 205 cells by 81% and WiDr cells by 42%. DHA inhibited the proliferation of WiDr cells by 41%, but not that of COLO 205 cells. DHA arrested the cell cycle at the G0/G1 and G2/M phases in WiDr and COLO 205 cells, respectively. DHA induced the apoptosis of COLO 205 but not WiDr cells. Although DHA showed differential effects on cell proliferation, cell cycle arrest phase, and apoptosis in colon cancer cells depending on p53 status, it is evident that DHA inhibits cancer growth by p53-dependent and -independent pathways [57]. Experimental studies have demonstrated that inflammation inactivates MMR function and increases mutation rates [58,59,60]. The ω-6 PUFA-derived pro-inflammatory product prostaglandin E2 has been shown to silence DNA repair genes by enhancing DNA methylation to promote colonic tumor growth [61]. Since there is a strong inverse association of ω-3 PUFA with proximal colon cancer associated with defective MMR, the protective effect of ω-3 PUFAs against CRC is suggested to be mediated through DNA mismatch repair. On the other hand, oxidative stress inactivates MMR gene expression by mutation [59,62] or epigenetic silencing and may directly damage MMR proteins [63]. Since a high amount of DHA produces oxidative stress in cancer cells, DHA may inactivate MMR in cancer cells by ROS-dependent pathways. 4. DHA Increases Chemosensitivity of Cancer Cells DHA has been found to enhance the activity of several anticancer drugs through an oxidative mechanism. Viet et al. [64] demonstrated that DHA increased the sensitivity of the breast cancer cell line MDA-MB-231 to doxorubicin, but it did not affect MCF-7 cells. In the MDA-MB-231 cells, DHA decreased the activity of cytosolic GSH peroxidase, an enzyme that protects against hydrogen and lipid peroxides. This modification of the GSH peroxidase response in the DHA-supplemented rats was associated with increased tumor sensitivity to anthracyclines. Therefore, the loss of the GSH peroxidase response due to oxidative stress in transformed cells may account for the ability of peroxidizable agents such as DHA to enhance tumor sensitivity to ROS-generating anticancer drugs. Wang et al. [65] demonstrated that combined treatment with DHA and the anticancer drug etoposide exhibited an additive effect on brain tumor cells. Compared to etoposide used alone, the combination of DHA and etoposide suppressed the expression of the DNA-dependent protein kinase catalytic subunit (DNA-PKcs), which is involved in non-homologous end joining (NHEJ) repair pathways [66]. Recently, Chauvin et al. [67] investigated whether DHA increased tumor sensitivity to docetaxel by downregulating the survival pathways. Taxanes induce drug resistance by increasing the activity of phosphoinositide 3-kinase (PI3K)/Akt and ERK signaling pathways, which promote survival and cell growth of human cancer cells. In docetaxel-treated MDA-MB-231 cells, ERK1/2 phosphorylation and protein kinase C (PKC) activity were increased compared to that in untreated cells. In DHA-supplemented cells, docetaxel was unable to increase PKC levels in the membrane and nuclear fractions and diminished ERK1/2 phosphorylation, resulting in increased docetaxel efficacy. DHA treatment affected the expression of the target proteins of cancer therapy such as p21, CDC25 homolog, and cyclin-dependent kinase 1, which led to cell cycle arrest at both the G1 and G2 phases in chemotherapy-resistant colon cancer SW620 cells [68]. The activity of P-glycoprotein (Pgp) and multidrug resistance-related protein 1 (MRP1), two membrane transporters involved in the multidrug resistance of colon cancer, is increased by high amounts of cholesterol in the plasma membrane and detergent-resistant membranes (DRMs). Multidrug-resistant (MDR) tumors, which overexpressed Pgp and MRP1, had a dysregulated cholesterol metabolism due to the lower expression of ubiquitin E3 ligase [69]. The ω-3 PUFAs were incorporated in the DRMs of MDR cells and they reduced the Pgp and MRP1 content of the DRMs, resulting in decreased transporter activity and restoration of the antitumor effects of the chemotherapeutic drugs. The ω-3 PUFA conjugates of anticancer drugs have been the focus of increasing attention due to their enhancement of curative effects, reduction of side effects, and tumor-targeting abilities in preclinical studies. The ω-3 PUFAs are readily incorporated into the lipid bilayer of tumor cells and, therefore, can be used as a useful carrier to increase the therapeutic efficacy of anticancer drugs. Wang et al. [34] reported that the DHA-paclitaxel has received phase III clinical trial approval. Further research is necessary to determine specific mechanisms by which ω-3 PUFAs increase chemotherapy efficacy and to determine the optimal cellular/membrane levels of ω-3 PUFAs that increase the chemosensitivity and therapeutic efficacy of anticancer drugs. 5. Conclusions The induction of cancer cell death is important in cancer therapy. Recent studies have highlighted DHA as an effective anticancer agent because it induces apoptosis and enhances the drug sensitivity of cancer cells but not normal cells. In the cancer cell environment, DHA is oxidized to form DNA adducts such as Acr-dG and induces oxidative DNA damage, which triggers apoptosis of some cancer cells. DNA damage is repaired by DNA repair processes such as NER, MMR, and NHEJ including DNA-PKcs. However, excess DNA damage, induced by oxidized DHA, beyond the capacity of the DNA repair processes, initiates apoptotic signaling in cancer cells. Moreover, DHA suppresses the expression of DNA-PKcs in some cancer cells. DHA inactivates MMR in some cancer cells by ROS-dependent pathways (Figure 1). Therefore, DHA treatment or supplementation may be beneficial for cancer prevention and therapy. In addition, DHA increases tumor sensitivity to anticancer drugs by enhancing drug uptake and inhibiting survival signaling in cancer cells, as well as reducing Pgp and MRP1 in MDR tumors. Acknowledgments This work was supported by a grant from the National Research Foundation (NRF) of Korea, which is funded by the Korean Government (NRF-2015 R1A2A2A01004855). Author Contributions Eun Ah Song performed the comprehensive literature search and wrote the paper. Hyeyoung Kim performed the initial literature search as well as revised and edited the paper. All authors agree with the edited version. Conflicts of Interest The authors declare no conflicts of interest. Figure 1 Proposed mechanism of docosahexaenoic acid (DHA)-induced DNA damage response in human cancer cells. DHA initially induces glutathione (GSH) extrusion and mitochondrial dysfunction, which increases reactive oxygen species (ROS) in cancer cells. DHA is oxidized, leading to DNA adduct formation and oxidative DNA damage which triggers cell cycle arrest and apoptosis in p53-dependent and p53-independent pathways. DNA damage is repaired by DNA repair processes such as nucleotide excision repair (NER), mismatch repair (MMR), and non-homologous end joining (NHEJ) mediated by DNA-dependent protein kinase catalytic subunit (DNA-PKcs). After the damaged DNA is repaired, the cells survive. However, excess DNA damage, induced by oxidized DHA, beyond the capacity of the DNA repair processes, initiates apoptotic signaling in cancer cells. Moreover, DHA suppresses the expression of DNA-PKcs and inactivates MMR in some cancer cells. ==== Refs References 1. Bassett J.K. Hodge A.M. English D.R. MacInnis R.J. Giles G.G. Plasma phospholipids fatty acids, dietary fatty acids, and breast cancer risk Cancer Causes Control 2016 27 1 15 10.1007/s10552-016-0753-2 26573845 2. Kimura Y. Kono S. Toyomura K. Nagano J. Mizoue T. Moore M.A. Mibu R. Tanaka M. Kakeji Y. Maehara Y. Meat, fish and fat intake in relation to subsite‐specific risk of colorectal cancer: The Fukuoka Colorectal Cancer Study Cancer Sci. 2007 98 590 597 10.1111/j.1349-7006.2007.00425.x 17425596 3. Hidaka A. Shimazu T. Sawada N. Yamaji T. Iwasaki M. Sasazuki S. Inoue M. Tsugane S. Japan Public Health Center-Based Prospective Study Group Fish, n-3 PUFA consumption, and pancreatic cancer risk in Japanese: A large, population-based, prospective cohort study Am. J. Clin. Nutr. 2015 102 1490 1497 10.3945/ajcn.115.113597 26537936 4. Theodoratou E. McNeill G. Cetnarskyj R. Farrington S.M. Tenesa A. Barnetson R. Porteous M. Dunlop M. Campbell H. Dietary fatty acids and colorectal cancer: A case-control study Am. J. Epidemiol. 2007 166 181 195 10.1093/aje/kwm063 17493949 5. Chavarro J. Stampfer M. Li H. Campos H. Kurth T. Ma J. A prospective study of polyunsaturated fatty acid levels in blood and prostate cancer risk Cancer Epidemiol. Biomark. Prev. 2007 16 1364 1370 10.1158/1055-9965.EPI-06-1033 17585059 6. Arem H. Neuhouser M.L. Irwin M.L. Cartmel B. Lu L. Risch H. Mayne S.T. Yu H. Omega-3 and omega-6 fatty acid intakes and endometrial cancer risk in a population-based case-control study Eur. J. Nutr. 2013 52 1251 1260 10.1007/s00394-012-0436-z 22915050 7. Song M. Chan A.T. Fuchs C.S. Ogino S. Hu F.B. Mozaffarian D. Ma J. Willett W.C. Giovannucci E.L. Wu K. Dietary intake of fish, omega-3 and omega-6 fatty acids and risk of colorectal cancer: A prospective study in U.S. men and women Int. J. Cancer 2014 135 2413 2423 10.1002/ijc.28878 24706410 8. Sasazuki S. Inoue M. Iwasaki M. Sawada N. Shimazu T. Yamaji T. Takachi R. Tsugane S. Japan Public Health Center-Based Prospective Study Group Intake of n-3 and n-6 polyunsaturated fatty acids and development of colorectal cancer by subsite: Japan Public Health Center-based prospective study Int. J. Cancer 2011 129 1718 1729 10.1002/ijc.25802 21120874 9. Thibodeau S.N. Bren G. Schaid D. Microsatellite instability in cancer of the proximal colon Science 1993 260 816 819 10.1126/science.8484122 8484122 10. Ward R. Meagher A. Tomlinson I. O’Connor T. Norrie M. Wu R. Hawkins N. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer Gut 2001 48 821 829 10.1136/gut.48.6.821 11358903 11. Popat S. Hubner R. Houlston R.S. Systematic review of microsatellite instability and colorectal cancer prognosis J. Clin. Oncol. 2005 23 609 618 10.1200/JCO.2005.01.086 15659508 12. Boland C.R. Goel A. Microsatellite instability in colorectal cancer Gastroenterology 2010 138 2073 2087 10.1053/j.gastro.2009.12.064 20420947 13. Song M. Nishihara R. Wu K. Qian Z.R. Kim S.A. Sukawa Y. Mima K. Inamura K. Masuda A. Yang J. Marine ω-3 polyunsaturated fatty acids and risk of colorectal cancer according to microsatellite instability J. Natl. Cancer Inst. 2015 107 10.1093/jnci/djv007 25810492 14. Thun M.J. DeLancey J.O. Center M.M. Jemal A. Ward E.M. The global burden of cancer: Priorities for prevention Carcinogenesis 2010 31 100 110 10.1093/carcin/bgp263 19934210 15. Czabotar P.E. Lessene G. Strasser A. Adams J.M. Control of apoptosis by the BCL-2 protein family: Implications for physiology and therapy Nat. Rev. Mol. Cell Biol. 2014 15 49 63 10.1038/nrm3722 24355989 16. Adams J.M. Cory S. Bcl-2-regulated apoptosis: Mechanism and therapeutic potential Curr. Opin. Immunol. 2007 19 488 496 10.1016/j.coi.2007.05.004 17629468 17. Baig S. Seevasant I. Mohamad J. Mukheem A. Huri H.Z. Kamarul T. Potential of apoptotic pathway-targeted cancer therapeutic research: Where do we stand? Cell Death Dis. 2016 7 e2058 10.1038/cddis.2015.275 26775709 18. Liu J.J. Lin M. Yu J.Y. Liu B. Bao J.K. Targeting apoptotic and autophagic pathways for cancer therapeutics Cancer Lett. 2011 300 105 114 10.1016/j.canlet.2010.10.001 21036469 19. Evan G.I. Vousden K.H. Proliferation, cell cycle and apoptosis in cancer Nature 2001 411 342 348 10.1038/35077213 11357141 20. Skender B. Hyrslova Vaculova A. Hofmanova J. Docosahexaenoic fatty acid (DHA) in the regulation of colon cell growth and cell death: A review Biomed. Pap. Med. Fac. Univ. Palacky Olomouc Czechoslov. 2012 156 186 199 10.5507/bp.2012.093 23069883 21. Calder P.C. Marine ω-3 fatty acids and inflammatory processes: Effects, mechanisms and clinical relevance Biochim. Biophys. Acta 2015 1851 469 484 10.1016/j.bbalip.2014.08.010 25149823 22. Berquin I.M. Edwards I.J. Chen Y.Q. Multi-targeted therapy of cancer by ω-3 fatty acids Cancer Lett. 2008 269 363 377 10.1016/j.canlet.2008.03.044 18479809 23. Serini S. Piccioni E. Merendino N. Calviello G. Dietary polyunsaturated fatty acids as inducers of apoptosis: Implications for cancer Apoptosis 2009 14 132 152 10.1007/s10495-008-0298-2 19130233 24. Gleissman H. Johnsen J.I. Kogner P. ω-3 fatty acids in cancer, the protectors of good and the killers of evil? Exp. Cell Res. 2010 316 1365 1373 10.1016/j.yexcr.2010.02.039 20211172 25. Vaughan V.C. Hassing M.R. Lewandowski P.A. Marine polyunsaturated fatty acids and cancer therapy Br. J. Cancer 2013 108 486 492 10.1038/bjc.2012.586 23299528 26. Zajdel A. Wilczok A. Tarkowski M. Toxic effects of n-3 polyunsaturated fatty acids in human lung A549 cells Toxicol. In Vitro 2015 30 486 491 10.1016/j.tiv.2015.09.013 26381084 27. Zhelev Z. Ivanova D. Lazarova D. Aoki I. Bakalova R. Saga T. Docosahexaenoic acid sensitizes leukemia lymphocytes to barasertib and everolimus by ROS-dependent mechanism without affecting the level of ROS and viability of normal lymphocytes Anticancer Res. 2016 36 1673 1682 27069145 28. Nikolakopoulou Z. Nteliopoulos G. Michael-Titus A.T. Parkinson E.K. Omega-3 polyunsaturated fatty acids selectively inhibit growth in neoplastic oral keratinocytes by differentially activating ERK1/2 Carcinogenesis 2013 34 2716 2725 10.1093/carcin/bgt257 23892603 29. Abdi J. Garssen J. Faber J. Redegeld F.A. Omega-3 fatty acids, EPA and DHA induce apoptosis and enhance drug sensitivity in multiple myeloma cells but not in normal peripheral mononuclear cells J. Nutr. Biochem. 2014 25 1254 1262 10.1016/j.jnutbio.2014.06.013 25277647 30. D’Eliseo D. Velotti F. Omega-3 fatty acids and cancer cell cytotoxicity: Implications for multi-targeted cancer therapy J. Clin. Med. 2016 5 15 10.3390/jcm5020015 26821053 31. Siddiqui R.A. Harvey K.A. Xu Z. Bammerlin E.M. Walker C. Altenburg J.D. Docosahexaenoic acid: A natural powerful adjuvant that improves efficacy for anticancer treatment with no adverse effects Biofactors 2011 37 399 412 10.1002/biof.181 22038684 32. Murphy R.A. Mourtzakis M. Chu Q.S. Baracos V.E. Reiman T. Mazurak V.C. Supplementation with fish oil increases first-line chemotherapy efficacy in patients with advanced non-small cell lung cancer Cancer 2011 117 3774 3780 10.1002/cncr.25933 21328326 33. Bougnoux P. Hajjaji N. Ferrasson M.N. Giraudeau B. Couet C. le Floch O. Improving outcome of chemotherapy of metastatic breast cancer by docosahexaenoic acid: A phase II trial Br. J. Cancer 2009 101 1011978 1011985 10.1038/sj.bjc.6605441 19920822 34. Wang J. Luo T. Li S. Zhao J. The powerful applications of polyunsaturated fatty acids in improving the therapeutic efficacy of anticancer drugs Expert Opin. Drug Deliv. 2012 9 1 7 10.1517/17425247.2011.618183 22171694 35. Merendino N. Molinari R. Loppi B. Pessina G. D’Aquino M. Tomassi G. Velotti F. Induction of apoptosis in human pancreatic cancer cells by docosahexaenoic acid Ann. N. Y. Acad. Sci. 2003 1010 361 364 10.1196/annals.1299.143 15033753 36. Kang K.S. Wang P. Yamabe N. Fukui M. Jay T. Zhu B.T. Docosahexaenoic acid induces apoptosis in MCF-7 cells in vitro and in vivo via reactive oxygen species formation and caspase 8 activation PLoS ONE 2010 5 1257 10.1371/journal.pone.0010296 20421971 37. Fukui M. Kang K.S. Okada K. Zhu B.T. EPA, an omega-3 fatty acid, induces apoptosis in human pancreatic cancer cells: Role of ROS accumulation, caspase-8 activation, and autophagy induction J. Cell. Biochem. 2013 114 192 203 10.1002/jcb.24354 22903547 38. Shin S. Jing K. Jeong S. Kim N. Song K. Heo J. Park J. Seo K. Han J. Park J. The Omega-3 polyunsaturated fatty acid DHA induces simultaneous apoptosis and autophagy via mitochondrial ROS-mediated Akt-mTOR signaling in prostate cancer cells expressing mutant p53 BioMed Res. Int. 2013 2013 1 11 10.1155/2013/568671 23841076 39. Jeong S. Jing K. Kim N. Shin S. Kim S. Song K. Heo J. Park J. Seo K. Han J. Docosahexaenoic acid-induced apoptosis is mediated by activation of mitogen-activated protein kinases in human cancer cells BMC Cancer 2014 14 1 11 10.1186/1471-2407-14-481 24383403 40. Jing K. Shin S. Jeong S. Kim S. Song K. Park J. Heo J. Seo K. Park S. Kweon G. Docosahexaenoic acid induces the degradation of HPV E6/E7 oncoproteins by activating the ubiquitin–proteasome system Cell Death Dis. 2014 5 e1524 10.1038/cddis.2014.477 25393480 41. Rohrbach S. Effects of dietary polyunsaturated fatty acids on mitochondria Curr. Pharm. Des. 2009 15 4103 4116 10.2174/138161209789909692 20041812 42. Pan J. Chung F.L. Formation of cyclic deoxyguanosine adducts from ω-3 and ω-6 polyunsaturated fatty acids under oxidative conditions Chem. Res. Toxicol. 2002 15 367 372 10.1021/tx010136q 11896684 43. Pan J. Keffer J. Emami A. Ma X. Lan R. Goldman R. Chung F.L. Acrolein-derived DNA adduct formation in human colon cancer cells: Its role in apoptosis induction by docosahexaenoic acid Chem. Res. Toxicol. 2009 22 798 806 10.1021/tx800355k 19341237 44. Roos W.P. Kaina B. DNA damage-induced cell death by apoptosis Trends Mol. Med. 2006 12 440 450 10.1016/j.molmed.2006.07.007 16899408 45. Chen Z.Y. Istfan N.W. Docosahexaenoic acid is a potent inducer of apoptosis in HT-29 colon cancer cells Prostaglandins Leukot. Essent. Fat. Acids 2000 63 301 308 10.1054/plef.2000.0218 11090257 46. Lee S.E. Lim J.W. Kim H. Activator protein-1 mediates docosahexaenoic acid-induced apoptosis of human gastric cancer cells Ann. N. Y. Acad. Sci. 2009 1171 163 169 10.1111/j.1749-6632.2009.04716.x 19723051 47. Merendino N. Loppi B. D’Aquino M. Molinari R. Pessina G. Romano C. Velotti F. Docosahexaenoic acid induces apoptosis in the human PaCa-44 pancreatic cancer cell line by active reduced glutathione extrusion and lipid peroxidation Nutr. Cancer 2005 52 225 233 10.1207/s15327914nc5202_12 16201853 48. Serini S. Trombino S. Oliva F. Piccioni E. Monego G. Resci F. Boninsegna A. Picci N. Ranelletti F.O. Calviello G. Docosahexaenoic acid induces apoptosis in lung cancer cells by increasing MKP-1 and down-regulating p-ERK1/2 and p-p38 expression Apoptosis 2008 13 1172 1183 10.1007/s10495-008-0246-1 18679798 49. Yang T. Fang S. Zhang H.X. Xu L.X. Zhang Z.Q. Yuan K.T. Xue C.L. Yu H.L. Zhang S. Li Y.F. N-3 PUFAs have antiproliferative and apoptotic effects on human colorectal cancer stem-like cells in vitro J. Nutr. Biochem. 2013 24 744 753 10.1016/j.jnutbio.2012.03.023 22854319 50. Calviello G. Resci F. Serini S. Piccioni E. Toesca A. Boninsegna A. Monego G. Ranelletti F.O. Palozza P. Docosahexaenoic acid induces proteasome-dependent degradation of beta-catenin, down-regulation of survivin and apoptosis in human colorectal cancer cells not expressing COX-2 Carcinogenesis 2007 28 1202 1209 10.1093/carcin/bgl254 17183061 51. Lii C.K. Chang J.W. Chen J.J. Chen H.W. Liu K.L. Yeh S.L. Wang T.S. Liu S.H. Tsai C.H. Li C.C. Docosahexaenoic acid inhibits 12-O -tetradecanoylphorbol-13- acetate-induced fascin-1-dependent breast cancer cell migration by suppressing the PKCδ- and Wnt-1/β-catenin-mediated pathways Oncotarget 2016 7 25162 25179 10.18632/oncotarget.7301 27036017 52. Xue M. Wang Q. Zhao J. Dong L. Ge Y. Hou L. Liu Y. Zheng Z. Docosahexaenoic acid inhibited the Wnt/β-catenin pathway and suppressed breast cancer cells in vitro and in vivo J. Nutr. Biochem. 2014 25 104 110 10.1016/j.jnutbio.2013.09.008 24290517 53. Fluckiger A. Dumont A. Derangère V. Rébé C. de Rosny C. Causse S. Thomas C. Apetoh L. Hichami A. Ghiringhelli F. Inhibition of colon cancer growth by docosahexaenoic acid involves autocrine production of TNF Oncogene 2016 10.1038/onc.2015.523 26853468 54. Pettersen K. Monsen V.T. Hakvåg Pettersen C.H. Overland H.B. Pettersen G. Samdal H. Tesfahun A.N. Lundemo A.G. Bjørkøy G. Schønberg S.A. DHA-induced stress response in human colon cancer cells - Focus on oxidative stress and autophagy Free Radic. Biol. Med. 2016 90 158 172 10.1016/j.freeradbiomed.2015.11.018 26585906 55. Wan X.H. Fu X. Ababaikeli G. Docosahexaenoic acid induces growth suppression on epithelial ovarian cancer cells more effectively than eicosapentaenoic acid Nutr. Cancer 2016 4 1 8 10.1080/01635581.2016.1142581 26942868 56. Rescigno T. Capasso A. Tecce M.F. Effect of docosahexaenoic acid on cell cycle pathways in breast cell lines with different transformation degree J. Cell. Physiol. 2016 231 1226 1236 10.1002/jcp.25217 26480024 57. Kato T. Kolenic N. Pardini R.S. Docosahexaenoic acid (DHA), a primary tumor suppressive omega-3 fatty acid, inhibits growth of colorectal cancer independent of p53 mutational status Nutr. Cancer 2007 58 178 187 10.1080/01635580701328362 17640164 58. Chang C.L. Marra G. Chauhan D.P. Ha H.T. Chang D.K. Ricciardiello L. Randolph A. Carethers J.M. Boland C.R. Oxidative stress inactivates the human DNA mismatch repair system Am. J. Physiol. Cell Physiol. 2002 283 C148 C154 10.1152/ajpcell.00422.2001 12055083 59. Gasche C. Chang C.L. Rhees J. Goe A. Boland C.R. Oxidative stress increases frameshift mutations in human colorectal cancer cells Cancer Res. 2001 61 7444 7448 11606378 60. Lee S.H. Chang D.K. Goel A. Boland C.R. Bugbee W. Boyle D.L. Firestein G.S. Microsatellite instability and suppressed DNA repair enzyme expression in rheumatoid arthritis J. Immunol. 2003 170 2214 2220 10.4049/jimmunol.170.4.2214 12574395 61. Xia D. Wang D. Kim S.H. Katoh H. DuBois R.N. Prostaglandin E2 promotes intestinal tumor growth via DNA methylation Nat. Med. 2012 18 224 226 10.1038/nm.2608 22270723 62. Jackson A.L. Chen R. Loeb L.A. Induction of microsatellite instability by oxidative DNA damage Proc. Natl. Acad. Sci. USA 1998 95 12468 12473 10.1073/pnas.95.21.12468 9770509 63. Edwards R.A. Witherspoon M. Wang K. Afrasiabi K. Pham T. Birnbaumer L. Lipkin S.M. Epigenetic repression of DNA mismatch repair by inflammation and hypoxia in inflammatory bowel disease-associated colorectal cancer Cancer Res. 2009 69 6423 6429 10.1158/0008-5472.CAN-09-1285 19638594 64. Vibet S. Goupille C. Bougnoux P. Steghens J.P. Goré J. Mahéo K. Sensitization by docosahexaenoic acid (DHA) of breast cancer cells to anthracyclines through loss of glutathione peroxidase (GPx1) response Free Radic. Biol. Med. 2008 44 1483 1491 10.1016/j.freeradbiomed.2008.01.009 18267129 65. Wang F. Bhat K. Doucette M. Zhou S. Gu Y. Law B. Liu X. Wong E.T. Kang J.X. Hsieh T.C. Docosahexaenoic acid (DHA) sensitizes brain tumor cells to etoposide-induced apoptosis Curr. Mol. Med. 2011 11 503 511 10.2174/156652411796268740 21663587 66. Morio T. Kim H. Ku, Artemis, and ataxia-telangiectasia-mutated: Signalling networks in DNA damage Int. J. Biochem. Cell Biol. 2008 40 598 603 10.1016/j.biocel.2007.12.007 18243767 67. Chauvin L. Goupille C. Blanc C. Pinault M. Domingo I. Guimaraes C. Bougnoux P. Chevalier S. Mahéo K. Long chain n-3 polyunsaturated fatty acids increase the efficacy of docetaxel in mammary cancer cells by downregulating Akt and PKCε/δ-induced ERK pathways Biochim. Biophys. Acta 2016 1861 380 390 10.1016/j.bbalip.2016.01.012 26821209 68. Slagsvold J.E. Pettersen C.H. Størvold G.L. Follestad T. Krokan H.E. Schønberg S.A. DHA alters expression of target proteins of cancer therapy in chemotherapy resistant SW620 colon cancer cells Nutr. Cancer 2010 62 611 621 10.1080/01635580903532366 20574922 69. Gelsomino G. Corsetto P.A. Campia I. Montorfano G. Kopecka J. Castella B. Gazzano E. Ghigo D. Rizzo A.M. Riganti C. ω-3 Fatty acids chemosensitize multidrug resistant colon cancer cells by down-regulating cholesterol synthesis and altering detergent resistant membranes composition Mol. Cancer 2013 12 137 10.1186/1476-4598-12-137 24225025

PMC005xxxxxx/PMC5000656.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081258ijms-17-01258ReviewBioactivities and Health Benefits of Wild Fruits Li Ya 1Zhang Jiao-Jiao 1Xu Dong-Ping 1Zhou Tong 1Zhou Yue 1Li Sha 2Li Hua-Bin 13*Battino Maurizio Academic Editor1 Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-Sen University, Guangzhou 510080, China; [email protected] (Y.L.); [email protected] (J.-J.Z.); [email protected] (D.-P.X.); [email protected] (T.Z.); [email protected] (Y.Z.)2 School of Chinese Medicine, The University of Hong Kong, Hong Kong, China; [email protected] South China Sea Bioresource Exploitation and Utilization Collaborative Innovation Center, Sun Yat-Sen University, Guangzhou 510006, China* Correspondence: [email protected]; Tel.: +86-20-873-323-9104 8 2016 8 2016 17 8 125824 5 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Wild fruits are exotic or underutilized. Wild fruits contain many bioactive compounds, such as anthocyanins and flavonoids. Many studies have shown that wild fruits possess various bioactivities and health benefits, such as free radical scavenging, antioxidant, anti-inflammatory, antimicrobial, and anticancer activity. Therefore, wild fruits have the potential to be developed into functional foods or pharmaceuticals to prevent and treat several chronic diseases. In the present article, we review current knowledge about the bioactivities and health benefits of wild fruits, which is valuable for the exploitation and utilization of wild fruits. wild fruitbioactivityantioxidantanticanceranti-inflammatory ==== Body 1. Introduction Fruits and vegetables, containing abundant dietary fiber, vitamins, and minerals, in particular large amounts of phytochemicals [1,2,3,4,5,6,7], are recommended by nutritionists because of their health benefits [8,9]. Phytochemicals in these natural products are considered to be responsible for positive health outcomes. Particularly, it is widely noted that plants produce a great deal of antioxidants to combat the oxidative stress induced by oxygen and light in the natural environment [10]. Oxidative stress performs an essential role in multiple chronic diseases [11,12,13]. Therefore, antioxidants in fruit and vegetables have been extensively explored for their effects on several diseases. Epidemiological and nutritional studies suggested that the higher one’s fruit and vegetable consumption, the lower the incidence of chronic diseases such as coronary heart problems, cancer, and Alzheimer’s disease [14,15]. Wild fruits are fruits of wild plants, and are often exotic, underutilized, or less known. Many wild fruits are safe to consume, and some have been developed as medicines. Due to different genotypes and environmental concerns, wild fruits contain rich phytochemicals such as anthocyanin and flavonoids. Therefore, wild fruits are often considered to be healthy foods. In recent years, wild fruits have attracted increasing attention, and accumulative investigations have been performed for their bioactive effects, such as antioxidant, antimicrobial anti-inflammatory, and anticancer effects. These studies pointed out that wild fruits could have the potential to prevent and treat some chronic diseases. This review summarizes the bioactivities and health benefits of wild fruits. 2. Bioactivities of Wild Fruits 2.1. Antioxidant Activity Free radicals are normally produced as a byproduct of cellular metabolism. Free radicals are capable of killing bacteria, damaging biomolecules, provoking immune responses, activating oncogenes, causing atherogenesis, and enhancing the ageing process [16]. The most important classes of radical species generated in living systems are reactive oxygen and nitrogen species (ROS and RNS). The excessive production of ROS and RNS could play a pivotal part in many human chronic diseases, including atherosclerosis, diabetes mellitus, cancer, rheumatoid arthritis, cataract, and Parkinson’s disease [17]. Various natural products have been proved to have antioxidant activities, such as fruits, vegetables, edible flowers, cereal grains, wine, herbal plants, and their tea infusions [18,19,20,21,22,23,24,25,26]. Therefore, natural resources of antioxidants have been considered as quite important. There have been several experiments both in vivo and in vitro proving that many wild fruits possess antioxidant activities, such as wild blueberries, wild apples, and wild hawthorn fruits. 2.1.1. In Vitro Studies Several studies have evaluated the antioxidant capacity of a certain species of wild fruit. The underutilized wild berry fruit Prunus mahaleb showed strong antioxidant activity [27]. The results of oxygen radical absorption capacity (ORAC) and 2,2′-azinobis-3-ethylbenzothiazoline-6-sulphonate (ABTS·+, expressed as trolox equivalent antioxidant capacity (TEAC) value) assays were 150 and 45 mmol Trolox equivalents/kg fresh weight, respectively. Furthermore, the P. mahaleb fruit had high anthocyanin content, which was comparable to that of some reported superfruits (bilberries and blackcurrants). Moreover, Araca-pera (Psidium acutangulum), an exotic guava fruit from the Amazon, was analyzed for antioxidant properties by 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical, ABTS free radical scavenging capacity (24.96 ± 0.75, 90.57 ± 0.63 mg of vitamin C/100 g fresh fruit, respectively), and cell-based assays (76%–100%) [28]. Results indicated that this guava fruit could be developed into functional foods for the prevention of chronic diseases due to its antioxidant activity. In another study, the antioxidant activities of water, ethyl acetate, acetone, and methanol extracts from the wild Sorbus torminalis fruit were assessed by DPPH, ABTS, superoxide anion radicals scavenging, and ferric reducing antioxidant power (FRAP) assays. The antioxidant activity and total phenolic concentration were both ranked as water > ethyl acetate > acetone > methanol extracts [29]. Another edible wild fruit, Ziziphus mistol, was analyzed for its antioxidant activity. All extracts showed strong antioxidant activity. As a hydrogen or electron donor, the ethanol extract (EME) was significantly more effective than the aqueous one (AME); when scavenging hydroxyl and superoxide radicals, AME was significantly more effective than EME. In addition, a dose-dependent relationship (R2 > 0.90) was found between polyphenols content and antioxidant capacity. These results suggested that consumption of Ziziphus mistol fruit could be encouraged due to its antioxidant activity [30]. The pulp of wild cherimoya fruits (Annona cherimola) was also assessed for the antioxidant capacities of its methanol, ethanol, and dimethyl formamide extracts. The three extracts all showed strong free radical capturing and antioxidant activities. Among them, the dimethyl formamide extract showed the highest DPPH and ABTS scavenging and FRAP activities, and the ethanol extract showed the strongest anti-lipid peroxidation activity [31]. In another study, a crude extract of Myrica esculenta fruit was assessed for antioxidant properties. Results showed that the extracts exhibited considerable antioxidant potential based on data from DPPH, ABTS, and FRAP assays. Moreover, the antioxidant capacity was positively correlated with total phenolic and total flavonoids contents [32]. The wild bilberry (Vaccinium meridionale) is an edible fruit from Colombia. Garzon et al. evaluated its antioxidant activity and the results of ABTS and FRAP assays proved its strong antioxidant activity [33]. In another study, the antioxidant activity of fruit from wild Lycium ruthenicum, a nutritional food that has been used in traditional Chinese medicine, was evaluated. The methanol extracts exhibited high antioxidant activity in ABTS, DPPH, and FRAP assays [34]. In addition, hydrophilic extracts of wild acerola (Malpighia emarginata) pulps and juices were analyzed for antioxidant activities. Results of DPPH, ABTS, and FRAP assays indicated that the antioxidant activity of acerola juice was stronger than that of the fruit juices reported in the literature, such as strawberry, grape, or apple. In addition, anthocyanins, flavonoids, and phenolic acids fractions were separated; among them, phenolic acids showed the highest antioxidant activity, indicating that phenolic acids contributed the most to the antioxidant property of wild acerola fruit [35]. Furthermore, Koca et al. analyzed the antioxidant activity of purple mulberry (Morus rubra) fruits growing wild in Turkey. FRAP assay was used and the average value was 33.90 μmol/g [36]. In addition, fruits of wild Bunium persicum, Elaeagnus latifolia, Solanum incanum, Rosa canina, Mespilus germanica, Aristotelia chilensis, Myrtus communis, Rubus hirsutus, Piper capense, Vitis coignetiae, Prunus spinosa, Syzygium cumini, and Vatis amurensis also showed strong antioxidant activities [37,38,39,40,41,42,43,44,45,46,47,48,49]; the related information is displayed in Table 1. Some studies compared the antioxidant activities between different genotypes of a certain species of wild fruit. Fourteen wild mandarin genotypes of Citrus reticulata were assessed for the antioxidant activities and phenolic compounds in the peels [50]. Results showed that antioxidant potency composite (APC) index varied from 58.84 to 98.89 in the studied wild genotypes, and among them, Nieduyeju showed the highest APC index. Furthermore, wild genotypes Guangxihongpisuanju, Nieduyeju, Cupigoushigan, and Daoxianyeju contained more phenolic compounds and exhibited higher antioxidant capacities than the commercial cultivars Satsuma and Ponkan. In another study, ethanol and ethyl acetate extracts of 10 crabapple varieties (Malus wild species) from China were analyzed for the antioxidant activities. Ethyl acetate extract showed higher contents of total phenolic and total flavonoids, and stronger DPPH and ABTS radical scavenging activities than ethanol extract, while ethanol extract had a significantly higher FRAP value (p < 0.01) than ethyl acetate extract. Results also showed that whole fruits of wild Malus species, particularly Malus rockii, exhibited stronger antioxidant activity than reported apple peel, indicating that Malus wild species could be rich sources of antioxidants [51]. Papaya, a fruit of the genus Chaenomeles, is an important source of functional food and traditional Chinese herbs. Du et al. evaluated the total polyphenol content (TPC) and antioxidant potential of five wild Chaenomeles genotypes [52]. Among them, the fruit of C. speciosa showed the highest free radical scavenging abilities by ABTS and FRAP assays while the C. thibetica extract was less effective. C. sinensis showed the highest DPPH scavenging capacity. Among them, DPPH values of extracts from four genotypes, C. sinensis (6.48 ± 0.23), C. speciosa (5.63 ± 0.17), C. thibetica (4.89 ± 0.21), and C. cathayensis (4.88 ± 0.25), were higher than Trolox (3.79 ± 0.07). In addition, wild genotypes of Vaccinium berries were evaluated for their differences in bioactivity on oxidative protection and minimum dosage to have a significant action [53]. Wild Vaccinium extracts are 3.04-fold more active than cultivated extracts by EC50, indicating that wild Vaccinium berries possessed stronger antioxidant activity than the cultivated ones. The results of six antioxidant assays showed a good relationship with anthocyanin and polyphenol content. In addition, the essential oil (EO) compositions and antioxidant activities of wild fruits Hypericum perforatum and Hypericum scabrum were analyzed. It was found that the antioxidant abilities of the EOs evaluated by β-carotene bleaching and DPPH assays might be due to their α-pinene contents [54]. Furthermore, the total phenolics and antioxidant activity of a group of Fragaria genotypes were determined and compared with the commercial genotype F. xananassa. The antioxidant capacity in the wild material was about three-fold higher than the commercial material [55]. In another study, wild bananas (Ensete superbum) had higher contents of phenolics and tannins, higher DPPH, ABTS, and FRAP activities than commercial ones [56]. Furthermore, all the investigated wild strawberry accessions (Fragaria vesca) showed higher antioxidant activity than the commercial cultivar (Camarosa) [57]. Results of another study indicated a significant difference between different wild strawberry fruits in their abilities to scavenge DPPH radicals [58]. Furthermore, a wild strawberry showed higher total phenolics and antioxidant activity than those cultivated samples [59]. Two wild raspberries also showed high antioxidant activity by FRAP, ABTS, and DPPH assays [60]. Besides, six genotypes of Diospyros kaki fruits were analyzed, and wild genotype D. kaki var. Silvestris Makino showed the highest content of phenolics and strongest antioxidant activity [61]. Additionally, eight wild genotypes of Rosa canina fruit showed great antioxidant activity, with a good relationship with total polyphenols and vitamin C content [62]. Several genotypes of wild bitter gourd (Momordica charantia) from Taiwan showed protective activity against Cu2+-induced low-density-lipoprotein peroxidation [63]. Furthermore, fruits of 10 wild almonds (Prunus amygdalus) were assessed, and two kinds (A. pabotti Browicz and A. orientalis Duhamel) exhibited the best antioxidant properties [64]. In addition, four wild almond species from Iran showed strong antioxidant activity [67]. Moreover, two wild blueberries showed higher total polyphenols content and antioxidant activity than three cultivated ones [65]. In addition, the fruit of the wild lime (Citrus hystrix) had higher antioxidant, flavonoid, and phenolic contents than cultivated ones [66]. In another study, wild blueberry exhibited stronger antioxidant activity than four cultivated ones [68]. Two wild berries showed potent antioxidant activity by ORAC assay [69]. However, in another study, the antioxidant capacities of fruits of wild and cultivated cranberries were similar, without a statistically significant difference (p < 0.05) [70]. Furthermore, wild cranberries exhibited a lower average antioxidant capacity than cultivated berries [71]. Similarly, hydroalcoholic extracts of wild murtilla (Ugni molinae) fruit showed weaker DPPH· and ABTS· scavenging capacity than cultivated ones [72]. Some studies screened different species of wild fruits for their antioxidant activities. In a study, 12 native Australian fruits were screened for the antioxidant activities and contents of phenolics, anthocyanins, and ascorbic acid, using ABTS and photochemiluminescence (PCL) assays [73]. Among them, five fruits exhibited significant stronger radical scavenging abilities (3.1- to 5.2-fold and 1.2- to 4.2-fold for ABTS and PCL assays, respectively) than blueberry (used as control). Six studied fruits showed higher total phenolics content (2.5- to 3.9-fold of control). Moreover, the Kakadu plum had the highest content of ascorbic acid (938-fold of control). These fruits could be a novel rich source of natural antioxidants. In another study, Fu et al. evaluated the antioxidant abilities of 56 exotic fruits from south China. Results of FRAP and ABTS·+ (expressed as TEAC value) assays showed that these fruits generally possessed high antioxidant capacities, which were strongly correlated with total phenolic content, indicating that phenolic compounds mainly contributed to their antioxidant activities [3]. In another study, 14 wild fruits were assessed for their antioxidant activities [74]. Results showed that among all the tested fruits, the acetone extract of Detarium microcarpum fruit possessed the highest DPPH free radical scavenging capacity, FRAP values, and ABTS free radical scavenging capacity. Meanwhile, antioxidant activities were strongly correlated with total phenolic and flavonoid levels. In addition, antioxidant activity was evaluated for seed residue extracts of wild Rubus ulmifolius and Sambucus nigra fruits [75]. The results of a DPPH assay showed significant antioxidant capacities of the extracts from all fruit seed residues. Meanwhile, the methanolic extract of Rubus seed residue exhibited a stronger antioxidant activity than that of Sambucus seed. In another study, Malta et al. tested the antioxidant activities of three wild cerrado fruits called gabiroba (Campomanesia cambessedeana), murici (Byrsonoma verbascifolia), and guapeva (Pouteria guardneriana) [76]. Results showed that gabiroba fruit was the richest source of total phenolics, and exhibited the highest antioxidant activity for both assays (ORAC, peroxyl radical scavenging capacity assays). In addition, ethanol extracts of three wild fruits, genipap (Genipa americana), umbu (Spondia tuberosa), and siriguela (Spondia purpurea), were analyzed for their antioxidant capacities. Siriguela and umbu (seeds and peels) extracts exhibited the highest antioxidant activities. Results of lipid peroxidation assay showed that pulp of genipap could be a promising source of antioxidant [77]. Furthermore, three wild fruits, Rubus megalococcus, Myrciaria aft cauliflora, and Hyeronima macrocarpa, were tested for their antioxidant activities. Results showed that the anthocyanin-rich extracts of Hyeronima macrocarpa exhibited stronger radical scavenging activity than the other extracts [78]. Moreover, 11 fresh exotic fruits from Brazil were analyzed for antioxidant activities by DPPH and ABTS assays. All the fruits showed considerable antioxidant activity, and the phenolic contents were positively correlated with total antioxidant activity by ABTS (R = 0.94, p ≤ 0.001) and DPPH (R = 0.88, p ≤ 0.001) assays [79]. In another study, 15 wild fruits were screened for their antioxidant activities [80]. Results showed that fruits of wild Terminalia bellirica, Terminalia chebula, Phyllanthus emblica, and Spondias pinnata possessed the strongest antioxidant activity based on the DPPH assay. Moreover, Spondias pinnata was more effective (16% radical scavenging activity) than vitamin C (5% radical scavenging activity), both at 5 μg/mL. Additionally, the peel and pulp of six wild fruits, sour plum (Ximenia caffra), marula (Sclerocarya birrea), mobola plum (Parinari curatellifolia), chocolate berry (Vitex payos), velvet sweet-berry (Bridelia molis), and red ivory (Berchemia zeyheri), were tested for their antioxidant activities. Both the peel and pulp of sour plum showed higher reducing capacities than all the other fruits, while velvet sweet-berry, the peel and pulp of sour plum, and chocolate berry peel showed high inhibitory effects on phospholipid peroxidation at high concentrations [81]. Additionally, the exotic Camu-camu fruit (Myrciaria dubia) presented the highest DPPH·scavenging capacity of all the fruits tested [82]. Furthermore, several wild blackberry fruit samples showed strong antioxidant activity with rich phenolic profile and content [83]. In another study, results showed the order of the antioxidant activity of five wild fruits was Rhus semialata > Docynia indica > Garcinia xanthochymus > Averrhoa carambola > Garcinia pedunculata [84]. In addition, wild Arbustus unedo fruit showed higher Folin–Ciocalteu values, vitamin C, and phenolic content than Rubus ulmifolius fruit [85]. Wild blackthorn (Prunus spinose) fruit exhibited higher antioxidant capacity than hawthorn (Crataegus monogyna) fruit [86]. In addition, several exotic tropical fruits (bacuri, caja, camu-camu, carnauba, gurguri, jabuticaba, jambolao, jucara, murta, black puca, and puca) showed strong antioxidant activity in a DPPH assay [87]. Additionally, fruit of wild Rosa canina showed higher efficacy towards ABTS· and H2O2 species than other tested wild fruits [88]. Moreover, methanolic extracts from jackal berry (Diospyros mespiliformis) showed higher DPPH radical scavenging capacity compared with other tested fruits [89]. In addition, fruits of Fragaria indica, Prunus armeniaca, Pyracantha crenulata, and Rubus ellipticus showed strong antioxidant activity [90]. Furthermore, 20 exotic fruits showed high antioxidant activity [91]. In addition, 24 exotic fruits were assessed, and the highest antioxidant activity and content of total phenolics were observed in banana passion fruits (Passiflora tarminiana and Passiflora mollisima) [92]. Similarly, exotic acerola showed the highest antioxidant values in the 10 exotic fruits investigated [93], and exotic dovialis showed the strongest antioxidant activity among the investigated exotic fruits [94]. In addition, wild Psidium cattleianum, Averrhoa carambola, Syzygium cumini, and Psidium guajava fruits showed the highest antioxidant capacities among the 17 exotic fruits from Mauritius [95]. Furthermore, polyphenolic extracts of three wild red berry fruits (Cornus mas, Prunus spinose, and Rubus fruticosus) showed strong scavenging ability on DPPH radical (IC50 values of 22.19 to 31.18 mL/g) [96]. 2.1.2. In Vivo Studies Several studies also evaluated the antioxidant activities of some wild fruits in vivo. In a study, wild snake fruit (Salacca edulis) and mangosteen (Garcinia mangostana) were analyzed for their influences on antioxidant activities and plasma lipids in rats fed with cholesterol. The rats were fed with diets supplemented with snake fruit and mangosteen for four weeks, and it was found that the increase in plasma lipids and the decline in antioxidant activity were both hindered, and snake fruit was more effective than mangosteen [97]. In another study, the effect of a polyphenol-rich extract (PrB) of Vaccinium angustifolium (wild blueberries) on brain oxidative status in adult, male, 3–4-month-old Balb-c mice was examined. Antioxidant status was determined by FRAP assay and levels of ascorbic acid, malondialdehyde, and reduced glutathione in whole brain homogenates. Lipid peroxidation products were decreased (38% and 79%) and brain ascorbic acid level was increased (21% and 64%) in both PrB30- and PrB60-treated groups. An increased glutathione level (28%) was observed in the PrB60-treated group. The results indicated that the fruit possessed strong brain antioxidant property [98]. In addition, the in vivo antioxidant activities of rare exotic Thai fruits, durian, snake fruit, and mangosteen, were investigated. Results showed that plasma lipid profile and antioxidant activity in rats fed with cholesterol-containing diets were positively influenced by diets supplemented with these exotic fruits [99]. These studies proved that abundant wild fruits could be potential sources of natural antioxidants, thus supporting their full utilization as bioactive elements in the food, pharmaceutical, and cosmetic industries. The antioxidant activity and possible functional components of extracts of some wild fruits are summarized in Table 1. 2.2. Antimicrobial Activity It is well known that various bacterial, fungal, and viral species could cause plant, animal, and human diseases, thereby causing the loss of crops, food spoilage, or even food poisoning that could damage human health [100,101]. Hence, it is important to develop natural effective antimicrobial agents. In recent years, wild fruits have exhibited potential antibacterial, antifungal, and antiviral activities in several studies. 2.2.1. Antibacterial and Antifungal Activities Some studies analyzed the antimicrobial activity of a certain species of wild fruit. An aqueous extract of wild fruit Nitraria retusa was tested for inhibition of microbial growth in beef patties. The results showed that the extract possessed strong antimicrobial activity against Salmonella typhimurium, Klebsiella pneumonia, and Bacillus thuringiensis [102]. In addition, extracts of wild yellow azarole fruit peel showed considerable antibacterial activity, especially against Staphylococcus aureus and Streptococcus faecalis [103]. Moreover, methanol and n-hexane extracts from fruits of wild mahaleb cherry (Prunus mahaleb) were screened by measuring their inhibitory activity on several bacteria (Escherichia coli, Pseudomonas aeruginosa, Proteus mirabilis, K. pneumoniae, Acinetobacter baumannii, S. aureus, Enterococcus faecalis, and Bacillus subtilis), as well as several fungi (Candida albicans, Candida parapsilosis, Candida tropicalis, and Candida krusei). The extracts showed antibacterial activity against both Gram (+) bacteria and Gram (−) bacteria tested, and the methanol and n-hexane extracts showed antifungal activity against C. krusei [104]. Moreover, a fresh fruit extract of the wild plant Clematis apiifolia exhibited minimum inhibitory concentrations (MIC) in the vicinity of 0.1% against various yeasts and non-lactic acid bacteria of ≤0.4%. MICs against lactic acid bacteria were about 2.0%. Results indicated that this fruit was even more effective in antibacterial activity than garlic, which has great antibacterial properties. Furthermore, the principal antimicrobial compound of C. apiifolia was isolated and identified as protoanemonin. The researchers suggested that the antimicrobial compound of C. apiifolia inhibited microorganisms by reacting with sulfhydryl groups of cellular proteins [105]. Furthermore, the antimicrobial activities of the essential oil hydrodistilled from wild pepper fruits were evaluated. Results showed medium inhibitory effect against the Gram (+) species E. faecalis, S. aureus, and the yeast C. albicans [59]. Moreover, inhibitory effects on the growth of Mycobacterium tuberculosis H(37)Rv was observed in fruits of wild ampalaya (Momordica charantia). The fruits of wild ampalaya showed higher antitubercular activity (90%) than that of the cultivated variety (81%) [106]. Malek et al. tested the antibacterial activities of oils separated from the fruit of Scabiosa arenaria, a wild plant growing in Tunisia. The 16 Gram (+) and Gram (−) bacteria and four Candida species were used. The oils exhibited significant inhibitory activities against these bacterial and Candida species, superior to thymol, which was used as a positive control [107]. Carissa opaca is a wild plant used widely in ethnomedicine. Thirty-four strains of Gram (+) and Gram (−) bacteria were used to determine the antibacterial activities of ethanol extracts of the fruits. The results exhibited a broad spectrum of efficacy [108]. Additionally, crude oils from ripe and unripe wild olive fruits were proved to have antibacterial activity against some of the Gram (+) and Gram (−) bacterial strains [109]. In another study, antimicrobial properties of extracts of fruits from wild melon (Citrullus lanatus) were tested. The researchers tested antimicrobial properties of crude chloroform, hexane, and ethanol extracts against five bacteria (E. coli, S. aureus, P. aeruginosa, B. subtilis, and Proteus vulgaris) and two fungi (Aspergillus nigar and C. albican). It was found that a chloroform extract of the fruit showed the highest antibacterial activity, while an ethanol extract of the fruit pulp exhibited the highest antifungal activity. It is worth mentioning that the fruit of this plant was as potent as standard antimicrobial drugs (clotrimazole and gentamici) against certain microorganisms [110]. Moreover, another study showed that wild strawberry guavas (Psidium cattleianum) possessed better antimicrobial activity than common guavas [111]. In addition, different wild clones of European cranberry were investigated for their antimicrobial activities. Results showed that extracts of wild European cranberry had inhibitory effects against the growth of varieties of human pathogenic bacteria, both Gram (+) and gram (−). Among them, the most sensitive bacteria were Listeria monocytogenes and Enterococcus faecalis (average inhibition zones of 20.35 and 19.71 mm, respectively), and S. typhimurium and S. aureus showed moderate resistance [112]. Some studies compared the antimicrobial activity between different species of wild fruits. In a study, polyphenolic extracts of three wild red berry fruits, European cornel (Cornus mas), blackthorn (Prunus spinosa), and blackberry (Rubus fruticosus), were assessed for their antimicrobial activities by the disc diffusion method. Almost all the tested bacterial strains (such as E. coli, P. aeruginosa, and Salmonella enteritidis) were inhibited by all extracts. S. enteritidis was the most sensitive among Gram (−) bacteria, while S. aureus was the most sensitive among Gram (+) bacteria. Blackthorn extract showed slightly higher antimicrobial activity compared with the other tested extracts [96]. In addition, Turker et al. tested the antimicrobial activity of eight wild fruits grown in Turkey [113]. Results showed that fresh fruits of wayfaring tree, firethorn, and hawthorn showed the highest antibacterial activity. In addition, ethanol extracts of these fruits exhibited strong inhibitory effects on S. aureus, Staphylococcus siepidermidis, and Streptococcus pyogenes [113]. Furthermore, fruits of three wild plants growing in Mexico, namely nanchi (Byrsonima crassifolia), arrayan (Psidium sartorianum), and ayale (Crescentia alata), were analyzed by Pio-Leon et al. They not only measured their antibacterial activities against 21 human pathogenic bacteria by the micro-dilution assay, but also established the minimum inhibitory concentration (MIC) and minimum bactericide concentration (MBC). Results showed that methanol extracts of arrayan exhibited the highest activity against the Gram (+) bacteria, being most sensitive to S. aureus. Meanwhile, hexane extracts of arrayan and ayale exhibited the highest inhibitory effects on enterobacteria (E. coli, Salmonella spp., and Shigella spp.) [114]. Moreover, it was found that essential oils isolated from the fruits of wild Hypericum perforatum and Hypericum scabrum exhibited higher antimicrobial activity against S. aureus and E. coli than their main constituent, α-pinene [41]. Additionally, methanol and hexane extracts from a pulp of wild tamarind fruit (Tamarindus indica) were tested for their inhibitory activities on human pathogenic microorganisms including five bacteria and three fungi. All the bacterial strains showed sensitivity to both extracts, while only Penicillium species were sensitive to hexane extract [115]. In addition, several water and methanol extracts of the 16 cultivars selected from Taiwanese indigenous wild bitter gourd (Momordica charantia) showed inhibitory activity against the growth of E. coli and Salmonella enterica [116]. Results of another study revealed that a petroleum ether extract of wild Atriplex inflata fruits possessed high inhibitory activity against Botrytis cinerea [117]. 2.2.2. Antiviral Activity Several wild fruits have exhibited antiviral activity. Knox et al. detected antiviral properties of crude extracts of wild Kurokarin (Ribes nigrum) fruit against influenza virus types A and B (VIA and VIB). At a concentration of 3.2 μg/mL, plaque formation of both IVA and IVB was inhibited by the extract by 50% (IC50). Additionally, when treating the host cells with 10 and 100 μg/mL of the extract for 6 h after infection, the growth of IVA could be completely suppressed. Virus titers in culture fluids of the cells were completely suppressed after treatment with 100 μg/mL of Kurokarin extract for 1 h after infection of 8 to 9 h, indicating that the extract inhibited the virus release from the infected cells [118]. Furthermore, extracts of a series of wild berry fruit from Bulgaria possessed great antiviral activities [119]. Four wild berries, strawberry, raspberry, bilberry, and lingonberry, were tested for their antiviral properties against some important human pathogens, poliovirus type 1 (PV-1), coxsackievirus B1 (CV-B1), human respiratory syncytial virus A2 (HRSV-A2), and influenza virus (A/H3N2), by virus cytopathic effect inhibition test. It was revealed that extracts of all berry fruits suppressed proliferation of CV-B1 and influenza virus A/H3N2. Meanwhile, anthocyanin fractions of all wild berries showed a considerable inhibitory effect against the replication of influenza virus A/H3N2. These studies proved that wild fruits could function as potent antibacterial, antifungal, and antiviral agents. The antimicrobial activities of some wild fruits are summarized in Table 2. 2.3. Anti-Inflammatory Activity Inflammation is closely related to various diseases, such as atherosclerosis, heart disease, stroke, cancer, diabetes mellitus, bone arthritis, asthma, migraine pain, periodontitis, irritable bowel syndrome, and chronic fatigue syndrome. Currently, drugs used to treat chronic inflammatory diseases are mainly various nonsteroidal drugs, which may exert side effects [120]. Therefore, the development of effective and natural sources of anti-inflammatory products has gained increasing attention. Evidence accumulated in recent years pointed out that several kinds of wild fruits possess anti-inflammatory activities, through various mechanisms of action. Nitric oxide (NO) is a marker of late inflammation formed during activation of inducible nitric oxide synthase (iNOS) [121], and chemokine (C–C motif) ligand 20 (CCL20) is an important chemokine for immune and inflammatory response [122]. Therefore, the inhibition of NO and CCL20 is an indicator of possible anti-inflammatory properties. In the research conducted by Fazio et al., in vitro anti-inflammatory activities of the methanol extracts from the seeds of wild blackberry (Rubus ulmifolius) and elderberry (Sambucus nigra) were analyzed [75]. They firstly evaluated the seeds’ ability to inhibit lipopolysaccharide (LPS) induced NO production in mouse macrophage cell line RAW264.7 macrophages. Results showed that wild blackberry extract decreased NO release with almost 60% inhibition at the highest dose (50 μg/mL). Meanwhile, it showed a concentration-dependent effect. Subsequently, the influence of both extracts on macrophage-inflammatory protein-3α/CCL20 were evaluated. Wild blackberry extract decreased CCL20 production in a concentration-dependent manner, with a more than 90% inhibition at 50 μg/mL. By comparison, wild elderberry extract did not show a significant effect on decreasing either NO or CCL20 production. The results confirmed that wild blackberry possessed a strong anti-inflammatory activity. Metabolites of the 5-lipoxygenase (5-LOX) pathway are important mediators of inflammation. LOX and its metabolites are shown to play a vital part in tumor formation and cancer metastasis. In some cancer cells, such as prostate, lung, colon, and breast, high expression of 5-LOX was found. In one study, the anti-inflammatory activity of Ziziphus mistol ripe berries, an exotic Argentinean fruit, was tested. The three tested extracts (ethanolic mistol extraction, aqueous mistol extraction, and acetone water mistol extract) were obtained after two different processes: boiling and hydroalcoholic extraction. They determined LOX activity to evaluate anti-inflammatory activity. In working conditions, only an ethanolic extract exhibited inhibition of LOX activity (IC50 = 183.80 μg gallic acid equivalents (GAE)/mL), while an aqueous extract showed no inhibitory effect at the tested concentrations (until 45.08 μg GAE/mL). These results suggested that bioactive compounds might be thermolabile, yet Ziziphus mistol ripe berries still had potent anti-inflammatory activity [30]. Cyclooxygenase-2 (COX-2) expression is an important pro-inflammatory response. Several studies have confirmed that COX-2, an important inflammatory mediator, is closely related to the occurrence and development of diabetes mellitus and diabetic nephropathy. Therefore, the inhibition of COX-2 is an indicator of possible anti-inflammatory properties. A study analyzed the anti-inflammatory activity of three wild Jamaica-grown fruits species (Rubus jamaicensis, Rubus rosifolius, and Rubus racemosus) and three wild Michigan-grown species (Rubus acuminatus, Rubus idaeus cv., and Rubus idaeus cv.). The COX-1 and COX-2 enzyme inhibitory activities were measured by monitoring the initial rate of O2 uptake. Aspirin, Celebrex, and Vioxx were used as positive controls. Results showed that all the hexane extracts of the Jamaica-grown Rubus berries were COX-active, inhibiting COX-2 by 18%–33%, while the Michigan-grown Rubus extracts were, in general, not COX-active [123]. In another study, eight compounds separated from the ethyl acetate extract of the Rubus rosifolius growing wild in elevated regions in Jamaica were identified as euscaphic acid, 1-b-hydroxyeuscaphic acid, hyptatic acid B, 19α-hydroxyasiatic acid, trachelosperogenin, 4-epi-nigaichigoside F1, nigaichigoside F1, and trachelosperoside B-1 by nuclear magnetic resonance (NMR) spectroscopy. In vitro COX-1 and COX-2 enzyme inhibitory assays were conducted to evaluate anti-inflammatory activity. Euscaphic acid, 1-b-hydroxyeuscaphic acid, and hyptatic acid B showed selective COX-1 enzyme inhibitory activity (13%, 25%, and 35% respectively) at 25 μg/mL. Similar COX inhibitory activity was demonstrated by compounds 4-epi-nigaichigoside F1 and trachelosperoside B-1, which showed moderate selectivity against the COX-1 enzyme [124]. In addition, the anti-inflammatory activity of Psidium cattleianum (strawberry guava) was analyzed using COX-1 and -2 enzyme inhibitory assays. Results showed that ethyl acetate extract of guava exhibited notable activity (56.4%) against the COX-2 isoform, followed by methanolic extract (44.1%) against the COX-1 enzyme at 250 μg/mL [111]. Furthermore, a polyphenol-rich fraction from lowbush cranberry, a wild Alaskan Vaccinium berry, showed effective inhibition of LPS-elicited induction of interleukin-1β (IL-1β) in RAW 264.7 cells [125]. Some wild fruits are rich in anthocyanins, which are known to possess antioxidant and anti-inflammatory activities. A study evaluated the inhibitory effects of wild blackberries on pro-inflammatory responses (NO production, iNOS expression, COX-2 expression, and prostaglandin E2 level). Results demonstrated that dietary consumption of wild blackberries (Rubus spp.) could decrease NO-generated oxidative stress and inhibit the expression of pro-inflammatory proteins, thus protecting the body against oxidation- or inflammation-related diseases [120]. The macrophage cell line RAW 264.7 was stimulated by LPS to cause pro-inflammatory responses. Different fractions from wild blackberry genotypes (WB-3, WB-7, WB-10, and WB-11) were tested separately. At 50 μM (cyanidin-3-O-glucoside or catechin equivalent), all markers were significantly (p < 0.05) inhibited by most fractions. The highest NO inhibition was observed in the anthocyanin-rich fraction from WB-10, the highest inhibitory activity on iNOS expression was presented by proanthocyanidin-rich fractions from the WB-10, and polyphenolic-rich fractions from WB-7 were identified as potent inhibitors of COX-2 expression. Nuclear factor-κB (NF-κB) plays an important part in immune, stress, inflammatory, proliferative, and apoptotic responses [126]. The inhibition of NF-κB is commonly considered as an effective strategy to treat inflammatory disorders [127]. Tumor necrosis factor-α (TNF-α) and interleukin (IL) are important inflammatory cytokines. In a study, the anti-inflammatory activities of wild lowbush blueberry were investigated. Effects of the phenolic acid (PA) mixture were firstly measured by the inhibition against LPS-induced NF-κB activation, and results showed that NF-κB activation was significantly inhibited (by 33.2% at 4 mg FBE/mL) by PA mixture. Based on the result, a concentration of 4 mg FBE/mL was used in TNF-α and IL-6 ELISA. The production of both TNF-α (36.7%) and IL-6 (37.5%) were significantly decreased by the PA mixture. In conclusion, a phenolic acid mixture of lowbush blueberry showed anti-inflammatory activities by inhibiting NF-κB activation and the production of inflammatory cytokines (TNF-α and IL-6) at a high dose [128]. In addition, Hsu et al. did a relatively comprehensive experiment on the anti-inflammatory property of wild bitter melon (WBM), including both in vitro and in vivo experiments [129]. Inflammation was induced by Propionibacterium acnes. Results showed that in vitro, an ethyl acetate (EA) extract of WBM fruit potently suppressed pro-inflammatory cytokine (IL-8, TNF-α, and IL-1β) and matrix metalloproteinase (MMP)-9 levels in P. acnes-stimulated THP-1 cells. As for in vivo, P. acnes-induced ear swelling and granulomatous inflammation in mice were effectively attenuated by concomitant intradermal injection of EA extract. This study indicated that wild bitter melon could produce an anti-inflammatory effect. Several subfractions of Aristotelia chilensis have shown a notable inhibition on the 12-deoxyphorbol-13-decanoate (TPA)- induced inflammation in ear of the mouse edema (EC50 of 0.3 to 11.8 μg/mL) [130]. In another study conducted by the same researchers, results showed that carrageenan-induced inflammation in the rat paw was inhibited by these samples [131]. Similarly, in the inflammatory pain mice models induced by acetic acid and formalin, abdominal constrictions and the inflammatory phase of nociception were significantly reduced by intraperitoneal administration of a fraction separated from tamarillo (Solanum betaceum), a tropical exotic fruit. The results suggested that the fraction had a possible antinociceptive effect on inflammatory pain models [132]. These studies strongly proved that some wild fruits could be good natural sources of anti-inflammatory materials through different mechanisms of action, such as inhibiting COX-2 and NF-κB, as well as decreasing NO and CCL20 release. The anti-inflammatory activities of some wild fruits are summarized in Table 3. 2.4. Anticancer Activity Cancer is known as a major cause of death all over the world. A relationship between fruit intake and a reduced risk of cancer has been found [133,134]. Various natural products, such as fruits, vegetables, and herbal plants, have been widely proved to possess antiproliferative activities [135,136,137]. Several wild fruits, such as wild red raspberry from Jamaica, and wild blueberry, have been proven to possess anticancer activities against breast, colon, prostate, and cervical cancer cells. Malta et al. tested the inhibitory activity on tumor cell proliferation of three kinds of exotic Brazilian fruits, gabiroba (Campomanesia cambessedeana), murici (Byrsonoma verbascifolia), and guapeva (Pouteria guardneriana), by the MTS assay [77]. The gabiroba, murici, and the pulp of guapeva inhibited growth of HepG2 cell in a dose-dependent manner, with EC50 values of 40.7 ± 4.8, 173.6 ± 18.2, and 37.9 ± 2.2 mg/mL, respectively. The extracts were nontoxic at the concentrations used in the experiments. In another study, hyptatic acid B and 4-epi-nigaichigoside F1 compounds separated from ethyl acetate extract of wild Rubus fruits inhibited the growth of human colon tumor cells by 56% and 40%, respectively [124]. In another study, eight different extracts of each wild fruit were tested for anticancer activity. Results showed that the greatest anticancer activity was obtained from a cold water extract of fresh R. caesius fruit (100% inhibition), followed by cold and hot ethanol extracts of fresh V. lantana fruit (90.5% and 95.2% inhibition, respectively) [114]. Proanthocyanidin is a general term for a large class of polyphenols, which is composed of catechin, epicatechin, and epicatechin gallate in forms of different degrees of polymerization (DPn). Some studies have proved that proanthocyanidin possessed various kinds of bioactivities, such as antioxidant and anticancer activities, preventing hepatic and brain lipid peroxidation and DNA damage in animals [138]. The antiproliferative activity of proanthocyanidin-rich extracts from wild blueberry (Vaccinium angustifolium) was tested. Results showed that the antiproliferative activity of different fractions was positively correlated with proanthocyanidin content, and the fraction with a DPn of 5.65 showed considerable antiproliferative activity against human prostate and mouse liver cancer cell lines [139]. Results also suggested that antiproliferative activity was associated with high molecular weight proanthocyanidin oligomers from wild blueberry fruits. Yellow Himalayan raspberry, as a wild edible fruit, was analyzed for antiproliferative activities. Results showed that acetone and methanol extracts exhibited inhibitory effects against human cervical cancer cells (C33A) (EC50 at 5.04 and 4.9 mg/mL fruit concentration respectively), and were nontoxic to normal peripheral blood mononuclear cells at the same time [140]. In addition, three wild species of strawberries (Fragaria virginiana, Fragaria Chiloensis, and Fragaria xananassa) were analyzed for antiproliferative activity. Extracts of the three fruits all significantly inhibited the proliferation of A549 human lung epithelial cancer cells [141]. In another study, Woguem et al. found that the volatile oil from the wild pepper could inhibit the growth of human tumor cells MDA-MB 231 (breast adenocarcinoma), A375 (malignant melanoma), and HCT116 (colon carcinoma), in a concentration-dependent manner [59]. Several kinds of water and methanol extracts of wild bitter gourd also showed similar cytotoxic activities on human fibrosarcoma HT 1080 cells to 10 μg/mL of doxorubicin, which was used as positive control in this study [103]. Finally, the anticancer activities of several wild fruits are summarized in Table 4. 2.5. Other Bioactivities of Wild Fruits In addition to the biological activities mentioned above, some wild fruits have shown other beneficial health effects. Some wild fruits have shown anti-acetylcholinesterase activity. The acetylcholinesterase inhibitory activity is a commonly used pharmacological model of Alzheimer’s disease. In one study, a water extract of Sorbus torminalis (wild service tree) fruit showed moderate ability to inhibit acetylcholinesterase [29]. Similarly, three exotic fruits from Brazil were tested for anti-acetylcholinesterase activities, namely genipap (Genipa americana), umbu (Spondia tuberosa) and siriguela (Spondia purpurea). Results showed that ethanol extracts of genipap pulp and siriguela seed could present a similar inhibitory effect on acetylcholinesterase compared with carbachol (positive control) [77]. In another study, an obvious cognitive enhancement was observed in the experimental mice after short-term intraperitoneal supplementation with a polyphenol-rich extract of wild blueberries (Vaccinium angustifolium) [98]. Researchers found that the brain antioxidant properties of mice were higher and acetylcholinesterase activity was inhibited after the treatment, indicating that bioactive components of wild blueberry are able to affect the brain function of mice in a positive way. Furthermore, larvicidal/insecticidal activities have been observed in several wild fruits. The researchers evaluated the insecticidal activity of wild Tetradium glabrifolium fruits against Aedes albopictus [142]. Essential oils and three compounds from the fruit showed strong larvicidal activities against the early fourth-instar larvae of A. albopictus. In another study, the antigiardial activities of wild watermelon (Citrullus lanatus) fruits were investigated [143]. Results revealed that two compounds from the fruits, cucurbitacin L 2-O-β-glucoside and cucurbitacin E, had potent antigiardial activity against Giardia lamblia in vitro. Meanwhile, all the extracts, including petroleum ether, ethyl acetate, and butanol crude extracts, were active against Giardia lamblia. The results indicated that this fruit might be a potential new resource for the control of giardiasis. Zanthoxylum schinifolium is a traditional wild Chinese medicinal plant. Researchers found that essential oils of the fruits exhibited strong fumigant toxicity against the maize weevil Sitophilus zeamais, a common grain storage insect [144]. Similarly, fruits of another wild Chinese medicinal plant called Carum carvi showed strong fumigant toxicity and contact toxicity against Sitophilus zeamais and Tribolium castaneum adults, which are both common grain storage insects [145]. There was a study proving that 70% methanol extract of Elaeagnus latifolia, a wild edible fruit, had a promising effect on protecting pUC18 DNA [37]. In addition, a methanol extract of wild Brenania brieyi fruit showed estrogenic effects by doubling the uterine weight and increasing the vaginal epithelial height of female rats [146]. Advanced glycation endproducts (AGE) is an important related pathophysiological feature common to many chronic diseases, such as cardio- and cerebrovascular diseases, diabetes mellitus, and Alzheimer’s disease. Inhibitory activity on AGE formation was related to radical scavenging activities. In a study, all samples of wild berries reduced AGE formation in a concentration-dependent way, with a positive correlation to each extract’s total phenolic content and, to a lesser degree, total anthocyanin content [147]. Moreover, it has been reported that methanol extracts of wild raspberry fruits had potassium-conservation diuretic activity in experimental rats [148]. The fruit of wild Aristotelia chilensis also showed gastroprotective effects and thus have great potential as nutraceuticals [131]. Other bioactivities of wild fruits are summarized in Table 5. All the bioactivities of wild fruits are displayed in Figure 1. 3. Bioactivities of Wild Berries The berries are an important group of fruits. Berries include members of several families, such as Rosaceae and Ericaceae [149]. It is well established that berries contain high contents of bioactive compounds, such as phenolic acids, anthocyanins, flavonols, and tannins [150,151]. Wild berries are so far underutilized, but they are often equal to or more valuable than commercial berries in terms of their bioactivities and health benefits, such as antioxidant, antimicrobial, anti-inflammatory, and anticancer activities [78,104,123,139]. The bioactivities of wild berries involved in this review are summarized in Table 6. 4. Conclusions The special genotype and formative environment create unique and abundant ingredients with health benefits in wild fruits. When a wild species is domesticated, the biological activities might decrease. In addition, wild fruits should not be excessively exploited, as this could cause a depauperation of the natural environment. Various kinds of wild fruits have shown numerous bioactivities, such as antioxidant, antimicrobial, anti-inflammatory, anticancer, and anti-acetylcholinesterase activities. Some wild fruits have more than one bioactivity. For example, Aristotelia chilensis possesses anti-inflammatory, antiedema, and gastroprotective activities. The consumption and utilization of some wild fruits have been increasing, and some wild fruits have been developed into functional foods. In the future, for full utilization of wild fruit resources, more bioactivities of wild fruits should be evaluated, and bioactive components should be isolated and identified. The mechanisms of action should be explored further. In addition, the toxicological evaluation of some wild fruits is also necessary for safe human consumption. Acknowledgments This work was supported by the National Natural Science Foundation of China (No. 81372976), a Key Project of the Guangdong Provincial Science and Technology Program (No. 2014B020205002), and the Hundred-Talents Scheme of Sun Yat-Sen University. Author Contributions Ya Li, Sha Li, and Hua-Bin Li conceived this paper; Ya Li, Jiao-Jiao Zhang, Dong-Ping Xu, Tong Zhou, and Yue Zhou wrote this paper; and Sha Li and Hua-Bin Li revised the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Some bioactivities of wild fruits. ijms-17-01258-t001_Table 1Table 1 Antioxidant activities of some wild fruits. Wild Fruits Bioactive Compounds Effects References 56 wild fruits from South China polyphenols antioxidant activity [3] Prunus mahaleb total anthocyanin and phenolics scavenging free radicals (oxygen radicals) [27] Psidium acutangulum phenolics, citric, annurcoic, ω3, ω6, ω9 fatty acids, and ascorbic acid scavenging free radicals (DPPH, ABTS) [28] Sorbus torminalis phenolic compounds scavenging free radicals (ABTS, superoxide anion radicals), antioxidant activity [29] Ziziphus mistol polyphenols scavenging free radicals (ABTS, DPPH, superoxide and hydroxyl radicals) [30] Annona cherimola not mentioned scavenging free radicals (DPPH, ABTS), antioxidant activity, inhibition of lipid peroxidation [31] Myrica esculenta polyphenols scavenging free radicals (DPPH, ABTS), antioxidant activity [32] Vaccinium meridionale phenolic compounds scavenging free radicals (ABTS), antioxidant activity [33] Lycium ruthenicum polyphenols scavenging free radicals (DPPH, ABTS), antioxidant activity [34] Malpighia ernarginata phenolic acids scavenging free radicals (DPPH, ABTS, oxygen radical) [35] Morus rubra not mentioned antioxidant activity [36] Bunium persicum Phenolics and flavonoids scavenging free radicals (DPPH), antioxidant activity [37] Elaeagnus latifolia phenolics and flavonoids scavenging hydroxyl radicals, superoxide radicals, singlet oxygen radicals, hypochlorous acid [38] Solanum incanum 3-O-acetyl-and4-O-acetyl-5-O-(E)-caffeoylquinic acids scavenging free radicals (ABTS, DPPH) and iron chelation activity [39] Rosa canina α-tocopherol, β-carotene, reducing sugar, and ascorbic acid scavenging free radicals (DPPH), reducing power, inhibition of β-carotene bleaching and lipid peroxidation [40] Mespilus germanica not mentioned scavenging nitric oxide and H2O2 radicals, inhibition of lipid peroxidation [41] Aristotelia chilensis phenolics scavenging free radicals (DPPH, superoxide radicals, oxygen radicals), antioxidant activity, inhibition of lipid peroxidation [42] Myrtus communis not mentioned scavenging free radicals (DPPH, β-carotene-linoleic acid) [43] Rubus hirsutus Phenolics and flavonoids scavenging free radicals (DPPH), antioxidant activity [44] Piper capense not mentioned scavenging free radicals (ABTS) [45] Vitis coignetiae anthocyanins scavenging free radicals (ABTS, DPPH) [46] Syzygium cumini phenolics, tannins, and anthocyanins scavenging free radicals (DPPH, hydroxyl radical and superoxide radical), inhibition of lipid peroxidation [48] Vatis amurensis catechin, epicatechin, 4-methyl-catechol, gallic, protocatechuic, chlorogenic, caffeic, p-coumaric, and syringic acids scavenging superoxide radicals [49] 14 wild genotypes of Citrus reticulata not mentioned scavenging free radicals (DPPH, ABTS, oxygen radicals), antioxidant activity [50] 10 crabapples (Malus wild species) polyphenols, flavonoids scavenging free radicals (DPPH, ABTS), antioxidant activity [51] C. speciosa, C. thibetica, C. cathayensis, C. sinensis, C. japonica polyphenols scavenging free radicals (DPPH, ABTS), antioxidant activity [52] wild genotype of Vaccinium spp. anthocyanin, polyphenols scavenging free radicals (ABTS, superoxide anion and hydroxyl radical) [53] Hypericum perforatum, Hypericum scabrum α-pinene scavenging free radicals (DPPH), inhibition of β-carotene bleaching [54] wild Fragaria genotypes not mentioned antioxidant activity [55] Ensete superbum Phenolics and tannin scavenging free radicals (DPPH, ABTS), antioxidant activity [56] Fragaria vesca phenolics scavenging free radicals (DPPH), antioxidant activity [57] Fragaria vesca not mentioned scavenging free radicals (DPPH) [58] wild strawberries not mentioned antioxidant activity [59] 2 wild raspberries not mentioned scavenging free radicals (DPPH, ABTS), antioxidant activity [60] 6 genotypes of Diospyros kaki gallic acid, vanillic acid, caffeic acid, syringic acid, and quercetin scavenging free radicals (DPPH, ABTS, hydroxyl radical), antioxidant activity [61] Rosa canina polyphenols and vitamin C scavenging free radicals (DPPH) [62] Momordica charantia not mentioned scavenging free radicals (DPPH, hydroxyl radicals), protection against Cu2+-induced low-density-lipoprotein peroxidation [63] Prunus amygdalus not mentioned scavenging free radicals (DPPH), reducing power [64] 2 wild blueberries polyphenols scavenging free radicals (DPPH, ABTS, oxygen radicals), antioxidant activity [65] Citrus hystrix phenolics scavenging free radicals (DPPH), antioxidant activity [66] Amygdalus lycioides, Amygdalus kotschyi, Amygdalus pabotti, Amygdalus trichamygdalus phenolics scavenging free radicals (nitrite, hydrogen peroxide, superoxide radicals), reducing power [67] Vaccinium miyrtillus phenolics scavenging free radicals (DPPH), antioxidant activity [68] Rubus croceacanthus and Rubus sieboldii anthocyanins, ascorbic acid scavenging oxygen radicals [69] wild cranberry not mentioned scavenging free radicals (ABTS) [70] wild blueberry and cranberry not mentioned scavenging free radicals (DPPH) [71] Ugni molinae polyphenols scavenging free radicals (DPPH, ABTS) [72] 12 native Australian fruits total phenolics scavenging free radicals [73] 14 species of wild fruits phenolics and flavonoids scavenging free radicals (DPPH, ABTS), antioxidant activity [74] Rubus ulmifolius and Sambucus nigra phenolics scavenging free radicals (DPPH) [75] Campomanesia cambessedeana, Byrsonoma verbascifolia, Pouteria guardneriana phenolics and flavonoids scavenging free radicals (oxygen radicals, peroxyl radicals), cellular antioxidant activity [76] Genipa americana, Spondia tuberose, Spondia purpurea chlorogenic acid scavenging free radicals (ABTS), antioxidant activity, inhibition of lipid peroxidation in a biomimetic membrane system and mouse liver, inhibition of lipid peroxidation in mouse liver [77] Rubus megalococcus, Myrciaria aft cauliflora, Hyeronima macrocarpa anthocyanin scavenging free radicals (ABTS, DPPH) [78] 11 exotic fruits from Brazil phenolics scavenging free radicals (DPPH, ABTS) [79] 15 wild fruits polyphenols scavenging free radicals (DPPH) [80] Ximenia caffra, Sclerocarya birrea, Parinari curatellifolia, Vitex payos, Bridelia molis, Berchemia zeyheri not mentioned scavenging free radicals (DPPH, superoxide anion radical), reducing power, inhibition of phospholipids peroxidation [81] cambuci, araca-boi, camu-camu, jaracatia, araca not mentioned scavenging free radicals (DPPH) [82] 23 wild blueberry fruits phenolic compounds scavenging free radicals (ABTS), antioxidant activity [83] Garcinia pedunculata, Garcinia xanthochymus, Docynia indica, Rhus semialata and Averrhoa carambola phenolics antioxidant activity [84] Arbutus unedo, Rubus ulmifolius phenolic acids, anthocyanins, ascorbic acid scavenging free radicals (ABTS, DPPH), antioxidant activity [85] Prunus spinosa and Crataegus monogyna phenolic compounds scavenging free radicals (DPPH, ABTS), antioxidant activity [86] wild bacuri, caja, camu-camu, carnauba, gurguri, jabuticaba, jambolao, jucara, murta, black puca and puca fruits not mentioned scavenging free radicals (DPPH) [87] Crataegus azarolus, Crataegus monogyna, Prunus spinosa, Rosa canina, Rubus ulmifolius, Sorbus domestica Phenolics and carotenoids scavenging free radicals (ABTS, H2O2) [88] Diospyros mespiliformis, Flacourtia indica, Uapaca kirkiana and Ziziphus mauritiana not mentioned scavenging free radicals (DPPH, superoxide anion radical), reducing power [89] Fragaria indica, Prunus armeniaca, Pyracantha crenulata and Rubus ellipticus not mentioned scavenging free radicals (DPPH, ABTS), antioxidant activity [90] 20 exotic fruits not mentioned scavenging free radicals (DPPH, ABTS, oxygen radicals), antioxidant activity [91] 24 exotic Colombian fruits soluble phenolics scavenging free radicals (ABTS), antioxidant activity [92] wild abiu, acerola, wax jambu, cashew, mamey sapote, carambola or star fruit, Surinam cherry, longan, sapodilla and jaboticaba fruits not mentioned scavenging free radicals (hypochlorous acid, ABTS, and DPPH) [93] exotic araca-boi, cajamanga, sirihuela, dovialis, landim, murici, tomatinho do mato fruits phenolics scavenging free radicals (ABTS, DPPH), antioxidant activity [94] 17 exotic fruits phenolics and proanthocyanidins antioxidant activity [95] Cornus mas, Prunus spinosa, Rubus fruticosus polyphenolics scavenging free radicals [96] Salacca edulis Reinw, Garcinia mangostana phenolics hindering the rise in plasma lipids and decrease of antioxidant activity in rats fed with cholesterol [97] Vaccinium angustifolium polyphenols improving brain antioxidant properties in mice (antioxidant activity, improving ascorbic acid concentration and glutathione levels, reducing lipid peroxidation products) [98] wild durian, snake fruit and mangosteen not mentioned scavenging free radicals (ABTS, DPPH), antioxidant activity [99] ABTS: 2,2′-azinobis-3-ethylbenzothiazoline-6-sulphonate; DPPH: 2,2-diphenyl-1-picrylhydrazyl. ijms-17-01258-t002_Table 2Table 2 Antimicrobial activities of some wild fruits. Wild Fruits Bioactive Compounds Effects References Hypericum perforatum, Hypericum scabrum not mentioned inhibition of S. aureus and E. coli [54] Cornus mas, Prunus spinosa, Rubus fruticosus polyphenols inhibition of all the tested bacterial strains [96] Piper capense not mentioned inhibition of S. aureus, E. faecalis, and C. albicans [45] Nitraria retusa not mentioned inhibition of S. typhimurium, K. pneumonia, and B. thuringiensis [102] Crataegus azarolus phenolics inhibition of S. aureus and S. faecalis [103] Prunus mahaleb not mentioned inhibition of some Gram (+) and Gram (−) bacteria and fungi [104] Clematis apiifolia protoanemonin inhibition of various yeasts and non-lactic acid bacteria [105] Momordica charantia not mentioned inhibition of Mycobacterium tuberculosis [106] Scabiosa arenaria not mentioned inhibition of some bacteria, Candida species, and phytopathogenic fungi [107] Carissa opaca not mentioned inhibition of some bacteria [108] Olea ferruginea not mentioned inhibition of some Gram (+) and Gram (−) bacteria [109] Citrullus lanatus not mentioned inhibition of S. aureus, B. subtilis, P. valgaris, and P. aerguinosa [110] Psidium cattleianum not mentioned inhibition of B. subtilis and S. aureus [111] Ribes nigrum L. not mentioned inhibition influenza virus types A and B [118] Viburnum lantana, Pyracantha coccinea, Crataegus monogyna not mentioned inhibition of S. aureus, S. epidermidis, and S. pyogenes [113] Byrsonima crassifolia, Psidium sartorianum, Crescentia alata not mentioned inhibition of E. coli, Salmonella spp., Shigella spp., and S. aureus [114] Tamarindus indica not mentioned inhibition of some human pathogenic microorganisms [115] Momordica charantia not mentioned inhibition of E. coli and Salmonella enterica [116] Atriplex inflata not mentioned inhibition of Botrytis cinerea [117] Fragaria vesca, Rubus idaeus, Vaccinium myrtillis, Vaccinium vitis-idaea anthocyanins inhibition of the replication of coxsackie virus B1 and influenza virus A/H3N2 [119] wild European cranberry not mentioned inhibition of E. coli and S. typhimurium, E. faecalis, Listeria monocytogenes, S. aureus, and B. subtilis [112] ijms-17-01258-t003_Table 3Table 3 Anti-inflammatory activities of some wild fruits. Wild Fruits Bioactive Compounds Effects References Ziziphus mistol not mentioned inhibition of LOX activity [30] Rubus ulmifolius, Sambucus nigra not mentioned inhibition of LPS-induced inflammatory mediators (NO, CCL20) [75] Psidium cattleianum not mentioned inhibition expression of COX-2 enzyme [111] wild blueberry (Rubus spp.) anthocyanin-rich, proanthocyanidin-rich, and polyphenolic-rich fraction inhibition expression of COX-2, NO, and iNOS [120] Rubus jamaicensis, Rubus rosifolius, Rubus racemosus not mentioned inhibition the expression of COX-1 and COX-2 enzymes [123] Rubus rosifolius ursolic acid analogues inhibition expression of COX-1 enzyme [124] Vaccinium vitis-idaea, Vaccinium uliginosum polyphenol-rich fraction inhibition of LPS-elicited induction of IL-1 β in RAW 264.7 cells [125] Vaccinium angustifolium phenolic acids inhibiting NF-κB activation and production of inflammatory cytokines (TNF-α and IL-6) [128] Momordica charantia phytol and lutein suppressing pro-inflammatory cytokine and MMP-9 levels, attenuating P. acnes-induced ear swelling and granulomatous inflammation in mice [129] Aristotelia chilensis not mentioned inhibition of carrageenan-induced inflammation in ear of the mouse edema in TPA inflammation mode [130,131] Solanum betaceum not mentioned antinociceptive effect on inflammatory pain mice models [132] LOX: lipoxygenase; LPS: lipopolysaccharide; NO: nitric oxide; COX-2: cyclooxygenase-2; iNOS: inducible nitric oxide synthase; TNF: Tumor necrosis factor, NF-κB: nuclear factor-κB; MMP: matrix metalloproteinase; TPA: 12-deoxyphorbol-13-decanoate; IL-1: interleukin-1; RAW 264.7: mouse macrophage cell line. ijms-17-01258-t004_Table 4Table 4 Anticancer activities of some wild fruits. Wild Fruits Bioactive Compounds Effects References Campomanesia cambessedeana, Byrsonoma verbascifolia, Pouteria guardneriana phenolic compounds inhibiting growth of HepG2 human liver cancer cells [77] Piper capense essential oil inhibiting growth of human breast adenocarcinoma, malignant melanoma, and colon carcinoma cells [45] Rubus caesius, Viburnum lantana, Crataegus monogyna, Crataegus tanacetifolia not mentioned inhibition of tumor cells [114] Momordica charantia not mentioned cytotoxic activities on human fibrosarcoma HT 1080 cells [117] Rubus rosifolius hyptatic acid B, 4-epi-nigaichigoside F1 inhibiting growth of colon tumor cells [124] Vaccinium angustifolium oligomeric proanthocyanidins fraction inhibiting growth of human prostate and mouse liver cancer cell lines [139] Rubus ellipticus not mentioned inhibiting growth of human cervical cancer cells (C33A) [140] Fragaria virginiana, F. chiloensis, F. xananassa not mentioned inhibiting growth of A549 human lung epithelial cancer cells [141] ijms-17-01258-t005_Table 5Table 5 Other bioactivities of some wild fruits. Wild Fruits Bioactive Compounds Effects References Sorbus torminalis not mentioned antiacetylcholinesterase activity [29] Genipa americana, Spondia tuberosa, Spondia purpurea chlorogenic acid antiacetylcholinesterase activity [78] Elaeagnus latifolia phenolic and flavonoid compounds protection of pUC18 DNA [37] Vaccinium angustifolium polyphenol-rich extract decreasing acetylcholinesterase activity and enhancing cognition in adult mice [98] Aristotelia chilensis aglycone and phenolic compounds inhibition of the carrageenan- induced inflammation in the paw rat and gastroprotective activity in rats [131] Tetradium glabrifolium 2-tridecanone, 2-undecanone and d-limonene larvicidal activity against the early fourth-instar larvae of A. albopictus [142] Citrullus lanatus cucurbitacin E, cucurbitacin L 2-O-β-glucoside antigiardial activities [143] Zanthoxylum schinifolium estragole, linalool and sabinene fumigant toxicity against S. zeamais [144] Carum carvi (R)-carvone and d-limonene contact toxicity against S. and T. castaneum adults [145] Brenania brieyi not mentioned estrogenic effects [146] 12 species of wild berries phenolics, anthocyanins antiglycation activity [147] Rubus idaeus not mentioned diuretic activity [148] ijms-17-01258-t006_Table 6Table 6 Bioactivities of some wild berries. Bioactivity Wild Berry Effects References antioxidant activity Rubus megalococcus scavenging free radical [78] Rubus ulmifolius scavenging free radicals (ABTS, DPPH, H2O2), antioxidant activity [85,88] Rubus hirsutus scavenging free radicals (DPPH), antioxidant activity [44] Rubus ellipticus scavenging free radicals (DPPH, ABTS), antioxidant activity [90] Rubus croceacanthus, Rubus sieboldii scavenging oxygen radicals [69] Rubus fruticosus scavenging free radical (DPPH) [96] Rubus caucasicus antioxidant activity in β-carotene-linoleic acid, DPPH free radical scavenging, and FRAP assays [152] Vaccinium meridionale scavenging free radical (ABTS), antioxidant activity [33] wild genotype of Vaccinium spp. scavenging free radicals (ABTS, superoxide anion, and hydroxyl radical) [53] Vaccinium angustifolium improving brain antioxidant properties in mice (antioxidant activity, improving ascorbic acid concentration, reducing glutathione levels, reducing lipid peroxidation products) [98] Vaccinium miyrtillus scavenging free radicals (DPPH), antioxidant activity [68] Sorbus torminalis scavenging free radicals (ABTS, superoxide anion radicals), antioxidant activity [29] Sambucus nigra scavenging free radicals (DPPH) [75] Fragaria vesca scavenging free radicals (DPPH), antioxidant activity [57,58] Sorbus domestica scavenging free radicals (ABTS, H2O2) [88] Fragaria indica scavenging free radicals (DPPH, ABTS), antioxidant activity [90] Vitis coignetiae scavenging free radicals (ABTS, DPPH) [46] antimicrobial activity wild European cranberry inhibition of E. coli and S. typhimurium, E. faecalis, Listeria monocytogenes, S. aureus, and B. subtilis [112] Rubus fruticosus inhibition of all the tested bacterial strains [96] Fragaria vesca, Rubus idaeus, Vaccinium myrtillis, Vaccinium vitis-idaea inhibition the replication of coxsackie virus B1 and influenza virus A/H3N2 [119] anti-inflammatory activity Rubus ulmifolius, Sambucus nigra inhibition of LPS-induced inflammatory mediators (NO, CCL20) [75] Rubus jamaicensis, Rubus rosifolius, Rubus racemosus inhibition the expression of COX-1 and COX-2 enzymes [123] Rubus rosifolius inhibition expression of COX-1 enzyme [124] Vaccinium vitis-idaea, Vaccinium uliginosum inhibition of LPS-elicited induction of IL-1 β in RAW 264.7 cells [125] Vaccinium angustifolium inhibiting NF-κB activation and production of inflammatory cytokines (TNF-α and IL-6) [128] anticancer activity Rubus caesius inhibition of tumor cells [114] Rubus rosifolius inhibiting growth of colon tumor cells [124] Vaccinium angustifolium inhibiting growth of human prostate and mouse liver cancer cell lines [139] Rubus ellipticus inhibiting growth of human cervical cancer cells (C33A) [140] Fragaria virginiana, F. chiloensis, F. xananassa inhibiting growth of A549 human lung epithelial cancer cells [141] ==== Refs References 1. Baliga M.S. Pai R.J. Bhat H.P. Palatty P.L. Boloor R. Chemistry and medicinal properties of the Bakul (Mimusops elengi Linn): A review Food Res. Int. 2011 44 1823 1829 10.1016/j.foodres.2011.01.063 2. Deng G.F. Lin X. Xu X.R. Gao L. Xie J.F. Li H.B. Antioxidant capacities and total phenolic contents of 56 vegetables J. Funct. Foods 2013 5 260 266 10.1016/j.jff.2012.10.015 3. Fu L. Xu B.T. Xu X.R. Qin X.S. Gan R.Y. Li H.B. Antioxidant capacities and total phenolic contents of 56 wild fruits from South China Molecules 2010 15 8602 8617 10.3390/molecules15128602 21116229 4. Xia E.Q. Deng G.F. Guo Y.J. Li H.B. Biological activities of polyphenols from grapes Int. J. Mol. Sci. 2010 11 622 646 10.3390/ijms11020622 20386657 5. Fu L. Xu B.T. Xu X.R. Gan R.Y. Zhang Y. Xia E.Q. Li H.B. Antioxidant capacities and total phenolic contents of 62 fruits Food Chem. 2011 129 345 350 10.1016/j.foodchem.2011.04.079 6. Deng G.F. Shen C. Xu X.R. Kuang R.D. Guo Y.J. Zeng L.S. Gao L.L. Lin X. Xie J.F. Xia E.Q. Potential of fruit wastes as natural resources of bioactive compounds Int. J. Mol. Sci. 2012 13 8308 8323 10.3390/ijms13078308 22942704 7. Li A.N. Li S. Zhang Y.J. Xu X.R. Chen Y.M. Li H.B. Resources and biological activities of natural polyphenols Nutrients 2014 6 6020 6047 10.3390/nu6126020 25533011 8. Giampieri F. Alvarez-Suarez J.M. Battino M. Strawberry and human health: Effects beyond antioxidant activity J. Agric. Food Chem. 2014 62 3867 3876 10.1021/jf405455n 24450925 9. Hoffmann G. Schwingshackl L. Mediterranean diet supplemented with extra virgin olive oil reduces the incidence of invasive breast cancer in a randomised controlled trial Evid. Based Med. 2016 21 72 10.1136/ebmed-2015-110366 26744413 10. Lobo V. Patil A. Phatak A. Chandra N. Free radicals, antioxidants and functional foods: Impact on human health Pharmacogn. Rev. 2010 4 118 126 10.4103/0973-7847.70902 22228951 11. Forbes-Hernandez T.Y. Giampieri F. Gasparrini M. Mazzoni L. Quiles J.L. Alvarez-Suarez J.M. Battino M. The effects of bioactive compounds from plant foods on mitochondrial function: A focus on apoptotic mechanisms Food Chem. Toxicol. 2014 68 154 182 10.1016/j.fct.2014.03.017 24680691 12. Huang W.J. Zhang X. Chen W.W. Role of oxidative stress in Alzheimer's disease (Review) Biomed. Rep. 2016 4 519 522 10.3892/br.2016.630 27123241 13. Halliwell B. Gutteridge J.M.C. Free Radicals in Biology and Medicine 3rd ed. Oxford University Press New York, NY, USA 1999 331 332 14. Lock K. Pamerleau J. Causer L. Altmann D.R. McKee M. The global burden of disease attributable to low consumption of fruit and vegetables: Implications for the global strategy on diet Bull. World Health Organ. 2005 83 100 108 15744402 15. Gescher A. Pastorino U. Plummer S.M. Manson M.M. Suppression of tumor development by substances derived from the diet-mechanisms and clinical implications Br. J. Clin. Pharmacol. 1998 45 1 12 10.1046/j.1365-2125.1998.00640.x 9489587 16. Ansari K.N. The free radicals—The hidden culprits: An update Indian J. Med. Sci. 1997 51 319 336 9567509 17. Halliwell B. Gutteridge J.M.C. Role of free radicals and catalytic metal ions in human disease: An overview Method Enzymol. 1990 186 1 85 18. Li A.N. Li S. Li H.B. Xu D.P. Xu X.R. Chen F. Total phenolic contents and antioxidant capacities of 51 edible and wild flowers J. Funct. Foods 2014 6 319 330 10.1016/j.jff.2013.10.022 19. Li S. Li S.K. Gan R.Y. Song F.L. Kuang L. Li H.B. Antioxidant capacities and total phenolic contents of infusions from 223 medicinal plants Ind. Crop Prod. 2013 51 289 298 10.1016/j.indcrop.2013.09.017 20. Deng G.F. Xu X.R. Guo Y.J. Xia E.Q. Li S. Wu S. Chen F. Ling W.H. Li H.B. Determination of antioxidant property and their lipophilic and hydrophilic phenolic contents in cereal grains J. Funct. Foods 2012 4 906 914 10.1016/j.jff.2012.06.008 21. Fu L. Xu B.T. Gan R.Y. Zhang Y. Xu X.R. Xia E.Q. Li H.B. Total phenolic contents and antioxidant capacities of herbal and tea infusions Int. J. Mol. Sci. 2011 12 2112 2124 10.3390/ijms12042112 21731430 22. Song F.L. Gan R.Y. Zhang Y. Xiao Q. Kuang L. Li H.B. Total phenolic contents and antioxidant capacities of selected Chinese medicinal plants Int. J. Mol. Sci. 2010 11 2362 2372 10.3390/ijms11062362 20640157 23. Alezandro M.R. Granato D. Genovese M.I. Jaboticaba (Myrciaria jaboticaba (Vell.) Berg), a Brazilian grape-like fruit, improves plasma lipid profile in streptozotocin-mediated oxidative stress in diabetic rats Food Res. Int. 2013 54 650 659 10.1016/j.foodres.2013.07.041 24. Margraf T. Santos E.N.T. de Andrade E.F. van Ruth S.M. Granato D. Effects of geographical origin, variety and farming system on the chemical markers and in vitro antioxidant capacity of Brazilian purple grape juices Food Res. Int. 2016 82 145 155 10.1016/j.foodres.2016.02.003 25. Macedo L.F.L. Rogero M.M. Guimaraes J.P. Granato D. Lobato L.P. Castro I.A. Effect of red wines with different in vitro antioxidant activity on oxidative stress of high-fat diet rats Food Chem. 2013 137 122 129 10.1016/j.foodchem.2012.10.017 23199999 26. Granato D. Karnopp A.R. van Ruth S.M. Characterization and comparison of phenolic composition, antioxidant capacity and instrumental taste profile of juices from different botanical origins J. Sci. Food Agric. 2015 95 1997 2006 10.1002/jsfa.6910 25213811 27. Blando F. Albano C. Liu Y.Z. Nicoletti I. Corradini D. Tommasi N. Gerardi C. Mita G. Kitts D.D. Polyphenolic composition and antioxidant activity of the under-utilised Prunus mahaleb L. fruit J. Sci. Food Agric. 2016 96 2641 2649 10.1002/jsfa.7381 26300229 28. Ramos A.S. Souza R.O.S. Boleti A.P.D.A. Bruginski E.R.D. Lima E.S. Campos F.R. Machado M.B. Chemical characterization and antioxidant capacity of the araca-pera (Psidium acutangulum ): An exotic Amazon fruit Food Res. Int. 2015 75 315 327 10.1016/j.foodres.2015.06.026 29. Hasbal G. Yilmaz-Ozden T. Can A. Antioxidant and antiacetylcholinesterase activities of Sorbus torminalis (L.) Crantz (wild service tree) fruits J. Food Drug Anal. 2015 23 57 62 10.1016/j.jfda.2014.06.006 30. Cardozo M.L. Ordonez R.M. Alberto M.R. Zampini I.C. Isla M.I. Antioxidant and anti-inflammatory activity characterization and genotoxicity evaluation of Ziziphus mistol ripe berries, exotic Argentinean fruit Food Res. Int. 2011 44 2063 2071 10.1016/j.foodres.2011.02.040 31. Barreca D. Lagana G. Ficarra S. Tellone E. Leuzzi U. Galtieri A. Bellocco E. Evaluation of the antioxidant and cytoprotective properties of the exotic fruit Annona cherimola Mill. (Annonaceae) Food Res. Int. 2011 44 2302 2310 10.1016/j.foodres.2011.02.031 32. Rawat S. Jugran A. Giri L. Bhatt I.D. Rawal R.S. Assessment of antioxidant properties in fruits of Myrica esculenta : A popular wild edible species in Indian Himalayan Region Evid. Based Complement. Altern. 2011 2011 1 8 10.1093/ecam/neq055 21785629 33. Garzon G.A. Narvaez C.E. Riedl K.M. Schwartz S.J. Chemical composition, anthocyanins, non-anthocyanin phenolics and antioxidant activity of wild bilberry (Vaccinium meridionale Swartz) from Colombia Food Chem. 2010 122 980 986 10.1016/j.foodchem.2010.03.017 34. Zheng J. Ding C. Wang L. Li G. Shi J. Li H. Wang H. Suo Y. Anthocyanins composition and antioxidant activity of wild Lycium ruthenicum Murr. from Qinghai-Tibet Plateau Food Chem. 2011 126 859 865 10.1016/j.foodchem.2010.11.052 35. Mezadri T. Villano D. Fernandez-Pachon M.S. Garcia-Parrilla M.C. Troncoso A.M. Antioxidant compounds and antioxidant activity in acerola (Malpighia emarginata DC.) fruits and derivatives J. Food Compos. Anal. 2008 21 282 290 10.1016/j.jfca.2008.02.002 36. Koca I. Ustun N.S. Koca A.F. Karadeniz B. Chemical composition, antioxidant activity and anthocyanin profiles of purple mulberry (Morus rubra ) fruits J. Food Agric. Environ. 2008 6 39 42 37. Chizzola R. Saeidnejad A.H. Azizi M. Oroojalian F. Mardani H. Bunium persicum : Variability in essential oil and antioxidants activity of fruits from different Iranian wild populations Genet. Resour. Crop Evol. 2014 61 1621 1631 10.1007/s10722-014-0158-6 38. Panja S. Chaudhuri D. Ghate N.B. Minh H.L. Mandal N. In vitro assessment of phytochemicals, antioxidant and DNA protective potential of wild edible fruit of Elaeagnus latifolia Linn Fruits 2014 69 303 314 10.1051/fruits/2014019 39. Ma C.H. Dastmalchi K. Whitaker B.D. Kennelly E.J. Two new antioxidant malonated caffeoylquinic acid isomers in fruits of wild eggplant relatives J. Agric. Food Chem. 2011 59 9645 9651 10.1021/jf202028y 21800866 40. Barros L. Carvalho A.M. Ferreira I.C.F.R. Exotic fruits as a source of important phytochemicals: Improving the traditional use of Rosa canina fruits in Portugal Food Res. Int. 2011 44 2233 2236 10.1016/j.foodres.2010.10.005 41. Nabavi S.F. Nabavi S.M. Ebrahimzadeh M.A. Asgarirad H. The antioxidant activity of wild medlar (Mespilus germanica L.) fruit, stem bark and leaf Afr. J. Biotechnol. 2011 10 283 289 42. Cespedes C.L. Valdez-Morales M. Avila J.G. El-Hafidi M. Alarcon J. Paredes-Lopez O. Phytochemical profile and the antioxidant activity of Chilean wild black-berry fruits, Aristotelia chilensis (Mol) Stuntz (Elaeocarpaceae) Food Chem. 2010 119 886 895 10.1016/j.foodchem.2009.07.045 43. Serce S. Ercisli S. Sengul M. Gunduz K. Orhan E. Antioxidant activities and fatty acid composition of wild grown myrtle (Myrtus communis L.) fruits Pharmacogn. Mag. 2010 6 9 12 10.4103/0973-1296.59960 20548930 44. Fu Y. Zhou X.Z. Chen S.G. Sun Y.J. Shen Y. Ye X.Q. Chemical composition and antioxidant activity of Chinese wild raspberry (Rubus hirsutus Thunb.) LWT Food Sci. Technol. 2015 60 1262 1268 10.1016/j.lwt.2014.09.002 45. Woguem V. Maggi F. Fogang H.P.D. Tapondjou L.A. Womeni H.M. Quassinti L. Bramucci M. Vitali L.A. Petrelli D. Lupidi G. Antioxidant, antiproliferative and antimicrobial activities of the volatile oil from the wild Pepper Piper capense used in Cameroon as a culinary spice Nat. Prod. Commun. 2013 8 1791 1796 24555300 46. Choi J.Y. Lee S.J. Lee S.J. Park S. Lee J.H. Shim J.H. El-Aty A.M.A. Jin J.S. Jeong E.D. Lee W.S. Analysis and tentative structure elucidation of new anthocyanins in fruit peel of Vitis coignetiae Pulliat (meoru) using LC-MS/MS: Contribution to the overall antioxidant activity J. Sep. Sci. 2010 33 1192 1197 20187032 47. Fraternale D. Giamperi L. Bucchini A. Ricci D. Antioxidant activity of Prunus spinosa L. fruit juice Ital. J. Food Sci. 2009 21 337 346 48. Banerjee A. Dasgupta N. De B. In vitro study of antioxidant activity of Syzygium cumini fruit Food Chem. 2005 90 727 733 10.1016/j.foodchem.2004.04.033 49. Koh K. Kim H. Hang S. Park Y. Lee C. Polyphenolic compounds and superoxide radical scavenging activity of Moru-Ju Food Sci. Biotechnol. 2003 12 290 297 50. Zhang Y.M. Sun Y.J. Xi W.P. Shen Y. Qiao L.P. Zhong L.Z. Ye X.Q. Zhou Z.Q. Phenolic compositions and antioxidant capacities of Chinese wild mandarin (Citrus reticulata Blanco) fruits Food Chem. 2014 145 674 680 10.1016/j.foodchem.2013.08.012 24128530 51. Li N. Shi J.L. Wang K. Profile and antioxidant activity of phenolic extracts from 10 crabapples (Malus Wild Species) J. Agric. Food Chem. 2014 62 574 581 10.1021/jf404542d 24392851 52. Du H. Wu J. Li H. Zhong P.X. Xu Y.J. Li C.H. Ji K.X. Wang L.S. Polyphenols and triterpenes from Chaenomeles fruits: Chemical analysis and antioxidant activities assessment Food Chem. 2013 141 4260 4268 10.1016/j.foodchem.2013.06.109 23993614 53. Braga P.C. Antonacci R. Wang Y.Y. Lattuada N. dal Sasso M. Marabini L. Fibiani M. lo Scalzo R. Comparative antioxidant activity of cultivated and wild Vaccinium species investigated by EPR, human neutrophil burst and COMET assay Eur. Rev. Med. Pharmacol. Sci. 2013 17 1987 1999 23884818 54. Akhbari M. Batooli H. Mozdianfard M. Comparative study of composition and biological activities of SDE prepared essential oils from flowers and fruits of two Hypericum species from central Iran Nat. Prod. Res. 2012 26 193 202 10.1080/14786419.2010.534994 21827289 55. Ozgen M. Serce S. Gunduz K. Yen F. Kafkas E. Paydas S. Determining total phenolics and antioxidant activity of selected Fragaria genotypes Asian J. Chem. 2007 19 5573 5581 56. Sasipriya G. Maria C.L. Siddhuraju P. Influence of pressure cooking on antioxidant activity of wild (Ensete superbum ) and commercial banana (Musa paradisiaca var. Monthan) unripe fruit and flower J. Food Sci. Technol. Technol. 2014 51 2517 2525 10.1007/s13197-012-0791-z 25328191 57. Yildiz H. Ercisli S. Hegedus A. Akbulut M. Topdas E.F. Aliman J. Bioactive content and antioxidant characteristics of wild (Fragaria vesca L.) and cultivated strawberry (Fragaria ananassa Duch.) fruits from Turkey J. Appl. Bot. Food. Qual. 2014 87 274 278 58. Dyduch-Sieminska M. Najda A. Dyduch J. Gantner M. Klimek K. The content of secondary metabolites and antioxidant activity of wild strawberry fruit (Fragaria vesca L.) J. Anal. Methods Chem. 2015 2015 1 8 10.1155/2015/831238 26539306 59. Mitic M.N. Kostic D.A. Pavlovic A.N. Dimitrijevic D.S. Veljkovic J.N. Effects of solvent extraction system on concentration and antioxidant activity of strawberry phenolics Agro FOOD Ind. Hi Tech 2014 25 24 28 60. Lee H.H. Moon Y.S. Yun H.K. Park P.J. Kwak E.J. Contents of bioactive constituents and antioxidant activities of cultivated and wild raspberries Korean J. Hortic. Sci. Technol. 2014 32 115 122 10.7235/hort.2014.13114 61. Pu F. Ren X.L. Zhang X.P. Phenolic compounds and antioxidant activity in fruits of six Diospyros kaki genotypes Eur. Food Res. Technol. 2013 237 923 932 10.1007/s00217-013-2065-z 62. Roman L. Stanila A. Stanila S. Bioactive compounds and antioxidant activity of Rosa canina L. biotypes from spontaneous flora of Transylvania Chem. Cent. J. 2013 7 73 10.1186/1752-153X-7-73 23618509 63. Lu Y.L. Liu Y.H. Liang W.L. Chyuan J.H. Cheng K.T. Liang W.L. Hou W.C. Antioxidant activities of different wild bitter gourd (Momordica charantia L. var. abbreviata Seringe) cultivars Bot. Stud. 2012 53 207 214 64. Esfahlan A. Jamei R. Properties of biological activity of ten wild almond (Prunus amygdalus L.) species Turk. J. Biol. 2012 36 201 209 65. Bunea A. Rugina D.O. Pintea A.M. Sconta Z. Bunea C.I. Socaciu C. Comparative polyphenolic content and antioxidant activities of some wild and cultivated blueberries from Romania Not. Bot. Horti Agrobot. Cluj-Napoca 2011 39 70 76 66. Ghafar M.F.A. Prasad K.N. Weng K.K. Ismail A. Flavonoid, hesperidine, total phenolic contents and antioxidant activities from Citrus species Afr. J. Biotechnol. 2010 9 326 330 67. Isfahlan A.J. Mahmoodzadeh A. Hassanzadeh A. Heidari R. Jamei R. Antioxidant and antiradical activities of phenolic extracts from Iranian almond (Prunus amygdalus L.) hulls and shells Turk. J. Biol. 2010 34 165 173 68. Giovanelli G. Buratti S. Comparison of polyphenolic composition and antioxidant activity of wild Italian blueberries and some cultivated varieties Food Chem. 2009 112 903 908 10.1016/j.foodchem.2008.06.066 69. Kubota M. Ishikawa C. Sugiyama Y. Fukumoto S. Miyagi T. Kumazawa S. Anthocyanins from the fruits of Rubus croceacanthus and Rubus sieboldii , wild berry plants from Okinawa, Japan J. Food Compos. Anal. 2012 28 179 182 10.1016/j.jfca.2012.09.002 70. Mazur B. Borowska E.J. Polak M. Content of vitamin C in and antioxidant capacity of wild and cultivated cranberry fruit and of their pulps Zywnosc Nauka Technol. Jakosc 2009 16 130 137 71. Vollmannova A. Musilova J. Toth T. Arvay J. Bystricka J. Medvecky M. Daniel J. Phenolic compounds, antioxidant activity and Cu, Zn, Cd and Pb content in wild and cultivated cranberries and blueberries Int. J. Environ. Anal. Chem. 2014 94 1445 1451 10.1080/03067319.2014.974588 72. Augusto T.R. Salinas E.S.S. Alencar S.M. D'Arce M.A.B.R. Camargo A.C.D. Vieira T.M.F.D. Phenolic compounds and antioxidant activity of hydroalcoholic extracts of wild and cultivated murtilla (Ugni molinae Turcz.) Food Sci. Technol. 2014 34 667 679 10.1590/1678-457X.6393 73. Netzel M. Netzel G. Tian Q.G. Schwartz S. Konczak I. Native Australian fruits - a novel source of antioxidants for food Innov. Food Sci. Emerg. Technol. 2007 8 339 346 10.1016/j.ifset.2007.03.007 74. Lamien-Meda A. Lamien C.E. Compaore M.M.Y. Meda R.N.T. Kiendrebeogo M. Zeba B. Millogo J.F. Nacoulma O.G. Polyphenol content and antioxidant activity of fourteen wild edible fruits from Burkina Faso Molecules 2008 13 581 594 10.3390/molecules13030581 18463567 75. Fazio A. Plastina P. Meijerink J. Witkamp R.W. Gabriele B. Comparative analyses of seeds of wild fruits of Rubus and Sambucus species from Southern Italy: Fatty acid composition of the oil, total phenolic content, antioxidant and anti-inflammatory properties of the methanolic extracts Food Chem. 2013 140 817 824 10.1016/j.foodchem.2012.11.010 23692771 76. Malta L.G. Tessaro E.P. Eberlin M. Pastore G.M. Liu R.H. Assessment of antioxidant and antiproliferative activities and the identification of phenolic compounds of exotic Brazilian fruits Food Res. Int. 2013 53 417 425 10.1016/j.foodres.2013.04.024 77. Omena C.M.B. Valentim I.B. Guedes G.D. Rabelo L.A. Mano C.M. Bechara E.J.H. Sawaya A.C.H.F. Trevisan M.T.S. da Costa J.G. Ferreira R.C.S. Antioxidant, anti-acetylcholinesterase and cytotoxic activities of ethanol extracts of peel, pulp and seeds of exotic Brazilian fruits Food Res. Int. 2012 49 334 344 10.1016/j.foodres.2012.07.010 78. Santacruz L. Carriazo J.G. Almanza O. Osorio C. Anthocyanin composition of wild Colombian fruits and antioxidant capacity measurement by electron paramagnetic resonance spectroscopy J. Agric. Food Chem. 2012 60 1397 1404 10.1021/jf2042533 22242913 79. Almeida M.M.B. de Sousa P.H.M. Arriaga A.M.C. do Prado G.M. Magalhaes C.E.D.C. Maia G.A. de Lemos T.L.G. Bioactive compounds and antioxidant activity of fresh exotic fruits from northeastern Brazil Food Res. Int. 2011 44 2155 2159 10.1016/j.foodres.2011.03.051 80. Chalise J.P. Acharya K. Gurung N. Bhusal R.P. Gurung R. Skalko-Basnet N. Basnet P. Antioxidant activity and polyphenol content in edible wild fruits from Nepal Int. J. Food Sci. Nutr. 2010 61 425 432 10.3109/09637481003591590 20187713 81. Ndhlala A.R. Mupure C.H. Chitindingu K. Benhura M.A. Muchuweti M. Antioxidant potentials and degrees of polymerization of six wild fruits Sci. Res. Essays 2006 1 87 92 82. Genovese M.I. Pinto M.D.S. Goncalves A.E.D.S. Lajolo F.M. Bioactive compounds and antioxidant capacity of exotic fruits and commercial frozen pulps from Brazil Food Sci. Technol. Int. 2008 14 207 214 10.1177/1082013208092151 83. Oszmianski J. Nowicka P. Teleszko M. Wojdylo A. Cebulak T. Oklejewicz K. Analysis of phenolic compounds and antioxidant activity in wild blackberry fruits Int. J. Mol. Sci. 2015 16 14540 14553 10.3390/ijms160714540 26132562 84. Sharma P.B. Handique P.J. Devi H.S. Antioxidant properties, physico-chemical characteristics and proximate composition of five wild fruits of Manipur, India J. Food Sci. Technol. 2015 52 894 902 10.1007/s13197-013-1128-2 25694698 85. Ruiz-Rodriguez B.M. Sanchez-Moreno C. de Ancos B. Sanchez-Mata M.D. Fernandez-Ruiz V. Camara M. Tardio J. Wild Arbutus unedo L. and Rubus ulmifolius Schott fruits are underutilized sources of valuable bioactive compounds with antioxidant capacity Fruits 2014 69 435 448 10.1051/fruits/2014035 86. Ruiz-Rodriguez B.M. de Ancos B. Sanchez-Moreno C. Fernandez-Ruiz V. Sanchez-Mata M.D. Camara M. Tardio J. Wild blackthorn (Prunus spinosa L.) and hawthorn (Crataegus monogyna Jacq.) fruits as valuable sources of antioxidants Fruits 2014 69 61 73 10.1051/fruits/2013102 87. Rufino M.S.M. Alves R.E. Fernandes F.A.N. Brito E.S. Free radical scavenging behavior of ten exotic tropical fruits extracts Food Res. Int. 2011 44 2072 2075 10.1016/j.foodres.2010.07.002 88. Egea I. Sanchez-Bel P. Romojaro F. Pretel M.T. Six edible wild fruits as potential antioxidant additives or nutritional supplements Plant Food Hum. Nutr. 2010 65 121 129 10.1007/s11130-010-0159-3 20198440 89. Ndhlala A.R. Chitindingu K. Mupure C. Murenje T. Ndhlala F. Benhura M.A. Muchuweti M. Antioxidant properties of methanolic extracts from Diospyros mespiliformis (jackal berry), Flacourtia indica (Batoka plum), Uapaca kirkiana (wild loquat) and Ziziphus mauritiana (yellow berry) fruits Int. J. Food Sci. Technol. 2008 43 284 288 10.1111/j.1365-2621.2006.01431.x 90. Saklani S. Badhani A. Mishra A.P. Chandra S. Health promoting phytochemicals their concentration and antioxidant activity of wild edible fruits of Uttarakhand, India Asian. J. Chem. 2012 24 5558 5560 91. Dembitsky V.M. Poovarodom S. Leontowicz H. Leontowicz M. Vearasilp S. Trakhtenberg S. Gorinstein S. The multiple nutrition properties of some exotic fruits: Biological activity and active metabolites Food Res. Int. 2011 44 1671 1701 10.1016/j.foodres.2011.03.003 92. Contreras-Calderon J. Calderon-Jaimes L. Guerra-Hernandez E. Garcia-Villanova B. Antioxidant capacity, phenolic content and vitamin C in pulp, peel and seed from 24 exotic fruits from Colombia Food Res. Int. 2011 44 2047 2053 10.1016/j.foodres.2010.11.003 93. De Assis S. Vellosa J.C.R. Brunetti I.L. Khalil N.M. Leite K.M.D.C. Martins A.B.G. Oliveira O.M.M.D. Antioxidant activity, ascorbic acid and total phenol of exotic fruits occurring in Brazil Int. J. Food Sci. Nutr. 2009 60 439 448 10.1080/09637480701780641 18785051 94. Ribeiro A.B. Bonafe E.G. Silva B.C. Montanher P.F. Santos Junior O.O. Boeing J.S. Visentainer J.V. Antioxidant capacity, total phenolic content, fatty acids and correlation by principal component analysis of exotic and native fruits from Brazil J. Braz. Chem. Soc. 2013 24 797 804 10.5935/0103-5053.20130105 95. Luximon-Ramma A. Bahorun T. Crozier A. Antioxidant actions and phenolic and vitamin C contents of common Mauritian exotic fruits J. Sci. Food Agric. 2003 83 496 502 10.1002/jsfa.1365 96. Radovanovic B.C. Andelkovic A.S.M. Radovanovic A.B. Andelkovic M.Z. Antioxidant and antimicrobial activity of polyphenol extracts from wild berry fruits grown in Southeast Serbia Trop. J. Pharm. Res. 2013 12 813 819 10.4314/tjpr.v12i5.23 97. Leontowicz H. Leontowicz M. Drzewiecki J. Haruenkit R. Poovarodom S. Park Y.S. Jung S.T. Kang S.G. Trakhtenberg S. Gorinstein S. Bioactive properties of snake fruit (Salacca edulis Reinw) and mangosteen (Garcinia mangostana ) and their influence on plasma lipid profile and antioxidant activity in rats fed cholesterol Eur. Food Res. Technol. 2006 223 697 703 10.1007/s00217-006-0255-7 98. Papandreou M.A. Dimakopoulou A. Linardaki Z.I. Cordopatis P. Klimis-Zacas D. Margarity M. Lamari F.N. Effect of a polyphenol-rich wild blueberry extract on cognitive performance of mice, brain antioxidant markers and acetylcholinesterase activity Behav. Brain Res. 2009 198 352 358 10.1016/j.bbr.2008.11.013 19056430 99. Gorinstein S. Poovarodom S. Leontowicz H. Leontowicz M. Namiesnik J. Vearasilp S. Haruenkit R. Ruamsuke P. Katrich E. Tashma Z. Antioxidant properties and bioactive constituents of some rare exotic Thai fruits and comparison with conventional fruits in vitro and in vivo studies Food Res. Int. 2011 44 2222 2232 10.1016/j.foodres.2010.10.009 100. Cilerdzic J. Vukojevc J. Stajic M. Stanojkovic T. Glamoclija J. Biological activity of Ganoderma lucidum basidiocarps cultivated on alternative and commercial substrate J. Ethnopharmacol. 2014 155 312 319 10.1016/j.jep.2014.05.036 24879959 101. Garcia M.E. Blanco J.L. Mycosis in domestic animals Rev. Iberoam. Micol. 2000 17 2 7 15762785 102. Mariem C. Sameh M. Nadhem S. Soumaya Z. Najiba Z. Raoudha E.G. Antioxidant and antimicrobial properties of the extracts from Nitraria retusa fruits and their applications to meat product preservation Ind. Crops Prod. 2014 55 295 303 10.1016/j.indcrop.2014.01.036 103. Belkhir M. Rebai O. Dhaouadi K. Congiu F. Tuberoso C. Amri M. Fattouch S. Comparative analysis of Tunisian wild Crataegus azarolus (Yellow Azarole) and Crataegus monogyna (Red Azarole) leaf, fruit, and traditionally derived syrup: Phenolic profiles and antioxidant and antimicrobial activities of the aqueous-acetone extracts J. Agric. Food Chem. 2013 61 9594 9601 10.1021/jf405285m 24070066 104. Ozcelik B. Koca U. Kaya D. Sekeroglu N. Evaluation of the in vitro bioactivities of mahaleb cherry (Prunus mahaleb L.) Romanian Biotech. Lett. 2012 17 7863 7872 105. Kyung K.H. Woo Y.H. Kim D.S. Park H.J. Kim Y.S. Antimicrobial activity of an edible wild plant, apiifolia Virgin's Bower (Clematis apiifolia DC) Food Sci. Biotechnol. 2007 16 1051 1054 106. Panlilio B.G. Franzblau S. Aguinaldo A.M. Variability of the antitubercular activity of Ampalaya (Momordica charantia L.) from selected provinces in Luzon, Philippines Philipp. Agric. Sci. 2007 90 337 340 107. Besbes M. Omri A. Cheraif I. Daami M. Ben Jannet H. Mastouri M. Aouni M. Selmi B. Chemical composition and antimicrobial activity of essential oils from Scabiosa arenaria Forssk growing wild in Tunisia Chem. Biodivers. 2012 9 829 839 10.1002/cbdv.201100191 22492499 108. Ahmed D. Waheed A. Chaudhary M.A. Khan S.R. Hannan A. Barkaat M. Nutritional and antimicrobial studies on leaves and fruit of Carissa opaca Stapf ex Haines Asian J. Chem. 2011 23 2072 2076 109. Hussain A. Qarshi I.A. Liaqat R. Akhtar S. Aziz I. Ullah I. Shinwari Z.K. Antimicrobial potential of leaf and fruit extracts and oils of wild and cultivated edible olive Pak. J. Bot. 2014 46 1463 1468 110. Hassan L.E.A. Sirat H.M. Yagi S.M.A. Koko W.S. Abdelwahab S.I. In vitro antimicrobial activities of chloroformic, hexane and ethanolic extracts of Citrullus lanatus var. citroides (wild melon) J. Med. Plants Res. 2011 5 1338 1344 111. McCook-Russell K.P. Nair M.G. Facey P.C. Bowen-Forbes C.S. Nutritional and nutraceutical comparison of Jamaican Psidium cattleianum (strawberry guava) and Psidium guajava (common guava) fruits Food Chem. 2012 134 1069 1073 10.1016/j.foodchem.2012.03.018 23107729 112. Cesoniene L. Jasutiene I. Sarkinas A. Phenolics and anthocyanins in berries of European cranberry and their antimicrobial activity Med. Lith. 2009 45 992 999 113. Turker A.U. Yildirim A.B. Karakas F.P. Antibacterial and antitumor activities of some wild fruits grown in Turkey Biotechnol. Biotechnol. Equip. 2011 26 2765 2772 10.5504/BBEQ.2011.0156 114. Pio-Leon J.F. Diaz-Camacho S.P. Lopez-Lopez M.A. Uribe-Beltran M.D. Willms K. Lopez-Angulo G. Montes-Avila J. Delgado-Vargas F. Antibacterial activity of extracts obtained from the nanchi (Byrsonima crassifolia (L.) Kunth), arrayan (Psidium sartorianum (O. Berg) Nied.) and ayale (Crescentia alata Kunth) fruits Bol. Latinoam. Caribe Plantas Med. 2013 12 356 364 115. Adeola A.A. Adeola O.O. Dosumu O.O. Comparative analyses of phytochemicals and antimicrobial properties of extracts of wild Tamarindus indica pulps Afr. J. Microbiol. Res. 2010 4 2769 2779 116. Lu Y.L. Liu Y.H. Liang W.L. Chuang J.H. Cheng K.T. Liang H.J. Hou W.C. Antibacterial and cytotoxic activities of different wild bitter gourd cultivars (Momordica charantia L. var. abbreviata Seringe) Bot. Stud. 2011 52 427 434 117. Boughalleb N. Trabelsi L. Harzallah-Skhiri F. Antifungal activity from polar and non-polar extracts of some Chenopodiaceae wild species growing in Tunisia Nat. Prod. Res. 2009 23 988 997 10.1080/14786410802168494 19452345 118. Knox Y.M. Suzutani T. Yosida I. Azuma M. Anti-influenza virus activity of crude extract of Ribes nigrum L. Phytother. Res. 2003 17 120 122 10.1002/ptr.1053 12601672 119. Nikolaeva-Glomb L. Mukova L. Nikolova N. Badjakov I. Dincheva I. Kondakova V. Doumanova L. Galabov A.S. In vitro antiviral activity of a series of wild berry fruit extracts against representatives of Picorna-, Orthomyxo- and Paramyxoviridae Nat. Prod. Commun. 2014 9 51 54 24660461 120. Cuevas-Rodriguez E.O. Dia V.P. Yousef G.G. Garcia-Saucedo P.A. Lopez-Medina J. Paredes-Lopez O. Gonzalez De Mejia E. Lila M.A. Inhibition of pro-inflammatory responses and antioxidant capacity of Mexican blackberry (Rubus spp.) extracts J. Agric. Food Chem. 2010 58 9542 9548 10.1021/jf102590p 20715775 121. Stamler J.S. Single D. Loscalzo J. Biochemistry of nitric oxide and its redox-activated forms Science 1992 258 1892 1902 10.1126/science.1281928 122. Varesio L. Battaglia F. Raggi F.L.B. Bosco M.C. Macrophage inflammatory protein-3α/CCL-20 is transcriptionally induced by the iron chelator desferrioxamine in human mononuclear phagocytes through nuclear factor (NF)-κB Mol. Immunol. 2010 47 685 693 10.1016/j.molimm.2009.10.031 19939449 123. Bowen-Forbes C.S. Zhang Y. Nair M.G. Anthocyanin content, antioxidant, anti-inflammatory and anticancer properties of blackberry and raspberry fruits J. Food Compos. Anal. 2010 23 554 560 10.1016/j.jfca.2009.08.012 124. Bowen-Forbes C.S. Mulabagal V. Liu Y. Nair M.G. Ursolic acid analogues: non-phenolic functional food components in Jamaican raspberry fruits Food Chem. 2009 116 633 637 10.1016/j.foodchem.2009.02.075 125. Grace M.H. Esposito D. Dunlap K.L. Lila M.A. Comparative analysis of phenolic content and profile, antioxidant capacity, and anti-inflammatory bioactivity in wild Alaskan and commercial Vaccinium Berries J. Agric. Food Chem. 2014 62 4007 4017 10.1021/jf403810y 24219831 126. Karin M. Ben-Neriah Y. Phosphorylation meets ubiquitination: The control of NF-kappa B activity Annu. Rev. Immunol. 2000 18 621 663 10.1146/annurev.immunol.18.1.621 10837071 127. Karin M. Yamamoto Y. Wang Q.M. The IKK NF-κB system: A treasure trove for drug development Nat. Rev. Drug Discov. 2004 3 17 26 10.1038/nrd1279 14708018 128. Kang J. Thakali K.M. Jensen G.S. Wu X.L. Phenolic acids of the two major blueberry species in the US market and their antioxidant and anti-inflammatory activities Plant Food Hum. Nutr. 2015 70 56 62 10.1007/s11130-014-0461-6 25535004 129. Hsu C. Tsai T.H. Li Y.Y. Wu W.H. Huang C.J. Tsai P.J. Wild bitter melon (Momordica charantia Linn. var. abbreviata Ser.) extract and its bioactive components suppress Propionibacterium acnes -induced inflammation Food Chem. 2012 135 976 984 10.1016/j.foodchem.2012.05.045 22953813 130. Cespedes C.L. Alarcon J. Avila J. Nieto A. Anti-inflammatory activity of Aristotelia chilensis Mol. (Stuntz) (Elaeocarpaceae) Bol. Latinoam. Caribe Plantas Med. 2010 9 127 135 131. Cespedes C.L. Alarcon J. Avila J. El-Hafidi M. Anti-inflammatory, antioedema and gastroprotective activities of Aristotelia chilensis extracts, Part 2 Bol. Latinoam. Caribe Plantas Med. 2010 9 432 439 132. Do Nascimento G.E. Hamm L.A. Baggio C.H. Werner M.F.D. Iacomini M. Cordeiro L.M.C. Structure of a galactoarabinoglucuronoxylan from tamarillo (Solanum betaceum ), a tropical exotic fruit, and its biological activity Food Chem. 2013 141 510 516 10.1016/j.foodchem.2013.03.023 23768387 133. Chu Y.F. Sun J. Wu X.Z. Liu R.H. Antioxidant and antiproliferative activities of common vegetables J. Agric. Food Chem. 2002 50 6910 6916 10.1021/jf020665f 12405796 134. Liu R.H. Health benefits of fruits and vegetables are from additive and synergistic combinations of phytochemicals Am. J. Clin. Nutr. 2003 78 517S 520S 12936943 135. Li F. Li S. Li H.B. Deng G.F. Ling W.H. Wu S. Xu X.R. Chen F. Antiproliferative activity of peels, pulps and seeds of 61 fruits J. Funct. Foods 2013 5 1298 1309 10.1016/j.jff.2013.04.016 136. Zhou Y. Li Y. Zhou T. Zheng J. Li S. Li H.B. Dietary natural products for prevention and treatment of liver cancer Nutrients 2016 8 156 10.3390/nu8030156 26978396 137. Li F. Li S. Li H.B. Deng G.F. Ling W.H. Xu X.R. Antiproliferative activities of tea and herbal infusions Food Funct. 2013 4 530 538 10.1039/c2fo30252g 23307138 138. Bagchi D. Swaroop A. Preuss H.G. Bagchi M. Free radical scavenging, antioxidant and cancer chemoprevention by grape seed proanthocyanidin: An overview Mutat. Res. Fundam. Mol. Mech. Mutagen. 2014 768 69 73 10.1016/j.mrfmmm.2014.04.004 24751946 139. Schmidt B.M. Howell A.B. McEniry B. Knight C.T. Seigler D. Erdman J.W. Lila M.A. Effective separation of potent antiproliferation and antiadhesion components from wild blueberry (Vaccinium angustifolium Ait.) fruits J. Agric. Food Chem. 2004 52 6433 6442 10.1021/jf049238n 15479003 140. Saini R. Dangwal K.A. Singh H. Garg V. Antioxidant and antiproliferative activities of phenolics isolated from fruits of Himalayan yellow raspberry (Rubus ellipticus ) J. Food Sci. Technol. 2014 51 3369 3375 10.1007/s13197-012-0836-3 26396333 141. Wang S.Y. Lewers K.S. Bowman L. Ding M. Antioxidant activities and anticancer cell proliferation properties of wild strawberries J. Am. Soc. Hortic. Sci. 2007 132 647 658 142. Liu X.C. Liu Q. Chen X.B. Zhou L. Liu Z.L. Larvicidal activity of the essential oil from Tetradium glabrifolium fruits and its constituents against Aedes albopictus Pest Manag. Sci. 2015 71 1582 1586 10.1002/ps.3964 25504672 143. Hassan L.E.A. Koko W.S. Osman E.B.E. Dahab M.M. Sirat H.M. In vitro antigiardial activity of Citrullus lanatus Var. citroides extracts and cucurbitacins isolated compounds J. Med. Plants Res. 2011 5 3338 3346 144. Wang C.F. Yang K. Zhang H.M. Cao J. Fang R. Liu Z.L. Du S.S. Wang Y.Y. Deng Z.W. Zhou L.G. Components and insecticidal activity against the Maize Weevils of Zanthoxylum schinifolium fruits and leaves Molecules 2011 16 3077 3088 10.3390/molecules16043077 21490556 145. Fang R. Jiang C.H. Wang X.Y. Zhang H.M. Liu Z.L. Zhou L. Du S.S. Deng Z.W. Insecticidal activity of essential oil of Carum Carvi fruits from China and its main components against two grain storage insects Molecules 2010 15 9391 9402 10.3390/molecules15129391 21173724 146. Nde C. Njamen D. Mbanya J.C. Zierau O. Vollmer G. Fommn Z.T. Estrogenic effects of a methanol extract of the fruit of Brenania brieyi de Wild (Rubiaceae) J. Nat. Med. 2007 61 86 89 147. Harris C.S. Cuerrier A. Lamont E. Haddad P.S. Arnason J.T. Bennett S.A.L. Johns T. Investigating wild berries as a dietary approach to reducing the formation of advanced glycation endproducts: Chemical correlates of in vitro antiglycation activity Plant Food Hum. Nutr. 2014 69 71 77 10.1007/s11130-014-0403-3 24448675 148. Zhang Y. Zhang Z.Y. Yang Y.N. Zu X.Y. Guan D. Wang Y.P. Diuretic activity of Rubus idaeus L. (Rosaceae) in rats Trop. J. Pharm. Res. 2011 10 243 248 10.4314/tjpr.v10i3.15 149. Skrovankova S. Sumczynski D. Mlcek J. Jurikova T. Sochor J. Bioactive compounds and antioxidant activity in different types of berries Int. J. Mol. Sci. 2015 16 24673 24706 10.3390/ijms161024673 26501271 150. Namiesnik J. Vearasilp K. Nemirovski A. Leontowicz H. Leontowicz M. Pasko P. Martinez-Ayala A.L. Gonzalez-Aguilar G.A. Suhaj M. Gorinstein S. In vitro studies on the relationship between the antioxidant activities of some berry extracts and their binding properties to serum albumin Appl. Biochem. Biotechnol. 2014 172 2849 2865 10.1007/s12010-013-0712-2 24449373 151. De Souza V.R. Pereira P.A.P. da Silva T.L.T. Lima L.C.D. Pio R. Queiroz F. Determination of the bioactive compounds, antioxidant activity and chemical composition of Brazilian blackberry, red raspberry, strawberry, blueberry and sweet cherry fruits Food Chem. 2014 156 362 368 10.1016/j.foodchem.2014.01.125 24629981 152. Yildiz H. Sengul M. Celik F. Hegedus A. Ercisli S. Tosun M. Some phytochemical and antioxidant characteristics of wild and cultivated blackberry (Rubus caucasicus ) fruits J. Food Agric. Environ. 2010 8 156 159

PMC005xxxxxx/PMC5000657.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081259ijms-17-01259ArticleFibroblast-Derived Extracellular Matrix Induces Chondrogenic Differentiation in Human Adipose-Derived Mesenchymal Stromal/Stem Cells in Vitro Dzobo Kevin 12*Turnley Taegyn 12Wishart Andrew 12Rowe Arielle 1Kallmeyer Karlien 3van Vollenstee Fiona A. 3Thomford Nicholas E. 4Dandara Collet 4Chopera Denis 5Pepper Michael S. 3Parker M. Iqbal 12Rahaman Mohamed N. Academic Editor1 International Centre for Genetic Engineering and Biotechnology (ICGEB), Cape Town Component, Anzio Road, Observatory, Cape Town 7925, South Africa; [email protected] (T.T.); [email protected] (A.W.); [email protected] (A.R.); [email protected] (M.I.P.)2 Division of Medical Biochemistry, Institute of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, Anzio Road, Observatory, Cape Town 7925, South Africa3 Department of Immunology, Institute for Cellular and Molecular Medicine, South African Medical Research Council (SAMRC) Extramural Unit for Stem Cell Research and Therapy, Faculty of Health Sciences, University of Pretoria, Pretoria 0002, South Africa; [email protected] (K.K.); [email protected] (F.A.v.V.); [email protected] (M.S.P.)4 Division of Human Genetics, Department of Pathology, Institute of Infectious Diseases and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, Anzio Road, Observatory, Cape Town 7925, South Africa; [email protected] (N.E.T.); [email protected] (C.D.)5 Division of Immunology, Faculty of Health Sciences, University of Cape Town, Anzio Road, Observatory, Cape Town 7925, South Africa; [email protected]* Correspondence: [email protected]; Tel.: +27-21-404-7689; Fax: +27-21-406-606003 8 2016 8 2016 17 8 125905 7 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Mesenchymal stromal/stem cells (MSCs) represent an area being intensively researched for tissue engineering and regenerative medicine applications. MSCs may provide the opportunity to treat diseases and injuries that currently have limited therapeutic options, as well as enhance present strategies for tissue repair. The cellular environment has a significant role in cellular development and differentiation through cell–matrix interactions. The aim of this study was to investigate the behavior of adipose-derived MSCs (ad-MSCs) in the context of a cell-derived matrix so as to model the in vivo physiological microenvironment. The fibroblast-derived extracellular matrix (fd-ECM) did not affect ad-MSC morphology, but reduced ad-MSC proliferation. Ad-MSCs cultured on fd-ECM displayed decreased expression of integrins α2 and β1 and subsequently lost their multipotency over time, as shown by the decrease in CD44, Octamer-binding transcription factor 4 (OCT4), SOX2, and NANOG gene expression. The fd-ECM induced chondrogenic differentiation in ad-MSCs compared to control ad-MSCs. Loss of function studies, through the use of siRNA and a mutant Notch1 construct, revealed that ECM-mediated ad-MSCs chondrogenesis requires Notch1 and β-catenin signaling. The fd-ECM also showed anti-senescence effects on ad-MSCs. The fd-ECM is a promising approach for inducing chondrogenesis in ad-MSCs and chondrogenic differentiated ad-MSCs could be used in stem cell therapy procedures. mesenchymal stromal/stem cellsregenerative medicinethree-dimensionalextracellular matrixdifferentiationchondrogenesis ==== Body 1. Introduction The extracellular matrix (ECM) is a complex network of a variety of components that interact to create a molecular scaffold [1,2]. The ECM is secreted by tissue-specific cells that are found mostly in connective tissue and is able to support cells within its reach [2,3]. It acts as a substrate for cell attachment and migration, while presenting physical and chemical cues to cells. ECM molecules include proteoglycans, collagens, glycosaminoglycans, and non-collagenous glycoproteins. Cells continue to interact with the ECM they produce and that produced by other cells. The composition of the ECM is in a constant flux and varies depending on factors such as cell and tissue type. The mechanical and biochemical properties of the ECM are the two key factors influencing cellular behavior [4,5,6]. The extracellular matrices (ECMs) of different tissues show differences that provide specific signaling for cells of that tissue. ECMs of specific tissues induce cell proliferation that is specific for that tissue and is involved in maintenance of that cell phenotype [7]. The ECM regulates the function of cells and drives cellular differentiation. Thus ECMs show bioactivities including inducing proliferation and differentiation [2,5,8,9]. The in vitro bioactivities of many synthetic and natural ECMs are not fully understood. Our previous work has demonstrated that ECMs play a role in the specification of cell fate, especially that of stem cells [10,11]. One of the greatest challenges in utilizing stem cells for tissue repair is directing gene regulation and guiding stem cell differentiation toward a specific lineage [1,4,9,10,12,13,14]. Mesenchymal stromal/stem cells (MSCs) have been shown to be able to proliferate considerably and can differentiate into various tissue lineages when induced. For the use of MSCs in regenerative or reparative therapeutic processes to become a reality, an understanding of the processes and signaling pathways that effectively induce MSCs differentiation along key lineages is imperative. Several studies have been able to exhibit this with varying degrees of success [9,10,13,14]. However, the lack of consistency and the inability to replicate the same results with repeat experiments is still a challenge. The current methods used for MSCs expansion in vitro limit the use of MSCs due to many reasons including impaired multi-lineage differentiation during monolayer culture [15]. Thus research aimed at directing MSC differentiation along certain lineages is ongoing. Recent work on the use of matrices to drive stem cell differentiation has focused on the use of combinatorial matrices and hydrogels [16,17,18,19]. There is a gradual realization that three-dimensional ECM (3D ECM) models better recapitulate the in vivo situation, as cells reside and differentiate within a 3D microenvironment in the body [4,5]. The use of 3D ECM models in tissue engineering studies will also provide more accurate data or outcomes regarding the relevance and potential use of the information in clinical applications. The incorporation of MSCs into synthetic biomaterial scaffolds such as hydrogels and polymers such as polyglycolic acid (PGA) are some of the techniques being used to expand and differentiate MSCs into the desired cells in vitro [16,17,18,19]. However, these synthetic polymer-based scaffolds lack biocompatibility and are not suitable when cells are intended to be used therapeutically. Therefore, more ECMs and their combinations are being researched in order to understand their effect on MSC proliferation and differentiation. Recent studies have shown that cell-derived extracellular matrices (cd-ECMs) can promote the proliferation of cells such as fibroblasts and MSCs, and are able to maintain a high responsiveness in vitro [4,5,20]. In vivo, MSCs are found within microenvironments where the ECM is also synthesized by cells such as fibroblasts. In particular, ad-MSCs are found in close proximity to several cell types collectively termed the stromal vascular fraction, which is rich in stromal cells such as preadipocytes, fibroblasts, and macrophages; therefore, the ECM will mostly be of stromal origin [7,21]. It is therefore reasonable to study how a fibroblast-derived ECM will affect ad-MSCs in vitro instead of using synthetic ECMs. In this study, therefore, we used a fibroblast-derived extracellular matrix (fd-ECM) to evaluate its effect on ad-MSCs, focusing specifically on ad-MSC proliferation, attachment, migration, and possible differentiation. We also determined the signaling pathways perturbed by the fd-ECM. We show in this study that both Notch1 and β-catenin signaling were activated by the presence of the fd-ECM and this led to chondrogenic differentiation of ad-MSCs. The fd-ECM also has an anti-senescence effect and may improve the survival of ad-MSCs in culture. 2. Results 2.1. Adipose-Derived Mesenchymal Stromal/Stem Cells (MSCs) Characterization Ad-MSCs were characterized by identification of cell surface markers via flow cytometric analysis. Ad-MSCs expressed CD73, CD90, and CD 105, and did not express CD34 and CD45 (Figure 1A–F). Classic phenotypic characterization of MSCs includes the expression of CD73+, CD90+, CD105+, CD34− and CD45− in ≥95% of the cell population (Figure 1F). Ad-MSCs were able to differentiate into multiple lineages (adipogenic, osteogenic, and chondrogenic) after cultivation in the respective differentiation media for 21 days (Figure 2A–F). Staining was done to show lipid vacuoles, calcium deposits, and glycosaminoglycans. Ad-MSCs were further characterized by evaluating the expression of several mesenchymal stromal/stem and self-renewal markers, such as vimentin, CD44, octamer-binding transcription factor 4 (Oct4), and Nanog via immunoblot analysis (Figure 3A–C). Ad-MSCs from three donors were found to express the MSC marker vimentin and CD44. Our data also show that ad-MSCs can differentiate into mesodermal cell lineage (Wnt Family Member 3a: WNT3A, N-CADHERIN mRNA), ectodermal differentiation (CALBINDIN2 and RECOVERIN mRNA), and endodermal differentiation (PANCREATIC AMYLASE mRNA, SOX17 mRNA) (Figure S1A–F). Sub-culturing the ad-MSCs did not affect the expression of Oct4 or Nanog, showing that ad-MSCs maintain their multipotency and there is no induction of apoptosis over several passages (Figure 3B–D). Flow cytometric analysis showed that, over several passages, the ad-MSCs maintained the same cell cycle pattern (Figure 3E). 2.2. Proliferative Capacity and Morphology of Adipose-Derived MSCs (Ad-MSCs) Cultured on a Fibroblast-Derived Extracellular Matrix (Fd-ECM) Ad-MSC proliferative capacity was analyzed by determining the growth kinetics of MSCs by direct cell counting (Figure 4A). Cell numbers at each time point indicated significant differences in proliferation between MSCs grown on control plastic dishes and those cultured on fd-ECM (Figure 4B). Cell growth rate was also determined by evaluating the population doubling time during successive subcultures. Indeed, the average population doubling time for ad-MSCs on the fd-ECM increased significantly (Figure 4C). There was no significant change in cellular adhesion to the fd-ECM compared to the control dishes (Figure 4C). The morphological differences between MSCs plated on plastic (control) dishes and those plated on fd-ECM (matrix) were observed by phase-contrast microscopy. Images were captured at the end of the 48-h time point at 100× magnification and no major changes in morphology were observed between MSCs grown on control dishes and those cultured on fd-ECM (Figure 4D). The expression of proliferative markers, proliferating cell nuclear antigen (PCNA), and Ki67 were analyzed by immunoblot analysis and ad-MSCs cultured on the fd-ECM show significant decrease in Ki67, PCNA, and CD44 protein levels after 48 h of incubation (Figure 4E). After 24 h of culture on the fd-ECM significant increases were observed in integrins α2 and β1 while integrin α3 was downregulated (Figure 4F). Significant decreases in integrins α2 and β1 were observed in ad-MSCs cultured on fd-ECM compared to controls after 48 h of culture on the fd-ECM (Figure 4F). In addition, a semi-quantitative assessment of whether the fd-ECM affects ad-MSCs’ cell cycle progression was made. Cell cycle analysis was done by flow cytometry on ad-MSCs plated on plastic (control) and fd-ECM after propidium iodide labeling. Significant differences in both the G1 and S phases were observed between control cells and those plated on fd-ECM. After 24 h of incubation there appear to be a G1 cell-cycle arrest when cells were cultured on fd-ECM (Figure 5A,B,E). Approximately 10% of the ad-MSCs were detected in the G2 phase of the cell cycle. After 48 h of incubation, there was also a significant increase in ad-MSCs in the G1 phase while ad-MSCs in the S phase decreased significantly (Figure 5C–E). Immunoblot analysis showed a downregulation of cyclin D1, cyclin B1, and cyclin A, and an increase in p27 (Figure 5F). This downregulation, coupled with the decrease in Ki67 expression, is not associated with actively cycling ad-MSCs but with those undergoing differentiation. These results suggest that the fd-ECM does not promote ad-MSC proliferation but induces differentiation of ad-MSCs. 2.3. Fd-ECM Directs Ad-MSCs Differentiation towards the Chondrogenic Lineage In order to explore the influence of the fd-ECM on ad-MSCs differentiation, we first analyzed the influence of the matrix on multipotency-associated endogenous gene expression in ad-MSCs over time. Immunoblot analysis and RT-qPCR analysis showed that ad-MSCs cultured on fd-ECM, from passage 6 to passage 16, displayed decreased Oct4, Sox2, and Nanog protein levels (Figure 6A–D). Further analysis to evaluate adipogenic, osteogenic, and chondrogenic differentiation of MSCs cultured on plastic and fd-ECM over the first eight days of culture was done. Differentiation was evaluated using RT-qPCR and immunoblot analysis on differentiation markers. On day 2 after culture, ad-MSCs seeded on fd-ECM showed significantly higher expression levels of Sox9, Notch1, β-catenin, Runx2, p-TGFβRII, and Jagged1 compared to controls (Figure 7A). We further assayed the expression of the same genes after four and eight days of incubation. Our results show that Sox9, Notch1, p-TGFβRII, and Jagged1 are upregulated in ad-MSCs cultured on fd-ECM after four and eight days of incubation (Figure 7B–F; Figure S2A–D). β-Catenin was significantly upregulated within the first two days of culture on the fd-ECM and thereafter the expression levels were similar to controls (Figure 8A,B; Figure S2A,C,D). Chondrogenic differentiation was also estimated by evaluating the expression of type I collagen and type II collagen. Our results show a gradual decrease in COL1A1 and integrin α2 protein levels and a gradual increase in COL2 synthesized by the ad-MSCs cultured on the fd-ECM (Figure S2C). Increase in the expression of Sox9, Notch1, Jagged1, COL2, and p-TGFβRII protein levels is associated with chondrogenic differentiation. This indicates that ad-MSCs seeded on fd-ECM were differentiating towards chondrogenic cell types compared to those on plastic. Our immunofluorescence data substantiated the above results, showing that β-catenin only increased on day 2 and decreased thereafter, whereas Sox9, Notch1, Jagged1, and p-TGFβRII increased from day 2 onwards in the presence of fd-ECM (Figure 8A,B; Figure S2A–D). Loss of function studies using Notch1 and β-catenin siRNA and a dominant negative Notch1 construct showed that both β-catenin and Notch1 are important in fd-ECM-mediated ad-MSC differentiation towards the chondrogenic lineage (Figure 8C,D; Figure S3A–D). However, it appears that β-catenin is required during the early stages of differentiation only and that during the later stages of differentiation only Sox9, Notch1, p-TGFβRII, and Jagged1 are important. 2.4. Anti-Senescence Effect of Fd-ECM on Ad-MSCs Cellular senescence is known to occur during the long-term culture of cells. Thus for ad-MSCs to be used in cell therapy there is a need to manage epigenetic modifications during clonal expansion. We evaluated the possible effect of fd-ECM on ad-MSC senescence. Our results suggest that there is no senescence occurring during the culture of ad-MSCs on fd-ECM. Several genes associated with senescence and anti-senescence were evaluated by RT-qPCR. Gene expression of several anti-senescence markers was significantly upregulated in ad-MSCs cultured on fd-ECM (Figure 9A,C). The expression of several genes such as P16, P21, and P53 was downregulated in the presence of fd-ECM (Figure 9B,C; Figure S4A–D). We then evaluated the cell-protective effect of fd-ECM on ad-MSCs. Ad-MSCs plated on the fd-ECM and control cells were exposed to H2O2 for an hour and a half and then the viability of the ad-MSCs was determined. Ad-MSCs cultured on fd-ECM were more viable compared to ad-MSCs exposed to H2O2. Immunoblot analysis substantiated these results as the increase in cleaved caspases 3 and 9 observed in the presence of H2O2 was slightly reduced in the presence of fd-ECM (Figure 9D; Figure S5A,B). Thus fd-ECM appears to reduce the effect of oxidizing agents on ad-MSCs. 2.5. Effect of Fd-ECM on Transformation Markers Since ad-MSCs are to be used for various therapeutic procedures, their biosafety needs to be evaluated. Therefore we analyzed the expression of several genes such as RAD51, ERCC3, XRCC4, and c-MYC in MSCs cultured on fd-ECM. For fd-ECM to be considered for further studies on tissue engineering applications, the transformation capacity of MSCs cultured on fd-ECM had to be evaluated. The mRNA levels of the transformation markers, ERCC3, RAD51, c-MYC, and XRCC4 were analyzed using RT-qPCR and the fd-ECM had no significant effect on the mRNA levels of the transformation markers (Figure 10A–D). 3. Discussion The differentiation potential and immunomodulatory properties of MSCs has attracted attention regarding their application in cellular therapies for many pathological conditions [22,23,24,25,26]. MSCs are now obtained from different tissues of the body, making it possible to use MSCs from a given patient in an autologous manner [27,28,29,30,31]. Besides differentiating into endodermal and mesodermal (osteocytes and chondrocytes) lineage cells, MSCs also have remarkable translineage differentiation capabilities into neuronal, retinal, and pancreatic β cells, thereby making them very attractive for regenerative medicine [25,31]. The stem cell niche has been documented to play a significant part in determining the fate of many cells including stem cells and the ECM is also a constituent of this niche [4,10]. In vivo, MSCs are found in environments in which other cells such as fibroblasts predominate [22,23,24,25,26]. In particular, ad-MSCs are found in close proximity to several cell types collectively termed the stromal vascular fraction; this environment is rich in stromal cells such as preadipocytes, fibroblasts, and macrophages and therefore the ECM will mostly be of stromal origin [7,21,32,33,34,35]. In this study, therefore, we determined the influence of fd-derived ECM on ad-MSCs. Cell-derived ECMs from cells such as fibroblasts recapitulate the microenvironments in which MSCs reside and function. Our studies and those of others have shown the importance of different ECMs in determining the fate of many cells including stem cells during differentiation [9,10,36,37]. However, the influence of the 3-D fd-ECM on ad-MSCs has not been studied. In order for ad-MSCs to be used in stem cell therapy, understanding the molecular processes involved in the differentiation of such cells is vital. Ad-MSCs can be obtained from the adipose tissue in abundant amounts and have proven multi-lineage differentiation abilities [23,24,26]. A large amount of adipose tissue is discarded as medical waste and therefore adipose-derived MSCs can be obtained in large quantities with minimal invasive procedures and can be cultured in vitro [32,38]. Of significance, our microscopic and flow cytometric analysis of ad-MSCs shows that ad-MSCs are not affected by subconfluent passaging. Subconfluent passaging resulted in ad-MSCs maintaining their MSC properties for several passages. Our data also show that when MSCs are subconfluent passaged they do not compromise their morphological features. They were also shown to maintain their multipotency and MSC properties such as expression of cell surface markers and lineage differentiation. Many studies have shown that MSCs express Oct4, Nanog, and CD44 [9,11,39,40]. Our results are in agreement with these studies as we observed the expression of stemness markers Oct4 and Nanog. The expression of proteins such as Oct4 is known to be affected by passaging over time [41,42]. The expression of these two stemness markers was maintained at least up to passage 22 in this study and the ad-MSCs had the usual spindle shape typical of fibroblasts. Beyond passage 22, we did not assess the characteristic profile of these ad-MSCs. This study showed that the fd-ECM reduces ad-MSC proliferation without compromising their morphological features. Cyclin D1 is known to be involved in the G1 and S phases of cell cycle. Our study shows that ad-MSCs cultured on the fd-ECM exhibited decreased expression of cyclin D1 and this might account for the reduced proliferation observed. The reduction in ad-MSC proliferation is coupled with a reduction in stemness and self-renewal gene expression. We further show that the reduction in stemness and self-renewal marker gene expression occurs at the same time as the increase in chondrogenic markers, such as Sox9, Notch1, type II collagen, p-TGFβRII, and β-catenin. There is also a slight increase in Runx2 protein levels on day 2 of incubation on the fd-ECM. Our results also show a change in the type of collagen synthesized by the ad-MSCs from type I collagen to type II collagen. This occurs at the same time as a change in integrin synthesis from integrin α2 and β1 to integrin α3. The differentiation of the ad-MSCs is in line with an increase in ad-MSCs in the G1 phase and reduced amounts of ad-MSCs in the S phase. This data illustrates that the fd-ECM induces chondrogenic differentiation of ad-MSCs compared to those on control plastic dishes. It has been reported that culture of MSCs affects their osteogenic and chondrogenic differentiation [36,39,40,43]. MSCs, like most cells, have been shown to be responsive to their environments, adapting their function and phenotype to the surrounding circumstances [11,32,39]. This study shows that the fd-ECM induces chondrogenic differentiation of ad-MSCs in vitro. Several lineage markers, such as osteopontin, Gata3, Runx2, β-catenin, Sox9, and Notch1, were analyzed using RT-qPCR and immunoblotting. We observed that Sox9, Notch1, Jagged1, β-catenin, and Runx2 were upregulated 48 h after culture of ad-MSCs on the ECM. Longer incubation of ad-MSCs on the ECM caused an upregulation of Sox9, p-TGFβRII, Notch1, Notch1 ligand, Jagged1, and Hes1, a Notch signaling pathway downstream gene. RT-qPCR substantiated these results and this was accompanied by a concomitant decrease in stemness markers Oct4, Sox2, and Nanog up to day 16. Our flow cytometric data show that during the differentiation of the ad-MSCs cell cycle-associated genes were downregulated and the population doubling time was increased. This is typical of cells undergoing differentiation. Sox9, p-TGFβRII, and Notch1 are known to be involved in the regulation of chondrogenesis and in preserving the chondrocyte phenotype [44,45,46,47]. Sox9 is well established as a master regulator of chondrogenesis and therefore our results are in agreement with the specific role for Sox9 in regulating chondrogenesis [46,47]. Sox9 is known to regulate chondrogenesis via the activation of Sox5 and Sox6, while Notch signaling has been identified as a regulator of Sox9 in chondrocytes [44,46]. It has been established that Notch signaling is necessary for normal onset of chondrocyte maturation [44,45,46,47]. Our data are in agreement with Notch1 being upstream of Sox9 and therefore regulating it. Notch signaling is known to influence proliferation [48,49] and we observed a decrease in proliferation of ad-MSCs in this study in the presence of fd-ECM. Long-term culture and passaging of cells have been shown to cause senescence of cells including MSCs [41,42]. These studies show that long-term culture of MSCs results in changes such as telomere shortening [41,42,43]. If this occurs during therapy it can result in reduced therapeutic efficacy [50,51,52]. Therefore we evaluated the influence of the ECM on the passaging of ad-MSCs. Based on senescence-associated gene expression, fd-ECM prevented senescence. In fact, we observed that ad-MSCs cultured on fd-ECM exhibited an increased expression of genes such as hTERT and bFGF and a decrease in the expression of genes such as P16 and P21. hTERT is known to play a significant part in the shortening of telomere during cellular senescence. These results demonstrate the fd-ECM does not cause senescence during ad-MSC culture. Reactive oxygen species can cause DNA damage to cells and this can lead to induction of senescence genes such as P53 and P21. Our results also show that the fd-ECM reduces the H2O2-mediated increase in cleaved caspase 3 and 9. In vitro culture of MSCs, under certain conditions such as stress and hypoxia, has been suggested to cause spontaneous cellular transformation [53], although this is not a commonly reported finding. Our analysis of specific transformation markers after culture of the MSCs for up to 24 days on fd-ECM shows that most genes involved in tumor activation, suppression and DNA repair remain unchanged in the control versus fd-ECM. In fact, the expression of genes such as P21 and P53 was reduced when ad-MSCs were cultured on fd-ECM. It is possible that fd-ECM can be used as a patch containing ad-MSCs and, based on our results, the fd-ECM will induce chondrogenic differentiation of the ad-MSCs. However, this might not be the case in vivo and the use of the patch might result in an immune reaction. Further experiments are therefore necessary to determine the usability of the fd-ECM as a patch and also to generate chondrogenically differentiated ad-MSCs for clinical use. 4. Materials and Methods 4.1. Reagents Most reagents were sourced from GIBCO BRL Life Technologies (Gaithersburg, MD, USA), BioRad (Foster City, CA, USA) and Merck Biosciences (Darmstadt, Germany). Professor Igor Prudovsky (Maine Medical Center Research Institute, Scarborough, ME, USA) kindly provided the dominant negative mutant Notch1 expression plasmid. 4.2. Preparation of Fd-ECM and Cell Culture WI-38 fibroblasts were used in the preparation of the fd-ECM as described before [5,6,20]. The WI38 cell line was sourced from American Type Culture Collection (ATCC, Manassas, VA, USA). Briefly, WI38 cells were grown in Dulbecco’s Modified Eagle’s Media (DMEM) containing 10% heat inactivated fetal bovine serum (FBS) (GIBCO, New York, NY, USA), 2 mM l-glutamine (GIBCO), 100 U/mL penicillin (Biochrom, Berlin, Germany), and 100 µg/mL streptomycin (Biochrom). Every alternate day there was the addition of ascorbic acid (50 μg/mL) (Sigma Aldrich, St. Louis, MO, USA). Confluent cells were grown for a further eight days and lysed through the addition of 20 mM ammonium hydroxide (Sigma Aldrich, Steinheim, Germany) for 1 min. Sterile phosphate-buffered saline (PBS) was used to wash the fd-ECM three times. The ECM was then air dried. The fd-ECM was used immediately or stored at 4 °C to maintain the protein integrity [5,20,54,55]. The fd-ECM was washed three times with sterile PBS before use. 4.3. Adipose Tissue Procurement and Processing Adipose tissue was sourced from patients undergoing liposuction procedures. Informed consent was obtained from all volunteers prior to sample collection according to institutional guidelines. All procedures were done according to the Declaration of Helsinki guidelines. Approval was obtained before commencement of the study from the Research Ethics Committee at the Faculty of Healthy Sciences, University of Pretoria, South Africa (Pretoria, South Africa; Register I.D.: FWA 00002567; IRB 00002235 IORG0001762; Protocol Number: 218/2010). Mesenchymal stromal/stem cells isolation was performed as previously described [11,13,14]. Prior to tissue digestion with 0.001 mg/mL Type I collagenase (GIBCO, Grand Island, NY, USA), PBS was used to wash the adipose tissue and it was centrifuged twice for 3 min at 1152× g in sterile PBS supplemented with penicillin and streptomycin. The pellets were collected as ad-MSCs and cultured in α-minimal essential medium (α-MEM) containing 10% FBS (v/v) (GIBCO, Life Technologies, New York, NY, USA), 50 U/mL penicillin, and 50 μg/mL streptomycin and cultured at 37 °C. Culture dishes (100 mm culture dish) were used and 5 × 105 ad-MSCs were cultured. Cells were cultured overnight and then washed with PBS supplemented with penicillin and streptomycin to remove non-adherent cells. Cells at 80% confluency were harvested using Trypsin-EDTA and designated as passage 0. Media were changed every 2–3 days. For the experiments described in this manuscript, ad-MSCs preparations were used at passages 6–10. 4.4. Immunophenotyping and Differentiation of Ad-MSCs In order to analyze surface marker expression profiles of mesenchymal stromal/stem cells, flow cytometric analysis was performed using mouse anti-human fluorochrome-conjugated monoclonal antibodies. Several monoclonal antibodies were sourced from Beckman Coulter (Miami, FL, USA), BioLegend (San Diego, CA, USA), and eBioscience (San Diego, CA, USA) and used in the characterization of the ad-MSCs: CD34-PC7/-ECD/-PE/-FITC, CD45-PC5/-PC7/-ECD, CD73-APC/-FITC, CD90-FITC/-PC5, and CD105-PE. For each sample, a 100-µL cell aliquot was incubated at 37 °C for 15 min in the dark after the addition of a panel of monoclonal antibodies. Following incubation, cells were washed three times with PBS supplemented with 2% FBS, re-suspended in PBS, and analyzed for antigen expression. Results were obtained using a FC500 MCL (5 colors, 1 laser configuration) and Gallios (10 colors, 3 lasers configuration) flow cytometer (from Beckman Coulter). A minimum of 5 × 103 intact cells were analyzed during data acquisition. The Kaluza Flow Cytometry analysis software 1.2 was used to analyze the data (Beckman Coulter). For adipogenic, osteogenic, and chondrogenic differentiation, ad-MSCs were incubated with the respective differentiation media for up to 21 days. Change of media was done twice every week. Once the cells were 70%–80% confluent, they were induced to differentiate. Adipogenic-inducing medium consisted of Dulbecco’s Modified Eagle’s Medium (DMEM 1× + GlutaMAX™, GIBCO by Life Technologies™, Grand Island, NY, USA) containing 10% FBS, 1% 1 μM dexamethasone (Sigma-Aldrich Chemie, Steinheim, Germany), 0.5 mM 3-isobutyl-methylxanthine (Sigma-Aldrich Chemie), 200 μM indomethacin (Sigma-Aldrich Chemie), and 10 μg/mL insulin (human recombinant Zinc, GIBCO by Life Technologies™). After 21 days of differentiation, 4% formaldehyde (Sigma-Aldrich Chemie) solution (Sigma-Aldrich Chemie) was used to fix the cells (60 min) and they were stored at 4 °C in PBS. Cells were stained with 0.3% Oil Red O (ORO, Sigma-Aldrich Chemie) solution and counter stained with 1 mL 0.01% Toluidine Blue O (TBO containing 0.01% Na2CO3 (Sigma-Aldrich Chemie)) for 5 min to detect accumulation of lipid droplets. Images were acquired using a fluorescence microscope (Zeiss Axiovert 200, Carl Zeiss Werke, Göttingen, Germany). Osteogenic-inducing medium consisted of DMEM supplemented with 10% FBS, 0.1 μM dexamethasone, 50 μM ascorbate-2-phosphate (Sigma-Aldrich Chemie), and 10 mM β-glycerophosphate (Sigma-Aldrich Chemie). After 21 days of differentiation, the cells were fixed using 4% formaldehyde solution (60 min) and stored at 4 °C in PBS. Mineralization was detected by staining the cells with 2% Alizarin Red S (ARS, Sigma-Aldrich Chemie). Images were acquired using a fluorescence microscope. The “pellet culture system” was used for chondrogenic differentiation with minor modifications. A chondrogenic-inducing medium consisting of DMEM containing 0.1 μM dexamethasone, 50 µg/mL ascorbate-2-phosphate, 10 ng/mL transformed growth factor β-3 (TGF-β3, GIBCO by Invitrogen™), 40 µg/mL proline (Merck, Darmstadt, Germany), 100 µg/mL pyruvate (Merck), and 1% insulin, human transferrin and selenous acid (ITS™) premix universal culture supplement (BD Biosciences, Bedford, OH, USA) was used for 21 days. After 21 days, the pellets were fixed in 4% formaldehyde and stored in PBS until further processing. Pellets were removed from the PBS and serially dehydrated (for 15 min per ethanol change) in 30%, 50%, 70%, and 90% ethanol, followed by dehydration in absolute ethanol. The sample was then infiltrated with a 50% London Resin (LR) white medium-grade acrylic resin (SPI supplies, West Chester, PA, USA) in absolute ethanol solution for one hour, followed by infiltration with 100% LR White Resin for a minimum of four hours. Sections were collected onto droplets of water on glass slides and dried on a slide warmer, stained with 1% TBO (1% Na2CO3), and images captured with a fluorescence microscope using a 10× magnification objective lens to confirm chondrogenic differentiation. Positive staining of the proteoglycans (purple) was an indication of chondrogenic differentiation. 4.5. RNA Preparation and RT-qPCR Total ribonucleic acid (RNA) was extracted from ad-MSCs as described before [5,6,56]. Complementary deoxyribonucleic acid (cDNA), generated from 5 μg of total RNA, was used in the real-time quantitative polymerase chain reactions (RT-qPCR). RT-qPCR was performed and monitored using the Light Cycler 480 II (Roche, Mannheim, Germany). Samples were analyzed using primers listed in Table S1 (all: Whitehead Scientific, Cape Town, South Africa). Thermocycling for all targets was carried out the following conditions: initial denaturation at 94 °C for 5 min followed by 35 cycles of 94 °C for 20 s, 55 °C for 20 s, and 72 °C for 20 s. Genes analyzed along with their respective primer sequences are given in Table S1. Expression levels of osteogenic markers (CBFA1, OC), chondrogenic markers (COL2A1, Sox9), and adipogenic markers (peroxisomal proliferator-activated receptor γ 2 (PPARg2) and LPL) were analyzed by RT-qPCR. RT-qPCR experiments were done in triplicate and significant differences are shown by * p < 0.05. 4.6. Cellular Adhesion Assays Ad-MSCs were seeded onto plastic and fd-ECM-coated six-well cell culture plates and allowed to attach for 2 h at 37 °C. Cells that were not attached were then washed off with PBS, leaving attached cells on the plastic dish surface or on the fd-ECM. Adherent cells were incubated with 0.5 mg/mL MTT solution 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (Promega, Madison, WI, USA) for 2 h at 37 °C. PBS was used to wash the cells and 0.30 mL acidic isopropanol was added. Plates were incubated to dissolve the converted dye. Absorbance was then measured at 570 nm on an enzyme-linked immunosorbent assay reader. The relative attachment potential of ad-MSCs on fd-ECM was compared to controls. 4.7. Ad-MSCs Doubling Time (Gt) Ad-MSCs doubling time (PD) is the time it takes the population of ad-MSCs to double and was calculated based on the following formula: PD = t × ln2/ln(FCC/SCC) where t equals time in h, ln represents the natural logarithm, FCC represents the final ad-MSC cell number, and SCC represents the starting ad-MSC cell number. 4.8. Immunoblot Analysis Immunoblot analysis was done as described before [10]. Ad-MSCs were washed twice with PBS, lysed in radioimmunoprecipitation assay (RIPA) buffer, and a protease inhibitors mixture was added. The protein concentration of the resulting lysates was determined using the bicinchoninic (BCA) assay. Proteins (50 μg) were electrophoresed on 10% SDS-PAGE gels in the presence of 50 mM β-mercaptoethanol. Transfer of proteins was done using a nitrocellulose membrane. Fat-free milk (5%, w/v) in Tris Buffered Saline (TBS) containing Tween-20 was used to block the membranes. Overnight incubation of the membranes was done at 4 °C with different primary antibodies: anti-Sox9, anti-Oct4, anti-CD44, anti-PCNA, anti-Nanog, anti-Gata 3, anti-Notch1, anti-Runx 2 (all from Cell Signaling Technology, Beverly, MA, USA), anti-cleaved caspase 3, anti-cleaved caspase 9, anti-caspase 3, anti-caspase 9, anti-β-catenin, anti-VEGF, anti-bFGF, anti-p21, anti-p27, anti-p53, anti-vimentin, ant-COL1A1, anti-COL II, anti-Ki67, anti-cyclin D1, anti-cyclin B1, anti-cyclin A, anti-p-β-catenin, anti-Osteopontin, anti-p-TGFβRII, anti-Jagged1, and anti-GAPDH (Santa Cruz, Biotechnology, Santa Cruz, CA, USA). The membranes were washed twice with TBS-T and incubation was done with secondary antibodies conjugated to horseradish peroxidase-conjugated (HRP) (BioRad). Detection was then done using Lumiglo substrate (KPL, Gaithersburg, MD, USA). All experiments were repeated three times. 4.9. Cell Cycle Analysis Approximately 5 × 104 ad-MSCs were cultured on control dishes and on fd-ECM for different incubation times up to a maximum of passage 16. For longer incubation times, passaging was done every four days. The cells were detached from the control dishes and the fd-ECM, and processed for flow cytometry analysis. Control and fd-ECM cultured ad-MSCs were washed with PBS and fixed in 70% ethanol for 1 h at 4 °C. Ad-MSCs were then stained with propidium iodide (50 µg/mL propidium iodide) and RNase A (10 µg/mL) was added for 3 h at 4 °C. Analysis was then done using a FACScan cell sorter (Becton Dickinson, Franklin Lakes, NJ, USA). Approximately 1 × 104 cells were analyzed. Cellquest software (Version 5.1, Becton Dickinson, Franklin Lakes, NJ, USA) was used to determine the cell cycle profiles. 4.10. Transient Transfection Assay The Transfectin reagent (BioRad, Munich, Germany) was used in the transient transfection assays. Ad-MSCs were cultured on plastic or on the fd-ECM. The media was then changed and the Notch1 siRNA, 2× β-catenin siRNA, dominant negative mutant Notch1 construct (dnNotch1), and control vectors were added to the cells. Incubation was continued for up to two days, after which cell lysates were obtained. RNA was extracted and used in RT-qPCR analysis. Immunoblot analysis was then performed to measure various proteins levels. 4.11. Statistical Analysis GraphPad Prism (Version 5, GraphPad Software Inc, La Jolla, CA, USA) was used to perform the statistical analysis. All data are presented as means ± standard deviation (S.D.). Statistical significance was evaluated using the paired Student’s t test. Statistical significance is shown by * p < 0.05. 5. Conclusions This study extends our understanding of how ad-MSCs behave and function in the context of a cellular niche provided by fd-ECM. How ad-MSCs differentiate when placed in different environments is vital prior to their use in tissue engineering and regenerative medicine. To the best of our knowledge, this study is the first study to evaluate the influence of an fd-ECM on ad-MSCs’ growth kinetics and differentiation. Based on our results, the fd-ECM directs ad-MSC differentiation along the chondrogenic lineage. We should mention that further analysis on the precise signaling pathways implicated in this process must still be elucidated. Further in vitro and in vivo experiments must be done to determine the long-term effect of fd-ECM use before it can be used in stem cell therapy procedures. With cartilage defects being one of the main problems experienced by many people, chondrogenically differentiated ad-MSCs represent a viable therapeutic option for the treatment of such defects. Acknowledgments This research and the publication thereof are the result of funding provided by the International Centre for Genetic Engineering and Biotechnology (ICGEB) (Grant Number: 2015/0001), the National Research Foundation (NRF) of South Africa (Grant Number: 91457: RCA13101656402), the Medical Research Council of South Africa in terms of the MRC’s Flagships Awards Project SAMRC-RFA-UFSP-01-2013/STEM CELLS, the SAMRC Extramural Unit for Stem Cell Research and Therapy and the Institute for Cellular and Molecular Medicine, University of Pretoria and the University of Cape Town. We would like to thank Professor Igor Prudovsky (Maine Medical Center Research Institute, Scarborough, ME, USA) for kindly providing the dominant negative mutant Notch1 expression plasmid. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1259/s1. Click here for additional data file. Author Contributions Kevin Dzobo, Taegyn Turnley, Andrew Wishart, Arielle Rowe, Karlien Kallmeyer, Nicholas E. Thomford, and Fiona A. van Vollenstee performed most of the experiments and analyzed the data. Denis Chopera was involved in several experiments including RT-qPCR and siRNA experiments. Kevin Dzobo, Collet Dandara, Michael S. Pepper, and M. Iqbal Parker developed the experimental design. All authors proofread and corrected the manuscript. Kevin Dzobo wrote the main body of the manuscript. Conflicts of Interest The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results. Figure 1 The phenotypic characterization of adipose-derived mesenchymal stromal/stem cells (ad-MSCs). (A–E) Flow cytometric analysis of ad-MSCs at passage 6–10 was performed as described in the “Materials and Methods” Section 4.9; (F) Average of cells (%) staining positive for MSC surface epitopes as described above. The classical MSC phenotype is when the cells have the phenotype CD73+, CD90+, CD105+, CD45− and CD34− in ≥95% of the cell population. Different ad-MSC preparations were used during the characterization; for the sake of brevity we show representative results for three donors. Mean% ± S.D. (standard deviation). Figure 2 Lineage-specific differentiation capacity of ad-MSCs. (A,B) RT-qPCR was done to evaluate adipogenesis markers, Peroxisome proliferator-activated receptor γ 2 (PPARG2), and Lipoprotein lipase (LPL). Quantification of lipid droplets is shown in (B) after staining undifferentiated and differentiated samples with Oil Red O. Scale bar: 20 µm; (C,D) RT-qPCR was done to evaluate osteogenesis markers, Core-binding factor subunit α1 (CBFA1) and Osteocalcin (OC). Undifferentiated and differentiated donor ad-MSCs were stained with Alizarin red S. In osteogenic cultures mineralization was visible as red-stained calcium deposition in (D). Scale bar: 20 µm; and (E,F) RT-qPCR was done to evaluate chondrogenesis markers such as type II collagen (COL2) and Sox9. Undifferentiated and differentiated donor ad-MSCs were stained with Toluidine blue O. In chondrogenic cultures staining of the proteoglycans (purple) was visible. Quantification of glycosaminoglycans production is shown. Scale bar: Undifferentiated: 500 µm; Differentiated: 200 µm. * p < 0.05. Figure 3 Ad-MSCs express pluripotency markers and are viable over several passages. (A) Vimentin and CD44 expression in ad-MSCs from three donors was determined by immunoblot analysis. GAPDH was used as a loading control; (B,C) Ad-MSCs express Octamer-binding transcription factor 4 (Oct4) over several passages. The expression of pluripotency markers Oct4 and Nanog expression in ad-MSCs was determined over several passages was determined by immunoblot analysis and RT-qPCR; (D) Cleaved caspase 3 and 9 in ad-MSCs was determined by immunoblot analysis over several passages; and (E) Flow cytometric analysis was done to determine the effect of prolonged culture on ad-MSCs’ cell cycling. Figure 4 Fibroblast-derived extracellular matrix (Fd-ECM) reduces ad-MSCs proliferation. (A) Schematic representation of the experimental setup. Ad-MSCs were cultured on control plastic dishes (−) and on dishes containing fd-ECM (+); (B) Ad-MSCs were cultured on control dishes (−) and on the fd-ECM (+) for the indicated time periods. Control ad-MSCs (−) and those plated on fd-ECM (+) were counted at the indicated times using the Countess Cell Counter; (C) Average ad-MSCs population doubling time was calculated using the method described in “Materials and Methods Section 4.7”. Relative adhesion of ad-MSCs cultured on plastic dishes and on fd-ECM was evaluated and expressed as a percent of control cells. Each experiment was performed in triplicate; (D) Ad-MSCs were cultured on control dishes and on the fd-ECM for 48 h and cell images were taken using an Olympus CKX41 microscope. Scale bar: 25 µm; (E) The expression of Ki67, proliferating cell nuclear antigen (PCNA), and CD44 decreased significantly in ad-MSCs cultured on fd-ECM (+) as determined by immunoblot analysis and (F) Immunoblot analysis of integrin α2, α3, β1, and β5 in ad-MSCs lysates after culture on plastic (−) and on fd-ECM (+) for 24 and 48 h. * p < 0.05. Figure 5 Fd-ECM downregulates cell cycling in ad-MSCs in vitro. Ad-MSCs were cultured on plastic dishes (−) or on fd-ECM (+) for the indicated time periods (A) Flow cytometric analysis of control ad-MSCs showed no apparent apoptosis after 24 h of incubation; (B) Ad-MSCs were cultured on fd-ECM for 24 h and cell cycle analysis was done by flow cytometry after labeling ad-MSCs with propidium iodide; (C) Ad-MSCs were cultured on tissue plastic dishes for 48 h, labeled with propidium iodide for flow cytometric analysis; (D) Ad-MSCs were cultured on fd-ECM for 48 h and flow cytometric analysis was done as described above; (E) Ad-MSCs (%) in each cell cycle stage after ad-MSCs were cultured on plastic and on fd-ECM for 24 and 48 h. Four different experiments were pooled. Results show mean ± standard deviation; and (F) Fd-ECM reduces cycling in ad-MSCs. Immunoblot analysis of cyclin D1, cyclin B1, cyclin A, and p27 in ad-MSCs lysates after culture on plastic (−) and fd-ECM (+) for 24 and 48 h. Figure 6 Fd-ECM downregulates pluripotency genes expression over several passages. Ad-MSCs were cultured on plastic dishes (−) or on fd-ECM (+) for the indicated passages (A) Protein and mRNA levels of Oct4, Sox2, and Nanog were determined in ad-MSCs lysates and mRNA at passage 6; (B) Protein and mRNA levels of OCT4, SOX2, and NANOG were evaluated in ad-MSCs lysates and mRNA at passage 8; (C) Protein and mRNA levels of OCT4, SOX2, and NANOG were evaluated in ad-MSCs lysates and mRNA at passage 12; and (D) Protein and mRNA levels of OCT4, SOX2, and NANOG were determined in ad-MSCs lysates and mRNA at passage 16. * p < 0.05. Figure 7 Fd-ECM induces chondrogenic differentiation in ad-MSCs. Ad-MSCs were cultured on plastic dishes (−) or on the fd-ECM (+) for the indicated days (A–C) Immunoblot analysis of Sox9, Notch1, β-catenin, Runx2, Gata3, Osteopontin, p-TGFβRII, and Jagged1 in ad-MSCs lysates after two, four, and eight days of incubation on plastic dishes and on the matrix; (D–F) RT-qPCR was used to analyze the levels of SOX9, NOTCH1, RUNX2, GATA3, OSTEOPONTIN, TNF-α, and HES1 in ad-MSCs mRNA after two, four, and eight days of incubation on plastic dishes and on fd-ECM. * p < 0.05. Figure 8 Chondrogenic differentiation of ad-MSCs in the presence of fd-ECM requires activation of Notch1 and β-catenin signaling. Ad-MSCs were cultured on plastic dishes (−) or on the fd-ECM (+) for the indicated number of days (A,B) Ad-MSCs cultured on plastic control dishes and on fd-ECM were evaluated for β-catenin expression using immunofluorescence assay. Ad-MSCs, cultured for two days and four days, were incubated with antibodies against β-catenin and DAPI was used to stain the DNA in the nucleus. Scale bar: 100 µm; (C) Ad-MSCs were transfected with Notch1 siRNA and the dominant negative Notch1 construct using Transfectin Lipid reagent. Evaluation of Notch1, Sox9 and β-catenin protein levels was done; and (D) ad-MSCs were transfected with β-catenin siRNA as described above and Notch1, Sox9, and β-catenin protein level was determined. Figure 9 Anti-senescence effect of fd-ECM on ad-MSCs. (A) Ad-MSCs were cultured on plastic dishes (−) and on fd-ECM (+) for 48 h, harvested, and total RNA extracted. RT-qPCR was performed to evaluate human Telomerase Reverse Transcriptase (hTERT), VEGF, and bFGF mRNA levels; (B) Ad-MSCs were cultured on plastic and on an fd-ECM for 48 h, harvested, and total RNA extracted. RT-qPCR was performed to evaluate P16, P21, and P53 mRNA levels; (C) Ad-MSCs were cultured on plastic (−) and on fd-ECM (+), harvested, and immunoblot analysis was performed to evaluate VEGF, bFGF, p21, and p53 protein levels; and (D) ad-MSCs were cultured on plastic (−) and on an fd-ECM (+) and immunoblot analysis was performed to evaluate cleaved caspases 3 and 9 protein levels. * p < 0.05. Figure 10 Fd-ECM does not affect transformation markers’ gene expression. (A) Ad-MSCs were cultured on plastic and on an fd-ECM for the indicated periods, harvested, and total RNA extracted. RT-qPCR was performed to evaluate RAD51 mRNA levels; (B) ERCC3 mRNA levels; and (C) XRCC4 mRNA levels; and (D) c-MYC mRNA levels. ==== Refs References 1. Bissell M.J. Barcellos-Hoff M.H. The influence of extracellular matrix on gene expression: Is structure the message? J. Cell Sci. 1987 8 327 343 10.1242/jcs.1987.Supplement_8.18 2. Hay E.D. Collagen and other matrix glycoproteins in embryogenesis Cell Biology of Extracellular Matrix 2nd ed. Hay E.D. Plenum Press New York, NY, USA 1991 419 456 3. Green J.A. Yamada K.M. Three-dimensional microenvironments modulate fibroblast signaling responses Adv. Drug Deliv. Rev. 2007 59 1293 1298 10.1016/j.addr.2007.08.005 17825946 4. Evans N.D. Gentleman E. Chen X. Roberts C.J. Polak J.M. Stevens M.M. Extracellular matrix-mediated osteogenic differentiation of murine embryonic stem cells Biomaterials 2010 31 3244 3252 10.1016/j.biomaterials.2010.01.039 20149448 5. Dzobo K. Leaner V.D. Parker M.I. Feedback regulation of the α2(1) collagen gene via the MEK-ERK signaling pathway IUBMB Life 2012 64 87 98 10.1002/iub.568 22131293 6. Dzobo K. Leaner V.D. Parker M.I. Absence of feedback regulation in the synthesis of COL1A1 Life Sci. 2014 103 25 33 10.1016/j.lfs.2014.03.006 24637022 7. Zhang Y. He Y. Bharadwaj S. Hammam N. Carnagey K. Myers R. Atala A. van Dyke M. Tissue-specific extracellular matrix coatings for the promotion of cell proliferation and maintenance of cell phenotype Biomaterials 2009 30 4021 4028 10.1016/j.biomaterials.2009.04.005 19410290 8. Hanagata N. Takemura T. Monkawa A. Ikoma T. Tanaka J. Pre-adsorbed type I collagen structure-dependent changes in osteoblastic phenotype Biochem. Biophys. Res. Commun. 2006 344 1234 1240 10.1016/j.bbrc.2006.04.021 16647688 9. Oberwallner B. Brodarac A. Anic P. Saric T. Wassilew K. Neef K. Choi Y.H. Stamm C. Human cardiac extracellular matrix supports myocardial lineage commitment of pluripotent stem cells Eur. J. Cardio-Thorac. Surg. 2015 47 416 425 10.1093/ejcts/ezu163 24778452 10. Dzobo K. Vogelsang M. Parker M.I. Wnt/β-catenin and MEK-ERK signaling are required for fibroblast-derived extracellular matrix-mediated endoderm differentiation of embryonic stem cells Stem Cell Rev. Rep. 2015 11 761 773 10.1007/s12015-015-9598-4 26022506 11. Dzobo K. Vogelsang M. Thomford N.E. Dandara C. Kallmeyer K. Pepper M.S. Parker M.I. Wharton’s Jelly-Derived mesenchymal stromal cells and fibroblast-derived extracellular matrix synergistically activate apoptosis in a p21-dependent mechanism in WHCO1 and MDA MB 231 cancer cells in vitro Stem Cells Int. 2016 2016 4842134 10.1155/2016/4842134 26880967 12. Arnold S.J. Robertson E.J. Making a commitment: Cell lineage allocation and axis patterning in the early mouse embryo Nat. Rev. Mol. Cell Biol. 2009 10 91 103 10.1038/nrm2618 19129791 13. Van Vollenstee F.A. Jackson C. Hoffmann D. Potgieter M. Durandt C. Pepper M.S. Human adipose derived mesenchymal stromal cells transduced with GFP lentiviral vectors: Assessment of immunophenotype and differentiation capacity in vitro Cytotechnology 2016 10.1007/s10616-016-9945-6 26815002 14. Durandt C. van Vollenstee F.A. Dessels C. Kallmeyer K. de Villiers D. Murdoch C. Potgieter M. Pepper M.S. Novel flow cytometric approach for the detection of adipocyte sub-populations during adipogenesis J. Lipid Res. 2016 57 729 742 10.1194/jlr.D065664 26830859 15. Robey P.G. Kuznetsov S.A. Ren J. Klein H.G. Sabatino M. Stroncek D.F. Generation of clinical grade human bone marrow stromal cells for use in bone regeneration Bone 2015 70 87 92 10.1016/j.bone.2014.07.020 25064527 16. Wilschut K.J. Haagsmanb H.P. Roelena B.A.J. Extracellular matrix components direct porcine muscle stem cell behavior Exp. Cell Res. 2010 316 341 352 10.1016/j.yexcr.2009.10.014 19853598 17. Liu Z. Scannell D.R. Eisen M.B. Tjian R. Control of embryonic stem cell lineage commitment by core promoter factor, TAF3 Cell 2011 146 720 731 10.1016/j.cell.2011.08.005 21884934 18. Irwin E. Gupta R. Dashti D. Healy K.E. Engineered polymer-media interfaces for the long-term self-renewal of human embryonic stem cells Biomaterials 2011 32 6912 6919 10.1016/j.biomaterials.2011.05.058 21774983 19. Cui Y. Lu C. Meng D. Xiao Z. Hou X. Ding W. Kou D. Yao Y. Chen B. Zhang Z. Collagen scaffolds modified with CNTF and BFGF promote facial nerve regeneration in minipigs Biomaterials 2014 35 7819 7827 10.1016/j.biomaterials.2014.05.065 24930851 20. Serebriiskii I. Castelló-Cros R. Lamb A. Golemis E.A. Cukierman E. Fibroblast-derived 3D matrix differentially regulates the growth and drug-responsiveness of human cancer cells Matrix Biol. 2008 27 573 585 10.1016/j.matbio.2008.02.008 18411046 21. Leonardi D. Oberdoerfer D. Fernandes M.C. Meurer R.T. Pereira-Filho G.A. Cruz P. Vargas M. Chem R.C. Camassola M. Nardi N.B. Mesenchymal stem cells combined with an artificial dermal substitute improve repair in full-thickness skin wounds Burns 2012 38 1143 1150 10.1016/j.burns.2012.07.028 22998897 22. Aggarwal S. Pittenger M.F. Human mesenchymal stem cells modulate allogeneic immune cell responses Blood 2005 105 1815 1822 10.1182/blood-2004-04-1559 15494428 23. Jiang X.X. Zhang Y. Liu B. Zhang S.X. Wu Y. Yu X.D. Mao N. Human mesenchymal stem cells inhibit differentiation and function of monocyte-derived dendritic cells Blood 2005 105 4120 4126 10.1182/blood-2004-02-0586 15692068 24. Rosenbaum A.J. Grande D.A. Dines J.S. The use of mesenchymal stem cells in tissue engineering: A global assessment Organogenesis 2008 4 23 27 10.4161/org.6048 19279711 25. Hocking A.M. Gibran N.S. Mesenchymal stem cells: Paracrine signaling and differentiation during cutaneous wound repair Exp. Cell Res. 2010 316 2213 2219 10.1016/j.yexcr.2010.05.009 20471978 26. Maxson S. Lopez E.A. Yoo D. Danilkovitch-Miagkova A. LeRoux M.A. Concise review: Role of mesenchymal stem cells in wound repair Stem Cells Transl. Med. 2012 1 142 149 10.5966/sctm.2011-0018 23197761 27. Rodriguez A.M. Pisani D. Dechesne C.A. Turc-Carel C. Kurzenne J.Y. Wdziekonski B. Villageois A. Bagnis C. Breittmayer J.P. Groux H. Transplantation of a multipotent cell population from human adipose tissue induces dystrophin expression in the immunocompetent mdx mouse J. Exp. Med. 2005 201 1397 1405 10.1084/jem.20042224 15867092 28. McIntosh K. Zvonic S. Garrett S. Mitchell J.B. Floyd Z.E. Hammill L. Kloster A. di Halvorsen Y. Ting J.P. Storms R.W. The immunogenicity of human adipose-derived cells: Temporal changes in vitro Stem Cells 2006 24 1246 1253 10.1634/stemcells.2005-0235 16410391 29. Corcione A. Benvenuto F. Ferretti E. Giunti D. Cappiello V. Cazzanti F. Risso M. Gualandi F. Mancardi G.L. Pistoia V. Human mesenchymal stem cells modulate B-cell functions Blood 2006 107 367 372 10.1182/blood-2005-07-2657 16141348 30. Stagg J. Pommey S. Eliopoulos N. Galipeau J. Interferon-γ-stimulated mar-row stromal cells: A new type of nonhematopoietic antigen-presenting cell Blood 2006 107 2570 2577 10.1182/blood-2005-07-2793 16293599 31. Casiraghi F. Azzollini N. Cassis P. Imberti B. Morigi M. Cuqini D. Cavinato R.A. Todeschini M. Solini S. Sonzoqni A. Pretransplant infusion of mesenchymal stem cells prolongs the survival of a semi allogeneic heart transplant through the generation of regulatory T cells J. Immunol. 2008 181 3933 3946 10.4049/jimmunol.181.6.3933 18768848 32. Gimble J. Guilak F. Adipose-derived adult stem cells: Isolation, characterization, and differentiation potential Cytotherapy 2003 5 362 369 10.1080/14653240310003026 14578098 33. Caplan A.I. Mesenchymal stem cells: Cell-based reconstructive therapy in orthopedics Tissue Eng. 2005 11 1198 1211 10.1089/ten.2005.11.1198 16144456 34. Gimble J.M. Katz A.J. Bunnell B.A. Adipose-derived stem cells for regenerative medicine Circ. Res. 2007 100 1249 1260 10.1161/01.RES.0000265074.83288.09 17495232 35. Meyerrose T.E. de Ugarte D.A. Hofling A.A. Herrbrich P.E. Cordonnier T.D. Schultz L.D. Eagon J.C. Wirthlin L. Sands M.S. Hedrick M.A. In vivo distribution of human adipose-derived mesenchymal stem cells in novel xenotransplantation models Stem Cells 2007 25 220 227 10.1634/stemcells.2006-0243 16960135 36. Flynn L.E. The use of decellularized adipose tissue to provide an inductive microenvironment for the adipogenic differentiation of human adipose-derived stem cells Biomaterials 2010 31 4715 4724 10.1016/j.biomaterials.2010.02.046 20304481 37. Rowland T.J. Blaschke A.J. Buchholz D.E. Hikita S.T. Johnson L.V. Clegg D.O. Differentiation of human pluripotent stem cells to retinal pigmented epithelium in defined conditions using purified extracellular matrix proteins J. Tissue Eng. Regen. Med. 2013 7 642 653 10.1002/term.1458 22514096 38. Zuk P.A. Zhu M. Ashjian P. de Ugarte D.A. Huang J.I. Mizuno H. Alfonso Z.C. Fraser J.K. Benhaim P. Hedrick M.H. Human adipose tissue is a source of multipotent stem cells Mol. Biol. Cell 2002 13 4279 4295 10.1091/mbc.E02-02-0105 12475952 39. Akiyama H. Chaboissier M.C. Martin J.F. Schedl A. de Crombrugghe B. The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6 Genes Dev. 2002 16 2813 2828 10.1101/gad.1017802 12414734 40. Tsai T.L. Wang B. Squire M.W. Guo L.W. Li W.J. Endothelial cells direct human mesenchymal stem cells for osteo-and chondro-lineage differentiation through endothelin-1 and AKT signaling Stem Cell Res. Ther. 2015 6 1 14 10.1186/s13287-015-0065-6 25559585 41. Flores I. Canela A. Vera E. Tejera A. Cotsarelis G. Blasco M.A. The longest telomeres: A general signature of adult stem cell compartments Genes Dev. 2008 22 654 667 10.1101/gad.451008 18283121 42. Donate L.E. Blasco M.A. Telomeres in cancer and ageing Philos. Trans. R. Soc. B: Biol. Sci. 2011 366 76 84 10.1098/rstb.2010.0291 21115533 43. Guo P. Yuan Y. Chi F. Biomimetic alginate/polyacrylamide porous scaffold supports human mesenchymal stem cell proliferation and chondrogenesis Mater. Sci. Eng: C 2014 42 622 628 10.1016/j.msec.2014.06.013 25063162 44. Lefebvre V. Li P. de Crombrugghe B. A new long form of Sox5 (L-Sox5), Sox6 and Sox9 are co-expressed in chondrogenesis and cooperatively activate the type II collagen gene EMBO J. 1998 17 5718 5733 10.1093/emboj/17.19.5718 9755172 45. Lefebvre V. de Crombrugghe B. Toward understanding Sox9 function in chondrocyte differentiation Matrix Biol. 1998 16 529 540 10.1016/S0945-053X(98)90065-8 9569122 46. Akiyama H. Lefebvre V. Unravelling the transcriptional regulatory machinery in chondrogenesis J. Bone Miner Metab. 2011 29 390 395 10.1007/s00774-011-0273-9 21594584 47. Kohn A. Dong Y. Mirando A.J. Jesse A.M. Honjo T. Zuscik M.J. O’Keefe R.J. Hilton M.J. Cartilage-specific RBPjκ-dependent and independent Notch signals regulate cartilage and bone development Development 2012 139 1198 1212 10.1242/dev.070649 22354840 48. Prasad S.M. Czepiel M. Cetinkaya C. Smigielska K. Weli S.C. Lysdahl H. Gabrielsen A. Petersen K. Ehlers N. Fink T. Continuous hypoxic culturing maintains activation of Notch and allows long-term propagation of human embryonic stem cells without spontaneous differentiation Cell Prolif. 2009 42 63 74 10.1111/j.1365-2184.2008.00571.x 19143764 49. Nekanti U. Dastidar S. Venugopal P. Totey S. Ta M. Increased proliferation and analysis of differential gene expression in human Wharton’s jelly-derived mesenchymal stromal cells under hypoxia Int. J. Biol. Sci. 2010 6 499 512 10.7150/ijbs.6.499 20877435 50. Heeschen C. Lehmann R. Honold J. Assmus B. Aicher A. Walter D.H. Martin H. Zeiher A.M. Dimmeler S. Profoundly reduced neovascularization capacity of bone marrow mononuclear cells derived from patients with chronic ischemic heart disease Circulation 2004 109 1615 1622 10.1161/01.CIR.0000124476.32871.E3 15037527 51. Oh Y.S. Jeong S.G. Cho G.W. Anti-senescence effects of DNA methyltransferase inhibitor RG108 in human bone marrow mesenchymal stromal cells Biotechnol. Appl. Biochem. 2015 62 583 590 10.1002/bab.1393 25952632 52. Choi Y.H. Kurtz A. Stamm C. Mesenchymal stem cells for cardiac cell therapy Hum. Gene Ther. 2010 22 3 17 10.1089/hum.2010.211 21062128 53. Serakinci N. Graakjaer J. Kolvraa S. Telomere stability and telomerase in mesenchymal stem cells Biochimie 2008 90 33 40 10.1016/j.biochi.2007.09.005 18029083 54. Cukierman E. Preparation of extracellular matrices produced by cultured fibroblasts Current Protocols in Cell Biology Bonifacino J.S. Dasso M. Lippincott-Schwartz J. Harford J.B. Yamada K.M. John Wiley and Sons New York, NY, USA 2002 10.09.01 10.09.14 10.1002/0471143030.cb1009s16 55. Cukierman E. Cell migration analyses within fibroblast-derived 3-D matrices Cell Migration: Developmental Methods and Protocols Guan J. Humana Press Totowa, NJ, USA 2005 70 93 56. Chomczynski P. Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction Anal. Biochem. 1983 132 6 13 10.1016/0003-2697(87)90021-2 6312838

PMC005xxxxxx/PMC5000658.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081260ijms-17-01260ArticleBioinformatics and Microarray Analysis of miRNAs in Aged Female Mice Model Implied New Molecular Mechanisms for Impaired Fracture Healing He Bing 12†Zhang Zong-Kang 3†Liu Jin 12†He Yi-Xin 1Tang Tao 4Li Jie 3Guo Bao-Sheng 12Lu Ai-Ping 12*Zhang Bao-Ting 3*Zhang Ge 12*Stathopoulos Constantinos Academic Editor1 Institute for Advancing Translational Medicine in Bone & Joint Diseases, School of Chinese Medicine, Hong Kong Baptist University, Hong Kong, China; [email protected] (B.H.); [email protected] (JinL.); [email protected] (Y.-X.H.); [email protected] (B.-S.G.)2 Institute of Integrated Bioinformedicine & Translational Science, HKBU Shenzhen Research Institute and Continuing Education, Shenzhen 518000, China3 School of Chinese Medicine, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, China; [email protected] (Z.-K.Z.); [email protected] (JieL.)4 Department of Obstetrics and Gynaecology, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, China; [email protected]* Correspondence: [email protected] (A.-P.L.); [email protected] (B.-T.Z.); [email protected] (G.Z.); Tel.: +852-3411-2457 (A.-P.L.); +852-3943-4285 (B.-T.Z.); +852-3411-2958 (G.Z.)† These authors contributed equally to this work. 03 8 2016 8 2016 17 8 126003 6 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Impaired fracture healing in aged females is still a challenge in clinics. MicroRNAs (miRNAs) play important roles in fracture healing. This study aims to identify the miRNAs that potentially contribute to the impaired fracture healing in aged females. Transverse femoral shaft fractures were created in adult and aged female mice. At post-fracture 0-, 2- and 4-week, the fracture sites were scanned by micro computed tomography to confirm that the fracture healing was impaired in aged female mice and the fracture calluses were collected for miRNA microarray analysis. A total of 53 significantly differentially expressed miRNAs and 5438 miRNA-target gene interactions involved in bone fracture healing were identified. A novel scoring system was designed to analyze the miRNA contribution to impaired fracture healing (RCIFH). Using this method, 11 novel miRNAs were identified to impair fracture healing at 2- or 4-week post-fracture. Thereafter, function analysis of target genes was performed for miRNAs with high RCIFH values. The results showed that high RCIFH miRNAs in aged female mice might impair fracture healing not only by down-regulating angiogenesis-, chondrogenesis-, and osteogenesis-related pathways, but also by up-regulating osteoclastogenesis-related pathway, which implied the essential roles of these high RCIFH miRNAs in impaired fracture healing in aged females, and might promote the discovery of novel therapeutic strategies. impaired fracture healingbioinformaticsmiRNA ==== Body 1. Introduction Fracture healing is a complex process, which is impacted profoundly by aging and osteoporosis [1,2]. Approximately one half of women aged 60 years or older are found to have osteoporosis [3], and over 50% of postmenopausal women will suffer an osteoporotic fracture [4]. Due to the impaired capacity of fracture repair, the incident of fracture nonunion or delayed union is high in aged women [5], leading to increased morbidity and mortality as well as high cost of caring for patients [4,6,7,8]. To date, the underlying mechanism responsible for the impaired fracture healing in aged women remains underexplored. Hence, it is highly desirable to understand the differences in the molecular and cellular events during the fracture healing progress between adult and aged women. MicroRNAs (miRNAs) are small non-coding RNAs of ~22 nucleotides which function as key post-transcriptional gene expression regulators by targeting the 3′-UTR of the mRNAs [9]. The miRNAs are important regulators of many key biological processes, such as cell proliferation, differentiation, and organ development [9]. They are also associated with human disease, including cancer [10]. In the field of skeletal biology, growing evidence suggests that miRNAs are key regulators of bone modeling and remodeling including angiogenesis [11,12], chondrogenesis [13,14], osteogenesis [14], and osteoclastogenesis [15,16,17], and they also participate in the regulation of fracture healing [11,18]. The fracture healing process is typically constructed by four stages, including initial inflammatory reaction, formation of soft callus formation, formation of hard callus, and remodeling to original bone contour [19]. Angiogenesis, chondrogenesis, osteogenesis, and osteoclastogenesis are essential components of fracture healing, while the impairment of these components has been reported to induce improper fracture healing [20,21,22,23]. Therefore, studying miRNAs during fracture healing will help identifying novel therapeutic targets for improving impaired fracture healing in aged women. Currently, fold change and t-test are common methods used to identify differentially expressed miRNAs in bone fracture [24,25]. Usually, dozens to hundreds of differentially expressed miRNAs are identified in these studies. It is hard to judge which differentially expressed miRNA is more important than others for fracture healing. Therefore, an algorithm scoring the contribution of every differentially expressed miRNA to fracture healing is needed. In this study, the miRNA expression profiles in fracture healing of adult and aged female mice were examined by microarray analysis. Differentially expressed miRNAs were identified between adult and aged female mice, followed by identification of target genes using an integrated method. We designed a novel bioinformatics scoring system to analyze the contribution of differentially expressed miRNAs to impaired fracture healing (RCIFH). Furthermore, pathway enrichment analyses were performed for the differentially expressed miRNAs with high RCIFH values to analyze their potential roles in impaired fracture healing of aged female mice. 2. Results 2.1. Fracture Healing Is Impaired in Aged Female Mice The micro computed tomography (micro-CT) images demonstrated remarkably different morphologies of the fracture callus between the aged and adult groups. The aged mice showed a smaller amount of newly mineralized callus at 2-week post-fracture and delayed bridging of the fracture site at 4-week post-fracture when compared to adult mice (Figure 1A). Quantitatively, although the bone volume (BVL), low-density bone volume fraction (BVL/TV), and bone mineral content (BMC) in both groups increased from 0- to 2-week after fracture and decreased gradually thereafter, the above three micro-CT parameters in aged female mice were all significantly lower than those in adult ones at 2- and 4-week post-fracture, respectively (Figure 1B). These evidences indicated that fracture healing was impaired in aged female mice. 2.2. MicroRNA (miRNA) Expression Profiles The expression of total 1079 mouse miRNAs was analyzed on a mouse SurePrint G3 miRNA Microarray. A total of 53 significantly differentially expressed miRNAs in bone fracture healing were identified. At 2-week post-fracture, 35 miRNAs were identified that were differentially expressed in aged female mice (17 up-regulated and 18 down-regulated), while there were 33 differentially expressed miRNAs in the adult group (23 up-regulated and 10 down-regulated). At 4-week post-fracture, 10 miRNAs were identified that were differentially expressed in aged female mice (5 up-regulated and 5 down-regulated) while there were 6 differentially expressed miRNAs in the adult group (4 up-regulated and 2 down-regulated) (Supplementary Data S1). As shown in Figure 2, the miRNA expression patterns were different between adult and aged mice during fracture healing. 2.3. Target Genes and Molecular Network Construction A total of 210,365 miRNA-target interactions were predicted for differentially expressed miRNAs (|log2FC| > 0). They were included in further selection using gene expression data of angiogenesis, chondrogenesis, and osteogenesis. Since miRNAs target genes and inhibit their expression, the expression pattern of a miRNA and its target gene should be opposite. Therefore, for up-regulated miRNAs in bone fracture healing, down-regulated target genes were selected, while for down-regulated miRNAs, up-regulated target genes were selected. Finally, 43,839 miRNA-target interactions were predicted to be involved in bone fracture healing, in which 5438 miRNA-target interactions are for the 53 significant differentially expressed miRNAs (Figure 3). 2.4. miRNA Contribution to Impaired Fracture Healing (RCIFH) and Pathway Enrichment Analysis The contribution of every differentially expressed miRNA to impaired fracture healing (RCIFH) was calculated in the context of molecular network and biological progresses using the algorithms described in the Materials and Methods section. The absolute value of RCIFH reveals the impact of the differentially expressed miRNA on fracture healing. A positive RCIFH indicates that the miRNA improves bone fracture healing, while a negative RCIFH indicates the miRNA impairs bone fracture. The top 10 RCIFH miRNAs at 2- or 4-week are presented in Table 1, more detail information is listed in Supplementary Data S1. Among these miRNAs, miR-142-5p [26], miR-223 [27], miR-22 [28,29] miR-24 [17], miR-497 [30], and miR-195 [30] have been found to participate fracture healing. The other 11 miRNAs are novel findings by this study. According to the RCIFH analysis, miR-494 shows to be the most important miRNA to impair fracture bone fracture healing at both 2- and 4-week post-fracture (Table 1). To investigate how much RCIFH miRNAs impact fracture healing, target genes with the 100 highest RCIFH miRNAs were analyzed (Supplementary Data S2). At 2-week after fracture, target genes of high RCIFH miRNAs were significantly enriched (p-value < 0.05) in 13 pathways (Figure 4A), while at 4-week after fracture, target genes were significantly enriched (p-value < 0.05) in 8 pathways (Figure 4B). These pathways are mainly involved in angiogenesis-, chondrogenesis-, osteogenesis-, and osteoclastogenesis-related functions, which are important for fracture healing. Interestingly, the top RCIFH miR-494 may impair fracture healing in aged females by inhibiting angiogenesis, chondrogenesis, and osteogenesis (Figure 5). 2.5. The miR-494 Inhibits Chondrogenic Differentiation in Vitro As mentioned above, the miR-494 is the miRNA with the top RCIFH and was predicted to impair fracture healing by inhibiting angiogenesis, chondrogenesis, and osteogenesis. Previous studies have shown that miR-494 inhibits angiogenesis [31,32]. To further investigate whether miR-494 has an impact on the progress of chondrogenic differentiation, miR-494 mimics or anti-miR-494 were transfected into C3H10T1/2 cells. The chondrogenic differentiation of C3H10T1/2 cells was induced by medium containing transforming growth factor-β (TGF-β3). Thereafter, according to the quantitative real-time polymerase chain reaction (QPCR) analysis, chondrogenesis markers—including Acan, Col2a1 and Col10a1—were significantly decreased at the mRNA level in C3H10T1/2 cells transfected with miR-494 mimics (Figure 6). Consistently, anti-miR-494, using the same induction conditions as mentioned above, enhanced the chondrogenic differentiation of C3H10T1/2 cells, as evidenced by significant increases of chondrogenesis markers at the mRNA level (Figure 6). These results indicated that miR-494 inhibits chondrogenic differentiation in vitro. 3. Discussion In order to extend our understanding on the pathogenesis of impaired fracture healing in aged females, this study focused on comparing the miRNA expression profiles at fracture site between adult and aged female mice during fracture healing. The differentially expressed miRNAs were identified and their dynamic expression patterns were described. A novel bioinformatics scoring system was designed to analyze the contribution of every differentially expressed miRNA to impaired fracture healing (RCIFH). In this study, fracture healing was remarkably impaired in aged female mice, as evidenced by the micro-CT data showing a smaller amount of newly mineralized calluses at the early stage and delayed bridging of the fracture gap at the later stage. Additionally, lower micro-CT parameters were at the fracture site during bone healing in the aged female mice when compared to the adult ones. Bone healing after fracture is a complicated and sequential process including inflammation, angiogenesis, progenitor cell recruitment, chondrogenesis, osteogenesis, and osteoclastogenesis [19,20,33,34]. In the aged population, the processes of angiogenesis, chondrogenesis, and osteogenesis have been down-regulated [35,36], whereas the process of osteoclastogenesis has been up-regulated [37], which results in impaired fracture healing. However, the regulatory mechanisms underlying these changes still have not been fully understood. The miRNAs are important regulatory molecules, which have been demonstrated to regulate angiogenesis, chondrogenesis, osteogenesis, and osteoclastogenesis by inhibiting gene expression [38,39,40]. The importance of miRNA on fracture healing highly relies on its impact on target genes, which cannot be fully illustrated by its own expression change. Sometimes, the most differentially expressed miRNA might not have the strongest impact on impaired fracture healing in the aged female mice. Therefore, we designed the RCIFH method to calculate the impact of miRNAs on impaired fracture healing. RCIFH equals fold changes difference between adult and aged group times the network power of the miRNA, which has more significance than differential expression only. A miRNA with a positive RCIFH means this miRNA improves bone fracture healing, while a negative RCIFH reveals the miRNA impairs bone fracture. Usually, the RCIFH value indicates the importance of the miRNA to the impaired fracture healing. According to the RCIFH in present study, miR-494 has been the most important miRNA at both 2- and 4-week post-fracture (Table 1). MiR-494 has been found to inhibit the vascular endothelial growth factor (VEGF), an angiogenic factor, in vitro [31], and the inhibition of miR-494 could increase neovascularization [32]. An angiogenesis-related signaling pathway—nitric oxide signaling—is the most significantly suppressed pathway at 2-week after fracture, indicating the suppression of angiogenesis in aged female mice might be one of the major reasons for impaired fracture healing, and miR-494 might play some role in this process (Figure 5). Our study further revealed that the miR-494 inhibits chondrogenic differentiation in vitro. The pathway analysis indicated that miR-494 inhibits genes in the retinoid acid receptor (RAR) pathway, which plays a fundamental role in chondrogenesis [41,42]. These results indicate the potential role of miR-494 in inhibiting chondrogenesis during fracture healing in aged female mice. In order to investigate molecular mechanisms underlying high RCIFH miRNAs in impaired fracture healing, we analyzed target genes with the top 100 RCIFH miRNAs. Although strict criteria have been used to increase the confidence of the predicted miRNA-target interactions in fracture healing, false positive miRNA-target interaction may still exist. Since there are more than 15,000 miRNA-target interactions predicted, as listed in Supplementary Data S2, we performed further functional analysis at the pathway level to eliminate the influence of scattered false positive miRNA-target interactions. Among these pathways, transforming growth factor-β (TGF-β) signaling and interleukin 6 (IL-6) signaling are most relevant to osteogenesis [43,44]. The target genes of miRNAs were significantly enriched in TGF-β signaling pathway at both 2-week and 4-week post-fracture. TGF-β and bone morphometric protein-2 (BMP-2) are the core signal proteins of the TGF-β signaling pathway [43]. In the present study, TGF-β is predicted to be inhibited by miR-425 at 2-week post-fracture. BMP-2 is predicted to be inhibited by miR-142-5p at 2- and 4-week post-fracture. In addition, the downstream markers of TGF-β signaling—including SMAD family member 9 (SMAD9) and transforming growth factor β-activated kinase 1 (TAK1)—were inhibited by the top RCIFH miRNA, miR-494 (Figure 5). These results indicate that the TGF-β signaling is modulated in aged female mice by the miRNAs. TGF-β signaling pathway could promote osteoblast differentiation and bone formation [43]. Therefore, the inhibition of TGF-β signaling mediated by miRNA might impair fracture healing in aged female mice. On the other hand, the IL-6 signaling pathway was the most significant pathway at 4-week after fracture. IL-6 signaling has been demonstrated to promote osteoclast differentiation [45,46,47], and inhibit osteoblast differentiation [44,48] and bone formation [49,50]. IL-6 negatively regulates osteoblast differentiation through phosphoinositide 3-kinase (PI3K)/Akt pathway [48]. Both genes of IL-6 and its downstream target, Akt, were inhibited by miR-203. Since miR-203 was down-regulated at 4-week after fracture in aged female mice, IL-6 signaling was activated in aged female mice at 4-week after fracture (Figure 5). The miRNA-mediated enhanced bone resorption at the later stage of fracture healing might also be one of the underlying mechanisms of fracture healing impairment in aged female mice. In summary, we identified differentially expressed miRNAs during impaired fracture healing in aged female mice. Additionally, we designed a novel RCIFH method to analyze the importance of differentially expressed miRNAs to impaired fracture healing. This method not only found miRNAs that are known to participate in bone fracture healing, but also identified novel candidate miRNAs that impair fracture healing. These high RCIFH miRNAs might influence the process of angiogenesis, chondrogenesis, osteogenesis, and, especially, osteoclastogenesis, thereby contributing to the impaired fracture healing in aged female mice. These results indicate clues for a deeper understanding of molecular mechanisms underlying high RCIFH miRNAs involved in impaired fracture healing. Moreover, the novel candidate miRNAs would be potential therapeutic targets for impaired fracture healing in aged women. In the future, further experimental investigations are required to promote the therapeutic strategies for impaired fracture healing. 4. Materials and Methods 4.1. Animal Model and Micro Computed Tomography (Micro-CT) Analysis Nine 24-month-old ovariectomized C57BL/6J female mice (Aged Group) and nine 3-month-old C57BL/6J female mice (Adult Group) were recruited in the study. All animal studies were performed in accordance with the guideline from the Animal Experimentation Ethics Committee of the Chinese University of Hong Kong (Hong Kong, China), and were approved by this committee (Reference No. 09/001/GRF). The animals underwent a transverse fracture in right femur. Briefly, prior to fracture induction, intramedullary fixation was used to stabilize the right femur. A small incision lateral to the patella was used to perform the retrograde nailing. The femoral notch was exposed using blunt dissection. Through the proximal metaphyseal zone, a 27 G needle was inserted into the intramedullary canal as described previously [51]. Then cannula was shortened under the cartilaginous surface. Simple interrupted sutures were used to close the wound. A standardized blunt guillotine device was used to induce the transverse femoral fracture. In vivo micro-CT analysis using vivaCT 40 (Scanco Medical, Brüttisellen, Switzerland) was performed at 0-, 2- and 4-week post-fracture, respectively. The contoured regions of interest (ROI) were selected from two-dimensional (2D) CT images. A low-pass Gaussian filter was used to do the three-dimensional (3D) reconstructions of mineralized tissues (Sigma = 1.2, Support = 2). To distinguish the newly mineralized callus from the old cortices, different thresholds (low attenuation = 130, high attenuation = 220) were determined in 2D images using the established evaluation protocol to reconstruct the low- and high-density mineralized tissues [52]. The low-density tissues indicated newly formed callus, while the high-density tissues indicated old cortices and highly mineralized callus. The quantitative analyses were performed covering the middle 400 slices. Morphometric parameters include low-density bone volume (BVL, mm3), low-density bone volume fraction (BVL/TV, %), and bone mineral content (BMC, mgHA) were calculated as indicators of callus mineralization. 4.2. Tissue Sample and RNA Isolation After 0-, 2- and 4-week post-fracture, three mice in each group were sacrificed, respectively. After sacrifice, callus from the right femurs were collected for RNA extraction. Since there was no callus at 0-week post-fracture, the tissue harvested from animals at 0-week was the fracture haematoma present at the fracture site at 3-day post-fracture. Trizol reagent (1 mL) (Invitrogen, Carlsbad, CA, USA) was added directly to the tissues after being crushed by pestle grinder in liquid nitrogen. The supernatant was collected after centrifugation at 8000 rpm for 5 min. According to commercialized protocol, the phase separation was performed. Total RNA was incubated in 0.5 mL isopropanol at −80 °C overnight. Thereafter it was centrifuged at 12,000 rpm for 10 min at 4 °C. RNA pellet was then washed by 75% ethanol twice and was centrifuged at 7500 rpm for 5 min at 4 °C. Total RNA pellet was put in a sterile hood to be briefly air-dried. Finally, it was dissolved in 100 µL RNase free water to be stored at −80 °C. 4.3. MicroRNAs (miRNAs) Microarray Labeling and Hybridization Each total RNA sample concentration was determined using NanoDrop ND-1000 spectrophotometer (NanoDrop, Wilmington, DE, USA). All RNA samples have 260/280 and 260/230 higher than 1.8 and 1.0 respectively. Samples were labeled and hybridized on Agilent 8 × 60 K Mouse miRNA Microarray (Agilent Technologies, Santa Clara, CA, USA) Release 17.0 according to manufacturer protocol. In brief, exactly 100 ng of RNA sample was used. The 3′ end of RNA was dephosphorylated by calf intestinal phosphatase (Agilent Technologies) and then ligated with pCp-Cy3 (Agilent Technologies) using T4 RNA ligase (Agilent Technologies). Then the labeled samples where hybridized for 20 h at 20 rotations per minute. After hybridization, the arrays were washed and scanned by Agilent scanner and then the image was analyzed by Agilent Feature Extraction 10.7 (Agilent Technologies). 4.4. Differentially Expressed miRNAs The miRNA microarrays’ data were qualified and normalized using limma package on R platform [53]. Two samples were dropped for low data quality (one is adult mice at 0-week, the other is aged mice at 0-week). A total of 16 samples (2 adult mice at 0-week, 3 adult mice at 2-week, 3 adult mice at 4-week, 2 aged mice at 0-week, 3 aged mice at 2-week, 3 aged mice at 4-week) were included in further analysis. Then, miRNAs that significantly differentially expressed in calluses of adult female mice at 2-week and 4-week post-fracture are identified using limma package on R platform [53] with the criterion p-value < 0.05 & |log2FC (fold change)| ≥ 1 (2-week vs. 0-week, 4-week vs. 0-week, t-test). The differentially expressed miRNAs in calluses of aged female mice after fracture were identified in the same method. 4.5. Gene Microarray Data Gene microarray data of mice angiogenesis, chondrogenesis, and osteogenesis was downloaded from the Gene Expression Omnibus (GEO) database (Data id: GDS1631, GDS1632, GDS1633, GSE64141 and GSE7507). Differentially expressed genes in angiogenesis, chondrogenesis, and osteogenesis were identified using limma package on R platform [53] with the criterion p-value < 0.05 (t-test). 4.6. Construction of Protein-Protein Interaction (PPI) Networks Interactions among products of the differentially expressed genes were identified using protein-protein interaction data downloaded from the BioGRID database [54] (version 3.4.130). Interactions supported by evidence from at least one wet-experiment were selected to construct the PPI networks involved in bone fracture healing. A total of 36,951 PPI interactions were used in further analysis. 4.7. Prediction of miRNA-Target Interactions Experimentally validated miRNA-target interactions were collected from miRTarBase database [55] (Release 6.0). More miRNA-target interactions were predicted using DIANA [56], miRanda [57], miRDB [58], and TargetScan [59]. The miRNA-target interactions supported by evidence from at least one wet-experiment or two prediction methods were selected for further analysis. Since miRNAs target genes and inhibit their expression, the expression pattern of a miRNA and its target gene should be opposite. Therefore, for up-regulated miRNAs in bone fracture healing, down-regulated target genes were selected, while for down-regulated miRNAs, up-regulated target genes were selected. Finally, 43,839 miRNA-target gene interactions were predicted to be involved in bone fracture healing. 4.8. miRNA Contribution to Fracture Healing Impairment The power of the miRNA on the PPI networks (PRN) is calculated as follows: PRN=∑i=1nTPNi where nT is the number of all targets of the miRNA and PNi is the number of proteins directly connected to the target protein i in the PPI network of mice. The miRNA contribution to impaired fracture healing (RCIFH) is calculated as follow: RCIFH=(log2ENHiENNi−log2EAHiEANi)PRNi ENHi is the average value of miRNA i expression in adult female mice at 2-/4-week after fracture. ENNi is the average value of miRNA i expression in adult female mice at 0-week after fracture. EAHi is the average value of miRNA i expression in aged female mice at 2-/4-week after fracture. EANi is the average value of miRNA i expression in aged female mice at 0-week after fracture. PRNi is the power of miRNA i in the molecular network of mice. 4.9. Pathway Enrichment Analysis Pathway enrichment analyses were performed with Ingenuity Pathway Analysis (IPA) tools [60] using Fisher’s exact test. Significant enriched pathways for the given genes were identified with the criterion p-value < 0.05. 4.10. Cell Culture and Transfection The C3H10T1/2 cell line was obtained from the ATCC (Manassas, WV, USA). These cells were preserved in the complete Dulbecco’s Modified Eagle Medium (DMEM), which was supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin, and 10% fetal bovine serum. These cells were maintained in a humidified 5% CO2 atmosphere at 37 °C. Before transfection, C3H10T1/2 cells were placed in 6-well plates at about 105 cell/well. The miR-494 mimics, anti-miR-494, or their inactive controls were transfected into C3H10T1/2 cells according to the manufacturer’s protocol, respectively. After an incubation for 24 h, these cells were trypsinized for further chondrogenic differentiation assay. 4.11. Chondrogenic Differentiation Assay After transfection and 24 h incubation, C3H10T1/2 cells were treated with high density micromass cultures. These cells were then trypsinized by 0.25% trypsin. Then the cells were modulated at the density of 107 cells/mL. Then the suspension was placed into a 12-well plate for 10 µL. Then it was incubated at 37 °C and 5% CO2 for 2 h. After that, it was flooded by the chondrogenic differentiation medium in the volume of 1 mL. The chondrogenic differentiation medium was replaced every 2 days. The medium was composed of ascorbate, dexamethasone, sodium pyruvate, proline, Insulin-Transferrin-Sodium selenite (ITS+) Supplement, and Transforming growth factor-β3 (TGF-β3). 4.12. Quantitative RT-PCR Analysis RNeasy Mini Kit (Cat no. 74106, QIAGEN, Hilden, Germany) was used to extract total RNAs from the cultured cells using the commercialized protocol. The cells were collected in a reaction tube and were treated with 700 µL QIAzol. Then they were mixed with 140 µL chloroform. After a 15 min centrifugation at 12,000 rpm at 4 °C, the upper aqueous phase was then transferred to the RNeasy Mini spin column using a 2 mL collection tube, and then was mixed with 100% ethanol. Thereafter, it was washed with 500 µL Buffer RPE and 700 µL Buffer RWT. After that, total RNAs were reverse-transcribed to cDNA using the previously established protocol. The solution contained 1 µL of cDNA product, 5 µL of 2× SYBR® Green Mix, 0.5 µL of each primer and 3 µL nuclease-free water. The fluorescence signal was collected by ABI PRISM® 7900HT System (Applied Biosystems, Foster City, CA, USA). 5. Conclusions Impaired fracture healing in aged females is still a challenge in clinics. In this study, we identified differentially expressed miRNAs during impaired fracture healing in aged female mice. Additionally, we designed a novel RCIFH method to analyze the importance of differentially expressed miRNAs to impaired fracture healing. High RCIFH miRNAs were found to potentially influence the process of angiogenesis, chondrogenesis, osteogenesis, and, especially, osteoclastogenesis in fracture healing. The results might improve our knowledge of impaired fracture healing. Moreover, the RCIFH method would promote the study of miRNAs in impaired fracture healing. Acknowledgments This work was supported by Natural Science Foundation Council (NSFC 81272045). Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1260/s1. Click here for additional data file. Author Contributions Ai-Ping Lu, Bao-Ting Zhang and Ge Zhang conceived the project; Bing He performed the calculation analysis, Zong-Kang Zhang, Jin Liu, Yi-Xin He., Tao Tang, Jie Li and Bao-Sheng Guo carried out the experiments; Bing He, Zong-Kang Zhang and Jin Liu wrote the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Fracture healing in aged and adult female mice (A) representative three-dimensional (3D) images of the fracture calluses in adult and aged groups at each time point after fracture; (B) time course changes in low-density bone volume (BVL) (left), low-density bone volume fraction (BVL/TV) (middle), and bone mineral content (BMC) (right) of the callus after fracture in each group; n = 3. Note: scale bar = 1 mm. * p < 0.05 for “Aged” vs. “Adult”. Figure 2 Heat map of differentially expressed microRNAs (miRNAs) in fracture healing. Red represents up-regulation and green represents down-regulation. Figure 3 miRNA-target gene interactions for the significantly differentially expressed miRNAs in bone fracture healing. Figure 4 Significantly enriched pathways of top 100 highest miRNA contribution to impaired fracture healing (RCIFH) miRNA targets at (A) 2-week after fracture and (B) 4-week after fracture. Figure 5 The effect of top 100 highest RCIFH miRNAs on the key genes of nitric oxide signaling, retinoid acid receptor (RAR) activation, transforming growth factor-β (TGF-β) signaling, and interleukin 6 (IL-6) signaling pathways in impaired fracture healing. Red: gene expression and function are activated in aged female mice compared to adult ones. Green: gene expression and function are inhibited in aged female mice compared to adult ones. Figure 6 The miR-494 inhibits chondrogenic differentiation in C3H10T1/2 cells. The miR-494, anti-miR-494, or inactive controls were transfected into C3H10T1/2 cells. Expressions of chondrogenic differentiation markers (Acan, Col2a1, and Col10a1) were detected by QPCR at day 1, 7, and 14 after transfection. Note: Data was represented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001. ijms-17-01260-t001_Table 1Table 1 The top 10 high miRNA contribution to impaired fracture healing (RCIFH) microRNAs (miRNAs) at 2- or 4-week (|log2FC| ≥ 1). A positive RCIFH miRNA promotes bone fracture healing, while a negative RCIFH miRNA inhibits it. 2-Week 4-Week miRNA RCIFH miRNA RCIFH miR-494 −1692.98 miR-494 −2038.91 miR-139-5p −428.294 miR-125a-3p −492.18 miR-142-5p −278.332 miR-24 326.6378 miR-206 269.8789 miR-144 −271.429 miR-181b 222.3468 miR-497 259.1864 miR-199a-5p 203.6746 miR-195 246.0054 miR-223 −192.034 miR-15b 230.7585 miR-144 −182.219 miR-23b 168.3962 miR-125a-5p 180.9605 let-7e 143.6061 miR-22 −151.984 miR-223 140.6847 ==== Refs References 1. Gruber R. Koch H. Doll B.A. Tegtmeier F. Einhorn T.A. Hollinger J.O. Fracture healing in the elderly patient Exp. Gerontol. 2006 41 1080 1093 10.1016/j.exger.2006.09.008 17092679 2. Virk M.S. Lieberman J.R. Biologic adjuvants for fracture healing Arthritis Res. Ther. 2012 14 225 10.1186/ar4053 23198865 3. Nguyen T.V. Center J.R. Eisman J.A. Osteoporosis: Underrated, underdiagnosed and undertreated Med. J. Aust. 2004 180 S18 S22 14984358 4. Kanis J.A. Johnell O. Oden A. Sembo I. Redlund-Johnell I. Dawson A. de Laet C. Jonsson B. Long-term risk of osteoporotic fracture in Malmo Osteoporos. Int. 2000 11 669 674 10.1007/s001980070064 11095169 5. Bliuc D. Nguyen N.D. Milch V.E. Nguyen T.V. Eisman J.A. Center J.R. Mortality risk associated with low-trauma osteoporotic fracture and subsequent fracture in men and women JAMA 2009 301 513 521 10.1001/jama.2009.50 19190316 6. Castronuovo E. Pezzotti P. Franzo A. Di Lallo D. Guasticchi G. Early and late mortality in elderly patients after hip fracture: A cohort study using administrative health databases in the Lazio region, Italy BMC Geriatr. 2011 11 1260 10.1186/1471-2318-11-37 21819551 7. Burge R. Dawson-Hughes B. Solomon D.H. Wong J.B. King A. Tosteson A. Incidence and economic burden of osteoporosis-related fractures in the United States, 2005–2025 J. Bone Miner. Res. 2007 22 465 475 10.1359/jbmr.061113 17144789 8. Blume S.W. Curtis J.R. Medical costs of osteoporosis in the elderly medicare population Osteoporos. Int. 2011 22 1835 1844 10.1007/s00198-010-1419-7 21165602 9. Bartel D.P. MicroRNAs: Genomics, biogenesis, mechanism, and function Cell 2004 116 281 297 10.1016/S0092-8674(04)00045-5 14744438 10. Kong Y.W. Ferland-McCollough D. Jackson T.J. Bushell M. MicroRNAs in cancer management Lancet Oncol. 2012 13 e249 e258 10.1016/S1470-2045(12)70073-6 22652233 11. Murata K. Ito H. Yoshitomi H. Yamamoto K. Fukuda A. Yoshikawa J. Furu M. Ishikawa M. Shibuya H. Matsuda S. Inhibition of miR-92a enhances fracture healing via promoting angiogenesis in a model of stabilized fracture in young mice J. Bone Miner. Res. 2014 29 316 326 10.1002/jbmr.2040 23857760 12. Pourrajab F. Vakili Zarch A. Hekmatimoghaddam S. Zare-Khormizi M.R. MicroRNAs; easy and potent targets in optimizing therapeutic methods in reparative angiogenesis J. Cell. Mol. Med. 2015 19 2702 2714 10.1111/jcmm.12669 26416208 13. Gurusinghe S. Strappe P. Gene modification of mesenchymal stem cells and articular chondrocytes to enhance chondrogenesis BioMed Res. Int. 2014 2014 369528 10.1155/2014/369528 24963479 14. Dong S. Yang B. Guo H. Kang F. MicroRNAs regulate osteogenesis and chondrogenesis Biochem. Biophys. Res. Commun. 2012 418 587 591 10.1016/j.bbrc.2012.01.075 22306817 15. Wang X. Guo B. Li Q. Peng J. Yang Z. Wang A. Li D. Hou Z. Lv K. Kan G. miR-214 targets atf4 to inhibit bone formation Nat. Med. 2013 19 93 100 10.1038/nm.3026 23223004 16. Van Wijnen A.J. van de Peppel J. van Leeuwen J.P. Lian J.B. Stein G.S. Westendorf J.J. Oursler M.J. Im H.J. Taipaleenmaki H. Hesse E. MicroRNA functions in osteogenesis and dysfunctions in osteoporosis Curr. Osteoporos. Rep. 2013 11 72 82 10.1007/s11914-013-0143-6 23605904 17. Seeliger C. Karpinski K. Haug A.T. Vester H. Schmitt A. Bauer J.S. van Griensven M. Five freely circulating miRNAs and bone tissue miRNAs are associated with osteoporotic fractures J. Bone Miner. Res. 2014 29 1718 1728 10.1002/jbmr.2175 24431276 18. Sun Y. Xu L. Huang S. Hou Y. Liu Y. Chan K.M. Pan X.H. Li G. miR-21 overexpressing mesenchymal stem cells accelerate fracture healing in a rat closed femur fracture model BioMed Res. Int. 2015 2015 412327 10.1155/2015/412327 25879024 19. Schindeler A. McDonald M.M. Bokko P. Little D.G. Bone remodeling during fracture repair: The cellular picture Semin. Cell Dev. Biol. 2008 19 459 466 10.1016/j.semcdb.2008.07.004 18692584 20. Saran U. Gemini Piperni S. Chatterjee S. Role of angiogenesis in bone repair Arch. Biochem. Biophys. 2014 561 109 117 10.1016/j.abb.2014.07.006 25034215 21. Nakajima F. Ogasawara A. Goto K. Moriya H. Ninomiya Y. Einhorn T.A. Yamazaki M. Spatial and temporal gene expression in chondrogenesis during fracture healing and the effects of basic fibroblast growth factor J. Orthop. Res. 2001 19 935 944 10.1016/S0736-0266(01)00024-9 11562144 22. Bahney C.S. Hu D.P. Miclau T. 3rd Marcucio R.S. The multifaceted role of the vasculature in endochondral fracture repair Front. Endocrinol. 2015 6 4 10.3389/fendo.2015.00004 25699016 23. Minkwitz S. Fassbender M. Kronbach Z. Wildemann B. Longitudinal analysis of osteogenic and angiogenic signaling factors in healing models mimicking atrophic and hypertrophic non-unions in rats PLoS ONE 2015 10 1260 10.1371/journal.pone.0124217 25910190 24. Hadjiargyrou M. Zhi J. Komatsu D.E. Identification of the microRNA transcriptome during the early phases of mammalian fracture repair Bone 2016 87 78 88 10.1016/j.bone.2016.03.011 27058875 25. Wu X. Zhang Y. Guo X. Xu H. Xu Z. Duan D. Wang K. Identification of differentially expressed microRNAs involved in non-traumatic osteonecrosis through microRNA expression profiling Gene 2015 565 22 29 10.1016/j.gene.2015.03.072 25863178 26. Tu M. Tang J. He H. Cheng P. Chen C. miR-142–5p promotes bone repair by maintaining osteoblast activity J. Bone Miner. Metab. 2016 10.1007/s00774-016-0757-8 27085967 27. Waki T. Lee S.Y. Niikura T. Iwakura T. Dogaki Y. Okumachi E. Kuroda R. Kurosaka M. Profiling microRNA expression in fracture nonunions: Potential role of microRNAs in nonunion formation studied in a rat model Bone Jt. J. 2015 97-B 1144 1151 10.1302/0301-620X.97B8.34966 26224835 28. Weilner S. Skalicky S. Salzer B. Keider V. Wagner M. Hildner F. Gabriel C. Dovjak P. Pietschmann P. Grillari-Voglauer R. Differentially circulating miRNAs after recent osteoporotic fractures can influence osteogenic differentiation Bone 2015 79 43 51 10.1016/j.bone.2015.05.027 26026730 29. Guerra-Garcia M.M. Rodriguez-Fernandez J.B. Puga-Sarmiento E. Charle-Crespo M.A. Gomes-Carvalho C.S. Prejigueiro-Santas A. [Incidence of hip fractures due to osteoporosis in relation to the prescription of drugs for their prevention and treatment in Galicia, Spain] Aten. Primaria 2011 43 82 88 20554353 30. Almeida M.I. Silva A.M. Vasconcelos D.M. Almeida C.R. Caires H. Pinto M.T. Calin G.A. Santos S.G. Barbosa M.A. miR-195 in human primary mesenchymal stromal/stem cells regulates proliferation, osteogenesis and paracrine effect on angiogenesis Oncotarget 2016 7 7 22 10.18632/oncotarget.6589 26683705 31. Chen S. Zhao G. Miao H. Tang R. Song Y. Hu Y. Wang Z. Hou Y. MicroRNA-494 inhibits the growth and angiogenesis-regulating potential of mesenchymal stem cells FEBS Lett. 2015 589 710 717 10.1016/j.febslet.2015.01.038 25660325 32. Welten S.M. Bastiaansen A.J. de Jong R.C. de Vries M.R. Peters E.A. Boonstra M.C. Sheikh S.P. La Monica N. Kandimalla E.R. Quax P.H. Inhibition of 14q32 microRNAs miR-329, miR-487b, miR-494, and miR-495 increases neovascularization and blood flow recovery after ischemia Circ. Res. 2014 115 696 708 10.1161/CIRCRESAHA.114.304747 25085941 33. Street J. Bao M. deGuzman L. Bunting S. Peale F.V. Jr. Ferrara N. Steinmetz H. Hoeffel J. Cleland J.L. Daugherty A. Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover Proc. Natl. Acad. Sci. USA 2002 99 9656 9661 10.1073/pnas.152324099 12118119 34. Hing K.A. Bone repair in the twenty-first century: Biology, chemistry or engineering? Philos. Trans. Ser. A Math. Phys. Eng. Sci. 2004 362 2821 2850 10.1098/rsta.2004.1466 15539372 35. Naik A.A. Xie C. Zuscik M.J. Kingsley P. Schwarz E.M. Awad H. Guldberg R. Drissi H. Puzas J.E. Boyce B. Reduced COX-2 expression in aged mice is associated with impaired fracture healing J. Bone Miner. Res. 2009 24 251 264 10.1359/jbmr.081002 18847332 36. Yuan H.F. Von Roemeling C. Gao H.D. Zhang J. Guo C.A. Yan Z.Q. Analysis of altered microRNA expression profile in the reparative interface of the femoral head with osteonecrosis Exp. Mol. Pathol. 2015 98 158 163 10.1016/j.yexmp.2015.01.002 25612520 37. Jevon M. Sabokbar A. Fujikawa Y. Hirayama T. Neale S.D. Wass J. Athanasou N.A. Gender- and age-related differences in osteoclast formation from circulating precursors J. Endocrinol. 2002 172 673 681 10.1677/joe.0.1720673 11874715 38. Plummer P.N. Freeman R. Taft R.J. Vider J. Sax M. Umer B.A. Gao D. Johns C. Mattick J.S. Wilton S.D. MicroRNAs regulate tumor angiogenesis modulated by endothelial progenitor cells Cancer Res. 2013 73 341 352 10.1158/0008-5472.CAN-12-0271 22836757 39. Bae Y. Yang T. Zeng H.C. Campeau P.M. Chen Y. Bertin T. Dawson B.C. Munivez E. Tao J. Lee B.H. miRNA-34c regulates notch signaling during bone development Hum. Mol. Genet. 2012 21 2991 3000 10.1093/hmg/dds129 22498974 40. Meng F. Zhang Z. Chen W. Huang G. He A. Hou C. Long Y. Yang Z. Zhang Z. Liao W. MicroRNA-320 regulates matrix metalloproteinase-13 expression in chondrogenesis and interleukin-1β-induced chondrocyte responses Osteoarthr. Cartil. 2016 24 932 941 10.1016/j.joca.2015.12.012 26774733 41. Underhill T.M. Sampaio A.V. Weston A.D. Retinoid signalling and skeletal development Novartis Found. Symp. 2001 232 171 185 11277079 42. Cash D.E. Bock C.B. Schughart K. Linney E. Underhill T.M. Retinoic acid receptor α function in vertebrate limb skeletogenesis: A modulator of chondrogenesis J. Cell Biol. 1997 136 445 457 10.1083/jcb.136.2.445 9015314 43. Chen G. Deng C. Li Y.-P. TGF-β and BMP signaling in osteoblast differentiation and bone formation Int. J. Biol. Sci. 2012 8 272 10.7150/ijbs.2929 22298955 44. Peruzzi B. Cappariello A. Del Fattore A. Rucci N. de Benedetti F. Teti A. c-Src and IL-6 inhibit osteoblast differentiation and integrate IGFBP5 signalling Nat. Commun. 2012 3 630 10.1038/ncomms1651 22252554 45. Tamura T. Udagawa N. Takahashi N. Miyaura C. Tanaka S. Yamada Y. Koishihara Y. Ohsugi Y. Kumaki K. Taga T. Soluble interleukin-6 receptor triggers osteoclast formation by interleukin 6 Proc. Natl. Acad. Sci. USA 1993 90 11924 11928 10.1073/pnas.90.24.11924 8265649 46. Axmann R. Bohm C. Kronke G. Zwerina J. Smolen J. Schett G. Inhibition of interleukin-6 receptor directly blocks osteoclast formation in vitro and in vivo Arthritis Rheum. 2009 60 2747 2756 10.1002/art.24781 19714627 47. Kudo O. Sabokbar A. Pocock A. Itonaga I. Fujikawa Y. Athanasou N.A. Interleukin-6 and interleukin-11 support human osteoclast formation by a rankl-independent mechanism Bone 2003 32 1 7 10.1016/S8756-3282(02)00915-8 12584029 48. Kaneshiro S. Ebina K. Shi K. Higuchi C. Hirao M. Okamoto M. Koizumi K. Morimoto T. Yoshikawa H. Hashimoto J. IL-6 negatively regulates osteoblast differentiation through the SHP2/MEK2 and SHP2/Akt2 pathways in vitro J. Bone Miner. Metab. 2014 32 378 392 10.1007/s00774-013-0514-1 24122251 49. Hughes F.J. Howells G.L. Interleukin-6 inhibits bone formation in vitro Bone Miner. 1993 21 21 28 10.1016/S0169-6009(08)80117-1 8324417 50. Kwan Tat S. Padrines M. Theoleyre S. Heymann D. Fortun Y. IL-6, RANKL, TNF-α/IL-1: Interrelations in bone resorption pathophysiology Cytokine Growth Factor Rev. 2004 15 49 60 14746813 51. Bonnarens F. Einhorn T.A. Production of a standard closed fracture in laboratory animal bone J. Orthop. Res. 1984 2 97 101 10.1002/jor.1100020115 6491805 52. He Y.X. Liu Z. Pan X.H. Tang T. Guo B.S. Zheng L.Z. Xie X.H. Wang X.L. Lee K.M. Li G. Deletion of estrogen receptor β accelerates early stage of bone healing in a mouse osteotomy model Osteoporos. Int. 2012 23 377 389 10.1007/s00198-011-1812-x 22037970 53. Smyth G.K. Limma: Linear models for microarray data Bioinformatics and Computational Biology Solutions Using R and Bioconductor Springer Berlin, Germany 2005 397 420 54. Oughtred R. Chatr-Aryamontri A. Breitkreutz B.J. Chang C.S. Rust J.M. Theesfeld C.L. Heinicke S. Breitkreutz A. Chen D. Hirschman J. Use of the biogrid database for analysis of yeast protein and genetic interactions Cold Spring Harb. Protoc. 2016 2016 10.1101/pdb.prot088880 26729913 55. Chou C.H. Chang N.W. Shrestha S. Hsu S.D. Lin Y.L. Lee W.H. Yang C.D. Hong H.C. Wei T.Y. Tu S.J. Mirtarbase 2016: Updates to the experimentally validated miRNA-target interactions database Nucleic Acids Res. 2016 44 D239 D247 10.1093/nar/gkv1258 26590260 56. Reczko M. Maragkakis M. Alexiou P. Grosse I. Hatzigeorgiou A.G. Functional microRNA targets in protein coding sequences Bioinformatics 2012 28 771 776 10.1093/bioinformatics/bts043 22285563 57. John B. Enright A.J. Aravin A. Tuschl T. Sander C. Marks D.S. Human microRNA targets PLoS Biol. 2004 2 1260 10.1371/journal.pbio.0020363 15502875 58. Wong N. Wang X. miRDB: An online resource for microRNA target prediction and functional annotations Nucleic Acids Res. 2015 43 D146 D152 10.1093/nar/gku1104 25378301 59. Lewis B.P. Burge C.B. Bartel D.P. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets Cell 2005 120 15 20 10.1016/j.cell.2004.12.035 15652477 60. Krämer A. Green J. Pollard J. Tugendreich S. Causal analysis approaches in ingenuity pathway analysis (IPA) Bioinformatic 2014 30 523 530 10.1093/bioinformatics/btt703 24336805

PMC005xxxxxx/PMC5000659.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081261ijms-17-01261ArticleCerebral Hyperperfusion after Revascularization Inhibits Development of Cerebral Ischemic Lesions Due to Artery-to-Artery Emboli during Carotid Exposure in Endarterectomy for Patients with Preoperative Cerebral Hemodynamic Insufficiency: Revisiting the “Impaired Clearance of Emboli” Concept Fujimoto Kentaro 1Matsumoto Yoshiyasu 1Oikawa Kohki 1Nomura Jun-ichi 1Shimada Yasuyoshi 1Fujiwara Shunrou 1Terasaki Kazunori 2Kobayashi Masakazu 1Yoshida Kenji 1Ogasawara Kuniaki 1*Henein Michael Academic Editor1 Department of Neurosurgery, School of Medicine, Iwate Medical University, 19-1 Uchmaru, 020-8505 Morioka, Japan; [email protected] (K.F.); [email protected] (Y.M.); [email protected] (K.O.); [email protected] (J.-i.N.); [email protected] (Y.S.); [email protected] (S.F.); [email protected] (M.K.); [email protected] (K.Y.)2 Cyclotron Research Center, School of Medicine, Iwate Medical University, 19-1 Uchmaru, 020-8505 Morioka, Japan; [email protected]* Correspondence: [email protected]; Tel.: +81-19-651-5111; Fax: +81-19-625-879903 8 2016 8 2016 17 8 126101 6 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The purpose of the present study was to determine whether cerebral hyperperfusion after revascularization inhibits development of cerebral ischemic lesions due to artery-to-artery emboli during exposure of the carotid arteries in carotid endarterectomy (CEA). In patients undergoing CEA for internal carotid artery stenosis (≥70%), cerebral blood flow (CBF) was measured using single-photon emission computed tomography (SPECT) before and immediately after CEA. Microembolic signals (MES) were identified using transcranial Doppler during carotid exposure. Diffusion-weighted magnetic resonance imaging (DWI) was performed within 24 h after surgery. Of 32 patients with a combination of reduced cerebrovascular reactivity to acetazolamide on preoperative brain perfusion SPECT and MES during carotid exposure, 14 (44%) showed cerebral hyperperfusion (defined as postoperative CBF increase ≥100% compared with preoperative values), and 16 (50%) developed DWI-characterized postoperative cerebral ischemic lesions. Postoperative cerebral hyperperfusion was significantly associated with the absence of DWI-characterized postoperative cerebral ischemic lesions (95% confidence interval, 0.001–0.179; p = 0.0009). These data suggest that cerebral hyperperfusion after revascularization inhibits development of cerebral ischemic lesions due to artery-to-artery emboli during carotid exposure in CEA, supporting the “impaired clearance of emboli” concept. Blood pressure elevation following carotid declamping would be effective when embolism not accompanied by cerebral hyperperfusion occurs during CEA. artery-to-artery embolismcarotid endarterectomycerebral hemodynamic insufficiencycerebral hyperperfusionischemic lesion ==== Body 1. Introduction It has been previously reported that hemodynamic and embolic mechanisms are closely linked, and they may act together to cause cerebral ischemia [1,2]. The authors suggested that clearance of emboli from a proximal lesion may be decreased by low blood flow velocity in a cerebral artery, which may lead to ischemia from emboli in poorly perfused areas of the brain. In support of this, several previous studies showed that there was a relationship between embolic and hemodynamic mechanisms, particularly in border zone regions having impaired wash-out due to artery-to-artery embolism in cases of middle cerebral artery stenosis [3,4] or in new postoperative neurological deficits caused by emboli developing during carotid artery stenting associated with intraprocedural lower middle cerebral artery blood flow velocity [5]. More than 70% of the intraoperative procedure-related strokes that occur during carotid endarterectomy (CEA) are due to surgical site embolisms [6]. When intraoperative transcranial Doppler (TCD) is used to monitor the middle cerebral artery (MCA), more than 90% of patients undergoing CEA are found to have microembolic signals (MES) [6,7,8]. However, the stage of CEA determines the quality and quantity of MES that are detected [6,8,9]. When the carotid arteries are being exposed, plaque that has not been removed is exposed to blood flow and can be a source of emboli. In such cases, emboli can be dislodged from the surgical site into the intracranial arteries during manipulation of the carotid arteries [8]. In addition, the MES that are detected are considered to represent solid masses, because the target vessel is closed while the vessel is being exposed [10]. However, once the carotid artery walls are incised for the endarterectomy, many harmless gaseous MES may be seen during carotid declamping as air enters the arterial lumina [8,11]. MES detection while the artery is being exposed has been shown to be correlated with the development of post-CEA cerebral ischemic lesions on diffusion-weighted imaging (DWI) [7,8,9,11,12,13]. Furthermore, these ischemic lesions that are related to the emboli generated during carotid artery exposure have been shown to be related to preoperative hemodynamic cerebral compromise, such as reduced cerebrovascular reactivity (CVR) to acetazolamide [14], providing support for the concept of “impaired clearance of emboli”. Cerebral hyperperfusion is defined as a major increase in ipsilateral cerebral blood flow (CBF) after surgical repair of carotid stenosis that is well above the metabolic demands of the brain tissue; it is another adverse event following CEA [15]. It occasionally evolves into cerebral hyperperfusion syndrome, whose characteristic manifestations include face and eye pain, unilateral headache, seizure, focal neurological symptoms, and disturbance of consciousness secondary to intracerebral hemorrhage or cerebral edema [15,16,17,18]. Impairment of the cerebral hemodynamic reserve before surgery may be related to post-CEA hyperperfusion, and quick normalization of perfusion pressure after CEA may produce hyperperfusion in brain regions with diminished autoregulation from chronic ischemia [17,18]. This hypothesis is consistent with the observation that reduced CVR to acetazolamide prior to surgery is a significant predictor of hyperperfusion after CEA [19,20,21]. Thus, both cerebral ischemic lesions due to artery-to-artery embolism and cerebral hyperperfusion may develop simultaneously during CEA in patients with preoperative cerebral hemodynamic impairment. When broadening the interpretation of the “impaired clearance of emboli” concept, blood flow greater than the normal level can inhibit development of ischemic lesions due to emboli in the brain, and research regarding the influence of cerebral hyperperfusion on the development of cerebral ischemic lesions due to artery-to-artery embolism seems interesting from the standpoint of this concept. The aim of the present study was to determine whether broadening the interpretation of the “impaired clearance of emboli” concept is correct, namely cerebral hyperperfusion after revascularization inhibits development of cerebral ischemic lesions due to intraoperative artery-to-artery emboli. In order to do this, the relationship between development of DWI-characterized postoperative cerebral ischemic lesions and cerebral hyperperfusion was investigated in patients with a combination of preoperatively reduced CVR to acetazolamide on brain perfusion single-photon emission computed tomography (SPECT) and MES on TCD during carotid artery exposure in CEA. 2. Results 2.1. Trial Profile Figure 1 shows the patient flow chart for this study. Six hundred and thirty patients with ipsilateral internal carotid artery (ICA) stenosis ≥70% and useful residual function were scheduled for CEA and consented to participate in the present study. Of these 630 patients, 190 were defined as having reduced CVR to acetazolamide. Of these 190 patients, five did not undergo CEA and were excluded from the analysis. Of the 185 patients who underwent CEA, 23 did not show reliable TCD monitoring during carotid exposure because of failure to obtain an adequate bone window, and 26 showed electroencephalography (EEG)-defined hemispheric ischemia during ICA clamping; these 47 patients (two had both conditions) were excluded from the analysis. Of the remaining 138 patients, 32 had MES during exposure of the carotid arteries and were finally analyzed. Data acquisition with brain perfusion SPECT was completed within 3 h after declamping of the ICA in all these 32 patients. They all underwent DWI 24 h after surgery. 2.2. Clinical Characteristics The mean age of the 32 patients (29 men, three women) was 71.7 ± 4.5 (mean ± standard deviation (SD)) years (range of 63–85 years). Twenty-six patients had preoperative hypertension, and 23 patients received antihypertensive drugs (calcium antagonist alone for five, angiotensin receptor blocker alone for 13 and both for five). Thirteen patients had preoperative diabetes mellitus, and all these patients received antidiabetic drugs. Fifteen patients had preoperative dyslipidemia, and 12 patients received a statin (strong statin for six). Seven patients had ischemic heart or valvular disease that did not satisfy the criteria for high-risk factors for CEA in the Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) study (congestive heart failure, abnormal stress test, or need for open-heart surgery) [22]. None of the 32 patients had atrial fibrillation. Twenty-five patients had ipsilateral carotid territory symptoms, and seven patients had asymptomatic ICA stenosis. The overall average degree of ICA stenosis was 85.4% ± 8.8% (range, 70%–99%), with nine patients showing >70% stenosis or occlusion in the contralateral ICA. Preoperative CBF and CVR to acetazolamide were 33.1 ± 6.1 mL/100 g/min (range of 22.6–44.5 mL/100 g/min) and 8.6% ± 6.9% (range of −8.3%–18.0%), respectively. Preoperative systolic blood pressure was 133.6 ± 11.7 mmHg (range of 111–156 mmHg). The number of MES ranged from one to 14 (4.3 ± 3.9). The interval from the first MES to ICA declamping ranged from 34 to 59 min (45.3 ± 6.4 min). The interval from the last MES to ICA declamping ranged from 35 to 65 min (46.8 ± 6.6 min). Mean systolic blood pressure during carotid exposure was 113.8 ± 10.7 mmHg (range of 95–138 mmHg). Mean duration of ICA clamping was 36.0 ± 5.7 min (range of 28–47 min). Mean systolic blood pressure in the post-carotid declamping period in surgery was 121.4 ± 11.9 mmHg (range of 105–141 mmHg). Mean systolic blood pressure in the postoperative period (within 24 h after surgery) was 127.8 ± 12.6 mmHg (range of 109–147 mmHg). The mean rate of blood pressure–measured points with successfully controlled blood pressure to all measured points in the post-carotid declamping period in surgery was 85.9% ± 3.9% (range of 72%–93%). The mean rate of blood pressure–measured points with successfully controlled blood pressure to all the measured points in the postoperative period was 76.1% ± 4.9% (range of 67%–92%). 2.3. Postoperative Events In the 32 patients studied, postoperative CBF was 52.9 ± 17.5 mL/100 g/min (range of 35.8–90.5 mL/100 g/min); 14 patients (44%) met CBF criteria for cerebral hyperperfusion. Sixteen patients (50%) developed new postoperative ischemic lesions on DWI 24 h after surgery in the cortex and/or white matter in the cerebral hemisphere ipsilateral to CEA. All new ischemic lesions were spotty, and their diameters were 1 cm or less. Five (16%) of 32 patients studied developed new neurological deficits after recovery from general anesthesia. All deficits included slight hemiparesis contralateral to the CEA. These deficits resolved completely within 12 h in these five patients, and they underwent additional DWI between 6 and 8 h after surgery and had new postoperative ischemic lesions on both the first (6 to 8 h after surgery) and second (24 h after surgery) postoperative DWI examinations. 2.4. Postoperative Cerebral Hyperperfusion vs. Diffusion-Weighted Imaging (DWI)-Characterized Postoperative Cerebral Ischemic Lesions Results of univariate analyses of factors related to the development of DWI-characterized postoperative cerebral ischemic lesions are shown in Table 1. The postoperative CBF and the incidence of postoperative cerebral hyperperfusion were significantly higher in patients without than in those with DWI-characterized postoperative cerebral ischemic lesions. Other variables were not significantly associated with DWI-characterized postoperative cerebral ischemic lesions. After eliminating variables that were closely related, the following items with values of p < 0.2 in univariate analyses were adopted as confounders in the logistic regression model for multivariate analysis: degree of ICA stenosis and postoperative CBF or postoperative cerebral hyperperfusion (since the latter two interacted, each item was adopted individually). This analysis showed that greater postoperative CBF (95% confidence interval, 0.616–0.938; p = 0.0104) or postoperative cerebral hyperperfusion (95% confidence interval, 0.001–0.179; p = 0.0009) was significantly associated with the absence of DWI-characterized postoperative cerebral ischemic lesions. Figure 2 shows the relationships between the number of MES, postoperative CBF, cerebral hyperperfusion, and the development of DWI-characterized postoperative cerebral ischemic lesions. Postoperative CBF in patients with cerebral hyperperfusion ranged from mean + 3.9 SD to mean + 11.8 SD of the control value. Whereas 15 (83%) of 18 patients without postoperative cerebral hyperperfusion showed DWI-characterized postoperative cerebral ischemic lesions, only one (7%) of 14 patients with hyperperfusion had these ischemic lesions. 2.5. Case Presentation Figure 3 shows images of brain perfusion SPECT, TCD, and DWI in a 74-year-old man with symptomatic ICA stenosis (90%) showing DWI-characterized postoperative cerebral ischemic lesions due to MES during exposure of the carotid arteries despite development of cerebral hyperperfusion after left CEA. 3. Discussion 3.1. Findings The present study demonstrated that cerebral hyperperfusion after revascularization inhibits the development of cerebral ischemic lesions due to artery-to-artery emboli during carotid exposure in CEA for patients with preoperatively impaired cerebral hemodynamics, supporting the “impaired clearance of emboli” concept when broadening its interpretation. 3.2. Reason of Patient Exclusion Hemodynamic cerebral ischemia due to hemispheric cerebral hypoperfusion during ICA clamping, as well as emboli from the surgical site, plays a significant role in the development of new ischemic lesions after CEA [11,23]. Intraoperative EEG monitoring is the most widely used and best documented method for the detection of hemispheric cerebral hypoperfusion due to carotid clamping [24]. To investigate the development of cerebral ischemic lesions caused by MES rather than by hemispheric cerebral hypoperfusion during ICA clamping, patients with EEG-defined hemispheric ischemia during ICA clamping were excluded from the present study. 3.3. Data Interpretation In the present study, 44% of patients with preoperatively reduced CVR to acetazolamide showed cerebral hyperperfusion immediately after surgery. This incidence was comparable to a previous study [19,20,21]. All postoperative ischemic lesions on DWI that were newly developed in the cortex and/or white matter in the cerebral hemisphere ipsilateral to CEA were spotty, and their diameters were 1 cm or less. Furthermore, the duration of ICA clamping did not differ between patients with and without DWI-characterized postoperative cerebral ischemic lesions. Thus, these ischemic lesions were possibly due to artery-to-artery embolism rather than cerebral hemispheric ischemia during ICA clamping. More than 80% of patients with a combination of preoperatively reduced CVR to acetazolamide and MES during carotid exposure, when they did not exhibit postoperative cerebral hyperperfusion, developed DWI-characterized postoperative cerebral ischemic lesions, which corresponded with previous findings [14]. In the present study, preoperative CBF, preoperative CVR to acetazolamide, mean systolic blood pressure during carotid exposure, and interval from the first or last MES to ICA declamping did not differ between patients with and without DWI-characterized postoperative cerebral ischemic lesions. Thus, perfusion in the cerebral hemisphere ipsilateral to surgery during carotid exposure and the duration of cerebral ischemia caused by emboli until ICA declamping might be equivalent between these two subgroups of patients. The duration of ICA clamping also did not differ between them. Nevertheless, greater postoperative CBF or postoperative cerebral hyperperfusion (postoperative CBF ≥ mean + 3.9 SD of the control value) was associated with the absence of DWI-characterized postoperative cerebral ischemic lesions, and only 7% of patients with a combination of MES during carotid exposure and postoperative cerebral hyperperfusion developed these lesions. These findings suggested that blood flow that increased far beyond the normal level might clear cerebral emboli generated from the surgical site, inhibiting the development of ischemic lesions. These support the “impaired clearance of emboli” concept if the interpretation of this concept is broadened. We have another hypothesis regarding the correlation between postoperative cerebral hyperperfusion and the development of ischemic lesions by emboli in patients with reduced CVR to acetazolamide. Reduced CVR to acetazolamide implies a chronic reduction in cerebral perfusion pressure and poor collateral blood flow [25,26,27]. When emboli generated from a lesion in the ICA acutely disturb blood flow in the cerebral artery, cerebral blood flow may be further decreased in the affected vascular territory with the pre-existing chronic reduction in cerebral perfusion pressure. However, if cerebral ischemic lesions have not yet formed between the onset of emboli and ICA declamping (the interval ranged from approximately 30 min to 60 min in the present study), hyperperfusion after ICA declamping may lead to an extreme increase in collateral blood flow to the affected vascular territory, inhibiting the postoperative development of new cerebral ischemic lesions. 3.4. Future Directions The present study suggests that CBF greater than the normal level after declamping of the ICA can inhibit development of cerebral ischemic lesions due to emboli from the surgical site during exposure of the carotid arteries. On the other hand, postoperative cerebral hyperperfusion, which is defined as a postoperative CBF increase of ≥100% when compared to preoperative values, occasionally evolves into cerebral hyperperfusion syndrome, leading to intracerebral hemorrhage [15,16,17,18]. Strict postoperative control of blood pressure (systolic blood pressure < 90 mmHg) reportedly prevents the development of intracerebral hemorrhage [16,20]. Further, intraoperative monitoring of MCA flow velocity using TCD or regional cerebral oxygen saturation using near-infrared spectroscopy are reliable methods of identifying patients with cerebral hyperperfusion following declamping of the ICA during CEA [28,29]. On the basis of these findings, we propose a practical clinical algorithm to prevent development of embolic ischemic events and hyperperfusion-related hemorrhage in CEA: when the intraoperative monitoring suggests development of cerebral hyperperfusion following declamping of the ICA, the blood pressure should then be reduced; when the intraoperative monitoring suggests development of embolism from the surgical site during carotid exposure that is not accompanied by cerebral hyperperfusion, the blood pressure should be elevated above the preoperative value following declamping of the ICA. Further investigation to determine whether the latter procedure prevents development of cerebral ischemic lesions would be of benefit, although, in the present study, blood pressure was reduced for all patients regardless of the presence or absence of cerebral hyperperfusion after declamping of the ICA. 3.5. Study Limitations Although TCD detects emboli generated from the surgical site of the carotid arteries as MES, it cannot provide information about the size and characteristics of each embolus, which may affect the development of postoperative cerebral ischemic lesions. The present results did not take into account these two factors. 4. Materials and Methods 4.1. Subjects The present study was designed as a prospective, observational study. This study was approved by the Regional Ethical Board in Iwate Medical University (H22-3) and was in compliance with the Helsinki Declaration, and written, informed consent was obtained from all patients or their next of kin prior to participation. Of symptomatic or asymptomatic patients with ipsilateral ICA stenosis ≥70%, as per the North American Symptomatic Carotid Endarterectomy Trial [30], on angiography/arterial catheterization, and useful residual function (modified Rankin scale score 0, 1, or 2) who were scheduled for CEA of the carotid bifurcation, those who satisfied the following inclusion criteria were prospectively selected for the present study: having preoperatively reduced CVR to acetazolamide according to the methods described below (see “4.2. CBF Measurements” section); undergoing CEA; and having MES during exposure of the carotid arteries under reliable TCD monitoring according to the methods described below (see “4.3. TCD Monitoring” section). Patients who showed electroencephalography (EEG)-defined cerebral hemispheric ischemia during ICA clamping according to the methods described below (see “4.5. Preoperative, Intraoperative, and Postoperative Management” section) were excluded from the present study. 4.2. CBF Measurements CBF was assessed using [123I]N-isopropyl-p-iodoamphetamine (IMP) and SPECT with a ring-type scanner (Headtome-SET 080; Shimadzu, Kyoto, Japan) within 14 days before and immediately after CEA. CBF measurement with acetazolamide challenge was also performed before CEA. The [123I]IMP SPECT study with and without acetazolamide challenge was performed as described previously [31,32]. After a 1 min intravenous infusion of 222 MBq of [123I]IMP (5 mL volume) at a constant rate of 5 mL/min and a 1 min infusion of physiologic saline at the same rate, data acquisition was performed at a midscan time of 30 min after the [123I]IMP administration for a scan duration of 20 min. At 10 min after the beginning of the [123I]IMP infusion, arterial blood (1 mL) was taken from the brachial artery. The whole-blood radioactivity of each blood sample obtained was measured using a well counter that was cross-calibrated to the SPECT scanner. All reconstructed SPECT images were corrected for the radioactive decay of 123I back to the [123I]IMP injection start time, normalized by the data collection time and cross-calibrated to the well counter system. The CBF images were calculated according to the [123I]IMP-autoradiography method [31,32]. The whole-blood radioactivity counts of the single blood sample were referred to the standard input function. All SPECT images were transformed into standard brain size and shape by linear and nonlinear transformations using statistical parametric mapping 2 software for anatomical standardization [33]. A three-dimensional stereotactic region-of-interest (ROI) template was used to automatically place 318 constant ROIs in both cerebral and cerebellar hemispheres [34]. ROIs were grouped into 10 segments (callosomarginal, pericallosal, precentral, central, parietal, angular, temporal, posterior, hippocampal, and cerebellar) in each hemisphere according to the arterial supply. Five (precentral, central, parietal, angular, and temporal) of these 10 segments were combined and defined as an ROI perfused by the middle cerebral artery (MCA) (Figure 4). The mean value of all pixels in the MCA ROI in the cerebral hemisphere ipsilateral to CEA was calculated. Preoperative CVR to acetazolamide in the cerebral hemisphere ipsilateral to CEA was calculated as follows: CVR (%) = [(CBF with acetazolamide challenge − CBF at the resting state)/CBF at the resting state] × 100. For CBF in the resting state and CVR to acetazolamide, data described previously ((mean ± SD), 35.9 ± 4.4 mL/100 g/min and 36.8% ± 9.2%, respectively) were used as control values, and decreased CVR to acetazolamide was defined as less than mean − 2 SD of the control value (18.4%) [31]. In each patient, cerebral hyperperfusion was defined as a postoperative CBF increase of ≥100% (i.e., a doubling) when compared to preoperative values in the MCA ROI ipsilateral to the side of surgery [20]. 4.3. Transcranial Doppler (TCD) Monitoring TCD was performed using a PIONEER TC2020 system (EME, Uberlingen, Germany; software version 2.50, 2 MHz probe; diameter, 1.5 cm; insonation depth, 40–66 mm; scale, −100 and +150 cm/s; sample volume, 2 mm; 64 point fast Fourier transform; fast Fourier transform length, 2 mm; fast Fourier transform overlap, 60%; high-pass filter, 100 Hz; detection threshold, 9 dB; minimum increase time, 10 ms) for insonation of the MCA ipsilateral to the carotid artery undergoing CEA. TCD data were stored on a hard disk using a coding system and later analyzed manually by a clinical neurophysiologist who was blinded to patient information. MES were identified during exposure of the carotid arteries (from skin incision to ICA clamping) according to the recommended guidelines [35]. 4.4. Magnetic Resonance Imaging DWI was performed using a 1.5 T whole-body imaging system (Signa MR/I; GE Healthcare, Milwaukee, WI, USA) within three days before and 24 h after surgery. A neuroradiologist who was blinded to patient clinical information analyzed the images and determined whether new ischemic lesions had developed postoperatively. 4.5. Preoperative, Intraoperative, and Postoperative Management Blood pressure was measured at the upper arm using an automatic sphygmomanometer with the oscillometric method, and mean systolic blood pressure in the morning for the three days before surgery was defined as the preoperative value for each patient. Patients received medications including antihypertensive and antidiabetic drugs and statins until the evening of the day before CEA was performed. All patients received a single antiplatelet drug until the morning of the day on which CEA was performed. For all patients, surgery was conducted under general anesthesia, which was induced with etomidate/fentanyl and maintained with O2/propofol. A bolus of heparin (5000 international units) was given prior to ICA clamping. Blood pressure was measured in the same fashion as preoperatively every 5 min throughout surgery. The EEG was recorded, and a clinical neurophysiologist monitored the recordings continuously during the surgical procedure. The presence of unilateral or bilateral decreases of alpha and beta activity during ICA clamping, with or without simultaneous increases of theta or delta activity, was defined as development of cerebral hemispheric ischemia by the clinical neurophysiologist [24]. In this situation, an intraluminal shunt was introduced. From declamping of the ICA to the third postoperative day, attempts were made to reduce systolic blood pressure to below 90% of the preoperative value using intravenous injection of the calcium antagonist nicardipine. Blood pressure was measured in the same fashion as preoperatively every 1 h until 24 h after surgery. When systolic blood pressure was <90% of the preoperative value at a blood pressure-measured point, the point was defined as having successfully controlled blood pressure. Patients received the same drugs as preoperative medications from the second postoperative day. 4.6. Statistical Analysis Data are expressed as means ± SD. The relationship between each variable and DWI-characterized postoperative cerebral ischemic lesions was evaluated by univariate analysis using the Mann-Whitney U test or the χ2 test. Hypertension was defined as preoperative systolic blood pressure ≥140 mmHg, preoperative diastolic blood pressure ≥90 mmHg or preoperatively receiving antihypertensive drugs; diabetes mellitus was defined as preoperative hemoglobin A1c ≥6.5% or preoperatively receiving antidiabetic drugs; dyslipidemia was defined as preoperative plasma low density lipoprotein (LDL) cholesterol ≥140 mg/dL, preoperative plasma high density lipoprotein (HDL) cholesterol <40 mg/dL, preoperative plasma triglyceride ≥150 mg/dL, or preoperatively receiving statins. Multivariate statistical analysis of factors related to DWI-characterized postoperative cerebral ischemic lesions was also performed using a logistic regression model. Variables with p < 0.2 on univariate analyses were selected for analysis in the final model. Differences were deemed significant for values of p < 0.05. 5. Conclusions The present study demonstrated that cerebral hyperperfusion after revascularization inhibits the development of cerebral ischemic lesions due to artery-to-artery emboli during carotid exposure in CEA for patients with preoperatively impaired cerebral hemodynamics, supporting the “impaired clearance of emboli” concept when its interpretation is broadened. Acknowledgments The author (Kuniaki Ogasawara) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Strategic Medical Science Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan, Grant Number S1491001; Scientific Research from Japan Society for the Promotion of Science, Grant Number JP15K10313. Author Contributions Kentaro Fujimoto and Kuniaki Ogasawara conceived and designed the study; Yoshiyasu Matsumoto, Kohki Oikawa and Kazunori Terasaki performed measurements and analyses of brain perfusion; Masakazu Kobayashi performed measurements and analyses of microembolic signals on transcranial Doppler; Shunrou Fujiwara statically analyzed the data; Jun-ichi Nomura, Yasuyoshi Shimada and Kenji Yoshida critically revised the manuscript and helped with results interpretation. Kentaro Fujimoto and Kuniaki Ogasawara wrote the paper. Conflicts of Interest The author (Kuniaki Ogasawara) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Consigned research fund (3,150,000 yen) from Nihon Medi-Physics Co., Ltd. Abbreviations CEA Carotid endarterectomy TCD Transcranial Doppler MCA Middle cerebral artery MES Microembolic signals DWI Diffusion-weighted imaging CVR Cerebrovascular reactivity CBF Cerebral blood flow SPECT Single-photon emission computed tomography ICA Internal carotid artery EEG Electroencephalography IMP [123I]N-isopropyl-p-iodoamphetamine ROI Region-of-interest SD Standard deviation Figure 1 Trial profile showing the flow chart of patient numbers from initial screening to final analysis. Patients who did not have preoperative reduced cerebrovascular reactivity (CVR), did not undergo carotid endarterectomy (CEA), did not have reliable intraoperative transcranial Doppler (TCD) monitoring, had hemispheric ischemia during carotid clamping, and did not have microembolic signals (MES) during carotid exposure were excluded from the study. Figure 2 Relationships between the number of microembolic signals (MES), postoperative CBF (cerebral blood flow), cerebral hyperperfusion, and the development of diffusion-weighted imaging (DWI)-characterized postoperative cerebral ischemic lesions. Closed and open circles indicate patients with and without DWI-characterized postoperative cerebral ischemic lesions, respectively. Red and black circles indicate patients with and without postoperative cerebral hyperperfusion (defined as postoperative CBF increase ≥100% compared with preoperative values), respectively. Whereas 15 (83%) of 18 patients without postoperative cerebral hyperperfusion showed DWI-characterized postoperative cerebral ischemic lesions, only one (7%) of 14 patients with hyperperfusion had these ischemic lesions. Figure 3 (A) Preoperative brain perfusion single-photon emission computed tomography in a 74-year-old man with symptomatic left internal carotid artery stenosis (90%) shows reduced cerebral blood flow (left) and reduced cerebrovascular reactivity to acetazolamide (center) in the left cerebral hemisphere where hyperperfusion develops immediately after surgery (right); (B) Transcranial Doppler recording during exposure of the carotid arteries in the patient of Figure 3A shows three microembolic signals (arrows) in the power spectrum display of left middle cerebral artery blood flow. This patient had a total of 10 microembolic signals during exposure of the carotid arteries; (C) A diffusion-weighted image 6 h after surgery in the patient of Figure 3A,B shows development of new postoperative multiple high-intensity lesions in the left cerebral hemisphere (right) when compared with a preoperative image (left). These lesions did not change on diffusion-weighted imaging 24 h after surgery. This patient suffered slight motor weakness in the right upper extremity after recovery from general anesthesia, and this deficit resolved completely within 12 h. Figure 4 Diagrams show the regions of interests (ROIs) for a three-dimensional, stereotactic ROI template to automatically place constant ROIs on brain perfusion single-photon emission computed tomography images. White ROIs indicate middle cerebral artery territories (precentral, central, parietal, angular, and temporal). ijms-17-01261-t001_Table 1Table 1 Univariate analysis of factors related to development of diffusion-weighted imaging (DWI)-characterized postoperative cerebral ischemic lesions. Variable DWI-Characterized Ischemic Lesions p Yes No (n = 16) (n = 16) Age (years, mean ± SD) 72.8 ± 5.4 70.7 ± 3.2 0.2557 Male sex 15 (94%) 14 (88%) >0.9999 Hypertension 12 (75%) 14 (88%) 0.6539    Preoperative antihypertensive drugs 10 (63%) 13 (81%) 0.4331    Preoperative calcium antagonist 5 (31%) 5 (31%) >0.9999    Preoperative angiotensin receptor blocker 8 (50%) 10 (63%) 0.7224 Diabetes mellitus 6 (38%) 7 (44%) >0.9999    Preoperative antidiabetic drugs 6 (38%) 7 (44%) >0.9999 Dyslipidemia 7 (44%) 8 (50%) >0.9999    Preoperative statins 5 (31%) 7 (44%) 0.7160    Preoperative strong statins * 2 (12%) 4 (25%) 0.6539 Preoperative aspirin 6 (38%) 4 (25%) 0.7043 Preoperative clopidogrel 10 (63%) 12 (75%) 0.7043 Ischemic heart or valvular disease 3 (19%) 4 (25%) >0.9999 Symptomatic lesion 14 (88%) 11 (69%) 0.3944 Degree of ICA stenosis (%, mean ± SD) 83.1 ± 9.0 87.7 ± 8.2 0.1258 Bilateral lesions 4 (25%) 5 (31%) >0.9999 Preoperative CBF (mL/100 g/min, mean ± SD) 31.9 ± 5.8 34.4 ± 6.4 0.2581 Preoperative CVR to acetazolamide (%, mean ± SD) 8.7 ± 5.9 8.6 ± 8.0 0.6783 Preoperative systolic blood pressure (mmHg, mean ± SD) 134.5 ± 15.8 132.5 ± 14.2 0.9254 Number of MES (mean ± SD) 4.6 ± 4.3 4.0 ± 3.6 0.8932 Interval from first MES to ICA declamping (min, mean ± SD) 46.1 ± 6.8 44.4 ± 6.1 0.4848 Interval from last MES to ICA declamping (min, mean ± SD) 47.2 ± 7.9 46.3 ± 5.2 0.8353 Mean systolic blood pressure during carotid exposure (mmHg, mean ± SD) 114.2 ± 14.8 113.4 ± 13.0 0.9849 Duration of ICA clamping (min, mean ± SD) 37.2 ± 5.6 34.8 ± 5.6 0.2191 Mean systolic blood pressure after carotid declamping (mmHg, mean ± SD) 122.2 ± 15.1 120.7 ± 13.1 0.9049 Successfully controlled blood pressure after carotid declamping ** (%, mean ± SD) 84.6 ± 5.3 87.2 ± 6.8 0.8954 Mean systolic blood pressure in postoperative period (mmHg, mean ± SD) 128.8 ± 16.8 127.1 ± 17.0 0.9241 Successfully controlled blood pressure in postoperative period *** (%, mean ± SD) 74.5 ± 6.8 78.3 ± 7.8 0.8037 Postoperative CBF (mL/100 g/min, mean ± SD) 39.6 ± 4.8 66.3 ± 15.0 <0.0001 Cerebral hyperperfusion 1 (6%) 13 (81%) <0.0001 SD, Standard deviation; ICA, Internal carotid artery; CBF, Cerebral blood flow; CVR, Cerebrovascular reactivity; MES, Microembolic signal; *, Including atorvastatin, pitavastatin, and rosuvastatin; **, Rate of blood pressure–measured points with systolic blood pressure <90% of the preoperative value in the post-carotid declamping period in surgery; ***, Rate of blood pressure–measured points with systolic blood pressure <90% of the preoperative value in the postoperative period (within 24 h after surgery). ==== Refs References 1. Caplan L.R. Hennerici M. Impaired clearance of emboli (washout) is an important link between hypoperfusion, embolism, and ischemic stroke Arch. Neurol. 1998 55 1475 1482 10.1001/archneur.55.11.1475 9823834 2. Caplan L.R. Wong K.S. Gao S. Hennerici M.G. Is hypoperfusion an important cause of strokes? If so, how? Cerebrovasc. Dis. 2006 21 145 153 10.1159/000090791 16401883 3. Wong K.S. Gao S. Chan Y.L. Hansberg T. Lam W.W. Droste D.W. Kay R. Ringelstein E.B. Mechanisms of acute cerebral infarctions in patients with middle cerebral artery stenosis: A diffusion-weighted imaging and microemboli monitoring study Ann. Neurol. 2002 52 74 81 10.1002/ana.10250 12112050 4. Schreiber S. Serdaroglu M. Schreiber F. Skalej M. Heinze H.J. Goertler M. Simultaneous occurrence and interaction of hypoperfusion and embolism in a patient with severe middle cerebral artery stenosis Stroke 2009 40 e478 e480 10.1161/STROKEAHA.109.549378 19498193 5. Orlandi G. Fanucchi S. Gallerini S. Sonnoli C. Cosottini M. Puglioli M. Sartucci F. Murri L. Impaired clearance of microemboli and cerebrovascular symptoms during carotid stenting procedures Arch. Neurol. 2005 62 1208 1211 10.1001/archneur.62.8.1208 16087759 6. Spencer M.P. Transcranial Doppler monitoring and causes of stroke from carotid endarterectomy Stroke 1997 28 685 691 10.1161/01.STR.28.4.685 9099179 7. Ackerstaff R.G. Moons K.G. van de Vlasakker C.J. Moll F.L. Vermeulen F.E. Algra A. Spencer M.P. Association of intraoperative transcranial doppler monitoring variables with stroke from carotid endarterectomy Stroke 2000 31 1817 1823 10.1161/01.STR.31.8.1817 10926940 8. Wolf O. Heider P. Heinz M. Poppert H. Sander D. Greil O. Weiss W. Hanke M. Eckstein H.H. Microembolic signals detected by transcranial Doppler sonography during carotid endarterectomy and correlation with serial diffusion-weighted imaging Stroke 2004 35 e373 e375 10.1161/01.STR.0000143184.69343.ec 15388901 9. Gavrilescu T. Babikian V.L. Cantelmo N.L. Rosales R. Pochay V. Winter M. Cerebral microembolism during carotid endarterectomy Am. J. Surg. 1995 170 159 164 10.1016/S0002-9610(99)80277-3 7631922 10. Rothwell P.M. Eliasziw M. Gutnikov S.A. Fox A.J. Taylor D.W. Mayberg M.R. Warlow C.P. Barnett H.J. Analysis of pooled data from the randomised controlled trials of endarterectomy for symptomatic carotid stenosis Lancet 2003 361 107 116 10.1016/S0140-6736(03)12228-3 12531577 11. Verhoeven B.A. de Vries J.P. Pasterkamp G. Ackerstaff R.G. Schoneveld A.H. Velema E. de Kleijn D.P. Moll F.L. Carotid atherosclerotic plaque characteristics are associated with microembolization during carotid endarterectomy and procedural outcome Stroke 2005 36 1735 1740 10.1161/01.STR.0000173153.51295.ee 16002764 12. Gaunt M.E. Martin P.J. Smith J.L. Rimmer T. Cherryman G. Ratliff D.A. Bell P.R. Naylor A.R. Clinical relevance of intraoperative embolization detected by transcranial Doppler ultrasonography during carotid endarterectomy: A prospective study of 100 patients Br. J. Surg. 1994 81 1435 1439 10.1002/bjs.1800811009 7820463 13. Jansen C. Ramos L.M. van Heesewijk J.P. Moll F.L. van Gijn J. Ackerstaff R.G. Impact of microembolism and hemodynamic changes in the brain during carotid endarterectomy Stroke 1994 25 992 997 10.1161/01.STR.25.5.992 7909393 14. Aso K. Ogasawara K. Sasaki M. Kobayashi M. Suga Y. Chida K. Otawara Y. Ogawa A. Preoperative cerebrovascular reactivity to acetazolamide measured by brain perfusion SPECT predicts development of cerebral ischemic lesions caused by microemboli during carotid endarterectomy Eur. J. Nucl. Med. Mol. Imaging 2009 36 294 301 10.1007/s00259-008-0886-y 18690436 15. Bernstein M. Fleming J.F. Deck J.H. Cerebral hyperperfusion after carotid endarterectomy: A cause of cerebral hemorrhage Neurosurgery 1984 15 50 56 10.1227/00006123-198407000-00010 6472594 16. Ogasawara K. Sakai N. Kuroiwa T. Hosoda K. Iihara K. Toyoda K. Sakai C. Nagata I. Ogawa A. Intracranial hemorrhage associated with cerebral hyperperfusion syndrome following carotid endarterectomy and carotid artery stenting: Retrospective review of 4494 patients J. Neurosurg. 2007 107 1130 1136 10.3171/JNS-07/12/1130 18077950 17. Piepgras D.G. Morgan M.K. Sundt T.M. Jr. Yanagihara T. Mussman L.M. Intracerebral hemorrhage after carotid endarterectomy J. Neurosurg. 1988 68 532 536 10.3171/jns.1988.68.4.0532 3351580 18. Sundt T.M. Jr. Sharbrough F.W. Piepgras D.G. Kearns T.P. Messick J.M. Jr. O’Fallon W.M. Correlation of cerebral blood flow and electroencephalographic changes during carotid endarterectomy, with results of surgery and hemodynamics of cerebral ischemia Mayo Clin. Proc. 1981 56 533 543 10.1097/00132586-198226040-00005 7266064 19. Hosoda K. Kawaguchi T. Shibata Y. Kamei M. Kidoguchi K. Koyama J. Fujita S. Tamaki N. Cerebral vasoreactivity and internal carotid artery flow help to identify patients at risk for hyperperfusion after carotid endarterectomy Stroke 2001 32 1567 1573 10.1161/01.STR.32.7.1567 11441203 20. Ogasawara K. Yukawa H. Kobayashi M. Mikami C. Konno H. Terasaki K. Inoue T. Ogawa A. Prediction and Monitoring of cerebral hyperperfusion after carotid endarterectomy by using single-photon emission computerized tomography scanning J. Neurosurg. 2003 99 504 510 10.3171/jns.2003.99.3.0504 12959438 21. Yoshimoto T. Houkin K. Kuroda S. Abe H. Kashiwaba T. Low cerebral blood flow and perfusion reserve induce hyperperfusion after surgical revascularization: Case reports and analysis of cerebral hemodynamics Surg. Neurol. 1997 48 132 139 10.1016/S0090-3019(97)90106-3 9242237 22. Yadav J.S. Wholey M.H. Kuntz R.E. Fayad P. Katzen B.T. Mishkel G.J. Bajwa T.K. Whitlow P. Strickman N.E. Jaff M.R. Protected carotid-artery stenting versus endarterectomy in high-risk patients N. Engl. J. Med. 2004 351 1493 1501 10.1056/NEJMoa040127 15470212 23. Van der Schaaf I.C. Horn J. Moll F.L. Ackerstaff R.G. Transcranial Doppler monitoring after carotid endarterectomy Ann. Vasc. Surg. 2005 19 19 24 10.1007/s10016-004-0146-y 15714362 24. Rutgers D.R. Blankensteijn J.D. van der Grond J. Preoperative MRA flow quantification in CEA patients: Flow differences between patients who develop cerebral ischemia and patients who do not develop cerebral ischemia during cross-clamping of the carotid artery Stroke 2000 31 3021 3028 10.1161/01.STR.31.12.3021 11108766 25. Grubb R.L. Jr. Derdeyn C.P. Fritsch S.M. Carpenter D.A. Yundt K.D. Videen T.O. Spitznagel E.L. Powers W.J. Importance of hemodynamic factors in the prognosis of symptomatic carotid occlusion JAMA 1998 280 1055 1060 10.1001/jama.280.12.1055 9757852 26. Kuroda S. Houkin K. Kamiyama H. Mitsumori K. Iwasaki Y. Abe H. Long-term prognosis of medically treated patients with internal carotid or middle cerebral artery occlusion: Can acetazolamide test predict it? Stroke 2001 32 2110 2116 10.1161/hs0901.095692 11546904 27. Schmiedek P. Piepgras A. Leinsinger G. Kirsch C.M. Einhüpl K. Improvement of cerebrovascular reserve capacity by EC-IC arterial bypass surgery in patients with ICA occlusion and hemodynamic cerebral ischemia J. Neurosurg. 1994 81 236 244 10.3171/jns.1994.81.2.0236 8027807 28. Ogasawara K. Konno H. Yukawa H. Endo H. Inoue T. Ogawa A. Transcranial regional cerebral oxygen saturation monitoring during carotid endarterectomy as a predictor of postoperative hyperperfusion Neurosurgery 2003 53 309 315 10.1227/01.NEU.0000073547.86747.F3 12925245 29. Ogasawara K. Inoue T. Kobayashi M. Endo H. Yoshida K. Fukuda T. Terasaki K. Ogawa A. Cerebral hyperperfusion following carotid endarterectomy: Diagnostic utility of intraoperative transcranial Doppler ultrasonography compared with single-photon emission computed tomography study AJNR Am. J. Neuroradiol. 2005 26 252 257 15709121 30. North American Symptomatic Carotid Endarterectomy Trial Collaborators Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis N. Engl. J. Med. 1991 325 445 453 1852179 31. Ogasawara K. Ito H. Sasoh M. Okuguchi T. Kobayashi M. Yukawa H. Terasaki K. Ogawa A. Quantitative measurement of regional cerebrovascular reactivity to acetazolamide using 123 I-N -isopropyl-p -iodoamphetamine autoradiography with SPECT: Validation study using H2 15 O with PET J. Nucl. Med. 2003 44 520 525 12679394 32. Iida H. Itoh H. Nakazawa M. Hatazawa J. Nishimura H. Onishi Y. Uemura K. Quantitative mapping of regional cerebral blood flow using iodine-123-IMP and SPECT J. Nucl. Med. 1994 35 2019 2030 7989987 33. Nishimiya M. Matsuda H. Imabayashi E. Kuji I. Sato N. Comparison of SPM and NEUROSTAT in voxelwise statistical analysis of brain SPECT and MRI at the early stage of Alzheimer’s disease Ann. Nucl. Med. 2008 22 921 927 10.1007/s12149-008-0211-7 19142712 34. Takeuchi R. Matsuda H. Yoshioka K. Yonekura Y. Cerebral blood flow SPET in transient global amnesia with automated ROI analysis by 3DSRT Eur. J. Nucl. Med. Mol. Imaging 2004 31 578 589 10.1007/s00259-003-1406-8 14722677 35. Consensus Committee of the Ninth International Cerebral Hemodynamic Symposium Basic identification criteria of Doppler microembolic signals Stroke 1995 26 1123 10.1161/01.STR.26.6.1123 7762033

PMC005xxxxxx/PMC5000660.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081262ijms-17-01262ArticleElectrospun Poly(l-lactide)/Poly(ethylene glycol) Scaffolds Seeded with Human Amniotic Mesenchymal Stem Cells for Urethral Epithelium Repair Lv Xiaokui Guo Qianping Han Fengxuan Chen Chunyang Ling Christopher Chen Weiguo *Li Bin *Hardy John G. Academic EditorDepartments of Urology and Orthopaedic Surgery, The First Affiliated Hospital, Orthopaedic Institute, Soochow University, 708 Renmin Rd., Suzhou 215007, China; [email protected] (X.L.); [email protected] (Q.G.); [email protected] (F.H.); [email protected] (C.C.); [email protected] (C.L.)* Correspondence: [email protected] (W.C.); [email protected] (B.L.); Tel.: +86-512-6778-0131 (W.C.); +86-512-6778-1163 (B.L.)09 8 2016 8 2016 17 8 126215 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Tissue engineering-based urethral replacement holds potential for repairing large segmental urethral defects, which remains a great challenge at present. This study aims to explore the potential of combining biodegradable poly(l-lactide) (PLLA)/poly(ethylene glycol) (PEG) scaffolds and human amniotic mesenchymal cells (hAMSCs) for repairing urethral defects. PLLA/PEG fibrous scaffolds with various PEG fractions were fabricated via electrospinning. The scaffolds were then seeded with hAMSCs prior to implantation in New Zealand male rabbits that had 2.0 cm-long defects in the urethras. The rabbits were randomly divided into three groups. In group A, hAMSCs were grown on PLLA/PEG scaffolds for two days and then implanted to the urethral defects. In group B, only the PLLA/PEG scaffolds were used to rebuild the rabbit urethral defect. In group C, the urethral defect was reconstructed using a regular urethral reparation technique. The repair efficacy was compared among the three groups by examining the urethral morphology, tissue reconstruction, luminal patency, and complication incidence (including calculus formation, urinary fistula, and urethral stricture) using histological evaluation and urethral radiography methods. Findings from this study indicate that hAMSCs-loaded PLLA/PEG scaffolds resulted in the best urethral defect repair in rabbits, which predicts the promising application of a tissue engineering approach for urethral repair. human amniotic mesenchymal cellselectrospinningurethral defecttissue engineeringpoly(l-lactide)poly(ethylene glycol) ==== Body 1. Introduction Urethral defects caused by urethral trauma, congenital malformation, and tumor are common causes of urological surgeries [1]. While regular defects can be repaired using end-to-end anastomosis or buccal mucosa tissues of patient, repairing larger segmental urethral defects (longer than 2 cm) remains challenging [2,3,4]. Encouraged by the success of a few recent clinical studies, tissue engineering-based urethral reconstruction is believed to be a promising treatment for urethral defect repair [5,6,7,8,9]. Three major approaches which involve cells alone, scaffolds alone, and cell-seeded scaffolds, respectively, have been used in urethral tissue engineering. Among them, cell-seeded scaffolds may lead to the most effective urethral regeneration [2,10,11]. To date, a variety of scaffold materials have been used for urethral tissue engineering. These include: (1) natural materials such as collagen, hyaluronic acid, and alginate; (2) biological matrices derived from decellularized tissues from the foreskin, bladder mucosa, small intestinal submucosa, and tunica vaginalis; and (3) synthetic polymers such as polyglycolic acid (PGA), polylactide (PLA), poly(l-lactide-co-glycolide) (PLGA), and polycaprolactone (PCL). Among them, natural materials are often of insufficient strength to support urethral repair [12]. Decellularized tissues, while containing cell signaling molecules which may favor tissue regeneration, suffer from source shortages and high complication rates [3,4,13]. Synthetic biodegradable polymers, on the other hand, have been widely used in tissue engineering [14,15]. A commonly used polymer, poly(l-lactide) (PLLA) possesses good mechanical properties and excellent biocompatibility. In order to achieve moderate hydrophilicity [16,17], PLLA can be blended with hydrophilic polymers such as poly(ethylene glycol) (PEG). Such a composite scaffold has been shown to possess adequate hydrophilicity to support cell adhesion and proliferation [18]. Cells are another critical component for urethral tissue engineering. Being the most important component of the urinary tract, epithelial cells have been used for urethral regeneration. However, the sources of urethral epithelial cells are limited. Harvesting these tissues involves a complicated operation to the genitourinary tract [2]. In addition, the proliferation of urothelial cells is limited. In contrast, mesenchymal stem cells (MSCs) which have multi-potential differentiation and self-renewal ability, including bone marrow mesenchymal stem cells (BMSCs) and adipose-derived mesenchymal stem cells (ADMSCs), are believed to be ideal cell sources for urethral tissue engineering [19,20,21]. However, the applications of such autologous adult MSCs are also limited due to the painful and invasive harvest procedure, limited number, and loss of stemness during in vitro expansion [22]. Recently, human amniotic mesenchymal stem cells (hAMSCs), which are derived from the amniotic membrane, have drawn much attention [23]. Similar to BMSCs and ADMSCs, hAMSCs can be induced into many lineages such as adipocytes, osteocytes, chondrocytes, and endothelial cells [24,25]. In addition, hAMSCs have strong proliferation capacity in vitro and are immunologically tolerant [26]. A major advantage of hAMSCs is their ready availability, which eliminates the invasive procedures and ethical concerns of cell harvesting [23]. Therefore, hAMSCs may be a good candidate cell source in urethral tissue engineering. In this study, we prepared fibrous PLLA/PEG composite scaffolds via electrospinning, a technique which enables the fabrication of highly porous structures with the diameter of fibers ranging from a few hundred nanometers to several microns [27,28]. We then cultured hAMSCs on the scaffolds to form cells-scaffold constructs in vitro, which we hypothesized would facilitate the repair of urethral defects. Following that, the cells-scaffold constructs were implanted to the urethral defects of rabbits to examine their repair capability. After up to three months from implantation, the urethral morphology, tissue reconstruction, luminal patency, and complication incidence of animals were checked using histological evaluation and urethral radiography approaches. 2. Results 2.1. Fabrication of Poly(l-lactide)/Poly(ethylene glycol) (PLLA/PEG) Scaffolds Fibrous scaffolds of PLLA/PEG composites were fabricated using electrospinning technique. As can be seen from the SEM images, the fiber diameter decreased with an increase of PEG fraction in the composites (Figure 1). The average fiber diameters of PLLA, PEG10, PEG20, PEG30, PEG40, and PEG50 samples were 1.6 ± 0.21, 1.5 ± 0.41, 1.48 ± 0.52, 1.3 ± 0.38, 0.9 ± 0.3, and 0.5 ± 0.24 μm, respectively. The scaffolds did not show apparent deformation upon immersion in cell culture medium, indicating that they had good dimensional stability. The wettability of PLLA/PEG scaffolds was determined using water contact angle measurement. Apparently, the hydrophilicity of the membrane was improved by increasing PEG content. The average water contact angle of PLLA, PEG10, PEG20, PEG30, PEG40, and PEG50 were 130.3 ± 2.1°, 124.8 ± 2.5°, 111.2 ± 3.1°, 76.6 ± 4.9°, 60.3 ± 2.8°, and 0°, respectively (Figure 1). When the PEG fraction reached 30%, the membrane became hydrophilic (i.e., water contact angle < 90°). Further increasing the PEG fraction to 50% resulted in the formation of super-hydrophilic membranes (i.e., water contact angle = 0°). The mechanical properties of electrospun PLLA/PEG scaffolds were determined using tensile tests (Figure 2). Among all the composite scaffolds, the PEG10 sample showed the best mechanical properties. Increasing the PEG fraction in the composites resulted in deterioration of their mechanical characteristics. For example, the tensile strength, Young’s modulus, and elongation of PEG10 samples were 5.3 ± 0.19 MPa, 132.8 ± 1.56 MPa, and 126.0% ± 14.26% respectively. However, the tensile strength, Young’s modulus, and elongation of PEG50 samples dropped to 1.4 ± 0.10 MPa, 31.4 ± 3.12 MPa, and 4.0% ± 1.16%, respectively. 2.2. Isolation and Characterizations of Human Amniotic Mesenchymal Cells (hAMSCs) hAMSCs were isolated using a combined trypsin-collagenase method and cultured in Dulbecco’s Modified Eagle’s medium (DMEM) medium supplemented with 10% fetal bovine serum (FBS). After 24 h, hAMSCs adhered to the plate and started to proliferate. The morphology of the hAMSCs was fibroblast-like, polygonal, or round after 48 h. Non-adhered cells were removed by changing the medium. Cells formed colonies 7 days later and reached 80%–85% confluence after 14 days. The cells were passaged every 3–4 days after P0 and could be passaged up to P15 without apparent morphological change. After passage, the cell morphology was fibroblast-like (Figure 3A). In order to identify the hAMSCs, cells at passage 3 were assayed by immunofluorescence and flow cytometry. The cells were stained positive for stem cell markers such as Oct-4 and nucleostemin (Figure 3B,C). In flow cytometry analysis, the cells showed expression of classic MSC surface markers CD29, CD90 and CD105. In addition, they also slightly expressed CD45 (Figure 3D). The multi-differentiation potential of hAMSCs was checked in vitro for adipogenesis, osteogenesis, and chondrogenesis. The results showed that hAMSCs were positive by Oil Red O staining, indicating that they secreted oil after adipogenic induction (Figure 3E). After osteogenic differentiation, the cells were then treated with Alizarin Red S solution to test for calcium deposits. The positive and strong staining of Alizarin Red S indicated osteogenesis (Figure 3F). The production of sulfated proteoglycan, as shown by Saffranin O staining solution, confirmed that the hAMSCs had undergone chondrogenesis (Figure 3G). 2.3. Biocompatibility of PLLA/PEG Scaffolds In order to test the cytotoxicity of PLLA/PEG scaffolds, the proliferation of hAMSCs on the scaffolds was examined using Cell Counting Kit-8 (CCK-8) assays. The cells were cultured on a plate and a PLLA scaffold as the control. As seen, the PLLA/PEG scaffolds had little cytotoxicity and well sustained hAMSC proliferation (Figure 4A). Compared to pure PLLA, hAMSCs appeared to proliferate faster on PLLA/PEG scaffolds. In addition, the SEM images showed that many fibroblast-like hAMSCs spread on the scaffolds with some of the pseudopodia approaching the inner part of scaffolds, again indicating the biocompatibility of PLLA/PEG scaffolds (Figure 4B). 2.4. Urethrography Analysis and Morphological Observation All rabbits had a retrograde urethrogram (RUG) before the surgery (Figure 5A) and showed normal morphology. After the operations, the RUG images indicated that strictures occurred in both group B and group C (Figure 5B,C). However, no signs of stricture and fistula were found in the group A at 12 weeks post-operation (Figure 5D). As shown in Table 1, the incidence of complications in the control group (group C) was 72.22% (13/18). This was significantly higher than the incidence of complications in group A (0%) and group B (5.55%). However, there was no statistically significant difference between the complication incidences of group A and group B. The urethral specimens were harvested and observed 4 weeks after the surgeries (Figure 5E–G). In group C, non-absorbable 4-0 vicryl marking sutures could be found. The urethral defect surface was uneven and covered by a scar. Mucous membrane contraction, luminal stricture, bladder crystallization and a large amount of the urethra were visible (Figure 5E). In group B, a part of the PLLA/PEG scaffold was exposed to the inside of the urethra lumen without a mucous membrane. Compared to group C, the luminal stricture in group B was relieved; however, scar formation, urethra and bladder crystallization was still obvious. In contrast, the implant in group A was covered with a mucous membrane, and no apparent scar or crystallization was observed (Figure 5G). 2.5. Histological Evaluation The histological analysis of the urethral specimens at 4, 8, and 12 weeks post-operation is shown in Figure 6. In group B, approximately half of the PLLA/PEG scaffolds degraded after 4 weeks. Some vascellum, collagen tissue, and lymphocytes formed, while no epithelial cells were present until 12 weeks (Figure 6A–C). Compared to group B, PLLA/PEG scaffolds in group A partially degraded and a layer of epithelial cells was observed on the surface of scaffolds. Moreover, the cellular layer increased over time. A multilayered urothelium was present in group A; vascellum, smooth muscle, and fibrous tissues were found to be arranged along the PLLA/PEG fibers (Figure 6D–F). In contrast, there were many lymphocytes invading the defected tissue, and a few vascellums were found 4 weeks after the surgery in group C. The number of cells increased in the defect region, but the cells were in an unorganized formation. In addition, many fibrous tissues containing a few smooth muscle fibers formed after 6 weeks. The urethral mucosa was still discontinuous, and some lymphocytes and fibrous tissue existed in the defect tissue after 8 weeks (Figure 6G–I). Further, the urethral epithelial cells were identified using a mouse anti-human cytokeratins AE1/AE3 monoclonal antibody. The results were negative in group B, indicating that no epithelial layer formed (Figure 7A–C). However, group A was positive for AE1/AE3 staining, indicating the presence of multilayered urothelium in the PLLA/PEG scaffold (Figure 7D–F). In the control group C, the thickness of the epithelial cell layer increased gradually from 4 to 12 weeks, and the cell layers became more and more uniform (Figure 7G–I). 3. Discussion Repair of long segmental urethral defects remains technically challenging. In this study, a tissue-engineering approach which combined the use of electrospun fibrous PLLA/PEG scaffolds and hAMSCs was explored for urethral defect repair in rabbits. A widely used technology for fabricating nano-/microfibrous scaffolds, electrospinning possesses a number of advantages [27,29]. First, the electrospun scaffolds have high porosity and complete interconnectivity, which is necessary for cell migration, nutrient diffusion, and vascellum in-growth. Second, the nano-/microfibrous structure of scaffolds well mimics the extracellular matrix (ECM) structure of native tissues and favors cell adhesion, proliferation, and differentiation. In a previous study, electrospun silk fibroin scaffolds seeded with urothelial cells were applied to the dorsal urethral mucosa defect of beagles. After up to 6 months, gradual epithelial cell development and stratified epithelial layers were clearly seen in the scaffolds, implying the potential of using electrospun scaffolds for urethral reconstruction [30]. In order to identify the optimal scaffold for urethral tissue engineering, PLLA/PEG scaffolds with different PEG fractions were prepared. The diameter of fibers in the scaffolds was in the range of 500–1500 nm, and it decreased with the increase of PEG content. Since PEG has lower molecular weight compared to PLLA, it may act as a lubricant in the PLLA/PEG mixed solution [18]. Therefore, higher PEG content contributed to thinner fibers as shown in Figure 2. Moreover, the hydrophilicity/hydrophobicity balance of scaffolds is very important for protein adsorption [17,31]. It has been reported that biomaterials with moderate hydrophilicity had better biocompatibility [32,33]. The introduction of PEG improved the hydrophilicity of PLLA scaffolds and consequently led to the faster proliferation of hAMSCs on PLLA/PEG scaffolds (Figure 4). In addition, the mechanical property of scaffolds plays a critical role in tissue engineering. The strength at rupture, Young’s modulus, and elongation of normal rabbit urethra are 0.2 ± 0.07, 1.98 ± 0.73, and 173.67 ± 50.67 MPa, respectively [34]. The addition of PEG dramatically deteriorated the mechanical properties of PLLA/PEG scaffold. Taking into consideration of both the hydrophilicity and mechanical properties of scaffolds, the PEG30 scaffold in which a mixture of PLLA and 30% PEG was used for electrospinning was chosen as the scaffolds for the following in vitro and in vivo studies. As shown in Figure 4, hAMSCs adhered and proliferated well on the PLLA/PEG scaffold (PEG30), indicating the possibility of using them for urethral reconstruction in vivo. In the in vivo tests, the hAMSCs–PLLA/PEG constructs were implanted into the urethral defects of rabbits. Clearly, the incidences of urethral stricture, urinary fistula, and complications associated with implantation using hAMSCs-seeded scaffolds (group A) were markedly lower than the two control groups (groups B and C). In addition, no urethral stricture was observed when hAMSCs-seeded scaffolds were implanted. The histological results also show that in group A, epithelial cells covered the defect gradually and formed multi-layer mucosa membranes which were similar to that of normal urethral tissue after 12 weeks. Further, immuofluoresence analysis revealed that the specimens implanted with hAMSCs-seeded scaffolds were positive in pan-cytokeratin AE1/AE3 staining, indicating the presence of urethral epithelial cells. Since urethral epithelial cells played an important role in protecting the underlying muscle tissues from the caustic properties of urine, urethral epithelium regeneration is critical for urethra reconstruction [35]. In this study, the urethral epithelial cells attached and formed multiple layers on hAMSCs–PLLA/PEG scaffold. In contrast, no urethral epithelium was seen in the specimens from the group using PLLA/PEG scaffold alone (group B). Together, these results indicate that PLLA/PEG scaffold supported the growth of epithelial cells, and the presence of hAMSCs promoted the regeneration of urethral epithelium. 4. Materials and Methods 4.1. Materials Twenty-seven male New Zealand white rabbits (3 month-old, 2.5–3.0 kg) were used in the experiment. Fetal bovine serum (FBS), 0.25% penicillin-streptomycin, and 0.05% trypsin contained 0.53 mM ethylenediaminetetraacetic acid (EDTA) were all purchased from Gibco (Grand Island, NY, USA). Dulbecco’s Modified Eagle’s medium (DMEM) was supplied by Hyclone (Logan City, UT, USA). PLLA (M¯w = 50 kDa) and PEG were provided by Daigang Biotechnology Company (Jinan, China) and Yarebio (Shanghai, China), respectively. The antibodies which included anti-CD29, CD45, CD90 and CD105 were purchased from Neomarkers (Thermo Fisher Lab Vision, Fremont, CA, USA). 4.2. Preparation of PLLA/PEG Scaffolds Using Electrospinning To prepare the PLLA/PEG fibers, a mixture of PLLA and PEG with various PEG fractions (0%, 10%, 20%, 30%, 40%, and 50%) was dissolved in chloroform/dimethylformamide (v/v = 7:3) to obtain an electrospinning solution. Five milliliters of the electrospun solution was fed into a plastic syringe fitted with a stainless-steel blunt needle of 0.21 mm in diameter and placed in the electrospinning machine. The solution was electrospun at a 0.5 mL/h flow rate, a voltage of 16.9 kV and the distance between the collection and capillary tip was 15–25 cm. The obtained fibers were recorded as PLLA, PEG10, PEG20, PEG30, PEG40, and PEG50, respectively. The morphology of the electrospun fibers was characterized by scanning electron microscope (SEM, S-4800, Hitachi, Tokyo, Japan) and the fiber diameter was measured from the SEM images. The hydrophilicity of the membrane was characterized via water contact angle measurement. For further applications in urethral tissue engineering, the mechanical properties of the scaffolds were tested in this experiment. The membrane with a thickness in the range of 0.2–0.4 mm was cut into 15 × 3 mm rectangles and the tensile tests of the electrospun membrane were performed using a tensile tester. The stress-strain curves of the samples were obtained from the load-deformation curves recorded at a cross-head speed of 5 mm/min. 4.3. Isolation of hAMSCs Fresh amniotic membrane was obtained from a maternal donor at the First Affiliated Hospital of Soochow University, Suzhou, China. Informed consent was given by the participant and the procedure for hAMSCs isolation was approved by the ethics committee of the hospital. The tissue was tested to be negative for Hepatitis A, Hepatitis B, HCV, HIV, syphilis, and influenza. After the blood was washed away by phosphate buffer solution (PBS), the amniotic membrane was separated from the chorion. hAMSCs were then isolated from the digesting amniotic membrane with combined trypsin-collagenase method and cultured in DMEM supplemented with 10% FBS. 4.4. Characterizations of hAMSCs The cell phenotype was characterized by differentiation and immunofluorescence measurements. hAMSCs at passage 0–3 were used for immunofluorescence. Cells were washed three times with PBS and fixed in cold 4% paraformaldehyde for 15 min. Following that, the cells were washed twice with PBS and treated with methanol at −20 °C for 5 min. For Oct-4 immunofluorescence, the fixed cells were washed with PBS three times and blocked with 4% BSA for 30 min before being incubated with mouse anti-human Oct-4 antibody (Cat. # MAB4401, Millipore, Billerica, MA, USA; 1:500 dilution in PBS) overnight at 4 °C. After being washed with PBS, the cells were incubated with Cy3-conjugated goat anti-mouse secondary antibody (A10521, Invitrogen, Carlsbad, CA, USA; 1:1000) for 1 h at room temperature. DNA was visualized by DAPI staining and cells were viewed under a fluorescence inverted microscope (EVOS f1, AMG, Carlsbad, CA, USA USA). For nucleostemin staining, goat anti-human nucleostemin antibody (GT15050, Neuromics, Edina, MN, USA; 1:250) along with Cy3-conjugated donkey anti-goat IgG secondary antibody (AP180C, Millipore; 1:1000) were used. hAMSCs at passage 0–3 were used for flow cytometry. The cells were digested by trypsin and collected. After washed with PBS three times, 1 × 106 cells were incubated with the antibody for 30 min and then washed twice with PBS. Analysis was performed after blending. The differentiation potentials of the hAMSCs were tested in vitro by adipogenic, osteogenic, and chondrogenic differentiation assays. Passage 2–4 hAMSCs were seeded at a density of 4 × 104 cells/well in a 24-well plate in a basic culture medium (DMEM supplemented with 10% FBS and 100 U/mL penicillin). For adipogenesis, cells were induced in an adipogenic medium consisting of basic culture medium supplemented with 0.1 μM dexamethasome (D4902, Sigma, Saint Louis, MO, USA), 1 mM isobutylmethylxanthine (IBMX) (I7018, Sigma), 10 ng/L insulin and 60 mM indomethacin (I7378, Sigma) after reaching full confluence at 2 days. For osteogenesis, cells were induced in an osteogenic medium consisting of a basic culture medium supplemented with 0.1 μM dexamethasone, 0.05 mM ascorbic acid 2-phospate (A8960, Sigma), and 10 mM β-glycerolphosphate (G8981, Sigma) when cells reached 80% confluence. For chondrogenesis, cells were induced in a chondrogenic medium consisting of a basic culture medium supplemented with 40 μg/mL proline (P5607, Sigma), 39 ng/mL dexamethasone, 10 ng/mL TGF-β3 (T5425, Sigma), 50 μg/mL ascorbate 2-phosphate, 100 μg/mL sodium pyruvate (P8574, Sigma), and 50 mg/ml insulin-transferrin-selenious acid mix (ITS) (I1884, Sigma) when the cells were 80% confluent. In the control groups, all the cells were cultured in a basic culture medium. The cell media was changed every three days. Oil Red O, Alizarin Red S and Safranin O staining were used for adipogenesis, osteogenesis and chondrogenesis assays, respectively. 4.5. Culture of hAMSCs on PLLA/PEG Scaffolds MC3T3-E1 mouse preosteoblasts (CRL-2594, subclone 14, ATCC, Rockville, MD, USA) were cultured in α-mininum essential medium (α-MEM, Gibco) with 10% fetal bovine serum (Gibco) and 1% penicillin/streptomycin under 37 °C, 5% CO2 environment. Cells were seeded onto different experimental substrates. The plates with dimensions 5.8 mm were placed on 96-well polystyrene plates. The concentration of the cells initially seeded onto the specimen substrate was 1 × 104 cells/well. The 13 mm plates were placed on 24-well polystyrene plates and the concentration was 3 × 104 cells/well. The 31 mm plates were placed on 6-well polystyrene plates and the concentration was 3 × 105 cells/well. In the osteogenic differentiation assay, after the cells were cultured for 24 h in the medium described earlier, 10 mM β-glycerol phosphate, 50 μg/mL ascorbic acid, and 10 nM dexamethasone were added for osteogenic induction. The media were refreshed every three days. 4.6. Cell Morphology PLLA/PEG scaffolds (PEG30) were fixed to 96-well plates or 12-well plates. The scaffolds were sterilized by 60Co and immersed in PBS for 10 min before cell seeding. In the cell proliferation assay, hAMSCs were seeded at a density of 1 × 104 cells/well to a 96-well plate and incubated at 37 °C with 5% CO2. After culturing for 1, 3, 5, and 7 days, the absorbance was measured according to the CCK-8 kit protocol. For SEM observation, hAMSCs were seeded at a density of 5 × 104 cells/well in a 12-well plate and cultured for 24 h. The cell-scaffold composite was fixed in 4% glutaraldehyde for 4 h and then freeze-dried. The cell morphology was examined via SEM (S-4800, Hitachi). 4.7. Preparation of hAMSC-PLLA/PEG Constructs for Implantation PLLA/PEG scaffolds (PEG30) were cut into 2 × 1.5 cm pieces and sterilized with 60Co at a dosage of 12–15 kGy. The pieces were immersed in DMEM medium for 24 h and then transferred to FBS for 12 h. The hAMSCs (passage 3) were seeded at a density of 3–5 × 106 cells/cm2 to the scaffold and cultured in DMEM containing 10% FBS at 37 °C with 5% CO2 for 48 h to obtain the hAMSC-PLLA/PEG constructs for implantation. 4.8. Animal Surgeries Twenty-seven rabbits were divided into 3 groups (n = 9). In Group A, the artificial urethral defects were covered with hAMSCs–PLLA/PEG scaffolds. In Group B, the defects were covered with PLLA/PEG scaffolds. Group C served as a control group in which the animals were subjected to sham surgeries. The rabbits (~3 kg) from all groups were anesthetized using urethane. Through a ventral longitudinal penile-skin incision, a dorsal urethral mucosa segment of about 2 × 1.5 cm which was about 0.5 cm from the external urethral orifice was excised. In group A or B, the hAMSCs–PLLA/PE scaffold or PLLA/PEG scaffold was trimmed and placed over the urethral defect. The urethral repair was continuously sutured with 7-0 vicryl suture. Multiple nonabsorbable 4-0 vicryl sutures were stitched at the four corners of the scaffold and were used as the markers for removing the tissue. In group C, the incision was closed immediately after the corpora cavernosa was blunt dissected. A catheter was inserted into the bladder and sewed with absorbable 5-0 vicryl sutures for 3–5 days. After operation, the rabbits were treated with gentamicin for 5–7 days. The animal surgery protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Soochow University. 4.9. Postoperative Evaluations Three rabbits were sacrificed at 4, 8, and 12 weeks post-operation. Retrograde urethrograms were performed before and after the sacrifices to observe the situation of urethra. Mucosa healing and calculus formation were observed after the urethral incision. Histological examinations were tested using hematoxylin–eosin (H&E) staining and immunohistochemistry. 4.10. H&E Staining Specimens were fixed in a 10% formalin solution and dehydrated in an ascending series of ethanol and embedded in paraffin. Sections which were 6 μm thick were prepared using a microtome and mounted on subbed glass slides. Slides containing four randomly selected sections from each implant were dipped in hematoxylin for 6 min and then in eosin for 1 min. The stained slides were observed using an inverted phase contrast microscope (Axiovert 200, Carl Zeiss, Oberkochen, Germany). 4.11. Immunohistochemistry The deparaffinized sections of urethra were thoroughly washed and non-specific endogenous peroxidase activity was quenched by immersing in 3% H2O2/methanol for 15 min. After being washed three times with PBS, the sections on the slides were incubated overnight with a mouse monoclonal pan-Cytokeratin antibody (AE1/AE3) (sc-81714, Santa Cruz, Santa Cruz, CA, USA; 1:500 dilution in PBS) at 4 °C. Slides incubated with isotype matched control antibodies or PBS without primary antibodies were used as negative controls. The slides were then washed with PBS and further incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (H+L) antibody (AB503, Novoprotein, Shanghai, China; 1:500 dilution in PBS) at 37 °C for 30 min. Finally, the HRP substrate was applied for visualization. 4.12. Statistics Analysis All quantitative data are presented as mean ± standard deviation with no less than three replicates for each experimental condition. Statistical analyses were performed by SPSS software. Kruskal-Wallis one-way analysis of variance (ANOVA) tests followed by Tukey post hoc tests were used. Unpaired Student’s t-tests were also used where appropriate. Difference between the groups is considered statistically significant if p is less than 0.05. 5. Conclusions In summary, PLLA/PEG fibrous scaffolds with various PEG contents have been fabricated via electrospinning in this study. The hAMSCs adhered and proliferated well on the scaffolds. After being seeded with hAMSCs, the scaffolds were implanted in the long segmental urethral defects of rabbits. Based on the results from urethral morphology, tissue reconstruction, luminal patency, and complication incidence (including stone formation, urinary fistula, and urethral stricture) among the animals, it is clear that PLLA/PEG scaffolds combined with hAMSCs led to the best repair of urethral defects. Moreover, histological evaluations show that while the cell-free scaffolds prevented re-epithelialization, hAMSCs facilitated re-epithelialization over the scaffolds. Findings from this study indicate that tissue engineering is a promising approach for urethral regeneration. Certainly, there are also limitations in this study. For example, the mechanisms of hAMSCs-based urethral regeneration remain to be elucidated. In addition, while this study examined the short-term effects of using the combination of hAMSCs and PLLA/PEG nanofibrous scaffolds on urethral repair, long-term studies should be performed following studies. Comparison of the effect of hAMSCs and other types of MSCs on urethral repair is also an area of interest. Acknowledgments This study was supported by the National Natural Science Foundation of China (81471790, 31400826), Jiangsu Provincial Special Program of Medical Science (BL2012004), Jiangsu Provincial Clinical Orthopedic Center, and the Priority Academic Program Development (PAPD) of Jiangsu Higher Education Institutions. Author Contributions Weiguo Chen and Bin Li conceived and designed the experiments; Xiaokui Lv and Qianping Guo performed the experiments; Xiaokui Lv and Fengxuan Han analyzed the data; Xiaokui Lv, Chunyang Chen, Christopher Ling and Bin Li wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Abbreviations ADMSCs adipose-derived mesenchymal stem cells α-MEM α-Mininum essential medium ANOVA analysis of variance BMSCs bone marrow mesenchymal stem cells CCK-8 Cell Counting Kit-8 DMEM Dulbecco’s Modified Eagle’s medium ECM extracellular matrix EDTA ethylenediaminetetraacetic acid FBS fetal bovine serum IACUC Institutional Animal Care and Use Committee IBMX isobutylmethylxanthine ITS insulin-transferrin-selenious acid mix hAMSCs human amniotic mesenchymal cells H&E hematoxylin-eosin HRP horseradish peroxidase MSCs mesenchymal stem cells PCL polycaprolactone PEG poly(ethylene glycol) PGA polyglycolic acid PLA polylactide PLGA poly(l-lactide-co-glycolide) PLLA poly(l-lactide) RUG retrograde urethrogram SEM scanning electron microscopy Figure 1 SEM and water contact angle measurements of electrospun poly(l-lactide)/poly(ethylene glycol) (PLLA/PEG) scaffolds with PEG fractions of 0% (A); 10% (B); 20% (C); 30% (D); 40% (E); and 50% (F), respectively. Both the fiber diameter and water contact angle decreased with the increase of PEG content. Scale bars, 10 μm. Figure 2 Stress-strain curves of electrospun PLLA/PEG scaffolds with various PEG fractions. Figure 3 Identification of human amniotic mesenchymal cells (hAMSCs). (A) The fibroblast-like morphology of hAMSCs at P3; (B,C) hAMSCs were positive for stem cell markers Oct-4; (B) and SSEA-4 (C); (D) Immunophenotypical characterization of hAMSCs. Cells at the 5th culture passage were trypsinized, labeled with antibodies against the antigens indicated and analyzed by flow cytometry. hAMSCs expressed CD29, CD90 and CD105, but did not express CD45. Black, isotype control; red, antibody; (E–G) Multi-lineage differentiations of hAMSCs in vitro. hAMSCs were stained with Oil Red O after being induced in adipogenic differentiation medium for 2 weeks (E). Small colonies with lipid secretion were clearly seen. hAMSCs were stained with Alizarin red S after being induced in osteogenic differentiation medium for 3 weeks (F). hAMSCs were stained Safranin O after being induced in chondrogenic differentiation medium for 3 weeks (G). Scale bars, (A–C) 200 µm; (E–G) 50 µm. Figure 4 Proliferation (A) and SEM images (B) of hAMSCs on electrospun PLLA/PEG scaffolds. * p < 0.05. Scale bar, 100 μm. Figure 5 Implantation of electrospun PLLA/PEG scaffolds and human amniotic mesenchymal cells (hAMSCs) in urethral defects of rabbits. (A) A urethral mucosa defect of about 2 × 1.5 cm was formed in a rabbit; (B–D) Retrograde urethrograms of the rabbits that were subjected to a mock operation (B), implanted with PLLA/PEG scaffold (C); and hAMSCs–PLLA/PEG construct (D); respectively; (E–G) Gross observation of mucosa healing and calculus formation in the rabbits that were subjected to a mock operation (E); implanted with PLLA/PEG scaffold (F); and hAMSCs–PLLA/PEG construct (G); respectively. Figure 6 Hematoxylin–eosin (H&E) staining of the urethral tissues at 4, 8, and 12 weeks post-operation. (A–C) implanted with PLLA/PEG scaffolds; (D–F) implanted with hAMSCs–PLLA/PEG constructs; (G–I) given mock operation. Scale bars, 100 μm. Figure 7 Immunohistochemical evaluation of urethral tissues by AE1/AE3 staining at 4, 8, and 12 weeks post-operation. (A–C) implanted with PLLA/PEG scaffolds; (D–F) implanted with hAMSCs–PLLA/PEG constructs; (G–I) given mock operation. Scale bars, 100 μm. ijms-17-01262-t001_Table 1Table 1 The incidences of post-operation complications including urethral stricture and urinary fistula in rabbits. Group Number of Urethral Stricture and Urinary Fistula Total Number Incidence (%) + − A 0 18 18 0.0 B 1 17 18 5.6 C 13 5 18 72.2 ==== Refs References 1. Marrocco G. Vallasciani S. Fiocca G. Calisti A. Hypospadias surgery: A 10-year review Pediatr. Surg. Int. 2004 20 200 203 10.1007/s00383-004-1147-1 15083330 2. Chapple C. Osman N. MacNeil S. Developing tissue-engineered solutions for the treatment of extensive urethral strictures Eur. Urol. 2013 63 539 541 10.1016/j.eururo.2012.09.046 23031675 3. Badylak S.F. Xenogeneic extracellular matrix as a scaffold for tissue reconstruction Transpl. Immunol. 2004 12 367 377 10.1016/j.trim.2003.12.016 15157928 4. Wu S. Liu Y. Bharadwaj S. Atala A. Zhang Y. Human urine-derived stem cells seeded in a modified 3D porous small intestinal submucosa scaffold for urethral tissue engineering Biomaterials 2011 32 1317 1326 10.1016/j.biomaterials.2010.10.006 21055807 5. El-Kassaby A. AbouShwareb T. Atala A. Randomized comparative study between buccal mucosal and acellular bladder matrix grafts in complex anterior urethral strictures J. Urol. 2008 179 1432 1436 10.1016/j.juro.2007.11.101 18295282 6. Bhargava S. Patterson J.M. Inman R.D. MacNeil S. Chapple C.R. Tissue-engineered buccal mucosa urethroplasty-clinical outcomes Eur. Urol. 2008 53 1263 1269 10.1016/j.eururo.2008.01.061 18262717 7. Raya-Rivera A. Esquiliano D.R. Yoo J.J. Lopez-Bayghen E. Soker S. Atala A. Tissue-engineered autologous urethras for patients who need reconstruction: An observational study Lancet 2011 377 1175 1182 10.1016/S0140-6736(10)62354-9 21388673 8. Osman N.I. Patterson J.M. MacNeil S. Chapple C.R. Long-term follow-up after tissue-engineered buccal mucosa urethroplasty Eur. Urol. 2014 66 790 791 10.1016/j.eururo.2014.07.007 25065526 9. Zhang K. Guo X. Zhao W. Niu G. Mo X. Fu Q. Application of wnt pathway inhibitor delivering scaffold for inhibiting fibrosis in urethra strictures: In vitro and in vivo study Int. J. Mol. Sci. 2015 16 27659 27676 10.3390/ijms161126050 26610467 10. De Kemp V. de Graaf P. Fledderus J.O. Ruud Bosch J.L. de Kort L.M. Tissue engineering for human urethral reconstruction: Systematic review of recent literature PLoS ONE 2015 10 1262 10.1371/journal.pone.0118653 25689740 11. Orabi H. AbouShwareb T. Zhang Y. Yoo J.J. Atala A. Cell-seeded tubularized scaffolds for reconstruction of long urethral defects: A preclinical study Eur. Urol. 2013 63 531 538 10.1016/j.eururo.2012.07.041 22877501 12. De Filippo R.E. Yoo J.J. Atala A. Urethral replacement using cell seeded tubularized collagen matrices J. Urol. 2002 168 1789 1793 10.1016/S0022-5347(05)64414-X 12352360 13. Yamzon J.L. Kokorowski P. Koh C.J. Stem cells and tissue engineering applications of the genitourinary tract Pediatr. Res. 2008 63 472 477 10.1203/PDR.0b013e31816a704a 18427290 14. Dong W. Jiang F. Zhao L. You J. Cao X. Li Y. PLLA microalloys versus PLLA nanoalloys: Preparation, morphologies, and properties ACS Appl. Mater. Interface 2012 4 3667 3675 10.1021/am3007577 22712765 15. Ulbricht J. Jordan R. Luxenhofer R. On the biodegradability of polyethylene glycol, polypeptoids and poly(2-oxazoline)s Biomaterials 2014 35 4848 4861 10.1016/j.biomaterials.2014.02.029 24651032 16. Lin Y. Wang L. Zhang P. Wang X. Chen X. Jing X. Su Z. Surface modification of poly(l -lactic acid) to improve its cytocompatibility via assembly of polyelectrolytes and gelatin Acta Biomater. 2006 2 155 164 10.1016/j.actbio.2005.10.002 16701873 17. Li B. Ma Y. Wang S. Moran P.M. A technique for preparing protein gradients on polymeric surfaces: Effects on PC12 pheochromocytoma cells Biomaterials 2005 26 1487 1495 10.1016/j.biomaterials.2004.05.004 15522750 18. Wang B.Y. Fu S.Z. Ni P.Y. Peng J.R. Zheng L. Luo F. Liu H. Qian Z.Y. Electrospun polylactide/poly(ethylene glycol) hybrid fibrous scaffolds for tissue engineering J. Biomed. Mater. Res. A 2012 100 441 449 10.1002/jbm.a.33264 22105865 19. Fu Q. Cao Y.L. Tissue engineering and stem cell application of urethroplasty: From bench to bedside Urology 2012 79 246 253 10.1016/j.urology.2011.08.043 22014966 20. Wang K. Guan Y. Liu Y. Zhu M. Li T. An D. Ou L. Che Y. Zhang G. Zhang J. Fibrin glue with autogenic bone marrow mesenchymal stem cells for urethral injury repair in rabbit model Tissue Eng. A 2012 18 2507 2517 10.1089/ten.tea.2011.0359 22779425 21. Wang D.J. Li M.Y. Huang W.T. Lu M.H. Hu C. Li K. Qiu J.G. Gao X. Repair of urethral defects with polylactid acid fibrous membrane seeded with adipose-derived stem cells in a rabbit model Connect. Tissue Res. 2015 56 434 439 10.3109/03008207.2015.1035376 25943462 22. Yuan H. Zhuang Y. Xiong J. Zhi W. Liu L. Wei Q. Han P. Human umbilical mesenchymal stem cells-seeded bladder acellular matrix grafts for reconstruction of bladder defects in a canine model PLoS ONE 2013 8 1262 10.1371/journal.pone.0080959 24278354 23. Diaz-Prado S. Muinos-Lopez E. Hermida-Gomez T. Cicione C. Rendal-Vazquez M.E. Fuentes-Boquete I. de Toro F.J. Blanco F.J. Human amniotic membrane as an alternative source of stem cells for regenerative medicine Differentiation 2011 81 162 171 10.1016/j.diff.2011.01.005 21339039 24. Seo M.-S. Park S.-B. Kim H.-S. Kang J.-G. Chae J.-S. Kang K.-S. Isolation and characterization of equine amniotic membrane-derived mesenchymal stem cells J. Vet. Sci. 2013 14 151 10.4142/jvs.2013.14.2.151 23388430 25. Alviano F. Fossati V. Marchionni C. Arpinati M. Bonsi L. Franchina M. Lanzoni G. Cantoni S. Cavallini C. Bianchi F. Term amniotic membrane is a high throughput source for multipotent mesenchymal stem cells with the ability to differentiate into endothelial cells in vitro BMC Dev. Biol. 2007 7 1262 10.1186/1471-213X-7-11 17313666 26. Tsuji H. Miyoshi S. Ikegami Y. Hida N. Asada H. Togashi I. Suzuki J. Satake M. Nakamizo H. Tanaka M. Xenografted human amniotic membrane-derived mesenchymal stem cells are immunologically tolerated and transdifferentiated into cardiomyocytes Circ. Res. 2010 106 1613 1623 10.1161/CIRCRESAHA.109.205260 20508201 27. Wang X. Ding B. Li B. Biomimetic electrospun nanofibrous structures for tissue engineering Mater. Today 2013 16 229 241 10.1016/j.mattod.2013.06.005 25125992 28. Zhao J. Dehbari N. Han W. Huang L. Tang Y. Electrospun multi-scale hybrid nanofiber/net with enhanced water swelling ability in rubber composites Mater. Des. 2015 86 14 21 10.1016/j.matdes.2015.07.105 29. Zhou L. Zhu C. Edmonds L. Yang H. Li B. Microsol-electrospinning for controlled loading and release of water-soluble drugs in nanofibrous membranes RSC Adv. 2014 4 43220 43226 10.1039/C4RA07518H 30. Xie M. Song L. Wang J. Fan S. Zhang Y. Xu Y. Evaluation of stretched electrospun silk fibroin matrices seeded with urothelial cells for urethra reconstruction J. Surg. Res. 2013 184 774 781 10.1016/j.jss.2013.04.016 23706393 31. Kim J.H. Kim S.H. Kim H.K. Akaike T. Kim S.C. Adhesion and growth of endothelial cell on amphiphilic PU/PS IPN surface: Effect of amphiphilic balance and immobilized collagen J. Biomed. Mater. Res. A 2002 62 613 621 10.1002/jbm.10340 12221710 32. Han D.K. Park K.D. Hubbell J.A. Kim Y.H. Surface characteristics and biocompatibility of lactide-based poly(ethylene glycol) scaffolds for tissue engineering J. Biomater. Sci. Polym. Ed. 1998 9 667 680 9686334 33. Li B. Ma Y. Wang S. Moran P.M. Influence of carboxyl group density on neuron cell attachment and differentiation behavior: Gradient-guided neurite outgrowth Biomaterials 2005 26 4956 4963 10.1016/j.biomaterials.2005.01.018 15769531 34. Feng C. Xu Y.M. Fu Q. Zhu W.D. Cui L. Chen J. Evaluation of the biocompatibility and mechanical properties of naturally derived and synthetic scaffolds for urethral reconstruction J. Biomed. Mater. Res. A 2010 94 317 325 10.1002/jbm.a.32729 20166222 35. Zhang Y. Atala A. Urothelial cell culture: Stratified urothelial sheet and three-dimensional growth of urothelial structure Methods Mol. Biol. 2013 945 383 399 23097119

PMC005xxxxxx/PMC5000661.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081263ijms-17-01263ArticleIntegrating Insect Life History and Food Plant Phenology: Flexible Maternal Choice Is Adaptive Fei Minghui 1Harvey Jeffrey A. 12Weldegergis Berhane T. 3Huang Tzeyi 3Reijngoudt Kimmy 3Vet Louise M. 13Gols Rieta 3*Maffei Massimo Academic Editor1 Department of Terrestrial Ecology, Netherlands Institute of Ecology, Droevendaalsesteeg 10, 6708 PB Wageningen, The Netherlands; [email protected] (M.F.); [email protected] (J.A.H.); [email protected] (L.M.V.)2 Section Animal Ecology, Department of Ecological Sciences, VU University Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands3 Laboratory of Entomology, Wageningen University, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands; [email protected] (B.T.W.); [email protected] (T.H.); [email protected] (K.R.)* Correspondence: [email protected]; Tel.: +31-317-48340003 8 2016 8 2016 17 8 126305 7 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Experience of insect herbivores and their natural enemies in the natal habitat is considered to affect their likelihood of accepting a similar habitat or plant/host during dispersal. Growing phenology of food plants and the number of generations in the insects further determines lability of insect behavioural responses at eclosion. We studied the effect of rearing history on oviposition preference in a multivoltine herbivore (Pieris brassicae), and foraging behaviour in the endoparasitoid wasp (Cotesia glomerata) a specialist enemy of P. brassicae. Different generations of the insects are obligatorily associated with different plants in the Brassicaceae, e.g., Brassica rapa, Brassica nigra and Sinapis arvensis, exhibiting different seasonal phenologies in The Netherlands. Food plant preference of adults was examined when the insects had been reared on each of the three plant species for one generation. Rearing history only marginally affected oviposition preference of P. brassicae butterflies, but they never preferred the plant on which they had been reared. C. glomerata had a clear preference for host-infested B. rapa plants, irrespective of rearing history. Higher levels of the glucosinolate breakdown product 3-butenyl isothiocyanate in the headspace of B. rapa plants could explain enhanced attractiveness. Our results reveal the potential importance of flexible plant choice for female multivoltine insects in nature. endoparasitoidforagingglucosinolateherbivoreherbivore induced plant volatile (HIPV)multivoltineovipositionrearing historyplant volatiles ==== Body 1. Introduction Herbivorous insects have to locate their host food plants often embedded in patches that may be species-rich and structurally and chemically complex [1,2]. Their co-evolved natural enemies, such as parasitoids and predators, are similarly challenged when they are searching for hosts or prey [1,2]. Location of these resources during foraging is often characterized by a gradual narrowing down of the area in which these resources can be found, and is described as a reliability-detectability problem [3]. For example, insect herbivores first have to find the proper habitat and then locate a suitable food plant within this habitat; their parasitoids must also overcome these same challenges to locate hosts that are often small and feeding in concealed locations on the food plant. Once potential food plants and/or hosts have been located, these can be accepted as oviposition sites or rejected, which is largely determined by differences in the suitability and quality of the resources for their development [4,5]. The first steps, i.e., habitat and host location, of this sequential process eventually leading to successful insect development primarily rely on visual and olfactory cues utilized by the insects [6,7,8]. In particular, volatiles that are released by plants in response to herbivore feeding (so-called herbivore induced plant volatiles (HIPVs)) have been extensively studied in relation to parasitoid foraging behaviour over the past 25 years [7,9,10,11,12,13]. The vast majority of insect herbivores are specialists that feed on only a few related plant species in nature [14,15]. Therefore, an herbivore and its specialist parasitoids are expected to rely on specific cues that are related to their food plants or hosts [7,16,17] such as phylogenetically conserved secondary (defensive) metabolites [18]. For example, larvae of the cabbage butterflies Pieris brassicae and P. rapae, primarily feed on plant species in the family Brassicaceae that produce inducible glycoside compounds called glucosinolates [19,20]. Gravid female butterflies of these two species use these glucosinolates to recognize suitable food plants for their offspring by using their tarsi to “scratch” plant tissues prior to oviposition [17,21]. Once they are in the appropriate habitat containing suitable plants or hosts, learning and subsequent experience may further influence the foraging behaviour of insect herbivores and their natural enemies when they are searching for resources. However, the strength of this effect often depends on the developmental stage at which the experience occurs. For example, oviposition experience of adult female parasitoids in the presence of characteristic volatile blends often enhances the parasitoid’s response to these volatiles when these are offered in the absence of hosts [22]. This behavioural adaptation is referred to as associative learning and has been observed in both insect herbivores and their natural enemies [23]. Natal experience, which is obtained during larval feeding and growth, may also affect habitat preferences later in life during the adult stage. Known as the Hopkins’s host selection principle (HHSP), it is, however, controversial as it implies that some form of imprinting is maintained during metamorphosis affecting later developmental stages [24] and also because evidence supporting the principle is thus far scarce [25]. Another major problem with the HHSP is that it does not take into account constraints imposed by temporal changes in diet that may be predicated by life history characteristics of the consumer and its resource. For insects that must switch plant diets from one generation to another (e.g., where the progeny exploit a different species of plant from their parents), it is clear that larval imprinting on a plant may be maladaptive if it hinders the ability of the insects to find and locate new resources that are chemically different from those on which they developed. Thus far, however, most studies tacitly assume that specialist herbivores exploit the same food plant species over many generations, making natal imprinting adaptive. Some studies have shown that pre-adult experience can affect later foraging behaviour in insects for oviposition sites, at least if these sites are the same or at least very similar to those on which the offspring developed [26,27]. Furthermore, when natal experience influences later habitat choice, it increases the acceptance of the natal habitat type [28]. Preference for the natal habitat type could be beneficial for insects, because natal experience can influence plastic traits, such as the response to cues used during foraging, which make them better adapted to exploit the same resources in similar habitats [29]. Such preference is adaptive because it reduces the costs associated with exploring multiple habitats and in assessing the suitability of these habitats [27]. However, the strength of adaptation also depends on the degree to which the environment changes across space and time in relation to the generation time of the insects. Many species of herbivorous insects are multivoltine and thus have two or more generations per year [30]. Moreover, some of these herbivores are known to feed on short-lived annual plants that are present in the field for only two or three months during the growing season [31]. Under these conditions, successive generations of herbivores that rely on short-lived annuals for food are obligated to leave the natal plant patch to locate and oviposit on a different plant species that may be different from the plant on which they developed and which grows a considerable distance (kilometers) away. Specialist multivoltine parasitoids of these herbivores are faced with the same constraints related to habitat and host location and thus must track them from one habitat patch to another. In this study, we investigate the effect of rearing history on oviposition preference for different related host plant species in a multivoltine herbivore Pieris brassicae L. (Lepidoptera: Pieridae) and host plant preference behaviour in its endoparasitoid Cotesia glomerata L. (Hymenoptera: Braconidae). Caterpillars of P. brassicae are specialized on brassicaceous plant species of which all native species over much of its range are short-lived annuals. In the Netherlands, the species has generally three generations per year depending on temperature. The three annual plants studied here, wild turnip, Brassica rapa L., charlock mustard, Sinapis arvensis L., and black mustard, Brassica nigra L., were grown in temporal sequence and are important wild food plants for successive generations of P. brassicae in The Netherlands [31]. These plant species tend to grow in dense stands, which is a prerequisite for survival of P. brassicae because females lay eggs in clusters that need several plants to sustain their larval development [32]. Cotesia glomerata is a specialized gregarious endoparasitoid, i.e., females lay several eggs in the host at a single oviposition event. It primarily attacks early caterpillar stages of P. brassicae and it has two to three generations in the Netherlands, also depending on temperature. The main aim of this study is to determine whether rearing history (i.e., insects reared on the different food-plant species [B. rapa, S. arvensis, B. nigra]) in one generation affects maternal preference of P. brassicae and foraging behaviour of C. glomerata for the three different plants infested with P. brassicae caterpillars. We hypothesize that the rearing history of the two insects will not affect preference for food plant species (herbivore) or volatile-mediated foraging (parasitoid) of future generations because pre-adult conditioning on the natal plant may confer costs e.g., the insects remaining within the natal patch may only encounter plants that are dying and are thus nutritionally unsuitable. 2. Results 2.1. Host-Plant Oviposition Preference of Pieris brassicae Butterflies There was a significant difference in oviposition preference when the P. brassicae butterflies were reared on B. oleracea (χ22 = 7.94, p = 0.02 Figure 1a), B. nigra (χ22 = 7.09, p = 0.03, Figure 1b) or B. rapa (χ22 = 9.80, p = 0.007, Figure 1c), whereas this was not the case for butterflies that were reared on S. arvensis (χ22 = 2.36, p = 0.31, Figure 1d). When reared on B. oleracea, female P. brassicae butterflies preferred to lay eggs on B. rapa, though this preference was only statistically significant for the pair-wise B. rapa–S. arvensis comparison. When reared on B. nigra, P. brassicae marginally preferred to lay eggs on B. rapa (B. rapa vs. S. arvensis (α = 0.0167): χ21 = 5.54, p = 0.019; B. rapa vs. B. nigra, χ21 = 3.57, p = 0.059, Figure 1b). With a rearing history on B. rapa or S. arvensis, butterfly oviposition preference ranked from low to high B. rapa < S. arvensis < B. nigra (Figure 1c,d). Statistically, preference was significant only for the B. rapa–B. nigra pair-wise comparison for butterflies reared on B. rapa. 2.2. Host-Plant Landing Preference of the Parasitoid, Cotesia glomerata In total, 1270 wasps made a choice in the wind tunnel, which was 92% of the wasps that were initially released in the wind tunnel. Rearing history had no effect on volatile-mediated foraging behaviour (Figure 2; B. nigra vs. S. arvensis: χ23 = 4.02, p = 0.26; B. rapa vs. S. arvensis: χ23 = 2.11, p = 0.55; B. nigra vs. B. rapa: χ23 = 3.68, p = 0.30 based on generalized linear model (GLM) analyses. Overall, wasps clearly preferred host-infested B. rapa plants over S. arvensis (t37 = 7.0, p < 0.001) and B. nigra plants (t39 = 6.2, p < 0.001), though, in the latter case, this preference was less pronounced when the wasps had been reared on B. nigra (Figure 2c). They also preferred host-infested B. nigra over S. arvensis plants (t38 = 2.7, p = 0.01), especially when they had been reared on B. nigra (Figure 2a). Furthermore, both plant architecture (GLM: χ21 = 0.12, p = 0.73, Figure 3) and early exposure to HIPV had no effect on wasp landing preference (GLM: χ21 = 1.66, p = 0.20, Figure 4). There were significant differences in the amount of leaf tissues consumed by P. brassicae larvae among the three plant species (F2,54 = 3.38, p = 0.041) (Figure 5). The damage inflicted to S. arvensis plants was marginally, though not statistically, greater than the damage inflicted to B. rapa (Tukey test: p = 0.06, Figure 5) and B. nigra (Tukey test: p = 0.08, Figure 5). Damage levels were similar on B. rapa and B. nigra plants (Tukey test: p = 0.99, Figure 5). 2.3. Headspace Analysis In the headspace of B. rapa, S. arvensis, and B. nigra that had been fed upon by P. brassicae larvae for 24 h, 33 different compounds were detected, of which 29 were present in the HIPV blend of all three host plant species (Table 1). Based on PCA analysis of the volatiles, samples from the three plant species clearly separated (Figure 6a). The first PC, explaining 28.94% of the variation, separated B. rapa from B. nigra, whereas the second PC, explaining an additional 24.76% of the variation, further separated S. arvensis from B. rapa and B. nigra plants (Figure 6a). This means that the volatile blends emitted by B. rapa and B. nigra were more dissimilar compared to the blend emitted by S. arvensis plants. There was a significant difference in the total amount of volatiles emitted by the three plant species (F2,29 = 15.3, p < 0.001). Brassica rapa and B. nigra emitted a larger volume of volatiles than S. arvensis (Tukey multiple comparison tests: B. rapa vs. S. arvensis and B. nigra vs. S. arvensis both p < 0.05, B. rapa vs. B. nigra p > 0.05). Compounds that were emitted in higher amounts by B. rapa were the two nitriles: 2-methylbutanenitrile (ID 1), and 3-methyl-3-butenenitrile (ID 2); the glucosinolate hydrolysis product: 3-butenyl isothiocyanate (ID 9); the two green leaf volatiles (Z)-3-hexen-1-ol (ID 4) and (Z)-3-hexen-1-ol-acetate (ID 11) and the sesquiterpene (E,E)-α-farnesene (ID 29) (Figure 6b, Table 1). B. nigra plants were characterized by the relatively high emissions of the glucosinolate breakdown product allyl isothiocyanate (ID 5), and silphiperfolene isomers (ID 19, 20, 22), which were absent or only emitted in very small amount by the other two plant species (Figure 6b, Table 1). S. arvensis plants produced relatively more of the sesquiterpernes α-and β-caryophyllene (ID 24 and 28) (Figure 6b, Table 1). 3. Discussion In this study, we show that rearing history only partially affected oviposition preference of P. brassicae butterflies and never resulted in a preference for the plant on which it had been feeding during larval development. It also had little or no effect on the foraging behaviour (i.e., plant preference) of its parasitoid, C. glomerata. Whereas P. brassicae butterflies reared on the different food plants did not exhibit any consistencies in oviposition preference behaviour, C. glomerata clearly preferred to alight on herbivore-damaged B. rapa plants. Preference of C. glomerata for B. rapa could not be explained by plant architecture, given that it is structurally similar, albeit slightly smaller, than the other two brassicaceous plant species studied here. Headspace analyses also revealed significant quantitative and qualitative differences among the HIPV blends emitted by the three plant species. Natal experience has been reported to affect adult habitat selection by insects [27,33,34], but few studies have investigated this in lepidopteran species (butterflies and moths). Larval feeding experience with a feeding deterrent modified oviposition responses of subsequent adults in the moth species, Ephestia cautella and Plodia interpunctella [35] Trichoplusia ni [36], Spodoptera littoralis [37] and Lobesia botran [38]. To affect choices made during oviposition preference, cues obtained in the natal habitat must be memorized during larval feeding and carried through pupation to the adult stage. However, in holometabolous insects, the nervous system is profoundly reorganized during metamorphosis [39,40], which makes it unlikely that experience learned during the larval stage is easily retained in adult insects [25]. In nature, P. brassicae has up to three generations per year across much of its native range in Western and central Europe [31,41]. Furthermore, P. brassicae larvae require many food plants to support the successful development of a single brood, a requirement that limits the number of suitable plant species as oviposition sites in nature to about six or seven [32]. These plants, including the three species studied here, grow in large tightly assembled populations that enable the caterpillars to disperse from the natal plant to adjacent plants later during larval development by moving through the canopy [32,42]. Importantly, qualitatively and quantitatively suitable food plants are annuals or biennials with short growing seasons, and many of these plants also exhibit discrete periods of growth during the season. For example, B. rapa generally grows between March and May, S. arvensis between May and July, and B. nigra between June and August [31]. This means that different generations of P. brassicae must search for different host plants that generally grow at different locations, often a considerable distance away from the natal patch. Therefore, it is adaptive for P. brassicae that oviposition preference is not affected by larval rearing history; otherwise, the adults would risk wasting time searching for food plants that are no longer present (or which are no longer nutritionally suitable) in the natal habitat. Natal experience is only expected to affect plastic traits when it benefits animal fitness [29]. The fact P. brassicae is multivoltine and a specialist on short-lived, clustered brassicaceous plant species may explain why the effect of natal imprinting on adult oviposition preference is weak or non-existent. Though butterflies never preferred the plant on which they had been reared, the rank order of oviposition preference differed with natal experience; females preferred B. rapa plants for oviposition when reared on B. oleracea and B. nigra and preferred B. nigra when reared on B. rapa or S. arvensis. Natal imprinting is not only found in lepidopteran species, but has also been observed in some species of Hymenoptera [28,43]. Several studies have shown that the response of adult female parasitoids to HIPV differs depending on the diet on which their host was feeding during larval parasitoid development inside of the host body [44,45]. For example, the ectoparasitoid Hyssopus pallid was more attracted to frass from its fruit-feeding host Cydia pomonella when the wasps had developed on hosts fed on apples compared to wasps reared on hosts fed on artificial diet [46]. The length of rearing history can also play a role in the wasp’s future plant volatile preferences [47]. For instance, when Plutella xylostella that are parasitized by Diadegma semiclausum were fed on snow pea for three successive generations, female wasps showed a relatively higher preference to snow pea volatiles in the third than in the first generation [47]. In our study, the parasitoid was only reared on hosts and plants for a single generation, reflecting conditions found in nature, where different generations generally must find hosts on different plant species. Consequently, we found that natal experience had no effect on volatile-mediated foraging behaviour in C. glomerata. The adaptive potential of natal imprinting clearly depends on such factors as the reliability of being associated with the same plant species or the degree of chemical and structural similarity of different plant species that may be used in successive generations by the herbivore and its parasitoid. Natal experience is adaptive when the environment is predictable over several generations in an insect. For multivoltine parasitic wasps, where different generations also need to search for hosts on different plant species in different habitats, it is important that natal experience exerts little effect on their landing preference. We also found that early exposure to HIPV at eclosion had no effect on wasp landing preference. By contrast, host plant stimuli have been reported to increase a parasitoid’s attraction to the natal host plant [48]. For example, the attraction of the parasitoid wasp Trichogramma brassicae to tomato plants increased when the wasps were allowed to emerge from their hosts in the presence of these plants. Similarly, attraction of C. congregata to cherry volatiles increased when the parasitoid had physical contact with the host plant at eclosion. Therefore, the importance of conditioning at eclosion appears to be association-specific and even differs amongst closely related taxa (e.g., C. glomerata and C. congregata). This could be due to differences in life-history traits among the plants and insects. In the case of C. congregata, its herbivore hosts (e.g., the larvae of sphingid moths) may associate with the same food plants and/or habitats over successive generations, making conditioning adaptive. Certainly more plant host–parasitoid associations need to be studied to extrapolate potential relationships between the life-history of the plants and hosts and conditioning/innate responses in their parasitoids. Female parasitoids clearly preferred B. rapa over the other two cruciferous species. Several factors could contribute to this preference. B. rapa grows early in the season and may therefore be one of the few plant species available in the Netherlands for P. brassicae when they emerge from winter diapause. Consequently C. glomerata may have evolved a strong sensitivity to (volatile) cues related to the first available food plant of its host. Alternatively, some of the volatiles emitted by B. rapa may trigger a stronger sensory response in the parasitoid than compounds in the blend of B. nigra and S. arvensis. Little is still known as to the identity of specific volatiles or volatile blends that are most attractive to parasitoids [45], although some compounds have been shown to play an important role in enhancing attractiveness of the blend [49,50]. The analysis of the HIPV blends showed that there were significant quantitative and qualitative differences among the three plant species. The total amounts of volatiles from B. rapa and B. nigra were significantly larger than from S. arvensis. When HIPV blends induced by different treatments of the same plant species are compared, quantitative aspects of these blends may to a large extent determine parasitoid attraction [49,51]. However, when parasitoid attractiveness to HIPV blends emitted by different plant species is compared, qualitative rather than quantitative aspects may be more important [45]. It is known that blends produced by species in the Brassicaceae vary dramatically across different species [52,53]. All brassicaceous plant species produce glucosinolates [54], which function as defensive compounds against a range of attackers such as pathogens and insect herbivores [20,55]. Deterrent or toxic activity only emerges after tissue damage, e.g., by caterpillar feeding and concomitant release of the enzyme myrosinase, which is stored in specialized cells. This enzyme catalyses the conversion of glucosinolates into toxic hydrolysis products, of which many are volatile [55]. These volatile breakdown products have been shown to serve as reliable signals for the parasitoid C. rubecula to their host, P. rapae [56]. If breakdown products of glucosinolates play a role in host plant selection by C. glomerata, the high amounts of 3-butenyl isothiocyanate, which is the breakdown product of gluconapin, the dominant glucosinolate in B. rapa, may explain its enhanced attraction to this plant. However, this does not explain why B. rapa is more attractive than B. nigra which emits allyl isothiocyanate in even larger amounts than B. rapa emits 3-butenyl isothiocyanate (allyl isothiocyanate is a hydrolysis product of sinigrin, the dominant glucosinolate in B. nigra). Diaeretiella rapae, a parasitoid of the aphid Brevicoryene brassicae that is a specialist of brassicaceous plants, was shown to be more attracted to synthetic 3-butenyl isothiocyanate than to 4-pentenyl isothiocyanate [50], although it is also attracted to synthetic allyl isocyanate [57]. This suggests that isothiocyanates are differentially attractive to parasitoids. The low volatility of hydrolysis products of sinabin, the dominant glucosinolate in S. arvensis, may be responsible for the absence of these compounds in the headspace of S. arvensis explaining the reduced attractiveness of these plants to C. glomerata. In summary, our study reveals that rearing history has little or no effect on oviposition preference of P. brassicae butterflies or landing preference of its major parasitoid C. glomerata. Oviposition preference of P. brassicae shifted between B. nigra and B. rapa, but the butterflies never displayed a clear preference for the plant species on which they had been reared. C. glomerata had a clear preference for host-infested B. rapa plants. For multivoltine insects, such as P. brassicae and C. glomerata that primarily rely on short-lived annuals for immature development, it is a challenge for different generations to locate suitable host plants, given that they are forced to leave the natal habitat to do so. Therefore, it is adaptive that these insects are labile in the cues that they use in host plant location behaviour and, thus, that it is not affected by natal imprinting. Furthermore, our study also shows that the herbivore and the parasitoid use different cues when searching for food or host plants. To better understand the mechanisms that underline these interactions, it is important to examine an array of ecophysiological constraints on the insects and the traits the insects exhibit to counter them. Clearly, the biology and phenology of the food plant(s) leave an indelible mark on their insects. 4. Materials and Methods 4.1. Plants B. rapa, B. nigra and S. arvensis seeds were collected from natural growing populations in Gelderland, The Netherlands. Seeds were germinated and seedlings were subsequently transferred to 1.1-L pots filled with peat soil (Lentse potgrond no.4; lent, The Netherlands). Plants were grown in a greenhouse at the Netherlands Institute of Ecology (NIOO) under the following conditions: 21 ± 2 °C (day) and 16 ± 2 °C (night), 50% relative humidity, and a photoperiod of at least 16 h. The plants were watered twice a week during the first 3 weeks of development. When the plants were 3 weeks old, they were fertilized once a week with Hoagland solution, which was applied to the soil. Watering and fertilization continued during the experiments. As the insects have been reared on Brussels sprout plants, Brassica oleracea L. var. gemmifera cv. Cyrus for many (>10) generations, this plant was used as a control. Brussels sprout plants were grown from seeds in peat soil in 1.1-L plastic pots in a greenhouse (50%–70% relative humidity, 20–25 °C, and a photoperiod of 16 h) and were 4 to 5 weeks old when used in the experiments. 4.2. Insects P. brassicae and C. glomerata were collected in experimental fields near Wageningen, The Netherlands. P. brassicae caterpillars were reared on Brussels sprout plants in a greenhouse at 50%–70% relative humidity, 20–25 °C, and a photoperiod of 16 h at Wageningen University (WU). C. glomerata was reared on young P. brassicae caterpillars feeding on Brussels sprouts. Once the fully developed larvae of C. glomerata emerged from P. brassicae hosts and had spun cocoons, they were collected for further rearing or experimental purposes. Approximately five days after cocoon formation, adult wasps emerged at which point they were provided with 10% sugar solution. 4.3. Preparation of Insects Used in Experiments 4.3.1. Herbivore P. brassicae were reared from egg-to-adult for one generation on one of the three host plant species: B. rapa, S. arvensis or B. nigra. We also determined oviposition preference of butterflies reared on B. oleracea on which they had been reared for many generations. Single four-week-old B. oleracea and three-week-old B. rapa, S. arvensis, or B. nigra plants were placed in the rearing cage with adult P. brassicae butterflies for 24 h. Plants with egg clusters were transferred to a cage with additional plants of the same species as the one on which the eggs were laid. Eggs were allowed to develop into pupae on their respective food plants. Eclosing butterflies were provided with a (20%) honey solution and were allowed to mate. Butterflies were 3–5 days old when they were used in the choice bioassays. 4.3.2. Parasitoid C. glomerata were reared for one generation on P. brassicae caterpillars feeding on one of the four host plant species: B. oleracea, B. rapa, S. arvensis or B. nigra. Caterpillars of P. brassicae were obtained and reared as described in the previous section until they reached the mid first instar stage. For parasitism, female wasps were collected from the general culture. First instar P. brassicae caterpillars were parasitized by presenting them individually to a female wasp. After parasitism by C. glomerata, caterpillars were introduced onto one of the four host plant species (B. oleracea, B. rapa, S. arvensis or B. nigra), which were maintained in separate cages until the larvae of the parasitoids emerged and formed cocoons. Parasitoid cocoons were collected in Petri dishes (9.5 cm) and were maintained in an incubator at 21 ± 1 °C until adult eclosion at which point they were transferred into 30 × 30 × 30 cm (Bugdorm) plastic cages and provided with 10% sugar, water, and honey. Female wasps used in the bioassays were 2–8 days old. 4.4. Host-Plant Oviposition Preference of Pieris brassicae Butterflies Oviposition preferences were assessed in three-choice experiments in six outdoor tents (3 × 4 × 2 m) placed on bare soil in an experimental field adjacent to WU. Plants from each species were prepared as described in the Plants section. Single plants of each of the three plant species were randomly placed in a triangle, approximately 1.5 m apart, in the experimental tents. One female and one male butterfly were released in the middle of the tent. A bioassay was terminated and its choice recorded when a female butterfly had laid the first egg clutch, which was checked three times a day. Females were used only once. The bioassay was repeated at least 30 times with butterflies being reared on the same plant species. Bioassays were conducted from June to August 2013. New plants were used for each replicate and the positioning of the plant species in the tent was randomized. 4.5. Host-Plant Landing Preference of the Parasitoid, Cotesia glomerata In a wind tunnel set-up (see below), we determined HIPV mediated landing preference of female C. glomerata parasitoids when reared for one generation from P. brassicae caterpillars developing on of the host plants, B. rapa, S. arvensis and B. nigra, respectively. In addition, we used wasps that had developed in P. brassicae feeding on B. oleracea, the food plant on which the insects had been reared for >10 generations. In the wind tunnel, plant pairs, i.e., all three combinations of host-infested B. rapa, S. arvensis and B. nigra plants were offered to parasitoids reared on the four different food plants. Individual plants were infested with 20 first instar P. brassicae or 10 second instar P. brassicae caterpillars, depending on caterpillar availability, and incubated in a greenhouse for 24 h at 50%–70% relative humidity and 20–25 °C with a photoperiod L:D of 16:8 h. Plant combinations used in single choice bioassays were always infested with the same number of caterpillars of the same instar. To determine whether differences in the amount of feeding damage affected landing preference, we determined for each of the three plant species (n = 19 per plant species) the amount of leaf tissue consumed from plants infested by 20 first-instar P. brassicae larvae for 24 h. Damaged areas were calculated using millimeter paper on transparent plastic sheets. The plant species differ in their architecture, which could affect landing preference of the wasps. For instance, S. arvensis and B. nigra grow taller than B. rapa, which has a shorter main stem and leaves that initially expand horizontally. In an additional wind tunnel experiment, we examined the architectural influence on HIPV preference using single detached leaves from B. nigra and B. rapa instead of intact plants. The wasps used in this experiment had been reared from P. brassicae larvae on B. nigra. Leaves infested by 20 first-instar P. brassicae larvae for 24 h were cut and put into vials with water, and were allowed to recover for 2–4 h before they were used in a wind tunnel experiment. This comparison was tested in 10 replicate bioassays. Furthermore, wasps can also be conditioned by exposure to HIPVs when they emerge from the host caterpillars prior to pupation and cocoon construction but in the presence of plant material [58]. As described above, one group of wasps was collected and separated from its host and the host plants prior to egression and cocoon construction, whereas another group of wasps was left with its host and host plants through egression and cocoon construction until adult eclosion. Landing preference (n = 10) was compared when wasps of these two groups were offered a host-infested S. arvensis and B. rapa plant, while the insects had been reared on S. arvensis. 4.6. Wind Tunnel Experiment Volatile-mediated foraging behaviour was studied in a wind tunnel set-up, which is described in detail in [59]. The environmental conditions were set as follows: wind speed, 0.1 m·s−1; light intensity 500–1000 Lux: temperature 25 ± 1 °C; relative humidity 60% ± 5%. To stimulate foraging of C. glomerata, females were exposed to a host-damaged Brussels sprout leaf from which the P. brassicae caterpillars had been removed. Female wasps were collected in 7-mL glass vials and wasps were released individually in a “release cylinder” located in the middle of the wind tunnel. Two test plants were placed approximately 60–70 cm up-wind from the release cylinder. Each wasp was observed for a maximum of 15 min. When a wasp did not land on one of the two plants within 15 min, it was recorded as “non-responding” and this data point was excluded from the statistical analysis. In each bioassay with one test plant combination, we tested 10 responding wasps, which served as a single data point. Each test plant combination in relation to the wasp’s rearing history was tested 8–10 times with a new set of plants and each wasp was used only once. The response of a total of 1170 wasps was recorded in the bioassays examining the effect of rearing history, whereas 100 wasps were tested in each of the two additional bioassays. 4.7. Volatile Collection and Analysis of Herbivore Infested Plants Volatiles emitted by B. rapa, S. arvensis, and B. nigra, which had been exposed to feeding by 20 first instar P. brassicae caterpillars for 24 h were collected and analysed. Plants were treated similarly as described for the behaviour bioassay. Volatiles were collected from individual plants, with 9–13 plants per species. The potting soil of the plants around the stem was wrapped in aluminium foil to reduce the release of plastic- and soil-related volatiles before the plants with the caterpillars remaining on them were transferred to a 30-L glass jar containers. Glass jars were sealed with viton-lined glass lids equipped with an air inlet and outlet. Pre-cleaned compressed air filtered through charcoal was led into the glass jars, and the plants were allowed to acclimatize for 40 min. Dynamic headspace volatile collection was carried out in a laboratory at 20 ± 2 °C, by sucking air out of the jar at a rate of 200 mL·min−1 for 2 h through a stainless steel cartridge containing 200 mg Tenax TA (20/35 mesh; CAMSCO, Houston, TX, USA). Immediately after volatile collection, foliar fresh weight of each plant was measured and the Tenax TA cartridges containing sample volatiles were dry-purged under a flow of nitrogen (50 mL·min−1) for 10 min at room temperature (21 ± 2 °C) to remove moisture and stored till analysis. Periodically, volatiles from just pots with soil wrapped in aluminium foil were collected and the compounds recorded, together with the volatiles originating from the Tenax TA adsorbent and the analytical instruments were excluded as artefacts from the data obtained for the plant samples as a correcting measure. Headspace volatile samples were analysed by using a Thermo Trace Ultra Gas Chromatography (GC) coupled to a Thermo Trace DSQ quadrupol mass spectrometer (MS) (both from Thermo (Thermo Fisher Scientific, Waltham, MA, USA) and were used for the separation and detection of volatile compounds. For details of the analytical protocol please refer to [60]. 4.8. Statistical Analysis To statistically analyse P. brassicae female butterfly oviposition preference, we used χ2-tests comparing the observed oviposition preference counts for the three plant species with an expected distribution of 1:1:1. When the test result was significant, we conducted pairwise χ2-square tests with α = 0.05/3 to correct for type I errors (Bonferroni correction). The response variable in the statistical analyses of the C. glomerata wind tunnel bioassays is the fraction of wasps out of the total of 10 responding wasps choosing one of the plant species that was set to be the focal odour source. We used logistic regression, i.e., a generalized linear model (GLM) with binomial variable distribution for errors and a logit link function to determine the effect of rearing history on wasp landing preference for each plant pair combination. In the case of over-dispersion, we corrected for this by allowing the variance functions of the binomial distribution to have a multiplicative over-dispersion factor. Plant species on which the wasps were reared was entered as the explanatory variable in the regression model. To determine whether there was a significant preference for one of the odour sources within a plant pair, we tested H0 = logit = 0 based on model term estimates of the GLM model. Additionally, we used one-sample t-tests to determine whether there was a preference for one of the two odour sources within a plant-pair comparison ignoring the effect of diet with H0: no preference, mean preference fraction is 0.5. We used a similar GLM approach for the data on the effect of plant architecture and the effect of early HIPV exposure on wasps landing preference, respectively. We used one-way ANOVA to examine differences in leaf damage on the three host plants. A multivariate statistical approach was used, i.e., principal component analysis (PCA), to visualize whether volatile profiles could be separated according to plant species and to determine which volatile compounds contributed the most to the separation. All volatiles compounds were included in the analysis. Univariate Kruskal–Wallis tests were employed to reveal significant differences in the emission of each volatile among the three plant species. We used ANOVA to analyse differences in total amounts of volatiles (log-transformed). PCA analysis on all volatile data was performed in Canoco version 5.03 (ter Braak and Šmilauer, Microcomputer Power, Ithaca, NY, USA). All the other analyses were performed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA). 5. Conclusions We have shown that, in an insect herbivore and its gregarious endoparasitoid, innate, conditioned responses are unimportant in terms of food plant preference. This is adaptive because of life-history traits in the herbivore and its parasitoid: the food plants are all short-lived annuals with little temporal overlap in seasonal phenology and the insects are multivoltine, with up to three generations per year. Therefore, in nature, different generations of the insects are obligatorily associated with different plant species that may grow some distance apart. Acknowledgments The authors wish to thank Leon Westerd and Andre Gidding at Wageningen University for the insect cultures and Roel Wagenaar at Netherlands Institute of Ecology (NIOO) for rearing Cotesia glomerata. This study was supported by a grant from the China Scholarship Council. This is NIOO publication number 6138. Author Contributions Minghui Fei, Rieta Gols, Jeffrey A. Harvey and Louise M. Vet designed the experiments; Minghui Fei, Kimmy Reijngoudt, Tzeyi Huang and Rieta Gols conducted the experiments; Minghui Fei, Rieta Gols, Berhane T. Weldegergis and Tzeyi Huang collected samples and performed the chemical analyses; and Minghui Fei, Rieta Gols, Berhane T. Weldegergis and Jeffrey A. Harvey wrote the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Oviposition preference of female Pieris brassicae that had been reared on: Brassica oleracea (a); Brassica nigra (b); Brassica rapa (c); and Sinapis arvensis (d) in a three-way choice assay with B. rapa (dashed bars), S. arvensis (white bars), and B. nigra (grey bars) plants. Bars represent the total preference number, and bars with the same letter are not significantly different from each other (pairwise χ2 test with a Bonferroni correction for multiple comparisons Type I errors). Samples sizes are given in brackets. Figure 2 Landing preference of female Cotesia glomerata that had been reared on Brassica oleracea, Brassica rapa, Sinapis arvensis, or Brassica nigra, in a pair-wise choice assays with: (a) B. nigra (white bars) and S. arvensis (dashed bars); (b) B. rapa (grey bars) and S. arvensis; or (c) B. rapa and B. nigra when infested with Pierisrs brassicae caterpillars for 24 h. Bars present the mean proportion (±standard error of the mean or SE) of choice based on 10 replicate bioassays each tested with 10 responding wasps. An asterisk indicates a significant preference within a plant pair. Figure 3 Landing preference of female Cotesia glomerata in a choice bioassay with intact plants (top bar) or leaves (bottom bar). Wasps had been reared on Brassica nigra and were given the choice between Brassica rapa (dashed bars) and B. nigra (grey bars) infested with 20 first instar Pieris brassicae for 24 h. Bars present the mean proportion (±SE) of choice based on 10 replicate bioassays each tested with 10 responding wasps. Figure 4 Landing preference of female Cotesia glomerata eclosing from cocoons produced by host caterpillars that had been removed from the plants before the parasitoid larvae egressed from the host (non-exposed group, top bar) and those eclosing from cocoons that were left on the plants on which the caterpillars had developed until adult eclosion (exposed group, bottom bar). Wasps had been reared on Sinapis arvensis and were given the choice between a Brassica rapa (dashed bars) and a S. arvensis plant (white bars) infested with 20 first instar Pieris brassicae for 24 h. Bars present the mean proportion (±SE) of choice based on 10 replicate bioassays each tested with 10 responding wasps. Figure 5 Leaf area consumed by 20 first instar Pieris brassicae feeding for 24 h on either Brassica rapa, Sinapis arvensis, or Brassica nigra plants. Bars present the means ± SE (n = 19). Figure 6 Principal component analysis (PCA) on the quantitative data of volatile compounds emitted by Brassica rapa (n = 13), Sinapis arvensis (n = 10), and Brassica nigra (n = 9) plants in response to Pieris brassicae feeding for 24 h. The score plots of the samples (a) depict separation of the different plant species along the first and second PC with the explained variance between brackets. The corresponding loading plot of the variables (b) shows the contribution of each volatile compound to the first two PCs and the sample groups (here plant species). For the identity of compounds presented as numbers in the loading plot (b), please refer to Table 1. ijms-17-01263-t001_Table 1Table 1 Volatile compounds emitted by Brassica rapa, Sinapis arvensis, and Brassica nigra plants damaged by 20 first instar Pieris brassicae caterpillars for 24 h. ID b Plant Species B. rapa a S. arvensis B. nigra Compound (n = 13) (n = 10) (n = 9) 1 2-Methylbutanenitrile *** 91.8 ± 24.4 3.7 ± 0.8 3.8 ± 0.4 2 3-Methyl-3-butenenitrile *** 6.1 ± 2.0 ND ND 3 (E)-2-Hexenal 0.7 ± 0.1 0.6 ± 0.2 0.3 ± 0.07 7,! 4 (Z)-3-Hexen-1-ol *** 20.3 ± 3.0 6.7 ± 1.4 7.5 ± 2.0 5 Allyl isothiocyanate *** 0.2 ± 0.1 7 1.4 ± 0.6 7 135.2 ± 22.6 6 Butane, 1-isothiocyanato *** 14.7 ± 5.3 0.5 ± 0.2 8 0.5 ± 0.08 8 7 (E)-4-Oxo-2-hexenal * 16.5 ± 2.7 14.9 ± 6.5 8 5.6 ± 2.4 7 8 Sabinene 0.9 ± 0.1 1.1 ± 0.2 9 1.0 ± 0.1 9 3-Butenyl isothiocyanate *** 20.1 ± 5.1 0.05 ± 0.08 2 0.7 ± 0.1 10 Myrcene 3.5 ± 0.3 2.9 ± 0.4 3.0 ± 0.3 11 (Z)-3-Hexen-1-ol, acetate 165.4 ± 20.8 49.7 ± 9.4 21.9 ± 7.9 12 Hexanoic acid, 2-ethyl-, methyl ester *** 1.6 ± 1.2 3.6 ± 2.7 0.1 ± 0.09 6 13 (E)-DMNT * 26.4 ± 11.0 4.3 ± 1.2 28.8 ± 7.7 14 Unknown 0.4 ± 0.02 0.3 ± 0.07 8 0.4 ± 0.04 15 Menthol 2.7 ± 0.9 2.0 ± 0.3 3.7 ± 0.8 16 Unknown 0.5 ± 0.1 1.1 ± 0.3 0.8 ± 0.3 17 Unknown 0.7 ± 0.2 1.6 ± 0.6 0.8 ± 0.3 18 Methyl salicylate *** 2.6 ± 0.7 0.3 ± 0.04 0.7 ± 0.2 8 19 Presilphiperfol-7-ene *** ND ND 0.4 ± 0.1 7 20 7-β-H-Silphiperfol-5-ene *** ND ND 0.5 ± 0.2 7 21 α-Terpinyl acetate * 0.3 ± 0.05 0.4 ± 0.06 9 0.2 ± 0.06 7 22 Silphiperfol-6-ene *** ND ND 0.2 ± 0.1 7 23 α-Funebrene ** 0.1 ± 0.05 11 0.09 ± 0.02 8 0.6 ± 0.1 24 β-Caryophyllene ** 2.2 ± 1.4 6 9.3 ± 3.7 0.02 ± 0.05 1 25 (E)-α-Bergamotene 0.09 ± 0.06 6 0.1 ± 0.04 7 0.2 ± 0.08 7 26 α-Guaiene 0.1 ± 0.03 0.1 ± 0.03 8 0.09 ± 0.03 8 27 (E)-β-Bergamotene 0.2 ± 0.2 5 0.05 ± 0.2 1 0.2 ± 0.1 5 28 α-Caryophyllene 0.5 ± 0.3 5 1.1 ± 0.6 6 0.03 ± 0.08 1 29 (E,E)-α-Farnesene *** 22.8 ± 10.6 0.3 ± 0.2 4 0.9 ± 0.3 7 30 α-Bulnesene 0.1 ± 0.04 8 0.06 ± 0.1 2 0.04 ± 0.1 1 31 Methyl cis-dihydrojasmonate 7.1 ± 1.3 9.5 ± 1.3 12.0 ± 1.1 32 Unknown * 20.0 ± 2.7 30.0 ± 6.7 43.1 ± 6.3 33 Unknown * 4.7 ± 0.6 6.8 ± 1.3 10.0 ± 1.3 Total *** 433.5 ± 50.4 152.4 ± 25.9 283.2 ± 32.7 a Volatile emissions are given as a mean peak area ± SE/g fresh weight of foliage divided by 105 with number of sample replicates (n) between brackets; b ID corresponds with the number presented in loading plot (Figure 6b); ! Numbers in superscript following emission quantities represent the number of samples in which a given compound was detected. ND = not detected in any of the samples; * Compounds with asterisks indicate significant differences in emission quantities of volatiles among the three plant species (Kruskal-Wallis one-way ANOVA, * p < 0.05; ** p < 0.01; *** p < 0.001); (E)-DMNT: (E)-4,8-dimethylnona-1,3,7-triene. ==== Refs References 1. Meiners T. Chemical ecology and evolution of plant-insect interactions: A multitrophic perspective Curr. Opin. Insect Sci. 2015 8 22 28 10.1016/j.cois.2015.02.003 2. Wäschke N. Hardge K. Hancock C. Hilker M. Obermaier E. Meiners T. Habitats as complex odour environments: How does plant diversity affect herbivore and parasitoid orientation? PLoS ONE 2014 9 1263 10.1371/journal.pone.0085152 24416354 3. Vet L.E.M. Wäckers F.L. Dicke M. How to hunt for hiding hosts: The reliability-detectability problem in foraging parasitoids Neth. J. Zool. 1990 41 202 213 10.1163/156854291X00144 4. Vinson S.B. Bin F. Vet L.E.M. Critical issues in host selection by insect parasitoids Biol. Cont. 1998 11 77 78 10.1006/bcon.1997.0600 5. Schoonhoven L.M. van Loon J.J.A. Dicke M. Insect-Plant Biology 2nd ed. Oxford University Press Oxford, UK 2005 6. Traynier R.M.M. Visual learning in assays of sinigrin solution as an oviposition releaser for the cabbage butterfly, Pieris rapae Entomol. Exp. Appl. 1986 40 25 33 10.1111/j.1570-7458.1986.tb02151.x 7. Vet L.E.M. Dicke M. Ecology of infochemical use by natural enemies in a tritrophic context Annu. Rev. Entomol. 1992 37 141 172 10.1146/annurev.en.37.010192.001041 8. De Boer J.G. Posthumus M.A. Dicke M. Identification of volatiles that are used in discrimination between plants infested with prey or nonprey herbivores by a predatory mite J. Chem. Ecol. 2004 30 2215 2230 10.1023/B:JOEC.0000048784.79031.5e 15672666 9. Turlings T.C.J. Tumlinson J.H. Lewis W.J. Exploitation of herbivore-induced plant odors by host-seeking parasitic wasps Science 1990 250 1251 1253 10.1126/science.250.4985.1251 17829213 10. Mumm R. Dicke M. Variation in natural plant products and the attraction of bodyguards involved in indirect plant defense Can. J. Zool. 2010 88 628 667 10.1139/Z10-032 11. Arimura G. Matsui K. Takabayashi J. Chemical and molecular ecology of herbivore-induced plant volatiles: Proximate factors and their ultimate functions Plant Cell Physiol. 2009 50 911 923 10.1093/pcp/pcp030 19246460 12. Dicke M. Baldwin I.T. The evolutionary context for herbivore-induced plant volatiles: Beyond the “cry for help” Trends Plant Sci. 2010 15 167 175 10.1016/j.tplants.2009.12.002 20047849 13. McCormick A.C. Unsicker S.B. Gershenzon J. The specificity of herbivore-induced plant volatiles in attracting herbivore enemies Trends Plant Sci. 2012 17 303 310 10.1016/j.tplants.2012.03.012 22503606 14. Loxdale H.D. Lushai G. Harvey J.A. The evolutionary improbability of “generalism” in nature, with special reference to insects Biol. J. Linn. Soc. 2011 103 1 18 10.1111/j.1095-8312.2011.01627.x 15. Loxdake H.D. Harvey J.A. The “generalism” debate: Misinterpreting the term in the empirical literature focusing on dietary breadth in insects Biol. J. Linn. Soc. 2016 10.1111/bij.12816 16. Fraenkel G.S. The raison d’Être of secondary plant substances: These odd chemicals arose as a means of protecting plants from insects and now guide insects to food Science 1959 129 1466 1470 10.1126/science.129.3361.1466 13658975 17. Van Loon J.J.A. Blaakmeer A. Griepink F.C. van Beek T.A. Schoonhoven L.M. de Groot A. Leaf surface compound from Brassica oleracea (Cruciferae) induces oviposition by Pieris brassicae (Lepidoptera: Pieridae) Chemoecology 1992 3 39 44 10.1007/BF01261455 18. Cornell H.V. Hawkins B.A. Herbivore responses to plant secondary compounds: A test of phytochemical coevolution theory Am. Nat. 2003 161 507 522 10.1086/368346 12776881 19. Gols R. Harvey J.A. Plant-mediated effects in the Brassicaceae on the performance and behaviour of parasitoids Phytochem. Rev. 2009 8 187 206 10.1007/s11101-008-9104-6 20. Hopkins R.J. van Dam N.M. van Loon J.J.A. Role of glucosinolates in insect-plant relationships and multitrophic interactions Annu. Rev. Entomol. 2009 54 57 83 10.1146/annurev.ento.54.110807.090623 18811249 21. Renwick J.A.A. Radke C.D. Sachdev-Gupta K. Städler E. Leaf surface chemicals stimulating oviposition by Pieris rapae (Lepidoptera: Pieridae) on cabbage Chemoecology 1992 3 33 38 10.1007/BF01261454 22. Vet L.E.M. de Jong A.G. Franchi E. Papaj D.R. The effect of complete versus incomplete information on odour discrimination in a parasitic wasp Anim. Behav. 1998 55 1271 1279 10.1006/anbe.1997.0686 9632510 23. Papaj D.R. Prokopy R.J. Ecological and evolutionary aspects of learning in phytophagous insects Annu. Rev. Entomol. 1989 34 315 350 10.1146/annurev.en.34.010189.001531 24. Emden H.F.V. Sponagl B. Wagner E. Baker T. Ganguly S. Douloumpaka S. Hopkins’ “host selection principle”, another nail in its coffin Physiol. Entomol. 1996 21 325 328 10.1111/j.1365-3032.1996.tb00873.x 25. Barron A. The life and death of Hopkins’ host-selection principle J. Insect Behav. 2001 14 725 737 10.1023/A:1013033332535 26. Immelmann K. Ecological significance of imprinting and early learning Ann. Rev. Ecol. System. 1975 6 15 37 10.1146/annurev.es.06.110175.000311 27. Davis J.M. Stamps J.A. The effect of natal experience on habitat preferences Trends Ecol. Evol. 2004 19 411 416 10.1016/j.tree.2004.04.006 16701298 28. Davis J.M. Patterns of variation in the influence of natal experience on habitat choice Q. Rev. Biol. 2008 83 363 380 10.1086/592851 19143336 29. Gotthard K. Nylin S. Adaptive plasticity and plasticity as an adaptation: A selective review of plasticity in animal morphology and life history Oikos 1995 74 3 17 10.2307/3545669 30. Gullan P.J. Cranston P.S. The Insects: An Outline of Entomology 4th ed. Wiley-Blackwell Oxford, UK 2010 31. Fei M. Gols R. Harvey J.A. Seasonal phenology of interactions involving short-lived annual plants, a multivoltine herbivore and its endoparasitoid wasp J. Anim. Ecol. 2014 83 234 244 10.1111/1365-2656.12122 24028469 32. Fei M. Gols R. Zhu F. Harvey J.A. Plant quantity affects development and survival of a gregarious insect herbivore and its endoparasitoid wasp PLoS ONE 2016 11 1263 10.1371/journal.pone.0149539 26963702 33. Liu S.S. Li Y.H. Liu Y.Q. Zalucki M.P. Experience-induced preference for oviposition repellents derived from a non-host plant by a specialist herbivore Ecol. Lett. 2005 8 722 729 10.1111/j.1461-0248.2005.00776.x 34. Anderson P. Anton S. Experience-based modulation of behavioural responses to plant volatiles and other sensory cues in insect herbivores Plant Cell Environ. 2014 37 1826 1835 10.1111/pce.12342 24689897 35. Olsson P.O.C. Anderbrant O. Löfstedt C. Experience influences oviposition behaviour in two pyralid moths, Ephestia cautella and Plodia interpunctella Anim. Behav. 2006 72 545 551 10.1016/j.anbehav.2005.10.023 36. Akhtar Y. Shikano I. Isman M.B. Topical application of a plant extract to different life stages of Trichoplusia ni fails to influence feeding or oviposition behaviour Entomol. Exp. Appl. 2009 132 275 282 10.1111/j.1570-7458.2009.00895.x 37. Anderson P. Sadek M. Larsson M. Hansson B. Thöming G. Larval host plant experience modulates both mate finding and oviposition choice in a moth Anim. Behav. 2013 85 1169 1175 10.1016/j.anbehav.2013.03.002 38. Moreau J. Rahme J. Benrey B. Thiéry D. Larval host plant origin modifies the adult oviposition preference of the female European grapevine moth Lobesia botrana Naturwissenschaften 2008 95 317 324 10.1007/s00114-007-0332-1 18066706 39. Armstrong J.D. de Belle J.S. Wang Z. Kaiser K. Metamorphosis of the mushroom bodies; large-scale rearrangements of the neural substrates for associative learning and memory in Drosophila Learn. Mem. 1998 5 102 114 10454375 40. Ray S. Survival of olfactory memory through metamorphosis in the fly Musca domestica Neurosci. Lett. 1999 259 37 40 10.1016/S0304-3940(98)00892-1 10027550 41. Feltwell J. Large White Butterfly: The Biology, Biochemistry and Physiology of Pieris Brassicae (Linnaeus) Springer Berlin, Germany 1982 42. Le Masurier A.D. Costs and benefits of egg clustering in Pieris brassicae J. Anim. Ecol. 1994 63 677 685 10.2307/5233 43. Morris R. Fellowes M. Learning and natal host influence host preference, handling time and sex allocation behaviour in a pupal parasitoid Behav. Ecol. Sociobiol. 2002 51 386 393 44. Hérard F. Keller M. Lewis W.J. Tumlinson J. Beneficial arthropod behavior mediated by airborne semiochemicals J. Chem. Ecol. 1988 14 1583 1596 10.1007/BF01012524 24276431 45. Gols R. Veenemans C. Potting R.P.J. Smid H.M. Dicke M. Harvey J.A. Bukovinszky T. Variation in the specificity of plant volatiles and their use by a specialist and a generalist parasitoid Anim. Behav. 2012 83 1231 1242 10.1016/j.anbehav.2012.02.015 46. Gandolfi M. Mattiacci L. Dorn S. Preimaginal learning determines adult response to chemical stimuli in a parasitic wasp Proc. R. Soc. B: Biol. Sci. 2002 270 2623 2629 10.1098/rspb.2003.2541 14728786 47. Rossbach A. Löhr B. Vidal S. Does a specialist parasitoid adapt to its host on a new host plant? J. Insect Behav. 2006 19 479 495 10.1007/s10905-006-9040-5 48. Bjorksten T. Hoffmann A. Separating the effects of experience, size, egg load, and genotype on host response in Trichogramma (Hymenoptera: Trichogrammatidae) J. Insect Behav. 1998 11 129 148 10.1023/A:1020826818884 49. Bradburne R.P. Mithen R. Glucosinolate genetics and the attraction of the aphid parasitoid Diaeretiella rapae to Brassica Proc. R. Soc. B: Biol. Sci. 2000 267 89 95 10.1098/rspb.2000.0971 10670958 50. Blande J. Pickett J. Poppy G. A comparison of semiochemically mediated interactions involving specialist and generalist Brassica-feeding aphids and the braconid parasitoid Diaeretiella rapae J. Chem. Ecol. 2007 33 767 779 10.1007/s10886-007-9264-7 17333371 51. Geervliet J.B. Vreugdenhil A.I. Dicke M. Vet L.E.M. Learning to discriminate between infochemicals from different plant-host complexes by the parasitoids Cotesia glomerata and C. rubecula Entomol. Exp. Appl. 1998 86 241 252 10.1046/j.1570-7458.1998.00286.x 52. Gols R. van Dam N.M. Raaijmakers C.E. Dicke M. Harvey J.A. Are population differences in plant quality reflected in the preference and performance of two endoparasitoid wasps? Oikos 2009 118 733 742 10.1111/j.1600-0706.2008.17231.x 53. Bukovinszky T. Gols R. Posthumus M. Vet L.E.M. van Lenteren J.C. Variation in plant volatiles and attraction of the parasitoid Diadegma semiclausum (Hellen) J. Chem. Ecol. 2005 31 461 480 10.1007/s10886-005-2019-4 15898495 54. Fahey J.W. Zalcmann A.T. Talalay P. The chemical diversity and distribution of glucosinolates and isothiocyanates among plants Phytochemistry 2001 56 5 51 10.1016/S0031-9422(00)00316-2 11198818 55. Halkier B.A. Gershenzon J. Biology and biochemistry of glucosinolates Annu. Rev. Plant Biol. 2006 57 303 333 10.1146/annurev.arplant.57.032905.105228 16669764 56. Gols R. Bullock J.M. Dicke M. Bukovinszky T. Harvey J.A. Smelling the wood from the trees: Non-linear parasitoid responses to volatile attractants produced by wild and cultivated cabbage J. Chem. Ecol. 2011 37 795 807 10.1007/s10886-011-9993-5 21748300 57. Read D.P. Feeny P.P. Root R.B. Habitat selection by the aphid parasite Diaeretiella rapae (Hymenoptera: Braconidae) and hyperparasite Charips brassicae (Hymenoptera: Cynipidae) Can. Entomol. 1970 102 1567 1578 10.4039/Ent1021567-12 58. Kester K. Barbosa P. Postemergence learning in the insect parasitoid, Cotesia congregata (Say)(Hymenoptera: Braconidae) J. Insect Behav. 1991 4 727 742 10.1007/BF01052227 59. Geervliet J.B. Vet L.E.M. Dicke M. Volatiles from damaged plants as major cues in long-range host-searching by the specialist parasitoid Cotesia rubecula Entomol. Exp. Appl. 1994 73 289 297 10.1111/j.1570-7458.1994.tb01866.x 60. Cusumano A. Weldegergis B.T. Colazza S. Dicke M. Fatouros N. Attraction of egg-killing parasitoids toward induced plant volatiles in a multi-herbivore context Oecologia 2015 179 163 174 10.1007/s00442-015-3325-3 25953114

PMC005xxxxxx/PMC5000662.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081264ijms-17-01264ArticleDroplet Digital PCR Based Androgen Receptor Variant 7 (AR-V7) Detection from Prostate Cancer Patient Blood Biopsies Ma Yafeng 1Luk Alison 1Young Francis P. 12Lynch David 13Chua Wei 4Balakrishnar Bavanthi 4de Souza Paul 1234Becker Therese M. 123*Marchetti Dario Academic Editor1 Centre for Circulating Tumor Cell Diagnostics and Research, Ingham Institute for Applied Medical Research, 1 Campbell St., Liverpool, NSW 2170, Australia; [email protected] (Y.M.); [email protected] (A.L.); [email protected] (F.P.Y.); [email protected] (D.L.); [email protected] (P.d.S.)2 South Western Clinical School, University of New South Wales, Goulburn St., Liverpool, NSW 2170, Australia3 Western Sydney University Clinical School, Elizabeth St., Liverpool, NSW 2170, Australia4 Department of Medical Oncology, Liverpool Hospital, Elizabeth St & Goulburn St., Liverpool, NSW 2170, Australia; [email protected] (W.C.); [email protected] (B.B.)* Correspondence: [email protected]; Tel.: +61-2-8738-903304 8 2016 8 2016 17 8 126430 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Androgen receptor splice variant V7 (AR-V7) was recently identified as a valuable predictive biomarker in metastatic castrate-resistant prostate cancer. Here, we report a new, sensitive and accurate screen for AR-V7 mRNA expression directly from circulating tumor cells (CTCs): We combined EpCAM-based immunomagnetic CTC isolation using the IsoFlux microfluidic platform with droplet digital polymerase chain reaction (ddPCR) to analyze total AR and AR-V7 expression from prostate cancer patients CTCs. We demonstrate that AR-V7 is reliably detectable in enriched CTC samples with as little as five CTCs, even considering tumor heterogeneity, and confirm detection of AR-V7 in CTC samples from advanced prostate cancer (PCa) patients with AR-V7 detection limited to castrate resistant disease status in our sample set. Sensitive molecular analyses of circulating tumor cells (CTCs) or circulating tumor nucleic acids present exciting strategies to detect biomarkers, such as AR-V7 from non-invasive blood samples, so-called blood biopsies. biomarkerandrogen receptorAR-V7prostate cancerddPCRCTC ==== Body 1. Introduction The insight that cancers, even of the same type, show strong inter- and intra-patient heterogeneity has emerged in recent years. Tumor biomarkers are most commonly conceptualized as specific cellular, biochemical or molecular alterations that characterize heterogeneous subcategories of cancers. Consequently, patient management increasingly relies on the detection of such tumor biomarkers to predict prognosis, guide therapies and monitor treatment response as well as the development of specific resistance mechanisms [1]. The number of identified actionable biomarkers (biomarkers that determine the best type of therapy) and the generation of novel targeted drugs are currently increasing faster than ever, leading to major changes in personalized therapy and clinical praxis. Some serum proteins, such as the prostate specific antigen (PSA) in prostate cancer (PCa) and carcinoembryonic antigen (CEA) in colorectal cancer, are well established biomarkers and clinically widely used, although their correlation with other disease progression parameters is not always ideal [2,3]. More recently, there has been a trend towards analyzing cancer associated gene expression, mutations, amplifications or gene expression variants, which can be very specific, and may predict response and resistance to certain therapies. Some of these assays have already been validated and adopted as part of clinical practice. For instance the detection of v-Raf murine sarcoma viral oncogene homolog B (BRAF) V600E mutation is now decisive for melanoma treatment options that involve targeted BRAF kinase inhibitors, and V-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations in colorectal cancer patients predict resistance to anti-epithelial growth factor receptor (EGFR) monoclonal antibody based therapy (panitumumab and cetuximab) [4,5]. Traditionally, such molecular biomarkers are examined in excised tumor tissue or fine needle biopsies at a single timepoint in the disease course. However, such biopsies are not always available or informative [6]. To progress targeted therapies more broadly into the clinic, it is desirable to detect biomarkers from non-invasive, easily accessible biopsies that are ideally of modest cost. Additionally, repeated biopsies should be feasible to enable continuous monitoring of changes in biomarkers during treatment and development of drug resistance. To that end, blood biopsies are becoming increasingly attractive because various tumor-derived biomarkers can be screened from blood. Tumors release entire circulating tumor cells (CTCs) and circulating tumor nucleic acids (ctDNA and ctRNA) into the blood stream, and screening these for tumor associated genetic changes is becoming increasingly feasible [6,7]. While these kinds of assays still need further validation before they can be adopted into clinical practice, blood-based assays are extremely appealing to clinicians and researchers. Importantly these assays are known to detect evidence of the common tumor heterogeneity, which needs to be accounted for in biomarker analysis [8]. In PCa, genetic changes in the androgen receptor (AR), such as point mutation and gene amplification, render the receptor independent of upstream testosterone levels and cause resistance against androgen deprivation therapy (ADT), which is the predominant first line therapy for advanced disease [9]. Thus far, AR amplification screening using fluorescent in situ hybridization (FISH) and point mutation screening with PCR-based methods from CTC enriched samples have been reported [10,11,12,13]. AR transcriptional variants, for example, AR-V7 and AR-V567es, which encode constitutively active, truncated receptor proteins, cause ligand independent AR activation and are clinically relevant [14]. ADT-drug exposure rapidly induces AR-V7 expression in in vivo models and patient PCa cells, likely to compensate loss of regular AR signaling [9,15]. More importantly, the detection of AR-V7 in PCa CTCs has been correlated with metastatic castrate resistant prostate cancer (CRPC) and resistance against enzalutamide and abiraterone, and potentially superior clinical outcomes for patients on taxane therapy, though response to cabazitaxel has been shown to be independent of AR-V7 status [16,17,18]. Taken together, this suggests that AR-V7 may be a useful biomarker on which to base therapy initiation or therapy changes. Here, we present development of a reliable droplet digital PCR based method to detect AR-V7 and total AR expression in PCa patient CTCs enriched by the IsoFlux system. Our assay has high specificity and sensitivity and detects AR-V7 expression in as little as five AR-V7 PCa cells spiked to produce a modeled CTC sample, and was confirmed in PCa patient CTCs. 2. Results and Discussion 2.1. Assay Optimization Droplet digital PCR (ddPCR), utilizing Taqman PCR principles, is a novel and sensitive method to detect rare mutations, gene expression and copy number variations in samples with limited amounts of nucleic acid templates [19]. To optimize the detection of total AR and AR-V7 transcript expression with ddPCR from total RNA we initially determined the appropriate annealing temperature for our assay using prostate cancer cell line total RNA as basis for ddPCR cDNA template synthesis. At 55 °C, annealing temperature for ddPCR reactions total AR and AR-V7 amplicon containing events (droplets) showed the best separation from empty baseline events and good PCR amplification was achieved (Figure 1). 2.2. Assay Specificity To test assay specificity we determined total AR and AR-V7 expression in a cohort of PCa cell lines: 22Rv1, VCaP, C4-2, LNCaP, C4-2B, LAPC4, and PC3 and the b-lymphocyte line WME-099 (Table 1). Amongst the PCa cell lines, 22Rv1 showed high AR-V7 levels and AR-V7 versus total AR expression ratios (26%). This agrees with previous studies, using immunoblots to discriminate between full length AR and AR-V7 [20,21]. VCaP, known to carry AR gene amplification [22], expectedly expressed the highest level of total AR transcript, which was two to seven fold the level of other AR positive cell lines. LNCaP and its derivates, C4-2 and C4-2B, had consistently less than one AR-V7 copy per cell, and interestingly C4-2 expressed reduced levels of total AR transcript compared to the related C4-2B and parental LNCaP cells. LAPC4 had low levels of total AR expression, as previously reported [23], and no detectable AR-V7. PC3 cells are known to be AR-negative [24] and expectedly expressed neither total AR nor AR-V7 as did the control WME-099 lymphocytes. Similarly, we assayed healthy donor peripheral blood mononuclear cells (PBMCs) (5 male, 1 female sample), which had negligible total AR and no AR-V7 expression, confirming data from normal blood cells in previous reports (Table 2) [13,25]. 2.3. Assay Sensitivity One mammalian cell contains approximately 10–30 pg of total RNA dependent on cell type and physiological state [26]. To test the sensitivity of our new ddPCR assay we titrated down the amount of purified total input RNA from AR-V7 positive 22Rv1 cells to approximately the amount expected from a single cell (2000 down to 15.5 pg) either alone, or by dilution in a constant amount of AR-V7-negative total RNA from WME-099 cells as would be expected in an RNA extract from a CTC sample with residual lymphocytes. As presented in Figure 2, the sensitivity of total AR and AR-V7 detection is able to capture the input RNA expected from a single cell regardless of lymphocyte RNA background. This was also validated in a similar dilution series with RNA template from VCaP cells despite the lower expression of AR-V7 in these cells (data not shown). To more thoroughly define what number of PCa CTCs is required in a typical IsoFlux-processed sample to reliably detect AR-V7 and total AR expression with our AR-V7 assay, we spiked defined numbers of 22Rv1 cells into 4000 PBMCs. This lymphocyte number is based on the average total cell count post IsoFlux CTC enrichment (see Table 3). The resulting samples were processed for ddPCR AR-V7 screening in the same way as IsoFlux enriched patient CTC samples, with Figure 3 demonstrating that our assay reliably detects relevant AR-V7 expression from one spiked cell into 4000 lymphocytes. However this is only true in two out of three replicates. Whether this reflects true technical assay limitations, such as ability to consistently reverse transcribe RNA and synthesize cDNA, or whether in fact not all 22Rv1 cells express AR/AR-V7, remains speculation at present. The latter is supported by our data which show that statistically some cell lines express less than one copy AR-V7 per cell (Table 1), as well as previous AR-V7 detection in only 74% of individual 22Rv1 cells analyzed by quantitative PCR (qPCR) [27]. Thus expression of AR and its variants may be regulated by physiological events, such as cell cycle dependent regulation; indeed AR has been proposed to be transcriptionally repressed by the retinoblastoma protein [28]. This may also help in explaining that although AR reportedly can be lost in vivo in advanced prostate cancer tissue, AR-loss was to our knowledge only found in a heterogeneous manner [29]. Consequently, a conservative interpretation of the assay in our hands is that we can confidently detect AR-V7 status in the presence of at least five prostate cancer CTCs. This allows, even in the presence of tumor cell heterogeneity, reproducible detection of enough ddPCR events positive for total AR, or importantly, AR-V7. Thus, our AR-V7 detection assay with ddPCR has high specificity and sensitivity, sufficient to reliably define AR-V7 expression for most of our advanced PCa patient samples, which have a median CTC number of 32 (range: 3–184, Table 3). By contrast, detection efficiency for other methods has not been documented in detail in most other studies but we note that AR-V7 was detected in samples with a range of one to a minimum of ten CTCs [18,27]. 2.4. AR-V7 Expression in Patient Circulating Tumor Cell (CTC) Samples The ddPCR assay was validated by testing twenty six CTC samples from twenty-four PCa patients after IsoFlux CTC isolation for expression of AR-V7. AR-V7 was detected in 30.8% of CTC samples (8/26): no AR-V7 was detected in any (0/10) of the HSPC (hormone sensitive prostate cancer) samples (note, HSPC patient 3 was with only three CTCs below our conservatively estimated AR-V7 detection limit) whereas AR-V7 was detected in 50% (8/16) of CRPC samples. Detection of AR-V7 in CTC positive patient samples is within the range reported by others for qPCR AR-V7 assays (29%–55%) and above that found by immunocytostaining (18%) [13,16,17,18,27]. As expected, residual lymphocyte counts had no effect on detectability of either form of the AR in our assay. AR-V7 expression in positive patient samples ranged from 8 to 1632 copies per eight mL blood. Patient 18 with eight AR-V7 copies had a CTC count of 47, highlighting tumor cell heterogeneity since clearly not all cells contribute to AR-V7 expression, a finding also observed for patient 19 who had sixteen AR-V7 copies detected from 70 CTCs. The lowest CTC count in a patient sample with detectable AR-V7 was ten (patient 20). Since we detected as many as 104 copies of AR-V7 from only 10 CTCs in this sample, it is likely our assay will detect AR-V7 in samples with considerably lower CTC counts as long as these cells express AR-V7. We will need to analyze larger patient cohorts to confirm this finding, but for now a minimum of five CTCs remains our conservatively estimated detection limit to confidently call a sample AR-V7-negative, based on our 22Rv1 spiking data and expected CTC heterogeneity. Interestingly, for two patients we tested samples taken approximately 3 month apart (patient 16 and 17). Both patients were considered to have CRPC at both time points, but patient 17 changed from AR-V7 negative to positive. He had a PSA of 76.6 at the first CTC sampling but showed disease progression radiologically, as well as resistance to Abiraterone with an increase of his PSA to 193.7 by the time his second sample was taken. Patient data are presented in Table 3. Despite the small patient cohort studied here for method validation, Table 4 summarizes that AR-V7 detection significantly correlated with CRPC (p = 0.008). 3. Materials and Methods 3.1. Cell Lines The human PCa cell lines LNCaP, VCaP, C4-2, C4-2B, PC3, LAPC, 22Rv1, as well as the human b-lymphocyte cell line WME-099, were either recently purchased (ATCC, in vitro Technologies, Lane Cove, Australia) or authenticated (AGRF, Melbourne, Australia). Cells were maintained in RPMI media (Lonza, Basel, Switzerland) supplemented with 10% FBS (Invitrogen, Carlsbad, CA, USA) in a humidified incubator with 5% CO2 at 37 °C. 3.2. Patients Twenty-four patients diagnosed with high risk PCa and positive for CTCs were recruited at Liverpool Hospital. Of the twenty-four patients, twenty-two had metastatic disease, with the remaining two having biochemical recurrent disease (Gleason grade 7 or greater). All patients and healthy blood donors gave informed consent to participate in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the South Western Sydney Local Health District Ethics Committee (Ref: HREC/13/LPOOL/158, 2nd September 2013). Whole blood from patients or healthy donors was drawn in 9 mL K3E K3EDTA-tubes (Greiner Bio-one, Kremsmünster, Austria) after discarding the first 2 mL blood to avoid contamination with epithelial skin cells. Patients were considered to have CRPC if they had experienced new metastases, progression of metastases or a rise in PSA despite adequate castration serum testosterone levels <1.7 nmol/L [30]. 3.3. CTC Enrichment CTC enrichment was performed with the IsoFlux CTC enrichment kit (Fluxion, San Francisco, CA, USA) according to manufacturer’s instructions. Briefly, PBMCs were separated from 8 mL of blood using 50 mL SepMate tubes and Lymphoprep (STEMCELL, Melbourne, Australia) according to the manufacturer’s instructions. PBMCs were once washed with PBS and transferred into a 1.5 mL Low-Protein-Bind tubes (Eppendorf, Hamburg, Germany) with 40 µL anti-human EpCAM conjugated immunomagnetic beads and 40 µL Fc blocker and incubated at 4 °C with slow agitation for 1.5 h. Cells were then loaded into isolation cartridges and CTCs were isolated using the IsoFlux CTC enrichment system and run using the standard separation protocol (Fluxion). Isolated CTCs were either enumerated or frozen at −80 °C for later down-stream analysis. 3.4. Immunocytostaining and CTC Enumeration CTCs were immunocytostained for the presence of cytokeratin, CD45 and nuclei by Hoechst dye using the IsoFlux CTC enumeration kit (Fluxion) according to the manufacturer’s instructions. CTCs mounted in 24-well glass bottom plates were visualized and scanned with Olympus Ix71 (Olympus, Tokyo, Japan) mounted with an automated stage ProScan III (PRIOR) (10× objective). The exposure time for Hoechst, CD45 and cytokeratin are 2, 200 and 400 ms, respectively. CD45 negative cells with positive Hoechst and cytokeratin staining were considered to be CTCs and counted manually from scanned images. Total cell numbers (i.e., including residual blood cells) were also determined in the Hoechst channel using the Olympus CellSens Software “count and measure” plugin (Tokyo, Japan). 3.5. RNA Extraction and cDNA Synthesis Total RNA from cell lines was extracted with the ISOLATE II RNA Mini Kit (Bioline, Sydney, Australia) and any residual genomic DNA contamination was removed by on-column DNAse I treatment for 15 min. RNA was eluted in 50 µL RNase-free H2O. RNA quality and quantity were measured using the NanoDrop 2000 (Thermo Scientific, Waltham, MA, USA). cDNA synthesis was performed from 1 µg total RNA with the SensiFAST cDNA Synthesis kit (Bioline, Sydney, Australia). Total RNA from IsoFlux CTC samples or healthy control PBMCs was extracted with the RNA purification Micro kit (Norgen Biotek Corp., Thorold, ON, Canada) and double-eluted in a total volume of 30 µL RNase-free H2O. 15 µL of this RNA was converted into cDNA with the SensiFAST cDNA Synthesis kit (Bioline). Healthy donor PBMCs consisted of 4000 cells to mimic IsoFlux CTC samples. 3.6. Droplet Digital PCR (ddPCR) Primers and Taqman probes were designed using NCI primer software (Table 5). ddPCR samples for total AR and AR-V7 were set up with 20 µL reaction mixture containing 10 µL ddPCR Supermix for Probes, no dUTP (Bio-Rad, Hercules, CA, USA), 500 nM of each forward primer (FP) and reverse primer (RP) and 250 nM probe (FAM and HEX). Droplets were generated with 70 µL oil using a QX200 droplet generator (Bio-Rad). Amplification was performed at 95 °C, 10 min; followed by 40 cycles of 94 °C, 30 s and 55 °C (or 60 °C for actin) 1 min using a C1000 Touch thermocycler (Bio-Rad). After amplification, the droplets were read on a QX200 droplet reader (Bio-Rad) and analyzed with QuantaSoft software V1.7.4 (Bio-Rad). The total error calculated by the software was used as the 95% confidence intervals. 3.7. Modeling CTC Samples and Single Cell Micromanipulation Single cells were isolated using the CellCelector (ALS GmbH, Jena, Germany) as described before [31]. In brief, 30 µm capillaries were used to pick live 22Rv1 cells under Brightfield at 20× magnification from 2% BSA-coated glass slides. Selected cells were aspirated with a volume of 20–100 nL and deposited into PCR tubes containing 100 µL RL buffer (ISOLATE II RNA Mini Kit, Bioline) and combined with 250 µL RL buffer containing 4000 healthy donor PBMCs. Samples were processed for RNA extraction and cDNA synthesis, as outlined for IsoFlux CTC samples. 3.8. Statistics To establish the relationship between hormone sensitivity and AR-V7 detection, we performed a Fisher’s Exact Test using SPSS Statistics (IBM) software package version 23 (New York, NY, USA). 4. Conclusions We developed a specific and sensitive ddPCR-based assay to determine AR-V7 expression in PCa CTC samples that can reliably detect AR-V7 expression from CTC positive patient samples. The advantage of the ddPCR based assay is that it determines not only AR-V7 positivity but actual transcript copy numbers and that allows in some instances to detect heterogeneity of AR-V7 gene expression without further separation of individual CTCs from residual lymphocytes and other CTCs after immunomagnetic CTC isolation using the IsoFlux platform. Although we were able to detect AR-V7 from a single cell, we conservatively estimate that AR-V7 can be reliably determined in samples with at least five CTCs, as this takes into account heterogeneity of AR-V7 expression as evident in two of our patient samples (patient 18 and 19). In our small proof of concept patient set, association of AR-V7 detection was significantly associated with development of CRPC and we are currently screening larger patient cohorts to confirm the robustness of our approach and association with clinical parameters. Our assay is likely adaptable for CTCs isolated by other methods or for circulating tumor RNA, which can be isolated from processed plasma samples. Acknowledgments This work was supported by the Cancer Institute New South Wales through the Centre for Oncology Education and Research Translation (CONCERT) and by a Sydney Southwest Local Health District Early Career Research Grant awarded to Wei Chua. Francis Young is recipient of an Ingham Research Institute Honours Scholarship. David Lynch is recipient of an Ingham Research Institute Director’s Ph.D. Scholarship. Human ethics approval, HREC/13/LPOOL/158, was obtained and managed by the CONCERT Biobank. Author Contributions Yafeng Ma was critically involved in project conception, design, experimental procedures and data analysis. Alison Luk, Francis P. Young and David Lynch were critically involved in experimental procedures and data analysis. Wei Chua, Bavanthi Balakrishna and Paul de Souza were involved in project design and patient recruitment. Therese M. Becker was central in project conception, design, and coordination. All authors contributed intellectually to the manuscript and approved the final version. Conflicts of Interest The authors declare no conflict of interest. Abbreviations AR androgen receptor PCa prostate cancer CRPC castrate resistant prostate cancer HSPC hormone sensitive prostate cancer CTC circulating tumor cell ctNA circulating tumor nucleic acid PSA prostate specific antigen CEA carcinoembryonic antigen EGFR epidermal growth factor receptor ADT androgen deprived therapy EpCam epithelial cell adhesion molecule CK cytokeratin PBMC peripheral blood mononuclear cell ddPCR droplet digital PCR FP forward primer RP reverse primer FISH fluorescent in situ hybridysation Figure 1 Annealing temperature optimization. Temperature gradient droplet digital PCR (ddPCR) was used to decide on the optimal annealing temperature for the assay. cDNA derived from 200 pg of cell line RNA was used per ddPCR reaction, multiplexed in the presence of probes and primers for both products. AR-V7 ddPCR products are shown in blue and total AR products in green. (A) Fluorescence product separation from background fluorescence from VCaP cell line AR-V7 FAM and total AR HEX reactions with indicated annealing temperatures is compared in 1-D graph presentation; (B) 2D separation of PCR products from VCaP; and (C) 22Rv1 prostate cancer (PCa) cell lines are depicted for optimized annealing temperature of 55 °C. Figure 2 Sensitivity of AR-V7 detection. (A,B) A dilution series of 22Rv1 input RNA to assay AR-V7 (FAM, blue) and total AR (HEX, green); (A) 1D ddPCR graph and (B) copy numbers; (C,D) a dilution series of 22Rv1 input RNA in 2000 pg lymphocyte WME-099 RNA to assay AR-V7 (FAM, blue) and AR (HEX, green); (C) 1D ddPCR graph and (D) copy numbers. Error bars represent 95% confidence intervals. In 1D dotblots, samples with the same amount of 22Rv1 input RNA are separated by black dotted lines, duplicates by yellow dotted lines; RNA concentration inputs within the range predicted for a single cell, ~10–30 pg, are highlighted in pink. Figure 3 Modeled CTC samples. Prostate cancer CTC samples were modeled by spiking indicated numbers of 22Rv1 cells into 4000 lymphocytes from healthy donors by micro-manipulation using the CellCelector (ALS, Jena, Germany). Data was derived from three independent experiments. (A) Representative 2D ddPCR plots from spike-in experiment (0, 1, 5, 10 22Rv1 cells, black circles highlight AR-V7 and total AR events); (B) copy numbers of total AR and AR-V7 dependent on 22Rv1 cell number. ijms-17-01264-t001_Table 1Table 1 Expression levels and ratio of AR-V7 and total AR in cell lines. 22Rv1 VCaP C4-2 LNCaP C4-2B LAPC4 PC3 WME099 AR-V7 (copies/cell) 4.6 1.2 0.5 0.4 0.4 0.0 0.0 0.0 AR-V7 95% CI 2.6–6.6 0.8–1.8 0.2–0.9 0.1–0.8 0.1–0.7 0.0–0.2 0.0–0.3 0.0–0.2 Total-AR (copies/cell) 17.6 97.0 24.7 44.4 47.8 13.4 0.0 0.1 Total-AR 95% CI 14.4–21.0 82.7–111.4 22.3–27.0 38.9–49.9 44.4–51.1 11.5–15.2 0.0–0.3 0.0–0.4 V7/AR (%) 26.0 1.3 1.9 0.9 0.7 0.0 0.0 0.0 Copies/cell were calculated from copies/µL droplet digital PCR (ddPCR) reaction data by accounting for the cDNA input corresponding to 500 pg RNA, and assuming 30 pg RNA/cell; CI: confidence interval. ijms-17-01264-t002_Table 2Table 2 Negligible expression levels of AR-V7 and total AR in blood cells. PBMC-1 PBMC-2 PBMC-3 PBMC-4 PBMC-5 PBMC-6 AR-V7 (copies/cell) 0 0 0 0 0 0 AR-V7 95% CI 0–0.004 0–0.003 0–0.003 0–0.003 0–0.003 0–0.002 Total-AR (copies/cell) 0 0 0.002 0 0.001 0 Total-AR 95% CI 0–0.004 0–0.003 0–0.00 0–0.003 0–0.006 0–0.003 Four thousand peripheral blood mononuclear cells (PBMCs) per healthy donor were processed the same way as patient circulating tumor cell (CTC) samples to determine background AR-V7 and total AR expression. Copies/cell was calculated from copies/µL ddPCR reaction data by accounting for the cDNA input; CI: confidence interval. ijms-17-01264-t003_Table 3Table 3 Patient data. Hormone Sensitivity Patient AR-V7 Copies Total AR Copies %AR-V7 of Total AR CTC Count Total Cell Number HSPC 1 0 96 0 31 2700 2 0 0 n/a 7 2897 3 * 0 0 n/a 3 6919 4 0 0 n/a 7 6800 5 0 40 0 9 3848 6 0 0 n/a 56 3400 7 0 80 0 8 3380 8 0 0 n/a 7 2732 9 0 8 0 6 5464 10 0 24 0 65 7366 CRPC 11 0 296 0 25 3182 12 0 360 0 28 6229 13 0 0 n/a 102 3997 14 0 88 0 35 1566 15 0 16 0 184 3224 16a 0 24 0 82 3715 16b 0 960 0 81 2058 17a 0 0 n/a 122 1163 17b 32 1152 2.5 12 3686 18 8 1000 0.8 47 1505 19 16 768 2.3 70 1820 20 104 5336 1.9 10 8900 21 264 37,008 0.7 39 1418 22 360 20,880 1.7 44 4600 23 880 153,120 0.6 12 2077 24 1632 74,824 2.2 56 4434 AR-V7 and total AR are normalized from template input to represent copy number per 8 mL blood sample. Patient 16 and 17 had consecutive samples evaluated (b) was analyzed ~3 months following sample (a). n/a: not applicable; * note, patient 3 has only 3 detected CTCs and is, thus, below our conservatively estimated AR-V7 detection limit. ijms-17-01264-t004_Table 4Table 4 AR-V7 status of circulating tumor cells (CTCs) correlates to hormone resistance. AR-V7 HSPC CRPC Total +ve 0 8 8 −ve 10 8 18 total 10 16 26 Association of AR-V7 with CRPC is statistically significant p = 0.008. Two of fourteen CRPC patients were analyzed at two time points (three month intervals) with one of them changing from AR-V7 negative to positive for the second time point. (total = total sample number; +ve: positive; −ve: negative). ijms-17-01264-t005_Table 5Table 5 Primers and probes. AR-Species Primers Probes Total AR FP: 5′-GGAATTCCTGTGCATGAAAGC-3′ 5′-[HEX]CTTCAGCATTATTCCAGTG[BHQ1]-3′ RP: 5′-CGATCGAGTTCCTTGATGTAGTTC-3′ AR-V7 FP: 5′-CGGAAATGTTATGAAGCAGGGATGA-3′ 5′-[6FAM]CGGAATTTTTCTCCCAGA[BHQ1]-3′ RP: 5′-CTGGTCATTTTGAGATGCTTGCAAT-3′ ==== Refs References 1. Gerlinger M. Rowan A.J. Horswell S. Larkin J. Endesfelder D. Gronroos E. Martinez P. Matthews N. Stewart A. Tarpey P. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing N. Engl. J. Med. 2012 366 883 892 10.1056/NEJMoa1113205 22397650 2. Gonzalez-Pons M. Cruz-Correa M. Colorectal cancer biomarkers: Where are we now? BioMed Res. Int. 2015 2015 1 14 10.1155/2015/149014 26106599 3. Pezaro C. Woo H.H. Davis I.D. Prostate cancer: Measuring PSA Intern. Med. J. 2014 44 433 440 10.1111/imj.12407 24816306 4. Chapman P.B. Hauschild A. Sondak V.K. Frontline approach to metastatic braf-mutant melanoma diagnosis, molecular evaluation, and treatment choice Am. Soc. Clin. Oncol. Educ. 2014 34 e412 e421 10.14694/EdBook_AM.2014.34.e412 24857132 5. Hecht J.R. Mitchell E. Chidiac T. Scroggin C. Hagenstad C. Spigel D. Marshall J. Cohn A. McCollum D. Stella P. A randomized phase iiib trial of chemotherapy, bevacizumab, and panitumumab compared with chemotherapy and bevacizumab alone for metastatic colorectal cancer J. Clin. Oncol. 2009 27 672 680 10.1200/JCO.2008.19.8135 19114685 6. Crowley E. di Nicolantonio F. Loupakis F. Bardelli A. liquid biopsy: Monitoring cancer-genetics in the blood Nat. Rev. Clin. Oncol. 2013 10 472 484 10.1038/nrclinonc.2013.110 23836314 7. Becker T.M. Caixeiro N.J. Lim S.H. Tognela A. Kienzle N. Scott K.F. Spring K.J. de Souza P. New frontiers in circulating tumor cell analysis: A reference guide for biomolecular profiling toward translational clinical use Int. J. Cancer 2014 134 2523 2533 10.1002/ijc.28516 24122526 8. Zhang C. Guan Y. Sun Y. Ai D. Guo Q. Tumor heterogeneity and circulating tumor cells Cancer Lett. 2016 374 216 223 10.1016/j.canlet.2016.02.024 26902424 9. Yu Z. Chen S. Sowalsky A.G. Voznesensky O.S. Mostaghel E.A. Nelson P.S. Cai C. Balk S.P. Rapid induction of androgen receptor splice variants by androgen deprivation in prostate cancer Clin. Cancer Res. 2014 20 1590 1600 10.1158/1078-0432.CCR-13-1863 24449822 10. Shaffer D.R. Leversha M.A. Danila D.C. Lin O. Gonzalez-Espinoza R. Gu B. Anand A. Smith K. Maslak P. Doyle G.V. Circulating tumor cell analysis in patients with progressive castration-resistant prostate cancer Clin. Cancer Res. 2007 13 2023 2029 10.1158/1078-0432.CCR-06-2701 17404082 11. Leversha M.A. Han J. Asgari Z. Danila D.C. Lin O. Gonzalez-Espinoza R. Anand A. Lilja H. Heller G. Fleisher M. Fluorescence in situ hybridization analysis of circulating tumor cells in metastatic prostate cancer Clin. Cancer Res. 2009 15 2091 2097 10.1158/1078-0432.CCR-08-2036 19276271 12. Jiang Y. Palma J.F. Agus D.B. Wang Y. Gross M.E. Detection of androgen receptor mutations in circulating tumor cells in castration-resistant prostate cancer Clin. Chem. 2010 56 1492 1495 20581083 13. Steinestel J. Luedeke M. Arndt A. Schnoeller T. Lennerz J.K. Maier C. Cronauer M. Steinestel K. Boegemann M. Schrader A.J. Detecting predictive androgen receptor modifications in circulating prostate cancer cells J. Clin. Oncol. 2015 10.18632/oncotarget.3925 14. Thadani-Mulero M. Portella L. Sun S. Sung M. Matov A. Vessella R.L. Corey E. Nanus D.M. Plymate S.R. Giannakakou P. Androgen receptor splice variants determine taxane sensitivity in prostate cancer Cancer Res. 2014 74 2270 2282 10.1158/0008-5472.CAN-13-2876 24556717 15. Qu Y. Dai B. Ye D. Kong Y. Chang K. Jia Z. Yang X. Zhang H. Zhu Y. Shi G. Constitutively active AR-V7 plays an essential role in the development and progression of castration-resistant prostate cancer Sci. Rep. 2015 5 7654 10.1038/srep07654 25563505 16. Antonarakis E.S. Lu C. Luber B. Wang H. Chen Y. Nakazawa M. Nadal R. Paller C.J. Denmeade S.R. Carducci M.A. Androgen receptor splice variant 7 and efficacy of taxane chemotherapy in patients with metastatic castration-resistant prostate cancer JAMA Oncol. 2015 1 582 591 10.1001/jamaoncol.2015.1341 26181238 17. Antonarakis E.S. Lu C. Wang H. Luber B. Nakazawa M. Roeser J.C. Chen Y. Mohammad T.A. Chen Y. Fedor H.L. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer N. Engl. J. Med. 2014 371 1028 1038 10.1056/NEJMoa1315815 25184630 18. Onstenk W. Sieuwerts A.M. Kraan J. Van M. Nieuweboer A.J. Mathijssen R.H. Hamberg P. Meulenbeld H.J. de Laere B. Dirix L.Y. Efficacy of cabazitaxel in castration-resistant prostate cancer is independent of the presence of AR-V7 in circulating tumor cells Eur. Urol. 2015 68 939 945 10.1016/j.eururo.2015.07.007 26188394 19. Reid A.L. Freeman J.B. Millward M. Ziman M. Gray E.S. Detection of BRAF-V600E and V600K in melanoma circulating tumour cells by droplet digital PCR Clin. Biochem. 2015 48 999 1002 10.1016/j.clinbiochem.2014.12.007 25523300 20. Sun H. Mediwala S.N. Szafran A.T. Mancini M.A. Marcelli M. Cudc-101, a novel inhibitor of full-length androgen receptor (FLAR) and androgen receptor variant 7 (AR-V7) activity: Mechanism of action and in vivo efficacy Horm. Cancer 2016 7 196 210 10.1007/s12672-016-0257-2 26957440 21. Liu C. Lou W. Zhu Y. Nadiminty N. Schwartz C.T. Evans C.P. Gao A.C. Niclosamide inhibits androgen receptor variants expression and overcomes enzalutamide resistance in castration-resistant prostate cancer Clin. Cancer Res. 2014 20 3198 3210 10.1158/1078-0432.CCR-13-3296 24740322 22. Makkonen H. Kauhanen M. Jaaskelainen T. Palvimo J.J. Androgen receptor amplification is reflected in the transcriptional responses of vertebral-cancer of the prostate cells Mol. Cell. Endocrinol. 2011 331 57 65 10.1016/j.mce.2010.08.008 20728506 23. Azad A.A. Volik S.V. Wyatt A.W. Haegert A. Le Bihan S. Bell R.H. Anderson S.A. McConeghy B. Shukin R. Bazov J. Androgen receptor gene aberrations in circulating cell-free DNA: Biomarkers of therapeutic resistance in castration-resistant prostate cancer Clin. Cancer Res. 2015 21 2315 2324 10.1158/1078-0432.CCR-14-2666 25712683 24. Yu Y. Zhang Y. Guan W. Huang T. Kang J. Sheng X. Qi J. Androgen receptor promotes the oncogenic function of overexpressed Jagged1 in prostate cancer by enhancing cyclin B1 expression via Akt phosphorylation Mol. Cancer Res. 2014 12 830 842 10.1158/1541-7786.MCR-13-0545 24574517 25. Mantalaris A. Panoskaltsis N. Sakai Y. Bourne P. Chang C. Messing E.M. David Wu J.H. Localization of androgen receptor expression in human bone marrow J. Pathol. 2001 193 361 366 10.1002/1096-9896(0000)9999:9999<::AID-PATH803>3.0.CO;2-W 11241417 26. Sambrook J. Russell D.W. Molecular Cloning—A Laboratory Manual 3rd ed. Cold Spring, Harbor Laboratory Press Cold Spring Harbor, NY, USA 2001 27. Scher H.I. Lu D. Schreiber N.A. Louw J. Graf R.P. Vargas H.A. Johnson A. Jendrisak A. Bambury R. Danila D. Association of AR-V7 on circulating tumor cells as a treatment-specific biomarker with outcomes and survival in castration-resistant prostate cancer JAMA Oncol. 2016 10.1001/jamaoncol.2016.1828 27262168 28. Sharma A. Yeow W.S. Ertel A. Coleman I. Clegg N. Thangavel C. Morrissey C. Zhang X. Comstock C.E. Witkiewicz A.K. The retinoblastoma tumor suppressor controls androgen signaling and human prostate cancer progression J. Clin. Investig. 2010 120 4478 4492 10.1172/JCI44239 21099110 29. Zhang X. Morrissey C. Sun S. Ketchandji M. Nelson P.S. True L.D. Vakar-Lopez F. Vessella R.L. Plymate S.R. Androgen receptor variants occur frequently in castration resistant prostate cancer metastases PLoS ONE 2011 6 1264 10.1371/journal.pone.0027970 22114732 30. Scher H.I. Morris M.J. Stadler W.M. Higano C. Basch E. Fizazi K. Antonarakis E.S. Beer T.M. Carducci M.A. Chi K.N. Trial design and objectives for castration-resistant prostate cancer: Updated recommendations from the prostate cancer clinical trials working group 3 J. Clin. Oncol. 2016 34 1402 1418 10.1200/JCO.2015.64.2702 26903579 31. Neumann M.H. Schneck H. Decker Y. Schomer S. Franken A. Endris V. Pfarr N. Weichert W. Niederacher D. Fehm T. Isolation and characterization of circulating tumor cells using a novel workflow combining the cellsearch(r) system and the cellcelector Biotechnol. Prog. 2016 10.1002/btpr.2294 27126369

PMC005xxxxxx/PMC5000663.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081265ijms-17-01265ReviewThe Role of Dietary Inflammatory Index in Cardiovascular Disease, Metabolic Syndrome and Mortality Ruiz-Canela Miguel 123*Bes-Rastrollo Maira 123Martínez-González Miguel A. 123Flood Vicki Academic Editor1 Department of Preventive Medicine and Public Health, University of Navarra, Pamplona 31008, Spain; [email protected] (M.B.-R.); [email protected] (M.A.M.-G.)2 IDISNA (Navarra Health Research Institute), Pamplona 31008, Spain3 Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y Nutrición (CIBERobn), Instituto de Salud Carlos III, Madrid 28029, Spain* Correspondence: [email protected]; Tel.: +34-948-425-60003 8 2016 8 2016 17 8 126514 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Inflammation is an underlying pathophysiological process in chronic diseases, such as obesity, type 2 diabetes mellitus and cardiovascular disease. In fact, a number of systematic reviews have shown the association between inflammatory biomarkers, such as CRP, IL-1β, IL-6, TNF-α, IL-4, or IL-10, and cardio-metabolic diseases. Diet is one of the main lifestyle-related factors which modulates the inflammatory process. Different individual foods and dietary patterns can have a beneficial health effect associated with their anti-inflammatory properties. The dietary inflammatory index (DII) was recently developed to estimate the inflammatory potential of overall diet. The aim of this review is to examine the findings of recent papers that have investigated the association between the DII, cardio-metabolic risk factors and cardiovascular disease. The relevance of the DII score in the association between inflammation and cardio-metabolic diseases is critically appraised, as well as its role in the context of healthy dietary patterns. We conclude that the DII score seems to be a useful tool to appraise the inflammatory capacity of the diet and to better understand the relationships between diet, inflammation, and cardio-metabolic diseases. dietary inflammatory indexinflammationcardiovascular diseasetype-2 diabetesmetabolic syndromemortality ==== Body 1. Introduction Inflammation is now widely believed to be a cause of atherosclerosis [1,2]. Moreover, inflammation is an underlying pathophysiological mechanism in many other chronic diseases, including obesity [3] and type 2 diabetes mellitus [4], in addition to cardiovascular disease (CVD) [5]. Multiple factors contribute to this inflammatory process, including age, sex, physical activity, smoking, the use of certain medications, and diet [6]. Therefore, lifestyle and diet have a significant impact on health which is based, at least in part, on its association with inflammation [7,8]. Diet is a major determinant of inflammation and a number of markers can be used to assess the inflammation in human nutrition studies [9,10]. Pro-inflammatory biomarkers, such as tumor necrosis factor-α (TNF-α), C-reactive protein (CRP), or cell adhesion molecules have been used as indices of the effect of dietary patterns on low-grade inflammatory status [11,12,13,14,15]. In fact, the inverse association of a Mediterranean-style diet with major chronic disease is partially attributed to the anti-inflammatory properties of some of their foods such as fruits, extra-virgin olive oil, red wine, or nuts [16,17], and some of their bioactive components, such as polyphenols [18]. In contrast, certain components of diet, such as red meat and processed foods, are considered to be pro-inflammatory stimulants [19,20]. Currently, there is an increasing interest in the anti-inflammatory properties of the food patterns for the prevention of cardio-metabolic and other chronic diseases [21]. The characterization of diet according to its inflammatory properties can be useful to investigate the link between diet and CVD. The dietary inflammatory index (DII) was developed to estimate the inflammatory potential of the overall food pattern [22]. The aim of this review is to examine the findings of all the studies that have investigated the association between the DII score, cardio-metabolic risk factors, and CVD. 2. Methods In May 2016, we conducted a search in PubMed and afterwards a hand-search of all references included in the identified articles. The search strategy included the terms “dietary inflammatory index”, “cardiovascular disease”, “metabolic syndrome”, and “death” or “mortality”. The eligibility criteria included any observational epidemiologic study, either cross-sectional or prospective, which had used the modified dietary inflammatory index designed in 2013 by Shivappa et al. [22], and the estimation of a multivariable-adjusted relative risk (and 95% confidence interval) comparing quantiles of the DII score with respect to the risk of CVD, metabolic syndrome or death, either as a primary or as a secondary outcome. 3. The Dietary Inflammatory Index (DII) Score The DII is a score used to determine the overall inflammatory potential of diet. This index is based on 1943 articles, published from 1950 and 2010, reporting the effect of 45 dietary parameters on six inflammatory biomarkers [22]. Each one of these dietary parameters received a positive score (+1) if its effect was pro-inflammatory (significantly increased IL-1β, IL-6, TNF-α, or CRP, or decreased IL-4 or IL-10), a negative score (−1) if its effect was anti-inflammatory and 0 if no significant change in biomarkers associated to that dietary parameter was found. Individuals’ intake of each food parameter was subtracted from a world global standard database and then divided by the world standard deviation for each food parameter. These values were converted to a percentile score, each percentile was doubled, and then 1 was subtracted to achieve a symmetrical distribution (from −1 to +1 and centered on 0). Afterwards, each one of these values was multiplied by the overall food parameter specific inflammatory score. Finally the sum of all the food parameter-specific DII scores provided the overall DII score for each individual. Thus, positive DII scores represent a pro-inflammatory diet and negative DII scores represent an anti-inflammatory diet. For example, this score can have values ranging from 7.98 to −8.87 in different scenarios in the design and developing of the DII score [22]. Several studies have shown the association between the DII score and inflammatory biomarkers, including the rate of telomere shortening [20,21,22,23]. In a longitudinal analysis of 559 healthy participants, higher DII scores were associated with values of high-sensitivity(hs)-CRP greater than 3 mg/L [23]. In a cross-sectional analysis with 2524 healthy participants, the DII score was positively associated with IL-6 (>1.6 pg/mL) and homocysteine (>15 μmol/L), although no significant association was found with high-sensitivity CRP and fibrinogen [24]. The DII score was also associated with IL-6, TNF-α, and hs-CRP among 2567 postmenopausal women [25]. In the PREDIMED-Navarra study, a greater DII score was associated with almost a two-fold higher risk of telomere shortening compared with the anti-inflammatory values of the DII score during a five-year follow-up period [26]. 4. Cardiovascular Disease and DII Score We found four studies assessing the association between the DII score and incident CVD [27,28,29,30], and another study with previous CVD [31]. Table 1 shows a summary of these studies with the relative risk for the main outcome in the most fully-adjusted model. Results were consistent showing in four of the five studies a direct association, meaning that a greater pro-inflammatory diet is related to a higher risk of CVD. There was some heterogeneity regarding several characteristics of the studies, as detailed below in Table 1. There was a variety regarding the number of food parameters used to calculate the DII score as well as the age of participants, duration of follow-up, and definition of cases. However, the relative risk estimates were consistent across the studies. The Geelong osteoporosis study was a population-based study including 1363 men with an age of 18 or older. Among them, 76 had CVD resulting in hospitalization over a period of five years of follow-up [27]. Participants were dichotomized according to the anti-inflammatory or pro-inflammatory capacity of diet (as captured by the DII score). CVD cases were collected retrospectively from medical records in 2011. Compared with participants following an anti-inflammatory diet pattern, the adjusted OR (95% confidence interval) for CVD was 2.00 (1.01–3.96) for those with positive values of the DII score (pro-inflammatory diet). The biggest impact was found when including only those events that occurred during the first three years of follow-up. The PREDIMED study assessed 7216 men (55 to 80 years) and women (60 to 80 years) at high risk of CVD [28]. The number of CVD events (MI, stroke or cardiovascular death) was 277 for a median follow-up of over 4.5 years. Using the lowest quartile of the DII score as the reference, the adjusted hazard ratio (95% confidence interval) for CVD was 1.73 (1.15–2.60) for those participants exposed to the highest quartile (most pro-inflammatory diet at baseline). A stronger association was found when cases occurring during the first year of follow-up were excluded from the analysis. The SUN cohort assessed the association between the DII score and incident CVD in 18,794 middle-age participants with a median follow-up of 8.9 years [29]. The number of new-onset CVD cases was 117 (MI, stroke and CVD death). The adjusted HR for participants in the highest (most pro-inflammatory) vs. the lowest quartile of the DII score was 2.03 (95% CI 1.06–3.88). The estimated risk was higher after excluding those with CVD events occurring after five years of follow-up. Figure 1 and Figure 2 show the cumulative incidence of CVD in the PREDIMED study [28] and the SUN cohort [29] according to the tertiles and quartiles of the DII score, respectively. Both curves indicate that the highest values of the DII score (the most pro-inflammatory diet in red color) had a higher incidence of CVD compared with lower values during the follow-up of the studies. The SU.VI.MAX study included 7743 women (aged 35–60 years) and men (aged 45–60 years) [30]. A total of 292 CVD cases (MI, stroke and angina pectoris, or revascularization intervention) were included during a mean follow-up of 11.4 years. In this study no statistically significant association between the DII score and the composite CVD outcome was observed (Table 1). However, a significant association was found for MI when comparing the highest vs. the lowest quartile of the DII score and this association was slightly higher when excluding MI cases occurring during the first two years of follow-up. The NHANES study explored the association between the DII score and a composite outcome of 1734 self-reported previously diagnosed CVD cases (including congestive heart failure, coronary heart disease, angina, heart attack, and stroke) [31]. In the analyses for each disease, significant direct associations were found for congestive heart failure, heart attack, stroke, and hypertension, but not for coronary heart disease and angina pectoris. In the stratified analysis by sex, a significant association was found in women, but not men (p for interaction < 0.01). 5. Metabolic Syndrome and the DII Score Regarding the association between DII score and metabolic syndrome (MetSyn), we found four studies (Table 2). No significant association was found in two of them with a cross-sectional design [32,33] and in a prospective study [34]. On the contrary, a significant association was found in a cohort study using a greater number of food parameters to obtain the DII score [35]. No association was found between the DII score and MetSyn in a cross-sectional analysis with 3862 participants from the Polish-Norwegian Study [32]. Participants in this study were men and women between 45 and 64 years and the prevalence of MetSyn was 30%. The HR for metabolic syndrome was 0.96 (95% CI: 0.77–1.19), when comparing quartile 4 vs. 1 of the DII score. Similarly, no significant association was found for the individual components of the MetSyn except for an inverse association between HDL cholesterol and the DII score (Q4 vs. Q1: OR = 0.62; 95% CI: 0.48–0.80). On the contrary, in the stratified analysis by sex, an inverse association was found between DII score and MetSyn among women [32]. This association might be due to reverse causation in this cross-sectional study. A lack of association between the DII score and MetSyn was also found in a cross-sectional study with 464 participants from the Buffalo Cardio-Metabolic Occupational Police Stress study [33]. The mean age of these participants was 42 years and the prevalence of MetSyn was 28%. There was also no association between the DII score and each of the individual components of the MetSyn except for the glucose intolerance component (OR = 2.03; 95% CI: 1.08–3.82 for quartile 4 vs. 1). Again, the cross-sectional design and the small sample size might explain this lack of association. Pimenta et al. studied the association between different dietary indexes and the MetSyn in participants from the SUN cohort [34]. The number of participants with incident MetSyn was 346 among 6851 participants followed during a mean of 8.3 years. They did not find a statistical significant association between the DII score and MetSyn after adjusting for other potential confounders. In the SU.VI.MAX study, 3726 participants (mean age 50 years) were included to assess the association between the DII score and the MetSyn [35]. After a mean follow-up of 12.4 years, 524 (14%) participants developed MetSyn. The odds of MetSyn was 39% higher among those with the greatest pro-inflammatory diet (quartile 4 of the DII score) compared with those with the most anti-inflammatory diet (quartile 1). A higher DII score was also significantly associated with higher diastolic and systolic blood pressure, higher triglycerides levels and lower HDL-cholesterol. Finally, in a cross-sectional analysis with 7236 participants from the PREDIMED study, a direct association was found between higher (pro-inflammatory) levels of the DII score and waist circumference which is one of the components of the MetSyn [36]. 6. Mortality and the DII Score Table 3 shows five studies studying the association between the DII score and all-cause mortality [37,38,39,40,41]. In all studies except one, a higher risk of all-cause death was found among those participants with the highest pro-inflammatory values of the DII score. Using the third National Health and Nutrition Examination Survey (NHANES III), Shivappa et al. found a direct association between a higher pro-inflammatory diet and mortality in participants with ages above 19 years [37]. In the same study, significant associations were also found between the DII score and all-cancer mortality (HR = 1.46; 95% CI: 1.10–1.96), digestive-tract cancer mortality (HR = 2.10; 95% CI: 1.15–3.84), and CVD mortality (HR = 1.46; 95% CI: 1.18, 1.81). Similar results were found in another analysis with participants from the NHANES III after stratifying by the diabetic status of participants [38]. In this case, the highest risk of all-cause mortality was found among participants with prediabetes (serum HbA1c between 5.7% and 6.4%), as it is shown in Table 3. The Iowa Women’s Health study found a more modest association between the DII score and total mortality [39]. This analysis included 37,525 women aged 55–69 at baseline and with a mean follow-up of 20.7 years. There were also significant and direct associations between a higher pro-inflammatory level in the DII score and cancer mortality (for all cancers and for digestive cancers), CVD mortality, and chronic obstructive pulmonary disease related mortality. Interestingly, the SU.VI.MAX study did not find any significant association between the DII score and all-cause mortality [40]. During a median follow-up of 12.4 years, 207 out of 8089 participants died. The DII score was positively associated with CVD mortality (HR = 1.53; 95% CI: 1.01–2.32) and cancer mortality (HR = 1.83; 95% CI: 1.12, 2.99) but no association was found with all-causes mortality (HR = 1.41; 95% CI: 0.97–2.04). However, in the stratified analysis by intervention group, a statistical association was found between the DII score and all-cause mortality in the placebo group but not in the antioxidant-supplemented group (Table 3). Finally, the Swedish Mammography Cohort including a large sample size (n = 33,747) found a direct association of the DII score with all-cause mortality as well as with CVD mortality (HR = 1.35; 95% CI: 1.01–1.81) when comparing extreme quintiles [41]. However, no significant association was found for cancer mortality or digestive-cancer mortality. 7. The Relevance of the DII Score in the Association between Inflammation and Cardio-Metabolic Diseases There is a clear association between the inflammatory biomarkers used to calculate the DII score (CRP, IL-1β, IL-6, TNF-α, IL-4, and IL-10) [22] and cardio-metabolic diseases such as obesity, diabetes, or CVD. Specifically, higher levels of CRP were correlated with BMI, waist circumference, or waist-to-hip ratio in a meta-analysis with 53 cross-sectional studies [42]. A cohort study with women also found this association [43]. Though two Mendelian randomization studies have suggested that probably adiposity causes inflammation but not vice versa [44,45], the bidirectional association is possible and a pro-inflammatory dietary exposure can precede the development of obesity [46,47]. Recently, a review on hypothalamic micro inflammation suggested a potential mechanism link between a pro-inflammatory diet and MetSyn [48]. Regarding diabetes, a meta-analysis with 16 studies found an association between hs-CRP and incident diabetes [49] although a nested case-control study in the European Prospective Investigation of Cancer (EPIC)-Norfolk cohort suggested that this association might be confounded by central adiposity [49]. With respect to CVD, higher hs-CRP was independently associated with carotid intima-media thickness [50]. An individual participant meta-analysis from 54 long-term prospective studies reported an association between hs-CRP concentration and the risk of coronary heart disease, stroke and both vascular and non-vascular mortality [51]. However, this association is not completely admitted because a Mendelian association meta-analysis of individual participant data did not confirm it [52]. However, Mendelian randomization studies are not free of biases and may have some important limitations mainly derived from their underlying assumptions that are not always valid [53]. Although CRP is clinically relevant for risk prediction, it is considered a surrogate biomarker of upstream cytokines, such as IL-6 and IL-1β [54]. A systematic review found that MetSyn is associated with elevated concentrations of IL-6 and TNF-α and with decreased levels of IL-10, an anti-inflammatory biomarker used in the calculation of the DII score [55]. However, this association between IL-10 and diabetes is still less studied and there are no available prospective studies to support this association [56]. The Leiden 85-Plus Study showed a cross-sectional association between a low IL-10 production capacity and both MetSyn and type-2 diabetes [57]. Regarding coronary disease, a meta-analysis of 29 prospective studies found that higher baseline levels of pro-inflammatory cytokines, including IL-6 and TNF-α, were associated with a greater risk of non-fatal myocardial infarction or CVD death [58]. Two mendelian randomization studies have suggested IL-6 receptors to have a causal role on coronary heart disease [59,60]. Although anti-inflammatory IL-10 was traditionally considered protective against atherosclerosis [61], a prospective study found that circulating IL-10 concentrations at baseline were positively associated with nonfatal myocardial infarction, stroke, or CVD death [62]. Similarly, a growing body of evidence suggests that IL-4, previously considered an anti-inflammatory cytokine, could be involved in the initiation and progression of atherosclerosis [63]. In general, all this scientific evidence is highly relevant to explain the association between dietary exposures and CVD, MetSyn, and mortality, which can be mediated by a pro-inflammatory exposure through the food pattern. As previously mentioned, the DII score seems to be able to capture this exposure because it is based on the role that 45 foods and dietary constituents have on six well-acknowledged inflammatory biomarkers [22]. Among these biomarkers, those with a pro-inflammatory effect (hs-CRP, IL-1β, IL-6 and TNF-α) are associated with cardiometabolic risk factors and CVD, as we have explained above. There is still a limited knowledge regarding the role of IL-10 and IL-4 and this is probably a limitation of the DII score. The construction of the DII score is based on scientific knowledge available until 2010 and thus, an update will be needed to refine the inflammatory capacity of these cytokines as well as other new inflammatory markers. 8. The DII Score in the Context of Healthy Dietary Patterns The use of dietary patterns is one of the best approaches to understand the relationship between diet and disease [64]. Dietary patterns are used to assess the general quality of a diet and takes into account the potential synergies between different foods [65]. One method for defining dietary patterns is the construction of dietary indices according to some specific dietary recommendations. For example, the Healthy Eating Index (HEI) followed the Dietary Guidelines for Americans 2005 [66] and it has been adapted according to changes in the dietary recommendations [67,68]. Other indices are more focused on some specific health benefit such as the Dietary Approaches to Stop Hypertension (DASH) diet [69], and other indices try to assess the level of adherence to traditional dietary patterns, like the Mediterranean diet (MedDiet) [70] or the vegetarian dietary pattern. Both the DASH and traditional MedDiet are the most well-studied dietary patterns [71]. All of these healthy dietary patterns share the protective effect against the most prevalent diseases such as cancer, cardiovascular diseases or diabetes. A meta-analysis of cohort studies found that the HEI, AHEI, and the DASH score were associated with a risk reduction for all-cause mortality, CVD, cancer, and type-2 diabetes [72]. The protective role of the MedDiet against these health outcomes has also been shown consistently in several systematic reviews [73,74]. Two systematic reviews [75,76] concluded that both the MedDiet and the DASH diet are effective for weight loss. The large and long-term PREDIMED randomized trial (www.predimed.es) recently also reported benefits for the Mediterranean diet for the prevention of age-related weight gain [77]. According to the 2015 Dietary Guidelines Advisory Committee (http://health.gov/dietaryguidelines/2015-scientific-report/) those dietary patterns associated with a decreased risk of CVD are characterized by a lower consumption of red and processed meat, lower intakes of sugar-sweetened foods and beverages and refined grains as well as higher consumption of fruits, vegetables, whole grains, low-fat dairies, and seafood [78]. In fact, several meta-analyses have shown the link between these individual foods and better cardiometabolic outcomes [71]. The DII score is inversely associated with healthy scores including the AHEI, HEI-2010, and DASH [79]. This is consistent with the fact that lower hs-CRP concentrations are associated with a higher intake of fruits and vegetables [80,81], legumes [82], nuts [83], and dietary fiber [84]. Western-style diets are positively associated with hs-CRP levels, whereas healthy diets are inversely associated with it [85,86]. A systematic review showed that higher adherence to the Mediterranean diet was associated with decreased levels of inflammatory biomarkers, including hs-CRP, interleukin-6, and intracellular adhesion molecule-1 [14]. Similarly, olive oil (a distinctive characteristic of the MedDiet) specially in its extra-virgin variety may also exert a powerful anti-inflammatory effect [87]. The DII score, in contrast to other dietary patterns, is focused on a pathophysiological process, i.e., how a number of foods modulate several markers involved in inflammation. In fact, the correlation with healthy indexes was not perfect [79] and, thus, the DII score probably accounts for other sources of variability related to inflammation and provides additional information beyond that provided by other dietary patterns. This makes very promising the construction of dietary patterns based on specific biological pathways (i.e., inflammation) and other alternative inflammatory indexes have been developed [88,89]. The strength of the DII score is the high number of studies showing an inverse association with cardiovascular-related diseases, as well as other chronic diseases, such as cancer [90,91,92,93]. 9. Conclusions Several studies have shown the important role of inflammation as a mechanism involved in the pathophysiological process of many chronic diseases. In this context, diet is a modifiable factor which is likely to exert a powerful pro-inflammatory or anti-inflammatory effect. The DII score seems to be a useful tool to appraise the inflammatory capacity of diet. A number of studies have consistently shown a direct association between the DII score and a higher risk of CVD, MetSyn, and overall mortality. The use of this score can be a recommended approach to understand the relationships between diet, inflammation, and cardio-metabolic diseases. Author Contributions Miguel Ruiz-Canela conceived and wrote the first draft of the paper. Maira Bes-Rastrollo and Miguel A. Martínez-González made substantial contributions in the review of the paper. All authors approved the last version of the paper. Conflicts of Interest The authors declare no conflict of interest. Abbreviations CRP C-Reactive Protein DII Dietary Inflammatory Index FFQ Food Frequency Questionnaire HR Hazard Ratio IL Interleukin MedDiet Mediterranean Diet MetSyn Metabolic Syndrome OR Odds Ratio TNF-α Tumor Necrosis Factor-α Figure 1 Cumulative incidence of CVD in the PREDIMED study [28] according to tertiles of the DII score. Figure 2 Cumulative incidence of CVD in the the SUN cohort [29] according to quartiles of the DII score (merging the two intermediate quartiles to build a medium category). The data are adjusted for sex, age, hypertension, dyslipidemia, diabetes, smoking, family history of premature CVD, and total energy intake. ijms-17-01265-t001_Table 1Table 1 Association between the DII score and the incidence of cardiovascular disease. Study Name Design # Food Parameters Follow-up 1 (Years) N Total (CVD Cases) Groups Adjusted Relative Risk (95% CI) Covariables GOS 2 [27] Cohort 22 using FFQ 5 1363 (76) Negative DII (ref) vs. Positive DII OR = 2.00 (1.01–3.96) Family history of CVD, blood pressure, sedentary, diabetes, smoking, waist circumference, age, total energy intake PREDIMED 3 [28] Cohort 32 using FFQ 4.7 7216 (277) Quartile 1 (ref) vs. Quartile 4 HR = 1.73 (1.15–2.60) Age, sex, overweight/obesity, waist-to-height ratio, total energy intake, smoking status, diabetes, hypertension, dyslipidemia, family history of premature cardiovascular disease, physical activity, educational level, intervention group, center SUN 4 [29] Cohort 28 using FFQ 8.9 18,794 (117) Quartile 1 (ref) vs. Quartile 4 HR = 2.03 (1.06–3.88) Age, sex, hypertension, dyslipidaemia, diabetes, smoking status, family history of cardiovascular disease, total energy intake, physical activity, body mass index, educational level, other cardiovascular diseases, special diet at baseline, snacking, average time sitting, average time spent watching television SU.VI.MAX 5 [30] Cohort 36 using 24-h dietary records 11.4 7743 (292) Quartile 1 (ref) vs. Quartile 4 HR = 1.16 (0.79–1.69) Sex, energy intake, supplementation group, number of 24-h records, education level, marital status, smoking status, physical activity, body mass index 11.4 7602 (93) Quartile 1 (ref) vs. Quartile 4 HR = 2.26 * (1.08–4.71) NHANES 6 [31] Cross-sectional 27 using 24-h dietary records NA 15,693 (1734) Quartile 1 (ref) vs. Quartile 4 OR = 1.30 (1.06–1.58) Family member smoking status, personal smoking status, age, body mass index 1 Mean or median except the GOS study; 2 GOS: Geelong Osteoporosis study; 3 PREDIMED: Prevention with Mediterranean Diet (PREvención con DIeta MEDiterránea); 4 SUN: University of Navarra Follow-up (Seguimiento Universidad de Navarra); 5 SU.VI.MAX: Antioxidant Vitamins and Minerals Supplementation (SUpplémentation en VItamines et Minéraux AntioXydants); 6 NHANES: National Health and Nutrition Examination Survey III follow-up; * HR for myocardial infarction in the stratified analysis; #: number; ref: reference. ijms-17-01265-t002_Table 2Table 2 Association between the DII score and the metabolic syndrome 1. Study Name Design # Food Parameters Follow-up (Years) N Total (Cases) Groups Adjusted Relative Risk (95% CI) Covariables PONS 2 [32] Cross-sectional 22 using FFQ NA 3862 (1159) Quartile 1 (ref) vs. Quartile 4 OR = 0.96 (0.77–1.19) Body mass index, age BCOPS 3 [33] Cross-sectional Not reported NA 464 (125) Quartile 1 (ref) vs. Quartile 4 OR = 0.87 (0.46–1.63) Age, sex SUN 4 [34] Cohort 28 using FFQ 8.3 6851 (346) Quintile 1 (ref) vs. Quintile 5 HR * = 0.86 (0.60–1.23) Age, sex, smoking, alcohol consumption, snacking between main meals, use of special diets, television watching, physical activity, changes in weight over the last 5 years prior, body mass index SU.VI.MAX 5 [35] Cohort 36 using 24-h dietary records 12.4 3726 (524) Quartile 1 (ref) vs. Quartile 4 HR = 1.39 (1.01–1.92) Age, sex, supplementation group, number of 24 h records, energy intake, education level, smoking status, physical activity, body mass index 1 MetSyn was defined as the presence of at least three of these components: Abdominal obesity; high blood pressure; low HDL cholesterol; high triglycerides and high glucose level; 2 PONS: Polish-Norwegian Study; 3 BCOPS: Buffalo Cardio-Metabolic Occupational Police Stress; 4 SUN: University of Navarra Follow-up (Seguimiento Universidad de Navarra); 5 SU.VI.MAX: Antioxidant Vitamins and Minerals Supplementation (SUpplémentation en VItamines et Minéraux AntioXydants); #: Number; ref: reference; * Data provided by the authors. ijms-17-01265-t003_Table 3Table 3 Association between the DII score and all-cause mortality. Study Name Design # Food Parameters Follow-up (Years) N Total (Cases) Groups Adjusted Relative Risk (95% CI) Covariables NHANES 1 III [37] Cohort 27 using 24-h dietary records 13.5 12,438 (2795) Tertile 1 (ref) vs. Tertile 3 HR = 1.34 (1.19, 1.51) Age, sex, race, diabetes status, hypertension, physical activity, BMI, poverty index, smoking NHANES 1 III [38] Cohort 27 using 24-h dietary records NA 2681 (896) Tertile 1 (ref) vs. Tertile 3 HR = 1.39 (1.13, 1.72) Age, sex, race, HbA1C, current smoking, physical activity, body mass index, systolic blood pressure Iowa Women’s Health study [39] Cohort 37 using FFQ 20.7 37,525 (17,793) Quartile 1 (ref) vs. Quartile 4 HR = 1.08 (1.03–1.13) Age, body mass index, smoking status, pack-years of Smoking, hormone replacement therapy use, education, diabetes, hypertension, heart disease, cancer, total energy intake SU.VI.MAX 2 [40] Cohort 36 using 24-h dietary records 1.24 8089 (207) Tertile 1 (ref) vs. Tertile 3 HR * = 2.10 (1.15–3.84) Age, sex, intervention group, number of 24-hour dietary records, body mass index, physical activity, smoking status, educational level, family history of cancer in first-degree relatives, family history of CVD in first-degree relatives, energy intake without alcohol, and alcohol intake HR ** = 1.09 (0.67–1.77) Swedish Mammography Cohort [41] Cohort 27 using FFQ 15 33,747 (7095) Quintile 1 (ref) vs. Quintile 5 HR = 1.41 (1.21–1.64) Age, energy intake, body mass index, education, smoking status, physical activity, alcohol intake 1 NHANES: National Health and Nutrition Examination Survey III follow-up; 2 SU.VI.MAX: Antioxidant Vitamins and Minerals Supplementation (SUpplémentation en VItamines et Minéraux AntioXydants); #: number; ref: reference; * HR for the placebo group; ** HR for the antioxidant supplementation group. ==== Refs References 1. Ross R. Atherosclerosis–An inflammatory disease N. Engl. J. Med. 1999 340 115 126 10.1016/S0002-8703(99)70266-8 9887164 2. Frostegård J. Immunity, atherosclerosis and cardiovascular disease BMC Med. 2013 11 1265 10.1186/1741-7015-11-117 23635324 3. Gregor M.F. Hotamisligil G.S. Inflammatory mechanisms in obesity Annu. Rev. Immunol. 2011 29 415 445 10.1146/annurev-immunol-031210-101322 21219177 4. Wellen K.E. Hotamisligil G.S. Inflammation, stress, and diabetes J. Clin. Investig. 2005 115 1111 1119 10.1172/JCI200525102 15864338 5. Hansson G.K. Inflammation, atherosclerosis and coronary artery disease N. Engl. J. Med. 2005 352 1685 1695 10.1056/NEJMra043430 15843671 6. Libby P. Inflammatory mechanisms: The molecular basis of inflammation and disease Nutr. Rev. 2007 65 S140 S146 10.1301/nr.2007.dec.S140-S146 18240538 7. Kantor E.D. Lampe J.W. Kratz M. White E. Lifestyle factors and inflammation: Associations by body mass index PLoS ONE 2013 8 1265 10.1371/journal.pone.0067833 23844105 8. Ahmadi-Abhari S. Luben R.N. Wareham N.J. Khaw K.-T. Distribution and determinants of C-reactive protein in the older adult population: European Prospective Investigation into Cancer-Norfolk study Eur. J. Clin. Investig. 2013 43 899 911 10.1111/eci.12116 23786220 9. Aggarwal B.B. David H. Immunonutrition: Interactions of Diet, Genetics and Inflammation CRC Press Boca Raton, FL, USA 2014 10. Calder P.C. Ahluwalia N. Albers R. Bosco N. Bourdet-Sicard R. Haller D. Holgate S.T. Jönsson L.S. Latulippe M.E. Marcos A. A consideration of biomarkers to be used for evaluation of inflammation in human nutritional studies Br. J. Nutr. 2013 109 S1 S34 10.1017/S0007114512005119 23343744 11. Barbaresko J. Koch M. Schulze M.B. Nöthlings U. Dietary pattern analysis and biomarkers of low-grade inflammation: A systematic literature review Nutr. Rev. 2013 71 511 527 10.1111/nure.12035 23865797 12. Esmaillzadeh A. Kimiagar M. Mehrabi Y. Azadbakht L. Hu F.B. Willett W.C. Dietary patterns and markers of systemic inflammation among Iranian women J. Nutr. 2007 137 992 998 17374666 13. Casas R. Sacanella E. Urpi-Sarda M. Chiva-Blanch G. Ros E. Martínez-González M. Covas M.-I.I. Lamuela-Raventos R.M. Salas-Salvadó J. Fiol M. The effects of the Mediterranean diet on biomarkers of vascular wall inflammation and plaque vulnerability in subjects with high risk for cardiovascular disease. A randomized trial PLoS ONE 2014 9 1265 10.1371/journal.pone.0100084 24925270 14. Schwingshackl L. Hoffmann G. Mediterranean dietary pattern, inflammation and endothelial function: A systematic review and meta-analysis of intervention trials Nutr. Metab. Cardiovasc. Dis. 2014 24 929 939 10.1016/j.numecd.2014.03.003 24787907 15. Neale E.P. Batterham M.J. Tapsell L.C. Consumption of a healthy dietary pattern results in significant reductions in C-reactive protein levels in adults: A meta-analysis Nutr. Res. 2016 36 391 401 10.1016/j.nutres.2016.02.009 27101757 16. Salas-Salvadó J. Garcia-Arellano A. Estruch R. Marquez-Sandoval F. Corella D. Fiol M. Gómez-Gracia E. Viñoles E. Arós F. Herrera C. Components of the Mediterranean-type food pattern and serum inflammatory markers among patients at high risk for cardiovascular disease Eur. J. Clin. Nutr. 2008 62 651 659 10.1038/sj.ejcn.1602762 17440519 17. Urpi-Sarda M. Casas R. Chiva-Blanch G. Romero-Mamani E.S. Valderas-Martínez P. Salas-Salvadó J. Covas M.I. Toledo E. Andres-Lacueva C. Llorach R. The Mediterranean diet pattern and its main components are associated with lower plasma concentrations of tumor necrosis factor receptor 60 in patients at high risk for cardiovascular disease J. Nutr. 2012 142 1019 1025 10.3945/jn.111.148726 22535754 18. Medina-Remón A. Casas R. Tressserra-Rimbau A. Ros E. Martínez-González M. Fitó M. Corella D. Salas-Salvadó J. Lamuela-Raventos R.M. Estruch R. Polyphenol intake from a mediterranean diet decreases inflammatory biomarkers related to atherosclerosis: A sub-study of The PREDIMED trial Br. J. Clin. Pharmacol. 2016 10.1111/bcp.12986 27100393 19. Ley S.H. Sun Q. Willett W.C. Eliassen A.H. Wu K. Pan A. Grodstein F. Hu F.B. Associations between red meat intake and biomarkers of inflammation and glucose metabolism in women Am. J. Clin. Nutr. 2014 99 352 360 10.3945/ajcn.113.075663 24284436 20. Herieka M. Faraj T.A. Erridge C. Reduced dietary intake of pro-inflammatory Toll-like receptor stimulants favourably modifies markers of cardiometabolic risk in healthy men Nutr. Metab. Cardiovasc. Dis. 2016 26 194 200 10.1016/j.numecd.2015.12.001 26803597 21. Marcason W. What is the anti-inflammatory diet? J. Am. Diet. Assoc. 2010 110 1780 10.1016/j.jada.2010.09.024 21034890 22. Shivappa N. Steck S.E. Hurley T.G. Hussey J.R. Hébert J.R. Designing and developing a literature-derived, population-based dietary inflammatory index Public Health Nutr. 2013 1 8 10.1017/S1368980013002115 23941862 23. Shivappa N. Steck S.E. Hurley T.G. Hussey J.R. Ma Y. Ockene I.S. Tabung F. Hébert J.R. A population-based dietary inflammatory index predicts levels of C-reactive protein in the Seasonal Variation of Blood Cholesterol Study (SEASONS) Public Health Nutr. 2014 17 1825 1833 10.1017/S1368980013002565 24107546 24. Shivappa N. Hébert J.R. Rietzschel E.R. De Buyzere M.L. Langlois M. Debruyne E. Marcos A. Huybrechts I. Associations between dietary inflammatory index and inflammatory markers in the Asklepios Study Br. J. Nutr. 2015 1 7 10.1017/S000711451400395X 25639781 25. Tabung F.K. Steck S.E. Zhang J. Ma Y. Liese A.D. Agalliu I. Hingle M. Hou L. Hurley T.G. Jiao L. Construct validation of the dietary inflammatory index among postmenopausal women Ann. Epidemiol. 2015 25 398 405 10.1016/j.annepidem.2015.03.009 25900255 26. Garcia-Calzon S. Zalba G. Ruiz-Canela M. Shivappa N. Hébert J.R. Martínez J.A. Fito M. Gómez-Gracia E. Martínez-González M. Marti A. Dietary inflammatory index and telomere length in subjects with a high cardiovascular disease risk from the PREDIMED-NAVARRA study: Cross-sectional and longitudinal analyses over 5 y Am. J. Clin. Nutr. 2015 102 897 904 10.3945/ajcn.115.116863 26354530 27. O’Neil A. Shivappa N. Jacka F.N. Kotowicz M.A. Kibbey K. Hébert J.R. Pasco J.A. Pro-inflammatory dietary intake as a risk factor for CVD in men: A 5-year longitudinal study Br. J. Nutr. 2015 114 2074 2082 10.1017/S0007114515003815 26450630 28. Garcia-Arellano A. Ramallal R. Ruiz-Canela M. Salas-Salvadó J. Corella D. Shivappa N. Schröder H. Hébert J.R. Ros E. Gómez-Gracia E. Dietary Inflammatory Index and Incidence of Cardiovascular Disease in the PREDIMED Study Nutrients 2015 7 4124 4138 10.3390/nu7064124 26035241 29. Ramallal R. Toledo E.E. Martínez-González M. Hernandez-Hernandez A. Garcia-Arellano A. Shivappa N. Hébert J.R. Ruiz-Canela M. Hernández-Hernández A. García-Arellano A. Dietary inflammatory index and incidence of cardiovascular disease in the SUN cohort PLoS ONE 2015 10 1265 10.1371/journal.pone.0135221 26340022 30. Neufcourt L. Assmann K.E. Fezeu L.K. Touvier M. Graffouillère L. Shivappa N. Hébert J.R. Wirth M.D. Hercberg S. Galan P. Prospective association between the dietary inflammatory index and cardiovascular diseases in the Supplémentation en Vitamines et Minéraux Antioxydants (SU.VI.MAX) Cohort J. Am. Heart Assoc. 2016 5 e002735 10.1161/JAHA.115.002735 27068628 31. Wirth M.D. Shivappa N. Hurley T.G. Hébert J.R. Association between previously diagnosed circulatory conditions and a dietary inflammatory index Nutr. Res. 2016 36 227 233 10.1016/j.nutres.2015.11.016 26923509 32. Sokol A. Wirth M.D. Manczuk M. Shivappa N. Zatonska K. Hurley T.G. Hébert J.R. Association between the dietary inflammatory index, waist-to-hip ratio and metabolic syndrome Nutr. Res. 2016 10.1016/j.nutres.2016.04.004 33. Wirth M.D. Burch J. Shivappa N. Violanti J.M. Burchfiel C.M. Fekedulegn D. Andrew M.E. Hartley T.A. Miller D.B. Mnatsakanova A. Association of a dietary inflammatory index with inflammatory indices and metabolic syndrome among police officers J. Occup. Environ. Med. 2014 56 986 989 10.1097/JOM.0000000000000213 25046320 34. Pimenta A.M. Toledo E. Rodriguez-Diez M.C. Gea A. Lopez-Iracheta R. Shivappa N. Hébert J.R. Martínez-González M. Dietary indexes, food patterns and incidence of metabolic syndrome in a Mediterranean cohort: The SUN project Clin. Nutr. 2015 34 508 514 10.1016/j.clnu.2014.06.002 24975512 35. Neufcourt L. Assmann K.E. Fezeu L.K. Touvier M. Graffouillère L. Shivappa N. Hébert J.R. Wirth M.D. Hercberg S. Galan P. Prospective association between the dietary inflammatory index and metabolic syndrome: Findings from the SU.VI.MAX study Nutr. Metab. Cardiovasc. Dis. 2015 25 988 996 10.1016/j.numecd.2015.09.002 26482566 36. Ruiz-Canela M. Zazpe I. Shivappa N. Hébert J.R. Sánchez-Tainta A. Corella D. Salas-Salvadó J. Fitó M. Lamuela-Raventós R.M. Rekondo J. Dietary inflammatory index and anthropometric measures of obesity in a population sample at high cardiovascular risk from the PREDIMED (PREvención con DIeta MEDiterránea) trial Br. J. Nutr. 2015 113 984 995 10.1017/S0007114514004401 25720588 37. Shivappa N. Steck S.E. Hussey J.R. Ma Y. Hébert J.R. Inflammatory potential of diet and all-cause, cardiovascular, and cancer mortality in National Health and Nutrition Examination Survey III Study Eur. J. Nutr. 2015 1 10 10.1007/s00394-015-1112-x 25296886 38. Deng F.E. Shivappa N. Tang Y. Mann J.R. Hébert J.R. Association between diet-related inflammation, all-cause, all-cancer and cardiovascular disease mortality, with special focus on prediabetics: Findings from NHANES III Eur. J. Nutr. 2016 10.1007/s00394-016-1158-4 26825592 39. Shivappa N. Blair C.K. Prizment A.E. Jacobs D.R. Steck S.E. Hébert J.R. Association between inflammatory potential of diet and mortality in the Iowa Women’s Health study Eur. J. Nutr. 2015 1 12 10.1007/s00394-015-0967-1 25296886 40. Graffouillère L. Deschasaux M. Mariotti F. Neufcourt L. Shivappa N. Hébert J.R. Wirth M.D. Latino-Martel P. Hercberg S. Galan P. Prospective association between the Dietary Inflammatory Index and mortality: Modulation by antioxidant supplementation in the SU.VI.MAX randomized controlled trial Am. J. Clin. Nutr. 2016 103 878 885 10.3945/ajcn.115.126243 26864363 41. Shivappa N. Harris H. Wolk A. Hébert J.R. Association between inflammatory potential of diet and mortality among women in the Swedish Mammography Cohort Eur. J. Nutr. 2015 55 1891 1900 10.1007/s00394-015-1005-z 26227485 42. Choi J. Joseph L. Pilote L. Obesity and C-reactive protein in various populations: A systematic review and meta-analysis Obes. Rev. 2013 14 232 244 10.1111/obr.12003 23171381 43. Gentile M. Panico S. Rubba F. Mattiello A. Chiodini P. Jossa F. Marotta G. Pauciullo P. Rubba P. Obesity, overweight, and weight gain over adult life are main determinants of elevated hs-CRP in a cohort of Mediterranean women Eur. J. Clin. Nutr. 2010 64 873 878 10.1038/ejcn.2010.69 20517327 44. Welsh P. Polisecki E. Robertson M. Jahn S. Buckley B.M. de Craen A.J.M. Ford I. Jukema J.W. Macfarlane P.W. Packard C.J. Unraveling the directional link between adiposity and inflammation: a bidirectional Mendelian randomization approach J. Clin. Endocrinol. Metab. 2010 95 93 99 10.1210/jc.2009-1064 19906786 45. Timpson N.J. Nordestgaard B.G. Harbord R.M. Zacho J. Frayling T.M. Tybjærg-Hansen A. Smith G.D. C-reactive protein levels and body mass index: Elucidating direction of causation through reciprocal Mendelian randomization Int. J. Obes. 2011 35 300 308 10.1038/ijo.2010.137 20714329 46. Ramallal R. Toledo E. Martínez J.A. Shivappa N. Hebert J.R. Martínez-González M. Ruiz-Canela M. Inflammatory potential of diet, weight gain and incidence of overweight/obesity: The SUN cohort Clin. Nutr. 2016 under review 47. Moreno-Aliaga M. Campión J. Milagro F. Berjón A. Martínez F. Adiposity and proinflammatory state: The chicken or the egg Adipocytes 2005 1 1 16 48. Tang Y. Purkayastha S. Cai D. Hypothalamic microinflammation: A common basis of metabolic syndrome and aging Trend. Neurosci. 2015 38 36 44 10.1016/j.tins.2014.10.002 25458920 49. Lee C.C. Adler A.I. Sandhu M.S. Sharp S.J. Forouhi N.G. Erqou S. Luben R. Bingham S. Khaw K.T. Wareham N.J. Association of C-reactive protein with type 2 diabetes: Prospective analysis and meta-analysis Diabetologia 2009 52 1040 1047 10.1007/s00125-009-1338-3 19326095 50. Willeit P. Thompson S.G. Agewall S. Bergström G. Bickel H. Catapano A.L. Chien K.-L. de Groot E. Empana J.-P. Etgen T. Inflammatory markers and extent and progression of early atherosclerosis: Meta-analysis of individual-participant-data from 20 prospective studies of the PROG-IMT collaboration Eur. J. Prev. Cardiol. 2016 23 194 205 10.1177/2047487314560664 25416041 51. Kaptoge S. di Angelantonio E. Lowe G. Pepys M.B. Thompson S.G. Collins R. Danesh J. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: An individual participant meta-analysis Lancet 2010 375 132 140 20031199 52. Wensley F. Gao P. Burgess S. Kaptoge S. di Angelantonio E. Shah T. Engert J.C. Clarke R. Davey-Smith G. Nordestgaard B.G. Association between C reactive protein and coronary heart disease: Mendelian randomisation analysis based on individual participant data BMJ 2011 342 d548 10.1136/bmj.d548 21325005 53. Burgess S. Thompson S.G. Bias in causal estimates from Mendelian randomization studies with weak instruments Stat. Med. 2011 30 1312 1323 10.1002/sim.4197 21432888 54. Ridker P.M. From C-Reactive Protein to Interleukin-6 to Interleukin-1: Moving upstream to identify novel targets for atheroprotection Circ. Res. 2016 118 145 156 10.1161/CIRCRESAHA.115.306656 26837745 55. Srikanthan K. Feyh A. Visweshwar H. Shapiro J.I. Sodhi K. Systematic Review of Metabolic Syndrome Biomarkers: A Panel for Early Detection, Management, and Risk Stratification in the West Virginian Population Int. J. Med. Sci. 2016 13 25 38 10.7150/ijms.13800 26816492 56. Herder C. Carstensen M. Ouwens D.M. Anti-inflammatory cytokines and risk of type 2 diabetes Diabetes Obes. Metab. 2013 15 39 50 10.1111/dom.12155 24003920 57. Van Exel E. Gussekloo J. de Craen A.J.M. Frolich M. Bootsma-van der Wiel A. Westendorp R.G.J. Low Production Capacity of Interleukin-10 Associates With the Metabolic Syndrome and Type 2 Diabetes : The Leiden 85-Plus Study Diabetes 2002 51 1088 1092 10.2337/diabetes.51.4.1088 11916930 58. Kaptoge S. Seshasai S.R.K. Gao P. Freitag D.F. Butterworth A.S. Borglykke A. di Angelantonio E. Gudnason V. Rumley A. Lowe G.D.O. Inflammatory cytokines and risk of coronary heart disease: New prospective study and updated meta-analysis Eur. Heart J. 2014 35 578 589 10.1093/eurheartj/eht367 24026779 59. I.-6R.M.R.A. (IL6R MR) Consortium The interleukin-6 receptor as a target for prevention of coronary heart disease: A mendelian randomisation analysis Lancet 2012 379 1214 1224 22421340 60. Sarwar N. Butterworth A.S. Freitag D.F. Gregson J. Willeit P. Gorman D.N. Gao P. Saleheen D. Rendon A. Nelson C.P. Interleukin-6 receptor pathways in coronary heart disease: A collaborative meta-analysis of 82 studies Lancet 2012 379 1205 1213 22421339 61. Mallat Z. Besnard S. Duriez M. Deleuze V. Emmanuel F. Bureau M.F. Soubrier F. Esposito B. Duez H. Fievet C. Protective Role of Interleukin-10 in Atherosclerosis Circ. Res. 1999 85 e17 e24 10.1161/01.RES.85.8.e17 10521249 62. Welsh P. Murray H.M. Ford I. Trompet S. de Craen A.J.M. Jukema J.W. Stott D.J. McInnes I.B. Packard C.J. Westendorp R.G.J. Circulating interleukin-10 and risk of cardiovascular events: A prospective study in the elderly at risk Arterioscler. Thromb. Vasc. Biol. 2011 31 2338 2344 10.1161/ATVBAHA.111.231795 21757655 63. Lee Y.W. Kim P.H. Lee W.H. Hirani A.A. Interleukin-4, Oxidative Stress, Vascular Inflammation and Atherosclerosis Biomol. Ther. 2010 18 135 144 10.4062/biomolther.2010.18.2.135 21072258 64. Hu F.B. Dietary pattern analysis: A new direction in nutritional epidemiology Curr. Opin. Lipidol. 2002 13 3 9 10.1097/00041433-200202000-00002 11790957 65. Jacobs D.R. Steffen L.M. Nutrients, foods, and dietary patterns as exposures in research: A framework for food synergy Am. J. Clin. Nutr. 2003 78 508S 513S 12936941 66. Guenther P.M. Reedy J. Krebs-Smith S.M. Development of the Healthy Eating Index-2005 J. Am. Diet. Assoc. 2008 108 1896 1901 10.1016/j.jada.2008.08.016 18954580 67. Guenther P.M. Casavale K.O. Reedy J. Kirkpatrick S.I. Hiza H.A.B. Kuczynski K.J. Kahle L.L. Krebs-Smith S.M. Update of the Healthy Eating Index: HEI-2010 J. Acad. Nutr. Diet. 2013 113 569 580 10.1016/j.jand.2012.12.016 23415502 68. Chiuve S.E. Fung T.T. Rimm E.B. Hu F.B. McCullough M.L. Wang M. Stampfer M.J. Willett W.C. Alternative dietary indices both strongly predict risk of chronic disease J. Nutr. 2012 142 1009 1018 10.3945/jn.111.157222 22513989 69. Miller P.E. Cross A.J. Subar A.F. Krebs-Smith S.M. Park Y. Powell-Wiley T. Hollenbeck A. Reedy J. Comparison of 4 established DASH diet indexes: Examining associations of index scores and colorectal cancer Am. J. Clin. Nutr. 2013 98 794 803 10.3945/ajcn.113.063602 23864539 70. Trichopoulou A. Costacou T. Bamia C. Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population N. Engl. J. Med. 2003 348 2599 2608 10.1056/NEJMoa025039 12826634 71. Mozaffarian D. Dietary and policy priorities for cardiovascular disease, diabetes, and obesity: A comprehensive review Circulation 2016 133 187 225 10.1161/CIRCULATIONAHA.115.018585 26746178 72. Schwingshackl L. Hoffmann G. Diet Quality as Assessed by the Healthy Eating Index, the Alternate Healthy Eating Index, the Dietary Approaches to Stop Hypertension Score, and Health Outcomes: A Systematic Review and Meta-Analysis of Cohort Studies J. Acad. Nutr. Diet. 2015 115 780 800 10.1016/j.jand.2014.12.009 25680825 73. Martínez-González M. Salas-Salvadó J. Estruch R. Corella D.D. Fitó M. Ros E. Benefits of the Mediterranean Diet: Insights from the PREDIMED Study Prog. Cardiovasc. Dis. 2015 58 50 60 10.1016/j.pcad.2015.04.003 25940230 74. Esposito K. Maiorino M.I. Bellastella G. Chiodini P. Panagiotakos D. Giugliano D. A journey into a Mediterranean diet and type 2 diabetes: A systematic review with meta-analyses BMJ 2015 5 e008222 10.1136/bmjopen-2015-008222 26260349 75. Mancini J.G. Filion K.B. Atallah R. Eisenberg M.J. Systematic Review of the Mediterranean Diet for Long-Term Weight Loss Am. J. Med. 2016 129 407 415 10.1016/j.amjmed.2015.11.028 26721635 76. Soltani S. Shirani F. Chitsazi M.J. Salehi-Abargouei A. The effect of dietary approaches to stop hypertension (DASH) diet on weight and body composition in adults: A systematic review and meta-analysis of randomized controlled clinical trials Obes. Rev. 2016 17 442 454 10.1111/obr.12391 26990451 77. Estruch R. Martínez-González M. Corella D. Salas-Salvadó J. Fitó M. Chiva-Blanch G. Fiol M. Gómez-Gracia E. Arós F. Lapetra J. Effect of a high-fat mediterranean diet on bodyweight and waist circumference: A prespecified secondary outcomes analysis of the PREDIMED randomised controlled trial Lancet Diabetes Endocrinol. 2016 10.1016/S2213-8587(16)30085-7 78. Dietary Guidelines Advisory Committe Scientific Report of the 2015 Dietary Guidelines Advisory Committee Proceedings of the USDA and US Department of Health and Human Services Dietary Guidelines Advisory Committe Washington, DC, USA 2015 79. Wirth M.D. Hébert J.R. Shivappa N. Hand G.A. Hurley T.G. Drenowatz C. McMahon D. Shook R.P. Blair S.N. Anti-inflammatory Dietary Inflammatory Index scores are associated with healthier scores on other dietary indices Nutr. Res. 2016 36 214 219 10.1016/j.nutres.2015.11.009 26923507 80. Esmaillzadeh A. Kimiagar M. Mehrabi Y. Azadbakht L. Hu F.B. Willett W.C. Fruit and vegetable intakes, C-reactive protein, and the metabolic syndrome Am. J. Clin. Nutr. 2006 84 1489 1497 17158434 81. Hermsdorff H.H.M. Zulet M.A. Puchau B. Martínez J.A. Fruit and vegetable consumption and proinflammatory gene expression from peripheral blood mononuclear cells in young adults: A translational study Nutr. Metab. 2010 7 42 10.1186/1743-7075-7-42 20465828 82. Hermsdorff H.H.M. Zulet M.Á. Abete I. Martínez J.A. A legume-based hypocaloric diet reduces proinflammatory status and improves metabolic features in overweight/obese subjects Eur. J. Nutr. 2011 50 61 69 10.1007/s00394-010-0115-x 20499072 83. Casas-Agustench P. López-Uriarte P. Bulló M. Ros E. Cabré-Vila J.J. Salas-Salvadó J. Effects of one serving of mixed nuts on serum lipids, insulin resistance and inflammatory markers in patients with the metabolic syndrome Nutr. Metab. Cardiovasc. Dis. 2011 21 126 135 10.1016/j.numecd.2009.08.005 20031380 84. Ma Y. Griffith J.A. Chasan-Taber L. Olendzki B.C. Jackson E. Stanek E.J. Li W. Pagoto S.L. Hafner A.R. Ockene I.S. Association between dietary fiber and serum C-reactive protein Am. J. Clin. Nutr. 2006 83 760 766 16600925 85. Nettleton J.A. Steffen L.M. Mayer-Davis E.J. Jenny N.S. Jiang R. Herrington D.M. Jacobs D.R. Dietary patterns are associated with biochemical markers of inflammation and endothelial activation in the Multi-Ethnic Study of Atherosclerosis (MESA) Am. J. Clin. Nutr. 2006 83 1369 1379 16762949 86. Julia C. Meunier N. Touvier M. Ahluwalia N. Sapin V. Papet I. Cano N. Hercberg S. Galan P. Kesse-Guyot E. Dietary patterns and risk of elevated C-reactive protein concentrations 12 years later Br. J. Nutr. 2013 110 747 754 10.1017/S0007114512005636 23302662 87. Schwingshackl L. Christoph M. Hoffmann G. Effe of Olive Oil on Markers of Inflammation and Endothelial Function—A Systematic Review and Meta-Analysis Nutrients 2015 7 7651 7675 26378571 88. Van Woudenbergh G.J. Theofylaktopoulou D. Kuijsten A. Ferreira I. van Greevenbroek M.M. van der Kallen C.J. Schalkwijk C.G. Stehouwer C.D. Ocke M.C. Nijpels G. Adapted dietary inflammatory index and its association with a summary score for low-grade inflammation and markers of glucose metabolism: The Cohort study on Diabetes and Atherosclerosis Maastricht (CODAM) and the Hoorn study Am. J. Clin. Nutr. 2013 98 1533 1542 10.3945/ajcn.112.056333 24153342 89. Tabung F.K. Smith-Warner S.A. Chavarro J.E. Wu K. Fuchs C.S. Hu F.B. Chan A.T. Willett W.C. Giovannucci E.L. Development and validation of an empirical Dietary Inflammatory Index J. Nutr. 2016 10.3945/jn.115.228718 27358416 90. Hodge A.M. Bassett J.K. Shivappa N. Hébert J.R. English D.R. Giles G.G. Severi G. Dietary inflammatory index, Mediterranean diet score, and lung cancer: A prospective study Cancer Causes Control 2016 27 907 917 10.1007/s10552-016-0770-1 27294725 91. Shivappa N. Hébert J.R. Rosato V. Rossi M. Montella M. Serraino D. la Vecchia C. Dietary inflammatory index and ovarian cancer risk in a large Italian case-control study Cancer Causes Control 2016 27 897 906 10.1007/s10552-016-0767-9 27262447 92. Shivappa N. Hébert J.R. Rosato V. Serraino D. la Vecchia C. Inflammatory potential of diet and risk of laryngeal cancer in a case-control study from Italy Cancer Causes Control 2016 27 1027 1034 10.1007/s10552-016-0781-y 27379989 93. Zucchetto A. Gini A. Shivappa N. Hébert J.R. Stocco C. dal Maso L. Birri S. Serraino D. Polesel J. Dietary inflammatory index and prostate cancer survival Int. J. Cancer 2016 10.1002/ijc.30208 27242333

PMC005xxxxxx/PMC5000664.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081266ijms-17-01266ReviewCirculating Tumor Cells in the Adenocarcinoma of the Esophagus Gallerani Giulia *Fabbri Francesco Marchetti Dario Academic EditorBiosciences Laboratory, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS, Via P. Maroncelli 40, Meldola 47014, FC, Italy; [email protected]* Correspondence: [email protected]; Tel.: +39-0543-739-230; Fax: +39-0543-739-22104 8 2016 8 2016 17 8 126629 6 2016 30 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Circulating tumor cells (CTCs) are elements of indisputable significance as they seem to be responsible for the onset of metastasis. Despite this, research into CTCs and their clinical application have been hindered by their rarity and heterogeneity at the molecular and cellular level, and also by a lack of technical standardization. Esophageal adenocarcinoma (EAC) is a highly aggressive cancer that is often diagnosed at an advanced stage. Its incidence has increased so much in recent years that new diagnostic, prognostic and predictive biomarkers are urgently needed. Preliminary findings suggest that CTCs could represent an effective, non-invasive, real-time assessable biomarker in all stages of EAC. This review provides an overview of EAC and CTC characteristics and reports the main research results obtained on CTCs in this setting. The need to carry out further basic and translational research in this area to confirm the clinical usefulness of CTCs and to provide oncologists with a tool to improve therapeutic strategies for EAC patients was herein highlighted. circulating tumor cellsesophagus adenocarcinomaliquid biopsy ==== Body 1. Introduction Esophageal carcinoma (EC) is one of the most common malignant tumors in the world and the sixth most common cause of death from cancer, with estimated 400,000 deaths in 2012 (4.9% of the total) and an overall 5-year survival rate ranging from 15% to 25% [1,2,3]. From the second half of the 1990s to the early 2000s the age-adjusted incidence of EC in Europe increased by 39.6% for men and 37.5% for women [4], indicating an alarming trend. EC is an extremely aggressive tumor characterized by very poor survival rates and by an epidemiologic pattern distinct from other cancers [5,6]. It is often diagnosed at an advanced stage (40% of patients), when the 5-year survival rate is lower than 3% [1,7,8]. EC typically occurs in one of two forms: squamous cell carcinoma (ESCC), arising from the stratified squamous epithelial lining of the organ, or adenocarcinomas (EAC) in which columnar glandular cells replace the squamous epithelium [9]. Despite intense research efforts, the pathogenesis of this disease is still widely debated. It is generally accepted that EAC is a direct consequence of the condition known as Barrett’s esophagus (BE), a condition in which the stratified epithelium is substituted by metaplastic columnar epithelium [10]. However, only a small fraction of individuals with BE go on to develop esophageal adenocarcinoma (about 0.22%) [11]. Conversely, some patients are diagnosed with EAC despite no prior finding of BE [12]. At present, repeated endoscopic biopsies with histological evaluation of dysplasia is the only means of evaluating the risk of progression to cancer. However, such an approach is hampered by all the challenges of these clinical determinations, e.g., inter-observer differences and lack of reproducible diagnostic classification [13], highly heterogeneous dysplasia in terms of progression to adenocarcinoma [14], or high number of biopsies required to reduce the risk of sampling error caused by primary tumor heterogeneity [15,16]. Moreover, once EAC has been diagnosed, there are few reliable methods to stratify patients and identify the most suitable therapeutic approach [17]. Thus, new reliable and reproducible diagnostic, prognostic and predictive markers must urgently be sought to improve the overall management of EAC patients. Circulating tumor cells (CTCs) are exceedingly rare, genetically and phenotypically heterogeneous cells found in the peripheral blood of cancer patients. Their presence is correlated with poor prognosis and progression-free survival, and they are considered indicators of treatment efficacy in different tumors [18,19,20,21]. CTCs can be considered as precursors of metastatic dissemination, one of the key element of the metastatic process, and are one of the main elements of liquid biopsy. However, their clinical use for tumor staging, disease monitoring and choice of treatment has yet to become a reality. Despite this, CTCs have the potential to serve as a biopsy for the “leukemic phase” of solid tumors [22]. Liquid biopsies enable patients with early or advanced disease to be stratified into prognostic groups. They could also potentially be used as a surrogate endpoint of survival for studies on therapeutic efficacy and for the molecular sub-classification of advanced cancer patients [23,24]. Thus, they represent a real opportunity to bring the science of personalized medicine to realization in clinical practice. This review focuses on the biology and clinical characteristics of EAC and on preliminary findings and potentialities of the use of CTCs as prognostic and predictive marker in this tumor. It also highlights the ability of CTCs to provide a real-time snapshot of tumor features (heterogeneity, aggressiveness, invasion capacity, etc.) that could be capable of offering clinically relevant information. An English-language literature search using PubMed/MEDLINE (up to the end of June 2016) was performed using the key words CTCs, CTC detection methods, EAC, survival, prognosis, progression, EAC-pathogenesis and EAC-staging system. All pertinent articles containing human clinical trial data and relevant information were evaluated and included if appropriate. 2. Esophageal Adenocarcinoma (EAC) Pathogenesis and Features One of the first steps in the development of EAC is the transition from normal esophageal epithelium to columnar and secretory epithelium, a process often associated with chronic inflammatory events triggered by gastro-esophageal reflux. The genesis of metaplasia is thought to be a response to chronic tissue inflammation [25] and is known as Barrett’s esophagus (BE) [26,27,28]. Although a number of papers has been published on this topic, there is still relatively little information available about tissue homeostasis in the epithelium of the esophagus. Furthermore, the role of candidate stem cells in the growth and regeneration of this tissue has not yet been fully defined in vivo [29]. However, research has shown that esophageal epithelium is maintained by a population of cells capable of both maintaining and repairing tissue and which divide to produce proliferating and differentiating daughter cells with equal likelihood, without the need for a slow-cycling stem cell pool [30]. Although in contrast to previous conventional theories regarding stem cells [31], this information is probably a key to identifying the mechanisms of pathogenesis of esophageal cancer and may also help to clarify the genetic heterogeneity of EAC. Recent findings on genomic abnormalities in EAC, occurring the early stages of disease [16,32], include conventional single nucleotide variants in 26 genes, recurrent deletions and focal amplifications, and mutations in chromatin-remodeling genes [33,34,35,36]. Within this context the linear model of carcinogenesis may not optimally define BE to EAC progression as it does not explain the genetic heterogeneity of EAC. Findings similar to those observed in breast and colorectal cancer [37] have demonstrated that all cancers are probably a spectrum of diseases and as such, EAC is no exception [17]. In addition, BE genetic heterogeneity may explain EAC clonal diversity and may predict transformation to adenocarcinoma [32]. Thus, the current challenge is to identify and validate new diagnostic, prognostic and predictive biomarkers that can unveil EAC genetic and cellular heterogeneity and, in doing so, help clinicians to select the best treatment option. CTCs are “multifaced” cells that actively or passively leave the primary tumor, follow potential dissemination pathways, and are able to reach distant localizations and adapt to different microenvironments [38,39]. They are a remarkably rare and heterogeneous cell population at the genetic and phenotypic level and potentially composed of a combination of subpopulations with dissimilar features [40]. An interesting hypothesis to explain the starting point of CTC spread and the mechanisms involved are those of the epithelial to mesenchymal transition (EMT) process [41,42]. EMT is considered a normal phenomenon implicated in embryogenesis and wound healing processes that may be activated during cancer progression and metastasis [43,44,45]. The process induces a significant change in cell phenotype associated with aggressive biological behavior in cancer cells, loss of cell junctions [46] and of apical-basal polarity [47], and enhanced CTC motility which facilitates intravasation into the bloodstream. EMT and its reverse process, mesenchymal to epithelial transition (MET), could enable CTCs to switch backwards and forwards between phenotypes, causing resistance to anoikis, to the physical stress induced by blood circulation, and also to chemo- and radio-therapy [43,48]. As EAC is a paradigm for inflammation-associated cancer [49], and inflammation processes have been reported to contribute to tumor progression, metastasis, EMT [50] and CTC spreading, it is tempting to hypothesize that an almost direct connection exists between all of these phenomena. Taken together, these hypotheses indicate that CTCs could be an extremely effective biomarker of EAC in terms of its genetic heterogeneity, clonal evolution and sensitivity or resistance to treatment. In the near future, once technical and methodological aspects of the CTC research area are sufficiently sensitive, specific and representative of systemic disease, these circulating cells could become a significant biomarker in EAC. 3. EAC Clinical Aspects Esophageal cancer staging has been defined by the American Joint Committee on Cancer (AJCC) Staging System which established a tumor-node-metastasis (TNM) classification with sub-classifications based on the depth of invasion of the primary tumor (T), lymph node involvement (N), and extent of metastatic disease (M) [51]. Current management of EAC is mainly based on complete preoperative assessment because accurate pre-treatment staging and subsequent stage-appropriate treatment is crucial to optimize EAC outcome. Diagnosis is made mainly by endoscopy and often includes multiple biopsies of the upper digestive tract. Once EAC has been histologically confirmed, clinical stage is determined by further instrumental tests. A computerized tomography (CT) scan provides valuable information about the longitudinal extension of the tumor and is useful in identifying the presence of distant metastases such as those of the lung and liver but somewhat limited in defining nodal involvement [52]. Positron-emission tomography (PET) scans represent an important aid to staging in that they detect previously unseen metastatic disease in up to 15%–20% of cases [53,54]. However, PET is generally considered a more suitable instrument for post-treatment assessment, especially that of neo-adjuvant therapy [55]. Despite these strategies, small secondary lesions may nevertheless be missed and patients may also have undetected pleural or peritoneal disease [56]. A more accurate staging system is thus urgently required to improve treatment strategies from the early stages of disease onwards. Novel tools for early tumor detection, adequate prognostic staging, and accurate treatment monitoring and selection, in particular in the neo-adjuvant setting, are also needed. A new staging category, M0(i+), was recently proposed by the AJCC [57]. M0(i+) is defined as “no clinical or radiographic evidence of distant metastases, but deposits of molecularly or microscopically detected tumor cells detected in circulating blood, bone marrow, or other non-regional nodal tissues, that are no larger than 0.2 mm in a patient without symptoms or signs of metastases”. This new M category could significantly improve cancer staging and consequently the therapeutic management of the disease. Using M0(i+) staging in EAC could also help to overcome the limitations of CT and PET in detecting minimal occult disease, often defined by the presence of CTCs in the blood. 4. Circulating Tumor Cells (CTCs) and EAC Recent advances in techniques to identify CTCs in the blood of patients with different types of cancer have generated interesting results. However, there is still a lack of methodological uniformity and a relatively high variability in detection rates, probably due to the markers used for CTC identification. The most widely used definition of a CTC, i.e., an EpCAM+/CK+/CD45− cell with a round or oval intracellular nucleus [23], refers mainly to epithelial markers, especially on EpCAM. However, this description falls short of including many potential CTC markers, subpopulations and clusters [40]. Further approaches that permit the identification of EpCAM-negative cells are thus urgently needed [58,59,60,61]. Non-EpCAM-based and EMT-related methods will be of unquestionable importance to decrease the risk of EpCAM-based analysis. In order to improve results, a combination of markers and enrichment strategies to identify EpCAM-negative cells is also needed. Immunocytological, molecular, densitometric and/or size-dependent CTC enrichment methods could lead to significantly different results from those obtained by EpCAM-based technologies [58,59,60,61,62,63]. Approaches that combine physical and cellular properties of CTCs have produced interesting results [64,65,66], as previously reported by our group [67,68]. Our approach, albeit improvable, was based on densitometric enrichment followed by an immunocytological detection step and allowed us to detect three CTC classes: EpCAM+/CKs+, EpCAM−/CKs+ and EpCAM+/CKs− cells. Although it was not possible to clearly distinguish between these three classes, the method enabled us to specifically detect a more wide-ranging “epithelial” phenotype. It can also detect CTCs that are positive for a “single marker”, such as EpCAM−/CKs+ or EpCAM+/CKs− cells, permitting the identification of EpCAM− cells and reducing the risk of inadequate EpCAM-based enrichment, which may lead to an underestimation of the significance of CTCs [58]. In a gastro-esophageal cancer setting, Kubisch et al. [69] reported an immuno-magnetic enrichment that included mucin-1 in addition to EpCAM, suggesting that a single marker is not enough to identify all CTCs. Despite these limits, preliminary studies on CTC status before and after surgery in patients with esophageal squamous cancer showed that the presence of CTCs was an independent predictor of disease recurrence [70]. Using reverse transcriptase-polymerase chain reaction assay, CTC detection rates ranging from 2% to 32.9% were found in patients [70,71]. Since these pioneering works were published, very few other investigations have been conducted in this setting. During this time, CTCs specifically classified as EpCAM+/CK+/CD45− cells and identified by the CELLSEARCH® System (Jannsen Diagnostics, Raritan, NJ, USA) have been definitively shown to be independent predictors of progression-free survival (PFS) and overall survival (OS) in patients with other metastatic cancers, e.g., breast cancer [72]. To date, the CELLSEARCH® System is the only standardized FDA-approved device and whose clinical validity has been confirmed. In 2008, Hiraiwa et al. [73] observed that about 21% (5/23) of patients with metastatic EC were CTC-positive and had a significantly shorter overall survival than CTC-negative cases. However, the number of patients was too small for the study to reach statistical significance. In 2013, a small pilot study on a cohort of 18 patients with advanced esophago-gastric cancer carried out using the CELLSEARCH® System reported that 44% of patients showed >2 CTCs/7.5 mL of blood before first-line chemotherapy [74]. Only 11 of the 18 patients had cancer of the esophagus itself (9 esophagogastric junction, 2 esophagus). Among these, only 4 with esophagogastric disease had >2 CTCs/7.5 mL of blood, whereas both cases of esophageal cancer showed ≤2 CTCs. Although this study confirmed the feasibility of CTC research in this clinical setting, it failed to draw any definitive conclusions about the relationship between CTCs and esophageal cancer due to the low number of cases studied. A larger, well-defined study to assess CTCs as a staging tool for non-metastatic esophageal cancer in prognostic subgroups was carried out by Reeh et al. [75]. The study enrolled 100 patients, including 29 with squamous cell carcinoma (SCC), 68 with adenocarcinoma, 2 with anaplastic carcinoma and 1 with a mixed-type cancer. Using a cutoff of one or more CTCs, the authors found that CTC-positive patients with non-metastatic disease had a significantly shorter overall and relapse-free survival than those without CTCs. This research demonstrated the clinical significance of CTCs as a preoperative staging factor in EAC, independently of other risk indicators such as histological subtype, tumor stage, lymph node (LN) invasion, and tumor grade. Interestingly, only 3 (10.3%) of the 29 patients with ESCC showed ≥1 CTC, whereas 14 (20.6%) of the 68 patients with EAC showed ≥1 CTC. This result could be due to dissimilar characteristics between esophageal cancer subtypes (SSC vs. EAC), e.g., varied EpCAM expression resulting in diverse CTC detection rates, indicating a potentially clinically relevant difference between the two histotypes [75]. Consistent with this hypothesis, an interesting study by Driemel et al. [76] reported that EpCAM expression in disseminated tumor cells (DTC) in early esophageal cancer may vary. In their study, analysis of EpCAM status in DTCs derived from lymph nodes and bone marrow showed that CK18-positive DTCs often lack EpCAM expression. CK18+ DTCs were detected in 38.9% of esophageal cancer patients, but co-expression of EpCAM was seen in only 37.1% of DTC-positive cases, whereas 62.9% of patients showed CK18+/EpCAM low/negative DTCs. A comparison of EpCAM expression in 14 pairs of primary tumors and their associated DTCs revealed an absence of EpCAM in DTCs in 64% of patients with EpCAM-overexpressing primary tumors. It was concluded that this discrepancy was not due to an intrinsic characteristic of the primary tumor, but most probably to EpCAM knockdown and/or EMT induction. Hence, it can be hypothesized that the metastatic progression of esophageal cancer may be supported by both EpCAM-positive and EpCAM-low/negative cancer cells in a context-dependent manner, and that this different phenotype may also be histotype-specific. Taking together, these findings suggest that CTC studies involving the CELLSEARCH® System may be limited due to the EpCAM down-regulation observed in DTCs. In agreement with literature data, EpCAM should not to be used as the only CTC identification marker as it may underestimate the actual amount of circulating and/or disseminated tumor cells. Although less specifically targeted at EAC, but in line with the results obtained by Reeh and et al., the paper by Kubisch et al. [69] suggests that the presence of CTCs is a predictor of outcome in patients with gastro-esophageal neoplasia, supporting the role of this biomarker as a marker of survival and progression in this setting. CTCs and DTCs could help to unravel the genetic heterogeneity of the tumor during its evolution. Monitoring genetic heterogeneity could provide significant information about cancer progression, potential new therapeutic targets, and tumor sensitivity or resistance to therapy. However, single tumor cell isolation and sequencing are still substantially hampered by methodological challenges such as cell isolation and manipulation, whole genome amplification, and genome-wide analysis [77]. Up to now, in addition to primary tumor tissue analysis, genetic heterogeneity in EAC has mainly been dissected in DTCs. Stoecklein et al. [78] observed that DTCs from lymph nodes and bone marrow differed from those of primary tumors in their genetic aberrations. The only commonly conserved region in cells disseminated lymphatically and hematogenously was the chromosomal region comprising HER2. The gain in just one DTC was enough to confer a poor prognosis. In agreement with a functional study on cell lines isolated from DTCs and primary tumors, data reported in the paper by Stoecklein et al. were suggestive of a new drug target during disease evolution. This has led to the possibility of extrapolating new common patterns of targetable alterations through the study of genetic heterogeneity. Despite the importance of such results, technical limitations still exist, hindering their implementation into clinical practice. In particular aCGH analysis which has a limited resolution hence it and may therefore have missed small genetic alterations and the real extension of the overall genetic heterogeneity. Next-generation sequencing (NGS) approaches developed more recently could potentially help to identify new and smaller genetic changes also [79,80] even in CTCs as well [81,82,83], paving the way towards the use of DTCs and CTCs as more effective tools to select treatments and to monitor disease progression and relapse. 5. Conclusions The usefulness of CTCs is still very much open to debate and their full implementation into clinical practice is unlikely in the near future. The lack of technical standardization for their identification, as well as limited data on their cellular and molecular characteristics, represent significant obstacles to the recognition of the clinical utility of this marker. Despite this, CTCs still hold great potential [84]. The detection of minimal occult disease is still an unmet clinical need in EAC and liquid biopsy could prove to be the solution to this problem. As already hypothesized by some authors, liquid biopsy could help us to better understand and manage EAC in terms of disease monitoring, choice of therapy and identification of drug resistance [85,86]. As already suggested [87], it is appealing to imagine that in the EAC setting, CTCs could be utilized for diagnosis, drug resistance identification, monitoring of occult disease, early relapse, and therapeutic efficacy, and selection of targeted drugs. However, their evaluation still presents several problems, especially with regard to their actual value and how they should be investigated, i.e., usefulness of evaluation of single CTCs, optimal number of cells to be tested, cell subpopulations worthy of a more in-depth analysis. Further studies are warranted to address these issues in an attempt to bring CTCs from the bench to the bedside in EAC. Acknowledgments The authors wish to thank Gráinne Tierney for editorial assistance. Author Contributions Giulia Gallerani and Francesco Fabbri were jointly responsible for writing the paper and agreed to its submission. Conflicts of Interest The authors declare no conflict of interest. ==== Refs References 1. Thrift A.P. The epidemic of oesophageal carcinoma: Where are we now? Cancer Epidemiol. 2016 41 88 95 10.1016/j.canep.2016.01.013 26851752 2. Rustgi A.K. El-Serag H.B. Esophageal carcinoma N. Engl. J. Med. 2014 371 2499 2509 25539106 3. GLOBOCAN Available online: http://globocan.iarc.fr/Default.aspx (accessed on 3 March 2016) 4. Pennathur A. Gibson M.K. Jobe B.A. Luketich J.D. Oesophageal carcinoma Lancet 2013 381 400 412 10.1016/S0140-6736(12)60643-6 23374478 5. Blot W.J. McLaughlin J.K. The changing epidemiology of esophageal cancer Semin. Oncol. 1999 26 2 8 10566604 6. Engel L.S. Chow W.-H. Vaughan T.L. Gammon M.D. Risch H.A. Stanford J.L. Schoenberg J.B. Mayne S.T. Dubrow R. Rotterdam H. Population attributable risks of esophageal and gastric cancers J. Natl. Cancer Inst. 2003 95 1404 1413 10.1093/jnci/djg047 13130116 7. Mao W.-M. Zheng W.-H. Ling Z.-Q. Epidemiologic risk factors for esophageal cancer development Asian Pac. J. Cancer Prev. 2011 12 2461 2466 22320939 8. Enzinger P.C. Mayer R.J. Esophageal cancer N. Engl. J. Med. 2003 349 2241 2252 10.1056/NEJMra035010 14657432 9. Blot W.J. Continuing climb in rates of esophageal adenocarcinoma: An update JAMA J. Am. Med. Assoc. 1993 270 1320 10.1001/jama.1993.03510110056031 10. Reid B.J. Li X. Galipeau P.C. Vaughan T.L. Barrett’s oesophagus and oesophageal adenocarcinoma: Time for a new synthesis Nat. Rev. Cancer 2010 10 87 101 10.1038/nrc2773 20094044 11. Bhat S. Coleman H.G. Yousef F. Johnston B.T. Mcmanus D.T. Gavin A.T. Murray L.J. Risk of malignant progression in Barrett’s esophagus patients: Results from a large population-based study J. Natl. Cancer Inst. 2011 103 1049 1057 10.1093/jnci/djr203 21680910 12. Wang D.H. Souza R.F. Biology of Barrett’s esophagus and esophageal adenocarcinoma Gastrointest. Endosc. Clin. N. Am. 2011 21 25 38 10.1016/j.giec.2010.09.011 21112495 13. Montgomery E. Bronner M.P. Goldblum J.R. Greenson J.K. Haber M.M. Hart J. Lamps L.W. Lauwers G.Y. Lazenby A.J. Lewin D.N. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: A reaffirmation Hum. Pathol. 2001 32 368 378 10.1053/hupa.2001.23510 11331953 14. Reid B.J. Levine D.S. Longton G. Blount P.L. Rabinovitch P.S. Predictors of progression to cancer in Barrett’s esophagus: Baseline histology and flow cytometry identify low- and high-risk patient subsets Am. J. Gastroenterol. 2000 95 1669 1676 10.1111/j.1572-0241.2000.02196.x 10925966 15. Wang K.K. Sampliner R.E. Updated guidelines 2008 for the diagnosis, surveillance and therapy of Barrett’s esophagus Am. J. Gastroenterol. 2008 103 788 797 10.1111/j.1572-0241.2008.01835.x 18341497 16. Stachler M.D. Taylor-Weiner A. Peng S. McKenna A. Agoston A.T. Odze R.D. Davison J.M. Nason K.S. Loda M. Leshchiner I. Paired exome analysis of Barrett’s esophagus and adenocarcinoma Nat. Genet. 2015 10.1038/ng.3343 26192918 17. Weaver J.M.J. Ross-Innes C.S. Fitzgerald R.C. The “-omics” revolution and oesophageal adenocarcinoma Nat. Rev. Gastroenterol. Hepatol. 2014 11 19 27 10.1038/nrgastro.2013.150 23982683 18. Cristofanilli M. Budd G.T. Ellis M.J. Stopeck A. Matera J. Miller M.C. Reuben J.M. Doyle G.V. Allard W.J. Terstappen L.W.M.M. Circulating tumor cells, disease progression, and survival in metastatic breast cancer N. Engl. J. Med. 2004 351 781 791 10.1056/NEJMoa040766 15317891 19. Cohen S.J. Punt C.J. Iannotti N. Saidman B.H. Sabbath K.D. Gabrail N.Y. Picus J. Morse M. Mitchell E. Miller M.C. Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer J. Clin. Oncol. 2008 26 3213 3221 10.1200/JCO.2007.15.8923 18591556 20. De Bono J.S. Scher H.I. Montgomery R.B. Parker C. Miller M.C. Tissing H. Doyle G.V. Terstappen L.W.W.M. Pienta K.J. Raghavan D. Circulating tumor cells predict survival benefit from treatment in metastatic castration-resistant prostate cancer Clin. Cancer Res. 2008 14 6302 6309 10.1158/1078-0432.CCR-08-0872 18829513 21. Rack B. Schindlbeck C. Jückstock J. Andergassen U. Hepp P. Zwingers T. Friedl T.W.P. Lorenz R. Tesch H. Fasching P.A. Fehm T. Circulating tumor cells predict survival in early average-to-high risk breast cancer patients J. Natl. Cancer Inst. 2014 106 1 11 10.1093/jnci/dju066 24832787 22. Mocellin S. Keilholz U. Rossi C.R. Nitti D. Circulating tumor cells: The “leukemic phase” of solid cancers Trends Mol. Med. 2006 12 130 139 10.1016/j.molmed.2006.01.006 16488189 23. Attard G. de Bono J.S. Utilizing circulating tumor cells: Challenges and pitfalls Curr. Opin. Genet. Dev. 2011 21 50 58 10.1016/j.gde.2010.10.010 21112767 24. Pantel K. Speicher M.R. The biology of circulating tumor cells Oncogene 2015 10.1038/onc.2015.192 26050619 25. Tosh D. Slack J.M.W. How cells change their phenotype Nat. Rev. Mol. Cell Biol. 2002 3 187 194 10.1038/nrm761 11994739 26. Jankowski J.A. Harrison R.F. Perry I. Balkwill F. Tselepis C. Barrett’s metaplasia Lancet 2000 356 2079 2085 10.1016/S0140-6736(00)03411-5 11145505 27. Kountourakis P. Papademetriou K. Ardavanis A. Papamichael D. Barrett’s esophagus: Treatment or observation of a major precursor factor of esophageal cancer? J. BUON 2011 16 425 430 22006743 28. Findlay J.M. Middleton M.R. Tomlinson I. Genetic biomarkers of Barrett’s esophagus susceptibility and progression to dysplasia and cancer: A systematic review and meta-analysis Dig. Dis. Sci. 2016 61 25 38 10.1007/s10620-015-3884-5 26445852 29. Kushner J. Esophageal stem cells, where art thou? Science 2012 337 1051 1052 10.1126/science.1227506 22936766 30. Doupé D.P. Alcolea M.P. Roshan A. Zhang G. Klein A.M. Simons B.D. Jones P.H. A Single progenitor population switches behavior to maintain and repair esophageal epithelium Science 2012 337 1091 1093 10.1126/science.1218835 22821983 31. Kountourakis P. Ajani J. Davila M. Barrett’s esophagus: A review of biology and therapeutic approaches Gastrointest. Cancer Res. 2012 5 49 57 22690258 32. Maley C.C. Galipeau P.C. Finley J.C. Wongsurawat V.J. Li X. Sanchez C.A. Paulson T.G. Blount P.L. Risques R.-A. Rabinovitch P.S. Genetic clonal diversity predicts progression to esophageal adenocarcinoma Nat. Genet. 2006 38 468 473 10.1038/ng1768 16565718 33. Dulak A.M. Stojanov P. Peng S. Lawrence M.S. Fox C. Stewart C. Bandla S. Imamura Y. Schumacher S.E. Shefler E. Exome and whole-genome sequencing of esophageal adenocarcinoma identifies recurrent driver events and mutational complexity Nat. Genet. 2013 45 478 486 10.1038/ng.2591 23525077 34. Agrawal N. Jiao Y. Bettegowda C. Hutfless S.M. Wang Y. David S. Cheng Y. Twaddell W.S. Latt N.L. Shin E.J. Comparative genomic analysis of esophageal adenocarcinoma and squamous cell carcinoma Cancer Discov. 2012 2 899 905 10.1158/2159-8290.CD-12-0189 22877736 35. Dulak A.M. Schumacher S.E. van Lieshout J. Imamura Y. Fox C. Shim B. Ramos A.H. Saksena G. Baca S.C. Baselga J. Gastrointestinal adenocarcinomas of the esophagus, stomach, and colon exhibit distinct patterns of genome instability and oncogenesis Cancer Res. 2012 72 4383 4393 10.1158/0008-5472.CAN-11-3893 22751462 36. Streppel M.M. Lata S. Delabastide M. Montgomery E.A. Wang J.S. Canto M.I. Macgregor-Das A.M. Pai S. Morsink F.H.M. Offerhaus G.J. Next-generation sequencing of endoscopic biopsies identifies ARID1A as a tumor-suppressor gene in Barrett’s esophagus Oncogene 2013 33 1 11 10.1038/onc.2012.586 23416974 37. Wood L.D. Parsons D.W. Jones S. Lin J. Sjoblom T. Leary R.J. Shen D. Boca S.M. Barber T. Ptak J. The genomic landscapes of human breast and colorectal cancers Science 2007 318 1108 1113 10.1126/science.1145720 17932254 38. Wan L. Pantel K. Kang Y. Tumor metastasis: Moving new biological insights into the clinic Nat. Med. 2013 19 1450 1464 10.1038/nm.3391 24202397 39. Krebs M.G. Metcalf R.L. Carter L. Brady G. Blackhall F.H. Dive C. Molecular analysis of circulating tumour cells-biology and biomarkers Nat. Rev. Clin. Oncol. 2014 11 129 144 10.1038/nrclinonc.2013.253 24445517 40. Barriere G. Fici P. Gallerani G. Fabbri F. Zoli W. Rigaud M. Circulating tumor cells and epithelial, mesenchymal and stemness markers: Characterization of cell subpopulations Ann. Transl. Med. 2014 2 109 25489583 41. Ledford H. Cancer theory faces doubts Nature 2011 472 273 10.1038/472273a 21512545 42. Tarin D. The fallacy of epithelial mesenchymal transition in neoplasia the fallacy of epithelial mesenchymal transition in neoplasia Cancer Res. 2005 65 5996 6001 10.1158/0008-5472.CAN-05-0699 16024596 43. Nieto M.A. Epithelial plasticity: A common theme in embryonic and cancer cells Science 2013 342 1234850 10.1126/science.1234850 24202173 44. Thiery J.P. Acloque H. Huang R.Y.J. Nieto M.A. Epithelial-mesenchymal transitions in development and disease Cell 2009 139 871 890 10.1016/j.cell.2009.11.007 19945376 45. Baum B. Settleman J. Quinlan M.P. Transitions between epithelial and mesenchymal states in development and disease Semin. Cell Dev. Biol. 2008 19 294 308 10.1016/j.semcdb.2008.02.001 18343170 46. Lin X. Shang X. Manorek G. Howell S.B. Regulation of the epithelial-mesenchymal transition by claudin-3 and claudin-4 PLoS ONE 2013 8 1266 10.1371/journal.pone.0067496 23805314 47. Ozdamar B. Bose R. Barrios-Rodiles M. Wang H.-R. Zhang Y. Wrana J.L. Regulation of the polarity protein Par6 by TGFβ receptors controls epithelial cell plasticity Science 2005 307 1603 1609 10.1126/science.1105718 15761148 48. Micalizzi D.S. Farabaugh S.M. Ford H.L. Epithelial-mesenchymal transition in cancer: Parallels between normal development and tumor progression J. Mammary Gland Biol. Neoplasia 2010 15 117 134 10.1007/s10911-010-9178-9 20490631 49. O’Sullivan K.E. Phelan J.J. O’Hanlon C. Lysaght J. O’Sullivan J.N. Reynolds J.V. The role of inflammation in cancer of the esophagus Expert Rev. Gastroenterol. Hepatol. 2014 8 749 760 10.1586/17474124.2014.913478 24857183 50. Liu J. Lin P.C. Zhou B.P. Inflammation fuels tumor progress and metastasis Curr. Pharm. Des. 2015 21 3032 3040 10.2174/1381612821666150514105741 26004407 51. Varghese T.K. Hofstetter W.L. Rizk N.P. Low D.E. Darling G.E. Watson T.J. Mitchell J.D. Krasna M.J. The society of thoracic surgeons guidelines on the diagnosis and staging of patients with esophageal cancer Ann. Thorac. Surg. 2013 96 346 356 10.1016/j.athoracsur.2013.02.069 23752201 52. WEBER W. Imaging of esophageal and gastric cancer Semin. Oncol. 2004 31 530 541 10.1053/j.seminoncol.2004.04.016 15297944 53. Downey R.J. Akhurst T. Ilson D. Ginsberg R. Bains M.S. Gonen M. Koong H. Gollub M. Minsky B.D. Zakowski M. Whole body 18FDG-PET and the response of esophageal cancer to induction therapy: Results of a prospective trial J. Clin. Oncol. 2003 21 428 432 10.1200/JCO.2003.04.013 12560430 54. Luketich J.D. Schauer P.R. Meltzer C.C. Landreneau R.J. Urso G.K. Townsend D.W. Ferson P.F. Keenan R.J. Belani C.P. Role of positron emission tomography in staging esophageal cancer Ann. Thorac. Surg. 1997 64 765 769 10.1016/S0003-4975(97)00624-3 9307471 55. Rebollo Aguirre A.C. Ramos-Font C. Villegas Portero R. Cook G.J.R. Llamas Elvira J.M. Tabares A.R. 18F-fluorodeoxiglucose positron emission tomography for the evaluation of neoadjuvant therapy response in esophageal cancer: Systematic review of the literature Ann. Surg. 2009 250 247 254 10.1097/SLA.0b013e3181b0c3a6 19638908 56. Luketich J.D. Friedman D.M. Weigel T.L. Meehan M.A. Keenan R.J. Townsend D.W. Meltzer C.C. Evaluation of distant metastases in esophageal cancer: 100 consecutive positron emission tomography scans Ann. Thorac. Surg. 1999 68 1133 1137 10543468 57. Compton C.C. Byrd D.R. Garcia-Aguilar J. Kurtzman S.H. Olawaiye A. Washington M.K. AJCC Cancer Staging Atlas Springer New York New York, NY, USA 2012 58. Grover P.K. Cummins A.G. Price T.J. Roberts-Thomson I.C. Hardingham J.E. Circulating tumour cells: The evolving concept and the inadequacy of their enrichment by EpCAM-based methodology for basic and clinical cancer research Ann. Oncol. 2014 25 1506 1516 10.1093/annonc/mdu018 24651410 59. Gabriel M.T. Calleja L.R. Chalopin A. Ory B. Heymann D. Circulating tumor cells: A review of non-EpCAM-based approaches for cell enrichment and isolation Clin. Chem. 2016 62 571 581 10.1373/clinchem.2015.249706 26896446 60. Zhang L. Ridgway L.D. Wetzel M.D. Ngo J. Yin W. Kumar D. Goodman J.C. Groves M.D. Marchetti D. The identification and characterization of breast cancer CTCs competent for brain metastasis Sci. Transl. Med. 2013 5 180ra48 10.1126/scitranslmed.3005109 23576814 61. Vishnoi M. Peddibhotla S. Yin W. Scamardo A.T. George G.C. Hong D.S. Marchetti D. The isolation and characterization of CTC subsets related to breast cancer dormancy Sci. Rep. 2015 5 17533 10.1038/srep17533 26631983 62. Alix-Panabières C. Pantel K. Challenges in circulating tumour cell research Nat. Rev. Cancer 2014 14 623 10.1038/nrc3820 25154812 63. Pailler E. Adam J. Oulhen M. Auger N. Valent A. Borget I. Planchard D. Taylor M. Andre F. Soria J.C. Detection of circulating tumor cells harboring a unique ALK rearrangement in ALK—positive non—small-cell lung cancer J. Clin. Oncol. 2014 31 2273 2282 10.1200/JCO.2012.44.5932 23669222 64. Yu M. Bardia A. Wittner B.S. Stott S.L. Smas M.E. Ting D.T. Isakoff S.J. Ciciliano J.C. Wells M.N. Shah A.M. Circulating breast tumor cells exhibit dynamic changes in epithelial and mesenchymal composition Science 2013 339 580 584 10.1126/science.1228522 23372014 65. Aceto N. Bardia A. Miyamoto D.T. Donaldson M.C. Wittner B.S. Spencer J.A. Yu M. Pely A. Engstrom A. Zhu H. Circulating tumor cell clusters are oligoclonal precursors of breast cancer metastasis Cell 2014 158 1110 1122 10.1016/j.cell.2014.07.013 25171411 66. Sarioglu A.F. Aceto N. Kojic N. Donaldson M.C. Hamza B. Engstrom A. Zhu H. Sundaresan T.K. David T. Luo X. A microfluidic device for label-free, physical capture of circulating tumor cell-clusters Nat. Methods 2016 12 685 691 10.1038/nmeth.3404 25984697 67. Maltoni R. Fici P. Amadori D. Gallerani G. Cocchi C. Zoli M. Rocca A. Cecconetto L. Folli S. Scarpi E. Circulating tumor cells in early breast cancer: A connection with vascular invasion Cancer Lett. 2015 10.1016/j.canlet.2015.06.020 26184997 68. Fabbri F. Carloni S. Zoli W. Ulivi P. Gallerani G. Fici P. Chiadini E. Passardi A. Frassineti G.L. Ragazzini A. Detection and recovery of circulating colon cancer cells using a dielectrophoresis-based device: KRAS mutation status in pure CTCs Cancer Lett. 2013 335 225 231 10.1016/j.canlet.2013.02.015 23419522 69. Kubisch I. de Albuquerque A. Schuppan D. Kaul S. Schaich M. Stölzel U. Prognostic role of a multimarker analysis of circulating tumor cells in advanced gastric and gastroesophageal adenocarcinomas Oncology 2015 89 294 303 10.1159/000437373 26315108 70. Kaganoi J. Shimada Y. Kano M. Okumura T. Watanabe G. Imamura M. Detection of circulating oesophageal squamous cancer cells in peripheral blood and its impact on prognosis Br. J. Surg. 2004 91 1055 1060 10.1002/bjs.4593 15286970 71. Ito H. Kanda T. Nishimaki T. Sato H. Nakagawa S. Hatakeyama K. Detection and quantification of circulating tumor cells in patients with esophageal cancer by real-time polymerase chain reaction J. Exp. Clin. Cancer Res. 2004 23 455 464 15595636 72. Bidard F.-C. Peeters D.J. Fehm T. Nolé F. Gisbert-Criado R. Mavroudis D. Grisanti S. Generali D. Garcia-Saenz J.A. Stebbing J. Clinical validity of circulating tumour cells in patients with metastatic breast cancer: A pooled analysis of individual patient data Lancet Oncol. 2014 15 406 414 10.1016/S1470-2045(14)70069-5 24636208 73. Hiraiwa K. Takeuchi H. Hasegawa H. Saikawa Y. Suda K. Ando T. Kumagai K. Irino T. Yoshikawa T. Matsuda S. Clinical significance of circulating tumor cells in blood from patients with gastrointestinal cancers Ann. Surg. Oncol. 2008 15 3092 3100 10.1245/s10434-008-0122-9 18766405 74. Sclafani F. Smyth E. Cunningham D. Chau I. Turner A. Watkins D. A pilot study assessing the incidence and clinical significance of circulating tumor cells in esophagogastric cancers Clin. Colorectal Cancer 2013 13 94 99 10.1016/j.clcc.2013.11.003 24332356 75. Reeh M. Effenberger K.E. Koenig A.M. Riethdorf S. Eichstädt D. Vettorazzi E. Uzunoglu F.G. Vashist Y.K. Izbicki J.R. Pantel K. Circulating tumor cells as a biomarker for preoperative prognostic staging in patients with esophageal cancer Ann. Surg. 2015 10.1097/SLA.0000000000001130 25607767 76. Driemel C. Kremling H. Schumacher S. Will D. Wolters J. Lindenlauf N. Mack B. Baldus S.A. Hoya V. Pietsch J.M. Context-dependent adaption of EpCAM expression in early systemic esophageal cancer Oncogene 2013 33 4904 4915 10.1038/onc.2013.441 24141784 77. Schmidt F. Efferth T. Tumor heterogeneity, single-cell sequencing, and drug resistance Pharmaceuticals 2016 10.3390/ph9020033 27322289 78. Stoecklein N.H. Hosch S.B. Bezler M. Stern F. Hartmann C.H. Vay C. Siegmund A. Scheunemann P. Schurr P. Knoefel W.T. Direct genetic analysis of single disseminated cancer cells for prediction of outcome and therapy selection in esophageal cancer Cancer Cell 2008 13 441 453 10.1016/j.ccr.2008.04.005 18455127 79. Navin N. Kendall J. Troge J. Andrews P. Rodgers L. McIndoo J. Cook K. Stepansky A. Levy D. Esposito D. Tumour evolution inferred by single-cell sequencing Nature 2011 472 90 94 10.1038/nature09807 21399628 80. Møller E.K. Kumar P. Voet T. Peterson A. van Loo P. Mathiesen R.R. Fjelldal R. Grundstad J. Borgen E. Baumbusch L.O. Next-generation sequencing of disseminated tumor cells Front. Oncol. 2013 3 1 10 10.3389/fonc.2013.00320 23373009 81. Heitzer E. Auer M. Gasch C. Pichler M. Ulz P. Hoffmann E.M. Lax S. Waldispuehl-Geigl J. Mauermann O. Lackner C. Complex tumor genomes inferred from single circulating tumor cells by array-CGH and next-generation sequencing Cancer Res. 2013 73 2965 2975 10.1158/0008-5472.CAN-12-4140 23471846 82. Salvianti F. Rotunno G. Galardi F. de Luca F. Pestrin M. Maria A. di Leo A. Pazzagli M. Pinzani P. Feasibility of a workflow for the molecular characterization of single cells by next generation sequencing Biomol. Detect. Quantif. 2015 5 23 29 10.1016/j.bdq.2015.07.002 27077040 83. De Luca F. Rotunno G. Salvianti F. Galardi F. Pestrin M. Gabellini S. Simi L. Mancini I. Maria A. Pazzagli M. Mutational analysis of single circulating tumor cells by next generation sequencing in metastatic breast cancer Oncotarget 2016 10.18632/oncotarget.8431 84. Alix-Panabieres C. Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy Cancer Discov. 2016 6 479 491 10.1158/2159-8290.CD-15-1483 26969689 85. Joosse S.A. Gorges T.M. Pantel K. Biology, detection, and clinical implications of circulating tumor cells EMBO Mol. Med. 2014 1 12 10.15252/emmm.201303698 25398926 86. Pantel K. Alix-Panabieres C. Real-time liquid biopsy in cancer patients: Fact or fiction? Cancer Res. 2013 73 6384 6388 10.1158/0008-5472.CAN-13-2030 24145355 87. Maltoni R. Gallerani G. Fici P. Rocca A. Fabbri F. CTCs in early breast cancer: A path worth taking Cancer Lett. 2016 376 205 210 10.1016/j.canlet.2016.03.051 27060205

PMC005xxxxxx/PMC5000665.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081267ijms-17-01267ArticleChallenges in Translating GWAS Results to Clinical Care Scheinfeldt Laura B. 123*Schmidlen Tara J. 3Gerry Norman P. 34Christman Michael F. 3Cho William Chi-shing Academic Editor1 Institute for Genomics and Evolutionary Medicine, Temple University, Philadelphia, PA 19122, USA2 Department of Biology, Temple University, Philadelphia, PA 19122, USA3 The Coriell Institute for Medical Research, Camden, NJ 08103, USA; [email protected] (T.J.S.); [email protected] (N.P.G.); [email protected] (M.F.C.)4 Advanced BioMedical Laboratories, Cinnaminson, NJ 08007, USA* Correspondence: [email protected]; Tel.: +1-215-204-181404 8 2016 8 2016 17 8 126710 6 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Clinical genetic testing for Mendelian disorders is standard of care in many cases; however, it is less clear to what extent and in which situations clinical genetic testing may improve preventive efforts, diagnosis and/or prognosis of complex disease. One challenge is that much of the reported research relies on tag single nucleotide polymorphisms (SNPs) to act as proxies for assumed underlying functional variants that are not yet known. Here we use coronary artery disease and melanoma as case studies to evaluate how well reported genetic risk variants tag surrounding variants across population samples in the 1000 Genomes Project Phase 3 data. We performed a simulation study where we randomly assigned a “functional” variant and evaluated how often this simulated functional variant was correctly tagged in diverse population samples. Our results indicate a relatively large error rate when generalizing increased genetic risk of complex disease across diverse population samples, even when generalizing within geographic regions. Our results further highlight the importance of including diverse populations in genome-wide association studies. Future work focused on identifying functional variants will eliminate the need for tag SNPs; however, until functional variants are known, caution should be used in the interpretation of genetic risk for complex disease using tag SNPs. precision medicinecomplex diseasegenetic risk ==== Body 1. Introduction Several factors contribute to health-related quality of life, including: healthcare quality and access, individual behavior and lifestyle choices, environment, and genetics. Many clinical genetic tests have been developed and routinely used for Mendelian disorders, which are generally rare, single-gene disorders [1]; however, the development and deployment of clinical genetic testing for common complex diseases faces many additional challenges due to multifaceted genetic, environmental and behavioral risk factors, and to the limited understanding of functional variation that is assumed to underlie the genetic associations identified in genome-wide association studies (GWAS) [2,3]. Since the sequencing of the human genome and the subsequent genotyping of worldwide population samples, GWAS have become feasible and accessible, and to date over 1000 GWAS have been reported [4,5]. The general goals of GWAS are to identify candidate genes and regions that are associated with disease and disease-related phenotypes to better understand the underlying biology of disease, and to identify genetic risk factors that are associated with disease. Moreover, the majority of GWAS that have been conducted to date have almost exclusively focused on a narrow set of human population samples [3,6,7,8] with two resulting limitations: these studies are likely missing important genetic factors involved in disease, and the extent to which results from these studies will generalize to diverse clinical communities, such as those common in the United States, is an open question. Here we have designed a study to explore the later limitation of generalizability of genetic risk factors for two common complex diseases in which genetic risk has been previously shown to motivate health behaviors [9,10]. The Coriell Personalized Medicine Collaborative (CPMC) is a prospective research study that began in 2007 and is focused on evaluating the potential clinical utility of personalized risk reports for complex disease and drug response [11,12,13]. Two complex disease risk reports, coronary artery disease (CAD) and melanoma, have been evaluated for the potential to motivate health behaviors [9,10]. Both personalized risk reports consist of non-genetic risk factors, and genetic risk estimated from single genetic risk variants, rs1333049 [14] and rs910873 [15], respectively. Personalized risk reports for both of these diseases (Figures S1 and S2 display example reports) have been delivered to and viewed by thousands of CPMC research participants, and many of these participants have also completed outcome surveys that include information on self-reported behavior changes after report viewing as well as self-reported motivations for behavior change. In both cases, our previous work has shown a significant increase in healthy behavior change after viewing risk reports in the subset of research participants that also reported having increased genetic risk for the disease [9,10]. Given that there is self-reported behavioral data suggesting that genetic risk for CAD and melanoma has the potential to motivate healthy behavior change, it is critical to insure that reported genetic risk is accurately capturing disease risk. In order to evaluate the accuracy of the genetic variants used in risk assessment for CAD and melanoma across diverse human populations, we have leveraged the publicly available Phase 3 1000 Genomes Project whole genome sequencing data [16] and performed a simulation study to test how often reported genetic risk factors are likely to correctly measure actual genetic risk variants. 2. Results We have focused on the performance of the two single nucleotide polymorphisms (SNPs) that have been previously shown to be associated with CAD (rs1333049 [14]) and melanoma (rs910873 [15]), have been used in CPMC personalized risk reports [9,10], and have been shown to motivate heath behaviors [9,10]. Both SNPs have relatively strong effect sizes relative to what is commonly identified in GWAS. More specifically, rs1333049 heterozygotes and rs1333049 homozygotes have relative risks of 1.3 and 1.7, respectively; and rs910873 heterozygotes and rs910873 homozygotes have relative risks of 1.7 and 3.0, respectively. rs1333049 is not present in a gene, and as previously described, the association signal spans over 80 kb [14]. There are several genes in the region surrounding rs910873, and as noted previously, the association signal spans a 400 kb region that contains multiple genes, including PIGU, which is the gene containing rs910873 [15]. The risk allele frequencies of rs1333049 (C) and rs910873 (T) range from 0.21 to 0.54 and from 0.00 to 0.05, respectively across the 1000 Genomes population samples [16]. We evaluated the performance of these SNPs as tag or proxy SNPs for what are assumed to be underlying functional variants that are in linkage disequilibrium (LD) with the tag SNP. Using the Phase 3 1000 Genomes Project sequence data surrounding each SNP, we defined four distance windows (5, 10, 50, and 100 kb) and simulated a “functional” variant located within each window. In each population sample (Shown in Figure S3 and described in Table S1), we asked whether the tag SNP was accurately tagging (R2 ≥ 0.8) the simulated functional variant in each of the 26 1000 Genomes population samples. The simulated “functional” variant was chosen in such a way that the tag SNP had to accurately tag (R2 ≥ 0.8) it in the CEU (Utah Residents (CEPH) with Northern and Western European Ancestry), which is the 1000 Genomes population sample that most closely resembles the population samples used in the studies that identified the tag SNPs for each disease [14,15], and most closely resembles the majority of population samples that have been included in published GWAS [3,6,7,8]. For each genetic risk variant, we performed 106 independent simulations for each distance window. In total, we performed 8 × 106 simulations. Figure 1 displays the results for the CAD risk variant for each of the four distance windows. For the smallest distance window, 5 kb, there is reasonably good consistency in tagging among European population samples. The Toscani in Italia population sample is the only European population sample in which the reported tag SNP does not accurately tag the simulated functional variant in every simulation, but rather in 91% of the simulations. The tag SNP performed with lower accuracy in all of the other regional population samples, tagging the “functional” variant in 73% of the simulations across South Asian population samples; 64%–100% of the simulations across Native American population samples; 64% of the simulations across East Asian population samples; and 10%–27% across African and African American population samples. As the distance window increases, the simulated accuracies decrease. The tag SNP reliably tagged the “functional” variant in 65%–100% of the simulations across European population samples; 53%–65% of the simulations across South Asian population samples; 41%–100% of the simulation across Native American population samples; 53% of the simulations across East Asian population samples; and 6%–18% across African and African American population samples. By the 50 kb distance window, the tag SNP is tagging the simulated functional variant less than 50% of the time, in every population sample except for the Peruvians from Lima, Peru, Iberian Population in Spain, Finnish in Finland, and British in England and Scotland. For the 5, 10, 50, and 100 kb distance windows, the number of potential “functional” SNPs are 11, 17, 56, and 56, respectively. Figure 2 displays the results for the melanoma risk variant for each of the four distance windows. One aspect of this case study is that the tag SNP is not polymorphic in any of the East Asian or African population samples, which means that there is no information contained in this tag SNP regardless of the distance window for these peoples. For all of the other population samples, at 5 kb, the tag SNP is tagging all of the simulated “functional” variants in the distance window. At 10 kb, tagging is reduced to less than 80% in three of the four European population samples, two of the four Native American population samples, and all of the South Asian population samples. We note that for the 50 kb distance window, tagging actually increases in several population samples. We believe the pattern across distance windows for the melanoma risk variant is due to the relatively small number of possible “functional” SNPs present in the region. For 5, 10, 50, and 100 kb, the number of potential “functional” SNPs are 1, 4, 5, and 7, respectively. Therefore, for the 5 kb window, there is one possible “functional” SNP that is either tagged 100% of the time or 0% of the time across population samples. For the 10 kb window, there are 4 possible “functional” SNPs that can be tagged, so the possible tagging percentages are: 0%, 25%, 50%, 75% or 100%. 3. Discussion For two common complex diseases, CAD and melanoma, we have found that a significant proportion of participants with increased genetic risk self-report increased healthy behavior change (heart health and sun protective behaviors, respectively) [9,10]. These participants also tend to self-report that genetic risk was a motivating factor in their reported behavior change [9,10]. Given that the CPMC does not evaluate clinical end points, we interpret these results to support personalized genetic risk as a potential motivational tool for healthy behavior change. This interpretation is also supported by a study of smoking cessation [17] and by a minority subset (obesity, breast cancer, and rheumatoid arthritis) of evaluated complex diseases in a recent meta-analysis [18]. However, we also note that previous work has not found significant results when generalizing across complex diseases [18]. We suggest that these studies may have been underpowered to identify relationships between behavior change and personalized genetic risk given that they included research participants with no increased genetic risk in their experimental sample. We also suggest that personalized genetic risk may only motivate behavior change to mitigate risk for some but not all complex diseases. If the preliminary self-reported behavioral change motivation results from the CPMC hold over longer time frames (>3 months), then personalized genetic risk factors for at least some complex diseases may be worth utilizing for health behavior motivation. However, this potential is limited by the degree to which genetic risk factors for complex disease are understood and accurate. Ascertainment bias in published GWAS is a known problem that is expected to limit the relevance of results to persons of non-European descent [3,6]. Here, we have explored the extent to which replicated genetic risk factors for CAD and melanoma generalize across diverse populations. We leveraged the publically available 1000 Genomes whole genome sequencing dataset for 26 worldwide population samples to infer the performance of genetic risk factors for CAD and melanoma as tag SNPs or proxy measurements of simulated “functional” risk variants (i.e., biomarkers of genetic risk for disease). Our results demonstrate that the simulated accuracy of reported genetic risk variants for CAD and melanoma as tag SNPs for underlying functional variation is not encouraging. These results are true across global population samples as well as for population samples of European descent. The extent to which other disease-associated SNPs are located in regions of the genome in which LD varies across population samples will determine the generalizability of our results to other single SNP models of disease risk. However, ascertainment bias may have an even larger negative impact on polygenic models of disease risk where single SNP inaccuracies will likely compound the inaccuracy of the overall model. Our simulation results therefore suggest that ascertainment bias in GWAS is a serious concern that needs to be addressed. Improving the representation of diverse clinical population samples in GWAS provides many benefits that have already been recognized [8]. Overall, improved representation is likely to contribute to a more comprehensive understanding of all of the genetic risk factors, gene/gene interactions, and gene/environmental interactions that influence complex disease given that important genetic risk factors are likely to be missing from many published studies. This general goal will also facilitate downstream applications such as targeted drug therapies and improved personalized disease risk assessment. Thus, reducing ascertainment bias and improving the quality of results generated from GWAS is likely to improve the accuracy of clinical genetic risk assessment thereby facilitating trust with patients and the general public at large. That is, for complex disease genetic risk information to contribute to health behavior motivation, genetic risk estimates must be accurate and clinically meaningful, and patients will have to trust that the genetic risk estimates are meaningful. There are several limitations to the current study that should be considered alongside the results presented here. We only included singe SNP genetic risk factors for CAD and melanoma despite the more recent identification of multiple genetic risk factors, and we did not consider the potential impact of SNP imputation on tag SNP performance. However, we anticipate that the incorporation of multiple genetic risk factors and imputation is likely to compound tag SNP performance error rates. In addition, we only considered two complex disease case studies in which existing data demonstrates health behavior motivation potential [19], and results are likely to vary across other diseases. As many have already noted [2,6,20], enrolling diverse communities in GWAS is an non-trivial challenge, and inclusion of clinical population samples that are more representative of the United States requires the inclusion of marginalized communities that commonly lack access to clinical resources. We applaud the efforts of various funding agencies (including NIH and the UK Wellcome Trust) that currently support GWAS that include diverse peoples and address health disparities and hope that these efforts expand in the future. This support must be structured to permit community outreach and ongoing relationships with marginalized communities (e.g., [21]). We argue that research funding should also support studies that follow-up on the strongest GWAS results with resequencing and functional validation so that the underlying functional variants can be identified and tested directly and replace tag SNPs. Once these functional variants are known, the problematic and error-prone reliance of clinicians on “Race” [22] as a proxy for ancestry in diagnosis and treatment decision making can be improved with individualized health care based on direct measurement of genetic and non-genetic risk factors. 4. Materials and Methods 4.1. Coriell Personalized Medicine Collaborative The CPMC is a prospective research study that evaluates the potential clinical utility of genetic risk factors for complex disease. More details on the study can be found in [11,12,13,23,24,25,26]; however, here we will briefly summarize the framework of the study. CPMC scientists and staff use published GWAS to identify replicated SNPs that are associated with complex diseases in at least two independent clinical population samples. These diseases and associated SNPs are evaluated by an external board of experts (ICOB) to determine whether the diseases are potentially actionable, which is defined by the potential for behavior change and/or clinical screening to either mitigate disease risk or contribute to early detection. The disease/SNP sets are also evaluated to insure robust statistical association. The subset of approved disease/SNP sets are then incorporated into personalized risk reports that include genetic and non-genetic risk factors for disease and are periodically provided to research participants through an online web portal. Participants may choose whether or not to view a given personalized risk report, and may optionally complete outcome surveys detailing what if anything they did with the information they chose to view. 4.2. Genetic Risk Variants For the current study we have focused on two complex disease case studies: CAD and melanoma. Personalized risk reports for both of these diseases (see Figures S1 and S2 for example risk reports) were viewed by over 1000 CPMC participants, and both reports were included in the earliest outcome surveys. Previous analysis of these outcome surveys identified significant increases in health behavior change after viewing personalized risk reports in participants that also reported increased genetic risk for the disease [9,10]. 4.3. Sequencing Data The data included in the current analysis come from the Phase 3 dataset provided by the 1000 Genomes Project [16]. For all analyses included in the current study, identified related individuals were removed. In addition, one of each pair of unknown relatives present in the 20130606_sample_info_sample_info.csv file downloaded from the 1000 Genomes website were also removed. The 20130502 version of the Phase 3 1000 Genomes data was downloaded, and 5 megabases (Mb) surrounding the CAD and melanoma genetic risk variants (rs1333049 and rs910873, respectively) was extracted with VCFtools [27]. Descriptions of the Phase 3 1000 Genomes population samples are included in Table S1, and a corresponding world map is displayed in Figure S3. 4.4. Simulation Study We used the software package PLINK [28] to calculate pairwise R2 among all of the 1000 Genomes SNPs present within 5 Mb of each tag SNP genetic risk variant for each of the 26 1000 Genomes population samples. For each genetic risk variant, we performed 1 million simulations each for 4 distance bins: 5, 10, 50, and 100 kb. For each simulation and for each distance bin, we randomly chose a simulated “functional” variant within the distance bin that was in LD (R2 ≥ 0.80) with the reported genetic risk variant in the CEU population sample. We then recorded how often this simulated “functional” variant was also in LD (R2 ≥ 0.80) with the reported genetic risk variant in the other 25 1000 Genomes population samples. 5. Conclusions In summary, large-scale genomic association studies have the potential to identify genes and genetic regions that harbor risk factors for disease; however, future research efforts should address ascertainment bias in clinical research participants with broader inclusion and follow up association signals to identify the underlying functional variants to mitigate the high error rates estimated for tag SNP performance. Acknowledgments This work was supported by grants from the RNR Foundation and the United States Air Force. In addition, the authors would like to thank Neda Gharani for helpful discussions. Finally, we are grateful to the participants of the Coriell Personalized Medicine Collaborative. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1267/s1. Click here for additional data file. Author Contributions Laura B. Scheinfeldt designed the study, performed the analysis, and wrote the manuscript. Tara J. Schmidlen, Norman P. Gerry and Michael F. Christman contributed to the study design and manuscript text. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Coronary artery disease (CAD) genetic risk tag single nucleotide polymorphism (SNP) performance. Figure 1 displays four distance windows, each showing the proportion of correctly (R2 ≥ 0.08) tagged functional variants on the y-axis and each 1000 Genomes population sample on the x-axis. Population samples are color coded by continental region such that East Asia is purple, Europe is orange, Africa is turquoise, America is dark blue, and South Asia is green. Explanation of population sample abbreviations for the x-axis are shown in Table S1. Figure 2 Melanoma genetic risk tag SNP performance. Figure 2 displays four distance windows, each showing the proportion of correctly (R2 ≥ 0.08) tagged functional variants on the y-axis and each 1000 Genomes population sample on the x-axis. Population samples are color coded by continental region such that East Asia is purple, Europe is orange, Africa is turquoise, America is dark blue, and South Asia is green. Explanation of population sample abbreviations for the x-axis are shown in Table S1. ==== Refs References 1. McPherson E. Genetic diagnosis and testing in clinical practice Clin. Med. Res. 2006 4 123 129 10.3121/cmr.4.2.123 16809405 2. Burke W. Korngiebel D.M. Closing the gap between knowledge and clinical application: Challenges for genomic translation PLoS Genet. 2015 11 1267 10.1371/journal.pgen.1004978 25719903 3. Carlson C.S. Matise T.C. North K.E. Haiman C.A. Fesinmeyer M.D. Buyske S. Schumacher F.R. Peters U. Franceschini N. Ritchie M.D. Generalization and dilution of association results from european GWAS in populations of non-european ancestry: The page study PLoS Biol. 2013 11 1267 10.1371/journal.pbio.1001661 24068893 4. Leslie R. O’Donnell C.J. Johnson A.D. Grasp: Analysis of genotype-phenotype results from 1390 genome-wide association studies and corresponding open access database Bioinformatics 2014 30 i185 i194 10.1093/bioinformatics/btu273 24931982 5. Welter D. MacArthur J. Morales J. Burdett T. Hall P. Junkins H. Klemm A. Flicek P. Manolio T. Hindorff L. The NHGRI GWAS Catalog, a curated resource of SNP-trait associations Nucleic Acids Res. 2014 42 D1001 D1006 10.1093/nar/gkt1229 24316577 6. Bustamante C.D. Burchard E.G. de la Vega F.M. Genomics for the world Nature 2011 475 163 165 10.1038/475163a 21753830 7. Scheinfeldt L.B. Tishkoff S.A. Recent human adaptation: Genomic approaches, interpretation and insights Nat. Rev. Genet. 2013 14 692 702 10.1038/nrg3604 24052086 8. Adeyemo A. Rotimi C. What does genomic medicine mean for diverse populations? Mol. Genet. Genom. Med. 2014 2 3 6 10.1002/mgg3.63 24498625 9. Diseati L. Scheinfeldt L.B. Kasper R.S. Zhaoyang R. Gharani N. Schmidlen T.J. Gordon E.S. Sessions C.K. Delaney S.K. Jarvis J.P. Common genetic risk for melanoma encourages preventive behavior change J. Pers. Med. 2015 5 36 49 10.3390/jpm5010036 25695399 10. Scheinfeldt L.B. Schmidlen T.J. Gharani N. MacKnight M. Jarvis J.P. Delaney S.K. Gordon E.S. Kronenthal C.J. Gerry N.P. Christman M.F. Coronary artery disease genetic risk awareness motivates heart health behaviors in the Coriell Personalized Medicine Collaborative Expert Rev. Precis Med. Drug Dev. 2016 1 407 413 10.1080/23808993.2016.1197039 11. Keller M.A. Gordon E.S. Stack C.B. Gharani N. Sill C.J. Schmidlen T.J. Joseph M. Pallies J. Gerry N.P. Christman M.F. Coriell Personalized Medicine Collaborative® : A prospective study of the utility of personalized medicine Pers. Med. 2010 7 301 317 10.2217/pme.10.13 12. Gharani N. Keller M.A. Stack C.B. Hodges L.M. Schmidlen T.J. Lynch D.E. Gordon E.S. Christman M.F. The Coriell Personalized Medicine Collaborative pharmacogenomics appraisal, evidence scoring and interpretation system Genome Med. 2013 5 93 10.1186/gm499 24134832 13. Stack C.B. Gharani N. Gordon E.S. Schmidlen T. Christman M.F. Keller M.A. Genetic risk estimation in the Coriell Personalized Medicine Collaborative Genet. Med. 2011 13 131 139 10.1097/GIM.0b013e318201164c 21233721 14. Schunkert H. Gotz A. Braund P. McGinnis R. Tregouet D.A. Mangino M. Linsel-Nitschke P. Cambien F. Hengstenberg C. Stark K. Repeated replication and a prospective meta-analysis of the association between chromosome 9p21.3 and coronary artery disease Circulation 2008 117 1675 1684 10.1161/CIRCULATIONAHA.107.730614 18362232 15. Brown K.M. Macgregor S. Montgomery G.W. Craig D.W. Zhao Z.Z. Iyadurai K. Henders A.K. Homer N. Campbell M.J. Stark M. Common sequence variants on 20q11.22 confer melanoma susceptibility Nat. Genet. 2008 40 838 840 10.1038/ng.163 18488026 16. The 1000 Genomes Project Consortium A global reference for human genetic variation Nature 2015 526 68 74 26432245 17. Hartz S.M. Olfson E. Culverhouse R. Cavazos-Rehg P. Chen L.S. DuBois J. Fisher S. Kaphingst K. Kaufman D. Plunk A. Return of individual genetic results in a high-risk sample: Enthusiasm and positive behavioral change Genet. Med. 2015 17 374 379 10.1038/gim.2014.110 25166427 18. Bloss C.S. Schork N.J. Topol E.J. Effect of direct-to-consumer genomewide profiling to assess disease risk N. Engl. J. Med. 2011 364 524 534 10.1056/NEJMoa1011893 21226570 19. Adzhubei I.A. Schmidt S. Peshkin L. Ramensky V.E. Gerasimova A. Bork P. Kondrashov A.S. Sunyaev S.R. A method and server for predicting damaging missense mutations Nat. Methods 2010 7 248 249 10.1038/nmeth0410-248 20354512 20. Wallerstein N.B. Duran B. Using community-based participatory research to address health disparities Health Promot. Pract. 2014 7 312 323 10.1177/1524839906289376 16760238 21. Taylor H.A. Jr. Wilson J.G. Jones D.W. Sarpong D.F. Srinivasan A. Garrison R.J. Nelson C. Wyatt S.B. Toward resolution of cardiovascular health disparities in African Americans: Design and methods of the Jackson Heart Study Ethn. Dis. 2005 15 Suppl. S6 4 17 22. Braun L. Fausto-Sterling A. Fullwiley D. Hammonds E.M. Nelson A. Quivers W. Reverby S.M. Shields A.E. Racial categories in medical practice: How useful are they? PLoS Med. 2007 4 1267 10.1371/journal.pmed.0040271 17896853 23. Scheinfeldt L.B. Gharani N. Kasper R.S. Schmidlen T.J. Gordon E.S. Jarvis J.P. Delaney S. Kronenthal C.J. Gerry N.P. Christman M.F. Using the Coriell Personalized Medicine Collaborative data to conduct a genome-wide association study of sleep duration Am. J. Med. Genet. B Neuropsychiatr. Genet. 2015 168 697 705 10.1002/ajmg.b.32362 26333835 24. Schmidlen T.J. Scheinfeldt L. Zhaoyang R. Kasper R. Sweet K. Gordon E.S. Keller M. Stack C. Gharani N. Daly M.B. Genetic knowledge among participants in the Coriell Personalized Medicine Collaborative J. Genet. Couns. 2016 25 385 394 10.1007/s10897-015-9883-z 26306685 25. Schmidlen T.J. Wawak L. Kasper R. Garcia-Espana J.F. Christman M.F. Gordon E.S. Personalized genomic results: Analysis of informational needs J. Genet. Couns. 2014 23 578 587 10.1007/s10897-014-9693-8 24488620 26. Shahabi P. Scheinfeldt L.B. Lynch D.E. Schmidlen T.J. Perreault S. Keller M.A. Kasper R. Wawak L. Jarvis J.P. Gerry N.P. An expanded pharmacogenomics warfarin dosing table with utility in generalised dosing guidance Thromb. Haemost. 2016 116 337 348 10.1160/TH15-12-0955 27121899 27. Danecek P. Auton A. Abecasis G. Albers C.A. Banks E. DePristo M.A. Handsaker R.E. Lunter G. Marth G.T. Sherry S.T. The variant call format and VCFtools Bioinformatics 2011 27 2156 2158 10.1093/bioinformatics/btr330 21653522 28. Purcell S. Neale B. Todd-Brown K. Thomas L. Ferreira M.A. Bender D. Maller J. Sklar P. de Bakker P.I. Daly M.J. PLINK: A tool set for whole-genome association and population-based linkage analyses Am. J. Hum. Genet. 2007 81 559 575 10.1086/519795 17701901

PMC005xxxxxx/PMC5000666.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081268ijms-17-01268ArticleParentage-Based Group Composition and Dispersal Pattern Studies of the Yangtze Finless Porpoise Population in Poyang Lake Chen Minmin 12Zheng Yang 13Hao Yujiang 1Mei Zhigang 1Wang Kexiong 1Zhao Qingzhong 1Zheng Jinsong 1*Wang Ding 1*Lin Li Academic Editor1 The Key Laboratory of Aquatic Biodiversity and Conservation of the Chinese Academy of Sciences, Institute of Hydrobiology of the Chinese Academy of Sciences, Wuhan 430072, China; [email protected] (M.C.); [email protected] (Y.Z.); [email protected] (Y.H.); [email protected] (Z.M.); [email protected] (K.W.); [email protected] (Q.Z.)2 Research Center of Aquatic Organism Conservation and Water Ecosystem Restoration in Anhui Province, School of Life Sciences, Anqing Normal University, Anqing 246133, China3 University of the Chinese Academy of Sciences, Beijing 100049, China* Correspondence: [email protected] (J.Z.); [email protected] (D.W.); Tel.: +86-27-8780-1331 (J.Z.); +86-27-6878-0178 (D.W.); Fax: +86-27-6878-0123 (J.Z. & D.W.)11 8 2016 8 2016 17 8 126823 5 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Social behaviors are poorly known for the critically endangered Yangtze finless porpoise (YFP, Neophocaena asiaeorientalis asiaeorientalis). Here, group composition and dispersal patterns of the YFP population living in the Poyang Lake were studied by parentage-based pedigree analyses using 21 microsatellite loci and a 597 bp segment of the mitochondrial DNA control region. In this study, 21 potential mother-offspring pairs and six potential father-offspring pairs (including two potential parents-offspring pairs) were determined, among which 12 natural mother-offspring groups and a maternal group of three generations were found. No genetically-determined fathers were found associated with their offspring. This study also found that maternally related porpoises at the reproductive state tend to group together. This suggest maternal relationship and reproductive state may be factors for grouping in the YFP population. In natural mother-offspring groups, male offspring were all younger than two years old, which suggest male offspring may leave their mothers at approximately two years of age, or at least they were not in tight association with their mothers as they may have been under two years old. However, female offspring can stay longer with their mothers and can reproduce in the natal group. Neophocaena asiaeorientalis asiaeorientalisparentage identificationsocial structurematrilinealdispersal pattern ==== Body 1. Introduction The social behavior of cetaceans is both complex and interesting. Advances in molecular techniques have led to an increasing number of studies that combine molecular, observational, and photo-ID data to reveal a variety of grouping and dispersing patterns in cetacean species [1,2]. Various cetacean social behaviors have been reported, including fluid fission-fusion societies described in small delphinid species (e.g., bottlenose dolphins (Tursiops aduncus) [3]; spinner dolphins (Stenella longirostris) [4]), matrilineal groups in larger toothed whales (e.g., killer whales (Orcinus orca) [5]; sperm whales (Physeter macrocephalus) [6]), and associations among individuals of both sexes or just a single sex that vary in size (from few to hundreds of individuals), duration (temporary or permanent), and composition (single or multiple generations) of Atlantic white-sided dolphins (Lagenorhynchus acutus) [7]. The social behavior of freshwater dolphins was also observed. Smith and Reeves [8] reviewed that Amazon River dolphins (Inia geoffrensis) sometimes form loose fishing groups and male-on-male aggression is common. Irrawaddy dolphins (Orcaella brevirostris) form fission-fusion group dynamics with frequent social interactions and cooperative feeding, and mother-young associations can be observed in the Ganges and Indus dolphins (Platanista spp.). Still, compared to marine cetaceans, knowledge about the social behavior of freshwater cetaceans remains very limited. The Yangtze finless porpoise (YFP, Neophocaena asiaeorientalis asiaeorientalis) is a small, freshwater toothed whale that occurs only in the middle and lower reaches of the Yangtze River (from Yichang to Shanghai) and its adjoining lakes (Poyang and Dongting) [9]. Due to its small population size, sharply declining population, and high probability of extinction, the YFP was recently reclassified a Critically Endangered (CR) population in the International Union for the Conservation of Nature (IUCN) Red List of Threatened Species [10,11]. Since the YFP is notoriously difficult to identify and track using normal observation methods (e.g., visual surveys and photo-identification), primarily due to their small size (≈1.5 m in length), lack of a dorsal fin, and behavior (e.g., only surface for 1–2 s at a time), little is known about their group composition and dispersal patterns. Understanding the social behaviors of the YFP will be instrumental for effective conservation and contribute to our general knowledge of the social behaviors of freshwater cetaceans [12,13,14]. Previous research suggests the YFP in the Yangtze main stream live either alone, or in groups of 2–20 individuals [15,16]. Groups consisting of two to three individuals have been suggested to be the most common and are generally called “core units”. Groups of >20 individuals are rare and likely consist of several core units. Similar grouping patterns have also been reported for porpoises living in semi-natural reserves [17,18,19]. Due to the limits of observational data, none of these previous studies have analyzed genetic relationships among individuals within YFP groups, nor investigated dispersal patterns. Poyang Lake is the most important limnic habitat of the YFPs with a population of ≈450 individuals (almost 50% of the total population of the YFPs [20]). To investigate the health and social structure of the Poyang Lake population, four capture-release surveys were conducted in the spring of 2009, 2010, 2011, and 2015, during which all captured individuals were marked with an internal ID and genetic samples were obtained. This dataset offers a unique opportunity to study aspects of social behaviors using both genetic and observational data. In this study, we used observational data collected during these four capture surveys and genetic data obtained from 21 microsatellite loci and a 597 bp highly variable segment of the mitochondrial DNA control region, to study the relationships of individuals in natural groups and the dispersal patterns of the YFP population living in the Poyang Lake. 2. Results 2.1. Genetic Variation A total of 171 alleles were detected at 21 microsatellite loci among 122 individuals. No evidence was found for null alleles, stuttering and allele dropout in each locus by the program MICRO-CHECKER, at a confidence level of 95%. The number of alleles (Na) ranged from 4 to 16 (mean 8.1), with observed heterozygosity (Ho) ranging from 0.374 to 0.854 (mean 0.661) and expected heterozygosity (He) ranging from 0.361 to 0.828 (mean 0.674; Table 1). The polymorphic information content (PIC) ranged from 0.345 to 0.801 (mean 0.629; Table 1). The combined non-exclusion probability for one candidate parent (Ne-1p) and for one candidate parent given the genotype of a known parent (Ne-2p) were 7.20 × 10−4 and 2.17 × 10−6, respectively (Table 1). In other words, the total exclusion probability of the combined loci when no parents were known (PE II) was 99.93%, and when one parent was known (PE I) was 99.99%. No deviation from Hardy-Weinberg equilibrium was detected for each locus. The inbreeding coefficient index (Fis) of each locus ranged from −0.182 to 0.178, with an average of 0.0002 (p > 0.05). Among the 122 individuals, three mitochondrial DNA (mtDNA) haplotypes were found: NAACR-Hap1, NAACR-Hap2, and NAACR-Hap8 (GenBank accession number: KC135874, KC135875 and KC135881). NAACR-Hap1, NAACR-Hap2 and NAACR-Hap8 were shared by 53, 64, and five individuals, respectively (Table S1). 2.2. Parentage Assignment and Relatedness Twenty-one potential mother-offspring pairs and six potential father-offspring pairs (including two potential mother-father-offspring families: 2015F11-2015M3-2015F12, 2009F7-2009M2-2009F8) were detected by CERVUS (Figure 1). Among the detected mother-offspring pairs, ten female offspring ranging 0.1–9.5 years old, were found associated with the mother according to the field observation record, of which four were aged >2 years old (Figure 1, Table S2), while six male calves ranging 0.3–1.7 years old were found associated with the mother (Table S2). Two female individuals, 2015F2 (5.2 years old) and 2015F10 (1.8 years old), and three male individuals, 2009M14 (6.8 years old), 2015M8 (5.4 years old), and 2011M6 (3.7 years old), were found not associated with the mothers (Table S2). All mothers and their offspring shared the same mtDNA haplotype (Table S2). No genetically-determined fathers were found associated with their offspring (Table S2). The mean pairwise relatedness index (r) value for the potential mother-offspring pairs was 0.4479 (0.2924–0.5551), and for the potential father-offspring pairs was 0.3863 (0.2295–0.5032) (Figure 1). The mean overall r value among all adults and among all individuals was 0.039 and 0.040, respectively. In general, individuals with relatedness values larger or equal than half-siblings (half-sibs) (the theoretical r ≥ 0.25) are deemed related, while all other individuals remain unrelated (the theoretical r < 0.25) [21]. However, in natural populations, the r values estimated from microsatellite loci may fluctuate around the theoretical value. Blouin et al. [22] suggested using the midpoint between the means of the two distributions as the cut-off value for classification. For example, if one wanted to distinguish between half-sibs (the theoretical r = 0.25) and lower grade related individuals (the theoretical r = 0.125) in a natural population, the cut-off value should be 0.1875. That means individual pairs with r ≥ 0.1875 would be classified as belonging to the category whose relatedness is larger than half-sibs. Using the cut-off values of r ≥ 0.1875, we found that 4.48% of all pairs of individuals were related to some extent. 2.3. Composition of Natural Groups During all four capture-release surveys (particularly those conducted in 2009 and 2010 in the channel), human disturbance due to the use of chasing boats and noise may have separated or centralized porpoise individuals. Therefore, most large groups captured in a single net were unlikely to be natural groups. In 2011 and 2015, the survey was conducted in sandpit areas. No chasing boats were used. Hence, human disturbance and sampling bias were largely reduced. We were fortunate enough to detect a maternal line consisting of three generations in a single group captured in 2011 (Figure 1). There are four mother-offspring pairs detected in this maternal line, and all members in this maternal line shared the same mtDNA haplotype (NAACR-Hap2; Table S1). Based on the detected genetic relationship and the observation data which showed that they grouped together (Table S2), we treated members in the maternal line as a natural group. Additionally, twelve mother-offspring groups (groups 2015F2-2015F1, 2015F11-2015F12, 2015F19-2015F20, 2011F2-2011F1, 2011F4-2011M3, 2011F12-2011F10, 2011F23-2011F22, 2011F24-2011M19, 2011F25-2011M20, 2010F11-2010F12, and 2009F7-2009F8, Group 2009F7-2015M7) were further detected (Figure 1). These were all considered to be natural mother-offspring groups based on a high degree of behavioral interaction (field observation data), and genetic relatedness. 3. Discussion 3.1. Parentage and Relatedness When using microsatellite markers for paternity analysis, the credibility of the results is highly dependent on the exclusion probability [14,23]. As revealed by previous studies, paternity results are credible when PE I and PE II values exceeded 99.9% and 99%, respectively [14,23,24,25,26]. In this study, PE I and PE II values were 99.99% and 99.93%, respectively, indicating that the 21 microsatellite loci we used were appropriate for parentage analysis. This was also supported by the fact that nearly all suspected mother-offspring pairs (those observed together and captured in one net) were genetically assigned mother-offspring status (Figure 1, Table S3). In recent years, serious human interference within the estuary area of the Poyang Lake (e.g., the construction of two bridges across the outlet area, and numerous sand-transport vessels) has caused the river-lake migration of the YFPs to be largely restricted. Thus, inbreeding within the Poyang Lake population has become a matter of great concern. The genetic diversity of this population, with a mean Ho = 0.661 and a mean He = 0.674, is moderate compared to other cetaceans [27,28]. Our relatedness analysis revealed that there is no genetic signature of inbreeding in the YFP population living in Poyang Lake. First, no significant Fis was detected in this study. Second, our relatedness index did not reflect inbreeding. In a free-range population, the average r between potential parents should be close to zero (e.g., −0.014 in Ursus arctos, [14]; 0.056 in Ursus americanus, [13]), and we found that the mean r between potential parents was very low (0.039). Third, as reported by Csilléry et al. [21], for natural populations, we found that the ratio of individual pairs with some relatedness was <10% (4.4%), further indicating little inbreeding in this population currently. Conversely, the population in Tian’ezhou ex situ reserve demonstrates 26.14% of individual pairs have some relatedness (r > 0.1875) [29]. Chen et al. [29] inferred this ex situ population was probably in high risk of, or has already been suffering from, inbreeding. 3.2. Group Composition Groups composed by maternal relationship are very common in marine odontocetes (e.g., long-finned pilot whales (Globicephala melas) [30]; belugas (Delphinapterus leucas) [31]; sperm whales (Physeter macrocephalus) [32]; killer whales (Orcinus orca) [33]). For instance, adult female sperm whales associate with sub-adults to form cohesive ‘social units’ that can remain together over several years [32]. Similarly, killer whales are also characterized either by multi-matrilines (resident killer whales [33]) or a single matriline (transient killer whales [5]). In this study, we also detected a maternal line in one YFP group living in the Poyang Lake. Individuals 2011F19, 2011F20, 2011F21, 2011M15, and 2011M16 were found to be grouping together and were subsequently captured in one net. Genetic results showed that this group comprised of four mother-offspring pairs belonging to one maternal linage. The oldest female, 2011F19, is at the top of the maternal linage and mother to two other adult females (2011F20, 2011F21) and a calf (2011M15), and grandmother to calf 2011M16, and the offspring of female 2011F21 (Figure 2). Furthermore, female 2011F20 was confirmed pregnant through B-type ultrasonic inspection. The matrilineal grouping pattern of the YFPs is further supported by the structure of another group (including individuals from 2009M1 to 2009F3) (Table S3). Groups captured in 2009 cannot be treated as natural groups because sound chasing operations during the survey may have disturbed the natural behavior in these porpoises. Nevertheless, our genetic results revealed that adult females 2009F1, 2009F2, and 2009F3 in this group were very closely related, with r values close to that of full-sibs (0.4012 to 0.5372). As these three females shared the same mtDNA haplotype (Table S1), we infer that they have high matrilineal relationship. B-type ultrasonic inspection further confirmed these females were all pregnant. It is interesting that maternally related female YFPs grouping together were all in various reproductive stages. This phenomenon has also been found in other cetacean species, where females with a dependent calf often form nursing groups to reduce unpredictable risk [34]. Indeed, females may form loose associations with related or unrelated females, preferentially associating with other females in similar reproductive states. For example, Möller et al. [35] found that reproductive state seemed to influence associations between female Indo-Pacific bottlenose dolphins (Tursiops aduncus), where females with same aged calves within social clusters usually exhibited strong association coefficient. 3.3. Dispersal Patterns Our results revealed that in natural mother-offspring groups, male calves were all <2 years old (0.3–1.7 years old), whereas the female calves ranged from 0.1 to 9.5 years of age. Additionally, three male calves not found in groups with their mothers were all older than two years (6.8-year-old 2009M14, 5.4-year-old 2015M8, and 3.7-year-old 2011M6). Furthermore, two mother-male offspring pairs found trapped in the Yangtze mainstream in January 2014 had also been identified as mothers with their male calves younger than two years old (0.58-year-old and 0.75-year-old, respectively; Ding Wang, unpublished data). This discrepancy could result if male calves disperse from their natal groups at approximately two years old, or at least are not in tight association with their mothers as they may have been at under two years old. Female offspring might associate with their mothers at an older age, and can reproduce or perhaps return to reproduce in the natal group. Still, this association may not be strict as female offspring also emigrate out and raise offspring alone. For example, in this study we found twelve mother-calf pairs that remained alone. 4. Materials and Methods 4.1. Study Location and Distribution of Porpoises To investigate the health and social structure of the YFP population living in Poyang Lake, four capture-release surveys were conducted in the spring of 2009–2011, and 2015 (20–24 February 2009; 2–11 March 2010; 21–25 February 2011; and 11–20 March 2015), during the dry season of the lake. During this season, Poyang Lake is reduced to a set of channels together with dispersed sandpit areas (Figure 2). In this season, porpoises mainly distribute along the main channel between Hukou and Kangshan, and also in some large sandpit areas between Duchang and Yongxiu (Figure 2). Since those sandpits are connected to the main channel by shallow waters (depth usually ≤2 m), some porpoises are restricted to those sandpit areas during the entirety of the dry season. In 2009 and 2010, capture-release surveys were conducted in the channel of Duchang County, and in 2011 and 2015, the surveys were conducted in the large sandpit areas located between Duchang and Yongxiu (Figure 2). 4.2. Sample Collection The well-developed “sound chase and net capture” method, which had been specially developed by the Institute of Hydrobiology of the Chinese Academy of Sciences to capture porpoises in the Yangtze main stream or natural reserves, was utilized to capture porpoises in the channel of Poyang Lake during the dry season. Once a group of porpoises (usually ≥2 individuals) was observed by the searching boat (a speed boat equipped with a 40 horsepower engine), chasing boats (consisting of 10–12 fishing boats, each about 12 m long equipped with a 15 horsepower engine) would subsequently arrange themselves in line or curve and move in synchrony keeping a distance of about 50 m to generate underwater noise and form an invisible sound barrier that would force the animals swimming slowly to a shallow water near the shore. Afterward, two net boats (driving either face-to-face or in the opposite direction) would quickly release large-meshed nets to form a large enclosure (about 1 km2) to surround the porpoises. The porpoises would then be driven to a smaller area with a radius of approximately 100 m, where small-meshed capture nets would be released quickly to surround the porpoises. All animals were allowed to swim freely in the small enclosure to have enough rest before any further manipulation. After approximately 30 min, the fishermen would draw the capture nets slowly to compress the enclosure until the animals were caught safely. Conversely, because the porpoises were restricted to a relatively small and shallow area (<1 km2) in the sandpit areas, the sound chasing operation was rendered unnecessary, leaving only net capture protocols to be used to catch the animals. In each capture operation, only a single or a small group of porpoises were captured. After the porpoises were successfully caught and sent to the examination platform, gender was identified, and then each male or female was given a serial ID number (e.g., 2009M1–2009M21, 2009F–2009F8) before proceeding. The serial ID numbers of those porpoises that had been captured in the same net were recorded. If an adult female and a calf had been captured in the same net, a suspected mother-calf pair was then recorded. Reproductive state (pregnancy or lactation) of the adult female was also noted. Body weight and length were measured and recorded. Blood samples were also drawn from the vein in the fluke using a disposable syringe. Blood was anti-coagulated with acid-citrate-dextrose (ACD), and then preserved in liquid nitrogen until DNA extraction. A total of 132 individuals (including 10 recaptured individuals) were captured and sampled: 29 in 2009, 22 in 2010, 46 in 2011 and 35 in 2015. All captured individuals were marked using a unique internal passive intergrated tag label (PIT label; HT850, Hongteng Company, Guangzhou, China), which could be identified via scanning. Group size and individual interaction information were all recorded during capturing (for details see Table S1). All capture-release surveys were authorized by the Poyang Lake Fishery and Fishing Administration Office of Jiangxi Province. All sampling was conducted in accordance with the Regulations of the People’s Republic of China for the Implementation of Wild Aquatic Animal Protection (promulgated in 1993), and adhering to all ethical guidelines and legal requirements in China. 4.3. DNA Extraction and PCR Amplification Genomic DNA was isolated using the Whole Genome DNA Extraction Kit (SBS, Shanghai, China) following the manufacturer’s instructions. Twenty-one polymorphic microsatellite loci (simple sequence repeats, SSR) were then used in parentage identification. Markers used included SSR1, SSR5, SSR8, SSR15, SSR22, SSR40, SSR41, SSR42, SSR51, SSR59, SSR63, SSR69, SSR71, SSR73, and SSR75 from Neophocaena phocaenoides asiaeorientalis [36,37,38]; PPHO130 from Phocoena phocoena [39], and NP391, NP404, NP409, NP464, and NP428 from Neophocaena phocaenoides [40,41]. PCR was performed in 15-μL reaction volumes containing 1 μL of template DNA, 1.5 μL 10× buffer, 0.7 μM of each primer, 0.25 mM deoxynucleotides (dNTPs), and 0.2 U of Taq DNA polymerase (Biostar; Wuhan Tianyuan Huida Biotech Company, China). Amplifications were carried out with conditions consisting of 95 °C for 5 min, followed by 33 cycles of denaturation at 95 °C for 30 s, annealing at 59.5 °C for 30 s and extension at 72 °C for 30 s, with a final extension step at 72 °C for 5 min. PCR products were separated by capillary electrophoresis on an ABI3130XL automated sequencer (Applied Biosystems, Foster City, CA, USA) and alleles were sized against the internal size standard (GeneScan ROX 500, ThermoFisher Scientific, Shanghai, China) using GeneMapperID v3.2 (Applied Biosystems). To minimize scoring error, samples that were homozygous, had low frequency alleles (only appeared in one or two individuals), or exhibited stutter bands, were amplified and genotyped at least three times. We used MICRO-CHECKER version 2.2.3 (Norwich Research Park, Norwick, UK) [42] to check for null alleles, stuttering error, and allele dropout for each locus, at a confidence level of 95%. A mtDNA control region segment of 597 bp, located at 84–680 bp of the complete control region, was selected due to highly variability [43]. The sequence was amplified with a forward primer (5′-GAA TTC CCC GGT CTT GTA AAC C-3′) and a reverse primer (5′-GGT TTG GGC CTC TTT GAG AT-3′) [44]. PCR amplifications were carried out in 25 μL reactions containing 10–100 ng genomic DNA, 0.6 μM of each primer, 2.5 μL 10× buffer, 0.25 mM dNTPs, and 1 U of Taq DNA polymerase (Biostar). Amplifications startedat 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 45 s, annealing at 60 °C for 45 s, and extension at 72 °C for 90 s, with a final extension step at 72 °C for 7 min. PCR products were then purified using a purification kit (PCR Product Purification Kit, BioTeke, Beijing, China) and sequenced in both directions using the PCR primers. Sequencing was performed on an ABI3130 DNA sequencer (Applied Biosystems). 4.4. Data Analysis We assessed measures of genetic diversity of microsatellite loci, including Na, Ho, He, and PIC [45,46], using CERVUS version 3.0 (Field Genetics Ltd., London, UK) [47,48]. The Ne-1p and Ne-2p for each microsatellite locus, and the combined values of microsatellite loci for Ne-1p and Ne-2p, were calculated by using CERVUS version 3.0 [47,48,49]. Hardy-Weinberg equilibrium across all microsatellite loci was assessed via an exact probability test implemented in GENEPOP version 4.0 (Laboratiore de Genetique et Environment, Montpellier, France) [50]. FSTAT version 2.9.3.2 (University of Lausanne, Lausanne, Switzerland) was used to calculate the inbreeding coefficient index Fis [51]. 4.5. Detecting Potential Parents We used the age-length formula established by Zhang [52] to calculate the age of each individual. The relationship between age (x) and body length (in cm) of male (Lm) and female (Lf) YFP can be calculated as follows: Lm = 114.4458 x0.1410 (♂ ≤ 13.0 years) (1) and Lf = 116.2519 x0.0947 (♀ ≤ 16.5 years) (2) Although the age at first reproduction in females remains unknown, Zhang [52] and Wu et al. [53] estimated that YFP females are sexually mature at approximately four years of age and males at 4.5 years of age. In order to not miss potential parents, we considered all female and male porpoises with calculated age ≥3 years old as potential parents. Accordingly, we considered individuals three years younger than the oldest within the population to be potential offspring. 4.6. Parentage Analysis Parentage analysis was conducted using the maximum likelihood (ML) method implemented in program CERVUS version 3.0 [47,48]. This method compares the likelihood of the two most likely mothers or fathers. For each offspring, the difference between the likelihoods of the two most probable mothers or fathers produces a Δ score. Simulations were conducted to estimate the critical values of Δ required to assign parentage with a certain degree of confidence, based on the assumptions made about the population. Parentage assigned at both the 95% and 80% confidence levels were reported, as determined by the critical Δ score. We set parameters for the simulation as follows: the proportion of loci was 1.0, the proportion of potential father and mother were both 20%, the level of potential mistyping was 1%. Simulations were conducted for 100,000 repetitions. The r was also calculated to analyze kinship. The theoretical values between parent-offspring and full-sibs are 0.5; those between half-sibs are 0.25 [21,54]. To calculate r estimates, we used the triadic likelihood estimator (TrioML [55]). This estimator computes the relatedness of a dyad in relation to a third reference individual in order to minimize errors stemming from identity-in-state rather than identity-by-descent. It further allows the specification of a genotyping error rate and is bounded between 0 and 1, a more legitimate range than that of other estimators. An evaluation using empirical and simulated data for seven different estimators showed that the TrioML estimator produced the most accurate estimates of kinship [55]. 5. Conclusions In this study, group composition and dispersal patterns were studied for the YFP population living in Poyang Lake. To avoid human disturbance to natural groups and guarantee the reliability of the results, we identified natural groups based on both genetic parentage and observational data. Results indicated maternal relationship and reproductive status may be important factors for group composition of females. The dispersal patterns of the YFP showed that male calves may disperse from their mothers at approximately two years old, or at least they were not in tight association with their mothers as they may have been under two years old. Female offspring are observed to stay longer with their mothers and can reproduce in the natal group. In the wild, the YFP is difficult to identify and track, and samples used for relatedness studies are difficult to obtain. Therefore, social behavior studies for this population in particular are poor. This study collected information from four capture-release surveys, however, accurate and valuable information is still restricted. For example, calculated parentage was limited, which may be because of the potential high mortality of calves in the Poyang Lake during the dry season, ultimately reducing detectable aspects of social behavior in this study. In addition, although we estimated the dispersal window of male calves, these observations were also limited to few datum points and they should be prudently treated. Acknowledgments We would like to thank our staff for collecting samples. We thank Wang Jinliang (Institute of Zoology, Zoological Society of London) for his kind help with the relatedness analysis. We also thank Kathryn Stewart for proofreading the revision. This work was supported by grants from the Special Fund for Agro-scientific Research in the Public Interest (No. 201203086) to Ding Wang, Jinsong Zheng and Yujiang Hao; the National Natural Science Foundation of China (No. 31430080 to Ding Wang, and No. 31000168 to Jinsong Zheng) and the Knowledge Innovation Program of Chinese Academy of Sciences (No. KSCX2-EW-Z-4) to Ding Wang. This work was also supported by Major Project of Natural Science Research in Anhui Province (No. KJ2016A863) to Minmin Chen. Supplementary Materials Supplementary materials can be found at http://www.mdpi.com/1422-0067/17/8/1268/s1. Click here for additional data file. Author Contributions Minmin Chen conducted the study, analyzed data and wrote the manuscript. Jinsong Zheng and Ding Wang designed the study and revised the manuscript. Yang Zheng, Yujiang Hao and Zhigang Mei conducted the study and supplied assistance. Kexiong Wang directed the capture-release surveys and revised the manuscript. Qingzhong Zhao helped collect samples. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Twenty-one mother-offspring pairs and six father-offspring pairs detected by CERVUS in the Yangtze finless porpoise population living in Poyang Lake. Parent-offspring pairs in dotted boxes were natural groups. The maternal line consisting of three generations was a natural maternal group captured in a sandpit in 2011. Blue represents offspring younger than two years old. Orange stands for offspring older than two years old. Pairwise relatedness index r was calculated by the triadic likelihood estimator (TrioML). Figure 2 Water coverage of the Poyang Lake in the early spring. Porpoises primarily distribute along the main channel between Hukou and Kangshan, and also in some large sandpit areas between Duchang and Yongxiu. The blue box represents the sampling area in 2009 and 2010. The red circle represents the sampling area in 2011 and 2015. ijms-17-01268-t001_Table 1Table 1 Characteristics of genetic diversity at 21 microsatellite loci for 122 Yangtze finless porpoises in the Poyang Lake. Locus Na Ho He PIC Ne-1p Ne-2p Fis SSR1 6 0.639 0.629 0.567 0.785 0.629 −0.017 SSR5 9 0.795 0.799 0.765 0.580 0.402 0.005 SSR8 6 0.800 0.740 0.691 0.678 0.503 −0.082 SSR15 11 0.746 0.716 0.660 0.708 0.541 −0.041 SSR22 4 0.689 0.665 0.591 0.777 0.628 −0.041 SSR40 10 0.746 0.746 0.705 0.657 0.477 0.000 SSR41 6 0.425 0.673 0.627 0.736 0.561 0.009 SSR42 5 0.610 0.671 0.608 0.751 0.592 0.092 SSR51 12 0.780 0.796 0.765 0.574 0.396 0.017 SSR59 10 0.854 0.828 0.801 0.522 0.349 −0.031 SSR63 7 0.742 0.632 0.559 0.787 0.641 −0.182 SSR69 7 0.694 0.611 0.575 0.784 0.606 −0.136 SSR71 7 0.529 0.523 0.490 0.848 0.682 −0.013 SSR73 6 0.782 0.793 0.756 0.599 0.420 0.013 SSR75 16 0.672 0.644 0.625 0.733 0.54 −0.044 NP391 12 0.415 0.475 0.441 0.877 0.723 0.128 NP404 5 0.504 0.555 0.459 0.844 0.736 0.077 NP409 9 0.664 0.748 0.706 0.652 0.476 0.113 NP428 5 0.374 0.361 0.345 0.930 0.792 −0.035 NP464 9 0.628 0.764 0.724 0.634 0.456 0.178 PPHO130 9 0.795 0.787 0.752 0.596 0.419 −0.010 Mean of 21 loci 8.1 0.661 0.674 0.629 Combined 7.20 × 10−4 Combined 2.17 × 10−6 0.0002 (p > 0.05) Na: Number of alleles per locus, Ho: observed heterozygosity, He: expected heterozygosity, PIC: polymorphic information content, Ne-1p: average non-exclusion probability when no parents were known, Ne-2p: or when one parent was known, Fis: inbreeding coefficient index. ==== Refs References 1. Mann J. Cetacean Societies: Field Studies of Dolphins and Whales University of Chicago Press Chicago, IL, USA 2000 2. Michaud R. Sociality and Ecology of the Odontocetes Sexual Segregation in Vertebrates: Ecology of the Two Sexes Cambridge University Press Cambridge, UK 2005 303 326 3. Möller L.M. Beheregaray L.B. Genetic evidence for sex-biased dispersal in resident bottlenose dolphins (Tursiops aduncus ) Mol. Ecol. 2004 13 1607 1612 10.1111/j.1365-294X.2004.02137.x 15140103 4. Karczmarski L. Würsig B. Gailey G. Larson K.W. Vanderlip C. Spinner dolphins in a remote Hawaiian atoll: Social grouping and population structure Behav. Ecol. 2005 16 675 685 10.1093/beheco/ari028 5. Baird R.W. Whitehead H. Social organization of mammal-eating killer whales: Group stability and dispersal patterns Can. J. Zool. 2000 78 2096 2105 10.1139/z00-155 6. Ortega-Ortiz J.G. Engelhaupt D. Winsor M. Mate B.R. Rus Hoelzel A. Kinship of long-term associates in the highly social sperm whale Mol. Ecol. 2012 21 732 744 10.1111/j.1365-294X.2011.05274.x 21951561 7. Mirimin L. Banguera-Hinestroza E. Dillane E. Hoelzel A.R. Cross T.F. Rogan E. Insights into genetic diversity, parentage, and group composition of Atlantic white-sided dolphins (Lagenorhynchus acutus ) off the west of Ireland based on nuclear and mitochondrial genetic markers J. Hered. 2011 102 79 87 10.1093/jhered/esq106 21059883 8. Smith B.D. Reeves R.R. River cetaceans and habitat change: Generalist resilience or specialist vulnerability? J. Mar. Biol. 2012 2012 1 11 10.1155/2012/718935 9. Gao A. Zhou K. Geographical variation of external measurements and three subspecies of Neophocaena phocaenoides in Chinese waters Acta Theriol. Sin. 1995 15 81 92 10. Mei Z. Huang S.L. Hao Y. Turvey S.T. Gong W. Wang D. Accelerating population decline of Yangtze finless porpoise (Neophocaena asiaeorientalis asiaeorientalis ) Biol. Conserv. 2012 153 192 200 10.1016/j.biocon.2012.04.029 11. Wang D. Turvey S.T. Zhao X. Mei Z. Neophocaena asiaeorientalis ssp. asiaeorientalis IUCN Red List of Threatened Species Version 2013.1 Available online: http://www.iucnredlist.org/ (accessed on 2 September 2013) 12. Taylor A. Horsup A. Johnson C. Sunnucks P. Sherwin B. Relatedness structure detected by microsatellite analysis and attempted pedigree reconstruction in an endangered marsupial, the northern hairy-nosed wombat Lasiorhinus krefftii Mol. Ecol. 1997 6 9 19 10.1046/j.1365-294X.1997.00146.x 9004516 13. Onorato D.P. Hellgren E.C. Van Den Bussche R.A. Skiles J. Raymond J. Paternity and relatedness of American black bears recolonizing a desert montane island Can. J. Zool. 2004 82 1201 1210 10.1139/z04-097 14. Itoh T. Sato Y. Kobayashi K. Mano T. Iwata R. Effective dispersal of brown bears (Ursus arctos ) in eastern Hokkaido, inferred from analyses of mitochondrial DNA and microsatellites Mamm. Study 2012 37 29 41 10.3106/041.037.0104 15. Zhang X. Liu R. Zhao Q. Zhang G. Wei Z. Wang X. Yang J. The population of finless porpoise in the middle and lower reaches of Yangtze River Acta Theriol. Sin. 1993 13 260 270 16. Wei Z. Wang D. Zhang X. Zhao Q. Wang K. Kuang X.A. Population size, behavior, movement pattern and protection of Yangtze finless porpoise at Balijiang section of the Yangtze River Resour. Environ. Yangtze Basin 2002 11 427 432 17. Yang J. Chen P. Movement and behavior of finless porpoise (Neophocaena phocaenoides ) at Swan Oxbow, Hubei province Acta Hydrobiol. Sin. 1996 20 32 40 18. Jiang W. Observation on the group of the ChangJiang finless porpoise conserved in semi-nature conditions J. Anhui Univ. (Nat. Sci.) 2000 24 106 111 19. Wei Z. Wang D. Zhang X. Aggregation and spatio-temporal distribution of the Yangtze finless porpoise Neophocaena phocaenoides asiaeorientalis in Tian-E-Zhou National Baiji Reserve Acta Hydrobiol. Sin. 2004 28 247 252 20. Mei Z. Zhang X. Huang S.-L. Turvey S.T. Gong W. Wang D. The Yangtze finless porpoise: On an accelerating path to extinction? Biol. Conserv. 2014 172 117 123 10.1016/j.biocon.2014.02.033 21. Csilléry K. Johnson T. Beraldi D. Clutton-Brock T. Coltman D. Hansson B. Spong G. Pemberton J.M. Performance of marker-based relatedness estimators in natural populations of outbred vertebrates Genetics 2006 173 2091 2101 10.1534/genetics.106.057331 16783017 22. Blouin M. Parsons M. Lacaille V. Lotz S. Use of microsatellite loci to classify individuals by relatedness Mol. Ecol. 1996 5 393 401 10.1111/j.1365-294X.1996.tb00329.x 8688959 23. Isberg S. Chen Y. Barker S. Moran C. Analysis of microsatellites and parentage testing in saltwater crocodiles J. Hered. 2004 95 445 449 10.1093/jhered/esh067 15388772 24. Haanes H. Rosef O. Veiberg V. Roed K.H. Microsatellites with variation and heredity applicable to genetic studies of Norwegian red deer (Cervus elaphus atlanticus ) Anim. Genet. 2005 36 454 455 10.1111/j.1365-2052.2005.01343.x 16167997 25. Cronin M. Shideler R. Hechtel J. Strobeck C. Paetkau D. Genetic relationships of grizzly bears (Ursus arctos ) in the Prudhoe Bay region of Alaska: Inference from microsatellite DNA, mitochondrial DNA, and field observations J. Hered. 1999 90 622 628 10.1093/jhered/90.6.622 10589512 26. Cronin M.A. Shideler R. Waits L. Nelson R.J. Genetic variation and relatedness in grizzly bears in the Prudhoe Bay region and adjacent areas in northern Alaska Ursus 2005 16 70 84 10.2192/1537-6176(2005)016[0070:GVARIG]2.0.CO;2 27. Mirimin L. Andrew W. Emer R. Patricia R. Andrew R. Jame C. Tom C. Population structure of short-beaked common dolphins (Delphinus delphis ) in the North Atlantic Ocean as revealed by mitochondrial and nuclear genetic markers Mar. Biol. 2009 156 821 834 10.1007/s00227-009-1147-8 28. Chen L. Bruford M.W. Xu S. Zhou K. Yang G. Microsatellite variation and significant population genetic structure of endangered finless porpoises (Neophocaena phocaenoides ) in Chinese coastal waters and the Yangtze River Mar. Biol. 2010 157 1453 1462 10.1007/s00227-010-1420-x 29. Chen M. Zheng J. Gong C. Zhao Q. Wang D. Inbreeding evaluation on the ex situ conserved Yangtze finless porpoise population in Tian’ezhou national natural reserve Chin. J. Zool. 2014 49 305 316 30. Fullard K. Early G. Heide-Jørgensen M. Bloch D. Rosing-Asvid A. Amos W. Population structure of long-finned pilot whales in the North Atlantic: A correlation with sea surface temperature? Mol. Ecol. 2000 9 949 958 10.1046/j.1365-294x.2000.00957.x 10886657 31. Palsbøll P.J. Heide-Jørgensen M.P. Berubé M. Analysis of Mitochondrial Control Region Nucleotide Sequences from Baffin Bay Belugas (Delphinapterus leucas ): Detecting Pods or Sub-Populations? Belugas in the North Atlantic and Russian Arctic Tromsø: North Atlantic Marine Mammal Commission Heide-Jørgensen M.P. Wiig Ø. NAMMCO Scientific Publications Nuuk, Greenland, Denmark 2002 39 50 32. Whitehead H. Sperm Whales: Social Evolution in the Ocean Chicago University Press Chicago, IL, USA 2003 33. Bigg M.A. Olesiuk P.F. Ellis G.M. Ford J.K.B. Balcomb K.C. Social organization and genealogy of resident killer whales (Orcinus orca ) in the coastal waters of British Columbia and Washington State Rep. Int. Whal. Commun. 1990 12 383 405 34. Gowans S. Würsig B. Karczmarski L. The social structure and strategies of delphinids: Predictions based on an ecological framework Adv. Mar. Biol. 2008 53 195 294 17936137 35. Möller L.M. Beheregaray L.B. Coastal bottlenose dolphins from southeastern Australia are Tursiops aduncus according to sequences of the mitochondrial DNA control region Mar. Mamm. Sci. 2011 17 249 263 10.1111/j.1748-7692.2001.tb01269.x 36. Zheng J.S. Liao X.L. Tong J.G. Du H.J. Milinkovitch M.C. Wang D. Development and characterization of polymorphic microsatellite loci in the endangered Yangtze finless porpoise (Neophocaena phocaenoides asiaeorientalis ) Conserv. Genet. 2008 9 1007 1009 10.1007/s10592-007-9435-7 37. Feng J. Zheng J. Zhou Z. Lin G. Wang D. Zheng B. Jiang W. Paternity determination of captivity-bred Yangtze finless porpoises Neophocaena phocaenoides asiaeorientalis by microsatellite genotyping Prog. Mod. Biomed. 2009 9 4015 4020 38. Zhou Z. Zheng J.S. Chen M.M. Zhao Q.Z. Wang D. Genetic Evaluation and Development Prognosis on Ex Situ Conserved Yangtze Finless Porpoise Living in Tian-e-Zhou National Natural Reserve Acta Hydrobiol. Sin. 2012 36 403 411 39. Rosel P. France S. Wang J. Genetic structure of harbour porpoise Phocoena phocoena populations in the northwest Atlantic based on mitochondrial and nuclear markers Mol. Ecol. 1999 8 S41 S54 10.1046/j.1365-294X.1999.00758.x 10703550 40. Chen L. Bruford M. Yang G. Isolation and characterization of microsatellite loci in the finless porpoise (Neophocaena phocaenoides ) Mol. Ecol. Notes 2007 7 1129 1131 10.1111/j.1471-8286.2007.01803.x 41. Chen L. Yang G. Development of tetranucleotide microsatellite loci for the finless porpoise (Neophocaena phocaenoides ) Conserv. Genet. 2008 9 1033 1035 10.1007/s10592-007-9443-7 42. Van Oosterhout C. Hutchinson W.F. Wills D.P. Shipley P. Micro-checker: Software for identifying and correcting genotyping errors in microsatellite data Mol. Ecol. Notes 2004 4 535 538 10.1111/j.1471-8286.2004.00684.x 43. Zheng J.S. Xia J.H. He S.P. Wang D. Population genetic structure of the Yangtze finless porpoise (Neophocaena phocaenoides asiaeorientalis ): Implications for management and conservation Biochem. Genet. 2005 43 307 320 10.1007/s10528-005-5222-7 16144307 44. Chen M. Zheng J. Wu M. Ruan R. Zhao Q. Wang D. Genetic diversity and population structure of the critically endangered Yangtze finless porpoise (Neophocaena asiaeorientalis asiaeorientalis ) as revealed by mitochondrial and microsatellite DNA Int. J. Mol. Sci. 2014 15 11307 11323 10.3390/ijms150711307 24968271 45. Botstein D. White R.L. Skolnick M. Davis R.W. Construction of a genetic linkage map in man using restriction fragment length polymorphisms Am. J. Hum. Genet. 1980 32 314 6247908 46. Hearne C.M. Ghosh S. Todd J.A. Microsatellites for linkage analysis of genetic traits Trends Genet. 1992 8 288 294 10.1016/0168-9525(92)90256-4 1509520 47. Marshall T. Slate J. Kruuk L. Pemberton J. Statistical confidence for likelihood-based paternity inference in natural populations Mol. Ecol. 1998 7 639 655 10.1046/j.1365-294x.1998.00374.x 9633105 48. Kalinowski S.T. Taper M.L. Marshall T.C. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment Mol. Ecol. 2007 16 1099 1106 10.1111/j.1365-294X.2007.03089.x 17305863 49. Jamieson A. Taylor C. Comparisons of three probability formulae for parentage exclusion Anim. Genet. 1997 28 397 400 10.1111/j.1365-2052.1997.00186.x 9616104 50. Rousset F. GENEPOP’007: A complete re-implementation of the GENEPOP software for Windows and Linux Mol. Ecol. Resour. 2008 8 103 106 10.1111/j.1471-8286.2007.01931.x 21585727 51. Goudet J. FSTAT 2.9.3.2, a Program to Estimate and Test Gene Diversities and Fixation Indices Available online: http://www2.unil.ch/popgen/softwares/fstat.htm (accessed on 23 February 2002) 52. Zhang X. Studies on the age determination, growth and reproduction of finless porpoise Neophocaena phocaenoides Acta Hydrobiol. Sin. 1992 16 289 298 53. Wu H.P. Hao Y.J. Li X. Zhao Q.Z. Chen D.Q. Kuang X.A. Kou Z.B. Feng K.K. Gong W.M. Wang D.B. Mode ultrasonographic evaluation of the testis in relation to serum testosterone concentration in male Yangtze finless porpoise (Neophocaena phocaenoides asiaeorientalis ) during the breeding season Theriogenology 2010 73 383 391 10.1016/j.theriogenology.2009.09.022 19945744 54. Queller D.C. Goodnight K.F. Estimating relatedness using genetic markers Evolution 1989 43 258 275 10.2307/2409206 55. Wang J. Triadic IBD coefficients and applications to estimating pairwise relatedness Genet. Res. 2007 89 135 153 10.1017/S0016672307008798 17894908

PMC005xxxxxx/PMC5000667.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081269ijms-17-01269ArticlePunicalagin Mollifies Lead Acetate-Induced Oxidative Imbalance in Male Reproductive System Rao Faiza 12Zhai Yiwen 12Sun Fei 3*Brennan Charles Academic Editor1 Institute of Immunology and CAS (Chinese Academy of Sciences) Key Laboratory of Innate Immunity and Chronic Disease, Innovation Center for Cell Biology, School of Life Sciences and Medical Center, University of Science and Technology of China, Hefei 230027, Anhui, China; [email protected] (F.R.); [email protected] (Y.Z.)2 Hefei National Laboratory for Physical Sciences at Microscale, Hefei 230027, Anhui, China3 International Peace Maternity & Child Health Hospital. School of Medicine, Shanghai Jiaotong University, Shanghai 200025, China* Correspondence: [email protected]; Tel./Fax: +86-21-6431-182111 8 2016 8 2016 17 8 126920 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Punicalagin (PU) is a known antioxidant. The present study examined PU to protect against lead-induced oxidative stress (OS) testicular damage in mice. Significant increase in lipid peroxidation (LPO) after intraperitoneal injection of lead acetate (LA) indicated enormous generation of reactive oxygen species (ROS). Lead-induced OS has a direct effect on the differentiation of spermatogonial cells, showing a significant decline in sperm count. Supplementation of PU significantly changes values of LPO and glutathione (GSH) with a concomitant increase in sperm count, a marked decrease in the abnormal sperms, and a decline in the morphologically abnormal sperm population. Moreover, the histopathological evaluation of testes and epididymides showed severe changes in mice treated with LA. PU significantly induced nuclear factor erythroid-2 related factor 2-like 2 (Nrf2) expression and phase II enzymes, and data suggest that PU may inhibit OS through Nrf2 activation. The fertility test proved that PU might play an important role in male infertility treatment, especially in the type of infertility induced by OS. fertilityoxidative stresspunicalaginlead acetatetesticular damage ==== Body 1. Introduction Lead compound toxicity has been studied in different human diseases. It is used for various industry and household products [1,2]. It has been previously studied that, compared to other body organs, testes are more sensitive to oxidative stress (OS) and are less defendable against OS due to the presence of high polyunsaturated fatty acids. Accumulated evidence has revealed that testicular physiology, which is basically characterized by the spermatogenesis process, gets disrupted by reactive oxygen-dependent mechanisms [3]. Lead acetate (LA) elicits toxic pathological changes in the testis, leading to atrophy of the organ [4,5,6,7]. Seminal cytology of lead-intoxicated animals normally depicts asthenospermia, hypospermia, teratospermia and remarkable changes in sperm count [8]. Studies in human and animals previously proved that sperm motility, DNA damage and sperm shape are affected by lead [9,10]. Previously, it was known that OS is induced by LA. Every organ has a different ability to bear oxidative stress. The testes are receptive to OS, so lead-induced OS is obvious in the testes. However, it has been shown that because the testis in particular readily succumbs to OS [11], reactive oxygen species (ROS) gather in the reproductive system, and with the triggering of lipid peroxidation (LPO), OS is induced [12,13]. It is well documented that ROS generation results in the activation of LPO. Many scientists have used different antioxidants, including vitamin C [14], vitamin E [15], curcumin [16], pomegranate juice (PJ) [17], and many others during their investigations. The beneficial effects of antioxidants on heavy metal-induced toxicity or OS have been documented previously [18]. Previous studies revealed that PJ contains certain constituents that appear to have beneficial therapeutic properties, including antioxidative effects [19,20]. The extracts contain a number of polyphenols, including anthocyanins, minor flavonoids and punicalagin (PU), which is the most important member of the ellagitannins family. PU is the largest polyphenol among the pomegranate ellagitannins and it is responsible for most of the antioxidant activity of the PJ [21]. PU as a strong antioxidant can also play a vital role in clinical and experimental studies, as other polyphenols have demonstrated their role [22]. In our previous study, we used PU of 98% purity against lipopolysaccharide (LPS)-induced OS testicular damage [23]. However, here LA is used to cause high damage in the testis and we want to evaluate the PU defensive capacity against high oxidative stress-induced damage. A previous study mentioned that pomegranate extracts improve semen quality in infertile men [24]. The biochemical and histopathological effects of daily PU consumption on the OS parameters and sperm concentrations of mice were subsequently evaluated. 2. Results 2.1. Evaluation of Testicular Oxidative Damage and Antioxidant Enzymes A significant decrease in the mean testicular weight was observed in the LA-treated group in comparison with the control group and PU + LA groups (Table 1). LA showed a significant increase in the LPO concentration as compared to the control while PU and LA co-administration showed a marked decrease in LPO (Table 1). PU and LA co-administration resulted in an increase in glutathione (GSH) level as compared to the LA group (Table 1). The 8-hydroxy-2′-deoxyguanosine (8-OHdG) values in the PU-treated group and control group were not different from each other. A marked increase of 8-OHdG in the LA group was observed as compared to the PU and LA co-administered group (Table 1). 2.2. Testicular Histology In the control and PU groups, the testes contained a well-organized seminiferous epithelium, along with the presence of mature spermatids at the luminal edge, demonstrating the normal process of spermatogenesis (Figure 1A,B). The epididymis of the control and PU groups showed a maximum number of sperms (Figure 2A,B). Some seminiferous tubules showed disorganization of their lining epithelium (Figure 1D,E). Most of the seminiferous tubules of the testes in the LA-treated group showed a complete absence of primary spermatocytes, secondary spermatocytes, spermatids and spermatozoa and a loss of the spermatogenesis process in comparison with the normal structure of the seminiferous tubules in the control mice. The epididymis of the LA group showed a marked reduction of sperms due to its toxicity (Figure 2C). The co-administration of PU and LA resulted in a normal structure of the seminiferous tubules of the testes and normal spermatogenesis (Figure 1C). The epididymis also appeared to have a maximum number of sperms which proved that PU could be used against OS–induced damage to testes and sperms (Figure 2D). 2.3. Nuclear Factor Erythroid-2 Factor 2 (Nrf2) Activation and Related Gene Expression The administration of PU in mice was able to induce Nrf2 activation; it was evidenced by the increased expression of Nrf2 target genes heme oxygenase-1 (HO-1) and glutamyl-cysteine ligase (GCL) (Figure 3A). The Western blot results for Nrf2 and HO-1 in the PU + LA group proved that PU helps in boosting the Nrf2 pathway, which helps to fight against OS damage in the testes (Figure 3B). 2.4. Sperm Assessment The sperm count in the LA group was noted to be significantly lower, while those of PU and the PU + LA groups were not significantly different from that of the control. The spermatozoa of the PU + LA group were found to be of significantly better morphology; those of the LA group were of significantly poor morphology (Figure 4) while those of the PU group were not significantly different from those of the control (Table 2). 2.5. Fertility Appraisal The fertility test ended with a remarkable difference in the number of pups of the LA group as compared to the control and PU + LA co-administered groups (Table 3). LA proved its toxicity on the reproductive system. It had an impact on the sperm count and morphology which resulted in a smaller number of pups. The number of LA group pups was also reduced as compared to the control and other groups. Six females were not pregnant at all; these were also not found with the plug, so we concluded that LA-induced OS may also affect sexual behavior due to which, within the exposure time, the males did not engage sexually. 3. Discussion The goal of the present study was to investigate the effect of PU (98% purity) in LA-induced testicular toxicity in mice. Previous studies describe the effect of heavy metals on male fertility [25]. The effect of LA on male reproduction has been documented well in various experimental species. Induced modification in sperm morphology, motility, and count, as well as biochemical disruptions of enzymes with the ROS mechanism are highly affected by this heavy metal exposure [26]. Environmental pollutants such as lead can threaten living creatures in different ways. Due to the increase of lead exposure in the environment, its toxic effect on different organs and their systems has been studied [27]. A direct or indirect effect of a low dose of lead on sexual development and reproduction is documented [28,29]. The study authenticated that the group which received LA (dose of 100 mg/kg) had significantly reduced levels of GSH in the testicular tissue homogenate when compared with control animals, and it had significantly increased LPO contents when compared with control group. These results correlate well with previous investigations [26,30]. A significant increase in levels of LPO activity in the testicular tissue homogenate when compared with the control group verifies the previous studies [31,32,33]. Lead toxicity leads to the degeneration of ROS, including hydroperoxide, singlet oxygen, and hydrogen peroxide, and this leads to the direct depletion of antioxidant reserves as lead has also been shown to suppress blood levels of the antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) [34,35]. High doses of LA caused a spermicidal effect, resulting in sperm count reduction, which supports the previous studies in which the testicular sperm count is an important indicator of the adverse effect of lead on spermatogenesis, and this can be accounted for by the direct influence of lead on testicular tissues. Sperm abnormality increases with the administration of high doses of lead. It has been reported that OS-induced toxicity disrupts spermatogenesis [36]. Pomegranate’s effect on sperm production in male mice treated with LA has been proved in previous studies. However, we used 98% pure PU and proved that PU alone could have great beneficial effects on sperm production. In the current study, the administration of PU (9 mg/kg) for four weeks with the co-administration of LA showed a significant increase in the levels of GSH and also a significant decrease in the LPO levels when compared with the LA-treated group. The antioxidant enzymes SOD and CAT also decreased in the LA group. Here it can be concluded that lead implicates testicular LPO due to free radical induction. PU decreased LPO and GSH increased due to its powerful antioxidant property. It has previously been documented that nuclear factor Nrf2 is bound to Kelch-like ECH-associated protein-1 (Keap1) in the cytoplasm under normal conditions. Under OS or other potentially damaging stimuli, Nrf2 is released from Keap1 and translocates to the nucleus, where it binds to antioxidant response element (ARE) sequences [37]. Nrf2’s protective potential role against different types of toxicity has also been studied [38]. In this study we tested whether Nrf2 was involved in the damage protection of PU in LA-induced OS testicular damage. We found that Nrf2 increased significantly in the testicular tissue upon PU treatment. Other studies showed some polyphenols also activate Nrf2 [19]. Since Nrf2 is a very important endogenous antioxidant, it is possible that increasing Nrf2 activity might mediate PU-induced attenuation of OS in the testes. To prove our findings, we organized a fertility test. This fertility test proved that maximum damage to sperm morphology resulted in a smaller number of pups with weight difference as compared to controls. PU and LA co-administration significantly increased the sperm count and improved sperm morphology which led to an increase in the number of pups and normal weight. The reason for the smaller number of pups in the LA group could be acrosome damage, which renders the sperm unable to fertilize the egg. The fertility test in this study can conclude that lead exposure decreased libido in male mice. Nonetheless, additional studies are recommended on this subject before clinical application can be endorsed. 4. Materials and Methods 4.1. Study Design Thirty-two (weighing between 25–28 g) adult male ICR mice (Mus musculus) were used for this study. Mice were obtained from the animal house of University of Science & Technology of China (USTC), Hefei, China. Mice were kept in plastic cages; four mice in each cage. Soft crushed wood shaving was used to cover the floor of all cages. The cages were kept clean, and mice were allowed an ad libitum approach to food and water. Mice were managed under standard laboratory conditions (22–24 °C, 12 h light/12 h dark cycle). All animal-based examinations were designed and performed with the recommendations in the Guide for the Care and Use of Laboratory Animals of National Institutes of Health and received approval from institutional review boards of the University of Science and Technology of China (USTC). Mice were randomly divided into four groups (eight ICR mice in each group) as follows: Control group mice were provided with water and fed with normal diet. PU group animals received PU (9 mg/kg/day) orally by gavage daily for four weeks. LA group mice were injected with LA (100 mg/kg intraperitoneally) for four weeks. PU + LA co-administered animals received PU (9 mg/kg) orally by gavage concurrently injected with LA (100 mg/kg intraperitoneally). All doses were prepared in water. Male treated mice were exposed to normal female mice on the last day of treatment. 4.2. Histopathological Examination of Testes Testes were weighed with a digital balance. Small pieces of testes fixed in Bouin’s solution and 70% ethanol, were taken and dehydrated in graded ethanol, embedded in paraffin, and sectioned (5 μm thickness). Hematoxylin-eosin was used to stain the sections and they were observed under light microscope. Epididymides of mice from all groups were collected and fixed in Bouin’s solution for histological study. 4.3. LPO, GSH and Endogenous Antioxidants Evaluation LPO in the testes homogenate was measured using a LPO Assay Kit (NJJCBio Nanjing, China). The testes were homogenized on ice using 0.2 g tissue in 1.8 mL physiological saline, centrifuged at 2500 rpm for 10 min (ThermoFisher Scientific Heraeus Pico 17, Osterode, Germany). The supernatant was collected. Then 200 µL was used in 96 well plate to detect optical density (OD) at 586 nm (U-3900/3900H HITACHI, Eppendorf AG, Hamburg, Germany). GSH in the testes homogenate was measured using a GSH Assay Kit (NJJCBio). The testes were homogenized on ice in 9 mL of normal saline per gram tissue, centrifuged at 2500 rpm (ThermoFisher Scientific Heraeus Pico 17, Osterode, Germany) for 10 min at 4 °C. The supernatant was used for the assay, according to the instructions in the kit and absorbance of each sample at 420 nm in a glass cuvette was measured. 4.4. The 8-OH-dG Assay DNA oxidative damage enzyme-linked immunosorbent assay (ELISA) kit (Cayman Chemical, Ann Arbor, MI, USA) was used to measure damage in all groups. The experiment was carried out according to the manual provided by the company. Shortly, testes samples were thawed and 5 mL of homogenization buffer per gram of tissue was added. They were centrifuged at 1000× g for 10 min and the supernatant was purified using a commercially available DNA extraction kit (Sangon, Shanghai, China). DNA was digested following the manufacturer’s instructions. One unit of alkaline phosphatase was added per 100 µg of DNA and incubated at 37 °C for 30 min. It was then boiled for 10 min. Reagents were added in 9-well plates along with DNA samples and incubated for 18 h at 4 °C. After adding other reagents like EIA buffer, DNA Oxidative Damage EIA standard, samples, DNA Oxidative Damage AChE Tracer and DNA Oxidative Damage EIA Monoclonal Antibody, the plate was read at a wavelength between 405–420 nm. 4.5. Western Blot For Western blot the samples were lysed with Western lysis buffer (Radioimmunoprecipitation assay (RIPA) buffer, inhibitor cocktail and phenylmethylsulfonyl fluoride (PMSF). They were homogenized and centrifuged at 14,000 rpm (ThermoFisher Scientific Heraeus Pico 17, Osterode, Germany) for 15 min at 4 °C. Supernatants were collected and loading buffer was added to them. Protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose, membranes. Then 5% nonfat milk in Tris-buffered saline-Tween (TBST) buffer was used to block them. The membranes were incubated with primary antibodies (Sangon Biotech, Shanghai, China) against Nrf2 and HO-1, later they were incubated with secondary antibody (anti-rabbit) (Promega, Beijing, China.) The membranes were visualized using enhanced chemiluminescence (ECL) (Kodak, Shanghai, China). Levels of protein were normalized to β-actin 4.6. Blood Lead Analysis Acid digestion method was used for the preparation of blood sample, i.e., the blood sample was digested in nitric acid:hypochlorite (6:1) mixture and the lead level was estimated using a spectrophotometer (U-3900/3900H HITACHI, Eppendorf AG, Hamburg, Germany) at 620–650 nm wavelength. Statistical significance was evaluated by calculating standard deviation followed by Student’s t-test [39]. 4.7. Reverse Transcription and Real-Time PCR The expression of Nrf2, HO-1, GCL and GAPDH (Glyceraldehyde-3-phosphate dehydrogenase) were detected by reverse transcript and real-time PCR. Total RNA was extracted from the tissues using Trizol total RNA isolation reagent (Sangon Shanghai, China) as detailed by instructions from manufacturers. PCR (SYBR Premix Ex Taq, ROX Reference Dye, primers and cDNA template) (TaKaRa Biotechnology Dalian, China) were run in triplicates in 20 μL total reaction volume. The amplification conditions were as follows: 95 °C for 5 min, 39 cycles of 30 s at 95 °C, 60 °C for 30 s, 72 °C for 30 s and ended with extension temperature of 72 °C for 10 min. The fold changes were calculated on the basis of relative quantification method. The primers were designed as Nrf2 Fw 5′-CAGTGCTCCTATGCGTGAA-3′; Rv 5′-GCGGCTTGAATGTTTGTC-3′: HO-1 Fw 5′-ACAGATGGCGTCACTTCG-3′; Rv 5′-TGAGGACCCACTGGAGGA-3′: GCL Fw 5′-GGATGATGCCAACGAGTC-3′; Rv 5′-GTGAGCAGTACCACGAATA-3′: GAPDH Fw 5′-TCTGACGTGCCGCCTGGAGA-3′; Rv 5′-GGGGTGGGTGGTCCAGGGTT-3′. 4.8. Sperm Quality Evaluation For normal testicular function, sperm count and morphology were considered as markers. Cauda epididymis was placed in Dulbecco’s modified Eagle’s medium, nutrient mixture F12 (DMEM/F12) with fetal bovine serum (FBS) 5% and cut into small pieces. The medium was balanced in incubator at temperature 37 °C and 5% CO2. Sperms from all four groups of mice were collected from their cauda epididymis and sperm number was determined by hemocytometer (Abcam Cambridge, MA, USA). Sperm morphology was also studied following a similar method with addition of eosin staining for a few minutes, then making a smear on a slide, air drying and mounting the slides for morphology study under microscope following the method of Majumder et al. [22] and WHO laboratory manual [40]. 4.9. Fertility Performance Adult ICR male mice were administered with LA and PU for four weeks, after which each of them was caged with untreated female ICR mice and provided with standard food. Males were separated from females after monitoring vaginal plugs within one week and if no plug was found still they were separated. Fertility performance of the four groups was analyzed on the basis of the time period when the females gave birth Pregnant female mice were inspected twice daily to appraise birth of offspring. 4.10. Statistical Analysis Data represent mean ± standard error of the mean (SEM) of at least three independent experiments. One-way analysis of variance (ANOVA) was used for statistical comparison of groups and followed by Tukey-Kramer for a multiple comparisons test. Acknowledgments This work was supported by the National Natural Science Foundation of China Grants 81430027 (to F.S.), the National Basic Research Program of China (2014CB943100) (to F.S.), Shanghai Municipal Education Commission-Gaofeng Clinical Medicine Grant Support (No. 20152531) (to F.S.). Author Contributions Faiza Rao and Fei Sun designed the study. Faiza Rao and Yiwen Zhai carried out experiments. Faiza Rao, Fei Sun and Yiwen Zhai analyzed experimental data, drafted the article and revised the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Testes histology. (A,B) Regular-shaped and normal spermatogenesis observed in control and punicalagin (PU)-injected mice testes; (C) Co-injection of PU and lead acetate (LA) protected testes against oxidative injury; normal seminiferous tubules with presence of sperm and less degenerative germ cells could be observed; (D) An amount of 100 mg/kg of LA administered intraperitoneally in mice; a single dose causes oxidative stress–induced damage in mice testes. The figure shows degeneration of germ cells, absence of sperm and vacuolation formation in testicular tissue; (E) Degeneration of seminiferous epithelium, vacuolization (arrow in figure), disordered epithelium and loss of sperms were evidenced in LA-treated group. Figure 2 Epididymis histology. (A,B) Control and PU groups show maximum number of sperms in epididymis; (C) LA-induced OS proved damage to sperms as the epididymis shows reduction of sperm count; (D) PU + LA co-administration presented increase in number of sperms in epididymis. Figure 3 Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) activation. (A) Real time reverse transcript-PCR was performed to detect expression of Nrf2 and its target genes heme oxygenase-1 (HO-1) and glutamyl-cysteine ligase (GCL); (B) Western blot analysis results show Nrf2 and HO-1 increase in LA + PU. (*, #, † p < 0.0001 values compared for two groups). Figure 4 Sperm morphology. (A,E) Normal sperm, pointed out by arrow showing acrosome formation as dark black color on sperm head; (B) abnormal sperm with bent head; (C) abnormal sperm with twisted head and neck; (D) abnormal sperm morphology with head damage and coiled tail; (F) arrow head showing abnormal formation of acrosome. ijms-17-01269-t001_Table 1Table 1 Effect of punicalagin (PU) on lipid peroxidation (LPO), glutathione GSH, 8-hydroxy-2′-deoxyguanosine (8-OHdG), superoxide dismutase (SOD) and catalase (CAT) in lead acetate (LA)- intoxicated mice testis (mean ± standard deviation (SD)). LPO (µmol/g Prot) GSH (mg/g Prot) 8-OHdG (ng/mL Tissue Homogenate) SOD (U/mg Prot) CAT (nmol/mg Prot) Control 1.93 ± 0.10 26.38 ± 0.90 1.16 ± 0.17 49.83 ± 0.98 72.91 ± 0.49 PU 1.95 ± 0.09 26.23 ± 0.96 1.18 ± 0.17 49.97 ± 1.17 72.97 ± 0.48 LA 10.45 ± 0.48 12.12 ± 1.38 6.70 ± 0.36 34.79 ± 0.52 52.83 ± 0.64 PU + LA 3.85 ± 0.35 22.14 ± 1.36 2.39 ± 0.32 47.92 ± 0.74 67.90 ± 0.49 ijms-17-01269-t002_Table 2Table 2 LA effect on testes weight grams (g), sperm abnormality (per 1000 sperm), blood lead level (µg/mL), sperm count (×106). All results shown as mean ± standard error (SE). Group 1 (Control) Group 2 (PU) Group 3 (LA) Group 4 (LA+ PU) Blood lead level 0.17 ± 0.014 0.168 ± 0.01 0.426 ± 0.035 * 0.20 ± 0.022 Abnormal sperm 112.5 ± 1.85 112.3 ± 1.40 209 ± 4.4 * 125 ± 4.9 Testes weight 0.22 ± 0.02 0.216 ± 0.01 0.14 ± 0.01 * 0.198 ± 0.01 Sperm count 24 ± 1.03 23 ± 1.1 14.9 ± 0.9 * 20.6 ± 1.03 * p < 0.0001 as compared group 3 with group 1 and group 4. ijms-17-01269-t003_Table 3Table 3 Mating test (mean ± SD). Control PU LA PU + LA No. of females pregnant 2 ± 0 2 ± 0 1.6 ± 1.03 * 2 ± 0 No. of pups 12 ± 0.92 12.2 ± 1.16 5.12 ± 4.45 * 10.25 ± 1.16 * p < 0.0001 as compared between control, LA, PU + LA groups. ==== Refs References 1. Saxena D.K. Hussain T. Lal B. Chandra S.V. Lead induced testicular dysfunction in weaned rats Ind. Health 1986 24 105 109 10.2486/indhealth.24.105 3759508 2. Kumar B.D. Krishnaswamy K. Detection of sub-clinical lead toxicity in monocasters Bull. Environ. Contam. Toxicol. 1995 54 863 869 10.1007/BF00197971 7647502 3. Koizumi T. Li Z.G. Tatsumoto H. DNA damaging activity of cadmium in Leydig cells, a target cell population for cadmium carcinogenesis in the rat testis Toxicol. Lett. 1992 63 211 220 1455453 4. Ghelberg N.W. Emeric B. Lead-induced experimental lesions of the testis and their treatment J. Appl. Toxicol. 2006 1 284 286 10.1002/jat.2550010509 7185889 5. Cullen M.R. Kayne R.D. Robins J.M. Endocrine and reproductive dysfunction in men associated with occupational inorganic lead intoxication Arch. Environ. Health 1984 39 431 440 10.1080/00039896.1984.10545877 6441528 6. Saxena D.K. Lal B. Murthy R.C. Chandra S.V. Lead induced histological changes in testes of rats Ind. Health 1984 22 225 260 10.2486/indhealth.22.255 7. Acharya U.R. Mishra P. Lead induced histological changes in the testes of Swiss mice Adv. Biosci. 1995 14 37 44 8. Bell J.U. Thomas J.A. Effect of Lead on Mammalian Reproduction Lead Toxicity Singhal R.L. Thomas J.A. Uraban and Schwarzenberg Baltimore, MD, USA 1980 169 185 9. Telisman S. Cvitkovic P. Jurasovic J. Pizent A. Gavella M. Semen quality and reproductive endocrine function in relation to biomarkers of lead, cadmium, zinc, and copper in men Environ. Health Perspect. 2000 108 45 53 10.1289/ehp.0010845 10620523 10. Bonde J.P. Joffe M. Apostoli P. Dale A. Kiss P. Sperm count and chromatin structure in men exposed to inorganic lead: Lowest adverse effect levels Occup. Environ. Med. 2002 59 234 242 10.1136/oem.59.4.234 11934950 11. Lisa M. Tomascik M. Matthew A.C. Francesco M. David O.N. Differential basal expression of genes associated with stress response, damage control, and DNA repair among mouse tissues Mutat. Res. Gen. Toxicol. Environ. Mutagen. 2004 561 1 14 12. Marchlewicz M. Michalska T. Wiszniewska B. Detection of lead-induced oxidative stress in the rat epididymis by chemiluminescence Chemosphere 2004 57 1553 1562 15519400 13. Mishra M. Acharya U.R. Protective action of vitamins on the spermatogenesis in lead-treated Swiss mice J. Trace Elem. Med. Biol. 2004 18 173 178 15646264 14. Hsu P. Liu M. Hsu C. Chen L. Yueliang L.G. Effect of vitamin E and/or C on reactive oxygen species-related lead toxicity in the rat sperm Toxicology 1998 128 169 179 9750040 15. Patra R.C. Rautray A.K. Swarup D. Oxidative stress in lead and cadmium toxicity and its amelioration Vet. Med. Int. 2011 2011 457327 21547215 16. Soliman M.M. Baiomy A.A. Yassin M.H. Molecular and Histopathological Study on the Ameliorative Effects of Curcumin Against Lead Acetate-Induced Hepatotoxicity and Nephrototoxicity in Wistar Rats Biol. Trace Elem. Res. 2015 167 91 102 10.1007/s12011-015-0280-0 25758718 17. Leiva K.P. Rubio J. Peralta F. Gonzales G.F. Effect of Punica granatum (pomegranate) on sperm production in male rats treated with lead acetate Toxicol. Mech. Methods 2011 21 495 502 21417630 18. Han S.G. Kim Y. Kashon M.L. Pack D.L. Castranova V. Vallyathan V. Correlates of oxidative stress and free-radical activity in serum from asymptomatic shipyard welders Am. J. Respir. Crit. Care Med. 2005 172 1541 1548 10.1164/rccm.200409-1222OC 16166614 19. Aviram M. Dornfeld L. Kaplan M. Coleman R. Gaitini D. Nitecki S. Hofman A. Rosenblat M. Volkova N. Presser D. Pomegranate juice flavonoids inhibit low-density lipoprotein oxidation and cardiovascular diseases: Studies in atherosclerotic mice and in humans Drugs Exp. Clin. Res. 2002 28 49 62 12224378 20. Atilqan D. Parlaktas B. Uluocak N. Gencten Y. Erdemir F. Ozyurt H. Erkorkmaz U. Aslan H. Pomegranate (Punica granatum ) juice reduces oxidative injury and improves sperm concentration in a rat model of testicular torsion-detorsion Exp. Ther. Med. 2014 8 478 482 25009604 21. Seeram N. Lee R. Hardy D. Heber D. Rapid large scale purification of ellagitannins from pomegranate husk, a by-product of the commercial juice industry Sep. Purif. Technol. 2005 41 49 55 10.1016/j.seppur.2004.04.003 22. Majumder G.C. Biswas R. Evidence for the occurrence of an ecto-(adenosine triphosphatase) in rat epididymal spermatozoa Biochem. J. 1979 183 737 743 10.1042/bj1830737 231971 23. Rao F. Tian H. Li W. Hung H. Sun F. Potential role of punicalagin against oxidative stress induced testicular damage Asian J. Androl. 2016 18 627 632 26763544 24. Fedder M.DK. Jakobsen H.B. Giversen I. Christensen L.P. Parner E.T. Fedder J. An Extract of Pomegranate Fruit and Galangal Rhizome Increases the Numbers of Motile Sperm: A Prospective, Randomised, Controlled, Double-Blinded Trial PLoS ONE 2014 9 1269 25. Roy C. Recent advances in heavy metals induced effect on male reproductive function—A retrospective Al Ameen J. Med. Sci. 2009 2 37 42 26. Patrick L. Lead toxicity, a review of the literature. Part 1: Exposure, evaluation, and treatment Altern. Med. Rev. 2006 11 2 22 16597190 27. Junaid M. Chowdhuri D.K. Narayan R. Shanker R. Saxena D.K. Lead-induced changes in ovarian follicular development and maturation in mice J. Toxicol. Environ. Health 1997 50 31 40 9015130 28. McMurry S.T. Lochmiller R.L. Chandra S.A. Qualls C.W. Sensitivity of selected immunological, hematological, and reproductive parameters in the cotton rat (Sigmodon hispidus ) to subchronic lead exposure J. Wildl. Dis. 1995 31 193 204 10.7589/0090-3558-31.2.193 8583637 29. Mohsen V. Derek R.S. Hsu P. How does lead induce male infertility? Iran J. Reprod. Med. 2011 9 1 8 25356074 30. Ahamed M. Verma S. Kumar A. Siddiqui M.K. Environmental exposure to lead and its correlation with biochemical indices in children Sci. Total Environ. 2005 346 48 55 10.1016/j.scitotenv.2004.12.019 15993681 31. Gurer-Orhan H. Sabir H.U. Ozqunes H. Correlation between clinical indicators of lead poisoning and oxidative stress parameters in controls and lead exposed workers Toxicology 2004 195 147 154 14751670 32. Patra R.C. Swarup D. Dwivedi S.K. Antioxidant effects of α tocopherol, ascorbic acid and l -methionine on lead induced oxidative stress to the liver, kidney and brain in rats Toxicology 2001 162 81 88 10.1016/S0300-483X(01)00345-6 11337108 33. Han J.M. Chang B.J. Li T.Z. Choe N.H. Quan F.S. Jang B.J. Cho I.H. Hong H.N. Lee J.H. Protective effects of ascorbic acid against lead-induced apoptotic neurodegeneration in the developing rat hippocampus in vivo Brain Res. 2007 1185 68 74 10.1016/j.brainres.2007.09.044 17959157 34. Sugawara E. Nakamura K. Miyake T. Fukumura A. Seki Y. Lipid peroxidation and concentration of glutathione in erythrocytes from workers exposed to lead Br. J. Ind. Med. 1991 48 239 242 2025589 35. Aruldhas M.M. Subramanian S. Seker P. Vengatesh G. Chandrahasan G. Govindarajulu P. Akbarasha M.A. Chronic chromium exposure-induced changes in testicular histoarchitecture are associated with oxidative stress: Study in a non-human primate (Macaca radiata Geoffroy) Hum. Reprod. 2005 20 2801 2813 15980013 36. Mann G.E. Niehueser-Saran J. Watson A. Gao L. Ishii T. De Winter T. Siow R.C. Nrf2/ARE regulated antioxidant gene expression in endothelial and smooth muscle cells in oxidative stress: Implications for atherosclerosis and preeclampsia Acta Physiol. Sin. 2007 59 117 127 37. Li J. Shen F. Guan C. Wang W. Sun X. Fu X. Huang M. Jin J. Huang Z. Activation of Nrf2 protects against triptolide-induced hepatotoxicity PLoS ONE 2014 9 1269 38. Wu J. Li Q. Wang X. Yu S. Li L. Wu X. Chen Y. Zhao J. Zhao Y. Neuroprotection by curcumin in ischemic brain injury involves the Akt/Nrf2 pathway PLoS ONE 2013 8 1269 10.1371/journal.pone.0059843 23555802 39. Ping-Chi H. Ming-Yie L. Chao-Chin H. Lih-Yuh C. Yueliang L.G. Lead exposure causes generation of reactive oxygen species and functional impairment in rat sperm Toxicology 1997 122 133 143 9274809 40. WHO Laboratory Manual for the Examination and Processing of Human Semen 5th ed. World Health Organization Press Geneva, Switzerland 2010

PMC005xxxxxx/PMC5000668.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081270ijms-17-01270ReviewUsing Proteomics to Understand How Leishmania Parasites Survive inside the Host and Establish Infection Veras Patrícia Sampaio Tavares 12*Bezerra de Menezes Juliana Perrone 1Sheehan David Academic Editor1 Laboratório de Patologia e Biointervenção, Instituto Gonçalo Moniz, FIOCRUZ, Salvador 40296-710, Brazil; [email protected] Instituto Nacional de Ciência e Tecnologia para Doenças Tropicais (INCT-DT), Salvador 40110-160, Brazil* Correspondence: [email protected]; Tel.: +55-71-3176-2263; Fax: +55-71-3176-229019 8 2016 8 2016 17 8 127001 5 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Leishmania is a protozoan parasite that causes a wide range of different clinical manifestations in mammalian hosts. It is a major public health risk on different continents and represents one of the most important neglected diseases. Due to the high toxicity of the drugs currently used, and in the light of increasing drug resistance, there is a critical need to develop new drugs and vaccines to control Leishmania infection. Over the past few years, proteomics has become an important tool to understand the underlying biology of Leishmania parasites and host interaction. The large-scale study of proteins, both in parasites and within the host in response to infection, can accelerate the discovery of new therapeutic targets. By studying the proteomes of host cells and tissues infected with Leishmania, as well as changes in protein profiles among promastigotes and amastigotes, scientists hope to better understand the biology involved in the parasite survival and the host-parasite interaction. This review demonstrates the feasibility of proteomics as an approach to identify new proteins involved in Leishmania differentiation and intracellular survival. proteomicsLeishmaniaintracellular survival ==== Body 1. Introduction Leishmania is a protozoan parasite that causes a broad range of different clinical symptoms in humans known as leishmaniasis. This disease represents a major public health problem and is endemic in 98 countries across five continents, Asia, Africa, Europe, North America and South America. Over 350 million people are at risk, with an estimated 12 million infected, and 0.9–1.6 million new cases emerging per year. More than 90% of global visceral leishmaniasis (VL) cases occur in six countries: Bangladesh, Brazil, Ethiopia, India, South Sudan and Sudan. In addition, ten countries with the highest estimated case counts for the cutaneous form of the disease are: Afghanistan, Algeria, Brazil, Colombia, Costa Rica, Ethiopia, Iran, Peru, Sudan, and Syria, together accounting for 70% to 75% of the global estimated cutaneous leishmaniasis (CL) incidence [1,2]. Leishmaniasis is caused by different species of protozoan parasites belonging to the genus Leishmania. Different species of Leishmania are responsible for varying clinical forms of leishmaniasis. Human leishmaniasis consists of a range of diseases which can manifest as a simple self-limiting or asymptomatic CL to a disfiguring and debilitating VL, the clinical form of the disease associated with higher mortality. Post-kala-azar dermal leishmaniasis (PKDL) is a dermal complication of VL and is considered a reservoir for Leishmania parasites [3]. Leishmania (L.) major or L. tropica causes localized cutaneous lesions that are usually self-healing [4]. South American species, such as L. braziliensis, manifest initially as cutaneous lesions that may metastasize resulting in mucocutaneous lesions or diffuse CL. On the other hand, infections caused by L. donovani or L. infantum may lead to chronic disseminating diseases, mainly in the liver and spleen, which are often fatal if left untreated [4]. The advances in large-scale technologies, such as proteomics, have allowed the identification and characterization of pathways, both in the parasite and the host, which have proven to be more effective than studying individual molecules. Proteomics is the large-scale characterization of the proteins in a cell line, tissue, or organism, with the goal to access a more global and integrated view of the biological processes by studying all the proteins in a cell rather than each one individually [5]. The use of proteomics tools has revolutionized several biomedical fields such as medicine and dentistry. Proteomics has contributed greatly to the dentistry field by helping in the identification of different biomarkers present in the oral fluids for early diagnosis of several diseases [6]. Also, proteomics has contributed to the understanding and identification of several medically important biomarkers for different diseases [7,8,9,10]. In the last decade, high-throughput techniques, which can process and analyze large amounts of diverse molecules using automated systems, has enabled us to identify molecules involved in the establishment of diseases caused by Leishmania parasites, development of parasite resistance [11,12,13], as well as the characterization of new chemotherapeutic targets [14,15]. The relatively weak correlation between mRNA and protein levels led to the conclusion that it is not possible to predict protein expression based on quantitative mRNA data [16,17]. The above reinforces the idea that proteomics should be considered as a large-scale critical tool to understand the host-Leishmania interactions better. Indeed, proteomic studies have been widely used to characterize molecules and pathways expressed in the parasite, as well as in the invertebrate [18,19,20,21], or mammalian [22] hosts. In the Leishmania research field, proteomic studies have provided valuable insights into the identification of molecules and pathways involved in host-parasite interactions in the parasite [18,19,20,21], and in the host [22,23]. Also, proteomics has significantly contributed to the identification of targets for prophylactic or chemotherapeutic treatment [22,24], as well as biomarkers that can be used for the diagnosis of the different diseases [25] (Figure 1). 2. Leishmania Adaptation to the Intracellular Life-Cycle: Modulation in Parasite Protein Expression 2.1. Modulation of Proteins during Axenic Differentiation of Leishmania Parasites During their life cycle, Leishmania spp. adapt to different environments in the insect and the mammalian host by undergoing a variety of morphological and biochemical changes [30,31,32]. These changes in environment correlate with the process of differentiation from promastigote, the motile form that proliferates inside the alimentary tract of Phlebotomine sandflies, to the amastigote form, the non-motile form that multiplies inside the acidified phagolysosomes of mammalian host macrophages [19,33,34,35,36] (Figure 2). The adaptation of the parasite to the host environment is crucial to the differentiation process. This adaptation includes changes in temperature and pH [32], as well as adjustments to the cytotoxic environment of the host. Furthermore, this adaptation is essential for the intracellular survival of the parasite, which requires a combination of survival factors expressed by parasites at distinct stages [37,38]. Until now, the cellular and molecular mechanisms involved in the differentiation of Leishmania parasites were poorly understood [19]. The proteomic studies published so far have been conducted using either L. donovani or L. infantum, probably due to the medical importance of the visceral forms arising from these parasites, which result in high mortality rate. The first study that compared the differentially expressed proteins of L. infantum parasites in differing life cycle stages employed comparative two-dimensional gel electrophoresis (2-DE) in addition to mass spectrometry. In this report, more than 62 differentially expressed proteins were detected in axenic amastigotes among ~2000 protein spots resolved by 2-DE. Two-dimensional gel electrophoresis has frequently been used as a protein separation method before mass spectrometry. Spots represent one or more proteins that have migrated to a particular location on the gel based on their biochemical properties. Spots of interest can then be subjected to in-gel digestion for further protein identification. Two of such proteins were identified as participating in energetic metabolism pathways, namely isocitrate dehydrogenase (IDH) and the glycolytic enzyme triosephosphate isomerase (TIM). Additionally, the authors demonstrated upregulated activity by these enzymes in amastigotes when compared to promastigote forms [18]. Isocitrate dehydrogenase is an enzyme that participates in the tricarboxylic acid cycle, a metabolic pathway by which acetate is oxidized to generate ATP. NADP-dependent IDH enzymes catalyze the decarboxylation of isocitrate to α-ketoglutarate, which is accompanied by the production of NADPH. This step is critical in the tricarboxylic acid cycle, and the α-ketoglutarate produced by this enzyme activity can contribute to the synthesis of glutamate, a precursor for amino acids. In this study, the authors showed that IDH-specific activity is approximately three times higher in amastigotes than in promastigotes. As IDH catalyzes the formation of α-ketoglutarate with the production of NADPH, its enhanced activity might be essential for meeting the increased demand for α-ketoglutarate at 37 °C. Triosephosphate isomerase, the other protein exhibiting higher expression in amastigotes, is a highly prevalent enzyme that plays a central role in glycolysis [39]. The authors showed that TIM activity in L. infantum amastigotes was two-fold higher compared to L. infantum promastigotes, probably because amastigotes require high levels of TIM activity to generate ATP via glycolysis within host cells [18]. In another study, a total of approximately 2000 protein spots were identified in the L. donovani proteome, 31 of which were exclusively present in promastigotes [19]. They found 65 proteins with increased expression resulting from heat-induced in vitro amastigote differentiation; however, four proteins exhibited decreased expression in amastigote differentiation. Further studies involving matrix-assisted laser desorption/ionization (MALDI)-time of flight (TOF) and peptide mass fingerprinting revealed 67 protein spots representing 41 different proteins previously identified by databases, in addition to eight hypothetical proteins. In this study, the authors showed that most of the stage-specific proteins identified in L. donovani promastigotes or axenic amastigotes can be divided into five groups of proteins with similar function: “(i) stress response (e.g., heat, oxidative stress); (ii) cytoskeleton and cell membrane; (iii) energy metabolism and phosphorylation; (iv) cell cycle and proliferation; and (v) amino acid metabolism” [19]. Although they found interesting data on protein modulation in amastigotes, the authors have yet to validate these data. Another study with the goal to better evaluate the differentiation of Leishmania parasites applied a comprehensive approach consisting of protein prefractionation, followed by global proteomics and targeted DNA microarray analysis. Using 2-D gels, the authors showed that over 2200 protein isoforms were identified in each developmental stage, corresponding to 10% more than what was previously identified by proteomic studies evaluating the in vitro differentiation process of Leishmania parasites. Of these, 6.1% were strongly increased or appeared exclusively in the promastigote stage, while 12.4% appeared in amastigotes. Although modest correlations between amastigote-specific protein isoform and mRNA expression (53%) were observed, these authors found no correlation with respect to promastigote-specific spots. They suggested that post-transcriptional controls at translational and post-translational levels may be critically involved in the Leishmania parasite differentiation process [40]. As shown by different studies [19,40,41], the major class of proteins exclusively identified or overexpressed in amastigotes was of those involved in stress response or protein folding. Metabolic enzymes were also frequently identified with higher levels of expression in axenic amastigotes compared to promastigotes. In addition, proteins involved in the proteolysis process were also modulated in the amastigote forms [19]. In the first proteomic study performed to evaluate in vitro differentiation of L. panamensis, the authors detected 75 differentially regulated protein spots in amastigotes, comprising 24 spots “uniquely” expressed during this life-stage, and 51 that were approximately one to five times overexpressed in comparison to promastigotes [42]. The spots were analyzed by mass spectrometry, and among 11 amastigote-specific spots, six spots were identified as seven distinct proteins. These proteins participate in different cellular processes such as carbohydrate/energy metabolism, stress response, cell membrane and cytoskeleton, amino acid metabolism and cell-cycle. Four additional over-expressed spots were identified as heat shock proteins (HSPs) 60 and 70, and HSP 70-related proteins [42]. Comparative proteomic studies have already shown that proteins involved in stress response and metabolic pathways are differentially expressed among promastigotes and amastigotes from L. donovani [19] and L. infantum [18]. In a more recent report, a different proteomic approach was performed to study the differentiation process of L. infantum. The authors applied protein fractionation by isoelectric point (pI) using free-flow electrophoresis to evaluate the expression of stage-specific proteins in this parasite. They identified 2469 protein spots in both life stages. This fractionation process allowed the identification and characterization of several proteins for the first time by proteomic analysis. Glycolytic enzymes and proteins expressed in the parasite flagellum were identified as upregulated in L. infantum promastigotes. On the other hand, enzymes involved in gluconeogenesis and fatty acid β-oxidation were upregulated in amastigotes [43]. Additionally, the authors also demonstrated that several proteins were identified in multiple spots, or as proteolytic fragments in both life stages, suggesting the occurrence of post-translational modification and processing. The first study using a modern quantitative proteomic approach to investigate the differentiation of Leishmania parasites, the isotope-coded affinity tag (ICAT) technology associated to mass spectrometry, aimed to identify differentially expressed proteins in L. infantum promastigotes and axenic amastigotes. In this report, the authors identified a relatively small number of total and stage-specific proteins. This limited number of proteins was also reported in other recent studies using L. infantum [18,44], L. donovani [19], and L. panamensis [42]. In this work, 8% of the 91 proteins identified were differentially expressed in amastigotes, 20% in promastigotes and 72% were considered constitutively expressed. Proteins with a higher level of expression in amastigotes included two novel proteins and enzymes involved in cell metabolism, as previously shown [45]. One of the branches of proteomics that have become increasingly popular in the last few years is phosphoproteomics. Leishmania parasite differentiation requires the activation of signaling cascades involving protein kinases and their downstream phosphoprotein substrates. These signaling pathways are highly adapted to the specific nutritional and physiological requirements of the cells. Therefore, the study of Leishmania phosphorylated proteins provided important insights into the parasite biology. Based on these findings, the authors sought to use a gel-based approach in a new study to investigate both qualitative and quantitative changes within the phosphoproteome during the L. donovani life cycle stages during in vitro differentiation process [20]. In this pioneering study, phosphoproteins were purified from parasites using immobilized metal affinity chromatography and then separated by 2-DE utilizing fluorescent multiplex staining. The identification of proteins was performed using matrix-assisted laser desorption/ionization-mass spectrometry (MALDI-MS) and mass spectrometry/mass spectrometry MS/MS [20], which identified proteins involved in stress and heat shock response, RNA/protein turnover, metabolism, and signaling. The identification of these proteins reinforces the idea already shown in previous studies that the modulation of proteins involved in stress response and metabolism is critical for Leishmania differentiation. 2.2. Modulation of Proteins during Intracellular Differentiation of Leishmania Parasites Until now, the majority of the studies have used axenic parasites grown under in vitro conditions that mimic the sand fly gut (26 °C, pH 7) and phagolysosome (37 °C, pH 5.5) environments to evaluate protein expression in the amastigote stage of the parasite. Although axenic amastigotes are morphologically similar to intracellular amastigotes [46,47,48,49,50,51], a constant concern has been the degree to which axenic amastigotes resemble intracellular ones. Recently, a group performed a comparative proteomic study that evaluated global protein expression in different life stages of Leishmania, using amastigotes that underwent the differentiation process from promastigotes to amastigotes intracellularly [52]. The authors used transgenic fluorescent L. mexicana parasites that were purified from infected cells combining isopycnic density centrifugation and fluorescent parasite sorting. In this study, a total of 509 different proteins were identified by mass spectrometry, of which 301 were exclusively detected in promastigotes, 31 were only identified in intracellular amastigotes, and 157 were common in both stages. Intracellular amastigotes demonstrated a greater profusion of enzymes involved in the catabolism of fatty acids, which may be the result of this parasite dwelling in acidic compartments, as well as its metabolic adaptation to scarce nutrient availability. These results corroborate those reported for the proteomic analysis of L. donovani in axenic amastigotes [53]. In addition, another study that investigated genes, whose products were expressed with higher levels in amastigotes, showed characteristic sequence motifs in 3′-untranslated regions that have been linked to translational control elements, suggesting that proteome data sets may be used to identify regulatory elements in mRNAs. These data support the notion that post-transcriptional processes are important for gene regulation in Leishmania parasites [11,12,13,14,17,54]. In a more recent report, an isobaric tagging method was used to quantify the differences among the proteome of promastigotes and amastigotes, which underwent differentiation within human monocyte-derived macrophages (THP-1). The proteins identified as differentially expressed between amastigotes and promastigotes are known to be involved in nutrient acquisition and energy metabolism, cell motility and cytoskeleton, transport, cell signaling and stress response. Upon investigating the proteins involved in vesicular trafficking and endocytosis, such as the rab7 GTP binding protein, GTP-binding proteins of the Ras superfamily, and developmentally regulated GTP-binding protein 1, the authors found enhanced expression in addition to a putative dynein heavy chain protein that was upregulated in amastigotes, which likely plays a role in cargo transport within vesicles. Furthermore, a protein involved in glucose transport exhibited significantly increased expression (8× to 15× higher) in intracellular amastigotes, while several proteins associated with cell motility and cytoskeleton had reduced levels [21]. Taken together, the studies published so far indicate a modulation of the parasite metabolism and molecules involved in stress response after the differentiation of promastigotes to amastigotes, probably favoring the intracellular survival of Leishmania. 3. Protein Expression by Macrophages in Response to Leishmania Infection in Vitro Only very few studies used proteomics to identify proteins expressed by the host cell in response to Leishmania. Tandem liquid chromatography-mass spectrometry (LC-MS/MS) was used to identify markers of resistance and susceptibility in macrophages during Leishmania infection in vitro. The CBA mouse macrophages have proven to be useful in identifying these markers because at early time points of infection, the cells present similar percentages of L. major- and L. amazonensis-infected macrophages. At later time points, a greater proportion of macrophages from the same strain became infected with L. amazonensis in comparison to L. major [22,55]. A total of 1352 proteins were found expressed in both infected and uninfected CBA macrophages, and only 62 proteins were predominantly expressed in infected macrophages. These proteins were previously described as involved in cell metabolism, or as carrier proteins, in addition to others that participate in cell signaling and cellular detoxification. Another group of proteins contributes to cell immune response, including immune receptors, scavenger receptor class B, and TNF receptor-associated protein. Interesting, only 10 out of the 62 proteins were exclusively identified in L. major infection: ribosomal protein S13; glutamate receptor ionotropic; guanine nucleotide binding protein (G protein, γ8 subunit); myosin; proteasome β3 subunit; ras homolog gene family, member B; cytochrome c-1; N-acetylglucosamine kinase; TNF receptor-associated protein 1 (TRAP1); and translin. By contrast, the unique protein found expressed in L. amazonensis infection was the succinate dehydrogenase, an enzyme involved in cell metabolism. The number of proteins identified in both L. major- and L. amazonensis-infected cells but that display differences in expression level was much higher, reaching a total of 162 proteins. A total number of 122 proteins were preferentially identified in L. major-infected macrophages while only forty of them showed higher expression in L. amazonensis infection. When the authors analyzed the greater differences in expression between these infected macrophages, they found a total of 15, 13 of which exhibited reduced expression in response to L. amazonensis infection, while two proteins showed increased expression in response to L. amazonensis infection. Considering the 15 proteins with significant levels of differential expression, 13 of these were found downregulated in L. amazonensis-infected macrophages, but these were upregulated in L. major-infected cells, and they were considered to be involved in several cell processes: coronin 1B, cytochrome C oxidase 6B (cox6B), heterogeneous nuclear ribonucleoprotein F (HNRPF), hypoxia-inducible factor 1-alpha (HIF-1α), osteoclast-stimulating factor-1 (OSTF1), programmed cell death protein 5 (PDCD5), protein phosphatase 2 (PP2), PYD And CARD domain-containing protein (PYCARD), RAB1, ribosomal protein S2 (RPS2), Serpin, peripheral benzodiazepine receptor (PBR), known as translocator protein (TSPO), and myosin light chain. The authors organized the identified proteins in networks using a biological network modeling, the Ingenuity Pathway Analysis (IPA)-Ingenuity Systems. This tool allows the organization of proteins detected in proteomic studies, as well as other proteins that are not identified by the mass spectrometric analysis, but may be involved in host response to infection. Interestingly, 14 out of 15 proteins with significant levels of differential expression were organized into a single network of cell development. Between the two highly expressed proteins in CBA macrophages infected with L. amazonensis, one of them, phospholipase D1 (PLD1), was proven to act on the membrane phospholipid, phosphatidylcholine. This protein causes the release of phosphatidic acid [56], as well as participates in the recruitment of additional membrane for the formation of nascent phagosomes. This protein can take part in the maintenance of early formed phagosome that will fuse with endocytic vesicles [57]. The authors suggested that the higher expression of PLD1 in L. amazonensis, but not in L. major infected macrophages, would contribute to the formation and maintenance of large parasitophorous vacuoles, characteristic of intracellular infection with L. amazonensis [58]. In this study [58], two out of the 15 proteins with a higher difference of expression were randomly selected for the validation of mass spectrometry results. Myosin light chain was validated as highly expressed in L. major-infected cells compared to L. amazonensis-infected macrophages using western blot and immunofluorescence staining for HIF-1α, which confirmed a higher expression of this protein in L. major-infected cells compared to those infected with L. amazonensis. The finding that myosin light chain was upregulated in L. major-infected macrophages was related to the formation of small individualized parasitophorous vacuoles induced and maintained throughout the maturation process in L. major infection [58], different from the large parasitophorous vacuoles that L. amazonensis induces in host cells. Previously, this protein has been implicated in the formation and maintenance of tight vacuoles formed around particles during phagocytosis [59]. Modulation of immune response was evidenced by HIF-1α, TRAP1, Serpin and PYDCARD that were upregulated in Leishmania-infected macrophages. TRAP1 and HIF-1α were found highly expressed in macrophages infected with L. major, and Serpin and PYDCARD exhibited reduced expression levels under L. amazonensis infection. TRAP1 has been shown to participate in the maintenance of cellular viability in cells subjected to H2O2-induced oxidative stress [60], and Serpin, a protein induced by TNF-α known to participate in conjunction with IL-1α in the inflammatory cascade [61]. Additionally, the PYDCARD adapter protein activates apoptosis by a mechanism dependent on NF-κB and caspases [62] that is initially induced by engagement of receptors in the TNF-α family. These proteomic findings can also be correlated with a previous study that CBA macrophages, which control L. major infection, express twice as much TNF-α when infected with L. major compared to those infected with L. amazonensis in response to IFN-γ stimulation [63]. Finally, the finding that HIF-1α is highly expressed in L. major-infected macrophages suggests a relationship between this transcription factor expression and a higher production of NO and expression of TNF-α, which are mediators known to act as regulators of HIF-1α when expressed under normoxic conditions [64]. The roles that myosin light chain and HIF-1α play in Leishmania infection deserve further investigation. Another study that evaluated the host cell response to Leishmania infection using proteomics was recently published. The study, which was a quantitative proteomic study, aimed to evaluate the effect of L. donovani parasites on the host cell response [23]. In this study, the authors infected the THP-1 with L. donovani promastigotes and following infection they determined relative and absolute quantification of protein expression using the isobaric tags (iTRAQ) method and LC-MS/MS. They compared the expression profiles of non-infected and L. donovani-infected THP-1 cells and found that proteins involved in major metabolic pathways, including glycolysis and fatty acid oxidation, are highly expressed after L. donovani infection, suggesting a parasite-induced global reprogramming of cell metabolism. The expression of proteins involved in gene transcription, RNA splicing (heterogeneous nuclear ribonucleoproteins (hnRNPs)), histones, and DNA repair and replication was also found to be increased in response to L. donovani infection in vitro. Proteins involved in cell survival and signal transduction were also shown to be more abundant in response to infection. Interestingly, several of the proteins that were identified as differentially expressed in this study had not been previously associated with the host response to the parasite, while some were related to proteins identified in previous studies. Quantitative polymerase chain reaction and immunoblot analysis of selected proteins identified in the study were used to validate proteins found differentially expressed in the mass spectrometry study. RNA interference (RNAi)-mediated gene knockdown of proteins was used to confirm the relevance of the observed host quantitative proteomic screening. Interestingly, this study shows that the mitochondrial antiviral signaling protein (MAVS) was significantly abundant in host cells after Leishmania infection. MAVS, the first mitochondrial protein to activate NF-κB and interferon (IFN) regulatory factors (IRF3 and IRF7), is known to synthesize type I interferons (IFN-α and IFN-β), which are essential in antiviral signaling. The silencing of endogenous MAVS expression by RNAi inhibits the activation of NF-κB, IRF3, and IRF7, thereby blocking the production of interferon and promoting viral infection [65]. These authors claim that there may be cross-talk between MAVS and the components of the NF-κB and IRF signaling pathways, which leads to the production of proinflammatory cytokines and type I IFN [66]. A recent study published using phosphoproteomics showed that Leishmania parasites respond to arginine pool reduction in the host cell by up-regulating expression and activity of the Leishmania arginine transporter (LdAAP3), as well as several other transporters. To study phosphoproteins involved in the signaling pathway that results in increased LdAAP3 expression and activity, the authors used a di-methylation tagging technique to evaluate changes in the phosphorylation profile of Leishmania promastigotes after five and 15 min of arginine deprivation. Phosphorylation analysis showed an increased phosphorylation of mitogen-activated protein kinase 2 (MPK2), suggesting that the arginine-deprivation response during Leishmania infection is mediated through a MPK2-dependent signaling cascade [67]. Taken together, these three studies show that Leishmania parasites modulate the host cell proteome during early stages of infection, providing evidence for the complex relations between the host and the parasite at the molecular level. These studies also reveal potential novel cellular targets which could modulate cell response to help control Leishmania infection. 4. Protein Profiling in Human Cutaneous Lesions A recent study generated interesting results in tegumentary leishmaniasis using 2-DE proteomics, as well as biological network analysis for protein profiling in cutaneous lesions comparing protein expression to normal skin samples [26]. In this study, the authors found that cutaneous lesions showed a composition similar to the inflammatory infiltrate with the presence of lymphocytes, macrophages, and plasma cells, as well as focal necrosis. Proteins that were extracted from lesions and normal skin samples were separated using 2-DE. Among a total of 150 differentially expressed spots of proteins from cutaneous lesions and normal skin samples, the authors identified 59 proteins. From these, 29 spots were identified only in CL samples, while 17 were found only in the normal skin. Besides, among the spots present in both cutaneous lesions and control samples, nine spots were upregulated, and four were downregulated in CL biopsies in comparison to normal skin. The proteins identified in the mass spectrometry analysis were organized according to biological functions, and the analysis revealed that those upregulated in CL biopsies participate in several cell processes including apoptosis (caspase-9 (CASP-9)), immune response (T cell receptor beta (TRB)), and biosynthetic process (transcription factor IIIB 90 kDa subunit (BRF1)). Previously, it was observed that an increased expression of TRB [68,69] could be involved in the recruitment of T cells to the infection site. The authors claimed that the increase in the apoptotic process, consequent to CASP-9 expression could be the result of the development of an intense inflammatory response commonly observed in cutaneous lesions of leishmaniasis. In this work, the authors performed biological network analysis to better understand how biological pathways were modulated in infected individuals. This approach was also used in the in vitro study using mouse macrophages [22]. To perform this analysis, the authors generated networks of interactions between the identified proteins, in addition to other proteins, which were included in the network by the Cytoscape software. A complex network was generated by Cytoscape, containing 505 nodes that contained many proteins involved in apoptosis, such as IL-23, transforming growth factor beta receptor 1 (TGFBR1), tumor necrosis factor receptor-I (TNFRI), caspase-3 (CASP-3), caspase-8 (CASP-8), and granzyme B (GZMB). Reinforcing the role that the apoptotic process plays in CL development, IPA generated the cytotoxic T lymphocyte-mediated apoptosis of target cells as the primary canonical pathway. This pathway included 34 proteins, 18 of which were differentially expressed among the samples. Nine out of these 18 proteins were exclusively expressed in CL biopsies, and five were upregulated, while four were downregulated, in CL biopsies when compared to normal skin. In the apoptosis cascade, granzyme B is a protein that has been shown to activate caspase-3, either directly or via the mitochondrial pathway, by way of induction of caspase-9 activation, which subsequently activates caspase-3 [70]. These authors validated the expression of three proteins, two of which were identified as differentially expressed in their proteomic study and another that was identified by IPA. Using immunohistochemistry, they demonstrated higher expression of caspase-9, caspase-3, and granzyme B in CL biopsies in comparison to normal skin. Also, they found that higher expression of these three proteins correlated positively with lesion size in CL patients. They conclude that apoptosis is probably involved in the mechanisms associated with the progression of tissue damage seen in CL lesions. 5. Protein Expression in Serum of Individuals with Visceral Leishmaniasis Very few studies have used large-scale analysis to better understand the aspects surrounding host immune inflammatory response to Leishmania infection. Recently, three studies have used proteomics to identify biomarkers that potentially participate in host immune response in the serum of VL patients [28,29], in addition to those that may help in the diagnosis of visceral disease [27]. Two proteomic studies employing either a quantitative [28] or a qualitative comparative [29] analysis identified differentially expressed proteins in human serum from patients and control groups. The patient study groups were diagnosed with VL as confirmed by the presence of Leishmania parasites in bone marrow aspirate [28,29] or by the presence of serum anti-rK39 antibody [28,29]. The first study using a quantitative proteomic approach evaluated the proteome in the serum of six VL patients and six healthy volunteers. The authors identified 26 differentially expressed spots, from these 15 were analyzed by mass spectrometry and only nine spots were identified with high confidence, corresponding to five different proteins. In the comparative proteomic study, the authors purified only plasma glycoproteins using a multi-lectin affinity column, followed by mass spectrometry analysis that allowed for the identification of 39 differentially expressed spots [29]. In both quantitative [28] and qualitative comparative [29] studies, some acute-phase proteins were found to be differentially expressed in sera from humans with VL compared to controls, supporting the notion that VL is a systemic disease. The glycoproteins identified as upregulated in only the qualitative comparative proteomic study were α-1-acid glycoprotein and C1 inhibitor, while, in the quantitative study, α-1-antitrypsin, α-1-B glycoprotein, and amyloid-A1 precursor were detected. On the other hand, the only downregulated protein identified in the qualitative glycoproteomic study, was the retinol binding protein, while, in the quantitative study, the vitamin-D binding protein was detected. Interestingly, in both studies, apolipoprotein A-I and transthyretin were found to be downregulated in VL sera [28,29]. Previously, α-1-acidglycoprotein was found to be elevated in sera of individuals with systemic tissue injury, or during inflammation and infection. This protein was also proven to be involved in other processes, such as neutrophil inactivation, chemotaxis and oxidative metabolism [71]. The authors claimed that the increased levels of the acute phase α-1-acid glycoprotein observed in serum from VL patients can result in the inhibition of neutrophil function, facilitating pyogenic infections in the skin and other tissues [28]. C1-inhibitor was the other acute-phase protein found to be overexpressed in serum from VL patients [28]. Inhibition of both complement and quinin generating cascades [72] by this plasma protease inhibitor has been associated with the anti-inflammatory effect of C1-inhibitor [73]. One of the proteins found to be downregulated in sera from VL patients was transthyretin [28], which is known to function as a transporter of thyroid hormones, as well as a negative acute phase protein with anti-inflammatory properties that subsequently results in inhibition of interleukin-1 production by monocytes and endothelial cells [74]. In agreement with this proteomic study, a previous one found transthyretin levels reduced in sera of patients with VL [75]. The other protein found downmodulated in the sera from VL patients was the retinol binding protein [28], known to transport retinol from the liver to the peripheral tissues. In plasma, retinol binding protein interacts with transthyretin, preventing retinol binding protein loss through filtration in the kidney [76]. Another proteomic study was performed to identify proteins potentially expressed by parasites in dogs with VL, as well as differences in protein expression in serum from infected and non-infected dogs [27]. The study performed in naturally infected dogs demonstrated that the mass spectrometry was not sensitive enough to detect parasite protein in dog sera. Nonetheless, labeling each sample of infected and non-infected individuals with different iTRAQ tag followed by the LC-MS/MS analysis allowed the authors to identify differentially expressed host proteins in serum of infected animals when compared to the non-infected ones. More than 105 proteins were identified in the serum of dogs, of which 22 were found in sera from infected dogs compared to controls. Of these 22 proteins, 17 were upregulated, and five were downregulated [27]. The number of proteins identified in the serum of dogs in this proteomic study [27] was greater than the number found in the proteomic studies using human serum [28,29]. Despite differences in the number of proteins expressed in dogs and humans with VL, increased levels of acute-phase proteins were identified as modulated in the three studies [28,29]. The proteins found in dog serum with VL, such as serum amyloid A (SAA) and haptoglobin, were different from those detected in human serum. The modulation of acute-phase proteins is consistent with results from previous studies, both in canine [77] and in human VL [78]. Furthermore, in patients with VL, the increased levels of such acute phase proteins in serum progressively decreased with effective therapy, while the high expression levels were maintained in those patients with parasite clearance delay [78]. In these proteomic studies, the finding that the inhibition of negative acute-phase proteins along with the enhancement of positive acute-phase protein expression in sera from humans and dogs with VL reinforces the notion that infection leads to an uncontrolled systemic inflammatory response. This response can account for the pathogenesis and clinical manifestations of VL [79]. The different studies postulate that the identification of these molecules could be useful as diagnostic/prognostic biomarkers and in the understanding of parasite survival in the host environment [29]. Additionally, these molecules could function as potential targets for future development of new treatment for controlling the clinical manifestations of the disease [28]. Finally, the study that found a reduction in the expression of acute-phase proteins in response to antileishmanial therapy [77] reinforces the potential use of these molecules as targets for monitoring the initial response to treatment and follow-up of humans and dogs with VL. 6. Concluding Remarks Comparative proteomics studies have demonstrated that the relation between Leishmania parasites and the host is extremely complex at the molecular level. Although very few proteins have been identified exclusively in Leishmania amastigotes, a consistent number of proteins modulated in this life stage were involved in metabolic pathways. In addition, during the host-parasite interaction, the expression of proteins in the host varies depending on the Leishmania spp., as well as the tissue or cell studied (Figure 3). However, most studies reported that the proteins involved in metabolic pathways and the immune and inflammatory response of the host are frequently modulated. To this date, none of the molecular targets identified in proteomic studies were used to develop new treatment approaches or as prognostic markers in a follow-up of different types of Leishmania infection. This review will provide a reference to conduct new studies to better understand the role these proteins and pathways play in the intracellular survival of Leishmania parasites and the outcome of leishmaniasis. Acknowledgments We thank Instituto Gonçalo Moniz/Fundação Oswaldo Cruz (IGM/FIOCRUZ); Coordenação de Aperfeiçoamento do Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB) for the financial support. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Proteomics approach process. The study of the proteome of Leishmania-infected cells and tissues, using mass spectrometry, can lead to the identification of targets for prophylactic and chemotherapeutic treatment and to the identification of biomarkers that can be used for the diagnosis of different diseases. The aim of the present report is to review the recent contributions of proteomics to the understanding of the various aspects of the Leishmania-mammalian host interaction. We will first describe the contributions made by large-scale proteomic studies on alterations of protein expression in parasites during their differentiation process from promastigotes to amastigotes, followed by studies identifying proteins differentially expressed by host cells and tissues in response to infection. These recent studies explored proteins expressed in macrophage-Leishmania interaction in vitro [22,23], in cutaneous lesions of infected humans [26], as well as in serum of individuals with VL [27,28,29]. Figure 2 The life cycle of Leishmania parasites. During blood feeding by female sandflies, metacyclic promastigotes are regurgitated. These promastigotes are then phagocytosed by cells at the site of the bite. Once inside the host cells, metacyclic promastigotes transform into amastigotes, which can survive and replicate inside phagolysosomes. Amastigote replication may lead to host cell rupture, allowing reinfection of other phagocytes. When infected phagocytes are taken up by another sandfly during the blood meal, amastigotes transform into procyclic promastigotes in the sandfly midgut. Leishmania procyclic promastigotes then differentiate into infective metacyclic promastigotes, completing the cycle. Figure 3 Proteins involved in the host response to Leishmania infection. Proteins identified as differently expressed in Leishmania infected host in proteomics studies. ==== Refs References 1. Alvar J. Velez I.D. Bern C. Herrero M. Desjeux P. Cano J. Jannin J. den Boer M. WHO Leishmaniasis Control Team. Leishmaniasis worldwide and global estimates of its incidence PLoS ONE 2012 7 1270 10.1371/journal.pone.0035671 22693548 2. WHO Leishmaniasis. Situation and Trends Available online: http://apps.who.int/gho/data/node.main.NTDLEISH?lang=en (accessed on 1 August 2016) 3. Zijlstra E.E. Musa A.M. Khalil E.A. El-Hassan I.M. El-Hassan A.M. Post-kala-azar dermal leishmaniasis Lancet Infect. Dis. 2003 3 87 98 10.1016/S1473-3099(03)00517-6 12560194 4. Desjeux P. Human leishmaniases: Epidemiology and public health aspects World Health Stat. Q. 1992 45 267 275 1462660 5. Graves P.R. Haystead T.A. Molecular biologist’s guide to proteomics Microbiol. Mol. Biol. Rev. 2002 66 39 63 10.1128/MMBR.66.1.39-63.2002 11875127 6. Khurshid Z. Zohaib S. Najeeb S. Zafar M.S. Rehman R. Rehman I.U. Advances of proteomic sciences in dentistry Int. J. Mol. Sci. 2016 17 728 10.3390/ijms17050728 27187379 7. Renella R. Clinically-oriented proteomic investigation of sickle cell disease: Opportunities and challenges Proteom. Clin. Appl. 2016 10.1002/prca.201500133 27193514 8. Gaudreau P.O. Stagg J. Soulieres D. Saad F. The present and future of biomarkers in prostate cancer: Proteomics, genomics, and immunology advancements Biomark. Cancer 2016 8 15 33 10.4137/BIC.S31802 27168728 9. Ahsan N. Rao R.S. Gruppuso P.A. Ramratnam B. Salomon A.R. Targeted proteomics: Current status and future perspectives for quantification of food allergens J. Proteom. 2016 143 15 23 10.1016/j.jprot.2016.04.018 27113134 10. Ferreira D. Seca A.M. Diana C.G.A. Silva A.M. Targeting human pathogenic bacteria by siderophores: A proteomics review J. Proteom. 2016 16 10.1016/j.jprot.2016.04.006 27109355 11. Charest H. Zhang W.W. Matlashewski G. The developmental expression of Leishmania donovani A2 amastigote-specific genes is post-transcriptionally mediated and involves elements located in the 3′-untranslated region J. Biol. Chem. 1996 271 17081 17090 10.1074/jbc.271.29.17081 8663340 12. Brooks D.R. Denise H. Westrop G.D. Coombs G.H. Mottram J.C. The stage-regulated expression of Leishmania mexicana CPB cysteine proteases is mediated by an intercistronic sequence element J. Biol. Chem. 2001 276 47061 47069 10.1074/jbc.M108498200 11592967 13. Kelly B.L. Nelson T.N. McMaster W.R. Stage-specific expression in leishmania conferred by 3′ untranslated regions of L. Major Leishmanolysin genes (GP63 ) Mol. Biochem. Parasitol. 2001 116 101 104 11463473 14. Clayton C. Shapira M. Post-transcriptional regulation of gene expression in trypanosomes and Leishmania s Mol. Biochem. Parasitol. 2007 156 93 101 10.1016/j.molbiopara.2007.07.007 17765983 15. Sodre C.L. Chapeaurouge A.D. Kalume D.E. de Mendonca Lima L. Perales J. Fernandes O. Proteomic map of Trypanosoma cruzi CL brener: The reference strain of the genome project Arch. Microbiol. 2009 191 177 184 10.1007/s00203-008-0439-6 19002435 16. Gygi S.P. Rochon Y. Franza B.R. Aebersold R. Correlation between protein and mRNA abundance in yeast Mol. Cell. Biol. 1999 19 1720 1730 10.1128/MCB.19.3.1720 10022859 17. Larreta R. Soto M. Quijada L. Folgueira C. Abanades D.R. Alonso C. Requena J.M. The expression of HSP83 genes in Leishmania infantum is affected by temperature and by stage-differentiation and is regulated at the levels of mrna stability and translation BMC Mol. Biol. 2004 5 1270 10.1186/1471-2199-5-3 15176985 18. El Fakhry Y. Ouellette M. Papadopoulou B. A proteomic approach to identify developmentally regulated proteins in Leishmania infantum Proteomics 2002 2 1007 1017 10.1002/1615-9861(200208)2:8<1007::AID-PROT1007>3.0.CO;2-G 12203896 19. Bente M. Harder S. Wiesgigl M. Heukeshoven J. Gelhaus C. Krause E. Clos J. Bruchhaus I. Developmentally induced changes of the proteome in the protozoan parasite Leishmania donovani Proteomics 2003 3 1811 1829 10.1002/pmic.200300462 12973740 20. Morales M.A. Watanabe R. Laurent C. Lenormand P. Rousselle J.C. Namane A. Spath G.F. Phosphoproteomic analysis of Leishmania donovani pro- and amastigote stages Proteomics 2008 8 350 363 10.1002/pmic.200700697 18203260 21. Biyani N. Madhubala R. Quantitative proteomic profiling of the promastigotes and the intracellular amastigotes of Leishmania donovani isolates identifies novel proteins having a role in leishmania differentiation and intracellular survival Biochim. Biophys. Acta 2012 1824 1342 1350 10.1016/j.bbapap.2012.07.010 22850196 22. Menezes J.P. Almeida T.F. Petersen A.L. Guedes C.E. Mota M.S. Lima J.G. Palma L.C. Buck G.A. Krieger M.A. Probst C.M. Proteomic analysis reveals differentially expressed proteins in macrophages infected with Leishmania amazonensis or Leishmania major Microbes Infect. 2013 15 579 591 10.1016/j.micinf.2013.04.005 23628411 23. Singh A.K. Pandey R.K. Siqueira-Neto J.L. Kwon Y.J. Freitas-Junior L.H. Shaha C. Madhubala R. Proteomic-based approach to gain insight into reprogramming of THP-1 cells exposed to Leishmania donovani over an early temporal window Infect. Immun. 2015 83 1853 1868 10.1128/IAI.02833-14 25690103 24. Costa M.M. Andrade H.M. Bartholomeu D.C. Freitas L.M. Pires S.F. Chapeaurouge A.D. Perales J. Ferreira A.T. Giusta M.S. Melo M.N. Analysis of Leishmania chagasi by 2-D difference gel electrophoresis (2-D dige) and immunoproteomic: Identification of novel candidate antigens for diagnostic tests and vaccine J. Proteome Res. 2011 10 2172 2184 10.1021/pr101286y 21355625 25. Paape D. Aebischer T. Contribution of proteomics of Leishmania spp. To the understanding of differentiation, drug resistance mechanisms, vaccine and drug development J. Proteom. 2011 74 1614 1624 10.1016/j.jprot.2011.05.005 21621022 26. Da Silva Santos C. Attarha S. Saini R.K. Boaventura V. Costa J. Khouri R. Barral-Netto M. Brodskyn C.I. Souchelnytskyi S. Proteome profiling of human cutaneous leishmaniasis lesion J. Investig. Dermatol. 2015 135 400 410 10.1038/jid.2014.396 25207817 27. Britti D. Gaspari M. Massimini G. Casalinuovo F. Morittu V.M. Cuda G. Proteomic analysis in canine leishmaniasis Vet. Res. Commun. 2010 34 Suppl. 1 S91 S96 10.1007/s11259-010-9410-5 20440645 28. Rukmangadachar L.A. Kataria J. Hariprasad G. Samantaray J.C. Srinivasan A. Two-dimensional difference gel electrophoresis (DIGE) analysis of sera from visceral leishmaniasis patients Clin. Proteom. 2011 8 4 10.1186/1559-0275-8-4 21906353 29. Bag A.K. Saha S. Sundar S. Saha B. Chakrabarti A. Mandal C. Comparative proteomics and glycoproteomics of plasma proteins in Indian visceral leishmaniasis Proteome Sci. 2014 12 48 10.1186/s12953-014-0048-z 25276097 30. MacFarlane J. Blaxter M.L. Bishop R.P. Miles M.A. Kelly J.M. Identification and characterisation of a Leishmania donovani antigen belonging to the 70-kDa heat-shock protein family Eur. J. Biochem. 1990 190 377 384 10.1111/j.1432-1033.1990.tb15586.x 2163842 31. Glaser T.A. Moody S.F. Handman E. Bacic A. Spithill T.W. An antigenically distinct lipophosphoglycan on amastigotes of Leishmania major Mol. Biochem. Parasitol. 1991 45 337 344 10.1016/0166-6851(91)90102-C 1710036 32. Zilberstein D. Shapira M. The role of pH and temperature in the development of Leishmania parasites Annu. Rev. Microbiol. 1994 48 449 470 10.1146/annurev.mi.48.100194.002313 7826014 33. Alexander J. Russell D.G. The interaction of leishmania species with macrophages Adv. Parasitol. 1992 31 175 254 1496927 34. Chappuis F. Sundar S. Hailu A. Ghalib H. Rijal S. Peeling R.W. Alvar J. Boelaert M. Visceral leishmaniasis: What are the needs for diagnosis, treatment and control? Nat. Rev. Microbiol. 2007 5 873 882 10.1038/nrmicro1748 17938629 35. Bates P.A. Rogers M.E. New insights into the developmental biology and transmission mechanisms of Leishmania Curr. Mol. Med. 2004 4 601 609 10.2174/1566524043360285 15357211 36. Kaye P. Scott P. Leishmaniasis: Complexity at the host-pathogen interface Nat. Rev. Microbiol. 2011 9 604 615 10.1038/nrmicro2608 21747391 37. Matlashewski G. Leishmania infection and virulence Med. Microbiol. Immunol. 2001 190 37 42 10.1007/s004300100076 11770107 38. Turco S.J. Spath G.F. Beverley S.M. Is lipophosphoglycan a virulence factor? A surprising diversity between Leishmania species Trends Parasitol. 2001 17 223 226 10.1016/S1471-4922(01)01895-5 11323305 39. Liaud M.F. Lichtle C. Apt K. Martin W. Cerff R. Compartment-specific isoforms of TPI and GAPDH are imported into diatom mitochondria as a fusion protein: Evidence in favor of a mitochondrial origin of the eukaryotic glycolytic pathway Mol. Biol. Evol. 2000 17 213 223 10.1093/oxfordjournals.molbev.a026301 10677844 40. McNicoll F. Drummelsmith J. Muller M. Madore E. Boilard N. Ouellette M. Papadopoulou B. A combined proteomic and transcriptomic approach to the study of stage differentiation in Leishmania infantum Proteomics 2006 6 3567 3581 10.1002/pmic.200500853 16705753 41. Nugent P.G. Karsani S.A. Wait R. Tempero J. Smith D.F. Proteomic analysis of Leishmania mexicana differentiation Mol. Biochem. Parasitol. 2004 136 51 62 10.1016/j.molbiopara.2004.02.009 15138067 42. Walker J. Vasquez J.J. Gomez M.A. Drummelsmith J. Burchmore R. Girard I. Ouellette M. Identification of developmentally-regulated proteins in Leishmania panamensis by proteome profiling of promastigotes and axenic amastigotes Mol. Biochem. Parasitol. 2006 147 64 73 10.1016/j.molbiopara.2006.01.008 16530278 43. Brotherton M.C. Racine G. Foucher A.L. Drummelsmith J. Papadopoulou B. Ouellette M. Analysis of stage-specific expression of basic proteins in Leishmania infantum J. Proteome Res. 2010 9 3842 3853 10.1021/pr100048m 20583757 44. Acestor N. Masina S. Walker J. Saravia N.G. Fasel N. Quadroni M. Establishing two-dimensional gels for the analysis of Leishmania proteomes Proteomics 2002 2 877 879 10.1002/1615-9861(200207)2:7<877::AID-PROT877>3.0.CO;2-D 12124933 45. Leifso K. Cohen-Freue G. Dogra N. Murray A. McMaster W.R. Genomic and proteomic expression analysis of Leishmania promastigote and amastigote life stages: The Leishmania genome is constitutively expressed Mol. Biochem. Parasitol. 2007 152 35 46 10.1016/j.molbiopara.2006.11.009 17188763 46. Pan A.A. Duboise S.M. Eperon S. Rivas L. Hodgkinson V. Traub-Cseko Y. McMahon-Pratt D. Developmental life cycle of Leishmania —Cultivation and characterization of cultured extracellular amastigotes J. Eukaryot. Microbiol. 1993 40 213 223 10.1111/j.1550-7408.1993.tb04906.x 8461895 47. Sereno D. Lemesre J.L. Axenically cultured amastigote forms as an in vitro model for investigation of antileishmanial agents Antimicrob. Agents Chemother. 1997 41 972 976 9145854 48. Puentes F. Diaz D. Hoya R.D. Gutierrez J.A. Lozano J.M. Patarroyo M.E. Moreno A. Cultivation and characterization of stable Leishmania guyanensis complex axenic amastigotes derived from infected U937 cells Am. J. Trop. Med. Hyg. 2000 63 102 110 11357988 49. Gupta N. Goyal N. Rastogi A.K. In vitro cultivation and characterization of axenic amastigotes of Leishmania Trends Parasitol. 2001 17 150 153 10.1016/S1471-4922(00)01811-0 11286801 50. Sereno D. Roy G. Lemesre J.L. Papadopoulou B. Ouellette M. DNA transformation of Leishmania infantum axenic amastigotes and their use in drug screening Antimicrob. Agents Chemother. 2001 45 1168 1173 10.1128/AAC.45.4.1168-1173.2001 11257031 51. Debrabant A. Joshi M.B. Pimenta P.F. Dwyer D.M. Generation of Leishmania donovani axenic amastigotes: Their growth and biological characteristics Int. J. Parasitol. 2004 34 205 217 10.1016/j.ijpara.2003.10.011 15037106 52. Paape D. Lippuner C. Schmid M. Ackermann R. Barrios-Llerena M.E. Zimny-Arndt U. Brinkmann V. Arndt B. Pleissner K.P. Jungblut P.R. Transgenic, fluorescent Leishmania mexicana allow direct analysis of the proteome of intracellular amastigotes Mol. Cell. Proteom. 2008 7 1688 1701 10.1074/mcp.M700343-MCP200 18474515 53. Rosenzweig D. Smith D. Myler P.J. Olafson R.W. Zilberstein D. Post-translational modification of cellular proteins during Leishmania donovani differentiation Proteomics 2008 8 1843 1850 10.1002/pmic.200701043 18398879 54. Purdy J.E. Donelson J.E. Wilson M.E. Regulation of genes encoding the major surface protease of Leishmania chagasi via mRNA stability Mol. Biochem. Parasitol. 2005 142 88 97 10.1016/j.molbiopara.2005.03.010 15876463 55. Gomes I.N. Calabrich A.F. Tavares Rda S. Wietzerbin J. de Freitas L.A. Veras P.S. Differential properties of CBA/J mononuclear phagocytes recovered from an inflammatory site and probed with two different species of Leishmania Microbes Infect. 2003 5 251 260 10.1016/S1286-4579(03)00025-X 12706438 56. Wang L. Cummings R. Usatyuk P. Morris A. Irani K. Natarajan V. Involvement of phospholipases D1 and D2 in sphingosine 1-phosphate-induced ERK (extracellular-signal-regulated kinase) activation and interleukin-8 secretion in human bronchial epithelial cells Biochem. J. 2002 367 751 760 10.1042/bj20020586 12149127 57. Corrotte M. Chasserot-Golaz S. Huang P. Du G. Ktistakis N.T. Frohman M.A. Vitale N. Bader M.F. Grant N.J. Dynamics and function of phospholipase D and phosphatidic acid during phagocytosis Traffic 2006 7 365 377 10.1111/j.1600-0854.2006.00389.x 16497229 58. Antoine J.C. Prina E. Lang T. Courret N. The biogenesis and properties of the parasitophorous vacuoles that harbour Leishmania in murine macrophages Trends Microbiol. 1998 6 392 401 10.1016/S0966-842X(98)01324-9 9807783 59. Araki N. Hatae T. Furukawa A. Swanson J.A. Phosphoinositide-3-kinase-independent contractile activities associated with Fcgamma-receptor-mediated phagocytosis and macropinocytosis in macrophages J. Cell Sci. 2003 116 247 257 10.1242/jcs.00235 12482911 60. Pridgeon J.W. Olzmann J.A. Chin L.S. Li L. PINK1 protects against oxidative stress by phosphorylating mitochondrial chaperone TRAP1 PLoS Biol. 2007 5 1270 10.1371/journal.pbio.0050172 17579517 61. Mishra S. Fujita T. Lama V.N. Nam D. Liao H. Okada M. Minamoto K. Yoshikawa Y. Harada H. Pinsky D.J. Carbon monoxide rescues ischemic lungs by interrupting MAPK-driven expression of early growth response 1 gene and its downstream target genes Proc. Natl. Acad. Sci. USA 2006 103 5191 5196 10.1073/pnas.0600241103 16551742 62. Reed J.C. Doctor K. Rojas A. Zapata J.M. Stehlik C. Fiorentino L. Damiano J. Roth W. Matsuzawa S. Newman R. Comparative analysis of apoptosis and inflammation genes of mice and humans Genome Res. 2003 13 1376 1388 10.1101/gr.1053803 12819136 63. Diefenbach A. Schindler H. Donhauser N. Lorenz E. Laskay T. MacMicking J. Rollinghoff M. Gresser I. Bogdan C. Type 1 interferon (IFNα/β) and type 2 nitric oxide synthase regulate the innate immune response to a protozoan parasite Immunity 1998 8 77 87 10.1016/S1074-7613(00)80460-4 9462513 64. Zhou J. Fandrey J. Schumann J. Tiegs G. Brune B. No and TNF-α released from activated macrophages stabilize HIF-1α in resting tubular LLC-PK1 cells Am. J. Physiol. Cell Physiol. 2003 284 C439 C446 10.1152/ajpcell.00294.2002 12388069 65. Yan H. Tsai M.D. Nucleoside monophosphate kinases: Structure, mechanism, and substrate specificity Adv. Enzymol. Relat. Areas Mol. Biol. 1999 73 103 134 10218107 66. Villa H. Perez-Pertejo Y. Garcia-Estrada C. Reguera R.M. Requena J.M. Tekwani B.L. Balana-Fouce R. Ordonez D. Molecular and functional characterization of adenylate kinase 2 gene from Leishmania donovani Eur. J. Biochem. 2003 270 4339 4347 10.1046/j.1432-1033.2003.03826.x 14622299 67. Goldman-Pinkovich A. Balno C. Strasser R. Zeituni-Molad M. Bendelak K. Rentsch D. Ephros M. Wiese M. Jardim A. Myler P.J. An arginine deprivation response pathway is induced in Leishmania during macrophage invasion PLoS Pathog. 2016 12 1270 10.1371/journal.ppat.1005494 27043018 68. Clarencio J. de Oliveira C.I. Bomfim G. Pompeu M.M. Teixeira M.J. Barbosa T.C. Souza-Neto S. Carvalho E.M. Brodskyn C. Barral A. Characterization of the T-cell receptor Vβ repertoire in the human immune response against Leishmania parasites Infect. Immun. 2006 74 4757 4765 10.1128/IAI.00265-06 16861664 69. Keesen T.S. Antonelli L.R. Faria D.R. Guimaraes L.H. Bacellar O. Carvalho E.M. Dutra W.O. Gollob K.J. CD4+ T cells defined by their Vβ T cell receptor expression are associated with immunoregulatory profiles and lesion size in human leishmaniasis Clin. Exp. Immunol. 2011 165 338 351 10.1111/j.1365-2249.2011.04430.x 21726211 70. Lord S.J. Rajotte R.V. Korbutt G.S. Bleackley R.C. Granzyme B: A natural born killer Immunol. Rev. 2003 193 31 38 10.1034/j.1600-065X.2003.00044.x 12752668 71. Pucadyil T.J. Tewary P. Madhubala R. Chattopadhyay A. Cholesterol is required for Leishmania donovani infection: Implications in leishmaniasis Mol. Biochem. Parasitol. 2004 133 145 152 10.1016/j.molbiopara.2003.10.002 14698427 72. Ben-Othman R. Flannery A.R. Miguel D.C. Ward D.M. Kaplan J. Andrews N.W. Leishmania -mediated inhibition of iron export promotes parasite replication in macrophages PLoS Pathog. 2014 10 1270 10.1371/journal.ppat.1003901 24497831 73. Chava A.K. Chatterjee M. Sharma V. Sundar S. Mandal C. Variable degree of alternative complement pathway-mediated hemolysis in Indian visceral leishmaniasis induced by differential expression of 9-O -acetylated sialoglycans J. Infect. Dis. 2004 189 1257 1264 10.1086/382752 15031795 74. Candiano G. Bruschi M. Musante L. Santucci L. Ghiggeri G.M. Carnemolla B. Orecchia P. Zardi L. Righetti P.G. Blue silver: A very sensitive colloidal coomassie G-250 staining for proteome analysis Electrophoresis 2004 25 1327 1333 10.1002/elps.200305844 15174055 75. Yan J.X. Wait R. Berkelman T. Harry R.A. Westbrook J.A. Wheeler C.H. Dunn M.J. A modified silver staining protocol for visualization of proteins compatible with matrix-assisted laser desorption/ionization and electrospray ionization-mass spectrometry Electrophoresis 2000 21 3666 3672 10.1002/1522-2683(200011)21:17<3666::AID-ELPS3666>3.0.CO;2-6 11271485 76. Friedman D.B. Wang S.E. Whitwell C.W. Caprioli R.M. Arteaga C.L. Multivariable difference gel electrophoresis and mass spectrometry: A case study on transforming growth factor-β and ERBB2 signaling Mol. Cell. Proteom. 2007 6 150 169 10.1074/mcp.D600001-MCP200 17028091 77. Martinez-Subiela S. Bernal L.J. Ceron J.J. Serum concentrations of acute-phase proteins in dogs with leishmaniosis during short-term treatment Am. J. Vet. Res. 2003 64 1021 1026 10.2460/ajvr.2003.64.1021 12926596 78. Wasunna K.M. Raynes J.G. Were J.B. Muigai R. Sherwood J. Gachihi G. Carpenter L. McAdam K.P. Acute phase protein concentrations predict parasite clearance rate during therapy for visceral leishmaniasis Trans. R. Soc. Trop. Med. Hyg. 1995 89 678 681 10.1016/0035-9203(95)90442-5 8594695 79. Cavalcanti A.S. Ribeiro-Alves M. Pereira Lde O. Mestre G.L. Ferreira A.B. Morgado F.N. Boite M.C. Cupolillo E. Moraes M.O. Porrozzi R. Parasite load induces progressive spleen architecture breakage and impairs cytokine mRNA expression in Leishmania infantum-naturally infected dogs PLoS ONE 2015 10 1270 10.1371/journal.pone.0123009 25875101

PMC005xxxxxx/PMC5000669.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081271ijms-17-01271ArticleThe FKBP5 Gene Affects Alcohol Drinking in Knockout Mice and Is Implicated in Alcohol Drinking in Humans Qiu Bin 1Luczak Susan E. 2Wall Tamara L. 345Kirchhoff Aaron M. 6Xu Yuxue 1Eng Mimy Y. 5Stewart Robert B. 7Shou Weinian 8Boehm Stephen L. II7Chester Julia A. 9Yong Weidong 18*Liang Tiebing 10*Singh Ashok K. Academic Editor1 Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences (CAMS), Peking Union Medical College (PUMC), Beijing 100021, China; [email protected] (B.Q.); [email protected] (Y.X.)2 Department of Psychology, University of Southern California, Los Angeles, CA 90089, USA; [email protected] Department of Psychiatry, University of California, San Diego, CA 92037, USA; [email protected] Psychology Service, Veterans Affairs San Diego Healthcare System, San Diego, CA 92161, USA5 Veterans Medical Research Foundation, San Diego, CA 92161, USA; [email protected] Immunology and Microbial Science Department, Research Technician, The Scripps Research Institute, Scripps Clinic South Driveway, La Jolla, CA 92037, USA; [email protected] Department of Psychology, Indiana University-Purdue University Indianapolis, Indianapolis, IN 46202, USA; [email protected] (R.B.S.); [email protected] (S.L.B.)8 Departments of Pediatrics and Medicine, Indiana University School of Medicine, Indianapolis, IN 46202, USA; [email protected] Department of Psychological Sciences, Purdue University, West Lafayette, IN 47907, USA; [email protected] Department of Medicine, Indiana University School of Medicine Gatch Hall, Indianapolis, IN 46202, USA* Correspondence: [email protected] (W.Y.); [email protected] (T.L.); Tel.: +86-01-6776-2060 (W.Y.); +1-317-274-7813 (T.L.)05 8 2016 8 2016 17 8 127115 6 2016 25 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).FKBP5 encodes FK506-binding protein 5, a glucocorticoid receptor (GR)-binding protein implicated in various psychiatric disorders and alcohol withdrawal severity. The purpose of this study is to characterize alcohol preference and related phenotypes in Fkbp5 knockout (KO) mice and to examine the role of FKBP5 in human alcohol consumption. The following experiments were performed to characterize Fkpb5 KO mice. (1) Fkbp5 KO and wild-type (WT) EtOH consumption was tested using a two-bottle choice paradigm; (2) The EtOH elimination rate was measured after intraperitoneal (IP) injection of 2.0 g/kg EtOH; (3) Blood alcohol concentration (BAC) was measured after 3 h limited access of alcohol; (4) Brain region expression of Fkbp5 was identified using LacZ staining; (5) Baseline corticosterone (CORT) was assessed. Additionally, two SNPs, rs1360780 (C/T) and rs3800373 (T/G), were selected to study the association of FKBP5 with alcohol consumption in humans. Participants were college students (n = 1162) from 21–26 years of age with Chinese, Korean or Caucasian ethnicity. The results, compared to WT mice, for KO mice exhibited an increase in alcohol consumption that was not due to differences in taste sensitivity or alcohol metabolism. Higher BAC was found in KO mice after 3 h of EtOH access. Fkbp5 was highly expressed in brain regions involved in the regulation of the stress response, such as the hippocampus, amygdala, dorsal raphe and locus coeruleus. Both genotypes exhibited similar basal levels of plasma corticosterone (CORT). Finally, single nucleotide polymorphisms (SNPs) in FKBP5 were found to be associated with alcohol drinking in humans. These results suggest that the association between FKBP5 and alcohol consumption is conserved in both mice and humans. Fkbp5 knockoutalcohol drinking behaviorhuman alcohol use disorder ==== Body 1. Introduction Alcohol use disorder displays a high level of comorbidity with several psychiatric disorders [1,2]. Dysregulation of a primary component of the stress response, the hypothalamic-pituitary-adrenal (HPA) axis, has been implicated in the pathophysiology of these psychiatric disorders [3,4,5,6] and mediates the transition from episodic drug and alcohol use to dependence [7,8,9,10]. Glucocorticoids, glucocorticoid receptor (GR) and its binding proteins (e.g., Fkbp5, FK506 binding protein 5, also known as FKBP51) are critical HPA axis regulatory elements. Diverse lines of research suggest a correlation between glucocorticoid levels and alcohol consumption. For instance, increases in HPA axis responsivity in young primates predicted higher levels of future alcohol consumption [11]. Alcohol consumption and withdrawal have both been shown to increase circulating glucocorticoids and to decrease GR availability [12], and the release of high levels of glucocorticoid peptides has been shown to sensitize the reward pathways in the brain [13,14,15]. Other research indicates that the GR plays an important role in the determination of alcohol abuse. Polymorphisms in the GR are associated with the onset of alcohol abuse in adolescents [6]; GR-mediated plasticity increased voluntary alcohol consumption in rats [16]; and a GR antagonist reduced alcohol intake in rats [17]. Thus, the GR is a worthy target of research aimed at identifying novel treatment strategies for alcohol use disorder. GR binding proteins are critical for GR function and GR-regulated behavior. However, limited research has been done to investigate the role of GR binding proteins in behaviors related to alcohol use disorder. FKBP5 is a GR binding protein that acts as a co-chaperone of heat shock protein 90 and is involved in regulating GR activity, nuclear translocation and transcriptional regulation of GR-targeted genes [18,19,20]. Functionally, FKBP5 is a potent inhibitor of GR activation and a determinant of HPA axis regulation [21]. As replicated in other studies, both Fkbp5 knockout (KO) mice and FKBP5 knockdown neuronal cell cultures exhibit elevated GR nuclear translocation [22,23]. High expression of Fkbp5 has been observed in the brain, especially, hippocampus [24], which is consistent with its function in the stress response. Increased hippocampal expression of Fkbp5 mRNA, as well as increased plasma corticosterone (CORT) levels and adrenal gland weight were observed after chronic social defeat in mice [25]. When compared to WT mice prior to chronic social defeat, Fkbp5-deficient mice exhibited lower basal CORT and lower adrenal weights [26]. Fkbp5 gene expression is not only induced by stress [27], but also by alcohol [28,29] and drug administration [30]. In addition, glucocorticoid treatment increased Fkbp5 expression in peripheral tissue [31] and the brain [32]. Consistent with the effects of alcohol on Fkbp5 gene expression in the CNS following acute alcohol injection, these findings suggest that increases in Fkbp5 expression following steroid receptor activation reduce GR sensitivity and, in turn, modulate GR-related behavior [28,33]. FKBP5 SNPs and gene expression levels are associated with the onset of posttraumatic stress disorder (PTSD) and anxiety disorder in humans [20,34]. Particularly, two SNPs, rs1360780 (C/T) and rs3800373 (T/G) are often used for association studies. In humans, alcohol use disorder is often comorbid with anxiety and other psychiatric disorders [2,35]. In rodents, anxiety is correlated with alcohol preference in various models of alcohol use disorder [36,37]. Genetic variations of FKBP5 are associated with an increased risk for depression [38,39], PTSD [40] and bipolar disorder [41] and are also associated with a greater risk for comorbid alcohol dependence and PTSD onset [42]. Moreover, FKBP5 SNPs are associated with the degree of cortisol response [20,21], response to antidepressants [38,39,40], heroin addiction [43] and alcohol withdrawal severity [5]. For example, heterozygous and homozygous carriers of the rs3800373 (T/G and G/G) or rs1360780 (C/T and T/T) variants are more likely to respond to antidepressant drugs [44]. Some findings suggest that the G allele of rs3800373 and the T allele of rs1360780 (minor alleles) may represent protective alleles for PTSD with a history of childhood abuse, while the major alleles, T allele of rs3800373 and C allele of rs1360780, represent risk alleles for comorbid alcohol dependence and PTSD onset [42]. However, correlations between the risk allele of FKBP5 and onset of PTSD have been inconsistent, and no direct research has investigated FKBP5 and its role in alcohol consumption. In the current study, we utilize the Fkbp5 KO mouse model to address the functional relevance of Fkbp5 in alcohol consumption and to map Fkbp5 gene expression in brain. Furthermore, we extend the mouse findings by determining associations between FKBP5 SNPs and alcohol drinking behaviors in humans, an effort critical for future translational research. Taken together, our findings are the first to demonstrate a role for Fkbp5 in the regulation of alcohol drinking in both mice and humans. 2. Results 2.1. Alcohol Consumption Is Increased in Fkbp5−/− Mice Studies suggest that alcohol consumption is directly affected by circulating corticoid and GR availability [12] and that FKBP5 plays a role in GR activation and HPA axis regulation [21]. In order to ascertain whether alcohol consumption was affected by Fkbp5 gene knockout, a series of drinking tests were performed in both male and female adult mice. In the established alcohol drinking test protocol, two indices were calculated including alcohol consumption (g EtOH/kg body weight/day) and alcohol preference (percentage of EtOH/total fluid, v/v). A significant genotypic effect was observed on male alcohol consumption using repeated measures two-way ANOVA with F (1,47) = 9.27, p = 0.0038. Sidak’s multiple comparisons post hoc test revealed significant differences at 9% (p = 0.0226), 12% (p = 0.0081) and 15% (p = 0.0002) (Figure 1A). However, while repeated measures two-way ANOVA revealed significant differences due to ethanol concentration (p < 0.0001), no genotypic effects were observed in female alcohol consumption or in ethanol preference of either sex (Figure 1B–D). During the drinking tests, the body weight was measured twice per week. No significant changes in body weight were detected within genotype, regardless of sex. When different concentrations of alcohol were presented with water, the total fluid intake (volume of alcohol plus water) of KO mice was lower than WT in general, but only in males, this difference found to be significant via repeated measures two-way ANOVA with F (1,47) = 11.07, p = 0.0017. Sidak’s multiple comparisons post hoc test revealed significant differences at 6% (p = 0.0043), 9% (p = 0.0495) and 15% (p = 0.0032) (Figure 1E,F). Taken together, the current results suggest that deficiency of Fkbp5 can enhance EtOH intake, at least in males. To exclude the possibility that taste sensitivity may have been influenced by Fkbp5 KO, animals were tested for quinine consumption (Figure 1G), but no significant difference was observed between KO and WT mice of either sex. Finally, we determined whether the observed differences in alcohol consumption might be due to differences in the rate of alcohol metabolism between genotypes. An ethanol dose of 2.0 g/kg body weight was injected intraperitoneally (IP), and the EtOH elimination rate was assessed. Regardless of sex, no significant difference in alcohol elimination rate was observed between Fkbp5 KO (1.2 ± 0.13 mg% EtOH per min) and WT mice (0.9 ± 0.23 mg% EtOH per min) (Figure 1H). 2.2. Blood Alcohol Concentration Is Higher in KO than WT Mice Blood alcohol concentration (BAC) was measured after 3 h of limited access to 15% EtOH. Fkbp5 KO mice consumed more alcohol than WT mice (mean = 2.24 g EtOH/kg body weight for KO vs. 0. 94 g/kg for WT), resulting in higher BACs for KO mice compared to WT mice (mean = 53 ± 1.3 mg% EtOH in KO vs. 25.9 ± 1.92 mg% EtOH in WT). 2.3. The Fkbp5 Gene Is Highly Expressed in Brain Regions Important for the Stress Response FKBP5 has been found to have reduced expression in individuals with PTSD [45]. Fkbp5-deficient mice were generated using a gene-trapping approach [46] in which the LacZ reporter gene was inserted. Previous studies have also shown that the murine Fkbp5 gene is expressed in various tissues, including brain and peripheral tissue. We therefore charted the Fkbp5 gene expression pattern in KO mice by staining for the LacZ gene product. In the brains of one-month-old Fkbp5 KO mice, LacZ staining revealed Fkbp5 expression in the hippocampus, striatum, dorsal raphe (DR) and locus coeruleus (LC) (Figure 2B). In four-month-old Fkbp5 KO mice, LacZ staining was observed in the hippocampus and additional brain regions, such as amygdala (Figure 2D). Whole brain LacZ staining followed by sectioning and eosin red staining in four-month-old KO mice revealed that Fkpb5 was highly expressed in the lateral septum, which includes lateral septal nuclei dorsal (LSD), ventral (LSV) and intermediate (LSI) (Figure 2F); bed nucleus of the striatum terminalis (BST) (Figure 2F); and hippocampus (Figure 2H,J). In the four-month-old hippocampus, LacZ staining suggested that Fkbp5 is highly expressed in CA1, CA2, CA3 and the dentate gyrus (DG) (Figure 2D,H,J). Consistent with the absence of the LacZ reporter gene, WT control mice did not exhibit any LacZ staining (Figure 2A,C,E,G,I). 2.4. Basal Corticosterone Is Not Different Between Fkbp5−/− and WT Mice Multiple lines of evidence demonstrate that Fkbp5 expression responds to CORT treatment [32,47]. Basal CORT was measured in blood samples of KO and WT mice to determine if there was a genotype and sex difference. No significant differences between KO and WT were detected by ANOVA (genotype × sex) (interaction term: F (1,21) = 3.4, p = 0.08) (Figure 3). However, there is trend of lower baseline CORT in female KO compared to WT. 2.5. SNPs in FKBP5 Are Associated with Alcohol Drinking Behavior in Humans Genotype frequency and allele frequency are shown in Table 1 split by ethnic group (including all 1162 participants regardless of drinking status). Genotype distributions were in Hardy-Weinberg equilibrium (a principle that the genetic variation in a population will remain constant from one generation to the next without disturbing factors) for all three ethnic groups (p > 0.40). Our results were consistent with previous findings that have found the two SNPs in the FKBP5 gene (rs3800373 in the 3’UTR and rs1360780 in intron 2) to be highly linked [38]. The two SNPs were linked among 92%–97% of all subjects studied (Table 1). Significant associations of both SNPs with alcohol measures were found in Chinese and Koreans, but not in Caucasians (Table 2) for raw means and standard deviations and regression statistics using transformed variables. For the rs1360780 SNP, Chinese with the CC genotype had significantly higher scores than those with CT/TT genotypes on average quantity, binge drinking episodes, lifetime maximum drinks and lifetime alcohol use disorder (AUD) symptoms, although lifetime maximum drinks was only significant in Chinese men (12.0 vs. 7.2, F = 9.32, p = 0.003, R2 change = 0.049). Koreans with the CC genotype also had significantly higher scores on AUD symptoms than Koreans with CT/TT genotypes. For the FKBP5 3’-UTR rs3800373 SNP, Chinese with the TT genotype had higher scores on lifetime maximum drinks and AUD symptoms than the TG/GG genotypes; again, lifetime maximum drinks was only significant in Chinese men (11.8 vs. 7.4, F = 13.37, p < 0.001, R2 change = 0.069). Koreans with the TT genotype also had a greater number of lifetime AUD symptoms than Koreans with TG/GG genotypes. No gene-binge drinking interaction terms were significant. 3. Discussion The current studies investigated the role that Fkbp5 plays in alcohol drinking in both mice and humans. Compared to WT mice, Fkbp5 KO mice exhibited increases in alcohol. These increases were not attributed to sensitivity to a bitter-tasting solution or to differences in the rate of alcohol metabolism. Relevant to its function in the stress response, Fkbp5 was found to be expressed in brain regions that are important in the stress response. Consistent with a previous study [26], basal differences in CORT levels were not observed between genotypes in male mice; although a trend toward lower basal CORT in female Fkbp5-deficient mice can be seen in Figure 3, consistent with those findings [26]. Finally, FKBP5 SNPs were associated with alcohol consumption in Asians, but not Caucasians. This is the first study to demonstrate a role for the Fkbp5 gene in regulating alcohol consumption in both mouse and human samples. 3.1. FKBP5 Is Associated with Psychiatric Disease Including Alcohol Use Disorder Previous research has focused on FKBP5 and its association with various psychiatric disorders [21,41,48,49]. This study demonstrated that the gene expression of Fkbp5 in the mouse plays at least a partial role in the regulation of alcohol drinking; KO mice with no expression of Fkbp5 consumed more alcohol. Even though we do not have RNA available for FKBP5 gene expression measurement in humans, the results from previous research also support the hypothesis that low FKBP5 expression is associated with greater alcohol consumption. Previous research has shown that rs1360780 (CC) homozygous subjects exhibit decreased FKBP5 mRNA expression compared to (TT) homozygous subjects [38]. Consistent with the assumption that subjects with CC genotypes also display lower FKBP5 mRNA expression than subjects with CT or TT genotypes, we found that Chinese and Koreans with CC had higher scores on alcohol-related variables than CT/TT carriers. FKBP5 SNPs and gene expression levels are associated with the onset of PTSD and anxiety disorder in humans [20,34]. These SNPs are also associated with altered response to antidepressants [39,40,41]. Particularly, heterozygous and homozygous carriers of the rs3800373 (TG and GG) or rs1360780 (CT and TT) variants are more likely to respond to antidepressant drugs [44]. A recent study also indicated that aged (>50 years) T allele carriers of rs1360780 showed significantly higher induction of FKBP5 mRNA expression by glucocorticoids in peripheral blood mononuclear cells [50]. Additionally, the effect of increasing severity of childhood abuse on the resultant level of adult PTSD appears to be carried by a subset of subjects with more common alleles of FKBP5 [34]. These findings suggest that the G allele of rs3800373 and the T allele of rs1360780 (minor alleles) may represent protective alleles for PTSD with a history of childhood abuse, while the major alleles, the T allele of rs3800373 and the C allele of rs1360780, represent risk alleles. In agreement with previous studies, these major alleles are associated with a higher risk for comorbid alcohol dependence and PTSD onset [42]. In our prior study, alcohol-dependent inpatients with the T allele of rs3800373 had more severe withdrawal symptoms [5]. In this study, this major allele is associated with higher scores on alcohol-related variables. Interestingly, the rs1360780 SNP is within an intronic region; a homologous region in rodent Fkbp5 is highly conserved and has a functional hormone response element [51]. We speculate that rs1360780 may be a functional SNP that regulates gene expression, which in turn alters alcohol drinking. In summary, the major alleles of FKBP5 SNPs are associated with higher levels of alcohol consumption, more frequent consumption and a greater likelihood for alcohol-related problems in this study, as well as with withdrawal symptoms in prior studies [5]. In the future, genotyping FKBP5 SNPs in heavy drinkers and the general population may explain the previous finding, which demonstrated that alcohol-dependent subjects had more withdrawal symptoms, higher alcohol intake and a higher maximum number of drinks compared to the general population [52]. FKBP5 can be used as a biomarker for the diagnosis of alcohol use disorder. This study demonstrates ethnic differences in the association of FKBP5 SNPs with alcohol-related phenotypes. In Korean and Chinese individuals, significant associations were found between FKBP5 SNPs and alcohol drinking-related variables, but no FKBP5 gene effects were detected in Caucasian participants. Previous research has found a combinatorial effect of the FKBP5 gene and childhood abuse on the risk for developing PTSD in African Americans, but not in European Americans [42]. Results of this study also indicated consistent findings across gender within each ethnicity with the exception of the association of FKBP5 SNPs with lifetime maximum number of drinks in Chinese, which only reached significance in the men. This lack of significance in the Chinese female sample may be due to statistical power given the lower levels of maximum drinks in the women overall, with the trends being similar in both genders. In addition, our analyses did not indicate that recent heavier drinking was associated with a differential effect of the genes on lifetime heavy drinking or problems, further indicating the consistency of the findings within each ethnic group despite differences between the ethnic groups. Future research should continue to investigate gender and ethnic group differences in the associations of these SNPs with alcohol-related behaviors to better understand their relationships in those with different allele prevalence and consumption patterns, including those with more severe alcohol-related problems. 3.2. No Basal Difference in CORT Level, but Fkbp5 Expressed in the Brain Regions Involving Stress Response Previous studies have indicated a correlation between alcohol consumption, glucocorticoid levels and GR activity [4,14,53]. Animal studies show that circulating levels of CORT rise after stress in Fkbp5 KO mice [54] and prenatally stressed rats [27]. Consistent with a previous study [26], we did not find basal differences in CORT levels between genotypes in male mice; although, a trend toward lower basal CORT in female Fkbp5-deficient mice can be seen in Figure 3. Based on our research and that of others, it is clear that eliminating Fkbp5 does not affect basal CORT. Given the tantalizing result that Fkbp5 is expressed in the brain regions that are relevant to stress response, it will be very interesting to study how stress affects CORT levels in Fkbp5 KO mice. In the periphery, as well as the CNS, Fkbp5 plays an important role in the stress response. Research has shown Fkbp5 gene expression in the adult brain [24,55] and in mice as young as three months [56]. The current study revealed Fkbp5 expression in the brain of one-month-old mice. Brain regions with high expression of Fkbp5 have been implicated in both the stress response and the development of alcohol dependence [57,58]. Fkbp5 expression is induced in the brain by stress [24,27], alcohol [28,29] and other drugs [30]. Our ongoing studies aim to understand how stress and alcohol affect Fkbp5 expression in the brain and circulating CORT levels. The limitation of this research is that because Fkbp5 is a stress response gene, other stressors could be confounders; for example, we did not include smoking as a covariate in our analyses and only assessed current consumption patterns and lifetime alcohol problems in these analyses. 4. Materials and Methods 4.1. Animal and Human Subjects All experimental protocols were reviewed and approved by the Animal Care and Use Committees at the Indiana University School of Medicine (protocol #DS0000871R, date of approval 4/24/2013), Purdue University (#1112000327, 10/11/2013), and the Institute of Laboratory Animal Science of Peking Union Medical College (#ILAS-PG-2014-013). These protocols were carried out in accordance with the NIH Guide for the Care and Use of Laboratory Animals. As described in a previous paper, Fkbp5 knockout (Fkbp5−/−) mice were generated using the gene trapping method [46]. Fkbp5−/− and WT littermates were bred through heterozygous mating and were backcrossed with C57BL/6J inbred mice for at least 5 generations. The human study (#041122, 9/2/2004; #080978, 5/22/2008; #100399, 2/25/2010) was approved by the University of California, San Diego (UCSD) Human Research Protections Program and had a Certificate of Confidentiality from the USA Department of Health and Human Services. UCSD college students of legal drinking age were recruited to participate through advertisements on campus. All participants provided written informed consent. 4.2. Alcohol Drinking Tests Drinking experiments were repeated three times using different cohorts of animals, in all a total of 82 Fkbp5−/− (male, n = 46; female, n = 36) and 92 WT (male, n = 52; female, n = 40) mice. Fluid intake and body weight were measured every other day throughout each experiment. The average body weights of the mice were WT = 30 g and KO = 27 g for males and WT = 24 g and KO = 22 g for females. At 13 weeks of age, mice were habituated to drinking in their home cages for 6 days with two 25-mL graduated cylinders containing water. Following the habituation phase, mice were given 24-h access to water and alcohol (EtOH). The concentration of EtOH (v/v) was increased every four days as follows: 3%, 6%, 9%, 12% and 15%. Average alcohol consumption per day was calculated and corrected for individual differences in body weight (BW) (g EtOH/kg BW/day). Alcohol preference ratios were then calculated (EtOH/total fluid). To exclude the possibility that KO and WT mice differed in taste reactivity, animals were tested for quinine (0.5 µM) intake as published [59]. 4.3. Alcohol Elimination Rate At 12 weeks of age, the EtOH elimination rate was determined using a separate cohort of Fkbp5−/− and WT mice (4 male and 4 female mice of each genotype). An EtOH dose of 2.0 g/kg was injected intraperitoneally (IP). Tail blood was collected at 30, 75 and 120 min after IP injection, and plasma samples were spun down and stored at −80 °C until further analysis [60]. Plasma alcohol concentrations were measured using gas chromatography according to the instructions of the manufacturer (HP Agilent, Santa Clara, CA, USA). 4.4. Blood Alcohol Concentration after 3-h Limited Access Mice (10 Fkbp5−/− and 10 WT mice of both sexes at 6 months of age) were maintained at reverse light cycle for 3 weeks before the experiment. The subjective dark period is the active time for rodents. In order to assess the blood EtOH levels achieved during the active period, intake volume of a 15% EtOH solution and subsequent blood EtOH concentrations were determined after a 3-h access period, occurring 3 h into the dark cycle. On the test day, a stacking design was used with 10-min intervals separating the presentation of EtOH to each mouse. Beginning at 7:00 a.m., the first animal received 15% EtOH, and tail blood was taken at 10:00 a.m., with the rest of the mice following in sequence at 10-min intervals. Blood was collected in an EDTA-coated tube and stored on ice until all samples were collected. After collection, blood samples were centrifuged at 1500× g for 10 min, and plasma was isolated for blood alcohol concentration measurement. The same GC platform was used as described above. 4.5. LacZ Staining Age-matched Fkbp5−/− (n = 6) and littermate control (n = 6) mice of both sexes were used for morphological and histological studies. Fkbp5 gene expression in the whole brain was determined at 1 and 4 months of age. Whole brains were collected, and the LacZ gene product, β-galactosidase activity, was detected using X-gal [48,61]. After images were collected, the whole brain or brain slices were fixed in 10% neutral-buffered formalin. Brains were paraffin embedded, sectioned (5 µm) and stained with hematoxylin and eosin following standard protocols [62,63]. 4.6. CORT Measurement Basal corticosterone (CORT) levels were measured in mice. Blood samples were collected from the submandibular vein (male: KO = 8, WT = 5; female: KO = 7, WT = 5) between 1200 and 1300 (light cycle on at 700–1900). The concentration of CORT in the blood was determined using a competitive enzyme immunoassay kit from Assay Designs (Ann Arbor, MI, USA), as previously published [64]. Optical densities were read on a MultiskanTM FC microplate reader (ThermoFisher Scientific Inc., Waltham, MA, USA). Standards and plasma samples were analyzed in duplicate. CORT concentrations were interpolated from standard curves generated using a four-parameter logistic curve fitting program (GraphPad Inc., San Diego, CA, USA). 4.7. Human Sample Phenotypes Participants were students at UCSD (n = 1162) 21–26 years of age (M = 22.0, SD = 1.34), with an average of 15.1 years of education (SD = 0.89), who reported that all 4 of their grandparents were entirely of Chinese (n = 360, 48% female), Korean (n = 343, 50% female) or Caucasian (n = 449, 48% female) ethnicity. Participants completed the time-line follow-back measure to evaluate alcohol consumption for the preceding 90 days [65,66]. A standard drink was defined as 12 oz (355 mL) of beer, 5 oz (150 mL) of wine or 1.5 oz (45 mL) of hard liquor. These amounts are the equivalent of approximately 14 g of pure ethanol. Participants also completed the Semi-Structured Assessment for the Genetics of Alcohol use disorder [67,68] with a trained research interviewer. During a portion of this interview, participants recounted the maximum number of drinks ever consumed in a 24-h period and were assessed for the 11 DSM-IV alcohol abuse and dependence symptoms (range 0–11; American Psychological Association, 1994) [69]. Five alcohol-related variables were calculated from these assessments: (1) average quantity of drinking during the previous three months (standard drinks/occasion); (2) average frequency of drinking during the previous three months (days/month); (3) number of binge drinking episodes (four or more drinks on an occasion for women and five or more on an occasion for men) during the previous three months; (4) lifetime maximum number of alcohol drinks ever consumed in a 24-h period; and (5) number of lifetime alcohol use disorder (AUD) symptoms (alcohol abuse and dependence). Lifetime non-drinkers (defined as never having had a standard drink of alcohol) were excluded (n = 30). 4.8. Human Sample Genotyping Blood samples were collected by fingertip puncture and delivered to the Genomics and Bioinformatics Core of Indiana Alcohol Research Center for genotyping. Genomic DNA was isolated by the “HotSHOT” method [70], and TaqMan probes were used for allelic discrimination (Life Technologies, Foster City, CA, USA). Genotyping procedures were reported previously [71]. Two SNPs, rs1360780 (major/minor allele, C/T) within intron 2 and rs3800373 (T/G) in the 3’UTR, were selected for genotyping. The two SNPs selected comprise one single large linkage disequilibrium (LD) block based on previous LD structure analysis [38]. 4.9. Statistical Analysis Total fluid intake and alcohol consumption and preference differences at tested EtOH concentrations between genotypes were assessed via 2-way repeated measures analysis of variance (ANOVA) with Sidak’s test for multiple comparisons. Differences in quinine intake and CORT data were analyzed using Student’s t-test. In humans, data were analyzed using linear regressions. All alcohol variables were log transformed to account for the non-normality of the distribution of raw scores. Based on the genotypic distributions (Table 1) and consistent with prior studies [44], we dichotomized the genotype groups, such that CT and TT were combined for the FKBP5 intron SNP (rs1360780) and TG and GG were combined for the FKBP5 3’-UTR SNP (rs3800373). Because of the strong association of alcohol metabolizing genes (e.g., ALDH2, ADH1B) with alcohol-related behaviors [72], we co-varied for these two genotypes (dichotomized variables for those with and without the variant ALDH2*2 and ADH1B*2 alleles) by entering them in linear regressions as the first step prior to entering the FKBP5 gene in the second step. Note that the ALDH2*2 allele was present only in the Chinese and Korean groups, whereas the ADH1B*2 was present in all three ethnic groups. We tested for gender differences in the relationship of the FKBP5 SNPs with alcohol-related behaviors by including gene-gender interaction terms in the regression models. For any interaction term with a p < 0.10, we then examined the relationship of the gene with the alcohol-related variable in each gender separately. We used a similar approach to examine the possibility that current heavy alcohol consumption might differentially affect the association of the genes with lifetime heavy drinking and AUD symptoms; in these analyses, we included interaction terms of the genes with having binged in the past 3 months. 5. Conclusions Despite these limitations, our study demonstrates that FKBP5 is associated with alcohol consumption phenotypes in mice and in humans of Asian, but not Caucasian, descent. The current study has established the foundation for future studies investigating the potential role of FKBP5 in responses to acute and chronic drinking, alcohol withdrawal symptoms and the interplay of stress and Fkbp5 expression on the development of alcohol dependence. FKBP5 may well be an interesting therapeutic target for the prevention and treatment of stress-related alcohol drinking behavior. Acknowledgments This research was supported by grants from the state high-tech program (863-2012AA022403), National Key Basic Research Program of China (2013CB945000), the National Institute on Alcohol Abuse and Alcohol use disorder (NIAAA) Grants R01AA10707, P60AA007611, R01AA11257, R01AA18179, and internal funding from the Indiana University, School of Medicine. We also would like to express our appreciation to Judy E. Powers, Hanying Chen and Tamara J. Graves for their technical support. Author Contributions Bin Qiu, Weidong Yong, and Tiebing Liang contributed to the experimental design, alcohol-related data collection and data analysis, and manuscript submission. Susan E. Luczak, Tamara L. Wall, and Mimy Y. Eng carried out human subject recruitment, data collection and data analysis, and manuscript preparation. Aaron M. Kirchhoff, Stephen L. Boehm II, and Julia A. Chester performed CORT measurement, data analysis, and manuscript preparation. Robert B. Stewart performed animal behavioral testing. Weinian Shou, Yuxue Xu, Weidong Yong, and Tiebing Liang carried out the creation and maintenance of KO animals and performed gene expression. Conflicts of Interest The authors declare no conflict of interest. Figure 1 EtOH consumption and preference, quinine consumption and EtOH metabolism in Fkbp5 KO and WT mice. A significant increase in alcohol consumption (A) in male mice was observed in the KO mice at 9%, 12% and 15% EtOH concentrations when compared to WT mice. No significant increases in alcohol consumption (C) in female mice or preference (B,D) in either sex were observed. Lower total fluid consumption was observed (E) in male mice, but not (F) in female mice during the EtOH consumption test. KO and WT mice did not differ in quinine consumption (G). No differences in alcohol metabolism were found between KO and WT mice (H). *, p < 0.05, **, p < 0.01, ***, p < 0.001. Figure 2 FKBP5 expression in the brain of WT and KO mice. Whole brain staining using the LacZ reporter gene. For each pairing of photomicrographs, the left panel is the WT control sample and the right panel is the KO sample. Fkbp5 gene expression was observed in the brain regions of mice at one month (A,B) and four months (C–J) of age. Whole brain staining for LacZ in fresh tissue was performed (A–D) followed by hematoxylin and eosin staining in the four-month-old sample (E–J). Hippo (hippocampus), DR (dorsal raphe nucleus), LC (locus coeruleus), DLG (dorsal lateral geniculate nucleus (nu)), Rt (reticular thalamic (nu)), PH (posterior hypothalamic area), Amg (amygdala), Pir (piriform cortex), LSD (lateral septal nucleus, dorsal part), LSI (intermediate part), LSV (ventral part), BST (bed nucleus of the stria terminalis), CA (field CA of Ammon’s horn) and DG (dentate gyrus). Scale bars (A,B) = 2 mm, scale bars (C,D) = 1.5 mm, scale bars (E–J) = 1 mm. Figure 3 Basal serum corticosterone (CORT) levels in male (A) and female (B) WT and KO mice. No significant difference was found between genotypes. ijms-17-01271-t001_Table 1Table 1 FKBP5 genotype distributions and Hardy–Weinberg equilibrium (HWE) values for each ethnic group. The chi-square statistic (X2) is used to test if the allele frequencies are in HWE for each ethnic group, that is they are consistent with the expected distribution for the general population. The significance tests (p-values), all being >0.05, show that the alleles are in HWE, indicating no deviation from the expected distribution of alleles in the population and no bias in our samples. SNP FKBP5 Intron (rs1360780) FKBP5 3’-UTR (rs3800373) Genotype Distribution HWE Genotype Distribution HWE Ethnicity CC CT TT X2 p TT TG GG X2 p Chinese 190 (53%) 141 (39%) 29 (8%) 0.2 0.69 192 (53%) 138 (38%) 29 (8%) 0.4 0.55 Korean 195 (57%) 131 (38%) 17 (5%) 0.7 0.40 205 (60%) 123 (36%) 15 (4%) 0.4 0.52 Caucasian 226 (50%) 189 (42%) 34 (8%) 0.4 0.52 240 (46%) 175 (33%) 33 (6%) 0.0 0.89 ijms-17-01271-t002_Table 2Table 2 Alcohol-related variables in Chinese, Korean and Caucasian college students for two FKBP5 genotypes after co-varying for ALDH2*2 and ADH1B*2. SNP FKBP5 Intron (rs1360780) FKBP51 UTR (rs3800373) Genotype CC CT/TT – – – TT TG/GG – – – Statistic M (SD) M (SD) F p Change in R2 M (SD) M (SD) F p Change in R2 Chinese n = 180 n = 160 – – – n = 183 n = 157 – – – Quantity 3.8 (6.02) 2.9 (1.52) 2.13 0.034 0.013 3.3 (2.11) 2.9 (1.53) 1.39 0.164 0.005 Frequency 10.9 (11.31) 9.7 (9.61) 0.69 0.491 0.001 10.7 (11.30) 9.8 (9.64) 0.34 0.731 0.001 Binges 3.3 (5.20) 2.8 (4.70) 1.96 0.051 0.010 3.2 (5.17) 2.8 (4.74) 1.60 0.110 0.007 Max drinks 9.1 (8.54) 6.3 (5.21) 3.69 <0.001 0.037 8.7 (8.46) 6.4 (5.21) 2.84 0.005 0.022 AUD symptoms 1.4 (1.00) 1.2 (0.65) 2.33 0.021 0.015 1.4 (0.97) 1.2 (0.65) 1.99 0.048 0.012 Korean n = 188 n = 146 – – – n = 198 n = 136 – – – Quantity 4.2 (2.48) 3.8 (2.21) 1.18 0.238 0.004 4.1 (2.44) 3.9 (2.20) 0.33 0.741 0.001 Frequency 15.1 (14.89) 12.2 (11.99) 1.63 0.105 0.008 14.8 (15.01) 12.5 (11.97) 0.89 0.374 0.002 Binges 6.1 (8.17) 5.2 (7.88) 1.37 0.172 0.006 5.8 (7.99) 5.4 (8.03) 0.58 0.560 0.001 Max drinks 12.1 (9.89) 10.4 (7.32) 1.57 0.118 0.007 11.7 (9.73) 10.7 (7.20) 0.52 0.603 0.000 AUD symptoms 2.1 (1.77) 1.7 (1.29) 2.40 0.017 0.017 2.0 (1.72) 1.7 (1.31) 1.95 0.052 0.011 Caucasian n = 217 n = 220 – – – n = 203 n = 235 – – – Quantity 3.1 (1.86) 3.3 (2.24) 0.21 0.65 0.000 3.2 (1.98) 3.2 (2.20) 0.00 0.98 0.000 Frequency 24.5 (18.80) 24.2 (17.19) 0.04 0.84 0.000 24.2 (18.40) 24.1 (17.42) 0.08 0.77 0.000 Binges 8.2 (12.16) 8.0 (10.51) 0.20 0.65 0.000 8.2 (12.28) 8.0 (10.63) 0.05 0.82 0.000 Max drinks 13.9 (10.13) 14.2 (11.83) 0.23 0.63 0.001 14.0 (10.35) 13.9 (11.42) 0.06 0.81 0.000 AUD symptoms 1.0 (1.71) 1.0 (1.61) 0.10 0.76 0.000 1.0 (1.75) 1.0 (1.58) 0.63 0.43 0.001 Raw scores reported for means and standard deviations. In the analyses, alcohol variables were log transformed to adjust for right-tailed distributions. ALDH2*2 and ADH1B*2 were covaried in all analyses, but no Caucasian possessed an ALDH2*2 allele. AUD, alcohol use disorder. Bolded p-value indicated statistically significant at p < 0.05. ==== Refs References 1. Kessler R.C. Nelson C.B. McGonagle K.A. Edlund M.J. Frank R.G. Leaf P.J. The epidemiology of co-occurring addictive and mental disorders: Implications for prevention and service utilization Am. J. Orthopsychiatry 1996 66 17 31 10.1037/h0080151 8720638 2. Hasin D.S. Grant B.F. The national epidemiologic survey on alcohol and related conditions (NESARC) Waves 1 and 2: Review and summary of findings Soc. Psychiatry Psychiatr. Epidemiol. 2015 50 1609 1640 10.1007/s00127-015-1088-0 26210739 3. Kellendonk C. Gass P. Kretz O. Schutz G. Tronche F. Corticosteroid receptors in the brain: Gene targeting studies Brain Res. Bull. 2002 57 73 83 10.1016/S0361-9230(01)00638-4 11827739 4. Roy A. Mittal N. Zhang H. Pandey S.C. Modulation of cellular expression of glucocorticoid receptor and glucocorticoid response element-DNA binding in rat brain during alcohol drinking and withdrawal J. Pharmacol. Exp. Ther. 2002 301 774 784 10.1124/jpet.301.2.774 11961084 5. Huang M.C. Schwandt M.L. Chester J.A. Kirchhoff A.M. Kao C.F. Liang T. Tapocik J.D. Ramchandani V.A. George D.T. Hodgkinson C.A. FKBP5 moderates alcohol withdrawal severity: Human genetic association and functional validation in knockout mice Neuropsychopharmacology 2014 39 2029 2038 10.1038/npp.2014.55 24603855 6. Desrivieres S. Lourdusamy A. Muller C. Ducci F. Wong C.P. Kaakinen M. Pouta A. Hartikainen A.L. Isohanni M. Charoen P. Glucocorticoid receptor (NR3C1) gene polymorphisms and onset of alcohol abuse in adolescents Addict. Biol. 2011 16 510 513 10.1111/j.1369-1600.2010.00239.x 20731635 7. Robinson T.E. Berridge K.C. The neural basis of drug craving: An incentive-sensitization theory of addiction Brain Res. Rev. 1993 18 247 291 10.1016/0165-0173(93)90013-P 8401595 8. Jose B.S. van Oers H.A. van de Mheen H.D. Garretsen H.F. Mackenbach J.P. Stressors and alcohol consumption Alcohol 2000 35 307 312 10.1093/alcalc/35.3.307 9. Koob G.F. Ahmed S.H. Boutrel B. Chen S.A. Kenny P.J. Markou A. O’Dell L.E. Parsons L.H. Sanna P.P. Neurobiological mechanisms in the transition from drug use to drug dependence Neurosci. Biobehav. Rev. 2004 27 739 749 10.1016/j.neubiorev.2003.11.007 15019424 10. Koob G.F. Le Moal M. Drug abuse: Hedonic homeostatic dysregulation Science 1997 278 52 58 10.1126/science.278.5335.52 9311926 11. Fahlke C. Lorenz J.G. Long J. Champoux M. Suomi S.J. Higley J.D. Rearing experiences and stress-induced plasma cortisol as early risk factors for excessive alcohol consumption in nonhuman primates Alcohol. Clin. Exp. Res. 2000 24 644 650 10.1111/j.1530-0277.2000.tb02035.x 10832905 12. Errico A.L. King A.C. Lovallo W.R. Parsons O.A. Cortisol dysregulation and cognitive impairment in abstinent male alcoholics Alcohol. Clin. Exp. Res. 2002 26 1198 1204 10.1111/j.1530-0277.2002.tb02656.x 12198394 13. Rouge-Pont F. Deroche V. Le Moal M. Piazza P.V. Individual differences in stress-induced dopamine release in the nucleus accumbens are influenced by corticosterone Eur. J. Neurosci. 1998 10 3903 3907 10.1046/j.1460-9568.1998.00438.x 9875367 14. Adinoff B. Iranmanesh A. Veldhuis J. Fisher L. Disturbances of the stress response: The role of the HPA axis during alcohol withdrawal and abstinence Alcohol. Health Res. World 1998 22 67 72 15706736 15. Nader K. Schafe G.E. Le Doux J.E. Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval Nature 2000 406 722 726 10.1038/35021052 10963596 16. Vendruscolo L.F. Barbier E. Schlosburg J.E. Misra K.K. Whitfield T.W. Logrip M.L. Rivier C. Repunte-Canonigo V. Zorrilla E.P. Sanna P.P. Corticosteroid-dependent plasticity mediates compulsive alcohol drinking in rats J. Neurosci. 2012 32 7563 7571 10.1523/JNEUROSCI.0069-12.2012 22649234 17. Koenig H.N. Olive M.F. The glucocorticoid receptor antagonist mifepristone reduces ethanol intake in rats under limited access conditions Psychoneuroendocrinology 2004 29 999 1003 10.1016/j.psyneuen.2003.09.004 15219650 18. Wochnik G.M. Ruegg J. Abel G.A. Schmidt U. Holsboer F. Rein T. Fk506-binding proteins 51 and 52 differentially regulate dynein interaction and nuclear translocation of the glucocorticoid receptor in mammalian cells J. Biol. Chem. 2005 280 4609 4616 10.1074/jbc.M407498200 15591061 19. Schiene C. Fischer G. Enzymes that catalyse the restructuring of proteins Curr. Opin. Struct. Biol. 2000 10 40 45 10.1016/S0959-440X(99)00046-9 10679456 20. Binder E.B. The role of FKBP5, a co-chaperone of the glucocorticoid receptor in the pathogenesis and therapy of affective and anxiety disorders Psychoneuroendocrinology 2009 34 Suppl. 1 S186 S195 10.1016/j.psyneuen.2009.05.021 19560279 21. Ising M. Depping A.M. Siebertz A. Lucae S. Unschuld P.G. Kloiber S. Horstmann S. Uhr M. Muller-Myhsok B. Holsboer F. Polymorphisms in the fkbp5 gene region modulate recovery from psychosocial stress in healthy controls Eur. J. Neurosci. 2008 28 389 398 10.1111/j.1460-9568.2008.06332.x 18702710 22. Tatro E.T. Everall I.P. Kaul M. Achim C.L. Modulation of glucocorticoid receptor nuclear translocation in neurons by immunophilins Fkbp51 and Fkbp52: Implications for major depressive disorder Brain Res. 2009 1286 1 12 10.1016/j.brainres.2009.06.036 19545546 23. Banerjee A. Periyasamy S. Wolf I.M. Hinds T.D. Yong W. Shou W. Sanchez E.R. Control of glucocorticoid and progesterone receptor subcellular localization by the ligand-binding domain is mediated by distinct interactions with tetratricopeptide repeat proteins Biochemistry 2008 47 10471 10480 10.1021/bi8011862 18771283 24. Scharf S.H. Liebl C. Binder E.B. Schmidt M.V. Muller M.B. Expression and regulation of the fkbp5 gene in the adult mouse brain PLoS ONE 2011 6 1271 10.1371/journal.pone.0016883 21347384 25. Wagner K.V. Marinescu D. Hartmann J. Wang X.D. Labermaier C. Scharf S.H. Liebl C. Uhr M. Holsboer F. Muller M.B. Differences in Fkbp51 regulation following chronic social defeat stress correlate with individual stress sensitivity: Influence of paroxetine treatment Neuropsychopharmacology 2012 37 2797 2808 10.1038/npp.2012.150 22871917 26. Hartmann J. Wagner K.V. Liebl C. Scharf S.H. Wang X.D. Wolf M. Hausch F. Rein T. Schmidt U. Touma C. The involvement of FK506-binding protein 51 (FKBP5) in the behavioral and neuroendocrine effects of chronic social defeat stress Neuropharmacology 2012 62 332 339 10.1016/j.neuropharm.2011.07.041 21839098 27. Szymanska M. Budziszewska B. Jaworska-Feil L. Basta-Kaim A. Kubera M. Leskiewicz M. Regulska M. Lason W. The effect of antidepressant drugs on the HPA axis activity, glucocorticoid receptor level and Fkbp51 concentration in prenatally stressed rats Psychoneuroendocrinology 2009 34 822 832 10.1016/j.psyneuen.2008.12.012 19195790 28. Kerns R.T. Ravindranathan A. Hassan S. Cage M.P. York T. Sikela J.M. Williams R.W. Miles M.F. Ethanol-responsive brain region expression networks: Implications for behavioral responses to acute ethanol in DBA/2J versus C57B1/6J mice J. Neurosci. 2005 25 2255 2266 10.1523/JNEUROSCI.4372-04.2005 15745951 29. Treadwell J.A. Singh S.M. Microarray analysis of mouse brain gene expression following acute ethanol treatment Neurochem. Res. 2004 29 357 369 10.1023/B:NERE.0000013738.06437.a6 15002731 30. McClung C.A. Nestler E.J. Zachariou V. Regulation of gene expression by chronic morphine and morphine withdrawal in the locus ceruleus and ventral tegmental area J. Neurosci. 2005 25 6005 6015 10.1523/JNEUROSCI.0062-05.2005 15976090 31. Vermeer H. Hendriks-Stegeman B.I. van der Burg B. van Buul-Offers S.C. Jansen M. Glucocorticoid-induced increase in lymphocytic Fkbp51 messenger ribonucleic acid expression: A potential marker for glucocorticoid sensitivity, potency, and bioavailability J. Clin. Endocrinol. Metab. 2003 88 277 284 10.1210/jc.2002-020354 12519866 32. Lee R.S. Tamashiro K.L. Yang X. Purcell R.H. Harvey A. Willour V.L. Huo Y. Rongione M. Wand G.S. Potash J.B. Chronic corticosterone exposure increases expression and decreases deoxyribonucleic acid methylation of fkbp5 in mice Endocrinology 2010 151 4332 4343 10.1210/en.2010-0225 20668026 33. McQuillin A. Rizig M. Gurling H.M. A microarray gene expression study of the molecular pharmacology of lithium carbonate on mouse brain mrna to understand the neurobiology of mood stabilization and treatment of bipolar affective disorder Pharmacogenet. Genom. 2007 17 605 617 10.1097/FPC.0b013e328011b5b2 17622937 34. Binder E.B. Bradley R.G. Liu W. Epstein M.P. Deveau T.C. Mercer K.B. Tang Y. Gillespie C.F. Heim C.M. Nemeroff C.B. Association of fkbp5 polymorphisms and childhood abuse with risk of posttraumatic stress disorder symptoms in adults J. Am. Med. Assoc. 2008 299 1291 1305 10.1001/jama.299.11.1291 18349090 35. Pacek L.R. Storr C.L. Mojtabai R. Green K.M. La Flair L.N. Alvanzo A.A. Cullen B.A. Crum R.M. Comorbid alcohol dependence and anxiety disorders: A national survey J. Dual. Diagn. 2013 9 271 280 10.1080/15504263.2013.835164 24223532 36. Izidio G.S. Ramos A. Positive association between ethanol consumption and anxiety-related behaviors in two selected rat lines Alcohol 2007 41 517 524 10.1016/j.alcohol.2007.07.008 17980788 37. Barrenha G.D. Chester J.A. Genetic correlation between innate alcohol preference and fear-potentiated startle in selected mouse lines Alcohol. Clin. Exp. Res. 2007 31 1081 1088 10.1111/j.1530-0277.2007.00396.x 17451404 38. Binder E.B. Salyakina D. Lichtner P. Wochnik G.M. Ising M. Putz B. Papiol S. Seaman S. Lucae S. Kohli M.A. Polymorphisms in fkbp5 are associated with increased recurrence of depressive episodes and rapid response to antidepressant treatment Nat. Genet. 2004 36 1319 1325 10.1038/ng1479 15565110 39. Lekman M. Laje G. Charney D. Rush A.J. Wilson A.F. Sorant A.J. Lipsky R. Wisniewski S.R. Manji H. McMahon F.J. The FKBP5 -gene in depression and treatment response—An association study in the sequenced treatment alternatives to relieve depression (STAR*D) cohort Biol. Psychiatry 2008 63 1103 1110 10.1016/j.biopsych.2007.10.026 18191112 40. Koenen K.C. Saxe G. Purcell S. Smoller J.W. Bartholomew D. Miller A. Hall E. Kaplow J. Bosquet M. Moulton S. Polymorphisms in fkbp5 are associated with peritraumatic dissociation in medically injured children Mol. Psychiatry 2005 10 1058 1059 10.1038/sj.mp.4001727 16088328 41. Willour V.L. Chen H. Toolan J. Belmonte P. Cutler D.J. Goes F.S. Zandi P.P. Lee R.S. MacKinnon D.F. Mondimore F.M. Family-based association of fkbp5 in bipolar disorder Mol. Psychiatry 2009 14 261 268 10.1038/sj.mp.4002141 18180755 42. Xie P. Kranzler H.R. Poling J. Stein M.B. Anton R.F. Farrer L.A. Gelernter J. Interaction of fkbp5 with childhood adversity on risk for post-traumatic stress disorder Neuropsychopharmacology 2010 35 1684 1692 10.1038/npp.2010.37 20393453 43. Levran O. Peles E. Randesi M. Li Y. Rotrosen J. Ott J. Adelson M. Kreek M.J. Stress-related genes and heroin addiction: A role for a functional fkbp5 haplotype Psychoneuroendocrinology 2014 45 67 76 10.1016/j.psyneuen.2014.03.017 24845178 44. Kirchheiner J. Lorch R. Lebedeva E. Seeringer A. Roots I. Sasse J. Brockmoller J. Genetic variants in fkbp5 affecting response to antidepressant drug treatment Pharmacogenomics 2008 9 841 846 10.2217/14622416.9.7.841 18597649 45. Yehuda R. Cai G. Golier J.A. Sarapas C. Galea S. Ising M. Rein T. Schmeidler J. Muller-Myhsok B. Holsboer F. Gene expression patterns associated with posttraumatic stress disorder following exposure to the world trade center attacks Biol. Psychiatry 2009 66 708 711 10.1016/j.biopsych.2009.02.034 19393990 46. Yong W. Yang Z. Periyasamy S. Chen H. Yucel S. Li W. Lin L.Y. Wolf I.M. Cohn M.J. Baskin L.S. Essential role for co-chaperone Fkbp52 but not Fkbp51 in androgen receptor-mediated signaling and physiology J. Biol. Chem. 2007 282 5026 5036 10.1074/jbc.M609360200 17142810 47. Malviya S.A. Kelly S.D. Greenlee M.M. Eaton D.C. Duke B.J. Bourke C.H. Neigh G.N. Estradiol stimulates an anti-translocation expression pattern of glucocorticoid co-regulators in a hippocampal cell model Physiol. Behav. 2013 122 187 192 10.1016/j.physbeh.2013.03.018 23541378 48. Tatro E.T. Everall I.P. Masliah E. Hult B.J. Lucero G. Chana G. Soontornniyomkij V. Achim C.L. Differential expression of immunophilins fkbp51 and fkbp52 in the frontal cortex of hiv-infected patients with major depressive disorder J. Neuroimmune Pharmacol. 2009 4 218 226 10.1007/s11481-009-9146-6 19199039 49. Ellsworth K.A. Moon I. Eckloff B.W. Fridley B.L. Jenkins G.D. Batzler A. Biernacka J.M. Abo R. Brisbin A. Ji Y. Fkbp5 genetic variation: Association with selective serotonin reuptake inhibitor treatment outcomes in major depressive disorder Pharmacogenet. Genom. 2013 23 156 166 10.1097/FPC.0b013e32835dc133 23324805 50. Fujii T. Hori H. Ota M. Hattori K. Teraishi T. Sasayama D. Yamamoto N. Higuchi T. Kunugi H. Effect of the common functional fkbp5 variant (rs1360780) on the hypothalamic-pituitary-adrenal axis and peripheral blood gene expression Psychoneuroendocrinology 2014 42 89 97 10.1016/j.psyneuen.2014.01.007 24636505 51. Hubler T.R. Scammell J.G. Intronic hormone response elements mediate regulation of fkbp5 by progestins and glucocorticoids Cell Stress Chaperones. 2004 9 243 252 10.1379/CSC-32R.1 15544162 52. Schumann G. Rujescu D. Kissling C. Soyka M. Dahmen N. Preuss U.W. Wieman S. Depner M. Wellek S. Lascorz J. Analysis of genetic variations of protein tyrosine kinase fyn and their association with alcohol dependence in two independent cohorts Biol. Psychiatry 2003 54 1422 1426 10.1016/S0006-3223(03)00635-8 14675807 53. Prasad C. Prasad A. A relationship between increased voluntary alcohol preference and basal hypercorticosteronemia associated with an attenuated rise in corticosterone output during stress Alcohol 1995 12 59 63 10.1016/0741-8329(94)00070-T 7748515 54. O’Leary J.C. 3rd Dharia S. Blair L.J. Brady S. Johnson A.G. Peters M. Cheung-Flynn J. Cox M.B. de Erausquin G. Weeber E.J. A new anti-depressive strategy for the elderly: Ablation of fkbp5/fkbp51 PLoS ONE 2011 6 1271 55. Soontornniyomkij V. Risbrough V.B. Young J.W. Wallace C.K. Soontornniyomkij B. Jeste D.V. Achim C.L. Short-term recognition memory impairment is associated with decreased expression of fk506 binding protein 51 in the aged mouse brain Age (Dordr.) 2010 32 309 322 10.1007/s11357-010-9145-9 20422297 56. Jinwal U.K. Koren J. 3rd Borysov S.I. Schmid A.B. Abisambra J.F. Blair L.J. Johnson A.G. Jones J.R. Shults C.L. O’Leary J.C. 3rd The hsp90 cochaperone, fkbp51, increases tau stability and polymerizes microtubules J. Neurosci. 2010 30 591 599 10.1523/JNEUROSCI.4815-09.2010 20071522 57. Bonkale W.L. Turecki G. Austin M.C. Increased tryptophan hydroxylase immunoreactivity in the dorsal raphe nucleus of alcohol-dependent, depressed suicide subjects is restricted to the dorsal subnucleus Synapse 2006 60 81 85 10.1002/syn.20278 16596624 58. Bardo M.T. Bhatnagar R.K. Gebhart G.F. Chronic naltrexone increases opiate binding in brain and produces supersensitivity to morphine in the locus coeruleus of the rat Brain Res. 1983 289 223 234 10.1016/0006-8993(83)90023-9 6318895 59. Grahame N.J. Li T.K. Lumeng L. Selective breeding for high and low alcohol preference in mice Behav. Genet. 1999 29 47 57 10.1023/A:1021489922751 10371758 60. Donovan J. Brown P. Blood collection Current Protocols in Immunology Coligan J. Bierer B. Margulies D. Shevach E. Strober W. Coico R. John Wiley and Sons New York, NY, USA 2006 61. Tojo H. Takami K. Kaisho Y. Nakata M. Abe T. Shiho O. Igarashi K. Analysis of neurotrophin-3 expression using the lacz reporter gene suggests its local mode of neurotrophic activity Neuroscience 1996 71 221 230 10.1016/0306-4522(95)00445-9 8834404 62. Chen H. Yong W. Hinds T.D. Yang Z. Zhou Y. Sanchez E.R. Shou W. Fkbp52 regulates androgen receptor transactivation activity and male urethra morphogenesis J. Biol. Chem. 2010 285 27776 27784 10.1074/jbc.M110.156091 20605780 63. McArthur T. Ohtoshi A. A brain-specific homeobox gene, Bsx , is essential for proper postnatal growth and nursing Mol. Cell. Biol. 2007 27 5120 5127 10.1128/MCB.00215-07 17485440 64. Chester J.A. Kirchhoff A.M. Barrenha G.D. Relation between corticosterone and fear-related behavior in mice selectively bred for high or low alcohol preference Addict. Biol. 2014 19 663 675 10.1111/adb.12034 23331637 65. Maisto S.A. Sobell M.B. Cooper A.M. Sobell L.C. Test-retest reliability of retrospective self-reports in three populations of alcohol abusers J. Behav. Assess. 1979 1 315 326 10.1007/BF01321373 66. Sobell L.C. Sobell M.B. Time-Line Follow-Back: A Technique for Assessing Self-Reported Alcohol Consumption, in Measuring Alcohol Consumption Humana Press Totowa, NJ, USA 1992 73 98 67. Bucholz K.K. Cadoret R. Cloninger C.R. Dinwiddie S.H. Hesselbrock V.M. Nurnberger J.I. Reich T. Schmidt I. Schuckit M.A. A new, semi-structured psychiatric interview for use in genetic linkage studies: A report on the reliability of the SSAGA J. Stud. Alcohol 1994 55 149 158 10.15288/jsa.1994.55.149 8189735 68. Hesselbrock M. Easton C. Bucholz K.K. Schuckit M. Hesselbrock V. A validity study of the SSAGA—A comparison with the scan Addiction 1999 94 1361 1370 10.1046/j.1360-0443.1999.94913618.x 10615721 69. APA Diagnostic and statistical manual of mental disorders (DSM-IV) APA Washington, DC, USA 1994 70. Truett G.E. Heeger P. Mynatt R.L. Truett A.A. Walker J.A. Warman M.L. Preparation of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT) Biotechniques 2000 29 52 54 10907076 71. Hendershot C.S. Neighbors C. George W.H. McCarthy D.M. Wall T.L. Liang T. Larimer M.E. ALDH2 , ADH1B and alcohol expectancies: Integrating genetic and learning perspectives Psychol. Addict. Behav. 2009 23 452 463 10.1037/a0016629 19769429 72. Luczak S.E. Glatt S.J. Wall T.L. Meta-analyses of ALDH2 and ADH1B with alcohol dependence in Asians Psychol. Bull. 2006 132 607 621 10.1037/0033-2909.132.4.607 16822169

PMC005xxxxxx/PMC5000670.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081272ijms-17-01272ArticleGene Set−Based Integrative Analysis Revealing Two Distinct Functional Regulation Patterns in Four Common Subtypes of Epithelial Ovarian Cancer Chang Chia-Ming 123Chuang Chi-Mu 234Wang Mong-Lien 25Yang Yi-Ping 345Chuang Jen-Hua 25Yang Ming-Jie 23Yen Ming-Shyen 23Chiou Shih-Hwa 1356Chang Cheng-Chang 7*Cho William Chi-shing Academic Editor1 Institute of Oral Biology, National Yang-Ming University, Taipei 112, Taiwan; [email protected] (C.-M.C.); [email protected] (S.-H.C.)2 School of Medicine, National Yang-Ming University, Taipei 112, Taiwan; [email protected] (C.-M.C.); [email protected] (M.-L.W.); [email protected] (J.-H.C.); [email protected] (M.-J.Y.); [email protected] (M.-S.Y.)3 Department of Obstetrics and Gynecology, Taipei Veterans General Hospital, Taipei 112, Taiwan; [email protected] Institute of Clinical Medicine, School of Medicine, National Yang−Ming University, Taipei 112, Taiwan5 Department of Medical Research, Taipei Veterans General Hospital, Taipei 112, Taiwan6 Department & Institute of Pharmacology, National Yang−Ming University, Taipei 112, Taiwan7 Department of Obstetrics and Gynecology, Tri-Service General Hospital, National Defense Medical Center, Taipei 114, Taiwan* Correspondence: [email protected]; Tel.: +886-228-757-394; Fax: +886-228-720-95905 8 2016 8 2016 17 8 127221 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Clear cell (CCC), endometrioid (EC), mucinous (MC) and high-grade serous carcinoma (SC) are the four most common subtypes of epithelial ovarian carcinoma (EOC). The widely accepted dualistic model of ovarian carcinogenesis divided EOCs into type I and II categories based on the molecular features. However, this hypothesis has not been experimentally demonstrated. We carried out a gene set-based analysis by integrating the microarray gene expression profiles downloaded from the publicly available databases. These quantified biological functions of EOCs were defined by 1454 Gene Ontology (GO) term and 674 Reactome pathway gene sets. The pathogenesis of the four EOC subtypes was investigated by hierarchical clustering and exploratory factor analysis. The patterns of functional regulation among the four subtypes containing 1316 cases could be accurately classified by machine learning. The results revealed that the ERBB and PI3K-related pathways played important roles in the carcinogenesis of CCC, EC and MC; while deregulation of cell cycle was more predominant in SC. The study revealed that two different functional regulation patterns exist among the four EOC subtypes, which were compatible with the type I and II classifications proposed by the dualistic model of ovarian carcinogenesis. epithelial ovarian cancerfunctionintegrative analysisgene expression microarraygene setmachine learning ==== Body 1. Introduction Epithelial ovarian carcinomas (EOC) are composed of a group of heterogeneous subtypes classified by their histology and the degree of epithelial proliferation and invasion. Clear cell (CCC), endometrioid (EC), mucinous (MC) and high-grade serous carcinoma (SC) are four common subtypes of EOC. Within the four subtypes, high-grade SC is the most common type accounting for 70% of EOC, followed by CCC, while MC is relatively rare. However, the carcinogenesis of EOC is still poorly understood. Based on the clinicopathological and molecular features, the dualistic model was proposed and divided EOCs into type I and II categories [1]. The type I EOC, including CCC, EC and MC, usually originating from the mutations of KRAS, BRAF, ERBB2, CTNNB1, PTEN and PIK3CA, is genetically stable and has a relatively indolent behavior [2]. The type II EOC, mainly high-grade SC, displays TP53 mutation in over 80% of the cases, exhibits impaired DNA damage repair and has a more uncontrolled cell differentiation and aggressive behavior. This hypothesis was based on the studies performed in the author’s laboratory and correlated with the clinical, pathologic and molecular features of the disease. However, there is no single study, nor integrative analysis to demonstrate this hypothesis and compare the pathogenesis among the four EOC subtypes. As a result, we conducted a gene set-based analysis integrating the microarray gene expression profiles of the four EOC subtypes from the publicly available database. Gene expression microarray is the primary tool for investigating cancers, the analysis of gene expression profiles usually starts with detecting the differentially expressed genes (DEG) by statistical methods, and then the aberrant Gene Ontology (GO) terms or signaling pathways are inferred from the DEGs. This workflow identifies the most significant disease-related genes, function or processes annotated by GO terms or signaling pathways, however, it will focus only on the significant ones and omit those whose p values do not reach statistical significance. In fact, genes or GO terms that did not reach the significance also play a role in the carcinogenesis of EOCs. Besides, only limited functions defined by the GO term or canonical pathways are analyzed; the complete information about the regulation of the functions i.e., functionome in EOC is not provided. To address these limitations, we investigated the pathogenesis of the four subtypes of EOC with microarray gene expression profiles of EOC and their functionomes. The biological function was quantized by converting the gene expression profiles to a gene set regularity (GSR) index computed by modifying the DIRAC algorithm [3], which measured the matching degree of gene expression rankings in a given gene set between two different phenotypes, i.e., EOC and the normal ovarian tissue control in this study. This model utilized the gene set definitions from the GO term [4] and Reactome pathway [5] databases downloaded from the Molecular Signatures Database (MSigDB) [6]. These two gene set definitions collect relatively comprehensive biological functions, processes or signaling pathways. We then utilized them to annotate human functionomes. The GO database contains 1454 gene sets, defining biological functions, process and cellular components; the canonical pathway database contains 1330 curated canonical signaling pathways. In our previous study [7], we demonstrated by the GSR indices a stepwise deterioration of cellular function regularity during SC progression from stage I to stage IV according to International Federation of Gynecology and Obstetrics (FIGO). The pathogenesis of SC centered on cell cycle deregulation accompanied with multi-functional aberrations and interactions. To further explore the pathogenesis and relationship among different subtypes of EOCs, we collected the gene expression datasets of the four common subtypes of EOC and normal ovarian samples from the publicly available databases and converted them into the GSR indices, ranging from 0 to 1 and reflecting the regularities of functions defined by the GO terms or Reactome pathways. Then, the pathogenesis of the four EOC subtypes was investigated and compared with the GSR indices by hierarchical clustering, statistical methods and exploratory factor analysis (EFA). 2. Results 2.1. DNA Microarray Gene Expression Datasets and Gene Sets DNA microarray gene expression datasets of the four EOC subtypes were downloaded from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database. Initially, 1855 potentially eligible microarray gene expression profiles were selected. We filtered out the datasets that resulted in the available common gene number less than 8000 during cross-platform integration. A total of 1452 samples, including 85 CCC, 90 EC, 48 MC, 1093 SC and 136 normal ovarian tissue control samples, were utilized in this study (Table 1). Most of the SC samples were not sub-divided into low- or high-grade SC in the GEO database. However, because high-grade SCs constitute around 90% of all SCs, it was reasonable to assume that the majority of the samples were high-grade SC. These samples data were collected from 38 datasets containing six different DNA microarray platforms without missing data. The 136 normal ovarian tissue gene expression profiles were used as controls for all of the four EOC subtypes. The detailed sample information, including the subtypes, platforms and accession numbers was available in Table S1. The 1454 GO term and 674 Reactome pathway gene set definitions were downloaded from the MSigDB, and the versions were “c5.all.v5.0.symbols.gmt” and “c2.cp.reactome.v5.0.symbols.gmt”, respectively. Because various genes were utilized in different microarray platforms, finally, 1446, 1445, 1446, 1350 GO terms and 669, 669, 669 and 614 Reactome pathways were used in computing the GSR indices for the CCC, EC, MC and SC groups, respectively. 2.2. Means and Histograms of the GSR Indices of the Four Subtypes The workflow of the GSR model was displayed on Figure 1, and the detailed procedures of computation were described in Methods. The GSR indices ranged from 0 to 1, 0 represented the most chaotic regulation of the function; while 1 represented the functional regulation of the EOC was completely unchanged in comparison with the most common gene expression ranking in the normal control population. The mean of total GSR indices for each subtype group was smaller than the normal control group, and the difference was statistically significant with a p value <0.001. The CCC and EC groups had similar means of the GSR indices, and the SC group has the smallest mean of the GSR indices, as listed in Table 1. It indicated the EOC groups exhibited more deregulated functions defined by the GO terms than the normal controls, while the SC group had the worst deregulation state. When displayed on the histograms (Figure 2), two distinguishable distributions of the GSR indices appeared in each EOC subtype; the distribution located on the left side consisted of the GSR indices for the EOC subtype had a smaller levels than the normal control distribution on the right side, indicating the biological functions were generally more deregulated in the EOC subtypes than the normal ovarian tissue controls. Especially, a second peak was observed on the left side of the histogram for the SC group, indicating the existence of a group of more severely deregulated functions in the SC group. 2.3. The Relationships of the Four Subtypes To discover the relationships of the four EOC subtypes, the GSR indices of each gene set for the four EOC subtypes were averaged then classified by hierarchical clustering and displayed on the heatmap and dendrogram (Figure 3). Grossly, the four EOC subtypes showed distinguishable patterns on the heatmap; the patterns between CCC and EC were more similar, while SC’s pattern was quite distinct from the others and showed the worst regularity of function. Their relationships were also demonstrated by the dendrogram (Figure 3). The CCC and EC groups had the closest relationship, followed by MC, while SC group was the farthest from the other three subtypes. 2.4. Functional Regulation Patterns Classified and Predicted by Machine Learning Machine learning can learn from data by building a model and recognizing patterns to make prediction. We trained support vector machine (SVM) [8], a high performance machine learning algorithm to classify and predict among the four EOC subtypes and the normal control datasets with their functional regulation patterns consisted of the GSR indices. The accuracies were tested by five-fold cross-validation. Of 1316 samples, 1052 samples were used for training, and the remaining 264 samples were used for classification and prediction. Each measurement was measured by the cumulative results of repeating 10 times classifications and predictions. The results were shown in Table 2. The accuracies of binary classification (each EOC subtype vs. control) ranged from 98.18% to 100.00%. The classification between the CCC and normal control groups had the best result. The AUC of the test for each subtype ranged from 0.9805 to 1.0000. The accuracy of multiclass classification among the four subtypes and normal control group was 95.55%. The SVM is a widely used, high-performance machine learning algorithm; this result revealed that the GSR indices could provide sufficient and adequate information for SVM to undergo accurate classification and prediction. 2.5. Deregulated GO Terms and Reactome Pathways of the Subtypes The GSR index is computed based on the extent of ranking change within a gene set defined by the GO terms or Reactome pathways between the case and control group, so the GSR index reflects the regulation of function defined by that gene set and can be utilized to evaluate the function regulation by comparing the difference between the EOC and normal control group. In order to compare the four EOC subtypes and normal controls based on the same standard, the GSR indices of the four subtype and normal control groups were computed after standardization by the baseline gene set template derived from the most common gene expression rankings in the normal ovarian gene expression profiles. The output of this calculation contained approximately 1400 or 670 GSR indices computed through the GO or Reactome pathway gene sets for each case and in each subtype. Table 3 displayed the top 15 deregulated GO terms ranked by the p values, and the full content was available in Table S2. The first deregulated GO term was “cofactor transport” for the CC and EC groups, “aldo-keto reductase” for MC and “protein tyrosine kinase activity” for the SC group. There were many recurring gene sets existing among the four subtype groups. For example, “oxidoreductase activity” was found in all of the four subtype groups, while “inositol or phosphatidylinositol phosphatase activity” appeared in the CCC, EC and MC groups. These recurring GO terms represented the commonly deregulated functions among the different EOC subtypes. In addition to oxidoreductase activity and cell adhesion, numerous deregulated GO terms in the SC group were associated with cell cycle, including “spindle”, “negative regulation of cell proliferation” and “double stranded DNA binding”, etc. The Reactome pathways ranked by the p values revealed the first and second significant deregulated pathways in the CCC and EC groups were “downregulation of ERBB2 ERBB3 signaling” and pathways related to PI3K-AKT, respectively (Table 4); the full content is available in Table S3. Obviously, numerous significantly deregulated Reactome pathways were involved in the PI3K-AKT pathway. In the SC group, the first, second and fourth deregulated pathways were associated with G protein. The first deregulated Reactome pathway was “Ca dependent events”; it was a downstream pathway of “G protein mediated events” and “PCL beta mediated events” (4th deregulated Reactome pathway). The second deregulated pathway was “DARPP 32 events”, which was a downstream of G protein coupled receptor (GPCR) signaling pathway and associated with neurotransmitter and steroid signaling. However, their roles in the carcinogenesis of EOC were unknown. Many of the subsequently deregulated pathways in the SC group were associated with cell cycle control, such as “G0 and early G1” and “cyclin A/B1 associated events during G2/M transition pathway”, etc. 2.6. The Commonly Deregulated GO Term and Reactome Pathway Gene Sets among the Four Subtypes Due to the existence of numerous recurring gene sets among the four subtype groups, we carried out set analysis for the top 200 deregulated GO or Reactome pathway gene sets to find out the similarities of deregulated functions among the four EOC subtypes. The p values of those selected gene sets were less than 0.001. The numbers of intersected gene set were displayed on the Venn diagram as shown in Figure 4. There were 27 commonly deregulated GO terms, accounting for 13.5% of all top 200 gene sets among the four subtype groups, including protein tyrosine kinase, cell adhesion, channel activity, oxidoreductase activity, DNA and protein binding etc. The number of common gene sets increased to 73%, or 36.5% of the top 200 gene sets among the CCC, EC and MC groups. Furthermore, the common gene set number was up to 114, or 57% of top 200 gene sets among the CCC and EC groups. It indicated the CCC and EC groups shared more than half of the most deregulated functions and implied a similar pathogenesis between CCC and EC. This finding was compatible with the relationship revealed by the dendrogram on Figure 3. In contrast, the deregulated functions of the SC group were quite different from the other three subtype groups; there were only 39 commonly deregulated gene sets between the MC and SC groups. The set analysis for the Reactome pathway gene sets among the four subtypes showed the number of commonly deregulated Reactome gene sets was 66, it accounted for 33% of the top 200 deregulated pathways. The number of commonly deregulated Reactome gene sets among the CCC, EC and MC groups was 101, or 50.5% of top 200 deregulated gene sets (Figure 5). 2.7. The Elements of Carcinogenesis Networks Discovered by Exploratory Factor Analysis Usually, the pathogenesis of complex diseases, such as EOC, involves a variety of functions” aberrations as well as interactions. EFA is a broadly applied statistical technique to discover the underlying structures, or networks among numerous variables. We carried out the EFA to find out the gene set elements contributing to the EOC carcinogenesis network among 1454 GO terms or 674 Reactome pathways with the gene sets of p value <0.0001. The number of “factors”, i.e., structure or network contributing to EOC carcinogenesis, was determined by the function “fa.parallel”. The numbers of factors was 6, 4, 4 and 11 for the CCC, EC, MC and SC groups, respectively. Taking the CCC group as an example, EFA found six networks (factors) of gene sets involved in the carcinogenesis of CCC selected from the deregulated GO terms of p value <0.0001; each of the six networks contained 118, 59, 40, 52, 35 and 22 gene set elements, respectively. The 118 deregulated GO terms in the first network were associated with oxidoreductase activity, transmembrane receptor protein tyrosine kinase activity, G protein coupled receptor binding, transcription coactivator activity, chromatin assembly, cell cycle, ion transport, binding and cell adhesion. The second network was composed of the elements associated with sterol binding, cell division, channel activity, oxidoreductase activity, chromatin assembly and inositol/phosphatidylinositol phosphatase activity. They represented two different but overlapped networks of EOC carcinogenesis. The sixth network containing 22 elements was a sub-network of the first one. Because of the similarity among the CCC, EC and MC groups revealed by the hierarchical clustering and set analysis, we merged the microarray gene expression datasets of the three subtypes (CCC-EC-MC group), recomputed the GSR indices for this group and carried out the EFA to discover the commonly deregulated functions among the three subtypes. The results of EFA showed seven networks of deregulated GO terms. The first network was composed of cell proliferation, oxidoreductase activity, protein binding, cell adhesion, steroid hormone, protein tyrosine kinase activity, GPCR, immune response, GTPase activity and metabolism. The second network was composed of oxidoreductase activity, cell adhesion, extracellular matrix, binding and GTPase activity. The third, fourth and fifth network was associated with channel activity, transport, G protein activity and chromatin assembly, respectively. We also utilized the EFA to analyze the Reactome pathways for the combined CCC-EC-MC group; the results showed the signaling cascades were primarily associated with the PI3K and ERBB pathways. The results of EFA for the SC group showed the deregulated GO terms were predominantly associated with cell cycle, apoptosis, cell proliferation and development. Especially, all of the elements in the 5th network were associated with cell cycle, including “spindle”, “mitotic cell cycle checkpoint”, “M phase of mitotic cell cycle”, “condensed chromosome”, “regulation of mitosis” and “microtubule organizing center”, indicating a series of cell cycle control deregulation. The full EFA results were available in Supplemental Materials (Table S4–S8, for CCC, EC, MC, SC and CCC-EC-MC groups, respectively) 2.8. Trees of Deregulated GO Terms for the Four Subtypes Because the GO terms are structured ontologies established according to their child-parent relationship, the deregulated GO gene set elements from the EFA could be organized and visualized on a directed acyclic graph according to their GO hierarchies. The redundant GO terms could be diminished and simplify the interpretation of EFA results. To establish the tree of deregulated GO terms for each subtype, the deregulated GO gene set elements collected from all factors were merged then remapped to the GO tree by the R package “RamiGO”, which would upload these GO terms to the AmiGO 2 web server for establishment of the GO trees. The deregulated GO tree of SC group is displayed in detail in Figure 6 as an illustration. The full deregulated GO trees of the four subtypes are available in Supplemental Materials (Figures S1–S4). This figure show the screenshot of the full GO tree of the SC group and some important deregulated GO terms. After mapping to the GO tree, the deregulated GO terms with similar functions or properties clustered together and were arranged by their GO hierarchies. Then, the group of clustered GO terms could be summarized by their common parental GO terms. Thus, the deregulated functions, processes or cellular components could be interpreted in a simplified way. Nine groups of clusters could be found in the deregulated GO terms of the SC group, including cell cycle, channel activity, oxidoreductase activity, chromosome, development, regulation of cell proliferation, regulation of programmed cell death and protein kinase activity. The GO tree provided an intuitive way to view the structure of deregulated functions in the carcinogenesis of EOCs. The GO trees of the CCC, EC and MC groups were relatively similar, including components of oxidoreductase activity, cell adhesion, binding, G protein activity, metabolism, channel activity and protein kinase activity. There were overlapping elements among the four EOC subtypes; however, the cell cycle-related GO terms were predominantly observed in the SC group. 2.9. Differentially Expressed Genes in the Four Subtypes of EOC We carried out integrative analysis for microarray gene expression datasets to discover and compare the differentially expressed genes (DEGs) in the four subtypes of EOC. The gene expressions of the samples in each dataset were rescaled to cumulative proportion before integration. Table 5 listed the top 100 down-regulated and up-regulated genes ranked by the p values. We found the CCC, EC and MC groups shared many common up-regulated or down-regulated DEGs. We then explored the relationship by set analysis of the top 100 DEGs to find out the similarities on deregulated functions among the four EOC subtypes. The numbers of common GEGs among subtypes were displayed on the Venn diagram (Figure 7). There were 38 commonly up-regulated DEGs, accounting for 38% of all top 100 DEGs among CCC, EC and MC groups; however, no commonly up-regulated DEGs among CCC, EC, MC and SC were found. There were 41% commonly down-regulated DEGs among CCC, EC and MC groups but only 21% among the CCC, EC, MC and SC groups. These findings indicated the distribution of pathogenic DEGs of EOC subtypes was similar among CCC, EC and MC, while SC exhibited a significantly different distribution from the other three subtypes. These results also provided additional evidence supporting the dualistic model of type I and II classifications for ovarian carcinogenesis. 3. Discussion Cancers are usually involved in multiple aberrations of gene and function as well as their interactions. In order to take these features into consideration, we utilized the GSR model to investigate the function regularities in cancers. Instead of detecting the DEGs, the model starts with converting the microarray gene expression profiles into quantized biological functions through a list of gene sets defined by the GO terms or Reactome pathways, and then the pathogenesis is evaluated by comparing the differences of functional regulation between the cases with the normal control groups. These quantized regularities of functions, i.e., the GSR indices, are computed by the modified DIRAC algorithm, which converts the gene expression levels to a gene expression ranking list in a gene set, and then measures the matching degree of gene expression rankings between two different phenotypes. We utilized a baseline gene set expression ranking template, defined as the most common gene expression ranking in the normal control populations for each gene set, as a standard to measure the regularity of gene ranking in either EOC or normal ovarian control sample. Then, the GSR index is computed by measuring the matching degree between the gene expression rankings of each ovarian cancer or normal ovarian control sample with the baseline gene set expression ranking template for each gene set. After being standardized by the baseline gene set template, the GSR indices of the four EOC subtypes can be compared based on the same standard. Besides, the GSR indices are computed based on the gene expression rankings; the gene expression levels are converted into ordinal data, and the ordinal data will encounter less cross-platform bias than the gene expression levels during integrating the datasets from different DNA microarray platforms. Computing the gene expression ranking in a gene set will take the gene interactions in a gene set into consideration. In contrast to the “genome” analyzed with gene expression microarray, this model investigates “functionome” with the GSR indices. By converting tens of thousands of gene expression profiles to approximately one thousand GSR indices, this approach will diminish the data noise, simplify the complexity of the subsequent analyses, and facilitate the performance of machine leaning. Besides, each GSR index is normalized to a value ranging from 0 to 1, in favor of the subsequent analyses. The functionome of each subtype was computed through either GO term or Reactome pathway gene set database, both databases collect relative comprehensive human biological functions and processes, and provide the browsers for viewing the hierarchy of GO terms (AmiGO 2) [9] and pathways (Reactome Pathway Browser) [10], facilitating the clarification of the relationships among numerous deregulated GO terms or pathways. The functionome was composed of approximately 1400 GO or 600 Reactome GSR indices for each case, when displayed on the heatmap, the functionomes of the four EOC subtypes could be visualized and show distinguishable patterns. These patterns could be recognized, classified and predicted by the machine learning. Our result revealed excellent binary or multiclass classification; it implied that the functionomes composed of GSR indices could be utilized as the basis of molecular classification by machine learning. Subsequently, the pathogenesis of the four subtypes was investigated by evaluating the GSR indexes. From the results of histograms and hierarchical clustering among the four subtypes, it could be found that CCC and EC had the closest relationship, followed by MC, and SC was relatively different from the others in terms of functional regulations. Indeed, the four subtypes shared quite a number of common deregulated functions, including cell adhesion, oxidoreductase activity, protein binding, channel activity and metabolism. However, deregulations of chromatin assembly, ERBB, PI3K-AKT pathways were more common among CCC, EC and MC but not in SC. In contrast, the predominant deregulated functions in SC were cell cycle control. We further explored the pathogenesis and the relationship among the four subtypes by the EFA. The results of EFA using GO terms disclosed that CCC, EC and MC shared a similar structure of pathogenesis, associated with binding, channel activity, cell adhesion, oxidoreductase activity, protein kinase activity, G protein activity and chromatin assembly. The results of EFA using Reactome pathway gene sets revealed the common deregulation of the PI3K-AKT and ERBB pathways. In contrast, the results of EFA for the SC group revealed the pathogenesis mainly involved in apoptosis, mitosis and cell division and cell cycle checkpoint. Overlapped deregulated functions among the four EOC subtype groups were also found, such as protein tyrosine kinase activity, carbohydrate biosynthetic process, immune response, channel activity, cell adhesion and oxidoreductase activity. The channel activity was demonstrated to be involved in the cell cycle control in the carcinogenesis of EOC [11], and cell adhesion played an important role in the metastasis of EOC [12]. These findings draw the conclusion that the two overlapped, but distinguishable function regulation patterns existing among the four subtypes of EOC. The first pattern observed in the CCC, EC and MC groups had moderate, deregulated functions involved in oxidoreductase activity, channel activity, binding activity, metabolism, chromatin assembly, cell adhesion, PI3K-AKT and ERBB signaling pathway. The secondary pattern, observed in the SC groups, had more severe functional regularity and was predominantly involved in the cell cycle deregulation. These two function regulation patterns were compatible with the type I and type II classifications proposed by the dualistic model of ovarian carcinogenesis: the type I EOCs, including CCC, EC and MC, usually originated from the mutation of KRAS, BRAF, ERBB2, PTEN and PIK3CA, are genetically stable and have a relatively indolent behavior; the type II EOCs, mainly high-grade SC, primarily exhibit a TP53 signature, have a more uncontrolled cell cycle and aggressive behavior. The type I and II EOCs were compatible with the first and second patterns of function regulation in our study, respectively. This study also showed evidence disclosing the relationship between deregulated functions and carcinogenesis. The association of CCC and EC with endometriosis has been repeated reported [13,14]. The cells in the endometriosis foci will be exposed to the reactive oxygen species (ROS) and are subjected to more DNA damage [15]. As the dendrogram showed in this study, the CCC and EC groups exhibited a relatively close relationship and shared many commonly deregulated GO terms, such as oxidoreductase activity and cell adhesion; both are the characteristic features of the pathogenesis of endometriosis. These findings provided the evidence supporting the role of endometriosis during the carcinogenesis of CCC and EC. Our results showed the PI3K-AKT signaling pathway was a key element of the pathogenesis of EOCs. PI3K-AKT has been demonstrated to play an important role in the carcinogenesis of EOC, especially in CCC and EC. The deregulation of this signaling pathway may be originated from the loss of PTEN in 40% cases [16], PIK3CA mutation in 33% cases [17] or AKT amplification in 14% cases [18] of CCC patients. PI3K is the major downstream effector of receptor tyrosine kinases (RTK) and GPCR. If PI3K is activated, apoptosis will be inhibited and leads to cell proliferation [19]. Both of PI3K-AKT and G protein deregulation were detected with statistical significances in this study. As the results of CCC-EC-MC combined analysis listed in the Table S9, the GO terms “inositol or phosphatidylinositol phosphatase activity” and “transmembrane receptor protein tyrosine kinase activity” were the first and sixth top deregulated GO gene sets. ERBB2 was the first deregulated pathways for CCC and EC, its expression in EOC varies widely, ranging from 20% to 30% of cases [20]. ERBB is a member of the epidermal growth factor receptor (EGFR) family, it can activate the PI3K-AKT pathway and may represent a prognostic factor in primary EOC [21]. The 9th deregulated Reactome pathway “PI3K events in ERBB2 signaling” in the CCC-EC-MC combined group indicated the interaction between the two important deregulated Reactome pathways in the carcinogenesis of EOC (Table S10). However, there are limitations when applying the GSR model to investigate the carcinogenesis of EOCs. As an illustration, the TP53 mutation is a common aberration in high-grade SC. The gene set related to TP53 could be found in the list of Reactome pathway database; however, they did not appear on the top of the significantly deregulated pathway list in this study; the first one appearing on the list was the 122th gene set “P53 dependent G1 DNA damage response” with a p value of 4.02 × 10−17. This finding illustrates the first limitation of this model: if the level of gene expression change does not reach the required extent, the gene expression ranking as well as the GSR index will remain unchanged and the aberration could not be detected. The second limitation is the incompleteness of gene set definitions. For example, there was no definition of PTEN gene set in the GO and Reactome gene set database, so no PTEN aberration was found in this study, although this model discovered a lot of PI3K related functions and pathway aberrations because the PI3K were the effector of PTEN. The third limitation is the false positivity. The third most deregulated Reactome pathway in the MC group was “olfactory signaling pathway” with a p value of 1.32 × 10−12, which should be independent of the carcinogenesis of MC. This situation can be checked and clarified via the Reactome Pathway Browser. When mapping to the browser, the hierarchy showed the “olfactory signaling pathway” was a member of the GPCR signaling pathway and contained elements involved with the regulation of G protein, and G protein was shown to play an important role in the carcinogenesis of EOC in this study. This false positivity happened because of the presence of the G protein-related gene elements in the gene set. Another limitation of this study was that the DEGs derived from the integrative analysis had not been validated. One of the best ways to validate these DEGs is RNA seq or protein expression for the samples of the four EOC subtypes. We attempted to validate the DEGs in our study by collecting the RNA seq datasets for the four EOC subtypes from two important publically available databases: The Cancer Genome Atlas (TCGA) and NCBI Sequence Read Archive (SRA). However, this validation was not feasible because the available samples of CCC, EC and MC were not enough to get significant statistical significance. Further investigation is still needed for validation of these DEGs. 4. Materials and Methods 4.1. Computing GSR Indices by Modified Differential Rank Conservation Algorithm The algorithm of computing the GSR indices was modified from the Differential Rank Conservation (DIRAC). DIRAC is designed to measure the perturbation of a gene set by converting gene expression levels to gene expression rankings, and quantifying the regularity of gene expression ranking in the gene set by computing the ranking matching score, which is a measurement of the degree of each sample’s gene expression ranking of each gene set matching the corresponding gene set ranking template. Instead of measuring the perturbation of gene expression ranking, the GSR index measures the extent of gene expression ranking change between two phenotypes in a gene set, i.e., EOC and normal controls in this study. For this purpose, the GSR indices for both EOC and the normal control are computed by comparing the sample’s gene expression ranking with a standard template derived from the most common gene expression ranking in a gene set among the entire normal ovarian tissue control samples. Then, the EOC pathogenesis was investigated by comparing the EOC and normal control GSR indices. The baseline gene set ranking template was defined as a template of the most common gene ranking among the unaffected controls in a gene set; it is used as a standard template for a gene set from the unaffected population. The baseline gene set raking template for each gene set is established by pairwise comparison between the expression levels of two genes for all possible combinations of gene pair. A gene set contains m gene G = {G1, Gm}, and the corresponding gene expression profile E = {E1, Em}, Ei denotes the expression level of gene Gi. Each sample is labeled by a phenotype of case (EOC) and unaffected control group, respectively. The baseline gene set raking template for each gene set is established by pairwise comparison between the expression levels of two genes for all possible combinations of gene pair. The baseline gene rank template B for a given gene set G is the binary vector composed of “A” or “B”, where each component is either “A” if the probabilities Pr(Ei < Ej | phenotype = control) >0.5 or “B” if Pr(Ei < Ej | phenotype = control) ≤0.5. For the expression profile of a given sample en, the GSR index for a given gene set is the fraction of the m × (m − 1)/2 pairs for which the observed gene expression ranking within en matches the baseline gene ranking template B. Establishment of the baseline gene set expression ranking template and measurement of GSR indices were executed in R environment, the code and the test datasets are available on the GitHub (https://github.com/carlzang/GSR-model.git). 4.2. Microarray Datasets Gene Set Definition and Data Processing Gene expression microarray datasets were downloaded in a SOFT format after comprehensively searching for all of the available microarray gene expression profiles in the NCBI GEO database. Ovarian carcinoma and normal ovarian tissue control datasets were selected only when the samples originated from the ovarian tissue and definite diagnosis was provided. The gene expression profile was discarded if containing missing data. The manipulation of genes and the corresponding gene expression data in each dataset was based on the HUGO Gene Nomenclature Committee (HGNC) gene symbols approved in 2013. The microarray gene expression datasets were utilized only if the corresponding gene symbol information was provided in the annotation table. The common genes and the corresponding gene expression profiles among all datasets were used in this study. The dataset were discarded if the number of the common genes became less than 8000 during intersecting with other datasets. The gene sets were discarded if the number of gene elements in the gene set is less the 3. 4.3. Statistical Analysis The differences between the four EOC subtypes and the control GSR indices were tested by Mann Whitney U test and corrected by multiple hypotheses using false discovery rate (Benjamini-Hochberg procedure). The significance level was set at <0.001. 4.4. Classification and Prediction by Machine Learning GSR index matrices computed through GO term and Reactome pathway gene sets were classified and predicted by the support vector machine (SVM) with kernlab [22], which is an R package for kernel-based machine learning methods and is used to classify patterns of the GSR indices with the setting of kernel = “rbfdot” (Radial Basis kernel “Gaussian”), type = “C−svc” (C classification). The performance of classification and prediction by SVM were measured by 5-fold cross-validation. Datasets were randomly sampled and divided into 5 parts, 4 parts were used for training sets and the remainder one part for prediction. The performance of binary classification was assessed with sensitivity, specificity, accuracy and area under curve (AUC), where Sensitivity = true positives/(true positives + false negatives) Specificity = true negatives/(true negatives + false positives) Accuracy = (true positives + true negatives)/(true positives + false positives + true negatives + false negatives) Sensitivity, specificity, accuracy and AUC were computed using the cumulative results of repeating classifications 10 times. AUC was computed by an R package pROC [23]. The performance of multiclass classification was assessed by the accuracy computed from the fraction of correct predictions within total prediction number. 4.5. Hierarchical Clustering Dendrogram and Heatmaps All of the GSR indices in each gene set for each subtype were averaged and underwent hierarchical clustering with the R function “heatmap” as default. This function would execute hierarchical clustering, and drew dendrogram and heatmaps. 4.6. Set Analysis All possible logical relations among the deregulated gene sets of the four EOC subtype groups was evaluated by set analysis and displayed by Venn diagram using an R package “VennDiagram” (version 1.6.16, downloaded from the comprehensive R archive network (CRAN), https://cran.r-project.org/index.html). 4.7. Exploratory Factor Analysis for the Deregulated GO Terms and Establishment of GO Trees The deregulated GO terms of p values <0.001 were selected for EFA. EFA was executed with the R package “psych” (version 1.5.8). The number of factors to be extracted was determined by the function “pa.parellel”. The setting of factoring method used in this study was “pa” and the correlation matrix rotation method was “promax”. The tree of the deregulated GO terms was constructed and visualized in Portable Network Graphics (PNG) format constructed by the “RamiGO” [24], an R package providing functions to interact with the AmiGO 2 web server and retrieves GO trees. 4.8. Detection of Differentially Expressed Genes in the Four Subtypes of EOC To discover the DEGs for each of the four EOC subtypes, we carried out integrative analysis with the downloaded DNA microarray datasets. The gene expression levels of all samples in each dataset were transformed and rescaled to cumulative proportion values from 0 (lowest expression) to 1 (highest expression) with an R package “YuGene” (version 1.1.5) before integration. The DEGs were discovered using linear model computed with empirical Bayes analysis by the functions “lmFit” and “eBayes” provided by the R package “limna” (version 3.26.9). 5. Conclusions Investigating the pathogenesis of diseases with the functionomes not only makes a clear distinction among the different subtypes, but also provides a comprehensive view of the deregulated functions in these diseases. Our study demonstrated two overlapped but distinguishable deregulated function patterns among the four EOC subtypes. The first pattern, observed in CCC, EC and MC, showed a relatively moderate deregulation of functions involving the PI3K-related functions and chromatin assembly. The second pattern, found in SC, showed more severely deregulated functions associated with the control of cell cycle. These findings were compatible with the type I and II classifications proposed by the dualistic model of ovarian carcinogenesis. This study provided solid evidences to support this classification and was the first integrative analysis demonstrating this model. Acknowledgments This work was supported by Healthbanks Biotech (R92-001-14) and Tri-Service General Hospital (TSGH-C104-006-008-S01). Supplementary Materials The following are available online at www.mdpi.com/1422-0067/17/8/1272/s1. Click here for additional data file. Author Contributions Chia-Ming Chang, Cheng-Chang Chang and Shih-Hwa Chiou designed the study. Chia-Ming Chang collected and characterized the samples. Chia-Ming Chang performed the experiments. Chia-Ming Chang and Mong-Lien Wang analyzed the data. Chia-Ming Chang, Chi-MuChuang, Mong-Lien Wang, Yi-Ping Yang, Jen-Hua Chuang, Ming-Jie Yang, Cheng-Chang Chang, Ming-Shyen Yen and Shih-Hwa Chiou wrote the paper. All authors have read and approved the submitted manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations EOC Epithelial Ovarian Carcinoma CCC Clear Cell Carcinoma EC Endometrioid Carcinoma MC Mucinous Carcinoma SC Serous Carcinoma GSR Gene Set Regularity DEG Differentially Expressed Gene DAG Directed Acyclic Graph GO Gene Ontology MSigDB Molecular Signatures Database EFA Exploratory Factor Analysis NCBI National Center for Biotechnology Information GEO Gene Expression Omnibus SD Standard Deviation SVM Support Vector Machine AUC Area under Curve GPCR G Protein Coupled Receptor HGNC HUGO Gene Nomenclature Committee DIRAC Differential Rank Conservation, RTK Receptor Tyrosine Kinases EGFR Epidermal Growth Factor Receptor CRAN Comprehensive R Archive Network Figure 1 Workflow of the gene set regularity model. The gene set regularity (GSR) index was computed by converting the gene expression rankings of each epithelial ovarian carcinoma (EOC) subtype or normal ovarian control sample through each gene Contrology (GO) term or Reactome pathway gene set. Machine learning algorithm was trained to recognize the patterns consisted of the GSR indices then executed the binary (EOC vs. control) or multiclass (four EOC subtypes + control) classifications. Functionome analyses were carried out by statistical methods, hierarchical clustering and exploratory factor analysis. Figure 2 Histograms of the four subtypes. The gene set regularity (GSR) indices for each subtype and normal control group were displayed on the histograms by density. The GSR indices for the two groups showed two distinguishable distributions on the histograms; the distribution consisted of the GSR indices for the EOC subtypes (orange) located on the left side had smaller levels, indicating the biological functions were generally more deregulated in the EOC subtypes than the normal control group (blue). Figure 3 Heatmap and dendrogram of the four subtypes. The heatmap and dendrogram (left side of the heatmap) demonstrated the relationships of the four EOC subtypes. The heatmap showed the CCC and EC groups were the closest, while the SC group exhibited farthest relationship from the others and the most seriously deregulated functions. The red color in the heatmap was correlated with lower, and yellow color with higher value of gene set regularity index. Figure 4 Venn diagram of the top 200 significantly deregulated GO terms for the four subtypes. The results of set analysis for the four ECO subtypes with the top 200 significantly deregulated GO terms ranked by the p values were displayed on the Venn diagram to show the gene set numbers of all possible logical relations among the four subtypes. The 27 common deregulated GO terms among the four subtypes were listed on the right side of the diagram. Figure 5 Venn diagram of the top 200 significantly deregulated Reactome pathways for the four subtypes. The results of set analysis for the four EOC subtypes with the top 200 significantly deregulated Reactome pathways ranked by the p values were displayed on the Venn diagram to show the gene set numbers of all possible logical relations among the four subtype groups. The 66 common deregulated Reactome pathways among the four subtype groups were listed on the right side of the diagram. Figure 6 GO tree of SC. This figure displayed the screenshot of the full GO tree for SC (middle). After mapping to the GO tree, the similar GO terms clustered together. Each cluster was circled (red) and some of the important deregulated GO terms (green boxes) in the cluster were magnified to view the details. Each cluster was labeled by the common parental GO term (orange rectangle). Figure 7 Venn diagram of the top 100 up- and down-regulated differentially expressed genes (DEGs) for the four subtypes. The results of set analysis for the four ECO subtypes with (A) the top 100 up-regulated; and (B) top 100 down-regulated DEGs were ranked by the p values, and the DEG numbers of all possible logical relationships among the four subtypes were shown. ijms-17-01272-t001_Table 1Table 1 Sample numbers and means of the gene set regularity indices for each subtype. The table displayed the sample numbers, means and SDs of GSR indices for the four EOC subtypes and the normal ovarian tissue controls computed through the GO term gene sets. The 136 normal ovarian tissue sample gene expression profiles were utilized as the control group for the all of the four EOC subtypes. EOC Subtype Sample Control Total Sample Mean (SD) Control Mean (SD) p Value * Clear cell 85 136 221 0.7438 (0.1171) 0.7727 (0.1329) <0.001 Endometrioid 90 136 226 0.7434 (0.1260) 0.7731 (0.1326) <0.001 Mucinous 48 136 184 0.7174 (0.1531) 0.7724 (0.1334) <0.001 Serous 1093 136 1229 0.6694 (0.1997) 0.7697 (0.1589) <0.001 SD: standard deviation; GSR: gene set regularity; EOC: epithelial ovarian carcinoma; GO: Gene Ontology; * Mann Whitney U test. ijms-17-01272-t002_Table 2Table 2 Accuracies of the binary and multiclass classification and prediction by machine learning. This table displayed the performances of the binary (each subtype vs. control group) and multiclass classification (among the four subtype groups) and prediction by SVM with the GSR indices computed through the GO terms. The sensitivities, specificities, AUC, accuracies and the SD were measured by five-fold cross-validation. Each measurement was computed by the cumulative 10 results of repeated classifications and predictions. SVM: support vector machine; GSR: gene set regularity; GO: Gene Ontology; AUC: area under curve; SD: standard deviation; NA: not available. EOC Subtype Sensitivity (SD) Specificity (SD) Accuracy (SD) AUC Clear Cell 1.0000 (0.0000) 1.0000 (0.0000) 1.0000 (0.0000) 1.0000 Endometrioid 0.9724 (0.0463) 1.0000 (0.0000) 0.9888 (0.0188) 0.9868 Mucinous 0.9582 (0.0559) 1.0000 (0.0000) 0.9818 (0.0139) 0.9805 Serous 0.9930 (0.0004) 0.9680 (0.0269) 0.9902 (0.0004) 0.9807 Multiclass NA NA 0.9555 (0.0112) NA ijms-17-01272-t003_Table 3Table 3 The top 15 deregulated Gene Ontology (GO) terms of the four subtype groups ranked by the p values. This table displayed the top 15 significantly deregulated GO terms of each subtype. GO: Gene Ontology. Clear Cell Endometrioid Mucinous Serous Cofactor Transport Cofactor Transporter Activity Aldo Keto Reductase Activity Protein Tyrosine Activity Inositol or Phosphatidylinositol Phosphatase Activity Secretin Like Receptor Activity Secretin Like Receptor Activity Oxidoreductase Activity Acting on The Aldehyde or OXO Group of Donors Rho Guanyl Nucleotide Exchange Factor Activity Carbohydrate Biosynthetic Process Vitamin Transport Homophilic Cell Adhesion Small Conjugating Protein Binding Regulation of Viral Reproduction Rho Guanyl Nucleotide Exchange Factor Activity Regulation of Actin Filament Length Ubiquitin Binding Calcium Independent Cell Adhesion Small Conjugating Protein Binding Regulation of Actin Polymerization and or Depolymerization Regulation of Viral Reproduction Coenzyme Binding Ubiquitin Binding Regulation of Cellular Component Size Vitamin Transport Sulfotransferase Activity Calcium Channel Activity Vitamin Metabolic Process Steroid Hormone Receptor Binding Inositol or Phosphatidylinositol Phosphatase Activity Negative Regulation of Immune System Process Spindle Pole Histone Deacetylase Binding Calcium Channel Activity Carbohydrate Biosynthetic Process Negative Regulation of Cellular Component Organization and Biogenesis Oxidoreductase Activity Acting on the CH NH Group of Donors Cofactor Binding Inositol or Phosphatidylinositol Phosphatase Activity Spindle Transmembrane Receptor Protein Tyrosine Kinase Activity Transferase Activity Transferring Sulfur Containing Groups Neuropeptide Binding Innate Immune Response Protein Tyrosine Kinase Activity Oxidoreductase Activity Acting on The Aldehyde or OXO Group of Donors Neuropeptide Receptor Activity Negative Regulation of Cell Proliferation Insoluble Fraction Vitamin Transport Transmembrane Receptor Protein Tyrosine Kinase Activity Regulation of Organelle Organization and Biogenesis Carbohydrate Biosynthetic Process Transmembrane Receptor Protein Tyrosine Kinase Activity Innate Immune Response Single Stranded DNA Binding Ras Guanyl Nucleotide Exchange Factor Activity Rho Guanyl Nucleotide Exchange Factor Activity Cofactor Transporter Activity Oxidoreductase Activity Acting on The Aldehyde or OXO Group of Donorsnad or Nadp As Acceptor ijms-17-01272-t004_Table 4Table 4 The top 15 deregulated Reactome pathways ranked by the p values of each EOC subtype group. This table partially displayed the significantly deregulated Reactome pathways of each subtype. Only the top 15 deregulated Reactome pathway gene sets were listed. Clear Cell Endometrioid Mucinous Serous Downregulation of ERBB2 ERBB3 Signaling Downregulation of ERBB2 ERBB3 Signaling Organic Cation Anion Zwitterion Transport Ca Dependent Events Negative Regulation of the PI3K AKT Network CD28 Dependent PI3K AKT Signaling Downregulation of ERBB2 ERBB3 Signaling DARPP 32 Events Activated AMPK Stimulates Fatty Acid Oxidation in Muscle Organic Cation Anion Zwitterion Transport Olfactory Signaling Pathway Signaling by Robo Receptor PERK Regulated Gene Expression Nef Mediated Downregulation of MHC Class I Complex Cell Surface Expression Digestion of Dietary Carbohydrate Plc Beta Mediated Events Ethanol Oxidation GABA Synthesis Release Reuptake and Degradation PI3K Events in ERBB2 Signaling COPI Mediated Transport Phospholipase C Mediated Cascade Negative Regulation of the PI3K AKT Network Regulation of Insulin Like Growth Factor Igf Activity by Insulin Like Growth Factor Binding Proteins Igfbps Sphingolipid De Novo Biosynthesis Regulation of Rheb Gtpase Activity By AMPK Inhibition of The Proteolytic Activity of APC C Required for The Onset of Anaphase By Mitotic Spindle Checkpoint Components Regulated Proteolysis of P75NTR DAG and IP3 Signaling PI3K Cascade NCAM1 Interactions Activated AMPK Stimulates Fatty Acid Oxidation in Muscle NCAM1 Interactions Beta Defensins GPVI Mediated Activation Cascade Nef Mediated Downregulation Of MHC Class I Complex Cell Surface Expression G0 and Early G1 FGFR Ligand Binding and Activation Phosphorylation of The APC C Peptide Ligand Binding Receptors Gα Z Signalling Events Common Pathway Termination of O Glycan Biosynthesis Class A1 Rhodopsin Like Receptors MHC Class II Antigen Presentation Activation of Genes by ATF4 Regulation of Rheb Gtpase Activity by AMPK Intrinsic Pathway Signaling by PDGF GPVI Mediated Activation Cascade Activated Ampk Stimulates Fatty Acid Oxidation in Muscle CD28 Dependent PI3K AKT Signaling HS GAG Biosynthesis PI3K Cascade Conversion From APC C CDC20 to APC C Cdh1 in Late Anaphase Endogenous Sterols Chondroitin Sulfate Dermatan Sulfate Metabolism Insulin Receptor Signalling Cascade APC C CDC20 Mediated Degradation of Cyclin B Formation of Fibrin Clot Clotting Cascade Abacavir Transport and Metabolism ijms-17-01272-t005_Table 5Table 5 The top 100 up- and down-regulated differentially expressed genes of the four EOC subtype groups. The genes were ranked by their p values. Clear Cell Endometrioid Mucinous Serous Down-Regulation Up-Regulation Down-Regulation Up-Regulation Down-Regulation Up-Regulation Down-Regulation Up-Regulation Ranking Gene p Value Gene p Value Gene p Value Gene p Value Gene p Value Gene p Value Gene p Value Gene p Value 1 EIF3F 7.35 × 10−109 TOMM7 1.67 × 10−118 EIF3F 8.76 × 10−114 TOMM7 9.66 × 10−107 EIF3F 2.22 × 10−101 RPL23 7.94 × 10−88 AOX1 3.51 × 10−133 C14orf2 8.15 × 10−78 2 RPL21 9.70 × 10−89 RPL24 1.23 × 10−109 RPS13 9.13 × 10−98 RPL34 1.96 × 10−96 TOMM7 7.54 × 10−94 PLS3 6.95 × 10−86 EIF3F 2.00 × 10−132 COX6B1 2.59 × 10−66 3 PRNP 1.88 × 10−81 RPS13 9.31 × 10−102 RPS11 5.65 × 10−95 RPL23 3.91 × 10−95 RPL34 2.75 × 10−89 FHL2 5.95 × 10−82 DFNA5 1.26 × 10−128 TRIAP1 3.44 × 10−65 4 RPL13 4.78 × 10−80 EIF3L 1.52 × 10−101 RPL27 2.10 × 10−93 ALDH9A1 5.65 × 10−95 RPS13 1.60 × 10−84 ALDH9A1 6.07 × 10−82 PTGIS 6.85 × 10−125 RBX1 9.37 × 10−63 5 CAV1 3.04 × 10−78 RPS11 1.71 × 10−98 DFNA5 4.74 × 10−92 PLS3 1.30 × 10−94 RPS11 2.85 × 10−83 RPS27L 8.49 × 10−80 TSPAN5 7.08 × 10−124 CGRRF1 1.25 × 10−61 6 DFNA5 1.76 × 10−76 ITM2B 6.57 × 10−98 RPL39 1.64 × 10−89 ITM2B 6.36 × 10−94 RPS15 7.86 × 10−82 SEC31A 8.73 × 10−80 BAMBI 2.13 × 10−108 LSM6 6.16 × 10−60 7 RPS28 4.73 × 10−74 RPL27 7.35 × 10−98 RPL41 1.38 × 10−86 RPS15 3.67 × 10−93 RPL27 2.31 × 10−80 RRAGA 2.54 × 10−78 SPOCK1 2.13 × 10−108 COX5A 1.71 × 10−59 8 CALD1 3.06 × 10−70 RPL17 5.33 × 10−97 SGK1 1.71 × 10−85 RPL36AL 2.03 × 10−90 DFNA5 2.26 × 10−79 YPEL5 3.86 × 10−78 GFPT2 8.91 × 10−107 TIMM8B 1.54 × 10−58 9 PMP22 5.06 × 10−69 RPS15 2.73 × 10−94 RPLP2 6.38 × 10−85 RPL32 1.64 × 10−89 RPL32 1.61 × 10−77 RPL36 2.04 × 10−77 C21orf62 1.35 × 10−106 SNX6 1.62 × 10−58 10 TPM1 8.35 × 10−69 RPL5 2.97 × 10−92 PRNP 3.01 × 10−84 LAPTM4A 1.91 × 10−88 RPL39 1.61 × 10−77 RPL36AL 4.12 × 10−77 FLRT2 5.29 × 10−104 IER3IP1 1.88 × 10−58 11 RPL10 1.07 × 10−67 PLS3 1.17 × 10−91 CAV1 5.20 × 10−84 SRP14 5.89 × 10−88 PRNP 3.65 × 10−77 LAPTM4A 1.20 × 10−76 NDN 2.35 × 10−103 MGST2 2.04 × 10−57 12 PTGIS 1.60 × 10−66 RPS3A 1.42 × 10−91 UROD 1.72 × 10−82 RPL36 1.43 × 10−87 SGK1 5.08 × 10−77 ANXA5 4.42 × 10−76 GPRASP1 5.93 × 10−103 METTL5 2.38 × 10−57 13 DCN 1.88 × 10−66 RPL39 7.65 × 10−91 RPS28 3.11 × 10−81 RPL6 2.48 × 10−86 RPL30 2.92 × 10−75 DSTN 5.00 × 10−74 IGFBP6 3.90 × 10−102 MRPS14 3.94 × 10−57 14 NDN 1.02 × 10−65 RPS27L 7.68 × 10−90 PMP22 4.40 × 10−81 RPL30 2.59 × 10−86 PMP22 3.66 × 10−74 OAT 5.21 × 10−74 RPS11 6.71 × 10−101 JMJD6 1.32 × 10−56 15 HNRNPA1L2 3.57 × 10−64 RPL23 9.84 × 10−90 TIMP2 5.47 × 10−81 GABARAP 8.47 × 10−86 RPL6 7.61 × 10−74 CD99 2.58 × 10−73 ZFPM2 4.41 × 10−96 NOP10 1.41 × 10−56 16 SH3BP4 1.22 × 10−63 RPL36AL 1.60 × 10−89 PTGIS 1.43 × 10−76 OAT 7.37 × 10−85 UROD 2.92 × 10−73 DPYSL2 4.47 × 10−73 RPS18 7.65 × 10−95 NFU1 1.52 × 10−56 17 RPS14 1.31 × 10−63 RPL34 1.98 × 10−89 NDN 2.18 × 10−76 LTA4H 1.92 × 10−84 RPLP2 9.80 × 10−72 CAMLG 4.55 × 10−73 ME1 9.97 × 10−94 PIGP 1.81 × 10−56 18 FHL1 6.68 × 10−63 ALDH9A1 2.31 × 10−89 GFPT2 1.20 × 10−73 FHL2 2.74 × 10−84 RPL41 1.66 × 10−71 GABARAPL2 5.06 × 10−73 RPL27A 1.68 × 10−93 ITGB3BP 2.15 × 10−55 19 HUWE1 7.18 × 10−63 RPL3 6.82 × 10−89 VCL 2.20 × 10−73 UBB 2.83 × 10−84 RPL15 3.77 × 10−70 FAU 7.09 × 10−73 SERPINE2 5.58 × 10−93 RNF139 2.65 × 10−55 20 SERPINE2 9.91 × 10−63 RPL36 1.59 × 10−88 AMIGO2 1.07 × 10−72 RRAGA 6.11 × 10−84 RPS28 1.07 × 10−69 SRP14 1.13 × 10−72 UROD 4.27 × 10−92 C19orf53 5.82 × 10−55 21 TACC1 3.81 × 10−62 RPL30 2.93 × 10−88 LXN 4.59 × 10−72 CD99 1.95 × 10−83 UBB 2.99 × 10−69 ST13 3.83 × 10−72 TRPC1 5.36 × 10−92 SEC22B 1.08 × 10−54 22 LXN 7.86 × 10−62 RPL6 9.56 × 10−88 MEIS2 3.01 × 10−71 RPS24 3.55 × 10−83 RPS27A 8.25 × 10−69 TCEAL4 5.30 × 10−72 AMIGO2 7.52 × 10−92 DDIT3 2.08 × 10−54 23 IL6ST 1.08 × 10−61 RPL32 1.05 × 10−86 CRIM1 7.32 × 10−70 ST13 3.57 × 10−83 RPL10A 9.99 × 10−69 HTRA1 9.66 × 10−72 ERH 1.00 × 10−91 NOSIP 8.18 × 10−54 24 ZFPM2 6.36 × 10−61 RPL31 3.52 × 10−86 TACC1 1.50 × 10−69 SEC31A 5.38 × 10−83 RPS27 1.85 × 10−68 NDUFA4 9.80 × 10−72 DAPK1 2.40 × 10−91 ELP4 1.23 × 10−53 25 VAPA 5.06 × 10−60 RPS16 5.06 × 10−86 ZFP36L1 3.10 × 10−69 DPYSL2 1.33 × 10−82 NDN 2.75 × 10−68 FTO 1.31 × 10−71 PMP22 5.50 × 10−90 ATP5G1 1.33 × 10−53 26 MEIS2 9.44 × 10−60 TPT1 6.28 × 10−86 SGCE 1.02 × 10−68 YPEL5 2.76 × 10−82 RPS18 2.91 × 10−68 RPS24 1.71 × 10−71 VCL 1.15 × 10−89 C14orf1 5.69 × 10−53 27 C1S 3.56 × 10−59 ACTG1 8.08 × 10−86 IGFBP6 1.69 × 10−68 FAU 3.91 × 10−82 ZFAND5 1.68 × 10−67 GABARAP 3.74 × 10−71 DIRAS3 1.56 × 10−89 SDC4 4.24 × 10−52 28 BAMBI 6.54 × 10−59 SNX3 9.76 × 10−86 ZFPM2 2.99 × 10−68 RPS27L 7.11 × 10−82 RPL27A 1.83 × 10−67 REEP5 9.51 × 10−71 PRKCDBP 6.25 × 10−89 PDCD10 8.22 × 10−52 29 CDH11 8.97 × 10−59 CCNI 1.18 × 10−85 SERPINE2 6.46 × 10−68 CAMLG 2.19 × 10−81 AMIGO2 3.98 × 10−66 GNB2L1 3.48 × 10−69 PDGFD 1.10 × 10−88 CCDC25 1.87 × 10−51 30 PDGFRA 4.98 × 10−58 RPL13A 4.88 × 10−85 GSTM3 1.76 × 10−67 RPL10A 5.37 × 10−81 PTGIS 1.35 × 10−64 LTA4H 3.48 × 10−69 CLIP4 1.39 × 10−88 NOC3L 3.10 × 10−51 31 CYBRD1 1.07 × 10−57 RPS20 2.26 × 10−84 PDGFRA 7.06 × 10−67 DSTN 1.02 × 10−80 C1S 1.02 × 10−63 ERH 5.04 × 10−69 RPL23 1.39 × 10−88 SDHD 4.27 × 10−51 32 IGFBP6 2.72 × 10−57 BTF3 2.72 × 10−84 PLSCR4 7.22 × 10−67 RPL15 1.02 × 10−80 GFPT2 4.97 × 10−63 TMSB4X 4.82 × 10−68 PLS3 2.25 × 10−88 FAM96B 4.47 × 10−51 33 ZMIZ1 3.24 × 10−57 COX7C 4.34 × 10−84 CYBRD1 1.44 × 10−66 RPS18 1.82 × 10−80 VCL 5.06 × 10−63 HNRNPK 4.82 × 10−68 PAPSS2 5.82 × 10−88 DCTPP1 8.65 × 10−51 34 7−Sep 3.73 × 10−57 RPS12 5.08 × 10−84 ARMCX1 3.81 × 10−66 GABARAPL2 1.88 × 10−80 SGCE 5.98 × 10−63 CRTAP 1.22 × 10−67 ST3GAL5 1.39 × 10−87 MRPS35 1.26 × 10−50 35 PLSCR4 1.04 × 10−56 SRP14 6.84 × 10−84 DAPK1 9.42 × 10−66 ANXA5 3.65 × 10−80 ATP5A1 3.58 × 10−62 PALLD 1.75 × 10−67 CAMLG 8.01 × 10−86 PPP1CB 1.46 × 10−50 36 CAPN2 2.02 × 10−56 RPL41 1.45 × 10−83 ZCCHC24 1.10 × 10−65 ERH 4.10 × 10−80 ZFP36L1 5.62 × 10−62 TMSB10 1.88 × 10−67 CALB2 2.38 × 10−85 ATIC 1.88 × 10−50 37 FLRT2 5.76 × 10−56 CAMLG 1.84 × 10−83 AOX1 1.15 × 10−65 TCEAL4 4.44 × 10−80 BNIP3 1.95 × 10−61 LEPROT 1.74 × 10−66 HOXC6 6.53 × 10−85 MRPS33 3.69 × 10−50 38 GFPT2 8.87 × 10−56 FAU 2.07 × 10−83 DDR2 2.29 × 10−65 HTRA1 5.70 × 10−80 BAMBI 4.34 × 10−61 MORF4L1 2.29 × 10−66 NT5E 1.07 × 10−84 RAB32 4.09 × 10−50 39 DDR2 1.19 × 10−55 ATP5L 2.53 × 10−83 IFFO1 4.57 × 10−65 TMSB4X 1.69 × 10−79 SERPINE2 2.87 × 10−60 ADH5 2.72 × 10−66 LXN 3.09 × 10−84 MYL6B 4.23 × 10−50 40 RGL1 1.89 × 10−55 RPS4X 6.11 × 10−83 FLRT2 5.39 × 10−65 REEP5 1.99 × 10−79 TUBA1A 4.49 × 10−60 UBA52 4.35 × 10−66 GALC 4.12 × 10−84 EIF2S1 4.48 × 10−50 41 DAB2 4.93 × 10−55 RPSA 9.43 × 10−83 PAPSS2 1.43 × 10−64 RPS27 2.52 × 10−79 RGL1 2.40 × 10−59 PNRC2 4.35 × 10−66 SGK1 4.67 × 10−84 SGCB 5.45 × 10−50 42 NR3C1 7.53 × 10−55 GNB2L1 9.98 × 10−83 PRKCDBP 1.59 × 10−64 UBA52 3.26 × 10−79 CCT8 4.12 × 10−59 EID1 5.10 × 10−66 ALDH1A3 7.40 × 10−84 SNAPC5 5.62 × 10−50 43 ZCCHC24 7.58 × 10−55 ATP6V0E1 1.35 × 10−82 PROS1 1.62 × 10−64 NPTN 3.81 × 10−79 CYBRD1 7.82 × 10−59 NPTN 5.13 × 10−66 PLSCR4 9.27 × 10−84 ZZZ3 1.37 × 10−49 44 PROS1 1.57 × 10−54 RPL18 1.91 × 10−82 FZD7 1.87 × 10−64 RPL27A 1.30 × 10−78 ZFPM2 1.02 × 10−58 RPS26 6.74 × 10−66 VGLL3 2.60 × 10−83 PSMB3 1.99 × 10−49 45 FSTL1 2.49 × 10−54 RPS24 3.78 × 10−82 IGFBP5 8.03 × 10−64 TMSB10 2.07 × 10−78 PLSCR4 1.89 × 10−58 SLC25A3 1.77 × 10−65 COX7A2 2.94 × 10−83 CISD1 2.63 × 10−49 46 MYLK 2.80 × 10−54 EEF1G 1.88 × 10−81 RGS2 1.71 × 10−63 SLC25A3 2.80 × 10−78 DYRK1A 4.52 × 10−58 EIF3E 1.81 × 10−65 ALDH9A1 5.65 × 10−83 RTN3 2.71 × 10−49 47 ARMCX1 3.45 × 10−54 RRAGA 6.51 × 10−81 TSPAN5 1.73 × 10−63 HNRNPK 7.96 × 10−78 FZD7 5.61 × 10−58 TAX1BP3 5.34 × 10−65 FHL2 6.12 × 10−82 TMED3 3.49 × 10−49 48 FZD7 4.01 × 10−54 RPL35A 1.11 × 10−80 BAMBI 1.81 × 10−63 GNB2L1 9.58 × 10−78 CAPN2 6.35 × 10−58 LDHA 6.73 × 10−65 CYBRD1 7.23 × 10−82 CCDC59 6.01 × 10−49 49 IGFBP5 6.73 × 10−54 RPS17 2.28 × 10−80 CLIP4 1.15 × 10−62 MYL6 1.30 × 10−77 FLRT2 7.64 × 10−58 MYL6 7.78 × 10−65 SEMA3C 1.20 × 10−81 POLR2L 6.23 × 10−49 50 GAS1 9.23 × 10−54 CIRBP 2.72 × 10−80 HOXC6 3.29 × 10−62 FTO 1.54 × 10−77 IGFBP6 1.30 × 10−57 HSP90AA1 8.80 × 10−65 ATP10D 3.70 × 10−81 FAM53C 8.50 × 10−49 51 SEMA3C 9.50 × 10−54 TMSB4X 3.69 × 10−80 TGFB1I1 3.77 × 10−62 RPS26 1.74 × 10−77 ARMCX1 2.18 × 10−57 KLHDC2 9.40 × 10−65 DPYSL2 3.70 × 10−81 GOLPH3L 9.76 × 10−49 52 TXNRD1 1.12 × 10−53 RPL10A 6.66 × 10−80 OPTN 4.04 × 10−62 NDUFA4 3.72 × 10−77 TSPAN5 2.64 × 10−57 ISCU 1.04 × 10−64 FOXO1 4.60 × 10−81 NDUFA13 1.14 × 10−48 53 RNASE4 1.60 × 10−53 FHL2 7.39 × 10−80 APPBP2 8.30 × 10−62 HSP90AA1 3.93 × 10−77 SDC2 2.86 × 10−57 PDLIM1 2.09 × 10−64 DSTN 8.10 × 10−81 DUSP22 1.27 × 10−48 54 TSPAN5 1.76 × 10−53 ANXA5 1.04 × 10−79 ST3GAL5 1.31 × 10−61 EIF3E 5.34 × 10−77 SEMA3C 3.16 × 10−57 SPCS1 2.52 × 10−64 TIMP2 2.78 × 10−80 BET1 1.32 × 10−48 55 CFH 2.76 × 10−53 NDUFA4 2.66 × 10−79 CLDN11 1.79 × 10−61 ADH5 7.61 × 10−77 DDR2 5.10 × 10−57 SPARC 4.80 × 10−64 ANXA5 3.30 × 10−80 SEH1L 1.35 × 10−48 56 ALCAM 5.43 × 10−53 SGK1 3.86 × 10−79 FBN1 5.06 × 10−61 SPCS1 9.99 × 10−77 ZCCHC24 6.92 × 10−57 LXN 5.79 × 10−64 DNAJB9 8.77 × 10−80 AMD1 1.46 × 10−48 57 PRKCDBP 1.19 × 10−52 EEF1A1 4.61 × 10−79 TCEAL2 1.17 × 10−60 MORF4L1 1.87 × 10−76 IGFBP5 9.53 × 10−57 ATF4 7.07 × 10−64 GHR 9.71 × 10−80 RALB 1.66 × 10−48 58 CLIP4 2.90 × 10−52 RPS27 6.26 × 10−79 HEG1 2.94 × 10−60 MTCH1 2.18 × 10−76 PRKCDBP 3.43 × 10−56 UXT 7.96 × 10−64 HTRA1 1.43 × 10−79 PLEKHA1 2.10 × 10−48 59 ANTXR1 3.16 × 10−52 OAT 1.18 × 10−78 RBPMS 6.53 × 10−60 RPS27A 2.98 × 10−76 IFFO1 1.12 × 10−55 SEPW1 8.64 × 10−64 SDC2 1.81 × 10−79 KIAA1598 2.21 × 10−48 60 GALC 3.85 × 10−52 YPEL5 2.06 × 10−78 AKT3 7.43 × 10−60 SEC11A 3.00 × 10−76 SPOCK1 2.09 × 10−55 COX7A2 1.60 × 10−63 COX6C 2.02 × 10−79 GGCT 2.51 × 10−48 61 EMP3 4.13 × 10−52 FTL 2.31 × 10−78 SPOCK1 9.65 × 10−60 LDHA 3.43 × 10−76 CFH 3.82 × 10−55 BTG1 1.61 × 10−63 FTO 2.75 × 10−79 MAGOH 3.31 × 10−48 62 IFFO1 6.17 × 10−52 RPS6 2.52 × 10−78 CFH 1.72 × 10−59 FTH1 7.91 × 10−76 STAT2 7.79 × 10−55 PGM1 1.68 × 10−63 NDUFA1 3.17 × 10−79 TBPL1 3.54 × 10−48 63 HOXC6 8.97 × 10−52 RPS18 1.35 × 10−77 RAB8B 7.02 × 10−59 ISCU 1.02 × 10−75 CLIC4 9.47 × 10−55 COX6C 2.69 × 10−63 IKBKAP 3.31 × 10−79 TSPAN31 3.79 × 10−48 64 SPOCK1 1.39 × 10−51 RPL27A 1.61 × 10−77 BNC2 1.21 × 10−58 EID1 1.46 × 10−75 ANTXR1 1.22 × 10−54 UQCRQ 3.22 × 10−63 LRRC49 4.20 × 10−79 BTN3A2 5.37 × 10−48 65 AOX1 2.07 × 10−51 RPS27A 1.68 × 10−77 GLT8D2 1.95 × 10−58 TAX1BP3 1.69 × 10−75 GALC 1.43 × 10−54 FTH1 3.25 × 10−63 TCF21 8.18 × 10−79 MEA1 6.93 × 10−48 66 B3GNT1 2.28 × 10−51 UBC 1.68 × 10−77 PDGFD 2.76 × 10−58 COX7A2 3.20 × 10−75 ZMIZ1 1.55 × 10−54 NPC2 3.31 × 10−63 AFF1 9.76 × 10−79 NUP37 8.14 × 10−48 67 RGS2 2.28 × 10−51 GABARAPL2 1.90 × 10−77 EMP3 4.25 × 10−58 CCNG1 4.06 × 10−75 SERPING1 2.10 × 10−54 DYNLL1 4.32 × 10−63 FSTL1 1.62 × 10−78 NXN 1.07 × 10−47 68 BNC2 3.57 × 10−51 LTA4H 2.69 × 10−77 MYLK 1.01 × 10−57 ATF4 4.81 × 10−75 PLSCR3 3.49 × 10−54 RWDD1 5.98 × 10−63 ADH5 2.40 × 10−78 ADNP2 1.08 × 10−47 69 ST3GAL5 8.15 × 10−51 C6orf48 3.35 × 10−77 TRPC1 1.35 × 10−57 PGAM1 9.94 × 10−75 CLIP4 4.46 × 10−54 YWHAQ 5.98 × 10−63 RPL36 4.09 × 10−78 EDEM1 1.39 × 10−47 70 AHNAK 9.52 × 10−51 EIF3E 3.47 × 10−77 OLFML1 1.86 × 10−57 PARK7 1.36 × 10−74 CLDN11 5.46 × 10−54 SKP1 8.57 × 10−63 GBE1 7.94 × 10−78 S100A6 1.66 × 10−47 71 TIPARP 9.98 × 10−51 ST13 1.04 × 10−76 COL16A1 2.92 × 10−57 PGM1 1.91 × 10−74 FYCO1 6.09 × 10−54 CTNNAL1 9.54 × 10−63 CUL3 8.09 × 10−78 FIS1 1.69 × 10−47 72 FBN1 3.72 × 10−50 ESD 1.25 × 10−76 ATP10D 6.19 × 10−57 LEPROT 4.87 × 10−74 MYLK 1.86 × 10−53 LAMP1 1.09 × 10−62 FGF2 2.12 × 10−77 RAB11FIP2 2.00 × 10−47 73 TCEAL2 5.69 × 10−50 UBA52 2.38 × 10−76 MAGEH1 6.93 × 10−57 NPC2 6.09 × 10−74 RNF38 2.24 × 10−53 IMPDH2 1.30 × 10−62 RRAGA 2.76 × 10−77 PPP1R8 2.10 × 10−47 74 SEPP1 1.54 × 10−49 MYL6 2.51 × 10−76 NAP1L3 8.78 × 10−57 RAC1 6.70 × 10−74 ST3GAL5 3.61 × 10−53 STX12 1.31 × 10−62 REEP1 3.46 × 10−77 NIPA2 2.37 × 10−47 75 TCF7L2 2.39 × 10−49 TIMP2 3.11 × 10−76 CAV2 5.95 × 10−56 PALLD 8.77 × 10−74 TGFB1I1 4.12 × 10−53 NDUFA1 1.37 × 10−62 HAS1 5.77 × 10−77 PNPO 2.38 × 10−47 76 AKT3 2.56 × 10−49 UBB 4.48 × 10−76 PDGFRL 5.01 × 10−55 PNRC2 9.95 × 10−74 TCEAL2 7.67 × 10−53 RNF11 1.40 × 10−62 RPL37 1.03 × 10−76 UBE2L6 2.77 × 10−47 77 CLDN11 2.84 × 10−49 COX6C 5.01 × 10−76 TGFBR2 5.55 × 10−55 SKP1 1.64 × 10−73 RBPMS 7.72 × 10−53 SEC11A 1.45 × 10−62 JAM3 1.16 × 10−76 ENY2 3.05 × 10−47 78 NFIB 2.88 × 10−49 TMSB10 7.86 × 10−76 GPR137B 6.65 × 10−55 COX6C 2.24 × 10−73 SULF1 8.83 × 10−53 LSM14A 1.75 × 10−62 RGL1 3.20 × 10−76 RBMX2 3.26 × 10−47 79 PDGFD 3.10 × 10−49 UROD 7.86 × 10−76 SULF1 7.43 × 10−55 TM2D3 2.24 × 10−73 AOX1 1.82 × 10−52 SCARB2 1.89 × 10−62 KLF2 4.96 × 10−76 NME4 4.03 × 10−47 80 RAB8B 7.57 × 10−49 PCNP 1.79 × 10−75 G0S2 7.82 × 10−55 PSAP 2.57 × 10−73 FOXJ3 2.84 × 10−52 TERF2IP 1.89 × 10−62 LDHA 5.59 × 10−76 TSN 5.05 × 10−47 81 HEG1 9.15 × 10−49 DSTN 3.82 × 10−75 ALDH1A3 8.41 × 10−55 NDUFA1 2.75 × 10−73 ZNF532 3.61 × 10−52 CRIM1 3.23 × 10−62 RAP1B 6.45 × 10−76 KPNA6 7.18 × 10−47 82 MAGEH1 4.66 × 10−48 RPL23A 4.49 × 10−75 PROCR 1.25 × 10−54 DYNLL1 2.83 × 10−73 FBN1 4.29 × 10−52 RASA1 3.64 × 10−62 VLDLR 1.40 × 10−75 COMMD8 8.03 × 10−47 83 GLT8D2 4.95 × 10−48 HSPA8 1.46 × 10−74 ANTXR1 3.39 × 10−54 CYB5R3 3.30 × 10−73 HEG1 1.58 × 10−51 LEPROTL1 4.46 × 10−62 TFPI 2.30 × 10−75 ASH1L 8.67 × 10−47 84 NT5E 5.68 × 10−48 HSP90AA1 1.76 × 10−74 ALDH1A2 3.79 × 10−54 KLHDC2 5.79 × 10−73 TNS3 1.99 × 10−51 TCF25 4.54 × 10−62 EHBP1 2.43 × 10−75 PEX11B 8.82 × 10−47 85 MAST4 1.08 × 10−47 ANP32B 1.97 × 10−74 PRKAR2B 4.07 × 10−54 LAMP1 6.47 × 10−73 BNC2 3.07 × 10−51 CCNG1 4.73 × 10−62 GPR176 2.50 × 10−75 MKKS 1.13 × 10−46 86 PTPRO 1.13 × 10−47 HINT1 3.85 × 10−74 CBX7 4.38 × 10−54 MXI1 7.22 × 10−73 MAP4 3.93 × 10−51 MXI1 5.81 × 10−62 RPS26 3.55 × 10−75 DUSP11 1.22 × 10−46 87 ZBED5 3.98 × 10−47 YWHAQ 5.81 × 10−74 MCC 4.49 × 10−54 ATP5J 1.12 × 10−72 AKT3 4.95 × 10−51 PSAP 8.88 × 10−62 CAPN2 3.59 × 10−75 ZMYND11 1.31 × 10−46 88 KLF2 6.13 × 10−47 EIF1 7.34 × 10−74 BEX1 1.01 × 10−53 RPL35 1.27 × 10−72 ROBO1 6.39 × 10−51 RPL35 9.78 × 10−62 EMP3 4.30 × 10−75 GCSH 1.39 × 10−46 89 OLFML1 7.40 × 10−47 RAC1 1.06 × 10−73 GPRASP1 1.21 × 10−53 YWHAQ 1.76 × 10−72 ARL3 7.05 × 10−51 MTCH1 1.06 × 10−61 SEC11A 4.40 × 10−75 MED7 1.41 × 10−46 90 RGS4 9.61 × 10−47 RPS26 1.22 × 10−73 RGS4 1.41 × 10−53 UQCRQ 1.94 × 10−72 CTSK 1.11 × 10−50 RPL8 1.08 × 10−61 MSRB2 4.95 × 10−75 C1orf54 1.48 × 10−46 91 ROBO1 1.00 × 10−46 SLC25A3 1.33 × 10−73 TBL1X 1.52 × 10−53 NEK7 2.39 × 10−72 MYH10 1.67 × 10−50 C14orf2 1.29 × 10−61 YPEL5 4.98 × 10−75 TSPO 1.55 × 10−46 92 BEX1 2.45 × 10−46 TCEAL4 1.35 × 10−73 STAT2 2.18 × 10−53 RPL37 2.57 × 10−72 EMP3 1.94 × 10−50 ARF4 1.34 × 10−61 MCC 7.04 × 10−75 ACVR2A 1.62 × 10−46 93 SULF1 2.81 × 10−46 RPS2 1.76 × 10−73 IKBKAP 2.43 × 10−53 CTNNAL1 3.40 × 10−72 MAGEH1 2.11 × 10−50 SH3BGRL 1.58 × 10−61 CAV2 9.72 × 10−75 GRSF1 1.96 × 10−46 94 FBXL7 2.96 × 10−46 RPLP2 1.77 × 10−73 NT5E 6.01 × 10−53 PTGES3 8.90 × 10−72 SALL2 2.54 × 10−50 MEIS2 1.60 × 10−61 TBC1D4 9.72 × 10−75 POLR2H 2.27 × 10−46 95 PROCR 8.86 × 10−46 DPYSL2 2.09 × 10−73 PSD3 6.66 × 10−53 UXT 9.38 × 10−72 RHOQ 2.56 × 10−50 AFF1 1.88 × 10−61 SLC25A3 1.19 × 10−74 THYN1 2.31 × 10−46 96 ABCA8 9.09 × 10−46 REEP5 2.23 × 10−73 PTRF 8.14 × 10−53 LSM14A 9.94 × 10−72 BEX1 5.32 × 10−50 PNMA1 2.20 × 10−61 LEPROTL1 1.42 × 10−74 UBE2V2 3.01 × 10−46 97 SALL2 9.31 × 10−46 MTCH1 2.59 × 10−73 PMM1 1.29 × 10−52 RPL8 1.09 × 10−71 SMARCA1 6.06 × 10−50 VAMP3 2.62 × 10−61 APPBP2 1.54 × 10−74 HMGCL 3.30 × 10−46 98 NAP1L3 9.44 × 10−46 LEPROT 3.68 × 10−73 GNAI1 7.00 × 10−52 ZFAND5 1.17 × 10−71 LDOC1 1.94 × 10−49 RPL37 3.27 × 10−61 MAPRE2 1.69 × 10−74 CSRP2 3.34 × 10−46 99 DKK3 1.89 × 10−45 UQCRQ 4.99 × 10−73 LAMA4 8.23 × 10−52 NCOA4 1.33 × 10−71 PDGFD 2.13 × 10−49 TIMP1 3.27 × 10−61 SPCS1 2.22 × 10−74 GPN3 3.68 × 10−46 100 ANXA6 2.16 × 10−45 SEC31A 5.53 × 10−73 ATP8B2 1.94 × 10−51 C14orf2 1.58 × 10−71 CTSF 2.31 × 10−49 PTGES3 5.14 × 10−61 TUBA1A 3.27 × 10−74 YIPF1 3.70 × 10−46 ==== Refs References 1. Kurman R.J. Shih Ie M. Pathogenesis of ovarian cancer: Lessons from morphology and molecular biology and their clinical implications Int. J. Gynecol. Pathol. 2008 27 151 160 10.1097/PGP.0b013e318161e4f5 18317228 2. Cho K.R. Shih Ie M. Ovarian cancer Annu. Rev. Pathol. 2009 4 287 313 10.1146/annurev.pathol.4.110807.092246 18842102 3. Eddy J.A. Hood L. Price N.D. Geman D. Identifying tightly regulated and variably expressed networks by Differential Rank Conservation (DIRAC) PLoS Comput. Biol. 2010 6 1272 10.1371/journal.pcbi.1000792 20523739 4. Ashburner M. Ball C.A. Blake J.A. Botstein D. Butler H. Cherry J.M. Davis A.P. Dolinski K. Dwight S.S. Eppig J.T. Gene ontology: Tool for the unification of biology Nat. Genet. 2000 25 25 29 10.1038/75556 10802651 5. Milacic M. Haw R. Rothfels K. Wu G. Croft D. Hermjakob H. D'Eustachio P. Stein L. Annotating cancer variants and anti-cancer therapeutics in reactome Cancers (Basel) 2012 4 1180 1211 10.3390/cancers4041180 24213504 6. Subramanian A. Tamayo P. Mootha V.K. Mukherjee S. Ebert B.L. Gillette M.A. Paulovich A. Pomeroy S.L. Golub T.R. Lander E.S. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles Proc. Natl. Acad. Sci. USA 2005 102 15545 15550 10.1073/pnas.0506580102 16199517 7. Chang C.M. Chuang C.M. Wang M.L. Yang M.J. Chang C.C. Yen M.S. Chiou S.H. Gene set-based functionome analysis of pathogenesis in epithelial ovarian serous carcinoma and the molecular features in different FIGO stages Int. J. Mol. Sci. 2016 17 886 909 10.3390/ijms17060886 27275818 8. Cortes C. Vapnik V. Support-vector networks Mach. Learn. 1995 20 273 297 10.1007/BF00994018 9. AmiGO 2 Available online: http://amigo2.berkeleybop.org/amigo (accessed on 9 January 2016) 10. Reactome Pathway Browser Available online: http://www.reactome.org/PathwayBrowser/ (accessed on 9 January 2016) 11. Frede J. Fraser S.P. Oskay-Ozcelik G. Hong Y. Ioana Braicu E. Sehouli J. Gabra H. Djamgoz M.B. Ovarian cancer: Ion channel and aquaporin expression as novel targets of clinical potential Eur. J. Cancer 2013 49 2331 2344 10.1016/j.ejca.2013.03.016 23683551 12. Yin B.W. Lloyd K.O. Molecular cloning of the CA125 ovarian cancer antigen: Identification as a new mucin, MUC16 J. Biol. Chem. 2001 276 27371 27375 10.1074/jbc.M103554200 11369781 13. Jiang X. Morland S.J. Hitchcock A. Thomas E.J. Campbell I.G. Allelotyping of endometriosis with adjacent ovarian carcinoma reveals evidence of a common lineage Cancer Res. 1998 58 1707 1712 9563487 14. Wiegand K.C. Shah S.P. Al-Agha O.M. Zhao Y. Tse K. Zeng T. Senz J. McConechy M.K. Anglesio M.S. Kalloger S.E. ARID1A mutations in endometriosis-associated ovarian carcinomas N. Engl. J. Med. 2010 363 1532 1543 10.1056/NEJMoa1008433 20942669 15. Meng A.X. Jalali F. Cuddihy A. Chan N. Bindra R.S. Glazer P.M. Bristow R.G. Hypoxia down-regulates DNA double strand break repair gene expression in prostate cancer cells Radiother. Oncol. 2005 76 168 176 10.1016/j.radonc.2005.06.025 16026872 16. Hashiguchi Y. Tsuda H. Inoue T. Berkowitz R.S. Mok S.C. PTEN expression in clear cell adenocarcinoma of the ovary Gynecol. Oncol. 2006 101 71 75 10.1016/j.ygyno.2005.09.047 16290000 17. Kuo K.T. Mao T.L. Jones S. Veras E. Ayhan A. Wang T.L. Glas R. Slamon D. Velculescu V.E. Kuman R.J. Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma Am. J. Pathol. 2009 174 1597 1601 10.2353/ajpath.2009.081000 19349352 18. Tan D.S. Iravani M. McCluggage W.G. Lambros M.B. Milanezi F. Mackay A. Gourley C. Geyer F.C. Vatcheva R. Millar J. Genomic analysis reveals the molecular heterogeneity of ovarian clear cell carcinomas Clin. Cancer Res. 2011 17 1521 1534 10.1158/1078-0432.CCR-10-1688 21411445 19. Hu L. Hofmann J. Lu Y. Mills G.B. Jaffe R.B. Inhibition of phosphatidylinositol 3′-kinase increases efficacy of paclitaxel in in vitro and in vivo ovarian cancer models Cancer Res. 2002 62 1087 1092 11861387 20. Leary J.A. Edwards B.G. Houghton C.R. Kefford R.F. Friedlander M.L. Amplification of HER-2/neu oncogene in human ovarian cancer Int. J. Gynecol. Cancer 1992 2 291 294 10.1046/j.1525-1438.1992.02060291.x 11576271 21. Tanner B. Hasenclever D. Stern K. Schormann W. Bezler M. Hermes M. Brulport M. Bauer A. Schiffer I.B. Gebhard S. ErbB-3 predicts survival in ovarian cancer J. Clin. Oncol. 2006 24 4317 4323 10.1200/JCO.2005.04.8397 16896008 22. Alexandros K. Alexandros S. Kurt H. Achim Z. Kernlab—An S4 package for kernel methods in R. journal of statistical software J. Stat. Softw. 2004 11 1 20 23. Robin X. Turck N. Hainard A. Tiberti N. Lisacek F. Sanchez J.C. Muller M. pROC: An open-source package for R and S+ to analyze and compare ROC curves BMC Bioinform. 2011 12 77 85 10.1186/1471-2105-12-77 21414208 24. Schroder M.S. Gusenleitner D. Quackenbush J. Culhane A.C. Haibe-Kains B. RamiGO: An R/Bioconductor package providing an AmiGO visualize interface Bioinformatics 2013 29 666 668 10.1093/bioinformatics/bts708 23297033

PMC005xxxxxx/PMC5000671.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081273ijms-17-01273ReviewCoadjuvants in the Diabetic Complications: Nutraceuticals and Drugs with Pleiotropic Effects Pereira Thiago Melo Costa 12Pimenta Fabio Silva 13Porto Marcella Lima 2Baldo Marcelo Perim 4Campagnaro Bianca Prandi 1Gava Agata Lages 56Meyrelles Silvana Santos 5Vasquez Elisardo Corral 15*Cai Lu Academic Editor1 Pharmaceutical Sciences Graduate Program, Vila Velha University (UVV), Av. Comissario Jose Dantas Melo 21, Boa Vista, 29102-920 Vila Velha, Brazil; [email protected] (T.M.C.P.); [email protected] (F.S.P.); [email protected] (B.P.C.)2 Federal Institute of Education, Science and Technology (IFES), 29106-010 Vila Velha, Brazil; [email protected] Burn Treatment Center, Children State Hospital, 29056-030 Vitoria, Brazil4 Department of Pathophysiology, Montes Claros State University, 39401-089, Montes Claros, Brazil; [email protected] Laboratory of Translational Physiology, Federal University of Espirito Santo (Ufes), 29047-100 Vitoria, Brazil; [email protected] (A.L.G.); [email protected] (S.S.M.)6 Division of Nephrology, McMaster University, Hamilton, ON L8N 4A6, Canada* Correspondence: [email protected]; Tel.: +55-027-999-79715205 8 2016 8 2016 17 8 127305 7 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Because diabetes mellitus (DM) is a multifactorial metabolic disease, its prevention and treatment has been a constant challenge for basic and clinical investigators focused on translating their discoveries into clinical treatment of this complex disorder. In this review, we highlight recent experimental and clinical evidences of potential coadjuvants in the management of DM, such as polyphenols (quercetin, resveratrol and silymarin), cultured probiotic microorganisms and drugs acting through direct/indirect or pleiotropic effects on glycemic control in DM. Among several options, we highlight new promising therapeutic coadjuvants, including chemical scavengers, the probiotic kefir and the phosphodiesterase 5 inhibitors, which besides the reduction of hyperglycemia and ameliorate insulin resistance, they reduce oxidative stress and improve endothelial dysfunction in the systemic vascular circulation. In the near future, experimental studies are expected to clear the intracellular pathways involving coadjuvants. The design of clinical trials may also contribute to new strategies with coadjuvants against the harmful effects of diabetic complications. diabetesquercetinpolyphenolsresveratrolsilymarinkefirprobioticsildenafilphosphodiesterase inhibitorsantioxidants ==== Body 1. Introduction Diabetes mellitus (DM) is an important public health issue because it is highly associated with increased morbidity and mortality [1]. In fact, the prevalence of diagnosed diabetes is increasing worldwide, as demonstrated by the rise from 6.5% (1999 to 2002) to 7.8% (2003 to 2006) of the population in just a few years [2]. Type 2 DM is the most common form of the disease and affects 90% to 95% of individuals with diabetes. The main issue of this pandemic is the increase in mortality associated with diabetes due to the risk of cardiovascular diseases (CVD), which are the leading cause of death in this population. This information became clear after an analysis of the First National Health and Nutrition Examination Survey (NHAHES), which covered the period of 1971–1993 and revealed that more than 65% of deaths of people with diabetes were due to CVD [3]. In addition, diabetes is a leading cause of morbidity and leads to microvascular and macrovascular complications [3,4,5]. Even with the already reported increase in the prevalence of diabetes over the years, only approximately 13% of individuals diagnosed as diabetic were in conformity with the control of the established levels of serum glucose, blood arterial pressure and total cholesterol at the same time [6]. Indeed, it is well known that most of the type 2 DM fail to control glycemia to normal levels when subjected only to diet and physical exercise and, consequently, it is necessary to treat them with anti-diabetic pharmacotherapy [1]. For instance, in a period of two years, among those patients who have been diagnosed as diabetics, the percentages of success in the control of glycated hemoglobin, blood pressure and total cholesterol are higher than 7%, 35% and 37%, respectively. As a result, good management of type 2 diabetes with pharmacological as well as non-pharmacological therapy (including reduction of caloric intake and intermittent fasting) is important [7,8]. Lately, with the mission of ameliorating this health problem, eight different classes of drugs for treatment of type 2 DM, with variations in their side effects and costs, have been approved by the US Food and Drug Administration (FDA) [9]. In this regard, investigators have been challenged to test potential therapies for DM based on functional foods, which are of low cost and very accessible (e.g., substances derived from marine algae [10]). In the present review, we discuss some epidemiological aspects of diabetic complications resulting from hyperglycemia and the therapeutic advances with antioxidant substances based on experimental and clinical studies. Among different alternatives discussed in this review, we highlight the putative coadjuvants in the management of DM, such as functional foods rich in polyphenols and the probiotic kefir. In addition, we discuss drugs with pleiotropic effects, such as phosphodiesterase 5 (PDE5) inhibitors, which lately have been the main focus of investigation in our laboratory. 2. The Impact of Chronic Hyperglycemia on Diabetic Complication 2.1. Epidemiological Aspects of Diabetes The high mortality and morbidity observed in DM patients characterized by chronic high levels of blood glucose and HbA1c, which compromises the function of the target organs heart and kidneys [11,12]. Therefore, the desired goal of treatment for diabetes is to maintain euglycemic levels as much as possible. Studies from early last century have also highlighted several effects of uncontrolled diabetes, such as dyslipidemia [13], reduced serum protein [14], skeletal muscle changes [15], and other complications. In the last decade, several epidemiological studies have been conducted to identify the risks associated to diabetes. It is well known that diabetes doubles the risk for acute coronary syndrome with an additional risk once the event has occurred. This risk was evident in the Tehran Lipid and Glucose Study, a population-based cohort study that took place in Iran. The authors found that in type 2 diabetic patients, hypercholesterolemia and central adiposity were independent risk factors for death by cardiovascular causes, and poor glycemic control is an independent risk factor for both cardiovascular and all-cause mortality [16]. In the United Kingdom, a cohort study of myocardial infarction risk in men and women with and without diabetes was carried out using a large, nationwide primary care database. The overall adjusted relative risk of myocardial infarction was higher in individuals with diabetes versus no diabetes and was greater in women compared to men [17]. After a 23-year follow-up to determine the prevalence of diabetes and associated characteristics, the Da Qing IGT and Diabetes Study showed that CVD was the leading cause of death in individuals with diabetes (47.5% in men and 49.7% in women), and almost half of the deaths were due to stroke [18]. This excessive risk of stroke associated with diabetes was significantly higher in women than men, and there were no sex differences for other major cardiovascular risk factors [19]. It is noteworthy that high glucose levels alone did not account for the increased risk associated with diabetes. A meta-analysis of 15 prospective studies in approximately 760,000 patients showed that people with pre-diabetes, which was defined as impaired fasting glucose of 110 to 124 mg/dL (6.1 to 6.9 mmol/L) or both, exhibited a moderate higher risk of stroke events [20]. Another condition associated with diabetes is chronic kidney disease. In fact, diabetes is one of the leading causes of chronic kidney disease in the United States [21], where the prevalence of diabetic nephropathy in the population of patients with type 1 and 2 DM is 20% to 40% [22,23]. The Madrid Diabetes Study, which is a prospective cohort study of 3443 type-2 diabetic outpatients, showed that the unadjusted hazard ratio for all-cause mortality in diabetic patients with eGFR < 60 mL/min/1.73 m2 was approximately 3 after five years of follow-up. Patients with chronic kidney disease at baseline had an increased risk of cardiovascular mortality [24]. 2.2. Toxic Effects of Hyperglycemia Chronic hyperglycemia can promote toxic effects in a myriad of tissues, especially in neurons, because they are more susceptible to glucose uptake [25]. Uncontrolled diabetes cam result in a pathological state characterized by severe hyperglycemia, elevation of plasma osmolarity and diabetic ketoacidosis [26]. Its classic manifestation consists of the biochemical triad of hyperglycemia, increased ketones in bloodstream, and metabolic acidosis, and it might be caused by several factors, including reduced secretion and action of insulin, and raised levels of anti-insulin hormones [27,28]. In general, patients with chronic hyperglycemia exhibit many other characteristics, such as altered expression of matrix degrading enzymes, increased synthesis and deposition of extracellular matrix (ECM), generation of advanced glycation end products (AGE), upregulation of pro-inflammatory cytokines and growth factors, and augmented flux of hexosamines and polyols [29]. Moreover, in chronic hyperglycemia conditions, the augmented glycation of intracellular proteins appears to attack other proteins and worsen the exacerbate formation of AGEs [30,31]. Consequently, it leads to the inhibition of mitochondrial respiration, increased production of reactive oxygen species (ROS) and inflammatory cytokines, culminating with marked alterations in the systemic vascular function. Also, it is well known that the augmented production of ROS causes DNA damage and results in alteration in the expression of ECM glycoproteins, which corroborates the concept that augmented oxidative stress accounts for DM complications [32,33]. In chronic hyperglycemia, aldose reductase is activated and catalyzes the first reaction in the polyol pathway, resulting in exacerbated productions and accumulation of sorbitol [34,35,36] and causing cellular toxicity by osmotic effects. NADPH is consumed and NADH is produced with accumulation of sorbitol and fructose that can also affect cellular osmosis. While there is an oversupply of NADH in individuals with diabetes due to chronic hyperglycemia and enhanced fatty acid oxidation, NAD+ could be depleted due to the activation of poly ADP ribose polymerase (PARP) by oxidative DNA damage during oxidative stress [37,38]. The activation of apoptosis in chronic hyperglycemia has received much attention in recent years. Several mechanisms regulate the complicated signaling pathways that mediate apoptosis by hyperglycemia. This process is initiated by interruption of mitochondrial electron transport, resulting in an incomplete reduction of molecular oxygen, generating superoxide anion (·O2−). This free radical can react with nitric oxide (NO), resulting in the production of peroxynitrite (ONOO−), which is a highly toxic molecule [39,40] that causes endothelial cell death. Dysfunction of endothelial cells, which causes loss of multiple endothelium-derived substances, has been hypothesized to play a key role in the progression of vascular disease in diabetes [41,42]. 2.3. Role of Oxidative Stress in Diabetic Complications Oxidative stress is induced by elevations in glucose and free fatty acid levels and has a key role in the pathogenesis of both types of DM and on diabetic complications, as has been reviewed by Wei et al. [43]. Recent evidence suggests oxidative stress is a key participant in the development and progression of diabetes as well as its micro- and macrovascular complications [44,45,46]. Paradoxically, not much attention has been given to other possible therapeutic interventions besides glucose reduction. ROS are a group of short-lived molecules derived from aerobic respiration and other oxygen reactions that include ·O2−, hydrogen peroxide (H2O2), hydroxyl radical (·OH), ONOO− and hydroxyl (OH−) [47,48]. The major sources of ROS are the mitochondria, NADPH oxidases, xanthine oxidase, uncoupled NO synthase (NOS), lipoxygenase, cyclooxygenases and CYP450, but they vary in their pathological role and importance depending on the disease and the organ [40,49]. Mitochondria and NADPH oxidases (Nox) are the most important sites for ROS production and are responsible for cardiovascular complications in diabetes [50]. In 1999, Ide et al. [51] observed enhanced cardiomyocyte mitochondrial ·O2− in the failing myocardium. Moreover, Selemidis et al. [52] suggested that NADPH is a primary ROS-producer not only in vascular smooth muscle cells but also in cardiomyocytes, vascular endothelial cells and adventitial fibroblasts. Furthermore, increased expression of Nox isoforms has been associated with myocardial hypertrophy and fibrosis in diabetes [52,53]. Hyperglycemia is characterized not only by a high-level production of ROS but also by an impairment of the intracellular antioxidant defense system, such as the nuclear factor (erythroid-derived 2)-like 2 (Nrf2), a master upregulator of several antioxidant enzymes [54,55]; consequently, the induction of genes encoding antioxidant molecules, including superoxide dismutase (SOD), glutathione peroxidase (GPx) and catalase is also affected [56]. Additionally, reduced SOD, catalase and GPx activity have been reported in both experimental and clinical diabetic conditions due to excessive glycation [57,58]. Batinic-Haberle et al. [59] found that diabetic blood vessels exhibited an improved endothelium-dependent relaxant response when treated with SOD. Interestingly, a recent study showed that the antioxidant curcumin may have a protective role against oxidative stress in diabetic mice (db/db) [60]. Therefore, it is important to emphasize that the nutraceutical compounds that require the activation of Nrf2 have been considered as relevant therapeutic strategy for prevention/treatment of diabetic complications [55,56]. Regardless of the imbalance between the generation of ROS and the activity/intracellular levels of the antioxidant defense mechanisms, excessive generation of ROS is a deleterious factor that leads to pathological consequences, including irreversible cellular damage by oxidation of proteins, lipids, carbohydrates and nucleic acids [61]. Recent evidence indicated that increased levels of urinary markers of oxidative DNA and RNA damage occur with diabetic complications [62]. Furthermore, Palem and Abraham [63] observed that diabetic patients taking both oral antidiabetic drugs and insulin still present high levels of oxidative stress, which emphasizes the need for adding antioxidants to reduce the impact of diabetic complications. In addition to the direct damage to cells, increased ROS levels also cyclically activates pathways associated with diabetes complications, such as the polyol pathway, increased production of AGEs, activation of PKC isoforms and the hexosamine pathway [54,64]. ROS overproduction and increased oxidative stress can also cause vascular endothelial and smooth muscle dysfunction. On the other hand, it has been shown that neutralization of reactive molecules in patients with diabetes was capable of preventing cardiomyopathy, retinopathy, nephropathy and neuropathy [65]. To avoid diabetes disorders, hyperglycemia should be treated promptly through stimulation of insulin secretion (not the best choice) or increasing insulin sensitivity. However, adopting a causal antioxidant therapeutic approach might be a modern adjuvant strategy to prevent the overproduction of ROS and consequently complications from diabetes. 3. Potential of Natural Products with Antioxidant Effects for Treating Diabetes Polyphenolic compounds are widely found in plants and provide several pharmacological properties, including antidiabetic effects [66,67,68]. Although not focused in the present discussion, it is important to recognize that the Chinese medicine has demonstrated the efficacy of several natural products that have been used in the treatment of DM as reviewed elsewhere [69]. In this subsection, the main polyphenols with potential antidiabetic activity investigated by us as well as others will be addressed. 3.1. Polyphenolic Compounds 3.1.1. Quercetin Quercetin (2-(3,4-dihydroxyphenyl)-3,5,7-trihydroxychromen-4-one) is the major flavonoid involved in vegetables and fruits, and it exhibits metabolic, anti-oxidative, anti-apoptotic and renoprotective effects at adequate doses [45,46,68,70]. Although this molecule is widely consumed in the diet, it was surprisingly reported as mutagenic in the 1970s in a study with unusual methods and with no reproducible results [70]. Only in 1999 did the International Agency for Research on Cancer conclude that quercetin should not be classified as carcinogenic to humans [70]. However, in parallel, the investigations with quercetin related to diabetes began in 1975 with an initial interest in preventing cataracts through the inhibition of the aldose reductase that blocks polyol accumulation in intact lenses [71]. Only in the 1990s did quercetin studies extend to other targets in diabetes complications. Many studies have demonstrated that this bioflavonoid may act through diverse pathways to decrease the tissue-damaging effects of chronic hyperglycemia, such as stimulation of glucose uptake via GLUT4 [72,73,74], inhibiting hepatic glycogenolysis and gluconeogenesis [72,75], and inhibiting α-glucosidase in the small intestine [76] or intestinal glucose transporter GLUT2 [77]. At the same time, another potential advantage is that quercetin exhibits all the characteristics of an adequate antioxidant for diabetes treatment: free radical scavenger ability [78,79], long half-life (~20 h in humans) [80,81], capacity to suppress pro-oxidant enzymes (NADPH oxidase, xanthine oxidase and CYP) [82,83,84] and the ability to stimulate antioxidant enzymes (SOD, catalase, glutathione peroxidase and glutathione reductase) [68,85,86] with high mitochondrial permeability [46,87], which are an important source of ROS in diabetes [88,89]. Given these multiple potential mechanisms, quercetin becomes an important protective molecule against the consequences of long-term diabetes (e.g., microvascular and macrovascular damage, nephropathy, neuropathy associated with the risks autonomic disturbance, amputations and foot ulcers) [67,90,91], as demonstrated in the experimental and clinical investigations as discussed below. In streptozotocin (STZ)-induced type 1 diabetes models, varying doses of quercetin have shown several benefits. At 50 mg/kg/day (oral dose), quercetin prevented retinal degeneration [92] and vascular complications by inhibiting NF-κB signaling [93]. In rats, quercetin ameliorated erectile dysfunction by inhibiting oxidative stress and upregulating eNOS [94], and it protected against the progression of neuropathy even with a low dose of quercetin (10 mg/kg) as well as attenuating cold allodynia and hyperalgesia [95]. Recently, for the first time, we demonstrated that the same low dose of quercetin attenuates hyperglycemia and nephropathy in STZ-induced diabetes in apolipoprotein E-deficient mice [46] and in C57BL/6J mice [45] (or in rats in a study conducted by others [96,97]), and quercetin treatment exhibited antioxidant benefits. With different doses (25 to 100 mg/kg/day), quercetin was also capable of suppressing the kidney inflammatory response at least partly via anti-hyperuricemic and anti-dyslipidemic effects [98]. In db/db mice (the most popular mouse model for type 2 DM), quercetin also demonstrated satisfactory effects [76]. At doses ranging between 50 and 100 mg/kg/day, quercetin treatment improved postprandial blood glucose (similarly to acarbose) [76] in addition to avoiding hyperglycemia and hyperlipidemia and increasing the antioxidant status [99]. Although experimental studies clearly support the protective effects of quercetin in diabetes, clinical data with this isolated compound are still insufficient and inconclusive. Recently, 500 mg of daily quercetin (for four weeks) was capable of reducing hyperuricemia in healthy men [100], which is a relevant factor associated with insulin resistance and progression of diabetic complications [91]. On the other hand, quercetin administered at the same dosage in women with type 2 DM, has been shown to decrease systolic arterial pressure, without significant effects on other cardiovascular risk factors [101]. Similarly, recent data from Brüll et al. [102] revealed that quercetin (162 mg/day) decreased day- and nighttime systolic blood pressure in overweight-to-obese patients without changing any other metabolic risk factor. More recently, another study reported no effect on flow-mediated dilation or insulin resistance with an analogue of quercetin (quercetin-3-glucoside, at 160 mg/day) in healthy men and women aged 40–80 years [103]. Therefore, more studies about quercetin will be necessary to establish the ideal dosage and to identify the real efficacy in diabetic patients. 3.1.2. Resveratrol This non-flavonoid polyphenolic compound (3,5,4′-trihydroxystilbene, notably present in peanuts, grapes, grape juice and red wine) might be the main molecule responsible for cardiovascular protective effects in the French population despite a high intake of saturated fats, which is known as “French Paradox” [66,104,105,106]. For that reason, this potent molecule (even with a short half-life) also would be highly beneficial as an adjuvant therapy for diabetes. Additionally, under in vitro [107,108] and in vivo [109,110,111] experimental conditions that mimic human diabetes, resveratrol has been shown to have a potential benefit in several multi-target mechanisms for diabetic complications, as presented below. Recently, Yan et al. [112] showed that 40 mg/kg/day of oral resveratrol (a high dose—according to Zhou et al. [112]) reduced proteinuria and attenuated the progress of renal fibrosis in db/db mice [112,113]. At the other extreme, it was demonstrated that a low dose of oral resveratrol (0.5 mg/kg) ameliorated classical DM symptoms (e.g., polydipsia, polyphagia, and body weight loss) and delayed the onset of insulin resistance in an STZ model [66], which probably occurred through improved glucose homeostasis. This evidence was supported by Palsamy et al., who in 2009 [114] showed decreased activity of key enzymes for gluconeogenesis by treating rats with mild diabetes (STZ-nicotinamide model) with a low dose (5 mg/kg) of resveratrol. Moreover, several in vitro studies have shown that resveratrol can increase glucose uptake by targeting insulin-affected cells (skeletal muscle, adipocytes and hepatocytes) [115,116,117,118,119], thereby improving the insulin signaling probably through improvement of insulin sensitivity in a SIRT1-dependent manner [120,121,122] or by other distinct mechanisms [66]. This stimulation of SIRT1 (a pivotal mediator of the metabolic effects of resveratrol) also may promote an increase of antioxidant enzymes (SOD, catalase, GPx and glutathione-S-transferase) in pancreatic β-cells and decrease the function of pro-inflammatory mediators (IL-6, NF-kB and COX-2) in many diabetic target tissues [119,123,124,125], which explains the relevant protective effects against apoptosis, neurodegeneration and cardiovascular complications [106,126,127]. Interestingly, resveratrol also seems to contribute to endothelial repair (which is an important tissue affected by chronic diabetes) through free-radical scavenging and/or restoration of eNOS functionality that culminates with increased bioavailability of NO [106,109,127,128] and consequently reduces diabetic complications. Recently, Neves et al. [129] showed another possible pathway of cellular protection through the regulation of cell membrane structure and fluidity (similar to cholesterol). In addition, it was observed that resveratrol might reduce endoplasmic reticulum stress by avoiding misglycosylation, depletion of calcium stores and DNA damage [127]. Therefore, resveratrol not only acts by glycemic control per se but also provides antioxidant and other pleiotropic effects [125,129,130]. Even though the preclinical evidence includes experimental evidence that clearly demonstrated that resveratrol has a significant antidiabetic effect in a wide dose range (0.1 to 1.500 mg/kg body weight), recommending resveratrol as a therapeutic supplement or treatment for diabetes patients is still controversial, and there is a similar controversy for quercetin [119,127,131,132]. This is a problem because it has generated serious doubts about the potential usefulness of these substances, particularly for dietary prevention strategies [133,134,135,136]. For instance, Thazhath et al. [137] have recently demonstrated that 1000 mg/day resveratrol in diet-controlled type-2 DM patients for five weeks did not change body weight, glycemic control or GLP-1 secretion. Similar data were also obtained by Poulsen et al. [132], who gave 1500 mg/day of resveratrol for four weeks to obese patients and found no effects on metabolic biomarkers, blood pressure or resting energy expenditure. These apparent unsuccessful studies also may be explained by variability between volunteers (age, body weight, nutrition, severity of diabetes) and/or duration of treatments [119]. In agreement with this hypothesis, another study of patients with metabolic syndrome treated the patients with 1500 mg/day of resveratrol for 90 days (~13 weeks) and revealed a significant reduction in body weight and insulin secretion [137]. Additionally, Goh et al. [138] showed improvement of insulin sensitivity via SIRT1 through 3000 mg of resveratrol for 12 weeks in type-2 diabetic patients. The advantage of this regulation is the promotion of survival and longevity, associated with telomere length [7]. Even for a shorter period of time, it was shown that resveratrol (1000 mg daily in first week followed by 2000 mg daily in second week) was able to reduce hepatic and intestinal lipoprotein production [139]. It is important to consider that although there are several investigations on the tolerability of resveratrol in humans, we cannot ignore the fact that studies about long-term resveratrol toxicity (or analogues such as pterostilbene) are still needed. 3.1.3. Silymarin Silymarin is a dry flavonoid mixture extracted after processing the seeds of Silybum marium with ethanol, methanol, and acetone [140], which contains seven major components: taxifolin (the most effective antioxidant), silychristin, silydianin, silybin A, silybin B, isosilybin A and isosilybin B [141,142]. Although silymarin has mainly been used to treat liver diseases [143], its antidiabetic activity was recently reported and is associated with an anti-glycation profile [144,145], inhibition of aldose reductase [143], partial agonist activity in peroxisome proliferator-activated receptor γ (PPARγ) [143], antioxidant capacity and radical scavenging [144,146]. All these characteristics make silymarin an interesting candidate for the prevention and treatment of diabetic complications, which has recently been demonstrated both in experimental models and in humans (the same as for quercetin and resveratrol). In 2013, Sheela et al. [146] demonstrated more fully that silymarin (60 and 120 mg/kg/day, i.m., for eight weeks) was able of reduce the classical signs of DM and attenuate the progression of the disease in a STZ-nicotinamide-induced nephropathy model (although the possible pathways were not investigated). However, in parallel, an in vitro study revealed that podocytes exposed to high glucose restored the ·O2− production and NADPH oxidase activity to basal levels through 10 μM of isolated compound silybin. In the same paper, Khazim et al. [147] obtained similar results in an in vivo experiment using 100 mg/kg/day of same substance (i.p., six weeks) in an advanced new model of diabetic nephropathy (OVE26 mice) with an additional reduction in albuminuria. These data also corroborated the findings of Vessal et al. [148], who used silymarin (100 mg/kg/day, i.p., for four weeks) in an STZ-rat model to obtain a reduction in kidney lipid peroxidation and increase the activity of catalase and GPx under hyperglycemia conditions. Further studies are necessary to explore the other cytoprotective effects of silymarin. For example, Tuorkey et al. [140] recently showed that this flavonoid mixture (120 mg/kg, i.p., for 10 days) could protect cardiomyocytes against apoptosis in diabetic (alloxan) rats via restoration of caspase-3 and Bcl-2 to control levels. For neuroprotection, silymarin (100 mg/kg/day for eight weeks) ameliorated hyperalgesia and sciatic motor nerve conduction velocity in STZ-diabetic neuropathic rat by reducing lipoperoxidation and increasing SOD activity [149]. Moreover, silibinin in db/db mice provided DNA protection and reduced oxidative stress in a brain-specific area in rodents [150]. Therefore, these preliminary studies also revealed the potential of silymarin as a valid tool to counteract oxidative stress in the central nervous system under diabetic conditions [151]. Although there are still only a few clinical studies with silymarin, the results have reflected the findings of the laboratory studies. Approximately two decades ago, a study with silymarin supplementation (600 mg/day for 12 months) was conducted in insulin-treated diabetics with alcoholic cirrhosis, and the study had encouraging results. Beyond the antioxidant effects, there was a reduction of insulin resistance, a decrease in endogenous insulin hypersecretion and a reduced need for exogenous insulin administration [152]. Corroborating this observation, 10 years later, 25 diabetic (but non-cirrhotic) patients treated with silymarin (600 mg/day, for 16 weeks) showed reductions in glycemia, glycated hemoglobin, and an improved lipid profile in liver biomarkers [153]. In addition, Hussain et al. [154] showed that silymarin (200 mg/day, for 12 weeks) could be an important adjuvant for improving the glycemic control target by increasing insulin sensitivity in peripheral tissues through sulfonylureas (glibenclamide). This finding was recently corroborated by a study in which type 2 DM patients aged 25–50 years old who were on stable medications were supplemented with silymarin (420 mg/day, for six weeks). These patients also showed improvements in some antioxidant indices (SOD, GPx and total antioxidant capacity) as well as decreased lipid peroxidation besides hs-CRP levels without reporting any adverse effects of silymarin treatment [155]. Based on these results, more studies are still needed for the evaluation of the possible synergistic effects of silymarin with other antidiabetic classes (e.g., biguanides/metformin, dipeptidyl peptidase-4 inhibitors/sitagliptin; glucagon-like peptide-1 analogues/liraglutide; sodium-glucose cotransporter 2 inhibitors/dapagliflozin). Interestingly, silymarin has also been demonstrated to be an alternative treatment for diabetic renal patients who are using the maximum doses of angiotensin-converting enzyme (ACE) inhibitors or AT1 antagonists; even after a short duration treatment with silymarin (420 mg/day, for 12 weeks), these patients showed a reduction of 50% in albuminuria, urinary TNFα levels besides serum and urinary lipid peroxidation [156,157], which reflects a potential nephroprotective activity. Although clinical trials with these polyphenols are still insufficient to define the optimal doses for treatment, the dosing range of silymarin used for diabetic patients is the closest to the ideal (when compared to diverse doses of resveratrol and quercetin) because it has clinically been investigated in several studies since the 1970s [143,158] compared to quercetin in 1995 [159] and resveratrol in 2007 [160]. The literature still describes that the therapeutic dose for the benefits of silymarin ranges between 210–800 mg/day, and silymarin appears to be safe and well tolerated up to 2100 mg/day [161], which reflects a wide therapeutic index [162]. Another additional advantage offered by silymarin in DM compared to other isolated polyphenols might be related to the relevance of silymarin use in combination with other antioxidants, which prevents individual antioxidant vulnerability and promotes synergistic effects against the chronic oxidative stress induced by diabetes [15]. 4. Beneficial Effects of Probiotics: Highlights of Treatments with Kefir In recent years, besides traditional drug treatments for DM, many efforts have been made in complementary or adjuvant therapy for the treatment of this complex disease [163]. Inadequate human dietary changes have been thought to be of major importance for the increased prevalence of DM. Overall, DM is estimated to afflict 350 million people globally and cost hundreds of billions of dollars annually [1,2]. Millions of cases could be prevented by including dietary modification to functional nutrition, which is a primary option for preventing metabolic disturbances and for reducing undesirable outcomes in DM [164,165]. Intestinal microbiota is a relevant therapeutic source for treatment of different diseases. Although there have been proposed different strategies including pre/probiotics and fecal microbiota transplantation interventions [166], in this section we review the main experimental and clinical studies that have focused on the beneficial effects of dairy cultured probiotics (live microorganisms) as coadjuvants in the prevention/treatment of this metabolic disorder. Although many studies have been focused on the identification and use of innately occurring dairy components for the prevention and correction of metabolic dysfunctions accompanying DM [167], there is a growing and remarkable body of research showing the beneficial effects of non-innately cultured probiotics or bioactive end products [168]. These health benefits are achieved by stimulating beneficial gastrointestinal indigenous microflora proliferation [169]. Fermented milk kefir, which originated in the Northern Caucasus Mountains, is now commercially available in some countries, and in others it has been domestically produced and is spreading hand-to-hand [170]. The probiotic kefir has been associated with a range of health benefits, which have been reviewed by others [171], and its continuous intake has been shown to modulate complex cardiovascular and metabolic dysfunctions, including arterial hypertension [172] and DM [164]. In contrast with non-cultured dairy products, kefir grains are small clusters of microorganisms held together by an exopolysaccharide matrix named kefiran, which is the main functional component of the beverage [169,173,174,175]. Kefir grains are produced during the fermentation of milk by a complex symbiotic mixture of yeasts as well as lactic and acetic acid bacteria [170,173,174,176]. The dominating populations of bacterial genera in cultured kefir are Lactobacillus, Lactococcus and Streptococcus [169]. Rats administered with STZ (type 1 DM) or fed a hypercaloric diet (type 2 DM) are experimental models of DM [177]. In STZ-induced DM, it has been shown that daily administration of kefir caused an improvement in the increased levels of glycemia and glucose tolerance compared to conventional fermented milk [168,178,179]. Interestingly, kefiran, which is an exopolysaccharide isolated from kefir grains, has been shown to decrease blood pressure and blood glucose in animal models of hypertension [180] and an animal model of intolerance to glucose overload [181]. Kefiran-kefir also enhanced glucose uptake into insulin-responsive muscle cells, probably through activation of PI 3-kinases or another related signaling pathway [182]. Kefir also decreased polyuria, polydipsia and polyphagia [178]. In this model of DM, it has also been shown that administration of kefir results in a decrease in total cholesterol, triglycerides, LDL-cholesterol and an increase of HDL-cholesterol levels [179]. Moreover, kefir treatment of type 1 DM rats led to a decrease in the pro-inflammatory cytokines IL-1 and IL-6 as well as an increase of anti-inflammatory IL-10 compared to control groups [167]. These studies support the concept that kefir can be useful as a complementary or adjuvant therapy for better control of glycemia. However, the mechanisms by which probiotic kefir modulate hyperglycemia are not fully understood. The beneficial effects of cultured probiotics have also been demonstrated in experimental type 2 DM. Administration of a strain of the probiotic microorganism Lactococcus lactis in rats with type 2 DM induced by a high-fructose diet resulted in significantly lower fasting blood glucose, HbA1c, insulin, free fatty acids and triglyceride levels than untreated DM rats [183]. By reducing the hyperglycemia, insulin resistance and hyperlipidemia, kefir actions were reflected in amelioration of the intracellular metabolic imbalance. In untreated DM animals, the excessive production of ROS overwhelmed the endogenous antioxidant defenses and resulted in oxidative stress, but this sequence of events can be attenuated through kefir treatment [183]. The use of antioxidant agents for therapeutic approaches in DM has been an attractive focus [184,185]. Accordingly, Friques et al. [172] observed through flow cytometry assays that kefir attenuated the endothelial dysfunction of spontaneously hypertensive rats by reducing the production of ·O2−, ONOO− and H2O2. Augmented oxidative stress has also been shown to play a role in DM [185,186,187,188] and arterial hypertension [172]. First, it was shown that kefir reduced the intracellular levels of ROS in insulin-responsive muscle cells [182]. Second, the antioxidative effects of kefir in STZ-induced DM led to an improvement in the ROS levels [178,179]. The antioxidative effect seems to be the main mechanism by which kefir reduced proteinuria and azotemia, which consequently improved the progression of renal injury in type 1 DM rats [178]. These results indicated that kefir treatment may exert beneficial effects on the oxidative stress that accompanies DM and suggests it could be used as a non-pharmacological adjuvant to delay the progression of this disease [178]. Recently, our laboratory has assessed the actions of kefir on cardiac dysautonomia and impaired baroreflex control of cardiovascular function in SHR [189]. The main action of kefir on cardiac autonomic imbalance and impaired baroreflex appears to be through attenuation of the cardiac and vascular sympathetic hyperactivity as well as augmenting cardiac parasympathetic hypoactivity [189]. Some of these effects are also expected to occur in animal models of diabetes because they present similar disturbances in the cardiovascular system [190]. For example, in the model of type 1 DM, an important imbalance of the cardiac autonomic nerves, located at both tissues and molecular pathways, has been observed. Recently, in the first 10 weeks of experimental DM, a marked cardiac dysfunction and an incomplete recovery of the cardioinhibitory vagal nerves, accompanied by a remodeling process in the stimulatory noradrenergic nerves [191], have been shown. Most clinical studies, including trials, have been conducted with patients who have type 2 DM, and most were treated with probiotic fermented milk kefir containing one, two or multi-strains of bacteria, such as Lactobacillus casei, L. acidophilus, L. bulgaricus, Streptococcus thermophiles and Bifidobacterium lactis [169]. The kefir effects observed on primary outcomes included decreased fasting blood glucose and HbA1c levels as well as improved insulin resistance [166,192,193,194]. The latter effect could be a consequence of a kefir-induced reduction in the inflammatory response [192]. In agreement with these results, it has been shown that kefir reduced pro-inflammatory cytokines, including TNFα, in DM [168,195]. The secondary outcomes included an improved lipid profile, blood pressure and hs-CRP, but kefir administration did not significantly change these parameters [191]. In contrast with experimental studies, it is still not clear whether kefir has beneficial effects on the lipid profile. An important finding after comparing the quantity of Lactobacillus and Bifidobacterium before and after the intervention was that there was successful passage of the probiotic supplement through the gastrointestinal tract [166,196]. The above studies support the concept that kefir can be useful as a complementary or adjuvant therapy for a better control of glycemia, insulin resistance and kidney function in diabetic individuals. An important characteristic of DM is endothelial dysfunction, which has been shown in experimental and clinical studies [197]. Our laboratory has demonstrated that kefir administered for at least 60 days in spontaneously hypertensive rats resulted in a significant attenuation of endothelial dysfunction [172]. Therefore, there is a need for more studies to test the hypothesis that kefir administration also exhibits benefits against this abnormality. A limitation in the therapeutic use of the probiotic kefir in DM is that this is a heterogeneous and a multiple systems-derived disease that results in multiple complications. Therefore, it makes it hard to prevent or to treat DM with traditional medicine and functional nutrition, especially when treating different people with different needs. However, there is evidence that kefir has great potential to become an adjuvant alternative for control of glycemia and other diabetes-related outcomes. Further studies are needed not only to clarify the mechanisms behind the effects of kefir but also to determine which microorganisms in kefir are responsible for its benefits. 5. Beneficial Effects of Phosphodiesterase Inhibitors in Diabetes Mellitus: New Insights Several investigations have demonstrated that increased cyclic GMP (cGMP) signaling might be an important strategy for reducing the progression of diabetes through multiple pathways [198]. Concerning glycemic control, even if an increase in intracellular calcium is the principal signal that activates insulin exocytosis, cGMP may also participate through distinct signals [199,200] and potentiating the stimulation of glucose [201,202,203]. In parallel, some in vitro studies have shown that cGMP may enhance insulin sensitivity in target organs (muscle and adipocytes) by stimulating GLUT4 recruitment into the plasma membrane [204,205,206]. In addition, because NO/cGMP signaling is fundamental to vascular protection [207,208,209], the increment of this pathway may be an attractive strategy to attenuate endothelial dysfunction development in diabetic complications, which is as major cause of disability and death in patients with DM [4,44]. Pharmacological strategies to increase cGMP signaling may be achieved through two main routes: (1) direct activation of guanylate cyclase directly by augmentation of NO; and/or (2) decreasing cGMP hydrolysis through PDE5 inhibitors (sildenafil/Viagra™, vardenafil/Levitra™, tadalafil/Cialis™, avanafil/Stendra™), which is currently considered an important tool to treat endothelial dysfunction in DM [209,210,211]. Because PDE5 is expressed in some tissues of the body (e.g., corpus cavernosum, platelets, systemic arteries and veins) [209,212], the diminished NO bioavailability in diabetic vasculature can be partially compensated through PDE5 inhibitors. Interestingly, although some recent studies demonstrated that association with antioxidants (e.g., polyphenols or vitamin E) potentiates vascular protection [213,214], sildenafil may also provide intrinsic antioxidant effects through NADPH oxidase activity inhibition [215]. This evidence was complemented by our research in various models of hypertension, nephropathy or atherosclerosis and demonstrated protective effects for endothelial, cardiac and kidney functions in physiological parameters, such as morphological analyses (Figure 1) [209,212,216,217,218,219]. Therefore, our studies and other experimental evidence support the clinical findings of improvement in endothelial function, reduction of markers of vascular inflammation, and beneficial effects for conditions beyond erectile dysfunction [220,221,222,223,224]. In relation to glycemic control, PDE5 inhibitors may have improved insulin sensitivity both in isolated human endothelial cells [225] and in high fat-fed mice [226], which corroborates the results related to cGMP that were previously discussed. Moreover, recent findings showed that chronic treatment with tadalafil reduced inflammatory cytokines, infarct size and oxidative stress in the hearts of diabetic mice by reducing NADPH oxidase activity, oxidized glutathione and lipid peroxidation [227,228]. These results show a potential role for PDE5 inhibitors in treating diabetes-related cardiac and inflammatory complications. Interestingly, clinical investigations confirmed that sildenafil could improve insulin sensitivity in addition to fibrinolytic balance and albuminuria in hyperglycemic patients [198,229]. These studies suggest that PDE inhibitors can be effective (and safely used) in patients with multiple comorbidities and therapies, except for patients treated with continuous nitrates [209,223,230]. 6. Conclusions In the present review, besides showing the importance of lifestyle modification, diet and weight control to prevent DM and its aggravation, we highlight recent experimental and clinical evidences of potential coadjuvants in the management of DM without compromising the function of β-cells via hyperinsulinism. Here, we have discussed the substances that exhibit direct/indirect or pleiotropic effects on glycemic control in DM and on oxidative stress that is one of the most contributors to the complications of this disease by affecting the different target organs. Among several options, we have highlighted new promising therapeutic coadjuvants, including the cultured probiotic microorganisms (such as kefir grains) and the PDE5 inhibitors (such as sildenafil), which besides the reduction of hyperglycemia and ameliorate insulin resistance, they have been shown to reduce the oxidative stress and improved the endothelial dysfunction in the systemic vascular circulation. In the near future, there are expected experimental studies designed to clear the intracellular pathways involving those coadjuvants discussed in this review as well as promoting clinical trials that may contribute to new strategies against the harmful effects of diabetic complications. Acknowledgments Elisardo Corral Vasquez is supported by National Council for the Development of Science and Technology (CNPq) (Grant 303001/2015-1) and State Foundation for Science and Technology (Fapes) (Grant Universal 2014 Proc 67597483). Silvana S. Meyrelles is supported by CNPq (307584/2015-1). Bianca P. Campagnaro is supported by CNPq (445736/2014-3). Thiago Melo C. Pereira is supported by CNPq (445080/2014-0). Author Contributions Elisardo Corral Vasquez supervised the analysis of publication data, and edited and approved the final version of this review. The contribution of the collaborators for each section is as follows. Section 1: Marcelo Perim Baldo and Silvana Santos Meyrelles; Section 2: Marcella Lima Porto and Agata Lages Gava; Section 3: Thiago Melo Costa Pereira and Bianca Prandi Campagnaro; Section 4: Fabio Silva Pimenta and Silvana Santos Meyrelles; Section 5: Thiago Melo Costa Pereira and Elisardo Corral Vasquez and creation of Figure 1: Thiago Melo Costa Pereira and Elisardo Corral Vasquez. Conflicts of Interest The authors declare no conflict of interest. Abbreviations ACE angiotensin-converting enzyme AGEs glycation end products cGMP cyclic guanosine monophosphate COX cyclooxygenase CRP C-reactive protein CVD cardiovascular diseases CYP450 cytochrome P450 DM diabetes mellitus ECM extracellular matrix eGFR estimated glomerular filtration rate eNOS endothelial nitric oxide synthase GLP-1 glucagon-like peptide-1 GLUT glucose transporter GPx glutathione peroxidase H2O2 hydrogen peroxide HbA1c glycated hemoglobin level HDL high density lipoprotein hs-CRP high-sensitivity C-reactive protein IGT impaired glucose tolerance IL interleukin iPDE PDE inhibitors LDL low density lipoprotein NADH nicotinamide adenine dinucleotide NADPH nicotinamide adenine dinucleotide phosphate NF-κB nuclear factor-κB NHAHES national health and nutrition examination survey NO nitric oxide NOS nitric oxide synthase Nox NADPH oxidases ·O2− superoxide anion ONOO− peroxynitrite OH− hydroxyl ·OH hydroxyl radical PARP poly ADP ribose polymerase PDE5 phosphodiesterase 5 PKC protein kinase C PPARγ proliferator-activated receptor γ ROS reactive oxygen species SHR spontaneously hypertensive rats SIRT sirtuin 1 SOD superoxide dismutase STZ streptozotocin TNFα tumor necrosis factor Figure 1 Schematic representation of metabolic complications of diabetes mellitus in two important target organs, and the main effects exhibited by three important coadjuvants currently under investigation with the aim of preventing and treating this complex disease. Arrows up (↑): increase; arrows down (↓): decrease. ==== Refs References 1. Garcia-Perez L.E. Alvarez M. Dilla T. Gil-Guillen V. Orozco-Beltran D. Adherence to therapies in patients with type 2 diabetes Diabetes Ther. 2013 4 175 194 10.1007/s13300-013-0034-y 23990497 2. Cheung B.M. Ong K.L. Cherny S.S. Sham P.C. Tso A.W. Lam K.S. Diabetes prevalence and therapeutic target achievement in the United States, 1999 to 2006 Am. J. Med. 2009 122 443 453 10.1016/j.amjmed.2008.09.047 19375554 3. Gu K. Cowie C.C. Harris M.I. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971–1993 Diabetes Care 1998 21 1138 1145 10.2337/diacare.21.7.1138 9653609 4. Potenza M.A. Gagliardi S. Nacci C. Carratu M.R. Montagnani M. Endothelial dysfunction in diabetes: From mechanisms to therapeutic targets Curr. Med. Chem. 2009 16 94 112 10.2174/092986709787002853 19149564 5. Santi D. Giannetta E. Isidori A.M. Vitale C. Aversa A. Simoni M. Therapy of endocrine disease. Effects of chronic use of phosphodiesterase inhibitors on endothelial markers in type 2 diabetes mellitus: A meta-analysis Eur. J. Endocrinol. 2015 172 R103 R114 10.1530/EJE-14-0700 25277671 6. Ong K.L. Cheung B.M. Wong L.Y. Wat N.M. Tan K.C. Lam K.S. Prevalence, treatment, and control of diagnosed diabetes in the U.S. National Health and Nutrition Examination Survey 1999–2004 Ann. Epidemiol. 2008 18 222 229 10.1016/j.annepidem.2007.10.007 18201902 7. Wegman M.P. Guo M.H. Bennion D.M. Shankar M.N. Chrzanowski S.M. Goldberg L.A. Xu J. Williams T.A. Lu X. Hsu S.I. Practicality of intermittent fasting in humans and its effect on oxidative stress and genes related to aging and metabolism Rejuvenation Res. 2015 18 162 172 10.1089/rej.2014.1624 25546413 8. Wang Y. Molecular Links between Caloric Restriction and Sir2/SIRT1 Activation Diabetes Metab. J. 2014 38 321 329 10.4093/dmj.2014.38.5.321 25349818 9. Kuecker C.M. Vivian E.M. Patient considerations in type 2 diabetes—Role of combination dapagliflozin-metformin XR Diabetes Metab. Syndr. Obes. 2016 9 25 35 26966383 10. Sharma B.R. Kim H.J. Rhyu D.Y. Caulerpa lentillifera extract ameliorates insulin resistance and regulates glucose metabolism in C57BL/KsJ-db /db mice via PI3K/AKT signaling pathway in myocytes J. Transl. Med. 2015 13 62 10.1186/s12967-015-0412-5 25889508 11. Shera A.S. Jawad F. Maqsood A. Jamal S. Azfar M. Ahmed U. Prevalence of chronic complications and associated factors in type 2 diabetes J. Pak. Med. Assoc. 2004 54 54 59 15134204 12. Siddiqui F.J. Avan B.I. Mahmud S. Nanan D.J. Jabbar A. Assam P.N. Uncontrolled diabetes mellitus: Prevalence and risk factors among people with type 2 diabetes mellitus in an Urban District of Karachi, Pakistan Diabetes Res. Clin. Pract. 2015 107 148 156 10.1016/j.diabres.2014.09.025 25451895 13. Hamwi G.J. Garcia O. Kruger F.A. Gwinup G. Cornwell D.G. Hyperlipidemia in uncontrolled diabetes Metabolism 1962 11 850 862 13904253 14. Moinat P. Nichols N. Tuller E.F. Changes in serum proteins and polysaccharides in rats with uncontrolled diabetes Diabetes 1956 5 468 474 13375444 15. Litchfield J.A. Biochemical changes in skeletal muscle in patients with uncontrolled diabetes mellitus Diabetes 1959 8 257 260 10.2337/diab.8.4.257 13663709 16. Afsharian S. Akbarpour S. Abdi H. Sheikholeslami F. Moeini A.S. Khalili D. Momenan A.A. Azizi F. Hadaegh F. Risk factors for cardiovascular disease and mortality events in adults with type 2 diabetes: A 10 year follow-up: Tehran lipid and glucose study Diabetes Metab. Res. Rev. 2016 10.1002/dmrr.2776 26787367 17. Mulnier H.E. Seaman H.E. Raleigh V.S. Soedamah-Muthu S.S. Colhoun H.M. Lawrenson R.A. de Vries C.S. Risk of myocardial infarction in men and women with type 2 diabetes in the UK: A cohort study using the General Practice Research Database Diabetologia 2008 51 1639 1645 10.1007/s00125-008-1076-y 18581091 18. An Y. Zhang P. Wang J. Gong Q. Gregg E.W. Yang W. Li H. Zhang B. Shuai Y. Chen Y. Cardiovascular and all-cause mortality over a 23-year period among chinese with newly diagnosed diabetes in the Da Qing IGT and Diabetes Study Diabetes Care 2015 38 1365 1371 10.2337/dc14-2498 25887356 19. Peters S.A. Huxley R.R. Woodward M. Diabetes as a risk factor for stroke in women compared with men: A systematic review and meta-analysis of 64 cohorts, including 775,385 individuals and 12,539 strokes Lancet 2014 383 1973 1980 10.1016/S0140-6736(14)60040-4 24613026 20. Lee M. Saver J.L. Hong K.S. Song S. Chang K.H. Ovbiagele B. Effect of pre-diabetes on future risk of stroke: Meta-analysis BMJ 2012 344 10.1136/bmj.e3564 22677795 21. Collins A.J. Foley R.N. Herzog C. Chavers B. Gilbertson D. Ishani A. Kasiske B. Liu J. Mau L.W. McBean M. US renal data system 2010 annual data report Am. J. Kidney Dis. 2011 57 e1 e526 10.1053/j.ajkd.2010.10.007 21184928 22. Radbill B. Murphy B. LeRoith D. Rationale and strategies for early detection and management of diabetic kidney disease Mayo Clin. Proc. 2008 83 1373 1381 10.4065/83.12.1373 19046557 23. de Boer I.H. Rue T.C. Hall Y.N. Heagerty P.J. Weiss N.S. Himmelfarb J. Temporal trends in the prevalence of diabetic kidney disease in the United States JAMA 2011 305 2532 2539 10.1001/jama.2011.861 21693741 24. Salinero-Fort M.Á. San Andrés-Rebollo F.J. de Burgos-Lunar C. Abánades-Herranz J.C. Carrillo-de-Santa-Pau E. Chico-Moraleja R.M. Jiménez-García R. López-de-Andrés A. Gómez-Campelo P. Cardiovascular and all-cause mortality in patients with type 2 diabetes mellitus in the MADIABETES Cohort Study: Association with chronic kidney disease J. Diabetes Complic. 2016 30 227 236 10.1016/j.jdiacomp.2015.10.007 26627635 25. Brownlee M. The pathobiology of diabetic complications: A unifying mechanism Diabetes 2005 54 1615 1625 10.2337/diabetes.54.6.1615 15919781 26. Kitabchi A.E. Umpierrez G.E. Miles J.M. Fisher J.N. Hyperglycemic crises in adult patients with diabetes Diabetes Care 2009 32 1335 1343 10.2337/dc09-9032 19564476 27. Kitabchi A.E. Nyenwe E.A. Hyperglycemic crises in diabetes mellitus: Diabetic ketoacidosis and hyperglycemic hyperosmolar state Endocrinol. Metab. Clin. N. Am. 2006 35 725 751 10.1016/j.ecl.2006.09.006 17127143 28. Davis S.N. Umpierrez G.E. Diabetic ketoacidosis in type 2 diabetes mellitus-pathophysiology and clinical presentation Nat. Clin. Pract. Endocrinol. Metab. 2007 3 730 731 10.1038/ncpendmet0641 17848925 29. Lewko B. Stepinski J. Hyperglycemia and mechanical stress: Targeting the renal podocyte J. Cell. Physiol. 2009 221 288 295 10.1002/jcp.21856 19562677 30. Di Mario U. Pugliese G. 15th Golgi lecture: From hyperglycaemia to the dysregulation of vascular remodelling in diabetes Diabetologia 2001 44 674 692 10.1007/s001250051676 11440360 31. Beisswenger P.J. Howell S.K. Nelson R.G. Mauer M. Szwergold B.S. α-Oxoaldehyde metabolism and diabetic complications Biochem. Soc. Trans. 2003 31 1358 1363 10.1042/bst0311358 14641063 32. Nishikawa T. Edelstein D. Du X.L. Yamagishi S. Matsumura T. Kaneda Y. Yorek M.A. Beebe D. Oates P.J. Hammes H.P. Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage Nature 2000 404 787 790 10783895 33. Lee H.B. Yu M.R. Yang Y. Jiang Z. Ha H. Reactive oxygen species-regulated signaling pathways in diabetic nephropathy J. Am. Soc. Nephrol. 2003 14 S241 S245 10.1097/01.ASN.0000077410.66390.0F 12874439 34. Dunlop M. Aldose reductase and the role of the polyol pathway in diabetic nephropathy Kidney Int. Suppl. 2000 77 S3 S12 10.1046/j.1523-1755.2000.07702.x 10997684 35. Tomlinson D.R. Gardiner N.J. Glucose neurotoxicity Nat. Rev. Neurosci. 2008 9 36 45 10.1038/nrn2294 18094705 36. Luo X. Wu J. Jing S. Yan L.J. Hyperglycemic stress and carbon stress in diabetic glucotoxicity Aging Dis. 2016 7 90 110 10.14336/AD.2015.0702 26816666 37. Ying W. NAD+ /NADH and NADP+ /NADPH in cellular functions and cell death: Regulation and biological consequences Antioxid. Redox Signal. 2008 10 179 206 10.1089/ars.2007.1672 18020963 38. Dolle C. Rack J.G. Ziegler M. NAD and ADP-ribose metabolism in mitochondria FEBS J. 2013 280 3530 3541 10.1111/febs.12304 23617329 39. Du X.L. Edelstein D. Rossetti L. Fantus I.G. Goldberg H. Ziyadeh F. Wu J. Brownlee M. Hyperglycemia-induced mitochondrial superoxide overproduction activates the hexosamine pathway and induces plasminogen activator inhibitor-1 expression by increasing Sp1 glycosylation Proc. Natl. Acad. Sci. USA 2000 97 12222 12226 10.1073/pnas.97.22.12222 11050244 40. Porto M.L. Rodrigues B.P. Menezes T.N. Ceschim S.L. Casarini D.E. Gava A.L. Pereira T.M. Vasquez E.C. Campagnaro B.P. Meyrelles S.S. Reactive oxygen species contribute to dysfunction of bone marrow hematopoietic stem cells in aged C57BL/6 J mice J. Biomed. Sci. 2015 22 97 10.1186/s12929-015-0201-8 26498041 41. Nakagami H. Morishita R. Yamamoto K. Yoshimura S.I. Taniyama Y. Aoki M. Matsubara H. Kim S. Kaneda Y. Ogihara T. Phosphorylation of p38 mitogen-activated protein kinase downstream of bax-caspase-3 pathway leads to cell death induced by high d -glucose in human endothelial cells Diabetes 2001 50 1472 1481 10.2337/diabetes.50.6.1472 11375350 42. Nakagami H. Kaneda Y. Ogihara T. Morishita R. Endothelial dysfunction in hyperglycemia as a trigger of atherosclerosis Curr. Diabetes Rev. 2005 1 59 63 10.2174/1573399052952550 18220582 43. Wei W. Liu Q. Tan Y. Liu L. Li X. Cai L. Oxidative stress, diabetes, and diabetic complications Hemoglobin 2009 33 370 377 10.3109/03630260903212175 19821780 44. Domingueti C.P. Dusse L.M. Carvalho M.D. de Sousa L.P. Gomes K.B. Fernandes A.P. Diabetes mellitus: The linkage between oxidative stress, inflammation, hypercoagulability and vascular complications J. Diabetes Complic. 2016 30 738 745 10.1016/j.jdiacomp.2015.12.018 26781070 45. Gomes I.B. Porto M.L. Santos M.C. Campagnaro B.P. Pereira T.M. Meyrelles S.S. Vasquez E.C. Renoprotective, anti-oxidative and anti-apoptotic effects of oral low-dose quercetin in the C57BL/6J model of diabetic nephropathy Lipids Health Dis. 2014 6 184 10.1186/1476-511X-13-184 25481305 46. Gomes I.B. Porto M.L. Santos M.C. Campagnaro B.P. Gava A.L. Meyrelles S.S. Pereira T.M. Vasquez E.C. The protective effects of oral low-dose quercetin on diabetic nephropathy in hypercholesterolemic mice Front. Physiol. 2015 6 247 10.3389/fphys.2015.00247 26388784 47. Holmström K.M. Finkel T. Cellular mechanisms and physiological consequences of redox-dependent signaling Nat. Rev. Mol. Cell Biol. 2014 15 411 421 10.1038/nrm3801 24854789 48. Kayama Y. Raaz U. Jagger A. Adam M. Schellinger I.N. Sakamoto M. Suzuki H. Toyama K. Spin J.M. Tsao P.S. Diabetic cardiovascular disease induced by oxidative stress Int. J. Mol. Sci. 2015 16 25234 25263 10.3390/ijms161025234 26512646 49. Münzel T. Gori T. Bruno R.M. Taddei S. Is oxidative stress a therapeutic target in cardiovascular disease? Eur. Heart J. 2010 31 2741 2748 10.1093/eurheartj/ehq396 20974801 50. Ceriello A. Testa R. Genovese S. Clinical implications of oxidative stress and potential role of natural antioxidants in diabetic vascular complications Nutr. Metab. Cardiovasc. Dis. 2016 26 285 292 10.1016/j.numecd.2016.01.006 27036849 51. Ide T. Tsutsui H. Kinugawa S. Utsumi H. Kang D. Hattori N. Uchida K. Arimura K. Egashira K. Takeshita A. Mitochondrial electron transport complex I is a potential source of oxygen free radicals in the failing myocardium Circ. Res 1999 85 357 363 10.1161/01.RES.85.4.357 10455064 52. Selemidis S. Sobey C.G. Wingler K. Schmidt H.H. Drummond G.R. NADPH oxidases in the vasculature: Molecular features, roles in disease and pharmacological inhibition Pharmacol. Ther. 2008 120 254 291 10.1016/j.pharmthera.2008.08.005 18804121 53. Zhou G. Li X. Hein D.W. Xiang X. Marshall J.P. Prabhu S.D. Cai L. Metallothionein suppresses angiotensin II-induced nicotinamide adenine dinucleotide phosphate oxidase activation, nitrosative stress, apoptosis, and pathological remodeling in the diabetic heart J. Am. Coll. Cardiol. 2008 52 655 666 10.1016/j.jacc.2008.05.019 18702970 54. Brownlee M. Biochemistry and molecular cell biology of diabetic complications Nature 2001 414 813 820 10.1038/414813a 11742414 55. Tan S.M. de Haan J.B. Combating oxidative stress in diabetic complications with Nrf2 activators: How much is too much? Redox Rep. 2014 19 107 117 10.1179/1351000214Y.0000000087 24559141 56. Xu X. Luo P. Wang Y. Cui Y. Miao L. Nuclear factor (erythroid-derived 2)-like 2 (NFE2L2) is a novel therapeutic target for diabetic complications J. Int. Med. Res. 2013 41 13 19 10.1177/0300060513477004 23569125 57. Murali R. Karthikeyan A. Saravanan R. Protective effects of d -limonene on lipid peroxidation and antioxidant enzymes in streptozotocin-induced diabetic rats Basic Clin. Pharmacol. Toxicol. 2013 112 175 181 10.1111/bcpt.12010 22998493 58. Ghattas M.H. Abo-Elmatty D.M. Association of polymorphic markers of the catalase and superoxide dismutase genes with type 2 diabetes mellitus DNA Cell Biol. 2012 31 1598 1603 10.1089/dna.2012.1739 22970972 59. Batinic-Haberle I. Reboucas J.S. Spasojevic I. Superoxide dismutase mimics: Chemistry, pharmacology, and therapeutic potential Antioxid. Redox Signal. 2010 13 877 918 10.1089/ars.2009.2876 20095865 60. Soto-Urquieta M.G. López-Briones S. Pérez-Vázquez V. Saavedra-Molina A. González-Hernández G.A. Ramírez-Emiliano J. Curcumin restores mitochondrial functions and decreases lipid peroxidation in liver and kidneys of diabetic db /db mice Biol. Res. 2014 47 74 10.1186/0717-6287-47-74 25723052 61. Pitocco D. Tesauro M. Alessandro R. Ghirlanda G. Cardillo C. Oxidative stress in diabetes: Implications for vascular and other complications Int. J. Mol. Sci. 2013 10.3390/ijms141121525 24177571 62. Liu X. Gan W. Zou Y. Yang B. Su Z. Deng J. Wang L. Cai J. Elevated levels of urinary markers of oxidative DNA and RNA damage in type 2 diabetes with complications Oxid. Med. Cell. Longev. 2016 2016 4323198 10.1155/2016/4323198 26770653 63. Palem S.P. Abraham P. A study on the level of oxidative stress and inflammatory markers in type 2 diabetes mellitus patients with different treatment modalities J. Clin. Diagn. Res. 2015 10.7860/JCDR/2015/14218.6449 26500896 64. Araki E. Nishikawa T. Oxidative stress: A cause and therapeutic target of diabetic complications J. Diabetes Investig. 2010 1 90 96 10.1111/j.2040-1124.2010.00013.x 24843413 65. Zatalia S.R. Sanusi H. The role of antioxidants in the pathophysiology, complications, and management of diabetes mellitus Acta Med. Indones 2013 45 141 147 23770795 66. Su H.C. Hung L.M. Chen J.K. Resveratrol, a red wine antioxidant, possesses an insulin-like effect in streptozotocin-induced diabetic rats Am. J. Physiol. Endocrinol. Metab. 2006 290 E1339 E1346 10.1152/ajpendo.00487.2005 16434553 67. Zhang Z. Li Y. Li Y. Grape seed proanthocyanidin extracts prevent hyperglycemia-induced monocyte adhesion to aortic endothelial cells and ameliorates vascular inflammation in high-carbohydrate/high-fat diet and streptozotocin-induced diabetic rats Int. J. Food Sci. Nutr. 2015 67 524 534 10.3109/09637486.2016.1154020 27121273 68. D’Andrea G. Quercetin: A flavonol with multifaceted therapeutic applications? Fitoterapia 2015 106 256 271 10.1016/j.fitote.2015.09.018 26393898 69. Liu J.Y. Chen X.X. Tang S.C. Sze S.C. Feng Y. Lee K.F. Zhang K.Y. Chinese medicines in the treatment of experimental diabetic nephropathy Chin. Med. 2016 11 6 10.1186/s13020-016-0075-z 26913057 70. Okamoto T. Safety of quercetin for clinical application (Review) Int. J. Mol. Med. 2005 16 275 278 10.3892/ijmm.16.2.275 16012761 71. Varma S.D. Mikuni I. Kinoshita J.H. Flavonoids as inhibitors of lens aldose reductase Science 1975 188 1215 1216 10.1126/science.1145193 1145193 72. Alam M.M. Meerza D. Naseem I. Protective effect of quercetin on hyperglycemia, oxidative stress and DNA damage in alloxan induced type 2 diabetic mice Life Sci. 2014 109 8 14 10.1016/j.lfs.2014.06.005 24946265 73. Xu M. Hu J. Zhao W. Gao X. Jiang C. Liu K. Liu B. Huang F. Quercetin differently regulates insulin-mediated glucose transporter 4 translocation under basal and inflammatory conditions in adipocytes Mol. Nutr. Food Res. 2014 58 931 941 10.1002/mnfr.201300510 24343960 74. Henagan T.M. Lenard N.R. Gettys T.W. Stewart L.K. Dietary quercetin supplementation in mice increases skeletal muscle PGC1α expression, improves mitochondrial function and attenuates insulin resistance in a time-specific manner PLoS ONE 2014 9 1273 10.1371/journal.pone.0089365 24586721 75. Eid H.M. Nachar A. Thong F. Sweeney G. Haddad P.S. The molecular basis of the antidiabetic action of quercetin in cultured skeletal muscle cells and hepatocytes Pharmacogn. Mag. 2015 11 74 81 25709214 76. Kim J.H. Kang M.J. Choi H.N. Jeong S.M. Lee Y.M. Kim J.I. Quercetin attenuates fasting and postprandial hyperglycemia in animal models of diabetes mellitus Nutr. Res. Pract. 2011 5 107 111 10.4162/nrp.2011.5.2.107 21556223 77. Kwon O. Eck P. Chen S. Corpe C.P. Lee J.H. Kruhlak M. Levine M. Inhibition of the intestinal glucose transporter GLUT2 by flavonoids FASEB J. 2007 21 366 377 10.1096/fj.06-6620com 17172639 78. Mahesh T. Menon V.P. Quercetin allievates oxidative stress in streptozotocin-induced diabetic rats Phytother. Res. 2004 18 123 127 10.1002/ptr.1374 15022163 79. Oršolić N. Gajski G. Garaj-Vrhovac V. Dikić D. Prskalo Z.Š. Sirovina D. DNA-protective effects of quercetin or naringenin in alloxan-induced diabetic mice Eur. J. Pharmacol. 2011 656 110 118 10.1016/j.ejphar.2011.01.021 21277296 80. Hollman P.C. van Trijp J.M. Buysman M.N. van der Gaag M.S. Mengelers M.J. de Vries J.H. Katan M.B. Relative bioavailability of the antioxidant flavonoid quercetin from various foods in man FEBS Lett. 1997 418 152 156 10.1016/S0014-5793(97)01367-7 9414116 81. De Boer V.C. Dihal A.A. van der Woude H. Arts I.C. Wolffram S. Alink G.M. Rietjens I.M. Keijer J. Hollman P.C. Tissue distribution of quercetin in rats and pigs J. Nutr. 2005 135 1718 1725 15987855 82. Bakhshaeshi M. Khaki A. Fathiazad F. Khaki A.A. Ghadamkheir E. Anti-oxidative role of quercetin derived from Allium cepa on aldehyde oxidase (OX-LDL) and hepatocytes apoptosis in streptozotocin-induced diabetic rat Asian Pac. J. Trop. Biomed. 2012 2 528 531 10.1016/S2221-1691(12)60090-2 23569964 83. Dodda D. Ciddi V. Plants used in the management of diabetic complications Indian J. Pharm. Sci. 2014 76 97 106 24843182 84. Umathe S.N. Dixit P.V. Kumar V. Bansod K.U. Wanjari M.M. Quercetin pretreatment increases the bioavailability of pioglitazone in rats: Involvement of CYP3A inhibition Biochem. Pharmacol. 2008 75 1670 1676 10.1016/j.bcp.2008.01.010 18329630 85. Sanders R.A. Rauscher F.M. Watkins J.B. Effects of quercetin on antioxidant defense in streptozotocin-induced diabetic rats J. Biochem. Mol. Toxicol. 2001 15 143 149 10.1002/jbt.11 11424224 86. Elbe H. Esrefoglu M. Vardi N. Taslidere E. Ozerol E. Tanbek K. Melatonin, quercetin and resveratrol attenuates oxidative hepatocellular injury in streptozotocin-induced diabetic rats Hum. Exp. Toxicol. 2015 34 859 868 10.1177/0960327114559993 26286521 87. Ortega R. García N. The flavonoid quercetin induces changes in mitochondrial permeability by inhibiting adenine nucleotide translocase J. Bioenerg. Biomembr. 2009 41 41 47 10.1007/s10863-009-9198-6 19296209 88. Sivitz W.I. Yorek M.A. Mitochondrial dysfunction in diabetes: From molecular mechanisms to functional significance and therapeutic opportunities Antioxid. Redox Signal. 2010 12 537 577 10.1089/ars.2009.2531 19650713 89. Kowluru R.A. Mishra M. Oxidative stress, mitochondrial damage and diabetic retinopathy Biochim. Biophys. Acta 2015 1852 2474 2483 10.1016/j.bbadis.2015.08.001 26248057 90. Gonzalez Suarez M.L. Thomas D.B. Barisoni L. Fornoni A. Diabetic nephropathy: Is it time yet for routine kidney biopsy? World J. Diabetes 2013 4 245 255 24379914 91. Kushiyama A. Tanaka K. Hara S. Kawazu S. Linking uric acid metabolism to diabetic complications World J. Diabetes 2014 5 787 795 10.4239/wjd.v5.i6.787 25512781 92. Kumar B. Gupta S.K. Nag T.C. Srivastava S. Saxena R. Jha K.A. Srinivasan B.P. Retinal neuroprotective effects of quercetin in streptozotocin-induced diabetic rats Exp. Eye Res. 2014 125 193 202 10.1016/j.exer.2014.06.009 24952278 93. Mahmoud M.F. Hassan N.A. El Bassossy H.M. Fahmy A. Quercetin protects against diabetes-induced exaggerated vasoconstriction in rats: Effect on low grade inflammation PLoS ONE 2013 8 1273 10.1371/journal.pone.0063784 23717483 94. Zhang W. Wang Y. Yang Z. Qiu J. Ma J. Zhao Z. Bao T. Antioxidant treatment with quercetin ameliorates erectile dysfunction in streptozotocin-induced diabetic rats J. Biosci. Bioeng. 2011 112 215 218 10.1016/j.jbiosc.2011.05.013 21664865 95. Anjaneyulu M. Chopra K. Quercetin attenuates thermal hyperalgesia and cold allodynia in STZ-induced diabetic rats Indian J. Exp. Biol. 2004 42 766 769 15573524 96. Anjaneyulu M. Chopra K. Quercetin, an anti-oxidant bioflavonoid, attenuates diabetic nephropathy in rats Clin. Exp. Pharmacol. Physiol. 2004 31 244 248 10.1111/j.1440-1681.2004.03982.x 15053821 97. Chen P. Chen J. Zheng Q. Chen W. Wang Y. Xu X. Pioglitazone, extract of compound Danshen dripping pill, and quercetin ameliorate diabetic nephropathy in diabetic rats J. Endocrinol. Investig. 2013 36 422 427 23211366 98. Wang C. Pan Y. Zhang Q.Y. Wang F.M. Kong L.D. Quercetin and allopurinol ameliorate kidney injury in STZ-treated rats with regulation of renal NLRP3 inflammasome activation and lipid accumulation PLoS ONE 2012 7 1273 10.1371/journal.pone.0038285 22701621 99. Jeong S.M. Kang M.J. Choi H.N. Kim J.H. Kim J.I. Quercetin ameliorates hyperglycemia and dyslipidemia and improves antioxidant status in type 2 diabetic db /db mice Nutr. Res Pract. 2012 6 201 207 10.4162/nrp.2012.6.3.201 22808343 100. Shi Y. Williamson G. Quercetin lowers plasma uric acid in pre-hyperuricaemic males: A randomised, double-blinded, placebo-controlled, cross-over trial Br. J. Nutr. 2016 115 800 806 10.1017/S0007114515005310 26785820 101. Zahedi M. Ghiasvand R. Feizi A. Asgari G. Darvish L. Does quercetin improve cardiovascular risk factors and inflammatory biomarkers in women with type 2 diabetes: A double-blind randomized controlled clinical trial Int. J. Prev. Med. 2013 4 777 785 24049596 102. Brüll V. Burak C. Stoffel-Wagner B. Wolffram S. Nickenig G. Müller C. Langguth P. Alteheld B. Fimmers R. Naaf S. Effects of a quercetin-rich onion skin extract on 24 h ambulatory blood pressure and endothelial function in overweight-to-obese patients with (pre-)hypertension: A randomised double-blinded placebo-controlled cross-over trial Br. J. Nutr. 2015 114 1263 1277 10.1017/S0007114515002950 26328470 103. Dower J.I. Geleijnse J.M. Gijsbers L. Zock P.L. Kromhout D. Hollman P.C. Effects of the pure flavonoids epicatechin and quercetin on vascular function and cardiometabolic health: A randomized, double-blind, placebo-controlled, crossover trial Am. J. Clin. Nutr. 2015 101 914 921 10.3945/ajcn.114.098590 25934864 104. Renaud S. de Lorgeril M. Wine, alcohol, platelets, and the French paradox for coronary heart disease Lancet 1992 339 1523 1526 10.1016/0140-6736(92)91277-F 1351198 105. Nishizuka T. Fujita Y. Sato Y. Nakano A. Kakino A. Ohshima S. Kanda T. Yoshimoto R. Sawamura T. Procyanidins are potent inhibitors of LOX-1: A new player in the French Paradox Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 2011 87 104 113 10.2183/pjab.87.104 21422743 106. Bonnefont-Rousselot D. Resveratrol and cardiovascular diseases Nutrients 2016 8 E250 10.3390/nu8050250 27144581 107. Lin Y.L. Chang H.C. Chen T.L. Chang J.H. Chiu W.T. Lin J.W. Chen R.M. Resveratrol protects against oxidized LDL-induced breakage of the blood-brain barrier by lessening disruption of tight junctions and apoptotic insults to mouse cerebrovascular endothelial cells J. Nutr. 2010 140 2187 2192 10.3945/jn.110.123505 20980646 108. Posadino A.M. Cossu A. Giordo R. Zinellu A. Sotgia S. Vardeu A. Hoa P.T. Nguyen le H.V. Carru C. Pintus G. Resveratrol alters human endothelial cells redox state and causes mitochondrial-dependent cell death Food Chem. Toxicol. 2015 78 10 16 10.1016/j.fct.2015.01.017 25656643 109. Zhang H. Zhang J. Ungvari Z. Zhang C. Resveratrol improves endothelial function: Role of TNFα and vascular oxidative stress Arterioscler. Thromb. Vasc. Biol. 2009 29 1164 1171 10.1161/ATVBAHA.109.187146 19478208 110. Petit M. Guihot A.L. Grimaud L. Vessieres E. Toutain B. Menet M.C. Nivet-Antoine V. Arnal J.F. Loufrani L. Procaccio V. Resveratrol improved flow-mediated outward arterial remodeling in ovariectomized rats with hypertrophic effect at high dose PLoS ONE 2016 11 1273 10.1371/journal.pone.0146148 26734763 111. Hu M. Liu B. Resveratrol via activation of LKB1-AMPK signaling suppresses oxidative stress to prevent endothelial dysfunction in diabetic mice Clin. Exp. Hypertens. 2016 5 1 7 10.3109/10641963.2015.1131288 27149559 112. Yan C. Xu W. Huang Y. Li M. Shen Y. You H. Liang X. HRD1-mediated IGF-1R ubiquitination contributes to renal protection of resveratrol in db /db mice Mol. Endocrinol. 2016 10.1210/me.2015-1277 27082896 113. Zhou Y. Chen K. He L. Xia Y. Dai W. Wang F. Li J. Li S. Liu T. Zheng Y. The Protective effect of resveratrol on concanavalin-A-induced acute hepatic injury in mice Gastroenterol. Res. Pract. 2015 2015 506390 10.1155/2015/506390 26089871 114. Palsamy P. Subramanian S. Ameliorative potential of resveratrol on proinflammatory cytokines, hyperglycemia mediated oxidative stress, and pancreatic β-cell dysfunction in streptozotocin-nicotinamide-induced diabetic rats J. Cell. Physiol. 2010 224 423 432 10.1002/jcp.22138 20333650 115. Breen D.M. Sanli T. Giacca A. Tsiani E. Stimulation of muscle cell glucose uptake by resveratrol through sirtuins and AMPK Biochem. Biophys. Res. Commun. 2008 374 117 122 10.1016/j.bbrc.2008.06.104 18601907 116. Borriello A. Cucciolla V. della Ragione F. Galletti P. Dietary polyphenols: Focus on resveratrol, a promising agent in the prevention of cardiovascular diseases and control of glucose homeostasis Nutr. Metab. Cardiovasc. Dis. 2010 20 618 625 10.1016/j.numecd.2010.07.004 20850033 117. Patel M.I. Gupta A. Dey C.S. Potentiation of neuronal insulin signaling and glucose uptake by resveratrol: The involvement of AMPK Pharmacol. Rep. 2011 63 1162 1168 10.1016/S1734-1140(11)70635-1 22180358 118. Gomez-Zorita S. Tréguer K. Mercader J. Carpéné C. Resveratrol directly affects in vitro lipolysis and glucose transport in human fat cells J. Physiol. Biochem. 2013 69 585 593 10.1007/s13105-012-0229-0 23315205 119. Szkudelski T. Szkudelska K. Resveratrol and diabetes: From animal to human studies Biochim. Biophys. Acta 2015 1852 1145 1154 10.1016/j.bbadis.2014.10.013 25445538 120. Sun C. Zhang F. Ge X. Yan T. Chen X. Shi X. Zhai Q. SIRT1 improves insulin sensitivity under insulin-resistant conditions by repressing PTP1B Cell Metab. 2007 6 307 319 10.1016/j.cmet.2007.08.014 17908559 121. Deng J.Y. Hsieh P.S. Huang J.P. Lu L.S. Hung L.M. Activation of estrogen receptor is crucial for resveratrol-stimulating muscular glucose uptake via both insulin-dependent and -independent pathways Diabetes 2008 57 1814 1823 10.2337/db07-1750 18426865 122. Wang Q. Sun X. Li X. Dong X. Li P. Zhao L. Resveratrol attenuates intermittent hypoxia-induced insulin resistance in rats: Involvement of Sirtuin 1 and the phosphatidylinositol-4,5-bisphosphate 3-kinase/AKT pathway Mol. Med. Rep. 2015 11 151 158 10.3892/mmr.2014.2762 25352008 123. Bagul P.K. Deepthi N. Sultana R. Banerjee S.K. Resveratrol ameliorates cardiac oxidative stress in diabetes through deacetylation of NFκB-p65 and histone 3 J. Nutr. Biochem. 2015 26 1298 1307 10.1016/j.jnutbio.2015.06.006 26298192 124. Diaz-Gerevini G.T. Repossi G. Dain A. Tarres M.C. Das U.N. Eynard A.R. Beneficial action of resveratrol: How and why? Nutrition 2016 32 174 178 10.1016/j.nut.2015.08.017 26706021 125. Granados-Soto V. Pleiotropic effects of resveratrol Drug News Perspect. 2003 16 299 307 10.1358/dnp.2003.16.5.829318 12942161 126. Rodrigo R. Miranda A. Vergara L. Modulation of endogenous antioxidant system by wine polyphenols in human disease Clin. Chim. Acta 2011 412 410 424 10.1016/j.cca.2010.11.034 21130758 127. Kakoti B.B. Hernandez-Ontiveros D.G. Kataki M.S. Shah K. Pathak Y. Panguluri S.K. Resveratrol and Omega-3 Fatty Acid: Its Implications in Cardiovascular Diseases Front. Cardiovasc. Med. 2015 11 38 10.3389/fcvm.2015.00038 26697434 128. Xia N. Förstermann U. Li H. Resveratrol and endothelial nitric oxide Molecules 2014 19 16102 16121 10.3390/molecules191016102 25302702 129. Neves A.R. Nunes C. Amenitsch H. Reis S. Effects of resveratrol on the structure and fluidity of lipid bilayers: A membrane biophysical study Soft Matter 2016 12 2118 2126 10.1039/C5SM02905H 26745787 130. Marques F.Z. Markus M.A. Morris B.J. Resveratrol: Cellular actions of a potent natural chemical that confers a diversity of health benefits Int. J. Biochem. Cell Biol. 2009 41 2125 2128 10.1016/j.biocel.2009.06.003 19527796 131. Cho S.J. Jung U.J. Choi M.S. Differential effects of low-dose resveratrol on adiposity and hepatic steatosis in diet-induced obese mice Br. J. Nutr. 2012 108 2166 2175 10.1017/S0007114512000347 22414733 132. Poulsen M.M. Vestergaard P.F. Clasen B.F. Radko Y. Christensen L.P. Stødkilde-Jørgensen H. Møller N. Jessen N. Pedersen S.B. Jørgensen J.O. High-dose resveratrol supplementation in obese men: An investigator-initiated, randomized, placebo-controlled clinical trial of substrate metabolism, insulin sensitivity, and body composition Diabetes 2013 62 1186 1195 10.2337/db12-0975 23193181 133. Walle T. Hsieh F. DeLegge M.H. Oatis J.E. Jr. Walle U.K. High absorption but very low bioavailability of oral resveratrol in humans Drug Metab. Dispos. 2004 32 1377 1382 10.1124/dmd.104.000885 15333514 134. la Porte C. Voduc N. Zhang G. Seguin I. Tardiff D. Singhal N. Cameron D.W. Steady-State pharmacokinetics and tolerability of trans-resveratrol 2000 mg twice daily with food, quercetin and alcohol (ethanol) in healthy human subjects Clin. Pharmacokinet. 2010 49 449 454 10.2165/11531820-000000000-00000 20528005 135. Kapetanovic I.M. Muzzio M. Huang Z. Thompson T.N. McCormick D.L. Pharmacokinetics, oral bioavailability, and metabolic profile of resveratrol and its dimethylether analog, pterostilbene, in rats Cancer Chemother. Pharmacol. 2011 68 593 601 10.1007/s00280-010-1525-4 21116625 136. Malhotra A. Bath S. Elbarbry F. An organ system approach to explore the antioxidative, anti-inflammatory, and cytoprotective actions of resveratrol Oxid. Med. Cell. Longev. 2015 2015 803971 10.1155/2015/803971 26180596 137. Thazhath S.S. Wu T. Bound M.J. Checklin H.L. Standfield S. Jones K.L. Horowitz M. Rayner C.K. Administration of resveratrol for 5 wk has no effect on glucagon-like peptide 1 secretion, gastric emptying, or glycemic control in type 2 diabetes: A randomized controlled trial Am. J. Clin. Nutr. 2016 103 66 70 10.3945/ajcn.115.117440 26607942 138. Goh K.P. Lee H.Y. Lau D.P. Supaat W. Chan Y.H. Koh A.F. Effects of resveratrol in patients with type 2 diabetes mellitus on skeletal muscle SIRT1 expression and energy expenditure Int. J. Sport Nutr. Exerc. Metab. 2014 24 2 13 10.1123/ijsnem.2013-0045 23918588 139. Dash S. Xiao C. Morgantini C. Szeto L. Lewis G.F. High-dose resveratrol treatment for 2 weeks inhibits intestinal and hepatic lipoprotein production in overweight/obese men Arterioscler. Thromb. Vasc. Biol. 2013 33 2895 2901 10.1161/ATVBAHA.113.302342 24072699 140. Tuorkey M.J. El-Desouki N.I. Kamel R.A. Cytoprotective effect of silymarin against diabetes-induced cardiomyocyte apoptosis in diabetic rats Biomed. Environ. Sci. 2015 28 36 43 25566861 141. Lee D.Y. Liu Y. Molecular structure and stereochemistry of silybin A, silybin B, isosilybin A, and isosilybin B, Isolated from Silybum marianum (milk thistle) J. Nat. Prod. 2003 66 1171 1174 10.1021/np030163b 14510591 142. Anthony K.P. Saleh M.A. Free radical scavenging and antioxidant activities of silymarin components Antioxidants 2013 2 398 407 10.3390/antiox2040398 26784472 143. Kazazis C.E. Evangelopoulos A.A. Kollas A. Vallianou N.G. The therapeutic potential of milk thistle in diabetes Rev. Diabet. Stud. 2014 11 167 174 10.1900/RDS.2014.11.167 25396404 144. Wu C.H. Huang S.M. Yen G.C. Silymarin: A novel antioxidant with antiglycation and antiinflammatory properties in vitro and in vivo Antioxid. Redox Signal. 2011 14 353 366 10.1089/ars.2010.3134 20578796 145. Shin S. Lee J.A. Kim M. Kum H. Jung E. Park D. Anti-glycation activities of phenolic constituents from Silybum marianum (Milk Thistle) flower in vitro and on human explants Molecules 2015 20 3549 3564 10.3390/molecules20033549 25706757 146. Sheela N. Jose M.A. Sathyamurthy D. Kumar B.N. Effect of silymarin on streptozotocin-nicotinamide-induced type 2 diabetic nephropathy in rats Iran J. Kidney Dis. 2013 7 117 123 23485535 147. Khazim K. Gorin Y. Cavaglieri R.C. Abboud H.E. Fanti P. The antioxidant silybin prevents high glucose-induced oxidative stress and podocyte injury in vitro and in vivo Am. J. Physiol. Ren. Physiol. 2013 305 F691 F700 10.1152/ajprenal.00028.2013 23804455 148. Vessal G. Akmali M. Najafi P. Moein M.R. Sagheb M.M. Silymarin and milk thistle extract may prevent the progression of diabetic nephropathy in streptozotocin-induced diabetic rats Ren. Fail. 2010 32 733 739 10.3109/0886022X.2010.486488 20540643 149. Baluchnejadmojarad T. Roghani M. Khastehkhodaie Z. Chronic treatment of silymarin improves hyperalgesia and motor nerve conduction velocity in diabetic neuropathic rat Phytother. Res. 2010 24 1120 1125 10.1002/ptr.3078 19960427 150. Marrazzo G. Bosco P. la Delia F. Scapagnini G. di Giacomo C. Malaguarnera M. Galvano F. Nicolosi A. Li Volti G. Neuroprotective effect of silibinin in diabetic mice Neurosci. Lett. 2011 504 252 256 10.1016/j.neulet.2011.09.041 21970972 151. Borah A. Paul R. Choudhury S. Choudhury A. Bhuyan B. Das Talukdar A. Dutta Choudhury M. Mohanakumar K.P. Neuroprotective potential of silymarin against CNS disorders: Insight into the pathways and molecular mechanisms of action CNS Neurosci. Ther. 2013 19 847 853 10.1111/cns.12175 24118806 152. Velussi M. Cernigoi A.M. De Monte A. Dapas F. Caffau C. Zilli M. Long-term (12 months) treatment with an anti-oxidant drug (silymarin) is effective on hyperinsulinemia, exogenous insulin need and malondialdehyde levels in cirrhotic diabetic patients J. Hepatol. 1997 26 871 879 10.1016/S0168-8278(97)80255-3 9126802 153. Huseini H.F. Larijani B. Heshmat R. Fakhrzadeh H. Radjabipour B. Toliat T. Raza M. The efficacy of Silybum marianum (L.) Gaertn. (silymarin) in the treatment of type II diabetes: A randomized, double-blind, placebo-controlled, clinical trial Phytother. Res. 2006 20 1036 1039 10.1002/ptr.1988 17072885 154. Hussain S.A. Silymarin as an adjunct to glibenclamide therapy improves long-term and postprandial glycemic control and body mass index in type 2 diabetes J. Med. Food 2007 10 543 547 10.1089/jmf.2006.089 17887949 155. Ebrahimpour Koujan S. Gargari B.P. Mobasseri M. Valizadeh H. Asghari-Jafarabadi M. Effects of Silybum marianum (L.) Gaertn. (silymarin) extract supplementation on antioxidant status and hs-CRP in patients with type 2 diabetes mellitus: A randomized, triple-blind, placebo-controlled clinical trial Phytomedicine 2015 22 290 296 10.1016/j.phymed.2014.12.010 25765835 156. Fallahzadeh M.K. Dormanesh B. Sagheb M.M. Roozbeh J. Vessal G. Pakfetrat M. Daneshbod Y. Kamali-Sarvestani E. Lankarani K.B. Effect of addition of silymarin to renin-angiotensin system inhibitors on proteinuria in type 2 diabetic patients with overt nephropathy: A randomized, double-blind, placebo-controlled trial Am. J. Kidney Dis. 2012 60 896 903 10.1053/j.ajkd.2012.06.005 22770926 157. Gueutin V. Gauthier M. Cazenave M. Izzedine H. Diabetic nephropathy: Emerging treatments Nephrol. Ther. 2014 10 210 215 10.1016/j.nephro.2014.02.004 24938412 158. Magliulo E. Gagliardi B. Fiori G.P. Results of a double blind study on the effect of silymarin in the treatment of acute viral hepatitis, carried out at two medical centres Med. Klin. 1978 73 1060 1065 353464 159. Hollman P.C. de Vries J.H. van Leeuwen S.D. Mengelers M.J. Katan M.B. Absorption of dietary quercetin glycosides and quercetin in healthy ileostomy volunteers Am. J. Clin. Nutr. 1995 62 1276 1282 7491892 160. Boocock D.J. Faust G.E. Patel K.R. Schinas A.M. Brown V.A. Ducharme M.P. Booth T.D. Crowell J.A. Perloff M. Gescher A.J. Phase I dose escalation pharmacokinetic study in healthy volunteers of resveratrol, a potential cancer chemopreventive agent Cancer Epidemiol. Biomark. Prev. 2007 16 1246 1252 10.1158/1055-9965.EPI-07-0022 17548692 161. Hawke R.L. Schrieber S.J. Soule T.A. Wen Z. Smith P.C. Reddy K.R. Wahed A.S. Belle S.H. Afdhal N.H. Navarro V.J. Silymarin ascending multiple oral dosing phase I study in noncirrhotic patients with chronic hepatitis C J. Clin. Pharmacol. 2010 50 434 449 10.1177/0091270009347475 19841158 162. Lecomte J. Pharmacologic properties of silybin and silymarin Rev. Med. Liege 1975 30 110 114 1224091 163. O'Connor L.M. Lentjes M.A. Luben R.N. Khaw K.T. Wareham N.J. Forouhi N.G. Dietary dairy product intake and incident type 2 diabetes: A prospective study using dietary data from a 7-day food diary Diabetologia 2014 57 909 917 10.1007/s00125-014-3176-1 24510203 164. Pasin G. Comerford K.B. Dairy foods and dairy proteins in the management of type 2 diabetes: A systematic review of the clinical evidence Adv. Nutr. 2015 6 245 259 10.3945/an.114.007690 25979490 165. Astrup A. Yogurt and dairy product consumption to prevent cardiometabolic diseases: Epidemiologic and experimental studies Am. J. Clin. Nutr. 2014 99 1235S 1242S 10.3945/ajcn.113.073015 24695891 166. Ejtahed H.S. Soroush A.R. Angoorani P. Larijani B. Hasani-Ranjbar S. Gut microbiota as a target in the pathogenesis of metabolic disorders: A new approach to novel therapeutic agents Horm. Metab. Res. 2016 48 349 358 10.1055/s-0042-107792 27203411 167. Aune D. Norat T. Romundstad P. Vatten L.J. Dairy products and the risk of type 2 diabetes: A systematic review and dose-response meta-analysis of cohort studies Am. J. Clin. Nutr. 2013 98 1066 1083 10.3945/ajcn.113.059030 23945722 168. Hadisaputro S. Djokomoeljanto R.R. Judiono Soesatyo M.H. The effects of oral plain kefir supplementation on proinflammatory cytokine properties of the hyperglycemia Wistar rats induced by streptozotocin Acta Med. Indones. 2012 44 100 104 22745139 169. Bourrie B.C. Willing B.P. Cotter P.D. The microbiota and health promoting characteristics of the fermented beverage kefir Front. Microbiol. 2016 7 647 10.3389/fmicb.2016.00647 27199969 170. Magalhães K.T. de Melo Pereira G.V. Campos C.R. Dragone G. Schwan R.F. Brazilian kefir: Structure, microbial communities and chemical composition Braz. J. Microbiol. 2011 42 693 702 10.1590/S1517-83822011000200034 24031681 171. De Oliveira Leite A.M. Miguel M.A. Peixoto R.S. Rosado A.S. Silva J.T. Paschoalin V.M. Microbiological, technological and therapeutic properties of kefir: A natural probiotic beverage Braz. J. Microbiol. 2013 44 341 349 10.1590/S1517-83822013000200001 24294220 172. Friques A.G. Arpini C.M. Kalil I.C. Gava A.L. Leal M.A. Porto M.L. Nogueira B.V. Dias A.T. Andrade T.U. Pereira T.M. Chronic administration of the probiotic kefir improves the endothelial function in spontaneously hypertensive rats J. Transl. Med. 2015 13 390 10.1186/s12967-015-0759-7 26715471 173. Zhou J. Liu X. Jiang H. Dong M. Analysis of the microflora in Tibetan kefir grains using denaturing gradient gel electrophoresis Food Microbiol. 2009 26 770 775 10.1016/j.fm.2009.04.009 19835760 174. Leite A.M. Mayo B. Rachid C.T. Peixoto R.S. Silva J.T. Paschoalin V.M. Delgado S. Assessment of the microbial diversity of Brazilian kefir grains by PCR-DGGE and pyrosequencing analysis Food Microbiol. 2012 31 215 221 10.1016/j.fm.2012.03.011 22608226 175. Hamet M.F. Londero A. Medrano M. Vercammen E. van Hoorde K. Garrote G.L. Huys G. Vandamme P. Abraham A.G. Application of culture-dependent and culture-independent methods for the identification of Lactobacillus kefiranofaciens in microbial consortia present in kefir grains Food Microbiol. 2013 36 327 334 10.1016/j.fm.2013.06.022 24010614 176. Marsh A.J. O’Sullivan O. Hill C. Ross R.P. Cotter P.D. Sequencing-based analysis of the bacterial and fungal composition of kefir grains and milks from multiple sources PLoS ONE 2013 8 1273 10.1371/journal.pone.0069371 23894461 177. Gonçalves N. Gomes-Ferreira C. Moura C. Roncon-Albuquerque R. Jr. Leite-Moreira A.F. Falcão-Pires I. Worse cardiac remodeling in response to pressure overload in type 2 diabetes mellitus Int. J. Cardiol. 2016 217 195 204 10.1016/j.ijcard.2016.04.178 27195451 178. Punaro G.R. Maciel F.R. Rodrigues A.M. Rogero M.M. Bogsan C.S. Oliveira M.N. Ihara S.S. Araujo S.R. Sanches T.R. Andrade L.C. Kefir administration reduced progression of renal injury in STZ-diabetic rats by lowering oxidative stress Nitric Oxide 2014 37 53 60 10.1016/j.niox.2013.12.012 24406684 179. Yadav H. Jain S. Sinha P.R. Oral administration of dahi containing probiotic Lactobacillus acidophilus and Lactobacillus casei delayed the progression of streptozotocin-induced diabetes in rats J. Dairy Res. 2008 75 189 195 10.1017/S0022029908003129 18474136 180. Maeda H. Zhu X. Omura K. Suzuki S. Kitamura S. Effects of an exopolysaccharide (kefiran) on lipids, blood pressure, blood glucose, and constipation Biofactors 2004 22 197 200 10.1002/biof.5520220141 15630283 181. Maeda H. Zhu X. Suzuki S. Suzuki K. Kitamura S. Structural characterization and biological activities of an exopolysaccharide kefiran produced by Lactobacillus kefiranofaciens WT-2B(T) J. Agric. Food Chem. 2004 52 5533 5338 10.1021/jf049617g 15315396 182. Teruya K. Yamashita M. Tominaga R. Nagira T. Shim S.Y. Katakura Y. Tokumaru S. Tokumaru K. Barnes D. Shirahata S. Fermented milk, Kefram-Kefir enhances glucose uptake into insulin-responsive muscle cells Cytotechnology 2002 40 107 116 10.1023/A:1023926407877 19003111 183. Yadav H. Jain S. Sinha P.R. Effect of Dahi containing Lactococcus lactis on the progression of diabetes induced by a high-fructose diet in rats Biosci. Biotechnol. Biochem. 2006 70 1255 1258 10.1271/bbb.70.1255 16717431 184. Asemi Z. Zare Z. Shakeri H. Sabihi S.S. Esmaillzadeh A. Effect of multispecies probiotic supplements on metabolic profiles, hs-CRP, and oxidative stress in patients with type 2 diabetes Ann. Nutr. Metab. 2013 10.1159/000349922 23899653 185. Rudich A. Kozlovsky N. Potashnik R. Bashan N. Oxidant stress reduces insulin responsiveness in 3T3-L1 adipocytes Am. J. Physiol. 1997 272 E935 E940 9176196 186. Rudich A. Tirosh A. Potashnik R. Hemi R. Kanety H. Bashan N. Prolonged oxidative stress impairs insulin-induced GLUT4 translocation in 3T3-L1 adipocytes Diabetes 1998 47 1562 1569 10.2337/diabetes.47.10.1562 9753293 187. Baynes J.W. Role of oxidative stress in development of complications in diabetes Diabetes 1991 40 405 412 10.2337/diab.40.4.405 2010041 188. Wolff S.P. Jiang Z.Y. Hunt J.V. Protein glycation and oxidative stress in diabetes mellitus and ageing Free Radic. Biol. Med. 1991 10 339 352 10.1016/0891-5849(91)90040-A 1855674 189. Klippel B.F. Duemke L.B. Leal M.A. Friques A.G.F. Dantas E.M. Dalvi R.F. Gava A.L. Pereira T.M.C. Andrade T.U. Meyrelles S.S. Effects of kefir on the cardiac autonomic tones and baroreflex sensitivity in spontaneously hypertensive rats Front. Physiol. 2016 7 211 10.3389/fphys.2016.00211 27375490 190. Liu Y. Liu S.X. Zheng F. Cai Y. Xie K.L. Zhang W.L. Cardiovascular autonomic neuropathy in patients with type 2 diabetes J. Diabetes Investig. 2015 10.1111/jdi.12438 27181277 191. Li X. Jiang Y.H. Jiang P. Lin H.Q. Yang J.L. Ma D.F. Wang X. Yang C.H. Analysis of heart rate variability and cardiac autonomic nerve remodeling in streptozotocin-induced diabetic rats Exp. Clin. Endocrinol. Diabetes 2015 123 272 281 10.1055/s-0035-1547258 25962404 192. Firouzi S. Majid H.A. Ismail A. Kamaruddin N.A. Barakatun-Nisak M.Y. Effect of multi-strain probiotics (multi-strain microbial cell preparation) on glycemic control and other diabetes-related outcomes in people with type 2 diabetes: A randomized controlled trial Eur. J. Nutr. 2016 10.1007/s00394-016-1199-8 26988693 193. Ostadrahimi A. Taghizadeh A. Mobasseri M. Farrin N. Payahoo L. Beyramalipoor Gheshlaghi Z. Vahedjabbari M. Effect of probiotic fermented milk (kefir) on glycemic control and lipid profile in type 2 diabetic patients: A randomized double-blind placebo-controlled clinical trial Iran J. Public Health 2015 44 228 237 25905057 194. Hulston C.J. Churnside A.A. Venables M.C. Probiotic supplementation prevents high-fat, overfeeding-induced insulin resistance in human subjects Br. J. Nutr. 2015 113 596 602 10.1017/S0007114514004097 25630516 195. Tonucci L.B. Olbrich Dos Santos K.M. Licursi de Oliveira L. Rocha Ribeiro S.M. Duarte Martino H.S. Clinical application of probiotics in type 2 diabetes mellitus: A randomized, double-blind, placebo-controlled study Clin. Nutr. 2015 10.1016/j.clnu.2015.11.011 26732026 196. Manaer T. Yu L. Zhang Y. Xiao X.J. Nabi X.H. Anti-diabetic effects of shubat in type 2 diabetic rats induced by combination of high-glucose-fat diet and low-dose streptozotocin J. Ethnopharmacol. 2015 169 269 274 10.1016/j.jep.2015.04.032 25922265 197. Kuschnerus K. Landmesser U. Kränkel N. Vascular repair strategies in type 2 diabetes: Novel insights Cardiovasc. Diagn. Ther. 2015 5 374 386 26543824 198. Ramirez C.E. Nian H. Yu C. Gamboa J.L. Luther J.M. Brown N.J. Shibao C.A. Treatment with sildenafil improves insulin sensitivity in prediabetes: A randomized, controlled trial J. Clin. Endocrinol. Metab. 2015 100 4533 4540 10.1210/jc.2015-3415 26580240 199. Gromada J. Høy M. Renström E. Bokvist K. Eliasson L. Göpel S. Rorsman P. CaM kinase II-dependent mobilization of secretory granules underlies acetylcholine-induced stimulation of exocytosis in mouse pancreatic B-cells J. Physiol. 1999 518 745 759 10.1111/j.1469-7793.1999.0745p.x 10420011 200. Fu Z. Gilbert E.R. Liu D. Regulation of insulin synthesis and secretion and pancreatic β-cell dysfunction in diabetes Curr. Diabetes Rev. 2013 9 25 53 10.2174/157339913804143225 22974359 201. Nyström T. Ortsäter H. Huang Z. Zhang F. Larsen F.J. Weitzberg E. Lundberg J.O. Sjöholm Å. Inorganic nitrite stimulates pancreatic islet blood flow and insulin secretion Free Radic. Biol. Med. 2012 53 1017 1023 10.1016/j.freeradbiomed.2012.06.031 22750508 202. Méndez-del Villar M. González-Ortiz M. Martínez-Abundis E. Pérez-Rubio K.G. Lizárraga-Valdez R. Effect of resveratrol administration on metabolic syndrome, insulin sensitivity, and insulin secretion Metab. Syndr. Relat. Disord. 2014 12 497 501 10.1089/met.2014.0082 25137036 203. Otunola G.A. Oloyede O.B. Oladiji A.T. Afolayan A.J. Selected spices and their combination modulate hypercholesterolemia-induced oxidative stress in experimental rats Biol. Res. 2014 47 5 25027235 204. Bergandi L. Silvagno F. Russo I. Riganti C. Anfossi G. Aldieri E. Ghigo D. Trovati M. Bosia A. Insulin stimulates glucose transport via nitric oxide/cyclic GMP pathway in human vascular smooth muscle cells Arterioscler. Thromb. Vasc. Biol. 2003 23 2215 2221 10.1161/01.ATV.0000107028.20478.8e 14615391 205. Tanaka T. Nakatani K. Morioka K. Urakawa H. Maruyama N. Kitagawa N. Katsuki A. Araki-Sasaki R. Hori Y. Gabazza E.C. Nitric oxide stimulates glucose transport through insulin-independent GLUT4 translocation in 3T3-L1 adipocytes Eur. J. Endocrinol. 2003 149 61 67 10.1530/eje.0.1490061 12824867 206. Lira V.A. Soltow Q.A. Long J.H. Betters J.L. Sellman J.E. Criswell D.S. Nitric oxide increases GLUT4 expression and regulates AMPK signaling in skeletal muscle Am. J. Physiol. Endocrinol. Metab. 2007 293 E1062 E1068 10.1152/ajpendo.00045.2007 17666490 207. Furchgott R.F. Zawadzki J.V. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine Nature 1980 288 373 376 10.1038/288373a0 6253831 208. Ignarro L.J. Nitric oxide: A unique endogenous signaling molecule in vascular biology Biosci. Rep. 1999 19 51 71 10.1023/A:1020150124721 10888468 209. Vasquez E.C. Gava A.L. Graceli J.B. Balarini C.M. Campagnaro B.P. Pereira T.M. Meyrelles S.S. Novel therapeutic targets for phosphodiesterase 5 inhibitors: Current state-of-the-art on systemic arterial hypertension and atherosclerosis Curr. Pharm. Biotechnol. 2016 17 347 364 10.2174/1389201017666151223123904 26696017 210. Behr-Roussel D. Oudot A. Caisey S. Coz O.L. Gorny D. Bernabé J. Wayman C. Alexandre L. Giuliano F.A. Daily treatment with sildenafil reverses endothelial dysfunction and oxidative stress in an animal model of insulin resistance Eur. Urol. 2008 53 1272 1280 10.1016/j.eururo.2007.11.018 18036727 211. Maiorino M.I. Bellastella G. Esposito K. Diabetes and sexual dysfunction: Current perspectives Diabetes Metab. Syndr. Obes. 2014 7 95 105 24623985 212. Leal M.A. Balarini C.M. Dias A.T. Porto M.L. Gava A.L. Pereira T.M. Meyrelles S.S. Vasquez E.C. Mechanisms of enhanced vasoconstriction in the mouse model of atherosclerosis: The beneficial effects of sildenafil Curr. Pharm. Biotechnol. 2015 16 517 530 10.2174/138920101606150407113458 25860063 213. Bai Y. An R. Resveratrol and sildenafil synergistically improve diabetes-associated erectile dysfunction in streptozotocin-induced diabetic rats Life Sci. 2015 135 43 48 10.1016/j.lfs.2015.04.020 26006044 214. Dalaklioglu S. Bayram Z. Tasatargil A. Ozdem S. Resveratrol reverses diabetes-related decrement in sildenafil-induced relaxation of corpus cavernosum in aged rats Aging Clin. Exp. Res. 2016 10.1007/s40520-016-0582-x 27170309 215. Bivalacqua T.J. Sussan T.E. Gebska M.A. Strong T.D. Berkowitz D.E. Biswal S. Burnett A.L. Champion H.C. Sildenafil inhibits superoxide formation and prevents endothelial dysfunction in a mouse model of secondhand smoke induced erectile dysfunction J. Urol. 2009 181 899 906 10.1016/j.juro.2008.10.062 19095260 216. Balarini C.M. Leal M.A. Gomes I.B. Pereira T.M. Gava A.L. Meyrelles S.S. Vasquez E.C. Sildenafil restores endothelial function in the apolipoprotein E knockout mouse J. Transl. Med. 2013 11 3 10.1186/1479-5876-11-3 23289368 217. Dias A.T. Cintra A.S. Frossard J.C. Palomino Z. Casarini D.E. Gomes I.B. Balarini C.M. Gava A.L. Campagnaro B.P. Pereira T.M. Inhibition of phosphodiesterase 5 restores endothelial function in renovascular hypertension J. Transl. Med. 2014 12 250 10.1186/s12967-014-0250-x 25223948 218. Fahning B.M. Dias A.T. Oliveira J.P. Gava A.L. Porto M.L. Gomes I.B. Nogueira B.V. Campagnaro B.P. Pereira T.M. Vasquez E.C. Sildenafil improves vascular endothelial structure and function in renovascular hypertension Curr. Pharm. Biotechnol. 2015 16 823 831 10.2174/1389201016666150610161330 26059106 219. de Almeida L.S. Barboza J.R. Freitas F.P. Porto M.L. Vasquez E.C. Meyrelles S.S. Gava A.L. Pereira T.M. Sildenafil prevents renal dysfunction in contrast media-induced nephropathy in Wistar rats Hum. Exp. Toxicol. 2016 10.1177/0960327115626582 26791539 220. Aversa A. Vitale C. Volterrani M. Fabbri A. Spera G. Fini M. Rosano G.M. Chronic administration of Sildenafil improves markers of endothelial function in men with Type 2 diabetes Diabet Med. 2008 25 37 44 10.1111/j.1464-5491.2007.02298.x 18199130 221. Kondoh N. Higuchi Y. Maruyama T. Nojima M. Yamamoto S. Shima H. Salvage therapy trial for erectile dysfunction using phosphodiesterase type 5 inhibitors and vitamin E: Preliminary report Aging Male 2008 11 167 170 10.1080/13685530802273624 19172547 222. Mandosi E. Giannetta E. Filardi T. Lococo M. Bertolini C. Fallarino M. Gianfrilli D. Venneri M.A. Lenti L. Lenzi A. Endothelial dysfunction markers as a therapeutic target for Sildenafil treatment and effects on metabolic control in type 2 diabetes Expert Opin. Ther. Targets 2015 19 1617 1622 10.1517/14728222.2015.1066337 26178526 223. Balhara Y.P. Sarkar S. Gupta R. Phosphodiesterase-5 inhibitors for erectile dysfunction in patients with diabetes mellitus: A systematic review and meta-analysis of randomized controlled trials Indian J. Endocrinol. Metab. 2015 19 451 461 10.4103/2230-8210.159023 26180759 224. Das A. Durrant D. Salloum F.N. Xi L. Kukreja R.C. PDE5 inhibitors as therapeutics for heart disease, diabetes and cancer Pharmacol. Ther. 2015 147 12 21 10.1016/j.pharmthera.2014.10.003 25444755 225. Mammi C. Pastore D. Lombardo M.F. Ferrelli F. Caprio M. Consoli C. Tesauro M. Gatta L. Fini M. Federici M. Sildenafil reduces insulin-resistance in human endothelial cells PLoS ONE 2011 6 1273 10.1371/journal.pone.0014542 21297971 226. Ayala J.E. Bracy D.P. Julien B.M. Rottman J.N. Fueger P.T. Wasserman D.H. Chronic treatment with sildenafil improves energy balance and insulin action in high fat-fed conscious mice Diabetes 2007 56 1025 1033 10.2337/db06-0883 17229936 227. Varma A. Das A. Hoke N.N. Durrant D.E. Salloum F.N. Kukreja R.C. Anti-inflammatory and cardioprotective effects of tadalafil in diabetic mice PLoS ONE 2012 7 1273 10.1371/journal.pone.0045243 23028874 228. Koka S. Das A. Salloum F.N. Kukreja R.C. Phosphodiesterase-5 inhibitor tadalafil attenuates oxidative stress and protects against myocardial ischemia/reperfusion injury in type 2 diabetic mice Free Radic. Biol. Med. 2013 60 80 88 10.1016/j.freeradbiomed.2013.01.031 23385031 229. Grover-Páez F. Villegas Rivera G. Guillén Ortíz R. Sildenafil citrate diminishes microalbuminuria and the percentage of A1c in male patients with type 2 diabetes Diabetes Res. Clin. Pract. 2007 78 136 140 10.1016/j.diabres.2007.02.006 17374416 230. Chrysant S.G. Chrysant G.S. The pleiotropic effects of phosphodiesterase 5 inhibitors on function and safety in patients with cardiovascular disease and hypertension J. Clin. Hypertens. 2012 14 644 649 10.1111/j.1751-7176.2012.00669.x 22947364

PMC005xxxxxx/PMC5000672.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081274ijms-17-01274Article1-(2-Hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedione Induces G1 Cell Cycle Arrest and Autophagy in HeLa Cervical Cancer Cells Tsai Jie-Heng 1†Hsu Li-Sung 12†Huang Hsiu-Chen 3Lin Chih-Li 45Pan Min-Hsiung 6Hong Hui-Mei 57Chen Wei-Jen 57*Wong Kwong-Kwok Academic Editor1 Institute of Biochemistry, Microbiology and Immunology, Chung Shan Medical University, Taichung 402, Taiwan; [email protected] (J.-H.T.); [email protected] (L.-S.H.)2 Clinical Laboratory, Chung Shan Medical University Hospital, Taichung 402, Taiwan3 Department of Applied Science, National Hsinchu University of Education, Hsinchu 300, Taiwan; [email protected] Institute of Medicine, Chung Shan Medical University, Taichung 402, Taiwan; [email protected] Department of Medical Research, Chung Shan Medical University Hospital, Taichung 402, Taiwan; [email protected] Institute of Food Science and Technology, National Taiwan University, Taipei 106, Taiwan; [email protected] Department of Biomedical Sciences, Chung Shan Medical University, Taichung 402, Taiwan* Correspondence: [email protected]; Tel.: +886-4-2473-0022 (ext. 11808)† These authors contributed equally to this work. 05 8 2016 8 2016 17 8 127404 6 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The natural agent, 1-(2-hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedione (HMDB), has been reported to have growth inhibitory effects on several human cancer cells. However, the role of HMDB in cervical cancer remains unclear. Herein, we found that HMDB dose- and time-dependently inhibited growth of HeLa cervical cancer cells, accompanied with G1 cell cycle arrest. HMDB decreased protein expression of cyclins D1/D3/E and cyclin-dependent kinases (CDKs) 2/4/6 and reciprocally increased mRNA and protein levels of CDK inhibitors (p15, p16, p21, and p27), thereby leading to the accumulation of hypophosphorylated retinoblastoma (Rb) protein. HMDB also triggered the accumulation of acidic vesicles and formation of microtubule-associated protein-light chain 3 (LC3), followed by increased expression of LC3 and Beclin-1 and decreased expression of p62, suggesting that HMDB triggered autophagy in HeLa cells. Meanwhile, suppression of the expression of survivin and Bcl-2 implied that HMDB-induced autophagy is tightly linked to apoptosis. Exploring the action mechanism, HMDB induced autophagy via the modulation of AMP-activated protein kinase (AMPK) and mTOR signaling pathway rather than the class III phosphatidylinositol 3-kinase pathway. These results suggest that HMDB inhibits HeLa cell growth by eliciting a G1 arrest through modulation of G1 cell cycle regulators and by concomitantly inducing autophagy through the mediation of AMPK-mTOR and Akt-mTOR pathways, and may be a promising antitumor agent against cervical cancer. 1-(2-hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedioneG1 arrestautophagylight chain 3 (LC3)Beclin-1p62 ==== Body 1. Introduction Cervical cancer is one of the leading causes of gynecologic cancer death in women worldwide and approximately 500,000 new cervical cancer cases are deduced, contributing to 280,000 deaths each year [1]. More than 80% of cervical cancer patients are diagnosed in developing countries [2], and new cases suffering from cervical cancer are approximately 150,000 in China per year, accounting for about 30% of new cases worldwide. More than 99.7% of cervical cancer cases contain one or more of the oncogenic human papillomavirus (HPV) genotypes that cause cervical cancer [3]. HPV infection has been considered as the most key factor responsible for the development of cervical cancer; especially, HPV 16, 18, 31, and 33 infections are characterized as the primary risk factors highly associated with cervical cancer. Among them, HPV-16 and -18 infections account for about 70% of cervical cancer cases [4]. Two primary HPV viral oncoproteins, E6 and E7, are required for the development of cervical cancer with the transformed phenotypes. For example, E6 protein induces p53 degradation by the ubiquitin-proteasome mediated pathway. E7 protein interacts with retinoblastoma (Rb) protein and preventing Rb binding to cell cycle-related transcription factor E2F [5,6], which give rise to the loss of Rb/E2F complexes, the release of E2F, and the subsequent progression of cell cycle from G1 to S phase [7,8]. Abnormal regulation of cell cycle is a result of cancer development [9]. In mammals, cell cycle progression is strictly for the regulation of a set of proteins, including cyclin-dependent kinases (CDKs), cyclins, and CDK inhibitors (CKIs) that control cell cycle progression at G1, S, and G2/M checkpoints [10]. In early stage of G1 phase, the D-type cyclins emerge and accumulate in the nucleus due to mitogenic signal firing, and then forms cyclin-CDK4/6 complexes, resulting in the initial phosphorylation of Rb proteins. In late stage of G1 phase, the cyclin E/CDK2 complex further promotes the formation of highly phosphorylated Rb protein, thus releasing E2F and eventually leading to the entry of the S phase. The mechanism how to regulate the activation of CDKs is well-established. Generally, the activity of CDKs can be mediated by altering their phosphorylation status on a conserved threonine residue or by interacting with CKIs [11]. CKIs consist of two families, including the Ink4 and the Cip/Kip families [12]. The p15, p16, p18, and p19 proteins which belong to the Ink4b family bind to CDK4/6 to prevent the formation of CDK4/6-cyclin complexes. The high protein levels of the Cip/Kip family (such as p21, p27, and p57) inactivate CDK2 activity, most probably by leading to the stoichiometry in the CDK2-cycle E complexes [12]. In addition, CDK inhibitors induce autophagy in cancer-associated fibroblasts and cancer cells [13,14]. Autophagy, a process for major intracellular degradation, occurs when cells undergo stress conditions, such as nutrient starvation, exposure of radiation or cytotoxic compounds, or suffering from cancer, to promote cell survival or to result in type II programmed cell death [15]. During autophagy, the cytoplasm components or organelles for determined degradation are conveyed to the double-membrane vesicle, known as autophagosome, and then acidified for maturation to pass into acidic vesicular organelles (AVOs) [16]. Eventually, the AVOs fused with lysosomes to form autophagolysosomes which digest their internal components by lysosomal hydrolases [17,18]. Beclin-1 and microtubule-associated protein 1A/1B-light chain 3 (LC3), two hallmarks of autophagy, regulate the beginning of mammalian autophagy [19]. Beclin-1 plays a role in involving in the signaling pathway required for the induction of autophagy and in the onset of autophagosome formation [19,20]. The overexpression of Beclin-1 inhibits the proliferation and growth of HeLa cells in vitro and in vivo, while inducing autophagy and subsequent apoptosis of HeLa cells [21]. LC3 consists of a soluble form, LC3I, and a lapidated form, known as LC3II. The presence of LC3II is directly linked to autophagy, because LCII is recruited in the formation of autophagosomes. The cellular stress triggers LC3I conjugated to phosphatidylethanolamine to constitute the lapidated LC3II, which is a component of autophagosomes and so far conceived as a marker of autophagy [20,22]. Compared to normal cervical epithelial tissues and cervical squamous carcinoma tissues, Beclin-1 is highly expressed in 96.2% (25/26) versus 28.0% (14/50) of cervical cancer patients, and LC3 is highly expressed in 76.9% (20/26) and 26.0% (13/50) of cervical cancer patients, respectively [23], suggesting the induction of autophagy may be an accessible tactic for cervical cancer therapy. 1-(2-Hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedione (HMDB) is a β-diketone structural compound that has growth inhibitory effects on several human cancer cells [24,25]. HMDB was suggested to function as an anticancer drug via modulating the mitochondrial functions that are regulated by reactive oxygen species, upregulating CCAAT/enhancer binding protein delta (CEBPD), growth arrest DNA damage-inducible gene 153 (GADD153), BAD, and p21, and downregulating BCL2L1 (BCL-XL) [26]. However, the growth-inhibitory and autophagy-inducing effects of HMDB on cervical cancer have not yet been elucidated. To this end, the HeLa cervical cancer cell line was employed as an in vitro model to explore the anti-cancer effect of HMDB, focusing on the induction of autophagy and the resultant growth inhibition. Moreover, the effects of HMDB on the activation of several kinases and the following signaling pathways critically responsible for cell autophagy were explored. The present study may provide novel evidence that HMDB may be a potent cancer chemopreventive agent against some types of cervical carcinomas. 2. Results 2.1. Inhibition of Cell Growth and Cell Cycle Progression at G1 Phase of HeLa Cells by 1-(2-Hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedione (HMDB) To investigate the growth inhibitory activity of HMDB (Figure 1a), we first examined the growth-inhibitory effect of HMDB on HeLa cells using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (MTT) and trypan blue exclusion assays, respectively. Human HeLa cervical cancer cells were treated with various concentrations of HMDB at the indicated time points. As shown in Figure 1b,c, HeLa cell growth was inhibited by HMDB in a dose- and time-dependent manner. HMDB also time-dependently reduced the expression of nuclear antigen proliferating cell nuclear antigen (PCNA), a hallmark of proliferation expressed in proliferating cells. Next, to determine whether HMDB has an effect on cell cycle, we analyzed the effect of HMDB on the cell cycle distribution by flow cytometry using PI staining. As shown in Figure 1d, HMDB at 40 μM caused G1 cell cycle arrest in a time-dependent manner. In particular, approximately 58.6% of the untreated HeLa cells were in the G1 phase, while the cells exposed to 40 μM HMDB exhibited a considerably greater proportion of G1 cells (approximately 80.02%). The increased number of cells in the G1 phase after HMDB treatment was tightly associated with the decreased number of cells in the S and G2/M phases compared to the control. 2.2. Modulation of the Expression of G1 Cell Cycle Checkpoint Regulators by HMDB in HeLa Cells Given that HMDB induces G1 cell cycle arrest in HeLa cells, we investigated whether HMDB treatment changes the expression profile of cell cycle regulatory proteins such as cyclin D, cyclin E, and their associated CDK4/6 and CDK2, required for G1 to S transition in cell cycle. HeLa cells were treated with 40 μM HMDB for the indicated time points and then cell extracts were harvested for Western blotting. As demonstrated in Figure 2a, HMDB distinctly reduced the protein expression of cyclin D1/D3/E, and CDK4/6/2 in a time-dependent manner. These results indicate that inhibition of the expression of G1 phase-related cyclins and CDKs might be a critical event in the HMDB-mediated growth arrest in HeLa cells. The phosphorylation of the Rb protein mediated by G1-related cyclin/CDK complexes is necessary for cell cycle progression from G1 to S phase. To assess whether the down-regulation of the expression of cyclins and CDKs by HMDB can lead to the dephosphorylation of the Rb protein, the phosphorylation status of the Rb protein was determined by Western blotting using specific antibodies against the phosphorylated Rb protein after exposure of exponentially-growing HeLa cells to HMDB. As illustrated in Figure 2b, the Rb phosphorylation at Ser780, 807, and 811, associated with the regulation of G1 cell cycle progression were time-dependent inhibited by HMDB from 6–24 h treatment, paralleled with a decrease in the protein levels of cyclin D1/D3/E and CDK4/6. These findings provide evidence that HMDB induces cell cycle arrest at G1 phase via downregulating the expression of cyclins (D1, D3, and E) and CDKs (CDK4 and CDK6). CKIs are well characterized to prevent the progression of cell cycle from binding and inactivating CDKs alone or cyclin/CDK complexes. To assess the effect of HMDB on the expression of CKIs, we incubated HeLa cells with 40 μM HMDB for the indicated times and then examined determined the protein and mRNA expression levels of CKIs (p15, p16, p21, and p27) by Western blotting and qPCR, respectively. As shown in Figure 2c,d, HMDB clearly resulted in the increase in both protein and mRNA expression of all these CKIs in a time-dependent manner. These results indicate that HMDB may cause the induction of steady-state levels of these CKIs by regulating the transcription of these proteins. 2.3. Induction of Cytoplasmic Vacuolation, Formation of Autolysosomes, and Accumulation of Acidic Vesicles in HMDB-Treated HeLa Cells As shown in Figure 3a, upper panel, we observed that HMDB induced a time-dependent increase in the formation of intracellular vacuoles in HeLa cells. In addition, we found that the vacuolar content was acidic through a neutral red staining shown in the lower panel, suggesting the presence of lysosomal content. Combined with the growth inhibition by HMDB, we suggested that HMDB-induced vacuolization in HeLa cells may be autophagic. To demonstrate that these acidic vesicles induced by HMDB are linked to autophagy, the formation of autolysosomes was detected by monodansylcadaverine (MDC) and acridine orange (AO) staining, respectively, which are two remarkable signs of autophagy. As shown in Figure 3b, HMDB elicited a time-dependent increase in MDC- and AO-stained cells, respectively; suggesting autophagy-mediated cell death may be, at least partly, involved in the action mechanism of HMDB-induced growth inhibition in HeLa cells. 2.4. Formation of Autophagic Vacuoles with the Increases in Microtubule-Associated Protein 1 Light Chain 3 (LC3) and Beclin-1 Expression Induced by HMDB Immunofluorescence staining disclosed that HMDB-treated HeLa cells accommodated the acquisition of numerous large autophagic vacuoles in the cytoplasm. To further ascertain that HMDB may induce autophagy in HeLa cells, we assessed the expression and distribution of LC3-II, a hallmark of autophagy present in the autophagosomal membrane. The results showed that HMDB resulted in a time-dependent increase in LC3-II expression in the cytoplasm of HeLa cells (Figure 4a). The protein levels of LC3-II, Beclin-1 (an autophagic mediator promoting the nucleation of the autophagosomal membrane) and p62 expressed during the early stage of autophagy were also monitored by Western blotting. As shown in Figure 4b, HMDB increased the expression of autophagosome-associated LC3-II and the primary pro-autophagic protein Beclin-1 in HeLa cells in a time-dependent manner after 40 μM HMDB treatment. Meanwhile, the levels of p62 protein, a maker of autophagic degradation, were decreased in response to HMDB treatment. Furthermore, the expression levels of B-cell lymphoma 2 (Bcl-2, an inhibitor of Beclin-1 as well as an anti-apoptotic factor) and survivin (a molecule that inhibits autophagy-dependent apoptosis) were determined by Western blotting. Incubation of HeLa cells with HMDB at different time intervals revealed that HMDB led to decreases in the expression levels of Bcl-2 and survivin (Figure 4c). These results suggest that HMDB-induced autophagy plays a suppressive role in HeLa cell survival through apoptosis. Then, we assess whether HMDB induces HeLa cells undergoing apoptosis, apoptosis was determined by the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. The results showed that HMDB resulted in a sustained increase in apoptotic cells in a dose-dependent manner (Figure 4d). Furthermore, HMDB dose-dependently increased cleaved forms of caspase-3 and Poly (ADP-ribose) polymerase (PARP), the hallmarks of apoptosis (Figure 4e). Collectively, these results provide evidence that HMDB-induced autophagy is tightly linked to apoptosis. 2.5. HMDB-Induced Autophagy Is Linked to the Mediation of AMPK/mTOR and Akt/mTOR Signaling The class III phosphatidylinositol 3-kinase (PI-3K), which complexes with Beclin-1 is necessary for the initiation of autophagy. Next, we used the class III PI-3K and autophagy inhibitor, 3-methyladenine, to examine the detailed mechanism by which HMDB induces autophagy. HeLa cells were pretreated with 1 mM 3MA for 1 h, followed by exposure to 40 μM HMDB for 24 h. Then the cell lysates were applied to Western blotting to monitor the protein expression of LC3-II, Beclin-1, and p62. As shown in Figure 5a, pretreatment with 3MA decreased the protein levels of Beclin-1 and LC3-II, and recovered p62 protein levels in HeLa cells; however, 3MA cannot prevent the elevation of Beclin-1 and LC3-II expression levels induced by HMDB in HeLa cells. The interplay of AMPK (a sensor and positive regulator of autophagy) and the mTOR pathway have been well-known to regulate the occurrence of autophagy. AMPK can phosphorylate Raptor (regulatory associated protein of mTOR), an essential component of mTOR complex 1 (mTORC1), the activity of which blocks autophagy or can phosphorylate TSC2 (tuberous sclerosis complex 2) that directly inhibits Ras homolog enriched in brain (Rheb)-mediated mTORC1activation. To evaluate the signaling pathways responsible for the induction of HMDB-mediated autophagy, we assessed the activation statuses of the main regulators involved in the mTOR signaling pathway in HeLa cells. HMDB triggered the phosphorylation/activation of AMPK reciprocally accompanied with the downregulation of the phosphorylation statuses of Raptor, mTOR and S6K (mTOR downstream substrate) in a time-related manner. HMDB also had an inhibitory effect on the phosphorylation of Akt that negatively regulate TSC2 to release its inhibition on Rheb (Figure 5b). These results suggested that the activation of AMPK and inhibitory modification of Akt-mTORC-S6K signaling axis by HMDB may contribute to autophagic induction and growth inhibition in HeLa cells. 3. Discussion Cervical cancer is developed from malignant cells that form in the cervix. Although a majority of cervical cancer patients have benefited from neoadjuvant chemotherapy together with concurrent chemotherapy and radiotherapy, the survival rate remains poor in cervical cancer patients with relapse or recurrence [27]. Resistance to chemotherapy is one of the common causes of treatment failure in patients suffering from cervical cancer [28,29]. HMDB has been proven to inhibit cell growth in various human cancer cells [25]. Here, we demonstrate that HMDB inhibits proliferation through inducing G1 cell cycle arrest and autophagy in HeLa cervical cancer cells. The results clearly indicate that HMDB induced a G1 cell cycle arrest through downregulating the expression of cyclin D1/D3/E and CDK2/4/6 and subsequently resulting in the hypophosphorylation of Rb protein. Meanwhile, the protein and mRNA levels of CKIs, including p15, p16, p21, and p27, were upregulated by HMDB treatment. These results suggest that the growth-inhibitory effect of HMDB might stem from the blockade of the G1 to S phase transition in HeLa cells through mediating the expression of CKIs binding to their relative cyclin/CDK complexes, a crucial event for inactivating the activity of cyclin/CDK complexes and restricting the progression of cell cycle [30]. Rb not only controls the G1 to S transition in the cell cycle, but also plays a critical role during cellular senescence in response to cancer therapeutics such as CDK inhibitors [31]. In addition, autophagy is an effector mechanism of senescence [32]. Given that HMDB retained the phosphorylation status of Rb in a low level and induced the occurrence of autophagy in HeLa cells, the growth inhibition of HeLa cells by HMDB might be involved to be a link between Rb-mediated autophagy and senescence followed by the consequential tumor suppression; however, this issue requires further research. Autophagy inhibits the growth of certain cancer cells [33]. In this study, we found that HMDB increased the formation of intracellular vesicles as well resulted in growth inhibition, suggesting that these presented vesicles are autophagic. By means of neutral red staining, HMDB-induced intracellular vesicles were composed of acidic content, presumably coming from lysosome. In additional, lysosome aggregation was present in HDMB-treated HeLa cells by MDC and AO staining. These results suggest that HDMB treatment promotes the combination of autophagosomes and acidic lysosomes arising at the late stage of autophagy. Moreover, our data provide additional evidence that HMDB-induced autophagosome formation is tightly linked to the modulation of protein expression of Beclin-1, LC3-II, and p62/SQSTM1. During autophagy, Beclin-1 and LC3-II are localized on the membrane of autophagosome accompanied with the downregulation of p62/SQSTM1 [34]. Beclin-1 has repeatedly been reported as a target for applied therapies because its low expression may be attributed to the development of human cancer. Inactivation of Beclin-1 was reported to enhance tumorigenesis in mice [35]. In cervical cancer, Beclin-1 expression was significantly decreased in samples of malignant cervical cancer tissues compared to that in normal or cervical intraepithelial neoplasia tissues [36]. Moreover, positive expression of Beclin-1 in human cervical carcinoma has benefits for patients, resulting in a better prognosis [37]. The proautophagic function of Beclin-1 can be negatively regulated by Bcl-2, another well-known anti-apoptosis factor. Beclin-1 contains a BH3 motif required to bind Bcl-2, Bcl-XL, and Mcl-1, and Bcl-2 binds to Beclin-1 from its BC groove. By interaction with Beclin-1, Bcl-2 can block Beclin-1 interacting with class III PI-3K and decrease class III PI-3K activity, thereby negatively regulating autophagy [38]. Although HMDB had no effect on class III PI-3K-mediated LC-3II expression, it showed a time-dependent inhibition on Bcl-2 expression that may release the pro-autophagic activity of Beclin-1. Meanwhile, downregulation of Bcl-2 by HMDB raises a possibility that HMDB could trigger the induction of autophagy-dependent apoptosis in HeLa cells. To this end, the effect of HMDB on survivin expression was investigated. Survivin has been reported to be overexpressed in cervical cancer and participates in the development and progression of cervical cancer by inhibiting autophagy-dependent apoptosis. Evidence that HMDB suppressed the expression of survivin as well as that of Bcl-2 in a time-dependent manner in HeLa cells (Figure 4c), accompanied with the increased number of TUNEL-positive cells and cleaved forms of caspase-3 and PARP (Figure 4d,e) suggested HMDB might at least partly inhibit cell growth of HeLa cells via inducing the autophagy-dependent apoptosis. AMPK and Akt are well known to negatively and positively regulate the mTOR pathway, respectively [39]. The mTOR signaling pathway is a regulator of several cellular processes, including proliferation, autophagy, and survival [40]. An intriguing finding based on our results is the induction of AMPK phosphorylation by HMDB as early as 6 h after the initiation of treatment accompanied with the suppression of phosphorylation of Akt, mTOR, Raptor, and S6K, suggesting that HMDB induces autophagy through inducing AMPK activation and subsequently blocking AKT and mTOR activation. The regulation of both Akt and mTOR signaling pathways by HMDB implicates another therapeutic benefit. The mechanism of resistance to mTORC1 inhibitors in endometrial cancer may result from a negative feedback loop emerging from receptor tyrosine kinase/PI3K/Akt/S6K signaling pathway [41,42]. This implies that losing the feedback inhibition of Akt/mTOR signaling resulted from the use of mTORC1 inhibitors may reduce their therapeutic capacities and lead to the following chemoresistance. Therefore, an agent with the ability to inhibit both mTORC1 and Akt, such as HMDB, could be beneficial for the treatment of cervical cancer. To sum up, our results show a paralleled event that HMDB modulates several cellular signaling pathways and targets cell proliferation in human HeLa cervical cancer cells by involving in cell cycle arrest and induction of cell death via autophagy and apoptosis. The mechanism (Figure 6) by which HMDB induces G1 cell cycle arrest is due to the mediation of G1 cell cycle regulators, including the changes in the expression of D- and E-type cyclins, CDKs, CKIs, and Rb phosphorylation. We provide additional evidence that the autophagic cell death induced by HMDB is due to the activation of AMPK and the subsequent inhibition of mTORC1 activity. On the basis of these findings, we conclude that HMDB might potentially serve as a therapeutic agent for cervical cancer. 4. Materials and Methods 4.1. Chemicals and Reagents 1-(2-Hydroxy-5-methylphenyl)-3-pheyl-1,3-propanedione (HMDB) was purchased from Aldrich Chemical Co. (Milwaukee, WI, USA). The purity of the compound is >97% by high performance liquid chromatography (HPLC), and dissolved in dimethyl sulfoxide (DMSO). Propidium iodide, RNaseA, and 3-methyladenine (3MA) were available from Sigma-Aldrich (St. Louis, MO, USA). Antibodies against β-actin, Bcl-2, PCNA, and survivin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against LC3A/B, phospho-AMPK (Thr172), AMPK, phospho-Akt (Ser473), Akt, phospho-mTOR (Ser2448), mTOR, phospho-Raptor (Ser792), Raptor, phospho-p70 S6K (Thr389), p70 S6K, and all G1 cell cycle regulators were from Cell Signaling Technology (Beverly, MA, USA). Antibodies against Beclin-1 and SQSTM1/p62 were purchased from GeneTex, Inc. (Irvine, CA, USA). 4.2. Cell Culture HeLa cells were purchased from American Type Culture Collection (ATCC) and were cultured in Dulbecco’s minimal essential medium (DMEM) supplemented with 10% fetal calf serum (Gibco BRL, Grand Island, NY, USA), 100 units/mL of penicillin, and 100 μg/mL of streptomycin (Gibco BRL, Grand Island, NY, USA), and kept at 37 °C in a humidified atmosphere of 5% CO2 in air according to ATCC recommendations. For all results, the cells cultured showed no more than 20 passages. 4.3. MTT Assay The cells were seeded into 96-well plates at 5 × 103 cells/well for 24 h. HMDB was added with various concentrations, and the cells were incubated for the indicated times. After treatment, cell viability was assessed by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay as follows: 20 μL of MTT solution (5 mg/mL, Sigma, St. Louis, MO, USA) was added to each well and incubated for 24 h at 37 °C. Then the supernatant was removed, and the MTT-formazan crystals formed by metabolically viable cells were dissolved in 200 μL of dimethyl sulfoxide (DMSO). The absorbance was monitored by a microplate reader at a wavelength of 570 nm. 4.4. Trypan Blue Dye Exclusion Assay The cells were cultured in 6-well plates at the density of 5 × 104 cells per well in triplicate. After 24 h, cells were treated with 40 μM HMDB for the indicated times. After treatment of HMDB, cells were washed with phosphate-buffered saline (PBS) and a solution of 0.125% trypsin, 0.05% ethylenediaminetetraacetic acid (EDTA) for 2 min, and then incubated with trypan blue solution (1:1 dilution) for 10 min. After staining, cells were transferred to a hemocytometer (Bright-line™; Hausser Scientific, Horsham, PA, USA) and counted by microscopy (Observer-A1; Carl Zeiss, Oberkochen, Germany). The cells stained with the trypan blue dye are defined as dead cells. The percentage of living cells represented the number of living cells divided by the total number of counted cells. 4.5. Cell Cycle Analysis Cell cycle population was determined by flow cytometry as follows. After exposing to HMDB for 0, 6, 12, and 24 h, HeLa cells were washed twice with PBS, and then fixed in 70% ethanol for additional 2 h at −20 °C. Following fixation, cells were washed with PBS again, incubated with 1 mL of PBS containing 0.5 μg/mL RNase A and 0.5% Triton X-100 for 30 min at 37 °C. Then the cells were stained with 50 μg/mL propidium iodide (PI). The stained cells were estimated by a FACScan laser flow cytometer equipped with Cell Quest software (Becton Dickinson, San Jose, CA, USA). 4.6. Western Blotting After HeLa cells were treated with the indicated concentration of HMDB for the indicated times, the cells were collected followed with PBS washing. Then, the cells were incubated in a lysis solution containing 50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 150 mM NaCl, 0.5% NP-40, 0.5 mM phenylmethanesulfonyl fluoride, and 0.5 mM dithiothreitol for 30 min at 4 °C. Equal amounts of total proteins (50 μg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by transferring to polyvinylidene difluoride (PVDF) membranes (Immobilon P, Millipore, Bedford, MA, USA) and incubating with primary antibodies. The membrane was further washed with PBST, incubated with corresponding peroxidase-conjugated goat anti-mouse or anti-rabbit secondary antibodies, and visualizing by enhanced chemiluminescence staining. The β-actin acts as an internal control to normalize protein loading. The density of the band on the blots was quantitated with a computerized densitometer (ImageQuant LAS4000 Digital System, GE Healthcare, Uppsala, Sweden). 4.7. RNA Extraction, cDNA Synthesis, and qRT-PCR Total RNA from HMDB-treated HeLa cells was prepared using TRIzol reagent (Sigma), followed by the manufacturer’s instructions as below. Total RNA (5 μg) was applied to reverse transcription to generate cDNA by incubating the reaction mixture (25 μL) at 42 °C for 90 min containing Moloney murine leukemia virus (M-MLV) reverse transcriptase and oligo (dT) 18 primer. Then, real-time qPCR was executed in a 20 μL final volume for each primer (as below) using the Fastart Universal SYBR Green Master Mix (Roche, Indianapolis, IN, USA) and detected by an ABI 7000 sequence detection system. The primer sequences for p15, p16, p21, p27, and β-actin (β-actin is internal control) are as follows: β-actin (5′-AGTTGCGTTACACCCTTTCTTG-3′, 5′-CACCTTCACCGTTCCAGTTTT-3′), p15 (5′-GGCAGTCGATGCGTTCACTC-3′, 5′-CAGGGCTTCCAGAGAGTGTC-3′), p16 (5′-TTCCTGGACACGCTGGT-3′, 5′-CAATCGGGGATGTCTGAG-3′), p21 (5′-GCGACTGTGATGCGCTAAT-3′, 5′-TAGGGCTTCCTCTTGGAGAA-3′), p27 (5′-ATGTCAAACGTGCGAGTGTCTAA-3′, 5′-TTACGTTTGACGTCTTCTGAGG-3′). The PCR program is designed including the first reaction at 50 °C for 2 min and at 95 °C for 10 min, and then incubating for 40 thermal cycles between 95 °C for 15 s and 60 °C for 1 min. The relative cDNA expression for each sample was computerized using the formula 2−∆∆Ct, where ∆∆Ct  =  ∆Ct(target gene) − ∆Ct(β-actin gene), which represents the target cDNA expression normalized to β-actin cDNA levels. 4.8. Neutral Red Staining Neutral red (Sigma-Aldrich, St. Louis, MO, USA) staining is designed to monitor the relative amounts of lysosomes or acidic vacuoles that are stained with the supravital dye neutral red. Briefly, following HMDB treatment, cells were washed and suspended in PBS. Next, cells were stained with neutral red (33 μg/mL) and visualized by phase contrast microscopy. 4.9. Monodansylcadaverine and Acridine Orange Staining HeLa cells were treated with 40 μM HMDB for 0, 6, 12 and 24 h. After washing with fresh culture medium, the cells were stained with the autofluorescent dye containing 0.05 mM monodansylcadaverine (MDC) (Sigma-Aldrich, St. Louis, MO, USA) in PBS at 37 °C for 1 h, and then fixed with cold 4% paraformaldehyde for 15 min. For acridine orange staining, the cells were stained with 1 μg/mL acridine orange (AO) (Sigma-Aldrich, St. Louis, MO, USA) at 37 °C for 15 min, and then fixed with cold 4% paraformaldehyde for 15 min. The stained cells were washed with PBS three times and visualized under a contrast microscope. 4.10. Immunofluorescent Staining HeLa cells were grown to approximately 70% confluence on a coverslip, and then incubated with to the indicated concentrations of HMDB for the indicated times. For immunostaining, the treated cells were washed with cold PBS, and fixed with 4% paraformaldehyde for 15 min at room temperature. Furthermore, the fixed cells were stained with rabbit anti-LC3 antibody for 24 h, rinsed with cold PBS, followed with incubation with goat anti-rabbit secondary antibody conjugated with fluorescein isothiocyanate or rhodamine (Sigma, St. Louis, MO, USA) for 30 min at room temperature. The immunolabeled cells were mounted on a glass slide with DAPI Dapi-Fluoromount-G (Southern Biotech, Birmingham, AL, USA) and observed using a fluorescence microscope (Carl Zeiss MicroImaging GmbH, Jena, Germany). 4.11. TUNEL Assay DNA fragmentation analysis was performed by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay using an in situ labeling cell death kit (Roche Applied Science, Indianapolis, IN, USA). As follows, cells were grown at a density of 5 × 104 cells on a coverslip (25 mm size), followed with the incubation with 40 μM HMDB for the indicated times. The treated cells were washed with cold PBS, and rinsed with 4% paraformaldehyde for 15 min at 37 °C. Then the cells were incubated with a permeabilization solution containing 0.1% Triton X-100 in 0.1% sodium citrate for 5 min at 4 °C, and then applied to the TUNEL reaction buffer for 60 min at 37 °C in a humidified atmosphere in the dark. Afterward, the results were observed using a fluorescence microscope (Carl Zeiss MicroImaging GmbH, Jena, Germany). 4.12. Statistical Analysis Quantitative data taken on the values from three or more replicates repeated experiments were representative as the mean value with the respective standard error of the mean (SE). One-way analysis of variance (ANOVA) using Tukey’s post hoc multiple comparisons was applied for multiple group comparison, and the analyzed values were considered statistically significant at p < 0.05. Acknowledgments This study was supported by grants from the Ministry of Science and Technology, China (MOST 103-2320-B-040-014-MY3). We are thankful for the Instrument Center of Chung Shan Medical University, which provides fluorescence microscope and flow cytometer supported by National Science Council, Ministry of Education and Chung Shan Medical University. Author Contributions Li-Sung Hsu, Hsiu-Chen Huang, Chih-Li Lin, Min-Hsiung Pan and Wei-Jen Chen conceived and designed the experiments; Jie-Heng Tsai and Li-Sung Hsu performed the experiments; Hui-Mei Hong and Chih-Li Lin analyzed the data; Wei-Jen Chen and Min-Hsiung Pan contributed reagents/materials/analysis tools; Jie-Heng Tsai and Wei-Jen Chen wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 HMDB inhibited proliferation of HeLa cells via inducing the G1 cell cycle arrest. (a) The chemical structure of HMDB; and (b) the effect of HMDB on cell viability of HeLa cells. Cells were treated with a variety of dosages of HMDB for 0–24 h or (c) with 40 µM HMDB for different time periods. Cell survival was determined using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and trypan blue exclusion assays, respectively. The protein levels of PCNA were determined by Western blotting; and (d) a histogram of the cell cycle distribution. HeLa cells were treated with 40 μM HMDB for 0, 6, 12, and 24 h. Cell distribution at G1, S and G2/M phase was determined using flow cytometry. All of the data resulted from repeating independent experiments three times and results are expressed as mean ± SE. Values were statistically significant (versus HMDB treatment) for * p < 0.05, ** p < 0.01, *** p < 0.001 as compared with the control group. Figure 2 Effects of HMDB on the expression of G1-related cyclins, cyclin-dependent kinases (CDKs), and CDK inhibitors (CKIs). (a) Relative protein expression levels of cyclin D1/D3/E, and CDK4/6/2 expressed in the G1 phase; (b) the total and phosphorylated forms of retinoblastoma (Rb) with specific antibodies for each; and (c) the change in the protein expression levels of CKIs (p15, p16, p21, and p27). HeLa cells were exposed to 40 μM HMDB for the indicated times. Then, cellular extracts were harvested and the protein levels were visualized by Western blotting using antibodies against G1 cell cycle regulators as indicated. The β-actin acts as an internal control for evaluating protein loading; and (d) the changes in mRNA expression levels of CKIs, including p15, p16, p21, and p27, by HMDB. The relative amounts of target mRNA, collected from HMDB-treated HeLa cells, were determined by qRT-PCR for the indicated time. All of the results that come from independent experiments three times are expressed as mean ± SE. The relative amounts of protein levels on the Western blots were quantitated with a computerized densitometer (ImageQuant LAS4000 Digital System, GE Healthcare, Uppsala, Sweden) compared to the control group. Values were statistically significant for * p < 0.05, ** p < 0.01, *** p < 0.001 as compared with the control group (without HMDB treatment). Figure 3 HMDB increased the number of massive vacuoles with acid content and the accumulation of autolysosomes in HeLa cells. (a) The cells were treated with 40 μM HMDB for the indicated times. Morphological changes and representative photographs of HeLa cells after neutral red staining in response to HMDB were observed by light contrast microscopy; (b) microphotograph of cells stained with monodansylcadaverine (MDC) and acridine orange (AO). Scale bar, 50 µm. Figure 4 HMDB induced autophagy and apoptosis in HeLa cells. (a) HeLa cells were treated with 40 μM HMDB for the indicated times, fixed and incubated with rabbit anti-LC3-II primary antibody. After incubation with Alexa Fluor 488 phalloidin (green) for conjugated anti-rabbit secondary antibodies, immune-labeled cells were monitored by microscopy; (b) HeLa cells were treated with 40 μM HMDB and then the protein expression levels of LC3-II, Beclin-1, and p62 were determined by Western blotting for the indicated times; (c) the cells were treated with 40 μM HMDB for different times. Cells were harvested and lysed for the detection of the indicated protein expression by Western blotting; (d) the cells were treated with 40 μM HMDB for the indicated times, and then the apoptotic cells were examined by TUNEL assay; and (e) the apoptosis-related proteins, cleaved caspase-3, and poly (ADP-ribose) polymerase (PARP), were assessed using Western blotting. The densities of the band on the Western blots from three independent experiments were calculated using a computerized densitometer (ImageQuant LAS4000 Digital System, GE Healthcare, Uppsala, Sweden). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Figure 5 Modulation of class III PI-3K and AMPK/Akt/mTOR signaling was linked to HMDB-induced cell cycle arrest and autophagy in HeLa cells. (a) The cells were pretreated with 1 mM autophagy inhibitor, 3-methyladenine, followed by 40 μM HMDB treatment for 24 h. The expression of the indicated proteins was determined by Western blotting; (b) HeLa cells were incubated in the presence of 40 μM HMDB for various time points. Cell extracts were harvested for determining the indicated protein expression by Western blotting. The densities of the band on the Western blots from three independent experiments were calculated using a computerized densitometer (ImageQuant LAS4000 Digital System). Figure 6 The proposed signal pathway activated by HMDB in HeLa cervical cancer cells. Red arrows represent increased protein expression by HMDB, while black arrows indicate a decrease. ==== Refs References 1. Jemal A. Bray F. Center M.M. Ferlay J. Ward E. Forman D. Global cancer statistics CA Cancer J. Clin. 2011 61 69 90 10.3322/caac.20107 21296855 2. Ferlay J. Shin H.R. Bray F. Forman D. Mathers C. Parkin D.M. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008 Int. J. Cancer 2010 127 2893 2917 10.1002/ijc.25516 21351269 3. Shi J.F. Canfell K. Lew J.B. Qiao Y.L. The burden of cervical cancer in China: Synthesis of the evidence Int. J. Cancer 2012 130 641 652 10.1002/ijc.26042 21387308 4. Roa J.C. Garcia P. Gomez J. Fernández W. Gaete F. Espinoza A. Lepetic A. Suarez E. HPV genotyping from invasive cervical cancer in Chile Int. J. Gynecol. Obstet. 2009 105 150 153 10.1016/j.ijgo.2008.12.017 19249046 5. Lee D. Kwon J.H. Kim E.H. Kim E.S. Choi K.Y. HMGB2 stabilizes p53 by interfering with E6/E6AP-mediated p53 degradation in human papillomavirus-positive HeLa cells Cancer Lett. 2010 292 125 132 10.1016/j.canlet.2009.11.015 20036050 6. Cornelio D.B. Roesler R. Schwartsmann G. Emerging therapeutic agents for cervical cancer Recent Pat. Anticancer Drug Discov. 2009 4 196 206 10.2174/157489209789206887 19522698 7. Choudhari A.S. Suryavanshi S.A. Kaul-Ghanekar R. The aqueous extract of Ficus religiosa induces cell cycle arrest in human cervical cancer cell lines SiHa (HPV-16 positive) and apoptosis in HeLa (HPV-18 positive) PLoS ONE 2013 8 1274 10.1371/journal.pone.0070127 23922932 8. Fan X. Chen J.J. Role of Cdk1 in DNA damage-induced G1 checkpoint abrogation by the human papillomavirus E7 oncogene Cell Cycle 2014 13 3249 3259 10.4161/15384101.2014.953879 25485505 9. Perez de Castro I. de Carcer G. Malumbres M. A census of mitotic cancer genes: New insights into tumor cell biology and cancer therapy Carcinogenesis 2007 28 899 912 10.1093/carcin/bgm019 17259655 10. Brenna S.M. Syrjanen K.J. Regulation of cell cycles is of key importance in human papillomavirus (HPV)-associated cervical carcinogenesis Sao Paulo Med. J. 2003 121 128 132 10.1590/S1516-31802003000300009 12920476 11. Kaldis P. The cdk-activating kinase (CAK): From yeast to mammals Cell. Mol. Life Sci. 1999 55 284 296 10.1007/s000180050290 10188587 12. Sherr C.J. Roberts J.M. CDK inhibitors: Positive and negative regulators of G1-phase progression Genes Dev. 1999 13 1501 1512 10.1101/gad.13.12.1501 10385618 13. Capparelli C. Chiavarina B. Whitaker-Menezes D. Pestell T.G. Pestell R.G. Hulit J. Andò S. Howell A. Martinez-Outschoorn U.E. Sotgia F. CDK inhibitors (p16/p19/p21) induce senescence and autophagy in cancer-associated fibroblasts, ”fueling” tumor growth via paracrine interactions, without an increase in neo-angiogenesis Cell Cycle 2012 11 3599 3610 10.4161/cc.21884 22935696 14. Jiang H. Martin V. Gomez-Manzano C. Johnson D.G. Alonso M. White E. Xu J. McDonnell T.J. Shinojima N. Fueyo J. The RB-E2F1 pathway regulates autophagy Cancer Res. 2010 70 7882 7893 10.1158/0008-5472.CAN-10-1604 20807803 15. Singletary K. Milner J. Diet, autophagy, and cancer: A review Cancer Epidemiol. Biomark. Prev. 2008 17 1596 1610 10.1158/1055-9965.EPI-07-2917 18628411 16. Liu Z. Lenardo M.J. Reactive oxygen species regulate autophagy through redox-sensitive proteases Dev. Cell 2007 12 484 485 10.1016/j.devcel.2007.03.016 17419989 17. Chang C.P. Yang M.C. Liu H.S. Lin Y.S. Lei H.Y. Concanavalin A induces autophagy in hepatoma cells and has a therapeutic effect in a murine in situ hepatoma model Hepatology 2007 45 286 296 10.1002/hep.21509 17256764 18. Kanzawa T. Kondo Y. Ito H. Kondo S. Germano I. Induction of autophagic cell death in malignant glioma cells by arsenic trioxide Cancer Res. 2003 63 2103 2108 12727826 19. Eskelinen E.L. Saftig P. Autophagy: A lysosomal degradation pathway with a central role in health and disease Biochim. Biophys. Acta 2009 1793 664 673 10.1016/j.bbamcr.2008.07.014 18706940 20. Miracco C. Cevenini G. Franchi A. Luzi P. Cosci E. Mourmouras V. Monciatti I. Mannucci S. Biagioli M. Toscano M. Beclin 1 and LC3 autophagic gene expression in cutaneous melanocytic lesions Hum. Pathol. 2010 41 503 512 10.1016/j.humpath.2009.09.004 20004946 21. Wang Z.H. Li L. Peng Z.L. Duan Z.L. Effect of autophagy gene Beclin 1 on the growth of cervical cancer HeLa cells in vitro and vivo Zhonghua Zhong Liu Za Zhi 2011 33 804 809 22335943 22. Klionsky D.J. Abeliovich H. Agostinis P. Agrawal D.K. Aliev G. Askew D.S. Baba M. Baehrecke E.H. Bahr B.A. Ballabio A. Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes Autophagy 2008 4 151 175 10.4161/auto.5338 18188003 23. Zhu W. Pan X. Li F. Zhang Y. Lu X. Expression of Beclin 1 and LC3 in FIGO stage I-II cervical squamous cell carcinoma and relationship to survival Tumour Biol. 2012 33 1653 1659 10.1007/s13277-012-0421-4 22648564 24. Pan M.H. Huang M.C. Wang Y.J. Lin J.K. Lin C.H. Induction of apoptosis by hydroxydibenzoylmethane through coordinative modulation of cyclin D3, Bcl-XL , and Bax, release of cytochrome c , and sequential activation of caspases in human colorectal carcinoma cells J. Agric. Food Chem. 2003 51 3977 3984 10.1021/jf034094i 12822933 25. Pan M.H. Sin Y.H. Lai C.S. Wang Y.J. Lin J.K. Wang M. Ho C.T. Induction of apoptosis by 1-(2-hydroxy-5-methylphenyl)-3-phenyl-1,3-propanedione through reactive oxygen species production, GADD153 expression, and caspases activation in human epidermoid carcinoma cells J. Agric. Food Chem. 2005 53 9039 9049 10.1021/jf051476a 16277400 26. Pan Y.C. Li C.F. Ko C.Y. Pan M.H. Chen P.J. Tseng J.T. Wu W.C. Chang W.C. Huang A.M. Sterneck E. CEBPD reverses RB/E2F1-mediated gene repression and participates in HMDB-induced apoptosis of cancer cells Clin. Cancer Res. 2010 16 5770 5780 10.1158/1078-0432.CCR-10-1025 20971808 27. Friedlander M. Grogan M. Guidelines for the treatment of recurrent and metastatic cervical cancer Oncologist 2002 7 342 347 12185296 28. Suh D.H. Kim J.W. Kim K. Kim H.J. Lee K.H. Major clinical research advances in gynecologic cancer in 2012 J. Gynecol. Oncol. 2013 24 66 82 10.3802/jgo.2013.24.1.66 23346316 29. Sun Y. Liu J.H. Jin L. Lin S.M. Yang Y. Sui Y.X. Shi H. Over-expression of the Beclin1 gene upregulates chemosensitivity to anti-cancer drugs by enhancing therapy-induced apoptosis in cervix squamous carcinoma CaSki cells Cancer Lett. 2010 294 204 210 10.1016/j.canlet.2010.02.001 20207475 30. Malumbres M. Barbacid M. Mammalian cyclin-dependent kinases Trends Biochem. Sci. 2005 30 630 641 10.1016/j.tibs.2005.09.005 16236519 31. Knudsen E.S. Wang J.Y. Targeting the RB-pathway in cancer therapy Clin. Cancer Res. 2010 16 1094 1099 10.1158/1078-0432.CCR-09-0787 20145169 32. Young A.R. Narita M. Ferreira M. Kirschner K. Sadaie M. Darot J.F. Tavaré S. Arakawa S. Shimizu S. Watt F.M. Autophagy mediates the mitotic senescence transition Genes Dev. 2009 23 798 803 10.1101/gad.519709 19279323 33. Deng L. Lei Y. Liu R. Li J. Yuan K. Li Y. Chen Y. Liu Y. Lu Y. Edwards C.K. III Pyrvinium targets autophagy addiction to promote cancer cell death Cell Death Dis. 2013 4 e614 23640456 34. Kelekar A. Autophagy Ann. N. Y. Acad. Sci. 2005 1066 259 271 10.1196/annals.1363.015 16533930 35. Hoyer-Hansen M. Bastholm L. Mathiasen I.S. Elling F. Jäättelä M. Vitamin D analog EB1089 triggers dramatic lysosomal changes and Beclin 1-mediated autophagic cell death Cell Death Differ. 2005 12 1297 1309 10.1038/sj.cdd.4401651 15905882 36. Wang Z.H. Peng Z.L. Duan Z.L. Liu H. Expression and clinical significance of autophagy gene Beclin 1 in cervical squamous cell carcinoma Sichuan Da Xue Xue Bao Yi Xue Ban 2006 37 860 863 17236580 37. Cheng H.Y. Zhang Y.N. Wu Q.L. Sun X.M. Sun J.R. Huang X. Expression of beclin 1, an autophagy-related protein, in human cervical carcinoma and its clinical significance Eur. J. Gynaecol. Oncol. 2012 33 15 20 22439399 38. Pattingre S. Tassa A. Qu X. Garuti R. Liang X.H. Mizushima N. Packer M. Schneider M.D. Levine B. Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy Cell 2005 122 927 939 10.1016/j.cell.2005.07.002 16179260 39. Cornu M. Albert V. Hall M.N. mTOR in aging, metabolism, and cancer Curr. Opin. Genet. Dev. 2013 23 53 62 10.1016/j.gde.2012.12.005 23317514 40. Xu Y. Liu C. Chen S. Chen S. Ye Y. Guo M. Ren Q. Liu L. Zhang H. Xu C. Activation of AMPK and inactivation of Akt result in suppression of mTOR-mediated S6K1 and 4E-BP1 pathways leading to neuronal cell death in in vitro models of Parkinson’s disease Cell Signal. 2014 26 1680 1689 10.1016/j.cellsig.2014.04.009 24726895 41. Slomovitz B.M. Coleman R.L. The PI3K/AKT/mTOR pathway as a therapeutic target in endometrial cancer Clin. Cancer Res. 2012 18 5856 5864 10.1158/1078-0432.CCR-12-0662 23082003 42. Suh Y. Afaq F. Khan N. Johnson J.J. Khusro F.H. Mukhtar H. Fisetin induces autophagic cell death through suppression of mTOR signaling pathway in prostate cancer cells Carcinogenesis 2010 31 1424 1433 10.1093/carcin/bgq115 20530556

PMC005xxxxxx/PMC5000673.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081275ijms-17-01275ReviewEnvironmental Microbial Community Proteomics: Status, Challenges and Perspectives Wang Da-Zhi *Kong Ling-Fen Li Yuan-Yuan Xie Zhang-Xian Woo Patrick C. Y. Academic EditorState Key Laboratory of Marine Environmental Science, College of the Environment and Ecology, Xiamen University, Xiamen 361102, China; [email protected] (L.-F.K.); [email protected] (Y.-Y.L.); [email protected] (Z.-X.X.)* Correspondence: [email protected]; Tel.: +86-592-218-601605 8 2016 8 2016 17 8 127524 5 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Microbial community proteomics, also termed metaproteomics, is an emerging field within the area of microbiology, which studies the entire protein complement recovered directly from a complex environmental microbial community at a given point in time. Although it is still in its infancy, microbial community proteomics has shown its powerful potential in exploring microbial diversity, metabolic potential, ecological function and microbe-environment interactions. In this paper, we review recent advances achieved in microbial community proteomics conducted in diverse environments, such as marine and freshwater, sediment and soil, activated sludge, acid mine drainage biofilms and symbiotic communities. The challenges facing microbial community proteomics are also discussed, and we believe that microbial community proteomics will greatly enhance our understanding of the microbial world and its interactions with the environment. microbesproteomicscommunity proteomicsmetaproteomics ==== Body 1. Introduction Microorganisms and their activities are of critical importance to virtually all biological systems on our planet. The composition and structure of microbial communities are diverse over a wide range of environments [1]. Moreover, microbial metabolic activities and cellular physiology frequently fluctuate along with environmental changes [2]. The ubiquity and complexity of microbial communities have been extensively studied for decades [3]; however, their ecological functions in the environment and their response to various environmental drivers has gained increasing attention in recent years. Microbial community proteomics (also termed metaproteomics), which characterizes all the proteins expressed at a given time within an ecosystem, plays a key role in exploring microbial functionality [4]. Investigation of the protein expression of a microbial community enables an unprecedented view of the adaptive responses of microbes to environmental stimuli or their interactions with other organisms or host cells [5]. Studies of the microbial community in natural environments have expanded our knowledge of microbial functions, such as nutrient cycling, mutualistic endosymbionts, organic matter degradation, metal utilization, and eutrophication response [6,7,8,9,10,11,12]. With the development of genome decoding techniques and high-throughput sequencing technologies, microbial community proteomics has quickly emerged over the past few years [13]. Much effort has been devoted to the microbial community proteomics in a variety of environments, including marine water [14,15,16,17], soils [1,18,19], sediments [8,9], activated sludge [20,21,22], groundwater [23], and acid mine biofilms [24,25]. These studies provide new insights into the outcome of gene expression regulation, protein synthesis, and the stability and turnover of mRNA and protein in response to environmental stress at a given time [26]. Moreover, these functional dimensions of the environmental proteomic database have facilitated the link of the individual microbial species to its ecological function [27]. In contrast to methods such as stable isotope probing [28], fluorescence in situ hybridization with microautoradiography [29], and full-cycle rRNA analysis [30], metaproteomics can reflect physiological activity and translational regulation of microbes in various environmental conditions. In this review, we highlight the advances of microbial community proteomics in the context of marine and freshwater, soil and sediment, activated sludge, acid mine drainage (AMD) biofilms as well as symbiotic communities. The challenges and perspectives of this field are also discussed. 2. Strategies for Microbial Community Proteomic Studies In the past few years, much effort has been devoted to exploring the strategies for microbial community proteomic studies, and several typical steps have been developed, including sample collection, protein extraction, protein separation and/or fractionation, mass spectrometry analysis, database searching and finally data interpretation (Figure 1). Because of the complex nature of environmental samples, specific approaches for either sample collection or protein extraction are developed when dealing with marine and freshwater as well as soil samples [31,32,33]. For protein separation and identification, two strategies have been established: one is the gel-based method. Traditionally, mixed proteins are separated using either one-dimensional or two-dimensional polyacrylamide gel electrophoresis (2-D PADE). Then, the target protein spots or bands are excised and proteins are digested into peptides with trypsin or other enzymes. Subsequently, the resulting peptides are subjected to mass spectrometry (MS) or tandem MS (MS/MS) analysis, database searching and bioinformatic analysis [5]. The other strategy is the liquid chromatography (LC)-based method, where the whole proteome is digested into a more complex peptide mixture using proteases without prior protein separation in gel. Then the resulting peptides are separated using strong cation exchange chromatography or microcapillary reverse-phase. In general, the separated peptides are analyzed using liquid chromatography coupled with MS/MS (LC-MS/MS). The produced MS data are interpreted for protein identification and then bioinformatic analysis. The second approach circumvents the limitations of the gel-based approach, and greatly increases the proteome coverage compared with the gel-based method, allowing high-throughput identification of thousands of proteins within a short time [34] and especially making detection of insoluble membrane proteins possible [35]. Thus, the LC-based approach has become the main stream of microbial community proteomic studies, although it still suffers from problems of reproducibility, dynamic range, and database availability. Reproducibility of metaproteomic analysis is critical to determine whether the variation of protein expression in the microbial community is environmentally relevant or the result of system errors. Usually, the technical reproducibility can be close to 50% across triplicates and over 67% between replicates using the same MS platform [36]. However, it should be noted that more replicates might improve the protein identification but the reproducibility may become worse, especially for the biological repeats. Besides qualitative analysis in proteomics, the output of a large scale of quantitative information with high reproducibility and accuracy is rather useful, especially for comparative and quantitative proteomics: their main goal aims at determining the differences in protein expression among different biological states (e.g., control vs. treatment, healthy vs. disease, specific genotype vs. wild type) or along environmental gradients (e.g., nutrient and salinity gradients). Recently, different labeling techniques for proteomics have been developed, such as stable isotope labeling using amino acids in cell culture [37], tandem mass tags [38], stable isotope-labeled peptides [39], isotope dilution [40], isotope-coded affinity tags [41] and, more recently, isobaric tags for relative and absolute quantification [42]. However, most label-based quantification approaches are limited in complex sample preparation, protein enrichment and incomplete labeling as well as in number of samples. With the development of suitable computational software, a label-free quantitative proteomic approach has emerged, which allows the profiling of a large scale of proteins with the flexibility of multiple different comparisons. The label-free method is a semi-quantification based on the comparison of either the peak intensity of the same peptide or the spectral count of the same protein, and abundant proteins produce more spectral counts or peptide intensities. In addition, it is cost-effective due to its non-labeling characteristic. As a result, the MS-based label-free approach has been more popular and has become the main research method in metaproteomics (Table 1). 3. Microbial Community Proteomics in Various Environments 3.1. Marine and Freshwater Metaproteomics As the Earth’s largest aquatic ecosystem, the marine habitat harbors diverse microbial communities which play a central role in regulating biogeochemical cycling of biogenic elements, including carbon, nitrogen and phosphorus, as well as various micronutrients and trace metals [68]. Deciphering metabolic activity and the ecosystem functioning of specific microbial assemblages in a variety of marine habitats provides new insights into carbon cycling as well as nutrient and energy utilization in the ocean. Since the pioneering metaproteomic work on the marine microbial community reported by Kan et al. [14], more efforts have been devoted to metaproteomic studies on the marine microbial communities. Marine microbes can adapt to different nutrient environments through expressing abundant transporter proteins with ATP binding cassette (ABC)-type and tripartite ATP-independent periplasmic (TRAP)-type being the most abundant [69]. Similar results are reported in a membrane quantitative metaproteomic study from the South Atlantic Ocean in which TonB-dependent transporters dominate the membrane proteins [43]. A quantitative proteomic investigation of the microbial community in the coastal northwest Atlantic Ocean is also characterized by the prevalent periplasmic-binding proteins (PBPs) of ABC transporters (751 proteins) and TRAP transporters (202) [15]. The proportion of transporters shows a seasonal variation, more obviously at the deep layer (from 17% in winter to 57% in spring), indicating fierce competition within the microbial community of deep waters in spring, when organic compounds (i.e., sugar, amino acids, taurine, dipeptides and glycine betaine) are replenished owing to phytoplankton production. It is interesting that approximately 91% of the transporter spectra belong to the SAR11 and Rhodobacterales clades, which is consistent with the abundance of the SAR11 clade throughout the ocean, especially in oligotrophic water as well as in the bathypelagic region [68]. In a recent study, most of the ABC-type sugar-, organic polyanion-, and glycine betaine-transport proteins are identified from Pelagibacter, indicating their important roles in marine carbon and nitrogen cycling [44]. In order to evaluate the microbial response to nitrogen limitation in the Pacific Ocean, targeted metaproteomics is applied to investigate the protein expression profiles of the major phytoplankton groups [45]. In this study, a specific peptide biomarker for nitrogen response regulator NtcA is identified abundantly in the oligotrophic region of the North Pacific, which was consistent with the prevalence of the Prochlorococcus urea transporter proteins (UrtA) in low-nitrogen areas. The Roseobacter clade contributes a large portion of the ABC transporter (13.7% of the total metaproteome) for amino acids and polyamines, suggesting that the Rosebacteria rely on these nitrogen-containing organic matters [46]. SAR11 is the dominant group of α-proteobacteria throughout most sections of the ocean, and its adaption to oligotrophic environments has attracted great attention. A large number of mass spectra disproportionately map the periplasmic substrate-binding proteins (PSPs) from SAR11; for example, two PSPs for phosphonate acquisition are the most frequently detected, suggesting the active expression of the phosphorus transporter of SAR11 in response to phosphorus limitation in the Sargasso Sea [70]. However, in another distinct ecosystem, a productive coastal upwelling system, the highly detected transporter proteins from SAR11 are involved in amino acid, taurine and polyamine transport, as well as highly abundant glutamine synthetase [46], which is in accordance with the nitrogen and carbon limitations in this region. In addition to the accumulation of transporters with a high affinity of nutrients, microorganisms have evolved distinct metabolic strategies to utilize hydrogen [16], one-carbon compounds [47], urea [48,71] and taurine [46], as well as other potential substrates as energy sources. A semi-quantitative metaproteomic analysis of the dissolved organic matter (DOM) from the surface and bathypelagic layers of the South China Sea indicates that the most abundant protein at the surface is the urea ABC transporter, whereas methylene tetrahydomethanopterin reductase dominates the proteome of the abyssal small-size fraction of DOM, suggesting that microbes can utilize urea as an alternative nitrogen source in the oligotrophic surface water [48]. Proteins involved in two chemolithoautotrophic pathways, the 3-hydroxypropionate/4-hydroxybutyrate cycle and the reverse tricarboxylic acid cycle, dominate the winter metaproteome of cold and dark polar water in the Western Antarctic Peninsula [46]. Consistent with the chemosynthesis cycle, ammonia is oxidized to available nitrate by the archaea and bacteria through ammonia monooxygenase. The genes of two ammonia-oxidizing Betaproteobacteria-associated RuBisCO enzymes are also detected in the winter metagenome. In addition, transporters and enzymes participating in taurine uptake and degradation, including taurine-pyruvate aminotransferase and sulfoacetaldehyde acetyltransferase, are abundantly detected, suggesting their important roles in regulating carbon and nitrogen utilization in the deep dark sea. The study of microbial communities from nutrient-enriched coastal systems shows the large subunit of methanol dehydrogenase from the OM43 clade in almost all the samples [47]. In a proof-of-concept experiment, the RuMP cycle is regarded as the main carbon assimilation pathway in the Methylophaga-like bacterium [49]. Moreover, hexulose-6-phosphate synthase, the key enzyme of the RuMP pathway of OM43, is also detected in all Atlantic Ocean samples. In addition, methanol oxidation proteins originating from the common OM43 marine clade are also identified in a deep and stratified estuary [50]. These results support the in situ activities of the OM43 clade using one-carbon compounds for energy production. Recently, Kleiner et al. combined metaproteomic quantification and metabolomic technologies to reveal that chemosynthetic symbionts can utilize carbon monoxide (CO) which has been previously thought to be unavailable for microbial nitrification due to its toxicity to aerobic organisms [16]. However, both aerobic and anaerobic CO dehydrogenases are detected in three types of Olavius algarvensis symbionts, indicating that they could utilize CO produced in the sediment at the sampling site. In addition, the identification of periplasmic uptake (NiFeSe) hydrogenases assigned to the δ-symbionts in the counterpart metaproteome demonstrates that energy production from hydrogen occurs in the sulfate-reducing symbionts. Notably, 544 previously unassigned proteins in the metagenomic analysis are annotated to a specific symbiont based on proteomics-based binning. Therefore, complementary information of the symbiotic community is obtained using the combined genomic and proteomic approaches, including the utilization of CO, sulphur and hydrogen in a certain specific symbiont. However, proteomic information of one symbiont, the spirochete, is completely missing owing to the lack of unambiguous metagenomic annotation for this species. Overall, phylogenetic analysis based on proteomics depends on genomic information. However, proteomic-based binning after the enrichment of a microbial group, to some extent, may overcome the obstacle and provide further functional insights. Trace metals are essential nutrients needed for bacteria to survive on the Earth, and metalloproteins play vital roles in catalyzing critical biogeochemical reactions [72,73]. Recently, metaproteomics has been applied to explore microbial adaptive strategies in metal acquisition and utilization in various environments [8,72,73]. The uptake of limiting metals is a key driver of the ongoing adaptive strategies by which microbes evolved. Either Iron (Fe) or zinc (Zn) could form the active center of alkaline phosphatases. Therefore, they are two essential metals involved in phosphate acquisition by microorganisms. When low phosphate is available, two types of alkaline phosphatase enzymes, PhoA and PhoX, collaboratively function based on the availability of Zn or Fe. Another study reveals that the flavodoxin protein, which is the equivalent alternative of the Fe-binding protein, is abundantly distributed in the low-Fe waters in the Pacific Ocean. Unlike numerous studies that focused on well-oxygenated oceanic waters and special microbial metalloproteomes [73], Glass et al. explored microbial metal utilization in a deep-sea methane seep ecosystem using the metaproteogenomic approach [8]. Their results indicate that the anaerobic oxidation of methane bacteria can produce nickel-binding ligands to release nickel from HS− outside the cells so as to increase nickel availability, which thereafter is captured by Ni-bound ligands. Similar to nickel, cobalt exists mainly in the form of Co(HS)2, which is less bioavailable for microbial cells. To deal with this, microbial consortia can produce high-affinity cobalt-binding ligands for acquiring the inaccessible forms. At the functional level, metaprotemics has improved our knowledge of nutrients and carbon utilization in the ocean, by providing notable information including dominant groups, transporter proteins and key enzymes involved in biogeochemical cycling. Extreme stress environments, including hypoxia, low-light intensity and polar regions, greatly challenge microbial survival. Therefore, microbes evolve specialized strategies, i.e., sulfur oxidation and syntrophic associations, to overcome these challenges. Recently, proteins of SUP05 related to sulfur oxidation were identified, suggesting that SUP05 is able to utilize reduced sulfur compounds, such as thiosulfate or elemental sulfur (S0), as an energy source in the hypoxic bottom water of the Northwest Atlantic Ocean [15]. In a green sulfur bacteria (GSBs)-dominated community in Ace Lake of Antarctica [51,52] many Chlorobia-like chlorosome envelope proteins were identified using metaproteogenomics, indicating that GSBs have the ability of light capture at a high efficiency which allows them to adapt to low-light conditions. Moreover, the GSBs may facilitate their essential metabolism through coupling carbonic fixation and sulfide oxidation in the Antarctic, given that many proteins related to the sulfur cycle, such as the dissimilatory sulfite reductase system, a polysulfide-reductase-like complex, as well as a number of sulfur metabolic proteins, are detected, implying this important adaptive mechanism of GSBs to sulfur-rich polar environments. However, a comparative analysis of Lake Cadagno in Switzerland [33], where the community is dominated by the GSB Chlorobium clathratiforme, reveals that proteins participating in sulfur metabolism are two-fold less abundant in the dark water column; therefore, the sulfur cycle is probably not active in this dark deep water. The metaomics study of Urich et al. suggests that microorganisms in deep-sea venting sediments are fueled by chemically fixed energy to maintain growth [53]. The dominant genera Sulfurimonas and Sulfurovum, as the primary producers of the upper sediment layers, can utilize H2S to drive CO2 fixation. The free-living anaerobic methanotrophicarchaea (ANME-1) is another dominant microbial species in methane-enriched cold seep sediments which plays a major role in the sulfur cycle and the biological sinking of methane [54]. Identification of cold-adaptation proteins and key metabolic enzymes involved in the reverse methanogenesis (i.e., methyl-Coenzyme M reductase) and sulfate reduction pathways (i.e., adenylyltransferase, and adenosine 5′-phosphosulfate-reductase (AprAB)) reveals the adaptive clues of the ANME-1 community to the marine cold seep systems. These metaproteomic studies provide new insights into the adaptive lifestyle of anaerobic bacteria in the anoxic and sulfur-rich regions of the dark ocean, which advances our knowledge of microbial life in extreme stress environments. Of the numerous adaptive strategies possessed by microbial communities, symbiotic combinations (especially between chemosynthetic bacteria and their hosts) are one of the key mechanisms in adaptation to low-nutrient and high-stress environments. In the gutless marine worm O. algarvensis, nutrient supply depends on its chemosynthetic bacteria, which allow it to grow well in the dark deep sea which features a high concentration of sulfide and CO2 [16]. In addition to expressing high-affinity transporters and utilizing alternative energy sources such as CO and hydrogen to maintain normal growth under stress conditions, significant activities of active mobile genetic elements are also found [9]. Through increasing transposase profiles (the enzyme-catalyzing movement of genetic elements), host-restricted bacteria experience an evolutionary adaptation process to rapidly changing environments. 3.2. Soil Metaproteomics Soil covers almost all of the terrestrial regions and harbors the most abundant and diverse microbiota on Earth, which make it into another complex and dynamic ecosystem. Soil microbial assemblages participate in the decomposition and transformation of soil materials, contaminant remediation, rhizospheric soils, semiarid soils, as well as the biogeochemical cycling of carbon, nitrogen and other biogenic elements [61]. Thus, qualitative and quantitative assessment of the protein complement in the soil environment might provide new insights into the interactions between microbes and the environment. Semiarid soils are composed of different soil carbon contents, vegetative communities and microbial communities. The quantitative metaproteomic approach has been applied to evaluate the functional and phylogenetic information regarding semiarid soils with distinct edaphic properties and degradation levels [18]. Proteins are identified which have the potential to participate in the biogeochemical cycling of elements as well as in the oxidation of organic matter in semiarid soils, i.e., a wide variety of dehydrogenases. Proteins involved in nitrogen cycling in semiarid soils are also identified, particularly proteases and peptidases, as well as enzymes directly participating in nitrogen fixation and nitrification. In contrast to poor soils, proteins related to phycocyanin and photosystemic apoproteins from diverse cyanobacteria are identified while superoxide dismutase and catalase are detected in a majority of semiarid soils. With regard to carbon cycling, the CO dehydrogenase and several hydrolases are also identified from Singulisphaera acidiphila (Planctomycete). Similarly, cyanobacteria plays important ecological roles in carbon fixation during soil erosion since multiple N-metabolic proteins are identified in semiarid areas [57]. Microbial decomposition of senesced-leaf litter plays an important role in the carbon and other nutrient cycling of terrestrial ecosystems [74]. The quantitative metaproteomic analysis of beech leaf litter indicates that environmental factors including nutrients influence the structure and function of decomposers during decomposition [58]. Fungi are the major producers of extracellular hydrolytic enzymes while no bacterial hydrolases have been detected. The litter nutrient content and stoichiometry affect microbial succession, together with decomposer community structure and activity. Moreover, microbial activity is stimulated by high litter nutrient content via high expression and high activity of extracellular enzymes. Rhizospheric microbes are another hot spot in the terrestrial research field that aims at uncovering the interactions between plants and microorganisms in the soil ecosystem. Plant root exudates significantly affect the diversity of the microbial community. Conversely, rhizospheric microbes provide a multitude of benefits to their host including promotion of plant growth, stimulation of pathogen resistance or direct defense against pathogens as well as nutrient supply [74]. Wang et al. characterizes the metaproteomes of different crop rhizospheric soils (CRS) using 2-DE coupled with MALDI-TOF/TOF-MS [19]. Among the successfully identified 189 protein spots, 72 derived from the microflora are involved in protein, energy, nucleotide and secondary metabolism, as well as signal transduction and resistance. Most of these biological processes are associated with the soil nutrient cycle, particularly carbon and nitrogen. These proteins might play crucial roles in the communication among plants, microbes and fauna, and induce metabolic changes inside the organisms. A comparison of the CRS subjected to increasing periods of Rehmannia glutinosa reveals that the identified proteins derived from plants and microorganisms actively participated in nutrient assimilation and energy transformation in the rhizospheric soil ecosystem [59]. They participate in protein, nucleotide and secondary metabolism, signal transduction and resistance, and 33 differentially expressed protein spots are shown to respond to an increase in the monoculture years. Among them, most of the upregulated plant proteins are involved in carbon and nitrogen metabolism and stress response, while the majority of the upregulated microbial proteins participate in protein metabolism and cell-wall biosynthesis. With an increase in the monoculture years, the phenylalanine ammonia-lyase significantly increases in total phenolic acid content, implying that it participates in the phenylpropanoid metabolism. These studies indicate that the consecutive monoculture of R. glutinosa changes the soil microbial ecology owing to the accumulation of exudates, which in turn might affect the nutrient cycle, resulting in plant growth and development retardation. The metaproteomic approach has also been employed to characterize microbial metabolic activities relevant to the bioremediation of pollutant-contaminated environments. Using the gel-based approach, the metaproteome of cadmium-contaminated soil is analyzed, although very limit protein information is obtained [75]. A proteomic-based study on the uranium-contaminated aquifer demonstrates the importance of the dominant Geobacter community members as well as their pathways involved in energy generation during biostimulation [60]. However, a recent study on the initial responses of the indigenous aquifer microbiota to biostimulation with emulsified vegetable oil at a uranium-contaminated site suggests that members of the Betaproteobacteria and the Firmicutes dominate the biostimulated aquifer community [76]. Organic pollutants in soils have also attracted great attention. Metaproteomic analysis of the microbial community from 2,4-dichlorophenoxy (2,4-D)-contaminated soils indicates that at least two species are linked to the biodegradation of chlorobenzene and that the 2,4-dichlorophenoxyacetate dioxygenase involved in 2,4-D degradation is expressed by autochthonous bacteria [23]. Recently, a culture-dependent community proteomic study traced changes in the microbial assemblies of a hydrocarbon-polluted soil [61]. The results suggest that the soil microbial community becomes more complex in hydrocarbon-polluted soil compared to that in untreated soil. Although Bacillus sp. dominates in both communities, other species, such as Ralstonia solanacearum, Synechococcus elongates and Clostridium sp., do not appear in the non-contaminated soil, suggesting their resistance to hydrocarbon contamination. A further study on the bioremediation of hydrocarbon contamination indicated that compost-assisted bioremediation is mainly driven by Sphingomonadales and uncultured bacteria through the high expression of catabolic enzymes such as catechol 2,3-dioxygenases, cisdihydrodiol dehydrogenase and 2-hydroxymuconic semialdehyde [62]. A similar metaproteomic survey of toluene-amended soil as well as enriched cultures containing toluene and soil extracts shows that many proteins are shared between the two toluene-amended communities [1]. Compared with glucose-amended soil as the control, a high expression of glutamine synthetase, ABC transporters, extracellular solute-binding proteins, and outer membrane proteins in both toluene-amended communities might be involved in the removal of toluene from the bacterial cells. Overall, metaproteomic approaches provide a valuable avenue to explore the roles of the major particular bacteria with specific functions in situ rather than in the traditional “artificial” laboratory experiments [77]. 3.3. Wastewater and Activated Sludge Metaproteomics Microbial communities play important roles in wastewater treatment and different microbial systems have been developed [78]. Metaproteomics provides an interesting functional insight into the complex microbial communities in the wastewater. Carla et al. employ the metaproteomic approach to investigate the response of an unsequenced bacterial community in a continuous-flow wastewater treatment bioreactor with an inhibitory level of cadmium [63]. The metaproteome in the bioreactor has a quick response (after 15 min) to cadmium exposure and shows a temporal change compared with the unexposed control at each time point (0.25, 1, 2 and 3 h). More than 100 unique differentially expressed proteins are identified, including ATPases, oxidoreductases, and transport proteins. Metaproteomics has also become a critical research component of activated sludge wastewater treatment. Wilmes et al. conducted a series of metaproteomic studies on the molecular mechanisms of enhanced biological phosphorus removal (EBPR) [4,21,22]. Although only several proteins are identified in their first effort, metaproteomics shows its potential in the study of activated sludge [4]. With both the improvement of the MS technique and the availability of metagenomic data, great achievements have made in the metaproteomics of the activated sludge system. Based on the gel-based proteomic approach, 46 proteins among the 111 excised spots are identified and many of them are closely matched to “Candidatus Accumulibacter phosphatis”, indicating that the Accumulibacter’s metabolic activities are related to the chemical transformations in EBPR. Furthermore, more than 700 proteins are identified from the A. phosphatis population using a non-gel-based proteomic approach, and these proteins are involved in many key metabolic pathways, such as denitrification, fatty acid cycling and the glyoxylate bypass, with significant importance in EBPR. The differences in protein abundance for enzyme variants related to core metabolism and EBPR-specific pathways, as well as genetic diversity, are crucial for maintaining the stable performance of EBPR systems. Park et al. investigate activated sludge extracellular proteins in sludge digestion using SDS-PAGE combined with LC-MS/MS [64]. The results suggest that activated sludge flocs contain different fractions of proteins and each fraction undergoes a different fate in anaerobic and aerobic digestion. Several bacterial proteins and sewage-derived polypeptides are identified, indicating that microbial interactions are mediated by extracellular enzymes [79]. In a recent study on the characterization of the microbial communities from continuous stirred tank reactors for digesting sewage sludge, a large number of proteins are identified as belonging to the “Candidatus Competibacter” group, suggesting that this microbial group play key roles in phosphorus removal [65]. Overall, meteproteomics enhances our understanding of the microbial communities and their functions in different sludge systems. 3.4. Acid Mine Drainage (AMD) Biofilm Metaproteomics AMD refers to the extremely acidic (pH < 3), metal-enriched waters derived from pyritic material [80]. AMD is regarded as the principle environmental problem in the global mining industry, and the water should be treated to remove metals and raise the pH before discharge. AMD is an enormous environmental issue associated with energy and metal resources; for example, the burning of sulfur-rich coal leads to the release of contaminants such as mercury and the formation of acid rain [81]. Microorganisms associated with AMD are of great concern, having many effects on the formation, the pollutant release, and the biological remediation of AMD. A pioneer study was conducted in 2005 using quantitative metaproteomic analysis to evaluate the in situ microbial activity of a natural AMD microbial biofilm community with low complexity [24]. In total, 2033 proteins were identified from the five most abundant microbial species and nearly half were derived from Leptospirillum group II. The high expression of proteins related to protein refolding and oxidative stress is regarded as a critical role for microbial survival. More excitingly, an abundant novel protein is determined to be a cytochrome central to iron oxidation and AMD formation. A similar proteogenomic strategy is employed to identify proteins in natural acidophilic biofilms [25]. With strain specificity, the proteomic results reveal a genome shaped by the recombination of two closely related bacterial populations. The confirmation of a large scale of inter-population genetic exchange indicates that this exchange is key to the adaptation to specific ecological niches partitioning the AMD biofilm. Denef et al. further analyzed the dominant Leptospirillum group II populations from 27 biofilms of the AMD system [66]. The results indicate that the specific environmental conditions select the particular recombinant variants, thus leading to a fine-scale tuning of microbial populations. Genes involved in motility, signal transduction and transport are over-expressed in tens to hundreds of kilobase recombinant blocks, whereas core metabolic functions are significantly down-expressed. Goltsman et al. also employ community genomic and proteomic approaches to investigate chemoautotrophic Fe-oxidizing Leptospirillum group II and III bacteria in AMD biofilms [67]. Leptospirillum groups II and III are responsible for 64.6% and 44.9% of the predicted proteins, respectively, and 20% of the proteins are identified as plasmid proteins. Among them, the proteins identified from both bacterial groups are involved in community-essential functions, including carbon fixation and the biosynthesis of vitamins, fatty acids, and biopolymers (including cellulose). Notably, these studies indicate that the AMD system is often dominated by Leptospirillum groups II and III. Signal transduction and methyl-accepting chemotaxis proteins are abundant in Leptospirillum group III, while Leptospirillum group III possesses a methyl-independent response pathway. 4. Challenges In general, the metaproteomic approach has been widely applied to study microbial communities from various environmental circumstances in the past few decades, and it has provided new insights into microbial diversity, metabolic potential, ecological function and microbe-environment interactions. However, because of the complexity and diversity of environmental samples, this technology still faces great challenges in the study of environmental microbial communities. To fully understand the role of microbial communities in the environment, we should obtain as wide a range of proteins or protein information as possible, especially for those low-abundance proteins. Up until now, there are still a few challenges from the technological point of view. The first challenge is sample collection and preservation. For most environmental samples, the density of the microbial population is very low, and, furthermore, the vast majority of microorganisms in the environment cannot be cultured. Various sample collection methods, such as ultrafiltration and flow cytometry sorting, have been developed, but these methods can neither separate different microorganisms in the populations nor obtain sufficient cell biomass of the different microorganism species. The protein information obtained from the metaproteome just reflects the abundant microbes in the populations but not the rare or sparse species, and this hinders our understanding of the function and role of different microbial species in the environment. On the other hand, in order to reflect the real microbial world, the in situ environmental sample should not be altered too much. For some samples, it is convenient and flexible to use quick fixation and preservation at a low temperature, such as liquid nitrogen, immediately after sample collection. However, it could become a problem when sampling is conducted under extreme environments, for example samples from habitats under oxygen deficit, extremely high or low temperatures or high pressure. Microbial communities are very sensitive to environmental alterations and respond quickly. When it is difficult to maintain the natural conditions, it is better to cut down the time cost of sample collection. Nevertheless, it is still a problem when the microbial biomass is low and has to be enriched, or a longer time is required to transport samples, such as in the case of deep ocean samples. In a word, care should be taken to maintain microbial communities at an in situ status, which is a big challenge for microbial metaproteomic studies. Another obstacle for microbial metaproteomics is protein extraction from complex environmental samples, which are mixtures of various organic and inorganic materials, such as humic acid, lignin, chemical chelation, cell exudation and various degradation products. Different protein extraction methods have been developed based on the features of the environmental samples, such as soils [82], sediments [54], sea water samples [15], activated sludge [83], biofilms [84], marine organic particles [85] and symbionts [16]. However, due to the heterogeneous species distribution, the wide range of protein abundance levels, and the unextractable proteins binding to the membrane or soil matrix, there is no standard and efficient protocol for extracting proteins from environmental samples [70]. Thus, it is essential to explore new extraction methods and optimize protein quantification efficiencies regarding the specific physical, chemical and biological properties of the individual sample. As the last but most important step, the identification of proteins is crucial for microbial metaproteomic studies. Several challenges hold back protein identification, which greatly depends on the database design, capacity and quality influencing the resulting peptide sequence matches. First, peptide sequence matching against such a large database suffers from the increased potential for false-positive matches, which lowers the number of highly confident true matches [86]. The second challenge is that the confidence of protein assignments to taxa is limited by the species present in the database, and functional assignments are often therefore more robust than taxonomic assignments of proteins. The last challenge concerns the explosive increase of the protein database which makes protein identification extremely time-consuming and demanding on hardware. To improve peptide identification, metagenomic data derived from the sampling sites can be used as the reference [26]. Another approach is to combine metagenomic with metaproteomic analysis, thus providing an enhanced means to reconstruct the microbial processes of a community [8]. 5. Perspectives Although challenges still exist in microbial community proteomics, the improvements in protein sample preparation and downstream MS technology along with the fast growth of bioinformatics tools and various databases might overcome these limitations and speed up microbial metaproteomic research. One of the trends of microbial community proteomics in the future will be the move from the qualitative analyses of function and activity surveys to the quantitative analyses of protein expressions and dynamics in environmental samples. Nowadays, to satisfy the demands of studies of systematic biology or biomarker discovery in environmental microbiology, it is necessary to obtain the precise quantitative information of a predefined set of proteins or all proteins in environmental samples. The most widely used technique is called selected reaction monitoring or multiple reaction monitoring (MRM) [87]. Recently, a group of protein biomarkers were reported to diagnose ocean metabolism in Pacific Ocean biomes using MRM [45], which provides a good example of a targeted proteomic study conducted in an environmental microbial community. However, this method could only target a limited number of proteins compared to a global proteomic approach. Recently, an alternative approach, the SWATH-MS technique, was introduced, in which fragment ion maps are generated using a data-independent acquisition method to give accurate global quantitative information. [87]. A proteome dataset with more accurate quantitative information combined with other omic (metagenomic and metatranscriptomic) datasets will provide an entire view of the functions, activities and interactions in real microbial communities. Another exciting new direction is the characterization of metabolic activities and the interpretation of adaptation for a specific microbial population or species in natural environments, given that many microbes cannot be cultivated and that the metabolism of those growing in monoculture are unlikely to be the same as those of organisms growing in assemblages. Improvements of the sensitivity and accuracy of the MS technique enable the possible MS-based characterization of amino acid substitutions. A proteogeonomic study in an AMD community [25] presents the strain-resolved capability of microbial metaproteomics. The key finding of this study is the possibility of identifying peptide sequences shared with sequenced organisms when multiple genomic data sets from closely related microbes in the community are available. The high resolution of the proteomic approach is further demonstrated by another study in a more complex community of surface seawater in the Sargasso Sea [56]. Moreover, proteomics-inferred genome typing reveals an adaption strategy of Leptospirillum group II to environmental stress through inter-population recombination [66]. Recently, a novel sequencing approach, single-cell sequencing [88,89], was developed, which provides cell-specific genetic information from a single cell of the non-cultured bacteria, even for the low-abundance organisms. In addition, both host sequence contamination and the difficulty of metagenomic assembly can be easily bypassed [3]. Therefore, elegant combinations of both single-cell sequencing and metaproteomics will provide new insights into the function and activity via species-specific protein identifications among a diverse community. Despite the currently high costs per sample and per depth of single-cell sequencing relative to metagenomics, the emergence of this technique will greatly advance the capability of cross-strain identification in community proteomics. With a high throughput of strain-specific proteome data, it also becomes possible to investigate the post-translational modifications in in situ environments, including phosphorylation, acetylation, glycosylation, ubiquitination and glutathionylation, which are extensively used by bacteria to transmit signals and to coordinate cellular functions. For example, bacterial protein phosphorylation is considered to be a signal transduction device which mainly links environmental factors to the regulation of important physiological processes [90]. Finally, besides metaproteomics, other omic approaches, such as metagenomics and metatranscriptomics, as well as metabolomics, are still being rapidly developed to pave the way for integrated multi-omic approaches in microbiology. With the advantages of computational tools, an understanding of the systems biology of the natural microbial community is the future trend to integrate and meta-analyze multiple data sets. Acknowledgments This study was supported by the National Natural Science Foundation of China through grant 41425021, and the Ministry of Science and Technology through grant 2015CB954003. Da-Zhi Wang was also supported by the “Ten Thousand Talents Program” for leading talents in science and technological innovation. John Hodgkiss is thanked for his help with English. Author Contributions Da-Zhi Wang and Zhang-Xian Xie conceptualized this review, Da-Zhi Wang, Ling-Fen Kong and Yuan-Yuan Li led the writing of this manuscript. All authors read and approved the manuscript, with Da-Zhi Wang being responsible for this article. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Typical workflow for microbial community proteomic analysis. ijms-17-01275-t001_Table 1Table 1 Survey of metaproteomic studies in different environments from a literature search. Environment Subject of Analysis Protein-Method Major Findings Refs/Year Separation Identification Marine and Freshwater South Atlantic Microbial membrane proteins from surface water LC-MS/MS TonB-dependent transporters dominated bacterial membrane proteins while bacterial rhodopsins were detected in every sample; Archaeal, ammonia monooxygenase proteins were identified in upwelling region. [43]/2010 The Northwest Atlantic Ocean Metabolic activity of microbial plankton in a seasonally hypoxic basin LC-MS/MS A seasonal increase in high-affinity membrane transport proteins involved in scavenging of organic substrates; Rhodobacterales transporters were strongly associated with the spring phytoplankton bloom, whereas SAR11 transporters were abundant in the underlying waters. [15]/2014 Six disparate aquatic habitats Microbial populations LC-MS/MS ABC-type sugar-, organic polyanion–, and glycine betaine–transport proteins were identified from Pelagibacter, and these transporters play important roles in carbon and nitrogen cycling. [44]/2014 The intersecting Pacific Ocean Multiple nutrients MS-based Multiple Reaction Monitoring Nitrogen response regulator NtcA was abundant, which was consistent with the prevalence of the Prochlorococcus urea transporter proteins (UrtA) in low-nitrogen areas. [45]/2014 The Antarctic Peninsula coast Bacterioplankton of winter and summer SDS–PAGE MS/MS Roseobacter clade contributed a large portion of ABC transporter for amino acids and polyamines; transporter proteins involved in amino acid, taurine and polyamine transport, as well as glutamine synthetase, were highly detected in SAR11; proteins involved in two chemolithoautotrophic pathways dominated the winter metaproteome of cold and dark polar water. [46]/2012 The Oregon coast Microbial plankton of upwelling region 2D-LC-MS/MS Thirty-six percent and 17% of detected proteins were from the SAR11 clade and Roseobacter clade; transporters for amino acids, taurine, polyamines and glutamine synthetase were highly detected; methanol dehydrogenase was detected. [47]/2011 Symbionts A gutless worm and its symbiotic microbial community 1D PAGE; 2D LC-MS/MS Sulfur oxidation proteins; aerobic and anaerobic CO dehydrogenases were detected in three types of Olavius algarvensis symbionts; high expression of periplasmic uptake (NiFeSe) hydrogenases and high-affinity uptake transport related proteins were detected. [16]/2013 The South China Sea Dissolved organic matter (DOM) from marine surface and bathypelagic region SDS-PAGE LC-MS/MS Archaea and Proteobacteria were the major contributors to bathypelagic proteome; protein compositions differed along the vertical water column, and urea ABC transporter was abundant in the surface DOM. [48]/2011 The English Channel Natural populations Protein-SIP LC-MS/MS RuMP cycle was the main carbon assimilation pathway in Methylophaga-like bacterium, and methanol dehydrogenase–encoding gene mxaF, as well as three out of four identified xoxF homologues were expressed. [49]/2015 The Lower St. Lawrence Estuary Microbial communities through the stratified water column LC-MS/MS Chemosynthetic production coupled to nitrification by MG-I Thaumarchaeota and Nitrospina was a dominant metabolic strategy; methanol oxidation proteins were detected from the OM43 marine clade; membrane transport proteins were assigned to the uncultivated MG-II Euryarchaeota. [50]/2015 Sulphidic marine sediments Trace metal utilization of methane-oxidizing microbial consortia LC-MS/MS Microbial consortia relied on the nickel metalloenzymes and transporters, cobalt metalloenzymes and transporters, molybdenum and tungsten enzymes to catalyze anaerobic oxidation of methane (AOM). [8]/2014 Ace Lake in Antarctica Green sulfur bacteria SDS-PAGE LC-MS/MS Proteins that participated in DNA processing, nucleic acid binding, folding/refolding of proteins and lipid biosynthesis were identified to be involved in cold adaption of green sulfur bacteria. [51]/2010 [52]/2011 The meromictic Lake Cadagno Green sulfur Bacterium Chlorobium clathratiforme LC-MS/MS Chlorobium clathratiforme contained enzymes for fixation of N(2) and oxidation of sulfide to sulfate, and they were not active in the dark; fermentation of polyglucose in the dark was the major pathway to obtain energy. [33]/2011 Hydrothermal venting sediments Microbial community structure and functioning SDS-PAGE LTQ Orbitrap-MS/MS Epsilonpro-teobacteria, δ- and γ-proteobacteria, ciliates, nematodes and various archaeal taxa were identified; high expressions of carbon fixation pathways as well as chemotaxis and flagella genes. [53]/2014 Marine seep sediments Free-living ANME-1; Sulfate-reducing bacteria 2-DE MS Anaerobic methanotrophic archaea dominated microbial species involved in the sulfur cycle and the biological sinking of methane; cold-adaptation proteins and key metabolic enzymes involved in the reverse methanogenesis and sulfate-reduction pathways were identified. [54]/2012 Symbionts Microbial community of the sponge Cymbastela LC-MS Proteins involved in cold adaptation and production of gas vesicles were abundant; high expressions of affinity transporters and alternative energy–utilizing proteins under stress conditions. [55]/2012 The Sargasso Sea Microbial membrane proteins of surface water; the SAR11 clade LC-MS/MS SAR11 periplasmic substrate-binding proteins (PBP) for phosphate were most abundant; proteins involved in amino acids, phosphonate, sugars and spermidine were detected. [56]/2009 The western South China Sea Particulate organic matters (POM) from marine surface and mesopelagic layers SDS-PAGE LC-MS/MS Cyanobacteria was the largest contributor; photosynthesis-associated proteins; porins, adenosine triphosphate synthases, nutrient transporters, molecular chaperones, and ectoenzymes were detected. [32]/2010 Soils Semiarid soils Functional and phylogenetic information SDS-PAGE LC-MS-MS Three protein extraction methods were examined, and the functional, phylogenetic and bio-geochemical information obtained by three methods in semiarid soils presented distinct edaphic properties. [18]/2014 Semiarid soils Deforestation fosters bacterial diversity and the cyanobacterial community SDS-PAGE LC-MS-MS Deforestation increased bacterial diversity in semiarid ecosystems and raised the abundance of cyanobacterial proteins involved in C-fixation in semiarid areas. [57]/2015 Beech leaf litter Environmental factors and nutrients on the decomposer structure and function SDS-PAGE LC-MS/MS Fungi were the main producers of extracellular hydrolytic enzymes, and microbial activity was stimulated at a higher litter nutrient content via a higher abundance and activity of extracellular enzymes. [58]/2012 Crop rhizospheric soil Crop soil metaproteomics 2-DE MALDI-TOF/TOF-MS Proteins involved in protein, energy, nucleotide, secondary metabolisms and signal transduction and resistance were identified; most upregulated plant proteins were involved in carbon and nitrogen metabolism and stress response, while the majority of the upregulated microbial proteins participated in protein metabolism and cell-wall biosynthesis. [19]/2011 [59]/2011 Toluene-amended soil The microbial community proteome SDS–PAGE MALDI-MS Glutamine synthetase (Gln), ABC transporters, extracellular solute-binding proteins, outer membrane proteins (Omp) were upregulated in toluene-amended soil; arginine deiminase (ArcA) and cold-shock protein presented in toluene-amended culture while superoxide dismutase (SodB) and chaperonin (GroEL) presented in toluene-amended soil. [1]/2010 Humic soil Enzymes connected with bacterial metabolic pathways SDS-PAGE; 2-DE LC-ESI-MS Protein extraction method from soil was developed, and 2,4-dichlorophenoxy acetate dioxygenase, chlorocatechol dioxygenases, molecular chaperons and transcription factors were identified. [23]/2007 Uranium-amended soil The subsurface microbial communities 2-DE LC-MS-MS The proteome was dominated by the enzymes converting acetate to acetyl-coenzyme A and pyruvate for central metabolism. Geobacter dominated the microbial community and they participated in energy generation during biostimulation. [60]/2009 Hydrocarbon-polluted soil Changes in the microbial community SDS-PAGE HPLC-MS/MS The complexity of the microbial community showed a relative increase in hydrocarbon-enriched cultures, and the majority of identified proteins were related to glycolysis pathways, structural or protein synthesis. [61]/2010 Hydrocarbon-polluted soil Changes in the microbial community SDS-PAGE LC-MS/MS Proteobacterial protein expressions increased while the abundance of Rhizobiales decreased during petroleum pollution; compost-assisted bioremediation was mainly driven by Sphingomonadales; abundances of catechol 2,3-dioxygenases, cis-dihydrodiol dehydrogenase and 2-hydroxymuconic semialdehyde were increased. [62]/2016 Wastewater and Activated Sludge Cadmium-polluted wastewater The response of a natural community 2-DE MALDI-TOF/TOF MS Significant community proteome responses to cadmium exposure were observed, and ATPases, oxidoreductases, and transport proteins played important roles in the cadmium shock. [63]/2007 Wastewater sludge Laboratory wastewater sludge microbial communities 2-D PAGE MALDI-TOF-MS Substantial differences in protein abundance for enzyme variants were uncovered among the A. phosphatis population, and these proteins were mainly involved in core metabolism, EBPR-specific pathways, energy generation. [4]/2004 [21]/2008 [22]/2008 Wastewater activated sludge Extracellular proteins in sludge digestion SDS-PAGE LC-MS/MS The proteins resistant to degradation and generated during anaerobic digestion were identified, including a limited number of bacterial and human polypeptides. [64]/2008 Sewage sludge Different proteins in two parallel anaerobic digestion lines SDS-PAGE LC-MS/MS Protein-inferred and 16S rDNA tags–based taxonomic community profiles were not consistent, and a high proportion of proteins belonged to “Candidatus Competibacter” group. [65]/2015 Acid Mine Drainage Biofilm Acid mine drainage biofilm(AMD) Gene expression, identified key activities, examined partitioning of metabolic functions 2-DE LC-MS/MS Half of the predicted proteins from the dominant biofilm organism Leptospirillum group II and protein involved in refolding and oxidative stress response presented high expressions; cytochrome played a central role in iron oxidation and AMD formation. [24]/2005 Acid mine drainage biofilm Biofilms growing at the liquid-air interface 2-DE LC-MS/MS Leptospirillum groups II and III dominated AMD system; signal transduction and methyl-accepting chemotaxis proteins were abundant in Leptospirillum group III, while Leptospirillum group III possessed a methyl-independent response pathway. [66]/2009 [67]/2009 ==== Refs References 1. Williams M.A. Taylor E.B. Mula H.P. Metaproteomic characterization of a soil microbial community following carbon amendment Soil Biol. Biochem. 2010 42 1148 1156 10.1016/j.soilbio.2010.03.021 2. Wilkins M.J. Wrighton K.C. Nicora C.D. Williams K.H. McCue L.A. Handley K.M. Miller C.S. Giloteaux L. Montgomery A.P. Lovley D.R. Fluctuations in species-level protein expression occur during element and nutrient cycling in the subsurface PLoS ONE 2013 8 1275 10.1371/journal.pone.0057819 23472107 3. Segata N. Boernigen D. Tickle T.L. Morgan X.C. Garrett W.S. Huttenhower C. Computational meta’omics for microbial community studies Mol. Syst. Biol. 2013 9 666 10.1038/msb.2013.22 23670539 4. Wilmes P. Bond P.L. The application of two-dimensional polyacrylamide gel electrophoresis and downstream analyses to a mixed community of prokaryotic microorganisms Environ. Microbiol. 2004 6 911 920 10.1111/j.1462-2920.2004.00687.x 15305916 5. Wang D.-Z. Xie Z.-X. Zhang S.-F. Marine metaproteomics: Current status and future directions J. Proteom. 2014 97 27 35 10.1016/j.jprot.2013.08.024 24041543 6. Baldrian P. Lopez-Mondejar R. Microbial genomics, transcriptomics and proteomics: New discoveries in decomposition research using complementary methods Appl. Microbiol. Biotechnol. 2014 98 1531 1537 10.1007/s00253-013-5457-x 24384749 7. Dyhrman S.T. Molecular approaches to diagnosing nutritional physiology in harmful algae: Implications for studying the effects of eutrophication Harmful Algae 2008 8 167 174 10.1016/j.hal.2008.08.016 8. Glass J.B. Yu H. Steele J.A. Dawson K.S. Sun S. Chourey K. Pan C. Hettich R.L. Orphan V.J. Geochemical, metagenomic and metaproteomic insights into trace metal utilization by methane-oxidizing microbial consortia in sulphidic marine sediments Environ. Microbiol. 2014 16 1592 1611 10.1111/1462-2920.12314 24148160 9. Kleiner M. Young J.C. Shah M. VerBerkmoes N.C. Dubilier N. Metaproteomics reveals abundant transposase expression in mutualistic endosymbionts mBio 2013 4 e00223 e00213 10.1128/mBio.00223-13 23781067 10. Moore C.M. Microbial proteins and oceanic nutrient cycles Science 2014 345 1120 1121 10.1126/science.1258133 25190779 11. Schneider T. Riedel K. Environmental proteomics: Analysis of structure and function of microbial communities Proteomics 2010 10 785 798 10.1002/pmic.200900450 19953545 12. Schneider T. Gerrits B. Gassmann R. Schmid E. Gessner M.O. Richter A. Battin T. Eberl L. Riedel K. Proteome analysis of fungal and bacterial involvement in leaf litter decomposition Proteomics 2010 10 1819 1830 10.1002/pmic.200900691 20198641 13. Ge Y. Wang D.-Z. Chiu J.-F. Cristobal S. Sheehan D. Silvestre F. Peng X. Li H. Gong Z. Lam S.H. Environmental omics: Current status and future directions J. Integr. OMICS 2013 3 75 87 10.5584/jiomics.v3i2.141 14. Kan J. Hanson T.E. Ginter J.M. Wang K. Chen F. Metaproteomic analysis of chesapeake bay microbial communities Saline Syst. 2005 1 7 10.1186/1746-1448-1-7 16176596 15. Georges A.A. El-Swais H. Craig S.E. Li W.K. Walsh D.A. Metaproteomic analysis of a winter to spring succession in coastal northwest Atlantic Ocean microbial plankton ISME J. 2014 8 1301 1313 10.1038/ismej.2013.234 24401863 16. Kleiner M. Wentrup C. Lott C. Teeling H. Wetzel S. Young J. Chang Y.J. Shah M. VerBerkmoes N.C. Zarzycki J. Metaproteomics of a gutless marine worm and its symbiotic microbial community reveal unusual pathways for carbon and energy use Proc. Natl. Acad. Sci. USA 2012 109 E1173 E1182 10.1073/pnas.1121198109 22517752 17. McCarren J. Becker J.W. Repeta D.J. Shi Y. Young C.R. Malmstrom R.R. Chisholm S.W. DeLong E.F. Microbial community transcriptomes reveal microbes and metabolic pathways associated with dissolved organic matter turnover in the sea Proc. Natl. Acad. Sci. USA 2010 107 16420 16427 10.1073/pnas.1010732107 20807744 18. Bastida F. Hernandez T. Garcia C. Metaproteomics of soils from semiarid environment: Functional and phylogenetic information obtained with different protein extraction methods J. Proteom. 2014 101 31 42 10.1016/j.jprot.2014.02.006 24530626 19. Wang H.B. Zhang Z.X. Li H. He H.B. Fang C.X. Zhang A.J. Li Q.S. Chen R.S. Guo X.K. Lin H.F. Characterization of metaproteomics in crop rhizospheric soil J. Proteome Res. 2011 10 932 940 10.1021/pr100981r 21142081 20. Martin H.G. Ivanova N. Kunin V. Warnecke F. Barry K.W. McHardy A.C. Yeates C. He S. Salamov A.A. Szeto E. Metagenomic analysis of two enhanced biological phosphorus removal (EBPR) sludge communities Nat. Biotechnol. 2006 24 1263 1269 10.1038/nbt1247 16998472 21. Wilmes P. Andersson A.F. Lefsrud M.G. Wexler M. Shah M. Zhang B. Hettich R.L. Bond P.L. VerBerkmoes N.C. Banfield J.F. Community proteogenomics highlights microbial strain-variant protein expression within activated sludge performing enhanced biological phosphorus removal ISME J. 2008 2 853 864 10.1038/ismej.2008.38 18449217 22. Wilmes P. Wexler M. Bond P.L. Metaproteomics provides functional insight into activated sludge wastewater treatment PLoS ONE 2008 3 1275 10.1371/journal.pone.0001778 18392150 23. Benndorf D. Balcke G.U. Harms H. von Bergen M. Functional metaproteome analysis of protein extracts from contaminated soil and groundwater ISME J. 2007 1 224 234 10.1038/ismej.2007.39 18043633 24. Ram R.J. VerBerkmoes N.C. Thelen M.P. Tyson G.W. Baker B.J. Blake R.C. Shah M. Hettich R.L. Banfield J.F. Community proteomics of a natural microbial biofilm Science 2005 308 1915 1920 10.1126/science.1109070 15879173 25. Lo I. Denef V.J. Verberkmoes N.C. Shah M.B. Goltsman D. DiBartolo G. Tyson G.W. Allen E.E. Ram R.J. Detter J.C. Strain-resolved community proteomics reveals recombining genomes of acidophilic bacteria Nature 2007 446 537 541 10.1038/nature05624 17344860 26. Williams T.J. Cavicchioli R. Marine metaproteomics: Deciphering the microbial metabolic food web Trends Microbiol. 2014 22 248 260 10.1016/j.tim.2014.03.004 24731505 27. Wilmes P. Bond P.L. Metaproteomics: Studying functional gene expression in microbial ecosystems Trends Microbiol. 2006 14 92 97 10.1016/j.tim.2005.12.006 16406790 28. Mazard S. Schäfer H. Stable isotope probing to study functional components of complex microbial ecosystems Environmental Microbiology Humana Press New York, NY, USA 2014 169 180 29. Amann R. Fuchs B.M. Single-cell identification in microbial communities by improved fluorescence in situ hybridization techniques Nat. Rev. Microbiol. 2008 6 339 348 10.1038/nrmicro1888 18414500 30. Ginige M.P. Hugenholtz P. Daims H. Wagner M. Keller J. Blackall L.L. Use of stable-isotope probing, full-cycle rRNA analysis, and fluorescence in situ hybridization-microautoradiography to study a methanol-fed denitrifying microbial community Appl. Environ. Microbiol. 2004 70 588 596 10.1128/AEM.70.1.588-596.2004 14711691 31. Chen S. Rillig M.C. Wang W. Improving soil protein extraction for metaproteome analysis and glomalin-related soil protein detection Proteomics 2009 9 4970 4973 10.1002/pmic.200900251 19743425 32. Dong H.-P. Wang D.-Z. Dai M. Hong H.-S. Characterization of particulate organic matter in the water column of the South China Sea using a shotgun proteomic approach Limnol. Oceanogr. 2010 55 1565 1578 10.4319/lo.2010.55.4.1565 33. Habicht K.S. Miller M. Cox R.P. Frigaard N.U. Tonolla M. Peduzzi S. Falkenby L.G. Andersen J.S. Comparative proteomics and activity of a green sulfur bacterium through the water column of lake Cadagno, Switzerland Environ. Microbiol. 2011 13 203 215 10.1111/j.1462-2920.2010.02321.x 20731699 34. VerBerkmoes N.C. Denef V.J. Hettich R.L. Banfield J.F. Systems biology: Functional analysis of natural microbial consortia using community proteomics Nat. Rev. Microbiol. 2009 7 196 205 10.1038/nrmicro2080 19219053 35. Wu C.C. Yates J.R. The application of mass spectrometry to membrane proteomics Nat. Biotechnol. 2003 21 262 267 10.1038/nbt0303-262 12610573 36. Elias J.E. Haas W. Faherty B.K. Gygi S.P. Comparative evaluation of mass spectrometry platforms used in large-scale proteomics investigations Nat. Methods 2005 2 667 675 10.1038/nmeth785 16118637 37. Lau H.T. Suh H.W. Golkowski M. Ong S.E. Comparing silac- and stable isotope dimethyl-labeling approaches for quantitative proteomics J. Proteome Res. 2014 13 4164 4174 10.1021/pr500630a 25077673 38. Ward M.A. Hyman B. Scherzer C. Trisini-Lipsanopoulos A. Ivinson A.J. Lößner C. Farztdinov V. Lovestone S. Pike I. Exploration of plasma biomarkers for alzheimer’s disease using isotopic tandem mass tags and a combined directed/data-dependent acquisition NLC-MS/MS method Alzheimer’s Dement. J. Alzheimer’s Assoc. 2014 10 P674 10.1016/j.jalz.2014.05.1215 39. Geiger T. Wisniewski J.R. Cox J. Zanivan S. Kruger M. Ishihama Y. Mann M. Use of stable isotope labeling by amino acids in cell culture as a spike-in standard in quantitative proteomics Nat. Protoc. 2011 6 147 157 10.1038/nprot.2010.192 21293456 40. Bettmer J. Application of isotope dilution ICP-MS techniques to quantitative proteomics Anal. Bioanal. Chem. 2010 397 3495 3502 10.1007/s00216-010-3861-y 20532485 41. Hägglund P. Bunkenborg J. Maeda K. Svensson B. Identification of thioredoxin disulfide targets using a quantitative proteomics approach based on isotope-coded affinity tags J. Proteome Res. 2014 7 5270 5276 19367707 42. Martínez-Esteso M.J. Casado-Vela J. Sellés-Marchart S. Pedreño M.A. Bru-Martínez R. Differential plant proteome analysis by isobaric tags for relative and absolute quantitation (iTRAQ) Plant Proteomics Humana Press New York, NY, USA 2014 155 169 43. Morris R.M. Nunn B.L. Frazar C. Goodlett D.R. Ting Y.S. Rocap G. Comparative metaproteomics reveals ocean-scale shifts in microbial nutrient utilization and energy transduction ISME J. 2010 4 673 685 10.1038/ismej.2010.4 20164862 44. Hanson B.T. Hewson I. Madsen E.L. Metaproteomic survey of six aquatic habitats: Discovering the identities of microbial populations active in biogeochemical cycling Microb. Ecol. 2014 67 520 539 10.1007/s00248-013-0346-5 24425229 45. Saito M.A. McIlvin M.R. Moran D.M. Goepfert T.J. DiTullio G.R. Post A.F. Lamborg C.H. Multiple nutrient stresses at intersecting pacific ocean biomes detected by protein biomarkers Science 2014 345 1173 1177 10.1126/science.1256450 25190794 46. Williams T.J. Long E. Evans F. Demaere M.Z. Lauro F.M. Raftery M.J. Ducklow H. Grzymski J.J. Murray A.E. Cavicchioli R. A metaproteomic assessment of winter and summer bacterioplankton from Antarctic Peninsula coastal surface waters ISME J. 2012 6 1883 1900 10.1038/ismej.2012.28 22534610 47. Sowell S.M. Abraham P.E. Shah M. Verberkmoes N.C. Smith D.P. Barofsky D.F. Giovannoni S.J. Environmental proteomics of microbial plankton in a highly productive coastal upwelling system ISME J. 2011 5 856 865 10.1038/ismej.2010.168 21068774 48. Wang D.-Z. Dong H.-P. Xie Z.-X. Dai M.-H. Hong H.-S. Metaproteomic characterization of dissolved organic matter in the water column of the South China Sea Limnol. Oceanogr. 2011 56 1641 1652 10.4319/lo.2011.56.5.1641 49. Grob C. Taubert M. Howat A.M. Burns O.J. Dixon J.L. Richnow H.H. Jehmlich N. von Bergen M. Chen Y. Murrell J.C. Combining metagenomics with metaproteomics and stable isotope probing reveals metabolic pathways used by a naturally occurring marine methylotroph Environ. Microbiol. 2015 17 4007 4018 10.1111/1462-2920.12935 26033676 50. Colatriano D. Ramachandran A. Yergeau E. Maranger R. Gelinas Y. Walsh D.A. Metaproteomics of aquatic microbial communities in a deep and stratified estuary Proteomics 2015 15 3566 3579 10.1002/pmic.201500079 26223443 51. Ng C. DeMaere M.Z. Williams T.J. Lauro F.M. Raftery M. Gibson J.A. Andrews-Pfannkoch C. Lewis M. Hoffman J.M. Thomas T. Metaproteogenomic analysis of a dominant green sulfur bacterium from Ace Lake, Antarctica ISME J. 2010 4 1002 1019 10.1038/ismej.2010.28 20237513 52. Lauro F.M. DeMaere M.Z. Yau S. Brown M.V. Ng C. Wilkins D. Raftery M.J. Gibson J.A. Andrews-Pfannkoch C. Lewis M. An integrative study of a meromictic lake ecosystem in Antarctica ISME J. 2011 5 879 895 10.1038/ismej.2010.185 21124488 53. Urich T. Lanzen A. Stokke R. Pedersen R.B. Bayer C. Thorseth I.H. Schleper C. Steen I.H. Ovreas L. Microbial community structure and functioning in marine sediments associated with diffuse hydrothermal venting assessed by integrated meta-omics Environ. Microbiol. 2014 16 2699 2710 10.1111/1462-2920.12283 24112684 54. Stokke R. Roalkvam I. Lanzen A. Haflidason H. Steen I.H. Integrated metagenomic and metaproteomic analyses of an ANME-1-dominated community in marine cold seep sediments Environ. Microbiol. 2012 14 1333 1346 10.1111/j.1462-2920.2012.02716.x 22404914 55. Liu M. Fan L. Zhong L. Kjelleberg S. Thomas T. Metaproteogenomic analysis of a community of sponge symbionts ISME J. 2012 6 1515 1525 10.1038/ismej.2012.1 22297557 56. Sowell S.M. Wilhelm L.J. Norbeck A.D. Lipton M.S. Nicora C.D. Barofsky D.F. Carlson C.A. Smith R.D. Giovanonni S.J. Transport functions dominate the SAR11 metaproteome at low-nutrient extremes in the sargasso sea ISME J. 2009 3 93 105 10.1038/ismej.2008.83 18769456 57. Bastida F. García C. von Bergen M. Moreno J.L. Richnow H.H. Jehmlich N. Deforestation fosters bacterial diversity and the cyanobacterial community responsible for carbon fixation processes under semiarid climate: A metaproteomics study Appl. Soil Ecol. 2015 93 65 67 10.1016/j.apsoil.2015.04.006 58. Schneider T. Keiblinger K.M. Schmid E. Sterflinger-Gleixner K. Ellersdorfer G. Roschitzki B. Richter A. Eberl L. Zechmeister-Boltenstern S. Riedel K. Who is who in litter decomposition? Metaproteomics reveals major microbial players and their biogeochemical functions ISME J. 2012 6 1749 1762 10.1038/ismej.2012.11 22402400 59. Wu L. Wang H. Zhang Z. Lin R. Zhang Z. Lin W. Comparative metaproteomic analysis on consecutively rehmannia glutinosa-monocultured rhizosphere soil PLoS ONE 2011 6 1275 10.1371/journal.pone.0020611 21655235 60. Wilkins M.J. Verberkmoes N.C. Williams K.H. Callister S.J. Mouser P.J. Elifantz H. N’Guessan A L. Thomas B.C. Nicora C.D. Shah M.B. Proteogenomic monitoring of geobacter physiology during stimulated uranium bioremediation Appl. Environ. Microbiol. 2009 75 6591 6599 10.1128/AEM.01064-09 19717633 61. Bastida F. Nicolás C. Moreno J. Hernández T. Garcia C. Tracing changes in the microbial community of a hydrocarbon-polluted soil by culture-dependent proteomics Pedosphere 2010 20 479 485 10.1016/S1002-0160(10)60037-9 62. Bastida F. Jehmlich N. Lima K. Morris B.E. Richnow H.H. Hernandez T. von Bergen M. Garcia C. The ecological and physiological responses of the microbial community from a semiarid soil to hydrocarbon contamination and its bioremediation using compost amendment J. Proteom. 2016 135 162 169 10.1016/j.jprot.2015.07.023 26225916 63. Carla M.R. Lacerda L.H.C. Kenneth F.R. Metaproteomic analysis of a bacterial community response to cadmium exposure J. Proteome Res. 2007 6 1145 1152 17284062 64. Chul Park J.T.N. Investigating the fate of activated sludge Water Environ. Fed. 2007 07 1408 1421 65. Bize A. Cardona L. Desmond-Le Quemener E. Battimelli A. Badalato N. Bureau C. Madigou C. Chevret D. Guillot A. Monnet V. Shotgun metaproteomic profiling of biomimetic anaerobic digestion processes treating sewage sludge Proteomics 2015 15 3532 3543 10.1002/pmic.201500041 26260998 66. Denef V.J. VerBerkmoes N.C. Shah M.B. Abraham P. Lefsrud M. Hettich R.L. Banfield J.F. Proteomics-inferred genome typing (PIGT) demonstrates inter-population recombination as a strategy for environmental adaptation Environ. Microbiol. 2009 11 313 325 10.1111/j.1462-2920.2008.01769.x 18826438 67. Goltsman D.S. Denef V.J. Singer S.W. VerBerkmoes N.C. Lefsrud M. Mueller R.S. Dick G.J. Sun C.L. Wheeler K.E. Zemla A. Community genomic and proteomic analyses of chemoautotrophic iron-oxidizing “leptospirillum rubarum” (group II) and “leptospirillum ferrodiazotrophum” (group III) bacteria in acid mine drainage biofilms Appl. Environ. Microbiol. 2009 75 4599 4615 10.1128/AEM.02943-08 19429552 68. Azam F. Malfatti F. Microbial structuring of marine ecosystems Nat. Rev. Microbiol. 2007 5 782 791 10.1038/nrmicro1747 17853906 69. Morris R.M. Rappé M.S. Connon S.A. Vergin K.L. Siebold W.A. Carlson C.A. Giovannoni S.J. SAR11 clade dominates ocean surface bacterioplankton communities Nature 2002 420 806 810 10.1038/nature01240 12490947 70. Taylor E.B. Williams M.A. Microbial protein in soil: Influence of extraction method and c amendment on extraction and recovery Microb. Ecol. 2010 59 390 399 10.1007/s00248-009-9593-x 19844650 71. Alonso-Saez L. Waller A.S. Mende D.R. Bakker K. Farnelid H. Yager P.L. Lovejoy C. Tremblay J.E. Potvin M. Heinrich F. Role for urea in nitrification by polar marine archaea Proc. Natl. Acad. Sci. USA 2012 109 17989 17994 10.1073/pnas.1201914109 23027926 72. Yong S.C. Roversi P. Lillington J. Rodriguez F. Krehenbrink M. Zeldin O.B. Garman E.F. Lea S.M. Berks B.C. A complex iron-calcium cofactor catalyzing phosphotransfer chemistry Science 2014 345 1170 1173 10.1126/science.1254237 25190793 73. Cvetkovic A. Menon A.L. Thorgersen M.P. Scott J.W. Poole F.L. 2nd Jenney F.E. Jr. Lancaster W.A. Praissman J.L. Shanmukh S. Vaccaro B.J. Microbial metalloproteomes are largely uncharacterized Nature 2010 466 779 782 10.1038/nature09265 20639861 74. Becher D. Bernhardt J. Fuchs S. Riedel K. Metaproteomics to unravel major microbial players in leaf litter and soil environments: Challenges and perspectives Proteomics 2013 13 2895 2909 10.1002/pmic.201300095 23894095 75. Singleton I. Merrington G. Colvan S. Delahunty J.S. The potential of soil protein-based methods to indicate metal contamination Appl. Soil Ecol. 2003 23 25 32 10.1016/S0929-1393(03)00004-0 76. Chourey K. Nissen S. Vishnivetskaya T. Shah M. Pfiffner S. Hettich R.L. Loffler F.E. Environmental proteomics reveals early microbial community responses to biostimulation at a uranium- and nitrate-contaminated site Proteomics 2013 13 2921 2930 10.1002/pmic.201300155 23894087 77. Wilmes P. Heintz-Buschart A. Bond P.L. A decade of metaproteomics: Where we stand and what the future holds Proteomics 2015 15 3409 3417 10.1002/pmic.201500183 26315987 78. Puigagut J. Salvadó H. García J. Short-term harmful effects of ammonia nitrogen on activated sludge microfauna Water Res. 2005 39 4397 4404 10.1016/j.watres.2005.08.008 16213541 79. Park C. Helm R.F. Novak J.T. Investigating the fate of activated sludge extracellular proteins in sludge digestion using sodium dodecyl sulfate polyacrylamide gel electrophoresis Water Environ. Res. 2008 80 2219 2227 10.2175/106143008X325791 19146099 80. Xie J. He Z. Liu X. Liu X. van Nostrand J.D. Deng Y. Wu L. Zhou J. Qiu G. Geochip-based analysis of the functional gene diversity and metabolic potential of microbial communities in acid mine drainage Appl. Environ. Microbiol. 2011 77 991 999 10.1128/AEM.01798-10 21097602 81. Banfield J.F. Verberkmoes N.C. Hettich R.L. Thelen M.P. Proteogenomic approaches for the molecular characterization of natural microbial communities Omics: J. Integr. Biol. 2005 9 301 333 10.1089/omi.2005.9.301 16402891 82. Keiblinger K.M. Wilhartitz I.C. Schneider T. Roschitzki B. Schmid E. Eberl L. Riedel K. Zechmeister-Boltenstern S. Soil metaproteomics—Comparative evaluation of protein extraction protocols Soil Biol. Biochem. 2012 54 14 24 10.1016/j.soilbio.2012.05.014 23125465 83. Hansen S.H. Stensballe A. Nielsen P.H. Herbst F.A. Metaproteomics: Evaluation of protein extraction from activated sludge Proteomics 2014 14 2535 2539 10.1002/pmic.201400167 25116144 84. Leary D.H. Hervey W.J.T. Li R.W. Deschamps J.R. Kusterbeck A.W. Vora G.J. Method development for metaproteomic analyses of marine biofilms Anal. Chem. 2012 84 4006 4013 10.1021/ac203315n 22468925 85. Dong H.-P. Wang D.-Z. Dai M. Chan L.L. Hong H.-S. Shotgun proteomics: Tools for analysis of marine particulate proteins Limnol. Oceanogr. Methods 2009 7 865 874 10.4319/lom.2009.7.865 86. Jagtap P. McGowan T. Bandhakavi S. Tu Z.J. Seymour S. Griffin T.J. Rudney J.D. Deep metaproteomic analysis of human salivary supernatant Proteomics 2012 12 992 1001 10.1002/pmic.201100503 22522805 87. Maiolica A. Junger M.A. Ezkurdia I. Aebersold R. Targeted proteome investigation via selected reaction monitoring mass spectrometry J. Proteom. 2012 75 3495 3513 10.1016/j.jprot.2012.04.048 22579752 88. Chitsaz H. Yee-Greenbaum J.L. Tesler G. Lombardo M.-J. Dupont C.L. Badger J.H. Novotny M. Rusch D.B. Fraser L.J. Gormley N.A. Efficient de novo assembly of single-cell bacterial genomes from short-read data sets Nat. Biotechnol. 2011 29 915 921 10.1038/nbt.1966 21926975 89. Thrash J.C. Temperton B. Swan B.K. Landry Z.C. Woyke T. DeLong E.F. Stepanauskas R. Giovannoni S.J. Single-cell enabled comparative genomics of a deep ocean SAR11 bathytype ISME J. 2014 8 1440 1451 10.1038/ismej.2013.243 24451205 90. Kobir A. Shi L. Boskovic A. Grangeasse C. Franjevic D. Mijakovic I. Protein phosphorylation in bacterial signal transduction Biochim. Biophys. Acta 2011 1810 989 994 10.1016/j.bbagen.2011.01.006 21266190

PMC005xxxxxx/PMC5000674.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081277ijms-17-01277ArticleNew Natural Pigment Fraction Isolated from Saw Palmetto: Potential for Adjuvant Therapy of Hepatocellular Carcinoma Tan Hor-Yue 1Wang Ning 1Takahashi Masao 2Feng Yigang 3Li Hongyun 3Feng Yibin 1*Choi Chang Won Academic Editor1 LKS Faculty of Medicine, School of Chinese Medicine, The University of Hong Kong, 10 Sassoon Road, Pokfulam, Hong Kong, China; [email protected] (H.-Y.T.); [email protected] (N.W.)2 Heimat Co., Ltd., Heimat Building, 1-21-3 Nihonbashi, Chuo-Ku, Tokyo 103-0027, Japan; [email protected] Guanghua School of Stomatology, Hospital of Stomatology, Sun Yat-sen University, Guangzhou 510055, China; [email protected] (Y.F.); [email protected] (H.L.)* Correspondence: [email protected]; Tel.: +852-2589-0482; Fax: +852-2872-547605 8 2016 8 2016 17 8 127723 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).For the first time, we discovered a small proportion of aqueous fraction from Saw Palmetto apart from the fatty acid-rich fraction exhibited pharmacological activity. Therefore, this study aims to explore the anti-tumor potential of red pigmented aqueous fraction of Saw Palmetto, NYG on human hepatocellular carcinoma and its possible targets. Subcutaneous xenograft and orthotopic implantation models of HCC were used to evaluate the tumor inhibitory effect of NYG. Human hepatocellular carcinoma (HCC) cell lines and human umbilical vein endothelial cells (HUVEC) were used as in vitro model. The mRNA expression was conducted by qPCR. Protein expression was monitored by immunoblotting and immunohistochemistry. Cell migration and blood vessel formation were determined by chamber assay and tube formation assay, respectively. Significant tumor inhibition of NYG in dose-dependent manner was observed on subcutaneous xenograft and orthotopic HCC model. NYG has no direct action on cell viability or VEGF secretion of HCC cells. However, NYG reduced in vitro migration and vessel formation activities of HUVEC cells, as well as in vivo intratumoral neovascularization. NYG attenuated extracellular signal-regulated kinases (ERK) activation in endothelial cells, which may be associated with the suppression of migration and tube formation of HUVEC. NYG suppressed tumor expansion of HCC via inhibiting neovascularization, and may be potential adjuvant treatment for HCC. saw palmettoNYGhepatocellular carcinomaHUVECangiogenesis ==== Body 1. Introduction Liver cancer is one of the most prevalent human malignancies all over the world. As the sixth most common cancer, there are 782,000 new cases of liver cancer diagnosed annually and the death rate is increasing annually [1]. High mortality rate due to liver cancer was reported in Asia and Africa regions, especially in those less developed countries [2]. Hepatocellular carcinoma (HCC) is the major form, accounting for 85% of liver cancer cases. Treatments for HCC are limited, and operation including liver transplantation and surgical resection is the best prognosis across different treatments [3]. However, only 15% of HCC patients diagnosed are suitable for surgical methods, and non-surgical treatment is still under great demand. Unfortunately, as most HCC are highly radio- and chemo-resistant, desirable therapeutic outcome is hardly achieved in clinical cases with general cancer treatment protocols [4]. Therefore, therapeutic agents that specifically and effectively target on HCC are always in need. Tumor neovascularization is the process in which endothelial cells of intra-tumoral blood vessel proliferates and migrates to form new vasculatures, in order to supply oxygen and nutrients for the rapid growth of tumor cells [5]. Agents in blocking tumor neovascularization in HCC are available, and sorafenib was recently approved by FDA in targeting this process for HCC treatment [6]. Although sorafenib may extend the life span of HCC patients, the treatment is associated with various adverse effects that cannot be neglected, and the high cost of sorafenib renders financial burden to the patients [7]. Therefore, searching for an alternative medicine targeting on tumor neovascularization for treatment of HCC is still of interest. Saw Palmetto is the fruit extract of Serenoa repens (Bartram) J.K.Small, an edible plant originated from Southeastern United States. Approximately 90% of its containing fatty acids are currently manufactured as a nutrient supplement for its therapeutic efficacy on benign prostatic hyperplasia (BPH) [8]. Apart from its well-known effect in BPH patient, the pharmacological effects of saw palmetto such as immune-modulatory effect [9] and inhibitory effect in lipid droplet and adipocyte accumulation are also extensively explored [10]. Previous studies showed the in vitro anti-cancer effect of lipidosterolic fraction of Saw Palmetto through inducing cell apoptosis in cancer cell lines [11]. In vivo study using transgenic prostate adenocarcinoma murine model also postulated that the administration of lipidosterolic extract (300 mg/kg/day) may prevent tumor occurrence [12]. However, epidemiology [13] and randomized trials [14] demonstrated that there is no association of prostate cancer risk between the intervention and placebo group. Although the result discrepancy between the studies may not conclude the effect of Saw Palmetto lipidosterolic extract in prostate cancer prevention, we could not preclude the potential anti-tumor effect of Saw Palmetto in other tumor models. Fatty acids are key components of Saw Palmetto and may majorly contribute to the abovementioned pharmacological actions; nonetheless, a small proportion of natural pigment compounds have also been isolated from the commercial Saw Palmetto Extract [15]. To date, no previous study has reported any bioactivity of these pigmented fractions from Saw Palmetto. In this study, we reported for the first time the anti-tumor activity of the Saw Palmetto red pigmented aqueous fraction named NYG (patent number: WO 2014174703 A1). We employed the murine subcutaneous and orthotopic HCC models to investigate any tumor regression after NYG treatment, and probed out its possible target with multiple cell models. We found that NYG exhibited potent inhibitory effect on xenograft and orthotopic growth of HCC as well as reduced the in vivo neovascularization. However, we did not observed any effect of NYG on the in vitro viability and proliferation of HCC cells; it also did not reduce the secretion of neovascularization-favoring factor, vascular endothelial growth factor (VEGF), from tumor cells. Instead, NYG reduced the migration and blood vessel formation of endothelial cells throughout tumor stroma. Suppression of VEGF-induced ERK activation may be involved in the pharmacological action of NYG on endothelial cells. Our results shed light on the potential of Saw Palmetto aqueous fraction as adjuvant treatment of HCC via targeting tumor neovascularization. 2. Results 2.1. NYG Suppressed Xenograft Growth of Hepatocellular Carcinoma (HCC) in Vivo The lipidosterolic fraction of Saw Palmetto is frequently studied and actively used as health supplement for prevention of BPH and hair loss. Its aqueous fraction is rarely investigated, however, a previous study showed that the acidic water extract of Saw Palmetto exhibited anti-oxidant and COX-2 inhibitory effect [16]. This observation also further supported another study on the inhibitory activity of Saw Palmetto berry extract on COX-2, which is associated with its prostate cancer cell growth suppression [17]. In accordance with the previous study, we hypothesized that the aqueous fraction of Saw Palmetto may exhibit tumor inhibitory effect. The water-soluble fraction, namely NYG, was prepared following stringent manufacturing practice by Heimat Co., Ltd. (Tokyo, Japan). Our preliminary compound characterization using thin layer chromatography has suggested that the containing compounds of NYG are mainly composed of proanthocyanidins and oil elements. Further study in compound characterization using other means is essentially needed to confirm the ingredients contained and serve as quality control of NYG. As NYG is a novel fraction isolated from Saw Palmetto, we first examined its toxicity by dose escalation method. Mice were treated with NYG at doses of 0.1, 1, 10 and 100 mg/kg via intraperitoneal injection on five consecutive days. One day after injection, four out of five mice at treatment group of 100 mg/kg died, while mice in the other groups exhibited normal behavior. After four-days of intervention, the mice in 100 mg/kg NYG treatment group died (Figure 1A). The LD50 of single treatment was calculated as approximately 66.3 mg/kg. NYG administration below LD50 is considered safe. Next, we examined the in vivo anti-tumor effect of NYG on xenograft growth of MHCC97L cell in nude mice. NYG treatment (5 mg/kg and 10 mg/kg every two days) was administrated intraperitoneally after one week of tumor inoculation. PBS and beta-cyclodextrin (B-CD) was given to negative control groups of mice while mice receiving doxorubicin (2.5 mg/kg) served as positive control. We observed that treatment of NYG exhibited least toxicity to the mice, as evidenced by maintenance of body weight during the whole treatment (Figure 1B). As postulated in Figure 1C, the mice group intervened with NYG (10 mg/kg) exhibited slower growth rate of HCC cells compared to vehicle receiving group. Both doses of NYG administration reduced tumor size by the end of four-week treatment; while NYG in 10 mg/kg demonstrated much potent tumor inhibitory effect suggested the dose-dependent efficacy of NYG (Figure 1D,E). Treatment of B-CD, the pharmaceutical excipient of NYG showed minimal effect on tumor growth and body weight of mice (Figure 1B,E), which reflected that the anti-tumor activity on human HCC is solely contributed by NYG itself. Immunostaining of CD31 on xenograft HCC tumor showed significant reduced micro-vessel formation after NYG intervention (Figure 1F). Overall, these results postulate that NYG inhibited HCC xenograft growth and the effect is mainly contributed by reduced angiogenesis in tumor environment. 2.2. NYG Inhibited Orthotopic Implanted HCC Growth in Vivo Xenograft model is often used as the first line model for investigating the anti-cancer efficacy of new therapeutic agents, however, this preclinical model has its limitation in reflecting the liver tumor microenvironment and renders poor prognostic outcome of drug efficacy [18]. Therefore, we established the orthotopic HCC implantation model in which the MHCC97L cells tagged with luciferase are implanted onto the right lobe of mice liver. The orthotopic HCC tumor growth will be monitored by live-animal imaging throughout the intervention period. After one-week of model establishment, the mice with observable luciferase intensity will be chosen and further randomized into two groups: Negative control group receiving PBS, and NYG (10 mg/kg) intervention group. As observed from the luciferase signal intensity plot, the orthotopic HCC tumor growth rate was decelerated in NYG intervened mice group after Week 2 of treatment (Figure 2A), while the tumor growth of vehicle-receiving mice group increased exponentially within five weeks. By the end of experiment, we observed significant reduction in liver tumor size of NYG-intervened group, which accounted for approximately 60% suppression of HCC tumor growth in NYG-treated mice as compared to control mice (Figure 2B). Similar to subcutaneously grown tumor, NYG intervention also significantly reduced CD31-positive cell populations in orthotopically-grown liver tumor, suggested the tumor inhibitory effect of NYG on orthotopic implanted HCC growth may be partly contributed by reduced neovascularization by NYG (Figure 2C). 2.3. NYG Exerted Minimal Effect on in Vitro Cultured HCC Cells To elucidate the effect of NYG, we further investigated whether the fraction suppresses HCC cell proliferation. The MTT assay was performed to examine the cytotoxic doses of NYG on individual HCC cells. Surprisingly, NYG up to 500 μg/mL exhibited no potent cytotoxicity to HepG2 and MHCC97L, the two human hepatocellular carcinoma cell lines, even if the incubation time is extended to 72 h (Figure 3A,B). As expected from the minimal toxicity incurred by NYG on HCC cells, NYG also exerted no cytotoxicity on normal hepatic cell line L-02, up to concentration of 500 μg/mL (Figure 3C). Prompted by the observation of reduced CD31-stained in vivo vascular cell density, we further examined the expression of VEGF, the angiogenic-favoring factor in NYG-treated HCC cells. The MHCC97L cells were supplemented with 250 and 500 μg/mL of NYG; the cells and culture supernatant were harvested after 48 h of incubation. The quantitative PCR analysis showed that NYG intervention has least effect on mRNA expression of VEGF, in both normoxia and hypoxic condition (Figure 3D). To validate, the secretion of VEGF protein by MHCC97L cells was determined by ELISA assay, and we did not observed any inhibition on VEGF secretion by NYG (Figure 3E). These results indicate that the tumor inhibitory effect of NYG was independent to its action on HCC cells. 2.4. NYG Reduced Migration and Tube Formation of HUVECs Angiogenesis is the formation of new blood vessels with the purpose of supplying nutrients and oxygen for tumor cell growth. The blood vasculature is mainly supported by the inter-connected endothelial cells with the properties of invasion from basement membrane, migration, proliferation and sprouts-formation [19]. Therefore, we further examined whether the in vivo tumor suppression by NYG is attributed by reduced initiation of neovascularization by endothelial cells. Briefly, we exposed the human umbilical vein endothelial cells (HUVEC) to VEGF, the angiogenesis-favoring factor, alone or in combination with NYG and the behaviors of HUVEC were further investigated. Similar to the results obtained from human cancer cells, we observed minimal cytotoxicity against HUVEC induced by NYG up to concentration of 500 μg/mL. The 50% inhibitory concentration (IC50) of NYG on HUVEC is approximately 2 mg/mL and 4 mg/mL with incubation time of 24 and 48 h, respectively (Figure 4A). We then examine whether NYG reduces motility of HUVEC. Using matrigel invasion chamber assay, we observed reduced percentage of migrated HUVEC from the apical side to basal side of the chamber in the presence of NYG at non-toxic doses (Figure 4B). The motility blockade of endothelial cells towards chemoattractant VEGF by NYG indicated that functions of endothelial cells may be interrupted in the presence of NYG. Observation from tube formation assay supported the hypothesis, as the formation of capillary-like tubular structure by HUVEC was completely attenuated by NYG treatment (Figure 4C). The inhibitory effects of NYG at 100 μg/mL on HUVEC motility and tube formation are comparable to 30 μM of suramin, the angiogenesis inhibitor. Altogether, our results indicated that NYG targeting on endothelial cells in suppressing the vasculature formation in tumor stroma. 2.5. Inhibition of Tumor Neovascularization by NYG May Be Related to Inactivation of ERK in Endothelial Cells The ERK/MAPK signaling pathway has been implicated in promoting tumor angiogenesis, primarily involved in endothelial cell survival, migration and sprouting [20]. We thus examined whether ERK participated in the NYG-mediated inhibitory effects on endothelial cells. As observed, NYG potently reduced phosphorylated activation of ERK induced by VEGF on HUVEC (Figure 5A). Erk pathway is generally quiescent and only transiently activated upon exposure to certain stimulus. In our study, ERK was activated in HUVECs upon being challenged by VEGF. Previous study showed that inhibition of ERK is responsible for the reduced angiogenesis in VEGF-treated HUVECs [21]. To further validate whether the reduced in vivo tumor neovascularization involves blockade of ERK activity by NYG, we co-stained the blood vessel network in tumor stroma with CD31 and phosphorylated-ERK (p-ERK) before being subjected to confocal microscopy analysis. We observed that p-ERK was highly expressed on CD31 stained vasculature in xenografted tumor and intervention of NYG suppressed the expression of p-ERK (Figure 5B). However, future study in observing the changes of pERK in the relationship of variety time points and doses as well as any possibility of affecting ERK downstream or upstream targets is definitely needed to confirm the role of ERK upon NYG intervention. Overall, our results indicated that inhibition of ERK may be associated with NYG-regulated inhibitory functions of endothelial cells. 3. Discussion Neovascularization is an essential process, allowing the expansion of tumor cells from the primary site. The process also involves the proliferation and migration of endothelial cells within tumor stroma; and formation of microvasculature by endothelial cells further supply sufficient oxygen and nutrients for the rapid tumor expansion [5]. Neovascularization is critically involved in progression of metastatic cancers [22]. Secretion of vascular endothelial growth factors (VEGF) by tumor cells has been postulated as the determinant factor [23]; however, in our study, we found that NYG has minor effect on the oncogenic property of HCC cells in vitro, and did not affect the expression and secretion of VEGF on HCC cells. This observation excludes the possibility that NYG target on upstream regulation of tumor neovascularization. The binding of VEGF to its membrane receptors, VEGFR, activates multiple intracellular signal transductions and allows transcriptional activation of proliferation, motility and permeability related genes [24]. Previous studies showed that activation of ERK signaling is majorly contributed to the functional expansion [25] and migration of endothelial cells [26]. We observed that in cultured endothelial cell line HUVEC, NYG potently suppressed the VEGF-induced phosphorylation of ERK signaling, indicating that activation of ERK by VEGF was blocked upon NYG intervention and this action may be VEGF-dependent (Figure 6). It is noted that many nature derived anti-cancer compounds have poor solubility, low stability and bioavailability, which hinder the progression of the use of natural product for cancer therapy [27]. The excipients of NYG, β-cyclodextrin (B-CD), has been widely used by pharmaceutical industries as complexing agents, which have been scientifically proven to improve the drug solubility, bioavailability, stability and safety to human body [28]. However, some studies also claimed that B-CD may cause in vitro and in vivo toxicities to cell lines and animals. Previous studies have suggested that the route of administration may render the toxicity of B-CD [29]. Our study therefore has included another group of experimental animals with the intervention of B-CD, which showed minimal toxicity to the animals as observed from the absence of obvious body weight decrease in the mice group with i.p. administration. The in vitro studies revealed that 1% (w/v) of B-CD on HaCaT keratinocytes resulted in 52.23% cell death [30] and 5 mmol/L (approximately 5.675 mg/mL) led to 60% death on P388 cells [31], further postulated the cytotoxicity of B-CD at certain high concentrations. Consistent with the previous studies, our results showed that B-CD concentration below 500 μg/mL exerted no toxicity to HepG2 and MHCC97L liver cancer cells. Therefore, we expected the concentration of 0.1 mg/mL of NYG with the content of approximately 0.095 mg/mL of B-CD is non-toxic to the cells. It is frequently observed in clinical practice in which the patients are associated with severe adverse side effects like nausea, vomiting, diarrhea and lack of appetite after chemotherapy. These side effects could be attributed to the toxic reaction on normal cells incurred by chemotherapeutics agents [32]. Therefore, it seems to be an evitable issue in the use of chemotherapeutic agents with direct toxicity to cancer cells. Besides, due to the chemo-resistant characteristics of HCC, the dose of chemotherapeutic agents used may be much higher resulting in greater risk of adverse reactions. In our observation, though NYG did not exhibit any toxicity to HCC cells, the fraction also presented minimal effect on normal hepatic cells, which may suggest the safety of NYG on patients. The observation of animal body weight changes under treatment of NYG also further supports the safe use of NYG, as there is no adverse reaction or body weight loss observed at certain dose of NYG. As observed in our study, the reduced tumor expansion by NYG is not due to shrinkage of tumor cell population, but related to the restricted motility of endothelial cells and vasculature formation within tumor stroma. Failure of constructing vascular network by endothelial cells limited the oxygen and nutrients supply to the tumor cells, resulting in tumor growth retardation in NYG-treated mice. NYG intervention may not completely eradicate tumor cells; however, the rapid growth and proliferation of solid tumor was retarded. The tumor cells may remain at lower proliferation rate and vulnerable state in which low dose of chemotherapeutic agents may be employed to gain desirable therapeutic outcome with minimal side effects. With these concerns, we regarded NYG as potential adjuvant therapy to HCC, by restricting tumor expansion via neovascularization inhibition, and may be used as complementary treatment to other first line cancer therapeutic agents. Future study focusing on efficacy of combination use of first line treatment with NYG and the possible herbal–drug interaction should be conducted to validate the hypothesis. 4. Materials and Methods 4.1. Preparation of NYG (Red Pigment from Saw Palmetto) Saw palmetto powder was extracted with hot 90% ethanol (v/v). After extraction, the residue was separated from the supernatant by filtration. The supernatant was then concentrated by evaporation. After that, an equivalent of water was added to the concentrated extract with stirring. The mixture allows settling at room temperature in order to achieve efficient separation between oil (upper) and water (lower) phases. The water layer was collected in a separating funnel and the wet crystalline component was harvested by filtration. This followed by addition of an equivalent weight of beta-cyclodextrin (B-CD) and 3 times volume of 90% ethanol (v/v) to the wet crystalline material and mixed well. The homogenized paste was dried under reduced pressure. The dried fraction was re-slurried with 90% ethanol (v/v), and the fats and other impurities were washed out. The washed slurry was dried again, and the net weight of crystalline material calculated. The final concentration was adjusted by B-CD (20 times trituration), which means 1 portion of natural pigment extract of Saw Palmetto was mixed with 19 portion of B-CD evenly. 4.2. Cell Line and Cell Culture Human hepatocellular carcinoma cell lines HepG2 was purchased from ATCC and MHCC97L was gifted by Dr. Man Kwan, Department of Surgery, The University of Hong Kong. L-02 cell line was purchased from Experimental Animal Center of Sun Yat-sen University, Guangzhou. Cell lines were maintained in DMEM (Gibco, Carlsbad, CA, USA) supplemented with 10% FBS and 1% penicillin/streptomycin. HUVEC cell line was purchased from ATCC and it was maintained in EGM-2 complete medium (Lonza, Basel, Switzerland) at 5% of CO2 incubator. The cell lines were passaged whenever they reached 80% confluency. 4.3. Cell Viability Assay The cell viability was determined using MTT assay according to the previous publication by Mosmann [33]. The cells were treated with serial concentrations of NYG for 24, 48 and 72 h. At the end of the incubation, 10 μL of MTT solution was added to each well and incubated for another 3 h. One hundred microliters of DMSO was added to dissolve formazan crystal before absorbance measurement at wavelength of 570 nm. 4.4. Migration and Tube Formation Assay For migration assay, HUVECs were seeded on transwell insert 8 μm (Costar). The receiving chamber was filled with serum free cell culture medium supplemented with 20 ng/mL VEGF with or without NYG and Suramin. The cells were allowed to migrate at 37 °C for 4 h. After that, the remaining cells at upper insert were removed by cotton swab and cells at lower chamber were fixed and stained with crystal violet. The number of migrating cells was counted in five fields per well under microscope. As for tube formation assay, HUVECs with or without treatment were seeded on a 96-well plate that was pre-coated with Matrigel basement membrane matrix (BD). The cells were incubated at 37 °C for 4 h before visualized under microscope. 4.5. Quantitative Real-Time PCR Total RNA was extracted from NYG-treated cells using RNA isoplus reagent (Takara, Tokyo, Japan). Reverse transcription was performed using cDNA reverse transcription kit (Takara). Quantitative real-time PCR was performed with SYBR premix Ex Taq (Takara) with Light Cycler 480 real time PCR system (Roche, Basel, Switzerland). The primer sequence is as followed: VEGF 5’-CCTCCGAAACCATGAACTTT-3’ (forward) and 5’-TTCTTTGGTCTGCATTCACATT-3’ (reverse). 4.6. Western Blotting The NYG-treated cells were lysed with RIPA buffer and protein concentration was determined using Bradford protein assay. Ten micrograms of protein samples were separated on 12% SDS-acrylamide gel before transferring to polyvinylidenedifluoride membranes. Membranes were blocked with 5% bovine serum albumin before incubated with primary antibodies to rabbit anti-GADPH and anti-phosphorylated ERK. The membrane were then incubated with horseradish peroxidase-conjugated rabbit antibody and visualized under chemiluminescence system (Biorad, CA, USA). 4.7. Immunohistochemistry Tumors were fixed in 30% sucrose, freezed under −20 °C in OCT and sectioned at thickness of 8 μm. The sections were blocked by 10% goat serum and incubated with anti-mouse CD31 and p-ERK overnight. The multi vessel density of each section was evaluated through staining with FITC conjugated antibody for 2 h and counterstained with DAPI before visualized under fluorescent microscope. 4.8. ELISA Assay on VEGF Secretion The VEGF secretion was measured using human VEGF ELISA kit (ExCell Biology) according to the manufacturer’s protocol. The cell supernatant was seeded onto the plate pre-coated with anti-VEGF monoclonal antibody. After that, the wells were incubated with HRP-conjugated streptavidin for 30 min before TMB substrate solution was added. Lastly, stop solution was added to terminate the enzyme/substrate reaction and absorbance was determined at 450 nm. 4.9. Animal Studies 4.9.1. Subcutaneous Xenograft Model MHCC97L cells (1 × 106) in PBS were subcutaneously injected onto the right flank of BALB/cAnN-nu athymic mice nude mice. After one week, serial doses of NYG and doxorubicin (n = 5) treatment were initiated via intraperitoneal administration three times every week for four weeks. Control group was administrated with the same volume of PBS. The weight and tumor size of each mice were closely monitored. After the mice were sacrificed, the tumors were removed and measured. The animal procedures were approved by the Committee on the Use of Live Animal in Teaching and Research (CULATR) in The University of Hong Kong. 4.9.2. Orthotopic Implantation Model The animal model establishment protocol has been described in our previous study [34]. In brief, 1 × 106 luciferase-tagged MHCC97L cells in PBS was subcutaneously injected onto the right flank of Balb/C nude mouse. The subcutaneous tumor was removed when it reached approximately 1 cm and it was cut into 1 mm3 in pieces before implanted into the left lobe of another mouse liver. After one week of laparotomy, the mice were subjected to luciferase imaging analysis for examination of any tumor growth. The mice presenting signal of luciferase (n = 5) were randomized into control saline group and NYG intervention group via intraperitoneal injection (10 mg/kg every 2 days). The weight and tumor growth of each mice were closely monitored every week. By the end of experiment, the mice were sacrificed and liver tumors were removed. The animal procedures with reference number of 3398-14 were approved on August, 2014 by the Committee on the Use of Live Animal in Teaching and Research (CULATR), The University of Hong Kong. 4.10. Statistical Analysis All data were statistically analyzed by unpaired Student’s t test. It was considered as significant when p-value < 0.05. 5. Conclusion In conclusion, we demonstrated the anti-tumor effect of NYG, the aqueous fraction of Saw Palmetto Extract, on human hepatocellular carcinoma. NYG exhibited potent inhibitory effect of HCC growth in both xenograft and orthotopic tumor models through suppression of neovascularization in tumor stroma. NYG did not exert toxicity to HCC cells, nor could it reduce VEGF expression in HCC cells. Instead, NYG suppressed the migration and sprout formation of endothelial cells in tumor stroma, which might be the major contribution to its anti-tumor effect. This is associated with the inactivation of ERK signaling on NYG-treated endothelial cells. Our study shed light on the novel use of the aqueous fraction of Saw Palmetto as potential adjuvant therapy of HCC. Acknowledgments This research was partially supported by the research council of the University of Hong Kong (Project codes: 104003422 and 104004092), Wong’s Donation for modern oncology of Chinese medicine (Project No. 200006267) and Hong Kong Government Matching Funding (Project No. 207060411). We also expressed our appreciation to Heimat Co., Ltd., Tokyo, Japan in providing the extract for the study. Author Contributions Hor-Yue Tan and Ning Wang conducted the experiment, collected the data and drafted the manuscript. Masao Takahashi, Yigang Feng, and Hongyun Li interpreted the data and commented on the manuscript. Yibin Feng designed the experiment, interpreted the data and finalized the manuscript. All authors read and approved the final manuscript. Conflicts of Interest Masao Takahashi is the research staff from Heimat Co., Ltd., Tokyo, Japan. Others have no conflict to disclose. Figure 1 NYG suppressed xenograft growth of HCC in vivo. (A) Mice were treated with 0.1, 1, 10 and 100 mg/kg of NYG (i.p.) on five consecutive days (n = 5); (B,C) Subcutaneous xenograft mice were treated with NYG, either 5 mg/kg or 10 mg/kg, via i.p. injection. The body weight was monitored and measured as average of replicates (n = 5) with standard deviation; (D) The tumors were excised and measured as average ± SD (mm3). The tumor size of NYG (10 mg/kg) treatment group is smaller than the saline-given mice group; The subcutaneous tumor growth is indicated by circle; (E) The tumor size was monitored and measured every week after subcutaneous injection of tumor cells; (F) Histological sections of excised hepatocellular carcinoma (HCC) tumor staining with cluster of differentiation 31 (CD31) antibody in control and NYG-treated mice (10 mg/kg). The vascular density of each section was measured as the mean number of microvessel in five histological areas (100× magnification); NYG intervention significantly reduced CD31-stained microvessel density in xenograft mice. *** p < 0.001. Figure 2 NYG inhibited orthotopic growth of HCC in vivo. (A) The orthotopic growth of HCC was monitored by imaging of luciferin signal every week after model establishment. The orthotopic HCC growth rate is slower in NYG treatment group as compared to saline-given group of mice; (B) The liver were excised after five weeks of NYG treatment. The representative photograph showed reduced orthotopic tumor growth in NYG-treated mice group as compared to saline-treated group. After NYG treatment, the significant reduced tumor volume is observed; The orthotopic grown tumor nodules are indicated by circle; (C) Histological sections of excised orthotopic HCC tumor staining with CD31 antibody in control and NYG-treated mice (10 mg/kg). The vascular density of each section was measured as the mean number of microvessels in five histological areas (100× magnification). NYG intervention significantly reduced CD31-stained microvessel density in orthotopic HCC implanted mice. ** p < 0.01. Figure 3 NYG exerted minimal effect on in vitro cultured HCC cells. (A) HepG2 and (B) MHCC97L cell lines were treated with NYG ranging from 0 to 500 μg/mL at 24, 48 and 72 h of incubation. Beta-cyclodextrin (B-CD), suramin and sorafenib served as control groups. Cell viability were measured as average of replicates (n = 3) with standard deviation (%); (C) The normal cell line L-02 was also treated with NYG in a dose and time dependent manner. NYG exerted no cytotoxicity on HepG2 and MHCC97L or L-02 cells; (D) The relative mRNA expression of VEGF in MHCC97L cells determined by qPCR. The VEGF mRNA was measured as fold change ± SD; (E) Amount of VEGF secretary protein of MHCC97L cells in both normoxia and hypoxia conditions. The VEGF concentration was measured as average ± SD (ng/nL). MHCC97L showed no significant changes in mRNA and protein concentration of VEGF after NYG treatment. Figure 4 NYG reduced migration and tube formation of HUVEC cells. (A) HUVECs were treated with VEGF and serial doses of NYG and suramin for 24 and 48 h. Cell viability were measured as average of replicates (n = 3) with standard deviation (%). The IC50 of NYG on HUVEC is approximately 2 mg/mL and 4 mg/mL with incubation time of 24 and 48 h, respectively; (B) HUVECs were seeded on the upper chamber of transwell, either treated with VEGF (20 ng/mL) alone or in combination with NYG or suramin. The mean number of migrating cells was counted in five randomized fields per well under microscope. NYG treatment reduced invasive potential of HUVECs in dose dependent manner (40× magnification); (C) HUVECs were seeded on matrigel pre-coated well, either treated with VEGF (20 ng/mL) alone or in combination with NYG or suramin. The mean number of sprouts formed was counted in five randomized fields per well under microscope. NYG inhibited sprout formation of HUVECs in dose-dependent manner (100× magnification). * p < 0.05, ** p < 0.01. Figure 5 Inhibition of tumor neovascularization by NYG may be related to ERK suppression in endothelial cells. (A) HUVECs were treated with VEGF alone or in combination with NYG. The lysates were immune-blotted against anti-pERK and GAPDH. The protein levels of pErk1 and pErk2 were normalized with GAPDH and quantified, as shown in the right panel. NYG intervention downregulated the expression of p-ERK; (B) Histological sections of excised HCC tumor staining with CD31 and pERK antibodies in control and NYG-treated mice (10 mg/kg) (400× magnification). NYG intervention significantly reduced CD31-stained microvessel density in xenograft mice. The expression of p-ERK was reduced on CD31 stained vasculature in NYG intervention group compared to control group of mice. * p < 0.05, *** p < 0.001. Figure 6 Schematic diagram of in vivo and in vitro experimental procedures demonstrate the anti-tumor effect of NYG through mediating HUVECs migration and sprout formation. ==== Refs References 1. Liver Cancer Incidence Statistics Available online: http://www.cancerresearchuk.org/cancer-info/cancerstats/types/liver/incidence/uk-liver-cancer-incidence-statistics#world (accessed on 12 June 2014) 2. IARC, W.H.O. Globocan 2012: Estimated Incidence, Mortality and Prevalence Worldwide in 2012 Available online: http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx (accessed on 10 June 2012) 3. American Cancer Society How is Liver Cancer Treated? Available online: http://www.cancer.org/cancer/livercancer/detailedguide/liver-cancer-treating-general-info (accessed on 10 June 2014) 4. Johnson P.J. Non-surgical treatment of hepatocellular carcinoma HPB 2005 7 50 55 10.1080/13651820410024076 18333161 5. Pang R. Poon R.T. Angiogenesis and antiangiogenic therapy in hepatocellular carcinoma Cancer Lett. 2006 242 151 167 10.1016/j.canlet.2006.01.008 16564617 6. National Cancer Institute Fda Approval for Sorafenib Tosylate Available online: http://www.cancer.gov/cancertopics/druginfo/fda-sorafenib-tosylate#Anchor-Live-50484 (accessed on 10 June 2014) 7. Childs M. Do Drugs Really Have to Be So Expensive? Available online: http://www.bbc.com/news/health-21834442 (accessed on 13 December 2015) 8. Bent S. Kane C. Shinohara K. Neuhaus J. Hudes E.S. Goldberg H. Avins A.L. Saw palmetto for benign prostatic hyperplasia N. Engl. J. Med. 2006 354 557 566 10.1056/NEJMoa053085 16467543 9. Sirab N. Robert G. Fasolo V. Descazeaud A. Vacherot F. Taille Ade L. Terry S. Lipidosterolic extract of serenoa repens modulates the expression of inflammation related-genes in benign prostatic hyperplasia epithelial and stromal cells Int. J. Mol. Sci. 2013 14 14301 14320 10.3390/ijms140714301 23846725 10. Villaverde N. Galvis A. Marcano A. Priestap H.A. Bennett B.C. Barbieri M.A. Saw palmetto ethanol extract inhibits adipocyte differentiation J. Nat. Med. 2013 67 619 625 10.1007/s11418-012-0723-2 23179316 11. Hostanska K. Suter A. Melzer J. Saller R. Evaluation of cell death caused by an ethanolic extract of serenoae repentis fructus (prostasan) on human carcinoma cell lines Anticancer Res. 2007 27 873 881 17465214 12. Wadsworth T.L. Worstell T.R. Greenberg N.M. Roselli C.E. Effects of dietary saw palmetto on the prostate of transgenic adenocarcinoma of the mouse prostate model (tramp) Prostate 2007 67 661 673 10.1002/pros.20552 17342743 13. Bonnar-Pizzorno R.M. Littman A.J. Kestin M. White E. Saw palmetto supplement use and prostate cancer risk Nutr. Cancer 2006 55 21 27 10.1207/s15327914nc5501_3 16965237 14. Andriole G.L. McCullum-Hill C. Sandhu G.S. Crawford E.D. Barry M.J. Cantor A. The effect of increasing doses of saw palmetto fruit extract on serum prostate specific antigen: Analysis of the camus randomized trial J. Urol. 2013 189 486 492 10.1016/j.juro.2012.09.037 23253958 15. Abe M. Ito Y. Suzuki A. Onoue S. Noguchi H. Yamada S. Isolation and pharmacological characterization of fatty acids from saw palmetto extract Anal. Sci.: Int. J. Jpn. Soc. Anal. Chem. 2009 25 553 557 10.2116/analsci.25.553 16. Raman P. Dewitt D.L. Nair M.G. Lipid peroxidation and cyclooxygenase enzyme inhibitory activities of acidic aqueous extracts of some dietary supplements Phytother. Res.: PTR 2008 22 204 212 10.1002/ptr.2287 17726737 17. Goldmann W.H. Sharma A.L. Currier S.J. Johnston P.D. Rana A. Sharma C.P. Saw palmetto berry extract inhibits cell growth and cox-2 expression in prostatic cancer cells Cell Biol. Int. 2001 25 1117 1124 10.1006/cbir.2001.0779 11913955 18. Tan H.Y. San-Marina S. Wang N. Hong M. Li S. Li L. Cheung F. Wen X.Y. Feng Y. Preclinical models for investigation of herbal medicines in liver diseases: Update and perspective Evid. Based Complement. Altern. Med. 2016 2016 4750163 10.1155/2016/4750163 26941826 19. Eilken H.M. Adams R.H. Dynamics of endothelial cell behavior in sprouting angiogenesis Curr. Opin. Cell Biol. 2010 22 617 625 10.1016/j.ceb.2010.08.010 20817428 20. Mavria G. Vercoulen Y. Yeo M. Paterson H. Karasarides M. Marais R. Bird D. Marshall C.J. Erk-mapk signaling opposes rho-kinase to promote endothelial cell survival and sprouting during angiogenesis Cancer Cell 2006 9 33 44 10.1016/j.ccr.2005.12.021 16413470 21. Prasadam I. Zhou Y. Du Z. Chen J. Crawford R. Xiao Y. Osteocyte-induced angiogenesis via vegf-mapk-dependent pathways in endothelial cells Mol. Cell. Biochem. 2014 386 15 25 10.1007/s11010-013-1840-2 24162672 22. Weidner N. Semple J.P. Welch W.R. Folkman J. Tumor angiogenesis and metastasis–correlation in invasive breast carcinoma N. Engl. J Med. 1991 324 1 8 10.1056/NEJM199101033240101 1701519 23. Goel H.L. Mercurio A.M. Vegf targets the tumour cell Nature Rev. Cancer 2013 13 871 882 10.1038/nrc3627 24263190 24. Hoeben A. Landuyt B. Highley M.S. Wildiers H. Van Oosterom A.T. De Bruijn E.A. Vascular endothelial growth factor and angiogenesis Pharmacol. Rev. 2004 56 549 580 10.1124/pr.56.4.3 15602010 25. Narasimhan P. Liu J. Song Y.S. Massengale J.L. Chan P.H. Vegf stimulates the erk 1/2 signaling pathway and apoptosis in cerebral endothelial cells after ischemic conditions Stroke 2009 40 1467 1473 10.1161/STROKEAHA.108.534644 19228841 26. Srinivasan R. Zabuawala T. Huang H. Zhang J. Gulati P. Fernandez S. Karlo J.C. Landreth G.E. Leone G. Ostrowski M.C. Erk1 and erk2 regulate endothelial cell proliferation and migration during mouse embryonic angiogenesis PLoS ONE 2009 4 1277 10.1371/journal.pone.0008283 20011539 27. Gidwani B. Vyas A. A comprehensive review on cyclodextrin-based carriers for delivery of chemotherapeutic cytotoxic anticancer drugs BioMed Res. Int. 2015 2015 198 268 10.1155/2015/198268 26582104 28. Tiwari G. Tiwari R. Rai A.K. Cyclodextrins in delivery systems: Applications J. Pharmacy Bioallied Sci. 2010 2 72 79 10.4103/0975-7406.67003 21814436 29. Irie T. Uekama K. Pharmaceutical applications of cyclodextrins. Iii. Toxicological issues and safety evaluation J. Pharm. Sci. 1997 86 147 162 10.1021/js960213f 9040088 30. Schonfelder U. Radestock A. Elsner P. Hipler U.C. Cyclodextrin-induced apoptosis in human keratinocytes is caspase-8 dependent and accompanied by mitochondrial cytochrome c release Exp. Dermatol. 2006 15 883 890 10.1111/j.1600-0625.2006.00481.x 17002685 31. Leroy-Lechat F. Wouessidjewe D. Andreux J.-P. Puisieux F. Duchêne D. Evaluation of the cytotoxicity of cyclodextrins and hydroxypropylated derivatives Int. J. Pharm. 1994 101 97 103 10.1016/0378-5173(94)90080-9 32. Kwok C. Management of Side Effects from Chemotherapy Available online: https://www.hkacs.org.hk/ufiles/Chemotherapy.pdf (accessed on 13 January 2016) 33. Mosmann T. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays J. Immunol. Method 1983 65 55 63 10.1016/0022-1759(83)90303-4 34. Tan H.Y. Wang N. Man K. Tsao S.W. Che C.M. Feng Y. Autophagy-induced relb/p52 activation mediates tumour-associated macrophage repolarisation and suppression of hepatocellular carcinoma by natural compound baicalin Cell Death Dis. 2015 6 e1942 10.1038/cddis.2015.271 26492375

PMC005xxxxxx/PMC5000675.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081278ijms-17-01278ReviewThe Ribonuclease A Superfamily in Humans: Canonical RNases as the Buttress of Innate Immunity Koczera Patrick 12Martin Lukas 1Marx Gernot 1Schuerholz Tobias 1*Boix Ester Academic Editor1 Department of Intensive Care and Intermediate Care, University Hospital Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen, Aachen 52074, Germany; [email protected] (P.K.); [email protected] (L.M.); [email protected] (G.M.)2 Department for Experimental Molecular Imaging, University Hospital RWTH Aachen and Helmholtz Institute for Biomedical Engineering, RWTH Aachen University, Aachen 52074, Germany* Correspondence: [email protected]; Tel.: +49-241-80-3587105 8 2016 8 2016 17 8 127811 5 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).In humans, the ribonuclease A (RNase A) superfamily contains eight different members that have RNase activities, and all of these members are encoded on chromosome 14. The proteins are secreted by a large variety of different tissues and cells; however, a comprehensive understanding of these proteins’ physiological roles is lacking. Different biological effects can be attributed to each protein, including antiviral, antibacterial and antifungal activities as well as cytotoxic effects against host cells and parasites. Different immunomodulatory effects have also been demonstrated. This review summarizes the available data on the human RNase A superfamily and illustrates the significant role of the eight canonical RNases in inflammation and the host defence system against infections. human RNasescanonical RNasessecreted RNaseshost defence proteinantimicrobial activity ==== Body 1. Introduction The protein ribonuclease A (RNase A) was initially extracted from the bovine pancreas and is one of the best-characterized mammalian proteins in the literature. Over time, several other proteins with significant sequence homology were identified in mammals and other vertebrates, allowing assembly of the vertebrate-specific RNase A superfamily [1,2]. The members of this superfamily can be distinguished from other exo- and endoribonucleases, which show different distributions and properties [3]. In humans, eight secreted RNases have been described and are generally referred to as the canonical RNases. The following paragraph will highlight selected common features of these RNases with regard to their sequences, conformations, phylogenesis, biochemical characterization and regulation. The proteins show a tertiary structure that is stabilized by eight disulphide bridges, with the exception of RNase 5, which has six cysteine residues. Two histidine residues and one lysine residue determine the catalytic activity of these RNases; the lysine residue lies within the common invariant sequence motif CKxxNTF. Each RNase initially contains an N-terminal signal sequence that directs protein biosynthesis within the endoplasmic reticulum, with its final form being secretory. Moreover, the N-terminal portion of the mature extracellular RNase appears to be required for antimicrobial activity [4]. This feature was demonstrated by generating N-terminus-derived peptides that showed similar antimicrobial activity. The ribonucleolytic activity, in contrast, appears to not be crucial for the activity against microbes [5]. The antibacterial activity of this protein family has been best characterized based on RNase 3 and is associated with disruption of the bacterial membrane. However, the mechanism of activity against viruses, fungi and parasites has not yet been resolved. Regarding antiviral activity, targeting of the virion is hypothesized. In addition to targeting the virion, the intracellular activity of these proteins in the cytosol might degrade viral RNA to inhibit viral replication or may induce host cell apoptosis [6,7]. As part of the Human Genome Project, corresponding genes were found to be located on chromosome 14, within cluster 14q11.2. Five additional proteins with relevant sequence similarity did not show ribonuclease activity or the characteristic N-terminal signal sequence [4,8]. The phylogenetic origin of the RNase A superfamily was extensively evaluated and further assessed with respect to the genome sequences of additional species. As a result, RNases 2 and 3 can be grouped together, as can RNases 7 and 8. Together with RNase 6, these RNases seem closely related. Meanwhile, RNases 1, 4 and 5 can be grouped, with a closer relationship between RNases 1 and 4 [9,10,11]. There is further information on the ancestral origin and role of this superfamily. While results regarding the function of RNases in zebrafish (Danio rerio) suggest an ancestral origin consisting of angiogenesis-related ribonucleases, a host defence-associated role was proposed in studies on birds and mammals [12,13,14,15]. This hypothesis was based on the proteins’ structure, biochemical properties and bactericidal functions. In comparison to other immune-associated proteins/genes, the RNase A superfamily also exhibits high rates of duplication and amino acid substitution. Additionally, these RNases have high isoelectric points and positive net charges (with theoretically calculated pIs ranging from 8.69–10.12) [8,16]. Both properties are associated with antibacterial activity and strong interactions with the respective substrates of the enzyme reaction, namely, negatively charged polynucleotides. For the canonical RNases, each varies in ribonuclease activity and nucleotide preference for substrate cleavage, without strict selectivity for cleavage and recognition sites [10]. Different patterns of regulation of RNases, e.g., on the level of transcription or secretion, have been recognized. Notably, the ribonuclease inhibitor (RI), which can be found in all mammalian cells, controls the activity of all RNases in different ways. The RI binds to ribonucleases with femtomolar affinity and inhibits or attenuates the biological effects of the RNases by generating an RNase:RI complex. The presence of cytosolic RI protects the host cells from the cytotoxic activity of RNases [17]. Although the biochemical properties of RNase 1 and the emergence of the RNase A gene superfamily have been evaluated extensively, the RNases’ physiological function needs further clarification. To date, different reports have suggested the relevance of the RNases to host defence, angiogenesis and digestion. The present review summarizes information regarding the significance of the canonical RNases for human host defence (Table 1) and aims to provide a synopsis of current data with regard to the antimicrobial activity of the RNases and their roles in host immune responses. 2. Ribonuclease (RNase) 1 RNase 1, also known as RNase A or pancreatic-type RNase, can be found in various organs, so its expression is not restricted to the exocrine pancreas [65]. This protein is known to undergo different post-translational modifications, and purified samples from the urine, seminal plasma, kidney and brain show different patterns of glycosylation [66,67,68,69]. The proposed functions of RNase 1 in immune defence are summarized in Figure 1. When taking into account the origin of these different modifications, the endothelial cells of the circulatory system appear to be one possible source, as they selectively produce and secrete RNase 1 in reasonable amounts, as demonstrated in vitro in human endothelial cells derived from veins, arteries and capillaries. Interestingly, human umbilical vein endothelial cells expressed and released the highest concentrations of RNase 1. Although endothelial cells constitutively secrete RNase 1, a fraction is stored in Weibel-Palade bodies, which are also known for storage and induced release of von Willebrand factor [70,71]. In ruminants, the RNase 1 secreted by the exocrine pancreas degrades dietary RNA to aid in nutrition, whereas in humans, digestion is not the main function [1,72]. The endothelial origin suggests association of this protein with vascular homeostasis [2]. RNase 1 has distinct ribonuclease activity, which allows degradation of single- and double-stranded polyRNA as well as DNA-RNA hybrids. Extracellular RNA is known as a dangerous molecule that can induce coagulation and endothelial permeability as well as modulation of the inflammatory response, partly through liberation of cytokines [18,73,74,75,76]. Therefore, RNase 1 is a potent polynucleotide scavenger that serves as an opposing force to vascular RNA in terms of coagulation, endothelial permeability and inflammation. Additionally, RNase 1 might be important for normalization of serum viscosity and clearance of perivascular pathogenic polynucleotides [70]. Although experimental data on humans are limited, the importance of extracellular RNA and RNase 1 has been illustrated very recently in different studies of animals [19]. Clinical proof of this concept was demonstrated by Cabrera-Fuentes et al. [77], who utilized upper-limb ischaemia for remote ischaemic preconditioning before cardiac surgery to protect the heart against ischaemia-reperfusion injury. By preconditioning, the blood levels of protective RNase 1 were increased, whereas vascular RNA and tumour necrosis factor α (TNFα) decreased. These findings indicate that enhancing the levels of RNase 1 by preconditioning before heart surgery may improve patient outcomes. The relevance of RNase 1 to host defence is supported by reports of its antiviral activity. RNase 1 extracts from the urine for human chorionic gonadotropin preparations as well as recombinant RNase 1 showed antiviral activity against human immunodeficiency virus (HIV)-1, suggesting possible protection of the foetus during pregnancy [20,21,22]. Additionally, RNase 1 was demonstrated to induce activation and maturation of dendritic cells as well as subsequent production of different cytokines (e.g., TNFα, interleukin (IL)-6 and IL-12) [23]. Interestingly, for endothelial cells, studies have shown somewhat contradictory results. Following incubation with IL-1β or TNFα, endothelial cells were reported to have decreased secretion and cellular expression of RNase 1 due to an epigenetic mechanism [26]. These results indicate a disturbance of the vascular RNA/RNase system as a result of exposure to inflammatory stimuli. 3. RNase 2 RNase 2, commonly known as eosinophil-derived neurotoxin (EDN), can be found in the secondary granules of eosinophil granulocytes and is one of the four major secretory proteins released upon activation of eosinophils [27,28,29]. The discovery of this RNase is connected to the Gordon phenomenon, a non-physiological syndrome in rabbits that is characterized by cerebellar dysfunction upon intrathecal injection of EDN, as also shown in guinea pigs [78,79]. EDN shows antiviral activity against HIV and respiratory syncytial virus (RSV) in vitro [20,21,22,24,25]. Through cell culture experiments, it was shown that EDN reduces the infectivity of HIV and RSV. The ribonuclease activity of EDN appears to be crucial for the antiviral activity, as the RI eliminated the antiviral effect of activated eosinophils. Similarly, sequence-altered recombinant EDN lost its antiviral activity upon loss of ribonuclease activity [24]. However, the activation of EDN-producing eosinophil granulocytes is also associated with allergic inflammation, such as that related to asthma, in the respiratory tract. These two facets of eosinophil activation in the respiratory system could be considered as a side effect of host defence. The antiviral response against RSV by eosinophil granulocytes could be associated with immunization against allergen [80,81]. Eosinophils are not the only cells capable of EDN secretion; human monocyte-derived macrophages also produce this RNase upon stimulation with lipopolysaccharide (LPS) and TNFα [23]. Although EDN does not exhibit potent anthelmintic or antibacterial activity, eosinophils express different pattern recognition receptors, such as toll-like receptors (TLRs) and nucleotide-binding oligomerization domain (NOD)-like receptors, which determine interaction with bacteria and helminths [38,82,83]. Studies have reported the release of EDN upon incubation of eosinophils with pathogenic bacteria, such as Clostridium difficile or Staphylococcus aureus, but not with Bifidobacteria, Hemophilus or Prevotella species [30,31]. However, the specific mechanism of distinction between different bacteria needs further investigation. Yang et al. [32] described EDN as an alarmin because the ribonuclease was shown to facilitate antigen recognition by binding to TLR2 and stimulating a type 2 helper T (Th2)-polarized response. Further studies illustrated the effect of EDN on dendritic cells, with EDN acting as a chemoattractant for immature human dendritic cells. Furthermore, EDN was found to induce maturation and activation of cultured dendritic cells [23,33]. In summary, there is a strong need for further investigations into the role of EDN in host defence. Current thinking suggests that EDN causes elimination of cells by triggering an inflammatory response that activates killer cells or cell death pathways. Alternatively, EDN may be internalized by infected cells to degrade viral RNA in the cytoplasm or may be secreted by eosinophils or monocytes/macrophages to further modulate the immune response. 4. RNase 3 RNase 3, or eosinophil cationic protein (ECP), is another ribonuclease found in the secondary granules of eosinophils and is released upon cell activation/stimulation. Similar to EDN, ECP displays antiviral and neurotoxic activities, and anthelmintic, bactericidal and cytotoxic effects have also been described [8]. For example, incubation of ECP with target cell cultures reduced the infectivity of RSV group B. As with EDN, the antiviral activity of ECP is dependent upon its enzymatic, ribonucleolytic function. However, ECP’s bactericidal activity remains unaltered in enzymatically inactive RNase. With regard to the complex of the RI and ECP, the issue of pathogen toxicity is less clear. Data illustrating this topic are only available for parasites, as incubation of ECP with the RI suppresses the antiparasitic nature of the RNase [84,85]. Similar antiviral activity was described for EDN, and the antiviral activities of both ECP and EDN are dependent on their ribonuclease activity [34]. However, enzymatic function is not necessary for the antibacterial activity of ECP. Gram-positive and Gram-negative bacteria as well as certain mycobacterial strains (Staphylococcus aureus, Escherichia coli and Mycobacterium vaccae) have been shown to be susceptible to the antibacterial activity of ECP in vitro [35,36]. The mechanism of the antibacterial activity was demonstrated in recent studies, and a stepwise process was suggested. For Gram-negative bacteria, ECP appears to display an amyloid-like aggregation. ECP binds to the bacterial surface, which causes conformational alterations in the protein. This rearrangement of the protein enables bacterial agglutination by binding of the protein to other rearranged ECP molecules attached to bacterial surfaces. Finally, these ECP aggregates disrupt the LPS bilayer, which may result in membrane disruption and bacterial [36,86]. In this context, it is worth mentioning that the Alzheimer’s disease-associated amyloid β (Aβ) peptide has been discussed as a host defence strategy against fungal challenge, as amyloids possess antimicrobial properties [8,44,86,87,88,89,90,91]. These observations, although not completely understood, stress the role of ECP in bacterial clearance. As a result, further inflammatory pathways are initialized directly via the protein’s immunomodulatory effects, such as mast cell degranulation, and indirectly by the aforementioned antibacterial activity [37]. Eosinophils are associated with host defence strategies against helminthic parasites, and for granular proteins, ECP is the most potent anthelmintic member. ECP’s high toxic activity has been demonstrated in vitro against Schistosoma mansoni [38]. In vivo, epidemiological studies have illustrated the importance of ECP. The ECP gene shows sequence polymorphism, with the respective proteins differing in toxic activity against Schistosoma mansoni. Ugandan populations, which live in regions where the parasites are endemic, were tested for their distribution of ECP alleles. In this study, homozygous carriers of the more cytotoxic allele of ECP (434GG) showed a lower prevalence of Schistosoma mansoni infection. This finding indicates that the more cytotoxic variant of ECP improves elimination of Schistosoma mansoni infection. Additionally, another sequence polymorphism-related allele, 371G, was associated with a higher susceptibility to cerebral malaria, the most severe manifestation of infection with the parasite Plasmodium falciparum [39,40]. In addition, ECP showed toxic activity in other diseases caused by parasites, and specifically Brugia pahangi and Trichinella spiralis [41,42]. However, the specific mechanism needs further investigation. Cytotoxic effects of ECP have been demonstrated against different host mammalian cell lines, including epithelial cells [43,44]. The process of the antimicrobial activity was reviewed previously [86,87], but further research needs to be performed to determine the significance of (host) cytotoxicity as an adverse effect of bactericidal, antiviral and anthelmintic activities or as a driver for tissue remodelling. In this regard, ECP has been shown to induce apoptosis in bronchial epithelial cells; however, the downstream mechanism of apoptosis needs further investigation. Intracellular as well as extracellular pathways were suggested for the apoptotic effect of ECP (heparan sulphate proteoglycan-mediated internalization for the former and cell surface aggregation for the latter), in addition to TNFα production and activation of the caspase-8 and caspase-3 pathways [92,93,94]. A summary of the activities of ECP is illustrated in Figure 2. 5. RNase 4 The significance of RNase 4 for physiological function in general and for host defence in particular is widely unknown because very few studies have been performed to investigate its function. RNase 4 mRNA was detected in several human somatic tissues, such as the skeletal muscle, pancreas, lung, kidney, placenta, liver and blood (i.e., in monocytes) [45,46,47]. The last two sources suggest a role in host defence, as these tissues play a significant role in this regard. No direct evidence for antimicrobial activity of RNase 4 in humans has been described to date, but two observations are worth noting. First, RNase 4 extracted from bovine milk reduced the viability of Candida albicans in vitro, and the antimicrobial effects of lactoferrin and lactoferricin against E. coli were enhanced by co-incubation with a mixture of RNases 4 and 5. Second, in vitro studies showed that the mRNA of RNase 4 was found in proliferating and differentiated human keratinocytes [48,95,96]. In these studies, RNase 4 expression was associated with the expression of RNase 5, another member of the RNase family with proven antimicrobial activity. 6. RNase 5 RNase 5, or angiogenin, was named after its angiogenic potency, which induces extensive blood vessel growth [97]. This angiogenic property requires ribonuclease activity, but in comparison to the enzymatic activity of RNase 1, RNase 5 has much lower activity levels (1 × 10−5–1 × 10−6) [8,98,99,100]. Both its enzymatic activity and its angiogenic effect are modulated by the RI. RNase 5 can be detected in different tissues and organs and is associated with a number of (patho) physiological processes, including neoplasia, reproduction and regeneration of damaged tissue [101,102,103]. The aforementioned processes include activation of both the immune system and angiogenesis, each happening at different stages, which hampers distinction between those processes. For RNase 5, the patterns of expression vary in comparison to those of other inducers, such as vascular endothelial growth factor. These observations hint at a broader physiological role for RNase 5 in addition to a significant role in inflammatory processes and host defence. The broad biological relevance of this protein was recently reviewed, with a focus on its function in host defence [100]. It was reported that serum levels of RNase 5 increase during acute-phase responses [49,50]. Similar to RNase 4, RNase 5 is secreted by proliferating and activated keratinocytes; in fact, in the cervical-vaginal lavage of women with a sexually transmitted disease, RNase 5 levels were elevated [48,51]. The antimicrobial and antiviral activities of RNase 5 were also demonstrated in vitro. RNase 5 inhibited the reproduction of HIV-1 and reduced the number of colony counts of Streptococcus pneumoniae and Candida albicans [22,52]. Although the latter observation has been questioned, research in other species supports the antimicrobial property of RNase 5 [53]. In bovine milk, RNase 5 showed antifungal activity against Candida albicans, and particularly the hyphal form. The antimicrobial effects of lactoferrin and lactoferricin against E. coli were enhanced by co-incubation with a mixture of RNases 4 and 5 [95,96]. Furthermore, the antifungal activity against Candida albicans depends on the ribonuclease activity because blocking the enzyme significantly reduced the capacity to kill Candida [48]. In addition, murine analogues of RNase 5 (Ang1 and Ang4) showed antimicrobial activity; Ang4 was secreted into the murine intestine by Paneth cells and goblet cells of the colon upon microbiological challenge with LPS or Salmonella species, for example [52,104,105]. The direct effects of RNase 5 on host defence cells have also been reported. Upon microbiological challenge, mast cells synthesize RNase 5 and store it in their granules. In particular, LPS, E. coli and peptidoglycans can induce the secretion of this protein. Incubation with RNase 5 stimulated leukocytes to synthesize pro-inflammatory cytokines such as IL-6 and TNFα. In contrast, degranulation of neutrophil granulocytes was inhibited by incubation with RNase 5, which in turn can inhibit hyper-inflammatory states during the immune response [54,55,56,57]. Together, these reports suggest that RNase 5 plays a role involving not only direct antimicrobial activity but also modulation of the immune response as part of the host defence system. 7. RNase 6 Little evidence for the significance of RNase 6 has been found to date. mRNA transcripts of this protein were found in different tissues, including those of the lung, heart, brain, placenta, liver, skeletal muscle, kidney and pancreas. Detection in neutrophil granulocytes and monocytes suggested a role in inflammation [106]. Very recently, disease-associated functions for RNase 6 were described. It was demonstrated that RNase 6 protein levels in the urinary tract were elevated upon infection in vivo. Monocytes/macrophages secreted RNase 6 upon E. coli challenge. The protein showed in vitro antimicrobial activity against different uropathogenic bacterial strains, including E. coli, Staphylococcus saprophyticus and Enterococcus faecalis [58]. The mechanism of this antimicrobial activity was evaluated recently. The protein is able to destabilize the membrane of Gram-negative bacteria and to agglutinate them [59]. As demonstrated for RNase 1, 2 and 5, also RNase 6 has impact on the HI virus. Incubation of RNase 6 with target cells in vitro inhibits HIV infection [107]. These findings highlight the role of RNase 6 for host defence. 8. RNase 7 RNase 7 was first detected and purified from human skin; in fact, it is the most abundant RNase found in the skin and is constitutively expressed. It has been proposed that this protein, together with additional components, such as the secreted antimicrobial proteins β-defensin and psoriasin, constitutes the host defence system of the cutaneous epithelia. In psoriatic lesions, the levels of these proteins are elevated. This finding suggests a possible cause for the low incidence of infections related to psoriatic lesions [48,108,109,110]. RNase 7 is also expressed by various epithelial tissues, such as the genitourinary tissues; respiratory tissues and, to a lesser extent, the cells of the gastrointestinal tract, and is associated with the host defence properties of these tissues in response to environmental and microbial challenges [60,61]. In skin, keratinocytes are the major source of RNase 7; the protein was shown to be secreted in combination with the RI. The proteolytic activity in the stratum corneum reveals the broad-spectrum antimicrobial activity of RNase 7 upon degradation of the RI [48]. For the urogenital tract, a distinct form of control was recently demonstrated, as shown by decreased RI expression upon infectious challenge causing pyelonephritis [48,60]. Although RNase 7 shows high constitutive expression in keratinocytes, the expression of RNase 7 in cultured primary keratinocytes can be further enhanced by incubation with TNFα, IL-1β, or interferon γ (IFNγ) or by microbial challenge with Pseudomonas aeruginosa, E. coli, Staphylococcus aureus and epidermidis, Streptococcus pyogenes or Trichophyton rubrum. It has also been demonstrated in vitro that cigarette smoke enhanced RNase 7 expression in the respiratory epithelium and that protozoan challenge with Acanthamoeba castellanii increased mRNA expression of RNase 7 in corneal epithelial cells [108,111,112]. In these studies, recognition and signal transduction were associated with the TLR2, epidermal growth factor receptor (EGFR), nuclear factor κ-light-chain enhancer of activated B cells (NFκB), signal transducer and activator of transcription (STAT) 3 and mitogen-activated protein kinase (MAPK) pathways [62,63,111,113]. In human umbilical vein endothelial cells, RNase 7 was induced upon incubation with inflammatory cytokines, including TNFα, or co-incubation with IL-1β and IFNγ, suggesting a role in host defence of the tissue-blood barrier [114]. Broad antimicrobial activity was demonstrated for RNase 7 against different bacterial strains and fungi, but antiviral activity has not yet been described. It was reported that the protein is active against both Gram-positive and Gram-negative bacteria of clinical interest (E. coli, Enterococcus faecium, Pseudomonas aeruginosa and Staphylococcus aureus) as well as against mycobacteria (Mycobacterium vaccae), which are emerging in the clinic. RNase 7 shows activity at micromolar concentrations, even against multidrug-resistant isolates of Enterococcus faecium (vancomycin-resistant Enterococci, or VRE). Additionally, fungi, such as Candida albicans or the dermatophyte Epidermophyton floccosum, are susceptible to RNase 7 [36,108,115,116]. The mechanism of the antimicrobial activity is not yet fully understood, but several studies have sought to address this question. The ribonuclease activity does not appear to be relevant to the antimicrobial activity; rather, it is proposed that RNase 7 has the ability to disrupt bacterial membranes. In contrast to the action of RNase 3, there is no agglutination of the bacteria. It has been reported that the cationic protein interacts electrostatically with synthetic lipid vesicles, which causes leakage of the spheres. It was also reported that the positively charged protein is capable of forming a complex with LPS as well as outer membrane protein I (OprI) from Pseudomonas aeruginosa. For this bacterium, the interaction causes permeation of the membrane and triggers cell death [64,117]. 9. RNase 8 Although there is a strong sequence homology between RNase 7 and RNase 8, their distinct patterns of expression imply different roles. RNase 8 is expressed in the placenta and later on in the spleen, lung and testis [118]. However, its physiological function is still unknown. The antimicrobial activity of this protein has been demonstrated by its killing of different clinically relevant Gram-positive and Gram-negative bacteria, including different multidrug-resistant strains (methicillin-resistant Staphylococcus aureus and VRE), as well as Candida albicans [119]. The antimicrobial properties and the expression of RNase 8 in the placenta suggest its significance during host defence in pregnant women. This implies an additional defence system between the mother and her sterile foetus because different pathogens are known to pass through the placenta to the foetus [120]. 10. Conclusions The scientific investigation of pancreatic RNase and the RNase A superfamily began with an evaluation of the biochemical properties of these proteins and was continued to obtain a better understanding of the biological significance of each family member in terms of homeostasis and pathophysiology. The canonical RNases are able to digest polynucleotides in addition to their antiviral, antibacterial, anthelmintic, antifungal and cytotoxic activities. They are secreted by different tissues and by cells of the immune system, so they also possess immunomodulatory properties (Figure 3) [8,121]. These findings demonstrate the significance of canonical RNases for human host defence in general and as a backbone of the innate immune system in particular. However, there is a strong need for further evaluation. In humans, RNases appear to strongly protect the body’s interfaces with the environment, which include the cutaneous, urogenital and respiratory epithelia. Information regarding the significant role of RNases in protection against infection and severe inflammatory diseases, such as sepsis, is very limited, but as recently shown by our group, these proteins appear to be important players in acute infections [122]. Another benefit to gaining a better understanding of secreted RNases is the pharmacological exploitation of this family of proteins. With the deluge of multidrug-resistant bacteria and a drought in the antibiotic pipeline, conventional antibiotics have lost some of their usefulness in the battle against infections. Together with antimicrobial peptides, for which resistance is less threatening, RNases could offer a new solution with concurrent use of conventional antibiotics [123]. Further research is needed to evaluate the usefulness of treating infections directly with RNases and of exploiting their antimicrobial activity for the creation of synthetic analogues for tailored applications. Acknowledgments The authors have no support or funding to report. We gratefully acknowledge the European Bioinformatics Institute for providing structural visualization of RNase 3. Author Contributions Patrick Koczera and Tobias Schuerholz wrote and designed the review; Patrick Koczera and Lukas Martin performed the literature search; and Patrick Koczera, Lukas Martin, Gernot Marx and Tobias Schuerholz critically revised the article and approved it for publication. Conflicts of Interest Tobias Schuerholz is the chief medical officer for Brandenburg Antiinfektiva, Borstel, Germany, a company developing antimicrobial peptides. All other authors declare no conflict of interest. Abbreviations RNase A Ribonuclease A RI Ribonuclease inhibitor RNA Ribonucleic acid DNA Deoxyribonucleic acid HIV Human immunodeficiency virus RSV Respiratory syncytial virus TNFα Tumour necrosis factor α IL Interleukin EDN Eosinophil-derived neurotoxin LPS Lipopolysaccharide TLR Toll-like receptor NOD Nucleotide-binding oligomerization domain Th2 Type 2 helper T ECP Eosinophil cationic protein E. coli Escherichia coli EGFR Epidermal growth factor receptor NFκB Nuclear factor κ-light-chain enhancer of activated B cells STAT Signal transducer and activator of transcription MAPK Mitogen-activated protein kinase IFNγ Interferon γ VRE Vancomycin-resistant enterococci OprI Outer membrane protein I Figure 1 Ribonuclease (RNase) 1 is required for vascular homeostasis. (A) Endothelial cells (ECs) are the main source of large amounts of RNase 1; (B) Due to the ribonuclease activity of single- and double-stranded RNA as well as DNA-RNA hybrids, RNase 1 serves as a potent RNA scavenger for the normalization of serum viscosity and the clearance of perivascular polynucleotides; RNase 1 (C) shows antiviral activity against HIV and (D) is able to stimulate and induce maturation of dendritic cells (DCs). TNFα: Tumour necrosis factor alpha; IL: Interleukin; HIV: Human immunodeficiency virus; RNA: Ribonucleic acid. Figure 2 Effects of the eosinophil cationic protein (ECP). (A) ECP is stored in the secondary granules of eosinophils and can be released upon eosinophil simulation; (B) ECP can induce degranulation of mast cells; (C) ECP has broad antimicrobial activity, including inhibition of viruses, Gram-positive and Gram-negative bacteria and fungal and helminthic pathogens; (D) dose-dependent cytotoxic effects have been described for ECP, including necrosis and apoptosis. Figure 3 A schematic overview of the human canonical ribonucleases (RNases) in the host defence system. (A,B) RNases show antiviral activity and cytotoxic properties in mammalian cells and degrade RNA; (C) Cells of the immune system (immune cells, or ICs) secrete and are modulated by RNases; (D,F) Different mature cells secrete RNases, including endothelial (EnC) and epithelial (EpC) cells, for pericellular homeostasis; (E) Antimicrobial activity against bacteria, fungi and parasites has been demonstrated for RNases. RNA: Ribonucleic acid; HIV: Human immunodeficiency virus. ijms-17-01278-t001_Table 1Table 1 Proposed functions of the canonical ribonucleases (RNases) in human host defence. Ribonuclease Proposed Impact on Host Defence Reference(s) RNase 1 Degradation of vascular polyRNA [18,19] Anti-HIV-1 activity [20,21,22] Induces maturation and activation of dendritic cells [23] RNase 2/EDN Antiviral activity against HIV-1 and RSV-B [20,21,22,24,25] Secretion by eosinophil granulocytes and monocyte-derived macrophages [23,26,27,28,29,30,31] TLR2 binding and Th2 polarization [32] Chemokine and cytokine induction for activation and maturation of dendritic cells [23,33] RNase 3/ECP Antiviral activity against RSV-B [34] Antibacterial activity against mycobacteria and Gram+ and Gram− bacteria [35,36] Induces degranulation of mast cells [37] Anthelmintic activity against Schistosoma mansoni, Brugia pahangi and Trichinella spiralis [38,39,40,41,42,43] Cytotoxic activity against mammalian cells [43,44] RNase 4 Expression in host defence-associated tissues [45,46,47,48] Coexpression with lactoferrin, lactoferricin and RNase 5 Enhances antimicrobial activity of lactoferrin and lactoferricin RNase 5/Angiogenin Increased serum levels during acute-phase response [49,50,51] Antiviral activity against HIV-1 [22] Activity against Candida [48,52,53] Activity against Streptococcus (controversial data) Synthesis and secretion by mast cells [54] Proinflammatory stimulation of leukocytes [55] Inhibition of degranulation of neutrophil granulocytes [56,57] RNase 6 Infection-induced secretion in urinary tract [58,59] Antibacterial activity against Gram+ and Gram− bacteria RNase 7 Synthesis upon microbial, inflammatory and physicochemical challenge in epithelial tissues [59,60,61,62] Antibacterial activity against mycobacteria and Gram+ and Gram− bacteria [36,59,63] RNase 8 Antibacterial and antifungal activity against Gram+ and Gram− bacteria and Candida [64] RNase: Ribonuclease; RNA: Ribonucleic acid; HIV-1: Human immunodeficiency virus 1; RSV-B: Respiratory syncytial virus B; EDN: Eosinophil-derived neurotoxin; TLR2: Toll-like receptor 2; Th2: Type 2 helper T-cell; ECP: Eosinophil cationic protein. ==== Refs References 1. Beintema J.J. Kleinedam R.G. The ribonuclease A superfamily: General discussion Cell. Mol. Life Sci. 1998 54 825 832 10.1007/s000180050211 9760991 2. Sorrentino S. The eight human “canonical” ribonucleases: Molecular diversity, catalytic properties, and special biological actions of the enzyme proteins FEBS Lett. 2010 584 2194 2200 10.1016/j.febslet.2010.04.018 20388512 3. Luhtala N. Parker R. T2 Family ribonucleases: Ancient enzymes with diverse roles Trends Biochem. Sci. 2010 35 253 259 10.1016/j.tibs.2010.02.002 20189811 4. Lander E.S. Linton L.M. Birren B. Nusbaum C. Zody M.C. Baldwin J. Devon K. Dewar K. Doyle M. FitzHugh W. Initial sequencing and analysis of the human genome Nature 2001 409 860 921 10.1038/35057062 11237011 5. Torrent M. Pulido D. Valle J. Nogués M.V. Andreu D. Boix E. Ribonucleases as a host-defence family: Evidence of evolutionarily conserved antimicrobial activity at the N-terminus Biochem. J. 2013 456 99 108 10.1042/BJ20130123 23962023 6. Malik A. Batra J.K. Antimicrobial activity of human eosinophil granule proteins: Involvement in host defence against pathogens Crit. Rev. Microbiol. 2012 38 168 181 10.3109/1040841X.2011.645519 22239733 7. Rosenberg H.F. Eosinophil-derived neurotoxin (EDN/RNase 2) and the mouse eosinophil-associated RNases (mEars): Expanding roles in promoting host defense Int. J. Mol. Sci. 2015 16 15442 15455 10.3390/ijms160715442 26184157 8. Boix E. Nogués M.V. Mammalian antimicrobial proteins and peptides: Overview on the RNase A superfamily members involved in innate host defence Mol. Biosyst. 2007 3 317 335 10.1039/b617527a 17460791 9. Dubois J.Y. Ursing B.M. Kolkman J.A. Beintema J.J. Molecular evolution of mammalian ribonucleases 1 Mol. Phylogenet. Evol. 2003 27 453 463 10.1016/S1055-7903(03)00014-9 12742750 10. Rosenberg H.F. RNase A ribonucleases and host defense: An evolving story J. Leukoc. Biol. 2008 83 1079 1087 10.1189/jlb.1107725 18211964 11. Premzl M. Comparative genomic analysis of eutherian ribonuclease A genes Mol. Genet. Genom. 2014 289 161 167 10.1007/s00438-013-0801-5 24337645 12. Dyer K.D. Rosenberg H.F. The RNase a superfamily: Generation of diversity and innate host defense Mol. Divers. 2006 10 585 597 10.1007/s11030-006-9028-2 16969722 13. Cho S. Beintema J.J. Zhang J. The ribonuclease A superfamily of mammals and birds: Identifying new members and tracing evolutionary histories Genomics 2005 85 208 220 10.1016/j.ygeno.2004.10.008 15676279 14. Pizzo E. Buonanno P. di Maro A. Ponticelli S. de Falco S. Quarto N. Cubellis M.V. D’Alessio G. Ribonucleases and angiogenins from fish J. Biol. Chem. 2006 281 27454 27460 10.1074/jbc.M605505200 16861230 15. Cho S. Zhang J. Zebrafish ribonucleases are bactericidal: Implications for the origin of the vertebrate RNase A superfamily Mol. Biol. Evol. 2007 24 1259 1268 10.1093/molbev/msm047 17347156 16. Artimo P. Jonnalagedda M. Arnold K. Baratin D. Csardi G. de Castro E. Duvaud S. Flegel V. Fortier A. Gasteiger E. ExPASy: SIB bioinformatics resource portal Nucleic Acids Res. 2012 40 W597 W603 10.1093/nar/gks400 22661580 17. Rutkoski T.J. Raines R.T. Evasion of ribonuclease inhibitor as a determinant of ribonuclease cytotoxicity Curr. Pharm. Biotechnol. 2008 9 185 189 10.2174/138920108784567344 18673284 18. Sorrentino S. Naddeo M. Russo A. D’Alessio G. Degradation of double-stranded RNA by human pancreatic ribonuclease: Crucial role of noncatalytic basic amino acid residues Biochemistry 2003 42 10182 10190 10.1021/bi030040q 12939146 19. Zernecke A. Preissner K.T. Extracellular ribonucleic acids (RNA) enter the stage in cardiovascular disease Circ. Res. 2016 118 469 479 10.1161/CIRCRESAHA.115.307961 26846641 20. Lee-Huang S. Huang P.L. Sun Y. Huang P.L. Kung H.F. Blithe D.L. Chen H.C. Lysozyme and RNases as anti-HIV components in β-core preparations of human chorionic gonadotropin Proc. Natl. Acad. Sci. USA 1999 96 2678 2681 10.1073/pnas.96.6.2678 10077570 21. Rugeles M.T. Trubey C.M. Bedoya V.I. Pinto L.A. Oppenheim J.J. Rybak S.M. Shearer G.M. Ribonuclease is partly responsible for the HIV-1 inhibitory effect activated by HLA alloantigen recognition Aids 2003 17 481 486 10.1097/00002030-200303070-00002 12598767 22. Bedoya V.I. Boasso A. Hardy A.W. Rybak S. Shearer G.M. Rugeles M.T. Ribonucleases in HIV type 1 inhibition: Effect of recombinant RNases on infection of primary T cells and immune activation-induced RNase gene and protein expression AIDS Res. Hum. Retrovir. 2006 22 897 907 10.1089/aid.2006.22.897 16989616 23. Yang D. Chen Q. Rosenberg H.F. Rybak S.M. Newton D.L. Wang Z.Y. Fu Q. Tchernev V.T. Wang M. Schweitzer B. Human ribonuclease A superfamily members, eosinophil-derived neurotoxin and pancreatic ribonuclease, induce dendritic cell maturation and activation J. Immunol. 2004 173 6134 6142 10.4049/jimmunol.173.10.6134 15528350 24. Domachowske J.B. Dyer K.D. Bonville C.A. Rosenberg H.F. Recombinant human eosinophil-derived neurotoxin/RNase 2 functions as an effective antiviral agent against respiratory syncytial virus J. Infect. Dis. 1998 177 1458 1464 10.1086/515322 9607820 25. Rosenberg H.F. Dyer K.D. Domachowske J.B. Respiratory viruses and eosinophils: Exploring the connections Antivir. Res. 2009 83 1 9 10.1016/j.antiviral.2009.04.005 19375458 26. Gansler J. Preissner K.T. Fischer S. Influence of proinflammatory stimuli on the expression of vascular ribonuclease 1 in endothelial cells FASEB J. 2014 28 752 760 10.1096/fj.13-238600 24174426 27. Rosenberg H.F. Dyer K.D. Foster P.S. Eosinophils: Changing perspectives in health and disease Nat. Rev. Immunol. 2013 13 9 22 10.1038/nri3341 23154224 28. Lacy P. Moqbel R. Signaling and degranulation Eosinophils in Health and Disease Lee J.J. Rosenberg H.F. Elsevier Press Waltham, MA, USA 2013 206 219 29. Abu-Ghazaleh R.I. Dunnette S.L. Loegering D.A. Checkel J.L. Kita H. Thomas L.L. Gleich G.J. Eosinophil granule proteins in peripheral blood granulocytes J. Leukoc. Biol. 1992 52 611 618 1464733 30. Hosoki K. Nakamura A. Nagao M. Hiraguchi Y. Tokuda R. Wada H. Nobori T. Fujisawa T. Differential activation of eosinophils by “probiotic” Bifidiobacterium bifidum and “pathogenic” Clostridium difficle Int. Arch. Allergy Immunol. 2010 152 83 89 10.1159/000312131 20523069 31. Hosoki K. Nakamura A. Kainuma K. Sugimoto M. Nagao M. Hiraguchi Y. Tanida H. Tokuda R. Wada H. Nobori T. Differential activation of eosinophils by bacteria associated with asthma Int. Arch. Allergy Immunol. 2013 161 16 22 10.1159/000350338 23711849 32. Yang D. Chen Q. Su S. Zhang P. Kurosaka K. Caspi R. Eosinophil-derived neurotoxin acts as an alarmin to activate the TLR2–MyD88 signal pathway in dendritic cells and enhances Th2 immune responses J. Exp. Med. 2008 205 79 90 10.1084/jem.20062027 18195069 33. Yang D. Rosenberg H.F. Chen Q. Dyer K.D. Kurosaka K. Oppenheim J.J. Eosinophil-derived neurotoxin (EDN), an antimicrobial protein with chemotactic activities for dendritic cells Blood 2003 102 3396 3403 10.1182/blood-2003-01-0151 12855582 34. Domachowske J.B. Dyer K.D. Adams A.G. Leto T.L. Rosenberg H.F. Eosinophil cationic protein/RNase 3 is another RNase A-family ribonuclease with direct antiviral activity Nucleic Acids Res. 1998 26 3358 3363 10.1093/nar/26.14.3358 9649619 35. Pulido D. Moussaoui M. Nogués M.V. Torrent M. Boix E. Towards the rational design of antimicrobial proteins single point mutations can switch on bactericidal and agglutinating activities on the RNase A superfamily lineage FEBS J. 2013 280 5841 5852 10.1111/febs.12506 23992292 36. Pulido D. Torrent M. Andreu D. Nogués M.V. Boix E. Two human host defense ribonucleases against mycobacteria, the eosinophil cationic protein (RNase 3) and RNase 7 Antimicrob. Agents Chemother. 2013 57 3797 3805 10.1128/AAC.00428-13 23716047 37. Venge P. Byström J. Carlson M. Hâkansson L. Karawacjzyk M. Peterson C. Sevéus L. Trulson A. Eosinophil cationic protein (ECP): Molecular and biological properties and the use of ECP as a marker of eosinophil activation in disease Clin. Exp. Allergy 1999 29 1172 1186 10.1046/j.1365-2222.1999.00542.x 10469025 38. Ackerman S.J. Gleich G.J. Logering D.A. Richardson B.A. Butterworth A.E. Comparative toxicity of purified human eosinophil granule cationic proteins for schistosomula of Schistosoma mansoni Am. J. Trop. Med. Hyg. 1985 34 735 745 4025686 39. Eriksson J. Reimert C.M. Kabatereine N.B. Kazibwe F. Ireri E. Kadzo H. Eltahir H.B. Mohamed A.O. Vennervald B.J. Venge P. The 434(G>C) polymorphism within the coding sequence of eosinophil cationic protein (ECP) correlates with the natural course of Schistosoma mansoni infection Int. J. Parasitol. 2007 37 1359 1366 10.1016/j.ijpara.2007.04.001 17507019 40. Adu B. Dodoo D. Adukpo S. Gyan B.A. Hedley P.L. Goka B. Adjei G.O. Larsen S.O. Christiansen M. Theisen M. Polymorphisms in the RNASE3 gene are associated with susceptibility to cerebral malaria in ghanaian children PLoS ONE 2011 6 1278 10.1371/journal.pone.0029465 22216286 41. Hamann K. Gleich G. Checkel J. Loegering D. McCall J. Barker R. In vitro killing of microfilariae of Brugia pahangi and Brugia malayi by eosinophil granule proteins J. Immunol. 1990 144 3166 3173 2324497 42. Hamann K. Barker R. Loegering D. Gleich G. Comparative toxicity of purified human eosinophil granule proteins for newborn larvae of Trichinella spiralis J. Parasitol. 1987 73 523 529 10.2307/3282130 3598802 43. Maeda T. Kitazoe M. Tada H. de Llorens R. Salomon D.S. Ueda M. Growth inhibition of mammalian cells by eosinophil cationic protein Eur. J. Biochem. 2002 269 307 316 10.1046/j.0014-2956.2001.02653.x 11784325 44. Carreras E. Boix E. Rosenberg H.F. Cuchillo C.M. Nogués M.V. Both aromatic and cationic residues contribute to the membranelytic and bactericidal activity of eosinophil cationic protein Biochemistry 2003 42 6636 6644 10.1021/bi0273011 12779318 45. Rosenberg H.F. Dyer K.D. Human ribonuclease 4 (RNase 4): Coding sequence, chromosomal localization and identification of two distinct transcripts in human somatic tissues Nucleic Acids Res. 1995 23 4290 4295 10.1093/nar/23.21.4290 7501448 46. Egesten A. Dyer K.D. Batten D. Domachowske J.B. Rosenberg H.F. Ribonucleases and host defense: Identification, localization and gene expression in adherent monocytes in vitro Biochim. Biophys. Acta 1997 1358 255 260 10.1016/S0167-4889(97)00081-5 9366257 47. Futami J. Tsushima Y. Murato Y. Tada H. Sasaki J. Seno M. Yamada H. Tissue-specific expression of pancreatic-type RNases and RNase inhibitor in humans DNA Cell Biol. 1997 16 413 419 10.1089/dna.1997.16.413 9150428 48. Abtin A. Eckhart L. Mildner M. Ghannadan M. Harder J. Schröder J.M. Tschachler E. Degradation by stratum corneum proteases prevents endogenous RNase inhibitor from blocking antimicrobial activities of RNase 5 and RNase 7 J. Investig. Dermatol. 2009 129 2193 2201 10.1038/jid.2009.35 19262607 49. Koutroubakis I.E. Xidakis C. Karmiris K. Sfiridaki A. Kandidaki E. Kouroumalis E.A. Serum angiogenin in inflammatory bowel disease Dig. Dis. Sci. 2004 49 1758 1762 10.1007/s10620-004-9565-4 15628698 50. Nittoh T. Hirakata M. Mue S. Ohuchi K. Identification of cDNA encoding rat eosinophil cationic protein/eosinophil-associated ribonuclease Biochim. Biophys. Acta 1997 1351 42 46 10.1016/S0167-4781(97)00024-9 9116043 51. Spear G.T. Kendrick S.R. Chen H.Y. Thomas T.T. Bahk M. Balderas R. Ghosh S. Weinberg A. Landay A.L. Multiplex immunoassay of lower genital tract mucosal fluid from women attending an urban STD clinic shows broadly increased IL1β and lactoferrin PLoS ONE 2011 6 1278 10.1371/journal.pone.0019560 21572958 52. Hooper L.V. Stappenbeck T.S. Hong C.V. Gordon J.I. Angiogenins: A new class of microbicidal proteins involved in innate immunity Nat. Immunol. 2003 4 269 273 10.1038/ni888 12548285 53. Avdeeva S.V. Chernukha M.U. Shaginyan I.A. Tarantul V.Z. Naroditsky B.S. Human angiogenin lacks specific antimicrobial activity Curr. Microbiol. 2006 53 477 478 10.1007/s00284-006-0033-6 17106799 54. Kulka M. Fukuishi N. Metcalfe D.D. Human mast cells synthesize and release angiogenin, a member of the ribonuclease A (RNase A) superfamily J. Leukoc. Biol. 2009 86 1217 1226 10.1189/jlb.0908517 19625371 55. Shcheglovitova O. Maksyanina E. Ionova I. Rustam’yan Y.L. Komolova G. Cow milk angiogenin induces cytokine production in human blood leukocytes Bull. Exp. Biol. Med. 2003 135 158 160 10.1023/A:1023871931764 12802423 56. Tschesche H. Kopp C. Hörl W. Hempelmann U. Inhibition of degranulation of polymorphonuclear leukocytes by angiogenin and its tryptic fragment J. Biol. Chem. 1994 269 30274 30280 7982938 57. Schmaldienst S. Oberpichler A. Tschesche H. Hörl W.H. Angiogenin: A novel inhibitor of neutrophil lactoferrin release during extracorporeal circulation Kidney Blood Press. Res. 2000 26 107 112 10.1159/000070992 12771535 58. Becknell B. Eichler T.E. Beceiro S. Li B. Easterling R.S. Carpenter A.R. James C.L. McHugh K.M. Hains D.S. Partida-Sanchez S. Ribonucleases 6 and 7 have antimicrobial function in the human and murine urinary tract Kidney Int. 2015 87 151 161 10.1038/ki.2014.268 25075772 59. Pulido D. Arranz-Trullén J. Prats-Ejarque G. Velázquez D. Torrent M. Moussaoui M. Boix E. Insights into the antimicrobial mechanism of action of human RNase6: Structural determinants for bacterial cell agglutination and membrane permeation Int. J. Mol. Sci. 2016 17 552 10.3390/ijms17040552 27089320 60. Spencer J.D. Schwaderer A.L. Eichler T. Wang H. Kline J. Justice S.S. Cohen D.M. Hains D.S. An endogenous ribonuclease inhibitor regulates the antimicrobial activity of ribonuclease 7 in the human urinary tract Kidney Int. 2014 85 1179 1191 10.1038/ki.2013.395 24107847 61. Laudien M. Dressel S. Harder J. Gläser R. Differential expression pattern of antimicrobial peptides in nasal mucosa and secretion Rhinology 2011 49 107 111 21468384 62. Wanke I. Steffen H. Christ C. Krismer B. Götz F. Peschel A. Schaller M. Schittek B. Skin commensals amplify the innate immune response to pathogens by activation of distinct signaling pathways J. Investig. Dermatol. 2010 131 382 390 10.1038/jid.2010.328 21048787 63. Simanski M. Rademacher F. Schröder L. Schumacher H.M. Gläser R. Harder J. IL-17A and IFN-γ synergistically induce RNase 7 expression via STAT3 in primary keratinocytes PLoS ONE 2013 8 1278 10.1371/journal.pone.0059531 23555696 64. Lin Y.M. Wu S.J. Chang T.W. Wang C.F. Suen C.S. Hwang M.J. Outer membrane protein I of Pseudomonas aeruginosa is a target of cationic antimicrobial peptide/protein J. Biol. Chem. 2010 285 8985 8994 10.1074/jbc.M109.078725 20100832 65. Morita T. Niwata Y. Ohgi K. Ogawa M. Irie M. Distribution of two urinary ribonuclease-like enzymes in human organs and body fluids J. Biochem. 1986 99 17 25 3514589 66. Iwama M. Kunihiro M. Ohgi K. Irie M. Purification and properties of human urine ribonucleases J. Biochem. 1981 89 1005 1016 6788751 67. De Prisco R. Sorrentino S. Leone E. Libonati M. A ribonuclease from human seminal plasma active on double-stranded RNA Biochim. Biophys. Acta 1984 788 356 363 10.1016/0167-4838(84)90049-9 6466685 68. Yasuda T. Nadano D. Takeshita H. Kishi K. Two distinct secretory ribonucleases from human cerebrum: Purification, characterization and relationships to other ribonucleases Biochem. J. 1993 296 617 625 10.1042/bj2960617 8280059 69. Mizuta K. Awazu S. Yasuda T. Kishi K. Purification and characterization of three ribonucleases from human kidney: Comparison with urine ribonucleases Arch. Biochem. Biophys. 1990 281 144 151 10.1016/0003-9861(90)90424-W 2383019 70. Landré J.B. Hewett P.W. Olivot J.M. Friedl P. Ko Y. Sachinidis A. Moenner M. Human endothelial cells selectively express large amounts of pancreatic-type ribonuclease (RNase 1) J. Cell. Biochem. 2002 86 540 552 10.1002/jcb.10234 12210760 71. Fischer S. Nishio M. Dadkhahi S. Gansler J. Saffarzadeh M. Shibamiyama A. Kral N. Baal N. Koyama T. Deindl E. Expression and localisation of vascular ribonucleases in endothelial cells Thromb. Haemost. 2011 105 345 355 10.1160/TH10-06-0345 21103661 72. Barnard E.A. Biological function of pancreatic ribonuclease Nature 1969 221 340 344 10.1038/221340a0 4974403 73. Kannemeier C. Shibamiya A. Nakazawa F. Trusheim H. Ruppert C. Markart P. Song Y. Tzima E. Kennerknecht E. Niepmann M. Extracellular RNA constitutes a natural procoagulant cofactor in blood coagulation Proc. Natl. Acad. Sci. USA 2007 104 6388 6393 10.1073/pnas.0608647104 17405864 74. Fischer S. Gerriets T. Wessels C. Walberer M. Kostin S. Stolz E. Zheleva K. Hocke A. Hippenstiel S. Preissner K.T. Extracellular RNA mediates endothelial-cell permeability via vascular endothelial growth factor Blood 2007 110 2457 2465 10.1182/blood-2006-08-040691 17576819 75. Fischer S. Grantzow T. Pagel J.I. Tschernatsch M. Sperandio M. Preissner K.T. Deindl E. Extracellular RNA promotes leukocyte recruitment in the vascular system by mobilising proinflammatory cytokines Thromb. Haemost. 2012 108 730 741 10.1160/TH12-03-0186 22836360 76. Cabrera-Fuentes H.A. Lopez M.L. McCurdy S. Fischer S. Meiler S. Baumer Y. Galuska S.P. Preissner K.T. Boisvert W.A. Regulation of monocyte/macrophage polarisation by extracellular RNA Thromb. Haemost. 2015 113 473 481 10.1160/TH14-06-0507 25589344 77. Cabrera-Fuentes H.A. Niemann B. Grieshaber P. Wollbrueck M. Gehron J. Preissner K.T. Böning A. RNase1 as a potential mediator of remote ischaemic preconditioning for cardioprotection Eur. J. Cardio-Thorac. Surg. 2015 48 732 737 10.1093/ejcts/ezu519 25564211 78. Gordon M.H. Remarks on Hodgkin’s disease: A pathogenic agent in the glands, and its application in diagnosis Br. Med. J. 1933 1 641 644 10.1136/bmj.1.3771.641 20777482 79. Durack D.T. Sumi S.M. Klebanoff S.J. Neurotoxicity of human eosinophils Proc. Natl. Acad. Sci. USA 1979 76 1443 1447 10.1073/pnas.76.3.1443 286329 80. Harrison A.M. Bonville C.A. Rosenberg H.F. Domachowske J.B. Respiratory syncytical virus-induced chemokine expression in the lower airways: Eosinophil recruitment and degranulation Am. J. Respir. Crit. Care Med. 1999 159 1918 1924 10.1164/ajrccm.159.6.9805083 10351940 81. Jacobsen E.A. Ochkur S.I. Lee N.A. Lee J.J. Eosinophils and asthma Curr. Allergy Asthma Rep. 2007 7 18 26 10.1007/s11882-007-0026-y 17504657 82. Kvarnhammar A.M. Cardell L.O. Pattern recognition receptors in human eosinophils Immunology 2012 136 11 20 10.1111/j.1365-2567.2012.03556.x 22242941 83. Lehrer R.I. Szklarek D. Barton A. Ganz T. Hamann K.J. Gleich G.J. Antibacterial properties of eosinophil major basic protein and eosinophil cationic protein J. Immunol. 1989 142 4428 4434 2656865 84. Rosenberg H.F. Recombinant human eosinophil cationic protein. Ribonuclease activity is not essential for cytotoxicity J. Biol. Chem. 1995 270 7876 7881 10.1074/jbc.270.14.7876 7713881 85. Molina H.A. Kierszenbaum F. Hamann K.J. Gleich G.J. Toxic effects produced or mediated by human eosinophil granule components on Trypanosoma cruzi Am. J. Trop. Med. Hyg. 1988 38 327 334 2451444 86. Boix E. Salazar V.A. Torrent M. Pulido D. Nogués M.V. Moussaoui M. Structural determinants of the eosinophil cationic protein antimicrobial activity Biol. Chem. 2012 393 801 815 10.1515/hsz-2012-0160 22944682 87. Torrent M. Pulido D. Nogués M.V. Boix E. Exploring new biological functions of amyloids: Bacteria cell agglutination mediated by host protein aggregation PLoS Pathog. 2012 8 1278 10.1371/journal.ppat.1003005 23133388 88. Pisa D. Alonso R. Rábano A. Rodal I. Carrasco L. Different brain regions are infected with fungi in Alzheimer’s disease Sci. Rep. 2015 5 15015 10.1038/srep15015 26468932 89. Young J.D.E. Peterson C.G.B. Venge P. Cohn Z.A. Mechanism of membrane damage mediated by human eosinophil cationic protein Nature 1986 321 613 616 10.1038/321613a0 2423882 90. Torrent M. Cuyás E. Carreras E. Navarro S. López O. de la Maza A. Nogués M.V. Reshetnyak Y.K. Boix E. Topography studies on the membrane interaction mechanism of the eosinophil cationic protein Biochemistry 2007 46 720 733 17223693 91. Torrent M. Navarro S. Moussaoui M. Nogués M.V. Boix E. Eosinophil cationic protein high-affinity binding to bacteria-wall lipopolysaccharides and peptidoglycans Biochemistry 2008 47 3544 3555 10.1021/bi702065b 18293932 92. Navarro S. Aleu J. Jiménez M. Boix E. Cuchillo C. Nogués M.V. The cytotoxicity of eosinophil cationic protein/ribonuclease 3 on eukaryotic cell lines takes place through its aggregation on the cell membrane Cell. Mol. Life Sci. 2008 65 324 337 10.1007/s00018-007-7499-7 18087674 93. Chang K.C. Lo C.W. Fan T. Chang M.D.T. Shu C.W. Chang C.H. Chung C.T. Fang S.L. Chao C.C. Tsai J.J. TNF-α Mediates eosinophil cationic protein-induced apoptosis in BEAS-2 B cells BMC Cell Biol. 2010 11 1278 10.1186/1471-2121-11-6 20089176 94. Fan T.C. Chang H.T. Chen I. A heparan sulfate-facilitated and raft-dependent macropinocytosis of eosinophil cationic protein Traffic 2007 8 1778 1795 10.1111/j.1600-0854.2007.00650.x 17944807 95. Harris P. Johannessen K.M. Smolenski G. Callaghan M. Broadhurst M.K. Kim K. Wheeler T.T. Characterisation of the anti-microbial activity of bovine milk ribonuclease4 and ribonuclease5 (angiogenin) Int. Dairy J. 2010 20 400 407 10.1016/j.idairyj.2009.12.018 96. Murata M. Wakabayashi H. Yamauchi K. Abe F. Identification of milk proteins enhancing the antimicrobial activity of lactoferrin and lactoferricin J. Dairy Sci. 2013 96 4891 4898 10.3168/jds.2013-6612 23769363 97. Fett J.W. Strydom D.J. Lobb R.R. Alderman E.M. Bethune J.L. Riordan J.F. Vallee B.L. Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells Biochemistry 1985 24 5480 5486 10.1021/bi00341a030 4074709 98. Harper J.W. Vallee B.L. A covalent angiogenin/ribonuclease hybrid with a fourth disulfide bond generated by regional mutagenesis Biochemistry 1989 28 1875 1884 10.1021/bi00430a067 2719939 99. Lee F.S. Vallee B.L. Characterization of ribonucleolytic activity of angiogenin towards tRNA Biochem. Biophys. Res. Commun. 1989 161 121 126 10.1016/0006-291X(89)91569-6 2730651 100. Sheng J. Xu Z. Three decades of research on angiogenin: A review and perspective Acta Biochim. Biophys. Sin. 2016 48 399 410 10.1093/abbs/gmv131 26705141 101. Tsuji T. Sun Y. Kishimoto K. Olson K.A. Liu S. Hirukawa S. Hu G.F. Angiogenin is translocated to the nucleus of HeLa cells and is involved in ribosomal RNA transcription and cell proliferation Cancer Res. 2005 65 1352 1360 10.1158/0008-5472.CAN-04-2058 15735021 102. Lee H.S. Lee I. Kang T.C. Jeong G.B. Chang S.I. Angiogenin is involved in morphological changes and angiogenesis in the ovary Biochem. Biophys. Res. Commun. 1999 257 182 186 10.1006/bbrc.1999.0359 10092530 103. King T.V. Vallee B.L. Neovascularisation of the meniscus with angiogenin. An experimental study in rabbits J. Bone Jt. Surg. 1991 73 587 590 104. Walker C.R. Hautefort I. Dalton J.E. Overweg K. Egan C.E. Bongaerts R.J. Newton D.J. Cruickshank S.M. Andrew E.M. Carding S.R. Intestinal intraepithelial lymphocyte-enterocyte crosstalk regulates production of bactericidal angiogenin 4 by Paneth cells upon microbial challenge PLoS ONE 2013 8 1278 10.1371/journal.pone.0084553 24358364 105. Forman R.A. deSchoolmeester M.L. Hurst R.J. Wright S.H. Pemberton A.D. Else K.J. The goblet cell is the cellular source of the anti-microbial angiogenin 4 in the large intestine post Trichuris muris infection PLoS ONE 2012 7 1278 10.1371/journal.pone.0042248 22970115 106. Rosenberg H.F. Dyer K.D. Molecular cloning and characterization of a novel human ribonuclease (RNase k6): Increasing diversity in the enlarging ribonuclease gene family Nucleic Acids Res. 1996 24 3507 3513 10.1093/nar/24.18.3507 8836175 107. Christensen-Quick A. Lafferty M. Sun L. Marchionni L. DeVico A. Garzino-Demo A. Human Th17 cells lack HIV-inhibitory RNases and are highly permissive to productive HIV infection J. Virol. 2016 10.1128/JVI.02869-15 27334595 108. Harder J. Schröder J. RNase 7, a novel innate immune defense antimicrobial protein of healthy human skin J. Biol. Chem. 2002 277 46779 46784 10.1074/jbc.M207587200 12244054 109. Zasloff M. Antimicrobial RNases of human skin J. Investig. Dermatol. 2009 129 2091 2093 10.1038/jid.2009.216 19809422 110. Harder J. Schröder J.M. Psoriatic scales: A promising source for the isolation of human skin-derived antimicrobial proteins J. Leukoc. Biol. 2005 77 476 486 10.1189/jlb.0704409 15629886 111. Firat Y.H. Simanski M. Rademacher F. Schröder L. Brasch J. Harder J. Infection of keratinocytes with Trichophytum rubrum induces epidermal growth factor-dependent RNase 7 and human β-defensin-3 expression PLoS ONE 2014 9 1278 10.1371/journal.pone.0093941 24747887 112. Otri A.M. Mohammed I. Abedin A. Cao Z. Hopkinson A. Panjwani N. Dua H.S. Antimicrobial peptides expression by ocular surface cells in response to Acanthamoeba castellanii : An in vitro study Br. J. Ophthalmol. 2010 94 1523 1527 10.1136/bjo.2009.178236 20679078 113. Mohammed I. Yeung A. Abedin A. Hopkinson A. Dua H.S. Signalling pathways involved in ribonuclease-7 expression Cell. Mol. Life Sci. 2010 68 1941 1952 10.1007/s00018-010-0540-2 20967562 114. Burgey C. Kern W.V. Römer W. Sakinc T. Rieg S. The innate defense antimicrobial peptides hBD3 and RNase7 are induced in human umbilical vein endothelial cells by classical inflammatory cytokines but not Th17 cytokines Microbes Infect. 2015 17 353 359 10.1016/j.micinf.2015.01.005 25637949 115. Köten B. Simanski M. Gláser R. Podschun R. Schröder J.M. Harder J. RNase 7 contributes to the cutaneous defense against Enterococcus faecium PLoS ONE 2009 4 1278 10.1371/journal.pone.0006424 19641608 116. Fritz P. Beck-Jendroschek V. Brasch J. Inhibition of dermatophytes by the antimicrobial peptides human β-defensin-2, ribonuclease 7 and psoriasin Med. Mycol. 2012 50 579 584 10.3109/13693786.2012.660203 22332906 117. Torrent M. Badia M. Moussaoui M. Sanchez D. Nogués M.V. Boix E. Comparison of human RNase 3 and RNase 7 bactericidal action at the Gram-negative and Gram-positive bacterial cell wall FEBS J. 2010 277 1713 1725 10.1111/j.1742-4658.2010.07595.x 20180804 118. Chan C.C. Moser J.M. Dyer K.D. Percopo C.M. Rosenberg H.F. Genetic diversity of human RNase 8 BMC Genom. 2012 13 1278 10.1186/1471-2164-13-40 22272736 119. Rudolph B. Podschun R. Sahly H. Schubert S. Schröder J.M. Harder J. Identification of RNase 8 as a novel human antimicrobial protein Antimicrob. Agents Chemother. 2006 50 3194 3196 10.1128/AAC.00246-06 16940129 120. Mylonakis E. Paliou M. Hohmann E.L. Calderwood S.B. Wing E.J. Listeriosis during pregnancy: A case series and review of 222 cases Medicine 2002 81 260 269 10.1097/00005792-200207000-00002 12169881 121. Gupta S.K. Haigh B.J. Griffin F.J. Wheeler T.T. The mammalian secreted RNases: Mechanisms of action in host defence Innate Immun. 2013 19 86 97 10.1177/1753425912446955 22627784 122. Martin L. Koczera P. Simons N. Zechendorf E. Hoeger J. Marx G. Schuerholz T. The human host defense ribonucleases 1, 3 and 7 are elevated in patients with sepsis after major surgery—A pilot study Int. J. Mol. Sci. 2016 17 294 10.3390/ijms17030294 26927088 123. Schuerholz T. Brandenburg K. Marx G. Antimicrobial peptides and their potential application in inflammation and sepsis Crit. Care 2012 16 207 10.1186/cc11220 22429567

PMC005xxxxxx/PMC5000676.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081279ijms-17-01279ReviewEpigenetic Modifications of Major Depressive Disorder Saavedra Kathleen 1Molina-Márquez Ana María 1Saavedra Nicolás 1Zambrano Tomás 1Salazar Luis A. 12*Bustin Stephen A. Academic Editor1 Center of Molecular Biology and Pharmacogenetics, Scientific and Technological Bioresource Nucleus, Universidad de La Frontera, Temuco 4811230, Chile; [email protected] (K.S.); [email protected] (A.M.M.-M.); [email protected] (N.S.); [email protected] (T.Z.)2 Millennium Institute for Research in Depression and Personality (MIDAP), Universidad de La Frontera, Temuco 4811230, Chile* Correspondence: [email protected]; Tel.: +56-45-259-672405 8 2016 8 2016 17 8 127920 6 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Major depressive disorder (MDD) is a chronic disease whose neurological basis and pathophysiology remain poorly understood. Initially, it was proposed that genetic variations were responsible for the development of this disease. Nevertheless, several studies within the last decade have provided evidence suggesting that environmental factors play an important role in MDD pathophysiology. Alterations in epigenetics mechanism, such as DNA methylation, histone modification and microRNA expression could favor MDD advance in response to stressful experiences and environmental factors. The aim of this review is to describe genetic alterations, and particularly altered epigenetic mechanisms, that could be determinants for MDD progress, and how these alterations may arise as useful screening, diagnosis and treatment monitoring biomarkers of depressive disorders. major depressive disorderdepressionepigenetic modificationsmethylationhistone modificationmiRNAsbiomarkers ==== Body 1. Introduction Major depressive disorder (MDD) is a chronic and debilitating disease that affects more than 350 million people worldwide, making it one of the most common mental illnesses [1]. MDD ranks second in terms of disease burden, accounting for 40.5% of disability-adjusted life years [2]. The World Health Organization (WHO) has estimated that MDD will be the second leading cause of disability throughout the world, preceded only by ischemic heart disease [3]. These data include MDD among major public health problems. MDD is defined by the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) as a complex and heterogeneous syndrome that covers a wide spectrum of symptoms, including anhedonia, disturbed sleep, reduced appetite and energy, depressed mood, reduced concentration and suicidal thoughts, among others [4,5,6]. Since MDD diagnosis continues to be based in detecting such clinical symptoms, current research efforts aim to identify specifics biomarkers that might facilitate depressive disorders diagnosis. However, such efforts have been unsuccessful, partly because depressive disorders neurobiological basis and pathophysiology remain poorly understood. Genetic factors, and their association with the environment, play important roles in MDD; therefore, exploring the genetic background might reveal important information about the mechanisms underlying MDD development. The present review describes how genetics and epigenetics alterations can be important determinants for MDD progress, and how these alterations may arise as interesting biomarkers for screening, diagnosis and treatment monitoring of depressive disorders. 2. Genetics of Major Depression Epidemiological studies have shown that major depression is a familial disorder. A meta-analysis derived from five twin studies including more than 21,000 subjects revealed a genetic contribution for MDD development of 37% (95% CI: 31%–42%), added to evidence obtained from family studies showing two- to three-fold increased MDD risk (Mantel–Haenszel odds ratio = 2.84, 95% CI = 2.31–3.49) during the lifetime among first-degree relatives [7]. Moreover, when considering disease severity, defined by relapse rate and early disease onset, depression heritability may increase by up to 70% [8,9]. The important heritability observed in this disorder has elevated the expectations of identifying key genes involved in MDD progress that might be considered potentials risk indicators. However, no specific genetic variants have been identified as robust contributors to major depression. Several linkage and association studies have been conducted with the hope to identify risk-associated genes. Both are complementary methods to locate susceptibility genes, but they have failed to identify universal genetic risk or causal factors for depression disorders. At present, and due to the rapid development of technological advances in the field of genomics, it is possible to perform large genome-wide association studies (GWAS). This method, although combining the advantage of the breadth of linkage with the power of association, has also failed, as few genetic variants have been strongly implicated in depressive disorders. Lee et al. [10] performed a meta-analysis including 4346 cases and 4430 controls, and found that genes involved in glutamatergic synaptic neurotransmission were significantly associated with MDD. However, they were not able to associate a genetic variant with MDD risk. Wray et al. [11] studied 5763 cases and 6901 controls, showing evidence of an association between MDD and polymorphisms at or near genes with plausible biological relevance, such as galanin (GAL) and adenylate cyclase 3 (ADCY3). Ripke et al. performed a meta-analysis from 9240 cases and 9519 controls, with replication in 6783 cases and 50,695 controls, but no genetics variants achieved genome-wide significance, neither in the MDD discovery phase nor in the MDD replication phase [11,12]. Despite these negative results, alterations in numerous genes have been linked to depression pathophysiology in different studies. Some of these genes include regulators of neurotransmitter signaling as serotonin transporter (SLC6A4; particularly the 5-HTTLPR polymorphism), monoamine oxidase A (MAOA), catechol-o-methyltransferase (COMT), regulators of neural plasticity and connectivity as the brain-derived neurotrophic factor (BDNF) and the enzyme tryptophan hydroxylase (TPH) that mediates serotonin synthesis on peripheral (TPH1) and cerebral (TPH2) level, among others. Studies of polymorphisms associated with these genes have showed that genetic variants might increase genetic susceptibility to develop depression, anxiety, stress or cognitive functions alterations [13,14,15,16,17,18,19,20,21]. However, again, these results lack consistency and have not shown the desired reproducibility. The discrepancy observed between evidence showing MDD as a family disease together with the impossibility to identify genetic alterations associated with MDD suggest that additional factors are involved in MDD development. In the report of Sullivan et al. [7], they demonstrated that common environmental influences had a minimal contribution of 0% (95% CI: 0%–5%), while individual-specific environmental factors showed a significant contribution of 63% (95% CI: 58%–67%). Since MDD cannot be attributed to a single genetic mutation or exposure to one specific environmental stimulus, MDD is proposed to arise from an interaction between genetic variations and environmental factors [22,23]. Regarding environmental factors influencing MDD development, exposure to environmental stressors, especially as traumatic events in early life, is one of the strongest risk factors described to date. Recently, it has been suggested that adverse environmental stimulus can stably alter gene expression in healthy subjects and encourage depression development through epigenetic mechanisms [24,25]. Moreover, reports show that epigenetic processes would be involved in the development of several human diseases, including psychiatric disorders as MDD [26]. 3. Epigenetic Modifications and Depressive Disorders Epigenetics refers to changes in gene expression that are not due to alterations in DNA sequence; these changes can be potentially heritable, but environmentally modifiable, and could explain different scenarios in which medical observations confront traditional genetics [27,28]. Epigenetic regulation is fundamental for many cellular processes including gene (mRNA) and microRNA (miRNA) expression, DNA–protein interactions, suppression of transposable elements, cellular differentiation, embryogenesis, X-chromosome inactivation and genomic imprinting. In the same way, epigenetic regulation not only regulates physiological but also pathological processes [29]. In fact, has been described that the epigenetics has an important role in the development of many mental illness, such as MDD [30,31]. Overall, epigenetic modifications can be grouped into three general categories: DNA methylation, histone modification and nucleosome positioning. Non-coding RNA (ncRNA)-mediated regulation is also considered an important epigenetic regulation in the pathophysiologic process of depression [32,33]. 3.1. DNA Methylation The most widely studied epigenetic modification in humans is DNA methylation. This mechanism consist in the addition of a methyl group at the 5′ position of cytosines in cytosine-phosphate-guanine dinucleotides (CpG), a process catalyzed by DNA methyltransferases (DNMTs) occurring almost exclusively in CpG dinucleotides usually clustered within the promoter region of genes, termed CpG islands [29]. Cytosine methylation reduces the access of transcription factors into regulatory elements; therefore, DNA methylation is associated with transcriptional repression. Evidence suggests that DNA methylation is responsive to environmental signals [34], and recently, a large number of studies conducted in animal models and humans support the idea that DNA methylation plays an important role in mediating stress effects. Important environmental factors to consider in the risk for developing MDD occur early during the gestational stage of an individual [35]. Accordingly, gestational stimuli play important roles in the development of various neuropsychiatric disorders, including MDD [36,37]. Intrauterine conditions can have long-term effects in terms of risk of neurological or psychiatric disorders, which would be mediated through epigenetic modifications such as DNA methylation (Figure 1). Nieratschker et al. [38] reported that MORC1 methylation, a gene that evokes a depression-like phenotype in mice, is a candidate marker for MDD development associated with early life stress in rodents, primates and humans undergoing prenatal stressed conditions [39,40]. Interesting models for the study of variations in DNA methylation profiles are the monozygotic twins, as they have almost the same DNA sequence, but frequently show phenotypic discordance [41,42,43,44,45]. Monozygotic twins methylation profile can be very similar, not only by the nearly identical DNA sequence that they possess, but also because both individuals are subjected to one common pre- and post-natal environment [46]. However, there are still differences in methylation profiles, which can be produced by exposure to environmental causes influencing one of the twins, or by stochastic factors [47,48]. Considering the aforementioned background, Córdova-Palomera et al. [49] evaluated differences in DNA methylation of monozygotic twins using two analytical strategies to identify differentially methylated probes (DMPs) and variably methylated probes (VMPs), showing associations with differences in the psychopathological status of twins. Most DMPs were located in genes previously related to neuropsychiatric disorders; one of these was the WD Repeat Domain 26 (WDR26) gene, implicated in MDD from GWAS data [11,50]. VMPs were also located in genes such as Calcium Channel, Voltage-Dependent, L Type, Alpha 1C (CACNA1C), Insulin-Like Growth Factor 2 (IGF2) and the p38 MAP kinase (MAPK11), showing enrichment for biological processes such as glucocorticoid signaling. DNA sequence variation of CACNA1C has been recognized as a susceptibility factor for depressive psychopathology development, and its methylation changes have been associated with risk factors for depressive disorders as early-life stress [11,51,52]. Additionally, the activity of MAPK11 has been associated with depression phenotypes [53]. While GWAS have failed in identifying sequence variations influencing MDD susceptibility, epigenetic marks such as DNA methylation have emerged as better candidates to be employed as depression biomarkers. Sabunciyan et al. [54] performed the first genome-wide DNA methylation (NMD) scan in MDD. In that study, 39 post-mortem frontal cortex MDD samples were compared to 26 controls using the Comprehensive High-throughput Arrays for Relative Methylation (CHARM) platform, covering 3.5 million CpGs. They identified 224 candidate regions having DNA methylation differences >10% in highly enriched regions for neuronal growth and development genes. Further experimental validation showed the greatest differences in Proline Rich Membrane Anchor 1 (PRIMA1), with 12%–15% increased DNA methylation in MDD individuals than controls [54]. PRIMA1 is important within MDD biology as it encodes a protein that organizes acetylcholinesterase (AChE) into tetramers, anchoring AChE to neural cell membranes [55,56]. Postmortem analysis from brain tissue of individuals with neuropsychiatric disorders have shown that DNA methylation is compromised in comparison to control individuals [27]. Brain tissue is considered an ideal sample for DNA methylation analyses in neuropsychiatric disorders. However, its accessibility is highly difficult; therefore, tissue sampling is restricted to postmortem collection. This critical issue drives the need to search for additional non-invasive samples with better accessibility that also reflects the biochemical and molecular changes occurring in the brain. Thus, numerous studies using peripheral blood as a non-invasive model for DNA methylation analyses in neuropsychiatric diseases have been performed, allowing the identification of potential circulating biomarkers for MDD diagnosis. An example is the study performed by Numata et al. [57], identifying DNA methylation markers able to distinguish between medication-free patients with MDD and non-psychiatric controls. In this study, significant diagnostic differences in DNA methylation were observed at 363 CpG sites, all of them demonstrating lower DNA methylation in patients with MDD than controls [57]. Some of these markers were Cell Cycle Associated Protein 1 (CAPRIN1), cAMP-response element binding protein/p300-interacting transactivator with Glu/Asp-rich carboxy-terminal domain-2 (CITED2), Diacylglycerol Kinase (DGKH), Glycogen Synthase Kinase 3 Beta (GSK3B) and Serum/Glucocorticoid Regulated Kinase 1 (SGK1), genes previously associated with MDD [40,58,59,60,61,62]. Moreover, it has been reported that DNA methylation status obtained from blood samples shows a correlation with the methylation status observed in post-mortem brain tissue. Stenz et al. [63] described a correlation between promoter methylation of Brain-Derived Neurotrophic Factor (BDNF) gene in blood and post-mortem brain tissue from depressed patients (Pearson, n = 98, 263 r = 0.48, p = 4.5 × 10−7). BDNF promotes proliferation, differentiation and survival of neurons and is crucial for neural plasticity and cognitive function [64]. Meanwhile, Januar et al. [65] proposed the detection of BDNF methylation in oral tissue as a potential depression biomarker (promoter I, Δmean = 0.4%, p = 0.0002). Taken together, DNA methylation can be one of the several epigenetic mechanisms by which stressors can have long-term effects through gene expression alteration of exposed individuals, influencing or determining the course of depressive disorders. Numerous studies have reported the identification of genes frequently methylated and related to depressive disorder pathophysiology in peripheral blood samples, allowing potential identification of biomarkers for MDD early detection and diagnosis. A list of these biomarkers is displayed in Table 1. 3.2. Histone Modification All histones undergo post-transcriptional modifications affecting the histone tail. Modifications include acetylation, methylation, phosphorylation, ubiquitination, SUMOylation and ADP-ribosylation, among others [29], and can change the DNA–histone core interaction, which is involved in gene expression regulation by chromatin remodeling. Histone acetylation involves the transferring of an acetyl group to histone tails by Histone acetyltransferases (HATS) enzymes. This process promotes histone units unfolding and chromatin decondensation, allowing transcription factors binding to genomic DNA and therefore, promoting gene expression. Conversely, enzymes known as histone deacetylases (HDACs) remove the acetyl group from the histone tail, causing chromatin condensation and preventing transcription factors access to genomic DNA, thus, decreasing gene expression [70]. Few studies investigating the effect of histone modification on depressive disorders development have been conducted (Table 2). Interestingly, the initial findings were achieved by using histone deacetylase inhibitors (HDACi), alone or in combination with antidepressants, in a variety of animal models [71,72,73,74]. An example is valproate, an HDACi commonly used as a mood stabilizer in bipolar disorder which function may be mediated through HDACs inhibition [75,76]. Thus, HDACs dysfunction can be involved in the pathophysiology of mood disorders. Covington et al. [71] explored the impact of chronic stress on histone acetylation in the nucleus accumbens (NAc), an important limbic brain region, in a chronic stress defeat model and postmortem tissue of depressed individuals. In this study, histone acetylation (H3K14ac) was transiently decreased and then stably increased in the NAc of mice after chronic social defeat stress, which was correlated with a reduction in HDAC2 levels and reproduced in postmortem tissue of depressive patients. Later, the effect of direct MS-275 infuse (a selective inhibitor of class I HDACs) in the NAc was also evaluated, resulting in a robustly antidepressant effect of chronic defeat stress on global patterns of gene expression, and suggesting that histone acetylation has an adaptive role in stress and depression [25]. HDAC5 expression studies in the NAc of mice susceptible to chronic social defeat stress have also been performed, finding that HDAC5 was repressed, whereas imipramine treatment (a chronic antidepressant) increased HDAC5 expression [77]. In addition, mice lacking HDAC5 exhibited increased depressive-like behaviors after chronic social defeat stress compared to control animals [77]. The results reported by Convington et al. [71] and Renthal et al. [77] suggest that histone modifications by HDACs play an adaptive role in chronic psychiatric illnesses response, and that HDAC5 targets may have a depressive role, while targets of HDAC2 may have an antidepressant role. Other studies evaluated gene expression of the histone acetylation machinery as potential biomarkers in peripheral blood cells of depressed patients. Hobara et al. [78] assessed gene expression of 11 HDACs (including HDAC2 and -5) in peripheral white blood cells of MDD and bipolar disorder (BPD) patients during depressive and remissive episodes. Experiments revealed that HDAC2 and HDAC5 expression was significantly increased in MDD patients in the depressive state compared to controls subjects (HDAC2 p < 0.001; HDAC5 p = 0.001), while during remissive state, expression of the same HDACs was comparable to controls subjects (HDAC2 p = 0.975; HDAC5 p = 0.506), suggesting a state-dependent alteration (Figure 2) [78]. These results are consistent with those previously reported by Iga et al. [79] in peripheral leucocytes of drug-free depressive patients, in which the expression of HDAC5 was higher compared to controls. On the other hand, due to the trimethylated forms of histone H3, lysines (K4, K9, and K2) can serve to distinguish between active/inactive chromatin, and remain stable during tissue autolysis, investigations performed on these epigenetic marks have raised special interest within postmortem research [81]. Consequently, one of the best histone modifications studied so far corresponds to the tri-methylation of the fourth lysine tail on histone 3 (H3K4me3) [82]. This modification opens chromatin, allowing the transcriptional machinery binding to the promoter region of genes, inducing transcription initiation [65]. Enrichments of this marker are highly associated with increased gene expression levels [83,84]. Cruceanu et al. [80] analyzed the expression of transcript variants for the three synapsin genes and investigated their relationship with H3K4me3 promoter enrichment in post-mortem brain samples from BPD (n = 13), MDD (n = 18) and controls (n = 14) patients with no psychiatric history. The SYN1, SYN2 and SYN3 genes encode neuronal phosphoproteins belonging to the synapsin family, which are reported to play crucial activities in different neuropsychiatric disorders [85,86,87]. Cruceanu and colleagues found that histone modification markers were significantly increased in MDD, and this effect was correlated with a significant increase in SYN2 gene expression [80]. 3.3. Non-Coding RNAs The term ncRNAs indicates different classes of RNAs that are not translated into a protein, but exert a functional role as well. There are five classes of ncRNAs distinguished so far: microRNAs (miRs), small nucleolar RNAs (snoRNAs), large intergenic non-coding RNAs (lincRNAS), PIWI-interacting RNAs (piRNAs) and transcribed ultraconserved regions (T-UCRs). However, the most widely studied class of ncRNAs corresponds to microRNAs (miRs), small ncRNAs of 22 nucleotides that mediates post-transcriptional gene silencing by controlling the translation of mRNA [88,89]. These ncRNAs are involved in many different regulatory processes, including proliferation, differentiation, apoptosis and development. They can regulate one particular target or may regulate the expression of hundreds of genes simultaneously [90]. miRs modulate mRNA expression depending on the number of mismatches between its own sequence and the sequence of the target mRNA, regulating expression by enzymatic target degradation or by preventing mRNA translation into protein due to steric hindrance of the protein synthesis machinery [91,92,93]. In recent years, the role of miRs in neuropsychiatric and neurodegenerative diseases development has gained significant attention, however, the implication of this kind of ncRNA in affective diseases, particularly MDD, is less clear. Nevertheless, recent studies have suggested that miRs play a key role in MDD pathophysiology, particularly at neurogenesis, synaptic plasticity and regulation of key genes that are critical components of signaling pathways involved in MDD. Similarly, various researches have shown miR profiles dysregulation when comparing depressed patients with normal controls (Figure 2). Table 3 shows a summary of miRs studies in depressive disorders patients. Uchida et al. [94] reported that miR-18a inhibits translation of the glucocorticoid receptor in neuron cell culture, and that its expression in the hypothalamic paraventricular nucleus is increased in F344 rats compared to Sprague-Dawley control rats, both with repeated restraint stress. This finding could explain, at least in part, the decreased expression of glucocorticoid receptors in depressed individuals, and why the majority of depressed patients would have elevated cortisol levels in plasma and cerebrospinal fluid (CSF) [95]. Vreugdenhil et al. [96] reported that miR-18a and miR-124a decreased glucocorticoid receptor protein expression by luciferase reporter assays in NS1 cells, confirming the results obtained by Uchida et al. [94]. Dwivedi et al. [97] were the first to examine global expression patterns of miRs in the dorsolateral prefrontal cortex (dlPFC) of depressed subjects. They found 21 miRs significantly downregulated in the prefrontal cortex of depressed patients compared to normal controls, many of them implicated in cellular growth and differentiation and showing high synaptic enrichment [97,98,99]. miRs regulation of BDNF, a critical gene for MDD physiopathology, has also been evaluated in serum of MDD patients. Li et al. [33] found two miRs (miR-182 and miR-132) as putative regulators of BDNF expression in MDD patients. In that study, miR-182 and miR-132 levels were upregulated, while the expression of BDNF was repressed. The Self-Rating Depression Scale score showed an inverse correlation with serum BDNF levels, while demonstrating a direct correlation with miR-132 levels [33]. In addition, Smalheiser et al. [98] demonstrated that miR-494 and miR-335 are downregulated in the prefrontal cortex of depressed suicide patients [98]. Antidepressant (AD) treatment affects miRs expression, allowing the identification of new miRs involved in MDD physiopathology. Bocchio-Chiavetto et al. [100] conducted a whole-miRNome quantitative analysis using qRT-PCR and evaluating miRs expression changes in the blood of 10 depressed subjects following 12 weeks of escitalopram treatment. Thirty miRs were differentially expressed after escitalopram treatment: 28 miRs were upregulated and two miRs were strongly downregulated. Among these miRs differentially regulated, miR-132 has been implicated in both neurogenesis and synaptic plasticity, whereas miR-26a, miR26b and miR-183 contribute to BDNF function in the brain [100,101]. The use of circulating miRs as potential clinical biomarkers has gained substantial interest in recent years. Numerous studies indicate that miRs can be detected in several body fluids, such as blood and cerebrospinal fluid, in addition to highlighting its great stability [32,102,103,104,105,106]. Other studies show that miRs, under health conditions, have a stable expression; while pathological conditions within the central nervous system can alter their expression greatly [107]. Changes in circulating miRs correlate with expression changes of miRs evaluated in neuronal tissues [108,109,110]. In this context, Belzeaux et al. [96] evaluated the expression of miRs in peripheral blood mononuclear cells (PBMC) of patients with and without MDD at baseline, and two and eight weeks after antidepressive treatments. The authors identified changes in several miRs (miR-107, miR-133a, miR-148a, miR-200c, miR-381, miR-425-3p, miR-494, miR-517b, miR-579, miR-589, miR-636, miR-652, miR-941, and miR-1243). Two of these miRs were overexpressed in MDD patients after an eight-week follow-up (miR-589 and miR-941). Furthermore, based on target profiling predicted for these miRs, a combination of four genes (PPT1, TNF, IL1B and HIST1H1E) showed potential as biomarkers that could have predictive value for treatment response [96]. Fan et al. [112] explored miRNAs expression in PBMC as specific blood-based biomarker for MDD patients, identifying 26 miRs with significant expression changes. After validating in a larger cohort, five miRs (miR-26b, miR-1972, miR-4485, miR-4498, and miR-4743) were found up-regulated, which would be controlling pathways associated with the nervous system and brain functions [112]. Wan et al. [113] examined the differential miRs expression profile in CSF and serum of MDD patients, finding three upregulated miRs (miR-221-3p, miR-34a-5p, let-7d-3p) and one repressed miR. These results were further validated in another 32 MDD patients. ROC analysis showed that the area under the curve (AUC) of let-7d-3p, miR-34a-5p, miR-221-3p and miR-451a was 0.94, 0.98, 0.97 and 0.94, with 90.48%, 95.24%, 90.48% and 90.48% specificity, and 93.75%, 96.88%, 90.63% and 84.85% sensitivity, respectively, suggesting that these miRs might serve as MDD biomarkers [113]. Most recently, Wang et al. [114] identified that miR-144-5p levels are associated with depressive symptoms, and miR-144-5p detection in plasma could be a potential biomarker for pathologic processes related to depression. 4. Epigenetics Modifications in MDD Therapy In addition to being involved in depressive diseases physiopathology, epigenetic modifications could be implicated in the mechanism of action of antidepressants [115] (Table 4). Perisic et al. [116] reported that treatment with valproic acid may results in stronger global chromatin modifications, including histone H3/H4 hyperacetylation, 2MeH3K9 hypomethylation, and DNA demethylation [116]. In the same way, amitriptyline treatment can induce slight cytosine demethylation, paralleled by a reduction in DNA methyltransferase enzymatic activity, without affecting global histone acetylation status [116,117]. AD treatment can regulate epigenetic modifications affecting the expression of genes involved in MDD pathology. Electroconvulsive therapy (ECT) has been shown to be an effective and safe treatment for MDD patients [118]. However, the physiological mechanisms of ECT and its effects on brain structure are still unclear. In addition, some reports suggest the participation of histone modification in the mechanism of AD treatment. In this context, Tsankova et al. [119] found that histone modification controls the expression of BDNF after electroconvulsive stimulation in the brain hippocampus from rats (animal model equivalent of ECT), depending on treatment duration (30 min, two hours or 24 h), post-treatment time and gene promoter region. These results suggest that epigenetic modulation could be important for the action mechanism of ECT. Iga et al. [79] evaluated the expression of HDAC5 and cyclic AMP response element-binding protein 1 (CREB) in 20 MDD patients after eight-week paroxetine treatment. They reported higher HDAC5 and CREB post-treatment levels, and the correlation between levels of HDAC5 and CREB was positive [79]. Tsankova et al. [73] demonstrated that chronic social defeat stress induced lasting downregulation of Bdnf transcripts III and IV in mice hippocampus, and robustly increased repressive histone methylation at their corresponding promoters. Imipramine treatment in this model resulted in downregulation reversal and increased histone acetylation within these promoters. Hyperacetylation by chronic imipramine was associated with selective Hdac5 downregulation [73]. miRs have been also involved in the antidepressant mechanism of various drugs. At the moment, the first-choice therapy prescribed for patients suffering from MDD corresponds to a class of drugs known as Serotonin-selective reuptake inhibitors (SSRIs) [120]. SSRIs downregulate the serotonin transporter (SERT) and serotonin 1A (5-HT1A) autoreceptors on serotonergic neurons in the raphe nuclei. Nevertheless, to date, the stable modifications induced by chronic SSRI medication in serotonergic transmission is lacking a clear mechanism that can explain SERT and 5-HT1A repression [120]. Baudry et al. [121] demonstrated that SERT is a miR-16 target. When comparing miR-16 expression patterns, reports show higher levels of this miR in noradrenergic vs. serotonergic cells, where miR-16 repression in noradrenergic neurons can produce de novo SERT expression. Using a mice model, fluoxetine therapy has been shown to induce higher miR-16 levels within the serotonergic raphe nuclei, which is consequently followed by SERT repression [122,123]. Furthermore, raphe nuclei exposed to fluoxetine releases the neurotrophic factor S100b, which acts on noradrenergic cells of the locus coeruleus [121,123]. Based on these results, Baudry et al. [121] proposed that miR-16 contributes to the therapeutic action of SSRI antidepressants in monoaminergic neurons. In the same way, miR-1202 is a primate-specific miRNA associated with MDD pathophysiology and SSRIs responsiveness [124]. Issler et al. [125] identified a strong interaction between miR-135 and 5HT transporter and 5-HT1A receptor transcripts. Interestingly, miR-135a levels were up-regulated after AD treatment administration. Using genetically modified mice expressing higher or lower miR-135 levels, Issler and colleagues demonstrated major alterations in anxiety- and depression-like behaviors, 5HT levels, and behavioral response to AD treatment. Finally, miR-135a levels in blood and brain of depressed human patients were also evaluated, identifying significant lower expression levels. These results suggest both a potential role for miR-135 in the pathophysiology of MDD and its use as an endogenous antidepressant [125]. To date, few studies in human have been conducted to investigate the effect of AD treatment on epigenetic modifications. Some clinical trials have been performed (Table 4), but none has achieved significant results. Independently to antidepressant treatment, it has shown that the use of epigenetic engineering can also be effective for treating patients with depression. Recently it has been shown that the use of zinc-finger proteins or transcription activator-like effectors, they can be used to control depression and addiction related behavior [126]. In the same way, Sun et al. demonstrate that the ATP-dependent chromatin remodeling can be a novel therapeutics targets in depressed patients to mediate depressive-like behavior [127]. These results represent a promising area of research in the treatment of patients with MDD. 5. Future Perspectives The described studies indicate that psychiatric disorders, including MDD, are complex multifactorial diseases that include chronic alterations in the structure and function of neural circuits. Despite reports stating genetics plays an important role in the etiology of these diseases, there are discordances when analyzing the results of investigations on identical twins. These differences clarify the participation of additional factors. Environmental influences, like early life stress, play an important role in the development of psychiatric diseases to induce expression changes in important genes associated with MDD physiopathology. These changes would be mediated through epigenetic modifications, promoting or suppressing gene expression through three main mechanisms: DNA methylation, histone modification and miRs. Evidence supporting this hypothesis is large, and allows a better understanding of previously unknown physiopathological processes. In addition, the study of this new field suggests a possible link between the long-term effects of adverse life events and changes in gene expression associated with depression. Even though investigation of epigenetics in depression is still in development, numerous studies have hypothesized epigenetic modifications as potential biomarkers for depression diagnosis. It is expected that in coming years, epigenetic profiling will allow us to predict future susceptibility and/or MDD onset, improve diagnosis and to achieve a superior understanding on depression pathophysiology. Furthermore, epigenetics appears to be important for the mechanism of action of antidepressants. Future perspectives will aim to detect epigenetics modifications following AD therapy, which in turn will permit identifying new therapeutic targets based on epigenetics modifications for depressive diseases, ultimately helpful for monitoring treated patients. Acknowledgments This study was supported by the Fund for Innovation and Competitiveness (FIC) of the Chilean Ministry of Economy, Development and Tourism, through the Millennium Scientific Initiative, Grant No. IS130005 and National Fund for Scientific and Technological Development (FONDECYT No. 1130810 and No. 1150166). Author Contributions Kathleen Saavedra and Luis A. Salazar designed the study. Kathleen Saavedra and Ana María Molina-Márquez extracted the data. Kathleen Saavedra wrote the manuscript. Nicolás Saavedra and Tomás Zambrano revised the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Demanding conditions in utero or during the first years can elevate the risk of both neurological and psychiatric disorders, possibly by mechanisms involving epigenetic dysregulation. Figure 2 Epigenetic alterations in major depressive disorder. Evidence from human and animal observations indicates that environmental stimuli, such as stressful experiences, are associated with a deregulation of epigenetic modifications. These modifications include deregulation in profiles of hypermethylated genes, miRNA profiles and histone modifications. In addition, it has been reported that antidepressant treatment may act via modulation of these epigenetic modifications, particularly histone modification. Likewise, during recessive episodes, epigenetic modification described in MDD patients can return to their baseline state (control patients without MDD). ijms-17-01279-t001_Table 1Table 1 DNA methylation studies in patients with depressive disorders. Reference Sample Characteristics Study Tissue Diagnosis Platform Gene Associated Potential Relevance of Gene in Depressive Disorders Córdoba-Palomera et al. [49] 2015 17 MZ pairs Caucasian Spanish adult twins Genome-wide DNA methylation Peripheral blood Anxious or depressive disorder Illumina Infinium HumanMethylation450 Beadchip WDR26 Prospective blood transcriptomic marker for depression [11,50]. CACNA1C Susceptibility factor for depressive psychopathology [11]. Methylation changes have been associated with risk factors for depressive disorders [51,52]. MAPK11 Associated with depression phenotypes [53]. Sabunciyan et al. [54] 2012 39 individuals with MDD from Stanley Medical Research Institute Genome-wide DNA methylation Post-mortem frontal cortex MDD Comprehensive High-throughput Arrays for Relative Methylation (CHARM) PRIMA1 Encodes a protein that functions to organize AChE into tetramers, and to anchor AChE to neural cell membranes [55,56]. Numata et al. [57] 2015 29 Medication-free patients with MDD Genome-wide DNA methylation Peripheral leukocytes MDD Infinium HumanMethylation450 BeadChips CAPRIN1 Potential blood marker of major depressive disorder [58]. CITED2 Differentially expressed in the mood disorder, associated with neurological or psychiatric diseases [40]. DGKH Risk gene for bipolar disorder [59,66]. Januar et al. [65] 2015 183 patients with MDD >65 years-old High-throughput DNA methylation profiling Buccal tissue MDD Sequenom MassARRAY BDNF Promotes the proliferation, differentiation and survival of neurons, crucial for neural plasticity and cognitive function [64]. Potential biomarker of depression [65]. Nieratschker et al. [38] 2014 8 mothers and their infants with prenatal stressed conditions. 9 pregnant rats with prenatal stressed conditions Genome-wide association Peripheral leukocytes and refrontal cortex of adult rats MDD Methylated DNA immunoprecipitation (MeDIP) and pyrosequencing MORC1 Candidate gene for major depressive disorder related to early life stress in rodents, primates and humans [38]. Evokes a depression-like phenotype in mice [39]. Davies et al. [67] 2014 50 monozygotic twin pairs from the UK and Australia discordant for depression Genome-wide DNA methylation Whole blood and brain tissue samples MDD MeDIP-Sequencing ZBTB20 Important for the hormonal hippocampal function, crucial for the regionalization and volume of archicortex, playing a role in depression [68,69]. MDD: Major depressive disorder; MZ: Monozygotic. ijms-17-01279-t002_Table 2Table 2 Histone modification studies in patients with depressive disorders. Reference Sample Characteristics Tissue Diagnosis Platform Epigenetic Modification Evaluated Gene and Histone Modification Associated Main Findings Cruceanu et al. [80] 2013 Individuals with bipolar disorder type I (n = 13) or MDD (n = 18) and controls (n = 14) with no psychiatric history Post-mortem prefrontal cortex (PCF) from Broadman Area (BA) 10 BD or MDD Chromatin immunoprecipitation (ChIP) and Quantitative real-time PCR Histone modification SYN2 H3K4me3 H3K4me3 increase in MDD patients and correlated with gene expression of SYN2 [80]. Covington et al. [71] 2009 C57BL/6J male mice with chronic social defeat stress (n = 6) and control mice (n = 10). Patients depress postmortem (n = 8) Brain tissue Depression Immunohistochemistry, Western blot and Illumina MouseWG-8 V2.0 array Histone modification H3K14ac Transiently decreased and then stably increased of H3K14ac in the NAc of mice after chronic social defeat stress, correlated with a reduction in HDAC2 levels [71]. Hobara et al. [78] 2010 Mood disorder patients in a depressive and remissive state Peripheral white cells MDD and BD Quantitative real-time PCR Expression of HDACs HDAC2 and HDAC5 Gene expression of HDAC2 and HDAC5 were significantly increased in MDD patients in depressive state compare to controls subjects, while during remissive state, HDACs expression was comparable to controls subjects, suggesting a state-dependent alteration [78]. Iga et al. [79] 2007 Patients diagnosed with MDD according to DSM-IV (n = 25) and controls (n = 25) Peripheral leucocytes MDD Quantitative real-time PCR Expression of HDACs HDAC5 HDAC5 mRNA levels were significantly higher in drug-free depressive patients than controls [79]. Renthal et al. [77] 2007 Mice with chronic social defeat stress Brain tissue Depression Immnunohistochemistry, ChIP and microarray Histone modification and expression of HDACs HDAC5 HDAC5 expression was decrease in a model with social defeat stress, imipramine treatment increased HDAC5 expression [77]. BD: Bipolar disorder; DSM-IV: Diagnostic and Statistical Manual of Mental Disorders IV. ijms-17-01279-t003_Table 3Table 3 microRNAs (miRs) studies in patients with depressive disorders. Reference Sample Characteristics Tissue Diagnosis Platform miRNAs Associated Main Findings Uchida et al. [94] 2008 SH-SY5Y cells and Male rats Fisher 344 (F344) and Sprague-Dawley (SD) control with repeated restraint stress Neuron cell lines Hypothalamic paraventricular nucleus ------ ----- miR-18a Overexpressed in repeated restraint stress model. Its expression inhibits translation of the glucocorticoid receptor in neuron cell culture. Vreugdenhil et al. [97,111] 2009 NS1 cells Neuron cell lines ------ Luciferase reporter assay miR-18a and miR-124a miR-18a and miR-124a decrease protein expression of glucocorticoid receptor by luciferase reporter assay in NS1 cells. Caputo et al. [101] 2011 HeLa cells Cervix epithelial cell line Schizophrenia Luciferase reporter assay miR-132 and miR-182 These miRNAs regulate the expression of BDNF by Allele-Specific Binding [101]. Smalheiser et al. [98] 2012 Antidepressant-free depressed suicide (n = 18) and well-matched non-psychiatric control subjects (n = 17) Tissue, prefrontal cortex (Brodmann Area 9) Depression PCR miltiplex miR-142-5p, miR-137, miR-489, miR-148b, miR-101, miR-324-5p, miR-301a, miR-146a, miR-335, miR-494, miR-20b, miR-376a*, miR-190, miR-155, miR-660, miR-130a, miR-27a, miR-497, miR-10a, miR-20a, miR-142-3p miRs significantly downregulated in the prefrontal cortex of depressed patients compared with normal controls, many of them implicated in cellular growth and differentiation and some of them showed high synaptic enrichment [98,99]. Belzeaux et al. [96] 2012 16 severe MDE patients and 13 matched controls Peripheral blood mononuclear cells Major depressive episode Microarray SurePrint G3 Human GE 8 x 60 K miR-107, miR-133a, miR-148a, miR-200c, miR-381, miR-425-3p, miR-494, miR-517b, miR-579, miR-589, miR-636, miR-652, miR-941, miR-1243 miRs significantly deregulated between MDE patients and controls. These miRs help finding a gene combination useful to predict treatment response [96]. Bocchio-Chiavetto et al. [100] 2013 10 patients with MD, the sample was extracted before and after treatment Blood MDD TaqMan Array Human MicroRNA A + B Cards Set v3.0 UP: miR-130b*, miR-505*, miR-29b-2*, miR-26b, miR-22*, miR-26a, miR-664, miR-494, let-7d, let-7g, let-7e, let-7f, miR-629, miR-106b*, miR-103, miR-191, miR-128, miR-502-3p, miR-374b, miR-132, miR-30d, miR-500, miR-589, miR-183, miR-574-3p, miR-140-3p, miR-335, miR-361-5p. DOWN: miR-34c-5p and miR-770-5p Associated with neuronal brain function, such as neuroactive ligand–receptor interaction, axon guidance, long-term potentiation and depression [100]. Li et al. [33] 2013 40 patients and 40 healthy controls Serum MDD Real time PCR miR-132 and miR-182 The expression of these miRs was negatively correlated with BDNF expression [33]. Fan et al. [112] 2014 81 MDD patients and 46 healthy controls Peripheral blood mononuclear cells MDD Affymetrix miRNA 3.0 array miRNA-26b, miRNA-1972, miRNA-4485, miRNA-4498, and miRNA-4743 Overexpressed in MDD patients, and would regulate pathways associated with nervous system and brain functions [112]. Wan et al. [113] 2015 1° cohort: 6 depressed and 6 control patients. 2° cohort: 32 MDD patients and 21 healthy individuals Peripheral blood mononuclear cells MDD microRNA PCR Panel (V3.M) let-7d-3p, miR-34a-5p, miR-221-3p, miR-451a Potential MDD biomarkers [113]. Wang et al. [114] 2015 169 patients and 52 controls Plasma Depression Serum/Plasma Focus microRNA PCR Panel miR-144-5p miR-144-5p levels are associated with depressive symptoms, and the detection of this miR in plasma could be a potential peripheral biomarker for pathologic processes related to depression [114]. MDE: Major depressive episode. ijms-17-01279-t004_Table 4Table 4 Clinical trials of antidepressant treatment associated epigenetic modifications. Study ClinicalTrials.gov Identifier Status Phase Aims Intervention Condition Publications Paliperidone and lithium in the treatment of suicidality—treatment indication and epigenetic regulation (AFSP) NCT01134731 Completed Phase 4 To identify an efficient pharmacotherapy for the acute management of suicidality and the epigenetic regulation associated with the treatment. Paliperidone and lithium MDD Suicidality Not provided Epigenetic regulation of brain-derived neurotrophic factor (BDNF) in major depression NCT01182103 Completed ----- To detect the associations between BDNF, DNA methylation, histone modification, depressive symptoms, suicidal behavior and antidepressant responses in MDD patients, check the correlation between blood BDNF protein and RNA and BDNF rs6265 gene, and discuss the possible mechanisms of epigenetic regulation of BDNF in Taiwanese MDD patients. ----- MDD Not provided A neuroimaging and epigenetic investigation of antidepressants in depression NCT00703742 Completed ----- To find out the structural or functional effects of SSRI in MDD; to explore the DNA methylation status in depression; to find special abnormalities in depression secondary to other disease (autoimmune disease like systemic lupus erythematosus). Escitalopram Depression secondary to other disease [128,129] miRNAs, suicide, and ketamine—plasma exosomal microRNAs as novel biomarkers for suicidality and treatment outcome NCT02418195 Recruiting participants Phase 2 To examine whether neural-derived exosomal miRNAs are differentially expressed that are specific to suicidal ideation or behavior, and which by affecting specific miRNA targets and pathways, are associated with suicidal behavior and response to ketamine. ketamine MDD Not provided Add-On Study of MSI-195 (S-adenosyl-l-methionine, SAMe) for patients with major depressive disorder (MDD) NCT01912196 Ongoing Phase 2 To determine the efficacy and safety of 800 mg MSI-195 in reducing symptoms of depression in Major Depressive Disorder (MDD) patients with inadequate response to current antidepressant therapy. MSI-195 and Placebo MDD Not provided ==== Refs References 1. Smith K. Mental health: A world of depression Nature 2014 515 181 10.1038/515180a 25391942 2. Whiteford H.A. Degenhardt L. Rehm J. Baxter A.J. Ferrari A.J. Erskine H.E. Charlson F.J. Norman R.E. Flaxman A.D. Johns N. Global burden of disease attributable to mental and substance use disorders: Findings from the global burden of disease study 2010 Lancet 2013 382 1575 1586 10.1016/S0140-6736(13)61611-6 23993280 3. WHO Mental Health and Older Adults WHO Ginebra, Switzerland 2013 4. Sun H. Kennedy P.J. Nestler E.J. Epigenetics of the depressed brain: Role of histone acetylation and methylation Neuropsychopharmacology 2013 38 124 137 10.1038/npp.2012.73 22692567 5. Uher R. Payne J.L. Pavlova B. Perlis R.H. Major depressive disorder in dsm-5: Implications for clinical practice and research of changes from dsm-iv Depress Anxiety 2014 31 459 471 10.1002/da.22217 24272961 6. Nemeroff C.B. Weinberger D. Rutter M. MacMillan H.L. Bryant R.A. Wessely S. Stein D.J. Pariante C.M. Seemuller F. Berk M. Dsm-5: A collection of psychiatrist views on the changes, controversies, and future directions BMC Med. 2013 11 1279 10.1186/1741-7015-11-202 24229007 7. Sullivan P.F. Neale M.C. Kendler K.S. Genetic epidemiology of major depression: Review and meta-analysis Am. J. Psychiatry 2000 157 1552 1562 10.1176/appi.ajp.157.10.1552 11007705 8. McGuffin P. Cohen S. Knight J. Homing in on depression genes Am. J. Psychiatry 2007 164 195 197 10.1176/ajp.2007.164.2.195 17267777 9. Menke A. Klengel T. Binder E.B. Epigenetics, depression and antidepressant treatment Curr. Pharm. Des. 2012 18 5879 5889 10.2174/138161212803523590 22681167 10. Lee P.H. Perlis R.H. Jung J.Y. Byrne E.M. Rueckert E. Siburian R. Haddad S. Mayerfeld C.E. Heath A.C. Pergadia M.L. Multi-locus genome-wide association analysis supports the role of glutamatergic synaptic transmission in the etiology of major depressive disorder Transl. Psychiatry 2012 2 e184 10.1038/tp.2012.95 23149448 11. Wray N.R. Pergadia M.L. Blackwood D.H. Penninx B.W. Gordon S.D. Nyholt D.R. Ripke S. MacIntyre D.J. McGhee K.A. Maclean A.W. Genome-wide association study of major depressive disorder: New results, meta-analysis, and lessons learned Mol. Psychiatry 2012 17 36 48 10.1038/mp.2010.109 21042317 12. Major Depressive Disorder Working Group of the Psychiatric GWAS Consortium Ripke S. Wray N.R. Lewis C.M. Hamilton S.P. Weissman M.M. Breen G. Byrne E.M. Blackwood D.H. Boomsma D.I. A mega-analysis of genome-wide association studies for major depressive disorder Mol. Psychiatry 2013 18 497 511 22472876 13. Mak K.K. Kong W.Y. Mak A. Sharma V.K. Ho R.C. Polymorphisms of the serotonin transporter gene and post-stroke depression: A meta-analysis J. Neurol. Neurosurg. Psychiatry 2013 84 322 328 10.1136/jnnp-2012-303791 23236014 14. Phillips J.L. Batten L.A. Tremblay P. Aldosary F. Du L. Blier P. Impact of monoamine-related gene polymorphisms on hippocampal volume in treatment-resistant depression Acta Neuropsychiatr. 2015 27 353 361 10.1017/neu.2015.25 25990886 15. Brummett B.H. Krystal A.D. Siegler I.C. Kuhn C. Surwit R.S. Zuchner S. Ashley-Koch A. Barefoot J.C. Williams R.B. Associations of a regulatory polymorphism of monoamine oxidase—A gene promoter (maoa-uvntr) with symptoms of depression and sleep quality Psychosom. Med. 2007 69 396 401 10.1097/PSY.0b013e31806d040b 17585061 16. Tadic A. Rujescu D. Szegedi A. Giegling I. Singer P. Moller H.J. Dahmen N. Association of a maoa gene variant with generalized anxiety disorder, but not with panic disorder or major depression Am. J. Med. Genet. B Neuropsychiatr. Genet. 2003 117B 1 6 10.1002/ajmg.b.10013 12555227 17. Tadic A. Muller M.J. Rujescu D. Kohnen R. Stassen H.H. Dahmen N. Szegedi A. The maoa t941g polymorphism and short-term treatment response to mirtazapine and paroxetine in major depression Am. J. Med. Genet. B Neuropsychiatr. Genet. 2007 144B 325 331 10.1002/ajmg.b.30462 17192957 18. Dannlowski U. Ohrmann P. Konrad C. Domschke K. Bauer J. Kugel H. Hohoff C. Schoning S. Kersting A. Baune B.T. Reduced amygdala-prefrontal coupling in major depression: Association with maoa genotype and illness severity Int. J. Neuropsychopharmacol. 2009 12 11 22 10.1017/S1461145708008973 18544183 19. Goenjian A.K. Noble E.P. Steinberg A.M. Walling D.P. Stepanyan S.T. Dandekar S. Bailey J.N. Association of comt and tph-2 genes with dsm-5 based ptsd symptoms J. Affect. Disord. 2014 172C 472 478 10.1016/j.jad.2014.10.034 25451452 20. Han L. Nielsen D.A. Rosenthal N.E. Jefferson K. Kaye W. Murphy D. Altemus M. Humphries J. Cassano G. Rotondo A. No coding variant of the tryptophan hydroxylase gene detected in seasonal affective disorder, obsessive-compulsive disorder, anorexia nervosa, and alcoholism Biol. Psychiatry 1999 45 615 619 10.1016/S0006-3223(98)00122-X 10088048 21. Goenjian A.K. Bailey J.N. Walling D.P. Steinberg A.M. Schmidt D. Dandekar U. Noble E.P. Association of tph1, tph2, and 5httlpr with ptsd and depressive symptoms J. Affect. Disord. 2012 140 244 252 10.1016/j.jad.2012.02.015 22483952 22. Kendler K.S. Gardner C.O. Prescott C.A. Toward a comprehensive developmental model for major depression in men Am. J. Psychiatry 2006 163 115 124 10.1176/appi.ajp.163.1.115 16390898 23. Kendler K.S. Gardner C.O. Prescott C.A. Toward a comprehensive developmental model for major depression in women Am. J. Psychiatry 2002 159 1133 1145 10.1176/appi.ajp.159.7.1133 12091191 24. Dalton V.S. Kolshus E. McLoughlin D.M. Epigenetics and depression: Return of the repressed J. Affect. Disord. 2014 155 1 12 10.1016/j.jad.2013.10.028 24238955 25. Lolak S. Suwannarat P. Lipsky R.H. Epigenetics of depression Prog. Mol. Biol. Transl. Sci. 2014 128 103 137 25410543 26. Mateus-Pinheiro A. Pinto L. Sousa N. Epigenetic (de)regulation of adult hippocampal neurogenesis: Implications for depression Clin. Epigenet. 2011 3 5 10.1186/1868-7083-3-5 22414227 27. Urdinguio R.G. Sanchez-Mut J.V. Esteller M. Epigenetic mechanisms in neurological diseases: Genes, syndromes, and therapies Lancet Neurol. 2009 8 1056 1072 10.1016/S1474-4422(09)70262-5 19833297 28. Wu C. Morris J.R. Genes, genetics, and epigenetics: A correspondence Science 2001 293 1103 1105 10.1126/science.293.5532.1103 11498582 29. Portela A. Esteller M. Epigenetic modifications and human disease Nat. Biotechnol. 2010 28 1057 1068 10.1038/nbt.1685 20944598 30. Nestler E.J. Pena C.J. Kundakovic M. Mitchell A. Akbarian S. Epigenetic basis of mental illness Neuroscientist 2015 10.1177/1073858415608147 26450593 31. Nestler E.J. Epigenetic mechanisms of depression JAMA Psychiatry 2014 71 454 456 10.1001/jamapsychiatry.2013.4291 24499927 32. Cogswell J.P. Ward J. Taylor I.A. Waters M. Shi Y. Cannon B. Kelnar K. Kemppainen J. Brown D. Chen C. Identification of mirna changes in alzheimer‘s disease brain and csf yields putative biomarkers and insights into disease pathways J. Alzheimers Dis. 2008 14 27 41 18525125 33. Li Y.J. Xu M. Gao Z.H. Wang Y.Q. Yue Z. Zhang Y.X. Li X.X. Zhang C. Xie S.Y. Wang P.Y. Alterations of serum levels of bdnf-related mirnas in patients with depression PLoS ONE 2013 8 1279 10.1371/journal.pone.0063648 23704927 34. Weber M. Schubeler D. Genomic patterns of DNA methylation: Targets and function of an epigenetic mark Curr. Opin. Cell Biol. 2007 19 273 280 10.1016/j.ceb.2007.04.011 17466503 35. Barua S. Junaid M.A. Lifestyle, pregnancy and epigenetic effects Epigenomics 2015 7 85 102 10.2217/epi.14.71 25687469 36. Dulawa S.C. Epigenetic programing of depression during gestation Bioessays 2014 36 353 358 10.1002/bies.201300089 24446085 37. Babenko O. Kovalchuk I. Metz G.A. Stress-induced perinatal and transgenerational epigenetic programming of brain development and mental health Neurosci. Biobehav. Rev. 2015 48 70 91 10.1016/j.neubiorev.2014.11.013 25464029 38. Nieratschker V. Massart R. Gilles M. Luoni A. Suderman M.J. Krumm B. Meier S. Witt S.H. Nothen M.M. Suomi S.J. Morc1 exhibits cross-species differential methylation in association with early life stress as well as genome-wide association with mdd Transl. Psychiatry 2014 4 e429 10.1038/tp.2014.75 25158004 39. Schmidt M. Brandwein C. Luoni A. Sandrini P. Calzoni T. Deuschle M. Cirulli F. Riva M.A. Gass P. Morc1 knockout evokes a depression-like phenotype in mice Behav. Brain Res. 2015 296 7 14 10.1016/j.bbr.2015.08.005 26275923 40. Savitz J. Frank M.B. Victor T. Bebak M. Marino J.H. Bellgowan P.S. McKinney B.A. Bodurka J. Kent Teague T. Drevets W.C. Inflammation and neurological disease-related genes are differentially expressed in depressed patients with mood disorders and correlate with morphometric and functional imaging abnormalities Brain Behav. Immun. 2013 31 161 171 10.1016/j.bbi.2012.10.007 23064081 41. Johnson W. Turkheimer E. Gottesman I.I. Bouchard T.J. Jr. Beyond heritability: Twin studies in behavioral research Curr. Dir. Psychol. Sci. 2010 18 217 220 10.1111/j.1467-8721.2009.01639.x 20625474 42. Boomsma D. Busjahn A. Peltonen L. Classical twin studies and beyond Nat. Rev. Genet. 2002 3 872 882 10.1038/nrg932 12415317 43. Van Dongen J. Slagboom P.E. Draisma H.H. Martin N.G. Boomsma D.I. The continuing value of twin studies in the omics era Nat. Rev. Genet. 2012 13 640 653 10.1038/nrg3243 22847273 44. Tan Q. Christiansen L. von Bornemann Hjelmborg J. Christensen K. Twin methodology in epigenetic studies J. Exp. Biol. 2015 218 134 139 10.1242/jeb.107151 25568460 45. Castillo-Fernandez J.E. Spector T.D. Bell J.T. Epigenetics of discordant monozygotic twins: Implications for disease Genome Med. 2014 6 60 10.1186/s13073-014-0060-z 25484923 46. Kaminsky Z.A. Tang T. Wang S.C. Ptak C. Oh G.H. Wong A.H. Feldcamp L.A. Virtanen C. Halfvarson J. Tysk C. DNA methylation profiles in monozygotic and dizygotic twins Nat. Genet. 2009 41 240 245 10.1038/ng.286 19151718 47. Haque F.N. Gottesman I.I. Wong A.H. Not really identical: Epigenetic differences in monozygotic twins and implications for twin studies in psychiatry Am. J. Med. Genet. C Semin. Med. Genet. 2009 151C 136 141 10.1002/ajmg.c.30206 19378334 48. Silva S. Martins Y. Matias A. Blickstein I. Why are monozygotic twins different? J. Perinat. Med. 2011 39 195 202 10.1515/jpm.2010.140 21142845 49. Cordova-Palomera A. Fatjo-Vilas M. Gasto C. Navarro V. Krebs M.O. Fananas L. Genome-wide methylation study on depression: Differential methylation and variable methylation in monozygotic twins Transl. Psychiatry 2015 5 e557 10.1038/tp.2015.49 25918994 50. Pajer K. Andrus B.M. Gardner W. Lourie A. Strange B. Campo J. Bridge J. Blizinsky K. Dennis K. Vedell P. Discovery of blood transcriptomic markers for depression in animal models and pilot validation in subjects with early-onset major depression Transl. Psychiatry 2012 2 e101 10.1038/tp.2012.26 22832901 51. Oh J.E. Chambwe N. Klein S. Gal J. Andrews S. Gleason G. Shaknovich R. Melnick A. Campagne F. Toth M. Differential gene body methylation and reduced expression of cell adhesion and neurotransmitter receptor genes in adverse maternal environment Transl. Psychiatry 2013 3 e218 10.1038/tp.2012.130 23340501 52. Heim C. Binder E.B. Current research trends in early life stress and depression: Review of human studies on sensitive periods, gene-environment interactions, and epigenetics Exp. Neurol. 2012 233 102 111 10.1016/j.expneurol.2011.10.032 22101006 53. Bruchas M.R. Schindler A.G. Shankar H. Messinger D.I. Miyatake M. Land B.B. Lemos J.C. Hagan C.E. Neumaier J.F. Quintana A. Selective p38alpha mapk deletion in serotonergic neurons produces stress resilience in models of depression and addiction Neuron 2011 71 498 511 10.1016/j.neuron.2011.06.011 21835346 54. Sabunciyan S. Aryee M.J. Irizarry R.A. Rongione M. Webster M.J. Kaufman W.E. Murakami P. Lessard A. Yolken R.H. Feinberg A.P. Genome-wide DNA methylation scan in major depressive disorder PLoS ONE 2012 7 1279 10.1371/journal.pone.0034451 22511943 55. Xie H.Q. Siow N.L. Peng H.B. Massoulie J. Tsim K.W. Regulation of prima: Membrane anchor of acetylcholinesterase (ache) in neuron and muscle Chem. Biol. Interact. 2005 432 157 158 10.1016/j.cbi.2005.10.093 56. Xie H.Q. Choi R.C. Leung K.W. Chen V.P. Chu G.K. Tsim K.W. Transcriptional regulation of proline-rich membrane anchor (prima) of globular form acetylcholinesterase in neuron: An inductive effect of neuron differentiation Brain Res. 2009 1265 13 23 10.1016/j.brainres.2009.01.065 19368807 57. Numata S. Ishii K. Tajima A. Iga J. Kinoshita M. Watanabe S. Umehara H. Fuchikami M. Okada S. Boku S. Blood diagnostic biomarkers for major depressive disorder using multiplex DNA methylation profiles: Discovery and validation Epigenetics 2015 10 135 141 10.1080/15592294.2014.1003743 25587773 58. Spijker S. van Zanten J.S. de Jong S. Penninx B.W. van Dyck R. Zitman F.G. Smit J.H. Ylstra B. Smit A.B. Hoogendijk W.J. Stimulated gene expression profiles as a blood marker of major depressive disorder Biol. Psychiatry 2010 68 179 186 10.1016/j.biopsych.2010.03.017 20471630 59. Weber H. Kittel-Schneider S. Gessner A. Domschke K. Neuner M. Jacob C.P. Buttenschon H.N. Boreatti-Hummer A. Volkert J. Herterich S. Cross-disorder analysis of bipolar risk genes: Further evidence of dgkh as a risk gene for bipolar disorder, but also unipolar depression and adult adhd Neuropsychopharmacology 2011 36 2076 2085 10.1038/npp.2011.98 21654738 60. Ronai Z. Kovacs-Nagy R. Szantai E. Elek Z. Sasvari-Szekely M. Faludi G. Benkovits J. Rethelyi J.M. Szekely A. Glycogen synthase kinase 3 beta gene structural variants as possible risk factors of bipolar depression Am. J. Med. Genet. B Neuropsychiatr. Genet. 2014 165B 217 222 10.1002/ajmg.b.32223 24677591 61. Yang C. Xu Y. Sun N. Ren Y. Liu Z. Cao X. Zhang K. The combined effects of the bdnf and gsk3b genes modulate the relationship between negative life events and major depressive disorder Brain Res. 2010 1355 1 6 10.1016/j.brainres.2010.07.079 20682308 62. Anacker C. Cattaneo A. Musaelyan K. Zunszain P.A. Horowitz M. Molteni R. Luoni A. Calabrese F. Tansey K. Gennarelli M. Role for the kinase sgk1 in stress, depression, and glucocorticoid effects on hippocampal neurogenesis Proc. Natl. Acad. Sci. USA 2013 110 8708 8713 10.1073/pnas.1300886110 23650397 63. Stenz L. Zewdie S. Laforge-Escarra T. Prados J. La Harpe R. Dayer A. Paoloni-Giacobino A. Perroud N. Aubry J.M. Bdnf promoter i methylation correlates between post-mortem human peripheral and brain tissues Neurosci. Res. 2015 91 1 7 10.1016/j.neures.2014.10.003 25450314 64. Autry A.E. Monteggia L.M. Brain-derived neurotrophic factor and neuropsychiatric disorders Pharmacol. Rev. 2012 64 238 258 10.1124/pr.111.005108 22407616 65. Januar V. Ancelin M.L. Ritchie K. Saffery R. Ryan J. Bdnf promoter methylation and genetic variation in late-life depression Transl. Psychiatry 2015 5 e619 10.1038/tp.2015.114 26285129 66. Barnett J.H. Smoller J.W. The genetics of bipolar disorder Neuroscience 2009 164 331 343 10.1016/j.neuroscience.2009.03.080 19358880 67. Davies M.N. Krause L. Bell J.T. Gao F. Ward K.J. Wu H. Lu H. Liu Y. Tsai P.C. Collier D.A. Hypermethylation in the zbtb20 gene is associated with major depressive disorder Genome Biol. 2014 15 R56 10.1186/gb-2014-15-4-r56 24694013 68. Mitchelmore C. Kjaerulff K.M. Pedersen H.C. Nielsen J.V. Rasmussen T.E. Fisker M.F. Finsen B. Pedersen K.M. Jensen N.A. Characterization of two novel nuclear btb/poz domain zinc finger isoforms: Association with differentiation of hippocampal neurons, cerebellar granule cells, and macroglia J. Biol. Chem. 2002 277 7598 7609 10.1074/jbc.M110023200 11744704 69. Rosenthal E.H. Tonchev A.B. Stoykova A. Chowdhury K. Regulation of archicortical arealization by the transcription factor zbtb20 Hippocampus 2012 22 2144 2156 10.1002/hipo.22035 22689450 70. Bagot R.C. Labonte B. Pena C.J. Nestler E.J. Epigenetic signaling in psychiatric disorders: Stress and depression Dialogues Clin. Neurosci. 2014 16 281 295 25364280 71. Covington H.E. 3rd Maze I. LaPlant Q.C. Vialou V.F. Ohnishi Y.N. Berton O. Fass D.M. Renthal W. Rush A.J. 3rd Wu E.Y. Antidepressant actions of histone deacetylase inhibitors J. Neurosci. 2009 29 11451 11460 10.1523/JNEUROSCI.1758-09.2009 19759294 72. Schroeder F.A. Lin C.L. Crusio W.E. Akbarian S. Antidepressant-like effects of the histone deacetylase inhibitor, sodium butyrate, in the mouse Biol. Psychiatry 2007 62 55 64 10.1016/j.biopsych.2006.06.036 16945350 73. Tsankova N.M. Berton O. Renthal W. Kumar A. Neve R.L. Nestler E.J. Sustained hippocampal chromatin regulation in a mouse model of depression and antidepressant action Nat. Neurosci. 2006 9 519 525 10.1038/nn1659 16501568 74. Weaver I.C. Cervoni N. Champagne F.A. D’Alessio A.C. Sharma S. Seckl J.R. Dymov S. Szyf M. Meaney M.J. Epigenetic programming by maternal behavior Nat. Neurosci. 2004 7 847 854 10.1038/nn1276 15220929 75. Arent C.O. Valvassori S.S. Fries G.R. Stertz L. Ferreira C.L. Lopes-Borges J. Mariot E. Varela R.B. Ornell F. Kapczinski F. Neuroanatomical profile of antimaniac effects of histone deacetylases inhibitors Mol. Neurobiol. 2011 43 207 214 10.1007/s12035-011-8178-0 21424678 76. Machado-Vieira R. Ibrahim L. Zarate C.A. Jr. Histone deacetylases and mood disorders: Epigenetic programming in gene-environment interactions CNS Neurosci. Ther. 2011 17 699 704 10.1111/j.1755-5949.2010.00203.x 20961400 77. Renthal W. Maze I. Krishnan V. Covington H.E. 3rd Xiao G. Kumar A. Russo S.J. Graham A. Tsankova N. Kippin T.E. Histone deacetylase 5 epigenetically controls behavioral adaptations to chronic emotional stimuli Neuron 2007 56 517 529 10.1016/j.neuron.2007.09.032 17988634 78. Hobara T. Uchida S. Otsuki K. Matsubara T. Funato H. Matsuo K. Suetsugi M. Watanabe Y. Altered gene expression of histone deacetylases in mood disorder patients J. Psychiatr. Res. 2010 44 263 270 10.1016/j.jpsychires.2009.08.015 19767015 79. Iga J. Ueno S. Yamauchi K. Numata S. Kinouchi S. Tayoshi-Shibuya S. Song H. Ohmori T. Altered hdac5 and creb mrna expressions in the peripheral leukocytes of major depression Prog. Neuropsychopharmacol. Biol. Psychiatry 2007 31 628 632 10.1016/j.pnpbp.2006.12.014 17258370 80. Cruceanu C. Alda M. Nagy C. Freemantle E. Rouleau G.A. Turecki G. H3k4 tri-methylation in synapsin genes leads to different expression patterns in bipolar disorder and major depression Int. J. Neuropsychopharmacol. 2013 16 289 299 10.1017/S1461145712000363 22571925 81. Akbarian S. Huang H.S. Epigenetic regulation in human brain-focus on histone lysine methylation Biol. Psychiatry 2009 65 198 203 10.1016/j.biopsych.2008.08.015 18814864 82. Berger S.L. The complex language of chromatin regulation during transcription Nature 2007 447 407 412 10.1038/nature05915 17522673 83. Bannister A.J. Kouzarides T. Regulation of chromatin by histone modifications Cell Res. 2011 21 381 395 10.1038/cr.2011.22 21321607 84. Kouzarides T. Chromatin modifications and their function Cell 2007 128 693 705 10.1016/j.cell.2007.02.005 17320507 85. Porton B. Kao H.T. Greengard P. Characterization of transcripts from the synapsin iii gene locus J. Neurochem. 1999 73 2266 2271 10.1046/j.1471-4159.1999.0732266.x 10582583 86. Cesca F. Baldelli P. Valtorta F. Benfenati F. The synapsins: Key actors of synapse function and plasticity Prog. Neurobiol. 2010 91 313 348 10.1016/j.pneurobio.2010.04.006 20438797 87. Fassio A. Raimondi A. Lignani G. Benfenati F. Baldelli P. Synapsins: From synapse to network hyperexcitability and epilepsy Semin. Cell Dev. Biol. 2011 22 408 415 21816229 88. He L. Hannon G.J. Micrornas: Small rnas with a big role in gene regulation Nat. Rev. Genet. 2004 5 522 531 15211354 89. Mendell J.T. Micrornas: Critical regulators of development, cellular physiology and malignancy Cell Cycle 2005 4 1179 1184 16096373 90. Esteller M. Non-coding rnas in human disease Nat. Rev. Genet. 2011 12 861 874 10.1038/nrg3074 22094949 91. Carthew R.W. Sontheimer E.J. Origins and mechanisms of mirnas and sirnas Cell 2009 136 642 655 10.1016/j.cell.2009.01.035 19239886 92. Krol J. Loedige I. Filipowicz W. The widespread regulation of microrna biogenesis, function and decay Nat. Rev. Genet. 2010 11 597 610 10.1038/nrg2843 20661255 93. Wu L. Fan J. Belasco J.G. Micrornas direct rapid deadenylation of mrna Proc. Natl. Acad. Sci. USA 2006 103 4034 4039 10.1073/pnas.0510928103 16495412 94. Uchida S. Nishida A. Hara K. Kamemoto T. Suetsugi M. Fujimoto M. Watanuki T. Wakabayashi Y. Otsuki K. McEwen B.S. Characterization of the vulnerability to repeated stress in fischer 344 rats: Possible involvement of microrna-mediated down-regulation of the glucocorticoid receptor Eur. J. Neurosci. 2008 27 2250 2261 10.1111/j.1460-9568.2008.06218.x 18445216 95. De Kloet E.R. Fitzsimons C.P. Datson N.A. Meijer O.C. Vreugdenhil E. Glucocorticoid signaling and stress-related limbic susceptibility pathway: About receptors, transcription machinery and microrna Brain Res. 2009 1293 129 141 10.1016/j.brainres.2009.03.039 19332027 96. Belzeaux R. Bergon A. Jeanjean V. Loriod B. Formisano-Treziny C. Verrier L. Loundou A. Baumstarck-Barrau K. Boyer L. Gall V. Responder and nonresponder patients exhibit different peripheral transcriptional signatures during major depressive episode Transl. Psychiatry 2012 2 e185 10.1038/tp.2012.112 23149449 97. Dwivedi Y. Evidence demonstrating role of micrornas in the etiopathology of major depression J. Chem. Neuroanat. 2011 42 142 156 10.1016/j.jchemneu.2011.04.002 21515361 98. Smalheiser N.R. Lugli G. Rizavi H.S. Torvik V.I. Turecki G. Dwivedi Y. Microrna expression is down-regulated and reorganized in prefrontal cortex of depressed suicide subjects PLoS ONE 2012 7 1279 10.1371/journal.pone.0033201 22427989 99. Lugli G. Torvik V.I. Larson J. Smalheiser N.R. Expression of micrornas and their precursors in synaptic fractions of adult mouse forebrain J. Neurochem. 2008 106 650 661 10.1111/j.1471-4159.2008.05413.x 18410515 100. Bocchio-Chiavetto L. Maffioletti E. Bettinsoli P. Giovannini C. Bignotti S. Tardito D. Corrada D. Milanesi L. Gennarelli M. Blood microrna changes in depressed patients during antidepressant treatment Eur. Neuropsychopharmacol. 2013 23 602 611 10.1016/j.euroneuro.2012.06.013 22925464 101. Caputo V. Sinibaldi L. Fiorentino A. Parisi C. Catalanotto C. Pasini A. Cogoni C. Pizzuti A. Brain derived neurotrophic factor (bdnf) expression is regulated by micrornas mir-26a and mir-26b allele-specific binding PLoS ONE 2011 6 1279 10.1371/journal.pone.0028656 22194877 102. Baraniskin A. Kuhnhenn J. Schlegel U. Maghnouj A. Zollner H. Schmiegel W. Hahn S. Schroers R. Identification of micrornas in the cerebrospinal fluid as biomarker for the diagnosis of glioma Neuro Oncol. 2012 14 29 33 10.1093/neuonc/nor169 21937590 103. Baraniskin A. Kuhnhenn J. Schlegel U. Chan A. Deckert M. Gold R. Maghnouj A. Zollner H. Reinacher-Schick A. Schmiegel W. Identification of micrornas in the cerebrospinal fluid as marker for primary diffuse large b-cell lymphoma of the central nervous system Blood 2011 117 3140 3146 10.1182/blood-2010-09-308684 21200023 104. Chen X. Ba Y. Ma L. Cai X. Yin Y. Wang K. Guo J. Zhang Y. Chen J. Guo X. Characterization of micrornas in serum: A novel class of biomarkers for diagnosis of cancer and other diseases Cell Res. 2008 18 997 1006 10.1038/cr.2008.282 18766170 105. Weber J.A. Baxter D.H. Zhang S. Huang D.Y. Huang K.H. Lee M.J. Galas D.J. Wang K. The microrna spectrum in 12 body fluids Clin. Chem. 2010 56 1733 1741 10.1373/clinchem.2010.147405 20847327 106. Mitchell P.S. Parkin R.K. Kroh E.M. Fritz B.R. Wyman S.K. Pogosova-Agadjanyan E.L. Peterson A. Noteboom J. O’Briant K.C. Allen A. Circulating micrornas as stable blood-based markers for cancer detection Proc. Natl. Acad. Sci. USA 2008 105 10513 10518 10.1073/pnas.0804549105 18663219 107. Jin X.F. Wu N. Wang L. Li J. Circulating micrornas: A novel class of potential biomarkers for diagnosing and prognosing central nervous system diseases Cell. Mol. Neurobiol. 2013 33 601 613 10.1007/s10571-013-9940-9 23633081 108. Gaughwin P.M. Ciesla M. Lahiri N. Tabrizi S.J. Brundin P. Bjorkqvist M. Hsa-mir-34b is a plasma-stable microrna that is elevated in pre-manifest huntington's disease Hum. Mol. Genet. 2011 20 2225 2237 10.1093/hmg/ddr111 21421997 109. Redell J.B. Moore A.N. Ward N.H. Hergenroeder G.W. Dash P.K. Human traumatic brain injury alters plasma microrna levels J. Neurotrauma 2010 27 2147 2156 10.1089/neu.2010.1481 20883153 110. Liu D.Z. Tian Y. Ander B.P. Xu H. Stamova B.S. Zhan X. Turner R.J. Jickling G. Sharp F.R. Brain and blood microrna expression profiling of ischemic stroke, intracerebral hemorrhage, and kainate seizures J. Cereb. Blood Flow Metab. 2010 30 92 101 10.1038/jcbfm.2009.186 19724284 111. Vreugdenhil E. Verissimo C.S. Mariman R. Kamphorst J.T. Barbosa J.S. Zweers T. Champagne D.L. Schouten T. Meijer O.C. de Kloet E.R. Microrna 18 and 124a down-regulate the glucocorticoid receptor: Implications for glucocorticoid responsiveness in the brain Endocrinology 2009 150 2220 2228 10.1210/en.2008-1335 19131573 112. Fan H.M. Sun X.Y. Guo W. Zhong A.F. Niu W. Zhao L. Dai Y.H. Guo Z.M. Zhang L.Y. Lu J. Differential expression of microrna in peripheral blood mononuclear cells as specific biomarker for major depressive disorder patients J. Psychiatr. Res. 2014 59 45 52 10.1016/j.jpsychires.2014.08.007 25201637 113. Wan Y. Liu Y. Wang X. Wu J. Liu K. Zhou J. Liu L. Zhang C. Identification of differential micrornas in cerebrospinal fluid and serum of patients with major depressive disorder PLoS ONE 2015 10 1279 10.1371/journal.pone.0121975 25763923 114. Wang X. Sundquist K. Hedelius A. Palmer K. Memon A.A. Sundquist J. Circulating microRNA-144–5p is associated with depressive disorders Clin. Epigenet. 2015 7 69 10.1186/s13148-015-0099-8 26199675 115. Vialou V. Feng J. Robison A.J. Nestler E.J. Epigenetic mechanisms of depression and antidepressant action Annu. Rev. Pharmacol. Toxicol. 2013 53 59 87 10.1146/annurev-pharmtox-010611-134540 23020296 116. Perisic T. Zimmermann N. Kirmeier T. Asmus M. Tuorto F. Uhr M. Holsboer F. Rein T. Zschocke J. Valproate and amitriptyline exert common and divergent influences on global and gene promoter-specific chromatin modifications in rat primary astrocytes Neuropsychopharmacology 2010 35 792 805 10.1038/npp.2009.188 19924110 117. Zimmermann N. Zschocke J. Perisic T. Yu S. Holsboer F. Rein T. Antidepressants inhibit DNA methyltransferase 1 through reducing g9a levels Biochem. J. 2012 448 93 102 10.1042/BJ20120674 22880885 118. Group U.E.R. Efficacy and safety of electroconvulsive therapy in depressive disorders: A systematic review and meta-analysis Lancet 2003 361 799 808 12642045 119. Tsankova N.M. Kumar A. Nestler E.J. Histone modifications at gene promoter regions in rat hippocampus after acute and chronic electroconvulsive seizures J. Neurosci. 2004 24 5603 5610 10.1523/JNEUROSCI.0589-04.2004 15201333 120. Heyer M.P. Kenny P.J. Microrna-mediated repression combats depression Neuron 2014 83 253 254 10.1016/j.neuron.2014.07.008 25033172 121. Baudry A. Mouillet-Richard S. Schneider B. Launay J.M. Kellermann O. Mir-16 targets the serotonin transporter: A new facet for adaptive responses to antidepressants Science 2010 329 1537 1541 10.1126/science.1193692 20847275 122. Lewis S. Small RNAs: The enigma of prozac resolved Nat. Rev. Neurosci. 2010 11 731 10.1038/nrn2941 20979321 123. Launay J.M. Mouillet-Richard S. Baudry A. Pietri M. Kellermann O. Raphe-mediated signals control the hippocampal response to sri antidepressants via mir-16 Transl. Psychiatry 2011 1 e56 10.1038/tp.2011.54 22833211 124. Lopez J.P. Lim R. Cruceanu C. Crapper L. Fasano C. Labonte B. Maussion G. Yang J.P. Yerko V. Vigneault E. Mir-1202 is a primate-specific and brain-enriched microRNA involved in major depression and antidepressant treatment Nat. Med. 2014 20 764 768 10.1038/nm.3582 24908571 125. Issler O. Haramati S. Paul E.D. Maeno H. Navon I. Zwang R. Gil S. Mayberg H.S. Dunlop B.W. Menke A. MicroRNA 135 is essential for chronic stress resiliency, antidepressant efficacy, and intact serotonergic activity Neuron 2014 83 344 360 10.1016/j.neuron.2014.05.042 24952960 126. Heller E.A. Cates H.M. Pena C.J. Sun H. Shao N. Feng J. Golden S.A. Herman J.P. Walsh J.J. Mazei-Robison M. Locus-specific epigenetic remodeling controls addiction- and depression-related behaviors Nat. Neurosci. 2014 17 1720 1727 10.1038/nn.3871 25347353 127. Sun H. Damez-Werno D.M. Scobie K.N. Shao N.Y. Dias C. Rabkin J. Koo J.W. Korb E. Bagot R.C. Ahn F.H. Acf chromatin-remodeling complex mediates stress-induced depressive-like behavior Nat. Med. 2015 21 1146 1153 10.1038/nm.3939 26390241 128. Cheng Y. Xu J. Yu H. Nie B. Li N. Luo C. Li H. Liu F. Bai Y. Shan B. Delineation of early and later adult onset depression by diffusion tensor imaging PLoS ONE 2014 9 1279 10.1371/journal.pone.0112307 25393297 129. Xu J. Cheng Y. Chai P. Lu Z. Li H. Luo C. Li X. Li L. Zhou Q. Chen B. White-matter volume reduction and the protective effect of immunosuppressive therapy in systemic lupus erythematosus patients with normal appearance by conventional magnetic resonance imaging J. Rheumatol. 2010 37 974 986 10.3899/jrheum.090967 20231206

PMC005xxxxxx/PMC5000677.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081280ijms-17-01280ArticleThe Study of Dynamic Potentials of Highly Excited Vibrational States of DCP: From Case Analysis to Comparative Study with HCP Wang Aixing 12Fang Chao 345*Liu Yibao 1Sauer Stephan P. A. Academic Editor1 Fundamental Science on Radioactive Geology and Exploration Technology Laboratory, East China University of Technology, Nanchang 330013, China; [email protected] (A.W.); [email protected] (Y.L.)2 School of Science, East China University of Technology, Nanchang 330013, China3 Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, China4 Key Laboratory of Advanced Reactor Engineering and Safety of Ministry of Education, Beijing 100084, China5 Collaborative Innovation Center of Advanced Nuclear Energy Technology, Beijing 100084, China* Correspondence: [email protected]; Tel./Fax: +86-10-6279-247422 8 2016 8 2016 17 8 128017 7 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The dynamic potentials of highly excited vibrational states of deuterated phosphaethyne (DCP) in the D–C and C–P stretching coordinates with anharmonicity and Fermi coupling are studied in this article and the results show that the D-C-P bending vibration mode has weak effects on D–C and C–P stretching modes under different Polyad numbers (P number). Furthermore, the dynamic potentials and the corresponding phase space trajectories of DCP are given, as an example, in the case of P = 30. In the end, a comparative study between deuterated phosphaethyne (DCP) and phosphaethyne (HCP) with dynamic potential is done, and it is elucidated that the uncoupled mode makes the original horizontal reversed symmetry breaking between the dynamic potential of HCP (q3) and DCP (q1), but has little effect on the vertical reversed symmetry, between the dynamic potential of HCP (q2) and DCP (q3). HCPDCPhighly excited vibrational statephase spacesymmetry of dynamic potentials ==== Body 1. Introduction Resonance coupling between the different vibrational modes of molecules, which typically increases with energy, makes triatomic molecules quite intricate [1]. The ways of studying the resonance coupling effect between the different modes in a triatomic molecule are ab initio calculations and semi-classical methods [2,3,4]. In recent years, a new semi-classical method, named dynamics potential [5,6,7], has been proposed and has been applied to study highly excited molecular vibrational states. This method could, not only verify the conclusions given by ab initio calculations, but also show visual physical pictures, including molecular isomerization [8,9], chaotic dynamics [10,11], dissociation dynamics [12], and other information. The internal interaction between D–C stretching and C–P stretching in DCP (deuterated phosphaethyne) has attracted a great deal of attention, since the information involved in the interaction is significant for understanding the mechanisms of chemical reactions. In previous articles, we analyzed the dynamic features of deuterated phosphaethyne (DCP) and phosphaethyne (HCP) using dynamic potentials [5]. Because of the drastic change of atomic masses of DCP compared with HCP, instead of the resonance between C–P stretching and D–C–P bending, a 2:1 D–C stretching and C–P stretching resonance governs the DCP spectrum. It is shown that there is dynamic symmetry between DCP and HCP systems, which is significant to analyze the features of homologous compounds. In this work, the dynamic potentials of highly excited vibrational states and phase space trajectories of DCP are studied. The effect of the D–C–P bending vibration mode on the D–C and C–P stretching modes, under different Polyad numbers, are also investigated. Finally, a comparative study between DCP and HCP is done to clarify the symmetry breaking of dynamic potentials in DCP and HCP systems with the effects of uncoupled modes, respectively. 2. The Semi-Classical Hamiltonian of the DCP The dynamic properties of DCP molecules’ highly-excited vibrational states, in the energy region 1.97 × 104–2.35 × 105 cm−1, are essential [5,9], and the corresponding Hamiltonian could be obtained as following: (1) H=ω1(n1+12)+ω2(n2+12)+ω3(n3+12)+X11(n1+12)2+X12(n1+12)(n2+1)+X13(n1+12)(n3+12)+X22(n2+1)2+X23(n2+1)(n3+12)+X33(n3+12)2+y111(n1+12)3+y112(n1+12)2(n2+1)+y113(n1+12)2(n3+12)+y122(n1+12)(n2+1)2+y123(n1+12)(n2+1)(n3+12)+y133(n1+12)(n3+12)2+y222(n2+1)3+y222(n2+1)3+y223(n2+1)2(n3+12)+y233(n2+1)(n3+12)2+y333(n3+12)3+z1111(n1+12)4+z1112(n1+12)3(n2+1)+z1222(n1+12)(n2+1)3+z1233(n1+12)(n2+1)(n3+12)2+z2222(n2+1)4+z2233(n2+1)2(n3+12)2+z2333(n2+1)(n3+12)3+z3333(n3+12)4+[k+λ1n1+λ3(n3+32)+μ11n12](a3+2a1+a32a1+) The corresponding coefficients of the DCP Hamiltonian are shown in Table 1, where subscripts 1, 2, and 3, correspond to the D–C stretching vibration mode, D–C–P bending vibration mode, and C–P stretching vibration mode, respectively. We will use n to denote the corresponding vibration mode, which will be indicated with qi in the position coordinate, and the momentum coordinate indicated with pi. ωi is the corresponding harmonic vibration coefficient, while Xij, yijm, zijmn denote the nonlinear coupling coefficients of different modes (Xij ~ coefficient of the two nonlinear coupling modes, yijm ~ coefficient of the three nonlinear coupling modes, zijmn ~ coefficient of the four nonlinear coupling modes). k, λ1, λ3, and μ11 represent the Fermi resonance strength coefficient, with regard to the quantum numbers of the three vibrational modes. Besides n2, there is another conserved action called Polyad number P = 2n1 + n3 (P number). Equation (1) is used to study the dynamic properties of highly excited vibrational states in the region of n1 ≤ 4, P ≤ 30 [9]. The coset space SU(2)/U(1) [13,14] could be used as the representing space of Hamiltonian and it could be rewritten in the coordinates (q1, p1) indicates with semi-classical representations as follows: (2) H(n2,q1,p1,P)=ω1(p12+q122+12)+ω2(n2+1)+ω3(P−(p12+q12)+12)+X11(p12+q122+12)2+X12(p12+q122+12)(n2+1)+X13(p12+q122+12)(P−(p12+q12)+12)+X22(n2+1)2+X23(n2+1)(P−(p12+q12)+12)+X33(P−(p12+q12)+12)2+y111(p12+q122+12)3+y112(p12+q122+12)2(n2+1)+y113(p12+q122+12)2(P−(p12+q12)+12)+y122(p12+q122+12)(n2+1)2+y123(p12+q122+12)(n2+1)(P−(p12+q12)+12)+y133(p12+q122+12)(P−(p12+q12)+12)2+y222(n2+1)3+y223(n2+1)2(P−(p12+q12)+12)+y233(n2+1)(P−(p12+q12)+12)2+y333(P−(p12+q12)+12)3+z1111(p12+q122+12)4+z1112(p12+q122+12)3(n2+1)+z1222(p12+q122+12)(n2+1)3+z1233(p12+q122+12)(n2+1)(P−(p12+q12)+12)2+z2222(n2+1)4+z2233(n2+1)2(P−(p12+q12)+12)2+z2333(n2+1)(P−(p12+q12)+12)3+z3333(P−(p12+q12)+12)4+[k+λ1(p12+q122）+λ3(P−(p12+q12)+32)+μ11(p12+q122）2]2(P−(p12+q12))q1 With the coordinate (q3,p3), the Hamiltonian can be written as: (3) H(n2,q3,p3,P)=ω1(P2−p32+q324+12)+ω2(n2+1)+ω3(p32+q322+12)+X11(P2−p32+q324+12)2+X12(P2−p32+q324+12)(n2+1)+X13(P2−p32+q324+12)(p32+q322+12)+X22(n2+1)2+X23(n2+1)(p32+q322+12)+X33(p32+q322+12)2+y111(P2−p32+q324+12)3+y112(P2−p32+q324+12)2(n2+1)+y113(P2−p32+q324+12)2(p32+q322+12)+y122(P2−p32+q324+12)(n2+1)2+y123(P2−p32+q324+12)(n2+1)(p32+q322+12)+y133(P2−p32+q324+12)(p32+q322+12)2+y222(n2+1)3+y223(n2+1)2(p32+q322+12)+y233(n2+1)(p32+q322+12)2+y333(p32+q322+12)3+z1111(P2−p32+q324+12)4+z1112(P2−p32+q324+12)3(n2+1)+z1222(P2−p32+q324+12)(n2+1)3+z1233(P2−p32+q324+12)(n2+1)(p32+q322+12)2+z2222(n2+1)4+z2233(n2+1)2(p32+q322+12)2+z2333(n2+1)(p32+q322+12)3+z3333(p32+q322+12)4+[k+λ1(P2−p32+q324）+λ3(p32+q322+32)+μ11(P2−p32+q324）2]P2−p32+q324(q32−p32) The semi-classical Hamiltonian, mentioned above, could further be used to obtain dynamic potentials, which are necessary for studying the dynamic nature of the DCP’s highly excited vibrational states. 3. The Dynamic Features of the Highly Excited Vibration States of DCP Two main parts would be addressed in the following: (1) the influences of bending modes to D–C and C–P stretching modes; (2) the phase space trajectories for each energy levels in the dynamic potentials when P = 30 (as a case study); and (3) the comparative study between DCP and HCP in the sense of symmetry of dynamic potential. 3.1. The Dynamic Potentials Corresponding Typical Polyad Number with Different Quantum Number n2 The case of small P number will be firstly discussed. We take P = 18 as instance and the dynamic potential is shown as Figure 1 (the rule of marking fixed points is the same with the literature [5,6,9]). Figure 1 shows that when n2 = 0,1,2,3, the dynamic potentials of q1 coordinates are simple inverse Morse potential. It is known that corresponding to a certain P, the stability of the lowest energy level in an inverse Morse potential is the worst, while that of the highest one is the best, which is totally different from the concept in general potential that the lower the energy level is, the worse the stability is. Furthermore, the shapes of dynamic potentials of q1 and q3 coordinates under different n2 are almost same, which elucidates that D-C-P bending has no effect on the stability of the highly excited vibrational states in DCP under the small P number. On the other hand, the dynamic potentials of q3 show that the three modes corresponding to the highest three energy levels are localized and this conclusion is consistent when n2 = 0,1,2,3. The dynamic potentials of q1 and q3 corresponding to different n2 is basically the same and all the fixed points are remained when n2 is different, which indicate that the effect of D-C-P bending mode has weak interaction with the two coupling modes, which are different from former studies of HOCl and HOBr systems [6,12]. Figure 2 shows the dynamic potentialins of q1 and q3 and it is shown that the results are different when P is large. For example, when P = 30, the dynamic potential of q1 becomes much more complex. There are three new fixed points emerging, [R13*], [r1] and [r1]¯ in the dynamic potential and the shape of dynamic potential of q1 becomes the combination of Morse and inverse Morse potentials. It is found that there is a phenomenon of fixed point-splitting in the dynamic potential of q3 . The original [r3] (the dynamic potentialins of q3 when P = 18 in Figure 1) becomes [R13*] and [r3] , which are similar with the results of HOBr and HOCl [6,12]. On the other hand, though the dynamic potentials of q2 and q3 when P = 30 are much more complex than the case of P = 18, but the shape of dynamic potentials of q2 and q3 remain the same when n2 = 0,1,2,3, respectively, which is consistent with the case of P = 18. Through the above study, it is found that the D–C–P bending mode weakly affects the resonant coupling of D–C and C–P stretching modes, thereby weakly affecting the dynamics features of DCP. It is shown that the geometrical shapes of the dynamic potentials and the corresponding fixed points are not sensitive to the D–C–P bending mode, but are sensitive to the P number, which are different to our previous studies [6,12]. Though the cases of P = 18 and P = 30 are shown here, these conclusions are also suitable for other cases. The reasons we address the cases of P = 18 and P = 30 are that the connotations of corresponding dynamic potentials are abundant and the shapes of dynamics potentials are typical. 3.2. The Trajectories of Phase Space Study for the Energy Levels under Specific Polyad Number (P = 30) For further quantitative analyzing the dynamic features of DCP, the representative trajectories of phase space in pi-qi for each energy level is studied when P = 30. The dynamic potentials and corresponding energy levels when P = 30 and n2 = 0 are shown in Figure 3. The trajectories of phase space for different energy levels in dynamic potentials of q1 and q3 are shown in Figure 4 and Figure 5. Based on previous studies, the envelope area of the trajectory in phase space shows the quantum environment of a series of energy levels [8,10]. In Figure 4, for L0–L13, it is found that the envelope area of the trajectory of phase space increases with the reduction of energy level because these energy levels lie in inverse Morse potential. In contrast, For L14–L15, the envelope area of the trajectory of phase space increases with the increase of energy level because these energy levels lie in Morse potential. Furthermore, because the L0 (L15) is tangential with the top (bottom) of the dynamic potential, the envelope area of the trajectory is zero. Particularly, the trajectories of L8 and L9 are divided into two separate trajectories, which show that these two energy levels are located in at a double-wells dynamic potential. The conclusions are similar in Figure 5. Because all energy levels (except L0 and L15) are in Morse potential of q3 so the envelope area of the trajectory in phase space increases with the increase of energy level. 3.3. Comparative Study between DCP and HCP with Dynamic Potential In previous work [5], it was shown that the dynamic potential of DCP in q3 coordinate is similar to the inverse of that of HCP in q2 coordinate and the dynamic potential of DCP in q1 coordinate is similar to that of HCP in q3 coordinate with −q3 transformation, which is called “dynamic symmetry”. This conclusion is available when the quantum number of the uncoupling mode (H–C stretching mode for HCP and D–C–P bending mode for DCP) is equal to 0. However, the dynamic symmetry will be broken when the quantum number of the uncoupling mode (nun, nun = n1 for HCP and nun = n2 for DCP [5]) is not equal to 0. From ergodic analysis of the P number, it is found that when nun > 0, the original horizontal reversed symmetry between the dynamic potentials of HCP (q3) and DCP (q1), mentioned in Reference [5], does not exist; however, for large P number (P > 22), the clockwise 180 °C rotation symmetry between the dynamic potentials of HCP (q3) and DCP (q1) emerge. This symmetry is not strict but it still could be recognized from the shapes of dynamic potentials and the above conclusion is consistent for nun = 1, 2, 3 (as shown in Figure 6, for instance). In contrast, the results of the dynamic potentials of HCP (q2) and DCP (q3) are different. As shown in Figure 7, for instance, the vertical reversed symmetry could remain for a small P number (10 < P < 20) but it is broken when P becomes large. From the above results, it is shown that the uncoupled mode has an effect on the dynamic symmetry. It is obvious that the nun makes the original horizontal reversed symmetry break between the dynamic potential of HCP (q3) and DCP (q1) but has little effect on the vertical symmetry breaking between the dynamic potential of HCP (q2) and DCP (q3). It is elucidated that the stability of dynamic symmetry is different under the effect of the uncoupling mode. 4. Conclusions In this study, the dynamic potentials of highly excited vibrational states of DCP with an harmonicity and Fermi coupling are studied. The results show that the D–C–P bending mode has weak effects on D–C and C–P stretching mode under different Polyad numbers. Just like previous studies, it is found that the vibrational energy levels could be classified by the quantum environments. From comparative studies, it shows that the uncoupled modes make the original horizontal reversed symmetry breaking between the dynamic potential of HCP (q3) and DCP (q1), but has little effect on the vertical symmetry between dynamic potential of HCP (q2) and DCP (q3). Considering the effect of n2 in DCP and n1 in HCP, the original dynamic similarities in these two systems disappear and the characteristics of symmetry become much more complex. The above results show that the method which enables us to understand the DCP dynamics simply from those of HCP without repeated elaboration are only available in some special conditions, and, on the other hand, there are some new dynamic symmetries appearing when the conditions are different, which indicate that the homologous compounds are intrinsically similar only if the coupling patterns of two systems are analogous. Acknowledgments This work was supported by the Natural Science Foundation of China (Grant No. 11505027, 11104156) and the Open Foundation of Fundamental Science on Radioactive Geology and Exploration Technology Laboratory (Grant No. RGET1516). Author Contributions Aixing Wang and Yibao Liu performed the calculations of dynamic potential and the corresponding analysis. Chao Fang supervised the work. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Dynamic potentials of DCP (P = 18) with n2 = 0,1,2,3, and the energy levels included in the dynamic potential are shown by the lines. Figure 2 Dynamic potentials of DCP (P = 30) with n2 = 0,1,2,3, and the energy levels included in the dynamic potential are shown by the lines. Figure 3 Dynamic potentials of DCP when P = 30 and n2 = 0 and the energy levels included in the dynamic potential are shown by the lines. Figure 4 Trajectories of phase space of L0–L15 when P = 30 (q1 coordinate). Figure 5 Trajectories of phase space of L0–L15 when P = 30 (q3 coordinate). Figure 6 Dynamic potentials of HCP (q3) and DCP (q1) when nun > 0 (P = 30). Figure 7 Dynamic potentials of HCP (q2) and DCP (q3) when nun > 0 (P = 12). ijms-17-01280-t001_Table 1Table 1 The coefficients of vibration Hamiltonian of deuterated phosphaethyne (DCP). Parameter Name Parameter Values (cm−1) Parameter Name Parameter Values (cm−1) ω1 2494.0412 y223 0.0482 ω2 539.1611 y233 0.2535 ω3 1237.0955 y333 −0.2447 X11 −24.0769 z1111 0.0510 X12 −11.1041 z1112 0.0806 X13 −4.6276 z1222 −0.0055 X22 −3.5142 z1233 0.0280 X23 −2.2082 z2222 −0.0014 X33 −2.2082 z2233 −0.0067 y111 −0.8896 z2333 −0.0132 y112 −0.4928 z3333 0.0092 y113 −0.5407 k 12.3422 y122 0.2167 λ1 0.5786 y123 −0.3655 λ3 0.1212 y133 −0.2167 μ11 −0.2990 y222 0.0884 ==== Refs References 1. Peterson K.A. Skokov S. Bowman J. A theoretical study of the vibrational energy spectrum of the HOCl/HClO system on an accurate ab initio potential energy surface J. Chem. Phys. 1999 111 7446 7456 10.1063/1.480069 2. Peterson K.A. An accurate global ab initio potential energy surface for the X(1)A′ electronic state of HOBr J. Chem. Phys. 2000 113 4598 4612 10.1063/1.1288913 3. Joyeux M. Sugny D. Lombardi M. Jost R. Schinke R. Skokov S. Bowman J. Vibrational dynamics up to the dissociation threshold: A case study of two-dimensional HOCl J. Chem. Phys. 2000 113 9610 9621 10.1063/1.1321031 4. Joyeux M. Grebenshchikov S.Y. Bredenbeck J. Schinke R. Farantos S.C. Phase space geometry of multi-dimensional dynamic systems and reaction processes Geometric Structures of Phase Space in Multidimensional Chaos: Applications to Chemical Reaction Dynamics in Complex Systems, Volume 130 John Wiley & Sons Hoboken, NJ, USA 2005 5. Fang C. Wu G.Z. Dynamic similarity in the highly excited vibrations of HCP and DCP: The dynamic potential approach Comp. Theor. Chem. 2009 910 141 147 6. Wang A.X. Sun L.F. Fang C. Liu Y.B. The study of dynamic potentials of highly excited vibrational states of HOBr Int. J. Mol. Sci. 2013 14 5250 5263 10.3390/ijms14035250 23462512 7. Fang C. Wu G.Z. Global dynamic analysis of vibrational manifolds of HOCl and HOBr under anharmonicity and Fermi resonance: the dynamic potential approach Chin. Phys. B 2010 19 010509 8. Fang C. Wu G.Z. Bending localization of nitrous oxide under anharmonicity and Fermi coupling: The dynamic potential approach Chin. Phys. B 2010 19 110513 9. Biczysko M. Tarroni R. Theoretical calculation of rovibronic energy levels and anharmonic resonances in the ground X (2)Pi state of HCP+ and DCP+ Phys. Chem. Chem. Phys. 2002 4 708 715 10.1039/b107282j 10. Fang C. Wu G.Z. Dynamic potential approach to dissociation of H–C bond in HCO highly excited vibration Chin. Phys. B 2009 18 130 135 11. Kellman M.E. Approximate constants of motion for vibrational-spectra of many-oscillator systems with multiple anharmonic resonances J. Phys. Chem. 1990 93 6630 6635 10.1063/1.458930 12. Wang A.X. Liu Y.B. Fang C. The study of the dynamic potentials of highly excited vibrational states in HOCl molecular system Acta Phys. Sin. 2012 61 053102 13. Zhang W.M. Feng D.H. Gilmore R. Coherent states: Theory and some applications Rev. Mod. Phys. 1990 62 867 927 10.1103/RevModPhys.62.867 14. Weiss J. Hauschildt J. Grebenshchikov S.Y. Duren R. Schinke R. Koput J. Stamatiadis S. Farantos S.C. Saddle-node bifurcations in the spectrum of HOCl J. Chem. Phys. 2000 112 77 79 10.1063/1.480563

PMC005xxxxxx/PMC5000678.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081281ijms-17-01281ArticleTranscriptomic Insights into the Response of Placenta and Decidua Basalis to the CpG Oligodeoxynucleotide Stimulation in Non-Obese Diabetic Mice and Wild-Type Controls Liu Xiao-Rui 1†Guo Yu-Na 1†Qin Chuan-Mei 1Qin Xiao-Li 1Tao Fei 2Su Fei 2Tian Fu-Ju 1Zhang Yan 3*Lin Yi 1*Ciccodicola Alfredo Academic Editor1 International Peace Maternity and Child Health Hospital, the Institute of Embryo-Fetal Original Adult Disease, School of Medicine, Shanghai Jiao Tong University, Shanghai 200030, China; [email protected] (X.-R.L.); [email protected] (Y.-N.G.); [email protected] (C.-M.Q.); [email protected] (X.-L.Q.); [email protected] (F.-J.T.)2 State Key Laboratory of Microbial Metabolism, School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China; [email protected] (F.T.); [email protected] (F.S.)3 Department of Obstetrics and Gynecology, Renmin Hospital of Wuhan University, Wuhan 430060, China* Correspondence: [email protected] (Y.Z.); [email protected] (Y.L.); Tel.: +86-27-8804-1911 (Y.Z.); +86-21-6407-0434 (Y.L.)† These authors contributed equally to this work. 05 8 2016 8 2016 17 8 128121 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Intrauterine infection is one of the most frequent causes of miscarriage. CpG oligodeoxynucleotide (CpG ODN) can mimic intrauterine infection. CpG ODN-induced embryo-resorption was observed consistently in the NK-cell deficient non-obese diabetic (NOD) mice but not in the wild-type (WT) mice. To elucidate the molecular mechanisms of differential pregnancy outcomes, differentially expressed genes (DEGs) in the placenta and decidua basalis was revealed by RNA-Seq with CpG ODN or control ODN treatment. Common DEGs in the WT and NOD mice were enriched in antimicrobial/antibacterial humoral responses that may be activated as a primary response to bacterial infection. The susceptibility to CpG ODN-induced embryo-resorption in the NOD mice might mainly be attributed to M1 macrophage polarization and the immunodeficient status, such as the down-regulation in antigen processing and presentation, allograft rejection, and natural killer cell mediated cytotoxicity. In contrast, the WT mice with normal immune systems could activate multiple immune responses and be resistant to CpG ODN-induced embryo-resorption, such as M2 macrophage differentiation and activation regulated by complement component C1q and peroxisome proliferation-activated receptor (PPAR) signaling pathways. Collectively, this study suggests that the immunodeficient status of NOD mice and the macrophage polarization regulated by C1q and PPAR signaling might be the basis for differential pregnancy outcomes between the NOD and WT mice. intrauterine infectionmiscarriageC1qmacrophage ==== Body 1. Introduction Reproductive success in mammals depends on coordinated interaction between the placenta and uterus [1]. In humans, miscarriage is a common complication of pregnancy [2]. One of the most frequent causes is intrauterine infection [3]. CpG oligodeoxynucleotide (CpG ODN) is a synthetic oligonucleotide containing non-methylated CpG dinucleotides (CpG motifs), which are present with 20-fold greater frequency in bacterial DNA than in mammalian DNA [4,5]. Systemic or intrauterine bacterial infection may produce excessive hypomethylated CpG DNA motifs, which are recognized by Toll-like receptor 9 (TLR9) [4,6,7]. The interaction initiates immune responses that impair pregnancy and result in embryo loss [7,8]. Non-obese diabetic (NOD) mice are NK-cell deficient and have impaired fertility with poor embryo implantation and low embryo viability [9,10]. The defect of young NOD mice in NK1+-like thymocytes is both quantitative and qualitative, involving lack of IL-4 and IFN-γ production [11]. Our previous studies used allogeneic mating BALB/c × C57BL/6 and NOD × C57BL/6 mouse models to evaluate the effects of CpG ODN on pregnancy. CpG ODN-induced embryo-resorption is consistently observed in the NOD mice but not in the WT mice [8,12]. In the maternal-fetal microenvironment of NOD mice, the percentage of IL-10+ cells in decidual CD45+ cell populations is significantly lower than that in the WT mice [13], and CpG ODN treatment triggered amplification of uterine macrophages and neutrophils [8]. These effects of CpG ODN on pregnancy outcomes in the NOD mice were also observed in the IL-10−/− mice [7]. Furthermore, by adoptive transfer of in vitro-induced regulatory T cells (Treg) into the NOD mice, the percentage of decidual IL-10+ cells was significantly increased and CpG-induced pregnancy failure could be rescued [12]. Differential pregnancy outcomes between the NOD and WT mice might be attributed to limited Treg cells and insufficient IL-10 expression in the NOD mice [12]. However, the detailed mechanisms remain unclear. Here, high-throughput RNA sequencing (RNA-Seq) was used to investigate the genomic responses to CpG ODN-simulated intrauterine infection and to identify differentially expressed genes (DEGs) in the maternal-fetal microenvironment between the NOD and WT mice. Deeper understanding of gene expression will lead to new insights into the mechanisms underlying adverse pregnancy outcomes and establish a valuable platform for developing improved strategies for normal pregnancy outcome. 2. Results 2.1. Effects of CpG ODN on Embryo Loss Based on our previous experience [8,12], we used CpG ODN (ODN1826) in 200 μL PBS at dose of 25 μg to activate TLR9 in this study. Figure S1 shows that the embryo resorption rate by day 10.5 of gestation (E10.5) was significantly increased in the NOD mice after CpG ODN treatment (from 4.2% (8 of 187) to 9.1% (14 of 148), p < 0.05). However, no significant difference was observed in the WT mice between control ODN and CpG ODN treatment at the same dose and time. This result indicates that the animal models used for the following RNA-seq and RT-qPCR are well built. The NOD mice are sensitive to intrauterine bacterial infection simulated by injection with CpG ODN, while the WT mice are resistant to CpG-induced embryo loss. 2.2. Illumina Sequencing and Gene Expression Profiles We performed high-throughput Illumina sequencing of four cDNA libraries from placenta with decidua basalis, including CpG ODN-treated groups (WT-CpG ODN and NOD-CpG ODN) and control ODN-treated groups (WT-control ODN and NOD-control ODN). Table S1 shows statistics for raw and mapped reads. After filtration of low quality and adapter sequences, the Q20 base call accuracies for the remaining sequences were >98%. Using TopHat software, over 95.3% of the sequencing reads were mapped to the Mus musculus genome. According to the studies of Anders et al. [14], we calculated the expression levels of all the genes remaining in our analysis using cuffdiff, which is part of the Cufflinks software package. Most genes had similar expression patterns in each of our samples, and their levels were as observed in the most Gene Expression Omnibus (GEO) experiments (Figure S2). Then, we checked the expression pattern of housekeeping genes such as PPIase, GAPDH, and β-actin, which can be used to estimate variability across samples in the experiment. We found no significant difference in expression levels of these genes between samples. Based on these analyses, 50 genes were found to be at least two-fold differentially expressed (p < 0.05) in the WT mice (CpG ODN vs. control ODN). Forty-five genes were upregulated and five were downregulated with CpG ODN treatment. In the NOD mice (CpG ODN vs. control ODN), there were 53 genes with at least two-fold differential expression (p < 0.05). Twenty-five genes were upregulated and 28 were downregulated with CpG ODN treatment. With injection of control ODN, the NOD mice had 77 genes expressed at least two-fold differentially in comparison to the WT mice. Fifty-four genes were upregulated and 22 were downregulated. Under the CpG ODN treatment, there were 83 genes with more than two-fold differential expression (p < 0.05). Forty-four genes were upregulated and 39 were downregulated. Figure 1 compares the number of differentially expressed genes observed in various conditions. There were 41, 36, 35 and 33 genes uniquely differentially expressed in the four comparisons. Three genes (H2-Q7/H2-Q9, IgJ and Ltf) were differentially expressed after treatment of both the WT and NOD mice with CpG (p < 0.05). Four genes (Itgam, Mmp7, Ctss and Cybb) were the common DEGs in both the WT mice (with CpG ODN or control ODN) and the WT and NOD mice (with CpG ODN treatment). 2.3. Gene Ontology and Pathway Analysis To identify the function of DEGs, DAVID functional annotation cluster analysis was performed for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. Figures S3–S5 shows enriched GO terms arranged according to biological processes (GO-BP), cellular components (GO-CC) and molecular functions (GO-MF). Compared with the control ODN-treated WT mice, upregulated genes in the CpG ODN-treated WT mice were mainly localized to MHC protein complexes, associated with antigen binding and important in multiple immune responses, including defense response, antigen processing and presentation of exogenous antigens, B cell/lymphocyte/leukocyte mediated immunity, and inflammatory responses. Downregulated genes related to hormone activity, response to organic substance/hormone or endogenous stimuli, fat cell differentiation, and lipid catabolic process (Figure S3a). With respect to analysis using KEGG, 15 pathways were significantly enriched (Figure 2a). Upregulated genes were mainly involved in immune responses, such as antigen processing and presentation, allograft rejection, systemic lupus erythematosus, autoimmune thyroid disease, graft-versus-host disease, complement and coagulation cascades, and natural killer cell mediated cytotoxicity. Downregulated genes (Adipoq and Plin1) were involved in the PPAR signaling pathway. GO analysis of DEGs in the CpG ODN-treated NOD mice (Figure S3b) showed that the upregulated genes (compared with the control ODN-treated mice) were mainly localized in extracellular regions and lipoprotein particles. They were associated with lipid/protein/lipoprotein binding and chiefly involved in lipid or lipoprotein transport/localization and lipoprotein metabolic processes. Downregulated genes were associated with peptidase activity and endopeptidase activity in extracellular regions and participated in proteolysis and secondary metabolic processes. In pathway analysis using KEGG, only three unregulated genes (Fga, Fgb, and Plg) were significantly enriched in complement and coagulation cascades (Figure 2b). In the CpG ODN treatment experiments, unregulated genes in the NOD mice compared to the WT mice located in the extracellular region and secretory granule, being associated with cell differentiation and metabolic processes (Figure S4). In addition, the downregulated genes were mainly in MHC protein complex, participating in antigen processing and presentation. Furthermore, as shown in Figure 2c, the upregulated KEGG pathways were the complement and coagulation cascades and cytokine–cytokine receptor interaction. Downregulated pathways were mainly involved in immune responses, including allograft rejection, natural killer cell mediated cytotoxicity. In control ODN treatment experiments, compared with the WT mice, the upregulated genes in the NOD mice located to extracellular regions, and were associated with epithelial cell differentiation and metabolic processes (Figure S5). Pathway analyses (Figure 2d) also indicated that the unregulated genes were enriched in sphingolipid metabolism. In accordance with the immunodeficient status of NOD mice, downregulated genes in the NOD mice mainly clustered in the MHC protein complexes involved in antigen processing and presentation, graft-versus-host disease and allograft rejection (Figure S5 and Figure 2d). 2.4. Functional Groups in Immune System Processes To further understand the biological relevance of the DEGs in immune system processes, we performed functional enrichment analysis using ClueGO, which facilitates the visualization of functionally related genes displayed as a clustered network and chart. Networks were constructed for DEGs in four comparisons: (1) DEGs between the CpG ODN-treated and control ODN-treated WT mice (Figure 3a,e); (2) DEGs between the CpG ODN-treated and control ODN-treated NOD mice (Figure 3b,f); (3) DEGs between the WT and NOD mice with injection of CpG ODN (Figure 3c,g); and (4) DEGs between the WT and NOD mice with injection of control ODN (Figure 3d,h). Nodes in the networks are the terms of functionally grouped networks. The size represents the term enrichment significance. Groups with related functions partially overlap. As shown in Figure 3a,b,e,f, on CpG ODN-simulated bacterial infection, the DEGs in both the WT and NOD mice were enriched in antimicrobial/antibacterial humoral gene responses, including Ltf, IgJ, Mmp7, Fga and Fgb. Furthermore, compared to the NOD mice, the WT mice had additional upregulated genes involved in antigen processing and presentation of exogenous peptide antigens (H2-Aa, H2-Ab1, and H2-Eb1), humoral immune responses mediated by circulating immunoglobulin (C1qa, C1qb, C1qc, and Cfb), negative regulation of leukocyte differentiation (Adipoq and Itgam), and microglial cell activation (Cx3cr1, Aif1, and Tlr1), especially macrophage activation and differentiation (C1qa, C1qb, C1qc, Adipoq, and Itgam) (Figure 3a). In addition, the DEGs between the WT and NOD mice treated with control ODN were mainly enriched in antigen processing and presentation of peptide antigens via MHC class I (H2-T10, H2-T22, and H2-T23) (Figure 3d). Except antigen processing and presentation, the biological relevance in immune system processes of the DEGs between the CpG ODN-treated WT and NOD mice mainly involved in lymphocyte mediated immunity (Figure 2g). 2.5. Validation of Gene Expression Most DEGs mentioned above and involved in clustered immune system processes were quantified by RT-qPCR. Twenty genes were quantified in the WT mice (CpG ODN vs. control ODN), such as downregulated genes (Adipoq and Plin1) in the PPAR signaling pathway, upregulated genes (H2-D1, Itgam, B2m, Cfb, and Cd74) involved in antigen processing and presentation, genes involved in macrophage activation (Tlr1) and antibacterial humoral response (Ltf) (Figure 4a). Twelve genes were quantified in the NOD mice (CpG ODN vs. control ODN), such as Fgb, IgJ, and Ltf involved in antimicrobial humoral response, and Apoa4 involved in mucosal immune response (Figure 4b). Ten genes were quantified in the CpG ODN treated WT and NOD mice, respectively (Figure 4c). Eleven genes were compared between the WT and NOD mice being injected with control ODN (Figure 4d). Expression of each gene was measured in triplicate. In total, 95.3% of them were significantly changed, which was consistent with the RNA-Seq results (p < 0.05) and indicated that data obtained from RNA-Seq were reliable. 2.6. Macrophage Polarization Macrophages are classified as pro-inflammatory/classically activated macrophages (M1) and proresolving/alternatively activated macrophages (M2). The expression of arginase or inducible nitric oxide synthase (iNOS) is associated with macrophage polarization [15,16]. The complement component C1q promotes M2 polarization by inducing the expression of arginase and limited inflammasome activation in human monocyte derived macrophages [17]. RT-qPCR was performed to explore the expression levels of TLR9, the three components of C1q, arginase, iNOS and IL10 in the WT and NOD mice with or without CpG ODN-simulated bacterial infection. As Figure 5a shows, with CpG ODN-treatment, the expression of TLR9 increased (p < 0.05). C1qa, C1qb and C1qc displayed higher expression in the WT mice than in the NOD mice. They were upregulated in the CpG-ODN treated WT mice relative to control ODN treatment (p < 0.01), but no significant change was observed in the NOD mice with the same treatment. In addition, both the WT and NOD mice expressed Arg1 (encoding liver-type arginase) at a higher level on treatment with CpG ODN; the expression of Arg2 (encoding kidney-type arginase) was downregulated while NOS2 (encoding iNOS) was upregulated in the NOD mice. In addition, compared to the control ODN-treatment, the expression of IL10 increased in the CpG ODN-treated WT mice, but no significant change was observed in the NOD mice. Taken together, these results indicate that macrophages in the placenta and decidua basalis of the WT mice had M2 activity, while the NOD mice had M1 polarized macrophages. 2.7. Effect of C1q Inhibition on Abortion in Wild-Type (WT) Mice To further confirm the function of C1q in preventing resorption in the CpG ODN-treated WT mice, anti-C1q antibody was injected intraperitoneally to block the function of the C1q in WT mice. As shown in Figure 6, the abortion rates of the CpG ODN-treated WT mice were increased significantly when the function of C1q was inhibited by the neutralizing antibody (from 6.4% to 23.7%, p < 0.01). Thus, the result in vivo indicates that C1q plays a positive role in preventing abortion in the WT mice responding to CpG ODN stimulation. 3. Discussion Intrauterine infection is one of most common causes of spontaneous miscarriage [3]. CpG ODN is a synthetic oligonucleotide containing non-methylated CpG dinucleotides (CpG motifs), which are present with 20-fold greater frequency in bacterial DNA than in mammalian DNA [4,5]. They are thus recognized by TLR9, leading to strong immune responses [18]. In the present study, the genomic responses to simulated intrauterine infection induced by CpG ODN were investigated in the maternal–fetal microenvironment of WT and NOD mice by RNA-Seq. Based on transcript assembly and abundance estimation using Cufflinks software [19], we identified DEGs with p < 0.05 and greater than two-fold change. Fifty-three genes were selected to be quantified by RT-qPCR. Among these, 96.2% were significantly altered (p < 0.05), indicating the reliability of our RNA-Seq data. Compared to control ODN treatment, CpG ODN-induced DEGs in both the WT and NOD mice were enriched in antimicrobial/antibacterial humoral responses, including Ltf, IgJ, Mmp7, Fga and Fgb. Lactoferrin (encoded by Ltf), an iron-binding glycoprotein in the transferrin family, has been detected in vaginal secretions and amniotic fluid [20,21,22,23,24,25,26,27,28]. This protein modulates inflammatory and immune responses to kill bacteria, viruses and fungi [29]. Recombinant human lactoferrin has a positive role in the prevention of bacteria-induced preterm delivery in rabbit and mouse models [23,25,26,28,30]. Furthermore, vaginal administration of lactoferrin plays a role in reducing the risk of preterm birth for women with shortened cervical length and elevated interleukin 6 levels [31], being especially effective for women with refractory vaginitis recurring preterm delivery [21]. Thus, there is a mechanistic link between lactoferrin and bacterial-induced embryo-resorption. The joining (J) chain (encoded by IgJ) is expressed by mucosal and glandular plasma cells, incorporates in the polymer of immunoglobulins such as IgM and IgA [32], and is also involved in mucosal immunity through the transport of Ig across epithelial surfaces by promoting binding with the poly Ig receptor (pIgR) [33,34,35,36]. It appears that mucosal immunity might be activated in response to bacterial infections in both the WT and NOD mice. Gene ontology and pathway analysis revealed that the susceptibility to CpG ODN-induced embryo-resorption in the NOD mice might mainly be attributed to the immunodeficient status, such as the downregulated in antigen processing and presentation, allograft rejection, and natural killer cell mediated cytotoxicity. In contrast, the WT mice with normal immune systems could activate multiple immune responses and appeared to be resistant to CpG ODN-induced embryo-resorption. Among the 50 genes exhibiting greater than two-fold change in the WT mice, most were upregulated and mainly localized to the MHC complex. They were associated with multiple immune responses (Figure 2a and Figure 3a,e and Figure S3a), including defense responses (such as antimicrobial/antibacterial humoral response), antigen processing, presentation of exogenous antigens, negative B cell/lymphocyte/leukocyte mediated immunity, and inflammatory responses. However, in the NOD mice, the great majority of DEGs were associated with biosynthesis, transport and localization of lipid or lipoprotein (Figure S3b). Only three genes (Fga, Fgb, and Plg) were significantly enriched in complement and coagulation cascades (Figure 2b). As indicated above, immune responses of the NOD mice involved antimicrobial/antibacterial humoral responses and mucosal immune response (Figure 3b,e). Taken together, antimicrobial/antibacterial humoral responses may be activated as a primary response to bacterial infection. On the CpG ODN treatment, except antigen processing and presentation, the most notable immune response in the WT mice was macrophage differentiation and activation (Figure 3a,e). Macrophages, essential components of the innate immune system, are specialized to respond to infectious microbes. They play a critical role in regulating inflammatory responses and host defense [37,38,39]. The DEGs Cx3cr1, Aif1 and Tlr1 genes appear related to macrophage activation, while Adipoq, C1qc and Itgam were involved in macrophage differentiation. Macrophages have two subtypes. M1 macrophages are often associated with an increased production of pro-inflammatory cytokines such as TNF-α and IL-1β, inhibit cell proliferation and cause tissue damage. M2 activity, the “default” activity in resident macrophages, promotes cell proliferation and tissue repair [15,16,40]. An increased production of anti-inflammatory cytokine IL-10 is associated with M2 macrophages [16]. The complement component C1q promotes M2 polarization by inducing the expression of arginase and limited inflammasome activation in human monocyte derived macrophages [17]. We previously found that the percentage of IL-10+ cells in the decidual CD45+ cell population derived from the WT mice was much higher than that in the NOD mice [13], and serum TNF-α levels were the same in the WT and NOD mice. In the present study, C1qa, C1qb and C1qc were expressed at a significantly higher level in the WT mice than in the NOD mice with control ODN treatment (Figure 5a), suggesting that macrophages in the WT mice might be predominantly M2 subtype. Upon CpG ODN stimulation, the three components of C1q in the WT mice were upregulated, then induced high expression of arginase (Figure 5a). In addition, the expression of IL-10 was higher than in the control. To clarify whether C1q plays a positive role in preventing resorption in the WT mice, neutralizing anti-C1q antibody was injected together with CpG ODN. The abortion rate increased markedly in the WT mice when CpG ODN challenge plus anti-C1q antibody injection (Figure 6). These results indicated that C1q might promoted M2 polarization in the WT mice and that the mice used a “repair” program in response to CpG ODN stimulation (Figure 5b). However, in the NOD mice with the same stimulation, iNOS was significantly upregulated. In our previous studies, CpG ODN induced a significant increase in uterine CD11b+F4/80+ macrophages of NOD mice [8]. Moreover, the NOD mice showed a substantial increase in serum and intracellular TNF-α, but the TNF-α level did not exhibit any change in the WT mice [8]. It is possible that NOD mouse macrophages have M1 activity and turn on a “killing” program to inhibit cell proliferation, causing increased embryo-resorption in the CpG ODN-treated NOD mice as compared with the control ODN-treated NOD mice (Figure 5c). In addition, among the only five genes downregulated markedly in the CpG ODN treated WT mice, Adipoq and Plin1 mainly participants in PPAR signaling pathways (Figure 2a). Adiponectin (APN), encoded by Adipoq, a hormone produced from adipose tissue, regulates various biological responses. Besides its metabolic functions, accumulating evidence indicates that APN exerts anti-inflammatory effects on macrophages, including stimulating the production of IL-10 and antagonist, decreasing phagocytic activity and inhibiting NF-κB to suppress the production of pro-inflammatory cytokines [41,42,43,44]. Expression of adiponectin can be enhanced by PPARs and suppressed by pro-inflammatory cytokines such as TNFα and IL-6 [44,45,46]. Rosiglitazone, an agonist for PPARγ, could increase adiponectin in adipose tissue and result in decreased inflammatory cytokines and macrophage infiltration [47]. It is interesting to explore the mechanism in detail of PPAR signaling pathway and pregnancy outcomes under intrauterine infection in future research. 4. Materials and Methods 4.1. Animal Administration and Sample Collection The wild-type (WT) female BALB/c mice, the female NOD mice of BALB/c background, and the C57BL/6 male mice (8–12 weeks old) were purchased from Beijing HFK Bioscience (Beijing, China). All mice were housed in a pathogen-free facility. The handing of the experimental animals was in accordance with national animal care guidelines and the Medical Ethics Committee of the International Peace Maternity and Child Health Hospital of China Welfare Institute (Shanghai, China, 2014-22, 27 February 2014) specifically approved this study. The female NOD and BALB/c mice were mated with the male C57BL/6 mice in natural cycling. The day of sighting a vaginal plug was considered as Embryonic Day 0.5 (E0.5). As shown in Figure S6, on E6.5, the WT and NOD mice were injected intraperitoneally with CpG ODN (ODN1826; InvivoGen, San Diego, CA, USA) at a dose of 25 μg in 200 μL PBS. The control mice were injected with control ODN (ODN 2138; InvivoGen) at the same dose and time. These were CpG-treated groups (WT-CpG ODN (n = 7) and NOD-CpG ODN (n = 14)) and control groups (WT-control ODN (n = 7) and NOD-control ODN (n = 16)). On E10.5 or E11.5, placentas with decidua basalis were collected separately and immediately frozen in liquid nitrogen. To investigate the effect of C1q on embryo resorption, the CpG ODN-treated WT mice were injected anti-C1q antibody [JL-1] (ab71940, Abcam, Cambridge, UK) intraperitoneally to block the function of C1q at a dose of 50 μg on E5.5 and E7.5 (n = 4). Control ODN-treated (n = 9) and CpG ODN-treated (n = 9) WT mice were used as control. Feature of resorbed embryos includes small size, haemorrhage and necrosis. Embryo-resorption was calculated as the ratio of resorbed embryos to total. 4.2. RNA Preparation and Construction of RNA-Seq Libraries One placenta with decidua basalis from each mouse was individually ground into powder by mortar and pestle in liquid nitrogen. Total RNA was extracted from each frozen powdered sample using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to standard protocols. RNA quality was evaluated by electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, San Diego, CA, USA). Samples with RNA integrity numbers (RINs) >9.4 and with 260/280 nm absorbance ratios from 1.9 to 2.1 were used for construction of RNA-Seq libraries. For RNA-seq, equal amounts of total RNAs from three mice within each treatment group were combined into a pooled sample. Four pooled RNA samples were used in cDNA library construction using the TruSeq™ RNA Sample Prep kit (Illumina, San Diego, CA, USA) following the manufacturer’s instructions. 4.3. Sequencing and Assembly Each RNA-Seq library was 100 bp pair end-sequenced on a HiSeq2000 instrument by Shanghai Majorbio Biopharm Biotechnology (Shanghai, China), and individually assessed for quality using FastQC. To avoid low-quality data negatively influencing downstream analysis, raw Illumina sequence reads were trimmed for low quality data (Phred < 20), ambiguous bases (N), sequencing adapters, primers, and poly(A)/(T) tails using the FastX Tool kit [48]. The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE69407 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE69407). 4.4. Data Analysis We aligned the quality checked reads to the mm10 build of the mouse genome (http://hgdownload.soe.ucsc.edu/goldenPath/mm10/chromosomes/) using TopHat version 2.0.13 with default parameters [49]. Transcript assembly and abundance estimation were performed using Cufflinks 2.2.1 [19]. Gene expression levels were expressed as fragments per kilobase per million (FPKM) mapped reads. A gene was considered to be expressed in a sample if its value in FPKM was ≥1. In this case, there is no biological replicate. In order to get all the differentially expressed genes, we use standard criteria for calling the DEGs, which is on the basis of past experience and used by many studies [14,50]. The significance of DEGs was identified with an adjusted p value (in Cuffdiff, the adjusted p value considers multiple testing using the Benjamini–Hochberg method) <0.05 (at 95% confidence) and with a fold change >2, which is recommended by the CuffDiff manual (false discovery rate <0.05). A Venn diagram was created using VENNY (http://bioinfogp.cnb.csic.es/tools/venny/). DAVID functional annotation cluster analysis (http://david.abcc.ncifcrf.gov/home.jsp) was performed on the list of DEGs for GO and KEGG pathway enrichment analysis [51,52]. The biological role of DEGs was analyzed and visualized using ClueGo (v2.1.6)/Clupedia (v1.1.6) [53] as a plug-in of Cytoscape (v3.1.1) [54,55,56]. DEG gene symbols were uploaded and analyzed employing default parameters. The statistical test used for the enrichment was based on the right-sided hypergeometric option with a Benjamini–Hochberg correction and κ score of 0.4. 4.5. Quantitative Real-Time PCR qRT-PCR was performed with RNA from one placenta with decidua basalis of another 6 individual mice in each group as biological replicates, not the pooled RNA used for RNA-seq. cDNA was synthesized from total RNA using a FastQuant RT kit with gDNase (Tiangen, Beijing, China). qPCR experiments were performed using SYBR® Premix Ex Taq™ (TaKaRa, Dalian, China) on a LightCycler® 480 real-time PCR system in conjunction with a 384 multi-well plate (Roche, Mannheim, Germany) per the manufacturers’ instructions. Each sample was run in triplicate reactions as technical replicates under the following conditions: initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 20 s. Relative gene expression was calculated by the 2−∆∆Ct method [57] and normalized to β-actin mRNA. Primer sequences used for qPCR of selected DEGs are shown in Table S2. 4.6. Statistical Analysis Statistical analysis was performed using GraphPad Prism software version 5.0 (GraphPad Software, CA, USA). We used K-S test to determine whether the data is normally distributed and used Levene’s test to determine homogeneity of variance. For the data of normal distribution, One-way ANOVA followed by Tukey’s test to compare means was used to compare more than two groups. And unpaired Student’s t-test was used to compare means in two groups. For the data that did not fit the normal distribution, nonparametric test (Mann-Whitney test) was used. Differences were identified as significant (* p < 0.05; ** p < 0.01; *** p < 0.001) or not significant (ns). Data represent means of the biological replicates ± SEM. 5. Conclusions In summary, we proposed a possible mechanism which can be used to explain the differential pregnancy outcomes between the WT and NOD mice responding to simulated intrauterine infection induced by CpG ODN based on transcriptomic analysis. Antimicrobial/antibacterial humoral responses may be activated as a primary response to bacterial infection. The M2 polarization regulated by C1q and PPAR signaling pathways in the WT mice and the immunodeficient status and M1 activity in the NOD mice are the fundamental basis of the differential pregnancy outcomes in response to infection. RT-qPCR was performed for verification, and the result was consistent with RNA-Seq data. These findings shed a light on the role of complement in reversing adverse pregnancy outcomes due to systemic or intrauterine bacterial infection. Acknowledgments This work was supported by grants from National Natural Science Foundation of China (31171439, 81125004, 81401274 and 81501277), National Basic Research Program of China (2013CB967404), Shanghai Jiao Tong University Medicine-Engineering Fund (YG2013ZD04 and YG2013MS68), and Natural Science Foundation of Hubei Province (2015CFB722). Supplementary Materials Supplementary materials can be accessed at: http://www.mdpi.com/1422-0067/17/8/1281/s1. Click here for additional data file. Author Contributions Xiao-Rui Liu and Yi Lin conceived and designed the project and experiments; Xiao-Rui Liu, Chuan-Mei Qin, Xiao-Li Qin and Yu-Na Guo performed the experiments; Xiao-Rui Liu, Fei Su, Fei Tao and Fu-Ju Tian analyzed the data. Xiao-Rui Liu, Fei Su and Yi Lin wrote the paper; Yan Zhang and Yi Lin supervised this study. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Unique or shared differentially expressed genes (DEGs). WT (CpG ODN vs. control ODN) refers to the DEGs between CpG ODN and control ODN-treated wild-type (WT) mice. NOD (CpG ODN vs. control ODN) refers to the DEGs between CpG ODN and control ODN-treated non-obese diabetic (NOD) mice. NOD vs. WT (CpG ODN) refers to the DEGs between WT and NOD mice with CpG ODN treatment. NOD vs. WT (Control ODN) refers to the DEGs between WT and NOD mice with control ODN treatment. Numbers of DEGs in the indicated comparisons are shown in the Venn diagram. The percentage numbers indicate the proportion of unique or shared DEGs to total DEGs. Figure 2 KEGG pathway analysis of DEGs: (a) based on the DEGs between CpG ODN and control ODN-treated WT mice; (b) based on the DEGs between CpG ODN and control ODN-treated NOD mice; (c) based on the DEGs between CpG ODN-treated WT and NOD mice; and (d) based on the DEGs between Control ODN-treated WT and NOD mice. Figure 3 Functionally grouped annotation in immune system process. Annotation network and overview chart indicating functional groups in immune system processes are analyzed using ClueGo. Nodes are the terms of the functionally grouped network. The sizes of nodes represent the term enrichment significance. The groups are visualized with different colors on the network. Groups with related functions partially overlap. (a,e) DEGs in WT mice (CpG ODN vs. control ODN); (b,f) DEGs in NOD mice (CpG ODN vs. control ODN); (c,g) DEGs between WT and NOD mice on CpG ODN treatment; and (d,h) DEGs between WT and NOD mice on control ODN treatment. Figure 4 RT-qPCR of selected DEGs involved in clustered immune system processes: (a) DEGs between CpG ODN and control ODN treatments in WT mice; (b) DEGs between CpG ODN and control ODN treatments in NOD mice; (c) DEGs between WT and NOD mice with CpG ODN treatment; and (d) DEGs between WT and NOD mice with control ODN treatment. qRT-PCR was performed with RNA from another six individual mice in each group as biological replicates. Each sample was run in triplicate reactions as technical replicates. The value on the y-axis represents the fold change value for each gene. Data represent means of the biological replicates ± SEM (n = 6); unpaired Student’s t-test: * p < 0.05; ** p < 0.01; *** p < 0.001; ns, not significant. Figure 5 Complement C1q and M1/M2 macrophage polarization: (a) the expression levels of TLR9, three components of C1q (C1qa, C1qb and C1qc), M1/M2 markers (arginase and inducible nitric oxide synthase), and IL10 in WT and NOD mice with CpG Oligodeoxynucleotide (CpG ODN) or control ODN treatment; (b) M1/M2 macrophage polarization in WT mice with CpG ODN treatment; and (c) M1/M2 macrophage polarization in NOD mice with CpG ODN treatment. qRT-PCR was performed with RNA from another six individual mice in each group as biological replicates. Each sample was run in triplicate reactions as technical replicates. Data represent means of the biological replicates ± SEM (n = 6), One-way ANOVA followed by Tukey’s test; ns, not significant; * p < 0.05; ** p < 0.01; *** p < 0.001. Figure 6 Effects of C1q on the pregnancy outcomes in CpG ODN-treated WT mice. The abortion rates of CpG ODN-treated WT mice were increased significantly when the function of C1q was inhibited by the neutralizing antibody. The abortion rates of pregnant mice were calculated. Data represent mean of the biological replicates ± SEM (n = 9 in control ODN-treated group, n = 9 in CpG ODN-treated group, and n = 4 in CpG ODN plus Anti-C1q treated group), One-way ANOVA followed by Tukey’s test; ns, not significant; *** p < 0.001. ==== Refs References 1. Erlebacher A. Immunology of the maternal-fetal interface Annu. Rev. Immunol. 2013 31 387 411 23298207 2. Salker M. Teklenburg G. Molokhia M. Lavery S. Trew G. Aojanepong T. Mardon H.J. Lokugamage A.U. Rai R. Landles C. Natural selection of human embryos: Impaired decidualization of endometrium disables embryo-maternal interactions and causes recurrent pregnancy loss PLoS ONE 2010 5 1281 3. Romero R. Espinoza J. Mazor M. Can endometrial infection/inflammation explain implantation failure, spontaneous abortion, and preterm birth after in vitro fertilization? Fertil. Steril. 2004 82 799 804 10.1016/j.fertnstert.2004.05.076 15482749 4. Krieg A.M. CpG motifs in bacterial DNA and their immune effects Annu. Rev. Immunol. 2002 20 709 760 10.1146/annurev.immunol.20.100301.064842 11861616 5. Krieg A.M. Yi A.K. Matson S. Waldschmidt T.J. Bishop G.A. Teasdale R. Koretzky G.A. Klinman D.M. CpG motifs in bacterial DNA trigger direct B-cell activation Nature 1995 374 546 549 10.1038/374546a0 7700380 6. Pedersen G. Andresen L. Matthiessen M.W. Rask-Madsen J. Brynskov J. Expression of toll-like receptor 9 and response to bacterial CpG oligodeoxynucleotides in human intestinal epithelium Clin. Exp. Immunol. 2005 141 298 306 10.1111/j.1365-2249.2005.02848.x 15996194 7. Thaxton J.E. Romero R. Sharma S. TLR9 activation coupled to IL-10 deficiency induces adverse pregnancy outcomes J. Immunol. 2009 183 1144 1154 10.4049/jimmunol.0900788 19561095 8. Sun Y. Qin X. Shan B. Wang W. Zhu Q. Sharma S. Wu J. Lin Y. Differential effects of the CpG-Toll-like receptor 9 axis on pregnancy outcome in nonobese diabetic mice and wild-type controls Fertil. Steril. 2013 99 1759 1767 10.1016/j.fertnstert.2013.01.121 23414919 9. Formby B. Schmid-Formby F. Jovanovic L. Peterson C.M. The offspring of the female diabetic “nonobese diabetic” (NOD) mouse are large for gestational age and have elevated pancreatic insulin content: A new animal model of human diabetic pregnancy Proceedings of the Society for Experimental Biology and Medicine Society for Experimental Biology and Medicine New York, NY, USA 1987 Volume 184 291 294 3547410 10. Kim J.H. Bugaj L.J. Oh Y.J. Bivalacqua T.J. Ryoo S. Soucy K.G. Santhanam L. Webb A. Camara A. Sikka G. Arginase inhibition restores NOS coupling and reverses endothelial dysfunction and vascular stiffness in old rats J. Appl. Physiol. 2009 107 1249 1257 10.1152/japplphysiol.91393.2008 19661445 11. Mehl A. Ghorbani P. Douda D. Huang H. Palaniyar N. Ratjen F. Grasemann H. Effect of arginase inhibition on pulmonary L-arginine metabolism in murine Pseudomonas pneumonia PLoS ONE 2014 9 1281 10.1371/journal.pone.0090232 24595185 12. Lin Y. Liu X. Shan B. Wu J. Sharma S. Sun Y. Prevention of CpG-induced pregnancy disruption by adoptive transfer of in vitro-induced regulatory T cells PLoS ONE 2014 9 1281 10.1371/journal.pone.0094702 24714634 13. Lin Y. Ren L. Wang W. Di J. Zeng S. Saito S. Effect of TLR3 and TLR7 activation in uterine NK cells from non-obese diabetic (NOD) mice J. Reprod. Immunol. 2009 82 12 23 10.1016/j.jri.2009.03.004 19560213 14. Anders S. McCarthy D.J. Chen Y. Okoniewski M. Smyth G.K. Huber W. Robinson M.D. Count-based differential expression analysis of RNA sequencing data using R and Bioconductor Nat. Protoc. 2013 8 1765 1786 10.1038/nprot.2013.099 23975260 15. Mills C.D. M1 and M2 Macrophages: Oracles of Health and Disease Crit. Rev. Immunol. 2012 32 463 488 10.1615/CritRevImmunol.v32.i6.10 23428224 16. Bohlson S.S. O’Conner S.D. Hulsebus H.J. Ho M.M. Fraser D.A. Complement, C1q, and C1q-related molecules regulate macrophage polarization Front. Immunol. 2014 5 402 10.3389/fimmu.2014.00402 25191325 17. Benoit M.E. Clarke E.V. Morgado P. Fraser D.A. Tenner A.J. Complement protein C1q directs macrophage polarization and limits inflammasome activity during the uptake of apoptotic cells J. Immunol. 2012 188 5682 5693 10.4049/jimmunol.1103760 22523386 18. Bauer S. Kirschning C.J. Hacker H. Redecke V. Hausmann S. Akira S. Wagner H. Lipford G.B. Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition Proc. Natl. Acad. Sci. USA 2001 98 9237 9242 10.1073/pnas.161293498 11470918 19. Trapnell C. Roberts A. Goff L. Pertea G. Kim D. Kelley D.R. Pimentel H. Salzberg S.L. Rinn J.L. Pachter L. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks Nat. Protoc. 2012 7 562 578 10.1038/nprot.2012.016 22383036 20. Teng C.T. Lactoferrin gene expression and regulation: An overview Biochem. Cell Biol. 2002 80 7 16 10.1139/o01-215 11908645 21. Otsuki K. Tokunaka M. Oba T. Nakamura M. Shirato N. Okai T. Administration of oral and vaginal prebiotic lactoferrin for a woman with a refractory vaginitis recurring preterm delivery: Appearance of lactobacillus in vaginal flora followed by term delivery J. Obstet. Gynaecol. Res. 2014 40 583 585 10.1111/jog.12171 24118573 22. Paesano R. Pietropaoli M. Berlutti F. Valenti P. Bovine lactoferrin in preventing preterm delivery associated with sterile inflammation Biochem. Cell Biol. 2012 90 468 475 10.1139/o11-060 22292525 23. Mitsuhashi Y. Otsuki K. Yoda A. Shimizu Y. Saito H. Yanaihara T. Effect of lactoferrin on lipopolysaccharide (LPS) induced preterm delivery in mice Acta Obstet. Gynecol. Scand. 2000 79 355 358 10.1080/j.1600-0412.2000.079005355.x 10830761 24. Giunta G. Giuffrida L. Mangano K. Fagone P. Cianci A. Influence of lactoferrin in preventing preterm delivery: A pilot study Mol. Med. Rep. 2012 5 162 166 21922138 25. Hasegawa A. Otsuki K. Sasaki Y. Sawada M. Mitsukawa K. Chiba H. Nagatsuka M.A. Okai T. Kato A. Preventive effect of recombinant human lactoferrin in a rabbit preterm delivery model Am. J. Obstet. Gynecol. 2005 192 1038 1043 10.1016/j.ajog.2005.01.013 15846177 26. Sasaki Y. Otsuki K. Hasegawa A. Sawada M. Chiba H. Negishi M. Nagatsuka M. Okai T. Preventive effect of recombinant human lactoferrin on lipopolysaccharide-induced preterm delivery in mice Acta Obstet. Gynecol. Scand. 2004 83 1035 1038 10.1111/j.0001-6349.2004.00587.x 15488117 27. Otsuki K. Yakuwa K. Sawada M. Hasegawa A. Sasaki Y. Mitsukawa K. Chiba H. Nagatsuka M. Saito H. Okai T. Recombinant human lactoferrin has preventive effects on lipopolysaccharide-induced preterm delivery and production of inflammatory cytokines in mice J. Perinat. Med. 2005 33 320 323 10.1515/JPM.2005.057 16207117 28. Saito H. Otsuki K. Mitsuhashi Y. Yoda A. Yanaihara A. Shimizu Y. Yanaihara T. Role of lactoferrin in the prevention of preterm delivery Int. Congr. Ser. 2000 1195 347 352 29. Teng C.T. Beard C. Gladwell W. Differential expression and estrogen response of lactoferrin gene in the female reproductive tract of mouse, rat, and hamster Biol. Reprod. 2002 67 1439 1449 10.1095/biolreprod.101.002089 12390874 30. Yakuwa K. Otsuki K. Nakayama K. Hasegawa A. Sawada M. Mitsukawa K. Chiba H. Nagatsuka M. Okai T. Recombinant human lactoferrin has a potential to suppresses uterine cervical ripening in preterm delivery in animal model Arch. Gynecol. Obstet. 2007 275 331 334 10.1007/s00404-006-0261-9 17031657 31. Locci M. Nazzaro G. Miranda M. Salzano E. Montagnani S. Castaldo C. De Placido G. Vaginal lactoferrin in asymptomatic patients at low risk for pre-term labour for shortened cervix: Cervical length and interleukin-6 changes J. Obstet. Gynaecol. 2013 33 144 148 10.3109/01443615.2012.740527 23445135 32. Koshland M.E. The coming of age of the immunoglobulin J chain Annu. Rev. Immunol. 1985 3 425 453 10.1146/annurev.iy.03.040185.002233 2415140 33. Hendrickson B.A. Conner D.A. Ladd D.J. Kendall D. Casanova J.E. Corthesy B. Max E.E. Neutra M.R. Seidman C.E. Seidman J.G. Altered hepatic transport of immunoglobulin A in mice lacking the J chain J. Exp. Med. 1995 182 1905 1911 10.1084/jem.182.6.1905 7500036 34. Johansen F.E. Braathen R. Brandtzaeg P. Role of J chain in secretory immunoglobulin formation Scand. J. Immunol. 2000 52 240 248 10.1046/j.1365-3083.2000.00790.x 10972899 35. Vaerman J.P. Langendries A.E. Giffroy D.A. Kaetzel C.S. Fiani C.M. Moro I. Brandtzaeg P. Kobayashi K. Antibody against the human J chain inhibits polymeric Ig receptor-mediated biliary and epithelial transport of human polymeric IgA Eur. J. Immunol. 1998 28 171 182 10.1002/(SICI)1521-4141(199801)28:01<171::AID-IMMU171>3.0.CO;2-# 9485197 36. Tacchi L. Larragoite E. Salinas I. Discovery of J chain in African lungfish (Protopterus dolloi , Sarcopterygii) using high throughput transcriptome sequencing: Implications in mucosal immunity PLoS ONE 2013 8 1281 10.1371/journal.pone.0070650 23967082 37. Gordon S. Alternative activation of macrophages Nat. Rev. Immunol. 2003 3 23 35 10.1038/nri978 12511873 38. Gordon S. Martinez F.O. Alternative activation of macrophages: Mechanism and functions Immunity 2010 32 593 604 10.1016/j.immuni.2010.05.007 20510870 39. Sica A. Mantovani A. Macrophage plasticity and polarization: In vivo veritas J. Clin. Investig. 2012 122 787 795 10.1172/JCI59643 22378047 40. Mills C.D. Kincaid K. Alt J.M. Heilman M.J. Hill A.M. M-1/M-2 macrophages and the Th1/Th2 paradigm J. Immunol. 2000 164 6166 6173 10.4049/jimmunol.164.12.6166 10843666 41. Wolf A.M. Wolf D. Rumpold H. Enrich B. Tilg H. Adiponectin induces the anti-inflammatory cytokines IL-10 and IL-1RA in human leukocytes Biochem. Biophys. Res. Commun. 2004 323 630 635 10.1016/j.bbrc.2004.08.145 15369797 42. Yamaguchi N. Argueta J.G. Masuhiro Y. Kagishita M. Nonaka K. Saito T. Hanazawa S. Yamashita Y. Adiponectin inhibits Toll-like receptor family-induced signaling FEBS Lett. 2005 579 6821 6826 10.1016/j.febslet.2005.11.019 16325814 43. Yokota T. Oritani K. Takahashi I. Ishikawa J. Matsuyama A. Ouchi N. Kihara S. Funahashi T. Tenner A.J. Tomiyama Y. Adiponectin, a new member of the family of soluble defense collagens, negatively regulates the growth of myelomonocytic progenitors and the functions of macrophages Blood 2000 96 1723 1732 10961870 44. Barnes M.A. Carson M.J. Nair M.G. Non-traditional cytokines: How catecholamines and adipokines influence macrophages in immunity, metabolism and the central nervous system Cytokine 2015 72 210 219 10.1016/j.cyto.2015.01.008 25703786 45. Bou M. Todorcevic M. Rodriguez J. Capilla E. Gutierrez J. Navarro I. Interplay of adiponectin, TNFalpha and insulin on gene expression, glucose uptake and PPARγ, AKT and TOR pathways in rainbow trout cultured adipocytes Gen. Comp. Endocrinol. 2014 205 218 225 10.1016/j.ygcen.2014.05.005 24846393 46. Chinetti G. Zawadski C. Fruchart J.C. Staels B. Expression of adiponectin receptors in human macrophages and regulation by agonists of the nuclear receptors PPARα, PPARγ, and LXR Biochem. Biophys. Res. Commun. 2004 314 151 158 10.1016/j.bbrc.2003.12.058 14715259 47. Hulsmans M. Geeraert B. Arnould T. Tsatsanis C. Holvoet P. PPAR agonist-induced reduction of Mcp1 in atherosclerotic plaques of obese, insulin-resistant mice depends on adiponectin-induced Irak3 expression PLoS ONE 2013 8 1281 10.1371/journal.pone.0062253 23620818 48. FASTX-Toolkit Available online: URL http://hannonlab.cshl.edu/fastx_toolkit/ (accessed on 2 February 2010) 49. Trapnell C. Pachter L. Salzberg S.L. TopHat: Discovering splice junctions with RNA-Seq Bioinformatics 2009 25 1105 1111 10.1093/bioinformatics/btp120 19289445 50. Trouw L.A. Seelen M.A. Duijs J.M. Benediktsson H. Van Kooten C. Daha M.R. Glomerular deposition of C1q and anti-C1q antibodies in mice following injection of antimouse C1q antibodies Clin. Exp. Immunol. 2003 132 32 39 10.1046/j.1365-2249.2003.02108.x 12653833 51. Huang da W. Sherman B.T. Lempicki R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources Nat. Protoc. 2009 4 44 57 10.1038/nprot.2008.211 19131956 52. Huang D.W. Sherman B.T. Lempicki R.A. Bioinformatics enrichment tools: Paths toward the comprehensive functional analysis of large gene lists Nucleic Acids Res. 2009 37 1 13 10.1093/nar/gkn923 19033363 53. Bindea G. Mlecnik B. Hackl H. Charoentong P. Tosolini M. Kirilovsky A. Fridman W.H. Pages F. Trajanoski Z. Galon J. ClueGO: A Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks Bioinformatics 2009 25 1091 1093 10.1093/bioinformatics/btp101 19237447 54. Cline M.S. Smoot M. Cerami E. Kuchinsky A. Landys N. Workman C. Christmas R. Avila-Campilo I. Creech M. Gross B. Integration of biological networks and gene expression data using Cytoscape Nat. Protoc. 2007 2 2366 2382 10.1038/nprot.2007.324 17947979 55. Saito R. Smoot M.E. Ono K. Ruscheinski J. Wang P.L. Lotia S. Pico A.R. Bader G.D. Ideker T. A travel guide to Cytoscape plugins Nat. Methods 2012 9 1069 1076 10.1038/nmeth.2212 23132118 56. Shannon P. Markiel A. Ozier O. Baliga N.S. Wang J.T. Ramage D. Amin N. Schwikowski B. Ideker T. Cytoscape: A software environment for integrated models of biomolecular interaction networks Genome Res. 2003 13 2498 2504 10.1101/gr.1239303 14597658 57. Livak K.J. Schmittgen T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2−∆∆C t Method Methods 2001 25 402 408 10.1006/meth.2001.1262 11846609

PMC005xxxxxx/PMC5000679.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081282ijms-17-01282ArticleMicroRNA-155 Mediates Augmented CD40 Expression in Bone Marrow Derived Plasmacytoid Dendritic Cells in Symptomatic Lupus-Prone NZB/W F1 Mice Yan Sheng 12Yim Lok Yan 1Tam Rachel Chun Yee 1Chan Albert 1Lu Liwei 3Lau Chak Sing 1*Chan Vera Sau-Fong 1*Taguchi Y-h. Academic EditorUI-TEI Kumiko Academic Editor1 Departments of Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China; [email protected] (S.Y.); [email protected] (L.Y.Y.); [email protected] (R.C.Y.T.); [email protected] (A.C.)2 School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China3 Departments of Pathology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China; [email protected]* Correspondence: [email protected] (C.S.L.); [email protected] (V.S.-F.C.); Tel.: +852-2255-3603 (C.S.L.); +852-2255-5995 (V.S.-F.C.); Fax: +852-2818-6474 (C.S.L. & V.S.-F.C.)06 8 2016 8 2016 17 8 128206 7 2016 02 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Systemic lupus erythematosus (SLE) is a chronic multi-organ autoimmune disease characterized by hyperactivated immune responses to self-antigens and persistent systemic inflammation. Previously, we reported abnormalities in circulating and bone marrow (BM)-derived plasmacytoid dendritic cells (pDCs) from SLE patients. Here, we aim to seek for potential regulators that mediate functional aberrations of pDCs in SLE. BM-derived pDCs from NZB/W F1 mice before and after the disease onset were compared for toll-like receptor (TLR) induced responses and microRNA profile changes. While pDCs derived from symptomatic mice were phenotypically comparable to pre-symptomatic ones, functionally they exhibited hypersensitivity to TLR7 but not TLR9 stimulation, as represented by the elevated upregulation of CD40, CD86 and MHC class II molecules upon R837 stimulation. Upregulated induction of miR-155 in symptomatic pDCs following TLR7 stimulation was observed. Transfection of miR-155 mimics in pre-symptomatic pDCs induced an augmented expression of Cd40, which is consistent with the increased CD40 expression in symptomatic pDCs. Overall, our results provide evidence for miR-155-mediated regulation in pDC functional abnormalities in SLE. Findings from this study contribute to a better understanding of SLE pathogenesis and ignite future interests in evaluating the molecular regulation in autoimmunity. systemic lupus erythematosusplasmacytoid dendritic cellsmicroRNAstoll-like receptor 7 ==== Body 1. Introduction Plasmacytoid dendritic cells (pDCs) constitute a unique subset of dendritic cells that have important immunoregulatory functions and play critical roles in autoimmunity. Similar to conventional or myeloid dendritic cells, pDCs upregulate the expressions of MHC class II antigen presentation molecules and T-cell costimulatory molecules such as CD80, CD86 and CD40 upon antigen stimulation, and serve as antigen presenting cells. Distinctively, pDCs specialize in producing type I interferon (IFN) [1]. pDCs express high levels of endosomal TLR7 and TLR9 which are the respective sensors for single stranded RNA and hypomethylated CpG DNA. Coupled with constitutive expression of the transcription factor interferon regulatory factor 7 (IRF7), pDCs are capable of rapidly secreting several hundred times more type I IFN than other leukocytes in response to viral or nucleic acids stimulations [2]. These unique features have granted pDCs a crucial role in the pathogenesis of systemic lupus erythematosus (SLE), a disease with multi-organ inflammatory manifestations mediated by autoantigens-autoantibodies immune complexes [3]. In human lupus disease, the role of pDCs is often inferred by the elevated serum level of IFN-α and upregulation of type I IFN-sensitive genes, which have been shown to correlate or associate with disease manifestations including the SLE hallmark anti-dsDNA autoantibodies, and the involvement of renal, hematological and central nervous systems [4,5,6]. Recent findings in animal studies, however, have specifically delineated pDCs’ functional contribution in SLE development. In the BXSB model, an early and transient depletion of pDCs before disease onset resulted in amelioration in lupus-associated pathology, which coincided with a reduced IFN-α/β-induced genes transcription [7]. Similar protective effects were also observed in Tcf4-haplodeficient Tlr7 transgenic and B6.Sle1.Sle3 lupus models in which the pDCs were functionally impaired by suppressing the E2-2 transcription factor [8]. In rheumatic diseases, a growing attention has been drawn to microRNAs (miRNAs) for their critical role in regulating immune cell functions [9,10]. These small non-coding RNA molecules post-transcriptionally regulate gene expression through complementary binding to their target messenger RNAs, leading to mRNA degradation or translation repression. Many regulatory miRNAs have been shown to link with SLE [11]. One interesting example is miR-146a. Its expression in peripheral blood mononuclear cells (PBMCs) of SLE patients negatively correlates with disease activity and IFN-sensitive genes expression [12]. Mechanistically, miR-146a acts as a negative regulator of type I IFN production by targeting multiple signaling components downstream of TLR7/9 and the retinoic acid-inducible gene-I pathways, thus its downregulated expression in lupus leukocytes promotes IFN-α production [12,13,14]. In addition, TLR7/9 and IFN-α stimulation of PBMCs induces miR-146a expression and apparently this negative feedback loop in IFN-α signaling pathway is dysfunctional in SLE patients [12]. Whether miR-146a-mediated IFN-α regulation is perturbed in lupus pDCs is unknown since there is no report on pDC-specific miRNA dysregulation in SLE so far. Nevertheless, a few recent studies have described the involvement of miR-155, miR-126, miR-29b, and miR-29c in regulating the functions of pDC in response to TLRs stimulation [15,16,17]. We previously demonstrated aberrations in frequencies, phenotypes and functions in DC subsets in SLE patients [18,19,20]. Intriguingly, bone marrow (BM)-derived pDCs from SLE patients were also found to have enhanced T-cell stimulatory ability and activated phenotypes [20], suggesting that abnormalities could originate from their precursors. Indeed, defects in BM and hematopoietic stem cells (HSCs) are also common in SLE patients [21,22,23]. Using the New Zealand Black/White F1 hybrid (NZB/W F1) lupus mouse model which mimics the spontaneous and multifactorial nature of human SLE disease, the present study aims to evaluate if SLE disease would impact on the generation and functional responses of BM-derived pDCs; and secondly to identify potential miRNA regulators in mediating functional irregularities in response to TLR stimulation. 2. Results 2.1. Lupus Disease Has Limited Impact on Plasmacytoid Dendritic Cells (pDC) Generation Potential of Bone Marrow (BM) Progenitor Cells in NZB/W F1 Mice To evaluate if the disease status in lupus has any effect on the development of pDCs, we isolated BM cells from the NZB/W F1 mice before (pre-symptomatic) and after (symptomatic) the onset of lupus symptoms for in vitro pDCs generation. In mice, the HSCs in BM are phenotypically identified as Lineage− (Lin−), Sca-1+ and c-Kit+ (LSK) cells and from which the progenitors of pDCs arise [24]. We analyzed the total numbers of BM cells as well as the frequencies of LSK cells from both groups of mice (Figure 1A). Consistently, an average of (2.4 ± 0.7) × 107 and (2.3 ± 0.8) × 107 of total BM cells could be harvested from the pre-symptomatic and symptomatic mice respectively. The LSK frequencies in the total BM cells varied from 0.27% to 0.85% in the pre-symptomatic mice, and from 0.18% to 1.7% in the symptomatic mice. Overall, lupus disease did not have significant effect on the total number of BM cells or the LSK frequency in NZB/W F1 mice. Cells derived from 8-day Flt3 ligand BM culture were evaluated for pDC generation. Mouse pDCs express the common DC marker CD11c and the B-cell lineage marker B220 [25]. Additionally, the BM stromal cell antigen 2 (BST2), also known as CD317 or PDCA-1, and the sialic acid-binding immunoglobulin (Ig)-like lectin H (Siglec-H) are specifically expressed on mouse pDCs at steady state [26,27]. Among these four markers, the former two can also be found on other immune cells while the expression of the latter two fluctuates upon stimulation. Therefore, a combination of CD11c, B220, CD317 and Siglec-H was used to identify pDCs from the heterogeneous BM culture (Figure 1B). The BM-derived DC culture constituted a mixture of cell populations expressing variable levels of these markers. Gating on the CD11c+B220+ cells, there remained a substantial proportion of cells that were CD317− and/or Siglec-H−. In contrast, the CD11c+B220hi subset were over 95% CD317+Siglec-H+, representing a more homogenous pDC population. Overall, approximately 20% of the cells generated from the Flt3 ligand-supplemented BM culture were CD11c+B220hi pDCs. The frequencies as well as the total numbers of pDCs generated from the pre-symptomatic and symptomatic BM cells were comparable (Figure 1C). 2.2. BM-Derived pDCs Display Similar Phenotypes Irrespective of Lupus Disease Stage Next, the surface expression of MHC class II and costimulatory molecules CD40 and CD86 were analyzed. Consistent with the reported pDC phenotypes at steady state [25,28], CD11c+B220hi pDCs expressed moderate level of MHC class II and very low levels for CD40 and CD86 (Figure 2A), and similar expression levels of these markers were observed between pre-symptomatic and symptomatic mice. Using quantitative RT-PCR, the expressions of Tlr7, Tlr9 and Irf7 were also compared as they are constitutively expressed by pDCs and are essential to pDC functions. As shown in Figure 2B, no significant difference in Tlr9 or Irf7 expression was observed between the two groups. Interestingly, the pDCs derived from symptomatic mice displayed a small but significant increase in Tlr7 expression when compared with the pre-symptomatic counterparts. However, further examination of TLR7 and TLR9 protein revealed high constitutive expression in these cells but no significant difference was observed between the two groups of pDCs (Figure 2C,D). In addition, the expression of three classical IFN-stimulated genes (ISGs) including IFN-induced protein with tetratricopeptide repeats 1 (Ifit1), IFN-inducible transmembrane protein 3 (Ifitm3) and 2′5′-oligoadenylate synthetase 1 (Oas1) were examined to check for any basal induction of type I IFN in lupus pDCs [29]. Among the three ISGs tested, no detectable expression was observed for Ifit1 and Oas1 (data not shown). Expression of Ifitm3 was detectable but again there was no significant difference between the two groups of pDCs (Figure 2B). Seemingly, these results appear to contradict to the established role of type I IFN in SLE. A recent report has demonstrated in disease NZB/W F1 mice the elevated expression of IFN signature genes such as Mx1, Ifit2 and Cxcl10 in pDCs isolated from spleen and BM, thus supporting the involvement of type I IFN in this murine model [30]. Likely, these differentiated pDCs have been continuously exposed to and stimulated by TLR7/9-activating immune complexes in vivo. In contrast, in our system, BM pDC precursors were cultured in vitro with Flt3L, which is known to promote pDC differentiation but not activation. It is likely that these BM-derived pDCs did not spontaneously produce IFNα in culture and thus had limited basal or increase in the expression of the ISGs. Overall, these data suggested that in vitro pDC development from BM cells was not grossly affected by the development of lupus. 2.3. BM-Derived pDCs from Symptomatic Lupus Mice Show Heightened TLR7-Mediated Antigen Presentation and Costimulatory Molecules Expressions To examine the functional responses of pDCs, purified CD11c+B220hi cells were activated by TLR agonists and compared. Upon R837 (TLR7 agonist) stimulation, MHC class II, CD40 as well as CD86 were clearly upregulated in both groups; and importantly, the induction was significantly higher in symptomatic mice (Figure 3). As depicted in Figure 3, pDCs from symptomatic group expressed augmented levels of MHC class II, and the percentage of cells expressing CD40 and CD86 was also significantly higher. In contrast to TLR7 response, no significant difference was observed in the induction of MHC class II, CD40 or CD86 expression upon TLR9 activation via CpG (Supplementary Figure S1). Additionally, TLR7/9-mediated IL-6 and IFN-α production was also compared. Similar amount of interleukin (IL)-6 was secreted by R837-activated pDCs from the two groups of mice (Supplementary Figure S2). Surprisingly, no detectable IFN-α was observed in the culture supernatant from R837-stimulated pDCs, whereas these pDCs were capable of producing IFN-α upon CpG stimulation. R837 is an imidazoquinoline amine analog to guanosine that commonly used for evaluation of TLR7 mediated IFN-α response in human cells and likely, R837 is inefficient in activating type I IFN production in mice [31,32]. Apart from disease status, the age difference between the pre-symptomatic and symptomatic NZB/W F1 mice could have contributed to the differential TLR7 response. To evaluate this, R837 responses of pDCs derived from parental non-lupus NZW young (12–16 weeks old) and old (>36 weeks old) mice were compared (Figure 4). Upregulation of MHC class II, CD40 as well as CD86 on pDCs from both young and old NZW mice were induced and no significant differences were observed between the two groups. These results suggested that the elevated R837 response in symptomatic NZB/W F1 group was not likely attributed by the age factor. 2.4. TLR7 Mediated Overexpression of miR-155 in Lupus pDCs Contributes to the Heightened Cd40 Expression To identify potential regulator(s) involved in the enhanced TLR7 response in lupus pDCs, we examined miRNA expression profiles upon R837 stimulation and compared between pre-symptomatic and symptomatic mice. The miRNA rodent set array included a total of 750 targets, among which 107 were expressed in pDCs. The basal expressions of miRNAs in unstimulated pDCs from symptomatic mice were first compared with the pre-symptomatic one and no consistent difference in specific miRNA was observed in three independent sets of profiling experiments (Supplementary Figure S3). Next, the relative quantity (RQ) of the expressed miRNAs in R837-activated pDCs was compared between the pre-symptomatic and symptomatic groups (Figure 5A). With a two-fold change as cutoff, six specific miRNAs were found differentially expressed. In particular, miR-155 was the most strongly and consistently upregulated upon activation, and with a significantly higher induction in the symptomatic group. A higher induction of miR-132 was also observed in symptomatic pDCs. The other four miRNAs, miR-339-3p, miR-694, miR-421 and miR-103 were downregulated in symptomatic group with only marginal changes in pre-symptomatic one. Since the differential induction of miRNAs was most pronounced in miR-155, we decided to focus on this miRNA by further validation with independent sets of stimulation samples using miR-155 specific qPCR assay. Consistent with the array data, miR-155 was expressed at similar levels in both groups of unstimulated pDCs (Figure 5B), while R837-induced miR-155 upregulation was significantly higher in the symptomatic group (23.44 ± 4.7 vs. 32.69 ± 4.5, Figure 5C). Furthermore, the R837-induced overexpression of miR-155 in pDCs of symptomatic NZB/W F1 mice was likely related to disease development since this difference was not observed when comparing young and old non-lupus NZW mice (Supplementary Figure S4). In order to study whether the elevated miR-155 expression contributes to pDC phenotypes observed in symptomatic mice, miR-155 mimics and scramble/non-targeting controls were transfected to pDCs derived from pre-symptomatic mice and Cd40 expression was measured. A significant increase was observed for Cd40 expression upon miR-155 overexpression (Figure 6A). This is consistent with the findings of an enhanced CD40 induction in symptomatic pDCs in response to R837 stimulation in association with a higher miR-155 induction (Figure 3 and Figure 5). In parallel, the expression of SH2-containing inositol phosphatase Ship1, a known primary target of miR-155 [33], was downregulated with a significant negative correlation with Cd40 in the miR-155 overexpressing cells (Figure 6B). 3. Discussion In this study, we have shown that in vitro derivation of pDCs from BM of symptomatic and pre-symptomatic NZB/W F1 mice was comparable phenotypically, including the expression of TLR7 and TLR9. However, hypersensitivity of symptomatic pDCs to TLR7 stimulation was illustrated by their increased upregulation of costimulatory molecules including CD40, CD86 as well as MHC class II. Screening of miRNAs in pDCs revealed an enhanced induction of miR-155 in symptomatic mice in response to TLR7 stimulation. Upon miR-155 overexpression, Cd40 expression was significantly upregulated with a negative correlation to the miR-155 primary target Ship1 expression. Overall, our findings suggest that the elevated miR-155 induction may contribute to the enhanced TLR7-induced CD40 expression in pDCs derived from BM of lupus mice. In SLE patients, BM abnormalities are not rare and reported features include hypocellularity, necrotic alterations, reduced CD34+ HSC frequency as well as elevation of markers like CD95, CD123 and CD166 [21,22,23]. Such aberrations could be resulted from the disease itself and/or treatment using immunosuppressive drugs. Based on our previous findings in BM-derived pDCs from patients [20], we speculated that hypersensitivity in lupus pDCs could originate from abnormalities in BM precursors as a result of the disease. In NZB/W F1 mice, despite the lack of gross aberrations in BM cells and HSC frequency in symptomatic mice, BM-derived pDCs were indeed more sensitive to TLR7 stimulation and showed higher CD40, CD86 and MHC class II induction. These observations are consistent with the phenotypes observed in Flt3L-induced BM DCs from SLE patients [20]. Furthermore, myeloid DCs derived from BM of old disease NZB/W F1 females also showed differential response to estrogen modulation, leading to higher inflammatory cytokines production when compared with young pre-disease mice [34]. Thus, it is likely that the common DC progenitors in BM of lupus patients are modulated by disease parameters (e.g., perturbed cytokines levels) to give rise to hypersensitive DCs that exit to the periphery. In lupus, activation of pDCs can be constantly triggered by self-nucleic acids in complex with autoantibodies or nuclear-binding proteins released from necrotic cells as well as activated neutrophils [35,36] via TLRs. The hyper-responsiveness to TLR7 stimulation in BM-derived pDCs in symptomatic lupus-prone mice likely represents one of the pathological attributes for SLE development. The pathogenic role of pDCs in lupus has been supported by recent evidence from selective depletion or functional blockade of this cell population in lupus-prone mouse models [7,8,37]. It was demonstrated that pDC deficiency in early disease development or the impairment of its function had beneficial effects including reduced splenomegaly, anti-nuclear autoantibody production, reduced glomerulonephritis and decreased levels of ISGs in kidney tissues [7,8]. In the tape-stripping model, depletion of pDCs protected NZB/W F1 mice from developing chronic skin lesions and so did the treatment of TLR7 and TLR9 inhibitors [37], suggesting that TLR7 and TLR9 signaling is important for lupus development. Indeed, the duplicated Tlr7 gene, and hence the overexpressed receptor, in the BXSB male mice promoted a biased response of autoreactive B cells toward nuclear self-antigens [38]. Introducing the Tlr7-bearing yaa gene segment to B6.Sle1 mice also accelerated autoimmunity, in which the disease severity positively correlated with the Tlr7 expression level [39]. The pathogenic role of TLR9, however, is controversial. In fact, disparate contribution of TLR7 and TLR9 in SLE pathogenesis has been reported. In B6.Nba2 congenic lupus mice, Tlr9 deletion led to accelerated lupus with an increased production of anti-nuclear antibodies and augmented lupus nephritis, while disease progression in the Tlr7/9 double-deficient mice was restored to a comparable or even slightly improved level as the parental strain [40]. Our finding is in agreement with the critical involvement of TLR7 signaling pathway in pDC malfunction in SLE pathogenesis while little implication can be derived from TLR9 response. Despite having little baseline difference, the expression levels of CD40, CD86 and MHC class II molecules were further augmented in symptomatic BM-derived pDCs upon TLR7 stimulation. Consistently, splenic and lymph node pDCs in lupic NZB/W F1 mice have also been shown to display elevated expression of costimulatory molecules, including CD40 and CD86 in pDCs when compared with mice prior to disease onset [41,42], suggesting that the BM-derived pDCs bear similarities to the peripheral circulating pDCs. Similarly, a recent report by Zhou et al. has also demonstrated in disease NZB/W F1 mice a higher expression of MHC-II and CD80 in ex vivo splenic and BM pDCs when compared with pre-disease mice [30]. Likely, these peripheral pDCs in disease mice have been continuously exposed to and stimulated by TLR7 stimulating immune complexes in vivo, thus leading to the hyperactivated phenotypes upon ex vivo examination. Functionally, it is conceivable that a higher expression of CD86 and MHC class II could lead to better T cell stimulation by symptomatic pDCs, and indeed this has been observed in SLE patients [19,20]. Enhanced CD40 expression in pDCs may promote SLE development through its interaction with CD40L. Both soluble and cell-bound forms of CD40L have been shown to increase significantly in SLE patients [43,44]. Interestingly, it has been shown that activated platelets in lupus patient sera interact with pDCs through CD40L-CD40 signaling and potentiate type I IFN secretion by pDCs [45]. CD40 signaling can also synergize TLR9-induced type I IFN response in pDCs [46]. With the increased pDC hypersensitivity we found here and reported elsewhere, we believe that pDCs could potentially function as an accelerator to promote SLE development, at least in part through the increased CD40 expression upon TLR7 activation. Previous miRNA profiling analyses in SLE were mainly done in total PBMCs from patients or total splenocytes and lymphocytes from murine models in unstimulated condition [12,47,48]. Here, changes in the miRNA expression were examined in purified pDCs upon TLR7 stimulation, and the heightened miR-155 induction was shown to mediate enhanced Cd40 expression in symptomatic pDCs. This finding is in line with another study, which reported a decrease in MHC class II, CD40, and CD86 expressions in miR-155 knockdown Kupffer cells [49]. It is not clear how miR-155 mediates an increase in CD40 expression in pDCs. We observed a significant negative correlation between Cd40 and Ship1 expression in pDCs upon miR-155 modulation. SHIP1 is a negative immuno-regulator and has been shown to suppress TLR4 response via MyD88 in DCs [50]. Consistently, a recent report shows that miR-155 upregulation in DCs can lead to breaking of T cell tolerance by negative regulation of SHIP1 [51]. However, pertaining to this study, whether there is a functional correlation between CD40 and SHIP1 in pDCs is yet to be determined. Our findings are consistent with the pro-autoimmune nature of miR-155, at least in the murine models. Notably, an upregulation of miR-155 was demonstrated in total splenocytes in different lupus mouse models, and was further shown in the splenic T and B lymphocyte compartments [48]. In MRL/lpr lupus mice, an enhanced miR-155 expression in CD4+CD25+Foxp3+ Treg cells was found in association with a compromised Treg suppressive function [52]; and the deletion of miR-155 in these mice alleviated lupus-like disease [53]. Besides, miR-155 appears to play a pathogenic role in other autoimmune diseases too. Silencing or depleting miR-155 in the mice protected them from developing experimental autoimmune encephalomyelitis or collagen-induced arthritis [54,55,56]. The miR-155-deficient mice in both of these models had impaired Th17 cell development and reduced proinflammatory Th1 and Th17 cytokines, suggesting a pathogenic contribution of miR-155 in T-cell mediated autoimmunity. Our study also suggests a pathogenic role of miR-155 in mediating pDC abnormality in the NZB/W F1 lupus model. We identified a novel regulation of CD40 expression by miR-155, by which may contribute to the hyperactivated TLR7 response in lupus pDCs. Pertaining to our findings, more questions arise and warrant further exploration. First, it is not clear how the symptomatic BM-derived pDCs mediate higher TLR7 responses. We evaluated TLR7 as well as its downstream transcription factor Irf7 expression levels and found no difference. Thus, other TLR7 downstream signaling networks (e.g., NF-κB and c-JNK pathways) should be examined. So far, the regulation of TLR7 (and TLR9) signaling network in pDCs remains elusive and can be influenced by other pDC-specific receptor-mediated signaling pathways [57]. A potential cross-regulation between TLR7 signaling and miR-155 would worth studying. In human pDCs, miR-155 and miR-155* are induced at different kinetics upon R837 stimulation [15]. Interestingly, miR-155 seems to act as a negatively regulator in type I IFN production by targeting TAB2 while miR-155* augments IFNα/β production by targeting IRAKM. In murine pDCs, R837 stimulation also upregulates miR-155 but cannot elicit IFNα production. We also found that miR-155 overexpression had minimal impact on Tab2 expression (data not shown). In our miRNA screening assay, miR-155-3p (star form) was not included, and whether its expression changes upon TLR7 stimulation remains to be determined. It is likely that if a cross regulatory mechanism between murine TLR7 and miR-155 exists, it would be different from that found in human cells. In contrast to costimulatory molecules expression, it is not clear why the R837-induced IL-6 appears not be affected in symptomatic pDCs. For TLR9 stimulation, different CpG oligonucleotides elicit disparate functional outcomes depending on the nature of the phosphorothioated backbone as well as the presence of palindromic sequences. Class A CpG is an efficient type I IFN inducer while class B is more superior in inducing costimulatory molecules and cytokines expression in B cells, each triggering distinct regulatory pathways [46]. It is possible that different TLR7 agonists (e.g., ss-PolyU) may yield different response outcomes in our study system. More studies are therefore needed to evaluate if a different regulatory mechanism exists for TLR7-mediated cytokine modulation in lupus pDCs. 4. Experimental Section 4.1. Mice The NZW/LacJ and NZB/BlNJ strains were purchased from The Jackson Laboratory (Bar Harbour, ME, USA). Breeding of the NZB/W F1 hybrid mice was carried out at The Laboratory Animal Unit, The University of Hong Kong. Licenses specifying experiments involved in this study were obtained from the Department of Health, Hong Kong. All experimental protocols and procedures were approved by the University Committee on the Use of Live Animals in Teaching and Research (CULATR 2213-10, approved on 5 August 2010 and CULATR 2449-11, approved on 25 May 2011). Lupus disease development in female NZB/W F1 mice and age-matched female NZW control mice was monitored. Serum levels of anti-dsDNA IgG autoantibodies were measured bi-weekly by ELISA. Proteinuria was measured weekly by Albutix stripes (Bayer HealthCare, Leverkusen, Germany). Aged (>25 weeks old) NZB/W F1 mice with persistent proteinuria (+++, 3 mg/mL or above for over 2 weeks) and anti-dsDNA IgG levels higher than two standard deviations (SD) over the mean of the age-matched NZW mice were considered as symptomatic. Young (8–15 weeks old) NZB/W F1 mice with anti-dsDNA levels comparable to non-lupus NZW control mice and negative for proteinuria were considered pre-symptomatic. 4.2. pDC Culture, Isolation and Stimulation BM-derived pDCs were cultured as previously described with some modifications [24,28]. Briefly, total BM cells were harvested by flushing humeri, tibias and femurs of mice. Red blood cells were lysed using ACK buffer (0.15 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA) and BM cells were seeded in 12-well plates at 5 × 106 cells/mL in 10% fetal bovine serum in complete IMDM supplemented with 100 ng/mL of mouse Flt3 ligand (Peprotech, Rocky Hill, NJ, USA). In some experiments, hematopoietic stem cells were enriched before culture using the mouse lineage cell depletion kit (Miltenyi Biotec, Bergisch Gladbach, Germany). On day-8, cells were stained with appropriate antibodies and pDCs were purified by fluorescence-activated cell sorting using the Beckman Coulter MoFlo™ XDP cell sorter for subsequent experiments. Purified pDCs were seeded at 5 × 104 cells/well in 96-well plates with 5 µg/mL of R837 (Imiquimod), 1 μM of CpG (InvivoGen, San Diego, CA, USA) or medium only (unstimulated) for 48 h, or specified otherwise. Stimulated cells were stained for activation markers and analyzed using the BD FACSCanto II Analyzer at Faculty Core Facilities, Li Ka Shing Faculty of Medicine, The University of Hong Kong. For staining of TLR7 and TLR9, day-8-BM cultures were first surface-stained for CD11c and B220, followed by fixation with 4% paraformaldehyde and permeabilization with 0.2% saponin. Intracellular staining was performed in the presence of 0.2% saponin throughout. The following anti-mouse monoclonal antibodies (BD Biosciences or Affymetrix eBiosciences, Franklin Lakes, NJ, USA) were used in this study: Sca-1-FITC (clone: D7); c-Kit-PE (2B8); CD11c-FITC (HL3); B220-APC or PECy7 (RA3-6B2); Siglec-H-PE (eBio440c); CD317-APC (eBio927); CD40-PE (3/23); CD86-PE or PECy7 (GL1); MHC class II—PECy7 (M5/114.15.2); TLR7-PE (A94B10) and TLR9-PE (J15A7). 4.3. Transfection of miRNA Mimics Double-stranded miRNA mimics or scramble controls (GenePharma, Shanghai, China) were transfected into BM-derived pDCs using the Lipofectamine 2000® reagent (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) at 100 nM for 4 h at 37 °C. Total RNAs of the transfected cells were extracted using TRI reagent (Sigma-Aldrich, St. Louis, MO, USA) and stored at −80 °C for subsequent experiments. 4.4. Quantitative Real-Time RT-PCR Quantitative gene expression analyses were performed following the Minimal Information for Publication of Quantitative Real-time PCR Experiments (MIQE) guidelines. Total cellular RNAs were extracted using TRI Reagent® (Sigma-Aldrich, St. Louis, MO, USA) and then reverse transcribed into cDNAs with the ThermoScript™ RT-PCR system (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s recommendation. Amplification of cDNAs was carried out using the StepOnePlus™ system and StepOne Software v2.3 (Applied Biosystems®, Thermo Fisher Scientific, Waltham, MA, USA) with KAPA SYBR® Fast qPCR kit (Kapa Biosystems, Wilmington, MA, USA) following the thermal conditions at: 95 °C for 3 min and 40 cycles of 95 °C, 60 °C and 72 °C for 20 s each. Relative quantity (RQ) of gene-of-interest was normalized with β-actin and calculated using the formula: RQ = 2−[ΔCq (test sample) − ΔCq (reference sample)]. The primers for specific targets tested in this study are listed in Table 1. The amplification efficiency of each pair of primers was validated and was comparable to that of the endogenous control. In all experiments, no template control and melting curve analyses of the amplification products were included. Independent quantification of miR-155 was carried out using the miRNA assay (assay ID 002571 specific for mmu-miR-155) purchased from Applied Biosystems® (Thermo Fisher Scientific, Waltham, MA, USA). Reactions and thermal conditions of quantitative real-time PCR for miR-155 were performed following manufacturer’s recommendation. 4.5. MicroRNA Profiling Extracted total RNAs were reverse transcribed using the Megaplex™ RT primers, Rodent pool A and B sets and the TaqMan® miRNA reverse transcription kit. Subsequently, cDNAs were pre-amplified using the Megaplex™ PreAmp Primers and the TaqMan® PreAmp Master Mix. Pre-amplified products were mixed with TaqMan® Universal PCR master mix (No AmpErase® UNG) in suggested ratio following the manufacturer’s recommendation and then loaded to the 384-well TaqMan® Array Rodent miRNA A and B Cards. The reactions were performed on 7900HT Real-Time PCR System at The Centre for Genomic Sciences, The University of Hong Kong. Results were analyzed using Sequence Detection Systems v2.4 and RQ Manager V1.21. The reagents and analysis software used for the arrays were purchased from Applied Biosystems® (Thermo Fisher Scientific, Waltham, MA, USA). The expression of each miRNA was normalized with the mammalian U6 internal control, and RQ was calculated using the formula: RQ = 2−[ΔCq (test sample) − ΔCq (reference sample)]. 4.6. Statistical Analysis Unless specified, all experimental data were presented as mean values ± SD. Statistical significance was determined by unpaired two-tailed Student’s t-test, and correlation analysis by linear regression using GraphPad Prism (GraphPad, Inc., San Diego, CA, USA). 5. Conclusions Using the NZB/W F1 murine lupus model, we unraveled a miR-155-associated hyperactive TLR7 response in BM-derived pDCs after the onset of lupus. The expression of Cd40 is upregulated upon overexpression of miR-155 in pDCs. In essence, the current study highlights the elevated pDC responses to TLR7 stimulation in lupus and suggests a potentially pathogenic role mediated by miR-155. We present here the connection between miRNA and the hyperactivation of pDCs including the upregulation of CD40. Findings in this study shed light on the potential roles of pDCs in the pathogenesis of SLE. Acknowledgments This study was supported by the Hong Kong Research Grant Council General Research Fund (770213). We thank Faculty Core Facility of The University of Hong Kong for the use of the flow cytometer. We are grateful to Nan Shen and Xinfang Huang for their helpful advices on miRNA analyses, Yong-jun Liu for his insightful comments on the pDC work. Thanks are also due to Derek Wong for his technical support in cell sorting. Supplementary Materials The following are available online at www.mdpi.com/1422-0067/17/8/1282/s1. Click here for additional data file. Author Contributions Sheng Yan was responsible for conducting majority of the experiments; She collected data, performed statistical analysis and drafted the manuscript; Lok Yan Yim, Rachel Chun Yee Tam and Albert Chan conducted part of the animal work and tissue culture work; Liwei Lu contributed to the initial planning of this study and gave critical revision to the manuscript; Chak Sing Lau and Vera Sau-Fong Chan were responsible for the conception of the study, designing experimental approach and the overall coordination of the study; They supervised data collection and analysis, and revised the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Gross development of pDCs is not affected by lupus. (A) Total numbers of bone marrow (BM) cells and frequencies of hematopoietic stem cells, marked by Lineage− (Lin−) Sca-1+c-Kit+ (LSK), from pre-symptomatic (Pre-sym) and symptomatic (Sym) mice were compared. Collective data from more than 10 independent experiments, totaling n = 16–20 in the respective charts shown. Each symbol represents one mouse; (B) Cells generated from Flt3-ligand-supplemented BM culture at day-8 were analyzed. The expression of CD11c and B220 on live-gated cells as well as the expression of CD317 and Siglec-H on CD11c+B220+- or CD11c+B220hi-gated cells are shown in representative contour plots. Isotype antibodies control was used for the gating of CD317 and Siglec-H positive cells. Numbers shown are the percent of double positive cells; (C) Collective data from at least 3 independent experiments on the total numbers and frequencies of CD11c+B220hi pDCs in the pre-symptomatic and symptomatic BM cultures are presented. Each symbol represents one mouse (n = 12). Figure 2 Bone marrow-derived pDCs from pre-symptomatic (Pre-sym) and symptomatic (Sym) lupus mice have similar phenotypes. (A) Expressions of MHC class II, CD40 and CD86 in CD11c+B220hi pDCs derived from day-8 BM culture were analyzed. Representative plots of n = 6–7 is shown; (B) Expression of Irf7, Tlr9, Tlr7 and Ifitm3 of purified pDCs derived from day-8 BM culture were examined by real time PCR, and were shown as relative quantities (RQ) with reference to the average expression of corresponding genes in the pre-symptomatic pDCs. Collective data of n = 6–7 mice were shown; (C) Expressions of TLR7 and TLR9 in BM-derived pDCs on day-8 were analyzed. Numbers shown are percentages of positive cells. Shaded histograms represent isotype antibody staining controls; (D) Collective data comparing the mean fluorescence intensity (MFI) of TLR7 and 9 expressed in pre-symptomatic and symptomatic pDCs. Each symbol represents one individual mouse (n = 6). Data were collected from at least three independent experiments. Bar: mean values; * p ≤ 0.05 (unpaired two-tailed Student’s t-test). Figure 3 pDCs response to TLR7 stimulation is augmented in symptomatic mice. Purified BM-derived pDCs from pre-symptomatic (Pre-sym) or symptomatic (Sym) NZB/W F1 mice were treated with (+) or without (−) 5 µg/mL of R837 for 48 h. Expression of MHC class II, CD40 and CD86 were examined and shown as representative histograms in (A). Grey: Isotype; dotted line: unstimulated; solid line: R837 activated. Collective data illustrating changes in the mean fluorescence intensity (MFI) of MHC class II, CD40 and CD86 or percentages (%) of the latter two are shown in (B,C) respectively. Each symbol represents one experimental mouse. For expression of CD40, n = 12–13; CD86, n = 4–5; MHC class II, n = 8. Data were collected from at least three independent experiments. Bar: mean value. * p ≤ 0.05; ** p ≤ 0.01; (unpaired two-tailed Student’s t-test). Figure 4 pDCs response to TLR7 stimulation is not affected in young and old NZW mice. Purified BM-derived pDCs from NZW mice age-matched with the pre-symptomatic (A, shown as Young) and symptomatic (B, shown as Old) NZB/W NZB/W F1 mice were treated with (+) or without (−) 5 µg/mL of R837 for 48 h. Expression of MHC class II, CD40 and CD86 were examined and shown as representative histograms in (A). Grey: Isotype; dotted line: unstimulated; solid line: R837 activated. Collective data illustrating changes in the mean fluorescence intensity (MFI) of MHC class II or percentages (%) of CD40 and CD86 are shown in (B). Each symbol represents one experimental mouse, n = 5. Data were collected from at least three independent experiments. Bar: mean value. Figure 5 Induction of miR-155 is enhanced in symptomatic pDCs upon TLR7 activation. (A) Differential miRNAs expression profiles in R837-activated pDCs relative to unstimulated pDCs were analyzed. Relative quantities (RQ) of expressed miRNAs in symptomatic (Sym) group were plotted against pre-symptomatic (Pre-sym) group. The two bold lines indicate a cutoff of less than two-fold differences in expression. Arrows indicate miRNAs that are differentially expressed, whose mean RQ of three independent repeats are presented in the table. Expression of miR-155 was independently verified by real time PCR on separate samples of (B) unstimulated pDCs; and (C) R837-stimulated pDCs from pre-symptomatic and symptomatic mice. Each symbol represents sample from one experimental mouse. Data were collected from three independent experiments, n = 5 for each group of F1 mice. Bar: mean value. ** p ≤ 0.01 (unpaired two-tailed Student’s t-test). Figure 6 Transfection of miR-155 induces upregulation of Cd40 in pre-symptomatic pDCs. (A) Pre-symptomatic pDCs were transfected with miR-155 mimics or scramble mimics control (control). The expression levels of Cd40 in miR-155-transfected pDCs were examined by real time PCR and presented as relative quantity (RQ) to their respective mean expression in the control group. Bar: mean value. *** p ≤ 0.001 (unpaired two-tailed Student’s t-test); (B) Correlation of Cd40 induction vs. Ship1 suppression upon transfection of miR-155 mimics was plotted. The expression levels are presented as RQ relative to their respective average of controls in each group. The coefficient of correlation (R2) and p value were determined by linear regression. Each symbol represents one experimental mouse. Data were collected from at least three independent experiments (n = 9). ijms-17-01282-t001_Table 1Table 1 List of primers used in this study. Gene Sequence Accession No. Tm Pdt Location β-actin Fwd TTG CTG ACA GGA TGC AGA AG NM_007392.3 58.2 147 1039–1058 Rev TGA TCC ACA TCT GCT GGA AG – 57.3 – 1185–1166 Irf7 Fwd GAT CTT CAA GGC CTG GGC TGT GG NM_016850.3 57.1 220 611–633 Rev TCC AAG CTC CCG GCT AAG TT – 55.0 – 830–811 Ifitm3 Fwd GAT CGG CTT CTG TCA GAA CTA NM_025378.2 57.2 154 209–229 Rev TTC CGA TCC CTA GAC TTC ACG GA – 62.5 – 362–340 Tlr7 Fwd TGT TAC TAT TCC ATA CCT GGC CAC NM_133211.4 60.1 179 2640–2663 Rev GGT GAC TTG TTG TCA TAA CTA CC – 57.1 – 2818–2796 Tlr9 Fwd CAA CAT GGT TCT CCG TCG AA NM_031178.2 58.2 243 103–122 Rev TTG TGC AGG TGG TGG ATA CGG T – 64.2 – 345–324 Cd40 Fwd TTG TTG ACA GCG GTC CAT CT NM_011611.2 59.6 154 112–131 Rev CTG AGT CAC ATG GGT GGC AT – 60.0 – 265–246 Ship1 Fwd CCA GGG CAA GAT GAG GGA GA NM_001110193.2 60.98 195 2766–2785 Rev GGA CCT CGG TTG GCA ATG TA – 60.04 – 2960–2941 Fwd: forward primer; Rev: reverse primer; Tm: melting temperature; Pdt: size of the amplification product in base pair. Final concentration of primer used in qPCR was 200 nM. ==== Refs References 1. Liu Y.J. IPC: Professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors Annu. Rev. Immunol. 2005 23 275 306 10.1146/annurev.immunol.23.021704.115633 15771572 2. Gilliet M. Cao W. Liu Y.J. Plasmacytoid dendritic cells: Sensing nucleic acids in viral infection and autoimmune diseases Nat. Rev. Immunol. 2008 8 594 606 10.1038/nri2358 18641647 3. Tsokos G.C. Systemic lupus erythematosus N. Engl. J. Med. 2011 365 2110 2121 10.1056/NEJMra1100359 22129255 4. Bengtsson A.A. Sturfelt G. Truedsson L. Blomberg J. Alm G. Vallin H. Ronnblom L. Activation of type i interferon system in systemic lupus erythematosus correlates with disease activity but not with antiretroviral antibodies Lupus 2000 9 664 671 10.1191/096120300674499064 11199920 5. Baechler E.C. Batliwalla F.M. Karypis G. Gaffney P.M. Ortmann W.A. Espe K.J. Shark K.B. Grande W.J. Hughes K.M. Kapur V. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus Proc. Natl. Acad. Sci. USA 2003 100 2610 2615 10.1073/pnas.0337679100 12604793 6. Bennett L. Palucka A.K. Arce E. Cantrell V. Borvak J. Banchereau J. Pascual V. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood J. Exp. Med. 2003 197 711 723 10.1084/jem.20021553 12642603 7. Rowland S.L. Riggs J.M. Gilfillan S. Bugatti M. Vermi W. Kolbeck R. Unanue E.R. Sanjuan M.A. Colonna M. Early, transient depletion of plasmacytoid dendritic cells ameliorates autoimmunity in a lupus model J. Exp. Med. 2014 211 1977 1991 10.1084/jem.20132620 25180065 8. Sisirak V. Ganguly D. Lewis K.L. Couillault C. Tanaka L. Bolland S. D’Agati V. Elkon K.B. Reizis B. Genetic evidence for the role of plasmacytoid dendritic cells in systemic lupus erythematosus J. Exp. Med. 2014 211 1969 1976 10.1084/jem.20132522 25180061 9. O’Connell R.M. Rao D.S. Chaudhuri A.A. Baltimore D. Physiological and pathological roles for micrornas in the immune system Nat. Rev. Immunol. 2010 10 111 122 10.1038/nri2708 20098459 10. Luo X. Ranade K. Talker R. Jallal B. Shen N. Yao Y. Microrna-mediated regulation of innate immune response in rheumatic diseases Arthritis Res. Ther. 2013 15 210 10.1186/ar4194 23582400 11. Yan S. Yim L.Y. Lu L. Lau C.S. Chan V.S. Microrna regulation in systemic lupus erythematosus pathogenesis Immune Netw. 2014 14 138 148 10.4110/in.2014.14.3.138 24999310 12. Tang Y. Luo X. Cui H. Ni X. Yuan M. Guo Y. Huang X. Zhou H. de Vries N. Tak P.P. Microrna-146a contributes to abnormal activation of the type I interferon pathway in human lupus by targeting the key signaling proteins Arthritis Rheum. 2009 60 1065 1075 10.1002/art.24436 19333922 13. Charrier E. Cordeiro P. Cordeau M. Dardari R. Michaud A. Harnois M. Merindol N. Herblot S. Duval M. Post-transcriptional down-regulation of toll-like receptor signaling pathway in umbilical cord blood plasmacytoid dendritic cells Cell Immunol. 2012 276 114 121 10.1016/j.cellimm.2012.04.010 22578600 14. Hou J. Wang P. Lin L. Liu X. Ma F. An H. Wang Z. Cao X. MicroRNA-146a feedback inhibits RIG-I-dependent type I IFN production in macrophages by targeting TRAF6, IRAK1, and IRAK2 J. Immunol. 2009 183 2150 2158 10.4049/jimmunol.0900707 19596990 15. Zhou H. Huang X. Cui H. Luo X. Tang Y. Chen S. Wu L. Shen N. miR-155 and its star-form partner miR-155* cooperatively regulate type i interferon production by human plasmacytoid dendritic cells Blood 2010 116 5885 5894 10.1182/blood-2010-04-280156 20852130 16. Agudo J. Ruzo A. Tung N. Salmon H. Leboeuf M. Hashimoto D. Becker C. Garrett-Sinha L.A. Baccarini A. Merad M. The miR-126–VEGFR2 axis controls the innate response to pathogen-associated nucleic acids Nat. Immunol. 2014 15 54 62 10.1038/ni.2767 24270517 17. Hong Y. Wu J. Zhao J. Wang H. Liu Y. Chen T. Kan X. Tao Q. Shen X. Yan K. miR-29b and miR-29c are involved in toll-like receptor control of glucocorticoid-induced apoptosis in human plasmacytoid dendritic cells PLoS ONE 2013 8 1282 10.1371/journal.pone.0069926 23894561 18. Jin O. Kavikondala S. Sun L. Fu R. Mok M.Y. Chan A. Yeung J. Lau C.S. Systemic lupus erythematosus patients have increased number of circulating plasmacytoid dendritic cells, but decreased myeloid dendritic cells with deficient CD83 expression Lupus 2008 17 654 662 10.1177/0961203308089410 18625638 19. Jin O. Kavikondala S. Mok M.Y. Sun L. Gu J. Fu R. Chan A. Yeung J. Nie Y. Lau C.S. Abnormalities in circulating plasmacytoid dendritic cells in patients with systemic lupus erythematosus Arthritis Res. Ther. 2010 12 R137 10.1186/ar3075 20618924 20. Nie Y.J. Mok M.Y. Chan G.C. Chan A.W. Jin O.U. Kavikondala S. Lie A.K. Lau C.S. Phenotypic and functional abnormalities of bone marrow-derived dendritic cells in systemic lupus erythematosus Arthritis Res. Ther. 2010 12 R91 10.1186/ar3018 20478074 21. Voulgarelis M. Giannouli S. Tasidou A. Anagnostou D. Ziakas P.D. Tzioufas A.G. Bone marrow histological findings in systemic lupus erythematosus with hematologic abnormalities: A clinicopathological study Am. J. Hematol. 2006 81 590 597 10.1002/ajh.20593 16823830 22. Sun L.Y. Zhou K.X. Feng X.B. Zhang H.Y. Ding X.Q. Jin O. Lu L.W. Lau C.S. Hou Y.Y. Fan L.M. Abnormal surface markers expression on bone marrow cd34+ cells and correlation with disease activity in patients with systemic lupus erythematosus Clin. Rheumatol. 2007 26 2073 2079 10.1007/s10067-007-0621-2 17447103 23. Wanitpongpun C. Teawtrakul N. Mahakkanukrauh A. Siritunyaporn S. Sirijerachai C. Chansung K. Bone marrow abnormalities in systemic lupus erythematosus with peripheral cytopenia Clin. Exp. Rheumatol. 2012 30 825 829 23101438 24. Brasel K. De Smedt T. Smith J.L. Maliszewski C.R. Generation of murine dendritic cells from flt3-ligand-supplemented bone marrow cultures Blood 2000 96 3029 3039 11049981 25. Nakano H. Yanagita M. Gunn M.D. Cd11c+ b220+ Gr-1+ cells in mouse lymph nodes and spleen display characteristics of plasmacytoid dendritic cells J. Exp. Med. 2001 194 1171 1178 10.1084/jem.194.8.1171 11602645 26. Blasius A.L. Giurisato E. Cella M. Schreiber R.D. Shaw A.S. Colonna M. Bone marrow stromal cell antigen 2 is a specific marker of type I IFN-producing cells in the naive mouse, but a promiscuous cell surface antigen following IFN stimulation J. Immunol. 2006 177 3260 3265 10.4049/jimmunol.177.5.3260 16920966 27. Zhang J. Raper A. Sugita N. Hingorani R. Salio M. Palmowski M.J. Cerundolo V. Crocker P.R. Characterization of siglec-H as a novel endocytic receptor expressed on murine plasmacytoid dendritic cell precursors Blood 2006 107 3600 3608 10.1182/blood-2005-09-3842 16397130 28. Gilliet M. Boonstra A. Paturel C. Antonenko S. Xu X.L. Trinchieri G. O′Garra A. Liu Y.J. The development of murine plasmacytoid dendritic cell precursors is differentially regulated by FLT3-ligand and granulocyte/macrophage colony-stimulating factor J. Exp. Med. 2002 195 953 958 10.1084/jem.20020045 11927638 29. Schoggins J.W. Rice C.M. Interferon-stimulated genes and their antiviral effector functions Curr. Opin. Virol. 2011 1 519 525 10.1016/j.coviro.2011.10.008 22328912 30. Zhou Z. Ma J. Xiao C. Han X. Qiu R. Wang Y. Zhou Y. Wu L. Huang X. Shen N. Phenotypic and functional alterations of pDCs in lupus-prone mice Sci. Rep. 2016 6 20373 10.1038/srep20373 26879679 31. Diebold S.S. Massacrier C. Akira S. Paturel C. Morel Y. Reis e Sousa C. Nucleic acid agonists for toll-like receptor 7 are defined by the presence of uridine ribonucleotides Eur. J. Immunol. 2006 36 3256 3267 10.1002/eji.200636617 17111347 32. Rajagopal D. Paturel C. Morel Y. Uematsu S. Akira S. Diebold S.S. Plasmacytoid dendritic cell-derived type i interferon is crucial for the adjuvant activity of toll-like receptor 7 agonists Blood 2010 115 1949 1957 10.1182/blood-2009-08-238543 20065291 33. O’Connell R.M. Chaudhuri A.A. Rao D.S. Baltimore D. Inositol phosphatase ship1 is a primary target of miR-155 Proc. Natl. Acad. Sci. USA 2009 106 7113 7118 10.1073/pnas.0902636106 19359473 34. Jiang B. Sun L. Hao S. Li X. Xu Y. Hou Y. Estrogen modulates bone marrow-derived DCs in sle murine model-(NZB × NZW) F1 female mice Immunol. Investig. 2008 37 227 243 10.1080/08820130801973328 18389442 35. Bave U. Magnusson M. Eloranta M.L. Perers A. Alm G.V. Ronnblom L. FCΓRIIa is expressed on natural IFN-α-producing cells (plasmacytoid dendritic cells) and is required for the IFN-α production induced by apoptotic cells combined with lupus IgG J. Immunol. 2003 171 3296 3302 10.4049/jimmunol.171.6.3296 12960360 36. Lande R. Ganguly D. Facchinetti V. Frasca L. Conrad C. Gregorio J. Meller S. Chamilos G. Sebasigari R. Riccieri V. Neutrophils activate plasmacytoid dendritic cells by releasing self-DNA-peptide complexes in systemic lupus erythematosus Sci. Transl. Med. 2011 3 73ra19 10.1126/scitranslmed.3001180 21389263 37. Guiducci C. Tripodo C. Gong M. Sangaletti S. Colombo M.P. Coffman R.L. Barrat F.J. Autoimmune skin inflammation is dependent on plasmacytoid dendritic cell activation by nucleic acids via TLR7 and TLR9 J. Exp. Med. 2010 207 2931 2942 10.1084/jem.20101048 21115693 38. Pisitkun P. Deane J.A. Difilippantonio M.J. Tarasenko T. Satterthwaite A.B. Bolland S. Autoreactive b cell responses to rna-related antigens due to Tlr7 gene duplication Science 2006 312 1669 1672 10.1126/science.1124978 16709748 39. Subramanian S. Tus K. Li Q.Z. Wang A. Tian X.H. Zhou J. Liang C. Bartov G. McDaniel L.D. Zhou X.J. A Tlr7 translocation accelerates systemic autoimmunity in murine lupus Proc. Natl. Acad. Sci. USA 2006 103 9970 9975 10.1073/pnas.0603912103 16777955 40. Santiago-Raber M.L. Dunand-Sauthier I. Wu T. Li Q.Z. Uematsu S. Akira S. Reith W. Mohan C. Kotzin B.L. Izui S. Critical role of Tlr7 in the acceleration of systemic lupus erythematosus in Tlr9 -deficient mice J. Autoimmun. 2010 34 339 348 10.1016/j.jaut.2009.11.001 19944565 41. Colonna L. Dinnall J.A. Shivers D.K. Frisoni L. Caricchio R. Gallucci S. Abnormal costimulatory phenotype and function of dendritic cells before and after the onset of severe murine lupus Arthritis Res. Ther. 2006 8 R49 10.1186/ar1911 16507174 42. Gleisner M.A. Reyes P. Alfaro J. Solanes P. Simon V. Crisostomo N. Sauma D. Rosemblatt M. Bono M.R. Dendritic and stromal cells from the spleen of lupic mice present phenotypic and functional abnormalities Mol. Immunol. 2013 54 423 434 10.1016/j.molimm.2013.01.011 23428837 43. Mehta J. Genin A. Brunner M. Scalzi L.V. Mishra N. Beukelman T. Cron R.Q. Prolonged expression of CD154 on CD4 T cells from pediatric lupus patients correlates with increased CD154 transcription, increased nuclear factor of activated T cell activity, and glomerulonephritis Arthritis Rheum. 2010 62 2499 2509 10.1002/art.27554 20506525 44. Goules A. Tzioufas A.G. Manousakis M.N. Kirou K.A. Crow M.K. Routsias J.G. Elevated levels of soluble CD40 ligand (sCD40l) in serum of patients with systemic autoimmune diseases J. Autoimmun. 2006 26 165 171 10.1016/j.jaut.2006.02.002 16621447 45. Duffau P. Seneschal J. Nicco C. Richez C. Lazaro E. Douchet I. Bordes C. Viallard J.F. Goulvestre C. Pellegrin J.L. Platelet CD154 potentiates interferon-α secretion by plasmacytoid dendritic cells in systemic lupus erythematosus Sci. Transl. Med. 2010 2 47ra63 10.1126/scitranslmed.3001001 20811042 46. Kerkmann M. Rothenfusser S. Hornung V. Towarowski A. Wagner M. Sarris A. Giese T. Endres S. Hartmann G. Activation with CpG-A and CpG-B oligonucleotides reveals two distinct regulatory pathways of type I IFN synthesis in human plasmacytoid dendritic cells J. Immunol. 2003 170 4465 4474 10.4049/jimmunol.170.9.4465 12707322 47. Dai Y. Huang Y.S. Tang M. Lv T.Y. Hu C.X. Tan Y.H. Xu Z.M. Yin Y.B. Microarray analysis of microRNA expression in peripheral blood cells of systemic lupus erythematosus patients Lupus 2007 16 939 946 10.1177/0961203307084158 18042587 48. Dai R. Zhang Y. Khan D. Heid B. Caudell D. Crasta O. Ahmed S.A. Identification of a common lupus disease-associated microRNA expression pattern in three different murine models of lupus PLoS ONE 2010 5 1282 10.1371/journal.pone.0014302 21170274 49. Li J. Gong J. Li P. Li M. Liu Y. Liang S. Gong J. Knockdown of microRNA-155 in kupffer cells results in immunosuppressive effects and prolongs survival of mouse liver allografts Transplantation 2014 97 626 635 10.1097/TP.0000000000000061 24531847 50. Cekic C. Casella C.R. Sag D. Antignano F. Kolb J. Suttles J. Hughes M.R. Krystal G. Mitchell T.C. Myd88-dependent ship1 regulates proinflammatory signaling pathways in dendritic cells after monophosphoryl lipid a stimulation of Tlr4 J. Immunol. 2011 186 3858 3865 10.4049/jimmunol.1001034 21339365 51. Lind E.F. Millar D.G. Dissanayake D. Savage J.C. Grimshaw N.K. Kerr W.G. Ohashi P.S. Mir-155 upregulation in dendritic cells is sufficient to break tolerance in vivo by negatively regulating ship1 J. Immunol. 2015 195 4632 4640 10.4049/jimmunol.1302941 26447227 52. Divekar A.A. Dubey S. Gangalum P.R. Singh R.R. Dicer insufficiency and microRNA-155 overexpression in lupus regulatory t cells: An apparent paradox in the setting of an inflammatory milieu J. Immunol. 2011 186 924 930 10.4049/jimmunol.1002218 21149603 53. Thai T.H. Patterson H.C. Pham D.H. Kis-Toth K. Kaminski D.A. Tsokos G.C. Deletion of microRNA-155 reduces autoantibody responses and alleviates lupus-like disease in the Faslpr mouse Proc. Natl. Acad. Sci. USA 2013 110 20194 20199 10.1073/pnas.1317632110 24282294 54. O'Connell R.M. Kahn D. Gibson W.S. Round J.L. Scholz R.L. Chaudhuri A.A. Kahn M.E. Rao D.S. Baltimore D. MicroRNA-155 promotes autoimmune inflammation by enhancing inflammatory T cell development Immunity 2010 33 607 619 10.1016/j.immuni.2010.09.009 20888269 55. Murugaiyan G. Beynon V. Mittal A. Joller N. Weiner H.L. Silencing microRNA-155 ameliorates experimental autoimmune encephalomyelitis J. Immunol. 2011 187 2213 2221 10.4049/jimmunol.1003952 21788439 56. Bluml S. Bonelli M. Niederreiter B. Puchner A. Mayr G. Hayer S. Koenders M.I. van den Berg W.B. Smolen J. Redlich K. Essential role of microRNA-155 in the pathogenesis of autoimmune arthritis in mice Arthritis Rheum. 2011 63 1281 1288 10.1002/art.30281 21321928 57. Bao M. Liu Y.J. Regulation of Tlr7/9 signaling in plasmacytoid dendritic cells Protein Cell 2013 4 40 52 10.1007/s13238-012-2104-8 23132256

PMC005xxxxxx/PMC5000680.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081283ijms-17-01283ArticleZeb1 Is a Potential Regulator of Six2 in the Proliferation, Apoptosis and Migration of Metanephric Mesenchyme Cells Gu Yuping 1†Zhao Ya 12†Zhou Yuru 13†Xie Yajun 1Ju Pan 1Long Yaoshui 1Liu Jianing 1Ni Dongsheng 1Cao Fen 1Lyu Zhongshi 1Mao Zhaomin 1Hao Jin 1Li Yiman 1Wan Qianya 1Kanyomse Quist 1Liu Yamin 1Ren Die 1Ning Yating 1Li Xiaofeng 1Zhou Qin 1Li Bing 14*Lemarié Anthony Academic Editor1 Division of Molecular Nephrology and The Creative Training Center for Undergraduates, The Ministry of Education Key Laboratory of Clinical Diagnostics, School of Laboratory Medicine, Chongqing Medical University, Chongqing 400016, China; [email protected] (Y.G.); [email protected] (Y.Z.); [email protected] (Y.Z.); [email protected] (Y.X.); [email protected] (P.J.); [email protected] (Y.L.); [email protected] (J.L.); [email protected] (D.N.); [email protected] (F.C.); [email protected] (Z.L.); [email protected] (Z.M.); [email protected] (J.H.); [email protected] (Y.L.); [email protected] (Q.W.); [email protected] (Q.K.); [email protected] (Y.L.); [email protected] (D.R.); [email protected] (Y.N.); [email protected] (X.L.); [email protected] (Q.Z.)2 Department of Laboratory Medicine, The First Hospital of Xi’an, Xi’an 710002, China3 Undergraduates Class of 2012 Entry, The Fifth Clinical College of Medicine, Chongqing Medical University, Chongqing 400016, China4 The First Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China* Correspondence: [email protected]; Tel.: +86-23-6848-5785† These authors contributed equally to this work. 06 8 2016 8 2016 17 8 128317 4 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Nephron progenitor cells surround around the ureteric bud tips (UB) and inductively interact with the UB to originate nephrons, the basic units of renal function. This process is determined by the internal balance between self-renewal and consumption of the nephron progenitor cells, which is depending on the complicated regulation networks. It has been reported that Zeb1 regulates the proliferation of mesenchymal cells in mouse embryos. However, the role of Zeb1 in nephrons generation is not clear, especially in metanephric mesenchyme (MM). Here, we detected cell proliferation, apoptosis and migration in MM cells by EdU assay, flow cytometry assay and wound healing assay, respectively. Meanwhile, Western and RT-PCR were used to measure the expression level of Zeb1 and Six2 in MM cells and developing kidney. Besides, the dual-luciferase assay was conducted to study the molecular relationship between Zeb1 and Six2. We found that knock-down of Zeb1 decreased cell proliferation, migration and promoted cell apoptosis in MM cells and Zeb1 overexpression leaded to the opposite data. Western-blot and RT-PCR results showed that knock-down of Zeb1 decreased the expression of Six2 in MM cells and Zeb1 overexpression contributed to the opposite results. Similarly, Zeb1 promoted Six2 promoter reporter activity in luciferase assays. However, double knock-down of Zeb1 and Six2 did not enhance the apoptosis of MM cells compared with control cells. Nevertheless, double silence of Zeb1 and Six2 repressed cell proliferation. In addition, we also found that Zeb1 and Six2 had an identical pattern in distinct developing phases of embryonic kidney. These results indicated that there may exist a complicated regulation network between Six2 and Zeb1. Together, we demonstrate Zeb1 promotes proliferation and apoptosis and inhibits the migration of MM cells, in association with Six2. Zeb1Six2metanephric mesenchyme cellscell proliferationcell apoptosiscell migration ==== Body 1. Introduction The kidney is a vital and complex organ that accomplishes multiple physiological functions in the body, such as metabolic waste excretion, water and electrolyte homeostasis control, acid-base balance and blood pressure maintenance. Nephrons is the major functional units for kidney to perform these tasks [1,2]. During kidney development, nephrons’ formation is mainly decided by the interaction of the ureteric bud (UB) and metanephric mesenchyme (MM) cells [3,4]. It begins at embryonic day 10.5 (E10.5) to E11.0 when the UB starts growing and branching under the induction of MM cells [5], and MM cells aggregate around the branched tips of UB. Then, the MM cells and UB form nephrons by two cellular processes: MET and tubule genesis [6]. These reports indicate that MM cells are the original cells of nephron generation and inductively interact with UB in kidney development [7,8]. In addition, self-renewal (proliferation) and consumption of MM cells determine the formation and complement of nephrons [8,9]. Consequently, the proliferation, apoptosis and migration of MM cells become especially important in the study of kidney development. Zeb1, a transcription factor containing 1117 amino acids, is an EMT marker in cancer metastasis of some tissues including kidney [10,11]. Zeb1 promotes EMT through suppression of CDH1 (encoding E-cadherin, an epithelial maker) and the microRNA-200 [10]. This process activates transforming growth factor-β1 (TGF-β1) signaling pathway and trigger cancer cell proliferation, invasiveness and stemness out of control [11,12]. In addition, Zeb1 also plays a critical role in animal organ development [13], cartilage development [14] and regulation of mesenchymal cell proliferation [15]. As an example, loss of Zeb1 results in MET and reduce the proliferation of progenitor cells at the sites of developmental defects in mouse embryos [15]. However, there is little reference about the concrete role of Zeb1 in the cellular regulation of MM cells. Six2, a MET-marker, maintains cap mesenchyme multipotent nephron progenitor cells at an undifferentiated state, promotes MM cell proliferation and restrains cell apoptosis during kidney development [8,9,16]. Deficiency of Six2 depletes cap mesenchyme progenitors, ectopic differentiation, and severe kidney hypoplasia and dysplasia [17,18]. However, EMT and MET are two distinct cellular processes that respectively function in cancer metastasis and development. Zeb1 and Six2 are the main markers of these two processes, respectively, but whether there exists a relationship between Zeb1 and Six2 in MM cells remains unknown. Here, we found that Zeb1 promoted cell proliferation and migration, but suppressed cell apoptosis in MM cells, and Zeb1 can bind to Six2 promoter to regulate its transcription by dual-luciferase assay and bioinformatics analysis. Our RT-PCR and Western blot results showed that Zeb1 increased the expression of Six2. Both of Zeb1 and Six2 had a high expression level in embryonic kidney at E13.5 and E18.5. These discoveries provided theoretical evidence for further studying the role of Zeb1—regulated Six2 in kidney development. 2. Results 2.1. Zeb1 Is Highly Conserved and Homologous across Different Mammalians To analyze the conservative of Zeb1 protein, we used CLUSTALW online [19]. The Zeb1 protein is highly conservative and homologous in evolution among mammal species such as Chimpanzee, Human, Rhesus monkey, Dog, Giant panda, Norway rat and House mouse (Figure 1A,B). Additionally, we compared the three types of Zeb1 function domains (seven C2H2 zinc finger, three Zinc finger double domain and a Homeodomain) in NCBI Protein Database [20]. Then, we found that the structure of Zeb1 protein across those mammal species is also highly conserved (Figure 1C). 2.2. Zeb1 Promotes the Proliferation and Migration but Inhibits the Apoptosis of MM Cells As noted above, the function of Zeb1 in metanephric mesenchymal cells remains unclear during kidney development, so we wonder whether Zeb1 plays a crucial role in the regulation of these cells. To investigate whether Zeb1 affects the proliferation, apoptosis and migration of MM cells, mK3 cells were used as a cell model. mK3 cells were transfected with Zeb1 overexpression or knock-down (Zeb1-shRNA) vector followed by EdU assay. As illustrated in Figure 2A,B and Figure A1, the ratio of EdU positive cells to the whole cells was promoted in mK3 cells transfected with the overexpression vector compared with the control. However, the ratio was reduced while Zeb1 was knocked down in mK3 cells. Meanwhile, to find out the effect of Zeb1 on cell apoptosis of the mK3 cells, we detected the apoptosis of mK3 cells transfected with Zeb1 overexpression vector, overexpression control vector, Zeb1-shRNA or control shRNA (pLKO.1). Then, mK3 cells transfected with Zeb1-shRNA or control shRNA (pLKO.1) were treated with 0.1 μM dexamethasone before cell apoptosis analysis. Overexpression of Zeb1 decreased the rate of mK3 cell apoptosis compared with the control cells (Figure 3A,B). Besides, Zeb1 silence and dexamethasone treatment increased the apoptosis rate of mK3 cells and knockdown of Zeb1 increased cell apoptosis induced by dexamethasone compared with the respective control cells (Figure 3C,D). These results demonstrate that Zeb1 inhibits MM cell apoptosis. In addition, we treated mK3 cells with the same methods used in the proliferation assay. Contrarily, we performed Wound Healing Assay and found that deficiency of Zeb1 resulted in the reduction of wound healing percentage compared with the control (Figure 4A,B). In contrast, mK3 cells transfected with the overexpression vectors shows promotion of cell healing percentage (Figure 4A,C). To some degree, these data concludes that the migration of mK3 cells is advanced by Zeb1. Moreover, to determine the efficiency of Zeb1 overexpression or knock-down, we conducted the RT PCR and Western-blot. As expected, Zeb1 was overexpressed and knocked down efficiently (Figure 2C,D). All these findings indicates that Zeb1 can promotes the proliferation of mK3 cells. 2.3. Zeb1 Binds to Six2 Promoter and Up-Regulates Six2 in Metanephric Mesenchymal Cell To find out the mechanism of the cellular regulation in addition to metanephric mesenchymal cells, we performed bioinformatics prediction of the interaction between Zeb1 and Six2. As exhibited in Figure 5A, Zeb1 binding motifs towards the potential promoter of Six2 were conserved among mammal species such as Human, Chimpanzee, Mouse, Norway rat, Dog and Rhesus. This result shows it is possible that Zeb1 can regulate the expression of Six2 gene by mediating the transcription of Six2. Therefore, we carried out dual-luciferase assay, RT-PCR and Western blot to verify Zeb1 affected the expression of Six2. We found that overexpression of Zeb1 significantly promotes Six2 promoter reporter activity (Figure 5B). Additionally, the deficiency of Zeb1 up-regulated both mRNA and protein expression of Six2 compared with the control (Figure 5C). In contrast, Zeb1 overexpression caused the up-regulation of Six2 at both mRNA and protein level (Figure 5D). It has been reported that Six2 suppression inhibits cell proliferation, but promotes cell apoptosis in MM cells and up-regulation of Six2 promotes cell migration [8,17]. These previous studies can explain the phenotype of Zeb1. 2.4. Zeb1 Regulates Cell Proliferation and Apoptosis of MM Cells by Working with Six2 To clarify the interaction between Zeb1 and Six2 in metanephric mesenchymal cells’ regulation, mK3 cells were transfected with negative control shRNA (pLKO.1), Zeb1-shRNA, Six2-shRNA or both Zeb1-shRNA & Six2-shRNA followed by EdU assay and apoptosis detection. As shown in Figure 6A,B, mK3 cells introduced with either Zeb1-shRNA or Six2-shRNA made cell proliferation significantly reduced, compared with the negative control cells. Moreover, mK3 cells transfected with both of Zeb1-shRNA and Six2-shRNA presented a smaller ratio compared with the mK3 cells transfected with one of the two shRNA. These results declare that down-regulation of Six2 restrains MM cell proliferation indeed and this suppression was enhanced when Zeb1 was knocked down with Six2 down-regulated (Figure 5C). What is more, we performed cell apoptosis detection in mK3 cells treated the same as in Figure 6. For cell apoptosis, when mK3 cells were introduced with either Zeb1-shRNA or Six2-shRNA, the rate of cell apoptosis was higher than the control cells. However, it is found that mK3 cells introduced with both of the two shRNAs had a lower apoptosis rate, compared with that transfected by only one shRNA (Figure 7A,B). From these data, it is implied that there is interaction between Zeb1 and Six2 in the apoptosis regulation of MM cells. The complicated relationship leads to the different apoptosis trends between single-knock-down of Zeb1, Six2 and double-knock-down of both. Meanwhile, to confirm Zeb1 mediates cell phenotypes above by affecting the expression of Six2, we quantified the mRNA expression of Six2 by RT-PCR. As illustrated in Figure 6C, the knock-down of Zeb1 or Six2 was all efficient and Zeb1 knock-down significantly down-regulated Six2 expression at the mRNA level. 2.5. The Expression Profile of Zeb1 and Six2 in Kidney Development As mentioned above, we found Zeb1 mediated cell proliferation, apoptosis and migration in MM cells associated with Six2, which was fundamental for kidney development. Thus, we wonder whether Zeb1 or Six2 was expressed differently in variant stages during kidney development. Accordingly, we conducted RT-PCR using cDNAs of embryonic mouse kidney at different times (E11.5, E12.5, E13.5, E14.5, E16.5, E18.5). As diagramed in Figure 8A,B, the mRNA expression of Zeb1 and Six2 changed with the kidney development stage, and the variant tendency of Zeb1 and Six2 was identical except for E16.5. This finding suggests that Zeb1 quite possibly plays an important role during kidney development in relation to Six2. Furthermore, we conducted the expression prediction of Zeb1 and Six2 by GUDMAP online to analyze Zeb1 and Six2 mRNA expression at different times, which presented variant mRNA levels (Figure 8C). 2.6. c-Myc Is Up-Regulated by Zeb1 in MM Cells To verify the mechanism that Zeb1 mediates MM cells by regulating Six2, we checked the protein expression of c-Myc, a transcription factor that has been reported to have interaction with Six2 during nephrogenesis [21]. The Western-blot data showed that the deficiency of Zeb1 down-regulated c-Myc (Figure 9A,B) and Zeb1 overexpression led to the up-regulation of c-Myc (Figure 9C,D). 3. Discussion Zeb1, a transcription factor, is highly conserved and homologous in evolution among different species at the protein level. Here, we first reported the role of Zeb1 in kidney development and its cellular function in MM cells. Knock-down of Zeb1 decreased cell proliferation and migration, but increased cell apoptosis in MM cells (Figure 2, Figure 3 and Figure 4 and Figure A1). Deficiency of Zeb1 down-regulated the expression of Six2 and c-Myc (Figure 5C,D and Figure 9A,B). Moreover, overexpression of Zeb1 promotes Six2 promoter reporter activity in luciferase assay (Figure 5B). On the contrary, overexpression of Zeb1 leads to the opposite results (Figure 2, Figure 3 and Figure 4, Figure 5C,D and Figure 9C,D). However, double knock down of Zeb1 and Six2 decreased cell proliferation more seriously (Figure 6A,B), but did not enhance cell apoptosis in MM cells compared with Six2 or Zeb1 knockdown alone (Figure 7A,B). Metanephric mesenchyme (MM) cells include the nephron progenitor cells marked by Six2 and form nephrons by balancing self-renewal and consumption [7,9]. Numerous transcription factors are involved in this process, such as c-Myc and Six2 [8,17]. In this study, we focused on a transcription factor Zeb1, which is widely expressed in development and cancer and acts as an inducer of EMT and a regulator of cell migration in cancer cells [10]. In addition, mutation of Zeb1 can induce MET and reduce the proliferation of progenitor cells at defects sites of developing mouse embryos [22]. Besides, knock-down of Zeb1 inhibits cell growth via activating the apoptosis pathway [23], and TXNIP/miR-200/Zeb1/E-cadherin signaling pathway is reported to function in beta cell apoptosis [24], which suggests that Zeb1 plays a crucial role in cell apoptosis. Meanwhile, our results suggested that the resistant of Zeb1-depleted mK3 cells to dexamethasone are decreased in comparison to control cells, which implied that Zeb1 is involved in the apoptosis process that is induced by agent dexamethasone (Figure 3C,D). So, we speculate that Zeb1 may be a regulator of MET and had an association with MET marker-Six2 in MM cells. Interestingly, knock-down of Zeb1 decreased cell proliferation and migration but increased cell apoptosis in MM cells as reported in other cells [10,22]. Furthermore, the expression of Six2 was down-regulated when Zeb1 was knocked down. Down-regulation of Six2 represses MM cell proliferation and migration [16,18] but enhances cell apoptosis [8], and overexpression of Zeb1 leads to the opposite results, reasonably. These suggested that Six2 is involved in the process that Zeb1 mediated cell proliferation, migration and apoptosis in MM cells. As is known to us, Six2 is an essential gene in kidney development and its expression is variant during renal development [25]. We further measured the mRNA expression of Six2 and Zeb1 in embryonic mouse kidney in vitro (Figure 8A,B). Although their expression pattern is not completely the same, both of Six2 and Zeb1 had high expression at the E11.5 to E13.5 stage, which is a key period in the mouse kidney development process. It has been reported that, in the process of mammalian cell apoptosis, caspases mediate 500–1000 proteins’ cleavage and generate many protein fragments, which can increase the probability of cell apoptosis [26,27,28]. Nevertheless, there are some reports that N-terminal truncated LynΔN that is generated by caspase cleavage has been demonstrated with anti-apoptotic roles [29]. Whatever the concrete role of this fragment in cell apoptosis is, these reports provide a direction for our subsequent study about Zeb1, Six2 and cell apoptosis in embryonic mouse kidney development. Here, our results demonstrated that Zeb1 depletion decreased Six2 expression and enhanced cell apoptosis in mk3 cells, but the connection between Six2 down-regulation and cell apoptosis promotion remains unknown. Because of the limit of time and materials, we did not test the protein expression of Zeb1 and Six2 in embryonic mouse kidney in the present study, but we will detect their protein level and further study the connection between Six2 down-regulation and cell apoptosis promotion during embryonic mouse kidney development. Interestingly, the recent study shows that Zeb1 can positively regulate the mTOR pathway and maintain its threshold level in wild-type MEFs that is required for Akt-S473 generation [30], and mTOR pathway can rescue many developmental defects of embryos due to the ESCO2 mutant [31], which indicates that the mTOR pathway is essential in development. All these studies indicate that mTOR pathway may be involved in Zeb1-regulated renal development. To find out the regulatory mode between Zeb1 and Six2, we combined the bioinformatics analysis with the dual-luciferase assay results and found that Zeb1 could bind to Six2 promoter to enhance the luciferase activity. Therefore, these may prove that Zeb1 binds to Six2 promoter to promote Six2 gene transcription and further promotes gene expression, which is identical to the up-regulation of Six2 in Zeb1 overexpressed cells. All these indicated that Zeb1 might up-regulate Six2 by transcriptional regulation. Perhaps post-translational modification is another regulation mode, which provides us valuable insight for further research on the regulation mechanism between Zeb1 and Six2. To make the mechanism clear, we detected the expression of c-Myc, a member of the Myc family of proto-oncogenes expressed in the MM progenitors and which are essential for progenitor cell proliferation and kidney growth [32,33]. Then, we found that a deficiency of Zeb1 down-regulated c-Myc and overexpression of Zeb1 up-regulated c-Myc in MM cells (Figure 9A–D). Zeb1 represses the expression of miR-34a and miR-34b/c, and miR-34a conversely down-regulated Zeb1 and c-Myc to decrease the migration and invasion of cancer cells [34]. Furthermore, Six2 mediates the nuclear translocation of Eya1, then Eya1 switches Myc between phosphorylation and dephosphorylation states to regulate MM cell multipotency, proliferation, apoptosis, and so on [21]. Deletion of c-Myc can reduce Six2-positive stem/progenitor populations and decrease cell proliferation [33]. Therefore, these evidences demonstrated Six2 and c-Myc may be new targets of Zeb1. However, the role of c-Myc needs to be clarified in further study, and elucidating the regulation network among them appears especially important in kidney development. Based on our results and the reported data, we formed a work model to illustrate the regulation between Zeb1 and Six2 during embryonic renal development (Figure 10). Zeb1 promotes MM cell proliferation (cell renewal) and cell migration but inhibits cell apoptosis (cell consumption) in association with Six2 up-regulation and c-Myc down-regulation. The MM cell renewal, consumption and migration are essential for the induced interaction between MM cells and UB. These contribute to UB branching morphogenesis and MET and elongation for nephrons’ formation in embryonic kidney development. Besides, Zeb1 and Six2 have similar expression patterns at different stages of developing kidney (Figure 8A,B). These results further suggested that Zeb1 is a potential regulator of Six2 and c-Myc in the proliferation, migration and apoptosis of MM cells. 4. Materials and Methods 4.1. Bioinformatic Analysis The species and evolutionary conservation of Zeb1 proteins was analyzed using Multiple Sequence Alignment by CLUSTALW online [19] and edited in BioEdit software. The amino acid sequences used for analysis were acquired from the NCBI GenPept Database [20]. Moreover, gene expression pattern (Developing Kidney MOE430 Microarray Analysis) of Zeb1 and Six2 was obtained from the GUDMAP Expression Database [35]. Additionally, the motifs where Zeb1 may bind to Six2 potential promoter (2000bp upstream of transcription start site) was predicted on the JASPAR Database [36], the six promoter sequence (NCBI reference sequence NC_000083.6) was retrieved from the Genbank Database. 4.2. Plasmids Construction The m.Zeb1 CDS was amplified from the cDNA of C57BL/6 embryonic mouse kidney by PCR using the forward primer: 5′-tagcgtttaaactta GATCATGGCGGATGGCCCCAGG TGTAAGC-3′, the reverse primer: 5′-tggactagtggatcc CTAAGCTTCATTTGTCTTCTCTTCA-3′. The amplification was inserted into the HindIII/BamHI site of pCDNA3.1 (+) (Invitrogen, Carlsbad, CA, USA) using the ligation-independent cloning (LIC) [37,38]. Then the m.Zeb1 CDS fragment was cut by the NheI/BamHI restriction enzyme and was cloned to the pCDH-CMV-MCS-EF1-copGFP vector (a lentivirus overexpression vector) at the same restriction site using T4 ligation cloning. The m.Zeb1 shRNA and m.Six2 shRNA sequences were acquired from the SIGMA ALORICH [39] with m.Zeb1 target: 5′-CCGGGTCAGTAAACATACCTA-3′; m.Six2 target: 5′-CCTCCACAAGAATGAAAGCGT-3′. Then anneal the shRNA oligo containing gene target and clone the annealed fragment to pLKO.1 vector at AgeI/EcoRI site according to the protocols [40] pGL3-basic vector (Promega, Madison, WI, USA) and pRL-SV40 was purchased from Promega. The promoter of murine Six2 was amplified from C57BL/6 mouse genomic DNA by PCR using the forward primer: CGTGCTAGCCCGGGCTATTTCCCAGGTCCCCTGGAATCCT and the reverse primer: CCGGAATGCCAAGCTCTTGCAGCTTTTTTAATAATATTAT. Then the fragment was inserted into the XhoI/HindIII site (upstream of fly luciferase gene) of the pGL3-luciferase vector to create pGL3-Six2 promoter-luciferase using LIC. All of these recombined vectors were sequenced and aligned in the NCBI Nucleotide Blast Database. 4.3. Cell Culture and Transfection The 293T cell and mK3 cell (a cell line cloned from mouse and representing the un-induced differentiation stage of metanephric mesenchyme [41,42]) were cultured in DMEM medium (Gibico, Carlsbad, CA, USA) with 10% FBS (Gibico, Carlsbad, CA, USA), 1000 units/mL of penicillin and 1000 μg/mL of streptomycin in 37 °C, 100% humidity and 10% CO2. The G401 cell were cultured in the same condition except the different base medium ATCC-formulated McCoy’s 5a Medium Modified (Cat. No.30-2007, ATCC). When the mK3 cell grew to 60% confluent in 6-well plates, lentivirus mediated cell transfection was performed, with the assist of polybrene (8 μg/mL). The lentivirus was packed in HEK293T cell line with 10 μg recombined vector in 10 cm dish, then it was harvested to infect mK3 cells after 48 h. The pCDH-CMV-MCS-EF1-copGFP-m.Zeb1 CDS, pLKO.1-m.Zeb1 shRNA, pLKO.1-m.Six2 shRNA and the corresponding control vectors were transfected alone while the m.Zeb1 shRNA was co-transfected with the m.Six2 shRNA in another well. 4.4. RNA Extraction and RT-PCR The total RNA was isolated using Trizol reagent (Invitrogen, Carlsbad, CA, USA) from 48 h post-transfected mK3 cells and the kidneys of C57BL/6 embryonic mice at different developmental stage. The cDNA was synthesized using the Invitrogen RT kit according to the manufacturer’s protocol (Invitrogen). The mRNA expression level of Zeb1, Six2 was detected by RT-PCR at 60 °C annealing temperature with Zeb1 real time PCR sense primer: 5′-CGAGTCAGATGCAGAAAATGAGCAA-3′ and the anti-sense primer: 5′-ACCCAGACTGCGTCACATGTCTT-3′; Six2 sense primer: 5′-GCCTGCGAGCACCTCCACAAGAAT-3′ and the anti-sense primer: 5′-CACCGACTTGCCACTGCCATTGAG-3′. The expression were normalized to the internal control (18s or GAPDH) and were quantified by Gray Scan using Image J software. 4.5. Western Blotting mK3 cells that were transfected for 48 h in 6-well plates were washed with PBS, pH 7.4 three times. Then, the cells were lysed with 300 μL of 1% SDS lysis buffer. The lysed complex were collected and boiled at 95 °C in a water bath for 10 min followed by centrifuging at 12,000 rpm for 10 min to gather the proteins in supernatant. The concentration of proteins was measured using the Pierce BCA Protein Assay Kit (Thermo Scientific, Waltham, MA, USA) based on the manufacturer instructions. 30 μg of each sample was used to conduct western in reference to the previous studies [8,43]. The primary antibodies respectively against Six2, Zeb1, c-Myc, β-tubulin were efficient with proper dilution (1:700; 1:800; 1:1000; 1:4000; Proteintech, Chicago, IL, USA). 4.6. 5-Ethynyl-2′-deoxyuridine (EdU) Assay mK3 cells were transfected with Zeb1 overexpression vectors or Zeb1-shRNA vectors or co-transfected with Zeb1-shRNA and Six2-shRNA vectors by lentivirus. Forty-eight hours later, the treated mK3 cells were seeded onto 96-well plates (10 thousand cells each well) and grew for about 8 h so that the cells were well adherent. Then the proliferation of mK3 cells was detected using the EdU DNA Proliferation in Detection kit (RiboBio, Guangzhou, China) according to the manufacturer instructions. 4.7. MTT Assay The MTT assay was carried out to test the changes of cellular viability. About 2000 mK3 cells transfected with Zeb1-shRNA or NC-shRNA were seeded in 96-well plates with five replicates. 12 h and 24 h later, the medium was aspirated, the cells were incubated with 100 μL fresh medium containing 0.5 mg/mL MTT for 4 h. Then, the medium was discarded and 100 μL of dimethyl sulfoxide was added into the well to dissolve the resulting formazan crystals. The absorbance was detected using a microplate reader (MULTISKAN GO, Thermo Scientific, Waltham, MA, USA) at a test wavelength of 590 nm. The average intensity of absorbance in relation to the formazan product indicated the number of cultured living cells with the equal cells at 0 h. 4.8. Flow Cytometry Assay and Reagent The apoptosis of mK3 cells was determined via the Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (KeyGEN BioTECH, Nanjing, China). mK3 cells were transfected when the cells grew to about 60% confluence and 0.1 μM dexamethasone was selected as an inducing agent in mK3 cells transfected with Zeb1-shRNA or negative control shRNA for 20 h [44,45,46]. Then, the cells were harvested 48 h later with EDTA-free trypsin. The cells were numbered by cell counting board and pipetted about 1 million for the later treatment according to the manufacturer instructions. The flow cytometry apoptosis detection was operated by the Institute of Pediatrics in Children’s Hospital of Chongqing Medical University. 4.9. Wound Healing Assay mK3 cells were plated in 6-well plates in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS. When cells confluence reached about 60%, pCDH-copGFP-Zeb1 CDS, pLKO.1-Zeb1-shRNA, or blank vector (pCDH-copGFP or pLKO.1) were introduced into mK3 cells. The monolayer cells were generated scratch wounds using a pipette tip when cells grew to 95%–100%, then washed twice with PBS, pH 7.4 to wipe off cell debris. Cells were incubated in completed DMEM in 37 °C, 100% humidity and 5% CO2. The images of wound width were taken at different time points (12 and 24 h) using a fluorescence microscope (ECLIPSE Ti-s, Nikon, Tokyo, Japan). Four images were collected from independent selected fields of each sample, and the width of wound areas were calculated by 2014 Microsoft PowerPoint. 4.10. Dual-Luciferase Assay The G401 cell (a human tumor epithelial cell of Caucasian male, aging 3 months) were cultured in 24-well plate (0.1 million each well) for 24 h and then transiently co-transfected with pCDH-m.Zeb1, pGL3-Six2 promoter-luciferase (500 ng/well) and plasmid pRL-SV40 (10 ng/well) utilizing Polyetherimide (PEI) (23966-2, polysciences, Warrington, PA, USA). 48 h later, luciferase activity was assayed using Dual-Luciferase Reporter assay kit (Promega). Levels of firefly luciferase were standardized to those of Renilla. 4.11. Embryonic Mouse Kidney Isolation Embryonic kidneys were isolated from the C57BL/6 mouse embryos at different developmental stage (E11.5, E12.5, E13.5, E14.5, E16.5, E18.5) as described [47]. The process was performed in PBS, pH 7.4 under a microscope and the kidneys were stored at −80 °C or put in Trizol solution for total RNA extraction. 4.12. Statistical Analysis All experiments were performed independently three times and the results were presented as the mean ± standard error of the mean (SEM) or SD. Data were assessed for the statistical significance using student’s t test. The GraphPad Prism 5 software (GraphPad, San Diego, CA, USA) was used to evaluate the statistical results. Statistical differences were considered significant with * p < 0.05, ** p < 0.01, *** p < 0.001. 5. Conclusions The results shown in this study indicate that Zeb1 up-regulates Six2 and promotes proliferation and apoptosis and inhibits the migration in MM cells. Acknowledgments We sincerely thank for the support of the National Natural Science Foundation of China (Grant No. 31271563 and Grant 81572076) to Qin Zhou and the National Basic Research Program of China (No. 2011CB944002) to Qin Zhou. We also acknowledge the materials support of the Students Innovation Laboratory of Chongqing Medical University. And we thank for the ScienLab team, an entrepreneur team of undergraduate students, for their help of lentivirus packing. Author Contributions Yuping Gu, Ya Zhao, Yuru Zhou, Yajun Xie Qin Zhou and Bing Li conceived and designed the experiments; Yuping Gu, Pan Ju, Yaoshui Long, Dongsheng Ni, Jianing Liu and Fen Cao performed the experiments, Zhongshi Lyu, Zhaomin Mao analyzed the data; Jin Hao, Yiman Li, Quist Kanyomse, Qianya Wan, Yamin Liu, Die Ren, Yating Ning and Xiaofeng Li contributed reagents/materials/analysis tools; Yuping Gu, Ya Zhao, Yuru Zhou, Yajun Xie, Qin Zhou and Bing Li wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Abbreviations Zeb1 Zinc finger E-box-binding homeobox 1 MM Metanephric mesenchymal UB Ureteric bud PA Pre-tubular aggregate UT Ureteric tree RV Renal vesicles CM Condensed/cap mesenchyme TGF-β Transforming growth factor-β FITC Fluorescein isothiocyanate fitc HEK Human embryonic kidney MET Mesenchymal-epithelial-transition EMT Epithelial- mesenchymal-transition DMEM Dulbecco’s modified Eagle’s medium Appendix A Figure A1 Knock-down of Zeb1 inhibits mK3 cell proliferation. mK3 cells were 48 h post-treated with vectors of target shRNA or the respective control vector, then the equal 2000 cells were seeded into the 96-well. Proliferating mK3 cells were tested in living cells using MTT at 12 and 24 h. The formazan product was detected by the intensity of absorbance at 590 nm wavelength to indicate cell vitality. The absorbance intensity of Zeb1 knocked down cells was smaller than the negative control cells at both 12 and 24 h. The absorbance present were examined by a microplate reader and results were displayed as mean ± SD (n = 3). ** p < 0.01, *** p < 0.001 negative control vectors. Figure 1 Bioinformatic analysis of Zeb1 protein. (A) Several tracks of entire amino acid sequences of Zeb1 across different mammal species. NCBI was used to get the sequences that were 1117aa in length and were highly conserved shown in gray shadow representing 100% matched sequences across different species; (B) Rooted phylogenetic tree (UPGMA) displayed Zeb1 is highly homologous among different mammalian. The identity is shown on the right; (C) Zeb1 protein structure contains seven C2H2 zinc finger domains, three zinc finger double domains and one homeodomain. Figure 2 Knock-down of Zeb1 inhibits mK3 cell proliferation while overexpression of Zeb1 promotes it. (A) Proliferating mK3 cells were labeled with EdU (red) and nucleuses of the whole cells were stained with DAPI or hoechst (blue). The cells were 48 h post-treated with vectors of overexpression or target shRNA or the respective control vectors. The images present were taken by fluorescent microscopy (200×) with a scale bar of 50 μm and the red and blue images were merged to the purple ones; (B) EdU positive percentage (EdU %) were quantified. Results were displayed as mean ± SD (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001 negative control vectors; (C,D) The RT-PCR and western-blot showed the significant efficiency of Zeb1 overexpression and knock-down. The expression was calculated by scanning gray in Image J software normalized to the internal mRNA control GAPDH or protein control β-tubulin. The result was shown with error bars representing mean ± SD (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001. Figure 3 Knock-down of Zeb1 promotes mK3 cell apoptosis while overexpression of Zeb1 inhibits it. (A) mK3 cell apoptosis was detected by flow cytometry with Annexin V-FITC/PI staining. The number of positive cells double stained by AnnexinV-FITC/PI in mK3 cells transfected with Zeb1 overexpression vector was dramatically smaller than the control vector. The detection was performed 48 h transfection later; (C) mK3 cells were transfected with Zeb1-shRNA or negative control shRNA for 20 h. Then one well cells of the two transfected with the same vector was treated with 0.1 μM dexamethasone for 30 h, following by cell apoptosis detection. The number of AnnexinV-FITC/PI-positive cells in Zeb1 knock-down mK3 cells was significantly larger than the negative control cells. And the dexamethasone at 0.1 μM concentration increased the number of AnnexinV-FITC/PI-positive mK3 cells; (B,D) Apoptosis rates (Q2% + Q4%) of mK3 cells detected in Figure 3A,B were respectively quantified. Results were displayed as mean ± SD (n = 3). ** p < 0.01, *** p < 0.001 negative control vector. Figure 4 Knock-down of Zeb1 inhibits mK3 cell migration while overexpression of Zeb1 promotes cell migration only after 24 h. (A) Cell migration of mK3 cells was measured via wound healing assay. The width of wound area was calculated at three time points (0, 12, 24 h) starting from the point when mK3 cells were transfected 48 h later. The transfection way was same with Figure 2. Zeb1 knock down led to lower apoptosis rate but Zeb1 overexpression resulted in higher rate, compared with the controls; (B,C) The quantification of wound healing percentage (100%) was exhibited with the error bars representing mean ± SD (n = 3). ** p < 0.01, negative control vector. Figure 5 Zeb1 binds to Six2 promoter and up-regulates Six2 in mK3 cells. (A) The predicted binding motifs of Zeb1 to Six2 promoter, which was acquired from the JASPAR Database. The binding motifs across mammal species are conservatively shown in gray shadow representing the matched sequence among several mammal species, for instance, seven base pairs in eight were matched between mouse and human; (B) G401 cells were co-transfected with pGL-SV40 (renilla control), firefly luciferase reporter pcDNA3.1-Six2 promoter-luciferase, and either the m.Zeb1 expression plasmid pCDH-copGFP-m.Zeb1 or the respective control vector. Luciferase activity was assayed using dual luciferase reporter assay 48 h after transfection, normalized to renilla control. The result was analyzed by student’s t test and displayed with error bars representing mean ± SD (n = 3), ** p < 0.01, *** p < 0.001; (C,D) RT-PCR was used to detect the expression of Six2 at both mRNA and protein level. Six2 was remarkably decreased in mK3 cells treated with Zeb1-shRNA compared with negative control shRNA. Overexpressing Zeb1 in mK3 cells led to the promotion of Six2 expression at mRNA level. The expression was quantified by scanning gray in Image J software normalized to the internal mRNA control 18S or protein control β-tubulin. The result was displayed with error bars representing mean ± SD (n = 3), ** p < 0.01, *** p < 0.001. Figure 6 Zeb1 is required for the proliferation of mK3 cells associated to Six2. (A) mK3 cells were 48 h post-introduced with control shRNA, Zeb1-shRNA, Six2-shRNA or both of the two target shRNA. Then EdU assay was performed refering to Figure 2 and the red and blue images were merged to the purple ones; the length of scale bar was 50 μm. Cell proliferation was decreased in mK3 cells treated with either single shRNA or double shRNA of Zeb1 and Six2 compared with respective controls; (B) The data of proliferation percentage (EdU %) were presented as mean ± SD (n = 3). * p < 0.05, ** p < 0.01, respective controls; (C) The significant knock-down efficiency of Zeb1 and Six2 was measured by the RT-PCR with the internal control 18S equal. Figure 7 Zeb1 is required for mK3 cells to survival associated with Six2. (A) mK3 cells were transfected same with Figure 6 and cell apoptosis was detected by reference to Figure 3. The rate of apoptosis in mK3 cells treated with single target shRNA is higher than the negative control shRNA, while mK3 cells treated double shRNAs had a lower rate than the single targeted mK3 cells; (B) Cell apoptosis rate was quantified and presented as mean ± SD (n = 3). * p < 0.05, ** p < 0.01, respective controls. Figure 8 Zeb1 and Six2 have a similar expression profile in mouse kidney development. (A) RT-PCR was performed using the cDNA library obtained from the embryonic mouse kidney at different embryonic days (E11.5, E12.5, E13.5, E14.5, E16.5, E18.5). The mRNA expression of both genes showed identical variation trends except for E16.5; (B) The quantification was analyzed by Gray Scan normalized to the internal control GAPDH and the results were acted as mean ± SD (n = 3); (C) Bioinformatics analysis of Zeb1 and Six2 mRNA expression was performed in GUDMAP Expression Database. It is the microarray data in different cells of developing renal, in which blue represents the median of each row, black means the expression value is smaller than median, while red represents a larger value. Figure 9 The expression of c-Myc in mK3 cells is regulated by Zeb1. (A) Western visualization shows that mK3 cells introduced with Zeb1-shRNA for 48 h expressed c-Myc less than cells treated with control shRNA at protein level; (B) The expression was calculated by scanning gray in Image J software normalized to the internal control GAPDH. The result was displayed with error bars representing mean ± SD (n = 3). ** p < 0.01, *** p<0.001; (C) The protein expression of c-Myc was increased in mK3 cells treated with Zeb1 overexpression for 48 h compared with the control cells (blank vector); (D) The expression was quantified the same way as in (B). Figure 10 The work model of Zeb1 in developing embryonic kidney. Zeb1 promotes MM cell proliferation (cell renewal), cell migration but inhibits cell apoptosis (cell consumption) in association with Six2 up-regulation and c-Myc down-regulation. As the green cycles show, MM cells migrate and aggregate at UB branches with the induction of UB, then interact with UB. So the MM cell renewal, consumption and migration are essential for the induced interaction between MM cells and UB. These contributes to UB branching morphogenesis and MET and elongation for nephrons formation during embryonic kidney development. ==== Refs References 1. McCampbell K.K. Springer K.N. Wingert R.A. Analysis of nephron composition and function in the adult zebrafish kidney J. Vis. Exp. 2014 9 e51644 10.3791/51644 25145398 2. Vainio S. How the developing mammalian kidney assembles its thousands of nephrons: Fgfs as stemness signals Dev. Cell 2012 22 1125 1126 10.1016/j.devcel.2012.05.016 22698278 3. Saxen L. Sariola H. Early organogenesis of the kidney Pediatr. Nephrol. 1987 1 385 392 10.1007/BF00849241 3153305 4. Nishita M. Qiao S. Miyamoto M. Okinaka Y. Yamada M. Hashimoto R. Iijima K. Otani H. Hartmann C. Nishinakamura R. Role of Wnt5a-Ror2 signaling in morphogenesis of the metanephric mesenchyme during ureteric budding Mol. Cell. Biol. 2014 34 3096 3105 10.1128/MCB.00491-14 24891614 5. Pope J.C.T. Brock J.W. 3rd. Adams M.C. Stephens F.D. Ichikawa I. How they begin and how they end: Classic and new theories for the development and deterioration of congenital anomalies of the kidney and urinary tract, CAKUT J. Am. Soc. Nephrol. 1999 10 2018 2028 10477156 6. Tanigawa S. Sharma N. Hall M.D. Nishinakamura R. Perantoni A.O. Preferential propagation of competent Six2+ nephronic progenitors by LIF/ROCKi treatment of the metanephric mesenchyme Stem Cell Rep. 2015 5 435 447 10.1016/j.stemcr.2015.07.015 26321142 7. Kobayashi A. Valerius M.T. Mugford J.W. Carroll T.J. Self M. Oliver G. McMahon A.P. Six2 defines and regulates a multipotent self-renewing nephron progenitor population throughout mammalian kidney development Cell Stem Cell 2008 3 169 181 10.1016/j.stem.2008.05.020 18682239 8. lyu Z. Mao Z. Wang H. Fang Y. Chen T. Wan Q. Wang M. Wang N. Xiao J. Wei H. MiR-181b targets Six2 and inhibits the proliferation of metanephric mesenchymal cells in vitro Biochem. Biophys. Res. Commun. 2013 440 495 501 10.1016/j.bbrc.2013.09.059 24055707 9. Self M. Lagutin O.V. Bowling B. Hendrix J. Cai Y. Dressler G.R. Oliver G. Six2 is required for suppression of nephrogenesis and progenitor renewal in the developing kidney EMBO J. 2006 25 5214 5428 10.1038/sj.emboj.7601381 17036046 10. Vannier C. Mock K. Brabletz T. Driever W. Zeb1 regulates E-cadherin and Epcam (epithelial cell adhesion molecule) expression to control cell behavior in early zebrafish development J. Biol. Chem. 2013 288 18643 18659 10.1074/jbc.M113.467787 23667256 11. Eggers J.C. Martino V. Reinbold R. Schafer S.D. Kiesel L. Starzinski-Powitz A. Schuring A.N. Kemper B. Greve B. Gotte M. microRNA miR-200b affects proliferation, invasiveness and stemness of endometriotic cells by targeting ZEB1, ZEB2 and KLF4 Reprod. Biomed. Online 2016 32 434 445 10.1016/j.rbmo.2015.12.013 26854065 12. Tang O. Chen X.M. Shen S. Hahn M. Pollock C.A. MiRNA-200b represses transforming growth factor-beta1-induced EMT and fibronectin expression in kidney proximal tubular cells Am. J. Physiol. Ren. Physiol. 2013 304 F1266 F1273 10.1152/ajprenal.00302.2012 23408168 13. Gong S.G. Eulenberg R.L. Palatal development in Twirler mice Cleft Palate Craniofac. J. 2001 38 622 628 10.1597/1545-1569(2001)038<0622:PDITM>2.0.CO;2 11681996 14. Nishimura G. Manabe I. Tsushima K. Fujiu K. Oishi Y. Imai Y. Maemura K. Miyagishi M. Higashi Y. Kondoh H. DeltaEF1 mediates TGF-β signaling in vascular smooth muscle cell differentiation Dev. Cell 2006 11 93 104 10.1016/j.devcel.2006.05.011 16824956 15. Liu Y. El-Naggar S. Darling D.S. Higashi Y. Dean D.C. Zeb1 links epithelial-mesenchymal transition and cellular senescence Development 2008 135 579 588 10.1242/dev.007047 18192284 16. Senanayake U. Koller K. Pichler M. Leuschner I. Strohmaier H. Hadler U. Das S. Hoefler G. Guertl B. The pluripotent renal stem cell regulator Six2 is activated in renal neoplasms and influences cellular proliferation and migration Hum. Pathol. 2013 44 336 345 10.1016/j.humpath.2012.05.021 22995329 17. Zhou P. Chen T. Fang Y. Wang H. Li M. Ma P. He L. Li Q. Liu T. Yang X. Down-regulated Six2 by knockdown of neurofibromin results in apoptosis of metanephric mesenchyme cells in vitro Mol. Cell. Biochem. 2014 390 205 213 10.1007/s11010-014-1971-0 24573885 18. Reidy K.J. Rosenblum N.D. Cell and molecular biology of kidney development Semin. Nephrol. 2009 29 321 337 10.1016/j.semnephrol.2009.03.009 19615554 19. Multiple Sequence Aligment by CLUSTALW Available online: http://www.genome.jp/tools/clustalw/ (accessed on 24 September 2015) 20. The Protein Database of National Center for Biotechnology Information Available online: http://www.ncbi.nlm.nih.gov/protein. (accessed on 14 August 2015) 21. Xu J. Wong E.Y. Cheng C. Li J. Sharkar M.T. Xu C.Y. Chen B. Sun J. Jing D. Xu P.X. Eya1 interacts with Six2 and Myc to regulate expansion of the nephron progenitor pool during nephrogenesis Dev. Cell 2014 31 434 447 10.1016/j.devcel.2014.10.015 25458011 22. Yu J. Carroll T.J. McMahon A.P. Sonic hedgehog regulates proliferation and differentiation of mesenchymal cells in the mouse metanephric kidney Development 2002 129 5301 5312 12399320 23. Takeyama Y. Sato M. Horio M. Hase T. Yoshida K. Yokoyama T. Nakashima H. Hashimoto N. Sekido Y. Gazdar A.F. Knockdown of ZEB1, a master epithelial-to-mesenchymal transition (EMT) gene, suppresses anchorage-independent cell growth of lung cancer cells Cancer Lett. 2010 296 216 224 10.1016/j.canlet.2010.04.008 20452118 24. Filios S.R. Xu G. Chen J. Hong K. Jing G. Shalev A. MicroRNA-200 is induced by thioredoxin-interacting protein and regulates Zeb1 protein signaling and beta cell apoptosis J. Biol. Chem. 2014 289 36275 36283 10.1074/jbc.M114.592360 25391656 25. Xu J. Liu H. Park J.S. Lan Y. Jiang R. Osr1 acts downstream of and interacts synergistically with Six2 to maintain nephron progenitor cells during kidney organogenesis Development 2014 141 1442 1452 10.1242/dev.103283 24598167 26. Crawford E.D. Wells J.A. Caspase substrates and cellular remodeling Annu. Rev. Biochem. 2011 80 1055 1087 10.1146/annurev-biochem-061809-121639 21456965 27. Pop C. Salvesen G.S. Human caspases: Activation, specificity, and regulation J. Biol. Chem. 2009 284 21777 21781 10.1074/jbc.R800084200 19473994 28. Piatkov K.I. Brower C.S. Varshavsky A. The N-end rule pathway counteracts cell death by destroying proapoptotic protein fragments Proc. Natl. Acad. Sci. USA 2012 109 E1839 E1847 10.1073/pnas.1207786109 22670058 29. Eldeeb M.A. Fahlman R.P. The anti-apoptotic form of tyrosine kinase Lyn that is generated by proteolysis is degraded by the N-end rule pathway Oncotarget 2014 5 2714 2722 10.18632/oncotarget.1931 24798867 30. Liu Y. Sanchez-Tillo E. Lu X. Clem B. Telang S. Jenson A.B. Cuatrecasas M. Chesney J. Postigo A. Dean D.C. Rb1 family mutation is sufficient for sarcoma initiation Nat. Commun. 2013 10.1038/ncomms3650 24150016 31. Xu B. Lee K.K. Zhang L. Gerton J.L. Stimulation of mTORC1 with L-leucine rescues defects associated with Roberts syndrome PLoS Genet. 2013 9 1283 10.1371/journal.pgen.1003857 24098154 32. Bates C.M. Kidney development: Regulatory molecules crucial to both mice and men Mol. Genet. Metab. 2000 71 391 396 10.1006/mgme.2000.3072 11001832 33. Couillard M. Trudel M. C-myc as a modulator of renal stem/progenitor cell population Dev. Dyn. 2009 238 405 414 10.1002/dvdy.21841 19161241 34. Siemens H. Jackstadt R. Hunten S. Kaller M. Menssen A. Gotz U. Hermeking H. miR-34 and SNAIL form a double-negative feedback loop to regulate epithelial-mesenchymal transitions Cell Cycle 2011 10 4256 4271 10.4161/cc.10.24.18552 22134354 35. The GenitoUrinary Development Molecular Anatomy Project Database Available online: http://www.gudmap.org/gudmap/pages/mastertablebrowse.html (accessed on 10 September 2015) 36. The JASPAR Database Available online: http://jaspar.genereg.net/cgi-bin/jaspar_db.pl. (accessed on 16 October 2015) 37. Sun H. Li Q.W. Lv X.Y. Ai J.Z. Yang Q.T. Duan J.J. Bian G.H. Xiao Y. Wang Y.D. Zhang Z. MicroRNA-17 post-transcriptionally regulates polycystic kidney disease-2 gene and promotes cell proliferation Mol. Biol. Rep. 2010 37 2951 2958 10.1007/s11033-009-9861-3 19821056 38. Balhana R. Stoker N.G. Sikder M.H. Chauviac F.X. Kendall S.L. Rapid construction of mycobacterial mutagenesis vectors using ligation-independent cloning J. Microbiol. Methods 2010 83 34 41 10.1016/j.mimet.2010.07.014 20650290 39. MISSION shRNA Clones for individual genes Available online: http://www.Sigmaaldrich.com/china-mainland/zh/life-science/functional-genomics-and-rnai/shrna.html (accessed on 3 December 2015) 40. pLKO.1-ShRNA Construction Available online:http://www.addgene.org/tools/protocols/plko/ (accessed on 12 December 2015) 41. Valerius M.T. Patterson L.T. Witte D.P. Potter S.S. Microarray analysis of novel cell lines representing two stages of metanephric mesenchyme differentiation Mech. Dev. 2002 112 219 232 10.1016/S0925-4773(02)00008-4 11850199 42. McLaughlin N. Yao X. Li Y. Saifudeen Z. El-Dahr S.S. Histone signature of metanephric mesenchyme cell lines Epigenetics 2013 8 970 978 10.4161/epi.25753 23867747 43. Tanigawa S. Wang H. Yang Y. Sharma N. Tarasova N. Ajima R. Yamaguchi T.P. Rodriguez L.G. Perantoni A.O. Wnt4 induces nephronic tubules in metanephric mesenchyme by a non-canonical mechanism Dev. Biol. 2011 352 58 69 10.1016/j.ydbio.2011.01.012 21256838 44. Suwanjang W. Abramov A.Y. Charngkaew K. Govitrapong P. Chetsawang B. Melatonin prevents cytosolic calcium overload, mitochondrial damage and cell death due to toxically high doses of dexamethasone-induced oxidative stress in human neuroblastoma SH-SY5Y cells Neurochem. Int. 2016 97 34 41 10.1016/j.neuint.2016.05.003 27155536 45. Zhang C. Li L. Zhao B. Jiao A. Li X. Sun N. Zhang J. Ghrelin protects against dexamethasone-induced INS-1 cell apoptosis via ERK and p38MAPK signaling Int. J. Endocrinol. 2016 10.1155/2016/4513051 27190513 46. Guo S. Mao L. Ji F. Wang S. Xie Y. Fei H. Wang X.D. Activating AMP-activated protein kinase by an alpha1 selective activator compound 13 attenuates dexamethasone-induced osteoblast cell death Biochem. Biophys. Res. Commun. 2016 471 545 552 10.1016/j.bbrc.2016.02.036 26891866 47. Brown A.C. Blank U. Adams D.C. Karolak M.J. Fetting J.L. Hill B.L. Oxburgh L. Isolation and culture of cells from the nephrogenic zone of the embryonic mouse kidney J. Vis. Exp. 2011 10.3791/2555 21540822

PMC005xxxxxx/PMC5000681.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081284ijms-17-01284ArticleAnti-Atherogenic Properties of Allium ursinum Liophylisate: Impact on Lipoprotein Homeostasis and Cardiac Biomarkers in Hypercholesterolemic Rabbits Bombicz Mariann 12Priksz Daniel 12Varga Balazs 12Gesztelyi Rudolf 1Kertesz Attila 3Lengyel Peter 4Balogh Peter 5Csupor Dezso 6Hohmann Judit 6Bhattoa Harjit Pal 7Haines David D. 1Juhasz Bela 12*Battino Maurizio Academic Editor1 Department of Pharmacology, Faculty of Pharmacy, University of Debrecen, Debrecen H-4032, Hungary; [email protected] (M.B.); [email protected] (D.P.); [email protected] (B.V.); [email protected] (R.G.); [email protected] (D.D.H.)2 Institute of Pharmacology and Pharmacotherapy, Faculty of Medicine, University of Debrecen, Debrecen H-4032, Hungary3 Department of Cardiology, Faculty of Medicine, University of Debrecen, Debrecen H-4032, Hungary; [email protected] Institute of Applied Informatics and Logistics, University of Debrecen, Debrecen H-4032, Hungary; [email protected] Department of Research Methodology and Statistics, Institute of Sectoral Economics and Methodology, University of Debrecen, Debrecen H-4032, Hungary; [email protected] Department of Pharmacognosy, Faculty of Pharmacy, University of Szeged, Szeged H-6720, Hungary; [email protected] (D.C.); [email protected] (J.H.)7 Department of Laboratory Medicine, Faculty of Medicine, University of Debrecen, Debrecen H-4032, Hungary; [email protected]* Correspondence: [email protected]; Tel.: +36-5242-7899 (ext. 56109)10 8 2016 8 2016 17 8 128408 6 2016 27 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The present investigation evaluates the capacity of Allium ursinum (wild garlic) leaf lyophilisate (WGLL; alliin content: 0.261%) to mitigate cardiovascular damage in hypercholesterolemic rabbits. New Zealand rabbits were divided into three groups: (i) cholesterol-free rabbit chow (control); (ii) rabbit chow containing 2% cholesterol (hypercholesterolemic, HC); (iii) rabbit chow containing 2% cholesterol + 2% WGLL (hypercholesterolemic treated, HCT); for eight weeks. At the zero- and eight-week time points, echocardiographic measurements were made, along with the determination of basic serum parameters. Following the treatment period, after ischemia-reperfusion injury, hemodynamic parameters were measured using an isolated working heart model. Western blot analyses of heart tissue followed for evaluating protein expression and histochemical study for the atheroma status determination. WGLL treatment mediated increases in fractional shortening; right ventricular function; peak systolic velocity; tricuspidal annular systolic velocity in live animals; along with improved aortic and coronary flow. Western blot analysis revealed WGLL-associated increases in HO-1 protein and decreases in SOD-1 protein production. WGLL-associated decreases were observed in aortic atherosclerotic plaque coverage, plasma ApoB and the activity of LDH and CK (creatine kinase) in plasma. Plasma LDL was also significantly reduced. The results clearly demonstrate that WGLL has complex cardioprotective effects, suggesting future strategies for its use in prevention and therapy for atherosclerotic disorders. Allium speciesatherosclerosislipoproteincardiovascular homeostasisechocardiography ==== Body 1. Introduction Wild garlic (Allium ursinum L.) is a wild plant belonging to the Amaryllidaceae family. It is distributed widely in Asia and Europe and known variously as bear’s garlic, buckrams, bear’s leek, wood garlic and ransoms [1]. The intense flavor of the plant makes it a popular flavoring and regular dietary component for people and animals living in regions where it grows [2]. The mild garlic-like scent of the plant is attributable to its content of sulfur-containing compounds. These include, prominently, sulfoxides and glutamyl peptides. The species also contains high levels of odorless, non-volatile metabolites: S-alk(en)yl-l-cysteine-sulfoxides, which hydrolyze under physiological conditions to volatile (poly)sulfides and thiosulfinates, imparting the characteristic odor and flavor of the plant [3]. Wild garlic also contains high levels of polyphenolic compounds, particularly in leaves and bulbs, which accounts substantially for the antioxidant and therapeutic properties of these sections of the plant [4,5,6]. It also combines two additional health-enhancing properties: the plant has approximately 20-times the level of adenosine as common garlic (Allium sativum), plus it has significantly higher levels of ajoene, both of which combine to stabilize blood pressure and cholesterol levels, reduce excessive thrombocyte aggregation and improve physiological control of cholesterol metabolism [7]. Indeed, the cardiovascular benefits of using this plant were observed to be so substantial that the Association for the Protection and Research on European Medicinal Plants designated it “Plant of the Year” for 1992 [7]. However, contrary to common garlic, wild garlic has not been studied in clinical trials, and although its cardiovascular effect may be hypothesized based on its chemical constituents, the preclinical confirmation is rather incomplete. A. ursinum was also selected as the subject of the present investigation based on the outcomes of previous work by this laboratory demonstrating cardioprotective properties of other plant extracts derived from traditional medicines [8,9]. Hypercholesterolemia, a syndrome characterized by abnormally elevated levels of blood cholesterol and lipoproteins [10,11], was chosen as a model disease for the present study due to its association with a wide range of pathologies, particularly atherosclerosis [12], with associated thrombosis, stroke and heart failure [13]. Although A. ursinum is not typically used as a stand-alone medication, its anti-atherogenic properties are well known, to the extent it is used as a dietary treatment for these disorders at Bucharest University Hospital in Romania [14]. In vitro evaluation for the effects of several A. ursinum fresh leaf extract preparations on the aggregation of human platelets revealed that ADP-induced aggregation was significantly suppressed by ethanolic extracts. The observed data suggested the similarity of pharmacological action to clopidogrel, a thienopyridine clot formation inhibitor that is a potent antiplatelet drug [15]. A likely explanation for this outcome is the known antiaggregatory properties of the β-sitosterol 3-O-β-d-glucopyranoside and 1,2-di-O-α-linolenoyl-3-O-β-d-galactopyranosyl-sn-glycerol (DLGG) components of A. ursinum [16]. It was further noted that 45-day administration of feed supplemented with 1% w/w wild garlic Allium ursinum (wild garlic) or alternatively with 1% w/w Allium sativum (cultivated garlic) to spontaneously hypertensive rats (SHR), in groups of 10 animals per experiment, mediated a significant reduction in final mean systolic blood pressure (SBP) [17]. The possible underlying mechanisms include the ability of ramson to inhibit the activity of angiotensin-converting enzyme (ACE). In vitro tests on the water extract from the leaves (at the concentration of 0.300 mg/mL) showed significantly increased activity on enzyme inhibition when compared to leaves with extract of garlic (58% versus 30%) [7]. Moreover, significantly lower levels of ACE activity were noted in the blood of animals fed for eight weeks with a standard rodent chow containing 2% pulverized whole leaf A. ursinum, versus untreated control rats [18]. The physiologic significance of hypercholesterolemia induced by elevated cholesterol in feed administered to animals is particularly well illustrated by the consideration of how such diets affect inflammatory processes, the dysregulation of which imposes increased oxidative stress on a wide range of tissues and to which cells of the cardiovascular system are particularly sensitive [19]. For example, pigs maintained on diets supplemented with 2% cholesterol exhibited impairment of coronary endothelial function associated with decreased capacity to neutralize free radicals, decreased expression of nitric oxide synthetase and elevated activation of nuclear factor-κβ, a pro-inflammatory transcription factor [20]. Outcomes of these investigations underscore the particular significance of hypercholesterolemia for the investigation of cardiac function, as was demonstrated in by the gene transfer studies conducted [21,22]. Previous work by the authors shows that methods of extraction used to recover, purify and concentrate plant products may cause some degradation in the bioactivity of component molecules [23]. For this reason, lyophilization was used to process the A. ursinum administered to animals in the present study. This method is easily accomplished and optimally preserves the native properties of extracted biological molecules [24]. 2. Results 2.1. Bioanalytical Analysis of Wild Garlic Leaf Lyophilisate Alliin (S-allyl-l-cysteine sulfoxide) is a non-protein amino acid abundant in most of the Allium species. It is the natural substrate of alliinase. Therefore, its content in the pure form is commonly analyzed by HPLC. The percentage of total alliin was analyzed by HPLC (Figure 1). Analysis of a representative lyophilized sample revealed the leaf to contain 0.261% alliin by weight (RSD% = 0.45%). The major peak at 3.8 min (detection at 204 nm) is identical to alliin based on its identical UV spectrum and detection time to those of a reference standard. 2.2. Echocardiographic Analyses All echocardiographic examinations were completed within a 20-min time interval with outcomes shown in Table 1. The end-systolic diameter (ESD) of the left ventricle measured in M-mode exhibited significant increases in hypercholesterolemic (HC) animals (1.242 ± 0.045 cm for HC, versus 1.016 ± 0.091 cm noted in the control group). Nevertheless, no change in this outcome was observed in wild garlic leaf lyophilisate (WGLL)-treated hypercholesterolemic (HCT) animals (1.184 ± 0.020 cm) in comparison with this parameter in the control group. Fractional shortening (FS) and ejection fraction (EF) data correlated strongly with measurements of both the parasternal long and short axis views. FS and EF of HC animals were significantly decreased in comparison with this outcome evaluated in animals in the control group (FSHC: 29.010 ± 1.056, versus FSControl: 32.310 ± 0.718; and EFHC: 49.810 ± 1.140, versus EFControl: 56.910 ± 1.294, respectively). Additionally, significant increases in fractional shortening were observed in the HCT group in comparison with the HC group (FSHCT: 32.970 ± 1.131 and EFHCT: 55.990 ± 1.756). The diastolic function of the left ventricle was evaluated by peak mitral early diastolic inflow velocity/peak atrial diastolic inflow velocity (E/A) ratios measured in Doppler (pulsed wave, PW) mode. E/A ratios were significantly lower in the HC group in comparison to the control animals (HC: 1.207 ± 0.037 versus the control: 1.376 ± 0.045). These results notwithstanding, no significant changes were observed in the E/A ratios of treated animals (HCT: 1.344 ± 0.076) in comparison to controls. Deceleration time of the E wave (DecT) exhibited significant lengthening in the HC animals (HC: 87.440 ± 3.534 ms versus the control: 71.250 ± 4.101 ms). However, DecT values of WGLL-treated animals were significantly lower compared to the HC rabbits (HCT: 69.540 ± 4.787 ms). Tissue velocity imaging (TDI) revealed a non-significant trend toward decreased lateral E’/A’ ratios in WGLL-treated animals. Surprisingly, right ventricle function characterized by measuring peak systolic velocity (S’) waves and tricuspidal annular plane systolic excursion (TAPSE) exhibited significant improvement in WGLL-treated animals. The amplitudes of S’ waves were significantly increased in WGLL-treated animals, compared to the HC group (HCT: 9.156 ± 0.210 cm/s versus HC: 8.103 ± 0.216 cm/s), and TAPSE values were also significantly elevated in the WGLL-treated HCT animals compared to the HC rabbits (HCT: 0.646 ± 0.020 cm versus HC: 0.5762 ± 0.015 cm). Additionally, right ventricle E’/A’ ratios of WGLL-treated animals were slightly decreased. 2.3. Cardiac Function in Isolated Working Hearts Figure 2 shows the effect on cardiac functional parameters of elevated dietary cholesterol-induced hypercholesterolemia and WGLL treatment. Cardiac functions evaluated included: aortic flow (AF, Figure 2A), coronary flow (CF, Figure 2B), aortic pressure (AoP, Figure 2C), heart rate (HR, Figure 2D), cardiac output (CO, Figure 2E) and stroke volume (SV, Figure 2F). Measurement of these functions in animals in the HC and HCT groups revealed decreases in AF, HR and CO for basal functions of the perfused hearts, compared to controls (p < 0.05). There were significant increases under preischemia AoP, both in hypercholesterolemic and WGLL-treated hypercholesterolemic groups, compared to the control group (p < 0.05). After 60 min of reperfusion, animals in all groups showed decreases in AF, CF, HR and CO compared to preischemic values. Significant increases in the recovery of AF and HR were observed in the WGLL-treated group, compared to the other groups (p < 0.05). Subsequent correlation with the results of echocardiographic measurements (Table 1) further supported the cardioprotective capacity of WGLL. 2.4. Western Blot Analysis for Biomarkers of Cardiac Tissue Function Myocardial tissue levels of four major mediators of cardiac homeostasis, measured by Western blot analysis, are shown in Figure 3. The outcomes of treatments administered to rabbits in these experiments revealed that expression of HO-1 protein was significantly greater in tissue harvested from HCT animals, compared to the levels observed in the HC group (Figure 3A, p < 0.05). Tissue expression of SOD-1 in the HC group was observed to be significantly higher compared to control and HCT animals (Figure 3C, p < 0.05). COXIII and VEGF proteins were expressed at lower levels both in HC and HCT groups versus quantities of these proteins found in hearts harvested from the control animals (Figure 3B,D, p < 0.05). 2.5. Rabbit Aortic Histology Histological sections of aortas stained with hematoxylin-oil red O from the three groups are shown in Figure 4. No atherosclerotic lesions were observed in sections of these blood vessels harvested from the control animals fed normal rabbit chow (Figure 4A). At the end of the eight-week treatment period, up to 35% of the total aortic area harvested from HC animals was oil red O positive (Figure 4B). The extent of atherosclerotic lesions observed in animals within the HC group was significantly increased (Figure 4D, 38.610% ± 0.146%) in comparison to lesional coverage in aortic sections taken from animals in the control group (p < 0.05). Discrete lesion formation was visualized by oil red O stain and consecutive quantitative analysis in aortas form HCT group rabbits (Figure 4C). Aortas harvested from WGLL-treated animals in the HCT group exhibited significantly reduced atherosclerotic lesional coverage in comparison to the HC group (Figure 4D, 16.710% ± 3.421%, p < 0.05). 2.6. Serological Correlates of Experimental Treatments The outcomes of the analyses of peripheral blood serum from animals treated with selected dietary regimens are shown in Table 2. Fasting plasma TC and LDL cholesterol were two orders of magnitude higher, and the HDL cholesterol concentration was eight-fold higher in the HC and six-fold higher in HCT groups, compared to the levels of these analytes in the serum of the control group animals (p < 0.05). However, significantly lower plasma TC and LDL cholesterol levels were observed in HCT, versus the HC groups (p < 0.05), showing a possible protective effect of the WGLL. Moreover, ApoA levels detected in blood of HC animals (0.022 ± 0.003) were significantly lower, versus those of the control rabbits fed diets with normal cholesterol content (0.042 ± 0.005) and WGLL-treated rabbits in the HCT group (0.056 ± 0.008, p < 0.05). No significant differences were noted between serum ApoA content of the serum from animals in the control group, versus the HCT group (p > 0.05). Serum levels of ApoB in rabbits from both the HCT and HC groups were significantly higher as compared to the control group (p < 0.05). Moreover, ApoB levels detected in the serum of rabbits in the HCT group (0.172 ± 0.019) were significantly lower compared to the content of this analyte in the HC group (0.280 ± 0.063, p < 0.05). Additionally, no significant differences were observed between the serum TG content of these three groups. It was further noted that the serum content of the pro-inflammatory biomarker c-reactive protein (CRP) was increased significantly in blood from HC animals, compared to the control group. GOT liver enzyme levels were elevated in the blood of HC group animals (48.8 ± 16.22), as compared to the control and HCT groups (29.670 ± 2.895, 24.910 ± 2.708; p < 0.05). LDH (316.6 ± 37.17) and creatine kinase (CK) (2851 ± 600.800) serum levels were significantly decreased in the HCT group compared to the HC group (791.90 ± 325.4 and 4955 ± 1353; p < 0.05). 3. Discussion As described in the Results Section 2.1 of this report, bioanalytical analysis of a representative WGLL sample revealed the leaf to contain 0.261% alliin by weight, a property of this material that contributes to its ability to mitigate the expression of other biomolecules described here, which are involved in the atherosclerotic pathophysiologic processes. This natural component of fresh garlic is a sulfoxide derived and formed from the amino acid cysteine [25] and is a major contributor to the capacity of garlic extracts to scavenge hydroxyl radicals, along with a wide variety of other antioxidant properties that counteract oxidative tissue damage [26]. Alliin has also been demonstrated to potently stabilize and strengthen immunoregulation, contributing to the well-known curative properties of garlic [27]. The outcomes of echocardiographic analyses conducted on live animals shown in Table 1 reveal the effect of elevated dietary cholesterol and WGLL treatment. The physiologic significance of these results may be stated according to four major interpretations of the data shown. These may be summarized as follows: (1) The observed stability throughout the evaluation period of heart rates, respiratory frequencies, M-mode and Doppler measurements demonstrate that basal cardiopulmonary activity was not substantially disrupted by hypercholesteremia, an outcome for which preliminary evidence was provided by an earlier study conducted in the laboratory of the authors [28]. (2) Moreover, in comparison to untreated control rabbits fed a normal diet, the left ventricular end-systolic diameter (ESD) measured in HC animals was significantly increased, with or without WGLL treatment, along with decreased fractional shortening and ejection fraction in HC animals, and a strong correlation was found between fractional shortening (FS) and ejection fraction (EF) data measured on both the parasternal long and short axis views. Furthermore, the effects of WGLL treatment included observations that, relative to HC animals not receiving the lyophilisate, HCT rabbits showed significant increases in fractional shortening. Pathologically-increased ESD is associated with greater risk of mortality in heart disease [29], and decreased FS and ES have recently been implicated as contributors a to fibrotic disease [30]. These results suggest that WGLL will contribute to the survival of cardiac patients and a lower propensity for the development of fibrosis. (3) Table 1 diastolic function data, generated in Doppler (PW) mode, also reveal that elevated dietary cholesterol resulted in significantly lower left ventricular E/A ratios relative to those observed in control animals. WGLL-treated animals showed values close to controls. Moreover, significant lengthening was observed in the deceleration time of the E wave (DecT) in HC animals, showing an abnormal diastolic filling pattern of the ventricle, which was counteracted by the WGLL-supplemented diet, indicating slightly improved diastolic function. (4) Finally, surprisingly significant increases were shown in Table 1, in right ventricular function mediated by the WGLL treatment of animals fed elevated cholesterol diets, which were obtained through the evaluation of peak systolic velocity (S’) waves and tricuspidal annular plane systolic excursion (TAPSE). Reduction of peak systolic velocity identifies the presence of right ventricle (RV) dysfunction with high sensitivity. This reduction was significant in HC animals, but was counteracted fully by WGLL treatment, showing that the aforementioned beneficial effects of WGLL supplementation can be seen on right ventricle function, as well. In heart failure patients, the reduction of tricuspidal annular systolic velocity is associated with the severity of RV dysfunction. Surprisingly, TAPSE values in the WGLL-treated group were significantly increased even compared to healthy animals. These findings indicates that diet supplemented with WGLL could have positive effects on right ventricle systolic function, but the relevance of these effects and the underlying mechanisms need further investigations. The evaluation of cardiac functions in Langendorff-mounted isolated working hearts shown in Figure 2 revealed significantly increased preischemic AoP values in both the hypercholesterolemic and WGLL-treated hypercholesterolemic groups, relative to untreated control rabbits. Furthermore, as shown in Figure 2, ischemic-reperfusion injury-associated decreases in AF, CF, HR and CO, versus preischemic values, along with significantly increased recovery of AF and HR in animals fed the lyophilisate further supported the anti-ischemic properties of WGLL. These effects are consistent with previous studies by the authors, in which interventions that decrease oxidative stress on cardiac tissue dramatically improved recovery from ischemic events [31,32,33]. An interpretation of the significance of these outcomes to the cardioprotective ability of WGLL should be considered in the context of the fact that myocardial ischemic events typically reduce cardiac aortic blood pressure (AoP), along with a reduction in myocardial metabolic requirements, coronary blood flow and left ventricular tension. For these reasons, influences that decrease AoP may be either detrimental or beneficial [34]. Thus, whereas increased preischemic AoP in HC animals indicates that such an increase is pathological for animals maintained on a high cholesterol diet, the failure of WGLL treatment to lower AoP suggests that the extract has negligible effect on this aspect of cardiovascular function. Data described in Section 2.4 of the Results Section of this article and shown in Figure 3 provide ventricular tissue expression profiles of proteins implicated in pathogenesis and adaptive response to atherosclerotic disease. Western blot analysis of myocardial tissue reveals significantly elevated content of HO-1 protein in tissue harvested from HCT animals, versus that taken from rabbits in the HC group. The heat shock protein HO-1 (hsp-32) is a major antioxidant defense enzyme, which is increased in response to trauma intrinsic to a wide range of diseases, including (and especially) atherosclerotic syndromes [35]. Often, the effects of a disease process overwhelm the protective capacity afforded by endogenous HO-1 expression [36,37,38]. However, its cardioprotective effects may be greatly amplified by the administration of pharmacological inducers, as demonstrated by the authors of the present report [28]. The cytoprotective effects correlating with increased expression of HO-1 are a likely effect of heme degradation by this enzyme to produce bilirubin and carbon monoxide (CO), both of which enhance the healthy function of cardiovascular tissue at the concentration normally produced by HO-1 activity during normal heme clearance. Therapeutic amplification of HO-1 in these studies was achieved using seed kernel extracts of Prunus cerasus (sour cherry). The present investigation demonstrates that lyophilisate of wild garlic leaf also mediates this effect. However, since this plant material also stimulates other protective effects, based on the data shown here, it cannot be determined to what extent WGLL-induced HO-1 expression contributes to the specific cardioprotective outcomes revealed. SOD1 levels measured by Western blot analysis in myocardial tissue of HC animals after ischemia/reperfusion injury were significantly elevated compared to the controls, while SOD1 expression in WGLL-treated animals was maintained at the normal (control) levels. Both apoptotic signaling and adaptive responses to oxidative stress involve processes for which SOD1 activity is a critical component. This enzyme produces molecular oxygen and hydrogen peroxide (H2O2) as end metabolites of its main activity, which is to neutralize superoxide [39]. H2O2 is itself a toxic reactive oxygen species (ROS) and may contribute to ischemia-reperfusion injury of myocardial tissue, through abnormally high apoptotic signaling and oxidative tissue damage in ischemic heart disease [40]. SOD1 is known to have a capacity to limit the detrimental effects of ROS by eliminating O2− to produce H2O2, which is eliminated by glutathione peroxidase or by catalase to harmless H2O and O2, but on the other hand, with free iron(II), H2O2 also can form free hydroxyl radicals by Fenton’s reaction (see the graphical abstract). High SOD levels along with considerable amounts of Fe2+ are associated with increased production of the highly toxic hydroxyl radical and may even enhance the extent of reperfusion injury [41]. An unbalance between the production of prooxidant H2O2 (SOD1) and antioxidants, such as glutathione peroxidase and catalase, in the cell might lead to a strengthened production of free radicals, which could lead to serious cellular damage. This is supported by the assessment of SOD activity in the blood of MI patients, which revealed that relative to healthy control individuals, SOD levels in the patient group were significantly higher [42]. This difference likely reflects a normal adaptive response to limit oxidative damage to the myocardium imposed by ischemic (and other) tissue injury. Western blot analyses conducted in this investigation revealed that COXIII and VEGF, which are both implicated in the pathophysiology of cardiovascular syndromes, were expressed at lower levels, both in HC and HCT groups, versus quantities of these proteins found in hearts harvested from the control animals. In COXIII and VEGF protein levels, there were no significant changes between WGLL-treated and hypercholesterolemic groups. Our results suggest that wild garlic may develop its cardioprotective activity via the heme/HO system and has no effect on COXIII and VEGF proteins. The extent of atherosclerotic plaque coverage on the intimal surface of hematoxylin-oil red O-stained rabbit aortas reveals negligible plaque on sections harvested from control animals maintained on diets with normal cholesterol content and a lesional extent of approximately 35% in sections from HC rabbits (Figure 4). The significantly reduced lesional coverage observed in WGLL-treated rabbits fed high cholesterol chow (HCT) is an effect also observed in previous work by these authors, in which HO-1 expression increased by adding sour cherry seed kernel extract to rabbit chow. This protected against dietary cholesterol-induced arterial plaque formation [28]. The blood of animals utilized in the present study was evaluated for serum analytes expressed at levels that may be used as diagnostic and therapy effect indicators for cardiovascular disease severity along a wide range of other severe inflammatory syndromes. The outcomes of serum parameter measurements are shown in Table 2. They reveal significantly elevated fasting plasma levels of TC and LDL cholesterol, which were two orders of magnitude higher, and HDL cholesterol concentration, which was eight-fold higher in HC and six-fold higher HCT rabbits versus the controls, effects that are an expected result of high cholesterol diets [43]. However, elevated levels of HDL cholesterol in WGLL-treated rabbits may indicate a cardioprotective property of the lyophilisate in the context of the beneficial effects of HDL. Significantly lower TC and LDL cholesterol levels were observed in WGLL-treated (HCT) animals versus groups fed with high cholesterol chow, but no WGLL (HC), demonstrating that the lyophilisate is protective with respect to the influence of these analytes. ApoA levels in the blood of HC animals were significantly lower versus rabbits fed normal chow (control). Thus, the lack of significant differences in the ApoA content of blood from animals fed normal diets (control) versus the content of this molecule in WGLL-treated rabbits maintained on high cholesterol (HCT) indicated a normalizing effect of the lyophilisate on this outcome. The significance of this result is that ApoA-I deficiency causes both hypertriglyceridemia and increased atherosclerosis in animal models [44], which can be counteracted by a WGLL-supported diet. Analysis of ApoB revealed that systemic levels of this analyte in rabbits from both HCT and HC groups were significantly higher versus its levels in the blood of animals fed chow with normal cholesterol content (controls). Moreover, ApoB levels detected in blood taken from rabbits in the HCT group were significantly lower in comparison to the content of this analyte in the HC group. This result is well correlated with LDL levels measured in the two groups. This finding was expected since ApoB is the primary apolipoprotein of chylomicrons, VLDL, IDL and LDL particles. Elevation in serum TG of the HC group was tendentious, but not significant compared to that of the controls. One interpretation is that triglyceride metabolism is unaffected by either influence within the constraints of the present study. Further analysis of blood from each of the three test groups revealed significant elevation of c-reactive protein (CRP) in HC animals versus the controls. Since CRP is a reliable systemic indicator of a wide range of inflammatory pathologies, this result implies that levels of dietary cholesterol administered to animals in this study managed to induce the onset of inflammatory processes. The analysis for serum liver enzyme activities demonstrated significantly elevated GOT in blood of the HC group animals versus the control and HCT groups, suggesting a hepatotoxic effect of elevated dietary cholesterol intake, an effect noted by previous investigators [45]. Finally, the significantly lower levels of LDH and CK observed in HCT animals, versus rabbits maintained on high cholesterol (HC), indicated that the effects of dietary supplementation with WGLL may have beneficial effects on impaired liver function caused by the hypercholesterolemic state. TNFα-induced ICAM-1 mRNA transcription, which has been demonstrated by in vitro studies to suppress the adhesion of monocytes to porcine arterial tissues and HUVECs, was significantly inhibited by treatment with alliin (S-allyl-l-cysteine sulfoxide). Moreover, alliin is also protected against the depolarization of mitochondrial membrane potential and overproduction of the superoxide anion, which occur as a downstream effect of TNFα, and may correlate with the suppression of NOX4 mRNA transcription by HUVECs. Additionally, treatment of HUVECs with alliin was also observed to reduce TNF-α-mediated ERK1/2 IjB and IjB (but not p38) phosphorylation. These results provide improved insight into the mechanisms by which alliin acts as a countermeasure to atherosclerotic pathomechanisms [46] and are consistent with the protective effects of the plant extract reported here. 4. Experimental Section (Methodology) 4.1. Sample Lyophilization and Bioanalytics Deep-frozen Allium ursinum leaves (Toltelekgyar Ltd., Zalakomar, Hungary) were lyophilized for 24 h in a Martin-Christ ALPHA 1-4 freeze dryer (Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, Germany) at an ambient pressure of 0.120 millibars (mb), with a condensor temperature of −50 °C and shelf temperature of 35 °C. The ratio of the frozen leaf to fresh and desiccated plant material was 5:6:1. HPLC analysis was accomplished using a Waters 600 system (Waters Corporation, Milford, CT, USA), equipped with a 2998 photodiode array detector, on-line degasser and auto sampler, using a reversed phase Phenomenex Synergi 4 μm Hydro-RP 80Å (250 mm × 4.6 mm) column (Phenomenex, Torrance, CA, USA). With a column temperature of 25 °C, a gradient elution was applied as follows: 0–15 min: 100% of Mobile Phase A (water + 0.1% phosphoric acid); 15–20 min: the ratio of Mobile Phase B (acetonitrile) increased to 100%; 20–25 min: 100% B; 25–27 min: A increased to 100%; 27–45 min: 100% A. The flow rate was 0.75 mL/min, and alliin was monitored at 204 nm. Alliin was detected at 4.6 min. Data acquisition and evaluation were performed using Empower Pro software. Alliin was purchased from LGC Standards. Acetonitrile used for chromatographic analysis (LiChrosolv® HPLC grade) was obtained from Merck (Merck Consumer Health Holding GmbH, Darmstadt, Germany). The Millipore Direct-Q UV3 clarifier (Merck Millipore, Molsheim, France) was used to produce purified water for HPLC measurements. Stock standard solutions of alliin were prepared with methanol and stored at 4 °C. The calibration range was 0.5–5 µg alliin/injection. The calibration standard was injected in triplicate at six volume levels. Extraction of alliin from the lyophilized plant material was carried out with 10 mL MeOH at room temperature for 3 min from a 1-g sample in an ultrasonic bath. After filtering through a filter membrane (Acrodisc® GHP 13 mm, 0.45 μm, Waters Corporation, Milford, CT, USA), 3 independently-prepared samples were analyzed in triplicate. 4.2. Animals The experiments were carried out using adult male New Zealand rabbits with an average body weight of 2.5–3.0 kg. The animals received human care in compliance with the “Principles of Laboratory Animal Care” by EU Directive 2010/63/EU for animal experiments. The duration of the adaptive feeding was 2 weeks. The rabbits were provided with laboratory rodent chow, or chow enriched with 2.0% cholesterol (Jurasko Ltd., Debrecen, Hungary), or cholesterol-supplemented chow containing 2% wild garlic leaf lyophilisate (WGLL) daily for 8 weeks ad libitum. Allium ursinum lyophilisate-containing chow was produced in the laboratory of Department of Pharmaceutical Technology, University of Debrecen. The comparison of the behavior and general health status of animals provided with unsupplemented rabbit chow versus feed containing other components showed no observable differences, with no indication that administration of feed acted as a confounder to the experiments conducted. 4.3. Echocardiography Echocardiographic examination of animals was conducted under light anesthesia (ketamine 15 mg/kg, xylazine 3 mg/kg (i.m.)) at the 8-week time point of the study [47]. The chest of each animal was shaved, and the rabbit was positioned in a dorsal decubitus position. Echocardiographic measurements were performed using a Siemens Acuson 512 sonograph (Siemens Healthcare GmbH, Erlangen, Germany), with a 7V3c probe at 7 MHz, with fundamental imaging (Figure 5). Conventional measurements were carried out in 2D and M-mode. Parasternal long axis views were obtained and recorded to ensure that the mitral and aortic valves, as well as the apex, were visualized. The parasternal short axis views were recorded at the mid-papillary muscle level. M-mode tracings were performed at the mid-papillary muscle level, either in parasternal long or short axis views. M-mode for visualization and quantification of wall motion in cardiovascular research was used; single line acquisition allows for the very high-temporal (1000 fps) resolution necessary for the analysis of LV function. Echocardiographic measurements included interventricular and left ventricular free-wall thickness in diastole and systole (IVSs, IVSd) and left ventricular internal diameter at end-diastole (LVIDd) and end-systole (LVIDs). End-systolic volume (ESV), end-diastolic volume (EDV), stroke volume (SV) and left ventricular mass were calculated. Fractional shortening was computed by using the equation FS = [(LVIDd − LVIDs)/LVIDd] × 100%, and the ejection fraction was calculated by using the following formula (Teicholz): EF = (LVEDD2 − LVESD2)/LVEDD2. In this latter formula LVEDD means Left Ventricular End Diastolic Dimension, while LVESD is Left Ventricular End Sistolic Dimension. Mitral and tricuspid annular peak systolic excursions (MAPSE and TAPSE) were assessed with M-mode, measuring the distance of mitral or tricuspid annular movement between end-diastole to end-systole. All measurements were averaged over three to five consecutive cardiac cycles. Doppler (PW) imaging of the mitral valve and aortic valve was obtained from the apical 4-chamber view and the apical 5-chamber view. From the mitral inflow velocity image, the following measurements were obtained: peak E and peak A waves (mitral early and late filling velocities), the E to A ratio (E/A) and deceleration time of early filling velocities (DecT). Aortic and left ventricular outflow tract (LVOT) parameters were also calculated: LVOTVmax, LVOT maxPG and LVOTEnvTi. Tissue velocity imaging (TVI) measurements were analyzed from the apical 4-chamber view and from the parasternal long axis and short axis views, as well. A 5-mm tissue sampling volume was obtained at the mitral annulus from both septal and lateral walls. From the acquired images, the following functional parameters were measured: S’, E’/A’ wave velocities, E/E’ (early diastolic mitral inflow velocity divided by average value of lateral and septal tissue Doppler early diastolic velocities) and E’/A’ (tissue Doppler early and late diastolic velocity ratio) [48]. 4.4. Measurement of Serum Parameters Blood samples were collected with EDTA-K2 evacuated tubes (BD Vacutainer, Becton Dickinson and Company, Franklin Lakes, NJ, USA) from the marginal ear vein of the animals, at the endpoint of the treatment. The samples were collected and processed aseptically to minimize hemolytic activity. The serum level of total cholesterol (TC), triglycerides (TG), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C) and the value of apolipoprotein A-I (ApoA), apolipoprotein B (ApoB), c-reactive protein (CRP), aspartate transaminase (also called glutamic oxaloacetic transaminase, GOT), lactate dehydrogenase (LDH) and creatine kinase (CK) were detected by automated analyzers in the Department of Laboratory Medicine at the University of Debrecen. 4.5. Isolated Working Heart Preparation (Langendorff Method) Each of the animals was anaesthetized with a mixture of ketamine/xylazine (50/5 mg/kg, intramuscularly). A bolus of heparin was administered (1000 U/kg of body weight, intravenously) 20 min before thoracotomy, to avoid thrombosis. Next, the chest cavity was opened and the pericardium incised. The heart was excised and immediately transferred to ice-cold modified Krebs–Henseleit (mKH) buffer (pH 7.4 on 37 °C, gassed with 95% O2 and 5% CO2 mixture) [49]. The aorta was then cannulated and the heart perfused for 10 min, retrogradely in Langendorff mode with mKH buffer to clear residual blood from each harvested organ. The perfusate had the following composition: NaCl, 118 mmol/L; NaHCO3, 25 mmol/L; KCl, 4.8 mmol/L; CaCl2, 1.8 mmol/L; Mg2SO4, 1.2 mmol/L; KH2PO4, 1.2 mmol/L; and 10 mM glucose. A dual-headed peristaltic pump controlled the rate of perfusion of mKH buffer. The left atrial appendage was incised, and the pulmonary veins were ligated. A small incision was made at the bifurcation of pulmonary arteries; thus all coronary effluent was collected by the pulmonary artery. Next, the circulation was switched to anterograde perfusion, in order to set the baseline parameters in working heart mode for 20 min. The following parameters were recorded and the resulting data analyzed using a pressure transducer attached to the aortic outflow line: aortic pressure (AoP), heart rate (beats/min), left ventricular developed pressure (LVDP). Aortic flow (AF, mL/min) and coronary flow (CF, mL/min) were measured by using a flowmeter. Hearts were then subjected to 30 min of global ischemia, then perfused for 15 min in Langendorff mode and converted to working heart mode for 105 min. The above-listed outcomes were measured and recorded during the reperfusion at the 30-, 60-, 90- and 120-min time points. Immediately following 120 min of reperfusion, small myocardial biopsies from LV heart tissue were removed and frozen for subsequent molecular biological analysis. 4.6. Histological Analysis of the Aortic Root Lipid staining was carried out with oil red O (Sigma Diagnostics, St. Louis, MO, USA) by use of the following protocol: aortic tissues were frozen in Optimal Cutting Temperature (OCT) medium (Thermo Fisher Scientific Inc., Waltham, MA, USA). Cryostated tissue sections were cut to a thickness of 6.0 µm and applied to Superfrost Plus slides (Daiggers, Vernon Hills, IL, USA). Atherosclerotic lesions in the aortic root were examined at 3 locations and each separated by 120 μm. Four to 5 serial sections were prepared from each location, starting beyond the aortic arch. The sections were stained, as described previously, with oil red O, followed by analysis of the lipid composition of the lesions, by calculating the percentage of oil red O-positive area, relative to the total cross-sectional vessel wall area. Nuclei were counterstained with hematoxylin (Sigma Diagnostics), using routine methods. All images were captured with a binocular light microscope (Carl Zeiss Microscopy GmbH, Jena, Germany) equipped with a video camera and analyzed using Scion Image software (Scion Corp., Torrance, CA, USA). 4.7. Extraction of Myocardial Protein Three hundred milligrams of frozen tissues from rabbit left ventricular myocardium were homogenized in 800 µL Buffer A (25 mM Tris-HCl, pH 8, 25 mM NaCl, 4 mM Na-orthovanadate, 10 mM NaF, 10 mM Na-pyrophosphate, 10 nM okadaic acid, 0.5 mM EDTA, 1 mM PMSF and protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA)) in a Polytron-homogenizer. Homogenates were centrifuged at 2000 rpm at 4 °C for 10 min. Supernatant from the above centrifugation was further centrifuged at 10,000 rpm at 4 °C for 20 min, and the resulting supernatant was used as the cytosolic extract. The nuclear pellet was resuspended in 400 µL of Buffer A with 0.1% Triton-X-100 and 500 mM NaCl, then lysed by incubation for one hour on ice. Homogenates were then centrifuged at 14,000 rpm at 4 °C for 10 min, and the supernatant was used as a mitochondrial lysate. Cytosolic mitochondrial extracts were aliquoted, snap frozen and stored at −80 °C for further investigations. The total protein concentration was assayed using the bicinchoninic acid (BCA) method with bovine serum albumin as the standard (Pierce, Rockford, IL, USA). 4.8. Western Blot Assays for Protein Expression in Cardiac Tissue Western blot analysis was used to evaluate left ventricular myocardial tissue for the expression level of the following proteins: heme-oxygenase 1 (HO-1), superoxide-dismutase 1 (SOD1), vascular endothelial growth factor A (VEGF), cytochrome c oxidase III (COXIII), cytochrome c oxidase IV (COXIV) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The total protein concentration in cytosolic and mitochondrial extract was determined using the BCA Protein Assay Kit. Next, the protein was diluted with Laemmli buffer and heated to 100 °C for 10 min. The denaturated samples were separated by SDS/polyacrylamide gel electrophoresis (SDS-PAGE) at 120 V for 90 min and transferred onto a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA) at 100 V for 1 h. Precision plus Protein Kaleidoscope standards (Bio-Rad Laboratories) were used as molecular-weight standards. The membranes were blocked in 5% low fat milk blocking buffer for 90 min and then incubated overnight at 4 °C with primary antibodies (Sigma-Aldrich). After being washed with Tris-buffered saline containing Tween 20 (TBS-T) three times for 10 min, the membranes were incubated with horseradish peroxidase-labeled secondary antibody diluted 1:2000 in TBS-T and 1% (wt/vol) nonfat dry milk for 90 min at room temperature. Enhanced chemiluminescent substrate (ECL, Litmus Scientific, Advansta Inc., Menlo Park, CA, USA) was used to identify bands. Detection was made by autoradiography for variable lengths of time with Medical X-Ray Film (Agfa-Gevaert N.V., Mortsel, Belgium). Quantitative analysis of scanned blots was carried out using the Scion for Windows Densitometry Image program Version Alpha 4.0.3.2 (Scion Corporation, Frederick, MD, USA). Signal intensity for bands corresponding to each protein of interest was estimated and reported in arbitrary units ± SEM. 4.9. Statistical Analysis All data are presented as the average magnitudes of each outcome in a group ± standard error of the mean (SEM). Statistical analysis was performed using one-way analysis of variance (ANOVA) followed by Kruskal–Wallis multiple comparison tests with GraphPad Prism software for Windows (GraphPad Software Inc., La Jolla, CA, USA). Probability values (p) less than 0.05 were considered statistically significant. 5. Conclusions Outcomes of the present report demonstrate that wild garlic leaf lyophilisate improves cardiac functions in isolated hearts harvested from WGLL-treated rabbits. The improvements observed include significantly better post-ischemic values of aortic flow in treated animals compared to the HC group (p < 0.05). Coronary flow measurements showed similar trends. Echocardiographic measurements showed improved diastolic heart functions in animals that ate an Allium ursinum lyophilisate-containing high-cholesterol diet. Impaired relaxation expressed as DecT and E/A ratios was found in HC animals, while parameters measured in WGLL-treated animals reached normal values. Systolic function expressed as FS and EF was also found significantly decreased in HC animals, and the process was greatly counteracted by WGLL treatment. Interestingly, better right ventricle functions were measured in treated animals (higher peak E-wave velocity and higher TAPSE values). Tissue staining showed significantly decreased atherosclerotic plaque formation in animals treated with HCT compared to the HC group. Total blood cholesterol levels in animals fed with 2% cholesterol-containing diet showed a dramatic increase after the eight-week period, while the values of the control group remain in the physiologic range. Cholesterol levels in animals treated with Allium ursinum lyophilisate were significantly lower compared to the HC group (p < 0.05). WGLL also had notable beneficial effects on the other monitored serum parameters (GOT, LDH, CK). Important novelties of this present report include the findings that increased dietary cholesterol intake may raise the level of SOD1 in cardiac tissue, which is associated with increased ROS-dependent tissue damage, and this may be counteracted by WGLL treatment; furthermore, our work revealed that WGLL supplementation could elevate the activity of the HO-1-mediated cardioprotective pathway in hypercholesterolemic circumstances. Acknowledgments This work was supported in part by the TÁMOP-4.2.2.D-15/1/KONV-2015-0016 project, implemented through the New Széchenyi Plan, and co-financed by the European Social Fund, and in part by KUTEGY (Bela Juhasz), University of Debrecen (Bela Juhasz), OTKA PD-78223, K109846, AGR-PIAC-13-1-2013-0008 and the TÁMOP-4.2.2.A-11/1/KONV-2012-0045 and TÁMOP-4.2.6.-15/1-2015-0001 (Daniel Priksz, Mariann Bombicz, Rudolf Gesztelyi, Balazs Varga, Bela Juhasz) projects co-financed by the European Union and the European Social Fund. This research was also supported by the European Union and the State of Hungary, co-financed by the European Social Fund in the framework of TÁMOP-4.2.4.A/2-11/1-2012-0001 ‘National Excellence Program (Mariann Bombicz, Balazs Varga.). The authors are sincerely grateful to Stephanie C. Fox of QueenBeeEdit in Bloomfield, CT, USA, for her hard work in organizing, formatting and editing this article. Author Contributions Mariann Bombicz: isolated working heart preparation, protein isolation and Western blot, histology; Daniel Priksz: echocardiography, isolated working heart preparation; Balazs Varga: animal treatment, histology, manuscript preparation; Rudolf Gesztelyi: isolated working heart preparation; Attila Kertesz: echocardiography; Peter Lengyel: statistical analysis, research plan; Peter Balogh: statistical analysis, research plan; Dezso Csupor: bioanalytics; Judit Hohmann: sample lyophilization; Harjit Pal Bhattoa: measurement of serum parameters; David D. Haines: native English speaker, data analysis and manuscript preparation; Bela Juhasz: echocardiography, corresponding author. Conflicts of Interest The authors declare no conflict of interest. Figure 1 High-performance liquid chromatography (HPLC) spectrum analysis: HPLC chromatogram of the wild garlic leaf lyophilisate (WGLL) reporting one major peak at 3.8 min. Figure 2 Effect of high cholesterol and WGLL on cardiac function. Hearts were isolated from three groups of animals (n = 6), defined as follows: non-hypercholesterolemic animals fed with normal chow (control); hypercholesterolemic group fed with 2% cholesterol-supplemented chow (HC); and a group of hypercholesterolemic animals treated with WGLL (HCT). Isolated working hearts harvested from each animal in each group were subjected to global ischemia followed by 120 min of reperfusion (I/R). Cardiac functions were registered before ischemia (preischemic) and 60 min after global ischemia (60 min of reperfusion). Results are shown as average values from each group of rabbits ± SEM of aortic flow (AF, mL/min, A); coronary flow (CF, mL/min, B); aortic pressure (AoP, Hgmm, C); heart rate (HR, beat/min, D); cardiac output (CO, mL/min, E); stroke volume (SV, mL, F). * p < 0.05 compared to the preischemic control; ** p < 0.05 compared to the 60 min of reperfusion control; ## p < 0.05 compared to the 60 min of reperfusion HC. Figure 3 Cardiac tissue biomarker expression: Western blot outcomes. The expression of HO-1 (A), COXIII (B), SOD-1 (C) and VEGF (D) protein in rabbit myocardial tissue was measured in homogenized left ventricular cardiac tissue samples drawn from three test groups (n = 6), defined as follows: I/R injured hearts from non-hypercholesterolemic animals fed with normal, non-cholesterol supplemented chow (control); I/R injured hearts from hypercholesterolemic animals fed with 2% cholesterol-supplemented chow (HC); and I/R-injured hearts harvested from hypercholesterolemic animals fed with 2% cholesterol and 2% wild garlic leaf lyophilisate-supplemented chow (HCT). GAPDH and COXIV expression levels were measured as reference proteins. Western blot analysis was conducted on each tissue homogenate in duplicate, and the signal intensity of the resulting bands corresponding to proteins of interest was measured using the Scion for Densitometry Image program, Alpha 4.0.2.3. The tissue content of each protein is shown in arbitrary units as the mean for each group of animal ± SEM. * p < 0.05 for comparison of the average expression levels of HO-1, COXIII, SOD-1 and VEGF in myocardium to the non-hypercholesterolemic group (control); # p < 0.05 for comparison to the hypercholesterolemic group (HC). Figure 4 Aortic histologic analysis. Histological sections of aortas from three groups of rabbits stained with hematoxylin-oil red O (A–C, magnification 25×). Bright red-stained lipid shows atherosclerotic plaques in the HC (B) and HCT groups (C). Internal elastic lamina is shown as IE; M, media; P, atherosclerotic plaques; and F, adventitial fatty tissue stained with oil red O. Comparison of plaque area in the three groups, expressed as a percentage of the total area (D). * p < 0.05 for comparison with the control group outcomes; # p < 0.05 for comparison with the HC group outcomes. Figure 5 Representative images of echocardiographic measurements. (A) Doppler image, mitral inflow velocities; (B) color Doppler image, velocity of left ventricle outflow tract; (C) M-mode image, tricuspidal annular plane systolic excursion (TAPSE); (D) M-mode image, parasternal long axis view (PLAX) of left ventricle. ijms-17-01284-t001_Table 1Table 1 Echocardiographic outcomes. Outcomes of echocardiographic evaluations on animals fed normal cholesterol-free rabbit chow (control); normal rabbit chow, containing 2% cholesterol (hypercholesterolemic, HC); rabbit chow containing 2% cholesterol + 2% WGLL (hypercholesterolemic treated, HCT). Outcomes evaluated included the following: heart rate (HR); beats per minute (bpm); aortic diameter (Ao); left ventricle (LV); right ventricle (RV); end-systolic diameter (ESD); end-diastolic diameter (EDD); parasternal long axis view (PLAX); short axis view (SAX); fractional shortening (FS) of the left ventricle; ejection fraction of the left ventricle (EF); calculated weight of the left ventricle (LV mass); peak mitral early diastolic inflow velocity/peak atrial diastolic inflow velocity (E/A); deceleration time of the E wave from maximum to baseline (DecT); peak mitral inflow velocity/average of spectral tissue Doppler peak early diastolic velocities at the septal and lateral corner of mitral annulus (E/E’); maximal velocity of left ventricle outflow (LVOTVmax); left ventricle outflow tract velocity time integral (LVOTVTI); peak early diastolic velocity of the lateral wall, spectral tissue Doppler/peak atrial diastolic velocity of the lateral wall, spectral tissue Doppler (E’/A’); peak systolic velocity (S); mitral annular plane systolic excursion (MAPSE); and tricuspidal annular plane systolic excursion (TAPSE). Mean ± SEM HR (bpm) Ao (cm) LV ESD (cm) LV EDD (cm) FS PLAX (%) Control 180.8 ± 4.145 0.946 ± 0.024 1.016 ± 0.091 1.655 ± 0.050 39.370 ± 5.021 HC 150.2 ± 4.303 * 0.919 ± 0.025 1.242 ± 0.045 * 1.756 ± 0.063 29.220 ± 0.803 * HCT 185.0 ± 7.053 ** 0.898 ± 0.012 1.184 ± 0.020 1.793 ± 0.031 33.820 ± 1.312 ** EF PLAX (%) LV mass PLAX (g) FS SAX (%) EF SAX (%) LV mass SAX (g) Control 56.910 ± 1.294 6.632 ± 0.478 32.310 ± 0.717 54.130 ± 0.961 6.573 ± 0.351 HC 49.810 ± 1.140 * 8.218 ± 0.628 29.010 ± 1.056 * 49.430 ± 1.517 * 8.315 ± 0.792 * HCT 55.990 ± 1.756 ** 8.769 ± 0.169 * 32.970 ± 1.131 ** 54.930 ± 1.522 ** 8.195 ± 0.226 * E/A DecT (ms) E/E’ LVOTVmax (cm/s) LVOTVTI (cm) Control 1.376 ± 0.045 71.250 ± 4.101 1.417 ± 0.058 84.280 ± 2.131 0.071 ± 0.002 HC 1.207 ± 0.037 * 87.440 ± 3.534 * 1.775 ± 0.101 87.940 ± 5.719 0.080 ± 0.005 HCT 1.344 ± 0.076 69.540 ± 4.787 ** 1.718 ± 0.155 77.150 ± 2.157 0.0685 ± 0.002 E’/A’ (lateral) MAPSE (cm) RV S′ (cm/s) RV E′/A′ TAPSE (cm) Control 1.303 ± 0.058 0.527 ± 0.018 8.935 ± 0.273 1.336 ± 0.051 0.576 ± 0.012 HC 1.109 ± 0.071 0.571 ± 0.025 8.103 ± 0.215 * 1.233 ± 0.092 0.576 ± 0.015 HCT 1.065 ± 0.117 0.592 ± 0.030 * 9.156 ± 0.210 ** 1.055 ± 0.077 * 0.644 ± 0.020 ** * p < 0.05 in comparison to mean the values of the control group; ** p < 0.05 in comparison to the mean values of the HC group. ijms-17-01284-t002_Table 2Table 2 Serum biomarkers of cardiac function. Average serum TC, TG, LDL-C, HDL-C (mmol/L), ApoA and ApoB (g/L), CRP (mg/L), GOT, LDH and CK (U/L) levels in 3 groups of rabbits (n = 6), were analyzed by an automated hematology analyzer. Each analysis was conducted on peripheral blood serum harvested from animals following the 8-week treatment periods, for: non-hypercholesterolemic (control), hypercholesterolemic rabbits (HC) and hypercholesterolemic animals receiving WGLL-supplemented chow (HCT). Results are shown as the average values from each group of animals ± SEM of serum total cholesterol (serum cholesterol, mmol/L) and low-density lipoprotein cholesterol (LDL-C, mmol/L). Groups TC LDL-C HDL-C ApoA ApoB Control 0.793 ± 0.067 0.230 ± 0.023 0.523 ± 0.045 0.042 ± 0.005 0.015 ± 0.003 HC 26.370 ± 3.660 *,# 23.550 ± 3.032 *,# 4.427 ± 0.656 * 0.022 ± 0.003 *,# 0.280 ± 0.063 *,# HCT 20.030 ± 1.947 *,** 17 ± 1.942 *,** 3.314 ± 0.369 * 0.056 ± 0.008 ** 0.172 ± 0.019 *,** TG CRP GOT LDH CK Control 0.788 ± 0.035 0.100 ± 0.014 29.670 ± 2.895 726.400 ± 170.700 4213 ± 623.200 HC 1.367 ± 0.335 0.728 ± 0.362 * 48.800 ± 16.220 * 791.900 ± 325.400 4955 ± 1353 HCT 1.052 ± 0.339 0.596 ± 0.231 24.910 ± 2.708 ** 316.600 ± 37.170 ** 2851 ± 600.800 ** * p < 0.05 compared to the control group values; ** p < 0.05 compared to the HC group values; # p < 0.05 compared to the HCT group values. ==== Refs References 1. Kilian R. Hanelt P. Research I.P.G.C.P. Kilian W. Mansfeld’s Encyclopedia of Agricultural and Horticultural Crops: Except Ornamentals Springer Berlin & Heidelberg, Germany 2001 2. Assessment G.F.I.F.R. Risk of mix-up with bear’s garlic Available online: http://www.bfr.bund.de/en/press_information/2005/10/risk_of_mix_up_with_bears_garlic-6228.html (accessed on 20 April 2016) 3. Yoo K. Pike L. Determination of flavor precursor compound s-alk(en)yl-l-cysteine sulfoxides by an HPLC method and their distribution in Allium species Sci. Hortic. 1998 75 1 10 4. Djurdjevic L. Dinic A. Pavlovic P. Mitrovic M. Karadzic B. Tesevic V. Allelopathic potential of Allium ursinum L. Biochem. Syst. Ecol. 2004 32 533 544 10.1016/j.bse.2003.10.001 5. Gitin L. Dinica R. Parnavel R. The influence of extraction method on the apparent content of bioactive compounds in romanian Allium spp. Leaves Not. Bot. Horti Agrobot. Cluj-Napoca 2012 40 93 97 6. Gođevac D. Vujisić L. Mojović M. Ignjatović A. Spasojević I. Vajs V. Evaluation of antioxidant capacity of Allium ursinum L. Volatile oil and its effect on membrane fluidity Food Chem. 2008 107 1692 1700 10.1016/j.foodchem.2007.10.017 7. Sendl A. Elbl G. Steinke B. Redl K. Breu W. Wagner H. Comparative pharmacological investigations of Allium ursinum and Allium sativum Planta Med. 1992 58 1 7 10.1055/s-2006-961378 1620734 8. Bak I. Lekli I. Juhasz B. Nagy N. Varga E. Varadi J. Gesztelyi R. Szabo G. Szendrei L. Bacskay I. Cardioprotective mechanisms of prunus cerasus (sour cherry) seed extract against ischemia-reperfusion-induced damage in isolated rat hearts Am. J. Physiol.. Heart Circ. Physiol. 2006 291 H1329 H1336 10.1152/ajpheart.01243.2005 16617126 9. Lekli I. Szabo G. Juhasz B. Das S. Das M. Varga E. Szendrei L. Gesztelyi R. Varadi J. Bak I. Protective mechanisms of resveratrol against ischemia-reperfusion-induced damage in hearts obtained from zucker obese rats: The role of glut-4 and endothelin Am. J. Physiol. Heart Circ. Physiol. 2008 294 H859 H866 10.1152/ajpheart.01048.2007 18065527 10. Durrington P. Dyslipidaemia Lancet 2003 362 717 731 10.1016/S0140-6736(03)14234-1 12957096 11. Posa A. Szabo R. Kupai K. Csonka A. Szalai Z. Veszelka M. Torok S. Daruka L. Varga C. Exercise training and calorie restriction influence the metabolic parameters in ovariectomized female rats Oxid. Med. Cell. Longev. 2015 2015 787063 10.1155/2015/787063 25874022 12. Bhatnagar D. Soran H. Durrington P.N. Hypercholesterolaemia and its management BMJ 2008 337 a993 10.1136/bmj.a993 18719012 13. Finn A.V. Nakano M. Narula J. Kolodgie F.D. Virmani R. Concept of vulnerable/unstable plaque Arterioscler. Thromb. Vasc. Biol. 2010 30 1282 1292 10.1161/ATVBAHA.108.179739 20554950 14. Epure L.I. Roman G.V. Mărăcineanu R. Studies on medicinal and aromatic plants used in the therapeutic recipes in the bucharest university hospital Sci. Pap. Univ. Agron. Sci. Vet. Med. Buchar. Ser. A Agron. 2011 LIV 304 313 15. Hiyasat B. Sabha D. Grotzinger K. Kempfert J. Rauwald J.W. Mohr F.W. Dhein S. Antiplatelet activity of Allium ursinum and Allium sativum Pharmacology 2009 83 197 204 10.1159/000196811 19174616 16. Sabha D. Hiyasat B. Grotzinger K. Hennig L. Schlegel F. Mohr F.W. Rauwald H.W. Dhein S. Allium ursinum L.: Bioassay-guided isolation and identification of a galactolipid and a phytosterol exerting antiaggregatory effects Pharmacology 2012 89 260 269 10.1159/000337380 22507958 17. Preuss H.G. Clouatre D. Mohamadi A. Jarrell S.T. Wild garlic has a greater effect than regular garlic on blood pressure and blood chemistries of rats Int. Urol. Nephrol. 2001 32 525 530 10.1023/A:1014417526290 11989540 18. Rietz B. Isensee H. Strobach H. Makdessi S. Jacob R. Cardioprotective actions of wild garlic (Allium ursinum ) in ischemia and reperfusion Mol. Cell. Biochem. 1993 119 143 150 10.1007/BF00926865 8455576 19. Haines D.D. Juhasz B. Tosaki A. Management of multicellular senescence and oxidative stress J. Cell. Mol. Med. 2013 17 936 957 10.1111/jcmm.12074 23789967 20. Rodriguez-Porcel M. Lerman L.O. Holmes D.R. Jr. Richardson D. Napoli C. Lerman A. Chronic antioxidant supplementation attenuates nuclear factor-κβ activation and preserves endothelial function in hypercholesterolemic pigs Cardiovasc. Res. 2002 53 1010 1018 10.1016/S0008-6363(01)00535-1 11922911 21. Gordts S.C. van Craeyveld E. Muthuramu I. Singh N. Jacobs F. de Geest B. Lipid lowering and HDL raising gene transfer increase endothelial progenitor cells, enhance myocardial vascularity, and improve diastolic function PLoS ONE 2012 7 1284 10.1371/journal.pone.0046849 23056485 22. Van Craeyveld E. Jacobs F. Gordts S.C. de Geest B. Low-density lipoprotein receptor gene transfer in hypercholesterolemic mice improves cardiac function after myocardial infarction Gene Ther. 2012 19 860 871 10.1038/gt.2011.147 21975462 23. Iloki-Assanga S.B. Lewis-Lujan L.M. Lara-Espinoza C.L. Gil-Salido A.A. Fernandez-Angulo D. Rubio-Pino J.L. Haines D.D. Solvent effects on phytochemical constituent profiles and antioxidant activities, using four different extraction formulations for analysis of bucida buceras l. And phoradendron californicum BMC Res. Notes 2015 8 1284 10.1186/s13104-015-1388-1 26323940 24. Halpern G.M. Anti-inflammatory effects of a stabilized lipid extract of Perna canaliculus (Lyprinol® ) Allerg. Immunol. 2000 32 272 278 25. Rybak M.E. Calvey E.M. Harnly J.M. Quantitative determination of allicin in garlic: Supercritical fluid extraction and standard addition of alliin J. Agric. Food Chem. 2004 52 682 687 10.1021/jf034853x 14969516 26. Ariga T. Seki T. Antithrombotic and anticancer effects of garlic-derived sulfur compounds: A review BioFactors 2006 26 93 103 10.1002/biof.5520260201 16823096 27. Salman H. Bergman M. Bessler H. Punsky I. Djaldetti M. Effect of a garlic derivative (alliin) on peripheral blood cell immune responses Int. J. Immunopharmacol. 1999 21 589 597 10.1016/S0192-0561(99)00038-7 10501628 28. Juhasz B. Kertesz A. Balla J. Balla G. Szabo Z. Bombicz M. Priksz D. Gesztelyi R. Varga B. Haines D.D. Cardioprotective effects of sour cherry seed extract (SCSE) on the hypercholesterolemic rabbit heart Curr. Pharm. Des. 2013 19 6896 6905 10.2174/138161281939131127120517 23590152 29. Tribouilloy C. Grigioni F. Avierinos J.F. Barbieri A. Rusinaru D. Szymanski C. Ferlito M. Tafanelli L. Bursi F. Trojette F. Survival implication of left ventricular end-systolic diameter in mitral regurgitation due to flail leaflets a long-term follow-up multicenter study J. Am. Coll. Cardiol. 2009 54 1961 1968 10.1016/j.jacc.2009.06.047 19909877 30. Carluccio E. Biagioli P. Zuchi C. Bardelli G. Murrone A. Lauciello R. D’Addario S. Mengoni A. Alunni G. Ambrosio G. Fibrosis assessment by integrated backscatter and its relationship with longitudinal deformation and diastolic function in heart failure with preserved ejection fraction Int. J. Cardiovasc. Imaging 2016 32 1 10 10.1007/s10554-016-0881-5 27076221 31. Csepanyi E. Czompa A. Haines D. Lekli I. Bakondi E. Balla G. Tosaki A. Bak I. Cardiovascular effects of low versus high-dose beta-carotene in a rat model Pharmacol. Res. 2015 100 148 156 10.1016/j.phrs.2015.07.021 26225824 32. Czompa A. Gyongyosi A. Czegledi A. Csepanyi E. Bak I. Haines D.D. Tosaki A. Lekli I. Cardioprotection afforded by sour cherry seed kernel: The role of heme oxygenase-1 J. Cardiovasc. Pharmacol. 2014 64 412 419 10.1097/FJC.0000000000000132 24949584 33. Haines D.D. Bak I. Ferdinandy P. Mahmoud F.F. Al-Harbi S.A. Blasig I.E. Tosaki A. Cardioprotective effects of the calcineurin inhibitor FK506 and the PAF receptor antagonist and free radical scavenger, EGb 761, in isolated ischemic/reperfused rat hearts J. Cardiovasc. Pharmacol. 2000 35 37 44 10.1097/00005344-200001000-00005 10630731 34. Isoyama S. Maruyama Y. Ashikawa K. Sato S. Suzuki H. Watanabe J. Shimizu Y. Ino-Oka E. Takishima T. Effects of afterload reduction on global left ventricular and regional myocardial functions in the isolated canine heart with stenosis of a coronary arterial branch Circulation 1983 67 139 147 10.1161/01.CIR.67.1.139 6847793 35. Haines D.D. Lekli I. Teissier P. Bak I. Tosaki A. Role of haeme oxygenase-1 in resolution of oxidative stress-related pathologies: Focus on cardiovascular, lung, neurological and kidney disorders Acta Physiol. 2012 204 487 501 10.1111/j.1748-1716.2011.02387.x 22118298 36. Kupai K. Szabo R. Veszelka M. Awar A.A. Torok S. Csonka A. Barath Z. Posa A. Varga C. Consequences of exercising on ischemia-reperfusion injury in type 2 diabetic Goto-Kakizaki rat hearts: Role of the HO/NOS system Diabetol. Metab. Syndr. 2015 7 85 10.1186/s13098-015-0080-x 26448786 37. Posa A. Kupai K. Menesi R. Szalai Z. Szabo R. Pinter Z. Palfi G. Gyongyosi M. Berko A. Pavo I. Sexual dimorphism of cardiovascular ischemia susceptibility is mediated by heme oxygenase Oxid. Med. Cell. Longev. 2013 2013 521563 24163720 38. Posa A. Pavo I. Varga C. Heme oxygenase contributes to estradiol and raloxifene-induced vasorelaxation in estrogen deficiency Int. J. Cardiol. 2015 189 252 254 10.1016/j.ijcard.2015.04.112 25897920 39. Danial N.N. Korsmeyer S.J. Cell death: Critical control points Cell 2004 116 205 219 10.1016/S0092-8674(04)00046-7 14744432 40. Maslov L.N. Naryzhnaia N.V. Podoksenov Iu K. Prokudina E.S. Gorbunov A.S. Zhang I. Pei Zh M. Reactive oxygen species are triggers and mediators of an increase in cardiac tolerance to impact of ischemia-reperfusion Rossiiskii fiziologicheskii zhurnal imeni IM Sechenova 2015 101 3 24 (In Russian) 41. Mao G.D. Thomas P.D. Lopaschuk G.D. Poznansky M.J. Superoxide dismutase (SOD)-catalase conjugates. Role of hydrogen peroxide and the fenton reaction in sod toxicity J. Biol. Chem. 1993 268 416 420 8380162 42. Bagatini M.D. Martins C.C. Battisti V. Gasparetto D. da Rosa C.S. Spanevello R.M. Ahmed M. Schmatz R. Schetinger M.R. Morsch V.M. Oxidative stress versus antioxidant defenses in patients with acute myocardial infarction Heart Vessels 2011 26 55 63 10.1007/s00380-010-0029-9 20978900 43. Stange E. Agostini B. Paenberg J. Changes in rabbit lipoprotein properties by dietary cholesterol, and saturated and polyunsaturated fats Atherosclerosis 1975 22 125 148 10.1016/0021-9150(75)90073-8 168909 44. Cohen E. Aviram M. Khatib S. Volkova N. Vaya J. Human carotid atherosclerotic plaque protein(s) change HDL protein(s) composition and impair HDL anti-oxidant activity BioFactors 2016 42 115 128 26662883 45. Kim E.J. Kim B.H. Seo H.S. Lee Y.J. Kim H.H. Son H.H. Choi M.H. Cholesterol-induced non-alcoholic fatty liver disease and atherosclerosis aggravated by systemic inflammation PLoS ONE 2014 9 1284 10.1371/journal.pone.0097841 24901254 46. Hui C. Like W. Yan F. Tian X. Qiuyan W. Lifeng H. S -Allyl-l -cysteine sulfoxide inhibits tumor necrosis factor-α induced monocyte adhesion and intercellular cell adhesion molecule-1 expression in human umbilical vein endothelial cells Anat. Rec. 2010 293 421 430 10.1002/ar.21070 20091890 47. Kertesz A. Bombicz M. Priksz D. Balla J. Balla G. Gesztelyi R. Varga B. Haines D.D. Tosaki A. Juhasz B. Adverse impact of diet-induced hypercholesterolemia on cardiovascular tissue homeostasis in a rabbit model: Time-dependent changes in cardiac parameters Int. J. Mol. Sci. 2013 14 19086 19108 10.3390/ijms140919086 24048247 48. Kovacs A. Fulop G.A. Kovacs A. Csipo T. Bodi B. Priksz D. Juhasz B. Beke L. Hendrik Z. Mehes G. Renin overexpression leads to increased titin-based stiffness contributing to diastolic dysfunction in hypertensive mRen2 rats Am. J. Physiol. Heart Circ. Physiol. 2016 310 H1671 H1682 10.1152/ajpheart.00842.2015 27059079 49. Juhasz B. Varga B. Czompa A. Bak I. Lekli I. Gesztelyi R. Zsuga J. Kemeny-Beke A. Antal M. Szendrei L. Postischemic cardiac recovery in heme oxygenase-1 transgenic ischemic/reperfused mouse myocardium J. Cell. Mol. Med. 2011 15 1973 1982 10.1111/j.1582-4934.2010.01153.x 20716121

PMC005xxxxxx/PMC5000682.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081285ijms-17-01285Article1α,25(OH)2D3 Suppresses the Migration of Ovarian Cancer SKOV-3 Cells through the Inhibition of Epithelial–Mesenchymal Transition Hou Yong-Feng 1†Gao Si-Hai 2†Wang Ping 3†Zhang He-Mei 4Liu Li-Zhi 5Ye Meng-Xuan 6Zhou Guang-Ming 7Zhang Zeng-Li 8*Li Bing-Yan 6*Perez-Fernandez Roman Academic Editor1 Department of Toxicology, School of Public Health, Soochow University, Suzhou 215123, China; [email protected] Department of Nutrition and Food Hygiene, Wenzhou Center for Disease Control and Prevention, Wenzhou 325000, China; [email protected] Organ Transplant Institute, Fuzhou General Hospital, Fuzhou 350025, China; [email protected] AIDS-STDs Prevention and Control Department, Wenzhou Center for Disease Control and Prevention, Wenzhou 325000, China; [email protected] State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou 510060, China; [email protected] Department of Nutrition and Food Hygiene, School of Public Health, Soochow University, Suzhou 215123, China; [email protected] Department of Radiation Biology, School of Radiation Medication and Protection, Soochow University, Suzhou 215123, China; [email protected] Department of Labor Hygiene and Environmental Health, School of Public Health, Soochow University, Suzhou 215123, China* Correspondence: [email protected] (Z.-L.Z.); [email protected] (B.-Y.L.); Tel.: +86-512-6588-2636 (B.-Y.L.)† These authors contributed equally to this work. 19 8 2016 8 2016 17 8 128511 5 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Ovarian cancer is the most lethal gynecological malignancy due to its high metastatic ability. Epithelial-mesenchymal transition (EMT) is essential during both follicular rupture and epithelium regeneration. However, it may also accelerate the progression of ovarian carcinomas. Experimental studies have found that 1α,25-dihydroxyvitamin-D3 [1α,25(OH)2D3] can inhibit the proliferation of ovarian cancer cells. In this study, we investigated whether 1α,25(OH)2D3 could inhibit the migration of ovarian cancer cells via regulating EMT. We established a model of transient transforming growth factor-β1(TGF-β1)-induced EMT in human ovarian adenocarcinoma cell line SKOV-3 cells. Results showed that, compared with control, 1α,25(OH)2D3 not only inhibited the migration and the invasion of SKOV-3 cells, but also promoted the acquisition of an epithelial phenotype of SKOV-3 cells treated with TGF-β1. We discovered that 1α,25(OH)2D3 increased the expression of epithelial marker E-cadherin and decreased the level of mesenchymal marker, Vimentin, which was associated with the elevated expression of VDR. Moreover, 1α,25(OH)2D3 reduced the expression level of transcription factors of EMT, such as slug, snail, and β-catenin. These results indicate that 1α,25(OH)2D3 suppresses the migration and invasion of ovarian cancer cells by inhibiting EMT, implying that 1α,25(OH)2D3 might be a potential therapeutic agent for the treatment of ovarian cancer. vitamin Dovarian cancermigrationEMT ==== Body 1. Introduction Ovarian cancer has the highest fatality rate among women, primarily due to advanced stage at diagnosis, the lack of effective therapies for late-stage, and the replase after chemotherapy and surgery, which result in poor overall survival for the patients [1]. About 90% of all ovarian cancers are epithelial ovarian cancer, which originated from ovarian surface epithelium. Hence, it is crucial to throw light on relevant molecular mechanisms of epithelial ovarian cancer progression for seeking targeted therapy that can help improve survival. Epithelial-mesenchymal transition (EMT) is a reversible cellular process by which epithelial cells depolarize, lose cell–cell contacts, and gain a spindle-mesenchymal morphology. It is characterized by loss of epithelial morphology and cytoskeletal reorganization, rendering cells more migratory and invasive [2]. This process is essential for embryonic development and wound healing. The ovarian surface epithelium can transform back and forth between epithelial and mesenchymal phenotypes in both follicular rupture and subsequent ovarian remodeling [3]. Moreover, tumor cells, through EMT, can enhance invasion and acquire properties of cancer stem-like cells, secondary tumor-initiating and chemoresistance [4,5,6]. Recent research suggests that EMT plays a critical role in the progression of ovarian carcinomas [7,8,9]. Therefore, it is essential to develop newer therapeutic methods with complete efficacy and low-toxicity toward metastatic cancer. Over the past two decades, vitamin D has been inspected preclinically for its efficacy in chemopreventive and anticancer therapy [7,8]. Experimental studies suggest that 1α,25(OH)2D3, active metabolite of vitamin D, and its synthetic derivatives, protect against ovarian cancer, manifesting anti-proliferative and pro-apoptotic effects in ovarian cancer cell lines [6,7,8,9,10,11,12] and anti-tumorigenesis in animal models [13,14,15]. However, the mechanisms of vitamin D inhibiting ovarian cancer remain largely unknown. The transforming growth factor (TGF)-β signaling pathway is a key inducer of EMT. In this study, we investigated whether 1α,25(OH)2D3 suppresses migration and invasion of human ovarian adenocarcinoma cell line SKOV-3 cells by regulating EMT. We found that TGF-β-induced EMT in ovarian cancer cells could be inhibited by 1α,25(OH)2D3. 2. Results 2.1. 1α,25(OH)2D3 Inhibits the Migration of Human Ovarian Cancer SKOV-3 Cells Our previous study has demonstrated that 1α,25(OH)2D3 inhibited the proliferation of SKOV-3 cells in a dose-dependent manner [12]. We wondered whether 1α,25(OH)2D3 could also inhibit the migration of these cells. After SKOV-3 cells were treated by 1, 10, or 100 nmol/L of 1α,25(OH)2D3, the cell migration decreased in both a time- and dose-dependent manner (Figure 1A, p < 0.05). Meanwhile, 1α,25(OH)2D3 shortened the moving distance and reduced the moving speed of SKOV-3 cells compared with control group (Supplementary Materials Figure S1). Furthermore, we analyzed the expression of E-cadherin and Vimentin, biomarkers of EMT. We found the increased expression of E-cadherin and decreased expression of Vimentin in the cytoplasm of SKOV-3 cells treated with 1α,25(OH)2D3 for 24 h (Figure 1C). The results of Western blotting also showed that the expression of E-cadherin was significantly increased, and Vimentin was clearly decreased when treated with 10 or 100 nmol/L of 1α,25(OH)2D3 (Figure 1C, p < 0.05). Collectively, 1α,25(OH)2D3 inhibited the migration of ovarian cancer cells, which was associated with the altered expression of E-cadherin and Vimentin. 2.2. Establishment of TGF-β1-Induced EMT in SKOV-3 Cells Our findings that 1α,25(OH)2D3 increases the expression of E-cadherin and decreases that of Vimentin prompted us to study whether 1α,25(OH)2D3 inhibited EMT or not. Thus, we first established a model of TGF-β1-induced EMT of SKOV-3 cells. After being stimulated with 10 ng/mL TGF-β1 for 24 h, cell morphology changed from pebble-like epithelial to spindle-like mesenchymal, and it gradually elongated in 72 h (Figure 2A). The administration of TGF-β1 promoted cell migration and pseudopodium stretching frequency in 36 h (Figure 2B). Western blotting analyses confirmed the EMT phenotype of TGF-β1-treated SKOV-3 cells with decreased expression level of E-cadherin but increased Vimentin, compared with untreated cells (Figure 2C, p < 0.05). The expression of Slug, a transcription factor of EMT, rapidly increased in 24 h after cells were treated with TGF-β1 (p < 0.05). Taken together, results of morphological changes and protein expression patterns strongly indicated that we successfully established an experimental model of TGF-β1-induced EMT in ovarian cancer SKOV-3 cells. 2.3. 1α,25 (OH)2D3 Inhibited the Migration and Invasion of SKOV-3 Cells during TGF-β1-Induced EMT One of the functional changes of EMT is the increase in migration and invasion capacities, typically characteristics of mesenchymal cells. Thus, we determined whether 1α,25(OH)2D3 decreases cell migration and invasion accompanied with the TGF-β1-induced EMT. Compared with negative control, 1α,25(OH)2D3 alone significantly decreased the migration of SKOV-3 cells, while TGF-β dramatically increased the cell migration. However, the elevation of migration by TGF-β1 was significantly decreased by the treatment together with 1α,25(OH)2D3 (Figure 3A, p < 0.05). 1α,25(OH)2D3 similarly reversed the shortened motion tracking during TGF-β1-induced EMT, compared to 1α,25(OH)2D3-untreated cells (Supplementary Materials, Figure S2). Subsequently, we determined the invasion ability of cells in vitro after being treated with 1α,25(OH)2D3. As showed in Figure 3B, TGF-β1 dramatically increased the invasion ability of SKOV-3 cells, which were the most important characteristics of a metastatic cell. In contrast, 1α,25(OH)2D3 substantially inhibited the invasion. These results were consistent with data obtained by the cell migration assay, indicating that 1α,25(OH)2D3 had a greater inhibitory ability for migration and invasion of ovarian cancer cells. 2.4. 1α,25(OH)2D3 Regulates the Expression of EMT Markers in TGF-β1-Treated SKOV-3 Cells Subsequently, we wanted to determine whether 1α,25(OH)2D3 regulates the protein level of key EMT markers in TGF-β1-treated cells. Figure 4A showed representative images of immunofluorescence staining for E-cadherin and Vimentin in SKOV-3 cells. Compared to TGF-β1-treated cells, E-cadherin, the marker of epithelial cells, increased, while Vimentin, the marker of mesenchymal cells, decreased. Quantification of Western blotting results revealed that these alterations were significant at 1α,25(OH)2D3-treated cells when compared with the corresponding controls (Figure 4D, p < 0.05). These results indicated that the promotion of TGF-β1- induced EMT could be inhibited by 1α,25(OH)2D3. One of the EMT characteristics is the increased expression of EMT-related transcription factors including Snail, Slug and β-catenin. Figure 4B showed that the expression of snail, slug, and β-catenin increased in TGF-β1-treated cells. However, the administration of 1α,25(OH)2D3 resulted in the decrease of these compared to TGF-β1-treated cells. Western blotting results also showed a similar pattern of EMT-related proteins (Figure 4E, p < 0.05). These results demonstrated that 1α,25(OH)2D3 could reverse TGF-β1-induced EMT in SKOV-3 cells. 1α,25(OH)2D3 mediates target genes by binding to the vitamin D receptor (VDR). The results from both immunofluorescence and Western blotting showed that the expression of VDR in SKOV-3 cells only treated by TGF-β1 was lower than that in cells treated with the combination of 1α,25(OH)2D3 and TGF-β1 (Figure 4C,F, p < 0.05). These results showed that 1α,25(OH)2D3 inhibited the TGF-β1-induced EMT accompanied with increased expression of VDR in ovarian cancer cells. 3. Discussion Ovarian cancer is the most lethal gynaecological cancer, and Epithelial-mesenchymal transition (EMT) was reported to be association with ovarian cancer cell dissemination and invasion [16]. To the best of our knowledge, this study is the first to report that 1α,25(OH)2D3 decreased the migration and invasion of human ovarian adenocarcinoma cell line SKOV-3 cells through inhibiting TGF-β1-induced EMT. This result is also in accordance with previous reports indicating the anti-metastasis potential of 1α,25(OH)2D3 in other types of cancer cells, including colon [15], breast [17], pancreatic [18], and lung cancers [19]. The normal ovarian surface epithelium exhibit epithelial and mesenchymal characteristics by the expression of both keratin and vimentin [20]. It is believed that ovarian surface epithelial cells adapt to changes by transitions between epithelial and mesenchymal stages during both follicular rupture and epithelium regeneration. Moreover, this plasticity may lie on the origin of ovarian cancer, an important initiating event in promoting tumor cell dissemination leading to metastasis [21]. We found that expression of both E-cadherin and Vimentin were observed in human ovarian cancer SKOV-3 cells. 1α,25(OH)2D3 inhibited migration of SKOV-3 cells accompanied with the decreased expression of E-cadherin but increased Vimentin. Furthmore, TGF-β1-induced EMT in ovarian cancer cells was also inhibited by 1α,25(OH)2D3. TGF-β is a multifunctional cytokine that acts as a tumor suppressor in early stages through stopping proliferation, inducing differentiation, or promoting apoptosis but promotes tumor progression in late stages through multiple mechanisms, including inducing EMT in cancer cells [22]. In addition, TGF-β is frequently used as a key inducer of EMT for experimental models. Other than changes of epithelial and mesenchymal markers, EMT is characterized by altered location of transcription factors, such as β-catenin, Slug, Snail, Twist and Sox10. In the present study, TGF-β1 promoted the expression of Slug, Snail and β-catenin but also increased their localization in the nuclei of ovarian cancer cells. For another study, TGF-β1-induced EMT promotes breast cancer cell migration toward lymphatic endothelial cells by activating CCR7 [23]. Snail1 is a cofactor for Smad3/4 during TGF-β-induced EMT, and a strong correlation was also found between loss of CAR and E-cadherin and nuclear co-expression of Snail1 and Smad3/4 in breast cancer [24]. TGF-β-induced CD59 expression during EMT is dependent on Smad3 but not on Smad2 in lung cancer A549 cells [25]. Liu et al reported that the JAK/STAT3 pathway is required for TGF-β-induced EMT, and the IL-6/JAK/STAT3 and TGF-β/Smad signaling synergistically gain EMT in lung cancer [26]. Therefore, TGF-β-induced EMT as a model promotes metastasis of tumor cells through activating many pathways. We also included an established model of transient TGF-β1-induced EMT in human ovarian cancer cells. Furthermore, TGF-β1-induced EMT in ovarian cancer cells was inhibited by treatment with 1α,25(OH)2D3. During recent years, vitamin D has been increasingly concerned as a potential for anti-cancer therapy, especially for the role of vitamin D in reducing risk and progression of colon, breast and prostate cancer [8,27,28]. However, there are fewer studies on the effect of vitamin D on proliferation and invasion of ovarian cancer. It is reported that Solar UVB irradiance, which resulted in higher level of 25-dihydroxyvitamin D [25(OH)D] in serum, a widely accepted biomarker of vitamin D status, was inversely associated with incidence rates of ovarian cancer in 175 countries in 2002. The high concentrations of the vitamin D receptor were demonstrated in ovarian cancer cells [27], and 1α,25(OH)2D3 has been shown to inhibit cell proliferation and induce apoptosis in ovarian cancer cell lines [11,12,27]. However, there is little convincing evidence for an association between 25(OH)D and the risk of developing ovarian cancer in a pooled analysis [28] and a meta-analysis [29]. Furthermore, the mechanism on antitumor of vitamin D on ovarian cancer is unclearly understood. Some experimental studies strongly suggest that 1α,25(OH)2D3 arrested ovarian cancer cells in G1 and G2/M phase by modulating GADD45 and p27 [6,30]. Our previous study indicated that 1α,25(OH)2D3 enhances the therapeutic effects of carboplatin by altering the cell cycle and increasing apoptosis through changes in reactive oxygen species and mitochondria membrane potential in SKOV-3 cells [12]. These findings demonstrated that vitamin D inhibited proliferation of ovarian cancer cells. The findings firstly presented in this study indicated that 1α,25(OH)2D3 decreased cell migration through inhibiting TGF-β1-induced EMT in human ovarian cancer SKOV-3 cells. It was in colon cancer cells that 1α,25(OH)2D3 inhibited TGF-β1/β2-increased invasion and migration by inhibiting the switch of cadherin and expression of EMT-related transcription factors. 1α,25(OH)2D3 also inhibited the secretion of MMP-2 and MMP-9 and increased expression of F-actin induced by TGF-β1/β2 in colon cancer cells [15]. 1α,25(OH)2D3 and its analogs inhibit the migration and invasion of tumor cells by regulating changes in the cell–extracellular matrix interaction as well as by promoting cell–cell contact in breast, prostate and colorectal cancer cells [31,32]. Taken together, the restraint of EMT might be one of the mechanisms underlying the anti-metastasis effect of 1α,25(OH)2D3 in cancer cells. Transcriptional factor β-catenin aggregating toward the nucleus is also considered as a sign of EMT [33]. In addition, E-cadherin could mediate the migration of β-catenin from nucleus to cytoplasm [34]. In this study, both β-catenin gathered to the nucleus and E-cadherin decreased in the cytoplasm were observed in SKOV-3 cells induced by TGF-β1. At the same time, the expression of Slug and Snail were increased in nuclei of ovarian cancer cells. Moreover, 1α,25(OH)2D3 significantly reversed the expression of EMT-related transcription factors induced by TGF-β1. Chen et al. also reported that increased TGF-β1/β2 expression of EMT-related transcription factors in colon cancer cells was also inhibited by 1α,25(OH)2D3 [15]. In addition, Snail repressed expression of VDR, resulting in reduction of the anticancer effects of 1α,25(OH)2D3 [35]. In another study for colon cancer, the author reported that expression of VDR was negatively correlated with those of snail and ZEB1 in the cancer tissue [36]. We found that 1α,25(OH)2D3 could change the localization of Slug, Snail, and β-cateinin and inhibited their expression in SKOV-3 cells exposed to TGF-β1, which was associated with the increase of VDR. The data suggested that vitamin D treatment strategies play their protective roles in ovarian tumor progression, by increasing VDR expression. 4. Experimental Section 4.1. Cell Culture and Treatment Human ovarian epithelial adenocarcinoma cell lines SKOV-3 were obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China), and maintained in Roswell Park Memorial Institute (RPMI) 1640 (Invitrogen Carlsbad, San Diego, CA, USA) with 10% fetal bovine serum (FBS, Sigma-Aldrich Chemie GmbHFBS, Steinheim, Germany), 100 U/mL penicillin, and 100 µg/mL streptomycin (Beyotime Biotechnology, Shanghai, China) in a humidified atmosphere of 5% CO2 at 37 °C. The cells were cultured at approximately 80% confluency and starved in serum-free RPMI 1640 overnight. After being removed from culture medium, SKOV-3 cells were treated with different factors, respectively. The cells in control group were treated with vehicle (0.1% ethanol), TGF-β1 group (10 ng/mL, PEPROTECH, Princeton, NJ, USA), VD group treated with different concentrations of 1α,25(OH)2D3 (1, 10, 100 nmol/L) , which was purchased from Sigma (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) and the cells in TGF-β1 + VD group were adminstrated with combination TGF-β1 (10 ng/mL) and 1α,25(OH)2D3 (100 nmol/L). 4.2. Wound Healing Assay The migration capacities of SKOV-3 cells were assessed by wound healing assay. Cells were plated, and serum-starved for 24 h after the cells adherent on culture dish about 12 h. A wound was created by scraping the cells with a sterile 1000 μL pipette tip in the middle of the culture well. Then, the dish was softly washed in phosphate-buffered saline (PBS) and put in a culture medium with different factors. The wound closure photographs were captured using a microscope (magnification of ×40, CKX41F, Olympus, Tokyo, Japan) equipped with a digital camera. A measure was taken of it, and then the average value was calculated. Data are presented as migration index (%) = [(the initialized width of the scratch) − (the final width of the scratch)]/(the initialized width of the scratch). Data points in figures represent three independent experiments. 4.3. Live Cell Imaging System SKOV-3 cells during logarithmic phase were planted in 24-well plates (Thermo Fisher Scientific, Waltham, MA, USA) for 12 h, and removed from culture medium. Then, cells were starved in serum-free RPMI 1640 for 24 h and treated with vehicle, 100 nmol/L of 1α,25(OH)2D3, 10 ng/mL of TGF-β1, or combination TGF-β1 with 1α,25(OH)2D3, respectively. SKOV-3 cells were incubated in the Live Cell Imaging System (Cell^R, Olympus, Tokyo, Japan) to monitor their growth in real time for 72 h. 4.4. Invasion Assay SKOV-3 cells were seeded in the the 24-well BD Biocoat Matrigel Invasion Chambers (BD Biosciences, Franklin Lakes, NJ, USA) with a vehicle, indicated concentration of TGF-β1 or 1α,25(OH)2D3 or combination of them, respectivedly. The lower chamber was supplemented with RPMI1640 culture containing 10% FBS. After the cells were cultured for 24 h, non-invading cells were carefully removed with a cotton swab. The cells on the bottom of inserts were fixed with 70% ethanol and were stained with Crystal Violet (c0121, Beyotime Biotechnology, Shanghai, China) for 3 min. The number of cells penetrating the membrane were calculated and pictures were taken under the microscope (×40, CKX41F, Olympus, Tokyo, Japan). Data points in figures represent three independent experiments. 4.5. Immunofluoresecence Staining SKOV-3 cells were grown on Cover slides (Thermo Fisher Scientific, Waltham, MA, USA), which were put into 24-well plates. After the cells adhered on the plate, cells were treated with a vehicle, indicating concentration of TGF-β1 or 1α,25(OH)2D3, or combination of them, respectively. After 24 h, cells were washed with cool PBS twice, and fixed in 4% paraformaldehyde for 20 min at 4 °C. Then, cells were permeabilized with 0.1% Triton at 4 °C for 15 min, and nonspecific binding was blocked with 1% FBS in confining liquid for 1 h at room temperature (RT). Next, the cells were incubated with primary antibodies: E-cadherin (32A8) Mouse mAb (1:100, #5296, Cell Signaling Technology, Irvine, CA, USA), Vimentin (D21H3) XP® Rabbit mAb (1:100, #5741, Cell Signaling Technology), β-catenin (6B3) Rabbit mAb (1:100, #9582, Cell Signaling Technology), Slug (C19G7) Rabbit mAb (1:100, #9585, Cell Signaling Technology), Snail (C15D3) Rabbit mAb (1:50, #3879, Cell Signaling Technology), mAb Anti-Vitamin D Receptor Antibody (1:100, NBP1-51322, Novus Biologicals, Littleton, CO, USA). After being incubated with primary antibodies at 4 °C overnight, the cells were stained with secondary antibody IGg-Cy5 (1:1000, #4412, Cell Signaling Technology) in the dark room for 1.5 h and washed with PBS for 3 min (3 times) in a horizontal earthquake shaking bed. Nuclei were labeled with DAPI for 20 min. Images were captured with Confocal Laser Scanning Microscope (TCS SP2, Leica, Wetxlar, Germany). 4.6. Western Blot After being treated with a vehicle, indicating concentration of TGF-β1 or 1α,25(OH)2D3, or a combination of them, respectively, SKOV-3 cells were harvested and lysed for total cellular protein extraction with RIPA buffer (p0013, Beyotime Biotechnology, Shanghai, China). The cells were centrifuged at 12,000 rpm for 30 min and the supernatants were collected. The protein concentration of lysate was quantified using a Bicinchonininc acid (BCA) protein assay kit (P0012, Beyotime Biotechnology, Shanghai, China). Equal amounts of total proteins (30 µg) were loaded onto 10% sodium dodecyl sulphate—polyacrylamide gel (SDS-PAGE) (P0012A, Beyotime Biotechnology) and the proteins were electrophoretically transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore, Boston, MA, USA). After being blocked with 5% skimmed milk, the membranes were incubated with primary antibodies of E-cadherin (1:700), Vimentin (1:800), Snail (1:700), Slug (diluted 1:700), β-catenin (1:100) or VDR (1:100) at 4 °C for overnight, respectively. Then, cells were washed 3 times by Phosphate Buffered Saline with Tween-20 (PBST) and incubated with anti-mouse or anti-rabbit horseradish peroxidase-conjugated secondary antibodies at RT for 1 h, and then washed 3 times by PBST again. The detection of the antigen–antibody complex was visualized using chemiluminescence (Immobilon ECL) reagent (Millipore). The indicated protein was quantified with gray value to identify the respective expression of targeted protein relative to β-actin (as the loading control). Data points in figures represent three independent experiments. 4.7. Statistical Analysis Statistical analysis was performed using SPSS, version 17.0 for Windows (SPSS, Inc., Chicago, IL, USA). The data of the experiments were presented as means and standard deviation and analyzed with Student’s t-test and ANOVA. 5. Conclusions 1α,25(OH)2D3 not only inhibited the invasion and the migration of SKOV-3 cells, but also promoted the acquisition of an epithelial phenotype of SKOV-3 cells treated with TGF-β1. We discovered that 1α,25(OH)2D3 increased the expression of epithelial marker E-cadherin while decreasing the level of mesenchymal marker, Vimentin, which was associated with the elevated expression of VDR. Moreover, 1α,25(OH)2D3 reduced the expression level of EMT-related transcription factors, such as slug, snail and β-catenin. These results indicate that 1α,25(OH)2D3 suppresses the metastasis of ovarian cancer cells by regulating EMT, implying that 1α,25(OH)2D3 might be a potential therapeutic agent for the treatment of ovarian cancer. Acknowledgments This study was supported by the National Natural Scientific Funding of China (grant No. 81072286, 81372979 and 11335011), in part, by the Collaborative Innovation Center of Radiation Medicine, Jiangsu Higher Education Institutions, and, Outstanding youth project of fuzhou general hospital (grant No. 2015Q03). Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1285/s1. Click here for additional data file. Author Contributions Yong-Feng Hou, Si-Hai Gao, Ping Wang, Zeng-Li Zhang and Bing-Yan Li conceived and designed the experiments; Yong-Feng Hou, Si-Hai Gao, Ping Wang, He-Mei Zhang, Li-Zhi Liu and Meng-Xuan Ye performed the experiments; Yong-Feng Hou, Si-Hai Gao, Ping Wang analyzed the data; Guang-Ming Zhou participated in writing the paper. Conflicts of Interest The authors declare no conflict of interest. Abbreviations EMT Epithelial-mesenchymal transition 1α,25(OH)2D3 1α,25-dihydroxyvitamin D VDR vitamin D receptor 25(OH)D 25-dihydroxyvitamin D Figure 1 1α,25(OH)2D3 inhibited the migration of human ovarian adenocarcinoma cell line SKOV-3 cells. (A) Left: representative pictures of the wound area obtained 24, 48 and 72 h after scratching. 100× magnification; Right: migration index (%) = [(the initialized width of the scratch) − (the final width of the scratch)]/(the initialized width of the scratch); (B) representative pictures of E-cadherin and Vimentin were captured by confocal laser scanner microscopy (CLSM) 24 h after being treated with 1α,25(OH)2D3. Nuclear DNA was visualized by 4′,6-diamidino-2-phenylindole (DAPI) staining. 200× magnification; (C) Left: Western blot analysis of the indicated proteins in SKOV-3 cells. β-actin served as a loading control; Right: the level of the indicated protein was quantified with gray value. The data represent the Mean ± SD. * p < 0.05 versus control. Figure 2 Transforming growth factor-β1 (TGF-β1) induces EMT of SKOV-3 cells. (A) SKOV-3 cells were exposed to 10 ng/mL of TGF-β1. Compared to the group of control, TGF-β1-treated SKOV-3 cells lost their cobblestone shape and adopted a fibroblast-like, spindle-shaped morphology. Morphology photographs were taken at 24, 48, and 72 h (magnification of 400×); (B) analyses with a Live Cell Imaging System showed that the movement distance increased further after SKOV-3 cells were treated with TGF-β1 for 36 h than control cells (Arrows refer to the movement track of SKOV-3 cells); (C) Left: Western blot analysis of the indicated proteins in SKOV-3 cells. β-actin served as a loading control; Right: the level of the indicated protein was quantified with gray value. The data represent the Mean ± SD. * p < 0.05 versus control. Figure 3 1α,25(OH)2D3 inhibited the migration and invasion of SKOV-3 cells during TGF-β1-induced EMT. (A) Left: representative pictures of the wound area obtained 24 h after scratching. 100× magnification; Right: migration index (%) = [(the initialized width of the scratch) − (the final width of the scratch)]/(the initialized width of the scratch); (B) Left: invasion assay was carried out using the 24-well Becton Dickinson (BD) Biocoat Matrigel Invasion Chambers(magnification of 400×); Right: the cells on the bottom of inserts were counted under microscope. # p < 0.05 versus negative control, * p < 0.05 versus TGF-β1, and ^ p < 0.05, 1α,25(OH)2D3 (VD) + TGF-β group versus VD group. Figure 4 1α,25(OH)2D3 regulated the expression of EMT-related markers in SKOV-3 cells exposed to TGF-β1. (A–C) representative pictures of indicated proteins E-cadherin, Vimentin, Snail, Slug, β-catenin and VDR were captured by confocal laser scanner microscopy (CLSM). Nuclear DNA was visualized by DAPI staining. 200× magnification; (D–F) Left: Western blot analysis of the indicated proteins in SKOV-3 cells. β-actin or GAPDH served as a loading control; Right: the level of the indicated protein was quantified with gray value. The data represent the Mean ± SD. # p < 0.05 versus negative control, * p < 0.05 versus TGF-β1-treated group, and ^ p < 0.05, VD + TGF-β group versus VD group. ==== Refs References 1. Siegel R. Ward E. Brawley O. Jemal A. Cancer statistics, 2011: The impact of eliminating socioeconomic and racial disparities on premature cancer deaths CA Cancer J. Clin. 2011 61 212 236 10.3322/caac.20121 21685461 2. Lamouille S. Xu J. Derynck R. Molecular mechanisms of epithelial-mesenchymal transition Nat. Rev. Mol. Cell Biol. 2014 15 178 196 10.1038/nrm3758 24556840 3. Ahmed N. Thompson E.W. Quinn M.A. Epithelial-mesenchymal interconversions in normal ovarian surface epithelium and ovarian carcinomas: An exception to the norm J. Cell. Physiol. 2007 213 581 588 10.1002/jcp.21240 17708542 4. Yan H. Sun Y. Evaluation of the mechanism of epithelial-mesenchymal transition in human ovarian cancer stem cells transfected with a WW domain-containing oxidoreductase gene Oncol. Lett. 2014 8 426 430 10.3892/ol.2014.2063 24959289 5. Abell A.N. Johnson G.L. Implications of mesenchymal cells in cancer stem cell populations: Relevance to emt Curr. Pathobiol. Rep. 2014 2 21 26 10.1007/s40139-013-0034-7 25530923 6. Li P. Li C. Zhao X. Zhang X. Nicosia S.V. Bai W. P27(Kip1) stabilization and G(1) arrest by 1,25-dihydroxyvitamin D(3) in ovarian cancer cells mediated through down-regulation of cyclin e/cyclin-dependent kinase 2 and Skp1-Cullin-F-box protein/Skp2 ubiquitin ligase J. Biol. Chem. 2004 279 25260 25267 10.1074/jbc.M311052200 15075339 7. Deeb K.K. Trump D.L. Johnson C.S. Vitamin D signalling pathways in cancer: Potential for anticancer therapeutics Nat. Rev. Cancer 2007 7 684 700 10.1038/nrc2196 17721433 8. Feldman D. Krishnan A.V. Swami S. Giovannucci E. Feldman B.J. The role of vitamin D in reducing cancer risk and progression Nat. Rev. Cancer 2014 14 342 357 10.1038/nrc3691 24705652 9. Ahonen M.H. Zhuang Y.H. Aine R. Ylikomi T. Tuohimaa P. Androgen receptor and vitamin D receptor in human ovarian cancer: Growth stimulation and inhibition by ligands Int. J. Cancer 2000 86 40 46 10.1002/(SICI)1097-0215(20000401)86:1<40::AID-IJC6>3.0.CO;2-E 10728592 10. Saunders D.E. Christensen C. Wappler N.L. Schultz J.F. Lawrence W.D. Malviya V.K. Malone J.M. Deppe G. Inhibition of c-myc in breast and ovarian carcinoma cells by 1,25-dihydroxyvitamin D3, retinoic acid and dexamethasone Anticancer Drugs 1993 4 201 208 10.1097/00001813-199304000-00012 8490200 11. Jiang F. Bao J. Li P. Nicosia S.V. Bai W. Induction of ovarian cancer cell apoptosis by 1,25-dihydroxyvitamin D3 through the down-regulation of telomerase J. Biol. Chem. 2004 279 53213 53221 10.1074/jbc.M410395200 15485861 12. Zhang Z. Zhang H. Hu Z. Wang P. Wan J. Li B. Synergy of 1,25-dihydroxyvitamin D3 and carboplatin in growth suppression of SKOV-3 cells Oncol. Lett. 2014 8 1348 1354 10.3892/ol.2014.2307 25120722 13. Lange T.S. Stuckey A.R. Robison K. Kim K.K. Singh R.K. Raker C.A. Brard L. Effect of a vitamin D(3) derivative (B3CD) with postulated anti-cancer activity in an ovarian cancer animal model Investig. New Drugs 2010 28 543 553 10.1007/s10637-009-9284-y 19582372 14. Kawar N. Maclaughlan S. Horan T.C. Uzun A. Lange T.S. Kim K.K. Hopson R. Singh A.P. Sidhu P.S. Glass K.A. Pt19c, another nonhypercalcemic vitamin D2 derivative, demonstrates antitumor efficacy in epithelial ovarian and endometrial cancer models Genes Cancer 2013 4 524 534 10.1177/1947601913507575 24386512 15. Chen S. Zhu J. Zuo S. Ma J. Zhang J. Chen G. Wang X. Pan Y. Liu Y. Wang P. 1,25(OH)2D3 attenuates Tgf-beta1/beta2-induced increased migration and invasion via inhibiting epithelial-mesenchymal transition in colon cancer cells Biochem. Biophys. Res. Commun. 2015 468 130 135 10.1016/j.bbrc.2015.10.146 26523511 16. Davidson B. Trope C.G. Reich R. Epithelial-mesenchymal transition in ovarian carcinoma Front. Oncol. 2012 2 33 10.3389/fonc.2012.00033 22655269 17. Chiang K.C. Chen S.C. Yeh C.N. Pang J.H. Shen S.C. Hsu J.T. Liu Y.Y. Chen L.W. Kuo S.F. Takano M. MART-10, a less calcemic vitamin D analog, is more potent than 1alpha,25-dihydroxyvitamin D3 in inhibiting the metastatic potential of MCF-7 breast cancer cells in vitro J. Steroid Biochem. Mol. Biol. 2014 139 54 60 10.1016/j.jsbmb.2013.10.005 24125734 18. Chiang K.C. Yeh C.N. Hsu J.T. Jan Y.Y. Chen L.W. Kuo S.F. Takano M. Kittaka A. Chen T.C. Chen W.T. The vitamin D analog, MART-10, represses metastasis potential via downregulation of epithelial-mesenchymal transition in pancreatic cancer cells Cancer Lett. 2014 354 235 244 10.1016/j.canlet.2014.08.019 25149065 19. Upadhyay S.K. Verone A. Shoemaker S. Qin M. Liu S. Campbell M. Hershberger P.A. 1,25-dihydroxyvitamin D3 (1,25(OH)2 D3 ) signaling capacity and the epithelial-mesenchymal transition in non-small cell lung cancer (NSCLC): Implications for use of 1,25(OH)2 D3 in NSCLC treatment Cancers 2013 5 1504 1521 10.3390/cancers5041504 24217116 20. Wong A.S. Auersperg N. Normal ovarian surface epithelium Cancer Treat. Res. 2002 107 161 183 11775449 21. Sundfeldt K. Piontkewitz Y. Ivarsson K. Nilsson O. Hellberg P. Brannstrom M. Janson P.O. Enerback S. Hedin L. E-cadherin expression in human epithelial ovarian cancer and normal ovary Int. J. Cancer 1997 74 275 280 10.1002/(SICI)1097-0215(19970620)74:3<275::AID-IJC7>3.0.CO;2-W 9221804 22. Zavadil J. Bottinger E.P. Tgf-beta and epithelial-to-mesenchymal transitions Oncogene 2005 24 5764 5774 10.1038/sj.onc.1208927 16123809 23. Pang M.F. Georgoudaki A.M. Lambut L. Johansson J. Tabor V. Hagikura K. Jin Y. Jansson M. Alexander J.S. Nelson C.M. TGF-beta1-induced EMT promotes targeted migration of breast cancer cells through the lymphatic system by the activation of CCR7/CCL21-mediated chemotaxis Oncogene 2016 35 748 760 10.1038/onc.2015.133 25961925 24. Vincent T. Neve E.P. Johnson J.R. Kukalev A. Rojo F. Albanell J. Pietras K. Virtanen I. Philipson L. Leopold P.L. A SNAIL1-SMAD3/4 transcriptional repressor complex promotes tgf-beta mediated epithelial-mesenchymal transition Nat. Cell Biol. 2009 11 943 950 10.1038/ncb1905 19597490 25. Goswami M.T. Reka A.K. Kurapati H. Kaza V. Chen J. Standiford T.J. Keshamouni V.G. Regulation of complement-dependent cytotoxicity by TGF-beta-induced epithelial-mesenchymal transition Oncogene 2016 35 1888 1898 10.1038/onc.2015.258 26148233 26. Liu R.Y. Zeng Y. Lei Z. Wang L. Yang H. Liu Z. Zhao J. Zhang H.T. JAK/STAT3 signaling is required for TGF-beta-induced epithelial-mesenchymal transition in lung cancer cells Int. J. Oncol. 2014 44 1643 1651 24573038 27. Leyssens C. Verlinden L. Verstuyf A. Antineoplastic effects of 1,25(OH)2D3 and its analogs in breast, prostate and colorectal cancer Endocr. Relat. Cancer 2013 20 R31 R47 10.1530/ERC-12-0381 23319494 28. Villena-Heinsen C. Meyberg R. Axt-Fliedner R. Reitnauer K. Reichrath J. Friedrich M. Immunohistochemical analysis of 1,25-dihydroxyvitamin-D3-receptors, estrogen and progesterone receptors and Ki-67 in ovarian carcinoma Anticancer Res. 2002 22 2261 2267 12174912 29. Zheng W. Danforth K.N. Tworoger S.S. Goodman M.T. Arslan A.A. Patel A.V. McCullough M.L. Weinstein S.J. Kolonel L.N. Purdue M.P. Circulating 25-hydroxyvitamin D and risk of epithelial ovarian cancer: Cohort consortium vitamin D pooling project of rarer cancers Am. J. Epidemiol. 2010 172 70 80 10.1093/aje/kwq118 20562186 30. Yin L. Grandi N. Raum E. Haug U. Arndt V. Brenner H. Meta-analysis: Circulating vitamin D and ovarian cancer risk Gynecol. Oncol. 2011 121 369 375 10.1016/j.ygyno.2011.01.023 21324518 31. Jiang F. Li P. Fornace A.J. Jr. Nicosia S.V. Bai W. G2/M arrest by 1,25-dihydroxyvitamin D3 in ovarian cancer cells mediated through the induction of GADD45 via an exonic enhancer J. Biol. Chem. 2003 278 48030 48040 10.1074/jbc.M308430200 14506229 32. Chiang K.C. Yeh T.S. Chen S.C. Pang J.H. Yeh C.N. Hsu J.T. Chen L.W. Kuo S.F. Takano M. Kittaka A. The vitamin D analog, MART-10, attenuates triple negative breast cancer cells metastatic potential Int. J. Mol. Sci. 2016 17 606 10.3390/ijms17040606 27110769 33. Lee J.M. Dedhar S. Kalluri R. Thompson E.W. The epithelial-mesenchymal transition: New insights in signaling, development, and disease J. Cell Biol. 2006 172 973 981 10.1083/jcb.200601018 16567498 34. Egan J.B. Thompson P.A. Vitanov M.V. Bartik L. Jacobs E.T. Haussler M.R. Gerner E.W. Jurutka P.W. Vitamin D receptor ligands, adenomatous polyposis coli, and the vitamin D receptor foki polymorphism collectively modulate beta-catenin activity in colon cancer cells Mol. Carcinog. 2010 49 337 352 20043299 35. Larriba M.J. Munoz A. Snail versus vitamin D receptor expression in colon cancer: Therapeutics implications Br. J. Cancer 2005 92 985 989 10.1038/sj.bjc.6602484 15770204 36. Pena C. Garcia J.M. Garcia V. Silva J. Dominguez G. Rodriguez R. Maximiano C. Garcia de Herreros A. Munoz A. Bonilla F. The expression levels of the transcriptional regulators p300 and CtBP modulate the correlations between SNAIL, ZEB1, E-cadherin and vitamin D receptor in human colon carcinomas Int. J. Cancer 2006 119 2098 2104 10.1002/ijc.22083 16804902

PMC005xxxxxx/PMC5000683.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081286ijms-17-01286ReviewPhylogenetic-Derived Insights into the Evolution of Sialylation in Eukaryotes: Comprehensive Analysis of Vertebrate β-Galactoside α2,3/6-Sialyltransferases (ST3Gal and ST6Gal) Teppa Roxana E. 1Petit Daniel 2Plechakova Olga 3Cogez Virginie 4Harduin-Lepers Anne 45*Kim Cheorl-Ho Academic Editor1 Bioinformatics Unit, Fundación Instituto Leloir, Av. Patricias Argentinas 435, C1405BWE Buenos Aires, Argentina; [email protected] Laboratoire de Génétique Moléculaire Animale, UMR 1061 INRA, Université de Limoges Faculté des Sciences et Techniques, 123 avenue Albert Thomas, 87060 Limoges, France; [email protected] FRABio-FR3688 CNRS, Univ. Lille, bât. C9, 59655 Villeneuve d’Ascq cedex, France; [email protected] Univ. Lille, CNRS, UMR 8576-UGSF-Unité de Glycobiologie Structurale et Fonctionnelle, 59000 Lille, France; [email protected] UGSF, Bât. C9, Université de Lille-Sciences et Technologies, 59655 Villeneuve d’Ascq, France* Correspondence: [email protected]; Tel.: +33-320-336-246; Fax: +33-320-436-55509 8 2016 8 2016 17 8 128628 6 2016 28 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Cell surface of eukaryotic cells is covered with a wide variety of sialylated molecules involved in diverse biological processes and taking part in cell–cell interactions. Although the physiological relevance of these sialylated glycoconjugates in vertebrates begins to be deciphered, the origin and evolution of the genetic machinery implicated in their biosynthetic pathway are poorly understood. Among the variety of actors involved in the sialylation machinery, sialyltransferases are key enzymes for the biosynthesis of sialylated molecules. This review focus on β-galactoside α2,3/6-sialyltransferases belonging to the ST3Gal and ST6Gal families. We propose here an outline of the evolutionary history of these two major ST families. Comparative genomics, molecular phylogeny and structural bioinformatics provided insights into the functional innovations in sialic acid metabolism and enabled to explore how ST-gene function evolved in vertebrates. evolutionsialyltransferasessialic acidmolecular phylogenyfunctional genomics ==== Body 1. Introduction Sialic acids (SA) represent a broad family of nine-carbon electro-negatively charged monosaccharides commonly described in the deuterostomes and some microorganisms [1,2,3,4,5]. Interestingly, SA show a discontinuous distribution across evolutionary metazoan lineages. Outside the deuterostome lineage (vertebrates, urochordates, echinoderms), SA are rarely described in some ecdysozoa and lophotrochozoa protostomes like in the Drosophila melanogaster nervous system during embryogenesis [6,7,8,9] or in larvae of the cicada Philaenus spumarius [10], or on glycolipids of the common squid and pacific octopus [11]. They are notably absent from plants, archaebacteria or the ecdysozoan Caenorhabditis elegans [12]. SA exhibit a huge structural diversity and species specific modifications. This family of compounds encompasses N-acetylneuraminic acid (Neu5Ac) and over 50 derivatives showing various substituents on carbon 4, 5, 7, 8 or 9, like Neu5Gc and Kdn, with Neu5Ac being the most prominent SA found in higher vertebrates (Figure 1A). In vertebrates, the SA hydroxyl group at position 2 is most frequently glycosidically-linked to either the 3- or 6-hydroxyl group of galactose (Gal) residues (Figure 1B) or the 6-hydroxyl group of N-acetylgalactosamine (GalNAc) residues and can form to a lesser extent di-, oligo- or poly-SA chains via their 8-hydroxyl group. In deuterostome lineages, sialoglycans are found in cellular secretions and on the outer cell surface, essentially as terminal residues of the glycan chains of glycoproteins and glycolipids [13,14] constituting the so-called siaLome [15], which varies according to animal species. Owing to their anionic charge and their peripheral position in glycans, SA play major roles in the various vertebrate biological systems ranging from protecting proteins from proteolysis, modulating cell functions to regulating intracellular communication [16]. For instance, the α2,3-linked SA contribute to the high viscosity of the mucin-type O-glycosylproteins found on the intestine endothelia or on the surface of fish or frog eggs [17]. Besides, some endogenous proteins specifically recognize sialylated molecules at the cell surface that act as receptors. Examples include selectin on endothelial cells mediating leucocytes and platelets trafficking, and siglecs playing a role in immune cell regulation [18,19]. Likewise, a number of pathogenic agents like toxins (cholera toxin), protozoa (Plasmodium), viruses (influenza virus), bacteria (Helicobacter pylori) use cell surface SA as ligands for cell adhesion [20] and have evolved this ability to distinguish a specific sialylated sugar code [21,22] distinguishing α2,3- or α2,6-linked SA in vertebrate tissues [23]. One of the most notable examples is the flu virus tropism: human strains of influenza A virus bind selectively to SAα2,6-Gal epitopes that prevail in the human tracheal mucosal epithelium, whereas chimpanzee strains bind selectively to SAα2,3-Gal epitopes primarily expressed in their tracheal mucosal epithelium [24,25,26], suggesting that the switch to α2,6-linked SA could give the human ancestor some resistance towards influenza viruses, which later on could have evolved and adapted to the modern humans. The SA metabolism is complex and requires a large panel of enzymes with various subcellular localization including the nuclear CMP-Neu5Ac synthase (CMAS), the cytosolic UDP-GlcNAc 2-epimerase/N-acetylmannosamine kinase (GNE), the cytosolic cytidine monophosphate-N-acetylneuraminic acid hydroxylase (CMAH), the Golgi CMP-Neu5Ac transporter (SLC35A1), the Golgi sialyltransferases (ST) and sialidases (Neu) (Figure 2A) [27]. The distribution of SA in the metazoans further suggests that this sialylation machinery has evolved at least in the last common ancestor (LCA) of the metazoans, well before the divergence of protostomes (Ecdysozoa and Lophotrochozoa) and deuterostomes. Very little is known pertaining to the evolutionary history of each orthologous gene. However, these genes show also an unusual and patchy phylogenetic distribution with a huge gene families’ expansion observed in the deuterostome lineages indicative of the prominent role of sialoglycoconjugates in the deuterostome ancestor [28,29,30,31] and selective loss in most non-deuterostome lineages as well as in some vertebrate lineages (e.g., CMAH gene). The humans cannot synthesize CMP-Neu5Gc from CMP-Neu5Ac because the human CMAH gene was inactivated 2 million years ago [32,33], an activity that was independently lost in the ferrets [34], birds and reptiles [35] (Figure 2B). Interestingly, a cmas gene was identified and characterized in the D. melanogaster genome [36,37] and moreover, 1 gne, 2 st and 2 neu genes were identified in the porifera Oscarella carmella [28,29,38,39], and a SLC35A1-related gene was identified in the tunicate Ciona intestinalis and C. elegans genomes (personal data) suggesting the ancient occurrence and subsequent divergent evolution of the sialylation machinery. The structural diversity of sialylated glycoconjugates is further ensured by a diverse set of STs consisting of 20 members described in the human tissues [40,41]. The STs reside and are strictly organized in the trans-Golgi network of eukaryotic cells as type II transmembrane proteins with a similar topology showing a short N-terminal cytoplasmic tail, a single transmembrane domain, a stem domain and a large C-terminal catalytic domain oriented in the Golgi lumen [42]. The STs use CMP-β-Neu5Ac, CMP-β-Neu5Gc or CMP-β-Kdn as activated sugar donors for the sialylation at terminal positions of oligosaccharide chains of glycoconjugates. These STs are categorized into 4 families (ST6Gal, ST3Gal, ST6GalNAc and ST8Sia) [41,43] found in the GT-29 of the Carbohydrate-Active enZYme (CAZy) database [44] and named according to the glycosidic linkage formed and the monosaccharide acceptor [45]. Each family catalyzes the formation of different glycosidic linkages, α2–3, or α2–6 to the terminal Gal residue in N- or O-glycans, α2–6 to the terminal GalNAc residue in O-glycans and glycolipids and α2–8 to terminal SA residues in N- or O-glycans or glycolipids). The ST enzymatic activities have been documented mainly in mouse and human tissues and more recently in chicken [46,47], and to a lesser extent in the invertebrates like the fly D. melanogaster [48], the silkworm Bombyx mori [49], the amphioxus Branchiostoma floridae [1] and the tunicate C. intestinalis [5]. Each member of the mammalian ST3Gal and ST6Gal families shows exquisite acceptor specificities (for reviews, see [41,50,51]. However, as most of the STs have not been experimentally characterized, it remains unclear how these diverse biochemical functions evolved and what were the biological consequences of the functional diversification of STs. In the post-genomic era, a major biological question remains to elucidate the multi-level protein function of STs (i.e., biochemical, cellular or developmental functions) which can be achieved through the simultaneous study of different levels of biological organization and the use of computational means. The most represented vertebrate β-galactoside α2,3/6-sialyltransferases (ST3Gal and ST6Gal) offer the unique opportunity to understand deuterostome innovations and the STs functional evolution. The ST3Gal and ST6Gal are well studied enzymes catalyzing the transfer of sialic acid residues to the terminal galactose residues of either the type-I, type-II or type-III disaccharides (Galβ1,3GlcNAc; Galβ1,4GlcNAc or Galβ1,3GalNAc, respectively) resulting in the formation of α2–3 or α2–6 glycosidic linkages on terminal galactose (Gal) residues. In previous reports, we deciphered key genetic events, which led to the various ST3Gal and ST6Gal subfamilies described in the vertebrates, we established the evolutionary relationships of newly described STs and provided insights into the structure-function relationships of STs [39,52] and into their various biological functions [38,53]. Focusing on β-galactoside α2,3/6-sialyltransferases (ST3Gal and ST6Gal), we explore in this review the molecular evolution of β-galactoside α2,3/6 sialyltransferases with the goal of bringing an evolutionary perspective to the study of SA-based interactions and contributing a powerful approach for a better understanding of sialophenotype in vertebrates. 2. Genome-Wide Search of STs Genes Decline or Expansion? A general strategy using conventional BLAST search approaches [54] was adopted for homologous ST sequences identification in the transcriptomic and genomic databases like NCBI or ENSEMBL to reconstruct the animal ST genes repertoire and assign orthologies [39]. Although the vertebrate ST amino acid sequences show very limited overall sequence identity (around 20%), conserved peptide motifs have been described within their catalytic domain, which are very useful hallmark for ST identification. Different sets of protein regions considering three levels of amino acid sequence conservation have been described in the past that are retrieved from multiple sequence alignments (MSA) of (1) all animal ST called sialylmotifs L (large), S (small), III and VS (very small); (2) each family of ST, called family motifs a, b, c, d and e; (3) in each vertebrate subfamily [55]. This strategy led to the identification of a total number of 750 st3gal- and st6gal-related sequences in the genome of 127 metazoan species that represent a significant sampling of metazoan diversity illustrated in Figure 2B. The st6gal and st3gal gene families show a broad phylogenetic distribution in Metazoan from sponges to mammals. The mRNA fragments identified from the Homosclerophore sponge O. carmella in the Porifera phylum suggested that ancestral st6gal1/2 and st3gal1/2/8 genes were already present in the earliest metazoans [38,53] and could represent the most ancient ST described in animals. This observation pointed also the early divergence of st3gal groups GR1 (st3gal1/2/8), GR2 (st3gal4/6/9), GR3 (st3gal3/3-r/5/7) and GRx (st3gal4/6/9/3/3-r/5/7) that far predates the divergence of protostomes and deuterostomes [53]. Interestingly, ST-related sequences possessing a conserved GT-29 protein domain Pfam00777 with sialylmotifs L, S, III and VS and no family motif could be identified in plants, in the green marine microalga Bathycoccus prasinos [56], in the haptophyte Emiliana huxleyi (XM_005778044) [57], in the cryptophyte alga Guillarda theta and in the red tide dinoflagelate Alexandrium minutum [29] suggesting the presence of an ancestral protist ST gene set. However, the evolutionary relationships of these more distantly related ST sequences are not yet clearly established and the origin of ST-related sequences in Metazoan remains enigmatic [13,31,43]. Since no ST-related sequence was identified in Choanoflagellates, the closest known relatives of metazoans [58], nor in fungi, it can be deduced the ancient origin of ST sequences and their subsequent disappearance in some metazoan branches like in the Nematoda C. elegans for both ST families, in protostome for the st3gal family, and echinoderms and tunicates for the st6gal family [38,53]. A large data set of β-galactoside α2,3/6-sialyltransferase related sequences was identified in vertebrate genomes and orthologs of the 8 known mammalian β-galactoside α2,3/6-sialyltransferase genes could be identified in fish and amphibian genomes with the notable exception of the st3gal6 gene that disappeared from fish genome (Table 1). An important indication of innovation was obtained in the genome of vertebrates suggesting the occurrence of as-yet not described ST-related homologs in fish and tetrapods. The ST sequence identification in vertebrate genomes with key phylogenetic position like the sea lamprey Petromyzon marinus [59] at the stem of vertebrates or the spotted gar Lepisosteus oculatus at the base of teleosts [60] was helpful to propose an evolutionary scenario. These novel ST-related sequences could originate from gene duplication events like the two whole-genome duplications (WGD R1–R2) that occurred deep in the ancestry of the vertebrate lineage (2R hypothesis) [61]. 3. Molecular Phylogeny of β-Galactoside α2,3/6-Sialyltransferases Molecular phylogeny, with the construction of phylogenetic trees has been used to get further insight into the orthology and structure/function relationships of the identified β-galactoside α2,3/6-sialyltransferase sequences. As a first step, multiple sequence alignments (MSA) using predicted protein sequences, clustal Omega or MUSCLE algorithms evidenced several informative amino acid sites in the catalytic domain of ST to construct phylogenetic trees [39]. Among these conserved motifs, sialylmotifs and family motifs detection helped establishing the global evolutionary relationships between the identified ST sequences and enabled sequence-based prediction of their molecular function. Phylogeny of β-galactoside α2,3/6-sialyltransferase sequences was reconstructed using various methods implemented in the Molecular Evolutionary Genetics Analysis (MEGA) software and the reliability of the branching pattern was assessed by the bootstrap method [39,62]. The topology of the trees indicated that the ST6Gal and ST3Gal sequences identified in invertebrates are orthologous to the common ancestor of vertebrate subfamily members as they branch out from the tree before the split into vertebrate ST subfamilies with the exception of the ST3Gal members of the GRx group, which disappeared during vertebrate evolution [38,53]. The molecular phylogeny of β-galactoside α2,3/6-sialyltransferases is displayed in Figure 3 using iTOL [63]. 4. When β-Galactoside α2,3/6-Sialyltransferase Evolutionary Studies Meet Genome Reconstruction Gene organization and gene localization studies were used to assign the newly described st3gal and st6gal orthologs and to reconstruct the genetic events that have led to ST functional diversification in vertebrates. At the gene level, β-galactoside α2,3/6-sialyltransferase genes are polyexonic with an overall conserved exon/intron organization in each family from fish to mammals, which support the model of the common ancestral origin of each family (ST3Gal and ST6Gal) [55]. Interestingly, analysis of exon/intron organization and composition in the st6gal gene family showed that the st6gal1 genes encoded by frogs and fish have independently undergone different insertion events inside the first exon. These genetic events have led to an extended stem region of fish and frog ST6Gal I with potential impact on their enzyme activities [38]. It is speculated that during metazoan evolution, β-galactoside α2,3/6-sialyltransferase genes were subject to several duplication events affecting single genes or chromosomes or whole genomes. As far as the st3gal genes are concerned, a first series of tandem duplication of an ancestral st3gal gene in proto-Metazoa stem led to the GR1 and GR2/GR3/GRx groups of α2,3-sialyltransferases before the Porifera emergence. As previously reported for α2,8-sialyltransferases [30], a second series of tandem duplication took place after the Porifera radiation that gave rise to the full diversity of α2,3-sialyltransferase groups, as confirmed using ancestral genome reconstruction data from Putnam et al. [53,64]. This further indicates that the functional diversity of st3gal groups was acquired well before vertebrate divergence. In addition, gene copy number variants (CNV) were described within various animal genomes that might have contributed to evolutionary novelties, although not much is known about the functional impact of CNVs [65]. For instance, 2 and 3 copies of the ancestral st6gal1/2 gene were identified in the amphioxus (B. floridae) and in the sea lamprey (P. marinus) genome, respectively. Similarly, 4 copies of the st3gal1 gene named st3gal1A, st3gal1B, st3gal1C and st3gal1D, which are not shared among other fish species could be identified in close chromosomal location in the zebrafish genome. In the vertebrate genomes, detection of conserved synteny (i.e., set of orthologous genes born by a chromosomal segment in different genomes) and of large sets of paralogons (i.e., pair of chromosomes bearing a set of paralogous genes in a given genome resulting from WGD) provided strong evidence of the 2 rounds of WGD, which likely occurred about 500 and 555 million years ago (MYA). These large scale genetic events generated a class of paralogs known as ohnologs [66] and the various st3gal and st6gal gene subfamilies described in Table 1. Interestingly, 5 out of the 16 β-galactoside α2,3/6-sialyltransferase genes subfamilies generated after the 2 WGD events were immediately lost in the early vertebrate genome, while 4 other st3gal subfamilies were independently lost later on, in various vertebrate genomes like st3gal6 in teleosts or st3gal7 in tetrapods and st3gal8 in mammals (Table 1). Similarly, almost all the ST duplicates generated after the teleost specific third WGD event at the base of Actinopterygii were lost with the exception of st6gal2-r and st3gal3-r genes conserved in the zebrafish genome. Finally, chromosomal localization and genome reconstruction studies [67] of the ST gene loci in the various vertebrate genomes indicated several major chromosomal rearrangement and translocations of ST genes like ST3GAL5 in the human genome or st6gal1 in the zebrafish genome, which likely have undergone chromosomal translocation from Hsa2 to Hsa4 and from Dre15 to Dre21, respectively [38,53] (Figure 4). To account for ST gene novelties found in vertebrates, a refined nomenclature was proposed in Petit et al [39] based on the gene symbols and names assigned by the HUGO Gene Nomenclature Committee (HGNC; http://www.genenames.org/cgi-bin/genefamilies/set/438) and the ST nomenclature initially established by Tsuji et al. [45]. As described above, the vertebrate genomes contain numerous ST-related genes that result from various duplication events. The newly identified ST subfamilies were named according to their phylogenetic relationship with previously described ST subfamilies as follows: (1) A genome-wide duplication event known as WGD-R3 took place in the ray fin fish lineage leading to two copies of a gene that is otherwise found as a single copy in tetrapods. The symbols used for these specific fish duplicated genes are identical to those used for the mouse ST orthologs followed by “-r” meaning “-related” (Table 1); (2) Genes resulting from lineage-specific small scale duplications are named according to the mouse ST orthologs symbol followed by A, B, C, D (Table 1); (3) Finally, duplicates that resulted from whole genome duplication events WGD-R1 and R2 before the emergence of the teleosts branch are given a new ST subfamily number and no additional suffix is attributed. For instance, see in Table 1 the newly described vertebrate st3gal7, st3gal8 and st3gal9 gene subfamilies. The invertebrate ST genes are orthologous to the common ancestor of the vertebrate subfamilies and are named accordingly. For instance, the D. melanogaster st6gal1/2 gene (also known as DSiaT) described in [48] and the C. intestinalis st3gal1/2 gene [5]. 5. Conservation versus Changes in the β-Galactoside α2,3/6-Sialyltransferase Sequences Even though a phylogenetic tree might not be adequate to reflect relatedness between all sequences and may not provide sufficient resolution, the branch lengths are indicative of the sequence changes. To deduce the evolutionary rates, the branch lengths have to be divided by the elapsed corresponding time, calculated from the calibrations available in Hedges et al. [68]. As illustrated in Figure 5, ST3Gal I, ST3Gal II, ST3Gal III and ST6Gal II have the most conserved sequences across the vertebrate lineages, whereas ST6Gal I and to a lesser extent ST3Gal VI and ST3Gal IV show a particularly high evolutionary rate in their catalytic domain during Amniotes differentiation [38,53]. It is useful to substantiate the proximity/divergence of sequences between the different subfamilies using other approaches like similarity network. Orthology inference and evolutionary relationships were analyzed using protein sequences and the approach of similarity network visualization in which the nodes represent proteins and the edges indicate similarity in amino acid sequence [70]. The generated network can be visualized in Cytoscape [71]. The similarity network of a larger set of β-galactoside α2,3/6-sialyltransferase protein sequences demonstrated a high degree of similarity between ST3Gal sequences belonging to the GR1 group (ST3Gal I/II/VIII) with the notable exception of the fish ST3Gal sequences and a lower degree of similarity for the sequences belonging to the GR2 and GR3 groups [53]. Similar analysis conducted for ST6Gal sequences illustrated in Figure 6 highlighted a higher degree of similarity between the invertebrate ST6Gal I/II and vertebrate ST6Gal II sequences and pointed to a stronger conservation of ST6Gal II sequence at a stringent threshold (E-value). 6. Fate of Vertebrate Duplicated ST Genes After a gene duplication event, the two paralogous genes are identical. Non-functionalization and loss of one of the duplicates by accumulation of deleterious mutations is the most frequent outcome [66,72] while the parental gene is maintained active (Figure 7). As mentioned previously, 5 out of the 16 β-galactoside α2,3/6-sialyltransferase genes subfamilies generated after the 2 WGD events that took place at the root of the vertebrate lineage were immediately lost in the early vertebrate genome. Similarly, only 2 duplicated ST genes, namely st3gal3-r and st6gal2-r were maintained in the ray-finned fish genome after the teleost-specific round of WGD that occurred about 350 MYA. A pseudogenization process can occur at larger evolutionary scales by the accumulation of loss-of-function mutations in previously established genes and might also influence the fate of the surviving paralogs [73,74]. Interestingly, the inactivation of 4 st3gal subfamilies occurred independently, in various vertebrate genomes like st3gal6 in teleosts or st3gal7 in tetrapods and st3gal8 in mammals, while st3gal9 was maintained mainly in birds. Substitution rate analysis in each st3gal gene subfamily indicated a weaker selective pressure on the st3gal7, st3gal8 and st3gal9 genes and acquisition of mutations that compromised their function in mammals [53]. Indeed, several st3gal pseudogenes could be identified in the human genome that likely result from pseudogenization of a once active gene like ST3GAL8P on human chromosome 20 (ENSG00000242507). It is suggested that inactivation of the st3gal8 gene in the mammalian ancestor became possible after alternative or more beneficial glycosyltransferase activity evolved in the stem lineage of mammals, which could have resulted in major adaptive changes in SA metabolism. As illustrated in Figure 7, the function of the duplicated genes may diverge either because one or both evolved new function (neofunctionalization) [75] or because both duplicates partition the ancestral gene function (subfunctionalization) and several models have been proposed [76]. To understand the evolutionary forces that have influenced the ST gene number and their functional fate, the expression profile of the various β-galactoside α2,3/6-sialyltransferase genes was studied across vertebrates. As a first step, screening of various tissue EST libraries and statistical analysis using principal component analysis (PCA) of the expression profile accessible from the Unigene data base were undertaken [39]. The data pointed to a wider expression of st3gal and st6gal1 genes in mammals and birds, whereas teleost and amphibian st6gal1 genes showed a restricted profile of expression comparable to the one of st6gal2 genes suggesting a change in the expression profile of st6gal1 genes in amniotes. The expression pattern of the various β-galactoside α2,3/6-sialyltransferase genes analyzed by means of RT-PCR in adult vertebrate tissues or using whole mount in situ hybridization (ISH) in the developing zebrafish embryo and comparative genomics approaches confirmed a relative conservation of the st6gal2 gene expression in vertebrate tissues, in particular in the central nervous system, and the expansion of st6gal1 gene expression in mammalian tissues [38]. In addition, rapid amplification of cDNA ends (5’-RACE) conducted in fish and frog tissues demonstrated the occurrence of a unique st6gal1 transcript [38], whereas numerous studies highlighted the 5’-untranslated region heterogeneity of the mammalian st6gal1 genes leading to several mRNA isoforms [55,77,78,79]. These data confirmed the increasing complexity in the st6gal1 gene expression profile in higher vertebrates and suggested that phenotypic differences in the siaLome between organisms could have arisen from changes in gene regulation and from alterations in the protein coding region of st6gal1 gene (e.g., a neofunctionalization of the st6gal1 gene in birds and mammals) [38]. As far as the st3gal genes are concerned, their functional fate could not be predicted on the basis of gene expression profile alone. However, st3gal gene losses were tentatively linked with relaxed gene evolution and reduced gene expression. These studies indicated that the most widely expressed st3gal genes like st3gal2 and st3gal3 were also the most evolutionary conserved, whereas st3gal genes losses were linked to high substitution rates and to restricted tissue expression [53]. 7. Functional Divergence and Molecular Evolution of STs To better understand the molecular basis of STs functional divergence after WGD, evolution of the function of the various β-galactoside α2,3/6-sialyltransferases was analyzed from a structural perspective. Despite sharing primary and secondary structural similarities, ST have different acceptor substrate specificities that can be ascribed to amino acid sites. A general method to predict functionally important sites or structural role of amino acid positions in a protein is to analyze their conservation level based on the assumption that highly conserved positions among member of the same family. To analyze the conservation level in the ST6Gal family sequences of ST6Gal I and ST6Gal II, protein sequences were aligned separately to build a profile and a MSA comprising the two subfamilies was obtained using profile-profile mode with ClustalW. The sequence conservation was calculated using ConSurf server [80] and mapped into the crystal structure of human ST6Gal I in complex with cytidine and phosphate [81]. As shown in Figure 8, the highest conserved residues of ST6Gal are mostly located in the active site. On the other hand, it is useful to decipher the specificity-determining positions (SDPs) of a family protein, i.e., the critical amino acids determining their functional specificity. These positions often play critical roles as they are involved in the molecular mechanisms ensuring functional diversity. Within a MSA, SDPs are amino acid positions that show a pattern of conservation in agreement with subfamily divergence. Two types of SDPs can be distinguished: a Type I corresponds to diverse amino acids in one group and a conserved one in the other(s) reflecting different levels of functional constraints between duplicated genes, whereas Type II positions are characterized by different conserved amino acids among groups associated to divergent constraints [82,83]. SDP prediction between the three vertebrate ST3Gal groups (GR1–GR3) led to the identification of five SDP, namely S197, Y233, V234, W304 and N307 in the reference porcine ST3Gal I structure (PDB: 2WNB) that are located in the active site [53]. These SDPs in close contact to the ST3Gal substrates are indicative of the functional divergence of each group of ST3Gal sequences in early vertebrates. As illustrated in Figure 9, SDP prediction between ST6Gal I and ST6Gal II was carried out using SPEER server [82,84]. Six SDPs corresponding to positions 95, 122, 169, 357, 359 and 380 in the reference sequence human ST6Gal I (PDB: 4js2) [81] localized at the protein surface and could reflect protein-protein interaction evolution. Two type II SDPs corresponding to L326 and F346 were found in the sialylmotif S, which is involved in acceptor and donor binding [85] and three others V352, Q357 and F359 were found in the mobile loop nearby sialylmotif III (Figure 9A,B). Interestingly, the highest scored SDP prediction corresponds to position Y122 in the reference structure that is located near the N-glycan binding site. A Tyr residue is conserved in ST6Gal I, whereas a His is conserved in ST6Gal II sequences. The hydrogen bonds involving the residues Y122, D274 and Y369 and the α1,3Man branch of the N-glycan, place the galactose (Gal) in the vicinity of CMP and the catalytic residue H370 [81] (Figure 9C). To visualize the impact of the amino acid change at this position, we performed in silico mutation Y122H and the Gal residue was modified to N-acetylgalactosamine (GalNAc), the monosaccharide acceptor for the ST6Gal II activity [86]. Point mutation was performed using the Dunbrack backbone-dependent rotamer library [87], the most probable rotamer was chosen and changes in the structure were followed by structure minimization (Figure 9D). The ST6Gal II H122 residue can participate in hydrogen bond with the Y369, whereas the D274 can interact with K121 and GalNAc. The model also shows that the N atom of GalNAc can be involved in a hydrogen bond with Y275. In summary they are no dramatic changes at the substrate binding site between ST6Gal I and ST6Gal II, however a semi-conservative mutation, such as Y122H, can impact in the substrate stabilization. 8. Conclusion This review has reported an overview of the evolutionary history of the β-galactoside α2,3/6-sialyltransferases. The human/mouse ST3Gal and ST6Gal families are comprised of eight members with low overall sequence similarities except for the conserved sialylmotif and family motifs in the catalytic domain that are hallmarks for homologs identification in databases. Genetics, molecular phylogeny and functional genomics approaches have been used to decipher their evolutionary relationships in the context of the dynamic remodeling of genome content (gene loss/gain, segmental and whole genome duplication events). Interestingly, the st3gal and st6gal genes could be identified in the sponge O. carmela suggesting their ancient occurrence in the metazoans and their expansion in deuterostome lineages. The 8 human st3gal and st6gal orthologs were also identified in all vertebrate genomes with the notable exception of the st3gal6 gene, which has been lost in bony fish. In addition, several novel and less conserved st3gal subfamilies have been described in non-mammalian vertebrates, some of which are restricted to birds and duck-billed platypus (e.g., st3gal9) or to fish (e.g., st3gal7) that could be associated with specialized or species-specific tasks. Finally, protein sequence and structural analyses shed light into the functional evolution of ST3Gal and ST6Gal, their enzymatic specificities and their role in cell-cell interactions and diseases. Acknowledgments This work has been funded by the Centre National de La Recherche Scientifique (CNRS), by the University of Lille Nord de France (PPF bioinformatique of the Lille1 University), by the ANR-2010-BLAN-120401 grant from the Agence Nationale de la Recherche, by the Ligue régionale de la recherche contre le cancer (2016). Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1286/s1. Click here for additional data file. Author Contributions Roxana E. Teppa, Daniel Petit and Anne Harduin-Lepers conceived the study, performed the experiments and analyzed the data; Olga Plechakova, Virginie Cogez and Anne Harduin-Lepers handled the sequences and sequence files; Roxana E. Teppa, Daniel Petit and Anne Harduin-Lepers wrote the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Sialic acids and sialylated molecules. (A) N-acetylneuraminic acid (Neu5Ac) is the major sialic acid molecule found in human tissues. Other commonly described sialic acids in vertebrates are N-glycolylneuraminic acid (Neu5Gc) and 2-keto-3-deoxy-nonulosonic acid (Kdn); (B) In vertebrates, the sialic acid hydroxyl group at position 2 is most frequently glycosidically-linked to either the 3- or 6-hydroxyl group of galactose (Gal) residues. These glycosidic linkages are formed by the β-galactoside α2,3/6-sialyltransferases described in this review. Figure 2 Evolution of the biosynthetic pathway of sialic acids in Metazoa. (A) Schematic representation of the vertebrate biosynthetic pathway of sialylated molecules. Key enzymes implicated in the biosynthetic pathway of sialic acids are indicated as follows: GNE: UDP-GlcNAc2epimerase/ManNAc kinase; NANS: Neu5Ac9-phosphate synthase; CMAS: CMP-Neu5Ac synthase: CMAH: CMP-Neu5Ac hydroxylase; SLC35A1: CMP-Neu5Ac transporter; ST: sialyltransferases, Neu: neuraminidase; (B) Illustration of the evolutionary history of the sialic acid biosynthetic pathway in the metazoans. Evidences of the occurrence of the biosynthetic pathway of sialylated molecules across the metazoans have been obtained based on BLAST search analysis of the various actors in genomic databases. Yellow stars indicate the two whole genome duplication events (WGD R1–R2) that took place at the base of vertebrates and the teleostean whole genome duplication event (WGD R3) that occurred in the stem of bony fish. Figure 3 Maximum Likelihood phylogenetic trees of protein sequences of β-galactoside α2,3/6-sialyltransferases (ST3Gal and ST6Gal). In both cases, the phylogenetic trees were inferred using the Maximum Likelihood method based on the Whelan and Goldman method with options G (gamma distribution) and I (Invariant sites present). Alignments were performed using Clustal Omega, available at the site http://www.ebi.ac.UK/Tools/msa/clustalo/ (Data S1). Evolutionary analyses were conducted in MEGA6. The trees were re-drawn using iTOL 3.2 [63] available at the URL http://itol.embl.de/. (A) ST3Gal tree was obtained from 124 sequences and 228 positions in the final data set; (B) ST6Gal tree was inferred from 50 sequences and 256 positions and 50 sequences in the final data set. Figure 4 Hypothetical scenario of the evolutionary history of β-galactoside α2,3/6-sialyltransferase genes in the WGR context. This drawing illustrates the proposed evolutionary scenario of st3gal and st6gal genes drawn in line with the 2R hypothesis [61]. The arrows indicate the two vertebrate whole genome duplication events (WGD-R1: ~555 MYA and WGD-R2: ~500 MYA) and the teleosts specific whole genome duplication events (WGD-R3: ~350 MYA). A single st3gal1/2/8 (GR1), st3gal3/5/7 (GR3), st3gal4/6/9 (GR2) and st6gal1/2 gene in the stem bilaterian was duplicated twice before and after the emergence of agnathans, raising 11 st3gal and st6gal subfamilies at the base of gnathostomes. 3 st3gal gene subfamilies (st3gal7, st3gal8 and st3gal9) were further lost in the mammalian lineage. In Actinopterygii, after WGD-R3 the two duplicated genes st3gal3-r and st6gal2-r are maintained in the zebrafish genome, whereas the st3gal6 and st3gal9 genes are secondarily lost. S. purpuratus = Strongylocentrotus purpuratus; C. intestinalis = Ciona intestinalis; B. floridae = Branchiostoma floridae; P. marinus = Petromyzon marinus; C. milii = Calorhinchus milii; D. rerio = Danio rerio. Figure 5 Evolutionary rates of β-galactoside α2,3/6-sialyltransferase subfamilies in Vertebrates. (A) Trees obtained from Minimum Evolution and JTT model implemented in MEGA 6.0 [69] using 91 ST3Gal and 27 ST6Gal sequences allowed calculation of the evolutionary rates of β-galactoside α2,3/6-sialyltransferase subfamilies for the major divisions in Vertebrates (Data S2). For Amniotes, 2 or 3 sequences were taken, including at least Man and a Marsupial in Mammals, Chicken and Ostrich in Avians, and Caroline Anole and Burmese Python in Lepidosaurians. Orange background denotes the highest values; (B) Mean evolutionary rates in the different β-galactoside α2,3/6-sialyltransferase subfamilies (±standard error). The mean evolutionary rates of ST3Gal and ST6Gal subfamilies were calculated from Teleosteans to Amniotes. The standard errors show variations in the different subfamilies, from the highest in ST3Gal VIII and ST6Gal I, to the lowest in ST3Gal II, ST3Gal I, ST3Gal III and ST6Gal II; (C) Highest evolution rates of β-galactoside α2,3/6-sialyltransferase in vertebrate evolutionary tree. They correspond to the cases where an elevated value is observed in one or two lineages. During the differentiation of Amniotes, we record three subfamilies particularly evolving their catalytic sequences, ST6Gal I and at a lesser extent ST3Gal III and ST3Gal IV. In the lineage of Tetrapods, ST3Gal IV and ST3Gal V present high evolutionary rates. In the ancestors of birds and Lepidosaurians (snakes and lizards, i.e., Sauropsides), there is only one subfamily where numerous changes occur in the catalytic domain, the ST3Gal VIII, as mentioned later on. Figure 6 Sequence similarity network of ST6Gal sequences. The Figure represents ST sequences as nodes (circles) and all pairwise sequence relationships (alignments) better than a BLAST E-value threshold of 1E-83 as edges (lines). The network is composed by 126 ST6Gal sequences and 9 ST3Gal sequences as control group (red circles). (A) The network is visualized using a Force Direct layout, where the length of the edges is inversely proportional to the sequence similarity. Sequences belonging to Invertebrates form a separate group from all ST6Gal sequences, pointing out the dissimilarity with ST6Gal I and ST6Gal II sequences. To better visualize the relationships of invertebrates sequences, we show only the edges that involve the Invertebrate sequences; In panel (B) sequences are clustered by groups without using edges information (the edges are not proportional to sequence similarity). The network shows that seven invertebrate sequences are related with sequences of ST6Gal2 and ST6Gal1 of the Bird and Fish groups. The names of the invertebrate sequences related to other groups and the number of edges are shown in the table. It is important to note that the Invertebrate sequences do not show relationships with the mammalian ST6Gal1 sequences at this threshold. Figure 7 Evolutionary fate of ST gene duplicates after WGD events. On the left side, schematic representation of ancestral polyexonic ST genes duplicates (exons are represented by colored boxes and genomic regulatory elements are represented by black and white circles). On the right side, the three major evolutionary fates of the various newly created ST gene subfamilies are indicated (1) pseudogenization: gene loss; (2) subfunctionalization: coding sequences and regulatory elements evolve and are partitioned according to specific molecular functions; (3) neofunctionalization: one of the newly duplicated gene accumulates mutations in its coding region and/or in its regulatory genomic elements giving rise to new molecular function. Figure 8 High sequence conservation near the binding site in ST6Gal. ST6Gal I and ST6Gal II show a high level of conservation in the region surrounding the ligand binding site. MSA were obtained using using profile-profile mode with clustalX and 101 vertebrate ST6Gal I and ST6Gal II sequences (Data S3). Sequence conservation was calculated with ConSurf server, the result was mapped into the PDB structure 4JS1, a crystal structure of human β-galactoside α2,6-sialyltransferase I (ST6Gal I) in a complex with cytidine (CTN) and phosphate (PO4). The structure is depicted in cartoon, colored by the conservation score. Ligand molecules are shown in ball and stick representation, residues at contact distance (<5 Å) from ligands are shown in sphere. Figure 9 Sequence and structural mapping of SDPs between ST6GalI and ST6GalII. (A) The human protein sequence of ST6Gal I corresponding to PDB 4js2 is shown as a reference sequence, at the top of the panel. The functionally important regions are indicated with colored boxes and 14 SDPs of type II are highlighted in red. SDP prediction between ST6Gal I and ST6Gal II was carried out using SPEER server [82] using the MSA described previously that was composed by 48 sequences belonging to ST6Gal I and 53 of ST6Gal II (Data S3). A total of 78 and 69 SDPs of Type I and Type II respectively were predicted (Data S4). At the bottom of the panel, a table is shown with the 14 Type II SDPs with a reliable score. The first column indicates the position in the multiple sequence alignment (MSA), the second column the position in the reference sequence, the 3rd and 4th columns indicate the conserved amino acid in the ST6Gal I and ST6Gal II sequences, respectively. The SDPs are sorted according to the predictive score, from higher to lower; from top to bottom and from left to right; (B) The functionally important regions and SDPs are shown in the structure of the reference human ST6Gal I sequence. In addition, the surfaces of the substrates are shown; (C) Close-up of the glycan binding site of ST6Gal I, hydrogen bonds are denoted by dashed lines; (D) Close-up of the glycan binding site, where the Gal was modeled to GalNAc to represent ST6Gal II binding site. ijms-17-01286-t001_Table 1Table 1 Vertebrate β-galactoside α2,3/6-sialyltransferases ohnologs: Vertebrate β-galactoside α2,3/6-sialyltransferase sequences belonging to the ST3Gal and ST6Gal families are grouped in 4 clades with distinct evolutionary origins (GR1, GR2 and GR3 for the ST3Gal and a unique group for the ST6Gal) encompassing 9 st3gal and 2 st6gal ohnologs. Acceptor substrate preferences of the mammalian enzymes and predicted acceptor substrate preference (in blue) of the novel vertebrate enzymes lost in mammals are indicated. Group Ancestral (Before 2nd WGDR) Ohnologs (After 2nd WGDR) Fish (After 3rd WGDR) Tetrapods Acc. Substrate Amphibians Birds Mammals GR1 st3gal1/2/8 st3gal1 st3gal1 st3gal1 st3gal1 st3gal1 Galβ1,3GalNAc-Ser st3gal2 st3gal2 st3gal2 st3gal2 st3gal2 Galβ1,3GalNAc-Ser st3gal8 st3gal8 st3gal8 st3gal8 lost Galβ1,3GalNAc-Ser GR2 st3gal3/5/7 st3gal3 st3gal3 st3gal3 st3gal3 st3gal3 Galβ1,3GalNAc-R st3gal3-r st3gal5 st3gal5 st3gal5 st3gal5 st3gal5 GM3 synthase st3gal7 st3gal7 lost lost lost GM4 synthase GR3 st3gal4/6/9 st3gal4 st3gal4 st3gal4 st3gal4 st3gal4 Galβ1,4GlcNAc-R st3gal6 lost st3gal6 st3gal6 st3gal6 Galβ1,4GlcNAc-R st3gal9 lost lost st3gal9 lost (except in platypus) Galβ1,4GlcNAc-R – st6gal1/2 st6gal1 st6gal1 st6gal1 st6gal1 st6gal1 Galβ1,4GlcNAc-R st6gal2 st6gal2 st6gal2 st6gal2 st6gal2 GalNAcβ1,4GlcNAc-R st6gal2-r ==== Refs References 1. Guerardel Y. Chang L.Y. Fujita A. Coddeville B. Maes E. Sato C. Harduin-Lepers A. Kubokawa K. Kitajima K. SiaLome analysis of the cephalochordate branchiostoma belcheri, a key organism for vertebrate evolution Glycobiology 2012 22 479 491 10.1093/glycob/cwr155 22057003 2. Schauer R. Sialic acids: Fascinating sugars in higher animals and man Zoology (Jena) 2004 107 49 64 10.1016/j.zool.2003.10.002 16351927 3. Schauer R. Kamerling J. Chemistry, biochemistry and biology of sialic acids Glycoproteins II Montreuil J. Vliegenthart J. Schachter H. Elsevier Amsterdam, The Netherlands 1997 Volume 29b 243 402 4. Varki A. Diversity in the sialic acids Glycobiology 1992 2 25 40 10.1093/glycob/2.1.25 1550987 5. Lehmann F. Kelm S. Dietz F. von Itzstein M. Tiralongo J. The evolution of galactose α2,3-sialyltransferase: Ciona intestinalis ST3GAL I/II and takifugu rubripes ST3GAL II sialylate Galbeta1, 3GalNAc structures on glycoproteins but not glycolipids Glycoconj. J. 2008 25 323 334 10.1007/s10719-007-9078-4 17973185 6. Corfield A.P. Berry M. Glycan variation and evolution in the eukaryotes Trends Biochem. Sci. 2015 40 351 359 10.1016/j.tibs.2015.04.004 26002999 7. Gagneux P. Varki A. Evolutionary considerations in relating oligosaccharide diversity to biological function Glycobiology 1999 9 747 755 10.1093/glycob/9.8.747 10406840 8. Koles K. Repnikova E. Pavlova G. Korochkin L.I. Panin V.M. Sialylation in protostomes: A perspective from drosophila genetics and biochemistry Glycoconj. J. 2009 26 313 324 10.1007/s10719-008-9154-4 18568399 9. Roth J. Kempf A. Reuter G. Schauer R. Gehring W.J. Occurrence of sialic acids in drosophila melanogaster Science 1992 256 673 675 10.1126/science.1585182 1585182 10. Malykh Y.N. Krisch B. Gerardy-Schahn R. Lapina E.B. Shaw L. Schauer R. The presence of N -acetylneuraminic acid in malpighian tubules of larvae of the cicada philaenus spumarius Glycoconj. J. 1999 16 731 739 10.1023/A:1007115627708 11003558 11. Saito M. Kitamura H. Sugiyama K. Occurrence of gangliosides in the common squid and pacific octopus among protostomia Biochim. Biophys. Acta 2001 1511 271 280 10.1016/S0005-2736(01)00282-6 11286970 12. Kim C. Sialic acid (N -acetylneuraminic acid) as the functional molecule for differentiation between animal and plant kingdom J. Glycom. Lipidom. 2014 4 e116 13. Angata T. Varki A. Chemical diversity in the sialic acids and related aketo acids: An evolutionary perspective Chem. Rev. 2002 102 439 469 10.1021/cr000407m 11841250 14. Corfield A. Schauer R. Sialic acids: Chemistry, metabolism and function Cell Biology Monograph Springer Berlin, Germany 1982 Volume 10 5 50 15. Cohen M. Varki A. The siaLome—Far more than the sum of its parts J. Integr. Biol. 2010 14 455 464 10.1089/omi.2009.0148 20726801 16. Schauer R. Sialic acids as regulators of molecular and cellular interactions Curr. Opin. Struct. Biol. 2009 19 507 514 10.1016/j.sbi.2009.06.003 19699080 17. Corfield A.P. Mucins: A biologically relevant glycan barrier in mucosal protection Biochim. Biophys. Acta 2015 1850 236 252 10.1016/j.bbagen.2014.05.003 24821013 18. Angata T. Brinkman-Van der Linden E. I-type lectins Biochim. Biophys. Acta 2002 1572 294 316 10.1016/S0304-4165(02)00316-1 12223277 19. Kelm S. Schauer R. Sialic acids in molecular and cellular interactions Int. Rev. Cytol. 1997 175 137 240 9203358 20. Schnaar R.L. Gerardy-Schahn R. Hildebrandt H. Sialic acids in the brain: Gangliosides and polysialic acid in nervous system development, stability, disease, and regeneration Physiol. Rev. 2014 94 461 518 10.1152/physrev.00033.2013 24692354 21. Gabius H.J. Biological information transfer beyond the genetic code: The sugar code Naturwissenschaften 2000 87 108 121 10.1007/s001140050687 10798195 22. Gabius H.J. Andre S. Kaltner H. Siebert H.C. The sugar code: Functional lectinomics Biochim. Biophys. Acta 2002 1572 165 177 10.1016/S0304-4165(02)00306-9 12223267 23. Lehmann F. Tiralongo E. Tiralongo J. Sialic acid-specific lectins: Occurrence, specificity and function Cell. Mol. Life Sci. 2006 63 1331 1354 10.1007/s00018-005-5589-y 16596337 24. Gagneux P. Cheriyan M. Hurtado-Ziola N. van der Linden E.C. Anderson D. McClure H. Varki A. Varki N.M. Human-specific regulation of α2–6-linked sialic acids J. Biol. Chem. 2003 278 48245 48250 10.1074/jbc.M309813200 14500706 25. Hayakawa T. Varki A. Human-specific changes in sialic acid biology Post-Genome Biology of Primates Hirai H. Imai H. Go Y. Springer Tokyo Tokyo, Japan 2012 123 148 26. Varki N.M. Strobert E. Dick E.J. Jr. Benirschke K. Varki A. Biomedical differences between human and nonhuman hominids: Potential roles for uniquely human aspects of sialic acid biology Annu. Rev. Pathol. 2011 6 365 393 10.1146/annurev-pathol-011110-130315 21073341 27. Li Y. Chen X. Sialic acid metabolism and sialyltransferases: Natural functions and applications Appl. Microbiol. Biotechnol. 2012 94 887 905 10.1007/s00253-012-4040-1 22526796 28. De Mendoza A. Ruiz-Trillo I. The mysterious evolutionary origin for the GNE gene and the root of bilateria Mol. Biol. Evol. 2011 28 2987 2991 10.1093/molbev/msr142 21616910 29. Giacopuzzi E. Bresciani R. Schauer R. Monti E. Borsani G. New insights on the sialidase protein family revealed by a phylogenetic analysis in metazoa PLoS ONE 2012 7 1286 10.1371/journal.pone.0044193 22952925 30. Harduin-Lepers A. Petit D. Mollicone R. Delannoy P. Petit J.M. Oriol R. Evolutionary history of the α2,8-sialyltransferase (ST8Sia ) gene family: Tandem duplications in early deuterostomes explain most of the diversity found in the vertebrate ST8Sia genes BMC Evol. Biol. 2008 8 1286 10.1186/1471-2148-8-258 18811928 31. Simakov O. Kawashima T. Marletaz F. Jenkins J. Koyanagi R. Mitros T. Hisata K. Bredeson J. Shoguchi E. Gyoja F. Hemichordate genomes and deuterostome origins Nature 2015 527 459 465 10.1038/nature16150 26580012 32. Chou H.H. Hayakawa T. Diaz S. Krings M. Indriati E. Leakey M. Paabo S. Satta Y. Takahata N. Varki A. Inactivation of CMP-N -acetylneuraminic acid hydroxylase occurred prior to brain expansion during human evolution Proc. Natl. Acad. Sci. USA 2002 99 11736 11741 10.1073/pnas.182257399 12192086 33. Varki A. Loss of N -glycolylneuraminic acid in humans: Mechanisms, consequences, and implications for hominid evolution Am. J. Phys. Anthropol. 2001 116 Suppl. 33 S54 S69 10.1002/ajpa.10018 11786991 34. Ng P.S. Bohm R. Hartley-Tassell L.E. Steen J.A. Wang H. LUKowski S.W. Hawthorne P.L. Trezise A.E. Coloe P.J. Grimmond S.M. Ferrets exclusively synthesize Neu5Ac and express naturally humanized influenza a virus receptors Nat. Commun. 2014 5 5750 10.1038/ncomms6750 25517696 35. Schauer R. Srinivasan G.V. Coddeville B. Zanetta J.P. Guerardel Y. Low incidence of N -glycolylneuraminic acid in birds and reptiles and its absence in the platypus Carbohydr. Res. 2009 344 1494 1500 10.1016/j.carres.2009.05.020 19541293 36. Sellmeier M. Weinhold B. Münster-Kühnel A. Cmp-sialic acid synthetase: The point of constriction in the sialylation pathway Sialoglyco Chemistry and Biology I Springer Berlin, Germany 2013 Volume 366 139 167 37. Viswanathan K. Tomiya N. Park J. Singh S. Lee Y.C. Palter K. Betenbaugh M.J. Expression of a functional drosophila melanogaster CMP-sialic acid synthetase: Differential localization of the drosophila and human enzymes J. Biol. Chem. 2006 281 15929 15940 10.1074/jbc.M512186200 16537535 38. Petit D. Mir A.M. Petit J.M. Thisse C. Delannoy P. Oriol R. Thisse B. Harduin-Lepers A. Molecular phylogeny and functional genomics of β-galactoside α2,6-sialyltransferases that explain ubiQuitous expression of st6gal1 gene in amniotes J. Biol. Chem. 2010 285 38399 38414 10.1074/jbc.M110.163931 20855889 39. Petit D. Teppa R.E. Petit J.M. Harduin-Lepers A. A practical approach to reconstruct evolutionary history of animal sialyltransferases and gain insights into the sequence-function relationships of golgi-glycosyltransferases Glycosyltransferases: Methods and Protocols Brockausen I. Springer, Humana Press New York, NY, USA 2013 Volume 1022 73 97 40. Harduin-Lepers A. Vertebrate sialyltransferases Sialobiology: Structure, Biosynthesis and Function; Sialic Acid Glycoconjugates in Health and Diseases Tiralongo J. Martinez-Duncker I. Bentham Science Schiphol, The Netherlands 2013 Volume 5 139 187 41. Harduin-Lepers A. Vallejo-Ruiz V. Krzewinski-Recchi M.A. Samyn-Petit B. Julien S. Delannoy P. The human sialyltransferase family Biochimie 2001 83 727 737 10.1016/S0300-9084(01)01301-3 11530204 42. Paulson J.C. Colley K.J. Glycosyltransferases. Structure, localization, and control of cell type-specific glycosylation J. Biol. Chem. 1989 264 17615 17618 2681181 43. Harduin-Lepers A. Mollicone R. Delannoy P. Oriol R. The animal sialyltransferases and sialyltransferase-related genes: A phylogenetic approach Glycobiology 2005 15 805 817 10.1093/glycob/cwi063 15843597 44. Cantarel B.L. Coutinho P.M. Rancurel C. Bernard T. Lombard V. Henrissat B. The carbohydrate-active enzymes database (CAZy): An expert resource for glycogenomics Nucleic Acids Res. 2009 37 233 238 10.1093/nar/gkn663 18838391 45. Tsuji S. Datta A.K. Paulson J.C. Systematic nomenclature for sialyltransferases Glycobiology 1996 6 5 7 46. Kidani S. Kaneoka H. Okuzaki Y. Asai S. Kojima Y. Nishijima K.I. Iijima S. Analyses of chicken sialyltransferases related to O -glycosylation J. Biosci. Bioeng. 2016 10.1016/j.jbiosc.2016.03.017 27150510 47. Kojima Y. Mizutani A. Okuzaki Y. Nishijima K. Kaneoka H. Sasamoto T. Miyake K. Iijima S. Analyses of chicken sialyltransferases related to N -glycosylation J. Biosci. Bioeng. 2015 119 623 628 10.1016/j.jbiosc.2014.11.009 25499752 48. Koles K. Irvine K.D. Panin V.M. Functional characterization of drosophila sialyltransferase J. Biol. Chem. 2004 279 4346 4357 10.1074/jbc.M309912200 14612445 49. Kajiura H. Hamaguchi Y. Mizushima H. Misaki R. Fujiyama K. Sialylation potentials of the silkworm, Bombyx mori; B. Mori possesses an active α2,6-sialyltransferase Glycobiology 2015 25 1441 1453 10.1093/glycob/cwv060 26306633 50. Harduin-Lepers A. Recchi M.A. Delannoy P. 1994, the year of sialyltransferases Glycobiology 1995 5 741 758 10.1093/glycob/5.8.741 8720072 51. Takashima S. Characterization of mouse sialyltransferase genes: Their evolution and diversity Biosci. Biotechnol. Biochem. 2008 72 1155 1167 10.1271/bbb.80025 18460788 52. Audry M. Jeanneau C. Imberty A. Harduin-Lepers A. Delannoy P. Breton C. Current trends in the structure-activity relationships of sialyltransferases Glycobiology 2011 21 716 726 10.1093/glycob/cwq189 21098518 53. Petit D. Teppa E. Mir A.-M. Vicogne D. Thisse C. Thisse B. Filloux C. Harduin-Lepers A. Integrative view of α2,3-sialyltransferases (ST3Gal) molecular and functional evolution in deuterostomes: Significance of lineage-specific losses Mol. Biol. Evol. 2015 32 906 927 10.1093/molbev/msu395 25534026 54. Altschul S.F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Gapped blast and PSI-blast: A new generation of protein database search programs Nucleic Acids Res. 1997 25 3389 3402 10.1093/nar/25.17.3389 9254694 55. Harduin-Lepers A. Comprehensive analysis of sialyltransferases in vertebrate genomes Glycobiol. Insights 2010 2 29 61 10.4137/GBI.S3123 56. Moreau H. Verhelst B. Couloux A. Derelle E. Rombauts S. Grimsley N. van Bel M. Poulain J. Katinka M. Hohmann-Marriott M.F. Gene functionalities and genome structure in bathycoccus prasinos reflect cellular specializations at the base of the green lineage Genome Biol. 2012 13 74 90 10.1186/gb-2012-13-8-r74 22925495 57. Read B.A. Kegel J. Klute M.J. Kuo A. Lefebvre S.C. Maumus F. Mayer C. Miller J. Monier A. Salamov A. Pan genome of the phytoplankton emiliania underpins its global distribution Nature 2013 499 209 213 10.1038/nature12221 23760476 58. King N. Westbrook M.J. Young S.L. Kuo A. Abedin M. Chapman J. Fairclough S. Hellsten U. Isogai Y. Letunic I. The genome of the choanoflagellate monosiga brevicollis and the origin of metazoans Nature 2008 451 783 788 10.1038/nature06617 18273011 59. Smith J.J. Kuraku S. Holt C. SaUKa-Spengler T. Jiang N. Campbell M.S. Yandell M.D. ManoUSAki T. Meyer A. Bloom O.E. Sequencing of the sea lamprey (petromyzon marinus) genome provides insights into vertebrate evolution Nat. Genet. 2013 45 415 421 10.1038/ng.2568 23435085 60. Braasch I. Postlethwait J.H. The teleost agouti-related protein 2 gene is an ohnolog gone missing from the tetrapod genome Proc. Natl. Acad. Sci. USA 2011 108 47 48 10.1073/pnas.1101594108 21406593 61. Ohno S. Evolution by Gene Duplication Springer-Verlag Heidelberg, Germany 1970 1 151 62. Kumar S. Stecher G. Tamura K. Mega7: Molecular evolutionary genetics analysis version 7.0 for bigger datasets Mol. Biol. Evol. 2016 33 1870 1874 10.1093/molbev/msw054 27004904 63. Letunic I. Bork P. Interactive tree of life (iTOL) v3: An online tool for the display and annotation of phylogenetic and other trees Nucleic Acids Res. 2016 44 242 245 10.1093/nar/gkw290 27095192 64. Putnam N.H. Butts T. Ferrier D.E. Furlong R.F. Hellsten U. Kawashima T. Robinson-Rechavi M. Shoguchi E. Terry A. Yu J.K. The amphioxus genome and the evolution of the chordate karyotype Nature 2008 453 1064 1071 10.1038/nature06967 18563158 65. Henrichsen C.N. Chaignat E. Reymond A. Copy number variants, diseases and gene expression Hum. Mol. Genet. 2009 18 1 8 10.1093/hmg/ddp011 18815198 66. Wolfe K.H. Yesterday’s polyploids and the mystery of diploidization Nat. Rev. Genet. 2001 2 333 341 10.1038/35072009 11331899 67. Yegorov S. Good S. Using paleogenomics to study the evolution of gene families: Origin and duplication history of the relaxin family hormones and their receptors PLoS ONE 2012 7 1286 10.1371/journal.pone.0032923 22470432 68. Hedges S.B. Marin J. Suleski M. Paymer M. Kumar S. Tree of life reveals clock-like speciation and diversification Mol. Biol. Evol. 2015 32 835 845 10.1093/molbev/msv037 25739733 69. Tamura K. Stecher G. Peterson D. Filipski A. Kumar S. Mega6: Molecular evolutionary genetics analysis version 6.0 Mol. Biol. Evol. 2013 30 2725 2729 10.1093/molbev/mst197 24132122 70. Atkinson H.J. Morris J.H. Ferrin T.E. Babbitt P.C. Using sequence similarity networks for visualization of relationships across diverse protein superfamilies PLoS ONE 2009 4 1286 10.1371/journal.pone.0004345 19190775 71. Shannon P. Markiel A. Ozier O. Baliga N.S. Wang J.T. Ramage D. Amin N. Schwikowski B. Ideker T. Cytoscape: A software environment for integrated models of biomolecular interaction networks Genome Res. 2003 13 2498 2504 10.1101/gr.1239303 14597658 72. Lynch M. Conery J.S. The evolutionary fate and consequences of duplicate genes Science 2000 290 1151 1155 10.1126/science.290.5494.1151 11073452 73. Albalat R. Canestro C. Evolution by gene loss Nat. Rev. Genet. 2016 17 379 391 10.1038/nrg.2016.39 27087500 74. Canestro C. Albalat R. Irimia M. Garcia-Fernandez J. Impact of gene gains, losses and duplication modes on the origin and diversification of vertebrates Semin. Cell Dev. Biol. 2013 24 83 94 10.1016/j.semcdb.2012.12.008 23291262 75. Force A. Lynch M. Pickett F.B. Amores A. Yan Y.L. Postlethwait J. Preservation of duplicate genes by complementary, degenerative mutations Genetics 1999 151 1531 1545 10101175 76. Innan H. Kondrashov F. The evolution of gene duplications: Classifying and distinguishing between models Nat. Rev. Genet. 2010 11 97 108 10.1038/nrg2689 20051986 77. Lo N.W. Lau J.T. Transcription of the β-galactoside α2,6-sialyltransferase gene in B lymphocytes is directed by a separate and distinct promoter Glycobiology 1996 6 271 279 10.1093/glycob/6.3.271 8724135 78. Svensson E.C. Soreghan B. Paulson J.C. Organization of the β-galactoside α2,6-sialyltransferase gene. Evidence for the transcriptional regulation of terminal glycosylation J. Biol. Chem. 1990 265 20863 20868 2249992 79. Wang X. O'Hanlon T.P. Young R.F. Lau J.T. Rat β-galactoside α2,6-sialyltransferase genomic organization: Alternate promoters direct the synthesis of liver and kidney transcripts Glycobiology 1990 1 25 31 10.1093/glycob/1.1.25 1983783 80. Ashkenazy H. Abadi S. Martz E. Chay O. Mayrose I. Pupko T. Ben-Tal N. ConSurf 2016: An improved methodology to estimate and visualize evolutionary conservation in macromolecules Nucleic Acids Res. 2016 44 344 350 10.1093/nar/gkw408 27166375 81. Kuhn B. Benz J. Greif M. Engel A.M. Sobek H. Rudolph M.G. The structure of human α2,6-sialyltransferase reveals the binding mode of complex glycans Acta. Crystallogr. D Biol. Crystallogr. 2013 69 1826 1838 10.1107/S0907444913015412 23999306 82. Chakrabarti S. Bryant S.H. Panchenko A.R. Functional specificity lies within the properties and evolutionary changes of amino acids J. Mol. Biol. 2007 373 801 810 10.1016/j.jmb.2007.08.036 17868687 83. Gu X. Maximum-likelihood approach for gene family evolution under functional divergence Mol. Biol. Evol. 2001 18 453 464 10.1093/oxfordjournals.molbev.a003824 11264396 84. Chakraborty A. Mandloi S. Lanczycki C.J. Panchenko A.R. Chakrabarti S. Speer-server: A web server for prediction of protein specificity determining sites Nucleic Acids Res. 2012 40 W242 W248 10.1093/nar/gks559 22689646 85. Datta A.K. Comparative sequence analysis in the sialyltransferase protein family: Analysis of motifs Curr. Drug Targets 2009 10 483 498 10.2174/138945009788488422 19519350 86. Rohfritsch P.F. Joosten J.A. Krzewinski-Recchi M.A. Harduin-Lepers A. Laporte B. Juliant S. Cerutti M. Delannoy P. Vliegenthart J.F. Kamerling J.P. Probing the substrate specificity of four different sialyltransferases using synthetic β-d -Galp-(1 → 4)-β-d -GlcpNAc-(1 → 2)-α-d -Manp-(1 → O) (CH2 )7 CH3 analogues general activating effect of replacing N -acetylglucosamine by N -propionylglucosamine Biochim. Biophys. Acta 2006 1760 685 692 10.1016/j.bbagen.2005.12.012 16439063 87. Dunbrack R.L. Jr. Rotamer libraries in the 21st century Curr. Opin. Struct. Biol. 2002 12 431 440 10.1016/S0959-440X(02)00344-5 12163064

PMC005xxxxxx/PMC5000684.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081287ijms-17-01287Articleβ-Ketoacyl-acyl Carrier Protein Synthase I (KASI) Plays Crucial Roles in the Plant Growth and Fatty Acids Synthesis in Tobacco Yang Tianquan 12Xu Ronghua 3Chen Jianghua 1Liu Aizhong 4*Iriti Marcello Academic Editor1 Key Laboratory of Tropical Plant Resource and Sustainable Use, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Xuefu Road 88, Kunming 650223, China; [email protected] (T.Y.); [email protected] (J.C.)2 University of Chinese Academy of Science, Beijing 100049, China3 College of Life Sciences, Anhui Science and Technology University, Fengyang 233100, China; [email protected] Key Laboratory of Economic Plants and Biotechnology, Kunming Institute of Botany, Chinese Academy of Sciences, Lanhei Road 132, Kunming 650201, China* Correspondence: [email protected]; Tel./Fax: +86-871-6514-042008 8 2016 8 2016 17 8 128721 6 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Fatty acids serve many functions in plants, but the effects of some key genes involved in fatty acids biosynthesis on plants growth and development are not well understood yet. To understand the functions of 3-ketoacyl-acyl-carrier protein synthase I (KASI) in tobacco, we isolated two KASI homologs, which we have designated NtKASI-1 and NtKASI-2. Expression analysis showed that these two KASI genes were transcribed constitutively in all tissues examined. Over-expression of NtKASI-1 in tobacco changed the fatty acid content in leaves, whereas over-expressed lines of NtKASI-2 exhibited distinct phenotypic features such as slightly variegated leaves and reduction of the fatty acid content in leaves, similar to the silencing plants of NtKASI-1 gene. Interestingly, the silencing of NtKASI-2 gene had no discernibly altered phenotypes compared to wild type. The double silencing plants of these two genes enhanced the phenotypic changes during vegetative and reproductive growth compared to wild type. These results uncovered that these two KASI genes had the partially functional redundancy, and that the KASI genes played a key role in regulating fatty acids synthesis and in mediating plant growth and development in tobacco. tobacco3-ketoacyl-ACP synthase Iover-expressiongene silencefatty acid synthesis ==== Body 1. Introduction Fatty acids (FAs) are major components for cell or organelle membrane lipids, and precursors of other significant complex molecules including waxes and cutin. Some FAs are also converted into messenger compounds such as jasmonic acid and phosphatidylinositol that play major roles in certain signal transduction pathways [1,2,3]. Furthermore, FAs are used as substrates for the synthesis of storage lipids (triacylglycerols, TAG), particularly in the cotyledon or endosperm of oilseeds, which are important materials for seed germination and food or energy supply for humankind [4,5]. Thus, understanding the biosynthesis of FAs in plants has significance with regard to careful control of plant development as well as practical implications. During FAs biosynthesis, the first committed step is catalyzed by acetyl-CoA carboxylase (ACCase), which converts acetyl-CoA to malonyl-CoA, and then is condensed by a set of β-ketoacyl-ACP synthases (KASs), resulting in FA chain elongation [6,7,8]. These KASs are crucial for carbon chain condensation and elongation from C4 to C18. In plants, several plastid or chloroplast specific types of KAS, including KASI, KASII, KASIII and KASIV have been characterized in diverse species [9,10,11,12,13,14]. KASIII is responsible for condensing the initial reaction of malonyl-acyl carrier protein and acetyl-CoA, resulting in a C4 FA molecule [15,16]. KASI has high activity when butyryl- to myristyl-ACP (C4:0–C14:0 ACP) is used as the substrate to produce hexanoyl- to palmitoyl-ACP (C6:0–C16:0 ACP), whereas KASII is a key enzyme that catalysis the last condensation reaction of palmitoyl-ACPs to stearoyl-ACPs [17]. KASIV is thought to participate in condensing the medium-chain FA (C10:0 or C12:0 ACP) in certain species such as Cuphea [14,18]. In addition, the mitochondrion-specific mtKAS participating in FA synthesis for forming mitochondrial membranes has been isolated and characterized in Arabidopsis [19,20], but still very little is known in other plants. Among these identified KASs, KASI, KASII and KASIII seem to be essential and exist broadly in plants. The genes encoding KASIII and KASII were extensively identified from various plants such as Spinach oleracea [21], Arabidopsis thaliana [22], Cuphea wrightii [23], Allium porrum [24], Pisum sativum [25], Helianthus annuus [12], Brassica napus [13] and Jatropha curcas [11,26]. In addition, their functions were partially documented in Jatropha curcas, sunflower and rapeseeds [11,12,13,26]. In particular, studies have demonstrated that KASIII was a rate-limiting enzyme in TAG accumulation [15] and KASII could cause significant changes of the FA composition in the conversion of from C16 to C18 in TAG biosynthesis [26]. KASI genes have been isolated from barley [27], groundnut [9] and rice [28] to date, but the functions of KASI in controlling FAs biosynthesis and regulating plant growth and development remained unclear in plants. Until recently, Wu and Xue demonstrated the functional characterization of KASI gene in Arabidopsis, and uncovered that AtKASI was not only crucial in controlling TAG biosynthesis in both leaf tissues and seeds but also critical in mediating chloroplast formation and division [10]. Besides, the mutants of OsKASI reduced fertility and altered the FA composition and contents in roots and seeds, suggesting that OsKASI is involved in regulating the root development in rice [28]. Apart from studies in Arabidopsis and rice the functions of KASI in controlling FAs biosynthesis and regulating growth and development have not been characterized in other plants. Tobacco (Nicotiana tabacum), as an alternative biofuel plant in recent years, has received a great attention because it possesses potent oil biosynthesis machinery and can accumulate up to 40% oil content in seed. In particular, tobacco leaves have been metabolically engineered as oil-bearing tissues, representing an attractive and promising “energy plant” platform and serving as a plausible system for manufacturing biodiesel production [29,30]. Tobacco oils have been successfully tested for its potential as a fuel for diesel engines [31]. Identification and dissection of key genes encoding the rate-limiting enzymes in FAs and TAG biosynthesis is essential to serve the genetic and metabolic engineering of tobacco for manufacturing oil production. In this study, two tobacco genes encoding KASI were isolated, and their function in mediating FAs and TAG biosynthesis as well as plant growth were characterized. Results obtained in this study provide fundamental and important information for understanding the molecular functions of KASI genes in tobacco. 2. Results 2.1. Identification of 3-Ketoacyl-ACP Synthase I Gene in Tobacco After the assembly of EST fragments, two putative KASI fragments were identified with full-length coding regions, named as NtKASI-1 and NtKASI-2, respectively. Subsequently, two full-length cDNA sequences of NtKASI-1 and NtKASI-2 were confirmed by the RT-PCR sequencing and were submitted to GenBank (KX033513 and KX033514). NtKASI-1 and NtKASI-2 contain a complete ORF (open reading frame) with 1410 bp and 1404 bp, encoding 469 and 467 amino acids, respectively. These two genes shared high similarities on nucleotide sequence (85%) and amino acid (88%) levels. Besides, there was approximately 83% amino acid sequence identity between tobacco NtKASI-1 and Arabidopsis KASI (AtKASI), and 81% sequence identity between tobacco NtKASI-2 and AtKASI. These results showed that tobacco KASIs may have similar functional roles to AtKASI. Multiple sequence alignments of amino acid sequences of KASI proteins from different plants revealed that KASIs were highly conserved in plants (Figure 1A). Some key functional sites were identified such as a substrate-binding cysteine (C) residue, two histidines (H) required for the decarboxylation, an essential lysine of uncertain function, one glycine (G) residue that allows entrance into the substrate-binding tunnel, and two threonine (T) residues that form hydrogen bond with the ACP phosphopantetheine moiety. In the C-terminal region, a conserved Gly-rich motif was also found that may act as forming oxide anion free radical [32]. The phylogenetic tree clearly demonstrated that NtKASI-1 and NtKASI-2 were clustered with two tomato KASIs (Solanum lycopersicum, which belongs to the same family as tobacco), respectively (Figure 1B), implying that NtKASI-1 and NtKASI-2 might have an independent evolution before the species differentiation between tobacco and tomato. Together, we isolated two KASI genes in tobacco with high sequences similarity, and whether these two genes have functional redundancy needs to be investigated. 2.2. Expression Patterns of NtKASI-1 and NtKASI-2 To gain insight into the possible roles of NtKASI-1 and NtKASI-2 in tobacco, we assayed their expression profiles in different tissues using a qRT-PCR technique. The results showed that both NtKASI-1 and NtKASI-2 were constitutively expressed in all tissues tested (Figure 2). It seemed that these two genes had nearly equal expression levels in stem, root, sepal and seed with low transcription abundance. Besides, NtKASI-1 exhibited higher expression level than NtKASI-2 in pistil, stamen and petal, whereas NtKASI-2 was highly expressed in leaf relative to NtKASI-1. In sum, we found that KASI genes, particular for NtKASI-1 gene, were highly transcribed in floral organ compared with vegetative tissues, implying their important function in reproductive stage. 2.3. Phenotypes of Over-Expression of NtKASI Genes in Tobacco Generally, over-expression of a gene is considered as a genetic tool to dissect the gene function. Here, we constructed two binary plant transformation vectors harboring NtKASI-1 and NtKASI-2 gene, respectively, with a cauliflower mosaic virus (CaMV) 35S promoter, and transformed them into wild type (WT) tobacco, respectively. The leaves from T0 transformed plants were screened for successfully over-expressed lines via a hygromycin gene-specific primer PCR (Table S1). The seeds from transgenic T0 plants were collected and germinated in a medium with hygromycin selection. The confirmed over-expression lines in T1 generation for these two genes were designed as KASI-1OE lines (for NtKASI-1 over-expressed) and KASI-2OE lines (for NtKASI-2 over-expressed), respectively. Subsequently, the expression levels of NtKASI genes were examined and the phenotypic changes were investigated for each over-expressed plant. As shown in Figure 3, the transcript levels of KASI-1 in KASI-1OE transgenic lines were at least nine-fold (9–25-fold) higher than that of the WT plants, and KASI-2 in KASI-2OE lines were at least eight-fold (8–20-fold) higher than that of the wild-type. Moreover, the expression levels of NtKASI-1 were not affected in KASI-2OE lines and vice versa. Morphologically, all KASI-1OE lines grew much better, for example, there was a higher plant height than WT plants (Figure 4A–C). Interestingly, we found the opposite phenotype in KASI-2OE lines, such as variegated leaves and slightly dwarf (Figure 4A–C). We speculated that NtKASI-1 or NtKASI-2 transcript levels may be down-regulated in KASI-2OE lines via an RNAi way due to high sequence similarity between two KASI genes. To test this, the expression level of these two genes was subjected to Northern blot in all over-expressed lines and WT. We found that all the KASI-2OE plants displayed significantly high expression levels of the NtKASI-2 gene compared to the WT plants, but the transcript level of KASI-1 in leaf was not changed, consistent with the qRT-PCR result (see Figure 4D,E). These results showed that NtKASI-1 gene can boost tobacco growth. However, for the KASI-2OE lines, the mechanism underlying the phenotypic changes remains unknown. 2.4. Silencing of NtKASI Genes in Tobacco The gene silence or mutant is a classical genetic approach for exploring gene functions that cause a phenotype of interest. In the current study, we performed the RNA interference (RNAi) to obtain gene-silenced plants. As previously described, these two NtKASI genes had high similarities, but in the N-terminal region the sequences were more variable. Therefore, partial sequences from the 5′ of NtKASI-1 (155 bp) and NtKASI-2 (162 bp) ORF, where NtKASI-1 and NtKASI-2 have the greatest sequence difference, were cloned into the vector in an inverted repeat orientation to create RNAi constructs, namely pCXSN-KASI-1 RNAi (kasI-1 RNAi) and pCXSN-KASI-2 RNAi (kasI-2 RNAi), respectively. To obtain double silencing lines of NtKASIs (kasI-1/2 RNAi) in tobacco, we generated the construct using the identical sequence between NtKASI-1 and NtKASI-2 gene. In the T0 generation, the silencing plants exhibited differentially phenotypic changes. For instance, 13 out of 20 independently transformed kasI-1 RNAi lines have kindly variegated leaves and semi-dwarf, whereas all kasI-2 RNAi plants of T0 generation were morphologically and developmentally similar to wild type. Furthermore, most of the double silence (kasI-1/2 RNAi) plants showed more obvious phenotypic changes and maldevelopment, such as severe variegated leaves, semi-dwarf and decreased seed production, and even death. Next, we collected the seeds from the NtKASIs silence plants and T1 generation seedlings were obtained for subsequent analysis. Similarly, the expression levels of NtKASI-1 and NtKASI-2 were examined. The results showed that the expression of NtKASI-1 was significantly decreased (approximately 70%) in kasI-1 RNAi lines, but there was a little decrease for the NtKASI-2 gene (approximately 20%). In Si-2 lines, the expression of NtKASI-2 exhibited a marked reduction (approximately 80%), but the NtKASI-1 expression was not changed. In the double silencing lines, NtKASI-1 (decreased approximately 85%) and NtKASI-2 (decreased approximately 80%) were significantly co-silenced (Figure 5). Similar to phenotypes observed in the T0 plants, both kasI-1 RNAi and kasI-1/2 RNAi plants of T1 generation showed mildly variegated leaves in early stage (Figure 6A). During the leaf development, kasI-1/2 RNAi lines exhibited more serious growth defects such as loss of apical dominance, highly stunted stems and curled leaves, while kasI-1 RNAi showed mildly variegated leaves and stunted stem relative to WT plants (Figure 6B–D). However, the growth and development of kasI-2 RNAi plants was not affected compared to WT plants. 2.5. NtKASI Genes Affected the Chloroplast Development It seems that the phenotype changes in all transgenic leaves were related to the chloroplast development. Thus, we investigated the chlorophyll (a and b) content in the leaves of the transgenic tobacco. We found that the content of chlorophyll a and b in transgenic lines of kasI-1 RNAi, kasI-1/2 RNAi and KASI-2OE lines was significantly decreased compared to the WT plants (Figure 7A), consistent with the variegated leaf phenotype, whereas the KASI-1OE and kasI-2 RNAi lines with normal development of leaf had similar levels of chlorophyll content relative to WT (Figure 7A). Further observation of chloroplast development in mesophyll cells using microscopy showed the presence of much fewer chloroplasts in the chlorotic sector of KASI-2OE, kasI-1 RNAi and kasI-1/2 RNAi leaves compared with many chloroplasts in wild type (Figure 7B). Consistent with the normal leaf phenotype, KASI-1OE and kasI-2 RNAi lines have no obvious defect in chloroplast development. These results showed that NtKASI genes may be involved into the chloroplast development. 2.6. NtKASI Genes Change Fatty Acids Composition in Tobacco Leaf To determine whether tobacco NtKASIs affect the FAs biosynthesis, the total FA content and composition in leaves from all transgenic lines were examined. In over-expressed lines, we found that KASI-1OE plants did not result in the obvious increase of the total FA content compared to WT. As shown in Table 1, the content of FAs extracted from WT and KASI-1OE leaves was approximately 0.96 mg·g−1 and 1.03 mg·g−1 of fresh weight, respectively. However, the FA content in the KASI-2OE plants was remarkably reduced to 0.76 mg·g−1 of fresh weight, similar to transgenic silence plants (0.76 mg·g−1 in kasI-1/2 RNAi), resulting in a decrease of 21% relative to WT plants. Besides, the kasI-1 RNAi and kasI-2 RNAi lines also showed a significant decrease of total FA content in leaves. We further performed the gas chromatographic (GC) analysis for detecting the FA profiles in all transgenic tobacco and WT leaves. In this study, we classified the FA species into three families according to the length of the carbon chain, namely medium-chain FAs (10C–14C), long-chain FAs (16C–18C) and very-long-chain FAs (>18C). Interestingly, although the FAs content was not changed notably in KASI-1OE lines, the FA composition was altered by the decrease of medium-chain and very long-chain FAs proportions and an increase of long-chain FAs proportions compared to WT plants. A similar change in FA profile was also observed in KASI-2OE lines. By the contrast, kasI-1/2 RNAi lines showed a significant increase in medium-chain FAs from 11.09% to 16.96%, and the reduction of long-chain FA proportions from 85.6% to 78.34%. Accompanied by these changes, all transgenic lines showed a dramatic change in very-long-chain FAs. The unsaturated FAs to saturated FAs ratio (US/S) showed a most significant reduction in kasI-1/2 RNAi. Taken together, our findings showed that the silencing of NtKASI decreases the total FA content while increases the medium-chain FAs ratio. 2.7. NtKASI Genes Affect the Seed Weight and Lipid Content In T1 plants, the transgenic lines showed phenotypic differences during the reproductive growth stage. In particular, due to the serious maldevelopment of kasI-1/2 RNAi lines during vegetative growth stage only about 20% of the transgenic plants produced flowers in the end. The kasI-1/2 RNAi plants exhibited significantly low fruiting rate with a mean flower number of 2 compared to WT lines with a mean flower number of 9 (Figure 8). The KASI-2OE (mean flower number of 6) and kasI-1 RNAi lines (mean flower number of 3) also exhibited relatively low fruiting rates, whereas the KASI-1OE and Si-2 lines showed a non-significant difference in fruiting rate compared to WT plants (Figure 8). We then examined whether seed weight, oil content and FA composition were changed in all transgenic lines. As shown in Figure 9A, measurements of seed weight revealed that the NtKASI-1 over-expressing lines had significantly higher thousand-seed weights than WT plants. In contrast, kasI-1/2 RNAi double silence lines exhibited markedly lower thousand-seed weights than WT plants. Other three transgenic lines (kasI-1 RNAi, kasI-2 RNAi and KASI-2OE) had no obvious difference in seed weights (Figure 9A). The lipid content in seeds of wild plants and various transgenic lines was also measured. Similar to seed weights, the KASI-1OE lines seeds had high levels of lipid content, whereas the kasI-1/2 RNAi seeds had low levels of lipid content relative to WT lines. Contents of total FA in the kasI-1 RNAi, kasI-2 RNAi and KASI-2OE lines seeds had no obvious changes compared to WT seeds (Figure 9B). As to FA composition, levels of all FA species from 16C–22C showed a non-significant change in all transgenic lines (Table S2). These results indicated that NtKASI genes significantly influenced the tobacco seed weight as well as lipid content, particularly for NtKASI-1 gene. 3. Discussion In plants, FAs are used for the synthesis of plastid and other cellular membranes in all cells and are also converted for producing various plant hormones, participating in regulating plant growth and development, cell signaling [33,34,35] and stress responses [33,36]. As mentioned above, Wu and Xue [10] firstly characterized the functions of KASI in regulating FAs biosynthesis and found that it affects multiple developmental processes such as altered chloroplast division and suppressed embryo development in Arabidopsis. Similarly, the mutant of rice KASI (OsKASI) also resulted in a remarkable change in fatty acid (FA) composition and contents and reduced fertility. In addition, OsKASI also is involved in regulating the root development in rice [28]. Generally speaking, the functions of KASI have major roles in controlling FAs biosynthesis and regulating growth and development in plants. In the current study, we isolated two NtKASI orthologs from Nicotiana tabacum that were highly conserved among various plants. Similar to their functions in Arabidopsis, the knock-down NtKASI-1 (kasI-1 RNAi lines) plants exhibited the variegated leaves and semi-dwarf phenotype. Moreover, the knock-down of both NtKASI-1 and NtKASI-2 (kasI-1/2 RNAi line) caused more seriously variegated leaves and a significant decrease in chlorophyll content, resulting in the vegetative stage maldevelopment compared to the kasI-1 RNAi lines and WT plants. These results imply that there is a partial functional redundancy between NtKASI-1 and NtKASI-2, and particularly NtKASI-1 has a stronger role in regulating growth and development in tobacco. In Arabidopsis, the mutant of KASI suppressed the expression of FtsZ and Min system genes, resulting in abnormal development of chloroplast division [10]. This implied that the similar mechanism of KASI may be present in tobacco leaf and needs to be investigated. In addition to abnormal vegetative growth, the NtKASI kasI-1/2 RNAi lines caused the reduction of plant height, seed weights and fertility during the reproductive stage similar to the mutation in OsKASI [28]. The higher expression of NtKASI genes in floral organs, especially pisitil and petal, strongly implied that NtKASI involved in regulating the flower development. Thus, it seems to be not difficult in an understanding of the flower development defect in silenced NtKASI lines. Besides, many evidences have shown that lipid genes involved in FA biosynthesis affected the vegetative and reproductive growth of plants. For instance, the knock-out of genes FATB (acyl-ACP thioesterases B), a major regulator for controlling saturated FAs fluxes, suppressed the rosettes size and delayed the bolting in Arabidopsis [1]. Interestingly, over-expressed NtKASI-2 plants exhibited a similar phenotype with the silenced NtKASI-1. Initially, we proposed that the over-expressed NtKASI-2 might result in the reduction of NtKASI-1 or NtKASI-2 expression via post-transcriptional gene silencing. However, this assumption should be ruled out because there was no obvious change of both NtKASI-1 and NtKASI-2 gene expression levels, which was confirmed by qRT-PCR and Northern blot tests. Thus, the potential mechanism of an NtKASI-2 gene in the control of the plant growth and development remains unknown in tobacco. The deficient NtKASI genes result in dramatic changes of FA content and profiles in tobacco leaves, consistent with the observation in Arabidopsis [10]. In leaf, the medium-chain FAs (10C-14C) in deficient NtKASI-1 and kasI-1/2 RNAi lines have a significant increase, suggesting that the elongation of FA chains be affected by deficient NtKASIs. In addition, the reduction in the content of unsaturated fatty acids in the kasI-1/2 RNAi line (see Table 1) means that NtKASI-1 and NtKASI-2 might participat in the processes of FA desaturation. The reduction of unsaturated/saturated FAs ratio (US/S) may be related to the variegated leaf phenotypes in kasI-1 RNAi and kasI-1/2 RNAi lines. Usually, the de novo FA synthesis provides FA sources for the normal membrane formation, which is necessary for maintaining the normal cell growth and development. The changes of unsaturated/saturated FAs ratios in leaves may be the potential cause of developmental abnormalities observed in the variegated and curled leaves and reduction in chloroplast division. Higher levels of saturated FAs may cause the formation of semi-crystalline gels in cell membranes, which impair the cell permeability thereby causing the leaf maldevelopment [37]. Besides, there are evidence showing that very long chain fatty acids (>C18) usually provide FA sources for maintaining plant growth, cell expansion and ethylene biosynthesis and signaling [2,38]. In addition, the leaf variegated in the kasI-1 RNAi and kasI-1/2 RNAi lines might be related to the increase proportion of very long chain fatty acids in leaf FA species. The over-expressed NtKASI-1 plants (in KASI-1OE line) slightly but not significantly increased the FA content in leaves, implying that some other factors are also required for enhancing FAs accumulation. As mentioned above, FAs synthesis is a complex process involving diverse enzymes and regulators, in particular, the rate limiting enzyme acetyl-CoA carboxylase responsible for the synthesis of malonyl-CoA [39,40]. Seed oils have been regarded as the potential feedstock for chemical industries and biodiesel production due to the raised concern of developing renewable and environment-friendly alternatives for crude oil [4,5]. As the initial carbon chain condensation enzyme of the de novo biosynthesis of FAs, KAS enzymes are indispensable for FA chain elongation. Usually, the over-expressed KAS genes positively influence the subsequent TAG assembly and oil content. Here, we also evaluated the specific roles of NtKASI-1 and NtKASI-2 in TAG biosynthesis in tobacco seeds. The results showed that the over-expressed NtKASI-1 significantly enhanced oil content and seed weight, but the silenced NtKAkasI-1/2 RNAi genes significantly decreased the oil content and seed weights, clearly suggesting that NtKASI-1 might play a critical role in regulating TAG accumulation in tobacco seeds. In addition, we noted that the silenced NtKASI-1 (kasI-1 RNAi lines) did not exhibit a significant reduction in the oil content and seed weight. Probably, this is related to the functional compensation of NtKASI-2 in the silenced NtKASI-1 plants. Besides, the over-expressed NtKASI-1 (KASI-1OE line) obviously promoted plant vegetative growth, which might provide more photosynthate feedstock in leaf for TAG biosynthesis in seeds. In addition, the FA profiles of seed oils among the transgenic lines and WT control did not display a significant difference, unlike FA composition’s dramatic change in leaves. Although the potential reasons remain uncertain, our current results are largely consistent with previous reports in which modifying the expression levels of genes involved in de novo fatty acid biosynthesis did not exert a significant change in fatty acids composition in seed oils [41,42,43]. 4. Materials and Methods 4.1. Isolation and Sequence Analysis of Tobacco KASI Genes In order to isolate the putative KASI genes in tobacco, we performed a BLASTN search using the CDS sequence of Arabidopsis KASI (At5g46290) in all tobacco EST (Expression Sequence Tag) database (NCBI; http://www.ncbi.nlm.nih.gov/dbest/). The fragments with high similarity score were collected and assembled. The full-length fragments of tobacco KASI genes were further confirmed by RT-PCR sequencing. Briefly, tobacco (Nicotiana tabacum) young leaves were harvested for RNA extraction using TaKaRa RNA extraction kit (TaKaRa, Dalian, China) with the manufacturer’s protocol. The first strand cDNA was synthesized from 1 μg of total RNA using a PrimeScriptTM RT-PCR Kit (TaKaRa, Dalian, China). The full lengths of ORF (open reading frame) were amplified using high fidelity PCR TransStart FastPfu DNA Polymerase (TransGen, Beijing, China) with gene-specific primers (see Table S1). The PCR products were cloned into pEASY-Blunt Cloning Vector (TransGen, Beijing, China) and sequenced (shenzhen-BGI, Shenzhen, China). Sequence similarity was analyzed by the BLAST search in GenBank and a multiple sequences alignment in CLASTAL W program [44]. Phylogenetic analysis was conducted using the Neighbor-Joining criteria in MEGA (version 5.0) [45]. Branch support of the phylogenetic tree was estimated on the basis of 10,000 bootstrap replicates of the data. 4.2. Vector Construction and Transformation For two KASI genes confirmed, we constructed two plant over-expression PCXSN vectors [46] using ORF sequence, respectively. Meanwhile, three RNAi transcriptional silencing vectors were designed using partial sequences of each KASI in an inverted-repeat fashion to create each gene suppressed (RNAi-KASI-1, RNAi-KASI-2) and both genes suppressed (RNAi-KASI-1/2) transformants. The primers used are listed in Supplementary Table S1. The authenticities of all recombinant vectors were verified by sequencing. Then, all vectors were transferred into tobacco using the Agrobacterium-mediated tobacco (Nicotiana tabacum) leaf disc transformation method [47] with a hygromycin (20 mg/L) selection. Transgenic individuals obtained were grown in a phytotron with 25 °C under a switch of 16-h-light/8-h-dark. The transgenic individual was confirmed by PCR with hygromycin-specific primers (see Table S1), and was moved to the greenhouse for further growth. 4.3. Expression Analyses of KASI Genes in Wild-Type and Transgenic Tobacco To inspect the expression profiles of NtKASI-1 and NtKASI-2 in wild-type (WT) tobacco, the different tissues including root, leaf, stem, pistil, stamen, sepal, petal tissues and developing seeds were collected. Total mRNAs were extracted from these tissues and were reversely transcripted using PrimeScriptTM RT reagent Kit with gDNA Eraser (TaKaRa, Dalian, China). Real-time PCR analysis was performed according to the SYBR Premix Ex TaqTM (Tli RNaseH Plus, TaKaRa, Dalian, China) manufacturer’s manual. For transgenic tobacco plants (T0), the qRT-PCR was also performed to test the expression levels of KASI genes using the mRNA from a young leaf. Quantitative RT-PCR was performed using the CFX96 machine (Bio-Rad, Hercules, CA, USA) following the manufacturer’s instructions. The amplification program was 95 °C for 10 s and 56 °C for 20 s. The relative quantification of each sample was determined by normalization to the amount of NtActin cDNA detected in the same sample. Primers used for quantitative RT-PCR were listed in Supplementary Table S1. Further, Northern blot analysis was performed on a 1.2% (w/v) agarose gel to confirm the gene expression levels of KASI genes in transgenic plants. Ethidium bromide staining was used to ensure the equal loading. The mRNA was then transferred to Hybond N+ nylon membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA) and fixed by drying at 80 °C for 1 h. mRNA gel blots were hybridized with DIG-dUTP-labeled probe prepared from the cDNA of the tobacco KASI genes using Roche PCR DIG Probe Synthesis Kit (Roche Cat. No. 1636090) (primer sequences were listed in Supplementary Table S1). Before addition to the filters in the hybridization solution, probes were denatured by dipping in boiling water for 5 min and then in ice. Pre-hybridization was done in 6× SSC (1× SSC is 150 mM NaCl and 15 mM sodium citrate), 5× Denhardt’s solution, 0.5% (w/v) SDS and 100 μg·mL−1 denatured salmon sperm DNA at 50 °C for 30 min, while the hybridization was done in 6× SSC, 0.5% (w/v) SDS and 100 μg·mL−1 denatured salmon sperm DNA at 50 °C for overnight. Afterward, the blot was subsequently washed in 2× SSC, 0.5% (w/v) SDS twice and in 0.5× SSC, 0.5% (w/v) SDS once, then exposed to X-ray film. Hybridization was visualized by autoradiography after exposure. Hybridization was performed using the DIG Northern starter kit (Roche Cat. No. 2039672). The gene-specific primers for detecting transcripts of NtKASI-1 and NtKASI-2 were listed in Supplementary Table S1. 4.4. Morphological Observation of Chloroplast and Measurement of Chlorophylls Chloroplast morphology of mesophyll cells was observed by OlympusBX51 microscope. Seventh and eighth rosette leaves at four months were collected. For all plants, The leaves of the same size were separated by a hole puncher and fixed in 3.5% glutaraldehyde in the dark for 60 min, and then were placed in 0.1 M Na2EDTA, pH 9.0, for overnight to allow the fixative to soften and then incubated at 60 °C with shaking for 2 to 3 h [48]. After that, the tissues were mounted in water, and cells were released by tapping on the cover slip. The chloroplasts were observed directly on the slide. Images were captured with ZEISS Discovery.V12 digital camera. Amounts of chlorophyll a and b were quantified using a simple method. Leaf filaments with known areas are soaked in 80% (v/v) acetone until the color of filaments changed from green to white, and then the supernatant is taken for light absorption measurement [49]. 4.5. Seed Weight Determination and Lipid Analysis To examine seed weight, four replicates of 100 tobacco seeds randomly selected from WT and transgenic lines were weighted. These seeds were dried in open tubes in desiccators for 3 days before weighing and counting. Total lipids were extracted from leaves and mature seeds of different transgenic lines and WT plants according to the method of Bligh and Dyer (1959) [50], and then fatty acid methyl esters (FAMEs) were prepared according to the description by Maisonneuve et al. (2010) [51]. In brief, fatty acids of total lipid were transmethylated with 2 mL of methanol containing 2.5% H2SO4 (v/v) and then heating at 85 °C for 90 min. In addition, we used a heptadecanoic acid as an internal standard at a final concentration of 50 ng·μL−1 for quantification. After cooling, 500 μL of hexane and 2.5 mL of 500 mM Na2SO4 were added. FAMEs were extracted into the hexane phase by vigorous shaking followed by centrifugation at 1500× g for 5 min. FAMEs were quantified by gas chromatography (Agilent 5975 system with an HP-INNOWax column). Individual methyl esters were identified by comparison with standards (Sigma-Aldrich, Shanghai, China). FAMEs and total lipids were calculated by comparing with the heptadecanoic acid methyl ester standard. 5. Conclusions In conclusions, we revealed that the NtKASI genes, particularly NtKASI-1, are crucial for regulating fatty acids synthesis in leaf and seeds, and play a key role in tobacco vegetative and reproductive growth. This work might establish a good basis for further studies on dissecting the functions of KASI genes in regulating oil accumulation in specific plant tissues, serving for modifying oil content and quality via genetic improvement and breeding in oil crops. Acknowledgments We thank Zhen Fang (Xishuangbanna Tropical Botanical Garden, Chinese Academy of Science) for kindly providing the Gas Chromatography and Jianqiang Wu (Kunming Institute of Botany, Chines Academy of Science) for helpful suggestions during conducting the experiments. This study is jointly supported by the Chinese National Key Technology R&D Program (2015BAD15B02) and the National Natural Science Foundation of China (31571709 and 31401421). Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1287/s1. Click here for additional data file. Author Contributions Aizhong Liu and Ronghua Xu conceived and designed research. Tianquan Yang conducted experiments. Tianquan Yang and Jianghua Chen analyzed data. Tianquan Yang, Jianghua Chen and Aizhong Liu wrote the manuscript. All authors read and approved the final manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations FA Fatty Acid ACCase Acetyl-CoA Carboxylase CoA Coenzyme A ACP Acyl Carrier Protein KAS 3-Ketoacyl-Acyl-Carrier Protein Synthase TAG Triacylglycerol GC Gas Chromatographic Figure 1 Multiple sequence alignment and phylogenetic analysis. (A) Comparison of amino acid sequences of KASI from different plants. The letters and red box represent the conserved residues and motif, respectively; (B) The phylogenetic tree of KAS proteins from diverse plants. The tree was generated with neighbor-joining methods and plant KASII proteins were used as the root of the KASI proteins phylogenetic tree. The blue dots represent the KASIs from tobacco. Potri, Populus trichocarpa; Rc, Ricinus communis; Jc, Jatropha curcas; Gorai, Gossypium raimondii; Lus, Linum usitatissimum; Solyc, Solanum lycopersicum; Nt, Nicotiana tabacum; At, Arabidopsis thaliana; Brara, Brassica rapa; Medtr, Medicago truncatula; Glyma, Glycine max; Os, Oryza sativa; Bradi, Brachypodium sylvaticum; GRMZM, Zea may; Phpat, Physcomitrella patents. Figure 2 Relative transcript levels of tobacco KASI genes in different tissues by qRT-PCR. The relative expression was detected in root, flower, the 5th leaf at 60 DAG (leaf1), the 18th leaf at middle stage (leaf2), the seed at 6 DAF (seed1), and the seed at 12 DAF (seed2). Error bars show the standard error with five biological replicates. The expression level of stem was normalized to 1. Figure 3 Relative transcript levels of KASI genes in all over-expressed lines and wild type (WT) tobacco. The expression of NtKASI-1 and NtKASI-2 gene in the 9th and 10th leaves in medium stage of wild type, KASI-1OE lines and KASI-2OE lines via qRT-PCR. The error bars represent the standard error with four independent lines. Figure 4 Phenotypic observation of NtKASI over-expression lines compared to WT plants: (A) phenotype at early-vegetative growth stage in WT, KASI-1OE and KASI-2OE line; (B) the leaf phenotype at early-vegetative growth stage (bar = 3 cm); (C) comparison of the growth at mid-vegetative growth stage (about four-month-old plants, bar = 15 cm); (D) total RNA were loaded on 1.2% agarose gels and stained with EtBr in buffer; and (E) Northern blot analysis for the expression of NtKASI-1 and NtKASI-2 gene in WT (Lane 1), KASI-1OE lines (Lane 2–5) and KASI-2OE lines (Lane 6–9). Figure 5 Relative transcript levels of KASI genes in the silence lines and WT tobacco. The expression of NtKASI-1 and NtKASI-2 gene in the 9th and 10th leaves in medium stage of wild type, kasI-1 RNAi, kasI-2 RNAi and kasI-1/2 RNAi lines using qRT-PCR. The error bars represent the standard error with four independent lines. The asterisk above the bars indicates the significant differences between the silence plants and WT (* p < 0.05, ** p < 0.01). Figure 6 Growth and morphology of Wild-Type (WT) tobacco and kasI-1 RNAi, kasI-2 RNAi, kasI-1/2 RNAi plants: (A) phenotypic observation of two-month-old plants; (B) phenotypic observation of three-month-old plants; (C) phenotypic observation of four-month-old plants (bar = 15 cm); and (D) comparison of leaf morphology between WT plant and kasI-1/2 RNAi plant (three-months-old plant, bar = 3 cm). Figure 7 Measurement of chlorophyll content: (A) content of chlorophyll a and b (mg/dm2 fresh weight) measured in 8th, 9th and 10th leaves at four month after germination. The asterisks above the bar represent significant differences in content of chlorophyll a and b between transgenic lines and WT (** p < 0.01). Error bars represent the standard errors with four biological replications; (B) Morphological observation of chloroplast in leaves of WT and all transgenic lines. Figure 8 Phenotype of reproductive growth. Figure 9 Seed weight and lipid content in tobacco: (A) thousand-seed weights in WT and all transgenic lines; and (B) the lipid contents in WT and all transgenic lines. Error bars represent the standard errors with four biological replications. The asterisk above the bars indicates the significant differences between transgenic plants and WT (** p < 0.01). ijms-17-01287-t001_Table 1Table 1 Analysis of fatty acid methyl esters (FAME) of the total lipid extract from leaves of wild type (WT) and transgenic tobacco. FA Species WT KASI-1OE KASI-2OE kasI-1 RNAi kasI-2 RNAi kasI-1/2 RNAi 10C 6.07 ± 0.48 4.66 ± 0.66 4.33 ± 0.65 6.56 ± 0.40 5.29 ± 0.28 4.39 ± 0.41 12C 3.31 ± 0.35 2.77 ± 0.35 2.44 ± 0.51 3.13 ± 0.75 3.34 ± 0.25 11.06 ± 0.61 14C 1.71 ± 0.56 1.67 ± 0.54 2.43 ± 0.60 1.81 ± 0.67 2.34 ± 0.11 1.51 ± 0.20 Total 11.09 9.10 9.20 11.50 10.97 16.96 ** 16C 11.07 ± 1.30 13.66 ± 1.83 11.66 ± 1.08 11.43 ± 0.18 11.32 ± 0.49 11.97 ± 2.38 16C1 2.80 ± 0.16 2.47 ± 0.70 2.08 ± 0.24 2.93 ± 0.13 2.56 ± 0.34 1.16 ± 0.31 16C2 8.88 ± 1.44 8.34 ± 0.12 8.72 ± 0.50 9.02 ± 0.14 8.66 ± 1.20 6.51 ± 0.23 16C3 2.14 ± 0.22 3.32 ± 0.30 2.90 ± 0.27 2.26 ± 0.16 2.05 ± 0.52 1.10 ± 0.10 18C 2.12 ± 0.22 3.76 ± 0.32 2.98 ± 0.28 2.63 ± 0.15 2.45 ± 0.14 3.07 ± 0.49 18C1 3.73 ± 0.13 4.73 ± 0.24 4.10 ± 0.20 4.19 ± 0.14 4.06 ± 0.19 4.51 ± 0.43 18C2 11.48 ± 1.26 10.49 ± 0.34 12.57 ± 0.25 12.76 ± 0.16 12.36 ± 0.90 14.04 ± 3.55 18C3 43.38 ± 1.34 43.50 ± 1.11 44.35 ± 1.31 38.63 ± 1.95 43.86 ± 1.29 35.98 ± 2.52 Total 85.60 90.27 * 89.36 83.85 87.32 78.34 ** 20C 2.74 ± 0.51 1.45 ± 0.60 ** 1.43 ± 0.22 ** 5.55 ± 0.19 ** 1.71 ± 0.61 * 4.7 ± 0.75 ** US/S 2.7 2.6 3.0 2.2 2.7 1.7 FA content 0.96 ± 0.13 1.03 ± 0.09 0.76 ± 0.05 ** 0.83 ± 0.03 * 0.88 ± 0.07 0.76 ± 0.07 ** Numbers in each column refer to the relative molar ratios of the different FA with the total being 100%. Means and standard deviation of four independent samples are presented; US/S refers to the ratio of unsaturated/saturated FA. FA content was measured by mg/g fresh weight. The asterisk indicates significant difference in fatty acids composition between transgenic plants and WT (* p < 0.05, ** p < 0.01). ==== Refs References 1. Bonaventure G. Salas J.J. Pollard M.R. Ohlrogge J.B. Disruption of the FATB gene in Arabidopsis demonstrates an essential role of saturated fatty acids in plant growth Plant Cell 2003 15 1020 1033 10.1105/tpc.008946 12671095 2. Zheng H.Q. Rowland O. Kunst L. Disruptions of the Arabidopsis enoyl-CoA reductase gene reveal an essential role for very-long-chain fatty acid synthesis in cell expansion during plant morphogenesis Plant Cell 2005 17 1467 1481 10.1105/tpc.104.030155 15829606 3. Roudier F. Gissot L. Beaudoin F. Haslam R. Michaelson L. Marion J. Molino D. Lima A. Bach L. Morin H. Very-long-chain fatty acids are involved in polar auxin transport and developmental patterning in Arabidopsis Plant Cell 2010 22 364 375 10.1105/tpc.109.071209 20145257 4. Durrett T.P. Benning C. Ohlrogge J. Plant triacylglycerols as feedstocks for the production of biofuels Plant J. 2008 54 593 607 10.1111/j.1365-313X.2008.03442.x 18476866 5. Dyer J.M. Stymne S. Green A.G. Carlsson A.S. High value oil from plants Plant J. 2008 54 640 655 10.1111/j.1365-313X.2008.03430.x 18476869 6. Gornicki P. Haselkorn R. Wheat acetyl-CoA carboxylase Plant Mol. Biol. 1993 22 547 552 10.1007/BF00015984 8101104 7. Jackowski S. Rock C.O. Acetoacetyl-acyl carrier protein synthase, a potential regulator of fatty acid biosynthsis in bacteria J. Biol. Chem. 1987 262 7927 7931 3294837 8. Shimakata T. Stumpf P.K. Purification and characterizations of beta-Ketacayl-[acyl-carrier-protein] reductase, beta-hydroxyacyl-[acyl-carrier-protein] dehydarse, and enoyl-[acyl-carrier-protein] reductase from Spinacia oleracea leaves Arch. Biochem. Biophys. 1982 218 77 91 10.1016/0003-9861(82)90323-X 6756317 9. Chi X.Y. Chen M.N. Yang Q.L. He Y.N. Pan L.J. Gao Y. Yu S.L. Isolation and expression analysis of a β-ketoacyl-acyl carrier protein synthaseⅠ gene from Arachis hypogaea L. Legume Genom. Genet. 2010 1 11 17 10. Wu G.Z. Xue H.W. Arabidopsis β-ketoacyl-[acyl carrier protein] synthase I is crucial for fatty acid synthesis and plays a role in chloroplast division and embryo development Plant Cell 2010 22 3726 3744 10.1105/tpc.110.075564 21081696 11. Wei Q. Li J. Zhang L. Wu P.Z. Chen Y.P. Li M.R. Jiang H.W. Wu G.J. Cloning and characterization of a β-Ketoacyl-acyl carrier protein Synthase II from Jatropha curcas J. Plant Physiol. 2012 169 816 824 10.1016/j.jplph.2012.02.006 22424763 12. González-Mellado D. von Wettstein-Knowles P. Garcés R. Martínez-Force E. The role of β-ketoacyl-acyl carrier protein synthase III in the condensation steps of fatty acid biosynthesis in sunflower Planta 2010 231 1277 1289 10.1007/s00425-010-1131-z 20221630 13. Stoll C. Lühs W. Zarhloul M.K. Friedt W. Genetic modification of saturated fatty acids in oilseed rape (Brassica napus ) Eur. J. Lipid Sci. Technol. 2005 107 244 248 10.1002/ejlt.200590021 14. Schutt B.S. Abbadi A. Loddenkotter B. Brummel M. Spener F. β-ketoacyl- acyl carrier protein synthase IV: A key enzyme for regulation of medium-chain fatty acid synthesis in Cuphea lanceolata seeds Planta 2002 215 847 854 10.1007/s00425-002-0803-8 12244451 15. Jaworski J.G. Clough R.C. Barnum S.R. A cerulenin insensitive short chain 3–ketoacyl–acyl carrier protein synthase in Spinacia oleracea leaves Plant Physiol. 1989 90 41 44 10.1104/pp.90.1.41 16666765 16. Clough R.C. Matthis A.L. Barnum S.R. Jaworski J.G. Purification and characterization of 3-ketoacyl-acyl carrier protein synthase III from spinach: A condensing enzyme utilizing acetyl-coenzyme-A to initiate fatty acid synthesis J. Biol. Chem. 1992 267 20992 20998 1328217 17. Murphy D.J. Plant Lipids: Biology, Utilization and Manipulation Blackwell Pubishing Oxford, UK 2005 18. Dehesh K. Edwards P. Fillatti J. Slabaugh M. Byrne J. KAS IV: A 3-ketoacyl-ACP synthase from Cuphea sp. is a medium chain specific condensing enzyme Plant J. 1998 15 383 390 10.1046/j.1365-313X.1998.00218.x 9750349 19. Yasuno R. von Wettstein-Knowles P. Wada H. Identification and molecular characterization of the β-ketoacyl-[acyl carrier protein] synthase component of the Arabidopsis mitochondrial fatty acid synthase J. Biol. Chem. 2004 279 8242 8251 10.1074/jbc.M308894200 14660674 20. Olsen J.G. Rasmussen A.V. von Wettstein–Knowles P. Henriksen A. Structure of the mitochondrial beta-ketoacyl-[acyl carrier protein] synthase from Arabidopsis and its role in fatty acid synthesis FEBS Lett. 2004 577 170 174 10.1016/j.febslet.2004.10.007 15527780 21. Tai H. Jaworski J.G. 3-ketoacyl-acyl carrier protein synthase III from spinach (Spinacia olerecea ) is not similar to other condensing enzyme of fatty acid synthase Plant Physiol. 1993 103 1361 1367 10.1104/pp.103.4.1361 8290632 22. Tai H. Post-Beiyyenmiller D. Jaworski J.G. Cloning of a cDNA encoding 3-ketoacyl-acyl carrier protein synthase III from Arabidopsis Plant Physiol. 1994 106 801 802 10.1104/pp.106.2.801 7991698 23. Slabaugh M.B. Tai H. Jaworski J.G. Knapp S.J. cDNA clones encoding β-ketoacyl-acyl carrier protein synthase III from Cuphea wrightii Plant Physiol. 1995 108 443 444 10.1104/pp.108.1.443 7784524 24. Chen J. Post-Beittenmiller D. Molecular cloning of a cDNA encoding 3-ketoacyl-acyl carrier protein synthase III from leek Gene 1996 182 45 52 10.1016/S0378-1119(96)00472-6 8982066 25. Jones A.L. Gane A.M. Herbert D. Willey D.L. Rutter A.J. Kille P. Dancer J.E. Harwood J.L. β-Ketoacyl-acyl carrier protein synthase III from pea (Pisum sativum L.): Properties, inhibition by a novel thiolactomycin analogue and isolation of a cDNA clone encoding the enzyme Planta 2003 216 752 761 12624762 26. Ye J. Qu J. Bui H.T. Chua N.H. Rapid analysis of Jatropha curcas gene functions by virus-induced gene silencing Plant Biotechnol. J. 2009 7 964 976 10.1111/j.1467-7652.2009.00457.x 19906247 27. Siggaard-Andersen M. Kauppinen S. von Wettstein-Knowles P. Primary structure of a cerulenin-binding beta-ketoacyl-[acyl carrier protein] synthase from barley chloroplasts Proc. Natl. Acad. Sci. USA 1991 88 4114 4118 10.1073/pnas.88.10.4114 2034657 28. Ding W. Lin L. Zhang B. Xiang X. Wu J. Pan Z. Zhu S. OsKASI: A β-ketoacyl-[acyl carrier protein] synthase I: is involved in root development in rice (Oryza sativa L.) Planta 2015 242 203 213 10.1007/s00425-015-2296-2 25893869 29. Vanhercke T. EI Tahchy A. Liu Q. Zhou X.R. Shrestha P. Divi U.K. Ral J.P. Mansour M.P. Nichols P.D. James C.N. Metabolic engineering of biomass for high energy density: Oilseed-like triacylglycerol yields from plant leaves Plant Biotechnol. J. 2014 12 231 239 10.1111/pbi.12131 24151938 30. Andrianov V. Borisjuk N. Pogrebnyak N. Brinker A. Dixon J. Spitsin S. Flynn J. Matyszczuk P. Andryszak K. Laurelli M. Tobacco as a production platform for biofuel: Overexpression of Arabidopsis DGAT and LEC2 genes increases accumulation and shifts the composition of lipids in green biomass Plant Biotechnol. J. 2010 8 277 287 10.1111/j.1467-7652.2009.00458.x 20051035 31. Usta N. Use of tobacco seed oil methyl ester in a turbocharged indirect injection diesel engine Biomass Bioenerg. 2005 28 77 86 10.1016/j.biombioe.2004.06.004 32. Zhang L. Joshi A.K. Hofmann J. Schweizer E. Smith S. Cloning; expression; and characterization of the human mitochondrial β-ketoacyl synthase J. Biol. Chem. 2005 280 12422 12429 10.1074/jbc.M413686200 15668256 33. Kachroo A. Lapchyk L. Fukushige H. Hildebrand D. Klessig D. Kachroo P. Plastidial fatty acid signaling modulates salicylic acid and jasmonic acid mediated defense pathways in the Arabidopsis ssi2 mutant Plant Cell 2003 15 2952 2965 10.1105/tpc.017301 14615603 34. Kachroo A. Venugopal S.C. Lapchyk L. Falcone D. Hildebrand D. Kachroo P. Oleic acid levels regulated by glycerolipid metabolism modulate defense gene expression in Arabidopsis Proc. Natl. Acad. Sci. USA 2004 101 5152 5157 10.1073/pnas.0401315101 15044700 35. Nandi A. Krothapalli K. Buseman C.M. Li M. Welti R. Enyedi A. Shah J. Arabidopsis sfd mutants affect plastidic lipid composition and suppress dwarfing; cell death; and the enhanced disease resistance phenotypes resulting from the deficiency of a fatty acid desaturase Plant Cell 2003 15 2383 2398 10.1105/tpc.015529 14507997 36. Ohlrogge J. Browse J. Lipid biosynthesis Plant Cell 1995 7 957 970 10.1105/tpc.7.7.957 7640528 37. Somerville C. Browse J. Jaworski J.G. Ohlrogge J. Lipids Biochemistry & Molecular Biology of Plants Buchanan B.B. Gruissem W. Jones R.L. Amercian Society of Plant Physiologists Rockville, MD, USA 2000 456 527 38. Qin Y.M. Hu C.Y. Pang Y. Kastaniotis A.J. Hiltunen J.K. Zhu Y.X. Saturated very-long-chain fatty acids promote cotton fiber and Arabidopsis cell elongation by activating ethylene biosynthesis Plant Cell 2007 19 3692 3704 10.1105/tpc.107.054437 17993622 39. Post-Beittenmiller D. Jaworski J.G. Ohlrogge J.B. In vivo pools of free and acylated acyl carrier proteins in spinach. Evidence for sites of regulation of fatty acid biosynthesis J. Biol. Chem. 1991 266 1858 1865 1988450 40. Post-Beittenmiller D. Roughan R.G. Ohlrogge J.B. Regulation of plant fatty acid biosynthesis: Analysis of acyl-CoA and acyl-ACP substrate pools in spinach and pea chloroplasts Plant Physiol. 1992 100 923 930 10.1104/pp.100.2.923 16653077 41. Lee J.M. Lee H. Kang S. Park W.J. Fatty acid desaturases, polyunsaturated fatty acid regulation and biotechnological advances Nutrients 2016 8 23 10.3390/nu8010023 26742061 42. Ruiz-Lopez N. Usher S. Sayanova O.V. Napier J.A. Haslam R.P. Modifying the lipid content and composition of plant seeds: Engineering the production of LC–PUFA Appl. Microbiol. Biotechnol. 2015 99 143 154 10.1007/s00253-014-6217-2 25417743 43. Xu R.H. Yang T.Q. Wang R.L. Liu A.Z. Characterisation of DGAT1 and DGAT2 from Jatropha curcas and their functions in storage lipid biosynthesis Funct. Plant. Biol. 2014 41 321 329 10.1071/FP12388 44. Larkin M.A. Blackshields G. Brown N.P. Chenna R. McGettigan P.A. McWilliam H. Valentin F. Wallace I.M. Wilm A. Lopez R. ClustalW and ClustalX version 2 Bioinformatics 2007 23 2947 2948 10.1093/bioinformatics/btm404 17846036 45. Tamura K. Peterson D. Peterson N. Stecher G. Nei M. Kumar S. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood; evolutionary distance; and maximum parsimony methods Mol. Biol. Evol. 2011 28 2731 2739 10.1093/molbev/msr121 21546353 46. Chen S. Songkumarn P. Liu J. Wang G.L. A versatile zero background T-vector system for gene cloning and functional genomics Plant Physiol. 2009 150 1111 1121 10.1104/pp.109.137125 19403729 47. Horsch R.B. Fry J.E. Hoffmann N. Eicholz D. Rogers S.G. Fraley R.T. A simple and general method for transferring genes into plants Science 1985 227 1229 1231 17757866 48. Pyke K.A. Leech R.M. Rapid image analysis screening procedure for identifying chloroplast number mutants in mesophyll cells of Arabidopsis thaliana (L.) Heynh Plant Physiol. 1991 96 1193 1195 10.1104/pp.96.4.1193 16668319 49. Arnon D.I. Copper enzymes in isolation chloroplasts. Polyphenoloxidase in Beta vulgaris Plant Physiol. 1949 24 1 15 10.1104/pp.24.1.1 16654194 50. Bligh E.G. Dyer W.J. A rapid method for total lipid extraction and purification Can. J. Biochem. Physiol. 1959 37 911 917 10.1139/o59-099 13671378 51. Maisonneuve S. Bessoule J.J. Lessire R. Delseny M. Roscoe T.J. Expression of rapeseed microsomal lysophosphatidic acid acyltransferase isozymes enhances seed oil content in Arabidopsis Plant Physiol. 2010 152 670 684 10.1104/pp.109.148247 19965969

PMC005xxxxxx/PMC5000685.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081288ijms-17-01288ReviewRole of Toll-Like Receptor Signaling in the Pathogenesis of Graft-versus-Host Diseases Tu Sanfang 1Zhong Danli 2Xie Weixin 2Huang Wenfa 2Jiang Yangyang 1Li Yuhua 1*Lemarié Anthony Academic EditorChan Vera Sau-Fong Academic Editor1 Department of Haematology, Zhujiang Hospital, Southern Medical University, Guangzhou 510282, China; [email protected] (S.T.); [email protected] (Y.J.)2 Second Clinical Medical College, Zhujiang Hospital, Southern Medical University, Guangzhou 510282, China; [email protected] (D.Z.); [email protected] (W.X.); [email protected] (W.H.)* Correspondence: [email protected]; Tel.: +86-135-3370-665611 8 2016 8 2016 17 8 128806 4 2016 03 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Graft-versus-host disease (GVHD) and infection are major complications after allogeneic hematopoietic stem cell transplantation (allo-HSCT) and the leading causes of morbidity and mortality in HSCT patients. Recent work has demonstrated that the two complications are interdependent. GVHD occurs when allo-reactive donor T lymphocytes are activated by major histocompatibility antigens or minor histocompatibility antigens on host antigen-presenting cells (APCs), with the eventual attack of recipient tissues or organs. Activation of APCs is important for the priming of GVHD and is mediated by innate immune signaling pathways. Current evidence indicates that intestinal microbes and innate pattern-recognition receptors (PRRs) on host APCs, including both Toll-like receptors (TLRs) and nucleotide oligomerization domain (NOD)-like receptors (NLRs), are involved in the pathogenesis of GVHD. Patients undergoing chemotherapy and/or total body irradiation before allo-HSCT are susceptible to aggravated gastrointestinal epithelial cell damage and the subsequent translocation of bacterial components, followed by the release of endogenous dangerous molecules, termed pathogen-associated molecular patterns (PAMPs), which then activate the PRRs on host APCs to trigger local or systemic inflammatory responses that modulate T cell allo-reactivity against host tissues, which is equivalent to GVHD. In other words, infection can, to some extent, accelerate the progression of GVHD. Therefore, the intestinal flora’s PAMPs can interact with TLRs to activate and mature APCs, subsequently activate donor T cells with the release of pro-inflammatory cytokines, and eventually, induce GVHD. In the present article, we summarize the current perspectives on the understanding of different TLR signaling pathways and their involvement in the occurrence of GVHD. TLRsGVHDPAMPsHSCT ==== Body Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is currently the only radical therapy for a wide range of benign and aggressive hematopoietic malignancies. However, the approach has limited efficacy due to lethal complications such as graft-versus-host disease (GVHD) and infection, the leading causes of morbidity and mortality in HSCT patients [1]. GVHD is considered as an accentuated inflammatory response triggered by donor T cells, and the immune system plays an essential role in the pathogenesis of GVHD [2]. Increasing evidence shows that the innate immune system bridges the association between infection and GVHD [3]. Pattern-recognition receptors (PRRs) play a key role in the innate immune system by sensing pathogen-associated molecular patterns (PAMPs) and danger-associated molecular patterns (DAMPs), and subsequently initiating immune responses. Among the innate immune receptors, PRRs, two families have been acknowledged as key molecules in the pathogenesis of GVHD: membrane-bound receptors termed Toll-like receptors (TLRs) and cytoplasmic nucleotide oligomerization domain (NOD)-like receptors (NLRs) which are localized in the cytoplasmatic compartment [4,5]. Most PRRs are present on host and donor antigen-presenting cells (APCs), including dendritic cell (DC) subsets, neutrophils, monocytes, macrophages and natural killer cells, while some are distributed on epithelia and organ parenchymal cells. Generally, PAMPs are small molecular motifs conserved within a class of microbes and are considered as exogenous microbial molecules including lipopolysaccharide (LPS), RNA, flagellen, etc., while DAMPs are associated with endogenous cell components that are released during cell damage or death, including high mobility group box 1, s100 proteins, elastase inhibitors (endogenous proteases inhibitors), defensins, cathelicidins, heat shock proteins, heparan sulfate proteoglycans, etc. [6,7]. Patients undergoing conditioning regimens including intensive chemotherapy and/or total-body irradiation (TBI) are susceptible to aggravated intestinal tissue damage caused by microflora [8,9], with subsequent translocation of gut microbes and microbial products, especially PAMPs, and the release of endogenous DAMPs. It leads to the production of pro-inflammatory cytokines and the recruitment of allo-reactive donor T cells to target tissues or organs, including the skin, gastrointestinal tract and liver, which are under long-term exposure to PAMPs [10]. In summary, infection can induce the occurrence and progression of GVHD, which is associated with the innate immune system. The innate immune system enhances adaptive immunity by activating donor T cells to produce massive cytokines, and then triggers or intensifies GVHD. Thus, it is essential to understand the role of innate immunity in the occurrence of GVHD for new approaches for prevention and therapies of GVHD. In the present article, we summarize the current research supporting the involvement of different TLR signaling pathways in the pathogenesis of GVHD. 1. Introduction to Toll-Like Receptors (TLRs) TLRs constitute a family of transmembrane innate PRRs that are broadly expressed by both non-hematopoietic and hematopoietic cells [11]. In mammals, TLRs are present on various immune cells, including natural killer cells, DC subsets, monocytes, macrophages, neutrophils, T lymphocytes and B lymphocytes. Some non-hematopoietic epithelial, endothelial and organ parenchymal cells also express these molecules. TLRs, as evolutionarily conserved molecules, were first described in vertebrates as proteins homologous to the insect molecule Toll, which could activate the secretion of antimicrobial peptides in Drosophila melanogaster [12]. Thirteen TLRs (TLR1 to TLR13) have been identified in both humans and mice, and various equivalent forms of these receptors have been discovered in other mammalian species. However, equivalent forms of certain TLRs found in humans are not present in all mammals [13]. For example, it has been found that mice express TLR11, TLR12 and TLR13, but none of them is represented in humans. Other non-mammalian species, for instance the Takifugu pufferfish, have been found to express TLR14 which cannot be found in mammals [13]. TLRs are now identified as key molecules that alert the immune system to the presence of microbial infections through signal transduction. TLR1 (functioning in TLR1/2 heterodimers), TLR2 (receptor for glycolipids, lipoteichoic acid, bacterial lipoprotein and components of mycobacterial walls), TLR4 (receptor for LPS, respiratory syncytial virus F protein, glycoinositolphospholipids, heat shock proteins and fibrinogen), TLR5 (receptor for bacterial flagellar protein), TLR6 (functioning in TLR1/2 heterodimers), TLR10 (receptor for undefined PAMPs) and TLR11 (receptor for uropathogenic bacteria and their products as well as Toxoplasma gondii) are cell surface receptors that primarily recognize bacterial structures. In contrast, TLR3 (receptor for double-stranded RNA and synthetic polysaccharides), TLR7 (receptor for single-stranded RNA), TLR8 (receptor for single-stranded RNA and imidazoquinolone, an antivirus drug) and TLR9 (receptor for unmethylated cytosine-phosphorothioate-guanine oligodeoxynucleotides (CpG-ODN)) are mainly localized intracellularly and recognize nucleic acids of microbial or viral origin. All TLRs can activate the nuclear factor-κB (NF-κB) signaling pathway, subsequently upregulating the expression of adhesion molecules and cytokines, and eventually leading to inflammatory responses [11,14,15]. The inflammatory cascades initiated by innate immune responses contribute to the occurrence of GVHD [7]. Besides, TLRs can intensify GVHD by subsequently inducing adaptive immune responses by recognition of exogenous microbial pathogens. In detail, TLRs expressed on recipient APCs can recognize and interact with the intestinal flora’s PAMPs, leading to the activation and maturation of APCs, which subsequently results in T cell migration or trafficking with an increasing production of inflammatory cytokines, including tumor necrosis factor-α (TNF-α), interleukin (IL)-1β, IL-6, IL-12 and interferon-γ (IFN-γ), which are all relevant to the activation of NF-κB signaling. In particular, IFN-γ can enhance antigen presentation by upregulating the expression of major histocompatibility complex (MHC) molecules on lymphoid and non-lymphoid tissues [16]. The activation of NF-κB signaling triggers the upregulation of MHC class II (MHCII) and costimulatory molecules B7, which present the first and second activation signals, respectively, to donor T lymphocytes for the production of cytokines (including IL-1, IL-6, IL-12, IL-21, IL-22, IL-23, TNF-α and IFN-γ). The cytokines then induce the differentiations of sub-populations of donor T cells, including helper T cell 1 (Th1), helper T cell 2 (Th2), helper T cell 17 (Th17) and regulatory T cell (Treg). Subsequently, allo-reactive donor-derived T cells attack target tissues or organs (including intestine, liver and skin) by direct cytotoxicity, eventually leading to GVHD (Figure 1). The current article reviews recent findings on the potential role of various TLRs in the occurrence and severity of GVHD, and explores new approaches for the prevention and therapies of GVHD. 2. TLR4 Signaling Pathway Activated by LPS LPS, also known as lipoglycan and endotoxin, is microbial breakdown products during pre-transplant conditioning and has been considered as a driver of GVHD in experimental models [15,17]. TLR4 is broadly expressed by immune cells, such as monocytes, macrophagocytes, DC subsets, B cells and T cells, as well as by non-immune cells, such as human skin keratinocytes, embryo kidney cells, bronchial epithelia and intestinal epithelia [14,18]. In addition to recognition of LPS, TLR4 can also be activated by respiratory syncytial virus F protein, glycoinositolphospholipids (Trypanosoma), heat shock proteins and fibrinogen [19]. Since heat shock proteins may be released from dead or dying epithelial cells, in addition to impaired TLR4 signaling in response to PAMPs, there may also be delayed epithelial healing signaling in response to necrotic debris [17,20]. It has been demonstrated that intestinal permeability increases significantly in allo-HSCT patients who have received pre-transplant regimens such as chemotherapy and/or TBI [8]. Disruption of the gut-mucosa barrier perturbs the intestinal bacterial balance, with subsequent translocation of LPS and microorganisms from damaged intestinal mucosa into the circulation. In response to PAMPs, the LPS-mediated TLR4 pathway initiates inflammatory cascades and activates adaptive immune responses by upregulating the expression of costimulatory molecules on host APCs, including lipopolysaccharide-binding protein (LBP), cluster of differentiation 14 (CD14) and myeloid differentiation factor 2 (MD-2). The TLR4 signaling pathway depends on Myeloid Differentiation Primary Response Protein-88 (MyD-88), the downstream events of which are shown in Figure 2 [21,22,23]. Specifically, LPS, the breakdown bacterial component, binds to LBP which localizes on the surface of host APCs. LPS is then released from the LPS-LBP complex and is presented to CD14 and TLR4 on APCs, leading to the activation of TLR4. With the assistance of MD-2, an important component of activated TLR4 termed Toll/IL-1 receptor (TIR) homologous domain, it binds to the C-terminus of cytoplasmic adaptor protein MyD88, while the death domain (DD) at the N-terminus of MyD88 interacts with cytoplasmic enzyme IL-1 receptor-associated kinase (IRAK), eventually triggering the phosphorylation of IRAK and the activation of TNF-α receptor-associated factor 6 (TRAF-6). The phosphorylated IRAK binds to the activated TRAF-6, and then the complex activates TGF-β-activated Kinase-1 (TAK1), triggering the activation of inhibitor of κ polypeptide gene enhancer in B-Cells (IκB) kinase, eventually activating the NF-κB signaling pathway. The activated NF-κB signaling pathway can upregulate the expression levels of target genes. The upregulated MHCII and costimulatory factor B7 present the first and second activation signals, respectively, to donor T cells for the production of cytokines (including IL-1, IL-6, IL-12, IL-21, IL-22, IL-23, TNF-α and IFN-γ). These cytokines then induce the differentiation of sub-populations of donor T cells (Th1, Th2, Th17 and Treg). Subsequently, the activated donor-derived T cells attack target organs by direct cytotoxicity. The LPS-induced TLR4 signaling plays an essential role in the occurrence of GVHD [24,25]. It seems that LPS plays an essential role in the initiation of GVHD by means of TLR4 signaling, but there should be caution when jumping to the conclusion that LPS can affect the progression of GVHD in HSCT patients. Lorenz et al. [26] found that it is tended to have a reduced incidence of grade II to IV acute GVHD (aGVHD) in TLR-mutant patients compared with TLR4-intact patients (33.3% versus 47.2%), but they found that there were no statistically significant results. Elmaagacli et al. [27] demonstrated that in HSCT patients, a trend toward an increased risk of severe aGVHD was associated with TLR4 gene mutations carried by both recipients and donors compared with recipients carrying the wild-type gene. However, the gene mutations are shown to have no effect on the transplant-related mortality, overall survival, and incidence of infectious complications. Due to the routine performance of bacterial gut decontamination in clinical HSCT patients, it may contribute, at least in part, to the observation of a strong effect of TLR4 signaling in mouse models and a weaker effect in human clinical studies. There is growing evidence on the role of TLR4 in the pathogenesis of GVHD, ranging from gene polymorphisms to expression levels. Emerging data show that polymorphisms in the gene encoding TLR4 are relevant to the susceptibility of aGVHD; interestingly, two single-nucleotide polymorphisms (SNPs) of TLR4, Asp299Gly and Thr399Ile, are relevant to the enhanced immune responses and increased genetic risk for aGVHD, but the association between the severity of aGVHD and the SNPs was not statistically significant [27,28]. A much larger study population was needed to confirm the role of TLR4 in the pathogenesis of human GVHD. Furthermore, an increased risk for intestinal GVHD and severe GVHD was proved to be attributed by mutations of the TLR4 (Thr399Ile) gene on both the patient and donor sides [27]. However, some data showed that mutations of TLR4 (Asp299Gly and Thr399Ile) in human leucocyte antigen (HLA)-matched sibling patients contributed to a reduced risk of GVHD, but an increased incidence for Gram-negative bacteremia tended to occur in HSCT patients with such TLR4 mutations [26]. Besides, LPS-induced TLR4 signaling plays a crucial role in the occurrence of chronic GVHD (cGVHD). Compared with non-GVHD patients after HSCT and healthy donor controls, TLR4-mediated NF-κB signaling-related genes including TLR4, NF-κB, IL-6 and intercellular adhesion molecules 1 were significantly increased in patients with cutaneous cGVHD. The possible mechanism is the involvement of inflammation-mediated fibrosis in cutaneous cGVHD, which is mediated by TLR4-mediated NF-κB signaling [29]. On the other hand, the disruption of LPS-induced TLR4 signaling can ameliorate GVHD. The depletion of certain bacteria by treatment with an LPS inhibitor or an anti-endotoxin-neutralizing antibody, such as metronidazole and ciprofloxacin, can remarkably attenuate the detrimental influence on GVHD [30]. This can be explained by the fact that the elimination of LPS following anaerobic bacterial eradication has a negative effect on donor T lymphocyte activation via the TLR4 signaling pathway. In comparison with previous results showing that intestinal GVHD was initiated by TLR4 signaling pathways, a recent survey found that host TLR4 mutations could significantly increase the severity of GVHD, according to the survival and intestinal histopathologic changes in TLR4 mutant host mice [31]. It has been discovered that host TLR4 is critical for the induction of tissue-protective factors including cyclooxygenase-2 (COX-2) and COX-2-derived prostaglandin E2 (PGE2) and for protection against intestinal injuries during aGVHD [31,32]. This hypothesis may improve the strategies in terms of prophylaxis and therapy of aGVHD. Fukata et al. [18] demonstrated that it failed to upregulate COX-2 expression for TLR4-deficient mice in response to intestinal epithelial injury on account of the administration of dextran sodium sulfate. Their study also demonstrated that oral supplementation with PGE2 resulted in increased proliferation and decreased apoptosis of intestinal epithelia in TLR4-deficient mice. Thus, the failure of TLR4 signaling in host GVHD mice may result in the decreased expression of tissue repair factors, such as PGE2 and hepatocyte growth factor, in response to intestinal damage [32,33]. In summary, LPS-induced TLR4 signaling plays a dual role in the pathogenesis of GVHD in that host TLR4 can protect recipients against GVHD while donor TLR4 is essential to the occurrence and development of GVHD. 3. The TLR9 Signaling Pathway As described above, TLR9 localizes within the endosomal compartment of immune cells and intestinal epithelial cells, and is activated by non-methylated CpG-ODN, which is derived from bacterial and viral DNAs such as malarial pigment hemozoin and herpes simplex virus DNA. The TLR9 signaling pathway has long been known to play a critical role in antitumor activity and immune responses; however, the underlying mechanism of the role of TLR9 in GVHD remains poorly understood. Since microbial breakdown products produced by conditioning regimens could include the TLR9 ligand CpG-ODN, the TLR9-mediated signaling pathway may play a critical role in the process of aGVHD [34]. In fact, the activation of the TLR9 signaling pathway promotes the transcriptional induction of genes that are important for host defense, producing Th1 which can promote chemokines and cytokine secretion including macrophage inflammatory protein-1, IFN-10, and other IFN-inducible proteins, thereby facilitating the regulation of antibody-dependent cell-mediated cytotoxicity induced by natural killer cells [16]. Similar to the LPS-induced TLR4 signaling pathway, the TLR9 transduction pathway also engages with the MyD88 adaptor protein. The role of TLR9 gene levels in the occurrence of GVHD has gained attention for many years. Emerging data showed that two SNPs in the human TLR9 gene, T1486C and T1237C, have been confirmed to downregulate the expression of TLR9 mRNA [35]. Duramad et al. found that injection of immunoregulatory DNA sequences, which act systemically to block TLR-9 stimulation, into mice could prevent the systemic inflammatory response syndrome (SIRS) [36], suggesting that the occurrence of SIRS in allo-HSCT patients carrying the TLR9 gene variant is significantly lower than that in patients with the wild-type TLR9 gene. Therefore, it is reasonable to conceive that the TLR9 gene variant can protect patients from CpG-ODN-induced lethal GVHD. However, based on an analysis of TLR9 gene variants in 413 allogenic transplant patients and donors, Elmaagacli et al. [37] claimed that TLR9 SNPs do not affect the occurrence or severity of GVHD. Nonetheless, different results have been found in patients harboring a 1486C variant. A recent survey showed that compared to transplant patients with the wild-type gene, those with a TLR9 gene variant at position 1486C had desirable outcomes, which was attributed to decreased transplant-associated mortality associated with GVHD [27]. Another study found that two tag SNPs at the donor side of the TLR9 gene, t1174 A/G (rs352139) and t1635 C/T (rs352140), influenced the risk of developing aGVHD and cytomegalovirus reactivation [16]. A systematic review or meta-analysis is necessary to solve the dispute of whether the TLR9 gene polymorphism can affect the occurrence or severity of GVHD. In addition, increasing studies are performed to interpret the roles of key molecules involved in TLR9 signaling. It has been demonstrated that a single injection of the TLR9 agonist CpG-ODN greatly upregulated the production of TNF-α, IL-6 and IFN-γ, which was lethal in GVHD mice. In contrast, the blocking of TNF-α, but not IL-6 or IFN-γ, rescued GVHD mice from CpG-induced mortality [38]. Furthermore, the severity of aGVHD was remarkably reduced in TLR9-knockout host mice compared to the control group, with improved survival rates [39,40]. Experiments performed in bone marrow chimeric mice have revealed that TLR9 in the non-hematopoietic system has a significant influence on the outcome of GVHD models, whereas the presence or absence of TLR9 in hematopoietic cells has no effect on GVHD. This phenomenon is reasonable because non-hematopoietic cells, such as intestinal epithelia, express high levels of TLR9, and TLR9 directly participates in antigen presentation during GVHD [40,41]. In addition, the development of cGVHD appears to be associated with high levels of TLR9 expressed by B cells in increasing number, resulting in increased sensitivity to microbe-derived immunostimulatory CpG [42]. 4. Other TLR Signaling Pathways Studies on the understanding of other TLR signaling pathways remain scarce, let alone those on their roles in the pathogenesis of GVHD. Initially, researchers focused on the role of TLR gene polymorphisms. Based on the analysis of 305 HSCT patients, the genetic SNPs rs4833079 in TLR1, rs4837656 and rs17582214 in TLR4, rs10737416 in TLR5, rs6531656 in TLR6, and rs337629 in TLR10 were associated with the occurrence of aGVHD. Interestingly, two SNPs in the TLR5 gene, rs2800230 and rs2800237, were associated with cGVHD [43]. In conclusion, it is reasonable to assume that these TLR signaling pathways are associated with the occurrence of GVHD. Since all TLRs can eventually initiate inflammatory responses by activating the NF-κB signaling pathway, with subsequent production of adhesion molecules and cytokines, it is thereby postulated that immune tolerance induction by TLR can make a difference in the occurrence and severity of GVHD. A new concept stating that frequent exposure to TLR stimulation may induce autoimmunity or tolerance, which is protective by limiting excessive inflammation, has gained attention. In vitro, repeated activation of TLRs induces unresponsiveness to the same TLR ligand in cell lines, B cells and plasmacytoid DCs. Repeated activation of TLR7 by low doses of TLR7 ligand-containing antigens can not only result in cross-tolerance, but also in enhanced responsiveness to other TLR ligands such as TLR2 and TLR9 [44]. Another study noted that following allo-HSCT, mice pretreated with 3M-011 (a TLR7/8 agonist) exhibited delayed GVHD and had significantly lower histological GVHD scores compared with those in the control group. The underlying mechanism is that IFN-γ produced by donor T cells leads to massive upregulation of indoleamine 2,3-dioxygenase (IDO) in host APCs, and IDO contributes to the reduced GVHD lethality [45]. Interestingly, GVHD inhibition was achieved by administering the TLR7/8 agonist before BM transplantation, in contrast to the GVHD acceleration with post-BM transplantation TLR agonist administration, with the latter likely caused by the release of overwhelming pro-inflammatory cytokines with inadequate control by IDO [46]. Thus, TLR7 signaling plays dual roles in local or systemic inflammatory responses. A TLR7 agonist can both induce the occurrence of GVHD and reduce the severity of GVHD. In addition, patients with GVHD have significantly increased expression of TLR5 mRNA, with the major TLR5 producers Lin-HLADR-CD33+CD16+ cells and CD14+CD16− monocytes [47]. In conclusion, these data increase our understanding of the mechanism of different TLR signaling pathways in the occurrence and severity of GVHD, which can provide new insights for developing new clinically applicable therapeutic strategies to prevent GVHD in allo-HSCT patients. 5. Inhibition of TLR Signaling Pathways The depletion of donor T lymphocytes or the inhibition of proliferative T lymphocytes can decrease the occurrence and severity of GVHD; however, there is a great risk of graft failure, a reduced graft-versus-leukemia (GVL) effect and an increased incidence of leukemic relapse, as well as an increased risk of severe infection. Therefore, an effective strategy for reducing GVHD without impairing the GVL effects or infection propensity is important. Emerging data showed that donor TLR4 and MyD88 deficiencies are protective against aGVHD [48], while activation of TLR9 with CpG-ODN in recipients markedly accelerates GVHD lethality [38,39]. Hence, novel therapeutic strategies that target TLR-mediated signaling pathways may be able to reduce the occurrence and severity of GVHD [49]. There is growing research in discovering new approaches relevant to TLR signaling pathways to intervene in the process of GVHD. LPS induces the release of massive inflammatory cytokines into the serum through the TLR4 signaling pathway with macrophage priming, thus leading to GVHD [50]. Conversely, antagonism of LPS (B975, a synthetic lipid-A analogue) significantly suppresses serum TNF-α levels and reduces both intestinal damage and systemic GVHD, without altering donor T cell activity toward host antigens in mice after experimental bone marrow transplantation [51]. In the meantime, the serum levels of LPS are relevant to the severity of GVHD in the target organs [45], while GVHD-associated lung injury can be significantly alleviated by treatment with an LPS antagonist [52]. However, the role of TLR4 on the donor side is distinct from that on the recipient side. A deficiency of activated TLR4 in donor hematopoietic cells prevents lung injury, whereas the presence or absence of TLR4 in recipient structural lung tissue has an insignificant effect on lung inflammation after transplantation. In conclusion, the signaling involved in lung injury thus may not be absolutely the same as that in the traditional TLR4 signaling pathway [52]. Heparan sulfate (HS), a ubiquitous component of the extracellular matrix, can activate TLR4 on DCs, leading to the enhancement of DC maturation and allo-reactive T cell responses [53]. Elevated serum HS levels aggravated GVHD both with regard to duration and severity of GVHD. In contrast, therapy with the serine protease inhibitor α1-antitrypsin was able to reduce the serum HS levels, leading to a reduction in donor allo-reactive T cell responses and GVHD severity. In conclusion, HS plays a crucial role in promoting GVHD, and the findings offer a new approach to preventing GVHD by lowering serum HS levels. Besides, Loiarro et al. synthesized the heptapeptide ST2825, which mimics the BB-loop of the MyD88-TIR domain, and found that ST2825 could block TLR signaling by interfering in MyD88 homodimerization [54]. In a co-immunoprecipitation assay, ST2825 inhibited the formation of the MyD88 dimerization by interacting with the TIR domains instead of the DD domains. ST2825 interfered with the recruitment of IRAK1 and IRAK4 by MyD88, attenuating the IL-1β-mediated activation of NF-κB. In addition, ST2825 suppressed the proliferation and differentiation of B lymphocytes in response to TLR9 stimulation. Thus, interference of MyD88 by ST2825 is a new approach to treating GVHD by inhibiting the recruitment of allo-reactive T lymphocytes activated by TLRs. Interestingly, a study demonstrated that in experimental allo-HSCT models, the absence of MyD88 in donor T cells diminishes the GVL effect without attenuating the severity of GVHD [18], which may be contradictory to the fact that all TLRs, except TLR3, can induce GVHD by activating NF-κB signaling through the MyD88-dependent pathway. The activation of several types of adaptive immune responses as well as the generation of effectors including Th1, Th2, Th17 and Treg could be triggered by TLRs. However, cell-type-specific functions of MyD88 signaling remain poorly understood. In fact, in the donor/recipient strain combination (B6 → B6D2F1), donor CD4+ T cells are critical effectors of GVHD, while the GVL effect is dependent on the CD8+ T cells that can mediate cytotoxicity [55,56]. In the early phase after transplantation, recipients of MyD88-deficient T cells are found to have increased levels of Treg and Th2 cells and decreased levels of Th1 cells in the spleen and tumor-draining lymph nodes, but after the lapse of time, diminished Treg and Th2 cells are detected in recipients. This observation may be associated with reducing the GVL effect rather than GVHD-associated tissue damage. In vitro, however, the absence of MyD88 in CD8+ T cells results in defective cytotoxic activity against tumor tissue and a reduced secretion of pro-inflammatory cytokines to host antigens. In conclusion, the cause of maintaining the severity of GVHD with a reduced GVL effect is associated with functional dissociation of the two T cell subsets according to MyD88 deficiency in T cells. The dissociation of the GVL effect from GVHD may be promising in designing new effective therapeutic approaches against hematologic malignancies, which can harness the benefit of the GVL effect and reduce the toxicity of GVHD [18,35,57]. 6. Conclusions Innate immune responses which can be initiated by TLR signaling pathways significantly contribute to inflammatory cascades which lead to the recruitment of allo-reactive donor T cells to target organs, which is equivalent to GVHD. In detail, pre-transplant regimens including radiotherapy and/or chemotherapy lead to aggravated intestinal epithelial damage, with the subsequent dislocation of intestinal bacteria and their breakdown components including LPS, CpG-ODN and other ligands, resulting in the activation of the MyD88-dependent signaling pathway by ligation with TLR4, TLR9 or other TLRs. The consequent activated NF-κB signaling upregulates the expression of target genes to trigger the activation of donor T cells with the production of massive cytokines including IFN-γ, IL-1β, IL-6, IL-12 and TNF-α, leading to the recruitment of allo-reactive donor T cells to target organs. The blocking of any key molecules in TLR signaling pathways can inhibit the occurrence of GVHD, however, often accompanied by some undesirable complication such as reduced GVL and bacteremia. MyD88 in donor CD8+ T cells is proven to be critical for the preservation of GVL activity, regardless of the occurrence of GVHD, which may be a promising therapeutic target. In addition, protective factors in recipients such as COX2, PEG2 and IDO also contribute to reduced GVHD. Indeed, the eradication of intestinal bacteria with antibiotics such as metronidazole and ciprofloxacin has been clinically applied in HSCT patients. Further experiments and clinical trials are needed to interpret the role of TLRs in the induction of GVHD, a prerequisite to exploiting the use of specific TLR agonists or antagonists for treating GVHD. Acknowledgments We wish to acknowledge the great contribution of Yuhua Li regarding her guidance and design of this article and also the contribution of Rui Huang, affiliated to Department of Haematology, Zhujiang Hospital of Southern Medical University regarding her comments and suggestions in this manuscript, for valuable contributions to the work. This research received grants from the National Natural Science Foundation of China (81372249 and 81300431), Science and Technology Planning Project of Guangdong Province, China (2016A020213005), Science and Technology Planning Project of Guangdong Province, China (2013B091500072) and Project of Department of Education of Guangdong Province, China (2014GKXM029). Author Contributions Sanfang Tu designed, carried out the literature and draft the manuscript; Danli Zhong carried out the literature and draft the manuscript; Weixin Xie drew the figures and evaluated the manuscript; Wenfa Huang and Yangyang Jiang evaluated the manuscript; Yuhua Li guided and critically evaluated the manuscript. All authors read and approved the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Schematic graph of GVHD initiated by intestinal flora. Intestinal flora enter the systemic circulation through damaged intestinal mucosa, and then interact with Toll-like receptors (TLRs), leading to the upregulation of major histocompatibility complex class II (MHCII) and costimulatory molecules B7 on host APCs. The upregulation of MHCII and costimulatory molecules B7 present the first and second activation signals, respectively, to donor T cells for the production of cytokines (including IL-1, IL-6, IL-12, IL-21, IL-22, IL-23, TNF-α and IFN-γ). The cytokines then induce the differentiation of a sub-population of donor T cells (Th1/Th2). Subsequently, the activated donor-derived T cells attack target tissues or organs (including intestine, liver and skin) by direct cytotoxicity, and eventually leading to GVHD. PAMPs = pathogen-associated molecular patterns; APCs = antigen-presenting cells; Th1 = helper T cell 1; Th2 = helper T cell 2. Figure 2 Diagram of LPS-induced TLR4 signaling. LPS, the breakdown bacterial component, binds to LBP, and then is released from the LPS-LBP complex, and is presented to CD14 and TLR4 on APCs, leading to the activation of TLR4. With the assistance of MD-2, an important component of activated TLR4, termed Toll/IL-1 receptor (TIR) homologous domain, binds to the C-terminus of cytoplasmic adaptor proteins MyD88, while the death domain (DD) at the N-terminus of MyD88 interacts with intercellular enzyme IL-1 receptor-associated kinase (IRAK), eventually resulting in the phosphorylation of IRAK and the activation of TNF-α receptor-associated factor 6 (TRAF-6). The phosphorylated IRAK binds to the activated TRAF, and the complex activates TGF-β-activated Kinase-1(TAK1), triggering the activation of inhibitor of κ polypeptide gene enhancer in B-cells (IκB) kinase, eventually activating the NF-κB signaling pathway. The NF-κB signaling pathway upregulates the expression levels of the target gene. The activated NF-κB signaling initiates the expression of target genes, leading to the damage of target organs. TLR4 = Toll-like Receptor-4; LBP = lipopolysaccharide-binding Protein; TIR = Toll-Interleukin-1-Receptor; TIRAP = Toll-Interleukin-1-Receptor Domain-containing Adapter Protein; MyD88 = Myeloid Differentiation Primary Response Protein-88; IRAK = Interleukin-1 Receptor-associated Kinase; IRAKM = Interleukin-1 Receptor-associated Kinase-M; TollIP = Toll-Interacting Protein; TRAF6 = Tumor Necrosis Factor Receptor-associated Factor-6; UbC13 = Ubiquitin-conjugating Enzyme-13; UEV1A = Ubiquitin-conjugating Enzyme E2-Variant-1; ECSIT = Evolutionarily Conserved Signaling Intermediate in Toll Pathways; TAK1 = TGF-β-activated Kinase-1; TAB1 = TAK1-binding Protein-1; TAB2 = TAK1-binding Protein-2; IKKs = Inhibitor of κ Light Polypeptide Gene Enhancer in B-Cells Kinase. ==== Refs References 1. Mo X.D. Huang X.J. Life quality related to health related after allogenic transplantation Chin. J. Hematol. 2012 33 968 971 2. Van Bekkum D.W. Knaan S. Role of bacterial microflora in development of intestinal lesions from graft-versus-host reaction J. Natl. Cancer Inst. 1977 58 787 790 14265 3. Poutsiaka D.D. Munson D. Price L.L. Chan G.W. Snydman D.R. Blood stream infection (BSI) and acute GVHD after hematopoietic stem cell transplantation (HSCT) are associated Bone Marrow Transplant. 2011 46 300 307 10.1038/bmt.2010.112 20479711 4. Reikine S. Nguyen J.B. Modis Y. Pattern recognition and signaling mechanisms of RIG-I and MDA5 Front. Immunol. 2014 5 342 10.3389/fimmu.2014.00342 25101084 5. Saxena M. Yeretssian G. NOD-like receptors: Master regulators of inflammation and cancer Front. Immunol. 2014 5 327 10.3389/fimmu.2014.00327 25071785 6. O’Neill L.A. Golenbock D. Bowie A.G. The history of Toll-like receptors—Redefining innate immunity Nat. Rev. Immunol. 2013 13 453 460 10.1038/nri3446 23681101 7. Ramadan A. Paczesny S. Various forms of tissue damage and danger signals following hematopoietic stem-cell transplantation Front. Immunol. 2015 6 14 10.3389/fimmu.2015.00014 25674088 8. Duran-Struuck R. Reddy P. Biological advances in acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation Transplantation 2008 85 303 308 10.1097/TP.0b013e318162d357 18301323 9. Johansson J.E. Ekman T. Gut toxicity during hemopoietic stem cell transplantation may predict acute graft-versus-host disease severity in patients Dig. Dis. Sci. 2007 52 2340 2345 10.1007/s10620-006-9404-x 17415646 10. Wells J.M. Loonen L.M.P. Karczewski J.M. The role of innate signaling in the homeostasis of tolerance and immunity in the intestine Int. J. Med. Microbiol. 2010 300 41 48 10.1016/j.ijmm.2009.08.008 19783476 11. Penack O. Holler E. van den Brink M.R. Graft-versus-host disease: Regulation by microbe-associated molecules and innate immune receptors Blood 2010 115 1865 1872 10.1182/blood-2009-09-242784 20042727 12. Takeda K. Takeuchi O. Akira S. Recognition of lipopeptides by Toll-like receptors J. Endotoxin Res. 2002 8 459 463 10.1179/096805102125001073 12697090 13. Lemaitre B. Nicolas E. Michaut L. Reichhart J.M. Hoffmann J.A. The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults Cell 1996 86 973 983 10.1016/S0092-8674(00)80172-5 8808632 14. Akira S. Takeda K. Kaisho T. Toll-like receptors: Critical proteins linking innate and acquired immunity Nat. Immunol. 2001 2 675 680 10.1038/90609 11477402 15. Kumar H. Kawai T. Akira S. Toll-like receptors and innate immunity Biochem. Biophys. Res. Commun. 2009 388 621 625 10.1016/j.bbrc.2009.08.062 19686699 16. Xiao H.W. Luo Y. Lai X.Y. Shi J.M. Tan Y.M. He J.S. Xie W.Z. Zheng W.Y. Ye X.J. Yu X.H. Donor TLR9 gene tagSNPs influence susceptibility to aGVHD and CMV reactivation in the allo-HSCT setting without polymorphisms in the TLR4 and NOD2 genes Bone Marrow Transplant. 2014 49 241 247 10.1038/bmt.2013.160 24121213 17. Fukata M. Michelsen K.S. Eri R. Thomas L.S. Hu B. Lukasek K. Nast C.C. Lechago J. Xu R. Najki Y. Toll-like receptor-4 is required for intestinal response to epithelial injury and limiting bacterial translocation in a murine model of acute colitis Am. J. Physiol. Gastrointest. Liver Physiol. 2005 288 1055 1065 10.1152/ajpgi.00328.2004 15826931 18. Lim J.Y. Lee Y.K. Lee S.E. Ju J.M. Eom K.S. Kim Y.J. Chung N.G. Jeong D.C. Park G. Choi E.Y. MyD88 in donor bone marrow cells is critical for protection from acute intestinal graft-versus-host disease Mucosal Immunol. 2016 9 730 743 10.1038/mi.2015.96 26442657 19. Abreu M.T. Toll-like receptor signaling in the intestinal epithelium: How bacterial recognition shapes intestinal function Nat. Rev. Immunol. 2010 10 131 144 10.1038/nri2707 20098461 20. Gong F.L. Xiong S.D. Li Z.Y. Medical Immunology 3rd ed. Science Press Wuhan, China 2008 134 135 21. Frazao J.B. Errante P.R. Condino-Neto A. Toll-like receptors’ pathway disturbances are associated with increased susceptibility to infections in humans Arch. Immunol. Ther. Exp. 2013 61 427 443 10.1007/s00005-013-0243-0 24057516 22. Kenny E.F. O’Neill L.A. Signalling adaptors used by Toll-like receptors: An update Cytokine 2008 43 342 349 10.1016/j.cyto.2008.07.010 18706831 23. Goldstein D.R. Toll-like receptors and other links between innate and acquired alloimmunity Curr. Opin. Immunol. 2004 16 538 544 10.1016/j.coi.2004.08.001 15341996 24. Huang Y. Ildstad S.T. A novel role of innate immune responses (toll-like receptor-4) in triggering graft-versus-host disease Transplantation 2010 90 1052 1053 10.1097/TP.0b013e3181f86988 20844466 25. Zhao Y. Cai Z. TLR4 and organ transplant rejection J. Zhejiang Univ. (Med. Sci.) 2011 40 495 500 26. Lorenz E. Schwartz D.A. Martin P.J. Gooley T. Lin M.T. Chien J.W. Hansen J.A. Clark J.G. Association of TLR4 mutations and the risk for acute GVHD after HLA-matched-sibling hematopoietic stem cell transplantation Biol. Blood Marrow Transplant. 2001 7 384 387 10.1053/bbmt.2001.v7.pm11529488 11529488 27. Elmaagacli A.H. Koldehoff M. Hindahl H. Steckel N.K. Trenschel R. Peceny R. Ottinger H. Rath P.M. Ross R.S. Roggendorf M. Mutations in innate immune system NOD2/CARD15 and TLR-4 (Thr399Ile ) genes influence the risk for severe acute graft-versus-host disease in patients who underwent an allogeneic transplantation Transplantation 2006 81 247 254 10.1097/01.tp.0000188671.94646.16 16436969 28. Smirnova I. Hamblin M.T. McBride C. Beutler B. Di Rienzo A. Excess of rare amino acid polymorphisms in the Toll-like receptor 4 in humans Genetics 2001 158 1657 1664 11514453 29. Weng J. Lai P. Geng S. Luo C. Wu S. Ling W. Deng C. Huang X. Lu Z. Du X. Role of Toll-like receptor 4 signaling in cutaneous chronic graft-versus-host disease Clin. Transplant. 2015 29 547 554 10.1111/ctr.12551 25845646 30. Beelen D.W. Elmaagacli A. Müller K.D. Hirche H. Schaefer U.W. Influence of intestinal bacterial decontamination using metronidazole and ciprofloxacin or ciprofloxacin alone on the development of acute graft-versus-host disease after marrow transplantation in patients with hematologic malignancies: Final results and long-term follow-up of an open-label prospective randomized trial Blood 1999 93 3267 3275 10233878 31. Imado T. Iwasaki T. Kitano S. Satake A. Kuroiwa T. Tsunemi S. Sano H. The protective role of host Toll-like receptor-4 in acute graft-versus-host disease Transplantation 2010 90 1063 1070 10.1097/TP.0b013e3181f86947 20975627 32. Fukata M. Chen A. Klepper A. Krishnareddy S. Vamadevan A.S. Thomas L.S. Xu R. Inoue H. Arditi M. Dannenberg A.J. Cox-2 is regulated by Toll-like receptor-4 (TLR4) signaling: Role in proliferation and apoptosis in the intestine Gastroenterology 2006 131 862 877 10.1053/j.gastro.2006.06.017 16952555 33. Kuroiwa T. Kakishita E. Hamano T. Kataoka Y. Seto Y. Iwata N. Kaneda Y. Matsumoto K. Nakamura T. Ueki T. Hepatocyte growth factor ameliorates acute graft-versus-host-disease and promotes hematopoietic function J. Clin. Investig. 2001 107 1365 1373 10.1172/JCI11808 11390418 34. Yasuda K. Wagner H. Takakura Y. Role of immunostimulatory DNA and TLR9 ingene therapy Crit. Rev. Ther. Drug Carr. Syst. 2006 23 89 110 10.1615/CritRevTherDrugCarrierSyst.v23.i2.10 35. Elmaagacli A.H. Steckel N. Ditschkowski M. Hegerfeldt Y. Ottinger H. Trenschel R. Koldehoff M. Beelen D.W. Toll-like receptor 9, NOD2 and IL23R gene polymorphisms influenced outcome in AML patients transplanted from HLA-identical sibling donors Bone Marrow Transplant. 2011 46 702 708 10.1038/bmt.2010.166 20622911 36. Duramad O. Fearon K.L. Chang B. Chan J.H. Gregorio J. Coffman R.L. Barrat F.J. Inhibitors of TLR-9 act on multiple cell subsets in mouse and man in vitro and prevent death in vivo from systemic inflammation J. Immunol. 2005 174 5193 5200 10.4049/jimmunol.174.9.5193 15843514 37. Elmaagacli A.H. Koldehoff M. Beelen D.W. Improved outcome of hematopoietic SCT in patients with homozygous gene variant of Toll-like receptor 9 Bone Marrow Transplant. 2009 44 295 302 10.1038/bmt.2009.32 19252531 38. Taylor P.A. Ehrhardt M.J. Lees C.J. Panoskaltsis-Mortari A. Krieg A.M. Sharpe A.H. Murphy W.J. Serody J.S. Hemmi H. Akira S. TLR agonists regulate alloresponses and uncover a critical role for donor APCs in allogeneic bone marrow rejection Blood 2008 112 3508 3516 10.1182/blood-2007-09-113670 18614760 39. Heimesaat M.M. Nogai A. Bereswill S. Plickert R. Fischer A. Loddenkemper C. Steinhoff U. Tchaptchet S. Thiel E. Freudenberg M.A. MyD88/TLR9 mediated immunopathology and gut microbiota dynamics in a novel murine model of intestinal graft-versus-host disease Gut 2010 59 1079 1087 10.1136/gut.2009.197434 20639251 40. Calcaterra C. Sfondrini L. Rossini A. Sommariva M. Rumio C. Ménard S. Balsari A. Critical role of TLR9 in acute graft-versus-host disease J. Immunol. 2008 181 6132 6139 10.4049/jimmunol.181.9.6132 18941203 41. Ewaschuk J.B. Backer J.L. Churchill T.A. Obermeier F. Krause D.O. Madsen K.L. Surface expression of Toll-like receptor 9 is upregulated on intestinal epithelial cells in response to pathogenicbacterial DNA Infect Immun. 2007 75 2572 2579 10.1128/IAI.01662-06 17325049 42. She K. Gilman A.L. Aslanian S. Shimizu H. Krailo M. Chen Z. Reid G.S. Wall D. Goldman F. Schultz K.R. Altered Toll-like receptor 9 responses in circulating B cells at the onset of extensive chronic graft-versus-host disease Biol. Blood Marrow Transplant. 2007 13 386 397 10.1016/j.bbmt.2006.12.441 17382246 43. Sivula J. Cordova Z.M. Tuimala J. Jaatinen T. Partanen J. Volin L. Turpeinen H. Toll-Like receptor gene polymorphisms confer susceptibility to graft-versus-host disease in allogenic hematopoietic stem cell transplantation Scand. J. Immunol. 2012 76 336 341 10.1111/j.1365-3083.2012.02737.x 22703024 44. Ehlers M. Ravetch J.V. Opposing effects of Toll-like receptor stimulation induce autoimmunity or tolerance Trends Immunol. 2007 28 74 79 10.1016/j.it.2006.12.006 17197239 45. Jasperson L.K. Bucher C. Panoskaltsis-Mortari A. Mellor A.L. Munn D.H. Blazar B.R. Inducing the tryptophan catabolic pathway, indoleamine 2,3-dioxygenase (IDO), for suppression of graft-versus-host disease (GVHD) lethality Blood 2009 114 5062 5070 10.1182/blood-2009-06-227587 19828695 46. Cooke K.R. Hill G.R. Crawford J.M. Bungard D. Brinson Y.S. Delmonte J. Jr. Ferrara J.L. Tumor necrosis factor-α production to lipopolysaccharide stimulation by donor cells predicts the severity of experimental acute graft-versus-host disease J. Clin. Investig. 1998 102 1882 1891 10.1172/JCI4285 9819375 47. Sawitzki B. Brunstein C. Meisel C. Schumann J. Vogt K. Appelt C. Curtsinger J.M. Verneris M.R. Miller J.S. Wagner J.E. Prevention of graft-versus-host disease by adoptive T regulatory therapy is associated with active repression of peripheral blood Toll-like receptor 5 mRNA expression Biol. Blood Marrow Transplant. 2014 20 173 182 10.1016/j.bbmt.2013.10.022 24184334 48. Lim J.Y. Ryu D.B. Lee S.E. Park G. Choi E.Y. Min C.K. Differential effect of MyD88 signal in donor T cells on graft-versus-leukemia effect and graft-versus-host disease after experimental allogeneic Stem cell transplantation Mol. Cells 2015 38 966 974 26552489 49. Zeiser R. Activation of innate immunity in graft-versus-host disease: Implications for Novel Targets? Oncol. Res. Treat. 2015 38 239 243 10.1159/000381296 25966771 50. Nestel F.P. Price K.S. Seemayer T.A. Lapp W.S. Macrophage priming and lipopolysaccharide-triggered release of tumor necrosis factor α during graft-versus-host disease J. Exp. Med. 1992 175 405 413 10.1084/jem.175.2.405 1732411 51. Cooke K.R. Gerbitz A. Crawford J.M. Teshima T. Hill G.R. Tesolin A. Rossignol D.P. Ferrara J.L. LPS antagonism reduces graft-versus-host disease and preserves graft-versus-leukemia activity after experimental bone marrow transplantation J. Clin. Investig. 2001 107 1581 1589 10.1172/JCI12156 11413166 52. Garantziotis S. Palmer S.M. Snyder L.D. Ganous T. Chen B.J. Wang T. Cook D.N. Schwartz D.A. Alloimmune lung injury induced by local innate immune activation through inhaled lipopolysaccharide Transplantation 2007 84 1012 1019 10.1097/01.tp.0000286040.85007.89 17989607 53. Brennan T.V. Lin L. Huang X. Cardona D.M. Li Z. Dredge K. Chao N.J. Yang Y. Heparan sulfate, an endogenous TLR4 agonist, promotes acute GVHD after allogeneic stem cell transplantation Blood 2012 120 2899 2908 10.1182/blood-2011-07-368720 22760779 54. Loiarro M. Capolunghi F. Fantò N. Gallo G. Campo S. Arseni B. Carsetti R. Carminati P. de Santis R. Ruggiero V. Pivotal Advance: Inhibition of MyD88 dimerization and recruitment of IRAK1 and IRAK4 by a novel peptidomimetic compound J. Leukoc. Biol. 2007 82 801 810 10.1189/jlb.1206746 17548806 55. Schmaltz C. Alpdogan O. Horndasch K.J. Muriglan S.J. Kappel B.J. Teshima T. Ferrara J.L. Burakoff S.J. van den Brink M.R. Differential use of Fas ligand and perforin cytotoxic pathways by donor T cells in graft-versus-host disease and graft-versus-leukemia effect Blood 2001 97 2886 2895 10.1182/blood.V97.9.2886 11313285 56. Yang Y.G. Sergio J.J. Pearson D.A. Szot G.L. Shimizu A. Sykes M. Interleukin-12 preserves the graft-versus-leukemia effect of allogeneic CD8 T cells while inhibiting CD4-dependent graft-versus-host disease in mice Blood 1997 90 4651 4660 9373279 57. Dubberke E.R. Hollands J.M. Georgantopoulos P. Augustin K. DiPersio J.F. Mundy L.M. Khoury H.J. Vancomycin-resistant enterococcal bloodstream infections on a hematopoietic stem cell transplant unit: Are the sick getting sicker? Bone Marrow Transplant. 2006 38 813 819 10.1038/sj.bmt.1705530 17057724

PMC005xxxxxx/PMC5000686.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081289ijms-17-01289ArticleAdvanced Glycation End-Products Enhance Lung Cancer Cell Invasion and Migration Hsia Te-Chun 12Yin Mei-Chin 34*Mong Mei-Chin 4Cho William Chi-shing Academic Editor1 Department of Respiratory Therapy, China Medical University, 40402 Taichung City, Taiwan; [email protected] Department of Internal Medicine, China Medical University Hospital, 40402 Taichung City, Taiwan3 Department of Nutrition, China Medical University, 40402 Taichung City, Taiwan4 Department of Health and Nutrition Biotechnology, Asia University, 41354 Taichung City, Taiwan; [email protected]* Correspondence: [email protected]; Tel.: +886-4-2205-3366 (ext. 7510); Fax: +886-4-2206-289109 8 2016 8 2016 17 8 128930 6 2016 04 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Effects of carboxymethyllysine (CML) and pentosidine, two advanced glycation end-products (AGEs), upon invasion and migration in A549 and Calu-6 cells, two non-small cell lung cancer (NSCLC) cell lines were examined. CML or pentosidine at 1, 2, 4, 8 or 16 μmol/L were added into cells. Proliferation, invasion and migration were measured. CML or pentosidine at 4–16 μmol/L promoted invasion and migration in both cell lines, and increased the production of reactive oxygen species, tumor necrosis factor-α, interleukin-6 and transforming growth factor-β1. CML or pentosidine at 2–16 μmol/L up-regulated the protein expression of AGE receptor, p47phox, intercellular adhesion molecule-1 and fibronectin in test NSCLC cells. Matrix metalloproteinase-2 protein expression in A549 and Calu-6 cells was increased by CML or pentosidine at 4–16 μmol/L. These two AGEs at 2–16 μmol/L enhanced nuclear factor κ-B (NF-κ B) p65 protein expression and p38 phosphorylation in A549 cells. However, CML or pentosidine at 4–16 μmol/L up-regulated NF-κB p65 and p-p38 protein expression in Calu-6 cells. These findings suggest that CML and pentosidine, by promoting the invasion, migration and production of associated factors, benefit NSCLC metastasis. CMLpentosidinenon-small cell lung cancermigrationinvasion ==== Body 1. Introduction Advanced glycation end-products (AGEs) such as carboxymethyllysine (CML) and pentosidine are reactive compounds formed from glycosylation of sugars and macromolecules like proteins or lipids. CML and pentosidine could be endogenously synthesized under certain pathological conditions such as diabetes or Alzheimer’s disease, and these AGEs are considered as endogenous AGEs. Many foods including sauces, canned meats, nuts or grain products contain CML, pentosidine and other AGEs [1,2]. Thus, these foods are an exogenous source of AGEs. It is reported that dietary intake of AGEs-rich foods increased circulating AGE levels in patients with diabetes or chronic kidney disease [3,4]. So far, the impact of endogenous or exogenous AGEs upon cancer progression has been a focus because AGE levels were found to be markedly elevated in both serum and tumors, especially in more aggressive tumors [5]. Kim et al. [6] and Sharaf et al. [7] indicated that the engagement of AGEs with their receptor (RAGE) further up-regulated RAGE protein expression, and subsequently activated mitogen-activated protein kinase (MAPK) and nuclear factor-κB (NF-κB) signaling pathways in myeloid leukemia and breast cancer cells. The studies of Weng et al. [8] and Tsao et al. [9] revealed that the activation of MAPK and NF-κB pathways in non-small cell lung cancer (NSCLC) cells promoted the massive production of oxidative, inflammatory and angiogenic factors including reactive oxygen species (ROS), tumor necrosis factor (TNF)-α, intercellular adhesion molecule (ICAM)-1, transforming growth factor (TGF)-β1, vascular endothelial growth factor (VEGF), fibronectin and matrix metalloproteinases (MMPs). Moreover, higher TGF-β1, ICAM-1 and MMP-2 levels in circulation and/or lung tissue were correlated with poor prognosis in NSCLC patients [10,11]. Actually, RAGE is constitutively expressed in lung tissue [12]. However, Marinakis et al. [13] reported that RAGE expression was lower in tumor of NSCLC, the most common type of lung cancer. Those authors indicated that the reduction of RAGE expression might contribute to interrupt cell-to-cell or cell-to-substrate communication, which favored cancer cell progression and migration. Therefore, it is highly possible that the presence of exogenous and/or endogenous AGEs, through rapidly stimulating RAGE expression and activating associated signaling pathways, may enhance the generation of oxidative, metastatic and inflammatory factors, and promote NSCLC progression. Takino et al. [14] indicated that RAGE was associated with cancer malignancy, and glyceraldehyde-derived AGEs facilitated the migration and invasion of A549 cells by activating Rac 1 and increasing ROS formation. Obviously, the influence of AGEs upon NSCLC progression could not be ignored. Since CML and pentosidine are two common AGEs in foods [1], the effects and action modes of these two AGEs upon NSCLC development are worthy of investigation. If CML and/or pentosidine benefit the proliferation, invasion and/or migration of NSCLC cells, they benefit NSCLC progression and deterioration. A549 and Calu-6 cells are human NSCLC cell lines, and have been widely used for NSCLC cell model researches [9,14]. In our present study, these two cell lines were also used to examine the effects of CML and pentosidine at various concentrations upon NSCLC cell proliferation, invasion and migration. Furthermore, the impact of these AGEs upon protein expression of RAGE, TGF-β1, ICAM-1, MMP-2, NADPH oxidase, NF-κB and MAPK was evaluated in order to understand the possible modes of action of AGEs upon NSCLC. 2. Results 2.1. Effects of CML and Pentosidine upon Invasion and Migration of Lung Cancer Cells As shown in Table 1 and Figure 1, CML or pentosidine treatments at test concentrations did not affect proliferation, and protein expression of Bcl-2, Bax, caspase-3 or caspase-8 in A549 cells and Calu-6 cells (p > 0.05). However, CML or pentosidine at 4–16 μmol/L enhanced invasion and migration of A549 cells (Table 2, p < 0.05). In Calu-6 cells, these two AGEs at 2–16 μmol/L promoted invasion and migration (p < 0.05). 2.2. Effects of CML and Pentosidine upon Oxidative and Inflammatory Factors CML at 4–16 μmol/L and pentosidine at 2–16 μmol/L increased the production of ROS, TNF-α, IL-6 and TGF-β1 in A549 and Calu-6 cells (Table 3, p < 0.05). Concentration-dependent manner was presented in raising TNF-α and TGF-β1 release in test cells (p < 0.05). CML at 4–16 μmol/L and pentosidine at 2–16 μmol/L up-regulated RAGE expression in A549 and Calu-6 cells (Figure 2, p < 0.05). In both cell lines, p47phox protein expression was enhanced by CML or pentosidine at 2–16 μmol/L (p < 0.05). However, CML or pentosidine at test concentrations did not alter gp91phox protein expression in two NSCLC cell lines (p > 0.05). 2.3. Effects of CML and Pentosidine upon VEGF, ICAM-1, Fibronectin, MMP-2 and MMP-9 Expression As shown in Figure 3, CML or pentosidine at 2–16 μmol/L up-regulated ICAM-1 and fibronectin protein expression in A549 and Calu-6 cells (p < 0.05). MMP-2 protein expression in A549 cells was increased by two AGEs at 4–16 μmol/L (p < 0.05). In Calu-6 cells, CML or pentosidine at 2–16 μmol/L enhanced MMP-2 protein expression (p < 0.05). CML and pentosidine at test concentrations did not change VEGF and MMP-9 protein expression in test NSCLC cell lines (p > 0.05). 2.4. Effects of CML and Pentosidine upon NF-κB and MAPK Pathways In A549 cells, CML or pentosidine at 2–16 μmol/L up-regulated NF-κB p65 protein expression (Figure 4, p < 0.05). These two AGEs at 1–16 μmol/L increased p38 phosphorylation in A549 cells (p < 0.05). In Calu-6 cells, CML and pentosidine promoted NF-κB p65 protein expression at 2–16 μmol/L and 4–16 μmol/L, respectively (p < 0.05). CML and pentosidine at 2–16 μmol/L enhanced p-p38 protein expression in Calu-6 cells (p < 0.05). CML and pentosidine at test concentrations failed to affect NF-κB p50, JNK and ERK1/2 protein expression or phosphorylation in test NSCLC cell lines (p > 0.05). As shown in Figure 5, CML and pentosidine at 2–16 μmol/L increased NF-κB p50/65 DNA binding activity in A549 and Calu-6 cells (p < 0.05). 3. Discussion Many foods, especially high temperature treated foods, have substantial levels of CML, pentosidine or other AGEs [15,16]. Furthermore, CML and/or pentosidine levels in human circulation could be increased by dietary intake of AGE-containing foods [3,4]. Consequently, these AGEs present in circulation could interact with cancer cells in lung tissue. Turner et al. [5] indicated that exogenous AGEs were linked to cancer risk and disparity. In our present study, CML and pentosidine at test concentrations did not alter cell proliferation and protein expression of Bcl-2, Bax, caspase-3 and caspase-8, apoptotic biomarkers, in two NSCLC cell lines. It seems that these AGEs at those concentrations might not be able to stimulate or inhibit lung tumor growth. However, we found that CML or pentosidine promoted invasion and migration, increased ROS and inflammatory cytokines production, up-regulated protein expression of RAGE, p47phox, ICAM-1, fibronectin and MMP-2, as well as activated NF-κB and MAPK pathways in A549 and Calu-6 cells. These findings suggest that these two AGEs might benefit NSCLC metastasis. Sharaf et al. [7] and Ko et al. [17] reported that AGEs enhanced proliferation, invasion and migration in oral and breast cancer cells. The study of Takino et al. [14] revealed that glyceraldehyde-derived AGEs at 20–100 μg/mL increased the migration capacity of A549 cells. Thus, our findings agreed and suggested that CML and pentosidine are promotive agents upon lung cancer cell migration and invasion. NADPH oxidase complex is responsible for ROS generation in lung tissue. Excessive ROS formation due to NADPH oxidase activation facilitates lung tumorigenesis [18,19]. Our data revealed that CML and pentosidine markedly up-regulated protein expression of p47phox, a cytosolic component of NADPH oxidase, in test NSCLC cells, which subsequently increased ROS production. TGF-β1 is an inflammatory mediator, and also an angiogenic inducer because it enhances epithelial-to-mesenchymal transition in late-stage tumor progression [20]. Saji et al. [21] indicated that TGF-β1, via its immunosuppressive action, facilitated pulmonary metastasis in NSCLC patients. Thus, the greater ROS and TGF-β1 production in CML or pentosidine treated A549 or Calu-6 cells partially explained that these AGEs enhanced oxidative, inflammatory and angiogenic stress in those cells, which might in turn favor the development of microvascular permeability and metastatic actions in lung cancer. Our Western blot data indicated that CML and pentosidine upregulated protein expression of RAGE, MAPK and NF-κB in two NSCLC cell lines. It is reported that the interaction between RAGE and AGEs could activate MAPK and NF-κB signaling pathways [22]. However, we found that CML or pentosidine at 1 μmol/L increased p38 phosphorylation in A549 cells, and both AGEs at 2 μmol/L up-regulated NF-κB p65 expression in A549 and Calu-6 cells; but CML at 4 μmol/L raised RAGE expression in two NSCLC cell lines. These findings suggest that these AGEs might be able to directly mediate NF-κB and MAPK expression, not RAGE dependent. In addition, CML or pentosidine treatments also elevated NF-κB p50/65 DNA binding activity in A549 and Calu-6 cells, which supported the activation of NF-κB. It is known that the activation of these pathways promotes the transcription of their target molecules including oxidants, inflammatory cytokines and even metastatic factors, and finally contributes to lung cancer migration and invasion [23]. Since CML and pentosidine markedly activated RAGE, NF-κB and MAPK pathways in test NSCLC cells, it was reasonable to observe the over-production of ROS, TGF-β1, TNF-α and IL-6 in those NSCLC cells. Besides MAPK and NF-κB pathways, Bao et al. [24] reported that AGE–RAGE interaction promoted prostate cancer cell proliferation via activating PI3K/Akt signaling pathway. Thus, it is highly possible that other pathways are involved in AGEs’ induced NSCLC progression. On the other hand, RAGE could bind to other ligands like HMGB1 and S100P, which subsequently activated TLR-4 and PI3K-Akt pathways, and favored lung cancer progression [25,26]. Thus, the up-regulated RAGE due to CML or pentosidine might in turn benefit NSCLC deterioration through engaging with other ligands and accelerating oxidative, inflammatory, angiogenic and metastatic reactions. Therefore, lowering exogenous AGEs via dietary restriction might be able to reduce RAGE protein expression, diminish the interaction of AGEs and RAGE, and decrease the formation of contributors toward NSCLC metastasis. ICAM-1, a cell adhesion factor, participates in intercellular and cell-extracellular matrix interactions of cancer cells [27]. It is reported that NSCLC patients had higher circulating ICAM-1 levels [28,29], which reflected poor prognosis and worse survival in those patients [28]. Fibronectin is an extra cellular matrix glycoprotein. The expression of fibronectin is increased with lung tumor growth, and is highly associated with resistance to lung cancer therapy [30]. MMPs degrade extracellular matrix components and allow cancer cells to approach vascular and lymphatic systems [31]. MMP-2 could demote type IV collagen, the basic component of the basement membrane in extracellular matrix [32]. We found that CML or pentosidine treatments markedly increased ICAM-1, fibronectin and MMP-2 protein expression in two NSCLC cell lines. Since those metastatic factors had been up-regulated, the observed greater invasion and migration in those cells could be explained. In addition, we notified that CML or pentosidine at test concentrations failed to affect VEGF and MMP-9, two crucial factors responsible for cancer metastasis. These results implied that those AGEs selectively mediated some molecules to promote the invasion and migration in those NSCLC cells. It is interesting to find that pentosidine at 4–16 μmol/L caused greater generation of ROS, TNF-α, IL-6 and TGF-β1 than CML at equal concentrations in A549 and Calu-6 cells. Also, pentosidine at 8 and 16 μmol/L induced greater RGAE protein expression than CML in those NSCLC cells. It is likely that pentosidine was more reactive than CML toward NSCLC cells. Consequently, pentosidine at 8 or 16 μmol/L led to greater migration than CML in A549 and Calu-6 cells. Although CML and pentosidine are AGEs, their impact upon those NSCLC cells does not seem to be identical. 4. Materials and Methods 4.1. Materials CML (95%) and pentosidine (90%) were purchased from Cayman Chemical Co. (Ann Arbor, MI, USA). Plates, medium, chemicals and antibiotics used for cell culture were purchased from Difco Laboratory (Detroit, MI, USA). Human lung cancer cell lines, A549 and Calu-6, were obtained from American Type Culture Collection (Rockville, MD, USA). 4.2. Cell Culture Cells were cultured in RPMI 1640 medium, containing 10% fetal bovine serum (FBS), 100 units/mL of penicillin and 100 units/mL of streptomycin (pH 7.4) at 37 °C in 5% CO2. The culture medium was changed every three days, and cells were subcultured once a week. A phosphate buffer saline (PBS, pH 7.2) was added to adjust the cell number to 105/mL for various experiments and analyses. The plasma concentrations of CML and pentosidine in healthy people were in the range of 0.053–0.49 μmol/L [33]. However, plasma level of these two AGEs in patients with diabetes or renal failure was in the range of 0.2–12.6 μmol/L [34,35]. Thus, CML or pentosidine at 1, 2, 4, 8 and 16 μmol/L were used in present study in order to examine the adverse and possible pathological impact of AGEs. Cells were treated with CML or pentosidine at those concentrations for 18 h at 37 °C, which resulted in 96.3% ± 2.1% incorporation of test agents. AGEs treated cells were washed twice by PBS. CML or pentosidine concentration in collected PBS was analyzed by a competitive ELISA kit (Roche Diagnostics, Penzberg, Germany) or HPLC method of Miyata et al. [35]. The left AGEs in cells were defined as incorporated. Our preliminary test showed that 6, 12, 18, 24 or 36 h incubation led to 43.2%, 68.4%, 96.5%, 95.4% and 94.2% incorporation of CML or pentosidine. Thus, 18 h incubation was used for this study. Control group contained no CML or pentosidine. 4.3. Cell Proliferation Cell proliferation was determined by using a bromodeoxyuridine ELISA colorimetric assay (Roche Diagnostics, Indianapolis, IN, USA). Cells were counted by using a hemocytometer. 4.4. Cell Invasion and Migration Cell invasion and migration were measured in transwell chambers by matrigel- and fibronectin-coated polycarbonate filters, respectively. In brief, cells (105/100 μL) were seeded into the upper chamber in 200 μL of serum-free medium; and the lower chamber was filled with 0.66 mL of RPMI 1640 media containing 10% of FBS as a chemoattractant. After 6 h incubation for migration assay or 16 h incubation for invasion assay at 37 °C, the cells on the upper surface of the filter were removed by a cotton swab. The migrated or invaded cells to the lower surface of the filter were stained with 0.2% crystal violet in 10% ethanol. Four independent fields of invasive or migratory cells per well were photographed under the microscope to count the cell numbers. Data were calculated as a percentage of the control groups. 4.5. Measurement of ROS, Interleukin (IL)-6, TNF-α and TGF-β1 Cells were washed and suspended in RPMI 1640 medium. ROS level was determined by 2’,7’-dichlorofluorescein diacetate, an oxidation sensitive dye. Cells were incubated with 50 μmol/L dye for 30 min and washed twice with PBS. After centrifugation at 412× g for 10 min, the medium was removed and cells were dissolved by 1% Triton X-100. Fluorescence value was measured at time 0 and 5 min by using a fluorescence microplate reader at excitation and emission wavelengths at 485 and 530 nm, respectively. Relative fluorescence unit (RFU) was the difference in fluorescence values obtained between time 0 and 5 min. Results are expressed as RFU/mg protein. The levels of IL-6, TNF-α and TGF-β1 in cell culture supernatant were measured by ELISA kits (R&D Systems, Minneapolis, MN, USA). Protein concentration was determined by an assay kit (Pierce Biotechnology Inc., Rockford, IL, USA), and bovine serum albumin was used as a standard. 4.6. Assay for NF-κB p50/65 DNA Binding Activity Nuclei pellets were isolated and re-suspended in a solution containing 20 mM HEPES, 1 mM EDTA, 0.4 M NaCl, 1 mM DTT and 25% glycerol. After incubation and centrifugation, supernatants were collected for protein concentration quantification by protein assay reagents (Bio-Rad Laboratories Inc., Hercules, CA, USA). NF-κB p50/65 DNA binding activity was determined by an assay kit (Chemicon International Co., Temecula, CA, USA). The binding of activated NF-κB was processed by a primary polyclonal antibody against NF-κB p50/p65, and followed by treating with an antibody conjugated with horseradish peroxidase. 3,3′,5,5′-tetramethylbenzidine was the substrate. The absorbance at 450 nm was recorded. Data are shown as optical density (OD)/mg protein. 4.7. Western Blot Analyses Cell was homogenized in protease-inhibitor cocktail containing 0.5% Triton X-100 (Sigma-Aldrich Chemical Co., St. Louis, MO, USA). This homogenate was further mixed with a buffer composed of 60 mM Tris-HCl, 2% β-mercaptoethanol and 2% SDS (pH 7.2), and boiled for 5 min. Sample at 40 μg protein was applied to 10% SDS-PAGE, and further transferred to a nitrocellulose membrane (Millipore, Bedford, MA, USA) for 1 h. After blocking with a solution containing 5% skim milk for 1 h to prevent non-specific binding of antibody, membrane was reacted with monoclonal antibody against Bcl-2, Bax, caspase-3, caspase-8 (1:1000), RAGE (1:500), p47phox, gp91phox, VEGF, ICAM-1, MMP-2, MMP-9, fibronectin, NF-κB (1:1000) or MAPK (1:2000) (Boehringer-Mannheim, Indianapolis, IN, USA) at 4 °C overnight, and followed by treating samples with horseradish peroxidase-conjugated antibody for 3.5 h at room temperature. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control, and the detected bands were quantified by normalizing against GAPDH. 4.8. Statistical Analysis The effect of each treatment was analyzed from 10 different preparations (n = 10). Data were expressed as means ± standard deviation (SD), and processed for analysis of variance. Differences among means were determined by Fisher’s Least Significance Difference Test with significance defined at p < 0.05. 5. Conclusions CML and pentosidine enhanced invasion and migration of A549 and Calu-6 cells. These AGEs increased the production of ROS and inflammatory cytokines, and up-regulated protein expression of NADPH oxidase, RAGE, ICAM-1, fibronectin, MMP-2, NF-κB p65 and p-p38 in both NSCLC cell lines. These findings suggest that CML and pentosidine benefit NSCLC metastasis. Acknowledgments This study was supported by a grant from China Medical University, Taichung City, Taiwan (CMU105-ASIA-12). Author Contributions Te-Chun Hsia and Mei-Chin Yin designed the experiments, Mei-Chin Yin and Mei-Chin Mong performed the experiments. Te-Chun Hsia discussed the data. Mei-Chin Yin wrote the manuscript. Conflicts of Interest The authors declare no conflict of interest. Abbreviations CML carboxymethyllysine AGE advanced glycation end-product NSCLC non-small cell lung cancer MMP matrix metalloproteinase RAGE receptor for advanced glycation end-product MAPK mitogen-activated protein kinase NF-κB nuclear factor κ-B ROS reactive oxygen species TNF tumor necrosis factor ICAM intercellular adhesion molecule TGF transforming growth factor VEGF vascular endothelial growth factor Figure 1 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon protein expression of Bcl-2, Bax, caspase-3 and caspase-8 in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10). Figure 2 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon protein expression of RAGE, p47phox and gp91phox in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10) and shown in the following table. a–d Means within a column without a common letter differ, p < 0.05. Figure 3 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon protein expression of VEGF, ICAM-1, fibronectin, MMP-2 and MMP-9 in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10) and shown in the following table. a–e Means within a column without a common letter differ, p < 0.05. Figure 4 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon protein expression of NF-κB and MAPK in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10), and shown in the following table. a–e Means within a column without a common letter differ, p < 0.05. Figure 5 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon NF-κB p50/65 DNA binding activity, determined as OD450 nm/mg protein, in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10). a–e Means among bars without a common letter differ, p < 0.05. ijms-17-01289-t001_Table 1Table 1 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon cell proliferation (% of control) in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10). A549 Cells Calu-6 Cells CML, 0 100 100 1 98 ± 4 101 ± 2 2 101 ± 5 97 ± 4 4 103 ± 3 100 ± 3 8 97 ± 5 103 ± 2 16 102 ± 4 99 ± 4 Pentosidine, 0 100 100 1 102 ± 3 99 ± 5 2 103 ± 4 102 ± 2 4 100 ± 2 104 ± 3 8 97 ± 3 101 ± 4 16 101 ± 5 98 ± 5 ijms-17-01289-t002_Table 2Table 2 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon cell invasion (% of control) and migration (% of control) in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10). a–e Means within a column without a common letter differ, p < 0.05. A549 Cells Calu-6 Cells Invasion Migration Invasion Migration CML, 0 100 a 100 a 100 a 100 a 1 102 ± 3 a 98 ± 4 a 99 ± 2 a 98 ± 5 a 2 107 ± 4 a 106 ± 5 a 127 ± 4 b 132 ± 3 b 4 135 ± 3 b 140 ± 7 b 155 ± 5 c 158 ± 4 c 8 142 ± 5 b 144 ± 5 b 182 ± 4 d 187 ± 6 d 16 166 ± 7 c 174 ± 4 c 190 ± 6 d 195 ± 5 d Pentosidine, 0 100 a 100 a 100 a 100 a 1 99 ± 2 a 104 ± 5 a 103 ± 4 a 107 ± 4 a 2 103 ± 5 a 108 ± 3 a 133 ± 2 b 147 ± 3 b 4 138 ± 4 b 142 ± 4 b 164 ± 5 c 173 ± 6 c 8 157 ± 5 c 168 ± 6 c 191 ± 7 d 205 ± 5 d 16 184 ± 7 d 201 ± 4 d 225 ± 6 e 232 ± 3 e ijms-17-01289-t003_Table 3Table 3 Effects of CML or pentosidine at 0 (control), 1, 2, 4, 8 or 16 μmol/L upon ROS (RFU/mg protein), TNF-α (pg/mg protein), IL-6 (pg/mg protein) and TGF-β1 (pg/mg protein) levels in human A549 and Calu-6 cells. Cells were exposed to CML or pentosidine for 18 h at 37 °C. Data are mean ± SD (n = 10). a–e Means within a column without a common letter differ, p < 0.05. A549 Cells Calu-6 Cells ROS TNF-α IL-6 TGF-β1 ROS TNF-α IL-6 TGF-β1 CML, 0 1.97 ± 0.18 a 156 ± 13 a 133 ± 10 a 141 ± 8 a 2.09 ± 0.21 a 160 ± 18 a 136 ± 8 a 130 ± 7 a 1 2.06 ± 0.21 a 161 ± 9 a 142 ± 8 a 147 ± 5 a 2.14 ± 0.15 a 158 ± 12 a 142 ± 14 a 139 ± 10 a 2 2.18 ± 0.25 a 167 ± 17 a 150 ± 16 a 163 ± 11 a 2.23 ± 0.24 a 166 ± 9 a 148 ± 10 a 145 ± 12 a 4 2.65 ± 0.17 b 194 ± 14 b 187 ± 11 b 197 ± 9 b 2.71 ± 0.17 b 201 ± 15 b 191 ± 16 b 186 ± 8 b 8 2.84 ± 0.20 b 237 ± 22 c 224 ± 19 c 245 ± 13 c 3.18 ± 0.23 c 242 ± 13 c 247 ± 19 c 217 ± 14 c 16 3.39 ± 0.28 c 280 ± 19 d 275 ± 23 d 291 ± 17 d 3.30 ± 0.28 c 293 ± 20 d 258 ± 25 c 266 ± 21 d Pentosidine, 0 2.08 ± 0.11 a 149 ± 15 a 136 ± 7 a 139 ± 9 a 2.21 ± 0.14 a 152 ± 12 a 130 ± 11 a 134 ± 9 a 1 2.13 ± 0.16 a 157 ± 12 a 145 ± 14 a 148 ± 10 a 2.18 ± 0.19 a 159 ± 8 a 127 ± 14 a 143 ± 13 a 2 2.62 ± 0.09 b 188 ± 10 b 176 ± 18 b 182 ± 13 b 2.57 ± 0.12 b 195 ± 16 b 178 ± 15 b 175 ± 15 b 4 2.75 ± 0.21 b 226 ± 19 c 213 ± 15 c 234 ± 8 c 3.06 ± 0.20 c 240 ± 23 c 201 ± 19 b 223 ± 10 c 8 3.55 ± 0.18 c 270 ± 22 d 259 ± 17 d 290 ± 14 d 3.59 ± 0.16 d 291 ± 26 d 263 ± 25 c 279 ± 14 d 16 3.69 ± 0.25 c 331 ± 27 e 304 ± 22 e 359 ± 19 e 3.74 ± 0.25 d 355 ± 21 e 310 ± 27 d 348 ± 16 e ==== Refs References 1. Chao P.C. Hsu C.C. Yin M.C. Analysis of glycative products in sauces and sauce-treated foods Food Chem. 2009 113 262 266 10.1016/j.foodchem.2008.06.076 2. Scheijen J.L. Clevers E. Engelen L. Dagnelie P.C. Brouns F. Stehouwer C.D. Schalkwijk C.G. Analysis of advanced glycation endproducts in selected food items by ultra-performance liquid chromatography tandem mass spectrometry: Presentation of a dietary AGE database Food Chem. 2016 190 1145 1150 10.1016/j.foodchem.2015.06.049 26213088 3. Chao P.C. Huang C.N. Hsu C.C. Yin M.C. Guo Y.R. Association of dietary AGEs with circulating AGEs, glycated LDL, IL-1α and MCP-1 levels in type 2 diabetic patients Eur. J. Nutr. 2010 49 429 434 10.1007/s00394-010-0101-3 20229096 4. Piroddi M. Palazzetti I. Quintaliani G. Pilolli F. Montaldi M. Valentina V. Libetta C. Galli F. Circulating levels and dietary intake of the advanced glycation end-product marker carboxymethyl lysine in chronic kidney disease patients on conservative predialysis therapy: A pilot study J. Ren. Nutr. 2011 21 329 339 10.1053/j.jrn.2010.06.024 21193322 5. Turner D.P. Advanced glycation end-products: A biological consequence of lifestyle contributing to cancer disparity Cancer Res. 2015 75 1925 1929 10.1158/0008-5472.CAN-15-0169 25920350 6. Kim J.Y. Park H.K. Yoon J.S. Kim S.J. Kim E.S. Ahn K.S. Kim D.S. Yoon S.S. Kim B.K. Lee Y.Y. Advanced glycation end product (AGE)-induced proliferation of HEL cells via receptor for AGE-related signal pathways Int. J. Oncol. 2008 33 493 501 10.3892/ijo_00000032 18695878 7. Sharaf H. Matou-Nasri S. Wang Q. Rabhan Z. Al-Eidi H. Al Abdulrahman A. Ahmed N. Advanced glycation endproducts increase proliferation, migration and invasion of the breast cancer cell line MDA-MB-231 Biochim. Biophys. Acta 2015 1852 429 441 10.1016/j.bbadis.2014.12.009 25514746 8. Weng Y. Cai M. Zhu J. Geng J. Zhu K. Jin X. Ding W. Matrix metalloproteinase activity in early-stage lung cancer Onkologie 2013 36 256 259 10.1159/000350304 23689219 9. Tsao S.M. Hsia T.C. Yin M.C. Protocatechuic acid inhibits lung cancer cells by modulating FAK, MAPK, and NF-κB pathways Nutr. Cancer 2014 66 1331 1341 10.1080/01635581.2014.956259 25356681 10. Kumar S. Guleria R. Mohan A. Singh V. Bharti A.C. Das B.C. Efficacy of plasma TGF-β1 level in predicting therapeutic efficacy and prognosis in patients with advanced non-small cell lung cancer Cancer Investig. 2011 29 202 207 10.3109/07357907.2010.543208 21280981 11. Kanoh Y. Abe T. Masuda N. Akahoshi T. Progression of non-small cell lung cancer: Diagnostic and prognostic utility of matrix metalloproteinase-2, C-reactive protein and serum amyloid A Oncol. Rep. 2013 29 469 473 10.3892/or.2012.2123 23135313 12. Buckley S.T. Ehrhardt C. The receptor for advanced glycation end products (RAGE) and the lung J. Biomed. Biotechnol. 2010 2010 917108 10.1155/2010/917108 20145712 13. Marinakis E. Bagkos G. Piperi C. Roussou P. Diamanti-Kandarakis E. Critical role of RAGE in lung physiology and tumorigenesis: A potential target of therapeutic intervention? Clin. Chem. Lab. Med. 2014 52 189 200 10.1515/cclm-2013-0578 24108211 14. Takino J. Yamagishi S. Takeuchi M. Cancer malignancy is enhanced by glyceraldehyde-derived advanced glycation end-products J. Oncol. 2010 2010 739852 10.1155/2010/739852 20631911 15. Cheng L. Jin C. Zhang Y. Investigation of variations in the acrylamide and N ε -(carboxymethyl) lysine contents in cookies during baking J. Food Sci. 2014 79 1030 1038 10.1111/1750-3841.12450 24734960 16. Chen G. Smith J.S. Determination of advanced glycation endproducts in cooked meat products Food Chem. 2015 68 190 195 10.1016/j.foodchem.2014.06.081 25172699 17. Ko S.Y. Ko H.A. Shieh T.M. Chang W.C. Chen H.I. Chang S.S. Lin I.H. Cell migration is regulated by AGE-RAGE interaction in human oral cancer cells in vitro PLoS ONE 2014 9 1289 18. Valavanidis A. Vlachogianni T. Fiotakis K. Loridas S. Pulmonary oxidative stress, inflammation and cancer: Respirable particulate matter, fibrous dusts and ozone as major causes of lung carcinogenesis through reactive oxygen species mechanisms Int. J. Environ. Res. Public Health 2013 10 3886 3907 10.3390/ijerph10093886 23985773 19. Liu X. Pei C. Yan S. Liu G. Liu G. Chen W. Cui Y. Liu Y. NADPH oxidase 1-dependent ROS is crucial for TLR4 signaling to promote tumor metastasis of non-small cell lung cancer Tumor Biol. 2015 36 1493 1502 10.1007/s13277-014-2639-9 25592377 20. Roberts A.B. Wakefield L.M. The two faces of transforming growth factor β in carcinogenesis Proc. Natl. Acad. Sci. USA 2003 100 8621 8623 10.1073/pnas.1633291100 12861075 21. Saji H. Nakamura H. Awut I. Kawasaki N. Hagiwara M. Ogata A. Hosaka M. Saijo T. Kato Y. Kato H. Significance of expression of TGF-β in pulmonary metastasis in non-small cell lung cancer tissues Ann. Thorac. Cardiovasc. Surg. 2003 9 295 300 14672525 22. Xie J. Méndez J.D. Méndez-Valenzuela V. Aguilar-Hernández M.M. Cellular signalling of the receptor for advanced glycation end products (RAGE) Cell Signal. 2013 25 2185 2197 10.1016/j.cellsig.2013.06.013 23838007 23. Mehta H.J. Patel V. Sadikot R.T. Curcumin and lung cancer-a review Target Oncol. 2014 9 295 310 10.1007/s11523-014-0321-1 24840628 24. Bao J.M. He M.Y. Liu Y.W. Lu Y.J. Hong Y.Q. Luo H.H. Ren Z.L. Zhao S.C. Jiang Y. AGE/RAGE/Akt pathway contributes to prostate cancer cell proliferation by promoting Rb phosphorylation and degradation Am. J. Cancer Res. 2015 5 1741 1750 10.21037/tau.2016.s044 26175942 25. Arumugam T. Logsdon C.D. S100P: A novel therapeutic target for cancer Amino Acids 2011 41 893 899 10.1007/s00726-010-0496-4 20509035 26. Xu X. Zhu H. Wang T. Sun Y. Ni P. Liu Y. Tian S. Amoah Barnie P. Shen H. Xu W. Exogenous high-mobility group box 1 inhibits apoptosis and promotes the proliferation of lewis cells via RAGE/TLR4-dependent signal pathways Scand. J. Immunol. 2014 79 386 394 10.1111/sji.12174 24673192 27. Gho Y.S. Kim P.N. Li H.C. Elkin M. Kleinman H.K. Stimulation of tumor growth by human soluble intercellular adhesion molecule-1 Cancer Res. 2001 61 4253 4257 11358852 28. Dowlati A. Gray R. Sandler A.B. Schiller J.H. Johnson D.H. Cell adhesion molecules, vascular endothelial growth factor, and basic fibroblast growth factor in patients with non-small cell lung cancer treated with chemotherapy with or without bevacizumab—An Eastern Cooperative Oncology Group Study Clin. Cancer Res. 2008 14 1407 1412 10.1158/1078-0432.CCR-07-1154 18316562 29. Guney N. Soydinc H.O. Derin D. Tas F. Camlica H. Duranyildiz D. Yasasever V. Topuz E. Serum levels of intercellular adhesion molecule ICAM-1 and E-selectin in advanced stage non-small cell lung cancer Med. Oncol. 2008 25 194 200 10.1007/s12032-007-9026-y 18008189 30. Ritzenthaler J.D. Han S. Roman J. Stimulation of lung carcinoma cell growth by fibronectin-integrin signaling Mol. BioSyst. 2008 4 1160 1169 10.1039/b800533h 19396378 31. Pritchard S.C. Nicolson M.C. Lloret C. McKay J.A. Ross V.G. Kerr K.M. Murray G.I. McLeod H.L. Expression of matrix metalloproteinases 1, 2, 9 and their tissue inhibitors in stage II non-small cell lung cancer: Implications for MMP inhibition therapy Oncol. Rep. 2001 8 421 424 10.3892/or.8.2.421 11182067 32. Hamano Y. Zeisberg M. Sugimoto H. Lively J.C. Maeshima Y. Yang C. Hynes R.O. Werb Z. Sudhakar A. Kalluri R. Physiological levels of tumstatin, a fragment of collagen IV α3 chain, are generated by MMP-9 proteolysis and suppress angiogenesis via αV β3 integrin Cancer Cell 2003 3 589 601 10.1016/S1535-6108(03)00133-8 12842087 33. Hoonhorst S.J. Lo Tam Loi A.T. Pouwels S.D. Faiz A. Telenga E.D. van den Berge M. Koenderman L. Lammers J.W. Boezen H.M. van Oosterhout A.J. Advanced glycation endproducts and their receptor in different body compartments in COPD Respir. Res. 2016 17 46 10.1186/s12931-016-0363-2 27117828 34. Misselwitz J. Franke S. Kauf E. John U. Stein G. Advanced glycation end products in children with chronic renal failure and type 1 diabetes Pediatr. Nephrol. 2002 17 316 321 10.1007/s00467-001-0815-9 12042886 35. Miyata T. Taneda S. Kawai R. Ueda Y. Horiuchi S. Hara M. Maeda K. Monnier V.M. Identification of pentosidine as a native structure for advanced glycation end products in β2-microglobulin-containing amyloid fibrils in patients with dialysis-related amyloidosis Proc. Natl. Acad. Sci. USA 1996 93 2353 2358 10.1073/pnas.93.6.2353 8637877

PMC005xxxxxx/PMC5000687.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081290ijms-17-01290ReviewSilk Spinning in Silkworms and Spiders Andersson Marlene 1Johansson Jan 12Rising Anna 12*Hardy John G. Academic EditorHolland Chris Academic Editor1 Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala 75651, Sweden; [email protected] (M.A.); [email protected] (J.J.)2 Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm 14157, Sweden* Correspondence: [email protected] or [email protected]; Tel.: +46-67211409 8 2016 8 2016 17 8 129017 6 2016 02 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Spiders and silkworms spin silks that outcompete the toughness of all natural and manmade fibers. Herein, we compare and contrast the spinning of silk in silkworms and spiders, with the aim of identifying features that are important for fiber formation. Although spiders and silkworms are very distantly related, some features of spinning silk seem to be universal. Both spiders and silkworms produce large silk proteins that are highly repetitive and extremely soluble at high pH, likely due to the globular terminal domains that flank an intermediate repetitive region. The silk proteins are produced and stored at a very high concentration in glands, and then transported along a narrowing tube in which they change conformation in response primarily to a pH gradient generated by carbonic anhydrase and proton pumps, as well as to ions and shear forces. The silk proteins thereby convert from random coil and alpha helical soluble conformations to beta sheet fibers. We suggest that factors that need to be optimized for successful production of artificial silk proteins capable of forming tough fibers include protein solubility, pH sensitivity, and preservation of natively folded proteins throughout the purification and initial spinning processes. spidroinfibroinBombyx morimajor ampullate glandcarbonic anhydrasepH gradientprotein conformation ==== Body 1. Introduction Spiders and silkworms produce silks with impressive properties. Not only do silk fibers represent the strongest fibers in nature [1], they are also well tolerated when implanted in pigs [2], sheep [3], and rats [4], and therefore represent interesting materials for a wide variety of applications. Spider and silkworm silks are produced in specific glands from unique proteins that are spun under ambient conditions. The domesticated silkworm Bombyx mori produces large amounts of silk for cocoons used during the metamorphosis from larvae to moths, while spiders spin silk for a variety of purposes, including web building, prey swathing, and reproduction, but only in small amounts. The ability to produce silk has apparently evolved multiple times, in silkworms and spiders but also in other members of the arthropod phylogeny [5]. The last common ancestor of silkworms and spiders lived around 500 million years ago [6] and although the origin of their silk producing organs is different, there are remarkable similarities in the way spiders and silkworms spin their silk, which are discussed herein. 2. Similar Gland Morphologies in Silkworms and Spiders The silk glands in B. mori originate from salivary glands, which explains why the silk fiber exits through a pore on the lower lip of the mouth of the silkworm. The silk glands are paired, but the two filaments produced are fused into a single fiber near the pore [7]. The B. mori silk gland can be divided into three macroscopic parts: the posterior silk gland (PSG), the middle silk gland (MSG) and the anterior silk gland (ASG) (Figure 1a) [7,8]. In contrast to silkworms, most spiders have several different types of glands, which produce silks with varying mechanical properties [11,12]. In spiders, the silk glands are thought to have evolved from epidermal invaginations of the opisthosoma (abdomen), hence the spigots are positioned in a caudal ventral position on the opisthosoma [11]. Herein, we focus on the major ampullate gland, which produces the dragline silk. The major ampullate gland is present in pairs and is, similarly to the silkworm silk gland, divided into three parts; the tail, the sac and the duct (Figure 1b) [9,13]. The cells in the PSG of the B. mori silk gland, and the corresponding tail of the major ampullate gland, produce proteins—fibroins and spidroins, respectively—that are stored in the lumen of the MSG/sac [13,14] in a soluble state at a very high concentration [15]. The silk proteins are then transported through the ASG/duct where they change conformation in response to several factors (cf. below). Upon spinning, the fiber is pulled out by the motion of the head of the silkworm or by the motion of the spider’s leg or by other means (e.g., gravity if the spider is suspended, movement of the spider in the web, wind) [16]. Both the B. mori silk gland and the major ampullate gland epithelium consist of different types of single columnar cells in the PSG/tail and MSG/sac, respectively. The cells in the most proximal parts of the glands are responsible for secreting the major constituent of the silk, fibroins/spidroins (purple in Figure 1c,d) [8,9,17]. The cells in the MSG of B. mori glands secrete different types of sericin (purple, Figure 1c) [7,8,14], which will form a coat surrounding the fibroin. In the major ampullate gland, the cell types present in the proximal part of the sac produce spidroins (purple, Figure 1d) [9] while the epithelium of the most distal part of the sac and the duct produces carbonic anhydrase (CA) (grey, Figure 1d). CA interconverts H2O + CO2 and H+ + HCO3− and is responsible for the generation and maintenance of the pH gradient of the gland (cf. below) [10]. In analogy, the final cell type in the MSG and the cells along the ASG found in B. mori silk glands likewise produce CA (grey, Figure 1c) [8]. At the funnel, a cuticular intima appears that lines the apical (interior) part of the ASG/duct epithelium all the way to the lip pore/spigot [7,8,9]. The cuticular intima in spiders has been hypothesized to protect the underlying cells of the duct from laceration caused by the appearing fiber [18]. It has also been suggested to act as a hollow dialysis membrane, aiding in dehydrating the spinning dope [19]. The ASG in B. mori and duct of the major ampullate gland are very similar in appearance and in a cross-section one can see epithelial cells with very abundant microvilli at the apical cell membrane, lined by the cuticular intima surrounding a lumen [8,9] (Figure 2). The diameter of the lumen in both the ASG and the duct is narrowing towards the lip pore/spigot. However, there is a big difference in size and geometry between spiders and silkworms, which is modeled to affect shear forces generated within the ASG/duct, cf. below [20]. The duct of the major ampullate gland has a slow and steady decrease in diameter going from around 100 µm to <10 µm [21], while the silkworm silk gland ASG diameter starts at 400 µm and after a sudden drop to 100 µm there is a slow decrease in diameter until it ends at around 50 µm [7]. 3. pH Gradients, Ions and Shear Forces Govern Silk Formation pH gradients have been inferred to be important for fiber formation in both silkworms and spiders, based on in vitro studies of regenerated silk proteins and recombinant spider silk proteins, but there are only a few reports on measured pH values in silk glands. The pH gradient in B. mori silk glands has been proposed to go from 6.9 in the PSG, to around 5 in the MSG and down to 4.8 in the ASG [22], but technical details for the underlying experiments were not described [17]. Another study, using injection of phenol red in the haemocoel of silkworms, showed that pH was 7–8 in the PSG and 5–6 in the MSG, while the dye could not penetrate the ASG [23]. Similar studies using pH sensitive dyes have been performed on major ampullate glands of spiders and have given pH values starting at ~7 in the sac going down to 6–6.5 in the duct [24,25]. pH sensitive dyes have low resolution, which likely explains why no difference in pH could be found between the beginning and end of the duct [24,25]. In two recent studies, pH was measured in silk glands of B. mori fifth instar larvae as well as in Nephila clavipes major ampullate glands using ion-selective microelectrodes [8,10]. The microelectrodes were very small, with tip sizes of 2–4 µm, which allowed for measurements also in the tiny PSG/tail and ASG/duct. In both the spider and the silkworm, pH was higher in the PSG/tail than previously reported, reaching >7.5 (Figure 1). In the major ampullate gland duct the pH was much lower than anticipated with a pH of 5.7 in the middle of the duct [10], while in the silkworm silk gland the lowest pH measured was 6.2 in the beginning of the ASG [8] (Figure 1). The pH gradients are apparently tightly controlled, since measured pH values are remarkably consistent, with less than 0.1 pH unit differences between individual spiders [10]. In none of the studies performed has the pH at the very end of the duct/ASG been possible to measure, since the duct/ASG is impenetrable for both dyes and microelectrodes. The exact mechanism whereby acidification of the dope occurs is still not fully elucidated. The generation and maintenance of the pH gradient has been attributed to CA activity in the ASG of silkworm silk glands [8] and in part of the sac and duct of the major ampullate gland [10] as well as to ATPase driven proton pumps in the epithelium of silkworm glands [26] and in the duct of major ampullate glands [27]. Ion concentration gradients have been suggested to play a role in fiber formation, and have been studied in both B. mori silk glands and major ampullate glands. In silkworms, copper levels in different parts of the silk gland were measured using proton induced X-ray emission, and an increase from PSG to ASG was measured [28]. Furthermore, proton induced X-ray emission, inductively coupled plasma mass spectroscopy and atomic adsorption spectroscopy showed that levels of sodium, potassium, magnesium and zinc also increase from PSG to ASG, while calcium levels decreased significantly [29]. Potassium, phosphorus and sulphur levels increase from proximal to distal parts of the major ampullate gland, while sodium and chloride levels decrease according to an analysis performed using cryo-SEM-EDAX (scanning electron microscope-energy dispersive X-ray) [24]. The cryo-SEM-EDAX data show statistically significant differences in ion levels at different locations, but there are large variations in the measurements, making interpretations difficult. Moreover, the final point measured is not in the duct, but in the fiber, and it is unclear if a correlation would be found if measurements were confined to the duct. For example, in B. mori silk glands, the copper levels were significantly higher in the fiber than in the end of the ASG [28]. The large total surface area provided by the microvilli at the apical end of the cells in the ASG/duct [8,9] probably allow for efficient absorption of water from the dope [19], which contributes to fiber formation. Shear forces are also important for fiber formation and a stress-induced phase transition of the spinning dope is evident in both spiders and silkworms [7,30,31]. Although the ASG and duct are similar in appearance and function, modeling suggests that the shear forces within the ASG and duct are different [20], with significantly stronger forces within the major ampullate gland duct, which might influence the tensile properties of the silk fibers. The difference in generated shear force between silkworms and spiders is ascribed to the differences in ASG and duct geometry [20]. 4. A Comparison of Fibroins and Spidroins B. mori fibers consist of several different proteins: the fibroin light (around 25 kDa) and heavy chain (around 350 kDa), fibroin p25 (a polypeptide which associates to the heavy and light chain by hydrophobic interactions), and sericin [32,33]. The fibroin heavy chain consists of the two hydrophilic N- and C-terminal domains (NT and CT), and a highly repetitive glycine and alanine-rich (GA) region consisting of 12 repetitive GA segments with conserved linker sequences in between. The fibroin heavy chain NT (FibNT) is 151 amino acid residues long [33] and adopts mainly a random coil conformation at neutral pH, while it folds into a double-layered anti-parallel beta sheet dimer at low pH [34]. The pH sensitivity is attributed to a cluster of acidic residues that are protonated around pH 6 and thereby the conformational change can take place [34]. The fibroin heavy chain CT (FibCT) is 58 residues long, its structure has not been characterized, and it is bound to the light chain by a disulfide bond. The fibroin light chain is a nonrepetitive protein rich in arginine and lysine. The exact function and importance of the fibroin light chain has been debated [35,36,37] but findings by Chen et al. [38] indicate that it plays an important role in lysine-mediated cross links in silk, which are common also in for example collagen and elastin. Dragline silk also consists of several proteins, mainly major ampullate spidroins (MaSps) 1 and 2 that are very similar in primary structure [6]. The overall architecture of the MaSps is reminiscent of the fibroin heavy chain, with a repetitive region rich in glycine and alanine amino acid residues, flanked by two non-repetitive terminal domains, NT and CT. However, the amino acid sequences of NT, CT and the repetitive part are very different between the fibroin heavy chain and MaSps. The FibNT [35] and the spidroin NT (SpNT) [39] are not homologous, nor are FibCT [35] and spidroin CT (SpCT) [40]. The SpNT is an approximately 130 amino acid residue domain that folds into a five helix bundle monomer with a dipolar charge distribution at high pH, while it dimerizes and forms an antiparallel dimer at low pH [41,42,43]. In analogy to the FibNT, specific carboxylates prevent the dimerization of SpNT at high pH, and protonation of these side chains at around pH 6.5 mediates dimerization [43]. However, in contrast to FibNT, SpNT undergoes further stabilization at pH 5–6, which is mediated by protonation of a specific carboxylate [43,44]. Moreover, SpNT is alpha helical at both high and low pH, while FibNT is in random coil conformation at high pH and forms a beta sheet rich dimer structure at low pH [34,43]. The SpCT is also a five helix bundle at high pH [10,45], and is a non-covalent or covalently linked, via a disulfide bond, constitutive dimer [40]. Upon a lowering of pH, SpCT unfolds and turns into beta sheet nuclei, which may trigger further beta sheet formation of the repetitive region of spidroins (SpRep) [10,45]. Shear forces affect the SpCT in a similar way [46], and most likely pH reduction and shear forces act together to mediate SpCT conformational changes during silk formation. There are apparent similarities between structural conversions in fibroins and spidroins since FibNT can convert to beta sheet structure at low pH [34] and SpCT unfolds and turns into amyloid-like fibrils, built up from β-sheets, upon lowering of pH [10]. However, the former transition is a bimolecular event that results in a defined soluble and globular structure, while the latter transition involves multiple intermolecular interactions and results in insoluble fibrils. The SpCT was recently found (using a web-based tool for prediction of amyloidogenic regions in proteins, waltz-switchlab.org) [47] to contain three amyloidogenic regions (Figure S1) [48], which can also be found using the Tango tool [49,50,51]. Analogous regions can be found in FibNT (Figure S1), further pointing to the possibility that FibNT and SpCT have similar functions in the control of fiber formation. Another candidate that may be responsible for nucleation dependent events in silkworm silk formation is the light chain, which also has a high amyloid-forming capacity predicted by the Waltz (Figure S1) and Tango tools. Although the amino acid composition of the repetitive region in silkworms and spiders is quite similar, their primary structures are very different. The repetitive part of fibroins (FibRep) and SpRep are both glycine- and alanine-rich, but FibRep have (GAGAGAGS)n motifs separated by non-repetitive linker regions [35], while SpRep have alternating poly-A and glycine rich repeats, mainly GGX (as in MaSp1) or GPGXX (as in MaSp2) [52,53]. The repetitive region confers the mechanical properties to the fiber, why these differences are likely responsible for the differences seen in secondary structure and tensile properties of the fibers (cf. below). For schematic illustrations of the different spidroin and fibroin domains and their structural conversions, see for example references [1,54,55,56,57]. 5. Are There Post-Translational Modifications of Silk Proteins? Post-translational modifications (PTMs), such as glycosylations and phosphorylations, of silkworm silk and spider silk have been studied to some extent, but it is unclear what PTMs are present, and also the effect of these potential PTMs on fiber formation and on the mechanical properties of fibers remain to be established. MaSp1 and MaSp2 isolated from N. clavipes dragline silk fibers contain phosphorylation sites in their respective repetitive regions [58,59], as does FibRep [38], the fibroin light chain and the P25 [60]. The impact of these phosphorylations on the silk formation process have not been thoroughly studied, but phosphorylation in general affect soluble proteins by altering conformational states. Michal et al. [61] studied N. clavipes dragline silk fibers and dope using solid state 31P-NMR, and found that the dragline silk proteins contain phosphotyrosine. In another study, a phosphorylation site was included in a recombinant construct of spider dragline silk. Phosphorylation of the expressed and purified protein increased the solubility level of the construct, while dephosphorylation caused beta sheet aggregation of the protein [62]. How this relates to a possible phosphorylation of the native silk proteins has not been studied. 6. Protein Structure Is Different in Solution and in Fibers We have not yet fully understood the mechanisms by which the high solubility in the spinning dopes is achieved. This feature has not been recapitulated in recombinant silk proteins or in regenerated silk, which likely is caused by the use of denaturing conditions, but additional factors may be involved as well. Both spidroins and fibroins have been hypothesized to be stored at high concentration in a soluble state in the form of protein micelles within their respective glands [21,34,55,63]. Another theory states that the dope is stored as a liquid crystalline phase [21]. These theories are not mutually exclusive and the exact state of storage of the highly aggregation prone proteins at high concentrations remains to be determined. The structure of soluble spidroins within the lumen of the major ampullate gland is believed to be mostly alpha helical and/or random coil and some studies show that in the end of the gland, beta sheets are starting to form [25,64]. In B. mori the poly-GA motifs in solution (silk I) contain repeated beta-turn structures as shown by both solid-state and solution NMR [65,66], while the repeat units containing tyrosines are in random coil conformation [67]. In both silkworm and spider silk fibers, the Ala-rich regions form crystalline beta sheet structures in the fiber, while the glycine-rich regions in SpRep and the linker regions in FibRep form a more amourphous and flexible strucure based on beta spirals and random coils [68]. In silkworm silk fibers (referred to as silk II), the main structures represented are beta sheet, distorted beta turns and distorted beta sheets [69,70]. It is still unclear whether or not the beta sheets in silk fibers are parallel or antiparallel [35,71,72]. The difficulties in conclusively showing a specific structure may be related to that the silk fibers are perhaps not as highly organized as we tend to believe. 7. Fiber Architecture B. mori fibers are 10–16 µm wide [73] while dragline silk fibers are significantly thinner, around 3–6 µm [74]. The B. mori fiber consists of two fibroin monofilaments originating from the two separate glands, with a coat of sericin [73,75]. The architecture of dragline silk fibers has been debated, and the fiber has been proposed to consist of two to five layers. A skin-core structure has been suggested [9,76,77,78], but also four-layered [79] or five-layered structures [80] have been proposed. The lack of conclusive data about the architecture of dragline silk probably relate to the fact that these studies involve extensive treatment of the silk, such as dehydrating, embedding in plastic, sectioning and staining [9,77,78], dipping silk fibers in urea and allowing them to supercontract [79] or treating fibers using ether, triton x-100 and freeze-thaw cycles, after which they were embedded and sectioned [80]. Most studies have, however, agreed upon the presence of micro- and nanofibrils in parallel to the fiber axis within the dragline silk [78,80,81,82,83], which is also the case for silkworm silk [75,83,84]. The presence of nano- and micro-fibrils within the fiber could be important for its tensile properties. Twisted microfibrils would significantly increase the toughness of the fiber, a feature which is frequently utilized in manmade fibers and ropes. Relaxation studies on native spider silk has pointed towards a torsional memory in dragline silk [85,86], which may in part explain the high toughness of spider silks. One aim of the production of artificial fibers has been to get fibers that are as homogenous as possible, an ambition that may be questionable since native dragline silk fibers are apparently quite heterogenous. A certain heterogeneity of the structure within the fiber may increase the strength of the fiber, while too much or too little heterogeneity could cause fibers to be less strong [87]. 8. Tensile Properties of Silk Fibers—Why Is Spider Silk Tougher? Fibers spun by spiders and silkworms both have a very high toughness, compared to other natural and manmade fibers, but spider silk is in general superior to silkworm silk. The stress/strain behaviors show that the dragline silk from Argiope trifasciata is much stronger (higher maximum stress before breaking) and also more extendible (higher strain) than B. mori fibers (Figure 3). There are likely several factors that lead to spider silk being tougher than silkworm silk, and several models of how the toughness is mediated have been put forward, e.g., [57,68,90,91,92]. One hypothesis states that the presence of poly-GA repeats in the repetitive part of fibroins, as compared to poly-A blocks present in spidroins, will create a lower binding strength of the beta sheets [68], which likely also affects the tensile properties. Hayashi and co-authors modelled that the poly-GA will form less stable beta sheets than poly-A since Gly lacks a side-chain that can mediate hydrophobic interactions [68]. Comparing the binding strength of potential beta sheets of SpRep and FibRep using the Zipper database [93], which can be used to predict amyloid fibril forming segments, one can see that the energies of the poly-GA motifs in the FibRep are much higher than those of the poly-A blocks in SpRep (Figure 4), which means that the poly-GA motifs are predicted to form much weaker beta sheets. This observation further argues that the nature of the repetitive parts to a large extent explains the lower toughness exhibited by silkworm silk. Neither the poly-GA nor the poly-A segments of the repetitive regions are intrinsically prone to forming beta sheets. In fact, Ala is the naturally occurring residue with the highest α-helical propensity and Gly is disfavoured in any secondary structure [94]. This may be a prerequisite for keeping the proteins in solution before spinning, and any other stretch of amino acid residues could pose a danger by being prone to aggregate prematurely [95]. 9. Conclusions and Future Perspectives By comparing silk spinning in two distantly related species, in which the ability to spin silk apparently evolved after they separated during evolution [5,6], we find some features that appear to be crucial for successful fiber formation. Firstly, the pH gradient is tightly controlled, is present in both silk production systems, and mediates dramatic secondary structure transitions of the silk proteins, via specific effects on the terminal domains; Secondly, the silk proteins are highly soluble at neutral pH. The high overall solubility may be attributed to the highly soluble terminal domains, but also the repetitive regions rich in alanine and glycine residues are water soluble, in contrast to stretches of hydrophobic residues [96]; Thirdly, silk proteins are large, >300 kDa, which may be necessary in order to obtain the great mechanical strength and extensibility displayed by the fibers, although alternative explanations to the large sizes, based on general genetic mechanisms, are also possible [97]. We believe that using a biomimetic approach, wherein silk proteins are not denatured throughout the purification and spinning processes, is vital for the generation of artificial silk fiber replicas with the same structure and toughness as native silk fibers. Supporting this are results from experiments on silkworm silk extracted from cocoons that are solubilized using high temperatures and lithium bromide [98] before spinning fibers, which results in silk that is very different in structure and inferior in toughness compared to native silkworm silk [88,99,100] likely due to the denaturation of the native silk proteins. In contrast to silkworms, spiders produce very small amounts of silk, and therefore largescale production of spider silk fibers requires expression in and purification from heterologous hosts. Expression of highly repetitive sequences such as the MaSp repetitive region is difficult in heterologous hosts, and therefore most constructs have been considerably smaller than the native spidroins, often including only a small part of the repetitive region and lacking one or both of the terminal domains. The water solubility levels of such constructs have still been very low (around 1% w/v [101,102]), and recombinant spidroins have therefore been dissolved in denaturing agents (reaching solubility levels of 5%–25% w/v [103,104,105]), after which spinning into tough fibers has proven difficult. In the same way that regenerated (denatured) silkworm silk cannot form fibers that are similar to native silk, it is highly unlikely that artificial fibers spun from denatured recombinant spidroins can capture the true structure and toughness of native spider silk. Although the physiology and biochemistry of silk spinning has been studied quite thoroughly over the past decade, there is apparently a lot that we still do not understand. For example, spidroins and fibroins produced in the epithelial cells are stored in secretory granula. How are these extremely pH sensitive and aggregation-prone proteins able to withstand the presumably quite low pH within the secretory granules without starting to assemble? What chaperone systems are involved in successfully transporting the silk proteins through the secretory pathway, a process that is demanding even for “normal” proteins [106]? We look forward to further detailed studies of silk proteins and the silk formation process, which can generate insights that have broad and important implications. Acknowledgments We thank Gustavo R. Plaza for providing the stress-strain graphs of dragline silk and B. mori silk fibers. Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1290/s1. Click here for additional data file. Author Contributions Marlene Andersson performed histology experiments; Marlene Andersson, Jan Johansson and Anna Rising analyzed data; Marlene Andersson, Jan Johansson and Anna Rising wrote the paper. Conflicts of Interest The authors declare no conflict of interest. Abbreviations PSG Posterior silk gland MSG Middle silk gland ASG Anterior silk gland CA Carbonic anhydrase Cryo-SEM-EDAX Cryo scanning electron microscope energy dispersive X-ray NT N-terminal domain CT C-terminal domain FibNT Heavy chain fibroin N-terminal domain FibCT Heavy chain fibroin C-terminal domain MaSp Major ampullate spidroin SpNT Spidroin N-terminal domain SpCT Spidroin C-terminal domain SpRep Spidroin repetitive region FibRep Heavy chain fibroin repetitive region PTM Post-translational modification NMR Nuclear magnetic resonance Figure 1 Macroscopic appearance of a B. mori silk gland with anterior silk gland (ASG), Funnel, middle silk gland (MSG) and posterior silk gland (PSG) identified (a) and a Major ampullate gland with Duct, Funnel, Sac and Tail indicated (b); Schematic image of a B. mori silk gland (c) and major ampullate gland (d) with pH values indicated in different parts. In (c) and (d) the regions containing epithelial cells with CA activity are shaded in grey, and fibroin/spidroin secreting parts are purple. Adapted from [8] (a,c) and [9,10] (b,d). Figure 2 Transmission electron micrograph of a cross-section of the third limb of the duct of an Euprosthenops australis major ampullate gland (a) and a hematoxylin-eosin stained light micrograph of a cross-section of the ASG from a B. mori silk gland (b). Mv: microvilli, Lu: lumen, CI: Cuticular intima. Scale bars (a) 2 µm (b) 15 µm. In (a) the microvilli appear detatched from the cuticular intima, likely due to processing of the section for transmission electron microscopy. Figure 3 Stress vs. strain curves of forcibly silked native dragline silk fibers from A. trifasciata (in red) and silk fibers from B. mori (in blue). Original data from [88,89]. Figure 4 Fibrillation propensity profiles of the first 100 amino acid residues of B. mori fibroin heavy chain repetitive part (AF226688.1) (top) and three repetitive blocks from E. australis MaSp1 (GenBank AM490183) (bottom). The amino acid residues are presented in one-letter format above each plot. The energies are color-coded, blue and green representing higher energies while orange and red represents lower energies that are deemed to have high fibrillation propensity. Profiles generated by ZipperDB [93]. ==== Refs References 1. Omenetto F.G. Kaplan D.L. New opportunities for an ancient material Science 2010 329 528 531 10.1126/science.1188936 20671180 2. Vollrath F. Barth P. Basedow A. Engstrom W. List H. Local tolerance to spider silks and protein polymers in vivo In Vivo 2002 16 229 234 12224131 3. Radtke C. Allmeling C. Waldmann K.H. Reimers K. Thies K. Schenk H.C. Hillmer A. Guggenheim M. Brandes G. Vogt P.M. Spider silk constructs enhance axonal regeneration and remyelination in long nerve defects in sheep PLoS ONE 2011 6 1290 10.1371/journal.pone.0016990 21364921 4. Allmeling C. Jokuszies A. Reimers K. Kall S. Choi C.Y. Brandes G. Kasper C. Scheper T. Guggenheim M. Vogt P.M. Spider silk fibres in artificial nerve constructs promote peripheral nerve regeneration Cell Prolif. 2008 41 408 420 10.1111/j.1365-2184.2008.00534.x 18384388 5. Sutherland T.D. Young J.H. Weisman S. Hayashi C.Y. Merritt D.J. Insect silk: One name, many materials Annu. Rev. Entomol. 2010 55 171 188 10.1146/annurev-ento-112408-085401 19728833 6. Sanggaard K.W. Bechsgaard J.S. Fang X. Duan J. Dyrlund T.F. Gupta V. Jiang X. Cheng L. Fan D. Feng Y. Spider genomes provide insight into composition and evolution of venom and silk Nat. Commun. 2014 5 3765 24801114 7. Asakura T. Umemura K. Nakazawa Y. Hirose H. Higham J. Knight D. Some observations on the structure and function of the spinning apparatus in the silkworm Bombyx mori Biomacromolecules 2007 8 175 181 10.1021/bm060874z 17206804 8. Domigan L.J. Andersson M. Alberti K.A. Chesler M. Xu Q. Johansson J. Rising A. Kaplan D.L. Carbonic anhydrase generates a pH gradient in Bombyx mori silk glands Insect Biochem. Mol. Biol. 2015 65 100 106 10.1016/j.ibmb.2015.09.001 26365738 9. Andersson M. Holm L. Ridderstrale Y. Johansson J. Rising A. Morphology and composition of the spider major ampullate gland and dragline silk Biomacromolecules 2013 14 2945 2952 10.1021/bm400898t 23837699 10. Andersson M. Chen G. Otikovs M. Landreh M. Nordling K. Kronqvist N. Westermark P. Jornvall H. Knight S. Ridderstrale Y. Carbonic anhydrase generates CO2 and H+ that drive spider silk formation via opposite effects on the terminal domains PLoS Biol. 2014 12 1290 10.1371/journal.pbio.1001921 25093327 11. Kovoor J. Iv comparative structure and histochemistry of silk-producing organs in arachnids Ecophysiology of Spiders Nentwig W. Springer Berlin, Germany 1987 160 186 12. Rising A. Johansson J. Toward spinning artificial spider silk Nat. Chem. Biol. 2015 11 309 315 10.1038/nchembio.1789 25885958 13. Peakall D.B. Synthesis of silk, mechanism and location Am. Zool. 1969 9 71 79 10.1093/icb/9.1.71 14. Gamo T. Inokuchi T. Laufer H. Polypeptides of fibroin and sericin secreted from the different sections of the silk gland in Bombyx mori Insect Biochem. 1977 7 285 295 10.1016/0020-1790(77)90026-9 15. Hijirida D.H. Do K.G. Michal C. Wong S. Zax D. Jelinski L.W. 13C NMR of Nephila clavipes major ampullate silk gland Biophys. J. 1996 71 3442 3447 10.1016/S0006-3495(96)79539-5 8968613 16. Vollrath F. Knight D.P. Liquid crystalline spinning of spider silk Nature 2001 410 541 548 10.1038/35069000 11279484 17. Magoshi J. Magoshi Y. Nakamura S. Mechanism of fiber formation of silkworm Silk Polymers American Chemical Society Saint Louis, MO, USA 1993 292 310 18. Davies G.J. Knight D.P. Vollrath F. Structure and function of the major ampullate spinning duct of the golden orb weaver, Nephila edulis Tissue Cell 2013 45 306 311 10.1016/j.tice.2013.04.001 23664309 19. Vollrath F. Knight D.P. Structure and function of the silk production pathway in the spider Nephila edulis Int. J. Biol. Macromol. 1999 24 243 249 10.1016/S0141-8130(98)00095-6 10342771 20. Breslauer D.N. Lee L.P. Muller S.J. Simulation of flow in the silk gland Biomacromolecules 2009 10 49 57 10.1021/bm800752x 19053289 21. Knight D.P. Vollrath F. Liquid crystals and flow elongation in a spider’s silk production line Proc. R. Soc. Lond. 1999 266 519 523 10.1098/rspb.1999.0667 22. Foo C.W.P. Bini E. Hensman J. Knight D.P. Lewis R.V. Kaplan D.L. Role of pH and charge on silk protein assembly in insects and spiders Appl. Phys. Mater. Sci. Process. 2006 82 223 233 10.1007/s00339-005-3426-7 23. Miyake S. Azuma M. Acidification of the silk gland lumen in Bombyx mori and samia cynthia ricini and localization of a H+ -translocating vacuolar-type ATPpase J. Insect Biotechnol. Sericol. 2008 77 9 16 24. Knight D.P. Vollrath F. Changes in element composition along the spinning duct in a nephila spider Naturwissenschaften 2001 88 179 182 10.1007/s001140100220 11480706 25. Dicko C. Vollrath F. Kenney J.M. Spider silk protein refolding is controlled by changing pH Biomacromolecules 2004 5 704 710 10.1021/bm034307c 15132650 26. Azuma M. Ohta Y. Changes in H+ -translocating vacuolar-type atpase in the anterior silk gland cell of Bombyx mori during metamorphosis J. Exp. Biol. 1998 201 479 486 9438824 27. Vollrath F. Knight D.P. Hu X.W. Silk production in a spider involves acid bath treatment Proc. R. Soc. Lond. 1998 265 817 820 10.1098/rspb.1998.0365 28. Zhou L. Chen X. Shao Z. Zhou P. Knight D.P. Vollrath F. Copper in the silk formation process of Bombyx mori silkworm FEBS Lett. 2003 554 337 341 10.1016/S0014-5793(03)01184-0 14623090 29. Zhou L. Chen X. Shao Z. Huang Y. Knight D.P. Effect of metallic ions on silk formation in the mulberry silkworm, Bombyx mori J. Phys. Chem. B 2005 109 16937 16945 10.1021/jp050883m 16853155 30. Knight D.P. Knight M.M. Vollrath F. Beta transition and stress-induced phase separation in the spinning of spider dragline silk Int. J. Biol. Macromol. 2000 27 205 210 10.1016/S0141-8130(00)00124-0 10828366 31. Moriya M. Roschzttardtz F. Nakahara Y. Saito H. Masubuchi Y. Asakura T. Rheological properties of native silk fibroins from domestic and wild silkworms, and flow analysis in each spinneret by a finite element method Biomacromolecules 2009 10 929 935 10.1021/bm801442g 19317399 32. Inoue S. Tanaka K. Arisaka F. Kimura S. Ohtomo K. Mizuno S. Silk fibroin of Bombyx mori is secreted, assembling a high molecular mass elementary unit consisting of H-chain, L-chain, and P25, with a 6:6:1 molar ratio J. Biol. Chem. 2000 275 40517 40528 10.1074/jbc.M006897200 10986287 33. Zhou C.Z. Confalonieri F. Medina N. Zivanovic Y. Esnault C. Yang T. Jacquet M. Janin J. Duguet M. Perasso R. Fine organization of Bombyx mori fibroin heavy chain gene Nucleic Acids Res. 2000 28 2413 2419 10.1093/nar/28.12.2413 10871375 34. He Y.X. Zhang N.N. Li W.F. Jia N. Chen B.Y. Zhou K. Zhang J. Chen Y. Zhou C.Z. N-terminal domain of Bombyx mori fibroin mediates the assembly of silk in response to pH decrease J. Mol. Biol. 2012 418 197 207 10.1016/j.jmb.2012.02.040 22387468 35. Zhou C.Z. Confalonieri F. Jacquet M. Perasso R. Li Z.G. Janin J. Silk fibroin: Structural implications of a remarkable amino acid sequence Proteins 2001 44 119 122 10.1002/prot.1078 11391774 36. Mori K. Tanaka K. Kikuchi Y. Waga M. Waga S. Mizuno S. Production of a chimeric fibroin light-chain polypeptide in a fibroin secretion-deficient naked pupa mutant of the silkworm Bombyx mori J. Mol. Biol. 1995 251 217 228 10.1006/jmbi.1995.0429 7643398 37. Long D. Lu W. Zhang Y. Guo Q. Xiang Z. Zhao A. New insight into the mechanism underlying fibroin secretion in silkworm, Bombyx mori FEBS J. 2015 282 89 101 10.1111/febs.13105 25302556 38. Chen W.Q. Priewalder H. John J.P. Lubec G. Silk cocoon of Bombyx mori : Proteins and posttranslational modifications—Heavy phosphorylation and evidence for lysine-mediated cross links Proteomics 2010 10 369 379 10.1002/pmic.200900624 20029844 39. Rising A. Hjalm G. Engstrom W. Johansson J. N-terminal nonrepetitive domain common to dragline, flagelliform, and cylindriform spider silk proteins Biomacromolecules 2006 7 3120 3124 10.1021/bm060693x 17096540 40. Chen G. Liu X. Zhang Y. Lin S. Yang Z. Johansson J. Rising A. Meng Q. Full-length minor ampullate spidroin gene sequence PLoS ONE 2012 7 1290 10.1371/journal.pone.0052293 23251707 41. Askarieh G. Hedhammar M. Nordling K. Saenz A. Casals C. Rising A. Johansson J. Knight S.D. Self-assembly of spider silk proteins is controlled by a pH-sensitive relay Nature 2010 465 236 238 10.1038/nature08962 20463740 42. Gaines W.A. Sehorn M.G. Marcotte W.R. Jr. Spidroin N-terminal domain promotes a pH-dependent association of silk proteins during self-assembly J. Biol. Chem. 2010 285 40745 40753 10.1074/jbc.M110.163121 20959449 43. Kronqvist N. Otikovs M. Chmyrov V. Chen G. Andersson M. Nordling K. Landreh M. Sarr M. Jornvall H. Wennmalm S. Sequential pH-driven dimerization and stabilization of the N-terminal domain enables rapid spider silk formation Nat. Commun. 2014 5 3254 10.1038/ncomms4254 24510122 44. Otikovs M. Chen G. Nordling K. Landreh M. Meng Q. Jornvall H. Kronqvist N. Rising A. Johansson J. Jaudzems K. Diversified structural basis of a conserved molecular mechanism for pH-dependent dimerization in spider silk n-terminal domains Chembiochem 2015 16 1720 1724 10.1002/cbic.201500263 26033527 45. Gauthier M. Leclerc J. Lefevre T. Gagne S.M. Auger M. Effect of pH on the structure of the recombinant C-terminal domain of Nephila clavipes dragline silk protein Biomacromolecules 2014 15 4447 4454 10.1021/bm501241n 25337802 46. Hagn F. Eisoldt L. Hardy J.G. Vendrely C. Coles M. Scheibel T. Kessler H. A conserved spider silk domain acts as a molecular switch that controls fibre assembly Nature 2010 465 239 242 10.1038/nature08936 20463741 47. Maurer-Stroh S. Debulpaep M. Kuemmerer N. Lopez de la Paz M. Martins I.C. Reumers J. Morris K.L. Copland A. Serpell L. Serrano L. Exploring the sequence determinants of amyloid structure using position-specific scoring matrices Nat. Methods 2010 7 237 242 10.1038/nmeth.1432 20154676 48. Tjernberg L.O. Rising A. Johansson J. Jaudzems K. Westermark P. Transmissible amyloid (Review Symposium) J. Intern. Med. 2016 10.1111/joim.12499 27002185 49. Fernandez-Escamilla A.M. Rousseau F. Schymkowitz J. Serrano L. Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins Nat. Biotechnol. 2004 22 1302 1306 10.1038/nbt1012 15361882 50. Linding R. Schymkowitz J. Rousseau F. Diella F. Serrano L. A comparative study of the relationship between protein structure and β-aggregation in globular and intrinsically disordered proteins J. Mol. Biol. 2004 342 345 353 10.1016/j.jmb.2004.06.088 15313629 51. Rousseau F. Schymkowitz J. Serrano L. Protein aggregation and amyloidosis: Confusion of the kinds? Curr. Opin. Struct. Biol. 2006 16 118 126 10.1016/j.sbi.2006.01.011 16434184 52. Xu M. Lewis R.V. Structure of a protein superfiber: Spider dragline silk Proc. Natl. Acad Sci. USA 1990 87 7120 7124 10.1073/pnas.87.18.7120 2402494 53. Hinman M. Dong Z. Xu M. Lewis R.V. Spider silk: A mystery starting to unravel Results Probl. Cell Differ. 1992 19 227 254 1289994 54. Rising A. Controlled assembly: A prerequisite for the use of recombinant spider silk in regenerative medicine? Acta Biomater. 2014 10 1627 1631 10.1016/j.actbio.2013.09.030 24090990 55. Eisoldt L. Thamm C. Scheibel T. Review the role of terminal domains during storage and assembly of spider silk proteins Biopolymers 2012 97 355 361 10.1002/bip.22006 22057429 56. Scheibel T. Spider silks: Recombinant synthesis, assembly, spinning, and engineering of synthetic proteins Microb. Cell Factories 2004 3 14 10.1186/1475-2859-3-14 15546497 57. Ebrahimi D. Tokareva O. Rim N.G. Wong J.Y. Kaplan D.L. Buehler M.J. Silk-its mysteries, how it is made, and how it is used ACS Biomater. Sci. Eng. 2015 1 864 876 10.1021/acsbiomaterials.5b00152 27398402 58. dos Santos-Pinto J.R. Lamprecht G. Chen W.Q. Heo S. Hardy J.G. Priewalder H. Scheibel T.R. Palma M.S. Lubec G. Structure and post-translational modifications of the web silk protein spidroin-1 from nephila spiders J. Proteom. 2014 105 174 185 10.1016/j.jprot.2014.01.002 24434585 59. Dos Santos-Pinto J.R.A. Arcuri H.A. Lubec G. Palma M.S. Structural characterization of the major ampullate silk spidroin-2 protein produced by the spider Nephila clavipes Biochim. Biophys. Acta 2016 10.1016/j.bbapap.2016.05.007 27208434 60. Zhang P. Aso Y. Yamamoto K. Banno Y. Wang Y. Tsuchida K. Kawaguchi Y. Fujii H. Proteome analysis of silk gland proteins from the silkworm, Bombyx mori Proteomics 2006 6 2586 2599 10.1002/pmic.200500348 16548058 61. Michal C.A. Simmons A.H. Chew B.G. Zax D.B. Jelinski L.W. Presence of phosphorus in Nephila clavipes dragline silk Biophys. J. 1996 70 489 493 10.1016/S0006-3495(96)79594-2 8770226 62. Winkler S. Wilson D. Kaplan D.L. Controlling β-sheet assembly in genetically engineered silk by enzymatic phosphorylation/dephosphorylation Biochemistry 2000 39 12739 12746 10.1021/bi001335w 11027155 63. Jin H.J. Kaplan D.L. Mechanism of silk processing in insects and spiders Nature 2003 424 1057 1061 10.1038/nature01809 12944968 64. Kenney J.M. Knight D. Wise M.J. Vollrath F. Amyloidogenic nature of spider silk Eur. J. Biochem. 2002 269 4159 4163 10.1046/j.1432-1033.2002.03112.x 12180993 65. Asakura T. Ashida J. Yamane T. Kameda T. Nakazawa Y. Ohgo K. Komatsu K. A repeated β-turn structure in Poly(Ala-Gly) as a model for silk I of Bombyx mori silk fibroin studied with two-dimensional spin-diffusion NMR under off magic angle spinning and rotational echo double resonance J. Mol. Biol. 2001 306 291 305 10.1006/jmbi.2000.4394 11237601 66. Suzuki Y. Yamazaki T. Aoki A. Shindo H. Asakura T. NMR study of the structures of repeated sequences, GAGXGA (X = S, Y, V), in Bombyx mori liquid silk Biomacromolecules 2014 15 104 112 10.1021/bm401346h 24266784 67. Asakura T. Yamane T. Nakazawa Y. Kameda T. Ando K. Structure of Bombyx mori silk fibroin before spinning in solid state studied with wide angle X-ray scattering and (13)C cross-polarization/magic angle spinning nmr Biopolymers 2001 58 521 525 10.1002/1097-0282(20010415)58:5<521::AID-BIP1027>3.0.CO;2-T 11241223 68. Hayashi C.Y. Shipley N.H. Lewis R.V. Hypotheses that correlate the sequence, structure, and mechanical properties of spider silk proteins Int. J. Biol. Macromol. 1999 24 271 275 10.1016/S0141-8130(98)00089-0 10342774 69. Asakura T. Yao J. 13C CP/MAS NMR study on structural heterogeneity in Bombyx mori silk fiber and their generation by stretching Protein Sci. 2002 11 2706 2713 10.1110/ps.0221702 12381852 70. Asakura T. Yao J. Yamane T. Umemura K. Ulrich A.S. Heterogeneous structure of silk fibers from Bombyx mori resolved by 13C solid-state NMR spectroscopy J. Am. Chem. Soc. 2002 124 8794 8795 10.1021/ja020244e 12137522 71. Yao J. Ohgo K. Sugino R. Kishore R. Asakura T. Structural analysis of Bombyx mori silk fibroin peptides with formic acid treatment using high-resolution solid-state 13C NMR spectroscopy Biomacromolecules 2004 5 1763 1769 10.1021/bm049831d 15360285 72. Rousseau M.E. Lefevre T. Beaulieu L. Asakura T. Pezolet M. Study of protein conformation and orientation in silkworm and spider silk fibers using raman microspectroscopy Biomacromolecules 2004 5 2247 2257 10.1021/bm049717v 15530039 73. Zhao H.-P. Feng X.-Q. Shi H.-J. Variability in mechanical properties of Bombyx mori silk Mater. Sci. Eng. C 2007 27 675 683 10.1016/j.msec.2006.06.031 74. Pérez-Rigueiro J. Elices M. Llorca J. Viney C. Tensile properties of argiope trifasciata drag line silk obtained from the spider´s web J. Appl. Polym. Sci. 2001 82 2245 2251 10.1002/app.2072 75. Poza P. Pérez-Rigueiro J. Elices M. Llorca J. Fractographic analysis of silkworm and spider silk Eng. Fract. Mech. 2002 69 1035 1048 10.1016/S0013-7944(01)00120-5 76. Work R.W. Duality in major ampullate silk fiber formation by Orb-web-spinning spiders Trans. Am. Microsc. Soc. 1984 103 113 121 10.2307/3226233 77. Frische S. Maunsbach A.B. Vollrath F. Elongate cavities and skin-core structure in nephila spider silk osberved by electron microscopy J. Microsc. 1998 189 64 70 10.1046/j.1365-2818.1998.00285.x 78. Augsten K. Muhlig P. Herrmann C. Glycoproteins and skin-core structure in Nephila clavipes spider silk observed by light and electron microscopy Scanning 2000 22 12 15 10.1002/sca.4950220103 10768384 79. Vollrath F. Holtet T. Thogersen H.C. Frische S. Structural organization of spider silk Proc. R. Soc. Lond. 1996 263 147 151 10.1098/rspb.1996.0023 80. Sponner A. Vater W. Monajembashi S. Unger E. Grosse F. Weisshart K. Composition and hierarchical organisation of a spider silk PLoS ONE 2007 2 1290 10.1371/journal.pone.0000998 17912375 81. Gould S.A. Tran K.T. Spagna J.C. Moore A.M. Shulman J.B. Short and long range order of the morphology of silk from latrodectus hesperus (black widow) as characterized by atomic force microscopy Int. J. Biol. Macromol. 1999 24 151 157 10.1016/S0141-8130(99)00003-3 10342759 82. Li S.F. McGhie A.J. Tang S.L. New internal structure of spider dragline silk revealed by atomic force microscopy Biophys. J. 1994 66 1209 1212 10.1016/S0006-3495(94)80903-8 8038392 83. Miller L.D. Putthanarat S. Eby R.K. Adams W.W. Investigation of the nanofibrillar morphology in silk fibers by small angle X-ray scattering and atomic force microscopy Int. J. Biol. Macromol. 1999 24 159 165 10.1016/S0141-8130(99)00024-0 10342760 84. Hakimi O. Knight D.P. Knight M.M. Grahn M.F. Vadgama P. Ultrastructure of insect and spider cocoon silks Biomacromolecules 2006 7 2901 2908 10.1021/bm060528h 17025368 85. Emile O. Le Floch A. Vollrath F. Time-resolved torsional relaxation of spider draglines by an optical technique Phys. Rev. Lett. 2007 98 167402 10.1103/PhysRevLett.98.167402 17501459 86. Emile O. Le Floch A. Vollrath F. Biopolymers: Shape memory in spider draglines Nature 2006 440 621 10.1038/440621a 16572162 87. Cranford S.W. Increasing silk fibre strength through heterogeneity of bundled fibrils J. R. Soc. Interface 2013 10 20130148 10.1098/rsif.2013.0148 23486175 88. Plaza G.R. Corsini P. Perez-Rigueiro J. Marsano E. Guinea G.V. Elices M. Effect of water on Bombyx mori regenerated silk fibers and its application in modifying their mechanical properties J. Appl. Polym. Sci. 2008 109 1793 1801 10.1002/app.28288 89. Plaza G.R. Perez-Rigueiro J. Riekel C. Perea G.B. Agullo-Rueda F. Burghammer M. Guinea G.V. Elices M. Relationship between microstructure and mechanical properties in spider silk fibers: Identification of two regimes in the microstructural changes Soft Matter 2012 8 6015 6026 10.1039/c2sm25446h 90. Guan J. Wang Y. Mortimer B. Holland C. Shao Z. Porter D. Vollrath F. Glass transitions in native silk fibres studied by dynamic mechanical thermal analysis Soft Matter 2016 12 5926 5936 10.1039/C6SM00019C 27320178 91. Porter D. Guan J. Vollrath F. Spider silk: Super material or thin fibre? Adv. Mater. 2013 25 1275 1279 10.1002/adma.201204158 23180482 92. Lee M. Kwon J. Na S. Mechanical behavior comparison of spider and silkworm silks using molecular dynamics at atomic scale Phys. Chem. Chem. Phys. 2016 18 4814 4821 10.1039/C5CP06809F 26806791 93. Goldschmidt L. Teng P.K. Riek R. Eisenberg D. Identifying the amylome, proteins capable of forming amyloid-like fibrils Proc. Natl. Acad. Sci. USA 2010 107 3487 3492 10.1073/pnas.0915166107 20133726 94. Johansson J. Nerelius C. Willander H. Presto J. Conformational preferences of non-polar amino acid residues: An additional factor in amyloid formation Biochem. Biophys. Res. Commun. 2010 402 515 518 10.1016/j.bbrc.2010.10.062 20971069 95. Johansson J. Harnessing the self-assembling properties of proteins in spider silk and lung surfactant, in Amyloid Fibrils and Prefibrillar Aggregates Molecular and Biological Properties Otzen D.E. Wiley-VCH Verlag Weinheim, Germany 2013 455 470 96. Hessa T. Kim H. Bihlmaier K. Lundin C. Boekel J. Andersson H. Nilsson I. White S.H. von Heijne G. Recognition of transmembrane helices by the endoplasmic reticulum translocon Nature 2005 433 377 381 10.1038/nature03216 15674282 97. Rising A. Johansson J. Larson G. Bongcam-Rudloff E. Engstrom W. Hjalm G. Major ampullate spidroins from euprosthenops australis: Multiplicity at protein, mRNA and gene levels Insect Mol. Biol. 2007 16 551 561 10.1111/j.1365-2583.2007.00749.x 17680798 98. Rockwood D.N. Preda R.C. Yucel T. Wang X. Lovett M.L. Kaplan D.L. Materials fabrication from Bombyx mori silk fibroin Nat. Protoc. 2011 6 1612 1631 10.1038/nprot.2011.379 21959241 99. Hossain K.S. Ochi A. Ooyama E. Magoshi J. Nemoto N. Dynamic light scattering of native silk fibroin solution extracted from different parts of the middle division of the silk gland of the Bombyx mori silkworm Biomacromolecules 2003 4 350 359 10.1021/bm020109u 12625731 100. Koebley S.R. Thorpe D. Pang P. Chrisochoides P. Greving I. Vollrath F. Schniepp H.C. Silk reconstitution disrupts fibroin self-assembly Biomacromolecules 2015 16 2796 2804 10.1021/acs.biomac.5b00732 26284914 101. Stark M. Grip S. Rising A. Hedhammar M. Engstrom W. Hjalm G. Johansson J. Macroscopic fibers self-assembled from recombinant miniature spider silk proteins Biomacromolecules 2007 8 1695 1701 10.1021/bm070049y 17402782 102. Xu L. Rainey J.K. Meng Q. Liu X.Q. Recombinant minimalist spider wrapping silk proteins capable of native-like fiber formation PLoS ONE 2012 7 1290 10.1371/journal.pone.0050227 23209681 103. Zhang H. Zhou F. Jiang X. Cao M. Wang S. Zou H. Cao Y. Xian M. Liu H. Microbial production of amino acid-modified spider dragline silk protein with intensively improved mechanical properties Prep. Biochem. Biotechnol. 2015 10.1080/10826068.2015.1084637 26460683 104. Lin Z. Deng Q. Liu X.Y. Yang D. Engineered large spider eggcase silk protein for strong artificial fibers Adv. Mater. 2013 25 1216 1220 10.1002/adma.201204357 23172740 105. Copeland C.G. Bell B.E. Christensen C.D. Lewis R.V. Development of a process for the spinning of synthetic spider silk ACS Biomater. Sci. Eng. 2015 1 577 584 10.1021/acsbiomaterials.5b00092 27064312 106. Dolfe L. Winblad B. Johansson J. Presto J. Brichos binds to a designed amyloid-forming β-protein and reduces proteasomal inhibition and aggresome formation Biochem. J. 2016 473 167 178 10.1042/BJ20150920 26578816

PMC005xxxxxx/PMC5000688.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081291ijms-17-01291ArticleA Genome-Wide Methylation Approach Identifies a New Hypermethylated Gene Panel in Ulcerative Colitis Kang Keunsoo 1Bae Jin-Han 2Han Kyudong 34Kim Eun Soo 5Kim Tae-Oh 6*Yi Joo Mi 2*Ciccodicola Alfredo Academic Editor1 Department of Microbiology, Dankook University, Cheonan 31116, Korea; [email protected] Research Center, Dongnam Institute of Radiological & Medical Sciences (DIRAMS), Busan 46033, Korea; [email protected] Department of Nanobiomedical Science Global Research Center for Regenerative Medicine, Dankook University, Cheonan 31116, Korea; [email protected] DKU-Theragen Institute for NGS Analysis (DTiNa), Dankook University; Cheonan 31116, Korea5 Division of Gastroenterology, Department of Internal Medicine, Kyungpook National University School of Medicine, Daegu 41931, Korea; [email protected] Department of Internal Medicine, Haeundae Paik Hospital, Inje University College of Medicine, Busan 48108, Korea* Correspondence: [email protected] (T.-O.K.); [email protected] (J.M.Y.); Tel.: +82-51-797-0200 (T.-O.K.); +82-51-720-5139 (J.M.Y.)09 8 2016 8 2016 17 8 129121 6 2016 03 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The cause of inflammatory bowel disease (IBD) is still unknown, but there is growing evidence that environmental factors such as epigenetic changes can contribute to the disease etiology. The aim of this study was to identify newly hypermethylated genes in ulcerative colitis (UC) using a genome-wide DNA methylation approach. Using an Infinium HumanMethylation450 BeadChip array, we screened the DNA methylation changes in three normal colon controls and eight UC patients. Using these methylation profiles, 48 probes associated with CpG promoter methylation showed differential hypermethylation between UC patients and normal controls. Technical validations for methylation analyses in a larger series of UC patients (n = 79) were performed by methylation-specific PCR (MSP) and bisulfite sequencing analysis. We finally found that three genes (FAM217B, KIAA1614 and RIBC2) that were significantly elevating the promoter methylation levels in UC compared to normal controls. Interestingly, we confirmed that three genes were transcriptionally silenced in UC patient samples by qRT-PCR, suggesting that their silencing is correlated with the promoter hypermethylation. Pathway analyses were performed using GO and KEGG databases with differentially hypermethylated genes in UC. Our results highlight that aberrant hypermethylation was identified in UC patients which can be a potential biomarker for detecting UC. Moreover, pathway-enriched hypermethylated genes are possibly implicating important cellular function in the pathogenesis of UC. Overall, this study describes a newly hypermethylated gene panel in UC patients and provides new clinical information that can be used for the diagnosis and therapeutic treatment of IBD. DNA methylation profilepromoter hypermethylationulcerative colitisbiomarker ==== Body 1. Introduction Inflammatory bowel disease (IBD) is a chronic, relapsing, remitting, and/or continuously active disease of the gastrointestinal tract that is occasionally associated with extra-intestinal manifestations [1]. IBD can be classified into Crohn’s disease (CD) and ulcerative colitis (UC). Each type exhibits distinct etiologies and clinical features [2]. Although the exact etiology of IBD remains unknown, numerous clinical and experimental reports including genome-wide association studies have suggested that IBD is a consequence of the complex, dysregulated interplay between genetic predispositions, environmental factors, and microbial compositions in the intestine [3,4,5,6]. These genetic studies have suggested that genetic discoveries can help understand the susceptibility to IBD, although environmental factors are also important in its pathogenesis. Regarding the interaction between the environment and genome, epigenetic mechanisms may contribute to the pathogenesis of IBD. Epigenetics is a widely investigated field of cancer biology. It deals with the regulation of gene expression, primarily through DNA methylation, miRNA or histone modifications such as acetylation, methylation, phosphorylation, or ubiquitination. DNA methylation is the most extensively studied epigenetic modification in mammals. Dysregulation of epigenetic processes can lead to altered gene function and malignant cellular transformation. During the last decade, DNA methylation has been the most studied epigenetic modification. It has been correlated with transcriptional gene silencing in human diseases, including its well established role in cancer [7]. Although numerous studies have associated DNA methylation with cancer development and dysplasia, it was only recently that DNA methylation has been implicated in the pathogenesis of IBD. To understand the relationship of hypermethylated genes with the pathogenesis and clinical aspects of IBD, most reports emphasized the studies of UC patients or IBD-associated colon cancers. However, there has been a lack of studies on genome-wide DNA methylation alterations in IBD patients. Genome-wide DNA methylation profiling technology with the Illumina Human Methylation 450K array facilitates query of >450,000 loci within the genome and to cover 99% of RefSeq genes [8]. This improved technology permits a more powerful and comprehensive analysis of DNA methylation changes. In the present study, we used this technology to identify the locus or gene-specific DNA methylation changes in UC patients. Identification of molecular pathways with aberrant epigenetic regulation by promoter hypermethylation could provide novel methylation biomarkers for UC and possibly suggest new interventions for therapeutic treatments for IBD disease. 2. Results 2.1. Genome-Wide DNA Methylation Changes in UC Patients Promoter CpG island hypermethylation of tumor suppressor genes is a common hallmark of various human cancers [9]. This epigenetic event can affect all cellular signaling pathways that contribute to tumorigenesis. The biological and clinical importance of hypermethylation-mediated gene silencing has been extensively studied in other human diseases such as developmental diseases, and recently in inflammatory diseases [10,11]. We have previously reported that several genes that are specifically hypermethylated in colon cancer are also hypermethylated in UC patients [12]. Although this observation suggested that DNA methylation might be useful as a diagnostic or prognostic marker for patients with UC, no IBD direct factors altered by epigenetic regulation have been reported so far. To identify novel DNA methylation markers mostly occurring in the CpG island, which can be used to distinguish UC patients and normal controls, we screened genome-wide methylation patterns of colon samples from control subjects (n = 3) and UC patients (n = 8) using the Human Methylation 450K BeadChip array (Illumina). The probe call rate was >99% for all samples and 454,215 CpG sites out of 485,577 were included in the analysis. The methylation level at the CpG site is measured by means of a continuous variable β-value. A value of 0 indicates a fully unmethylated site while a value of 1 indicates a fully methylated site. After eliminating unreliable probes (difference of β-values < 0.2), we identified a large number of probes, which showed significant changes in DNA methylation between UC patients and normal colons (Figure 1A). Among the probes, the majority of them (n = 4397) were hypermethylated, while 420 probes were hypomethylated in UC samples when compared with normal colon samples. Notably, most of the differentially methylated CpG sites were located in the CpG islands (Figure 1B). According to gene annotation, intron, intergenic and promoter regions were the major parts of the genome that contained susceptible CpG sites specific to UC patients. To identify relevant increasing methylation level in UC patients, the identified probes were further filtered with a strict criteria (>1.7-fold change of β-values in UC patients compared to normal colon). Unsupervised hierarchical cluster analyses distinguished normal colon and UC patient tissues. A total of 237 hypermethylated probes were identified that represented hypermethylation patterns in UC patients when compare with normal colons (Figure 1C). We used HCT116 colon cancer cells as a positive control for global methylation analysis. As expected, all of the candidates were hypermethylated in HCT116 colon cancer cells. Collectively, we identified 237 CpG sites that were significantly hypermethylated in UC patients compared to a normal colon. 2.2. Validation of Selected Candidate Genes in UC Patient Samples As shown in Figure 1C, 48 out of 237 probes have typical CpG islands, defined as 200 bp sequences containing greater than 50% CpG dinucleotides [13], in their promoter regions, which could regulate their transcriptional gene expression. The gene list is summarized in Table S1. To identify newly hypermethylated genes in UC patients, we performed the validation for these genes using MSP and bisulfite sequencing analyses in a large series of UC samples (n = 79) (Table 1). We designed primers for methylation analysis for all 45 genes (Table S2). Since our methylation profile showed candidate genes were relatively hypermethylated pattern in HCT116 cells, we therefore first tested the methylation levels of these genes in HCT116 cells by MSP (Figure 1C). We confirmed that 26 out of 45 genes were hypermethylated in HCT116 cells and the rest of them were illuminated. To filter down genes are increasing methylation level in UC compare to normal controls, the following experimental validation criteria was based on our previous studies were used [14,15]: (i) the presence of gene expression in normal colon tissue; (ii) low or no methylation in normal colon tissues; and (iii) the presence of methylation in primary UC samples. The three best candidate genes, including RIB43A domain with coiled-coiles 2 (RIBC2), family with sequence similarity 217 mermber B (FAM217B) and KIAA1614, fulfilled the above criteria on our validation criteria. We performed massive MSP analysis for this validation and found that all three genes, KIAA1614, RIBC2, and FAM217B, showed 91%, 64% and 62% methylation in the UC samples, respectively (n = 79) (Figure 2). We next confirmed the CpG island methylation level of the three genes via bisulfite sequencing analysis in representative UC samples along with normal colon tissues (Figure 3). Notably, all three genes were more densely methylated in UC samples compared to normal colon tissues. Moreover, we also verified the methylation levels of the three genes in representative UC patient samples using quantitative MSP analysis (Figure 4A), suggesting that the methylation levels are increased in UC samples (n = 8) compare to normal colon tissues (n = 8). Overall, these results strongly suggested that the final candidate genes showed increasing DNA methylation levels in UC samples compare with normal controls. Next, we investigated whether hypermethylation of these genes correlated with downregulation of transcriptional gene expression levels in UC samples. Quantitative RT-PCR (qRT-PCR) showed that all three genes were significantly downregulated in representative UC samples (n = 15) when compared with normal control samples (n = 10) (Figure 4B). Combining the results of methylation and transcriptional gene expression patterns suggested that hypermethylation of the candidate gene panel was correlated to their transcriptional repression in UC samples. 2.3. Functional Implications of the UC Hypermethylated Gene Panel It has been suggested that abnormal signaling pathways play an important role in the inflammatory processes leading to dysregulation of the inflammatory responses that are involved in the pathogenesis of IBD [16]. In the present study, we investigated how these hypermethylated genes were associated with the cellular pathway network. Using our hypermethylated gene panel, we performed gene ontology analysis with well curated databases (e.g., KEGG) including a variety biological processes, molecular functions, and cellular components. Redundant terms were removed using the REVIGO application [17]. The most significant pathway affected by the hypermethylated genes was related to tissue morphogenesis (Figure 5). The ANKRD1, HES1, HGF, LHX1, LRP5, NRP1, PRICKLE1, PROX1, SOX11, SRF, TGFBR3 and TPM1 genes were associated with the pathway. In addition, integrin activation-related genes including CXCL13, PIEZO1 and PTGER4 became hypermethylated in UC. Further investigations of those genes are required to unveil the relationship between the hypermethylated genes and ulcerative colitis (Figure 5). 3. Discussion The etiology of UC sometimes involves changes in genetic loci, which causes over 16% of the cases of this disease [5]. Epigenetic modifications involving other factors may explain the remaining causes or risk of the disease [18,19,20]. There is increasing evidence that environmental factors regulate the epigenetic alterations and therefore contribute to disease susceptibility, manifestation, and progression [21]. Although multiple studies have found aberrant promoter methylation of a number of genes in human patients with UC [22], IBD-related DNA methylation studies have emphasized the development of colon cancer associated with IBD [14,23,24]. To specifically identify hypermethylated genes or loci associated with UC, recent technological advances have facilitated the assessment of global DNA methylation patterns of patients with UC. Very recently, 25 genes associated with inflammation have been differenetially methylated during inflammation process, which implicated in cancer development in UC [25]. However, very few epigenetic factors regulated by hypermethylation in a large cohort of UC patients have been reported correlated to the transcriptional repression. Here, we analyzed genome-wide methylation profiles of patients with UC using an Illumina Human Methylation 450K array. After identifying a subset of hypermethylated genes (n = 45) with a strict criteria of >1.7-fold β-values in patients with UC, we validated the methylation levels of these genes using multiple methylation analyses and correlations with transcriptional expressions in primary samples of UC patients by qRT-PCR. We finally identified three genes (FAM217B, KIAA1614 and RIBC2) that were hypermethylated in UC patient samples associated with transcriptional repression. The three genes were also strongly hypermethylated in most colon cancer cell lines we tested (Figure S1). None of these three genes have been previously shown to be hypermethylated in human diseases including UC or cancer, and a biological function in human diseases. To the best of our knowledge, this is the first study to report that the expression of these genes is regulated by promoter DNA hypermethylation in UC. Thus, the epigenetic effects by DNA hypermethylation could be useful for prognosis of UC as a molecular biomarker. Although we used a small set of UC patient samples for screening with a methylation array, we still provided promising candidate genes that showed frequently hypermethylated and significantly increasing the methylation level in UC patient samples. Epigenetic regulation of tumor suppressor genes by promoter CpG hypermethylation is well established as an important mechanism for gene inactivation [26]. In addition, epigenetic alterations have become established in recent years as one of the most important molecular signatures of human diseases. Our candidate genes can be linked to disruption of many cellular pathways involving DNA repair, apoptosis and the cell cycle, especially in cancer. Signaling pathway analyses with our candidate genes implicated hypermethylated genes in patients with UC, which might contribute to identifying key cellular pathways leading to inflammatory diseases (Figure 5). Taken together, these results provide the basis for using the methylation profile of UC patients to obtain a better understanding of the molecular pathways in the pathogenesis of UC. Numerous potential clinical applications of epigenetics for diagnostic and therapeutic applications have been reported [27]. DNA methylation is an attractive biomarker, because it has been related to many different clinical aspects, such as disease severity, duration, phenotype, inflammation and dysplasia. Recent advances in our understanding of IBD-associated DNA methylation provide many promising clinical applications, such as the use of molecular biomarkers for diagnosis and prognosis of the disease, as well as prediction of treatment outcomes. Karatzas et al. have observed the specific DNA methylation signature in UC and CD patients using EpiTect Methyl II Signature PCR Array profiles [28]. Other group have identified the differentially methylated locus in blood samples with CD using HumanMethylation 450 K BeadChip platform [29]. In addition, emerging epigenetic and genetic approach can be useful for improving potential application for IBD surveillance [30]. However, these applications were limited by the absence of suitable targets, because DNA methylation frequencies of many candidate genes are not high enough to be clinically used. Earlier reports have demonstrated that ESR-1 (Oestrogen receptor-1) and N-33 (tumor suppressor candidate-3) have increased their promoter methylation in patients with UC [30]. Moreover, it has recently been suggested that detecting methylation of FOXE1 and SYNE1 genes in colitis-associated colorectal neoplasia could be a promising biomarker [31]. However, both genes have shown lower methylation frequencies (less than 50%) in our samples (Figure S2). In this study, we show that three novel genes are frequently methylated (>60%) in a larger validation set, strongly suggesting that this gene panel could be a valuable methylation biomarker for prognosis or diagnosis of UC patients. More extensive studies will be necessary to validate this gene panel in a larger cohort of UC patients, which could strongly support its clinical application in the treatment of IBD. Because a large number of studies have reported the feasibility of detecting promoter hypermethylation of multiple genes in serum, stool, or body fluids in a broad spectrum of tumor types [32,33,34], further studies will be necessary to confirm that our candidate genes can be useful in screening serum or stool samples of UC patients. Although the clinical relevance of epigenetic aspects of IBD have not been thoroughly characterized, several studies have reported that epigenetic mechanisms are implicated in the pathogenesis of UC. For example, p16INK4a methylation was observed in regions negative for dysplasia, as well as during neoplastic progression in UC [23]. Furthermore, other studies have reported that promoter methylations of E-cadherin, CDH1 and GDNF are more frequently detected in long-standing UC [35,36]. We therefore investigated whether our hypermethylation profile was associated with clinical outcomes of UC patients, such as disease duration, location, or other clinical features. In contrast with previous studies, our methylation profile using 45 hypermethylated genes was associated with inflammatory activity. Clustering analyses suggested that UC patients with active mucosal inflammation, as assessed by endoscopy, displayed increasing methylation levels when compared with inactive mucosal inflammation. The relationship between samples using multiimentional scaling (MDS) analysis also supports that DNA methylation profile is associated to the mucosal inflammation activity (Figure S3A). Notably, we did not identify any correlations between the hypermethylation profiles and long-standing UC samples which had been reported by previous studies. While there was no correlation between disease duration of UC patients and hypermethylated genes, UC patients with active (UC 6 and 8) and inactive (UC 1, 3 and 5) mucosal inflammation were significantly divided by hierarchical analysis using CpG promoter hypermethylated genes, suggesting that our selective hypermethylated gene profiles were clinically relevant loci during UC development (Figure S3B). Our results showed that a global DNA methylation screening platform can identify a novel hypermethylated gene panel in patients with UC. Using various methylation analyses, the methylation status of candidate genes (FAM217B, KIAA1614 and RIBC2) was validated in a larger panel of UC patient samples. This gene hypermethylation pattern was also correlated with transcriptional silencing in the mRNA of UC patients. We therefore propose that this novel hypermethylated gene panel could be a valuable biomarker for diagnosis and prognosis of UC patients. In addition, the integrative DNA methylation profile of UC patients was associated with disruption of cellular pathways contributing to IBD pathogenesis, suggesting the potential importance of epigenetic mechanisms involving promoter hypermethylation in the modulation of IBD. Overall, our findings may lead to the use of DNA methylation data for novel clinical applications, diagnoses and treatments for IBD. 4. Materials and Methods 4.1. Tissue Samples Tissue samples were collected from the rectum at the time of colonoscopy. All patients had a confirmed diagnosis of UC based on their clinical symptoms, and endoscopic and pathological findings. The inflammatory status was classified as either active or inactive, based upon endoscopy. The main clinicopathological characteristics of the patients for experimental validation are described in Table 1. Normal colon samples (n = 8, median age: 36 years) as control were used for transcriptional expression and methylation analyses. The UC and normal colon biospecimens for this study were provided by the Keimyung Human Bio-Resource Bank (KHBB) and Inje Biobank, a member of the National Biobank of Korea, which is supported by the Ministry of Health and Welfare. This study was approved by the respective institutioal review board of the participating institutions of the National Biobank of Korea and written, informed consent was obtained for all study participants prior to data collection. 4.2. Ethic Statement This study was approved by the respective institutional review board of the participating institutions of the National Biobank of Korea and written, informed consent was obtained for all study participants prior to data collection. All samples or specimen derived from the Inje Biobank were obtained with informed consent under the institutional review board (IRB)-approved protocols (129792-2014-012). 4.3. Genome-Wide DNA Methylation Analysis DNA was extracted from three normal colon (NC) and eight UC patients, as well as one HCT116 cell line. Genome-wide DNA methylation was assessed using the Infinium Human Methylation 450K BeadChip kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. It contained over 480,000 CpGs which covered approximately 2% of the CpGs in the human genome. Preprocessing of raw data was conducted using the GenomeStudio application by Macrogen (Macrogen). Background corrections and dye bias equalizations were performed using methylumi and lumi R packages, respectively [37,38]. Signals were further normalized using the beta-mixture quantile normalization method (BMIQ) [39]. Briefly, CpGs with high quality (p < 0.05) were only used for analyses. Each methylation data point was represented by fluorescent signals from the M (methylated) and U (unmethylated) alleles. Background intensity computed from a set of negative controls was subtracted from each data point. The ratio of fluorescent signals was then computed from the two alleles using the beta (β)-value. The β-values of each CpG site ranging from 0 to 1 reflected percentage methylation levels from 0%–100%, respectively. Differentially methylated CpG sites were defined as sites with a p value (independent samples t-test) <0.05 and a difference of average β-values between UC and N of >0.2. A total of 4820 CpGs were identified. Among them, 237 CpGs were hypermethylated (>1.7-fold increase) in UC samples compared to NC samples. 4.4. DNA Methylation Analysis For methylation analyses, genomic DNA was isolated from 8 normal colon tissues and 79 UC primary tissues using phenol/chloroform. Primer pairs for methylation analyses were preferentially designed for CpG islands of the target genes. For methylation-specific PCR (MSP) and bisulfate sequencing, all primers were designed by MethPrimer (http://www.urogene.org/methprimer). The primer sequences are listed in Table S2. Methylation analyses, including MSP and bisulfite sequencing analyses, were performed as previously described [40]. 4.5. Quantitative Methylation-Specific PCR (qMSP) Bisulfite modification of 2 μg genomic DNA was performed using the EZ DNA Methylation kit (Zymo Research, Irvine, CA, USA). For positive and negative controls, in vitro methylated DNA (IVD) and H2O were used, respectively. For methylation analyses of target genes, quantitative MSP amplification was performed on bisulfite treated samples and normalized using the Alu element. All primers were designed using MethPrimer and listed in Table S2. Real-time PCR was performed by a CFX96TM real-time system (Bio-Rad, Hercules, CA, USA). 4.6. Quantitative Real-Time RT-PCR (qRT-PCR) Total RNA was isolated from human normal colon and UC patient tissues using TRI-Solution (BioScience Technology, Rockaway, NJ, USA) following the manufacturer’s protocol. RNA quantity was measured using a NanoDrop 2000/2000c instrument (Thermo Scientific, Waltham, MA, USA) and 1 µg of total RNA was reverse-transcribed into cDNA using the iScriptTM cDNA Synthesis kit (BioRad, Hercules, CA, USA). For expression studies, primers were designed using the Primer3 web tool (http://frodo.wi.mit.edu/primer3), and listed in Table S2. Quantitative RT-PCR was performed on a CFX96TM Real-Time PCR Detection System (Bio-Rad) using a Syber Green master mix (Thermo Scientific, Waltham, MA, USA). The expression levels of target genes were normalized by β-actin levels, and all relative quantifications of expressions were calculated using the ΔΔCt method. 4.7. Computational Analysis The CpG islands of target genes were predicted and determined by the University of California Santa Cruz (UCSC, Santa Cruz, CA, USA) genome browser (http://www.genome.ucsc.edu) and the CpG Island Searcher (http://cpgislands.usc.edu), following basic limit values. Signal pathways of target genes were predicted by the Super-pathway category within the GeneCard database (http://www.genecards.org/cgi-bin) that included KEGG (http://www.kegg.jp/kegg/pathway.html). For in silico biological functional analyses of target genes, gene ontology (GO) was analyzed with the GeneMANIA application [41]. The p value threshold was limited to 10−3. 4.8. Statistical Analysis Nel Quantified data are expressed as the mean ± standard deviation (SD). Significance testing was conducted using the Student’s t-test. 4.9. Availability of Data and Materials Raw files of the Infinium Human Methylation 450K array supporting the results of this study are available in the Gene Expression Omnibus (GEO) repository under the accession number GSE81211. Acknowledgments This study was also supported by the “Leades INdustry-university Cooperation” Project, supported by the Ministry of Education, Science & Technology (MEST) (2015-B-0016-010112) and by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2015R1D1A1A02061794). Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1291/s1. Click here for additional data file. Author Contributions Keunsoo Kang, Tae-Oh Kim and Joo Mi Yi designed the experiments. Keunsoo Kang, Jin-Han Bae, and Joo Mi Yi analyzed the data. Tae-Oh Kim, Kyudong Han and Eun Soo Kim contributed for interpretating the data. Keunsoo Kang, Tae-Oh Kim and Joo Mi Yi wrote the paper. All authors reviewed and approved the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Genome-wide DNA methylation profiles of ulcerative colitis (UC) patients. (A) Comparison of genome-wide DNA methylation levels between UC tissues and normal control. 4397 probes showed significantly increased methylation level (hypermethylation) in UC tissues than normal colon (NC). However, 420 probes showed decreased methylation level (hypomethylation) in UC tissues than NC. p-values were calculated by Mann-Whitney U test; (B) Distributions of CpGs according to genome annotation. Most altered CpG sites in UC are located in CpG islands near the promoter regions; (C) Unsupervised clustering analysis of the normal colon (NC) and UC patient tissues (UC). The color gradient from white to red displays the β-values ranging from 0 (unmethylated) to 1 (fully methylated). Figure 2 DNA methylation frequencies of three best candidate genes (FAM217B, KIA1614 and RIBC2) in UC samples (n = 79) and normal colon (NC) (n = 8). Figure 3 Bisulfite sequencing analyses of the CpG island in FAM217B, KIA1614, and RIBC2 gene promoter regions. A schematic representation of each gene CpG island (box). The regions analyzed using methylation-specific PCR (MSP) and bisulfite sequencing are indicated by black bars below the CpG island. Individual CpG sites are indicated as vertical lines. Representative bisulfite sequencing analyses were performed for all three genes in representative UC patient samples (n = 5) and controls (n = 5). Each box represents a CpG dinucleotide. Black boxes represent methylated cytosines and gray boxes represent unmethylated cytosines. Figure 4 Correlation between promoter hypermethylation and transcriptional silencing in UC patient tissues and in controls. (A) Quantitative MSP and (B) RT-PCR analyses of FAM217B, KIAA1614 and RIBC2 genes in selective UC patient samples and controls. HCT116 cells were used as a positive control for methylation levels. All quantitative methylation levels were normalized by the Alu element. Human GAPDH was used for expression normalization. Statistical significance (p < 0.001) for all three genes is shown between UC patient tissue samples and controls. Figure 5 Gene ontology analysis and functional implication of hypermethylated genes in UC. Gene ontology (GO) analysis was performed using GeneMANIA with nearby genes around 48 hypermethylated CpG sites which coincides with CpG islands. The x and y axis represent arbitrary numbers. Similar terms tend to be clustered in the plot. The size of circle depicts whether a given term is a more general GO term (larger) or a more specific one (smaller). ijms-17-01291-t001_Table 1Table 1 Basic characteristics of the UC patient samples in this study. Characteristics Number (Mean) Total no. of patients 79 Age (years) median (range) 42.4 (16–68) Gender, n (%) Male 48 (60.8) Female 31 (39.2) Duration of disease ≤1 year 41 (51.9) 1–8 year 24 (30.4) >9 year 14 (17.7) Lesion location, n (%) Proctitis 44 (55.7) Left sided colitis 25 (31.6) Pancolitis 10 (12.7) Mayo endoscopic score, n (%) Normal or inactive 3 (3.8) Mild disease 23 (29.1) Moderate disease 44 (55.7) Severe disease 9 (11.4) Clinical type, n (%) Only one episode 39 (49.4) Chronic relapsing 35 (44.3) Chronic continuous 5 (6.3) ==== Refs References 1. Kaser A. Zeissig S. Blumberg R.S. Inflammatory bowel disease Annu. Rev. Immunol. 2010 28 573 621 10.1146/annurev-immunol-030409-101225 20192811 2. Fiocchi C. Inflammatory bowel disease: Etiology and pathogenesis Gastroenterology 1998 115 182 205 10.1016/S0016-5085(98)70381-6 9649475 3. Franke A. Balschun T. Karlsen T.H. Hedderich J. May S. Lu T. Schuldt D. Nikolaus S. Rosenstiel P. Krawczak M. Replication of signals from recent studies of crohn’s disease identifies previously unknown disease loci for ulcerative colitis Nat. Genet. 2008 40 713 715 10.1038/ng.148 18438405 4. Franke A. McGovern D.P.B. Barrett J.C. Wang K. Radford-Smith G.L. Ahmad T. Lees C.W. Balschun T. Lee J. Roberts R. Genome-wide meta-analysis increases to 71 the number of confirmed crohn’s disease susceptibility loci Nat. Genet. 2010 42 1118 1125 10.1038/ng.717 21102463 5. Anderson C.A. Boucher G. Lees C.W. Franke A. D’Amato M. Taylor K.D. Lee J.C. Goyette P. Imielinski M. Latiano A. Meta-analysis identifies 29 additional ulcerative colitis risk loci, increasing the number of confirmed associations to 47 Nat. Genet. 2011 43 246 252 10.1038/ng.764 21297633 6. Jostins L. Ripke S. Weersma R.K. Duerr R.H. McGovern D.P. Hui K.Y. Lee J.C. Philip Schumm L. Sharma Y. Anderson C.A. Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease Nature 2012 491 119 124 10.1038/nature11582 23128233 7. Herman J.G. Baylin S.B. Gene silencing in cancer in association with promoter hypermethylation N. Engl. J. Med. 2003 349 2042 2054 10.1056/NEJMra023075 14627790 8. Bibikova M. High density DNA methylation array with single cpg site resolution Genomics 2011 98 288 295 10.1016/j.ygeno.2011.07.007 21839163 9. Jones P.A. Baylin S.B. The epigenomics of cancer Cell 2007 128 683 692 10.1016/j.cell.2007.01.029 17320506 10. Robertson K.D. DNA methylation and human disease Nat. Rev. Genet. 2005 6 597 610 10.1038/nrg1655 16136652 11. Karatzas P.S. Gazouli M. Safioleas M. Mantzaris G.J. DNA methylation changes in inflammatory bowel disease Ann. Gastroenterol.: Quart. Pub. Hellenic Soc. Gastroenterol. 2014 27 125 132 12. Kim J.-G. Park M.-T. Heo K. Yang K.-M. Yi J.M. Epigenetics meets radiation biology as a new approach in cancer treatment Int. J. Mol. Sci. 2013 14 15059 15073 10.3390/ijms140715059 23873297 13. Portela A. Esteller M. Epigenetic modifications and human disease Nat. Biotechnol. 2010 28 1057 1068 10.1038/nbt.1685 20944598 14. Kim T.-O. Park J. Kang M.J. Lee S.H. Jee S.R. Rye D.Y. Yang K. Yi J.M. DNA hypermethylation of a selective gene panel as a risk marker for colon cancer in patients with ulcerative colitis Int. J. Mol. Sci. 2013 31 1255 1261 15. Yi J.M. Dhir M. van Neste L. Downing S.R. Jeschke J. Glöckner S.C. de Freitas Calmon M. Hooker C.M. Funes J.M. Boshoff C. Genomic and epigenomic integration identifies a prognostic signature in colon cancer Clin. Cancer Res. 2011 17 1535 1545 10.1158/1078-0432.CCR-10-2509 21278247 16. Khor B. Gardet A. Xavier R.J. Genetics and pathogenesis of inflammatory bowel disease Nature 2011 474 307 317 10.1038/nature10209 21677747 17. Supek F. Bošnjak M. Škunca N. Šmuc T. Revigo summarizes and visualizes long lists of gene ontology terms PLoS ONE 2011 6 1291 10.1371/journal.pone.0021800 21789182 18. Petronis A. Epigenetics as a unifying principle in the aetiology of complex traits and diseases Nature 2010 465 721 727 10.1038/nature09230 20535201 19. Stoll M. Corneliussen B. Costello C.M. Waetzig G.H. Mellgard B. Koch W.A. Rosenstiel P. Albrecht M. Croucher P.J.P. Seegert D. Genetic variation in dlg5 is associated with inflammatory bowel disease Nat. Genet. 2004 36 476 480 10.1038/ng1345 15107852 20. Van Overveld P.G.M. Lemmers R.J.F.L. Sandkuijl L.A. Enthoven L. Winokur S.T. Bakels F. Padberg G.W. van Ommen G.-J.B. Frants R.R. van der Maarel S.M. Hypomethylation of d4z4 in 4q-linked and non-4q-linked facioscapulohumeral muscular dystrophy Nat. Genet. 2003 35 315 317 10.1038/ng1262 14634647 21. Ventham N.T. Kennedy N.A. Nimmo E.R. Satsangi J. Beyond gene discovery in inflammatory bowel disease: The emerging role of epigenetics Gastroenterology 2013 145 293 308 10.1053/j.gastro.2013.05.050 23751777 22. Hartnett L. Egan L.J. Inflammation, DNA methylation and colitis-associated cancer Carcinogenesis 2012 33 723 731 10.1093/carcin/bgs006 22235026 23. Hsieh C.J. Klump B. Holzmann K. Borchard F. Gregor M. Porschen R. Hypermethylation of the p16ink4a promoter in colectomy specimens of patients with long-standing and extensive ulcerative colitis Cancer Res. 1998 58 3942 3945 9731506 24. Wang F. Arisawa T. Tahara T. Takahama K. Watanabe M. Hirata I. Nakano H. Aberrant DNA methylation in ulcerative colitis without neoplasia Hepatogastroenterology 2008 55 62 65 18507080 25. Gould N.J.D. Davidson K.L.M. Nwokolo C.U. Arasaradnam R.P. A systematic review of the role of DNA methylation on inflammatory genes in ulcerative colitis Epigenomics 2016 8 667 684 10.2217/epi-2016-0006 27096966 26. Esteller M. Fraga M.F. Paz M.F. Campo E. Colomer D. Novo F.J. Calasanz M.J. Galm O. Guo M. Benitez J. Cancer epigenetics and methylation Science 2002 297 1807 1808 10.1126/science.297.5588.1807d 12229925 27. Laird P.W. The power and the promise of DNA methylation markers Nat. Rev. Cancer 2003 3 253 266 10.1038/nrc1045 12671664 28. Karatzas P.S. Mantzaris G.J. Safioleas M. Gazouli M. DNA methylation profile of genes involved in inflammation and autoimmunity in inflammatory bowel disease Medicine 2014 93 e309 10.1097/MD.0000000000000309 25526479 29. Li Yim A.Y.F. Duijvis N.W. Zhao J. de Jonge W.J. D’Haens G.R.A.M. Mannens M.M.A.M. Mul A.N.P.M. te Velde A.A. Henneman P. Peripheral blood methylation profiling of female crohn’s disease patients Clin. Epigenet. 2016 8 1 13 10.1186/s13148-016-0230-5 27279921 30. Johnson D.H. Taylor W.R. Aboelsoud M.M. Foote P.H. Yab T.C. Cao X. Smyrk T.C. Loftus E.V. Jr. Mahoney D.W. Ahlquist D.A. DNA methylation and mutation of small colonic neoplasms in ulcerative colitis and crohn’s colitis: Implications for surveillance Inflamm. Bowel Dis. 2016 22 10.1097/MIB.0000000000000795 27104828 31. Papadia C. Louwagie J. del Rio P. Grooteclaes M. Coruzzi A. Montana C. Novelli M. Bordi C. de’ Angelis G.L. Bassett P. Foxe1 and syne1 genes hypermethylation panel as promising biomarker in colitis-associated colorectal neoplasia Inflamm. Bowel Dis. 2014 20 271 277 10.1097/01.MIB.0000435443.07237.ed 24280874 32. DeVos T. Tetzner R. Model F. Weiss G. Schuster M. Distler J. Steiger K.V. Grützmann R. Pilarsky C. Habermann J.K. Circulating methylated sept9 DNA in plasma is a biomarker for colorectal cancer Clin. Chem. 2009 55 1337 1346 10.1373/clinchem.2008.115808 19406918 33. Glockner S.C. Dhir M. Yi J.M. McGarvey K.E. van Neste L. Louwagie J. Chan T.A. Kleeberger W. de Bruïne A.P. Smits K.M. Methylation of tfpi2 in stool DNA: A potential novel biomarker for the detection of colorectal cancer Cancer Res. 2009 69 4691 4699 10.1158/0008-5472.CAN-08-0142 19435926 34. Li M. Chen W.-D. Papadopoulos N. Goodman S.N. Bjerregaard N.C. Laurberg S. Levin B. Juhl H. Arber N. Moinova H. Sensitive digital quantification of DNA methylation in clinical samples Nat. Biotechnol. 2009 27 858 863 10.1038/nbt.1559 19684580 35. Azarschab P. Porschen R. Gregor M. Blin N. Holzmann K. Epigenetic control of the e-cadherin gene (CDH1 ) by CpG methylation in colectomy samples of patients with ulcerative colitis Genes Chromosome Cancer 2002 35 121 126 10.1002/gcc.10101 12203775 36. Saito S. Kato J. Hiraoka S. Horii J. Suzuki H. Higashi R. Kaji E. Kondo Y. Yamamoto K. DNA methylation of colon mucosa in ulcerative colitis patients: Correlation with inflammatory status Inflamm. Bowel Dis. 2011 17 1955 1965 10.1002/ibd.21573 21830274 37. Du P. Kibbe W.A. Lin S.M. Lumi: A pipeline for processing illumina microarray Bioinformatics 2008 24 1547 1548 10.1093/bioinformatics/btn224 18467348 38. Du P. Davis S. Bilke T. Triche J.T. Bootawalla M. Methylumi: Handle illumina methylation data Available online: https://www.scienceopen.com/document?vid=4d66aeae-d817-400f-a84c-0b0df0066f3e (accessed on 30 June 2015) 39. Teschendorff A.E. Marabita F. Lechner M. Bartlett T. Tegner J. Gomez-Cabrero D. Beck S. A beta-mixture quantile normalization method for correcting probe design bias in illumina infinium 450 k DNA methylation data Bioinformatics 2013 29 189 196 10.1093/bioinformatics/bts680 23175756 40. Kim J.-G. Kim T.-O. Bae J.-H. Shim J.-W. Kang M.J. Yang K. Ting A.H. Yi J.M. Epigenetically regulated mir941 and mir1247 target gastric cancer cell growth and migration Epigenetics 2014 9 1018 1030 10.4161/epi.29007 24785261 41. Warde-Farley D. Donaldson S.L. Comes O. Zuberi K. Badrawi R. Chao P. Franz M. Grouios C. Kazi F. Lopes C.T. The genemania prediction server: Biological network integration for gene prioritization and predicting gene function Nucleic Acid. Res. 2010 38 W214 W220 10.1093/nar/gkq537 20576703

PMC005xxxxxx/PMC5000689.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081292ijms-17-01292ArticleSynchronized Cell Cycle Arrest Promotes Osteoclast Differentiation Kwon Minsuk 1†Kim Jin-Man 1†Lee Kyunghee 1Park So-Young 2Lim Hyun-Sook 3Kim Taesoo 4Jeong Daewon 1*Zhang Ge Academic Editor1 Laboratory of Bone Metabolism and Control, Department of Microbiology, Yeungnam University College of Medicine, Daegu 42415, Korea; [email protected] (M.K.); [email protected] (J.-M.K.); [email protected] (K.L.)2 Department of Physiology, Yeungnam University College of Medicine, Daegu 42415, Korea; [email protected] Department of Public Health Administration, Hanyang Women’s University, Seoul 04763, Korea; [email protected] Clinical Research Division, Korea Institute of Oriental Medicine, Daejeon 34054, Korea; [email protected]* Correspondence: [email protected]; Tel.: +82-53-640-6944; Fax: +82-53-653-6628† These authors contributed equally to this study. 09 8 2016 8 2016 17 8 129229 6 2016 04 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Osteoclast progenitors undergo cell cycle arrest before differentiation into osteoclasts, induced by exposure to macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor-κB ligand (RANKL). The role of such cell cycle arrest in osteoclast differentiation has remained unclear, however. We here examined the effect of synchronized cell cycle arrest on osteoclast formation. Osteoclast progenitors deprived of M-CSF in culture adopted a uniform morphology and exhibited cell cycle arrest at the G0–G1 phase in association with both down-regulation of cyclins A and D1 as well as up-regulation of the cyclin-dependent kinase inhibitor p27Kip1. Such M-CSF deprivation also promoted the differentiation of osteoclast progenitors into multinucleated osteoclasts expressing high levels of osteoclast marker proteins such as NFATc1, c-Fos, Atp6v0d2, cathepsin K, and integrin β3 on subsequent exposure to M-CSF and RANKL. Our results suggest that synchronized arrest and reprogramming of osteoclast progenitors renders them poised to respond to inducers of osteoclast formation. Further characterization of such effects may facilitate induction of the differentiation of heterogeneous and multipotent cells into desired cell lineages. cell cycle arrestcell synchronizationosteoclast differentiation ==== Body 1. Introduction Temporal coupling of cell cycle arrest and cell differentiation appears to be universal during organismal development [1]. Cell cycle arrest thus occurs prior to the differentiation of preadipocytes into adipocytes [2]. The transcription factor Prospero simultaneously regulates the expression of multiple cell cycle regulatory genes and neuronal lineage developmental genes in Drosophila [3]. The antiproliferative protein B cell translocation gene 1 (BTG1) is expressed at cell confluence as well as at the onset of myoblast differentiation, and its overexpression concurrently induces cell cycle arrest and terminal differentiation [4]. MyoD, a skeletal muscle-specific transcriptional regulator, coordinates skeletal muscle differentiation during cell cycle arrest in the G0–G1 phase by inducing the expression of the cyclin-dependent kinase (CDK)1 inhibitor p21 [5,6]. Additionally, forced silencing of proliferative signaling stimulates the differentiation of embryonic stem cells [7]. The precise nature of the relation between cell cycle arrest and the induction of differentiation has remained unclear, however. Osteoclast differentiation in mammals is mediated by two osteoclastogenic factors: Macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor-κB ligand (RANKL), a member of the TNF family of proteins. Both op/op mutant mice (which are deficient in M-CSF) and RANKL-deficient mice manifest osteopetrotic bone defects as a result of the impaired formation of bone-resorptive osteoclasts [8,9]. M-CSF and RANKL play distinct roles in osteoclast formation by contributing to the regulation of osteoclast progenitor proliferation and the differentiation of these cells into multinucleated mature osteoclasts, respectively [8,9]. RANKL induces cell cycle arrest in G0–G1 in association with up-regulation of the CDK inhibitor p27Kip1 in a manner dependent on the interaction of RANKL with its cognate receptor RANK and the recruitment of TRAF6 (TNF receptor-associated factor 6) to the intracellular domain of RANK [10]. It has also been reported that RANKL-induced CDK6 down-regulation or RANKL-induced cell cycle arrest with both up-regulation of both p21CIP1 and p27KIP1 may be implicated in osteoclast differentiation [11,12]. Further, TNF-α—another osteoclastogenic factor—is known to induce G1 arrest in endothelial cells in association with the down-regulation of cyclin D1 and CDK2 and with up-regulation of the CDK inhibitors p16INK4a, p21Waf, and p27Kip1 [13]. To shed light on the role of cell cycle arrest during osteoclast differentiation, we have examined whether such arrest directly influences the differentiation process. We found that synchronized G0–G1 arrest induced by withdrawal of the proliferative factor M-CSF promotes osteoclast differentiation. 2. Results and Discussion 2.1. M-CSF Deprivation Induces G0–G1 Cell Cycle Arrest To induce cell cycle synchronization, we cultured osteoclast progenitors in the absence of M-CSF for 12 h. Whereas cells cultured in the presence of M-CSF manifested a spindle and salverform morphology, those deprived of M-CSF for 12 or 24 h adopted a more spherical shape (Figure 1A). The surface area of the M-CSF-deprived cells decreased with time, in contrast with the increase apparent for cells cultured with M-CSF (Figure 1B). The uniformity of cell size was evaluated by calculation of the SD for the average area per cell, with a lower SD denoting a greater uniformity. The SD was markedly lower for cells cultured in the absence of M-CSF than for those maintained in its presence (Figure 1B). These results thus indicated that M-CSF-deprived cells were largely homogeneous in terms of cell morphology and size. We next measured the proliferation of osteoclast progenitors, both with the MTT assay and by measurement of [3H]thymidine incorporation into chromosomal DNA during the S phase of the cell cycle [14]. Both approaches confirmed that withdrawal of M-CSF for 12 h resulted in inhibition of cell proliferation (Figure 2A). Flow cytometric analysis of cells stained with propidium iodide also revealed that the proportion of cells in the S or G2-M phases of the cell cycle was reduced, whereas the proportion of those in the G0–G1 phase was increased in response to M-CSF deprivation (Figure 2B). Consistent with these results, immunoblot analysis of cell cycle regulators showed that the abundance of positive regulators of G1-phase CDKs—including cyclin A and cyclin D1—was reduced, whereas that of the G1-phase CDK inhibitor p27Kip1 was increased in cells deprived of M-CSF compared with those maintained in its presence (Figure 3). Furthermore, M-CSF withdrawal inhibited phosphorylation of histone H3 at Ser10, an event associated with S-phase entry [15]. The G0–G1 cell cycle arrest induced by M-CSF deprivation in osteoclast progenitors thus appeared to be due to down-regulation of cyclins A and D1 as well as up-regulation of the CDK inhibitor p27Kip1. 2.2. Cell Synchronization Promotes Osteoclast Formation A molecular link between cell cycle withdrawal and cell differentiation has previously been suggested [7]. Furthermore, we found that osteoclast progenitors deprived of M-CSF appear to be homogeneous in terms of cell morphology and cell cycle progression. These observations together with others suggest a possible link between cell synchronization and osteoclast differentiation. Regarding the induction of osteoclast differentiation after culture in the absence of M-CSF, reprogrammed osteoclast progenitors manifested a marked increase in the formation of tartrate-resistant acid phosphatase-positive multinucleated cells (TRAP(+) MNCs) containing ≥3 or ≥10 nuclei compared with progenitors not previously deprived of M-CSF (Figure 4A). In addition, the expression of bone resorption-related proteases (MMP-9 and cathepsin K) in cell lysates and culture media was highly up-regulated in the bone-resorbing process of osteoclasts induced by M-CSF withdrawal than in the control (Figure S1). However, the osteoclastic cell area of TRAP(+) MNCs containing ≥3 nuclei was not different between control and M-CSF-withdrawal cells. Additionally, bone resorption pit formation of mature osteoclasts and the content of bone-resorptive end product from type I collagen (deoxypyridinoline, DPD) showed no significant difference between control and M-CSF-deprived cells, due to the survival and longevity of osteoclasts with a full actin ring during bone resorption (Figures S2 and S3). We also observed that the differentiation of synchronized cells upon cell–cell contact inhibition or serum withdrawal showed a similar result as compared to cells synchronized by M-CSF deprivation alone (Figure S4). Further, enhanced osteoclast differentiation by prior M-CSF withdrawal was confirmed by real-time PCR (Figure 4B) or immunoblot analysis (Figure 4C), showing increased expression of various osteoclastic markers, including TRAP, osteoclast-associated immunoglobulin-like receptor (OSCAR), the osteoclastogenic transcription factors NFATc1 (nuclear factor of activated T cells c1) and c-Fos (component of AP-1), as well as dendrocyte-expressed seven transmembrane protein (DC-STAMP), osteoclast stimulatory transmembrane protein (OC-STAMP), and Atp6v0d2 (fusion factor of mononuclear osteoclast precursors), cathepsin K (bone-resorptive cysteine protease), and integrin β3 chain (subunit of integrin αvβ3). Our findings show that M-CSF deprivation induces cell cycle arrest at the G0–G1 phase, elicits the adoption of a uniform cell morphology, and promotes the subsequent induction of osteoclast formation in osteoclast progenitors. This concept would be especially important in the differentiation of synchronized osteoclast progenitors into dendritic cells in the future. In addition to its induction of cell cycle arrest, deprivation of M-CSF may silence intracellular signaling networks and thereby increase cell sensitivity to new extracellular cues, rendering osteoclast progenitors poised to respond to the induction of osteoclast differentiation by RANKL. More generally, withdrawal of nutrients (such as glucose and amino acids), growth factors, or other receptor ligands may serve to reprogram cells to confer enhanced susceptibility to inducers of differentiation. Further studies are warranted to determine the effects of such cell synchronization on differentiation efficiency in multipotent stem cells, cancer cells, graft cells, and tissue resident cells. Such knowledge may serve to facilitate the induction of the differentiation of multipotent or heterogeneous cells into specific cell types of interest. 3. Materials and Methods 3.1. Induction of Synchronized Cell Cycle Arrest and Osteoclast Differentiation Mononuclear osteoclast progenitors were isolated from bone marrow of mice as described previously [16], and were cultured in α-minimum essential medium (α-MEM; Invitrogen, Carlsbad, CA, USA) supplemented with antibiotics, 10% FBS, and recombinant human M-CSF (30 ng/mL). Osteoclast progenitors at 70% confluence (5 × 104 cells per well in 48-well culture plates) were induced to undergo cell cycle arrest by exposure to culture medium lacking M-CSF for 12 h. For induction of osteoclast differentiation, the M-CSF-deprived or control cells were cultured in medium containing M-CSF (30 ng/mL) and recombinant mouse RANKL (100 ng/mL) for 4 days, with replenishment of the medium after 2 days. Osteoclast differentiation was assessed by staining of the cells for tartrate-resistant acid phosphatase (TRAP) with the use of a leukocyte acid phosphatase staining kit (Sigma-Aldrich, St. Louis, MO, USA). TRAP-positive multinucleated cells (TRAP(+) MNCs) containing ≥3 or ≥10 nuclei were counted under a light microscope. 3.2. Analysis of Cell Area Osteoclast progenitors (5 × 104 cells per well in 48-well culture plates) were incubated in culture medium with or without M-CSF (30 ng/mL) for the indicated times. The cells were then fixed with 3.7% formalin for 10 min, stained with 0.5% crystal violet for 30 min, washed with PBS, and photographed under a light microscope for measurement of cell surface area with the use of Image-Pro plus software version 6.0 (Media Cybernetics, Silver Spring, MD, USA). 3.3. Assay of Cell Proliferation and Cell Cycle Analysis For assay of cell proliferation, osteoclast progenitors were incubated in culture medium with or without M-CSF (30 ng/mL) for 12 h before exposure to 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). The formazan product was then dissolved in DMSO and quantitated spectrophotometrically at a wavelength of 595 nm. Alternatively, cells (1 × 105 cells per well in 24-well culture plates) were incubated in culture medium containing [3H]thymidine (1 μCi per well; PerkinElmer, Waltham, MA, USA) in the absence or presence of M-CSF for 12 h, washed with ice-cold PBS, treated with ice-cold 5% trichloroacetic acid (TCA), and washed again with ice-cold PBS before preparation of cell lysates with a lysis solution containing 0.5% SDS and 0.5 M NaOH. The lysates were mixed with liquid scintillation cocktail (DCC-BIONET, Seongnam, Korea), and the amount of [3H]thymidine-labeled DNA was measured with a liquid scintillation counter (Tri-Carb 3110 TR, PerkinElmer, Santa Clara, CA, USA). For cell cycle analysis, osteoclast progenitors were incubated in culture medium with or without M-CSF (30 ng/mL) for 12 h and were then detached from the plate by exposure to trypsin and isolated by centrifugation at 1000× g for 5 min. The cells were suspended in PBS containing 5 mM EDTA, fixed with 70% ethanol for 12 h at 4 °C, isolated again by centrifugation, resuspended in PBS containing 5 mM EDTA, treated with RNase A (50 μg/mL) for 30 min at room temperature, and stained with propidium iodide (50 μg/mL) for 10 min in the dark. The stained cells were analyzed by flow cytometry with a FACSCalibur instrument (Becton Dickinson, San Jose, CA, USA). 3.4. Immunoblot Analysis For analysis of cell cycle regulators or osteoclast marker proteins, cells incubated in culture medium with or without M-CSF (30 ng/mL) and then exposed (or not) to M-CSF and RANKL for the indicated times were lysed and subjected to immunoblot analysis with antibodies to cyclin A, to cyclin D1, to cyclin E, to CDK2, to CDK4, to p21Waf1/Cip1, to p27Kip1, to Ser10-phosphorylated histone H3, to α-tubulin, to NFATc1, to Atp6v0d2, to integrin β3, and to β-actin (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA); with those to histone H3 and to cathepsin K (Abcam, Cambridge, MA, USA); and with those to c-Fos (Cell Signaling Technology, Boston, MA, USA). 3.5. Statistical Analysis Quantitative data are presented as means ± SD from at least three independent experiments. Differences between two groups were analyzed with Student’s t test. For statistical analysis for multiple comparisons, means between multiple groups were performed using one-way ANOVA analysis using the Microsoft 2010 Excel program. A p value of <0.05 was considered statistically significant. Acknowledgments This study was supported by grants from the Korea Healthcare Technology R&D Project, Ministry for Health, Welfare, Family Affairs, Republic of Korea (no. HI15C2164), the National Research Foundation of Korea (2015R1A5A2009124), and the 2013 Yeungnam University. Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1292/s1. Click here for additional data file. Author Contributions Minsuk Kwon, Jin-Man Kim, and Taesoo Kim performed the experiments; Kyunghee Lee, Hyun-Sook Lim, and Daewon Jeong analyzed the data; So-Young Park contributed reagents and Daewon Jeong designed the research and wrote the manuscripts. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Effects of macrophage colony-stimulating factor (M-CSF) deprivation on the morphology and size of osteoclast progenitors. (A) Cells were cultured in the absence or presence of M-CSF for the indicated times and then stained with crystal violet. Scale bar: 50 µm; (B) Relative average cell surface area was determined by dividing the total cell area by the number of cells (left panel), and SD of the average area per cell was determined by measuring the area of individual cells (right panel), in photographs similar to those in (A). * Differences compared with control were statistically significant (p < 0.01, ANOVA). Figure 2 Induction of G0–G1 cell cycle arrest in osteoclast progenitors by M-CSF deprivation. (A) Cells were cultured in the absence or presence of M-CSF for 12 h, after which cell proliferation was determined with the MTT assay (left panel) or by measurement of [3H]thymidine incorporation (right panel); (B) Cells cultured as in (A) were stained with propidium iodide and subjected to cell cycle analysis by flow cytometry. Data are means ± SD for a representative experiment run in triplicate. * p < 0.01 (Student’s t test). Figure 3 Change in expression levels of cell cycle regulators during M-CSF deprivation. Cells cultured for the indicated times were lysed and subjected to immunoblot analysis with antibodies to the indicated proteins. Figure 4 Prior M-CSF deprivation promotes osteoclast differentiation. (A) Osteoclast progenitors were cultured in the absence or presence of M-CSF for 12 h and were then exposed to M-CSF and receptor activator of nuclear factor-κB ligand (RANKL) for 4 days to induce osteoclast differentiation. The cells were then stained for tartrate-resistant acid phosphatase (TRAP, upper panels), and the number of TRAP(+) MNCs with ≥3 or ≥10 nuclei were counted (lower panels). Scale bar: 200 µm; (B) Osteoclast precursors with or without M-CSF were differentiated into osteoclasts for the indicated times. The mRNA levels of osteoclastogenic marker genes, including TRAP, OSCAR, NFATc1, DC-STAMP, OC-STAMP, ATP6v0d2, and cathepsin K (Ctsk). Quantitative data are means ± SD; * p < 0.01 (Student’s t test); (C) Immunoblot analysis of osteoclast marker proteins for osteoclast progenitors cultured in the absence or presence M-CSF for 12 h and then exposed to M-CSF and RANKL for the indicated times. ==== Refs References 1. Myster D.L. Duronio R.J. To differentiate or not to differentiate? Curr. Biol. 2000 10 R302 R304 10801410 2. Scott R.E. Florine D.L. Wille J.J. Jr. Yun K. Coupling of growth arrest and differentiation at a distinct state in the G1 phase of the cell cycle: Gd Proc. Natl. Acad. Sci. USA 1982 79 845 849 10.1073/pnas.79.3.845 6174983 3. Li L. Vaessin H. Pan-neural prospero terminates cell proliferation during drosophila neurogenesis Genes Dev. 2000 14 147 151 10652268 4. Rodier A. Marchal-Victorion S. Rochard P. Casas F. Cassar-Malek I. Rouault J.P. Magaud J.P. Mason D.Y. Wrutniak C. Cabello G. BTG1: A triiodothyronine target involved in the myogenic influence of the hormone Exp. Cell Res. 1999 249 337 348 10.1006/excr.1999.4486 10366433 5. Halevy O. Novitch B.G. Spicer D.B. Skapek S.X. Rhee J. Hannon G.J. Beach D. Lassar A.B. Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by myoD Science 1995 267 1018 1021 10.1126/science.7863327 7863327 6. De la Serna I.L. Roy K. Carlson K.A. Imbalzano A.N. Myod can induce cell cycle arrest but not muscle differentiation in the presence of dominant negative SWI/SNF chromatin remodeling enzymes J. Biol. Chem. 2001 276 41486 41491 10.1074/jbc.M107281200 11522799 7. Li V.C. Kirschner M.W. Molecular ties between the cell cycle and differentiation in embryonic stem cells Proc. Natl. Acad. Sci. USA 2014 111 9503 9508 10.1073/pnas.1408638111 24979803 8. Kong Y.Y. Yoshida H. Sarosi I. Tan H.L. Timms E. Capparelli C. Morony S. Oliveira-dos-Santos A.J. Van G. Itie A. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis Nature 1999 397 315 323 9950424 9. Yoshida H. Hayashi S. Kunisada T. Ogawa M. Nishikawa S. Okamura H. Sudo T. Shultz L.D. Nishikawa S. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene Nature 1990 345 442 444 10.1038/345442a0 2188141 10. Bharti A.C. Takada Y. Shishodia S. Aggarwal B.B. Evidence that receptor activator of nuclear factor (NF)-kappaB ligand can suppress cell proliferation and induce apoptosis through activation of a NF-kappaB-independent and TRAF6-dependent mechanism J. Biol. Chem. 2004 279 6065 6076 10.1074/jbc.M308062200 14645259 11. Ogasawara T. Katagiri M. Yamamoto A. Hoshi K. Takato T. Nakamura K. Tanaka S. Okayama H. Kawaguchi H. Osteoclast differentiation by RANKL requires NF-kappaB-mediated downregulation of cyclin-dependent kinase 6 (cdk6) J. Bone Miner. Res. 2004 19 1128 1136 10.1359/jbmr.2004.19.7.1128 15176996 12. Mizoguchi T. Muto A. Udagawa N. Arai A. Yamashita T. Hosoya A. Ninomiya T. Nakamura H. Yamamoto Y. Kinugawa S. Identification of cell cycle-arrested quiescent osteoclast precursors in vivo J. Cell Biol. 2009 184 541 554 10.1083/jcb.200806139 19237598 13. Dormond O. Lejeune F.J. Ruegg C. Modulation of cdk2, cyclin D1, p16INK4a, p21WAF and p27Kip1 expression in endothelial cells by TNF/IFN gamma Anticancer Res. 2002 22 3159 3163 12530059 14. Nowakowski R.S. Lewin S.B. Miller M.W. Bromodeoxyuridine immunohistochemical determination of the lengths of the cell cycle and the DNA-synthetic phase for an anatomically defined population J. Neurocytol. 1989 18 311 318 10.1007/BF01190834 2746304 15. De Souza C.P. Osmani A.H. Wu L.P. Spotts J.L. Osmani S.A. Mitotic histone H3 phosphorylation by the nima kinase in aspergillus nidulans Cell 2000 102 293 302 10.1016/S0092-8674(00)00035-0 10975520 16. Lee K. Chung Y.H. Ahn H. Kim H. Rho J. Jeong D. Selective regulation of mapk signaling mediates RANKL-dependent osteoclast differentiation Int. J. Biol. Sci. 2016 12 235 245 10.7150/ijbs.13814 26884720

PMC005xxxxxx/PMC5000690.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081293ijms-17-01293ArticleOleuropein Prevents Neuronal Death, Mitigates Mitochondrial Superoxide Production and Modulates Autophagy in a Dopaminergic Cellular Model Achour Imène 1Arel-Dubeau Anne-Marie 1Renaud Justine 1Legrand Manon 1Attard Everaldo 2Germain Marc 1Martinoli Maria-Grazia 13*Battino Maurizio Academic Editor1 Cellular Traffic Research Group, Department of Medical Biology, Université du Québec à Trois-Rivières, Trois-Rivières, QC G9A 5H7, Canada; [email protected] (I.A.); [email protected] (A.-M.A.-D.); [email protected] (J.R.); [email protected] (M.L.); [email protected] (M.G.)2 Institute of Earth Systems, University of Malta, Msida MSD 2080, Malta; [email protected] Department of Psychiatry and Neuroscience, U. Laval and CHU Research Center, Québec, QC G9A 5H7, Canada* Correspondence: [email protected]; Tel.: +1-819-376-5011 (ext. 3994); Fax: +1-819-376-508409 8 2016 8 2016 17 8 129305 7 2016 02 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Parkinson’s disease (PD) is a progressive neurodegenerative disorder, primarily affecting dopaminergic neurons in the substantia nigra. There is currently no cure for PD and present medications aim to alleviate clinical symptoms, thus prevention remains the ideal strategy to reduce the prevalence of this disease. The goal of this study was to investigate whether oleuropein (OLE), the major phenolic compound in olive derivatives, may prevent neuronal degeneration in a cellular dopaminergic model of PD, differentiated PC12 cells exposed to the potent parkinsonian toxin 6-hydroxydopamine (6-OHDA). We also investigated OLE’s ability to mitigate mitochondrial oxidative stress and modulate the autophagic flux. Our results obtained by measuring cytotoxicity and apoptotic events demonstrate that OLE significantly decreases neuronal death. OLE could also reduce mitochondrial production of reactive oxygen species resulting from blocking superoxide dismutase activity. Moreover, quantification of autophagic and acidic vesicles in the cytoplasm alongside expression of specific autophagic markers uncovered a regulatory role for OLE against autophagic flux impairment induced by bafilomycin A1. Altogether, our results define OLE as a neuroprotective, anti-oxidative and autophagy-regulating molecule, in a neuronal dopaminergic cellular model. oleuropeinpolyphenolneuroprotectionneurodegenerationoxidative stresscellular deathauthophagyParkinson’s disease ==== Body 1. Introduction In the last decade, a growing body of evidence has supported a role for oxidative stress as a mediator of nerve cell death in neurodegenerative diseases such as Parkinson’s disease (PD) and Alzheimer's disease (AD). Increased generation of reactive oxygen species (ROS) observed in PD and AD occurs as a consequence of mitochondrial dysfunction or inflammation, resulting in protein oxidation and aggregation as well as apoptosis [1,2]. In addition to directly damaging proteins, such as alpha-synuclein in PD [3,4], ROS also disrupt the autophagy-lysosomal pathway [5,6], the main cellular mechanism required to degrade the protein aggregates and dysfunctional mitochondria that characterize both PD and AD [7,8,9,10,11]. Failure to clear these noxious cellular components in turn leads to ROS generation [1,2] and further autophagy-lysosomal pathway impediment [12,13]. In other words, dysfunctional autophagic processes and excessive ROS production propel a self-perpetuating cycle that ultimately leads to neuronal death. Seeing as autophagy impairment has been observed in several neurodegenerative diseases, including PD [14,15,16,17], the autophagy-lysosome pathway is currently deemed a highly interesting therapeutic target against neurodegeneration. During autophagy, cytoplasmic content is packaged into a vesicle (autophagosome) and delivered to lysosomes (autophagolysosome) for degradation [14]. Thus, re-establishment of autophagic flux, which can be impaired either due to blockage of initiation of autophagy (formation of autophagosome) or obstruction of the clearance endpoint (degradation of cargo-filled autophagolysosome), is a viable neuroprotective strategy currently under pre-clinical evaluation (see for review [18]). However, while enhancing autophagy could provide neuroprotection by promoting aggregate degradation, on the other hand its activation can also be detrimental to neurons [19,20]. Therefore, while autophagy activation can promote survival when the accumulation of toxic cellular components is the primary issue, triggering it may also bear detrimental consequences for neurons. Neurodegenerative diseases such as PD and AD are challenging to treat due to the tardy appearance of clinical symptoms, typically occurring when neuronal depletion is already important and irreversible [21,22]. Although current pharmacological treatments aim to appease symptoms and to slow disease progression, tremendous efforts are presently deployed to formulate strategies that address neuronal death, either by preventing it (neuroprotection) or stopping it (neurorescue) [21,23,24,25,26]. Following this line of evidence, several natural phenolic compounds have already demonstrated their neuroprotective properties in the context of neurodegeneration. These natural molecules have been shown to exert their beneficial effects by several means including the modulation of key signaling pathways, with downstream effects such as reduction of oxidative stress, limitation of neuroinflammation and inhibition of apoptosis [27,28,29,30,31,32]. Among them, oleuropein (OLE), a phenolic compound found throughout the entire spectrum of products derived from Olea europaea, is reported to exert numerous pharmacological benefits. These include anti-oxidative, anti-inflammatory, anti-atherogenic, hypoglycemiant, antitumor and antiviral activities (for a recent review, see [28]). In particular, OLE neutralizes ROS and enhances the activity of anti-oxidant enzymes such as superoxide dismutase (SOD), catalase and glutathione peroxidase in animal models [33,34,35,36]. OLE is also reported to reduce apoptosis in catechoaminergic cells [37] and to ameliorate cognitive deficits in transgenic mice models mimicking AD [38,39]. Besides, OLE, together with resveratrol, is nowadays considered one of the most promising caloric restriction mimetics owing to its capacity to modulate autophagy via activation of AMP-activated protein kinase (AMPK) and mechanistic target of rapamycin (mTOR) pathways [28,39,40,41]. In the present study, we evaluated the neuroprotective effects of OLE administered preventively in a known cellular dopaminergic model of PD, nerve-growth-factor (NGF)-differentiated PC12 cells (neuronal PC12 cells) exposed to the potent parkinsonian toxin 6-hydroxydopamine (6-OHDA) [42,43]. We also studied its anti-oxidative capacity by appraising its effect on mitochondrial superoxide generation as well as investigated its role in the autophagy-lysosomal pathway by examining the formation/degradation of autophagic vesicles following the administration of OLE. Our results illustrate that OLE treatment decreases oxidative stress and regulates autophagy, as well as prevents in part neuronal death provoked by 6-OHDA administration. 2. Results 2.1. OLE Prevents 6-OHDA-Induced Neuronal Death The neuroprotective effects of OLE in neuronal PC12 cells treated with 40 µM 6-OHDA was assessed by measuring lactate dehydrogenase (LDH) release by damaged cells, an index of cytotoxicity. Figure 1A shows significant neuronal cell death after a 24 h exposure to 6-OHDA compared to the control condition (CTRL). A 3 h pre-treatment with 10−12 M OLE partially though significantly reduced the cytotoxic effect of 6-OHDA in neuronal PC12 cells (OLE + 6-OHDA), while treatment with OLE alone had no effect. Neuronal death was also detected by an apoptosis-specific DNA denaturation assay (Figure 1B). Levels of DNA denaturation were duly increased by 6-OHDA treatment compared to CTRL, whereas a pre-treatment with OLE successfully prevented apoptosis (6-OHDA + OLE). OLE administered alone caused no significant changes (Figure 1B). In order to further support these findings, we analyzed the expression of specific proteins acting in the apoptotic cascade. Western blotting was performed on total proteins extracted from neuronal PC12 cells treated with 6-OHDA, with or without OLE. We analyzed the ratio of pro-apoptotic Bax and anti-apoptotic Bcl-2 proteins (Figure 2) reported to correlate with apoptosis [42,44,45]. Our results demonstrate that the administration of 6-OHDA significantly increases the Bax/Bcl-2 ratio compared to CTRL and that this ratio is preserved at control levels in neuronal PC12 cells treated with OLE. 2.2. OLE Mitigates Mitochondrial Superoxide Production As OLE has been suggested to possess anti-oxidative properties, we investigated its capacity to decrease mitochondrial ROS in our cellular paradigm. We thus measured the production of superoxide anion (O2•−) with MitoSOX™ Red, an ethidium bromide fluorogenic dye enhanced with a triphenyl phosphonium moiety responsible for its specific mitochondrial localization. MitoSOX™ Red is oxidized and emits red fluorescence strictly upon reacting with O2•− in the mitochondria of live cells [43,46]. Neuronal PC12 cells treated for 3 h with 80 µM of a potent SOD inhibitor, N,N-diethyldithiocarbamate (DDC), indeed exhibited greater fluorescence due to impairment of superoxide anion detoxification (Figure 3A,B). This time period was considered since mitochondrial O2•− generation and subsequent oxidative stress are early events in the causative process of cellular death. Figure 2B and Figure 3A show the highest level of fluorescence after administration of DDC, while the pre-treatment with OLE strongly mitigated the DDC-induced production of O2•−. CTRL and OLE conditions both displayed low levels of red fluorescence (Figure 3A,B). Our results thus indicate that OLE can act as a potent antioxidant in our neuronal cellular system. 2.3. OLE Modulates Autophagy Autophagy is an important maintenance pathway with a focal role in cell fate and its dysfunction constitute both a possible cause and effect of oxidative stress and accumulation of misfolded proteins [1,2,5,6,12,13]. To determine whether OLE may modulate autophagy in our dopaminergic cell model of PD, we first measured autophagosome formation by quantifying the lipidated form of microtubule-associated protein 1A/1B-light chain 3-II (LC3-II). Following initiation of autophagy, LC3-II is lipidated and incorporated to the membrane of the autophagosome, therefore serving as a sensitive marker of their formation [47]. The presence of autophagosomes can be observed by immunofluorescence as small LC3-II-positive puncta within the cytoplasm. When neuronal PC12 cells were treated with 10−12 M OLE, we did not observe any alteration in LC3-II staining compared to control cells (Figure 4). These results were further supported by measuring protein expression of LC3. Lipidated LC3 (LC3-II) is observed as a faster migrating band by Western blotting compared to the non-lipidated form (LC3-I). Consistent with immunofluorescence data, LC3-II levels were similar in control and OLE-treated cells (Figure 5A, CTRL and OLE). Altogether, these results indicate that OLE does not stimulate autophagosome accumulation. In contrast to these results, higher doses of OLE have previously been suggested to induce autophagy [40]. Therefore, to clearly establish the role of OLE in the regulation of autophagy, we measured autophagic flux. During the autophagic process, a portion of LC3-II associated with the autophagosome is degraded in lysosomes along with the autophagosome cargo. As a consequence, blocking the fusion of autophagosomes with lysosomes using bafilomycin A1 (BAF) results in an increase in LC3-II, as detected by Western blotting analyses (Figure 5A, BAF). No increase was observed when cells were treated with OLE (Figure 5A, OLE). Since LC3-II accumulation is not observed, this signifies that picomolar doses of OLE as administered in this study as a pre-treatment inhibits autophagy initiation. Thus explaining that a subsequent treatment with BAF (Figure 5A, BAF + OLE) fails to increase levels of the autophagosome marker: there are no autophagosomes to accumulate. Together with our previous optical observations that OLE does not favor LC3-II accumulation (Figure 4), these data propose that picomolar doses of OLE inhibit rather than activate autophagy in our cellular paradigm. To broaden our understanding of these results, we measured expression levels of p62, a protein that links LC3-II to ubiquitinated substrates. While it is incorporated into the completed autophagosome, it is degraded in autolysosomes [48]. As a consequence, p62 is an indicator of inhibition of autophagy as it faithfully accumulates when autophagosome formation or its lysosomal degradation are impaired [48]. Consistent with decreased autophagic flux, p62 levels were similarly increased in OLE-alone and BAF-alone conditions (Figure 5B, OLE and BAF). However, the failure of BAF to increase p62 levels in the presence of OLE (BAF + OLE) suggests that OLE could affect lysosomes in addition to autophagosomes. We also determined the effect of OLE treatment on 6-OHDA-induced changes in autophagy. Consistent with oxidative stress reducing lysosomal function [5,6], 6-OHDA treatment of neuronal PC12 cells inhibited autophagic flux as measured by the increase in p62 levels despite LC3-II levels similar to control cells (Figure 5A,B). Importantly, OLE did not further modulate the decrease in autophagic flux caused by 6-OHDA (OLE + 6-OHDA). Finally, to determine whether OLE directly affects lysosomes, we measured the presence of lysosomes and acidic vacuoles in OLE-treated neuronal PC12 cells (Figure 6). The specific lysosomal marker lysosome-associated membrane protein 2-a (LAMP2) was used to label lysosomes or any organelle fused with a lysosome, including autophagolysosomes. On the other hand, fluorescent acridine orange dye is uptaken by acidic vesicles in general, which include lysosomes, endolysosomes, autophagolysosomes, late endosomes and any other acidic cisterns dwelling in the cytoplasm. While OLE did not affect the expression of the lysosomal marker LAMP2, it significantly decreased the number of acidic vesicles present in neuronal PC12 cells, as measured by acridine orange staining (Figure 6, red fluorescence and histogram). These results suggest that while OLE does not affect lysosome numbers, it decreases their acidification. 3. Discussion It is now apparent that success in preventing neuronal degeneration and protecting neurons against injuries will depend upon the control of free radical formation, inflammatory processes and autophagic mechanisms. Multiple lines of evidence suggest that several phytochemicals activate animal adaptive cellular stress response pathways. These induce the expression of gene networks encoding anti-oxidative enzymes, protein chaperones, neurotrophic factors and other cytoprotective proteins [49,50]. In particular, polyphenols and phytosterols have been widely studied for their neuroprotective properties often mediated by their anti-oxidative potential [40,51,52]. Their proven pro-survival effects in the context of neuronal death might lay the foundation for the development of novel preventive strategies for complementing current therapies in neurodegenerative diseases. In this context, the multiple beneficial properties of OLE [40] are under vigorous investigation because of its high consumption in the Mediterranean diet and its fairly significant bioavailability as a nutraceutical [53,54]. In this respect, the aim of this study was to evaluate the neuroprotective effect of OLE in a cellular dopaminergic model of PD, NGF-differentiated PC12 cells. Following NGF administration, PC12 cells adopt a neuronal-like phenotype as manifested by secretion of high levels of dopamine and the expression of tyrosine hydroxylase, dopamine transporter, neurofilaments as well as estrogen receptor-alpha and -beta [55,56,57,58]. This cellular paradigm has been extensively used by us as well as others to demonstrate that several polyphenols are indeed neuroprotective by reducing apoptosis, oxidative stress and neuroinflammation [46,59,60,61]. The data presented herein showed that a picomolar dose of OLE reduces neuronal death when administered prior to 6-OHDA, a potent parkinsonian toxin whose oxidative byproduct mediates its intracellular toxicity [62]. OLE also lowered 6-OHDA-induced apoptosis, as established by assessing levels of specific DNA denaturation by formamide, a robust marker of apoptosis, as well as the ratio of pro-apoptotic Bax and anti-apoptotic Bcl-2 expression. Indeed, a high Bax/Bcl-2 ratio favors the release of mitochondrial factors leading to the activation of effector caspases in the apoptotic cascade and consequent neuronal death [42,44,63]. Thus, these results endorse OLE as a pro-survival molecule playing a preventive pro-survival role in our cellular paradigm. To further characterize OLE as a neuroprotective phytochemical, we examined its potential anti-oxidative actions in our cellular paradigm and found a significant reduction of mitochondrial superoxide anion levels when neuronal PC12 cells were pre-treated with a picomolar dose OLE prior to treatment with DDC, a potent inhibitor of the enzyme responsible for detoxifying this ROS. These results are consistent with previous works sustaining that OLE is a potent natural anti-oxidant in a neuronal environment [36,39]. One event that possibly arises as a consequence of ROS overproduction in PD is autophagy-lysosomal pathway dyfunction [5,6]. Indeed, mitochondrial dysfunction with ensuing inhibition of the electron transport chain impairs lysosomal activity resulting in hampered autophagic clearance. Rigacci and collaborators [40] have recently proposed that OLE induces autophagy via the AMPK/mTOR cascade in neuroblastoma cells as well as in a mouse model of AD. We thus studied the effect of a picomolar dose of OLE on autophagosome/lysosome dynamics. Initially, we measured the expression of specific markers of the autophagy process. The detection of LC3-II expression by Western blot or fluorescence assays directly correlates with autophagy induction [47,48], while p62 protein serves as a link between LC3 and ubiquitinated substrates and is degraded in autophagolysosomes [48]. Its levels will characteristically rise if autophagy is blocked after it is linked to mature autophagosomes but before they merge with lysosomes and become autophagolysosomes. Changes in p62 levels therefore inversely correlate with autophagic clearance. In these specific experimental conditions, picomolar doses of OLE did not change the expression of LC3-II but significantly increased p62 expression, suggesting that OLE inhibits initiation of autophagy. Other investigation using double immunofluorescence localization for p62 and LC3 would better precise this premise. Consistent with oxidative stress reducing lysosomal function, our results report that 6-OHDA treatment inhibited autophagic flux as illustrated by the increase in p62 levels. In our experiments LC3-II levels are similar to control levels, after 24 h of 6-OHDA administration. It should be noted however that LC3-II expression has been reported to increase at maximal levels at 12 h after 6-OHDA administration in human neuroblastoma SH-SY5Y cells and return to control levels at 24 h [64]. Importantly, OLE did not further modulate the decrease in autophagic flux caused by 6-OHDA, suggesting that the neuroprotective effect of OLE shown previously does not depend on the modulation of autophagy, at least in our experimental condition. We also observed that while OLE treatment alone did not affect the expression of the lysosomal marker LAMP2, a protein found in the membrane of lysosome [65], it clearly decreased the number of acidic vesicles present in neuronal PC12 cells, as measured by the presence of acridine orange-stained staining. These observations may rather be due to loss of acidification in components usually targeted by acridine orange, or to lower formation and/or higher clearance rates of other acidic vesicles not stained by LAMP2, for example late endosomes or amphisomes. This may indicate a novel role for OLE on lysosome dynamics possibly related with the potent OLE anti-oxidant effects. Indeed, further investigations are required to ascertain the accurate role of OLE on lysosomal dynamics and the acidification of these types of vesicles. Overall, these results support a role for OLE as a modulator of the autophagic flux. It is realistic to speculate that physiologically plausible picomolar concentrations of OLE as tested in this study, may be sufficient to activate adaptive responses and favor protein homeostasis by activating complex yet intriguing vitagene pathways [66,67]. Interestingly, recent data report that OLE can induce mitochodrial biogenesis as well as activate the cellular anti-oxidant defense system in avian muscle cells [68]; thus sustaining its potential beneficial effects in neurodegenerative disease where the balance between oxidant and anti-oxidant systems is compromised. Finally, we would like to stress that although autophagy is usually considered a pro-survival mechanism in neurons, its activation is also linked to neuronal death under some circumstances as in ischemic brain damage [69,70]. As a consequence, autophagy regulation as a means of neuroprotection is rather promising, though it is important to consider its “yin-yang” duality. Indeed, as much as the recycling of dysfunctional organelles and misfolded proteins might benefit neurons, excessive activating of autophagy may also result in autophagic cell death. Therefore, the use of molecules targeting the autophagic pathways will require important fine-tuning. In summary, our data demonstrate that OLE possesses neuroprotective effects in an in vitro model of PD when administered preventively as a pre-treatment. In addition, OLE displays anti-oxidative and intriguing autophagy-modulating. Taken together, these data solidify OLE as a candidate for the development of novel preventive therapies in neurodegenerative diseases with a facet of oxidative stress and/or impairment of autophagy, such as in PD. 4. Materials and Methods 4.1. Drugs and Chemicals All reagents were purchased from Sigma (St. Louis, MO, USA) unless noted otherwise. 4.2. Cell Culture and Treatments A rat pheochromocytoma cell line (PC12 cells) was obtained from American Type Culture Collection (ATCC, Rockville, MD, USA) and maintained in a humidified environment at 37 °C and 5% CO2 atmosphere. They were grown in RPMI 1640 medium supplemented with 10% heat-inactivated horse serum, 5% heat-inactivated fetal bovine serum (FBS, Corning Cellgro, Manassas, VA, USA) and gentamicin (50 µg/mL). Neuronal differentiation was evoked by NGF-7S (50 ng/mL) in RPMI 1640 supplemented with 1% FBS for 7 days, as already described [42,43,46]. Fully NGF-differentiated PC12 cells (neuronal dopaminergic cellular phenotype) were pre-treated with 10−12 M OLE aglycone for 3 h, followed by addition of either 40 µM 6-OHDA for 24 h, 80 µM DDC for 3 h, or 100 nM BAF for 1 h. BAF was administered on live cells 1 h prior to protein extraction or cell fixation as a tool to evaluate autophagic flux because it prevents fusion of lysosome and autophagosome due to its inhibitory effect on V-ATPase proton pumps necessary for vesicle acidification. Therefore, it prevents the degradation of autophagic cargo and provokes the accumulation of autophagic vesicles in the cytoplasm [70,71,72]. All of these experimental conditions were selected after time course and dose response studies [43]. All experiments were performed in phenol red-free RPMI medium supplemented with 1% charcoal-stripped serum to avoid any intereference issuing from endogenous sera steroids. 4.3. Cytotoxicity Measurements Cytotoxicity was evaluated by a colorimetric assay based on the measurement of LDH activity released from damaged cells into the supernatant, as already described [55]. LDH, a stable cytoplasmic enzyme constitutively expressed in all cells, is rapidly released into the cell culture upon plasma membrane damage. Enzyme activity in the cell culture correlates with the proportion of lysed cells [73]. Briefly, 100 μL of cell-free supernatant served to quantify LDH activity by indirectly measuring transformation of lactate to pyruvate, which ultimately leads to the reduction of a tetrazolium salt whose absorbance is read at 490 nm in a microplate reader (Thermo Lab Systems, Franklin, MA, USA). Total cellular LDH was determined by lysing the cells with 1% Triton X-100 (high control); the assay medium was used as a low control and was subtracted from all absorbance measurements: (1) Cytotoxicity (%) = Experimental value−Low controlHigh control−Low control × 100 4.4. Apoptosis-Specific DNA Denaturation Apoptosis-specific detection of DNA denaturation by formamide was assessed with the single-stranded DNA (ssDNA) apoptosis enzyme-linked immunosorbent assay (ELISA) kit (Chemicon International, Billerica, MA, USA) as already described [43,63]. This procedure is based on selective DNA denaturation by formamide in apoptotic cells but not in necrotic cells or in cells with DNA damage in the absence of apoptosis [74]. Specificity of this technique relies on the particular chromatic configuration of DNA in cells undergoing apoptosis. The detection of denatured DNA was performed with a monoclonal antibody highly specific to ssDNA and a peroxidase-labeled secondary antibody on fixed neuronal PC12 cells, seeded at 25,000 cells/cm2 in 96-well plates. The reaction was then stopped with a hydrochloric acid solution and ssDNA was quantified by measuring absorbance at 405 nm in a microplate reader (ThermoFisher Scientifics, Ottawa, ON, Canada). ssDNA was analyzed with reference to control conditions. Absorbance of positive (wells coated with provided ssDNA) and negative controls (wells treated with S1 nuclease that digests ssDNA) served as quality controls for the ELISA assay, as previously described [42,46]. 4.5. Detection of Mitochondrial Superoxide Anion MitoSOX™ Red (Invitrogen, Burlington, ON, Canada) was used to estimate intracellular superoxide anion production as already described [42,43,46]. Neuronal PC12 cells were seeded at 25,000 cells/cm2, differentiated and treated on collagen-coated circular glass coverslips. After treatment, NGF-differentiated PC12 cells were washed with Hank’s buffered salt solution (HBSS) and incubated for 10 min at 37 °C with a 5 µM solution of MitoSOX™ Red. Nuclei were counterstained with Hoescht 33342 (5 μg/mL) for 15 min at 37 °C, and then cells were fixed with 4% paraformaldehyde (PFA), mounted on glass slides with Prolong Antifade kit (Invitrogen), examined under a Leitz Orthoplan fluorescence microscope (Leica, Wetzlar, Germany) and photographed with a QImaging camera (Nikon, Mississauga, ON, Canada). Fluorescence intensity was measured using NIS Elements 2.2 software (Nikon, Mississauga, ON, Canada). 4.6. Detection of Acidic Vesicles by Acridine Orange The vital dye acridine orange is a lipophilic, lysotropic stain that accumulates in lysosomes, late acidic autophagic vesicles (autophagolysosomes) and late endosomes [75], but not in autophagosomes who are not acidic. Neuronal PC12 cells were seeded at 25,000 cells/cm2, differentiated and treated on collagen-coated circular glass coverslips in 24-well plates. Acridine orange staining was performed immediately after experimental treatments on live cells. All coverslips were rinsed with PBS and nuclei counterstained with Hoescht 33342 (5 µg/mL) for 10 min at 37 °C. Then, cells were fixed with PFA and mounted on glass slides with Prolong Antifade kit (Invitrogen). Cells were observed and photographed with Images were acquired with an Olympus Corp (Olympus, Richmond Hill, ON, Canada). FV1200S confocal microscope using Fluoview10-ASW 4.0 software (Olympus, Richmond Hill, ON, Canada). 4.7. Immunofluorescence Microscopy Neuronal PC12 cells were seeded at 25,000 cells/cm2, differentiated and treated on collagen-coated coverslips in 24-well plates. Briefly, NGF-differentiated PC12 cells were fixed with PFA, then washed and incubated for 1 h at room temperature in a blocking and permeabilizing solution containing 1% BSA, 0.18% fish skin gelatin, 0.1% Triton X-100 and 0.02% sodium azide, as already described [42,43]. In order to monitor autophagy-related processes, cells were exposed to a primary antibody raised against LC3 (anti-LC3, Cell Signaling, Danvers, MA, USA) overnight at 4 °C. Then, cells were washed with PBS and incubated with Cy3-conjugated secondary antibody. To examine the relation between lysosomes and mitochondria, cells were incubated with primary antibody anti-LAMP2 (Novus Biologicals, Littleton, CO, USA) for 1 h at 37 °C. Then, slides were washed with PBS and stained with an alexafluor 488-conjugated secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA). Hoescht 33342 counterstained all nuclei. Images were acquired with an Olympus Corp. FV1200S confocal microscope using Fluoview10-ASW 4.0 software (Olympus, Richmond Hill, ON, Canada). 4.8. Western Blotting Assays Neuronal cells were seeded at 30,000 cells/cm2, differentiated and treated in collagen-coated 6-well plates. Total proteins were extracted (Nuclear Extraction Kit, Active Motif, Carlsbad, CA, USA) and concentrations were determined by bicinchoninic acid quantification (BCA protein assay kit, Pierce Biotechnology Inc., Rockford, IL, USA). Equal amounts of protein were loaded onto a 12% SDS-polyacrylamide gel. After electrophoretic separation (125 V, for 1.5 h), proteins were transferred onto PVDF membranes (0.22 μm pore size, BioRad) at 25 V overnight. The membranes were blocked for 30 min to 1 h and incubated overnight at 4 °C with primary antibodies anti-Bax, anti-Bcl-2, anti-LC3, anti-p62 (Progen Biotechnik GmbH, GP62-C) and anti-actin, (1:200, 1:100, 1:500, 1:1000, and 1:2000, respectively). The blots were then incubated with the appropriate peroxidase-conjugated secondary antibody (1:10,000) for 2 h at room temperature and finally developed with an enhanced chemiluminescence substrate solution (ThermoFisher Scientifics, Ottawa, ON, Canada). Immunopositive chemiluminescent signals were visualized with the AlphaEase FC imaging system (Alpha Innotech, San Leandro, CA, USA) and analyzed using AlphaEase FC (Alpha Innotech San Leandro, CA, USA) and ImageJ (https://imagej.nih.gov/ij/) software packages. 4.9. Statistical Analysis Significant differences between groups were ascertained by one-way analysis of variance (ANOVA), followed by Tukey’s post-hoc analysis, achieved with the GraphPad InStat program, version 3.06 for Windows (http://www.graphpad.com/). All data, analyzed at the 95% confidence interval, were expressed as means ± SEM. from at least 3 independent experiments. Asterisks indicate statistical differences between 6-OHDA or BAF or DDC and control (CTRL) conditions (*** p < 0.001, ** p < 0.01, and * p < 0.05) and plus signs (+) denote statistical differences between the treatment and 6-OHDA or BAF or DDC conditions (+++ p < 0.001, ++ p < 0.01, and + p < 0.05). Acknowledgments This work was funded by a Natural Sciences and Engineering Research Council of Canada (NSERC) grant to Maria-Grazia Martinoli. Justine Renaud is a NSERC-Vanier student fellow. Author Contributions Maria-Grazia Martinoli and Marc Germain conceived and designed the experiments; Imène Achour and Manon Legrand performed the experiments; Imène Achour, Anne-Marie Arel-Dubeau and Justine Renaud analyzed the data; Everaldo Attard contributed reagents; Maria-Grazia Martinoli, Marc Germain wrote the paper; and Justine Renaud edited the paper. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Oleuropein (OLE) prevents 6-hydroxydopamine (6-OHDA)-induced neuronal death. (A) Effect of OLE on 6-OHDA-induced toxicity was evaluated by measuring lactate dehydrogenase (LDH) activity in supernatants of damaged cells. Neuronal PC12 cells were pre-treated with or without 10−12 M OLE for 3 h, then 40 µM 6-OHDA were added for 24 h. Pre-treatment with OLE partially prevented neuronal PC12 cell death induced by 6-OHDA. Values are the average of six samples from three independent experiments for a total of 18 measurements. Data are expressed as means ± SEM (standard error of the mean). *** p ˂ 0.001 compared to control condition (CTRL) and + p ˂ 0.05 compared to 6-OHDA; (B) Levels of apoptotic cells were assessed by detecting apoptosis-specific DNA denaturation by formamide using a monoclonal antibody against single-stranded DNA. Neuronal PC12 cells treated with 40 µM 6-OHDA alone show a significant increase in apoptosis compared to CTRL. Administration of 10−12 M OLE 3 h prior to 6-OHDA treatment promotes a decrease in DNA denaturation. Values are the average of six samples from three independent experiments for a total of 18 measurements. Data are expressed as means ± SEM. ** p ˂ 0.01 compared to CTRL and ++ p ˂ 0.01 compared to 6-OHDA. Figure 2 OLE modulates the expression of Bax/Bcl-2 ratio. CTRL and OLE conditions do not show modulation of the pro-apoptotic Bax/anti-apoptotic Bcl-2 ratio. Treatment with 40 µM 6-OHDA for 24 h significantly increases the Bax/Bcl-2 ratio whereas a 3 h pre-treatment with 10−12 M OLE prevents this increment. Bottom: Bax and Bcl-2 bands, as revealed by Western blotting. Bax and Bcl-2 optical densities were measured on the same membrane. Ratios were performed between Bax and Bcl-2 of the same condition. Data are expressed as means ± SEM, n = 3. *** p < 0.001 compared with CTRL; and + p < 0.05 compared with 6-OHDA. Figure 3 OLE mitigates mitochondrial superoxide anion production. (A) Mitochondrial superoxide anion, as revealed by specific mitochondrial MitoSOX™ Red fluorescent signal, is observed in cells exposed for 3 h to 80 µM N,N-diethyldithiocarbamate (DDC), a superoxide dismutase inhibitor used to evoke superoxide accumulation. When pre-treated for 3 h with 10−12 M OLE, neuronal PC12 cells exhibit reduced levels of red fluorescence. Untreated control and OLE-treated cells show similarly low fluorescence levels. Nuclei were counterstained with Hoescht 33342. Scale bar = 10 μm; (B) Histogram represents semi-quantitative measures of MitoSOX™ Red fluorescence. OLE significantly dampens mitochondrial reactive oxygen species (ROS) generation induced by DDC. Data are expressed as relative fluorescence units (R.F.U.) and are means ± SEM, n = 3. * p ˂ 0.05 compared to CTRL and + p ˂ 0.05 compared to DDC. Figure 4 OLE administered alone does not affect the amount of LC3-II-positive vesicles. (A) Representative microphotographs show similar numbers of fluorescent vesicles between CTRL and neuronal PC12 cells treated for 3 h with 10−12 M OLE alone, as revealed by LC3-II immunofluorescence (red). Nuclei were counterstained with Hoescht 33342. Scale bar = 10 μm; (B) Histogram represents semi-quantitative measures of LC3-II red fluorescence. OLE and CTRL show no significant differences. Figure 5 OLE regulates LC3-II and p62 protein expression. (A) Western blotting analyses demonstrate that a treatment with 10−12 M OLE alone does not significantly increase LC3-II protein expression compared to CTRL. Treatment of neuronal PC12 cells for 1 h with 100 nM bafilomycin A1 (BAF), an inhibitor of autophagic vesicle clearance, dramatically increases LC3-II protein expression. Pre-treatment with OLE prior to BAF administration reduces the amount of LC3-II. Data are expressed as means ± SEM, n = 3. ** p ˂ 0.01 compared to CTRL and ++ p ˂ 0.01 compared to BAF; (B) Contrarily to LC3-II, p62 expression is significantly increased by OLE alone. Treatment with BAF strongly increases p62 expression while a pre-treatment with OLE partially rescues p62 levels. β actin served as internal control. Data are expressed as means ± SEM, n = 3. ** p ˂ 0.01 compared to CTRL, * p ˂ 0.05 compared to CTRL and ++ p ˂ 0.01 compared to BAF. Figure 6 OLE decreases the number of acidic vesicles (red) but not lysosomes (green). (A) Representative microphotographs show that a 3 h treatment with 10−12 M OLE increases the number of acidic vesicles in neuronal PC12 cells as revealed by acridine orange staining (red) compared to CTRL. Conversely, immunofluorescence labeling of lysosome-associated membrane protein 2-a LAMP2 (green), a lysosome-specific marker, suggests no effect on the number of lysosomal vesicles. Scale bar = 10 μm; (B) Histogram represents semi-quantitative measures of acridine orange (red) and LAMP2 (green) fluorescence. Data are expressed as means ± SEM. n = 3. ** p < 0.01. ==== Refs References 1. Twig G. Elorza A. Molina A.J. Mohamed H. Wikstrom J.D. Walzer G. Stiles L. Haigh S.E. Katz S. Las G. Fission and selective fusion govern mitochondrial segregation and elimination by autophagy EMBO J. 2008 27 433 446 10.1038/sj.emboj.7601963 18200046 2. Muñoz P. Huenchuguala S. Paris I. Segura-Aguilar J. Dopamine oxidation and autophagy Parkinson’s Dis. 2012 10.1155/2012/920953 3. Xiang W. Schlachetzki J.C. Helling S. Bussmann J.C. Berlinghof M. Schäffer T.E. Marcus K. Winkler J. Klucken J. Becker C.M. Oxidative stress-induced posttranslational modifications of alpha-synuclein: specific modification of alpha-synuclein by 4-hydroxy-2-nonenal increases dopaminergic toxicity Mol. Cell. Neurosci. 2013 54 71 83 10.1016/j.mcn.2013.01.004 23369945 4. Kleinknecht A. Popova B. Lázaro D.F. Pinho R. Valerius O. Outeiro T.F. Braus G.H. C-terminal tyrosine residue modifications modulate the protective phosphorylation of serine 129 of α-synuclein in a yeast model of Parkinson’s disease PLoS Genet. 2016 12 10.1371/journal.pgen.1006098 27341336 5. Dehay B. Bové J. Rodríguez-Muela N. Perier C. Recasens A. Boya P. Vila M. Pathogenic lysosomal depletion in Parkinson’s disease J. Neurosci. 2010 30 12535 12544 10.1523/JNEUROSCI.1920-10.2010 20844148 6. Demers-Lamarche J. Guillebaud G. Tlili M. Todkar K. Bélanger N. Grondin M. Nguyen A.P. Michel J. Germain M. Loss of Mitochondrial Function Impairs Lysosomes J. Biol. Chem. 2016 291 10263 10276 10.1074/jbc.M115.695825 26987902 7. More S.V. Kumar H. Kim I.S. Song S.Y. Choi D.K. Cellular and molecular mediators of neuroinflammation in the pathogenesis of Parkinson’s disease Mediators Inflamm. 2013 10.1155/2013/952375 23935251 8. Tsang A.H. Chung K.K. Oxidative and nitrosative stress in Parkinson’s disease Biochim. Biophys. Acta 2009 1792 643 650 10.1016/j.bbadis.2008.12.006 19162179 9. Tai H.C. Schuman E.M. Ubiquitin, the proteasome and protein degradation in neuronal function and dysfunctions Nat. Rev. Neurosci. 2008 9 826 838 10.1038/nrn2499 18931696 10. Patten D.A. Germain M. Kelly M.A. Slack R.S. Reactive oxygen species: Stuck in the middle of neurodegeneration J. Alzheimers Dis. 2010 20 S357 S367 20421690 11. Tramutola A. Lanzillotta C. Perluigi M. Allan Butterfield D. Oxidative stress, protein modification and Alzheimer disease Brain Res. Bull. 2016 in press 10.1016/j.brainresbull.2016.06.005 27316747 12. Tanik S.A. Schultheiss C.E. Volpicelli-Daley L.A. Brunden K.R. Lee V.M. Lewy body-like α-synuclein aggregates resist degradation and impair macroautophagy J. Biol. Chem. 2013 288 15194 15210 10.1074/jbc.M113.457408 23532841 13. Soper J.H. Kehm V. Burd C.G. Bankaitis V.A. Lee V.M. Aggregation of alpha-synuclein in S. cerevisiae is associated with defects in endosomal trafficking and phospholipid biosynthesis J. Mol. Neurosci. 2011 43 391 405 10.1007/s12031-010-9455-5 20890676 14. Banerjee R. Beal M.F. Thomas B. Autophagy in neurodegenerative disorders: pathogenic roles and therapeutic implications Trends Neurosci. 2010 33 541 549 10.1016/j.tins.2010.09.001 20947179 15. Son J.H. Shim J.H. Kim K.H. Ha J.Y. Han J.Y. Neuronal autophagy and neurodegenerative diseases Exp. Mol. Med. 2012 44 89 98 10.3858/emm.2012.44.2.031 22257884 16. Pan T. Kondo S. Le W. Jankovic J. The role of autophagy-lysosome pathway in neurodegeneration associated with Parkinson’s disease Brain 2008 131 1969 1978 10.1093/brain/awm318 18187492 17. Redmann M. Darley-Usmar V. Zhang J. The role of autophagy, mitophagy and lysosomal functions in modulating bioenergetics and survival in the context of redox and proteotoxic damage: Implications for neurodegenerative diseases Aging Dis. 2016 7 150 162 10.14336/AD.2015.0820 27114848 18. Tan C.C. Yu J.T. Tan M.S. Jiang T. Zhu X.C. Tan L. Autophagy in aging and neurodegenerative diseases: implications for pathogenesis and therapy Neurobiol. Aging 2014 35 941 957 10.1016/j.neurobiolaging.2013.11.019 24360503 19. Gao L. Jiang T. Guo J. Liu Y. Cui G. Gu L. Su L. Zhang Y. Inhibition of autophagy contributes to ischemic postconditioning-induced neuroprotection against focal cerebral ischemia in rats PLoS ONE 2012 7 10.1371/journal.pone.0046092 23029398 20. Xu F. Li J. Ni W. Shen Y.W. Zhang X.P. Peroxisome proliferator-activated receptor-γ agonist 15d-prostaglandin J2 mediates neuronal autophagy after cerebral ischemia-reperfusion injury PLoS ONE 2013 8 1293 10.1371/journal.pone.0055080 23372817 21. Sarkar S. Raymick J. Imam S. Neuroprotective and Therapeutic Strategies against Parkinson’s Disease: Recent Perspectives Int. J. Mol. Sci. 2016 17 904 10.3390/ijms17060904 27338353 22. Dubois B. Hampel H. Feldman H.H. Scheltens P. Aisen P. Andrieu S. Bakardjian H. Benali H. Bertram L. Blennow K. Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria Alzheimers Dement. 2016 12 292 323 10.1016/j.jalz.2016.02.002 27012484 23. Virmani A. Pinto L. Binienda Z. Syed A. Neuroprotection by what you eat Mol. Neurobiol. 2013 48 353 362 10.1007/s12035-013-8498-3 23813102 24. Bousquet M. Calon F. Cicchetti F. Impact of omega-3 fatty acids in Parkinson’s disease Ageing Res. 2011 10 453 463 10.1016/j.arr.2011.03.001 21414422 25. Koppula S. Kumar H. More S.V. Lim H.W. Hong S.M. Choi D.K. Recent updates in redox regulation and free radical scavenging effects by herbal products in experimental models of Parkinson’s disease Molecules 2012 17 11391 11420 10.3390/molecules171011391 23014498 26. Hang L. Basil A.H. Lim K.L. Nutraceuticals in Parkinson’s Disease Neuromol. Med. 2016 10.1007/s12017-016-8398-6 27147525 27. Vauzour D. Rodriguez-Mateos A. Corona G. Oruna-Concha M.J. Spencer J.P. Polyphenols and human health: Prevention of disease and mechanisms of action Nutrients 2010 2 1106 1131 10.3390/nu2111106 22254000 28. Rigacci S. Stefani M. Nutraceutical properties of olive oil polyphenols: An itinerary from cultured cells through animal models to humans Int. J. Mol. Sci. 2016 17 10.3390/ijms17060843 27258251 29. Umeno A. Horie M. Murotomi K. Nakajima Y. Yoshida Y. Antioxidative and Antidiabetic Effects of Natural Polyphenols and Isoflavones Molecules 2016 21 708 10.3390/molecules21060708 27248987 30. Attard E. Martinoli M.G. Cucurbitacin E, An experimental lead triterpenoid with anticancer, immunomodulatory and novel effects against degenerative diseases Curr. Top. Med. Chem. 2015 15 1708 1713 10.2174/1568026615666150427121331 25915611 31. Renaud J. Nabavi S.F. Daglia M. Nabavi S.M. Martinoli M.G. Epigallocatechin-3-gallate, a promising molecule for Parkinson’s Disease? Rejuvenation Res. 2015 18 257 269 10.1089/rej.2014.1639 25625827 32. Renaud J. Martinoli M.G. Resveratrol as a protective molecule for neuroinflammation: A review of mechanisms Curr. Pharm. Biotechnol. 2014 15 318 329 10.2174/1389201015666140617101332 24938890 33. Rahmani H. Albutti A.S. Aly S.M. Therapeutics role of olive fruits/oil in the prevention of diseases via modulation of anti-oxidant, anti-tumour and genetic activity Int. J. Clin. Exp. Med. 2014 7 799 808 24955148 34. Efentakis P. Iliodromitis E.K. Mikros E. Papachristodoulou A. Dagres N. Skaltsounis A.L. Andreadou I. Effects of the olive tree leaf constituents on myocardial oxidative damage and atherosclerosis Planta Med. 2015 81 648 654 10.1055/s-0035-1546017 26018920 35. Andreadou I. Iliodromitis E.K. Mikros E. Constantinou M. Agalias A. Magiatis P. Skaltsounis A.L. Kamber E. Tsantili-Kakoulidou A. Kremastinos D.T. The olive constituent oleuropein exhibits antiischemic, antioxidative, and hypolipidemic effects in anesthetized rabbits J. Nutr. 2006 136 2213 2219 16857843 36. Sarbishegi M. Mehraein F. Soleimani M. Antioxidant role of oleuropein on midbrain and dopaminergic neurons of substantia nigra in aged rats Iran. Biomed. J. 2014 18 16 22 24375158 37. Pasban-Aliabadi H. Esmaeili-Mahani S. Sheibani V. Abbasnejad M. Mehdizadeh A. Yaghoobi M.M. Inhibition of 6-hydroxydopamine-induced PC12 cell apoptosis by olive (Olea europaea L.) leaf extract is performed by its main component oleuropein Rejuvenation Res. 2013 16 134 142 10.1089/rej.2012.1384 23394606 38. Pantano D. Luccarini I. Nardiello P. Servili M. Stefani M. Casamenti F. Oleuropein aglycone and polyphenols from olive mill wastewater ameliorate cognitive deficits and neuropathology Br. J. Clin. Pharmacol. 2016 10.1111/bcp.12993 27131215 39. Casamenti F. Grossi C. Rigacci S. Pantano D. Luccarini I. Stefani M. Oleuropein aglycone: A possible drug against degenerative conditions: In vivo evidence of its effectiveness against Alzheimer’s disease J. Alzheimers Dis. 2015 45 679 688 25649656 40. Rigacci S. Miceli C. Nediani C. Berti A. Cascella R. Pantano D. Nardiello P. Luccarini I. Casamenti F. Stefani M. Oleuropein aglycone induces autophagy via the AMPK/mTOR signalling pathway: A mechanistic insight Oncotarget 2015 6 35344 35357 26474288 41. Van Cauwenberghe C. Vandendriessche C. Libert C. Vandenbroucke R.E. Caloric restriction: Beneficial effects on brain aging and Alzheimer’s disease Mamm. Genome 2016 27 300 319 10.1007/s00335-016-9647-6 27240590 42. Renaud J. Bournival J. Zottig X. Martinoli M.G. Resveratrol protects DAergic PC12 cells from high glucose-induced oxidative stress and apoptosis: Effect on p53 and GRP75 localization Neurotox. Res. 2014 25 110 123 10.1007/s12640-013-9439-7 24218232 43. Arel-Dubeau A.M. Longpre F. Bournival J. Tremblay C. Demers-Lamarche J. Hasvoka P. Attard E. Germain M. Martinoli M.G. Cucurbitacin E has neuroprotective properties and autophagic modulating activities on dopaminergic neurons Oxid. Med. Cell. Longev. 2014 10.1155/2014/425496 25574337 44. Kang M.H. Reynolds C.P. Bcl-2 inhibitors: Targeting mitochondrial apoptotic pathways in cancer therapy Clin. Cancer Res. 2009 15 1126 1132 10.1158/1078-0432.CCR-08-0144 19228717 45. Cory S. Adams J.M. The Bcl2 family: regulators of the cellular life-or-death switch Nat. Rev. Cancer 2002 2 647 656 10.1038/nrc883 12209154 46. Carange J. Longpré F. Daoust B. Martinoli M.G. 24-Epibrassinolide, a phytosterol from the brassinosteroid family, protects dopaminergic cells against MPP+-induced oxidative stress and apoptosis J. Toxicol. 2011 1293 10.1155/2011/392859 21776258 47. Barth S. Glick D. Macleod K.F. Autophagy: Assays and artifacts J. Pathol. 2010 221 117 124 10.1002/path.2694 20225337 48. Klionsky D.J. Abeliovich H. Agostinis P. Aliev G. Askew D.S. Baba M. Baehrecke E.H. Bahr B.A. Ballabio A. Bamber B.A. Guidelines for the use and interpretation of assays for monitoring autophagy in higher eukaryotes Autophagy 2008 4 151 175 10.4161/auto.5338 18188003 49. Calabrese V. Cornelius C. Dinkova-Kostova A.T. Iavicoli I. di Paola R. Koverech A. Cuzzocrea S. Rizzarelli E. Calabrese E.J. Cellular stress responses, hormetic phytochemicals and vitagenes in aging and longevity Biochim. Biophys. Acta 2012 1822 753 783 10.1016/j.bbadis.2011.11.002 22108204 50. Nile S.H. Park S.W. Edible berries: bioactive components and their effect on human health Nutrition 2014 30 134 144 10.1016/j.nut.2013.04.007 24012283 51. Costa S.L. Silva V.D. Dos Santos Souza C. Santos C.C. Paris I. Muñoz P. Segura-Aguilar J. Impact of Plant-Derived Flavonoids on Neurodegenerative Diseases Neurotox. Res. 2016 30 41 52 10.1007/s12640-016-9600-1 26951456 52. Reglodi D. Renaud J. Tamas A. Tizabi Y. Socías S.B. del-Bel E. Raisman-Vozari R. Novel tactics for neuroprotection in Parkinson’s disease: Role of antibiotics, polyphenols and neuropeptides Prog. Neurobiol. 2015 10.1016/j.pneurobio.2015.10.004 26542398 53. De Bock M. Thorstensen E.B. Derraik J.G. Henderson H.V. Hofman P.L. Cutfield W.S. Human absorption and metabolism of oleuropein and hydroxytyrosol ingested as olive (Olea europaea L.) leaf extract Mol. Nutr. Food Res. 2013 57 2079 2085 10.1002/mnfr.201200795 23766098 54. García-Villalba R. Larrosa M. Possemiers S. Tomás-Barberán F.A. Espín J.C. Bioavailability of phenolics from an oleuropein-rich olive (Olea europaea ) leaf extract and its acute effect on plasma antioxidant status: comparison between pre- and postmenopausal women Eur. J. Nutr. 2014 53 1015 1027 10.1007/s00394-013-0604-9 24158653 55. Gélinas S. Martinoli M.G. Neuroprotective effect of estradiol and phytoestrogens on MPP+-induced cytotoxicity in neuronal PC12 cells J. Neurosci. Res. 2002 70 90 96 10.1002/jnr.10315 12237867 56. Kadota T. Yamaai T. Saito Y. Akita Y. Kawashima S. Moroi K. Inagaki N. Kadota N. Expression of dopamine transporter at the tips of growing neurites of PC12 cells J. Histochem. Cytochem. 1996 44 989 996 10.1177/44.9.8773564 8773564 57. Nilsen J. Mor G. Naftolin F. Raloxifene induces neurite outgrowth in estrogen receptor positive PC12 cells Menopause 1998 5 211 216 10.1097/00042192-199805040-00005 9872486 58. Chiasson K. Lahaie-Collins V. Bournival J. Delapierre B. Gélinas S. Martinoli M.G. Oxidative stress and 17 alpha and 17 beta estradiol modulate neurofilaments differently J. Mol. Neurosci. 2006 30 297 309 10.1385/JMN:30:3:297 17401155 59. Bournival J. Francoeur M.A. Renaud J. Martinoli M.G. Quercetin and sesamin protect neuronal PC12 cells from high-glucose-induced oxidation, nitrosative stress and apoptosis Rejuvenation Res. 2012 15 322 333 10.1089/rej.2011.1242 22524206 60. Blasina M.F. Vaamonde L. Morquio A. Echeverry C. Arredondo F. Dajas F. Differentiation induced by Achyrocline satureioides (Lam) infusion in PC12 cells Phytother. Res. 2009 23 1263 1269 10.1002/ptr.2607 19370542 61. El Omri A. Han J. Yamada P. Kawada K. Ben Abdrabbah M. Isoda H. Rosmarinus officinalis polyphenols activate cholinergic activities in PC12 cells through phosphorylation of ERK1/2 J. Ethnopharmacol. 2010 131 451 458 10.1016/j.jep.2010.07.006 20633629 62. Olatunji O.J. Feng Y. Olatunji O.O. Tang J. Ouyang Z. Su Z. Cordycepin protects PC12 cells against 6-hydroxydopamine induced neurotoxicity via its antioxidant properties Biomed. Pharmacother. 2016 81 7 14 10.1016/j.biopha.2016.03.009 27261571 63. Bournival J. Quessy P. Martinoli M.G. Protective effects of resveratrol and quercetin against MPP+-induced oxidative stress act by modulating markers of apoptotic death in dopaminergic neurons Cell. Mol. Neurobiol. 2009 29 1169 1180 10.1007/s10571-009-9411-5 19466539 64. Li L. Gao L. Song Y. Qin Z.-H. Liang Z. Activated cathepsin L is associated with the switch from autophagy to apoptotic death of SH-SY5Y cells exposed to 6-hydroxydopamine Biochem. Biophys. Res. Commun. 2016 470 579 585 10.1016/j.bbrc.2016.01.102 26797274 65. Saftig P. Beertsen W. Eskelinen E.L. LAMP2, a control step for phagosome and autophagosome maturation Autophagy 2008 4 510 512 10.4161/auto.5724 18376150 66. Menendez J.A. Joven J. Aragonès G. Barrajón-Catalán E. Beltrán-Debón R. Borrás-Linares I. Camps J. Corominas-Faja B. Cufí S. Fernández-Arroyo S. Xenohormetic and anti-aging activity of secoiridoid polyphenols present in extra virgin olive oil: a new family of gerosuppressant agents Cell Cycle 2013 12 555 578 10.4161/cc.23756 23370395 67. Dattilo S. Mancuso C. Koverech G. Di Mauro P. Ontario M.L. Petralia C.C. Petralia A. Maiolino L. Serra A. Calabrese E.J. Heat shock proteins and hormesis in the diagnosis and treatment of neurodegenerative diseases Immun. Ageing 2015 12 10.1186/s12979-015-0046-8 26543490 68. Kikusato M. Muroi H. Uwabe Y. Furukama K. Toyomizu M. Oleuropein induces mitochondrial biogenesis and decreases reactive oxygen species generation in cultured avian muscle cells, possibly via an up-regulation of peroxisome proliferator-activated receptor γ coactivator-1α Anim. Sci. J. 2016 10.1111/asj.12559 26916829 69. Kim T. Mehta S.L. Kaimal B. Lyons K. Dempsey R.J. Vemuganti R. Poststroke induction of α-synuclein mediates ischemic brain damage J. Neurosci. 2016 36 7055 7065 10.1523/JNEUROSCI.1241-16.2016 27358461 70. Yin W.Y. Ye Q. Huang H.J. Xia N.G. Chen Y.Y. Zhang Y. Qu Q.M. Salidroside protects cortical neurons against glutamate-induced cytotoxicity by inhibiting autophagy Mol. Cell. Biochem. 2016 419 53 64 10.1007/s11010-016-2749-3 27357827 71. Yamamoto A. Tagawa Y. Yoshimori T. Moriyama Y. Masaki R. Tashiro Y. Bafilomycin A1 prevents maturation of autophagic vacuoles by inhibiting fusion between autophagosomes and lysosomes in rat hepatoma cell line, H-4-II-E cells Cell Struct. Funct. 1998 23 33 42 10.1247/csf.23.33 9639028 72. Ganley I.G. Wong P.M. Gammoh N. Jiang X. Disctinct autophagosomal-lysosomal fusion mechanism revealed by thapsigargin-induced autophagy arrest Mol. Cell 2011 42 731 743 10.1016/j.molcel.2011.04.024 21700220 73. Decker T. Lohmann-Matthes M.L. A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity J. Immunol. Methods 1998 115 61 69 10.1016/0022-1759(88)90310-9 3192948 74. Frankfurt O.S. Krishan A. Identification of apoptotic cells by formamide-induced DNA denaturation in condensed chromatin J. Histochem. Cytochem. 2001 49 369 378 10.1177/002215540104900311 11181740 75. Lin C. Tsai S.-H. Tseng M.T. Peng S.-F. Kuo C.-S. Lin M.-W. Hsu Y.-M. Lee M.-R. Amagaya S. Huang W.-W. AKT serine/threonine protein kinase modulates baicalin-triggered autophagy in human bladder cancer T24 cells Int. J. Oncol. 2013 42 993 1000 23354080

PMC005xxxxxx/PMC5000691.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081294ijms-17-01294ArticleEvidence of Decoupling Protein Structure from Spidroin Expression in Spider Dragline Silks Blamires Sean J. 1*Kasumovic Michael M. 1Tso I-Min 2Martens Penny J. 3Hook James M. 4Rawal Aditya 4Hardy John G. Academic Editor1 Evolution & Ecology Research Centre, School of Biological, Earth & Environmental Sciences, University of New South Wales, Sydney 2052, Australia; [email protected] Department of Life Science, Tunghai University, Taichung 40704, Taiwan; [email protected] Graduate School of Biomedical Engineering, University of New South Wales, Sydney 2052, Australia; [email protected] NMR Facility, Mark Wainwright Analytical Centre, University of New South Wales, Sydney 2052, Australia; [email protected] (J.M.H.); [email protected] (A.R.)* Correspondence: [email protected]; Tel.: +61-2-9385-126109 8 2016 8 2016 17 8 129430 6 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).The exceptional strength and extensibility of spider dragline silk have been thought to be facilitated by two spidroins, major ampullate spidroin 1 (MaSp1) and major ampullate spidroin 2 (MaSp2), under the assumption that protein secondary structures are coupled with the expressed spidroins. We tested this assumption for the dragline silk of three co-existing Australian spiders, Argiope keyserlingi, Latrodectus hasselti and Nephila plumipes. We found that silk amino acid compositions did not differ among spiders collected in May. We extended these analyses temporally and found the amino acid compositions of A. keyserlingi silks to differ when collected in May compared to November, while those of L. hasselti did not. To ascertain whether their secondary structures were decoupled from spidroin expression, we performed solid-state nuclear magnetic resonance spectroscopy (NMR) analysis on the silks of all spiders collected in May. We found the distribution of alanine toward β-sheet and 3,10helix/random coil conformations differed between species, as did their relative crystallinities, with A. keyserlingi having the greatest 3,10helix/random coil composition and N. plumipes the greatest crystallinity. The protein secondary structures correlated with the mechanical properties for each of the silks better than the amino acid compositions. Our findings suggested that a differential distribution of alanine during spinning could decouple secondary structures from spidroin expression ensuring that silks of desirable mechanical properties are consistently produced. Alternative explanations include the possibility that other spidroins were incorporated into some silks. amino acidshigh performance liquid chromatographymechanical propertiesorb weaving spidersprotein secondary structuressilk spinningsolid-state nuclear magnetic resonance spectroscopy ==== Body 1. Introduction The exceptional strength, extensibility and toughness of spider dragline, or major ampullate (MA), silk make it a desirable material for multiple industrial uses [1]. Dragline silk is traditionally thought to be comprised of two proteins (spidroins), conventionally called major ampullate spidroin 1 (MaSp1) and major ampullate spidroin 2 (MaSp2). Generally, the structures of these proteins are considered critical to the mechanical function of dragline silk. Techniques that can probe protein structures include small and wide angle X-ray scattering, Fourier Transform Infrared Spectroscopy (FTIR) and various forms of nuclear magnetic resonance (NMR) spectroscopy [2,3,4,5,6,7]. Small and wide angle X-ray scattering are used explicitly for examining the size, density and orientation of crystalline and non-crystalline structures [5]. Since NMR detects magnetically active isotopes such as 1H, 13C and 15N, it is used to identify the molecular orientations and bonding arrangements within protein secondary structures [8]. As such, solid state nuclear magnetic resonance (ssNMR) spectroscopy has been effectively used to assess the relationship between spider dragline silk amino acid compositions and the formation of protein secondary structures, as well as protein chain dynamics [7,8,9,10,11,12,13,14,15]. ssNMR has shown MaSp1 to consist of multiple (GA)n, (GGX)n and (A)n repeated amino acid sequences (G = glycine, A = alanine and X = other amino acids) and it is assumed that these sequences promote the formation of crystalline β-sheet structures in assembled fibers [3,9,16]. MaSp2, on the other hand, consists of multiple (GPGXX)n repeated sequences (P = proline) [10,17]. This sequence is currently assumed to self-assemble into β-spirals and type-II β-turns in the silk fibers [6,18]. Since MaSp2 contains (GPGXX)n repeated sequences, the proline composition of dragline silk has been used as an indicator of the presence of MaSp2. Thus, the combined expression of MaSp1 and MaSp2 are assumed to be coupled to the presence of various quantities of crystalline β-sheets, β-spirals and type-II β-turns, which are in turn thought to provide dragline silk with its great strength and extensibility [1]. The positioning of the alanine residues is an important indicator of dragline silk structure that is detectable using ssNMR [13,19,20,21]. Generally, the majority (>80%) of the alanine in MaSp1 resides in β-sheets, predominantly as (A)n repeated sequences [13,19,22,23]. The rest of the MaSp1 alanine lies in α-helices, 3,10helices and type-II β-turns [23]. The distribution of alanine in MaSp2 varies substantially among silks from different spider species [13,21,22,23]. Since the C=O segments bond weakly with adjacent amine segments [24], alanine is mobile during spinning and this mobility is identifiable using ssNMR techniques [19]. Hence the repositioning of alanine within the spidroins might be a detectable mechanism influencing protein secondary structures within dragline silk fibers. Glycine residues are, likewise, important identifiers of dragline silk structure when using ssNMR. In MaSp1 glycine primarily forms either (GA)n, repeating sequences within β-sheets or (GGX)n repeating sequences within GlyII-helices [25]. Glycine may be mobile during spinning, albeit to a lesser extent than alanine, so variability in glycine distribution between individual dragline silks might also be expected. Proline is found within the (GPGXX)n repeating sequences of MaSp2 and is identified using ssNMR by its tendency to form kinks [18]. It is thought to facilitate extensibility by forming weak hydrogen bonding between the structural components that enable the chains to freely slide past each other [26,27,28,29]. ssNMR studies of dragline silk have to date mostly focused on elucidating the amino acid compositions, secondary structures, and molecular dynamics of the spidroins from a few model species, such as Nephila clavipes, Latrodectus hesperus and Argiope aurantia [17,23,25,30]. One study [13] used ssNMR to compare the structures and properties of dragline silks from a range of species from four different spider genera: Araneus, Argiope, Nephila and Latrodectus. Among these genera Araneus and Argiope had dragline silks high in MaSp2 compositions and proline molar compositions of ~11%–14%, while Nephila and Latrodectus were low in MaSp2 and proline compositions of only ~1%–2% [13]. Studies comparing amino acid compositions in the dragline silks of different species [16,27,31,32,33] have revealed similar dichotomies in silks produced among different spiders, i.e., spiders seem to produce silks that are either high in MaSp2 or relatively devoid of MaSp2. There, however, does not seem to be any phylogenetic relationship between the spiders expressing silks that are relatively high or low in MaSp2 [32,33]. An explanatory hypothesis for the apparent dichotomy in dragline silk spidroin expression among spiders is that factors such as ecological circumstances, or spider body size, condition, or aging affect spidroin expression. For instance, MaSp2 expression may be down-regulated when a spider is starved or deprived of nutrients [31,34,35]. The finding that silk amino acid composition can differ in the dragline silks from the same species of spider from different regions [36,37] also supports this prediction. Nevertheless, the mechanical performance of a spider’s silk may be unaffected by variations in spidroin expression [33,38,39,40]. Thus, there seem to be several ways by which variations in silk mechanical property can be induced among different spiders without sacrificing functional effectiveness. As silk flows through the major ampullate gland as a crystalline liquid the actions of pH change, salt concentrations and shear forces in the duct effect the formation of the secondary structures [41,42,43,44,45]. Accordingly, a possible reason why the mechanical properties of dragline silk sometimes vary independent of spidroin expression is that the silk spinning processes decouple spidroin expression from secondary structure formation through an “on-the-fly” redistribution of alanine or glycine residues [46]. This explanation nonetheless remains to be empirically tested in different spiders using multiple techniques. We therefore determined herein the spidroin expression, protein secondary structure and mechanical properties of the dragline silks of three co-existing species of spider from Sydney, Australia: Argiope keyserlingi, Nephila plumipes and Latrodectus hasselti using high performance liquid chromatography (HPLC), ssNMR and tensile testing techniques. All of their silks were collected in May 2014. We determined the structural formations induced by the alanine and glycine residues in the silks from the ssNMR measurements. Any incongruencies between amino acid compositions, protein secondary structures and mechanical properties among spiders were construed as indicating that spinning processes decoupled spidroin expression from protein secondary structures and, thus, mechanical properties [44,45]. We additionally investigated temporal variations in spidroin expression by performing further HPLC analyses on Argiope keyserlingi and Latrodectus hasselti silks collected in November. We regarded any similarities in amino acid composition between species at any particular time of year or any differences within species at different times of year as indicating that ecological circumstances or demographic factors influenced spidroin expressions. 2. Results and Discussion 2.1. Amino Acid Compositions We found that the dragline silk amino acid compositions did not differ between the three species examined in May (MANOVA: Wilk’s λ = 0.161; F = 2.387; df = 10, 16; p = 0.058) (Table 1a). We then expanded our spidroin expression analysis by comparing the amino acid compositions of Latrodectus hasselti and Argiope keyserlingi silks collected in November and found differences between species (MANOVA: Wilk’s λ = 0.022; F = 25.670; df = 12,52; p < 0.001) (Table 1b). Furthermore, the amino acid composition of A. keyserlingi dragline silk collected in May differed from that collected in November (MANOVA: Wilk’s λ = 0.048; F = 15.99; df = 5,4; p = 0.009), with the silks from spiders collected in November having a greater percentage alanine (p = 0.038) and proline (p < 0.001) than those collected in May. On the other hand the amino acid composition of L. hasselti dragline silk did not differ between May and November (MANOVA: Wilk’s λ = 0.313; F = 1.578; df = 5, 4; p = 0.302). By comparing the compositions attained here with those derived from Latrodectus hesperus [47] and Argiope bruennichi [48] dragline silk gene sequences (Table 1c), our results suggested that each of the spiders produced dragline silk comprising of primarily MaSp1 in May, while in November A. keyserlingi produced dragline silk that likely comprised of a greater proportion of MaSp2. Thus, we concluded that spidroin expression in the dragline silk of A. keyserlingi differs at different times of year, possibly as a result of changing ecological or demographic circumstances. We found less variability around the mean compositions for silk collected in May than for silks collected in November (see Table 1). Work and Young [49] noted that some segments within individual silk fibers could differ in amino acid composition and suggested that within a single fiber there may be differences in the ratio of MaSp1 and MaSp2. We accounted for such within fiber variability in amino acid composition here by analyzing the amino acid composition of entire dragline threads from each individual of each species. Accordingly, we considered this source of variation as not being responsible for any of the variation in dragline silk amino acid compositions between species or within species at different times of year. The spiders that we used were all found in similar urban habitats in Sydney, Australia. L. hasselti, however, was found in more sheltered microhabitats than A. keyserlingi (L. hasselti was mostly found underneath rigid structures, e.g., park benches, whereas A. keyserlingi was more frequently found exposed among vegetation). Accordingly, it seems that the amino acid composition of A. keyserlingi silk could vary in concordance with relatively small-scale environmental fluctuations. Another possibility for the fluctuations in amino acid composition in A. keyserlingi silk is that MaSp1 and MaSp2 are not the only spidroins being expressed. Recent proteomic data using mass spectrometry of solubilized dragline silk from L. hesperus followed by in-solution tryptic cleavage supported the presence of another spidroin, AcSp1 (acinform spidroin 1), within the major ampullate gland as well as in dragline silk fibers [50]. The presence of AcSp1 might explain the unusually high serine compositions in the dragline silks of all species, the lower than predicted (based on sequences of A. bruennichi dragline silk [48]) glycine compositions in A. keyserlingi’s dragline silk in May and November, and the low alanine composition of L. hasselti’s dragline silk (compared to that of L. hesperus [47]) in May (see Table 1). We found a seasonal difference in spidroin expression in A. keyserlingi dragline silk but an absence of any change in the composition in L. hasselti dragline silk, which is consistent with studies showing that amino acid compositions in spider dragline silks that are relatively low in MaSp2 vary less extensively across environments than those high in MaSp2 [31,32,35]. We controlled for diet and nutritional effects on spidroin expression here by pre-feeding the spiders a standardized solution so recent diet did not affect dragline spidroin expression in any species at any time. A likely explanation for the seasonal variation in the amino acid composition of A. keyserlingi dragline silk is that MaSp2 was considerably down-regulated in May because it is synthesized at a considerable metabolic expense [34,35]. MaSp1 synthesis, on the other hand, is significantly less expensive, so is largely unregulated [31]. The three species examined here have different seasonal activity patterns, so their MaSp2 expression is likely to be regulated differently. Studies incorporating more environmental measurements across seasons and habitats are nonetheless required to better elucidate the particular circumstances that induce changes in dragline silk spidroin expression among co-existing web building spiders. 2.2. Solid-State Nuclear Magnetic Resonance Spectroscopy For 13C cross -polarization magic-angle spin (CPMAS) NMR experiments the 13C signal of a given functional group is dependent on the 1H–13C cross-polarization dynamics. The 13C signal intensity in a CPMAS spectrum will depend upon: (a) the protonation state of the carbon; (b) the local molecular mobility of the 13C site; and (c) the concentration of the 13C species in the material. Since the signals for the alanine and glycine functional groups were constant among the three spider’s silk samples here and we compared the same kind of 13C species, i.e., the methyl species, our spectra depended only on local molecular motions and concentration. We had similar materials in all samples, i.e., large spidroins, so any difference in the local molecular motions will be a function of the degree of hydration. Although we were careful not to expose the silks to the atmosphere, we took the additional precaution of measuring the 1H spectra of the different silks. While the signals from the water peak (5–10 ppm [21]) indicated that some water binding was present, the signals were generally broad and the peaks relatively small (see Figure S1) compared to what would be expected if the material had absorbed a large amount of water, for instance during supercontraction [11,19]. Thus, we were sure that humidity did not strongly influence our spectra attained. The tall, narrow peaks associated with each of the curves of Figure S1 were all found around 0 ppm so attributed to large amino acids or molecules associated with the silk’s skin [6,8]. Subsequently, we were confident that we attained highly reliable 13C CPMAS NMR spectra, enabling us to compare the relative intensities of the 13C signals among the different dragline silks and to compare secondary structures between individuals and species. The 13C cross-polarization magic-angle spectra for the three species’ silks are presented in Figure 1. All spectra were scaled to equal the intensity of the C=O peak. An additional plot of the variation of the spider silk among individuals of each species is presented in the Supplementary Materials (Figure S2). The peaks were assigned in accordance with Creager et al. [13] to enable estimation of the different residues and domains. The most striking difference that our ssNMR analyses found between the spidroin structures of the three species silks collected in May was the relative intensities of the Cβ signal on alanine in β-sheets (peaking at 21.5 ppm) versus 3,10helices/random coil (peaking at 17.5 ppm). We found that A. keyserlingi silk collected in May had the lowest β-sheet signal intensities relative to that of 3,10helices/random coil, while the N. plumipes and L. hasselti silk collected in May had much stronger β-sheet signal intensities relative to 3,10helices/random coil (see Figure 1). Assuming similar cross polarization of the methyl sites in the alanine β-sheet or 3,10helices/random coil conformations, we estimated that the ratio of alanine in β-sheets: Alanine in 3,10helices/random coil is 1:7 in A. keyserlingi silk and 7:1 in N. plumipes silk and 2:1 in L. hasselti silk. Thus, while the amino acid compositions suggested that all of the spiders produced dragline silk comprising of primarily MaSp1 in May, ssNMR revealed that the secondary structures of the proteins within each species’ dragline silk differed profoundly. While L. hasselti dragline silk was structurally more similar to N. plumipes’ silk than A. keyserlingi’s silk, there was more structural variation between individual fibers in this species compared to the other two species (Figure S2). This may be because there were considerably high signal-to-noise ratios in the resonances attained for the L. hasselti silks because of their exceptional thinness (<2 µm). On the other hand spinning conditions within the glands of individual L. hasselti may have induced a wider array of structural variations among the individual dragline silks, or the presence of AcSp1 in L. hasselti dragline silks was variable among individuals. The secondary structures within A. keyserlingi dragline silk resembled those expected for MaSp2 predominant silk, while those within N. plumipes and L. hasselti dragline silk more resembled those expected for MaSp1 predominant silk. N. plumipes’ dragline silk yielded the narrowest signal at the alanine Cα site (70 ppm), indicating a higher degree of order and crystallinity in their silks compared to A. keyserlingi or L. hasselti silks. Since there were incongruencies between the amino acid compositions revealed by HPLC and the protein secondary structures revealed by ssNMR we concluded that silk spinning processes may have acted to dissociate the protein secondary structures from spidroin expression, or other spidroins may have been incorporated into some of the silks. 2.3. Influences Affecting Mechanical Properties Representative stress–strain curves for three individual L. hasselti, A. keyseringi and N. plumipes silks are shown in the Supplementary Materials (Figure S3). One feature of the curves is what appears to be slippage of the fibers at the early stages of extension. We expect that this was a by-product of the low resolution of measurements of our testing machine, which we found to show high variability when measuring some exceptionally fine fibers. Our procedures, nevertheless, were consistent for all of the silks analyzed and the results were detailed enough to make comparisons between species. We found that the native state silks of each species had significantly greater ultimate strength but were less extensible than their supercontracted silks, indicating that alignment of the crystalline and amorphous region proteins affected the mechanical properties of each species’ native silks independent of spidroin expression [38,51]. We suggest that studies involving intraspecific comparisons require more precise mechanical testing methods. The mechanical properties of the dragline silks from all three species differed (Figure 2). These differences reconciled well with our ssNMR structural analysis. For instance, we found N. plumipes silk to have the highest ultimate strength, which can be attributed to the high crystallinity of its silk. On the other hand, A. keyserlingi’s silk had the greatest toughness and extensibility, properties that can be attributed to its low crystallinity and greater distribution of alanine toward 3,10helices/random coils rather than β-sheets. Moreover, despite having similar amino acid compositions to the other two species A. keyserlingi silk had the greatest percent shrinkage, which may be attributable to its high proportion of helical/coiled structures, as these structures are considered more amenable to extension under tension compared to the crystalline structures [45,52,53]. The theoretical premise for this prediction is that the weak hydrogen bonds between the amino acid residues in random coil structures would likely have been accessed more readily by water resulting in loss of alignment in the amorphous region [12,28,29,54]. The amino acid composition of A. keyserlingi dragline silk collected in May did not differ from the silk amino acid compositions of the other two species, so spidroin expression does not explain the variations in mechanical properties found between the three species’ silks. We therefore deduced that the incongruencies between the amino acid compositions and protein secondary structures and mechanical properties among the three species is evidence of decoupling between silk protein secondary structures and spidroin expression. Glandular processes that might induce decoupling may include the shear forces experienced during the final phase of silk spinning inducing crystalline and amorphous region self-alignment in the spun silk [44,45,46]. Additional amorphous region alignment may be extenuated at the valve by the influence of friction during drawing [45,55]. These actions could explain why the mechanical properties of the silks in all three species varied independent of their spidroin expressions and why A. keyserlingi’s dragline silk in May resembled MaSp1 predominant silk in amino acid composition but resembled a MaSp2 predominant silk in secondary structure. An alternative possibility is that additional spidroins, such as AcSp1 [50], were additionally expressed in A. keyserlingi dragline silk, rendering the influence of amino acid composition on mechanical properties inconsistent with that expected if only MaSp1 and MaSp2 were expressed. 3. Experimental Section 3.1. Spiders and Silk Collecting We collected five adult females of three species, Argiope keyserlingi, Nephila plumipes and Latrodectus hasselti, from similar urban habitats in Sydney, New South Wales, Australia, in May 2014, as at this time all three species were active. We collected a further 5 L. hesselti and 5 A. keyserlingi in November 2014 (the time of year when the activities of these two spiders peaked) to determine whether spidroin expression varied in similar spiders at different times of year. N. plumipes was not collected in November as it was not active at this time. Upon collection of the spiders, we measured their body length to ±0.1 mm using digital Vernier calipers (Caliper Technologies Corp., Mountain View, CA, USA) and their mass to ±0.001 g using an electronic balance (Ohaus Corp., Pine Brook, NY, USA) before placing them in 115 mm (wide) × 45 mm (high) plastic circular containers. The containers had perforated wire mesh lids with a 20 mm long slit cut into them using a Stanley knife to facilitate feeding with a 50 μL micropipette. We fed all of the spiders 20 μL (L. hesselti and A. keyserlingi) or 50 μL (N. plumipes) of an unlabeled 30% w/v glucose solution daily over five days (for details of solution preparation see Blamires et al. [31,38]) to standardize the recent diet of all spiders prior to collecting their silk. We reweighed the spiders after the 5 days of feeding and individuals whose mass deviated >50% from the mean for the species (one L. hasselti and one N. plumpes collected in May) were discarded. We anaesthetized each spider using CO2 gas and carefully pulled a single dragline fiber from the spinnerets using tweezers and wrapped it around a glass tube connected to a mechanical spool. The spool was spun at a constant speed (1 m·min−1) for ~1 h whereupon 10–15 mg of silk was collected. We estimated that this amount of silk closely represents the complete store of silk from the major ampullate gland for each of these species. All silks were extracted under controlled temperature (~25 °C) and humidity (~50% R.H.) in still air, so reeling and the post-spin environment did not influence their subsequent chemical or mechanical properties. 3.2. High Performance Liquid Chromatography (HPLC) We used HPLC to determine amino acid compositions using 1–5 mg of silk collected from the 13 individual spiders (4 or 5 individual × 3 species) collected in May and 1–5 mg of silk from the 10 (five L. hasselti and five A. keyserlingi) spiders collected in November. Thus, 25 samples were prepared for HPLC analysis in total. We weighed all of the silk samples to the nearest 0.001 mg on an electronic balance, before placing them into 100-µL Eppendorf tubes, submerging them in 99% hexaflouro-isopropanol solvent (500 µL of per mg of silk) and leaving them overnight. After removal of the solvent, the samples were then placed in glass tubes and hydrolyzed in 6 M HCl for 24 h in a furnace at 115 °C. The amino acids were separated in a Pico-Tag amino acids column (Waters, Milford CA, USA) and we determined the mole percentages of glutamine, serine, proline, glycine, and alanine (as these amino acids represent >80% of the total amino acids of dragline silks in most spiders [56]). We used a single-factor multivariate analysis of variance (MANOVA) and Fisher’s Least Significant Difference post-hoc analysis to compare the mean (±standard errors) mole percentages of glutamine, serine, proline, glycine, and alanine between species when collected in May. We additionally performed between species (L. hasselti versus A. keyserlingi) comparisons of the amino acid mole percentages for silks collected in May and November, as well as within species comparisons of the amino acid mole percentages of the silks collected in November. 3.3. Solid-State NMR (ssNMR) Spectroscopy We used 13C cross-polarization magic-angle spin (CPMAS) solid-state NMR analyses to examine and compare the protein secondary structures of 3 of the L. hasselti, A. keyserlingi and N. plumipes dragline silks collected in May to ascertain whether the amino acid compositions predictably represented protein secondary structures. The silks were not enriched with isotopes since metabolic processing of 13C or other isotopes may confound the spectra attained using 13C CPMAS ssNMR. We used 5–10 mg of silk from each spider collected. We performed a pre-assessment using a different set of spiders and found 5–10 mg of unlabeled silk to be adequate to attain reliable NMR spectra. The silk samples were packed in 2.5 mm zirconia MAS rotors with vespel caps. We were careful not to expose the silks to the atmosphere by immediately packing the silks into airtight containers post extraction, and minimizing the exposure to air when packing the sample holders. The ssNMR spectra were acquired using a Bruker Avance III NMR spectrometer (Bruker Pty Ltd., Melbourne, Australia), with a 16.4 Tesla superconducting magnet operating at 175 MHz and 700 MHz frequencies for the detection of 13C and 1H isotopes respectively. The 2.5 mm NMR rotor was spun at 30 kHz MAS in a 2.5 mm triple resonance probehead. The 13C CPMAS ssNMR spectra were acquired with 1H to 13C cross-polarization with a 2 ms contact time, SPINAL-64 decoupling for 1H decoupling at a field strength of 100 kHz, and a recycle delay of 4 s. Then, 16k–64k transients were co-added to reduce the signal-to-noise ratio. The C=O peak for pure glycine was set to 176 ppm as an external reference for the 13C chemical shifts [21]. The width of the alanine Cα signal (49.6 ppm) provided insights for determining the overall crystallinity of the ordered domains. Under the acquisition conditions assigned (i.e., 30 kHz MAS), suppression of the 1H–13C dipolar coupling ensured that only signals from rigid domains could have been detected using our technique. 3.4. Tensile Testing Tensile testing was done on a single thread of dragline silk from each of the 15 spiders collected in May to ascertain whether the amino acid compositions or secondary structures explained the mechanical properties of each of the three species’ silks. To do this, we connected a revolving headframe to the mechanical spool. We attached a 240 mm long × 40 mm wide cardboard (for native silk testing) or plastic (for supercontracted silk testing) strip, which had six 30 mm × 30 mm square holes punched at 10 mm intervals to the headframe. Double sided sticky tape was stuck onto the cardboard at the border of the holes. A single thread was pulled from the spinnerets of an anaesthetized spider and stuck to one of the pieces of sticky tape. The headframe was rotated once ensuring the silk traversed all of the holes and adhered to the tape. The strip was then removed from the headframe and a drop of water based glue applied to the position where the silk was fastened to the cardboard/plastic. Another frame of equal size with identically positioned holes punched into it was placed on top. The two strips were squeezed together with forceps ensuring that they stuck together. We then cut the strip in the regions between the holes perpendicular to the silk thread, leaving six 30 mm × 30 mm frames holding a single thread of silk. The above procedure was repeated for each individual from each of the three species (we accordingly collected 30 frames per species: 6 frames × 5 individuals). We taped one randomly selected frame of silk collected from each spider frame to a microscope slide and examined and photographed it under 1000× magnification using a polarized light microscope (CKX41, Olympus, Tokyo, Japan) connected to a SPOT Idea 5 Mp digital camera (Spot Imaging Solutions, Sterling Heights, MI, USA). The images were digitized using the program Spot Basic 4.7 (Spot Imaging Solutions, Sterling Heights, MI, USA) and the width of each thread determined as a mean of 12 measurements using the program Image J (NIH, Bethesda, MD, USA). These measurements enabled us to calculate the cross-sectional area of each individual thread used in the ensuing tensile tests. We performed the following tensile tests under controlled temperature (~25 °C) and humidity (~50% R.H.) in still air within 10 days of silk collection. We performed native state tensile tests on 15 frame-mounted silk samples (3 frames each from 5 individuals) per species. To do this, we placed the 30 mm × 30 mm frames containing a single fiber within the grips of an Instron 5543 tensile testing machine (Instron Machines, Melbourne, Australia) with a resolution to approximately 2 μN. We ensured that the grips held the silks firmly at the upper and lower frame edges. The left and right sides of the frames were cut away and the silks stretched at a rate of 0.1 mm·s−1 until the fiber ruptured. Stress (σ) and strain (ε) were calculated using equations [57]: (1) σ=FA (2) ε=logeLL0 where F is the force applied to the specimen measured using the program Bluehill 3.0 (Instron Machines, Melbourne, Australia); A is the cross-sectional area of the thread calculated from the thread diameter assuming a constant thread volume; L is the instantaneous length of the fiber at a given extension value measured using Bluehill 3.0; and L0 is the original gauge length of the fiber. Stress versus strain curves were determined for each silk tested by a standard trapezoidal method from which we calculated the following mechanical properties for each specimen: (1) ultimate strength, the stress at rupture; (2) extensibility, the strain at rupture; (3) toughness, the Area under the stress strain curve; and (4) Young’s modulus (stiffness), the slope of the stress–strain curve during its initial elastic phase. We performed supercontracted tensile tests on 8–10 frame-mounted silk samples (two samples each from four to five individuals) per species to calculate the parameters wet tension and percentage shrink. We submersed the fibers within a perspex water bath while the samples were held within the grips of the tensile testing machine without tension applied. We ascertained how much stress was generated by the restrained silks, then the fibers were relaxed while wet and the percentage shrink calculated as the proportional difference between the pre-shrunk (l0) and post-shrunk (l1) fiber lengths [29,51]. Since immersion in water is expected to interrupt protein alignment in the amorphous region [11,54], these parameters can be presumed to indicate approximately the amount of amorphous chain alignment in the native silks [51]. We then dried the fibers in air at maximum relaxation, upon which they were subjected to the same tensile testing procedures described for native silks. We compared the ultimate strength, extensibility, toughness and stiffness in the native and supercontracted states, and the wet tension and percentage shrink in the supercontracted state, between species using Multivariate analysis of variance MANOVA. All statistical analyses were performed after checking the variances for heterogeneity using Levene’s tests. 4. Conclusions We determined herein dragline silk spidroin expression and structural (i.e., β-sheet and 3,10helix alanine conformations and crystallinity) variations between and within three spider species using HPLC and ssNMR respectively. Our HPLC analysis of amino acid composition and ssNMR analysis of protein secondary structures found conflicting results for A. keyserlingi dragline silks collected in May. We tentatively considered this as evidence that silk spinning processes decouple protein secondary structures from spidroin expression. It is nevertheless possible that spidroins other than MaSp1 and MaSp2 also appeared in the dragline silks. The relatively high proline compositions in A. keyserlingi silks from spiders collected in November indicated that it was likely that MaSp2 was predominantly expressed, while the compositions of their silks from those collected in May indicated that it was likely that MaSp1 was predominantly expressed. In May, glycine and alanine within A. keyserlingi dragline silk were distributed more toward 3,10helices/random coil than β-sheets, which contrasted with the conformations within the MaSp1 predominant dragline silks of N. plumipes and L. hasselti. In the latter species’ silks, glycine and alanine were distributed more toward β-sheets. The greater proportion of β-sheet formations within N. plumipes silk is explainable by their high glycine and alanine compositions, with the majority of the alanine forming (A)n sequences and most of the glycine forming (GA)n sequences that conformed into β-sheets. Our tensile tests and HPLC analyses suggested that spidroin expression had little influence on silk mechanical properties. ssNMR, on the other hand, showed that L. hasselti and N. plumipes dragline silks had greater alanine β-sheet conformations and crystallinity than did that of A. keyserlingi. The greater helical conformations in A. keyserlingi dragline silks explained its extensibility. Our finding that alanine compositions did not differ between species while β-sheet conformations differed significantly suggested that “on-the-fly” repositioning of alanine during spinning may be a mechanism for inducing protein secondary structural variations, but this needs verification. We showed here that while amino acid compositions varied over time in different spiders, decoupling of protein secondary structures from spidroin expression could ensure that functionally effective dragline silks are consistently produced. Our findings improve our understanding of the biological processes that induce dragline silk to vary in chemical, structural and mechanical properties across environments. Such understanding is important if spider dragline silk is to be successfully synthesized commercially. Acknowledgments Research was supported by an Australian Research Council (Discovery Early Career Researcher Award DE140101281) grant to Sean J. Blamires and National Science Council, Taiwan, grants (NSC-102-2311-B-029-001-MY3 and NSC-102-2811-B-029-001) to I-Min Tso and Sean J. Blamires. Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1294/s1. Click here for additional data file. Author Contributions Sean J. Blamires, Michael M. Kasumovic and Aditya Rawal conceived and designed the experiments. Sean J. Blamires collected, fed and silked the spiders. I-Min Tso performed HPLC. Sean J. Blamires and Penny J. Martens conducted the mechanical testing and analyses. James M. Hook and Aditya Rawal performed the NMR experiments and analyzed the spectra. All authors were involved in the preparation of the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 13C CPMAS ssNMR of silk from individual: (a) Argiope keyserlingi (Argiope); (b) Nephila plumipes (Nephila); and (c) Latrodectus hasselti (Latrodectus). The spectral peaks shaded green represent poly-alanine β-sheets, while those shaded grey represent poly-alanine 3,10helices/random coils. Accordingly, the spectra show that there are differences in the ratios of alanine in β-sheets: alanine in 3,10helices/random coil between A. keyserlingi (1:7), N. plumipes (7:1) and L. hasselti (2:1) silks. Spectra of individual spiders within each species are shown in the Supplementary Materials (Figure S1). Figure 2 Comparisons of the ultimate strength, extensibility, stiffness and toughness of native silks, and wet tension and percent shrink of supercontracted silks of L. hasselti, A. keyserlingi and N. plumpes dragline silks taken from spiders collected in May. Numbers associated with the respective species’ bars show the β-sheets (green): 3,10helices/random coil (grey) ratios ascertained by ssNMR (see Figure 1). ijms-17-01294-t001_Table 1Table 1 Comparisons of: (a) The glutamine, serine, glycine, alanine and proline compositions (means with standard errors in parentheses) of L. hasselti, A. keyserlingi and N. plumpes dragline silks from spiders collected in May; (b) The glutamine, serine, glycine, alanine and proline compositions (means with standard errors in parentheses) of L. hasselti and A. keyserlingi dragline silks from spiders collected in November; (c) Compositions from dragline silk genetic sequences for Latrodectus hesperus (from reference [47]) and Argiope bruennichi [48]; * Indicates a statistically significant difference in composition was found between for silks collected in November compared to silks collected in May. Amino Acids (Percentage Composition) Glutamine Serine Glycine Alanine Proline (a) May Argiope keyserlingi 7.90 (0.88) 7.02 (0.72) 35.39 (1.52) 26.46 (1.87) 4.17 (0.84) Nephila plumipes 6.39 (0.92) 4.77 (0.33) 41.09 (0.45) 30.23 (0.56) 2.71 (0.42) Latrodectus hasselti 9.61 (0.20) 6.78 (0.62) 37.56 (1.67) 26.0 (1.92) 3.13 (0.65) (b) November Argiope keyserlingi 8.27 (0.84) 7.34 (2.29) 34.61 (6.77) 19.34 (6.76) * 12.53 (1.67) * Latrodectus hasselti 8.99 (0.38) 7.47 (0.46) 32.28 (2.75) 29.55 (2.92) 2.79 (0.65) (c) Sequenced compositions Latrodectus hesperus MaSp1 6.9 – 33.5 31.1 0.4 Latrodectus hesperus MaSp2 11.3 – 42.3 32.7 8.6 Argiope bruennichi MaSp1 4.38 5.67 45.05 30.61 0 Argiope bruennichi MaSp2 14.50 3.04 38.47 22.51 12.48 ==== Refs References 1. Heim M. Keerl D. Scheibel T. Spider silk: From soluble protein to extraordinary fiber Angew. Chem. Int. Ed. 2009 48 3584 3596 10.1002/anie.200803341 19212993 2. Van Krevelen D.W. Properties of Polymers: Their Correlation with Chemical Structure, Their Numerical Estimation and Predictions from Additive Group Contributions Elsevier Amsterdam, The Netherlands 2009 3. Parkhe A.D. Seeley S.K. Gardner A. Thompson L. Lewis R.V. Structural studies of spider silk proteins in the fiber J. Mol. Recog. 1997 10 1 6 10.1002/(SICI)1099-1352(199701/02)10:1<1::AID-JMR338>3.0.CO;2-7 4. Erdogan H. Apaydin M.S. Incorporating amino acid typing into nuclear magnetic resonance protein structure-based assignments Proteom. Bioinform. 2012 5 116 121 10.4172/jpb.1000222 5. Fratzl P. Jacob H.F. Rimmerthaler S. Roschger P. Klaushofer K. Position resolved small angle X-ray scattering of complex biological materials J. Appl. Crystal. 1997 30 765 769 10.1107/S0021889897001775 6. Ene R. Papadopoulos P. Kramer F. Combined structural model of spider dragline silk Soft Matter 2009 5 4568 4574 10.1039/b911159j 7. Rathore O. Sogah D.Y. Nanostructure formation through β-sheet self-assembly in silk-based materials Macromolecules 2001 34 1477 1486 10.1021/ma001553x 8. Bovey F.A. Miran P.A. NMR of Polymers Academic Press San Diego, CA, USA 1996 9. An B. Hinman M. Holland G.P. Yarger J.L. Lewis R.V. Inducing β-Sheets formation in synthetic spider silk fibers by aqueous post-spin stretching Biomacromolecules 2011 12 2375 2381 10.1021/bm200463e 21574576 10. Shi X. Yarger J.L. Holland G.P. Elucidating proline dynamics in spider dragline silk fibre using 2 H–13 C HETCOR MAS NMR Chem. Commun. 2014 50 4856 4859 10.1039/c4cc00971a 24686512 11. Yang Z. Liivak O. Seidel A. LaVerde G. Zax D.B. Jelinski L.W. Supercontraction and backbone dynamics in spider silk: 13 C and 2 H NMR Studies J. Am. Chem. Soc. 2000 122 9019 9025 10.1021/ja0017099 12. Li X. Eles P.T. Michal C.A. Water permeability of spider dragline silk Biomacromolecules 2009 10 1270 1275 10.1021/bm900103n 19331319 13. Creager M.S. Jenkins J.E. Thagard-Yeaman L.A. Brooks A.E. Jones J.A. Lewis R.V. Holland G.P. Yarger J.L. Solid-state NMR comparison of various spiders' dragline silk fiber Biomacromolecules 2010 11 2039 2043 10.1021/bm100399x 20593757 14. Jenkins J.E. Creager M.S. Lewis R.V. Holland G.P. Yarger J.L. Quantitative correlation between the protein primary sequences and secondary structures in spider dragline silks Biomacromolecules 2010 11 192 200 10.1021/bm9010672 20000730 15. Xu D. Yarger J.L. Holland G.P. Exploring the backbone dynamics of native spider silk proteins in black widow silk glands with solution-state NMR spectroscopy Polymer 2014 55 3879 3885 10.1016/j.polymer.2014.06.018 16. Shi X. Holland G.P. Yarger J.L. Molecular dynamics of spider dragline silk fiber by 2H MAS NMR Biomacromolecules 2015 16 852 859 10.1021/bm5017578 25619304 17. Izdebski T. Akhenblit P. Jenkins J.E. Yarger J.L. Holland G.P. Structure and dynamics of aromatic residues in spider silk: 2D carbon correlation NMR of dragline fibers Biomacromolecules 2010 11 168 174 10.1021/bm901039e 19894709 18. Rauscher S. Baud S. Miao M. Keeley F.W. Pomes R. Proline and glycine control protein self-organization into elastomeric or amyloid fibrils Structure 2006 14 1667 1676 10.1016/j.str.2006.09.008 17098192 19. Holland G.P. Yarger J.L. Lewis R.V. WISE NMR characterization of nanoscale heterogeneity and mobility in supercontracted Nephila clavipes spider dragline silk J. Am. Chem. Soc. 2004 126 5867 5872 10.1021/ja031930w 15125679 20. Addison J.B. Ashton N.N. Weber W.S. Stewart R.J. Holland G.P. Yarger J.L. β-Sheet nanocrystalline domains formed from phosphorylated serine-rich motifs in caddisfly larval silk: A solid state NMR and XRD study Biomacromolecues 2013 11 1140 1148 10.1021/bm400019d 23452243 21. Holland G.P. Jenkins J.E. Creager M.S. Lewis R.V. Yarger J.L. Solid-state NMR investigation of major and minor ampullate spider silk in the native and hydrated states Biomacromolecules 2008 9 651 657 10.1021/bm700950u 18171016 22. Xu D. Shi X. Thompson F. Weber W.S. Mou Q. Yarger J.L. Protein secondary structure of Green Lynx spider dragline silk investigated by solid-state NMR and X-ray diffraction Int. J. Biol. Macromol. 2015 81 171 179 10.1016/j.ijbiomac.2015.07.048 26226457 23. Jenkins J.E. Sampath S. Butler E. Kim J. Henning R.W. Holland G.P. Yarger J.L. Characterizing the secondary protein structure of black widow dragline silk using solid-state NMR and X-ray diffraction Biomacromolecules 2013 14 3472 3483 10.1021/bm400791u 24024617 24. Shen Y. Johnson M.A. Martin D.C. Microstructural characterization of Bombyx mori silk fibers Macromolecules 1998 31 8857 8864 10.1021/ma980281j 25. Asakura T. Yang M. Kawase T. Nakazawa Y. 13 C solid-state NMR Study of structural heterogeneity in peptides containing both polyalanine and repeated GGA sequences as a local structural model of Nephila clavipes dragline silk (Spidroin 1) Macromolecules 2005 38 3356 3363 10.1021/ma047660z 26. Craig C.L. Riekel C. Comparative architecture of silks, fibrous proteins and their encoding genes in insects and spiders Comp. Biochem. Physiol. B 2002 133 493 507 10.1016/S1096-4959(02)00095-7 12470814 27. Savage K.N. Gosline J.M. The role of proline in the elastic mechanism of hydrated spider silks J. Exp. Biol. 2008 211 1948 1957 10.1242/jeb.014225 18515725 28. Liu Y. Shao Z.Z. Vollrath F. Elasticity of spider silks Biomacromolecules 2008 9 1782 1786 10.1021/bm7014174 18529075 29. Boutry C. Blackledge T.A. Evolution of supercontraction in spider silk: Structure–function relationship from tarantulas to orb-weavers J. Exp. Biol. 2010 213 3505 3514 10.1242/jeb.046110 20889831 30. Van Beek J.D. Hess S. Vollrath F. Meier B.H. The molecular structure of spider dragline silk: Folding and orientation of the protein backbone Proc. Nat. Acad. Sci. USA 2002 99 10266 10271 10.1073/pnas.152162299 12149440 31. Blamires S.J. Wu C.L. Tso I.M. Variation in protein intake induces variation in spider silk expression PLoS ONE 2012 7 1294 10.1371/journal.pone.0031626 22363691 32. Blamires S.J. Wu C.L. Blackledge T.A. Tso I.M. Environmentally induced post-spin property changes in spider silks: Influences of web type, spidroin composition and ecology Biol. J. Linn. Soc. 2012 106 580 588 10.1111/j.1095-8312.2012.01884.x 33. Blamires S.J. Wu C.L. Blackledge T.A. Tso I.M. Post-secretion processing influences spider silk performance J. R. Soc. Interface 2012 9 2479 2487 10.1098/rsif.2012.0277 22628213 34. Craig C.L. Hsu M. Kaplan D.L. Pierce M.E. A comparison of the composition of silk proteins produced by spiders and insects Int. J. Biol. Macromol. 1999 24 109 118 10.1016/S0141-8130(99)00006-9 10342754 35. Guehrs K.H. Schlott B. Grosse F. Weisshart K. Environmental conditions impinge on dragline silk prtotein composition Insect Mol. Biol. 2008 17 553 564 10.1111/j.1365-2583.2008.00826.x 18828841 36. Craig C.L. Riekel C. Herberstein M.E. Weber R.S. Kaplan D. Pierce N.E. Evidence for diet effects on the composition of silk proteins produced by spiders Mol. Biol. Evol. 2000 17 1904 1913 10.1093/oxfordjournals.molbev.a026292 11110907 37. Tso I.M. Wu H.C. Hwang I.R. Giant wood spider Nephila pilipes alters silk protein in response to prey variation J. Exp. Biol. 2005 208 1053 1061 10.1242/jeb.01437 15767307 38. Blamires S.J. Liao C.P. Chang C.K. Chuang Y.C. Wu C.L. Blackledge T.A. Sheu H.S. Tso I.M. Mechanical performance of spider silk is robust to nutrient-mediated changes in protein composition Biomacromolecules 2015 16 1218 1225 10.1021/acs.biomac.5b00006 25764227 39. Marhabaie M. Leeper T.C. Blackledge T.A. Protein composition correlates with the mechanical properties of spider (Argiope trifasciata ) dragline silk Biomacromolecules 2014 15 20 29 10.1021/bm401110b 24313814 40. Liao C.P. Chi K. Tso I.M. The effects of wind on trap structural and material properties of a sit-and-wait predator Behav. Ecol. 2009 20 1194 1203 10.1093/beheco/arp119 41. Dicko C. Vollrath F. Kennedy J.M. Spider silk protein folding is controlled by changing pH Biomacromolecules 2004 5 704 710 10.1021/bm034307c 15132650 42. Hedhammer M. Rising A. Grip S. Saenz Martinez A. Nordling K. Casals C. Stark M. Johansson J. Structural properties of recombinant nonrepetitive and repetitive parts of major ampullate spidroin 1 from Euprosthenops australis : Implications for fiber formation Biochemistry 2008 47 3407 3417 10.1021/bi702432y 18293938 43. Knight D.P. Knight M.M. Vollrath F. Beta transitions and stress-induced phase separation in the spinning of spider dragline silk Int. J. Biol. Macromol. 2000 27 205 210 10.1016/S0141-8130(00)00124-0 10828366 44. Vollrath F. Knight D.P. Liquid crystalline spinning of spider silk Nature 2001 410 541 548 10.1038/35069000 11279484 45. Lefevre T. Paquet-Mercier F. Rioux-Dube J.F. Pezolet M. Structure of silk by Raman spectromicroscopy: From the spinning glands to the fibers Biopolymers 2011 97 322 335 10.1002/bip.21712 21882171 46. Giesa T. Perry C.C. Buehler M.J. Secondary structure and critical stress for a model of spider silk assembly Biomacromolecules 2016 17 427 436 10.1021/acs.biomac.5b01246 26669270 47. Ayoub N.A. Garb J.E. Tinghitella R.M. Collin M.A. Hayashi C.Y. Blueprint for a high-performance biomaterial: Full-length spider dragline silk genes PLoS ONE 2007 2 1294 10.1371/journal.pone.0000514 17565367 48. Zhang Y. Zhao A.C. Sima Y.H. Lu C. Xiang X.H. Nakagaki M. The molecular structures of major ampullate silk proteins of the wasp spider, Argiope bruennichi : A second blueprint for synthesizing de novo silk Comp. Biochem. Physiol. B 2013 164 151 158 10.1016/j.cbpb.2012.12.002 23262065 49. Work R.W. Young T.C. The amino acid compositions of major ampullate and minor ampullate silks of certain orb-web building spiders (Araneae, Araneidae) J. Arachnol. 1987 15 65 80 50. Chaw R.C. Correa-Garhwal S.M. Clarke T.H. Ayoub N.A. Hayashi C.Y. Proteomic evidence for components of spider silk synthesis from black widow silk glands and fibers J. Proteome Res. 2015 14 4223 4231 10.1021/acs.jproteome.5b00353 26302244 51. Elices M. Plaza G.R. Perz-Riguero J. Guinea G.V. The hidden link between supercontraction and mechanical behaviour of spider silks J. Mech. Behav. Biomed. Mater. 2011 4 658 669 10.1016/j.jmbbm.2010.09.008 21565714 52. Gosline J.M. Denny M.W. De Monte M.E. Spider silk as rubber Nature 1984 309 551 552 10.1038/309551a0 53. Vollrath F. Strength and structure of spider’s silk Rev. Mol. Biotechnol. 2000 74 67 83 10.1016/S1389-0352(00)00006-4 54. Perz-Riguero J. Elices M. Guinea G.V. Controlled supercontraction tailors the tensile behaviour of spider silk Polymer 2003 44 3733 3736 10.1016/S0032-3861(03)00245-3 55. Grubb D.T. Jelinski L.W. Fiber morphology of spider silk: The effects of tensile deformation Macromolecules 1997 30 2860 2867 10.1021/ma961293c 56. Lewis R.V. Bionanotechnology: Proteins to Nanodevices Renugopalakrishnan V. Lewis R.V. Springer Dordrecht, The Netherlands 2006 61 78 57. Meyers M.A. Chawla K.K. Mechanical Behavior of Materials Cambridge University Press Cambridge, UK 2009

PMC005xxxxxx/PMC5000692.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081295ijms-17-01295ReviewTowards Effective Photothermal/Photodynamic Treatment Using Plasmonic Gold Nanoparticles Bucharskaya Alla 1*Maslyakova Galina 1Terentyuk Georgy 12Yakunin Alexander 3Avetisyan Yuri 3Bibikova Olga 2456Tuchina Elena 7Khlebtsov Boris 89Khlebtsov Nikolai 89Tuchin Valery 2310Sivakov Vladimir Academic Editor1 Research Institute for Fundamental and Clinical Uronephrology, Saratov State Medical University, n.a. V.I. Razumovsky, 410012 Saratov, Russia; [email protected] (G.M.); [email protected] (G.T.)2 Research-Education Institute of Optics and Biophotonics, Saratov National Research State University, 410012 Saratov, Russia; [email protected] (O.B.); [email protected] (V.T.)3 Institute of Precision Mechanics and Control, RAS, 410028 Saratov, Russia; [email protected] (A.Y.); [email protected] (Y.A.)4 Artphotonics GmbH, 12489 Berlin, Germany5 Optoelectronics and Measurement Techniques Laboratory, University of Oulu, 90014 Oulu, Finland6 Institute of Analytical and Bioanalytical Chemistry, Ulm University, 89081 Ulm, Germany7 Department of Biology, Saratov National Research State University, 410012 Saratov, Russia; [email protected] Institute of Biochemistry and Physiology of Plants and Microorganisms, RAS, 410049 Saratov, Russia; [email protected] (B.K.); [email protected] (N.K.)9 Department of Nano- and Biomedical Technologies, Saratov National Research State University, 410012 Saratov, Russia10 Interdisciplinary Laboratory of Biophotonics, National Research Tomsk State University, 634050 Tomsk, Russia* Correspondence: [email protected]; Tel.: +7-845-266-974809 8 2016 8 2016 17 8 129513 6 2016 29 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Gold nanoparticles (AuNPs) of different size and shape are widely used as photosensitizers for cancer diagnostics and plasmonic photothermal (PPT)/photodynamic (PDT) therapy, as nanocarriers for drug delivery and laser-mediated pathogen killing, even the underlying mechanisms of treatment effects remain poorly understood. There is a need in analyzing and improving the ways to increase accumulation of AuNP in tumors and other crucial steps in interaction of AuNPs with laser light and tissues. In this review, we summarize our recent theoretical, experimental, and pre-clinical results on light activated interaction of AuNPs with tissues and cells. Specifically, we discuss a combined PPT/PDT treatment of tumors and killing of pathogen bacteria with gold-based nanocomposites and atomic clusters, cell optoporation, and theoretical simulations of nanoparticle-mediated laser heating of tissues and cells. Au nanoparticlesnanorodsnanostarssilicananocompositesphotothermal therapy (PPT)photodynamic therapy (PDT)pathogenstumorscell optoporation ==== Body 1. Introduction Currently, gold nanoparticles (AuNPs) of different size and shape are often used as photosensitizers for cancer plasmonic photothermal (PPT) therapy [1,2], for cell optoporation [3,4] and pathogen killing [5]. It is mandatory to analyze the underlying mechanisms to increase the selectivity of AuNP accumulation in tumors, to enhance treatment efficiency and photo-cytotoxicity, as well as to study other critical effects of laser radiation and AuNPs interaction with tissues and cells that are of great importance for laser treatments. Progress in the application of plasmonic AuNPs for laser hyperthermia of tumor tissues is due to the multiplicative effects of increased local absorption of laser radiation by plasmonic AuNPs [6] and their targeted delivery [7]. The combined impact of these factors has revolutionized the traditional and widespread laser hyperthermia of tissues. To take advantage of these mechanisms, it is necessary to develop multi-scale models and methods for calculation of the temperature fields on macro- and micro-scales. Furthermore, the models should be adopted for practical application to large size objects at the second- and minute-time scale, whereas the nano-, pico- and femtosecond laser pulses are applied to individual particles. Recently, significant advances were achieved in the development of mathematical modeling methods of photothermal effects of nanoparticles (NPs) in tissues and cells [8,9]. It was shown that local inhomogeneities of the temperature fields lead to new laws of the distribution and dynamics of the Arrhenius damage integral [10,11], which provide a more appropriate selection of nanoparticle dimension and light exposure to gain precise control of hyperthermia with minimal energy consumption. Laser induced cell optoporation by local heating of AuNPs was proven to be a promising approach to deliver exogenous molecules into cells. Theoretical estimations were used to build a strategy to increase laser cell membrane optoporation efficiency that strongly depends on the laser beam intensity and AuNP optical and thermal properties [12,13]. To achieve the optimal AuNP–cell interaction, sphere-, rod- and star-shaped AuNPs with the different plasmon-resonant peaks were fabricated and functionalized with different ligands [14,15]. One of the potential applications of AuNPs is to kill pathogens, as well as to regulate the number of opportunistic microorganisms. Many differently structured AuNPs have been studied to potentiate antimicrobial PDT by improving photosensitizer solubility, photochemistry, photophysics and targeting [16,17]. Potential ways to improve method are to increase local concentration of the photosensitizer via targeted delivery of nanoparticles, to provide selective interaction with the cell wall of bacteria, and the resonance NP heating under laser radiation. Currently, the practice of cancer plasmonic photothermal (PTT) and photodynamic (PDT) therapies has advanced in leaps and bounds due to the wide application of multifunctional plasmonic nanocomposites (NCs) and fluorescent photodynamic molecules. Unfortunately, a porphyrin-based PDT can be practical only for tumors on or under the skin or mucosa of the oral or internal organs, as it absorbs light wavelengths of less than 640 nm [18]. The usage of gold NCs for effective laser heating and photodynamic impact on tumor tissue was demonstrated recently [19,20]. Nevertheless, further studies are needed to improve the therapeutic protocols by adjusting light delivery, its power density, and irradiation doses. In this review, we summarize our recent theoretical, experimental, and preclinical results on light activated interaction of AuNPs with tissues and cells. Specifically, we discuss a combined PPT/PDT treatment of tumors and killing of pathogenic bacteria with gold-based nanocomposites and atomic clusters, cell optoporation, and theoretical simulations of nanoparticle-mediated laser heating of tissues and cells. 2. Nanocomposites Based on Au Nanoparticles and Nanoclusters The PTT is a promising strategy to destroy selectively tumor tissue [21]. This form of therapy is achieved by the conversion of light to thermal energy to heat up tissues by using light-absorbing materials. In order to maximize the photothermal response of the materials to laser irradiation, it is essential to maximize their absorption in the near-infrared (NIR) region [22]. Gold nanostructures are good candidates for photothermal cancer treatment, due to their high biocompatibility, ease of surface functionality and tunable localized surface plasmon resonance (LSPR) absorption band. Various geometries of gold nanoparticles have been previously synthesized, including nanorods, nanoshells and nanocages, in order to use them for PTT in the NIR region [23,24,25], where tissue light scattering and absorption are minimal. However, this therapy may still be challenged for the tumors located deep in the tissues. Many papers have been focused on destruction of deep-located tumors using optical fibers. Recently, Bhatia et al. suggest implanting an 808 nm-laser source for effective intratissue PTT using Au nanorods (AuNRs) as a photothermal agent [26]. Despite recent progress in this field and successful experiments in vitro and in vivo using tumor-breeding mice models [27], it has now become obvious that using PTT alone is an ineffective way to totally eliminate large, solid tumors and additional treatment is usually needed to stop tumor progression. For such a purpose, Hauck et al. [28] used AuNRs combined with chemotherapeutic drug cisplatin. In this study, the cell culture was incubated with AuNRs and cisplatin was added separately before laser irradiation. The cancer cells’ viability after this treatment was 78% less than that after chemotherapy alone and 84% less than after PTT alone. This fact confirms the synergistic effect of combined PPT and chemotherapy. In the past few years, hybrid Au nanoparticle systems have attracted significant interest, as they combine the outstanding optical properties of nanoparticles with dyes or drugs [29]. By use of smart bioconjugation techniques [30], Au nanoparticles can be functionalized with a set of different molecules, enabling them to perform targeting, diagnostic, and therapeutic functions in a single treatment procedure. This class of multifunctionalized nanoparticles has found exciting applications in proof-of-concept theranostic experiments [31]. For example, Tingting Wang et al. [32] have demonstrated DOX conjugated PaaPEG-AuNRs for combined cancer PTT and drug delivery. Jianliang Shen et al. [33] suggested novel dual-modality nanoparticles with the capability of gene silencing through the incorporation of siRNA and AuNRs mediated PTT. In this paper, the effective suppression of mRNA and corresponding protein expression was due to intracellular delivery of a siRNA against PKM2. The combined activity of PKM2 inhibition and plasmonic heating dramatically reduced the viability of breast cancer cells. In paper [34], the Al(III) phthalocyanine chloride tetrasulfonic acid functionalized AuNRs were applied for fluorescent imaging and PTT in vivo for mouse model. Three-modality nanocomplex based on AuNRs and protoporphyrin IX was recently applied for in vivo SERS detection, fluorescence imaging, and PDT [35]. Shouju Wang et al. reported on single NIR laser induced PDT/PTT therapy using chlorin e6 functionalized gold nanostars [36]. The problem of dye quenching near the metal surface together with low adsorption capacity of colloidal particles are strong limitation factors towards synthesis of effective complexes of metallic nanoparticles with PDT dyes. Recently, we suggested [37,38] composite nanoparticles consist of AuNRs that are coated with a mesoporous silica shell functionalized with a photosensitizer. This approach enables one to overcome the limitations concerning dye quenching and the low loading capacity of the nanocomposites. By using a NIR absorbing Au core and a mesoporous silica shell doped by hematoporphirin (HP) molecules, one enables the preparation of multifunctional nanoparticles for combined PTT/PDT therapy and enhanced photodynamic efficiency. The three basic steps of nanocomposite (NC) synthesis together with TEM images of resultant particles are presented in Figure 1a. Step 1 is the fabrication of Au nanorods with plasmon resonance in NIR spectral region. In Step 2, the AuNRs were coated with the primary SiO2 shell, which serves as a spacer between the Au nanorods and dye molecules. Finally, an additional layer of a mesoporous silica with HP molecules embedded inside was formed. As a result, we obtain AuNR/SiO2-HP composite particles consisting of the plasmonic AuNR core, the first HP-free silica layer, and the second HP-loaded mesoporous silica layer. Figure 1b shows photographs of cuvettes with silica coated nanorods, nanocomposites and HP solution under visible and UV illumination. The NR and NC samples have a red-brown color due to characteristic transversal mode in extinction spectra of AuNRs located at 510 nm while the HP solution looks water-white. The extinction spectra of the nanoparticle solution (Figure 1c) had an intense longitudinal resonance in the region 810–820 nm. In addition, as distinct from the AuNRs spectra, the NC spectrum has a characteristic peak at 400 nm related to absorption of HP molecules inside a nanocopmosite. Taking into account molecular absorption coefficient, we can assume the concentration of included HP molecules to be 7 mg/mL or 3.8 × 103 HP molecules per particle. The visual inspection of the cuvettes with samples (Figure 1b) under UV light showed red fluorescence of nanocomposites and the HP solution, whereas the silica coated Au nanorod suspension looks colorless. To indicate the difference in the fluorescence for HP molecules in solution and in the NC, the emission spectra of diluted solutions were measured. Figure 1c shows spectra taken during excitation with 405 nm-light. We observed the difference in both spectral shapes and intensities for the HP solution and the NC. For example, fluorescence spectra for HP solution has two peaks at 615 and 675 nm while the NC fluorescence spectrum exhibits three peaks located at 630, 650 and 690 nm. On the other hand, the intensity integral over all wavelengths is approximately the same for both cases. The photo-oxidation of the 9,10-anthracenediyl-bis (methylene) dimalonic acid (ABDA) during the illumination of samples with 625 nm light was measured to detect the singlet oxygen generation. The results of measurements for singlet oxygen generation are shown in Figure 1d. It is clearly seen that the characteristic absorbance peaks of ABDA gradually decrease with an increase of irradiation time. Since the silica shell serves as a protection layer between the metal surface and dye molecules, we did not observe a significant difference between the photodynamic activities of free HP and nanocomposite solutions. To compare photothermal conversation mediated by nanocomposites and plasmonic core (AuNRs), we measured the in-depth temperature distributions and the time dependent temperature changes in nanoparticle solutions under NIR light irradiation. During plasmonic heating of nanocomposites, the maximal temperature was about of 72 °C (Figure 1d). In the control test of tubes with saline, the temperature only reached 25 °C during 300 s of irradiation with a laser power density of 2 W/cm2. On a whole, we can conclude that our synthetic procedures allow us to obtain nanostructures possessing three important optical modalities—fluorescence under UV irradiation, generation of singlet oxygen under 625 nm irradiation and NIR mediated photothermal conversion. Au nanoclusters are a new type of luminescent nanomaterials, usually comprising Au nanoparticles smaller than 2 nm, and are typically composed of a few to about 100 gold atoms. The AuNCs are distinguished from the other nanomaterials by their strong photoluminescence, large Stokes shift and high emission rates [22]. The luminescent gold nanoclusters provide the bridge between atomic and nanoparticles behavior in noble metals and exhibit molecule-like photophysical properties, large surface-to-volume ratios, easy fictionalization and color tunability [39]. In contrast to many publications on multifunctional composites based on plasmonic nanoparticles [40], analogous nanocluster-based multifunctional theranostic nanocomposites have been developed in a few reports only. Perhaps, the first report on theranostic application of Au nanoclusters was published by Haiyan Chen et al. [41]. In this paper, Au-BSA-FA nanoclusters were loaded by a chemotherapeutic drug doxorubicin and a near infrared fluorescent dye MPA. The Au-BSA-FA-MPA and Au-BSA-FA-DOX were successfully applied for in vitro and in vivo tumor diagnostics and therapy with cancer cells and a xenografted mice model. Ding and Tian [42] applied Au-BSA-FITC-FA nanoclusters to specific bioimaging and biosensing of cancer cells, where Au nanoclusters produced a reference fluorescent signal, FITC allowed for pH monitoring, and FA acted as targeting molecules. In the Ref. [43], unique nucleus-targeting gold nanoclusters were made and applied for in vitro and in vivo fluorescence imaging, RNA delivery, and PDT of cancer cells. The main advantage of the TAT peptide-Au nanoclusters is their high accumulation rate into the cytoplasm region and a significant accumulation into the nucleus. Finally, Cui and co-workers [44] developed multifunctional NCs comprising GSH-cupped Au nanoclusters that further were coupled with FA and PEG followed by embedding photosensitizer chlorin e6 into PEG shell. The obtained Au-GSH-FA-PEG-Ce6 complexes were successfully applied to simultaneous in vitro and in vivo imaging and photodynamic therapy of cancer cells and tumors in mice. 3. Multiscale Mathematical Modeling of Temperature Field of Tissues and Cells Doped by Plasmonic Nanoparticles Recently, there has been significant progress in the development of theoretical analysis methods for examining the photothermal effects of nanoparticles (NPs) on tissues and cells [45,46,47,48,49,50,51,52,53,54,55,56]. The general scheme of the theoretical modeling [47,52,53,54,55] is shown in Figure 2. The thermal response on the laser irradiation can be found from the solution of the heat equation [47,53,55] (1) cρ∂T∂t=div [k⋅grad(T)]+U Here, c, ρ and k are the values of specific heat, mass density and thermal conductivity, respectively; T is the temperature; t is the time. The solution of Equation (1) must satisfy the condition of continuity of temperature and normal component of heat flux at the boundaries of the contacting media. A two-scale approach for calculation of kinetics of photo-induced temperature fields has been recently proposed [55,56]: (i) Macroscale model is valid for mean temperature fields analysis in the spatially extended regions of tissues doped by assembles of plasmonic nanoparticles. In this case, the values T, c, ρ, k in Equation (1) should be considered as corresponding variables averaged over physically small volumes containing, at the same time, a sufficiently large number of nanoparticles. (ii) Microscale model is valid for calculation of the small-scale spatial inhomogeneity of the temperature field within a nanoparticle itself and its vicinity. It means the exact local values of the variables T and other variables should be considered in Equation (1). This is important, e.g., for the study of cell membrane optoporation or transfection. The example of simulations for Model (i) is presented in Figure 3. Figure 4a represents [52] the distribution of the dimensionless value—efficiency of absorption Qabs=∫VUdv/(πR2I) versus radius R of Au spherical nanoparticle of the volume V = 4πR3/3 irradiated by laser light of wavelength λ and intensity I. The corresponding distribution of the steady state temperature increment ΔT within the Model (ii) for I = 100 kW/cm2 is depicted in Figure 4b. The short laser pulse model (ii) was developed as well. Under this approach, a novel thermal effect, which is a hoop-shaped hot zone formation on the surface of irradiated nanoshell, was found [53] (see Figure 5). The time of “life” of the hot zone is less than a nanosecond for the considered nanoshell size. 4. Arrhenius Damage Integral The irreversible thermal damage of a particular type of a tissue or a cell is described by the condition for Arrhenius damage function or integral [46,47]: (2) Ω(r,τ)=A∫0τexp(−EaRgT(r,t))dt≥1 Here, τ is the exposure time, Rg = 8.314 J/(mol·K) is the gas constant. For calculations, parameter A = 3.1 × 1098 s−1 and the activation energy Ea = 6.3 × 105 J/mol that characteristic for damage of porcine skin were used. Laser-irradiated AuNP creates a localized heated region in the surrounding medium. From Equation (2), the maximal allowed exposure time τ exceeding which leads to local destruction (damage) of cell membrane or tissue component can be determined [47]. A study of impact of AuNP size and laser pulse duration on the temperature jump and damage function is of undoubted practical interest. Figure 6 shows temporal dependences of the temperature on the surface of a 50 nm-AuNP at different pulse duration τp which varies from 0.0001 to 1 µs [52]. The time t in Figuare 6a is normalized to pulse duration τp. This allows for a graphical presentation in a single scale of relative time, time/pulse duration, and differently scaled temporal processes. The reduction of τp leads to a monotonic, however disproportional, decrease of maximal temperature jump ΔTmax at a similar laser intensity. It is evident that the Arrhenius function Ω for various τp differs significantly due to two major factors, the maximal temperature rise ΔTmax and duration of the elevated temperature of the biological object, which both contribute simultaneously. It should be noted that the analysis would be more objective if the comparison were to be carried out for a similar ΔTmax. A natural way to increase ΔTmax with a decrease of τp is the use of lasers with higher power densities. The required increase of the laser power to create an equivalent temperature effect can be estimated from data presented in Figure 6b. It is equally important to estimate the laser pulse energy E, which is required for the equivalent thermal exposure. Figure 6c, built in the coordinates “Normalized energy E/E0—Pulse duration τp”, shows that larger AuNPs have the advantage of better energy efficiency compared to small-sized nanoparticles. The results of calculations of ΔT and Ω for AuNPs (R = 50 nm, τp = 1 µs) and (R = 10 nm, τp = 0.1 µs) are shown in Figure 7 [52]. These AuNPs have a similar maximal pulse temperature received for the identical energy of the laser pulse. However, the increment of the Arrhenius function Ω, provided by AuNP with R = 50 nm and a longer pulse is almost an order of magnitude higher. Results of this study show that the choice of NP size and laser pulse duration could provide a precise control of a local tissue/cell hyperthermia. At the same time, for both, the criteria for optimization are energy transform efficiency and economic feasibility. We see that the similar effects are achievable when using expensive femtosecond or relatively cheap nanosecond laser systems. 5. PPT/PDT Pathogen Killing Using AuNPs The potential applications of AuNPs to kill pathogens, as well as to regulate the number of opportunistic microorganisms, are of great interest [57,58,59,60]. The use of AuNPs as light-activated agents, also in combination with other photosensitizers (PS), such as nanocomposites (NC) (see Section 2), increases the effectiveness of antimicrobial PDT. The AuNPs-mediated PTT/PDT action of lasers working in the red and/or infrared spectral ranges is known to possess pronounced antimicrobial properties [61]. The main focus is on the suppression of growth of such clinically significant bacteria as Staphylococcus aureus (S. aureus) and Pseudomonas aeruginosa [16,62], which are characterized normally by a high resistivity to many present-day antibiotics. For targeted delivery of AuNPs, containing antibiotics or photosensitizers, such as chlorine e6 or toluidine blue, immunoglobulin molecules were successfully used [63,64]. In the study [65], use of gold(III) complex on nanoporous materials such as mobile composite material number 41 (MCM-41) at irradiation by 532 nm-laser light led to selective reduction of the number of fungi Saccharomyces cerevisiae. The AuNP-PS conjugates and 540 nm-irradiation allowed one to inactivate fungal biofilm of genus Candida [66,67]. A significant reduction of colonies Candida albicans and Escherichia coli (E. coli) was shown after their photoinactivation using AuNP complexes (nanorods and bipyramids) with aluminum phthalocyanine [66]. The AuNPs in combination with a polyurethane, phthalocyanine and methylene blue resulted in a significant decrease in the number of S. aureus [62,67,68]. At using a combination of gold nanorods with indocyanine green (ICG) photosensitizer (PS) and pulsed laser light (810 nm), a reduction in size of the E. coli bacteria was recently achieved [69]. In our studies [37,70,71,72,73,74], we used various modifications of AuNPs (nanorods, nanoshells, nanocages, nanoclusters) and their conjugates with a PS (ICG, HP, Photosens™ (PhS)) for the selective destruction of microorganisms by light exposure (see, Table 1). Gold nanoparticles are unique objects for the targeted delivery of a variety of biologically active substances. Therefore, in the first phase of our research, the gold nanorods that conjugated with ICG were used for light-activated destruction of microbial cells [70]. We investigated the combination of NIR PTT/PDT against different bacterial strains (S. aureus and S. epidermidis, both methicillin susceptible and resistant). PTT was provided mostly at NIR light absorption by AuNPs, as PDT, by ICG molecules. The absorption spectrum of ICG bound to cell structures overlaps well with the plasmonic resonance of the gold nanorods (~800 nm). Therefore, diode laser 808 nm-radiation with a power density of 50 mW/cm2 inhibited growth of S. aureus by 65% and S. epidermidis by 80% via combined PPT/PDT treatment with a very low overheating of the cell suspension, which did not exceed 6 °C. Other types of gold nanoparticles conjugated with ICG molecules were also used; for instance, silica/gold nanoshells covered by additional mesoporous silica shell (about 10–15 nm) that was functionalized with ICG molecules [70,71]. The IGG molecules were absorbed on the particle surface due to the electrostatic interaction with positively charged amine groups. Additionally, the use of gold-silver nanocages with an inner cavity for containment of ICG molecules was described in [70]. Laser radiation of 805 nm (46 mW/cm2) for nanoshells and 808 nm (60 mW/cm2) for nanocages demonstrated a similar inhibitory effect for S. aureus. It should be noted that the concentrations of ICG used in some studies in combination with NIR laser radiation (810 nm, 48 mW/cm2, 30 min) were relatively high (25–250 mg/mL) [75]. Under these conditions, the suppression of bacterium growth by 90%–99% was observed for S aureus, P. aeruginosa, and Streptococcus pyogenes. In our experiments, the dye concentration was 1–2 orders of magnitude lower (2.5 mg/mL). Even at such low concentration of ICG, we observed a pronounce suppression (by 75%) of the meticilline-sensitive S. aureus 209 P. In search of new effective and low-toxic conjugates of nanoparticles and PDT dyes, we have chosen further studies of gold nanorods and gold-silver nanocages in complex with hematoporphyrin (see Figure 1) [37]. The effective suppression of S. aureus by dual PPT and PDT action was studied and compared with the activity of the control solutions (1—hematoporphyrin, 2—plasmonic nanoparticles coated by silica with concentrations similar to use in conjugates). To irradiate nanocomposites, we used a red LED (625 nm, 33 mW/cm2) and a NIR laser (808 nm, 100 mW/cm2), respectively. At irradiation with 625-nm light, the nanoconjugates have an enhanced PDT action toward S. aureus bacteria (80%–97% cell death), as compared with a molecular solution of HP of the similar concentration. An additional photoinactivation of bacteria can be provided by PPT action under irradiation of nanoconjugates with NIR light at a wavelength close to the plasmon resonance. Incubation of the bacteria with nanoconjugates and their irradiation at 808 nm also lead to decrease of bacterial survivability (65%–90% cell death). Because the mean overall heating of the cell suspensions was insignificant in this case, a possible mechanism responsible for the injury to the bacteria can be the local heating of nanoconjugates at the cell wall because of light absorption at the wavelength of the plasmon resonance of the composite’s core. The improvement of efficiency and selectivity of the photothermal effect of laser (808 nm) radiation against both antibiotic-sensitive and antibiotic-resistant strains of S. aureus is possible by using novel modifications of the “antibody–nanoparticle” system. In our studies, we offered a system of gold nanorods that carry on their surface inverted Fc-fragments of human immunoglobulin A and G [71,73]. The Fc-fragments of IgG in conjunction with nanoparticles have demonstrated a greater effectiveness than IgA (Figure 8). The number of MSSA cells in suspension with gold nanorods decreased by 78% (CFU reduction) after 15 min of NIR radiation exposure, and by 95% after 30 min exposure. For the MRSA strain, the number of cells incubated with IgG-conjugates of gold nanorods decreased by 96% after 15 min of light exposure and by 97% after 30 min. The increase of mean temperature in the bacterial suspension depended on the light exposure, the maximal values being recorded after 30 min of NIR irradiation. In the experiments on the impact of NIR radiation in combination with gold nanorods, the number of microorganisms of two studied strains decreased proportionally to the increase in the mean temperature. This means that the major contribution to the damage of bacterial cells can be related to the local photothermal plasmon resonance heating rather than to the total heating of suspension. Under laser irradiation, the increase of mean temperature of suspensions of 12.7 °C for nanoparticles conjugated with IgA and of 15.2 °C for nanoparticles conjugated with IgG was found, as compared to heating of a pure NP suspension. This phenomenon can be associated with the better stability of the colloidal system due to formation of clusters of nanoparticles and bacterial cell wall via immunoglobulin Fc-fragments, which may lead not only to better efficiency of the local photothermal effects, but also to elevation of suspension mean temperature. Functionalization of nanoparticles by antibodies allows one to use a wide range of auxiliary components and various synthesis conditions. Recently, we described Au-BSA nanoclusters [74] functionalized with targeting molecules (human anti-staphylococcal immunoglobulin, IgG) and PDT dye Photosens™ (PS) [76] for selective detection and effective PDT inactivation of both methicillin sensitive and resistant S. aureus. The synthesis of Au-BSA-IgG-PS complexes included three consecutive steps shown schematically in Figure 9a. Initially, Au-BSA nanoclusters were prepared from a mixture of BSA and Au3+ at high pH and boiling temperature [77]. The main characteristics of received Au-BSA nanoclusters are given in Figure 9b—extinction (1), excitation (2) and emission spectra (3). In general, the extinction spectrum is similar to the spectrum of small AuNPs (<3 nm), where the plasmonic peak around 510–520 nm is also missing. The large or aggregated nanoparticles were absent as demonstrated in TEM images of nanoclusters (Figure 9c), and the average size of Au-BSA nanoclusters is about 2 nm. In aqueous solution, the as-prepared Au-BSA are highly dispersed and they exhibit strong red fluorescence (FL) under UV illumination and brown color under white light, as shown in the insert of Figure 9c. The curve 2 in Figure 9d demonstrates two peaks around 405 and 514 nm in fluorescence excitation spectrum of Au-BSA nanoclusters. The emission maxima are found near 470 and 660 nm. As defined by using hematoporphyrin as a benchmark standard, the quantum yield of Au-BSA nanoclusters was about 14% at optimum conditions. The application of BSA as a coating agent is preferred because it contains various functional groups which could be used to bind with different ligands. Furthermore, the high sorption capacity of BSA for PS dyes, drugs and various therapeutic agents enables the use of BSA-capped nanoclusters as a promising nanoplatform for theranostic purposes. One of the possible applications has been demonstrated in Step 2, in which Au-BSA nanoclusters were functionalized with human anti-staphylococcal IgG, which are also known as antibodies against S. aureus endotoxin. In the last stage, Au-BSA and Au-BSA-IgG complexes were conjugated with photosensitizer (PS). Due to the high affinity of PS to BSA and other serum proteins, PS molecules were successfully incorporated into the BSA matrix and multifunctional Au-BSA-IgG-PS complexes were formed. The photo-oxidation of ABDA in the presence of AuNCls and Au-BSA-IgG-PS complexes was measured under LED laser irradiation (660 nm) for 60 min. As shown in Figure 9e, the characteristic absorbance maxima of ABDA, in particular the main peak at 402 nm, gradually decreased with an increase in light exposure. A weak photooxidation ability of Au-BSA nanoclusters itself was also revealed (data not shown here). The synthesized Au-BSA-IgG-PS complexes show enhanced PDT activity that is comparable or even a bit higher than that for free PS solution. Thus, such Au–BSA–anti-SAIgG complexes, due to their biospecific targeting and intense red fluorescence, may identify pathogenic microorganisms in bacterial suspensions by FL microscopy or even by the naked eye in the investigation of sediments under UV illumination. The proposed nanocomposites may be applied at a physiological pH of 7, as opposed to nonspecific electrostatic binding of Au—human serum albumin AuNCls to S. aureus at pH < 5–6. It was shown that exposure to red (660 nm) radiation and Au–BSA–anti-SAIgG complexes leads to clearly marked destruction of two studied Staphylococcus strains (MSSA and MRSA). If the red (660 nm) light in the absence of photoactive agents ensured destruction of microorganisms at a rate of 75% and 63%, respectively, then the adding of total Au–BSA–anti-SAIgG complexes to the suspension resulted in the death of up to 90% of bacterial cells. Protein- or glutathione-coated AuNCls appear to be a more convenient platform for nanomedicines than commonly used large AuNPs. Really, the presence of stable fluorescence in AuNCls makes them an indispensable tool for nanodiagnostics, without the need to functionalize them with FL dyes as well. These NCs were used as sensitive markers for the detection of pathogenic bacteria; for example, selective sensitivity was found against MSSA and MRSA in [74], where we established pH-dependent selective binding of Au—HSA NCs to MSSA and MRSA. Thus, the gold nanoparticles in various modifications significantly increase the effectiveness of PDT treatment of bacterial infections. Various mechanisms are apparently involved in killing of bacteria, including the local increase in the concentration of the photosensitizer through targeted delivery of nanoparticles, selective interaction with the cell wall of bacteria, and the resonance heating of AuNPs under laser light irradiation. 6. Photothermal and Photodynamic Therapy for Transplanted Tumors To demonstrate the utility of fabricated Au nanorods and Au nanocomposites for in vivo applications, we investigated the PTT and PDT treatments for big solid tumors in rats. Recently, several research groups reported the use of various gold nanoparticles: nanoshells, nanorods, nanocages, and other nanocubes for the plasmon resonance hyperthermia [23,78,79,80,81,82]. Use of gold nanorods (AuNRs) for PTT is preferred due to the colloidal stability and easy customization of nanorods plasmon resonance in accordance with the laser wavelength by changing the axial ratio of nanoparticles [83]. To improve the biocompatibility of the nanoparticles and to enhance their stability, different biocompatible polymers are applied [84]. A longer circulation time and better accumulation in tumors show nanoparticles coated with neutrally charged polymers, including polyethylene glycol [85]. We have previously tested a method of photothermal plasmon-resonance therapy in tumor-bearing rats with alveolar liver cancer PC-1 rats with the intratumoral introduction at an amount of 30% of the tumor volume was intratumorally injected. The length of AuNRs was 41 nm ± 8, diameter was 10 ± 2 nm, concentration was 400 µg/mL, and a maximum absorption was noted at a wavelength of 808 nm corresponding to the plasmon resonance of the gold nanorods. One hour after, laser irradiation was carried out percutaneously over the surface of the tumor for 15 min. The 808-nm CW dide laser LS-2-N-808-10000 (Laser Systems, Ltd., St. Petersburg, Russia) with a power density 2.3 W/cm2 was used for the laser hyperthermia. Temperature control of tumor heating was performed every 30 s using infrared thermograph IRI4010 (IRYSYS, Northampton, UK). During the laser irradiation a significant rise in temperature (55 ± 2 °C) was noted, most pronounced in the first 2 min of irradiation. Twenty-four hours after laser hyperthermia, animals were removed from the experiment, and the marked changes were revealed at morphological study of tumors. The necrotic zones occupy 80%–90% of the slice area in tumors. Nevertheless, survived tumor cells with degenerative changes were detected only in the subcapsular area of the tumors. The aim of the next study was to investigate the combined PDT and PTT treatment of tumor-bearing rats [38]. We were interested in the impact of large tumors, therefore, white outbred male rats with implanted cholangiocarcinoma PC-1 were taken in experiments on reaching their tumor volume of about 3 cm3. Five animal groups were formed randomly (6 rats per group): the control group received only saline injection (group I), comparison group received saline injection and was treated with laser 808 nm irradiation (group II), group III received NC injection and was treated with laser 808 nm irradiation (PTT), group IV received NC injection and was treated with laser 633 nm irradiation (PDT), group V received NC injection and was treated with synchronous irradiation of laser 808 nm and laser 633 nm (PDT + PTT). All injections were made intratumorally. The power density of 808-nm CW diode laser LS-2-N-808-10000 (Laser Systems, Ltd.) was 2.3 W/cm2 (groups II, III and V), the power density of 633-nm CW laser (GN-5P, “Plasma” Corp., Ryazan, Russia) was 160 mW/cm2 (group IV and V). Finally, animals of group V were simultaneously irradiated with both lasers. Each irradiation treatment continued for 20 min. The surface temperature profile over the tumor was captured using infrared camera IR Imager IRI4010, Infrared Integrated System (IRISYS). The tumor biopsies were sampled three days after the laser exposure, hematoxylin and eosin (H & E)-staining was used for morphological examination. In comparison group II, only NIR irradiation caused a slight increase in the surface tumor temperature from 30 °C to about 40 °C, thereby, necrotic changes were not observed in tumor tissues. Nevertheless, at 808-nm laser irradiation in rats with NC-inection, the ablative values of tumor temperature rapidly exceeded 60 °C and were maintained at about 75 °C thereafter. It is known that rapid coagulative necrosis and irreversible cell and tissue damage were observed at temperatures above 70 °C. The simultaneous treatment of tumors with NIR and He–Ne lasers also results in an increase of surface tumor temperature, which was slightly higher compared to PTT group. Therefore, PDT contribution to the total thermal effects was insignificant in group V. Furthermore, the temperature of the NC-treated tumors did not increase at only 633-nm irradiation, indicating that no thermal effects were induced under PDT. Morphological investigations of treated tumor tissues revealed the negligible effect of only NIR laser irradiation on cancer cells in the comparison group (Figure 10a). In group III (NCs injected + 633 nm laser irradiation), an increased number of brown spots were noted in tumor tissue, indicating some apoptotic damage of cells after PDT treatment (Figure 10b). In PTT-treated group (NCs injection + NIR laser irradiation), large areas of necrosis appeared in tumor tissue after NIR-irradiation (Figure 10c). Finally, in group V, marked necrotic changes were revealed in tumor tissue and significant tissue loss was observed after combined PDT + PTT treatments (Figure 10d). Simulated and experimental studies demonstrated that photothermal heating effects could be quite complicated and depend on the nanoparticle design and irradiation conditions [21]. The nanoparticle characteristics such as size, plasmonic resonance value, the type of the surface coating, and the particle concentration affect increase of tumor temperature and thus determine photothermal treatment efficiency. In addition, the heating also depends on irradiation parameters including laser wavelength, laser power and treatment time. To achieve appropriate temperature increments in xenografted tumors in vivo, the laser power density in the range of 1–5 W/cm2 is used and the treatment exposure varies between 1 and 10 min [80,82,83]. A few previously performed in vivo studies of combined PDT/PTT therapy with using different types of GNPs focused on the treatment of small tumors (typically, less than 0.5 cm3) [80,82,83]. The effective therapy of larger tumors poses new challenges related to the route of nanoparticle administration and their tumor accumulation, deep penetration of laser radiation inside the tumor, and optimization of NC and irradiation doses [86]. The NIR laser power density of 2.3 W/cm2 applied during 15–20 min has been shown to be effective for PTT damage of large tumors at intratumoral injection of gold nanorods [38,87]. Recently, an interesting solution was proposed by Boseung Jang et al. [19] for PTT/PDT treatments to increase the damage to SCC7 squamous cell carcinoma tumors in Balb/c-nu mice in vivo. The nanocomposites based on 34 nm AuGNRs (aspect ratio (AR) 3.7, λmax ~800 nm) and functionalized by RRLAC peptide were further conjugated with a photosensitizer AlPcS4 (absorbance at 675 nm) via electrostatic immobilization. The photothermal heating of tumors caused the release of bound photosensitizers from nanocomposites. PTT (810 nm, 3.82 W/cm2) followed up by PDT treatment (670 nm, 331 mW/cm2) which caused an increase of tumor temperature up to 65 °C. The lower laser power density is usually applied for PDT, because the main effects of photodynamic treatment associated with singlet oxygen generation but not with tissue hyperthermia. In our work [38], we have showed effective combined PDT/PTT treatment for large solid tumors, whereas only PDT treatment was ineffective for antitumor therapy. Nevertheless, possible tumor recurrence, probably caused by the limited light penetration and none-optimal spatial distribution of the NIR laser radiation within the tumor, remains an unsolved problem. Thus, further studies are needed to improve the therapeutic protocols by correct selection of nanoparticle administration techniques and irradiation modes for deep light penetration and adequate damage to tumor tissue. 7. AuNP Mediated Optoporation Recently, a new AuNPs mediated technique for permeabilizing cells was introduced [88]. In this method, AuNPs were placed on cells and treated by a weakly focused laser beam, which leads to a significant increase of membrane porosity in the vicinity of AuNPs. Spectral localization of LSPR enormously enhances laser absorption leading to photothermal and related phenomena such as heating of the surrounding medium/tissue, microbubble formation and acoustic shockwave generation [89,90]. Notably, that for treatment efficiency and locality, the laser treatment of cells should be optimized. From this point of view, it is crucial to distinguish the difference between impact on the cell/tissue caused by an enhanced localized thermal laser action mediated by AuNPs and non-targeted laser irradiation of the surroundings [91,92]. The mechanism of gold nanoparticle mediated (GNOME) optoporation and transfection is based on laser perforation caused by short laser pulses of sufficiently high intensity, which leads to formation of vapor nanobubbles (NBs) [93,94,95]. One of the first applications of laser-induced transient vapor NBs around overheated gold nanoparticles (called also photothermal or plasmonic NBs) in biology and medicine with a focus on integration of cancer and pathogen detection and treatment was performed by Zharov’s group in 2003 [90]. NBs emerge due to a rapid increase of the AuNPs’ temperature to several hundred degrees and evaporation of water surrounding AuNPs [96,97]. Collapse of NBs causes local damage to cell membranes. For successful delivery of medical drugs, DNA, or RNA molecules into the cytoplasm, cell membranes can be perforated, without any permanent membrane damage [98]. Due to the extremely short lifetime of NBs, the diffusion of heat from the AuNPs into the surrounding medium is negligible. The applicability of GNOME laser optoporation and transfection for cell manipulation was intensively studied during the last decade [88]. Typically, to mark perforated cell membranes under laser treatment, one adds fluorescent dyes, which are unable to introduce untreated cell membrane, in cell suspension [99]. Pitsillides et al. in 2003 were among the first to demonstrate the increased permeability of cell membrane in the presence of AuNSps under 10-kDa fluorescein–dextran conjugate in the presence of AuNSps under 20-ns, 532-nm laser pulses with an energy density of 0.5 J/cm2 [100]. The biological consequences of GNOME were investigated in detail by Heinemann’s research group, with analysis of the potential changes to the cell membrane [88]; study of cell volume and area and ion exchange [101]; calculation of the kinetics of fluorescent dyes perforation [102]; and evaluation of perforation process of fluorescent dextrans in a size range of 10–2000 kDa under irradiation of a 532 nm picosecond laser as a function of irradiation time and repetition steps [103]. For all research works demonstrated below, AuNSps were attached to the cell membrane due to a sedimentation process. To achieve selective targeting of cell membranes by nanoparticles, functionalization by specific antibodies is needed. Addressing this, Cuiping Yao et al. functionalized 15 nm and 30 nm AuNSps by antibodies to the plasma membrane of the Hodgkin’s disease cell line and/or the human large-cell anaplastic lymphoma cell line [104,105]. Strong absorption ability allowed AuNSps with LSPR at 532 nm to perforate the cell membrane even when the irradiation wavelength of the nanopulsed laser is off resonance at NIR (1064 nm), which maximizes the penetration depth and opens up the possibility to reach sublayer cells in vivo [106,107]. Notably, the AuNPs’ morphology and colocalization with cells (either target the cellular membrane or are endocytosed) should also be taken into account. AuNRs and hollow AuNSps induce less cell damage than AuNSs under picosecond irradiation [108]. Anisotropic AuNP polarization should be taken into account as well. Recently, the importance of AuNRs’ orientation to the incident electric field was acknowledged, where local defects in the phospholipid cell membrane were only induced when nanorods were placed in parallel to the polarization of the electric field of the laser beam, a normal orientation of AuNRs to the polarization field [109]. The energy density of the laser pulse that is necessary for effective cell perforation strongly depends on the molecular weight of the delivered material [4]. However, in spite of the differences between laser sources used by different research groups, energy densities providing cell optoporation are in the range of 100–200 mJ/cm2 with the threshold value of 15–30 mJ/cm2 [99,100,101,102,103,104,105,106]. Recently, we demonstrated different optoporation abilities in a comparative study of AuNPs with variable morphology for evaluation of their impact on cell membrane permeability under irradiation by three laser sources operating at different modes and wavelengths [110,111]. These lasers were used to irradiate the HeLa cells: CW-laser (Aculaser, Inc., Las Vegas, NV, USA), with a central wavelength 808 nm, power 2.5 W, and beam diameter 5 mm; a ns-laser (Opotek Tunable Laser Systems, OPOTEK, Inc., Carlsbad, CA, USA) generated laser pulses at 532 nm with a pulse duration of 5 ns, repetition rate 20 Hz, maximal pulse energy 0.1 mJ, mean power 2 mW, and beam diameter 5 mm; a nanosecond ytterbium fiber laser (scan-ns-laser) (Mini Marker 2TM, Laser Center, St. Petersburg, Russia) in 3D scanning mode, with the wavelength 1064 nm, pulse duration of 4 ns, repetition rate up to 20 kHz, pulse energy up to 1 mJ, mean power 20 W and sharply focused beam with diameter 6 µm. For 3-D scanning process, a laser beam scanned the sample surface area of 3 × 3 mm2 in the horizontal XY plane with the step of 20 μm. Then, the focal point was stepped down along Z-axis with the step of 25 μm to provide a three-dimensional irradiation matrix. The total number of laser pulses on one focal plane was 4 × 104 with the scanning time of 2 s, and the scanning speed of 0.4 m/s, i.e., one pulse per irradiation point. The scanned depth along Z-axis was 2 mm. Single pulse energy was taken as 0.1 µJ. For the experiments, we used three types of AuNPs: nanostars (AuNSts) with LSPR 805 nm [112], AuNSps with LSPR 520 nm [113], AuNRs with PR 805 nm [114]. The biocompatibility of synthesized AuNPs was discussed elsewhere [115,116], the chosen concentration of AuNPs 17 μg/mL was sufficient for cell perforation and was harmless to living cells according to previous reports [107]. Investigation of optoporation kinetics was performed of the fluorescent dye propidium iodide (PI), which becomes fluorescently detectable as a result of binding to nucleic acids after membrane permeabilization [117]. The exposure time to PI on the cells was essentially shorter in comparison with the classical staining protocol [118], to find only perforated cells as PI-positive. A similar method was applied in Meunier’s research group [107], as well as Kalies et al.’s [102] who studied the efficacy of Lucifer Yellow dye uptake depending on laser exposure to evaluate membrane permeabilization. MTT test was performed under irradiation conditions similar to the PI test to determine cell viability. To evaluate the heating properties of the AuNPs surrounding media, temperature measurements were carried out every 15 s during the laser irradiation. After irradiation by CW laser, samples of cells incubated with AuNRs demonstrated the highest percentage of PI-positive cells under laser treatment, thereat MTT assay showed almost total cell death as a result of strong heating of the surrounding medium in the presence of AuNPs (increase up to 72 °C after 2-min irradiation) leading to cell damage and death [110]. In contrast, we have not observed any temperature increase of cell medium in the presence of AuNPs under ns-laser irradiation. Figure 11 demonstrates the fluorescent images of cells irradiated by ns-laser generated laser pulses at 532 nm with a pulse duration of 5 ns, repetition rate of 20 Hz, maximal pulse energy of 0.1 mJ, mean power of 2 mW without AuNSps (a) and with AuNSps (b); and combination of bright-field and fluorescent images of cell samples irradiated by the scan-ns-laser with a wavelength 1064 nm, pulse duration of 4 ns, repetition rate up to 20 kHz, mean power of 20 W, and single pulse energy of 0.1 µJ: without AuNSts (c); with AuNSts (d). In the upper row, fluorescent dye Calcein AM stained only undamaged cells. For both types of nanosecond lasers, the border of the collimated laser beam is clearly seen for cell suspensions incubated with AuNPs: inside the irradiated area, only perforated cells are present. Effectiveness of AuNSts mediated optoporation with nanosecond laser in 3D scanning mode resulted from the broad plasmonic resonance of AuNSts, where, at a laser irradiation wavelength of 1064 nm, absorption is still high for successful optoporation, and a relatively high contribution of absorption to extinct parts of the AuNsts spectrum can be made. Figure 12 shows typical spectra of AuNSts’ extinction (μext), absorption (μabs), and scattering (μsct), coefficients, the peaks of which can vary from 650 to 1000 nm dependent on nanoparticle size. Recently the mutual dependence between the diameter of NSts and absorption-scattering cross-section ratios (Cabs/Csca) was shown experimentally. For the relatively small NSts (50 nm diameter), the contribution of scattering to absorption was much smaller, than for large NSts (diameter more than 100 nm) [119], which corresponds to the theoretical investigation undertaken by Yan et al. for nanostars and Prashant Jain et al. for AuNPs with variable morphology [120]. Therefore, to maximize the potential influence of laser irradiation on the AuNPs, it is necessary to fabricate AuNPs with high absorbance and LSPR corresponding to the wavelength of the laser source. Effective cell permeabilization with a precise control of cell treatment allows applying AuNPs’ optical transfection: an effective and accurate method with many potential applications both in vivo and in the clinic. The ability to deliver DNA/RNA/siRNA into mammalian cells using AuNSps and AuNRs was demonstrated by Ripken’s [121,122], Chia-Chun Chen et al.’s [123] and Dholakia’s research groups [124]. Even more precise treatment can be achieved by single-cell transfection techniques [125,126], which enable the individual monitoring and performing of genetic changes in a specific cell, without treatment of other cells. 8. Conclusions The therapeutic efficacy of only PDT in cancer treatment is limited because of the inadequate selectivity of most photosensitizers and their poor solubility and rapid destruction under light exposure. A combination of photodynamic therapy (PDT) and photothermal therapy (PTT) may become a universal and gentle tool for biomedical applications in the near future, which allows one to avoid side effects and the development of resistance to drugs. Development of AuNP-based nanocomposites combined with PDT agents can improve the effectiveness of both PDT and combined therapies. Currently, integrative approaches are required for precise control of laser thermal action of PTT with nanoscale/femtosecond resolution, which combined existing cutting-edge and newly developed methods for mathematical modeling of the optical and thermal properties of nanoparticle-containing medium. A general conclusion can be made about the significance of size-dependent distribution of AuNPs to provide optimal local laser hyperthermia of cells and tissues. The use in numerical simulations of the Arrhenius damage integral allowed us to make a more reasonable choice of the nanoparticle size and irradiation mode to precisely control hyperthermia with minimal energy consumption. The proposed approaches can be used for prognosis and monitoring of local hyperthermia caused by application of nanophotosensitizers of different shapes/structures and laser irradiation. Furthermore, the considered theoretical and experimental reports can be useful for a better understanding of new applications of AuNPs to pathogen killing or PTT/PDT treatment of tumors. Further studies are needed to develop a robust and effective strategy for delivery of AuNPs and related composites to tumor tissue, thus ensuring an effective dual PTT/PDT effect. We have evaluated the cell membrane optoporation percentage in in vitro experiments in terms of fluorescent dye permeability under CW and nanosecond (ns) pulse laser treatments. Differently-shaped AuNPs as nanospheres, nanorods and nanostars with various plasmon-resonant peaks were fabricated and functionalized with different ligands to achieve the optimal AuNP–cell interaction. More than 85% of cells can be permeabilized within the illuminated area. Nanostars demonstrated the highest optoporation efficacy under pulse laser irradiation at 1064 nm. The optoporation technology based on AuNPs with high absorbance seems to be an effective method for cell permeabilization with a precise control of cell treatment. This makes this novel technique a prospective tool for highly efficient transfection of extracellular substances into cells. Acknowledgments This study was supported by grants No. 14-13-01167 (designing of nanoparticles, multifunctional gold-based nanocomposites, and fluorescent atomic nanoclusters) and No. 14-15-00186 (nanoparticle mediated laser cell/tissue interactions) from the Russian Scientific Foundation. BNK was supported by RFBR grant 15-33-20248. Authors are grateful for Alexey Bashkatov, Elina Genina, Prateek Singh, Alexey Popov, Ilya Skovorodkin, Veli-Pekka Ronkainen, Seppo Vainio, Igor Meglinski for the discussions and collaborative work described in this paper. Author Contributions Alla Bucharskaya performed experiments on PPT/PDT for transplanted tumors and wrote Section 1, Section 6, and Section 8; Galina Maslyakova and Georgy Terentyuk analyzed the data on PPT/PDT for transplanted tumors; Alexander Yakunin performed modeling of optical, thermal and damage processes, and wrote Section 1, Section 3, and Section 4; Yuri Avetisyan performed modeling of optical, thermal and damage processes, and wrote Section 3 and Section 4; Olga Bibikova performed experiments on laser induced cell optoporation by local heating of AuNPs in vitro and wrote Section 7; Elena Tuchina performed experiments on PPT/PDT pathogen killing using AuNPs and wrote Section 5; Boris Khlebtsov and Nikolai Khlebtsov designed gold nanoparticles, multifunctional nanocomposites and fluorescent atomic nanoclusters (Section 2 and Section 5), and wrote Section 1 and Section 2; Valery Tuchin conceived and designed the experiments on nanoparticle mediated laser cell/tissue interactions and analyzed the data on modeling of temperature fields and damaging of tissues and cells doped by plasmonic nanoparticles, and wrote Section 1, Section 3, Section 4, Section 5, Section 7 and Section 8. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Schematic representation of the basic steps in fabrication of nanocomposites AuNR/SiO2-HP containing a plasmonic core, a primary silica shell, and a secondary mesoporous silica shell doped with HP molecules. Scale bars are 50 nm (a); Photos of cuvettes with silica-coated AuNRs (NR), nanocomposites (NCs), and an HP solution taken under visible (VIS) and UV light illuminations (b); Extinction spectra of HP (curve 1), NC (curve 2), and NR (curve 4) solutions; the spectrum 3 (dotted curve 3) was obtained by the superposition of spectra 1 and 4. The difference ∆A400 is roughly equal to the extinction maximum of HP with a concentration of 7 mg/L (curve 1) (c); Fluorescence spectra of HP (1) and NC (2) solutions measured under 405-nm excitation (d); For these measurements, all solutions were diluted 1:16. Absorption spectra of a mixture of ABDA and NCs under illumination by a 633-nm laser for different exposures from 0 to 60 min (e); Time-dependent increase in the temperature of NC suspension (1) and saline (2) under NIR laser (808 nm, 2 W/cm2) irradiation (f). Reproduced from [37,38] with permission from the Springer and Wiley. Figure 2 The general scheme of mathematical modeling of nanoparticle mediated laser photothermal treatment of cells or tissues. Here U = ε″ω|E|2/8π is the local value of intensity of heat sources caused by the absorption of laser radiation(see, e.g., [55]); ε″ is the imaginary part of dielectric constant of NP, ω and E are the angular frequency and the local value of electric field amplitude diffracted into NP, respectively. Figure 3 Three-dimensional finite-element grid in macro-domain (a) and distribution of temperature increment (°C) in the longitudinal sectional view of the macro-domain passing through the beam axis at the time moment of 300 s, corresponding to the end of light exposure (b). These figures are from [55] with permission of the Publisher. Figure 4 Distribution of the efficiency of absorption of the AuNP (a) [52] and of the temperature increment ΔT (b) on the surface of the AuNP versus wavelength λ and nanoparticle radius R; (b) is from [47] with permission of the Publisher. Figure 5 The temperature T distributions in the spherical nanoshell (with SiO2 core of a radius R = 70 nm and a gold coating of thickness 15 nm) in the water environment irradiated by a series of five pulses of a rectangular shape. The duration of each pulse is 50 ps, repetition rate is 10 ns, the intensity of radiation at a wavelength 800 nm is 4.5 MW/cm2: radial dependences of T at the times of turn-off of j-th irradiating pulse, j = 1/5 (а); two-dimensional distribution of T at the end of the 5th irradiating pulse (due to symmetry only 1/4 of the total allocation is presented) (b). The angle θ is measured from the axis z, directed along the polarization vector of the incident field. These Figures are from [53] with permission of the Publisher. Figure 6 Temporal temperature evolution for 50 nm-AuNPs under the action of pulse laser radiation (λ = 525 nm) (a); The normalized power density (b) and pulse energy (c) of laser beam necessary for achieving the same temperature on the surface of AuNP, depending on pulse duration and nanoparticle radius R [52]. Figure 7 The temperature increments vs. time (a) and the Arrhenius integral Ω increments vs. time (b). Both curves correspond to the same laser pulse energy [52]. Figure 8 Action of NIR laser light (808 nm) and gold nanorods conjugates with immunoglobulin A and G on S. aureus survival rate: methicillin sensitive strain (MSSA) (a); methicillin-resistant strain (MRSA) (b). Red columns—laser light, green columns—laser light and AuNR3 with IgA, violet columns—laser light and AuNR3 with IgG. Figure 9 Scheme of the preparation of the Au-BSA-IgG-PS complexes and their fluorescent and PDT properties under 405, 515 nm and 660 nm-excitation, respectively (a); extinction (1), excitation (2), and emission (3) spectra of Au-BSA NCs (b); TEM images of Au-BSA NCs, the insets show the photos of the solutions under white light (left) and UV light (right) irradiation, The scale bar is 20 nm (c); absorption spectra of a mixture of ABDA and Au-BSA-IgG-Phs under illumination by a 633-nm laser for different exposures from 0 to 60 min (d); the absolute changes in ABDA absorbance at 402 nm and as a function of irradiation time; the gray, blue, and red columns correspond to incubation of bacteria in Au-BSA NCs, Photosens solution, and Au-BSA-IgG-PhS NCs, respectively (e). Reproduced from [74] with permission from The Royal Society of Chemistry. Figure 10 Liver tumor slices after different treatments: Comparison group with 808 nm laser irradiation only (a); PDT-group (b); PTT group (c); PDT + PTT group (d). H & E staining, ×246.4. Reproduced from [38] with permission from the Springer and Wiley. Figure 11 Upper row: the fluorescent images of cells stained by Calcein AM (green color) and PI (red color) irradiated by ns-laser generated laser pulses at 532 nm with a pulse duration of 5 ns, repetition rate 20 Hz, maximal pulse energy 0.1 mJ, mean power 2 mW: samples of pure cells (a); cells incubated with AuNSps under NS-laser treatment (b); Lower row: combination of bright-field and fluorescent images of cell samples irradiated by the scan-NS-laser with a wavelength of 1064 nm, pulse duration of 4 ns, repetition rate up to 20 kHz, mean power 20 W, and single pulse energy of 0.1 µJ: without AuNSts (c); with AuNSts (d). PI-perforated cells stained red color. Figure 12 Typical spectra of AuNSts extinction (μext), absorption (μabs), and scattering (μsct), coefficients and representative TEM image of single AuNSt (inner part). ijms-17-01295-t001_Table 1Table 1 Nanoparticles, PSs and experimental conditions for treatment of pathogens. Abbr. Nanoparticle Shape Photosensitizer (PS) Functional Component Average Size, nm Type of Radiation Maximal Inhibition of S. aureus 209 P after 30 Min-Light Exposure; CFU, % (Reference) AuNRd1 Nanorods ICG – 30 × 10 808 nm, 50 mW/cm2 65 [70] AuNS Nanoshells ICG – 140 805 nm, 46 mW/cm2 55 [71] AuNCg Nanocages ICG – 53 808 nm, 60 mW/cm2 64 [71] AuNR2 Nanorods HP – 50 × 10 808 nm, 100 mW/cm2 90 [37] AuNCg2 Nanocages HP – 50 625 nm, 100 mW/cm2 97 [37] AuNR3 Nanorods – FcIgA, FcIgG 45 × 13 808 nm, 100 mW/cm2 95 [72,73] AuNCl Nanoclusters PhS BSA + IgG 1.8 (25 Au atoms) 660 nm, 50 mW/cm2 90 [74] ==== Refs References 1. Huang X. El-Sayed M.A. Plasmonic photo-thermal therapy (PPTT) Alex. J. Med. 2011 47 1 9 10.1016/j.ajme.2011.01.001 2. Mackey M.A. Ali M.R.K. Austin L.A. Near R.D. El-Sayed M.A. The most effective gold nanorod size for plasmonic photothermal therapy: Theory and in vitro experiments J. Phys. Chem. B 2014 118 1319 1326 10.1021/jp409298f 24433049 3. Bergeron E. Boutopoulos C. Martel R. Torres A. Rodriguez C. Niskanen J. Lebrun J.-J. Winnik F.M. Sapieha P. Meunier M. Cell-specific optoporation with near-infrared ultrafast laser and functionalized gold nanoparticles Nanoscale 2015 7 17836 17847 10.1039/C5NR05650K 26459958 4. Krawinkel J. Torres-Mapa M.L. Werelius K. Heisterkamp A. Rüttermann S. Romanos G.E. Gerhardt-Szép S. Gold nanoparticle-mediated delivery of molecules into primary human gingival fibroblasts using ns-laser pulses: A pilot study Materials 2016 9 397 10.3390/ma9050397 5. Pissuwan D. Cortie C.H. Valenzuela S.M. Cortie M.B. Functionalised gold nanoparticles for controlling pathogenic bacteria Trends Biotechnol. 2009 28 207 213 10.1016/j.tibtech.2009.12.004 20071044 6. Tzarouchis D.C. Ylä-Oijala P. Ala-Nissila T. Sihvola A. Shape effects on surface plasmons in spherical, cubic, and rod-shaped silver nanoparticles Appl. Phys. A 2016 122 298 10.1007/s00339-016-9851-y 7. Dreaden E.C. Austin L.A. Mackey M.A. El-Sayed M.A. Size matters: Gold nanoparticles in targeted cancer drug delivery Ther. Deliv. 2012 3 457 478 10.4155/tde.12.21 22834077 8. Nabil M. Decuzzi P. Zunino P. Modelling mass and heat transfer in nano-based cancer hyperthermia R. Soc. Open Sci. 2015 2 150447 10.1098/rsos.150447 26587251 9. Peeters S. Kitz M. Preisser S. Wetterwald A. Rothen-Rutishauser B. Thalmann G.N. Brandenberger C. Bailey A. Frenz M. Mechanisms of nanoparticle-mediated photomechanical cell damage Biomed. Opt. Express 2012 3 435 446 10.1364/BOE.3.000435 22435092 10. Li D. Wang G.X. He Y.L. Wu W.J. Chen B. A three-temperature model of selective photothermolysis for laser treatment of port wine stain containing large malformed blood vessels Appl. Therm. Eng. 2014 65 308 321 10.1016/j.applthermaleng.2013.12.042 11. Clark C.D. Denton M.L. Thomas R.J. Mathematical model that describes the transition from thermal to photochemical damage in retinal pigment epithelial cell culture J. Biomed. Opt. 2011 16 020504 10.1117/1.3544504 21361660 12. Schomaker M. Heinemann D. Kalies S. Willenbrock S. Wagner S. Nolte I. Ripken T. Murua Escobar H. Meyer H. Heisterkamp A. Characterization of nanoparticle mediated laser transfection by femtosecond laser pulses for applications in molecular medicine J. Nanobiotechnol. 2015 13 10 10.1186/s12951-014-0057-1 25645721 13. Schomaker M. Fehlauer H. Bintig W. Ngezahayo A. Nolte I. Murua Escobar H. Lubatschowski H. Heisterkamp A. Fs-laser cell perforation using gold nanoparticles of different shapes Proc. SPIE 2010 10.1117/12.842446 14. Khlebtsov B.N. Panfilova E.V. Khanadeev V.A. Bibikova O.A. Terentyuk G.S. Ivanov A. Rumyantseva V. Shilov I. Ryabova A. Loshchenov V. Nanocomposites containing silica-coated gold-silver nanocagesand Yb-2,4-dimethoxyhematoporphyrin: Multifunctional capability of IR-luminescence detection, photosensitization, and photothermolysis ACS Nano 2011 5 7077 7089 10.1021/nn2017974 21838309 15. Khlebtsov N. Bogatyrev V. Dykman L. Khlebtsov B. Staroverov S. Shirokov A. Matora L. Khanadeev V. Pylaev T. Tsyganova N. Analytical and theranostic applications of gold nanoparticles and multifunctional nanocomposites Theranostics 2013 3 167 180 10.7150/thno.5716 23471188 16. Huang W.-C. Tsai P.-J. Chen Y.-C. Functional gold nanoparticles as photothermal agents for selective-killing of pathogenic bacteria Nanomedicine 2007 2 777 787 10.2217/17435889.2.6.777 18095845 17. Yin R. Agrawal T. Khan U. Gupta G.K. Rai V. Huang Y.-Y. Hamblin M.R. Antimicrobial photodynamic inactivation in nanomedicine: Small light strides against bad bugs Nanomedicine 2015 10 2379 2404 10.2217/nnm.15.67 26305189 18. Yoon I. Li J.Z. Shim Y.K. Advance in photosensitizers and light delivery for photodynamic therapy Clin. Endosc. 2013 46 7 23 10.5946/ce.2013.46.1.7 23423543 19. Jang B. Park J.Y. Tung C.H. Kim I.H. Choi Y. Gold nanorod-photosensitizer complex for near-infrared fluorescence imaging and photodynamic/photothermal therapy in vivo ACS Nano 2011 5 1086 1094 10.1021/nn102722z 21244012 20. Lin J. Wang S. Huang P. Wang Z. Chen S. Niu G. Li W. He J. Cui D. Lu G. Photosensitizer-loaded gold vesicles with strong plasmonic coupling effect for imaging-guided photothermal/photodynamic therapy ACS Nano 2013 7 5320 5329 10.1021/nn4011686 23721576 21. Abadeer N.S. Murphy C.J. Recent progress in cancer thermal therapy using gold nanoparticles J. Phys. Chem. C 2016 120 4691 4716 10.1021/acs.jpcc.5b11232 22. Weissleder R.A. Clearer vision for in vivo imaging Nat. Biotechnol. 2001 19 316 317 10.1038/86684 11283581 23. Huang X. El-Sayed I.H. Qian W. El-Sayed M.A. Cancer cell imaging and photothermal therapy in the near-infrared region by using gold nanorods J. Am. Chem. Soc. 2006 128 2115 2120 10.1021/ja057254a 16464114 24. Hirsch L.R. Stafford R.J. Bankson J.A. Sershen S.R. Rivera B. Price R.E. Hazle J.D. Halas N.J. West J.L. Nanoshell-mediated near-infrared thermal therapy of tumors under magnetic resonance guidance Proc. Natl. Acad. Sci. USA 2003 100 13549 13554 10.1073/pnas.2232479100 14597719 25. Skrabalak S.E. Chen J. Au L. Lu X. Li X. Xia Y. Gold nanocages for biomedical applications Adv. Mater. 2007 19 3177 3184 10.1002/adma.200701972 18648528 26. Bagley A.F. Hill S. Rogers G.S. Bhatia S.N. Plasmonic photothermal heating of intraperitoneal tumors through the use of an implanted near-infrared source ACS Nano 2013 7 8089 8097 10.1021/nn4033757 23961973 27. Shanmugam V. Selvakumar S. Yeh C.-S. Near-infrared light-responsive nanomaterials in cancer therapeutics Chem. Soc. Rev. 2014 43 6254 6287 10.1039/C4CS00011K 24811160 28. Hauck T.S. Jennings T.L. Yatsenko T. Kumaradas J.C. Chan W.C.W. Enhancing the toxicity of cancer chemotherapeutics with gold nanorod hyperthermia Adv. Mater. 2008 20 3832 3838 10.1002/adma.200800921 29. Sanchez C. Belleville P. Popall M. Nicole L. Applications of advanced hybrid organic-inorganic nanomaterials: From laboratory to market Chem. Soc. Rev. 2011 40 696 753 10.1039/c0cs00136h 21229132 30. Jiao P.F. Zhou H.Y. Chen L.X. Yan B. Cancer-targeting multifunctionalized gold nanoparticles in imaging and therapy Curr. Med. Chem. 2011 18 2086 2102 10.2174/092986711795656199 21517767 31. Liang R. Wei M. Evans D.G. Duan X. Inorganic nanomaterials for bioimaging, targeted drug delivery and therapeutics Chem. Commun. 2014 50 14071 14081 10.1039/C4CC03118K 24955443 32. Wang T. Zhang X. Pan Y. Miao X. Su Z. Wang C. Li X. Fabrication of doxorubicin functionalized gold nanorod probes for combined cancer imaging and drug delivery Dalton Trans. 2011 40 9789 9794 10.1039/c1dt10565e 21879078 33. Shen J. Kim H.-C. Mu C. Gentile E. Mai J. Wolfram J. Ji L. Ferrari M. Mao Z. Shen H. Multifunctional gold nanorods for siRNA gene silencing and photothermal therapy Adv. Healthc. Mater. 2014 10 1629 1637 10.1002/adhm.201400103 24692076 34. Wang J. Tang H.Y. Yang W.L. Chen J.Y. Aluminum phthalocyanine and gold nanorod conjugates: The combination of photodynamic therapy and photothermal therapy to kill cancer cells J. Porphyr. Phthalocyanines 2012 16 802 808 10.1142/S108842461250099X 35. Zhang Y. Qian J. Wang D. Wang Y. He S. Multifunctional gold nanorods with ultrahigh stability and tunability for in vivo fluorescence imaging, SERS detection, and photodynamic therapy Angew. Chem. Int. Ed. 2013 52 1148 1151 10.1002/anie.201207909 23233455 36. Wang S. Huang P. Nie L. Xing R. Liu D. Wang Z. Lin J. Chen S. Niu G. Lu G. Single continuous wave laser induced photodynamic/plasmonic photothermal therapy using photosensitizer-functionalized gold nanostars Adv. Mater. 2013 25 3055 3061 10.1002/adma.201204623 23404693 37. Khlebtsov B.N. Tuchina E.S. Khanadeev V.A. Panfilova E.V. Petrov P.O. Tuchin V.V. Khlebtsov N.G. Enhanced photoinactivation of Staphylococcus aureus with nanocomposites containing plasmonic particles and hematoporphyrin J. Biophotonics 2013 6 338 351 10.1002/jbio.201200079 22736550 38. Terentyuk G. Panfilova E. Khanadeev V. Chumakov D. Genina E. Bashkatov A. Tuchin V. Bucharskaya A. Maslyakova G. Khlebtsov N. Gold nanorods with hematoporphyrin-loaded silica shell for dual-modality photodynamic and photothermal treatment of tumors in vivo Nano Res. 2014 7 325 337 10.1007/s12274-013-0398-3 39. Jin R. Atomically precise metal nanoclusters: Stable sizes and optical properties Nanoscale 2015 7 1549 1565 10.1039/C4NR05794E 25532730 40. Cabral R.M. Baptista P.V. Anti-cancer precision theranostics: A focus on multifunctional gold nanoparticles Expert Rev. Mol. Diagn. 2014 14 1041 1052 10.1586/14737159.2014.965683 25314939 41. Chen H. Li S. Li B. Ren X. Li S. Mahounga D.M. Cui S. Gu Y. Achilefu S. Folate-modified gold nanoclusters as near-infrared fluorescent probes for tumor imaging and therapy Nanoscale 2012 4 6050 6064 10.1039/c2nr31616a 22930451 42. Ding C. Tian Y. Gold nanocluster-based fluorescence biosensor for targeted imaging in cancer cells and ratiometric determination of intracellular pH Biosens. Bioelectron. 2015 65 183 190 10.1016/j.bios.2014.10.034 25461156 43. Vankayala R. Kuo C.-L. Nuthalapati K. Chiang C.-S. Hwang K. Nucleus-targeting gold nanoclusters for simultaneous in vivo fluorescence imaging, gene delivery, and NIR-light activated photodynamic therapy Adv. Funct. Mater. 2015 25 5934 5945 10.1002/adfm.201502650 44. Zhang C. Li C. Liu Y. Zhang J. Bao C. Liang S. Wang Q. Yang Y. Fu H. Wang K. Gold nanoclusters-based nanoprobes for simultaneous fluorescence imaging and targeted photodynamic therapy with superior penetration and retention behavior in tumors Adv. Funct. Mater. 2015 25 1314 1325 10.1002/adfm.201403095 45. Pustovalov V. Analytical modeling of the light-to-heat conversion by nanoparticle ensemble under radiation action Adv. Mater. Sci. Appl. 2015 4 10 22 10.5963/AMSA0401002 46. Jasi’nski M. Majchrzak E. Turchan L. Numerical analysis of the interactions between laser and soft tissues using generalized dual-phase lag equation Appl. Math. Model. 2016 40 750 762 10.1016/j.apm.2015.10.025 47. Yakunin A.N. Avetisyan Y.A. Tuchin V.V. Quantification of laser local hyperthermia induced by gold plasmonic nanoparticles J. Biomed. Opt. 2015 20 051030 10.1117/1.JBO.20.5.051030 25629389 48. Pustovalov V.K. Astafyeva L.G. Fritzsche W. Plasmonic and thermooptical properties of spherical metallic nanoparticles for their thermoplasmonic and photonic applications J. Nanopart. Res. 2014 2014 893459 10.1155/2014/893459 49. Pustovalov V.K. Astafyeva L.G. Fritzsche W. Selection of thermo-optical parameter of nanoparticles for achievement of their maximal thermal energy under optical irradiation Nano Energy 2013 2 1137 1141 10.1016/j.nanoen.2013.04.005 50. Mezeme M.E. Brosseau C. Engineering nanostructures with enhanced thermoplasmonic properties for biosensing and selective targeting applications Phys. Rev. E Stat. Nonlinear Soft Matter Phys. 2013 87 012722 10.1103/PhysRevE.87.012722 23410374 51. Letfullin R.R. George T.F. Plasmonic nanomaterials for nanomedicine Handbook of Nanomaterials Springer Berlin, Germany 2013 1063 1098 52. Yakunin A.N. Avetisyan Y.A. Bykov A.A. Tuchin V.V. Spatio-temporal thermal processes induced by pulsed laser irradiation of medium doped by nanoparticles Proceedings of the IEEE International Conference on BioPhotonics (BioPhotonics) Florence, Italy 20–22 May 2015 157 162 53. Avetisyan Y.A. Yakunin A.N. Tuchin V.V. Novel thermal effect at nanoshell heating by pulsed laser irradiation: Hoop-shaped hot zone J. Biophotonics 2012 5 734 744 10.1002/jbio.201100074 22076925 54. Avetisyan Y.A. Yakunin A.N. Tuchin V.V. Thermal energy transfer by plasmon-resonant composite nanoparticles at pulse laser irradiation Appl. Opt. 2012 51 C88 C94 10.1364/AO.51.000C88 22505118 55. Avetisyan Y.A. Yakunin A.N. Tuchin V.V. On the problem of local tissue hyperthermia control: Multiscale modelling of pulsed laser radiation action on a medium with embedded nanoparticles Quantum Electron. 2011 40 1081 1088 10.1070/QE2010v040n12ABEH014502 56. Richardson H.H. Carlson M.T. Tandler P.J. Hernandez P. Govorov A.O. Experimental and theoretical studies of light-to-heat conversion and collective heating effects in metal nanoparticle solutions Nano Lett. 2009 9 1139 1146 10.1021/nl8036905 19193041 57. Tiwari P.M. Vig K. Dennis V.A. Singh S.A. Functionalized gold nanoparticles and their biomedical applications Nanomaterials 2011 1 31 63 10.3390/nano1010031 58. Rotello V.M. Gold nanoparticles: Preparation, properties, and applications in bionanotechnology Nanoscale 2012 4 1871 1880 22076024 59. Yeh Y.-C. Creran B. Dreaden E. Alkilany A.M. Huang X. Murphy C.J. El-Sayed M.A. The golden age: Gold nanoparticles for biomedicine Chem. Soc. Rev. 2012 41 2740 2779 22109657 60. Raghavendra R. Arunachalam K. Annamalai S.K. Arunachalam A.M. Diagnostics and therapeutic application of gold nanoparticles Int. J. Pharm. Pharm. Sci. 2014 6 74 87 61. Kuo W. Chang C.N. Chang Y.T. Yeh C.S. Antimicrobial gold nanorods with dual-modality photodynamic inactivation and hyperthermia Chem. Сommun. 2009 32 4853 4855 10.1039/b907274h 62. Bresee J. Maier K.E. Boncella M.E. Melander C. Feldheim D.L. Growth inhibition of Staphylococcus aureus by mixed monolayer gold nanoparticles Small 2011 7 2027 2031 10.1002/smll.201100420 21630443 63. Embleton M.L. Nair S.P. Cookson B.D. Wilson M. Selective lethal photosensitization of methicillin-resistant Staphylococcus aureus using an IgG-tin (IV) chlorin e6 conjugate Microb. Drug Resist. 2004 10 92 10.1089/1076629041310000 15256023 64. Perni S. Piccirillo C. Kafizas A. Uppal M. Pratten J. Wilson M. Parkin I.P. Antibacterial activity of light-activated silicone containing methylene blue and gold nanoparticles of different sizes J. Clust. Sci. 2010 21 427 438 10.1007/s10876-010-0319-5 65. Fazaeli Y. Amini M.M. Ashourion H. Heydari H. Majdabadi A. Jalilian A.R. Abolmaali S. Grafting of a novel gold (III) complex on nanoporous MCM-41 and evaluation of its toxicity in Saccharomyces cerevisiae Int. J. Nanomed. 2011 6 3251 3257 66. Zhu X. Xie Y. Zhang Y. Huang H. Huang S. Hou L. Zhang H. Li Z. Shi J. Zhang Z. Photoinactivation of Candida albicans and Escherichia coli using aluminium phthalocyanine on gold nanoparticles J. Biomater. Appl. 2014 29 769 779 10.1177/0885328214543211 25033825 67. Mohd A.S. Tufail S. Khan A.A. Owais M. Gold nanoparticle-photosensitizer conjugate based photodynamic inactivation of biofilm producing cells: Potential for treatment of C. albicans infection in BALB/c Mice PLoS ONE 2015 10 1 20 68. Nombona N. Antunes E. Chidawanyika W. Kleyi P. Tshentu Z. Nyokong T. Synthesis, photophysics and photochemistry of phthalocyanine-E-polylysine conjugates in the presence of metal nanoparticles against Staphylococcus aureus J. Photochem. Photobiol. A 2012 233 24 33 10.1016/j.jphotochem.2012.02.012 69. Jijie R. Dumych T. Chengnan L. Bouckaert J. Turcheniuk K. Hage C.-H. Heliot L. Cudennec B. Dumitrascu N. Boukherroub R. Particle-based photodynamic therapy based on indocyanine green modified plasmonic nanostructures for inactivation of a Crohn’s disease-associated Escherichia coli strain J. Mater. Chem. B 2016 4 2598 2605 10.1039/C5TB02697K 70. Ratto F. Tuchina E.S. Khlebtsov B.N. Centi S. Matteini P. Rossi F. Fusi F. Khlebtsov N.G. Pini R. Tuchin V.V. Combined near infrared photothermolysis and photodynamic therapy by association of gold nanoparticles and an organic dye Proceedings of the Plasmonics in Biology and Medicine VIII San Francisco, CA, USA 22 January 2011 Volume 7911 79111C 71. Tuchina E.S. Tuchin V.V. Khlebtsov B.N. Khlebtsov N.G. Phototoxic effect of conjugates of plasmon-resonance nanoparticles with indocyanine green dye on Staphylococcus aureus induced by IR laser radiation Quantum Electron. 2011 41 354 359 10.1070/QE2011v041n04ABEH014595 72. Tuchina E.S. Petrov P.O. Kozina K.V. Ratto F. Centi S. Pini R. Tuchin V.V. Using gold nanorods labelled with antibodies in photothermal impact of IR laser radiation on Staphylococcus aureus Quantum Electron. 2014 44 683 688 10.1070/QE2014v044n07ABEH015497 73. Tuchina E.S. Petrov P.O. Ratto F. Centi S. Pini R. Tuchin V.V. The action of NIR (808 nm) laser radiation and gold nanorods labeled with IgA and IgG human antibodies on methicillin-resistant and methicillin sensitive strains of Staphylococcus aureus Proceedings of the Biophotonics and Immune Responses X SPIE Bellingham, WA, USA 2015 Volume 9324 93240X 74. Khlebtsov B.N. Tuchina E.S. Tuchin V.V. Khlebtsov N.G. Multifunctional Au nanoclusters for targeted bioimaging and enhanced photodynamic inactivation of Staphylococcus aureus RSC Adv. 2015 5 61639 61649 10.1039/C5RA11713E 75. Omar G.S. Wilson M. Nair S.P. Lethal photosensitization of wound-associated microbes using indocyanine green and near-infrared light BMC Microbiol. 2008 8 111 120 10.1186/1471-2180-8-111 18593460 76. Josefsen L.B. Boyle R.W. Unique diagnostic and therapeutic roles of porphyrins and phthalocyanines in photodynamic therapy, imaging and theranostics Theranostics 2012 2 916 966 10.7150/thno.4571 23082103 77. Xie J. Zheng Y. Ying J.Y. Protein-directed synthesis of highly fluorescent gold nanoclusters J. Am. Chem. Soc. 2009 131 888 889 10.1021/ja806804u 19123810 78. Kennedy L.C. Bickford L.R. Lewinski N.A. Coughlin A.J. Hu Y. Day E.S. West J.L. Drezek R.A. A new era for cancer treatment: Gold-nanoparticle-mediated thermal therapies Small 2011 7 169 183 10.1002/smll.201000134 21213377 79. Sau T.K. Rogach A.L. Jackel F. Klar T.A. Feldmann J. Properties and applications of colloidal nonspherical noble metal nanoparticles Adv. Mater. 2010 22 1805 1825 10.1002/adma.200902557 20512954 80. Xia Y. Li W. Cobley C.M. Chen J. Xia X. Zhang Q. Yang M. Cho E.C. Brown P.K. Gold nanocages: From synthesis to theranostic applications Acc. Chem. Res. 2011 44 914 924 10.1021/ar200061q 21528889 81. Zhang J. Biomedical applications of shape-controlled plasmonic nanostructures: A case study of hollow gold nanospheres for photothermal ablation therapy of cancer J. Phys. Chem. Lett. 2010 1 686 695 10.1021/jz900366c 82. Gobin A.M. Lee M.H. Halas N.J. James W.D. Drezek R.A. West J.L. Near-infrared resonant nanoshells for combined optical imaging and photothermal cancer therapy Nano Lett. 2007 7 1929 1934 10.1021/nl070610y 17550297 83. Maltzahn G. Park J.-H. Agrawal A. Bandaru N.K. Das S.K. Sailor M.J. Bhatia S.N. Computationally guided photothermal tumor therapy using long-circulating gold nanorod antennas Cancer Res. 2009 69 3892 3900 10.1158/0008-5472.CAN-08-4242 19366797 84. Hirn S. Semmler-Behnke M. Schleh C. Wenk A. Lipka J. Schäffler M. Takenaka S. Möller W. Schmid G. Simon U. Particle size-dependent and surface charge-dependent biodistribution of gold nanoparticles after intravenous administration Eur. J. Pharm. Biopharm. 2011 77 407 416 10.1016/j.ejpb.2010.12.029 21195759 85. Wang J. Bai R. Yang R. Liu J. Tang J. Liu Y. Li J. Chaic Z. Chen C. Size- and surface chemistry-dependent pharmacokinetics and tumor accumulation of engineered gold nanoparticles after intravenous administration Metallomics 2015 7 516 524 10.1039/C4MT00340C 25671498 86. Terentyuk G.S. Maslyakova G.N. Khlebtsov N.G. Akchurin G.G. Tuchin V.V. Maksimova I.L. Khlebtsov B.N. Suleymanova L.V. Laser-induced tissue hyperthermia mediated by gold nanoparticles: Toward cancer phototherapy J. Biomed. Opt. 2009 10.1117/1.3122371 19405729 87. Bucharskaya A.B. Maslyakova G.N. Afanasyeva G.A. Terentyuk G.S. Navolokin N.A. Zlobina O.V. Chumakov D.S. Bashkatov A.N. Genina E.A. Khlebtsov N.G. The morpho-functional assessment of plasmonic photothermal therapy effects on transplanted liver tumor J. Innov. Opt. Health Sci. 2015 8 1541004 10.1142/S1793545815410047 88. Heinemann D. Schomaker M. Kalies S. Schieck M. Carlson R. Escobar H.M. Ripken T. Meyer H. Heisterkam A. Gold nanoparticle mediated laser transfection for efficient siRNA mediated gene knock down PLoS ONE 2013 8 1295 10.1371/journal.pone.0058604 23536802 89. Terentyuk G.S. Akchurin G.G. Maksimova I.L. Maslyakova G.N. Khlebtsov N.G. Tuchin V.V. Cancer laser thermotherapy mediated by plasmonic nanoparticles Handbook of Photonics for Biomedical Science Tuchin V.V. CRC Press, Taylor & Francis Group London, UK 2010 763 797 90. Zharov V.P. Galitovsky V. Viegas M. Photothermal detection of local thermal effects during selective nanophotothermolysis Appl. Phys. Lett. 2003 83 4897 4899 10.1063/1.1632546 91. Dressler C. Schwandt D. Beuthan J. Mildaziene V. Zabarylo U. Minet O. Thermally induced changes of optical and vital parameters in human cancer cells Laser Phys. Lett. 2010 7 817 823 10.1002/lapl.201010070 92. Pustovalov V.K. Modeling of the processes of laser-nanoparticle interaction taking into account temperature dependences of parameters Laser Phys. 2011 21 906 912 10.1134/S1054660X11090234 93. Hleb E.Y. Hafner J.H. Myers J.N. Hanna E.Y. Rostro B.C. Zhdanok S.A. Lapotko D.O. LANTCET: Elimination of solid tumor cells with photothermal bubbles generated around clusters of gold nanoparticles Nanomedicine 2008 3 647 667 10.2217/17435889.3.5.647 18817468 94. Lukianova-Hleb E. Hu Y. Latterini L. Tarpani L. Lee S. Drezek R.A. Hafner J.H. Lapotko D.O. Plasmonic nanobubbles as transient vapor nanobubbles generated around plasmonic nanoparticles ACS Nano 2010 4 2109 2123 10.1021/nn1000222 20307085 95. Arita Y. Ploschner M. Antkowiak M. Gunn-Moore F. Dholakia K. Laser-induced breakdown of an optically trapped gold nanoparticle for single cell transfection Opt. Lett. 2013 38 3402 3405 10.1364/OL.38.003402 23988969 96. Qin Z.P. Bischof J.C. Thermophysical and biological responses of gold nanoparticle laser heating Chem. Soc. Rev. 2012 41 1191 1217 10.1039/C1CS15184C 21947414 97. Lukianova-Hleb E.Y. Mutonga M.B.G. Lapotko D.O. Cell-specific multifunctional processing of heterogeneous cell systems in a single laser pulse treatment ACS Nano 2012 6 10973 10981 10.1021/nn3045243 23167546 98. Boulais E. Lachaine R. Hatef A. Meunier M. Plasmonics for pulsed-laser cell nanosurgery: Fundamentals and applications J. Photochem. Photobiol. C Photochem. Rev. 2013 17 26 49 10.1016/j.jphotochemrev.2013.06.001 99. Umebayashi Y. Miyamoto Y. Wakita M. Kobayashi A. Nishisaka T. Elevation of plasma membrane permeability on laser irradiation of extracellular latex particles J. Biochem. 2003 134 219 224 10.1093/jb/mvg132 12966070 100. Pitsillides C.M. Joe E.K. Wei X. Anderson R.R. Lin C.P. Selective cell targeting with light-absorbing microparticles and nanoparticles Biophys. J. Cell 2003 84 4023 4032 10.1016/S0006-3495(03)75128-5 101. Schomaker M. Killian D. Willenbrock S. Heinemann D. Kalies S. Ngezahayo A. Nolte I. Ripken T. Junghanß C. Meyer H. Biophysical effects in off-resonant gold nanoparticle mediated (GNOME) laser transfection of cell lines, primary- and stem cells using fs laser pulses J. Biophotonics 2015 8 646 658 10.1002/jbio.201400065 25302483 102. Kalies S. Antonopoulos G.C. Rakoski M.S. Heinemann D. Schomaker M. Ripken T. Meyer H. Investigation of biophysical mechanisms in gold nanoparticle mediated laser manipulation of cells using a multimodal holographic and fluorescence imaging setup PLoS ONE 2015 10 1295 10.1371/journal.pone.0124052 25909631 103. Kalies S. Keil S. Sender S. Hammer S.C. Antonopoulos G.C. Schomaker M. Ripken T. Escobar H.M. Meyer H. Heinemann D. Characterization of the cellular response triggered by gold nanoparticle–mediated laser manipulation J. Biomed. Opt. 2015 10.1117/1.JBO.20.11.115005 26562032 104. Yao C. Rahmanzadeh R. Endl E. Zhang Z. Gerdes J. Hüttmann G. Elevation of plasma membrane permeability by laser irradiation of selectively bound nanoparticles J. Biomed. Opt. 2005 10 064012 10.1117/1.2137321 16409077 105. Yao C. Qu X. Zhang Z. Hüttmann G. Rahmanzadeh R. Influence of laser parameters on nanoparticle-induced membrane permeabilization J. Biomed. Opt. 2009 14 054034 10.1117/1.3253320 19895136 106. Baumgart J. Humbert L. Boulais É. Lachaine R. Lebrun J.J. Meunier M. Off-resonance plasmonic enhanced femtosecond laser optoporation and transfection of cancer cells Biomaterials 2012 33 2345 2350 22177619 107. St-Louis Lalonde B. Boulais É. Lebrun J.-J. Meunier M. Visible and near infrared resonance plasmonic enhanced nanosecond laser optoporation of cancer cells Biomed. Opt. Express 2013 4 490 499 10.1364/BOE.4.000490 23577284 108. Bartczak D. Muskens O.L. Millar T.M. Sanchez-Elsner T. Kanaras A.G. Laser-induced damage and recovery of plasmonically targeted human endothelial cells Nano Lett. 2011 20 1358 1363 10.1021/nl104528s 21322611 109. Palankar R. Pinchasik B.-E. Khlebtsov B.N. Kolesnikova T.A. Mohwald H. Winterhalter M. Skirtach A.G. Nanoplasmonically-induced defects in lipid membrane monitored by ion current: transient nanopores versus membrane rupture Nano Lett. 2014 14 4273 4279 10.1021/nl500907k 24961609 110. Bibikova O. Singh P. Popov A. Akchurin G. Skovorodkin I. Khanadeev V. Kinnunen M. Bogatyrev V. Khlebtsov N. Vainio S. The effect of laser irradiation on living cells incubated with gold nanoparticles Proc. SPIE 2015 10.1117/12.2083023 111. Bibikova O. Singh P. Popov A. Akchurin G. Skaptsov A. Skovorodkin I. Khanadeev V. Mikhalevich D. Kinnunen M. Akchurin G. Shape-dependent interaction of plasmonic gold nanoparticles with cultured cells at laser exposure J. Biomed. Opt. 2016 submitted 112. Yuan H. Khoury C.G. Hwang H. Wilson C.M. Grant G.A. Vo-Dinh T. Gold nanostars: surfactant-free synthesis, 3D modelling, and two-photon photoluminescence imaging Nanotechnology 2012 23 075102 10.1088/0957-4484/23/7/075102 22260928 113. Grabar K.C. Freeman R.G. Hommer M.B. Natan M.J. Preparation and characterization of Au colloid monolayers Anal. Chem. 1995 67 735 743 10.1021/ac00100a008 114. Khlebtsov B. Khanadeev V. Pylaev T. Khlebtsov N. New T-matrix solvable model for nanorods: TEM-based ensemble simulations supported by experiments J. Phys. Chem. C 2011 115 6317 6323 10.1021/jp2000078 115. Khlebtsov B. Khanadeev V. Panfilova E. Pylaev T. Bibikova O. Staroverov S. Bogatyrev V. Dykman L. Khlebtsov N. New types of nanomaterials: Powders of gold nanospheres, nanorods, nanostars and gold-silver nanocages Nanotechnol. Russ. 2013 8 209 219 10.1134/S1995078013020092 116. Khlebtsov N. Dykman L. Biodistribution and toxicity of engineered gold nanoparticles: A review of in vitro and in vivo studies Chem. Soc. Rev. 2011 40 1647 1671 10.1039/C0CS00018C 21082078 117. Fan Z. Liu H. Mayer M. Deng C.X. Spatiotemporally controlled single cell sonoporation Proc. Natl. Acad. Sci. 2012 109 16486 16491 10.1073/pnas.1208198109 23012425 118. Bibikova O. Popov A. Skovorodkin I. Prilepskyi A. Pylaev T. Bykov A. Staroverov S. Bogatyrev V. Tuchin V. Kinnunen M. Plasmon-resonant gold nanoparticles with variable morphology as optical labels and drug carriers for cytological research Proc. SPIE 2013 10.1117/12.2032547 119. Bibikova O. Popov A. Bykov A. Fales A. Yuan H. Skovorodkin I. Kinnunen M. Vainio S. Vo-Dinh T. Tuchin V. Plasmon-resonant gold nanostars with variable size as contrast agents for imaging applications IEEE J. Sel. Top. Quantum Electron. 2016 22 1 8 10.1109/JSTQE.2016.2526602 120. Jain P.K. Lee K.S. El-Sayed I.H. El-Sayed M.A. Calculated absorption and scattering properties of gold nanoparticles of different size, shape, and composition: Applications in biological imaging and biomedicine J. Phys. Chem. B 2006 110 7238 7248 10.1021/jp057170o 16599493 121. Kalies S. Heinemann D. Schomaker M. Gentemann L. Meyer H. Ripken T. Immobilization of gold nanoparticles on cell culture surfaces for safe and enhanced gold nanoparticle-mediated laser transfection J. Biomed. Opt. 2014 19 70505 10.1117/1.JBO.19.7.070505 25069006 122. Heinemann D. Kalies S. Schomaker M. Ertmer W. Escobar H.M. Meyer H. Ripken T. Delivery of proteins to mammalian cells via gold nanoparticle mediated laser transfection Nanotechnology 2014 25 245101 10.1088/0957-4484/25/24/245101 24859743 123. Chen C.-C. Lin Y.-P. Wang C.-W. Tzeng H.-C. Wu C.-H. Chen Y.-C. Chen C.-P. Chen L.-C. Wu Y.-C. DNA-gold nanorod conjugates for remote control of localized gene expression by near infrared irradiation J. Am. Chem. Soc. 2006 128 3709 3715 10.1021/ja0570180 16536544 124. Antkowiak M. Dholakia K. Gunn-Moore F. Optical transfection of mammalian cells Understanding Biophotonics: Fundamentals, Advances and Applications Tsia K. CRC Press, Taylor & Francis Group London, UK 2014 693 734 125. Gu L. Koymen A.R. Mohanty S.K. Crystalline magnetic carbon nanoparticle assisted photothermal delivery into cells using CW near-infrared laser beam Sci. Rep. 2014 4 5106 10.1038/srep05106 24870227 126. Arita Y. Torres-Mapa M.L. Lee W.M. Čižmár T. Campbell P. Gunn F.J. Dholakia K. Spatially optimized gene transfection by laser-induced breakdown of optically trapped nanoparticles Appl. Phys. Lett. 2011 98 093702 10.1063/1.3554415

PMC005xxxxxx/PMC5000693.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081296ijms-17-01296ReviewBinding and Fusion of Extracellular Vesicles to the Plasma Membrane of Their Cell Targets Prada Ilaria 1Meldolesi Jacopo 2*Drummen Gregor Academic Editor1 CNR Institute of Neuroscience, 20133 Milan, Italy; [email protected] San Raffaele Scientific Institute, DIBIT, via Olgettina 58, 20132 Milan, Italy* Correspondence: [email protected]; Tel.: +39-02-2643-277009 8 2016 8 2016 17 8 129620 6 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Exosomes and ectosomes, extracellular vesicles of two types generated by all cells at multivesicular bodies and the plasma membrane, respectively, play critical roles in physiology and pathology. A key mechanism of their function, analogous for both types of vesicles, is the fusion of their membrane to the plasma membrane of specific target cells, followed by discharge to the cytoplasm of their luminal cargo containing proteins, RNAs, and DNA. Here we summarize the present knowledge about the interactions, binding and fusions of vesicles with the cell plasma membrane. The sequence initiates with dynamic interactions, during which vesicles roll over the plasma membrane, followed by the binding of specific membrane proteins to their cell receptors. Membrane binding is then converted rapidly into fusion by mechanisms analogous to those of retroviruses. Specifically, proteins of the extracellular vesicle membranes are structurally rearranged, and their hydrophobic sequences insert into the target cell plasma membrane which undergoes lipid reorganization, protein restructuring and membrane dimpling. Single fusions are not the only process of vesicle/cell interactions. Upon intracellular reassembly of their luminal cargoes, vesicles can be regenerated, released and fused horizontally to other target cells. Fusions of extracellular vesicles are relevant also for specific therapy processes, now intensely investigated. exosomesectosomesextracellular vesicles (EVs)vesicle cargomultivesicular bodies (MVBs)plasma membraneretroviral-type membrane fusionsreceptorsdischarge of luminal cargoes ==== Body 1. Introduction During the last 15 years, exosomes and ectosomes, the two types of extracellular vesicles (EVs) endowed with physiological and pathological functions, have been among the structures attracting the greatest attention in the scientific community. Exosomes were identified in 2000 as the vesicles trapped within multivesicular bodies (MVBs). Their release to the extracellular space takes place upon the exocytosis of the latter organelles [1]. Considered from the very beginning as vesicles expressed by all types of cells, exosomes started to be recognized and characterized shortly after discovery. Ectosomes, referred to also with other names (shedding vesicles, microparticles, microvesicles and others), are generated at the plasma membrane as small cytoplasmic protrusions covered by specialized membrane rafts which rapidly grow up and are released by fission of their stalk. First described in 1991 in human neutrophils [2], ectosomes have been since investigated separately in various cell types. For years, therefore, little attention was paid to their expression and function in all cells. General interest about these EVs emerged after 2005 and grew progressively over the next years. At present, both exosomes and ectosomes are being intensely investigated, however mostly separate from each other. Time has come, therefore, to consider the properties of the two types of EVs together, focusing in particular on the mechanisms of their generation and function [3]. Numerous aspects of exosome and ectosome generation are known to be different. Among these are the sites of their origin, at the MVBs and the plasma membrane; the length of their intracellular life before discharge, long for exosomes (tens of minutes or more, necessary for their accumulation within MVBs and for their release) and much shorter for ectosomes (a few tens of seconds for their generation at the plasma membrane); their size (diameters of 50–150 and 100–350 nm, respectively); the mechanisms for the accumulation, within their luminal content, of proteins and other macromolecules indicated here as their cargo. In addition, the two types of EVs are often considered molecularly different. Thus, quite a few proteins were proposed as markers of exosomes. Recent studies, however, have shown many such molecules to be present also in ectosomes, although at lower levels. Real markers now recognized, specific of either type of EV, are few. Nevertheless, they remain useful for the identification of the two EVs [3]. Once generated and released, the destinies of the two types of EVs become similar, if not identical. Fractions of both types of EVs undergo dissolution of their membrane. As a consequence, the factors accumulated in their cargo are released to the extracellular space. Most other EVs, however, do not dissolve, but pursue their navigation in the extracellular fluid. In vivo, the navigation of some of them is long-term, sufficient to reach the large fluids of the bodies. Other EVs, however, remain local, establishing interactions with cells defined here as target cells. Such interactions are at least of two types. Some EVs, taken up by various forms of endocytosis, can either be discharged into lysosomes or fuse with endocytic membranes, with ensuing release of their cargo into the cytoplasm. Alternatively, the binding of EVs to target cells is followed by their fusion with the plasma membrane. The different conditions of the two pathways, acidic within endosomes, near neutral in the extracellular space, appear critical for the EV membrane fusion. The process occurring upon uptake of EVs (especially exosomes) into endosomes has been extensively investigated and recently illustrated in comprehensive reviews [4,5]. In contrast, the interaction/binding/fusion of EVs with the plasma membrane, although widely accepted, has never been described in detail. The present mini-review is focused exclusively on the latter process at the surface of target cells. The processes occurring with exosomes and ectosomes appear very similar [3]. Therefore, they will be considered together, under the common definition of EVs. 2. Vesicle (EV) Interactions with the Plasma Membrane of Target Cells As already mentioned, exosomes and ectosomes are generated and released by all types of cells. This property may suggest the two types of EVs to be homogeneous in terms of composition. This is not the case. The two EV types are, in fact, distinct in composition. This can be the case of EVs released from different cells. Even when EVs originate from the same type of cell, some of their properties can be different, depending on reasons such as the state of development of their cells of origin, their functions, and the degree of their stimulation. Large differences have been reported also between wild-type EVs and the analogous EVs of cancer cells [3]. These differences may account for the specificity of EV interactions with target cells. For example, the ectosomes released from platelets are known to interact with macrophages and endothelial cells, but not with neutrophils; the ectosomes from neutrophils to interact not only with platelets but also with macrophages and dendritic cells [6,7]. Analogously, the exosomes released by neuroblastoma cells bind indiscriminately to neurons and glial cells, whereas the exosomes released from stimulated cortical neurons bind only to neurons [8]. In order to clarify in detail the properties of the EV-cell interactions, the investigation has been expanded using new approaches by which the temporality and the dynamics of the process can be established [9]. Excellent results have been obtained by employing optical tweezers. By this approach, a single EV of known nature and origin is captured and transferred to the surface of various cell types in primary culture. Interestingly, sliding results can be different. In microglial cells, EVs from astrocytes were found to exhibit directional, long movements, whereas in astrocytes they exhibited only minor oscillations close to the site of first adhesion. After some duration of surface sliding, the EVs were seen to markedly reduce their movement (Figure 1), establishing a binding that soon evolved into fusion [10]. In view of the specificities summarized above, the variable properties of the EV-cell interactions can be attributed to the genes expressed by the original and target cells. Specifically, the interactions appear due to the cell surface proteins necessary for EV binding and compatible for the ensuing fusion to get started. 3. EV Binding to the Plasma Membrane of Target Cells Binding and fusion of EVs to the external surface of their target cells should be first distinguished from the apparently analogous fusion processes taking place inside the cell, during organelle traffic, exocytosis, endocytosis and other events. Intracellular binding and fusion require the involvement of various proteins, including actin, other cytoskeletal proteins and numerous forms of soluble N-ethylmaleimide-sensitive protein receptor (SNARE), which have nothing to do with the surface binding and fusion processes. Among the latter processes, the ones previously intensely investigated are virus fusions, which occur with the participation of four classes of proteins [11]. Proteins of class I and II have been identified as possibly involved also in the binding and fusion of EVs. Here we will briefly consider the possible involvement of class I proteins in the EV-cell binding processes. The ensuing fusion processes will be considered in the next section. The initial interactions of EVs are expected to require specific, high affinity binding of at least two surface proteins, one protruding from the EVs, the other from the plasma membrane of the target cells. The surface proteins of class I, syncytin-1 and syncytin-2, were discovered on placental trophoblast plasma membrane, where they participate in the cell-to-cell fusion process [12,13]. These proteins are not limited to the plasma membrane of placental cells. They have been found also on human gametes [14], blood cells [13], osteoclasts [15], differentiating myoblasts [16], pituitary gland cells and various types of tumor cells [17,18]. Moreover, they are exposed by trophoblast exosomes [13] and may be present also on the EVs of other cell types. The interest in these proteins was greatly increased by the identification of two high affinity binding proteins, the receptors Major Facilitator Superfamily Domain 2a (MFSD2a) and Soluble Carrier Family 1 (ASCT-2). These proteins belong to families of carbohydrate and neutral amino acid transporters, respectively [14,19,20]. Properties of the syncytin-2/MFSD2a binding were altered by some, but not all, single-nucleotide mutations of the syncytin gene, and by N-glycosylations of the protein [21]. Interestingly, syncytin-2 and MFSD2a were found to exhibit a critical distribution in the human gametes destined to fuse [14]. Moreover, the syncytin-1-equipped exosomes from placental trophoblasts were shown to bind and fuse with blood cells [13]. Summing up, syncytins are known proteins possibly involved in the binding of EVs to target cells, i.e., in the process that precedes fusion (Figure 2). 4. Present Knowledge about EV-Target Cell Surface Fusions The knowledge about EV/target cell fusion (Figure 2), analogous to the fusions taking place between the plasma membranes of two cells, is based on two types of information, the first from the fusions of viruses with cells [11], the other from the data about the binding processes reported in the preceding section [12,13,14,15,16,17,18,19,20,21]. In order for fusions to take place, several events need to take place, including the insertion, in the target cell plasma membrane, of the hydrophobic segments of fusogenic proteins, followed by lipid reorganization, protein restructuring and membrane dimpling [11]. Considered from the viral point of view, the two proteins involved in the fusion of at least some EVs belong to the classes I and II. Class I proteins, syncytin-1 and syncytin-2, are composed of α-helix-rich pre-fusion trimers that insert their hydrophobic fusion peptides in the target membranes. The two proteins then refold as post-fusion trimers. Class II includes the Epithelial Fusion Failure 1 (EFF-1) protein [22]. This protein, not considered here for binding because its possible receptor is unknown, includes β-sheet-rich pre-fusion homodimers and heterodimers that include loops destined to be inserted in the target membrane. At the end of the process the dimers refold into post-fusion trimers [11]. During the time between the pre-folding and post-folding of their fusogenic proteins, the EV and cell plasma membranes become continuous (Figure 2). This process is critical because, upon their discharge into the cytosol, it makes the molecules of the luminal cargo of EVs start carrying out their functions. Bioactive macromolecules of the cargo include proteins, various mRNAs, several miRNAs, and some DNA sequences. Within the cytoplasm the environment is crowded. The diffusion rate of the cargo macromolecules is variable, and their effects occur at different times. Among the discharged proteins, some are common to many eukaryotic cells, others to few cells only. Additional proteins are synthesized according to the mRNA and the DNA specificity. The relevance of protein functions is variable, from ordinary to unique. Highly important are the miRNAs that contribute significantly to the turnover of many proteins, thus inducing effects that may be stimulatory or inhibitory for target cells [23,24]. Additional consequences of the fusion, dependent on the properties of the donor and recipient cells, include changes of the gene expression. Taken together, the processes triggered by the macromolecules released by the EVs lead to the reprogramming of the target cell structure and function. EV transfer from the cell of origin to a target cell does not necessarily account for a complete interaction process. In target cells, in fact, the cargo and membrane molecules transferred upon EV fusion can accumulate again into vesicles that are released and then fuse to other target cells. This horizontal transfer, taking place among the cells of a tissue, could be expanded also to other cell types, contributing to many physiological processes such as coagulation, stem cell renewal and expansion, and also to the pathogenesis of diseases including inflammation diseases (atherosclerosis, angiogenesis and others), diabetes, and the growth, invasion and metastases of tumors [5,23,24,25,26,27]. Based on the analysis of EVs accumulated in urine, blood sera, saliva or cerebrospinal fluids, the molecules released by patients can also be employed for their diagnosis and prognosis [5,27,28,29,30]. Finally, manipulated vesicles can be employed in new prospective therapies. This issue is discussed in the following section. 5. Perspectives of Therapy Knowledge about their origin and properties stimulated the idea of employing EVs for therapy [5]. A problem that was initially considered was the nature of the vesicles to be employed for the task. Stem cell-derived EVs, loaded with exogenous genetic cargoes, appeared naturally equipped for human genetic therapies [31,32]. This approach, however, has remained without adequate development. More recently, the interest switched towards the delivery of drugs. For this purpose a few technical problems were considered, such as the need for EV stabilization, for example by glycosylation of their surface peptides [33,34]. The present perspectives are based on the design of exogenous EVs, loaded with clearly defined therapeutic cargoes and appropriately engineered with surface markers, to assure their targeting to diseased cells. These vesicles are expected to become useful for the development of a drug nano-delivery system, appropriate for future targeting to specific tissues, such as the human brain [28,35]. 6. Conclusions The task of the present review was the demonstration that specific binding and fusion of EVs with the plasma membrane of target cells is a process of great interest for cell physiology and pathology. Binding and fusion of EVs does occur also upon their uptake by the endocytic system, extensively presented in [4,5]. However, not only the timing but also the processing of the internalized EVs could be different from that of EVs undergoing direct fusion with the plasma membrane. The EVs are often defined as structures that have displaced the external borders of cells, from the plasma membrane to the limit of their navigation. The findings reported here have documented additional properties of EVs including, among others, their horizontal intercellular transfer of macromolecules; their target cell signaling induced by the intercellular traffic of macromolecules; their long-term control of gene expression. The identification of these and other extraordinary processes has introduced new concepts and ideas, not only in cell biology and physiology, but also in medicine. At present, in fact, the role of EVs is being evaluated also by procedures employed for the diagnosis, prognosis, evaluation and therapy of single patients. Based on the results of their intense, ongoing investigation, the EVs can therefore be envisaged as new tools, employed for the progress of biomedical sciences, destined to be further expanded in the next few years. Acknowledgments The original work of the two labs was supported by grants from Telethon (GGGP09066) and FISM (2012/R/17). Ilaria Prada is supported by a fellowship of the Umberto Veronesi Foundation, Milan. We are grateful to Claudia Verderio and Emanuele Cocucci for their key role in the original work of the Milan Labs included in this mini-review, together with the work of many other groups. Author Contributions Both Ilaria Prada and Jacopo Meldolesi have participated in the critical analysis of the literature. The review, written by Jacopo Meldolesi, has been reviewed by both authors, and read by a few expert colleagues. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Optical tweezer-induced direct interaction of a vesicle (EV) with the surface of a microglial target cell. The figure illustrates the interaction and binding events occurring in a cultured microglial cell interacting with a single EV (marked by arrows) trapped by optical tweezers (marked by triangles) (A) and then transferred near the periphery of the cell surface (B); Upon its release, the EV exhibited a number of oscillations ()F)), after which it started sliding over the cell surface towards the central area of the cell (C,D); After reaching a critical site of the cell surface, the EV sliding decreased markedly (E); (F) reports the whole pathway (blue area) followed by the EV, illustrated by the separate steps of (C–E). Scale bar in A = 5 µm. The morphological images are illustrated also quantitatively in (G) and (H). Notice in G the initial oscillations reported also by the cell in (F); in (H) the strong reduction of the speed, evident after approximately 400 s. Such a reduction is probably due to the receptor binding that anticipates the EV/cell fusion, occurring at the surface or upon internalization into endosomes. Modified with permission from Reference [10]. Figure 2 Illustration of the binding and ensuing fusion of an EV to the plasma membrane of its target cell. The EVs (such as the one shown in Figure 1) include, at their surface, domains of the trans-membrane proteins syncytins. (A) An EV with syncytin-2 is approaching a target cell which, in the plasma membrane, exhibits the syncytin 2-specific receptor, HFSD2a; (B) The EV protein and its receptor appear bound to each other; Hydrophobic loops of the vesicle protein begin to deepen into the plasma membrane, contributing to its molecular re-arrangement, with protein depletion of its external layer (C); This is followed by the hemifusion of the EV membrane with the cell target plasma membrane (D); followed by the re-organization of the two, closely attached membranes, with their dissolution at the fusion site (E); The fusion induced the ensuing insertion of the EV membrane in the target plasma membrane, and the release to the cytoplasm of the luminal cargo molecules: proteins, RNAs and small DNA sequences (F). ==== Refs References 1. Denzer K. Kleijmeer M.J. Heijnen H.F. Stoorvogel W. Geuze H.J. Exosome: From internal vesicle of the multivesicular body to intercellular signaling device J. Cell Sci. 2000 113 3365 3374 10984428 2. Stein J.M. Luzio P. Ectocytosis caused by sublytic autologous complement attack on human neutrophils. The sorting of endogenous plasma membrane proteins and lipids into shed vesicles Biochem. J. 1991 274 381 386 10.1042/bj2740381 1848755 3. Cocucci E. Meldolesi J. Ectosomes and endosomes: Shedding the confusion between extracellular vesicles Trends Cell Biol. 2015 25 364 372 10.1016/j.tcb.2015.01.004 25683921 4. Mulcahy L.A. Pink R.C.M. Carter D.R.F. Routes and mechanisms of extracellular vesicle uptake J. Extracell. Vesicles 2014 3 10.3402/jev.v3.24641 25143819 5. Christianson H.C. Svensson K.J. Belting M. Exosome and microvesicle mediated phene transfer in mammalian cells Semin. Cancer Biol. 2014 28 31 38 10.1016/j.semcancer.2014.04.007 24769057 6. Gasser O. Hess C. Miot S. Deon C. Sanchez J.C. Schifferli J.A. Characterization and properties of ectsosomes released by human polymorphonuclear neutrophils Exp. Cell Res. 2003 285 243 257 10.1016/S0014-4827(03)00055-7 12706119 7. Pluskota E. Woody N.M. Szpak D. Ballantyne C.M. Soloviev D.A. Simon D.I. Plow E.F. Expression, activation, and function of integrin αMβ2 (Mac1) on neutrophil-derived microparticles Blood 2008 112 2327 2335 10.1182/blood-2007-12-127183 18509085 8. Chivet M. Javalet C. Laulagnier K. Blot B. Hemming F.J. Sadoul R. Exosomes secreted by cortical neurons upon glutamatergic synapse activation specifically interact with neurons J. Extracell. Vesicles 2014 3 1296 10.3402/jev.v3.24722 25398455 9. Tian T. Zhu Y.L. Hu F.H. Wang Y.Y. Huang N.P. Xiao Z.D. Dynamics of exosome internalization and trafficking J. Cell. Physiol. 2013 228 1487 1495 10.1002/jcp.24304 23254476 10. Prada I. Amin L. Furlan R. Legname G. Verderio C. Cojoc D. A new approach to follow a single extracellular vesicle-cell interaction using optical tweezers BioTechniques 2016 60 35 41 10.2144/000114371 26757810 11. Podbilewicz B. Virus and cell fusion mechanisms Annu. Rev. Cell Dev. Biol. 2014 30 111 139 10.1146/annurev-cellbio-101512-122422 25000995 12. Potgens A.J. Drewlo S. Kokozidou M. Kaufmann P. Syncytin: The major regulator of trophoblast fusion? Recent developments and hypotheses on its action Hum. Reprod. Update 2004 10 487 496 10.1093/humupd/dmh039 15333590 13. Tolosa J.M. Schjenken J.E. Clifton V.L. Vargas A. Barbeau B. Lowry P. Maiti K. Smith R. The endogenous retroviral envelope protein syncytin-1 inhibits LPS/PHA-stimulated cytokine responses in human blood and is sorted into placental exosomes Placenta 2012 33 933 941 10.1016/j.placenta.2012.08.004 22999499 14. Bierregaard B. Lemmen J.G. Petersen M.R. Østrup E. Iversen L.H. Almstrup K. Larsson L.I. Ziebe S. Syncytin-1 and its receptor are present in human gametes J. Assist. Reprod. Genet. 2014 31 533 539 10.1007/s10815-014-0224-1 24687878 15. Søe K. Andersen T.L. Hobolt-Pedersen A.S. Bjerregaard B. Larsson L.I. Delaissé J.M. Involvement of human endogenous retroviral syncytin-1 in human osteoclast fusion Bone 2011 48 837 846 10.1016/j.bone.2010.11.011 21111077 16. Bierregard B. Ziomkiewicz I. Schulz A. Larsson L.I. Syncytin-1 in differentiating human myoblasts: Relationship to caveolin-3 and myogenin Cell Tissue Res. 2014 31 533 539 17. Buslei R. Strissel L. Henke C. Schey R. Lang N. Ruebner M. Stolt C.C. Fabry B. Buchfelder M. Strick R. Activation and regulation of endogenous retroviral genes in the human pituitary gland and related endocrine tumora Neuropathol. Appl. Neurobiol. 2014 41 180 200 10.1111/nan.12136 24635849 18. Maliniemi P. Vincendeau M. Mayer J. Frank O. Hahtola S. Karenko L. Carlsson E. Mallet F. Seifarth W. Leib-Mösch C. Expression of human endogenous retrovirus-w including syncytin-1 in cutaneous T-cell lymphoma PLoS ONE 2013 8 1296 19. Esnault C. Priet S. Ribet D. Vernochet C. Bruls T. Lavialle C. Weissenbach J. Heidmann T. A placenta-specific receptor for the fusogenic, endogenous retroviral-directed, human syncytin-2 Proc. Natl. Acad. Sci. USA 2008 106 17532 17537 10.1073/pnas.0807413105 18988732 20. Toufaily C. Vargas A. Lemire M. Lafond J. Rassart E. Barbeau B. MLSD2a, the syncytin-2 receptor, is important for trophoblast fusion Placenta 2013 34 85 88 10.1016/j.placenta.2012.10.012 23177091 21. Cui L. Wang H. Lu X. Wang R. Zheng R. Li Y. Yang X. Jia W.T. Zhao Y. Wang Y. Effects of individually silenced N -glycosylation sites and non-synonymous single-nucleotide polymorphisms on the fusogenic function of human syncytin-2 Cell Adhes. Migr. 2016 10 39 55 10.1080/19336918.2015.1093720 26853155 22. Pérez-Vargas J. Krey T. Valansi C. Avinoam O. Haouz A. Jamin M. Raveh-Barak H. Podbilewicz B. Rey F.A. Structural basis of eukaryotic cell-cell fusion Cell 2014 157 407 419 10.1016/j.cell.2014.02.020 24725407 23. Li J. Zhang Y. Liu. Y. Dai X. Li W. Cai X. Yin Y. Wang Q. Xue Y. Wang C. Microvesicle-mediated transfer of microRNA-150 from monocytes to endothelial cells promote angiogenesis J. Biol. Chem. 2013 288 23586 23596 10.1074/jbc.M113.489302 23766514 24. He W.A. Calore F. Londhe P. Canella A. Guttridge D.C. Croce C.M. Microvesicles containing miRNAs promote muscle cell death in cancer cachexia via TLR7 Proc. Natl. Acad. Sci. USA 2014 111 4525 4529 10.1073/pnas.1402714111 24616506 25. Muraluidharan-Chiari V. Clancy J.W. Sedgwick A. D’Souza-Schorey C. Microvesicles: Mediators of extracellular communication during cancer progression J. Cell Sci. 2010 123 1603 1611 10.1242/jcs.064386 20445011 26. Ohno S. Ishikawa A. Kuroda M. Role of exosomes and microvesicles in disease pathogenesis Adv. Drug Deliv. Rev. 2013 65 398 401 10.1016/j.addr.2012.07.019 22981801 27. Wu Z.H. Ji C.L. Li H. Qiu G.X. Gao C.J. Weng X.S. Membrane microparticles and diseases Eur. Rev. Med. Pharmacol. Sci. 2013 17 2420 2427 24089218 28. De Toro J. Herschlik L. Waldner C. Mongini C. Emerging roles of exosomes in normal and pathological conditions: New insights for diagnosis and therapeutic applications Front. Immunol. 2015 6 1296 10.3389/fimmu.2015.00203 25999947 29. Loyer X. Vion A.C. Tedgui A. Boulanger C.M. Microvesicles as cell-cell messengers in cardiovascular diseases Circ. Res. 2014 114 345 353 10.1161/CIRCRESAHA.113.300858 24436430 30. Martins V.R. Dias M.S. Hainaut P. Tumor-cell-derived microvesicles as carriers of molecular information in cancer Curr. Opin. Oncol. 2013 25 66 75 10.1097/CCO.0b013e32835b7c81 23165142 31. Lee Y. El Andaloussi S. Wood M.J. Exosomes and microvesicles: Extracellular vesicles for genetic information transfer and gene therapy Hum. Mol. Genet. 2012 21 R125 R134 10.1093/hmg/dds317 22872698 32. El Andaloussi S. Mäger I. Breakefield X.O. Wood M.J. Extracellular vesicles: Biology and emerging therapeutic opportunities Nat. Rev. Drug Discov. 2012 12 347 357 10.1038/nrd3978 23584393 33. Huang M.E. Leonard J.N. Stabilization of exosome-tageted peptides via engineered glycosylation J. Biol. Chem. 2015 290 8166 8172 10.1074/jbc.M114.621383 25657008 34. Van Dommelen S.M. Vader P. Lakhal S. Kooijmans S.A. van Solinge W.W. Wood M.J. Schiffelers R.M. Microvesicles and exosomes: Opportunities for cell-derived membrane vesicles in drug delivery J. Control. Release 2012 161 635 644 10.1016/j.jconrel.2011.11.021 22138068 35. Aryani A. Denecke B. Exosomes as a nanodelivery system: A key to the future of neuromedicine Mol. Neurobiol. 2016 53 818 834 10.1007/s12035-014-9054-5 25502465

PMC005xxxxxx/PMC5000694.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081297ijms-17-01297ArticleHow the Proximal Pocket May Influence the Enantiospecificities of Chloroperoxidase-Catalyzed Epoxidations of Olefins Morozov Alexander N. *Chatfield David C. *De Visser Samuel Academic EditorDepartment of Chemistry and Biochemistry, Florida International University, 11200 SW 8th St., Miami, FL 33199, USA* Correspondence: [email protected] (A.N.M.); [email protected] (D.C.C.); Tel.: +1-305-348-1198 (A.N.M.); +1-305-348-3977 (D.C.C.)09 8 2016 8 2016 17 8 129727 6 2016 01 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Chloroperoxidase-catalyzed enantiospecific epoxidations of olefins are of significant biotechnological interest. Typical enantiomeric excesses are in the range of 66%–97% and translate into free energy differences on the order of 1 kcal/mol. These differences are generally attributed to the effect of the distal pocket. In this paper, we show that the influence of the proximal pocket on the electron transfer mechanism in the rate-limiting event may be just as significant for a quantitatively accurate account of the experimentally-measured enantiospecificities. heme-thiolate enzymeschloroperoxidasecytochrome P450Compound Iproximal pockethydrogen bondinghelix dipolecatalytic reactivityepoxidationdensity functional theory ==== Body 1. Introduction Chloroperoxidase (CPO), an enzyme secreted by the marine fungus Caldariomyces fumago, is a glycosylated heme-thiolate protein known for its exceptional versatility [1]. Along with its native function of halogenating organic substrates using chloride, bromide and iodide ions [2,3,4], CPO is also capable of many promiscuous activities, including peroxidase, catalase and cytochrome P450 (P450) types of reactions [1]. CPO-catalyzed reactions are of biotechnological and environmental importance [5], therefore attracting current research interest on both the experimental [6,7,8,9,10,11,12,13,14] and theoretical sides [15,16,17,18,19,20,21]. The catalytic cycle requires a two-electron oxidation of the ferric heme center, using hydrogen peroxide or other suitable peroxide, and the glutamic acid side chain (E183) in the distal pocket as a general acid-base catalyst, to form Compound I (CPO-I), a highly reactive oxyferryl porphyrin π-cation radical intermediate [15,20,22,23]. This first step is followed by the oxidation of a substrate, and the reaction cycle ends as CPO regains the native state. The active center of CPO has both peroxidase and P450 features: like a peroxidase, CPO has a polar pocket on the distal side of the heme; like P450, CPO possesses a cysteine-derived thiolate axial ligand on the proximal side [24]. CPO’s distal pocket is connected to the protein surface by a roughly 10-Å wide channel, similar to P450s’ and with no analog in peroxidases [24,25]. In a previous computational study, we showed that CPO has the ability to mimic both peroxidase-like and P450-like distal pockets to tune the catalytic efficiency [18]. A key for the oxidative catalytic functions of CPO and P450s is the proximal thiolate ligand [24,26], which is a strong electron donor “pushing” electrons [27]. It took more than fifty years from the discovery of P450-type chemistry to establish that the “push” of the proximal thiolate results in a trade-off between the redox potential of the heme active center and the basicity of a distal axial ligand and, hence, constitutes a smart adaptation to the different stages of the reaction cycle [28,29], facilitating both O–O bond scission of a ferric hydroperoxy species [15,20,30] and subsequent oxygen transfer into a C=C double bond [31] or an inert C–H bond [28,29,32], as well as other oxygen transfer reactions [1,33]. In heme-thiolate enzymes, the thiolate “push” is modulated by adjusting the thiolate-to-thiol character of the proximal ligand via sulfur/backbone-amide (NH–S) hydrogen bonds, a conserved feature of heme-thiolate proteins [34]. In CPO, the proximal thiolate ligand forms A31:NH–C29:S and L32:NH–C29:S hydrogen bonds (Figure 1) [24], made possible by the backbone conformation of the proximal pocket’s C–P–A–L peptide fragment [35]. P450cam contains the peptide fragment C–L–G–X, conserved in P450s, which forms L358:NH–C357:S, G359:NH–C357:S and Q360:NH–C357:S thiolate sulfur/backbone-amide hydrogen bonds via a similar backbone conformation [35]. Recently, we have summarized an extensive set of experimental data available on the effect of proximal NH-S hydrogen bonding in P450cam [21]. Briefly, there is accumulating evidence that NH-S hydrogen bonds play a chemical role: (1) by modulating the thiolate “push” effect through the alteration of π-electron donation by the proximal thiolate [36]; (2) by modulating the heme reduction potential by up to ~200 mV per NH–S hydrogen bond [36,37,38]. NH–S hydrogen bonds also play a structural role, in protecting and stabilizing Fe-S coordination [39]. On the theoretical side, calculations of the effect of proximal NH–S hydrogen bonds on oxidations catalyzed by a model heme-thiolate Compound I, using (SH)− to represent the proximal ligand, indicate that NH–S bonding has an influence on the chemoselectivity of Compound I toward hydroxylation vs. epoxidation [40,41]. Theoretical simulations of proximal pocket/sulfur hydrogen bonding in nitric oxide synthase showed that such bonding may affect the distribution of the thiolate/porphyrin unpaired spin [42]. The effect of hydrogen bonding to the distal oxygen of model manganese(IV)-oxo and iron(IV)-oxo oxidants on the oxidants’ ability for oxygen atom transfer was studied [43]. However, to the best of our knowledge, there are no published theoretical results regarding the possibility that the proximal helix and the NH–S bonds may affect the enantiospecificity of oxidations catalyzed by heme-thiolate enzymes. In addition to NH–S hydrogen bonding, the electropositivity of the N-terminus of the proximal α-helix also modulates the anionic character of the proximal thiolate ligand in CPO and P450 [34]. The helix dipole moment is in the range of 3.5–5 D per peptide residue, according to both experiment and theoretical calculation [44,45]. Elements of secondary structure are known to play mechanistic roles in folding and post-translational modification of heme-containing proteins [46,47,48,49]. The electric field at helix termini is strong enough to influence protein folding, ligand binding and enzymatic reactions [44,50]. Experimental studies of heme-thiolate model complexes showed that adding the proximal helix to a model compound increases the FeIII/II redox potential by 130 mV for a CPO model and by 70 mV for a P450cam model [51]. In view of this, it is not surprising that the proximal helix may amplify the NH–S effect in heme-thiolate enzymes, as experiments suggest [51]. CPO’s ability to catalyze P450-type enantiospecific epoxidations of olefin substrates [52,53] is of particular interest, as chiral epoxides are synthons for various pharmaceutical and industrial applications. The experimental data described above suggest that the heme thiolate’s secondary coordination sphere and the proximal helix are of importance for epoxidations catalyzed by heme-thiolate enzymes. Our recent DFT studies provide theoretical evidence that the combined effect of the proximal NH–S bonds and the dipole moment of the proximal helix significantly enhances CPO’s reactivity toward the epoxidation of olefinic substrates [21]. When the environment of the proximal thiolate ligand (Figure 2B) is included in the model, the rate limiting barrier for C–O bond formation on the doublet spin surface for CPO-catalyzed epoxidation of cis-β-methylstyrene (CBMS) to form the 1S2R enantiomer is lowered by ~4.6 kcal/mol relative to the ~15 kcal/mol barrier for the bare-thiolate model (Figure 2A). In this nomenclature, the alpha carbon is labeled 1 and followed by its chirality, here S, while the beta carbon is labeled 2 and followed by its chirality, here R. It was estimated that the dipole moment of the proximal helix contributes ~1/3 of the decrease [21]. For the 1S2R enantiomer on the doublet spin surface, we found that the effects of CPO’s proximal pocket change the preferred electron transfer mechanism. For the bare thiolate model of CPO-I, the first and rate-limiting event is the reduction of FeIV to FeIII by an electron transferred from the C=C π bond to the oxyferryl π* orbital. With the proximal pocket added to the model, though, the reduction of the porphyrin moiety via electron transfer to the a2u+σS orbital is the first and rate-limiting event [21]. A detailed description of the orbitals important for the reactivity of heme thiolate enzymes can be found elsewhere [21,54]. The study of the 1S2R reaction pathway described above [21] clearly indicates that the effect of the secondary coordination sphere and proximal helix on the electron donating properties of the proximal thiolate is anisotropic in nature and hence may interfere differently with different chiral transition states. This observation provides a motivation for asking a rather unconventional question: Is it possible that the proximal environment may also have a significant effect on the enantiospecificity of a heme-thiolate-catalyzed epoxidation reaction? Herein, we address this question by comparing the combined effect of NH–S hydrogen bonding and the proximal helix’s dipole moment on the rate-limiting barrier for the 1R2S doublet pathway of the CPO-catalyzed epoxidation of CBMS with the result for the 1S2R pathway obtained previously. CPO-I converts CBMS into the epoxide with 96% enantiomeric excess of the 1S2R over the 1R2S product [52], which corresponds to a free energy difference of ~2.0 kcal/mol in favor of the 1S2R enantiomer at 300 K. This underscores that significant levels of enantiospecificity can result from small free energy differences, making it difficult to identify the underlying causes of enantiospecificity with certainty. It is generally assumed that the distal binding site controls enantiospecificity by providing steric or specific interactions. In this paper, we shall show that in the case of CPO-catalyzed epoxidation of CBMS, the effect of the proximal pocket on the enantiospecificity is of the same order of magnitude (~1.0 kcal/mol) as that of the distal pocket. Thus, it is possible that the effect of the proximal environment on the enantiospecificity of a heme-thiolate catalyzed epoxidation reaction may be significant. The underlying reason for this is that the secondary coordination sphere of the proximal thiolate and the dipole of the proximal helix determine the preferred electron transfer mechanism for the rate-limiting event. We found that in the case of CPO-I catalyzed epoxidation of CBMS, electron transfer to different oxyferryl π* orbitals results in the energy splitting of the transition states leading to the 1R2S and 1S2R product epoxides. This difference was not observed when the heme thiolate’s secondary coordination sphere and the proximal helix were included in the model. In this case, the proximal NH–S bonds and helix dipole reduce the electron donating properties of the proximal thiolate. This reduction in electron donation, in effect an electron withdrawal, is greater for π than for σ orbitals and results in the electron transfer being to the a2u+σS rather than to a π* orbital. Due to this change in the electron transfer mechanism, the 1R2S and 1S2R prechiral transition states become degenerate. We conclude that for heme-thiolate enzymes, the effect of the secondary coordination sphere and of the proximal helix on the electron-donating properties of the proximal thiolate ligand has to be included to provide quantitatively reliable energy differences for prechiral transition states. Our work employs model systems, described in the Computational Methods section, chosen to isolate the influence of the heme thiolate’s secondary coordination sphere and the proximal helix on CPO-I catalyzed epoxidation of CBMS. Thus, the model systems do not explicitly represent the distal binding pocket. We note that a large amount of conformational sampling is needed to quantify the influence of protein-CBMS steric interactions directly within meaningful margins of error. This is because steric interactions can fluctuate substantially relative to the small energy differences causing the enantiomeric excess. Our model-system approach allows us to focus on the influence of the proximal region without the conformational sampling issue. 2. Computational Details 2.1. Methods Unrestricted DFT calculations of the doublet spin surfaces were carried out without symmetry restrictions using the B3LYP [55,56] hybrid density functional (UB3LYP) with the LANL2DZ effective core potential (ECP) double-ζ basis set for Fe [57] and the 6-31G* basis set for H, C, N, O [58] and S [59] atoms (basis set B0) using NWChem 6.1 software [60]. The stability of the density functions obtained was checked with the B0 basis set using Gaussian-09 [61]. The LANL2TZ + ECP triple-ζ basis set for Fe [62] and the 6-311++G** basis set for H, C, N, O [63] and S [64] atoms (basis set B1) were used for energy refinement. Frequency calculations with basis set B0 were used to obtain zero point energy (ZPE) corrections, which are included in the energies of all stationary points given. Natural population analysis [65] (NPA) of spin/charge densities was carried out using NBO 6.0 software [66]. 2.2. CPO-I Models CPO-I was modeled as an R−-Fe4+O2−(N4C20H12)− species in which the heme moiety lacks the vinyl and propionate side chains. This 43-atom model of CPO-I, CPO-I-A, employed R− = (SCH3)− (Figure 2) and constitutes the thiolate model, which lacks the axial sulfur’s secondary coordination sphere and the proximal helix. The simulations of CPO-I-A were performed without geometric constraints. All transition state structures have one imaginary normal mode frequency, while all stable structures have only positive, real frequencies. The CPO-I model used to study the proximal pocket effect (CPO-I-B) employed the R− = CH3–NH–Asn–Leu–Ala–Pro–Cys−–Pro–Ala–CO–CH3 peptide fragment (Figure 2B), which includes the proximal amino acid residues having Cα atoms within 8 Å of the proximal sulfur. Model B consists of 141 atoms and includes the secondary coordination sphere of the axial sulfur provided by the A31:NH-C29:S and L32:NH-C29:S hydrogen bonds and the immediate steric environment of the axial sulfur, as well as the proximal α-helix. The proximal α-helix was included so that the dipole moment deriving from the C29:CO–N33:NH and P30:CO–A34:NH backbone hydrogen bonds is represented in the model. The full proximal α-helix of CPO has 3 more hydrogen bonds formed by the backbone of residues 35 to 38. These residues were not included due to computer resource limitations. Thus, the effect of the proximal helix dipole in our Model System B represents a lower bound to the actual effect. The proximal peptide fragment of CPO-I-B was constrained to maintain backbone and side-chain hydrogen bonds, backbone ϕ,ψ dihedrals and the orientation of the proximal helix relative to the heme moiety as in the crystal structure of CPO (PDB Code 1CPO) [24]. The stationary points of the CBMS/CPO-I-B reactant and transition state complexes have extra normal modes with imaginary frequencies caused by these constraints. The NH–S hydrogen bonds were not constrained. The details of the constraints applied are given in the Supplementary Materials (SM) (Data S1). 2.3. Initial Structures for 1R2S and 1S2R CPO-I/CBMS Transition State Complexes For both Models A and B, the C4 rotational symmetry of the porphyrin moiety results in four CPO-I/CBMS transition state complex conformations that are nearly degenerate in energy. In our previous work, we developed a molecular mechanics (MM) parameterization of the transition state using the quantum mechanics to molecular mechanics (Q2MM) method [67] and used it to dock the low spin CPO-I/CBMS transition state complex into the apoenzyme scaffold of CPO [18]. It was found that three of the four possible 1S2R CPO-I/CBMS transition state conformations were unrealistic because of significant steric overlap between CBMS and the distal pocket residues [18]. In this work, we used the same method to carry out the docking of four possible 1R2S CPO-I/CBMS low spin prechiral transition state conformations and found that, as for the 1S2R pathway, only one conformation is allowed by the steric restrictions of the distal pocket (Figure S1). The Cartesian coordinates of the 1R2S and 1S2R docked structures are given in the SM (Data S2A). The pertinent portions of these structures were used to initiate the fully quantum mechanical (QM) transition state searches on Model Systems A and B. The Cartesian coordinates of all stationary points are given in the SM (Data S2B). 3. Results We have shown previously that, on the doublet surface, the epoxidation reaction takes place in two steps. The first and rate-limiting step is the formation of the Cβ–O bond, which is followed by the facile formation of the Cβ–O bond and ring closure [18]. The UB3LYP/B1//B0 potential energy surfaces (PES) connecting the reactant state (R) and the rate-limiting transition state (TS) leading to the formation of a Cβ–O bond were calculated for the doublet 1R2S pathway of CPO-I-A and CPO-I-B catalyzed epoxidation of CBMS. The transition states were found by scanning the Cβ–O distance followed by saddle-point optimization. The reactant states were found by optimizing along the transition mode of the TS in the direction of increasing Cβ–O distance. The results calculated here for the 1R2S pathway were compared to those for 1S2R from our previous work [21]. For both Models A and B of CPO-I, the reactant states leading to the 1R2S and 1S2R products are CPO-I + CBMS bound complexes degenerate in energy and characterized by the weak interaction of the methyl group of CBMS and the oxygen of CPO-I; the transition states are the results of the oxyferryl radical attack on the C=C double bond of the substrate (Figure 3 and Figure 4). The natural spin densities and charges (Table 1) show that the 1R2S and 1S2R electron transfer mechanisms are identical. Furthermore, the 1R2S spin densities and charges confirm our previous result for the 1S2R reaction [21], namely that for the CPO-I-B model the rate-limiting kinetic event is an electron transfer from the C=C π bond to the a2u+σS thiolate-porphyrin orbital, while for the bare-thiolate CPO-I-A model, the electron is transferred to one of the oxyferryl π* orbitals. The important bond lengths of the stationary points are given in Table 1. The PESs and optimized structures for CPO-I-A and CPO-I-B catalyzed epoxidation of CBMS are shown in Figure 3 and Figure 4, respectively. The PESs include ZPE corrections calculated at the UB3LYP/B0 level. For the bare-thiolate model, CPO-I-A, the calculations show that the 1R2S reaction pathway is favored by 1 kcal/mol (Figure 3). For CPO-I-B, the calculations show that the proximal pocket lowers the rate-limiting TS barrier by 3.7 kcal/mol on the doublet spin 1R2S surface, as compared to 4.6 kcal/mol calculated previously [21] for the doublet spin 1S2R surface. The influence of the proximal pocket thus removes the relative favorability that the 1R2S reaction pathway TS has in the absence of the apoprotein (Figure 4). 4. Discussion The experimentally-measured enantiomeric excesses of CPO-catalyzed epoxidations of various alkenes vary in the range 66%–97% [52]. This translates into free energy differences of ~0–3 kcal/mol at room temperature, which is a small number compared to the thermal fluctuations of a typical ~20,000-atom molecular model of the solvated CPO/substrate complex in an NPT simulation (number of atoms, pressure, and temperature constant), leading to the difficulty of accurate sampling. A comprehensive study of a stereoselective reaction should produce an accurate free energy potential surface that includes appropriate conformational averaging and connects substrate/enzyme, transition-state/enzyme and product/enzyme complexes. With current computational means, a molecular dynamics (MD) trajectory of ~100 ns–~1 μs is achievable if a classical MM force field is employed, allowing the testing of the binding hypothesis, i.e., the notion that the favorability of substrate binding conformations parallels the enantiomeric excess of the epoxide product [68]. Previously, we carried out an extensive MM simulation to distinguish binding potential wells of the CPO/CBMS complex from which the reaction to 1S2R and 1R2S epoxide products may occur; the flatness of the calculated free energy landscape rules out the binding hypothesis [17]. This is because CBMS floats relatively freely in the active site, as it is oriented by nonpolar interactions. It follows that the reaction is under kinetic control, and QM/MM transition state calculations are needed to identify the source of stereoselectivity. This presents a computational challenge, foremost because a substantial amount of conformational sampling is likely to be required to adequately reproduce the net influence of the distal region. For this purpose, a Q2MM model might be sufficient. However, as the present study shows, for heme-thiolate enzymes, an explicitly QM level of theory is required to account for the effect of the proximal region on the electron transfer mechanism. The calculations presented here for CPO-I-A catalyzed epoxidation of CBMS on the doublet spin surface show that for the bare-thiolate model without the influence of the proximal pocket, CPO-I favors the formation of the 1R2S over the 1S2R enantiomer by 1 kcal/mol. Since the bound complexes were found to be degenerate in energy, the transition states are the key to understanding this phenomenon. On the 1R2S pathway, the C=C π bond is attacked by the oxyferryl πxz* electron, while on the 1S2R pathway, the attack is by the πyz* electron (Figure 5). This stereo difference is the result of the restrictions imposed by the distal pocket on the possible orientation of the substrate in the 1R2S and 1S2R transition states with respect to the πxz* and πyz* orbitals (Figure 5). The electron transfer to the different π* orbitals results in the energy splitting of the 1R2S and 1S2R transition states because of uneven π-donation by the proximal thiolate ligand to these orbitals. When the proximal NH-S bonds and helix dipole are included by means of the CPO-I-B model, there is a larger withdrawing effect on π-donation than on σ-donation by the proximal thiolate. As a result, the electron from the C=C π bond is transferred to the a2u+σS thiolate-porphyrin orbital, rather than to a π* orbital, for both the 1R2S and 1S2R channels. Hence, one would expect the chiral transition states to be degenerate, as we have found with the calculations presented. 5. Conclusions For heme-thiolate enzymes, contrary to the general understanding, the proximal region can have nearly as great an effect on the enantiospecificity of the epoxidation reaction as the distal region. This is possible because the proximal pocket determines the preferred electron transfer mechanism, which is either to the a2u+σS orbital or to one of the π* orbitals. The latter case allows an interplay of the substrate orientation in the distal pocket with the asymmetry of π-donation by the proximal thiolate. In the case of CPO, the calculated 1 kcal/mol difference in favor of the 1R2S enantiomer for the bare-thiolate (CPO-I-A) catalyzed epoxidation of CBMS on the low spin surface results in a difference of ~3 kcal/mol with respect to the experimental value of 1.96 kcal/mol in favor of the 1S2R epoxide product, determined from the enantiomeric excess [52]. The experimental value reflects the total effect of the enzyme in favor of the 1S2R channel. The results for CPO-I-B catalyzed epoxidation of CBMS show that ~1 kcal/mol out of the net ~3 kcal/mol difference is due to the combined influence of the proximal thiolate’s secondary coordination sphere (especially NH–S hydrogen bonding) and the proximal helix dipole. The inclusion of these factors reproduces the correct electron transfer mechanism. The remaining ~2 kcal/mol presumably reflects the net influence of the distal pocket. Since the expected effect of the distal and proximal pockets of CPO are of the same order of magnitude, we conclude that proper modeling of the proximal pocket is necessary to correctly calculate the enantiospecificities of heme-thiolate-catalyzed epoxidations. Acknowledgments The authors would like to acknowledge the Instructional and Research Computing Center (IRCC) at Florida International University for providing computing resources that have contributed to the research results reported within this paper. Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1297/s1. Click here for additional data file. Author Contributions Alexander N. Morozov contributed to simulation design and analysis, carried out the simulations, and drafted the initial manuscript. David C. Chatfield contributed to simulation design and analysis and to manuscript preparation, and he was responsible for manuscript submission. Conflicts of Interest The authors declare no conflict of interest. Figure 1 Proximal pockets of CPO and P450cam. Figure 2 Bare-thiolate, CPO-I-A (a); and CPO-like, CPO-I-B (b) proximal pocket models of CPO-I. Figure 3 The UB3LYP/B1//B0 potential energy surfaces (in kcal/mol) connecting the reactant states R and the rate-limiting transition states TS leading to the formation of a Cβ–O bond on the doublet potential energy surfaces for epoxidation of cis-β-methylstyrene (CBMS) by CPO-I-A to give 1R2S and 1S2R products. Figure 4 The UB3LYP/B1//B0 potential energy surfaces (in kcal/mol) connecting the reactant states R and the rate-limiting transition states TS leading to the formation of a Cβ–O bond on the doublet potential energy surfaces for 1R2S and 1S2R epoxidation of CBMS by CPO-I-B. Figure 5 Oxyferryl π* attack on C=C bond for model CPO-I-A: (a) LUMO in β manifold of 1R2S TS; (b) LUMO in β manifold of 1S2R TS. ijms-17-01297-t001_Table 1Table 1 Natural group spin densities/charges and bond lengths (Å) of the optimized structures on the doublet spin potential energy surfaces (PES). Natural Spin Densities/Natural Atomic Charges Bond Lengths S–R Por Fe O CβH CαH R1+ R2+ S–Fe Fe–O O–Cβ 1S2R R A −0.75/−0.05 −0.30/−0.49 1.10/0.91 0.95/−0.37 0.00/0.06 0.00/−0.03 0.00/−0.05 0.00/0.02 2.619 1.623 – B −0.60/−0.27 −0.50/−0.32 1.15/0.95 0.95/−0.36 0.00/0.06 0.00/−0.03 0.00/−0.05 0.00/0.02 2.776 1.619 – TS A −0.70/−0.08 −0.30/−0.58 0.95/0.90 0.75/−0.44 −0.10/0.15 0.30/0.04 0.10/−0.02 0.00/0.03 2.554 1.705 1.985 B −0.30/−0.40 −0.25/−0.51 1.40/0.96 0.50/−0.42 −0.05/0.18 −0.20/0.08 −0.10/0.07 0.00/0.04 2.570 1.658 2.099 1R2S R A −0.76/−0.05 −0.30/−0.49 1.10/0.91 0.96/−0.37 0.00/0.06 0.00/−0.03 0.00/−0.05 0.00/0.02 2.624 1.623 – B −0.59/−0.27 −0.48/−0.32 1.15/0.95 0.95/−0.36 0.00/0.06 0.00/−0.03 0.00/−0.05 0.00/0.02 2.776 1.619 – TS A −0.65/−0.08 −0.28/−0.59 0.91/0.89 0.76/−0.43 −0.11/0.15 0.27/0.05 0.10/−0.01 0.00/0.03 2.512 1.702 1.996 B −0.30/−0.41 −0.26/−0.50 1.38/0.96 0.52/−0.42 −0.06/0.18 −0.17/0.08 −0.11/0.07 0.00/0.04 2.576 1.658 2.116 S-R: proximal sulfur together with rest of R— moiety (SCH3 for model A; sulfur with proximal helix for model B); Por: porphyrin; R1+: benzylic group of CBMS; R2+: methyl group of CBMS. ==== Refs References 1. Poulos T.L. Heme enzyme structure and function Chem. Rev. 2014 114 3919 3962 10.1021/cr400415k 24400737 2. Shaw P.D. Hager L.P. Biological chlorination. IV. Peroxidative nature of enzymatic chlorination J. Am. Chem. Soc. 1959 81 1011 1012 10.1021/ja01513a069 3. Shaw P.D. Hager L.P. Biological Chlorination J. Biol. Chem. 1961 236 1626 1630 4. Hager L.P. Morris D.R. Brown F.S. Eberwein H. Chloroperoxidase II: Utilization of halogen ions J. Biol. Chem. 1966 241 1769 1777 5945851 5. Hofrichter M. Ullrich R. Heme-thiolate haloperoxidases: Versatile biocatalysts with biotechnological and environmental significance Appl. Microbiol. Biotechnol. 2006 71 276 288 10.1007/s00253-006-0417-3 16628447 6. Osborne R.L. Coggins M.K. Terner J. Dawson J.H. Caldariomyces fumago chloroperoxidase catalyzes the oxidative dehalogenation of chlorophenols by a mechanism involving two one-electron steps J. Am. Chem. Soc. 2007 129 14838 14839 10.1021/ja0746969 17990879 7. Gruia F. Ionascu D. Kubo M. Ye X. Dawson J. Osborne R.L. Sligar S.G. Denisov I. Das A. Poulos T.L. Low-frequency dynamics of caldariomyces fumago chloroperoxidase probed by femtosecond coherence spectroscopy Biochem. 2008 47 5156 5167 10.1021/bi7025485 18407660 8. Tzialla A.A. Kalogeris E. Gournis D. Sanakis Y. Stamatis H. Enhanced catalytic performance and stability of chloroperoxidase from Caldariomyces fumago in surfactant free ternary water-organic solvent systems J. Mol. Catal. B Enzym. 2008 51 24 35 10.1016/j.molcatb.2007.10.006 9. Hager L.P. A lifetime of playing with enzymes J. Biol. Chem. 2010 285 14852 14860 10.1074/jbc.X110.121905 20215109 10. Ayala M. Hernandez-Lopez E.L. Perezgasga L. Vazquez-Duhalt R. Reduced coke formation and aromaticity due to chloroperoxidase-catalyzed transformation of asphaltenes from Maya crude oil Fuel 2012 92 245 249 10.1016/j.fuel.2011.06.067 11. Zhang R. He Q.H. Chatfield D. Wang X.T. Paramagnetic nuclear magnetic resonance relaxation and molecular mechanics studies of the chloroperoxidase-indole complex: Insights into the mechanism of chloroperoxidase-catalyzed regioselective oxidation of indole Biochemistry 2013 52 3688 3701 10.1021/bi4002437 23634952 12. Liu Y. Wang Y. Jiang Y. Hu M. Li S. Zhai Q. Biocatalytic synthesis of C3 chiral building blocks by chloroperoxidase-catalyzed enantioselective halo-hydroxylation and epoxidation in the presence of ionic liquids Biotechnol. Prog. 2015 31 724 729 10.1002/btpr.2076 25826799 13. Muñoz-Guerrero F.A. Aguila S. Vazquez-Duhalt R. Campos C.H. Torres C.C. Alderete J.B. Substrate ionization energy influences the epoxidation of m-substituted styrenes catalyzed by chloroperoxidase from Caldariomyces fumago Catal. Commun. 2016 77 52 54 10.1016/j.catcom.2016.01.017 14. Zhang R. He Q. Huang Y. Wang X. Spectroscopic and QM/MM investigations of chloroperoxidase catalyzed degradation of orange G Arch. Biochem. Biophys. 2016 596 1 9 10.1016/j.abb.2016.02.026 26926259 15. Chen H. Hirao H. Derat E. Schlichting I. Shaik S. Quantum mechanical/molecular mechanical study on the mechanisms of Compound I formation in the catalytic cycle of chloroperoxidase: An overview on heme enzymes J. Phys. Chem. B 2008 112 9490 9500 10.1021/jp803010f 18597525 16. Lai W. Chen H. Cho K.B. Shaik S. Effects of substrate, protein environment, and proximal ligand mutation on Compound I and compound 0 of chloroperoxidase J. Phys. Chem. A 2009 113 11763 11771 10.1021/jp902898s 19572690 17. Morozov A.N. D’Cunha C. Alvarez C.A. Chatfield D.C. Enantiospecificity of chloroperoxidase-catalyzed epoxidation: Biased molecular dynamics study of a cis-β-methylstyrene/chloroperoxidase-Compound I complex Biophys. J. 2011 100 1066 1075 10.1016/j.bpj.2010.12.3729 21320452 18. Morozov A.N. Chatfield D.C. Chloroperoxidase-catalyzed epoxidation of cis -β-methylstyrene: Distal pocket flexibility tunes catalytic reactivity J. Phys. Chem. B 2012 116 12905 12914 10.1021/jp302763h 23020548 19. D’Cunha C. Morozov A.N. Chatfield D.C. Theoretical study of HOCl-catalyzed keto-enol tautomerization of cyclopentanedione in an explicit water environment J. Phys. Chem. A 2013 117 8437 8448 10.1021/jp401409y 23902476 20. Pardillo A.D. Morozov A.N. Chatfield D.C. Proximal pocket hydrogen bonds significantly influence the mechanism of chloroperoxidase Compound I formation J. Phys. Chem. B 2015 119 12590 12602 10.1021/acs.jpcb.5b06324 26339752 21. Morozov A.N. Pardillo A.D. Chatfield D.C. Chloroperoxidase-catalyzed epoxidation of cis -β-methylstyrene: NH–S hydrogen bonds and proximal helix dipole change the catalytic mechanism and significantly lower the reaction barrier J. Phys. Chem. B 2015 119 14350 14363 10.1021/acs.jpcb.5b06731 26452587 22. Rutter R. Hager L.P. The detection of two electron paramagnetic resonance radical signals associated with chloroperoxidase Compound I J. Biol. Chem. 1982 257 7958 7961 6282864 23. Kim S.H. Perera R. Hager L.P. Dawson J.H. Hoffman B.M. Rapid freeze-quench ENDOR study of chloroperoxidase Compound I: The site of the radical J. Am. Chem. Soc. 2006 128 5598 5599 10.1021/ja060776l 16637602 24. Sundaramoorthy M. Terner J. Poulos T.L. The crystal structure of chloroperoxidase: A heme peroxidase-cytochrome P450 functional hybrid Structure 1995 3 1367 1378 10.1016/S0969-2126(01)00274-X 8747463 25. Sundaramoorthy M. Kishi K. Gold M.H. Poulos T.L. The crystal-structure of manganese peroxidase from phanerochaete-chrysosporium at 2.06-angstrom resolution J. Biol. Chem. 1994 269 32759 32767 7806497 26. Poulos T.L. Finzel B.C. Howard A.J. High-resolution crystal structure of cytochrome P450cam J. Mol. Biol. 1987 195 687 700 10.1016/0022-2836(87)90190-2 3656428 27. Dawson J.H. Probing structure-function relations in heme-containing oxygenases and peroxidases Science 1988 240 433 439 10.1126/science.3358128 3358128 28. McQuarters A.B. Wolf M.W. Hunt A.P. Lehnert N. 1958–2014: After 56 years of research, cytochrome P450 reactivity is finally explained Angew. Chem. Int. Ed. 2014 53 4750 4752 10.1002/anie.201402404 24711286 29. Groves J.T. Enzymatic C–H bond activation: Using push to get pull Nat. Chem. 2014 6 89 91 10.1038/nchem.1855 24451580 30. Harris D.L. Loew G.H. Theoretical investigation of the proton assisted pathway to formation of cytochrome P450 Compound I J. Am. Chem. Soc. 1998 120 8941 8948 10.1021/ja981059x 31. Kumar D. Karamzadeh B. Sastry G.N. de Visser S.P. What factors influence the rate constant of substrate epoxidation by Compound I of cytochrome P450 and analogous iron(IV)-oxo oxidants? J. Am. Chem. Soc. 2010 132 7656 7667 10.1021/ja9106176 20481499 32. Rydberg P. Sigfridsson E. Ryde U. On the role of the axial ligand in heme proteins: A theoretical strudy J. Biol. Inorg. Chem. 2004 9 203 223 10.1007/s00775-003-0515-y 14727167 33. Sono M. Roach M.P. Coulter E.D. Dawson J.H. Heme-containing oxygenases Chem. Rev. 1996 96 2841 2888 10.1021/cr9500500 11848843 34. Poulos T.L. The role of the proximal ligand in heme enzymes J. Biol. Inorg. Chem. 1996 1 356 359 10.1007/s007750050064 35. Ueno T. Nishikawa N. Moriyama S. Adachi S. Lee K. Okamura T. Ueyama N. Nakamura A. Role of the invariant peptide fragment forming nhs hydrogen bonds in the active site of cytochrome P450 and chloroperoxidase: Synthesis and properties of cys-containing peptide Fe(III) and Ga(III) (octaethylporphinato) complexes as models Inorg. Chem. 1999 38 1199 1210 10.1021/ic980710u 11670903 36. Yoshioka S. Tosha T. Takahashi S. Ishimori K. Hori H. Morishima I. Roles of the proximal hydrogen bonding network in cytochrome P450cam-catalyzed oxygenation J. Am. Chem. Soc. 2002 124 14571 14579 10.1021/ja0265409 12465966 37. Yoshioka S. Takahashi S. Ishimori K. Morishima I. Roles of the axial push effect in cytochrome P450cam studied with the site-directed mutagenesis at the heme proximal site J. Inorg. Biochem. 2000 81 141 151 10.1016/S0162-0134(00)00097-0 11051559 38. Dey A. Jiang Y. Ortiz de Montellano P. Hodgson K.O. Hedman B. Solomon E.I. S K-edge XAS and DFT calculations on cytochrome P450: Covalent and ionic contributions to the cysteine-Fe bond and their contribution to reactivity J. Am. Chem. Soc. 2009 131 7869 7878 10.1021/ja901868q 19438234 39. Galinato M.G.I. Spolitak T. Ballou D.P. Lehnert N. Elucidating the role of the proximal cysteine hydrogen-bonding network in ferric cytochrome P450cam and corresponding mutants using magnetic circular dichroism spectroscopy Biochemistry 2011 50 1053 1069 10.1021/bi101911y 21158478 40. De Visser S.P. Ogliaro F. Sharma P.K. Shaik S. Hydrogen bonding modulates the selectivity of enzymatic oxidation by P450: Chameleon oxidant behavior by Compound I Angew. Chem. Int. Ed. 2002 124 11809 11826 41. De Visser S.P. Ogliaro F. Sharma P.K. Shaik S. What factors affect the regioselectivity of oxidation by cytochrome P450? A DFT study of allylic hydroxylation and double bond epoxidation in a model reaction J. Am. Chem. Soc. 2002 130 12961 12974 10.1021/ja8010995 18774806 42. De Visser S.P. Tan L.S. Is the bound substrate in nitric oxide synthase protonated or neutral and what is the active oxidant that performs substrate hydroxylation? J. Am. Chem. Soc. 2008 130 12961 12974 10.1021/ja8010995 18774806 43. Latifi R. Sainna M.A. Rybak-Akimova E.V. de Visser S.P. Does hydrogen-bonding donation to manganese(IV)-oxo and iron(IV)-oxo oxidants affect the oxygen-atom transfer ability? a computational study Chem. Eur. J. 2013 19 4058 4068 10.1002/chem.201202811 23362213 44. Hol W.G.J. van Duijnen P.T. Berendsen H.J.C. The α -Helix dipole and the properties of proteins Nature 1978 273 443 446 10.1038/273443a0 661956 45. Park C. Goddard W.A. Stabilization of α -helices by dipole-dipole interactions within α -helices J. Phys. Chem. B 2000 104 7784 7789 10.1021/jp0001743 46. Hol W.G.J. Halie L.M. Sander C. Dipoles of the α -helix and β -sheet: Their role in protein folding Nature 1981 294 532 536 10.1038/294532a0 7312043 47. Morozov A.N. Shiu Y.J. Liang C.T. Tsai M.Y. Lin S.H. Nonadditive interactions in protein folding: The zipper model of cytochrome C J. Biol. Phys. 2007 33 255 270 10.1007/s10867-008-9062-7 19669517 48. Tsai M.Y. Morozov A.N. Chu K.Y. Lin S.H. Molecular dynamics insight into the role of tertiary (foldon) interactions on unfolding in cytochrome C Chem. Phys. Lett. 2009 475 111 115 10.1016/j.cplett.2009.05.027 49. Morozov A.N. Roach J.P. Kotzer M. Chatfield D.C. A possible mechanism for redox control of human neuroglobin activity J. Chem. Inf. Model. 2014 54 1997 2003 10.1021/ci5002108 24855999 50. Lockhart D.J. Kim P.S. Internal stark effect measurement of the electric field at the amino terminus of an α -helix Science 1992 257 947 951 10.1126/science.1502559 1502559 51. Ueno T. Kousumi Y. Yoshizawa-Kumagaye K. Nakajima K. Ueyama N. Okamura T. Nakamura A. Role of α -helix conformation cooperating with NH···S hydrogen bond in the active site of cytochrome P-450 and chloroperoxidase: Synthesis and Properties of [MIII (OEP)(Cys-Helical Peptide)] (M = Fe and Ga) J. Am. Chem. Soc. 1998 120 12264 12273 10.1021/ja980016d 52. Allain E.J. Hager L.P. Deng L. Jacobsen E.N. Highly enantioselective epoxidation of disubstituted alkenes with hydrogen peroxide catalyzed by chloroperoxidase J. Am. Chem. Soc. 1993 115 4415 4416 10.1021/ja00063a091 53. Hu S. Hager L.P. Asymmetric epoxidation of functionalized cis -olefins catalyzed by chloroperoxidase Tetrahedron Lett. 1999 40 1641 1644 10.1016/S0040-4039(99)00056-8 54. Shaik S. Cohen S. Wang Y. Chen H. Kumar D. Thiel W. P450 Enzymes: Their structure, reactivity, and selectivity modeled by QM/MM calculations Chem. Rev. 2010 110 949 1017 10.1021/cr900121s 19813749 55. Becke A.D. Density-functional thermochemistry. III. The role of exact exchange J. Chem. Phys. 1993 98 5648 5652 10.1063/1.464913 56. Lee C. Yang W. Parr R.G. Development of the colle-salvetti correlation-energy formula into a functional of the electron density Phys. Rev. B 1988 37 785 789 10.1103/PhysRevB.37.785 57. Hay P.J. Wadt W.R. Ab initio effective core potentials for molecular calculations. Potentials for K to Au including the outermost core orbitals J. Chem. Phys. 1985 82 299 310 10.1063/1.448975 58. Hariharan P.C. Pople J.A. The influence of polarization functions on molecular orbital hydrogenation energies Theor. Chim. Acta 1973 28 213 222 10.1007/BF00533485 59. Francl M.M. Pietro W.J. Hehre W.J. Binkley J.S. Gordon M.S. DeFrees D.J. Pople J.A. Self-consistent molecular orbital methods. XXIII. A polarization-type basis set for second-row elements J. Chem. Phys. 1982 77 3654 3665 10.1063/1.444267 60. Valiev M. Bylaska E.J. Govind N. Kowalski K. Straatsma T.P. van Dam H.J.J. Wang D. Nieplocha J. Apra E. Windus T.L. NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations Comput. Phys. Commun. 2010 181 1477 1489 10.1016/j.cpc.2010.04.018 61. Frisch M.J. Trucks G.W. Schlegel H.B. Scuseria G.E. Robb M.A. Cheeseman J.R. Scalmani G. Barone V. Mennucci B. Petersson G.A. Gaussian 09, Revision A.1 Gaussian, Inc. Wallingford, UK 2009 62. Hay P.J. Wadt W.R. Ab initio effective core potentials for molecular calculations. Potentials for the transition metal atoms Sc to Hg J. Chem. Phys. 1985 82 270 283 10.1063/1.448799 63. Krishnan R. Binkley J.S. Seeger R. Pople J.A. Self‐consistent molecular orbital methods. XX. A basis set for correlated wave functions J. Chem. Phys. 1980 72 650 654 10.1063/1.438955 64. McLean A.D. Chandler G.S. Contracted gaussian basis sets for molecular calculations. I. Second row atoms, Z = 11–18 J. Chem. Phys. 1980 72 5639 5648 10.1063/1.438980 65. Foster J.P. Weinhold F. Natural Hybrid Orbitals J. Am. Chem. Soc. 1980 102 7211 7218 10.1021/ja00544a007 66. Glendening E.D. Landis C.R. Weinhold F. NBO 6.0: Natural Bond Orbital Analysis Program J. Comput. Chem. 2013 34 1429 1437 10.1002/jcc.23266 23483590 67. Norrby P.O. Selectivity in asymmetric synthesis from QM-guided molecular mechanics J. Mol. Struct. Theochem. 2000 506 9 16 10.1016/S0166-1280(00)00398-5 68. Fruetel J.A. Collins J.R. Camper D.L. Loew G.H. Ortiz de Montellano P.R. Calculated and experimental absolute stereochemistry of the styrene and β-methylstyrene epoxides formed by cytochrome P450cam J. Am. Chem. Soc. 1992 114 6987 6993 10.1021/ja00044a006

PMC005xxxxxx/PMC5000695.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081298ijms-17-01298ArticleThe Antitumor Effect of Metformin Is Mediated by miR-26a in Breast Cancer Cabello Paula 1Pineda Begoña 1Tormo Eduardo 1Lluch Ana 12Eroles Pilar 1*Taguchi Y-h. Academic Editor1 Biomedical Research Institute INCLIVA, 46010 Valencia, Spain; [email protected] (P.C.); [email protected] (B.P.); [email protected] (E.T.); [email protected] (A.L.)2 Oncology and Hematology Department, Hospital Clinico Universitario, 46010 Valencia, Spain* Correspondence: [email protected]; Tel.: +34-96-3864100 (ext. 51920)10 8 2016 8 2016 17 8 129820 5 2016 04 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Metformin, a drug approved for diabetes type II treatment, has been associated with a reduction in the incidence of breast cancer and metastasis and increased survival in diabetic breast cancer patients. High levels of miR-26a expression have been proposed as one of the possible mechanisms for this effect; likewise, this miRNA has also been associated with survival/apoptosis processes in breast cancer. Our aim was to evaluate if miR-26a and some of its targets could mediate the effect of metformin in breast cancer. The viability of MDA-MB-231, MDA-MB-468, and MCF-7 breast cancer cell lines was evaluated with an MTT assay after ectopic overexpression and/or downregulation of miR-26a. Similarly, the expression levels of the miR-26a targets CASP3, CCNE2, ABL2, APAF1, XIAP, BCL-2, PTEN, p53, E2F3, CDC25A, BCL2L1, MCL-1, EZH2, and MTDH were assessed by quantitative polymerase chain reaction (PCR). The effect of metformin treatment on breast cancer cell viability and miR-26a, BCL-2, PTEN, MCL-1, EZH2, and MTDH modulation were evaluated. Wound healing experiments were performed to analyze the effect of miR-26a and metformin treatment on cell migration. MiR-26a overexpression resulted in a reduction in cell viability that was partially recovered by inhibiting it. E2F3, MCL-1, EZH2, MTDH, and PTEN were downregulated by miR-26a and the PTEN (phosphatase and tensin homolog) protein was also reduced after miR-26a overexpression. Metformin treatment reduced breast cancer cell viability, increased miR-26a expression, and led to a reduction in BCL-2, EZH2, and PTEN expression. miR-26a inhibition partly prevents the metformin viability effect and the PTEN and EZH2 expression reduction. Our results indicate that metformin effectively reduces breast cancer cell viability and suggests that the effects of the drug are mediated by an increase in miR-26a expression and a reduction of its targets, PTEN and EHZ2 Thus, the use of metformin in breast cancer treatment constitutes a promising potential breast cancer therapy. miR-26ametforminbreast cancer ==== Body 1. Introduction Breast cancer is the most frequent cancer among women worldwide and the leading cause of death by cancer in women [1]. Breast cancer is a clinically, morphologically, and molecularly-heterogeneous disease [2,3]. In the year 2000, Perou et al. [4] classified it into five molecular subtypes according to its intrinsic genetic signature, however, immunohistochemical classification is still used in clinics. Treatment is based on the differential characteristics of breast cancer subtypes and is largely successful in human epidermal growth factor receptor 2 (HER2) and estrogen receptor (ER) positive (luminal) cancers (using anti-HER2+ and hormonal therapies, respectively). However, in triple negative breast cancer (TNBC; HER2, ER, and progesterone receptor negative), representing about 15%–20% of all breast cancer patients, there are no well-defined molecular targets. This subtype is related to an elevated recurrence rate, worse prognosis and a lower survival rate compared to other types of breast cancer [5,6] due, among other reasons, to the heterogeneity and aggressive nature of TNBC [6]. This peculiarity, and the fact that targeted treatments in HER2+ and ER+ breast cancers are not always beneficial, has led to the continued search for alternative therapies. However, extending the use of already approved drugs to new pathologies is a promising and rapid strategy for expanding the clinical therapeutic arsenal. Metformin (1,1-dimetilbiguanide hydrochloride) is a hypoglycemic oral biguanide drug that is prescribed and commercialized worldwide to treat diabetes type II. Epidemiological studies have revealed that oral use of metformin has a protective effect against tumors, reducing their incidence and improving the prognosis of cancer patients [7,8]. This drug can inhibit cancer cell proliferation, although its molecular mechanisms of action are not yet completely understood [9]. MicroRNAs (miRNAs or miRs) are small non-coding endogenous RNA molecules (approximately 22 nucleotides long) encoded in the introns of protein-coding genes and in the introns and exons of non-protein coding genes, which regulate gene expression at the post-transcriptional level [10,11]. MicroRNAs have already proven to be reliable biomarkers for predicting therapeutic response in several cancers [12,13,14,15,16] or as therapeutic tools in other types of cancer, not only by inhibiting them with drugs [17] but also, for example, by introducing them into liposomal vehicles for systemic distribution in lung cancer or hepatocellular carcinoma model mice [18,19]. In a previous study from our group [20], we observed significant changes in miR-26a levels when treating breast cancer cell lines with doxorubicin. This miRNA has also been previously studied in other types of cancer including renal and lung cancer. In 2014, Yang et al. [21] demonstrated that metformin inhibits renal cancer cell growth by inducing overexpression of the oncogenic microRNA, miR-26a, which has BCL-2 and PTEN among its targets. In this study we aimed to determine if miR-26a and/or some of its effector targets are implicated in the antitumor effect of metformin in TNBC and ER+ breast cancer, particularly in cell viability and/or migration. 2. Results 2.1. miR-26a Expression Modulates Cell Viability We studied the effect of exogenous miR-26a on cell viability by transfecting three cell lines (MDA-MB-231, MDA-MB-468, and MCF-7) with a miR-26a mimetic. Compared to the miRNA control (Cy3), miR-26a overexpression decreased cell viability in all three cell lines at all of the time-points we assayed (1, 4, and 7 days), and the difference was statistically significant at four days (36% (p = 0.0004), 31.11% (p = 0.0009), and 73.89% (p = 4.7 × 10−10) decrease in viability for MDA-MB-231, MDA-MB-468, and MCF-7 cells, respectively) and 7 days (75.47% (p = 9.66 × 10−5), and 92.32% (p = 3.78 × 10−6) decrease in viability for MDA-MB-468 and MCF-7 respectively) (Figure 1). However, rather than reducing cell viability, miR-26a inhibition increased viability, in some cases significantly (for example at 24 h in MDA-MB-468). This effect was partially reverted when the mimetic and miR-26a inhibitor were combined (Figure 2). These data support a role for miR-26a in breast cancer cell viability/apoptosis pathways acting as a tumor suppressor. 2.2. Effect of miR-26a on Cell Migration To evaluate the physiological impact of miR-26a regulation we also studied the migration capacity of MDA-MB-231, MDA-MB-468, and MCF-7 cell lines when miR-26a was overexpressed in a wound-healing assay. From 25 h, cell migration was higher in MDA-MB-231 and MDA-MB-468 cells transfected with the miR-26a mimetic compared to control cells, and the effect was more dramatic at 45 h, especially in MDA-MB-231 cells (Figure 3). 2.3. Evaluation of Potential miR-26a Targets We searched for and selected theoretical miR-26a target-genes in the miRTarBase and used DAVID Bioinformatics Resources and miRBase. From 950 potential targets detected in the miRTarBase, we selected 11 based of their relevance to cancer and cell viability and apoptosis processes: CASP3, CCNE2, ABL2, APAF1, XIAP, BCL-2, PTEN, p53, E2F3, CDC25A, and BCL2L1 ligand (Table 1). We differentiate between theoretical and proven targets, and note if the targets have been validated in cancer. We evaluated the expression of these 11 genes in the three breast cancer cell lines, after transfecting them with a miR-26a mimetic or inhibitor, by RT-qPCR. CASP3 expression significantly increased (p = 0.012) in the MDA-MB-468 cell line, and CCNE2 expression significantly increased in both the MDA-MB-231 (p = 0.002) and MDA-MB-468 (p = 0.0008) cell lines after transfection with the miR-26a inhibitor (Figure 4); in MDA-MB-231 cells ABL-2 was diminished after miR-26a transfection (p = 0.06) and increased after its inhibition (p = 0.027; Figure 4); APAF1 expression was higher in MDA-MB-231 (p = 0.00015) and MDA-MB-468 (p = 0.04) cells with miR-26a inhibition, and decreased in MCF-7 cells upon mimetic transfection (not significant); XIAP expression increased with miR-26a inhibition in the MDA-MB-231 and MCF-7 cell lines, significantly in the last (p = 0.00025); BCL-2 increased in the MDA-MB-231 (p = 0.018) and MDA-MB-468 (p = 0.023) cell lines in the presence of the miR-26a inhibitor; PTEN expression diminished upon miR-26a transfection in the MDA-MB-231 (p = 0.006) and MCF-7 (p = 0.002) cell lines, and significantly increased when miR-26a was inhibited in MDA-MB-468 cells (p = 2.45 × 10−5); miR-26a inhibition increased TP53 expression in the MDA-MB-468 cell line (p = 0.0009); E2F3 expression diminished in all three cell lines after miR-26a transfection and significantly increased in MCF-7 cells when miR-26a was inhibited (p = 0.0005); both CDC25A and BCL2L1 expression significantly increased after inhibiting miR-26a in the MDA-MB-468 cell line (p = 0.024 and p = 0.001, respectively) (Figure 4). For us, PTEN and E2F3 downregulation after miR-26a transfection was the most relevant finding because it suggests that this miRNA directly targets genes; this was especially interesting for PTEN as it has proven relevance in cancer processes. Thus, we focused on studying the implications of miR-26a in the modulation of this gene in the context of breast cancer. 2.4. Phosphatase and Tensin Homolog (PTEN) Regulation by miR-26a In order to validate PTEN as a miR-26a target in breast cancer, we transfected miR-26a mimetic into the MDA-MB-231 cell line and analyzed PTEN protein expression. Western blot analysis showed that levels of this protein significantly decreased (by 35.2% vs. control, p = 0.008) after miR-26a overexpression (Figure 5), in concordance with downregulation of this gene after miR-26a transfection (Figure 4). 2.5. Effect of Metformin on Breast Cancer Cells In order to study if metformin induces miR-26a overexpression in breast cancer, as previously described in renal [21] and pancreatic [22] cancer, we tested mRNA and protein expression in metformin-treated MDA-MB-231 cells. First we assessed the effect of metformin on cell viability using a MTT viability test at five different concentrations (1, 5, 10, 20, and 40 mM) at 24, 48, and 72 h of metformin treatment. At concentrations of 10 mM or higher, metformin decreased cell viability at 48 and 72 h (58% and 66% decrease, respectively) (Figure 6). We further analyzed the effect of metformin on the expression of miR-26a and its proposed targets, PTEN and BCL-2. MiR-26a was significantly increased after treating the cells with metformin (p = 0.012), however, both its potential targets showed a significant decrease in expression with the same treatment (p = 0.038 and p < 0.001 for PTEN and BCL-2, respectively) compared to non-treated cells (Figure 7A). PTEN protein levels were also lower after treatment with the drug (Figure 7B). These data correlate with our results from RT-qPCR. Finally, cell migration during metformin treatment was also checked using a wound healing assay to see if treatment with this drug reproduced the effects seen with miR-26a. Treating MDA-MB-231 cells with metformin increased migration, as shown by faster gap-closing at 24 and 30 h compared to non-treated cells (Figure 8). 2.6. Effect of Metformin through miR-26a on Breast Cancer Cells To clarify if the effect of metformin is mediated by upregulation of miR-26a we performed viability and expression analysis combining the miR-26a inhibitor and metformin. The cell viability reduction induced by metformin was partially rescued by miR-26a inhibitor (Figure 9A). We further analyzed the effect of metformin on the miR-26a proposed targets: MCL-1, EZH2, and MTDH. MCL-1 (myeloid cell leukemia 1) is a pro-survival member of the Bcl-2 (B-cell CLL/lymphoma 2) family [23], MTDH facilitates malignant transformation of normal immortal cloned rat embryo fibroblast cells [24], and EZH2 promotes anchorage-independent growth and invasion of immortalized human mammary epithelial cells [24]. MCL-1, EZH2, and MTDH showed significantly decreased expression (Figure 9B) after transfection with miR-26a mimetic (p = 0.02, p = 0.0004, and p = 0.0004, respectively) as was seen also for PTEN (Figure 4 and Figure 9B). Metformin significantly reduced EZH2 expression (p = 0.02) and the combination of miR-26a inhibitor with the drug significantly reversed the expression levels of EZH2 and PTEN (p = 0.019 and p = 0.05 respectively) (Figure 9C). 3. Discussion Evidence for the implication of miRNAs in cancer processes has been growing over the last decade, and many miRNAs have been described as being deregulated in cancer [25]. Consequently, investigation focusing on these small RNAs has increasingly focused on their therapeutic uses. Hence, several strategies have been designed based on miRNA inhibition or enhancement by ectopic expression [26]. We focused on miR-26a because it has been described to play an important role in several cancers, including hepatocellular carcinoma and lung and breast cancer [23,27,28,29]. Furthermore, some authors suggest that metformin, already used to treat diabetes, may modulate the expression of miR-26a. This miRNA is reported to be a tumor suppressor [23,24,27] whose increase leads to better outcomes in tamoxifen-treated breast cancer metastasis patients [30]. However, in lung cancer elevated miR-26a levels have been related to higher levels of tumor cell migration and invasion [28]. Low levels of miR-26a have been associated with TNBC [29] and with stimulating proliferation in ER+ breast cancer [31], and miR-26a expression levels have been associated with lymph node metastases in breast [29] and lung cancer [28]. Here we aimed to elucidate the role of miR-26a in modulating TNBC and ER+ breast cancer cell viability. Furthermore, we evaluated some of its targets, and finally, we assayed the effect of metformin on miR-26a and these targets. The use of drugs already on the market for new medical applications significantly streamlines their incorporation into the clinical armamentarium. Therefore, compounds already approved for certain treatments are being re-evaluated to discover their mechanisms of action and to search for possible new therapeutic applications. Metformin, a compound approved for treating type II diabetes, is being evaluated in cancer with the rationale that the incidence of breast cancer is decreased in diabetic patients [7,8] and the risk of metastasis and death by cancer is reduced in breast cancer patients treated with this drug [32,33,34]. The molecular mechanisms of metformin in diabetes control are not completely understood; activation of AMP-activated protein kinase (AMPK), inhibition of the mitochondrial respiratory chain (complex I) and mitochondrial glycerol-3-phosphate dehydrogenase, and a reduction in protein kinase A (PKA) activation have all been proposed as potential mechanisms [9]. The mechanisms by which metformin affects cancers are also unknown, although a large number of publications have shown that metformin could exert its antitumor effect by targeting AMPK/mTOR, anti-inflammatory, cell cycle/apoptosis, insulin/IGF-1R, and angiogenesis pathways in cancers [35,36,37,38]. It has also been shown that it can inactivate cells similar to breast cancer stem cells [39]. Metformin has potent growth-inhibitory and proapoptotic effects in pancreatic cancer [40], and several authors suggest that its biological effects are mediated through miRNA expression [22,40]. Some authors believe that metformin inhibits proliferation by upregulating miR-26a expression which consequently downregulates the targets of this miRNA [21]. Yang et al. [21] first described the involvement of miR-26a and its targets in the metformin mechanism of antitumor action in renal cancer. Our results confirmed previous data [23] in which miR-26a overexpression reduced cell viability, which was rescued with a miR-26a inhibitor to reverse the effect of the mimetic. Although this effect was slight, it reinforces the potential importance of miR-26a in cell viability/apoptosis processes in breast cancer. We detected a bigger reduction in cell viability in MCF-7 (luminal/ER+) cells than in the MDA-MB-468 and MDA-MB-231 (TNBC) cell lines. After evaluating miR-26a expression in different mammary cell line subtypes, one group reported that this miRNA is highly expressed in non-cancerous mammary cell lines but at lower levels in some breast cancer cell lines, in particular TNBC cells [29]. Differences in miR-26a expression in distinct breast cancer subtypes may also lead to different effects when it is exogenously overexpressed or downregulated. Similarly, other authors showed that miR-26a expression is higher in ER+ breast cancer [29]. In contrast with its effect on cell viability, miR-26 has been identified as a key mediator of estrogen-stimulated cell proliferation in ER+ breast cancer cells [31]. Furthermore, miR-26a seems to be strongly implicated in regulating ER+ breast cancer. Chen et al. [27] showed that miR-26a significantly downregulates ERα and prevents the stimulation of hepatoma cell growth by E2. Moreover, in MCF-7 cells, transient transfection of miR-26a initiates apoptosis [24]. Using bioinformatics tools we selected some theoretical miR-26a targets based on their relevance in cancer and their implication in viability/apoptosis processes (in which we and others have found miR-26a to be involved). Interestingly, PTEN and E2F3 were downregulated after transfection with the miR-26a mimetic; PTEN is one of the most commonly mutated tumor suppressors in cancer and has been shown to negatively regulate the AKT/PKB signaling pathway, favoring tumor development and progression. Our data agree with studies showing that PTEN is a miR-26a target in glioma [41,42] and in lung cancer [28]; we also showed that PTEN is downregulated at the protein level in breast cancer cells overexpressing miR-26a. The evaluation of metformin cellular effects reveals, according to other authors in kidney, pancreas and renal cancer [21,22], that the drug reduces cell viability in a dose-dependent manner in the breast cancer cell line MDA-MB-231 and that its administration increases miR-26a and reduces BCL-2 and PTEN expression. However, although the beneficial effects of metformin on breast cancer patient survival rates has been described by several authors [32,33,34,43,44,45,46], little is known about the mechanism. We show here that metformin up-regulated miR-26a and also downregulated its direct target PTEN. It is difficult to explain how metformin can have an anti-proliferative effect since PTEN is a tumor suppressor gene. Trying to understand how the PTEN inhibition by miR-26a can result in an anti-proliferative effect, we checked the effect of miR-26a overexpression in some additional targets of this miRNA. We identified EZH2, a miR-26a target that is downregulated by the miRNA and by metformin. EZH2 is a bona fide oncogene and acts as a dual function transcription regulator (not only repressor but also activator) [47] by converging on the methyltransferase-activity silencing tumor suppressor genes, which are implicated in neoplastic development and the transactivation property-activating genes involved in the late-stage process of cancer [48,49]. This gene has been implicated in promoting anchorage-independent growth and invasion of immortalized human mammary epithelial cells [24]. Retrospective studies from clinical breast cancer patients indicate that high expression of EZH2 is associated with short survival [50]. Therefore, it can be one of the effectors involved in the decrease in cell viability after treatment with metformin and overexpression of miR-26a and thus justify their anti-proliferative effect. It has been shown that metformin reduces cell migration. In general, we confirmed these data, although we observed slightly increased migration in the TNBC-model MDA-MB-231 cell line when the cells were treated with either a miR-26a mimetic or metformin. Our data are consistent with observations about the different behavior of breast cancer subtypes when miR-26a was upregulated. Boning Liu et al. [28] demonstrated that miR-26a increases lung cancer cell migration and the risk of metastasis by modulating activation of the AKT pathway by suppressing PTEN, data which agrees with our results in MDA-MB-231 cells. PTEN loss promotes cell migration in cancer cells, as previously described in breast cancer [51,52,53]. Simultaneous downregulation of PTEN and DLC1 in MCF-7 cells does not enhance cell proliferation, however, enhances cell migration [51]. DLC1 is negatively regulated by miRNAs in colorectal cancer [54,55], however, the possible DLC1 regulation by miR-26a has not been evaluated. We believe that PTEN inhibition by metformin via miR-26a could explain the increased cell migration under the treatment, but the metformin antitumor activity must be due to other miR-26a targets. We checked the gene expression levels of MCL-1, MTDH, and EZH2 which are proven targets of miR-26a in breast cancer and could be responsible of its anti-proliferative effect. We observed that EZH2 expression under metformin treatment was also lower than in non-treated cells, suggesting that metformin antitumor effect could involve this gene [23,24]. We also checked EZH2 expression under metformin treatment when transfecting miR-26a inhibitor and their levels were partially rescued, suggesting not only that metformin antitumor effect can be through this gene, but also that it happens via miR-26a overexpression. 4. Materials and Methods 4.1. Cell Lines and Culture The human MDA-MB-231 and MDA-MB-468 (both TNBC), and MCF-7 (luminal) cell lines were obtained from ATCC (ATCC, Manassas, VA, USA). The TNBC cell lines were maintained in DMEM/F12 medium with 10% fetal bovine serum (FBS) and MCF-7 cells were maintained in DMEM with 10% FBS; all the cell lines were cultured with 1% antibiotics (100 U/mL penicillin and 100 mg/L streptomycin) and were maintained in a humidified atmosphere with 5% CO2 at 37 °C. 4.2. Cell Viability Assay Cell viability was measured using a MTT-based cell growth [56] determination kit (#GDC1; Sigma-Aldrich, St. Louis, MO, USA). At the indicated intervals, MTT was added to each well and incubated for four hours at 37 °C. The medium was then carefully discarded and 50 µL MTT solvent was added to each well to dissolve the formazan crystals. Purple formazan crystals are formed from yellow MTT by succinate dehydrogenase in viable cells. The absorbance at 570 and 690 nm was measured using a microplate spectrophotometer and the percentage of surviving cells from each group relative to controls were calculated in triplicate. For the viability assays, cells were seeded at an initial density of 3 × 103 cells/mL in a 96-well plate and incubated with medium, transfection reagents Cy3 miRNA and 50 nM miR-26a mimic or inhibitor for different time periods at 37 °C. For metformin (Sigma-Aldrich) treatment, 24 h after seeding, the cells were treated with metformin (0, 1, 5, 10, 20, 40 mM) and viability was measured at 24, 48, and 72 h as described above. 4.3. MicroRNA Transfection To increase or reduce the miR-26a levels, 50 nM of miR-26a mimetic miRNA or inhibitor (Applied Biosystems, Foster City, CA, USA) were transfected using a TransIT-X2TM [57] polymeric non-liposomal system (Mirus Bio Corporation (Madison, WI, USA) following the manufacturer’s instructions; 50 nM of CyTM3 dye-labeled Pre-miRTM negative control was transfected as a negative control. 4.4. Analysis of miRNA and mRNA Expression by Quantitative Real-Time PCR RNA from cell lines was harvested using a miRNA isolation kit (mirVana, Ambion, Inc., 2130 Woodward Street, Austin, TX, USA) for miRNA and a TRIzol procedure for mRNA. The concentration and quality of the extracted RNA were determined by measuring OD260 and the OD260:OD280 ratio. First, 150 ng RNA were reverse transcribed to cDNA with specific stem-loop RT primers using a TaqMan microRNA reverse transcription kit (Applied Biosystems) for miRNA, and then 150 ng RNA were reverse transcribed to cDNA with random primers using a high-capacity cDNA reverse transcription kit (Applied Biosystems) from TaqMan for mRNA samples. RT-qPCR was performed using an ABI 7900HT fast RT-qPCR system and a TaqMan universal master mix (Applied Biosystems). All the primers were obtained from the TaqMan miRNA and mRNA assay kits (Applied Biosystems). The endogenous microRNA RNU43 [58] was used as an internal control for miRNA expression, and the housekeeping gene GAPDH was used as an internal control for mRNA expression. 4.5. Western Blot Assays Cells were lysed in a radio-immunoprecipitation assay (RIPA) lysis buffer containing protease inhibitor. Protein concentrations were determined using the Lowry-Folin method. Following SDS-PAGE separation, 50 µg of protein were transferred to polyvinylidene difluoride membranes (Bio-Rad). The membranes were blocked in tris-buffered saline (TBS) containing 5% non-fat milk and were subsequently incubated at 4 °C overnight with specific PTEN (1:500 in TBS with 0.1% Tween 20 (TBS-T) and 5% bovine serum albumin (BSA)) or β-actin (1:1000 5% non-fat milk in TBS-T) primary antibodies, and then washed repeatedly for 5 min with 1% BSA-TBS-T followed by incubation with anti-rabbit-horseradish peroxidase (HRP; 1:2500 5% non-fat milk in TBS-T) or anti-mouse-HRP (1:7500 5% non-fat milk in TBS-T) secondary antibodies for 1 h at room temperature (all antibodies were from Cell Signaling, Beverly, MA, USA). ECL reagent (Amersham Life Science, Piscataway, NJ, USA) was used for detection and the membranes were developed in an ImageQuant LAS 4000 (GE Healthcare Life Sciences, Princeton, NJ, USA). 4.6. Wound Healing Assay Cell migration was examined using a wound-healing assay [59]. Cells were cultured in six-well plates to 100% confluence. A plastic pipette tip was used to generate a wound area across the center of each well and after the wells were washed with PBS the medium was replaced and they were allowed to migrate. Micrograph images were taken with a microscope at 40× magnification at the indicated time points. All these experiments were repeated in triplicate. 4.7. Bioinformatic Databases miRBase database (http://www.mirbase.org/) was used to obtain information about miR-26a, and miRTarBase (http://mirtarbase.mbc.nctu.edu.tw/) was used to look for its gene targets. The first 950 genes were selected and analyzed in DAVID [60] (the database for annotation, visualization, and integrated discovery) bioinformatics resources to select targets involved in the most relevant pathways for cancer such as proliferation and apoptosis. 4.8. Statistical Analysis The data were presented as the mean ± standard deviation (SD) of the triplicate experiments. We analyzed the significance of any difference between the control and treatment groups using the Student t test and the level of statistical significance was set at 95% confidence (p < 0.05). 5. Conclusions We have confirmed the anti-proliferative effect of metformin in breast cancer. Our results suggest that upregulation of miR-26a and downregulation of some of these miRNA-targets are part of the action mechanisms of this drug. miR-26a is at least partially responsible for the metformin antineoplastic effect in breast cancer, even in triple negative breast cancer where there is no treatment other than chemotherapy, and this drug could result in a real improvement in treating the disease. Author Contributions Pilar Eroles and Ana Lluch designed the research and wrote the manuscript; Paula Cabello, Begoña Pineda, and Eduardo Tormo performed the molecular techniques; Paula Cabello and Begoña Pineda analyzed the data; All authors revised and approved the final version of the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 The effect of miR-26a on cell viability. MDA-MB-231, MDA-MB-468, and MCF-7 at days 1, 4, and 7 after transfection with miR-26a. C: non-treated control, CT: cells with transfection reagents, CY3: control with CY3 miRNA, 26a: miR-26a mimetic; 50 nM of pre-miRNA were transfected in all cases. Error bars represent the standard deviation of three experiments. Statistically significant differences comparing cells transfected with miR-26a to CY3 cells are shown at the respective time points. (Student t test: # p < 0.001). Figure 2 Determination of the miR-26a mimetic (26), the inhibitor (i26), and their combined effects on the viability of the (A) MDA-MB-231; (B) MDA-MB-468; and (C) MCF-7 cell lines at 24, 48, and 72 h after transfection; C: non-treated control, CT: cells with transfection reagents, CY3: control CY3 miRNA, 26: cells treated with 50 nM miR-26a mimetic; i26: cells treated with 50 nM miR-26a inhibitor, 26 + i26 is a combination of 25 + 25 nM or 50 + 50 nM. Error bars show the standard deviation of three experiments. (Student t test: * p < 0.05; # p < 0.001). Figure 3 Wound healing cell migration assay comparing cells transfected with 50 nM miR-26a or CY3 (control). Cells transfected with miR-26a closed the wound before the CY3-transfected cells (24 vs. 30 h). Figure 4 Gene expression of different miR-26a targets four days after transfection with either 50 nM miR-26a or its inhibitor in MDA-MB-231, MDA-MB-468, and MCF-7 cell lines, as measured by RT-qPCR. CY3: CY3 control miRNA. Error bars represent triplicate experiments. (Student t test: * p < 0.05; ** p < 0.005; # p < 0.001). Figure 5 PTEN protein expression was measured by Western blot after transfection with miR-26a or CY3 in MDA-MB-231 cell line. β-Actin was used as a control. Figure 6 The effect of metformin on MDA-MB-231 cell line viability at different time-points and concentrations. Error bars represent the standard deviation of triplicate experiments. Statistically significant differences are shown for the comparison between treated cells and the control (not treated) at respective time-points (Student t test: # p < 0.001). Figure 7 (A) Expression of miR-26a and two of its targets in MDA-MB-231 measured by RT-qPCR four days after treatment with 10 mM metformin or vehicle (PBS). Error bars represent triplicate experiments; (B) PTEN protein expression measured by Western blot. MDA-MB-231 was treated with 10 mM metformin or PBS, and β-actin was used as a control. Figure 8 Wound healing cell migration assay comparing metformin-treated (10 mM) MDA-MB-231 cells with non-treated cells at 24 and 30 h. (magnification 100×). Figure 9 The effect of metformin through miR-26a on MDA-MB-231 cells. (A) Cell viability at 48 h after 10 mM metformin treatment with or without the miR-26a inhibitor (50 nM); (B) Gene expression of miR-26a targets after transfection with either 50 nM miR-26a mimetic or its inhibitor, as measured by RT-qPCR. CY3: CY3 control miRNA; (C) Expression of miR-26a targets measured by RT-qPCR after treatment with 10 mM metformin (in presence or not of miR-26a inhibitor) or vehicle (PBS). Error bars represent triplicate experiments. (Student t test: * p < 0.05; # p < 0.001). ijms-17-01298-t001_Table 1Table 1 Theoretical and demonstrated miR-26a selected gene targets obtained with miRTarBase, and selected by DAVID Bioinformatics Resources and miRBase. Demonstrated targets are experimentally proveN targets by techniques such as RT-qPCR or Western Blot, although none of them are proven in breast cancer. miR-26a Target Genes CASP3 Theoretical CCNE2 Demonstrated ABL2 Theoretical APAF1 Theoretical XIAP Theoretical BCL2 Theoretical PTEN Demonstrated TP53 Theoretical E2F3 Theoretical CDC25A Theoretical BCL2L1 Theoretical ==== Refs References 1. Torre L.A. Bray F. Siegel R.L. Ferlay J. Lortet-Tieulent J. Jemal A. Global cancer statistics, 2012 CA Cancer J. Clin. 2015 65 87 108 10.3322/caac.21262 25651787 2. Eroles P. Bosch A. Perez-Fidalgo J.A. Lluch A. Molecular biology in breast cancer: Intrinsic subtypes and signaling pathways Cancer Treat. Rev. 2012 38 698 707 10.1016/j.ctrv.2011.11.005 22178455 3. Rivenbark A.G. O'Connor S.M. Coleman W.B. Molecular and cellular heterogeneity in breast cancer: Challenges for personalized medicine Am. J. Pathol. 2013 183 1113 1124 10.1016/j.ajpath.2013.08.002 23993780 4. Perou C.M. Sorlie T. Eisen M.B. van de Rijn M. Jeffrey S.S. Rees C.A. Pollack J.R. Ross D.T. Johnsen H. Akslen L.A. Molecular portraits of human breast tumours Nature 2000 406 747 752 10.1038/35021093 10963602 5. Bauer K.R. Brown M. Cress R.D. Parise C.A. Caggiano V. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: A population-based study from the California cancer Registry Cancer 2007 109 1721 1728 10.1002/cncr.22618 17387718 6. Carey L.A. Dees E.C. Sawyer L. Gatti L. Moore D.T. Collichio F. Ollila D.W. Sartor C.I. Graham M.L. Perou C.M. The triple negative paradox: Primary tumor chemosensitivity of breast cancer subtypes Clin. Cancer Res. 2007 13 2329 2334 10.1158/1078-0432.CCR-06-1109 17438091 7. Libby G. Donnelly L.A. Donnan P.T. Alessi D.R. Morris A.D. Evans J.M. New users of metformin are at low risk of incident cancer: A cohort study among people with type 2 diabetes Diabetes Care 2009 32 1620 1625 10.2337/dc08-2175 19564453 8. Quinn B.J. Kitagawa H. Memmott R.M. Gills J.J. Dennis P.A. Repositioning metformin for cancer prevention and treatment Trends Endocrinol. Metab. 2013 24 469 480 10.1016/j.tem.2013.05.004 23773243 9. Madiraju A.K. Erion D.M. Rahimi Y. Zhang X.M. Braddock D.T. Albright R.A. Prigaro B.J. Wood J.L. Bhanot S. MacDonald M.J. Metformin suppresses gluconeogenesis by inhibiting mitochondrial glycerophosphate dehydrogenase Nature 2014 510 542 546 10.1038/nature13270 24847880 10. Kim V.N. Nam J.W. Genomics of microRNA Trends Genet. 2006 22 165 173 10.1016/j.tig.2006.01.003 16446010 11. Lin S.L. Kim H. Ying S.Y. Intron-mediated RNA interference and microRNA (miRNA) Front. Biosci. 2008 13 2216 2230 10.2741/2836 17981704 12. Caramuta S. Egyhazi S. Rodolfo M. Witten D. Hansson J. Larsson C. Lui W.O. MicroRNA expression profiles associated with mutational status and survival in malignant melanoma J. Investig. Dermatol. 2010 130 2062 2070 10.1038/jid.2010.63 20357817 13. Dillhoff M. Liu J. Frankel W. Croce C. Bloomston M. MicroRNA-21 is overexpressed in pancreatic cancer and a potential predictor of survival J. Gastrointest. Surg. 2008 12 2171 2176 10.1007/s11605-008-0584-x 18642050 14. Li X. Zhang Y. Ding J. Wu K. Fan D. Survival prediction of gastric cancer by a seven-microRNA signature Gut 2010 59 579 585 10.1136/gut.2008.175497 19951901 15. Rossi S. Shimizu M. Barbarotto E. Nicoloso M.S. Dimitri F. Sampath D. Fabbri M. Lerner S. Barron L.L. Rassenti L.Z. microRNA fingerprinting of CLL patients with chromosome 17p deletion identify a miR-21 score that stratifies early survival Blood 2010 116 945 952 10.1182/blood-2010-01-263889 20393129 16. Yanaihara N. Caplen N. Bowman E. Seike M. Kumamoto K. Yi M. Stephens R.M. Okamoto A. Yokota J. Tanaka T. Unique microRNA molecular profiles in lung cancer diagnosis and prognosis Cancer Cell 2006 9 189 198 10.1016/j.ccr.2006.01.025 16530703 17. Shi Z. Zhang J. Qian X. Han L. Zhang K. Chen L. Liu J. Ren Y. Yang M. Zhang A. AC1MMYR2, an inhibitor of dicer-mediated biogenesis of Oncomir miR-21, reverses epithelial-mesenchymal transition and suppresses tumor growth and progression Cancer Res. 2013 73 5519 5531 10.1158/0008-5472.CAN-13-0280 23811941 18. Bader A.G. miR-34—A microRNA replacement therapy is headed to the clinic Front. Genet. 2012 3 120 10.3389/fgene.2012.00120 22783274 19. Trang P. Wiggins J.F. Daige C.L. Cho C. Omotola M. Brown D. Weidhaas J.B. Bader A.G. Slack F.J. Systemic delivery of tumor suppressor microRNA mimics using a neutral lipid emulsion inhibits lung tumors in mice Mol. Ther. 2011 19 1116 1122 10.1038/mt.2011.48 21427705 20. Tormo E. Pineda B. Serna E. Guijarro A. Ribas G. Fores J. Chirivella E. Climent J. Lluch A. Eroles P. MicroRNA Profile in Response to Doxorubicin Treatment in Breast Cancer J. Cell. Biochem. 2015 116 2061 2073 10.1002/jcb.25162 25802200 21. Yang F.Q. Wang J.J. Yan J.S. Huang J.H. Li W. Che J.P. Wang G.C. Liu M. Zheng J.H. Metformin inhibits cell growth by upregulating microRNA-26a in renal cancer cells Int. J. Clin. Exp. Med. 2014 7 3289 3296 25419360 22. Bao B. Wang Z. Ali S. Ahmad A. Azmi A.S. Sarkar S.H. Banerjee S. Kong D. Li Y. Thakur S. Metformin inhibits cell proliferation, migration and invasion by attenuating CSC function mediated by deregulating miRNAs in pancreatic cancer cells Cancer Prev. Res. (Phila) 2012 5 355 364 10.1158/1940-6207.CAPR-11-0299 22086681 23. Gao J. Li L. Wu M. Liu M. Xie X. Guo J. Tang H. MiR-26a inhibits proliferation and migration of breast cancer through repression of MCL-1 PLoS ONE 2013 8 1298 10.1371/journal.pone.0065138 23750239 24. Zhang B. Liu X.X. He J.R. Zhou C.X. Guo M. He M. Li M.F. Chen G.Q. Zhao Q. Pathologically decreased miR-26a antagonizes apoptosis and facilitates carcinogenesis by targeting MTDH and EZH2 in breast cancer Carcinogenesis 2011 32 2 9 10.1093/carcin/bgq209 20952513 25. Eroles P. Tormo E. Pineda B. Espin E. Lluch A. MicroRNAs in breast cancer: One More Turn in Regulation Curr. Drug Targets 2015 in press 26. Kasinski A.L. Slack F.J. Epigenetics and genetics. MicroRNAs en route to the clinic: Progress in validating and targeting microRNAs for cancer therapy Nat. Rev. Cancer 2011 11 849 864 10.1038/nrc3166 22113163 27. Chen L. Zheng J. Zhang Y. Yang L. Wang J. Ni J. Cui D. Yu C. Cai Z. Tumor-specific expression of microRNA-26a suppresses human hepatocellular carcinoma growth via cyclin-dependent and -independent pathways Mol. Ther. 2011 19 1521 1528 10.1038/mt.2011.64 21610700 28. Liu B. Wu X. Wang C. Liu Y. Zhou Q. Xu K. MiR-26a enhances metastasis potential of lung cancer cells via AKT pathway by targeting PTEN Biochim. Biophys. Acta 2012 1822 1692 1704 10.1016/j.bbadis.2012.07.019 22885155 29. Liu P. Tang H. Chen B. He Z. Deng M. Wu M. Liu X. Yang L. Ye F. Xie X. miR-26a suppresses tumour proliferation and metastasis by targeting metadherin in triple negative breast cancer Cancer Lett. 2015 357 384 392 10.1016/j.canlet.2014.11.050 25434799 30. Jansen M.P. Reijm E.A. Sieuwerts A.M. Ruigrok-Ritstier K. Look M.P. Rodriguez-Gonzalez F.G. Heine A.A. Martens J.W. Sleijfer S. Foekens J.A. High miR-26a and low CDC2 levels associate with decreased EZH2 expression and with favorable outcome on tamoxifen in metastatic breast cancer Breast Cancer Res. Treat. 2012 133 937 947 10.1007/s10549-011-1877-4 22094936 31. Tan S. Ding K. Li R. Zhang W. Li G. Kong X. Qian P. Lobie P.E. Zhu T. Identification of miR-26 as a key mediator of estrogen stimulated cell proliferation by targeting CHD1, GREB1 and KPNA2 Breast Cancer Res. 2014 16 R40 10.1186/bcr3644 24735615 32. Kim H.J. Kwon H. Lee J.W. Lee S.B. Park H.S. Sohn G. Lee Y. Koh B.S. Yu J.H. Son B.H. Metformin increases survival in hormone receptor-positive, HER2-positive breast cancer patients with diabetes Breast Cancer Res. 2015 17 64 10.1186/s13058-015-0574-3 25935404 33. Bayraktar S. Hernadez-Aya L.F. Lei X. Meric-Bernstam F. Litton J.K. Hsu L. Hortobagyi G.N. Gonzalez-Angulo A.M. Effect of metformin on survival outcomes in diabetic patients with triple receptor-negative breast cancer Cancer 2012 118 1202 1211 10.1002/cncr.26439 21800293 34. He X. Esteva F.J. Ensor J. Hortobagyi G.N. Lee M.H. Yeung S.C. Metformin and thiazolidinediones are associated with improved breast cancer-specific survival of diabetic women with HER2+ breast cancer Ann. Oncol. 2012 23 1771 1780 10.1093/annonc/mdr534 22112968 35. Anisimov V.N. Berstein L.M. Egormin P.A. Piskunova T.S. Popovich I.G. Zabezhinski M.A. Kovalenko I.G. Poroshina T.E. Semenchenko A.V. Provinciali M. Effect of metformin on life span and on the development of spontaneous mammary tumors in HER-2/Neu transgenic mice Exp. Gerontol. 2005 40 685 693 10.1016/j.exger.2005.07.007 16125352 36. Ben Sahra I. Laurent K. Loubat A. Giorgetti-Peraldi S. Colosetti P. Auberger P. Tanti J.F. Le Marchand-Brustel Y. Bost F. The antidiabetic drug metformin exerts an antitumoral effect in vitro and in vivo through a decrease of cyclin D1 level Oncogene 2008 27 3576 3586 10.1038/sj.onc.1211024 18212742 37. Dowling R.J. Zakikhani M. Fantus I.G. Pollak M. Sonenberg N. Metformin inhibits mammalian target of rapamycin-dependent translation initiation in breast cancer cells Cancer Res. 2007 67 10804 10812 10.1158/0008-5472.CAN-07-2310 18006825 38. Martin-Castillo B. Dorca J. Vazquez-Martin A. Oliveras-Ferraros C. Lopez-Bonet E. Garcia M. del Barco S. Menendez J.A. Incorporating the antidiabetic drug metformin in HER2-positive breast cancer treated with neo-adjuvant chemotherapy and trastuzumab: An ongoing clinical-translational research experience at the Catalan Institute of Oncology Ann. Oncol. 2010 21 187 189 10.1093/annonc/mdp494 19884247 39. Hirsch H.A. Iliopoulos D. Tsichlis P.N. Struhl K. Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission Cancer Res. 2009 69 7507 7511 10.1158/0008-5472.CAN-09-2994 19752085 40. Li W. Yuan Y. Huang L. Qiao M. Zhang Y. Metformin alters the expression profiles of microRNAs in human pancreatic cancer cells Diabetes Res. Clin. Pract. 2012 96 187 195 10.1016/j.diabres.2011.12.028 22245693 41. Huse J.T. Brennan C. Hambardzumyan D. Wee B. Pena J. Rouhanifard S.H. Sohn-Lee C. le Sage C. Agami R. Tuschl T. The PTEN-regulating microRNA miR-26a is amplified in high-grade glioma and facilitates gliomagenesis in vivo Genes Dev. 2009 23 1327 1337 10.1101/gad.1777409 19487573 42. Kim H. Huang W. Jiang X. Pennicooke B. Park P.J. Johnson M.D. Integrative genome analysis reveals an oncomir/oncogene cluster regulating glioblastoma survivorship Proc. Natl. Acad. Sci. USA 2010 107 2183 2188 10.1073/pnas.0909896107 20080666 43. Hou G. Zhang S. Zhang X. Wang P. Hao X. Zhang J. Clinical pathological characteristics and prognostic analysis of 1,013 breast cancer patients with diabetes Breast Cancer Res. Treat. 2013 137 807 816 10.1007/s10549-012-2404-y 23292119 44. Lega I.C. Austin P.C. Gruneir A. Goodwin P.J. Rochon P.A. Lipscombe L.L. Association between metformin therapy and mortality after breast cancer: A population-based study Diabetes Care 2013 36 3018 3026 10.2337/dc12-2535 23633525 45. Peeters P.J. Bazelier M.T. Vestergaard P. Leufkens H.G. Schmidt M.K. de Vries F. de Bruin M.L. Use of metformin and survival of diabetic women with breast cancer Curr. Drug Saf. 2003 8 357 363 10.2174/15680266113136660069 24215316 46. Xiao Y. Zhang S. Hou G. Zhang X. Hao X. Zhang J. Clinical pathological characteristics and prognostic analysis of diabetic women with luminal subtype breast cancer Tumour Biol. 2014 35 2035 2045 10.1007/s13277-013-1270-5 24096546 47. Shi B. Liang J. Yang X. Wang Y. Zhao Y. Wu H. Sun L. Zhang Y. Chen Y. Li R. Integration of estrogen and Wnt signaling circuits by the polycomb group protein EZH2 in breast cancer cells Mol. Cell. Biol. 2007 27 5105 5119 10.1128/MCB.00162-07 17502350 48. Bracken A.P. Kleine-Kohlbrecher D. Dietrich N. Pasini D. Gargiulo G. Beekman C. Theilgaard-Monch K. Minucci S. Porse B.T. Marine J.C. The Polycomb group proteins bind throughout the INK4A-ARF locus and are disassociated in senescent cells Genes Dev. 2007 21 525 530 10.1101/gad.415507 17344414 49. Schuettengruber B. Chourrout D. Vervoort M. Leblanc B. Cavalli G. Genome regulation by polycomb and trithorax proteins Cell 2007 128 735 745 10.1016/j.cell.2007.02.009 17320510 50. Kleer C.G. Cao Q. Varambally S. Shen R. Ota I. Tomlins S.A. Ghosh D. Sewalt R.G. Otte A.P. Hayes D.F. EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells Proc. Natl. Acad. Sci. USA 2003 100 11606 11611 10.1073/pnas.1933744100 14500907 51. Heering J. Erlmann P. Olayioye M.A. Simultaneous loss of the DLC1 and PTEN tumor suppressors enhances breast cancer cell migration Exp. Cell Res. 2009 315 2505 2514 10.1016/j.yexcr.2009.05.022 19482022 52. Tamura M. Gu J. Matsumoto K. Aota S. Parsons R. Yamada K.M. Inhibition of cell migration, spreading, and focal adhesions by tumor suppressor PTEN Science 1998 280 1614 1617 10.1126/science.280.5369.1614 9616126 53. Tamura M. Gu J. Takino T. Yamada K.M. Tumor suppressor PTEN inhibition of cell invasion, migration, and growth: Differential involvement of focal adhesion kinase and p130Cas Cancer Res. 1999 59 442 449 9927060 54. Wu P.P. Zhu H.Y. Sun X.F. Chen L.X. Zhou Q. Chen J. MicroRNA-141 regulates the tumour suppressor DLC1 in colorectal cancer Neoplasma 2015 62 705 712 10.4149/neo_2015_084 26278151 55. Zhang G.J. Li J.S. Zhou H. Xiao H.X. Li Y. Zhou T. MicroRNA-106b promotes colorectal cancer cell migration and invasion by directly targeting DLC1 J. Exp. Clin. Cancer Res. 2015 34 73 10.1186/s13046-015-0189-7 26223867 56. Mosmann T. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays J. Immunol. Methods 1983 65 55 63 10.1016/0022-1759(83)90303-4 6606682 57. Shopsowitz K.E. Wu C. Liu G. Dreaden E.C. Hammond P.T. Periodic-shRNA molecules are capable of gene silencing, cytotoxicity and innate immune activation in cancer cells Nucleic Acids Res. 2016 44 545 557 10.1093/nar/gkv1488 26704983 58. Gyongyosi A. Docs O. Czimmerer Z. Orosz L. Horvath A. Torok O. Mehes G. Nagy L. Balint B.L. Measuring expression levels of small regulatory RNA molecules from body fluids and formalin-fixed, paraffin-embedded samples Methods Mol. Biol. 2014 1182 105 119 25055905 59. Rodriguez L.G. Wu X. Guan J.L. Wound-healing assay Methods Mol. Biol. 2005 294 23 29 15576902 60. Huang D.W. Sherman B.T. Tan Q. Kir J. Liu D. Bryant D. Guo Y. Stephens R. Baseler M.W. Lane H.C. DAVID Bioinformatics Resources: Expanded annotation database and novel algorithms to better extract biology from large gene lists Nucleic Acids Res. 2007 35 W169 W175 10.1093/nar/gkm415 17576678

PMC005xxxxxx/PMC5000696.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081299ijms-17-01299ArticleAnalysis of Different Ploidy and Parent–Offspring Genomic DNA Methylation in the Loach Misgurnus anguillicaudatus Zhou He 1Ma Tian-Yu 1Zhang Rui 1Xu Qi-Zheng 1Shen Fu 2Qin Yan-Jie 1Xu Wen 1Wang Yuan 1Li Ya-Juan 1*Lin Li Academic Editor1 Key Laboratory of Mariculture and Stock Enhancement in North China’s Sea, Ministry of Agriculture, Dalian Ocean University, Dalian 116023, China; [email protected] (H.Z.); [email protected] (T.-Y.M.); [email protected] (R.Z.); [email protected] (Q.-Z.X.); [email protected] (Y.-J.Q.); [email protected] (W.X.); [email protected] (Y.W.)2 Fisheries Technology Extension Station of Beijing, Beijing 101105, China; [email protected]* Correspondence: [email protected]; Tel.: +86-411-8476-3639; Fax: +86-411-8476-262022 8 2016 8 2016 17 8 129906 6 2016 26 7 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).In this study, we selected natural polyploidy loach (diploid, triploid and tetraploid) and hybrid F1 generation obverse cross (4 × 2) and inverse cross (2 × 4) by diploids and tetraploids as the research model. The MSAP (methylation-sensitive amplified polymorphism) reaction system was established by our laboratory to explore methylation levels and pattern diversification features at the whole genome level of the polyploidy loach. The results showed that the total methylation and full methylation rates decreased on increased ploidy individuals; moreover, the hemimethylation rate showed no consistent pattern. Compared with diploid loach, the methylation patterns of tetraploid sites changed 68.17%, and the methylation patterns of triploid sites changed 73.05%. The proportion of hypermethylation genes is significantly higher than the proportion of demethylation genes. The methylation level of reciprocal cross F1 generation is lower than the male diploid and higher than the female tetraploid. The hemimethylation and total methylation rate of the cross hybrid F1 generation is significantly higher than the orthogonal F1 generation (p < 0.01). After readjusting, the methylation pattern of genome DNA of reciprocal hybrids changed 69.59% and 72.83%, respectively. Misgurnus anguillicaudatusdiploidtriploidtetraploidparent–offspring generationDNA methylation ==== Body 1. Introduction In the process of biological evolution, polyploidy (the duplication of the whole genome) is a universal and natural phenomenon. It leads to an increase of gene dosage and genome, which provides the space and potential for biological evolution. Polyploids generally exist among many plants and animals. Especially in freshwater fish, more than 30 kinds of polyploidy types have been found in China [1,2]. Researches in recent years show that the process of polyploidy and stabilization can lead to comprehensive changes in the structure, expression and function of genes. It affects the processes and the mechanisms of all genetic and epigenetic products [3]. Epigenetics is the genetic change in gene expression that does not involve the occurrence of DNA sequence change, which has important effects on the formation and evolution of multiples. The phenomenon and mechanism of epigenetics is researched widely and related to the polyploidy incorporation of DNA methylation, gene state, nucleolar dominance, etc. [4]. DNA methylation as a sort of modification process and reaction that commonly occurs in cells is a major epigenetic modification of genome DNA. It is also an important means of regulating gene function [5]. In recent years, methylation-sensitive amplified polymorphism (MSAP) technology has widely been used in animal and plant genome DNA methylation level and pattern analysis. However, there are few reports on the application in fish [6]. Loach (Misgurnus anguillicaudatus), besides being unique fish with high economic value, has multiple reproduction patterns and a phenomenon of ploidy variation [7]. Research shows that natural diploid, triploid, and tetraploid loaches exist in the same area in China [8,9,10]. Therefore, loach is an ideal material to explore the mechanism of homologous polyploids and allopolyploids. In the last 10 years, our laboratory has devised systematic research into the distribution, origin and formation mechanism of China’s natural polyploidy loach and has investigated the cellular and molecular patterns. China’s unique natural tetraploid loach is a homologous polyploidy in which both male and female are fertile [11,12,13]. At the same time, we also used natural tetraploid and diploid loach to hybridize a new triploid to research the composition of the gamete chromosomes [14]. To the best of our knowledge, there is no report into the epigenetic mechanism among different polyploid loach, hybrid triploid loach and parents. Therefore, in this research, we selected different polyploid loach (diploid, triploid and tetraploid) and a reciprocal cross generation of diploid and tetraploid parents as subjects. Our laboratory has established an MSAP (methylation-sensitive amplified polymorphism) reaction system to explore the changes on DNA methylation levels and patterns of different polyploidy and parent–offspring loaches, and to evaluate the regulation mechanism at gene expression changes of polyploidy in loach. This study aims to explore homologous and heterologous polyploidy mechanisms in fish. 2. Results 2.1. Ploidy Identification of Loach Flow cytometry analysis indicates that the diploid, triploid and tetraploid cell population DNA content is 2C DNA (Figure 1A), 3C DNA (Figure 1B) and 4C DNA (Figure 1C), respectively. The ratio of DNA content in individual blood cells of diploid, triploid and tetraploid is 1:1.5:2. We used the chromosome counting method for further determination of ploidy in loaches because it is currently the most direct and accurate assay method. Chromosome observation showed that the chromosomes number is 2n = 50 in diploid (Figure 1D), 3n = 75 in triploid (Figure 1E) and 4n = 100 in tetraploid (Figure 1F). 2.2. The Levels of Genomic DNA Methylation in Different Ploidy of Loaches The total bands and DNA methylation level, obtained by MSAP of diploid, triploid and tetraploid loaches, are shown in Table 1. The results show 5892 total bands from the selective amplification of eight primer pairs. Each pair of primers could amplify an average of 737 bands. There was a difference in methylation level between different ploidy loaches. Total methylation rate from high to low was diploid, 41.21% (full methylation 23.50% and hemimethylation 17.71%); triploid, 38.78% (full methylation 21.02% and hemimethylation 17.76%); and tetraploid, 37.03% (full methylation 22.22% and hemimethylation 14.81%). The results indicate a ploidy effect occurs with an increase of ploidy, the total methylation and full methylation ratio gradually reduced, while there is no specific rule in hemimethylation ratio. Statistical analysis showed that there was a significant difference in methylated level between triploid, tetraploid and diploid (p < 0.05), while there was no significant difference between triploid and tetraploid loaches (p > 0.05). 2.3. The Pattern of Genomic DNA Methylation on Different Ploidy for Loaches We divided the pattern of DNA methylation of different ploidy loaches using the method of Bian [15], which brings amplification bands between different ploidy in loquat into four types as follows: A-type is a monomorphic site with the same methylation status between two ploidies, that is, both ploidies are hemimethylated or fully methylated; B-type is a demethylation type in which methylation exists in a control sample while the ploidy of loquat has a variation of demethylation in this site; C-type is an over- or hypermethylation type, the methylation level of some ploidy loquats were higher than the control; and D-type is the methine type, where the methylation level of some ploidy loquats are lower than the control. The four types from Bian [15] in our experiment are summarized as follows (Table 2): A-type is a monomorphic site. The DNA methylation level is the same among the different ploidy loaches. B-type is a demethylation type. Methylation occurred in the diploid, but in the triploid and tetraploid demethylation variation occurred at this site. C-type is the over- or hypermethylation type. Triploid and tetraploid methylation is higher than diploid. D-type is the methine type. Triploid and tetraploid methylation is lower than diploid, but there still exists a methylation status. The results of this study showed that compared with the diploid, the triploid’s DNA had 73.05% patterns of methylation variation; and the tetraploid’s DNA had 68.17% patterns of methylation variation (Table 2). Polymorphic sites in loaches with different ploidy show that the over- or hypermethylation type (C-type) was the highest, followed by the demethylation type (B-type), and sub-methylated type (D-type) is characterized as the lowest. This shows that many adjustments occur in loach methylation patterns, mainly based on over- or hypermethylation. 2.4. Level of Genomic DNA Methylation in Parents–Offspring of Loaches The results of the amplification with the eight primer pairs show that the DNA methylation level of F1 was between their parents. It was lower than the male (female) diploid, but higher than the female (male) tetraploid. Using statistical analysis, the result of positive hybridization (4 × 2) showed that the female tetraploid, the male diploid and their offspring have significant differences in full methylation level (p < 0.01) and the male diploid and their offspring have no difference (p > 0.05). Comparing the female tetraploid and their offspring with the male diploid, there was a significant difference in hemimethylation level (p < 0.01). There is no difference between the female tetraploid and their offspring (p > 0.05). There was a significant difference in total methylation level (p < 0.01). The result in the hybrid (2 × 4) showed a significant difference between the female diploid and the male tetraploid and their offspring (p < 0.01) (Table 3). There is no significant difference in full methylation level among the F1, but there is significant difference in total methylation level and hemimethylation among the F1 (Table 4). 2.5. The Pattern of Genomic DNA Methylation in Parents-Offspring of Loaches The diploid, tetraploid and their F1 offspring have been adjusted again for methylation pattern. They were divided into four types named A, B, C, and D (Table 5). A-type accounted for 30.41% and 27.17% of total methylation sites of orthogonal and anti-cross F1 generation. B-type accounted for 17.04% and 17.36% of total methylation sites of orthogonal and anti-cross F1 generation. C-type accounted for 39.33% and 39.95% of total methylation sites of orthogonal and anti-cross F1 generation. D-type accounted for 13.22% and 15.52% of total methylation sites of obverse and inverse F1 generation. Thus, the orthogonal F1 generation genomic DNA hypomethylate pattern had 69.59% mutations (demethylation into 17.04%, over- or hypermethylation into 39.33%, hypomethylation into 13.22%); anti-cross F1 hybrids genomic DNA also has 72.83% methylation pattern mutations (demethylation into 17.36%, over- or hypermethylation into 39.95%, methylene into 15.52%). Orthogonal or anti-cross hybrids are the main hypermethylation (Table 5). 3. Discussion After species polyploidy, homologous polyploids and allopolyploids overcome the effects of genome doubling, some genetic and epigenetic changes will soon be produced, making it faster and better adapted to the new environment. The variation of epigenetics plays an important role in improving the diversity of polyploid gene expression, inducing the diploidization of genetics and cytology and promoting mutual coordination between the genome, and so on [16]. As an important form of epigenetic modifications, DNA methylation plays an important role in controlling gene expression, maintaining the stability of the genome, etc. Much research shows that after genome polyploidy the changes of levels and the adjustment of DNA methylation patterns is closely related to maintaining the stability of the genome, and balancing the reconstruction of nucleoplasm in polyploidy [17,18,19]. Generally speaking, DNA methylation and gene expression show a negative correlation, no matter whether in diploid or polyploid, DNA demethylation can lead to gene activation or translocation activation. High levels of methylation always lead to certain gene silencing or inhibition, expression and the expression level of related gene are likely to change [20]. In studies of Arabidopsis thaliana [21,22], when the genome DNA is at high level of methylation, it has been discovered that the expression of some genes are silenced. When methylation of DNA is processed, the level of methylation is reduced, and the related gene silencing phenomena is then lifted [19]. As a result, the relation of ploidy and DNA methylation levels in the previous are not the same. There are three main trends: first, along with the increase of ploidy, DNA methylation levels gradually increase, which shows a positive correlation [16,23]; second, along with the increase of ploidy, DNA methylation levels gradually reduce, which show a negative correlation [24,25]; and, third, no specific rule is shown [26,27,28]. In this research, DNA methylation levels of different ploidy also change; the total methylation rates of diploid, triploid and tetraploid were 42.21%, 38.78% and 37.03%, respectively. Methylation levels show a significant difference between triploid, tetraploid and diploid loach (p < 0.01). However, there was no significant difference between the triploid and tetraploid loach. This indicates that in the process of loach doubling its genome DNA methylation modification changes have taken place. The study found that the polyploid methylation levels did not increase with increasing polyploid levels, but gradually decreased with the increase of polyploid levels. This belongs to the second type of characteristic. After loach polyploidy, because of the increased genome, the methylation level is relatively lower, which means that some gene silencing is lifted, then related transcription can be reactivated. This means that the polyploid loach can regulate the redundant genes by DNA methylation. It also indicates that auto tetraploid loach has greater improvement potential and wider adaptability, shows advantageous characteristics of quick growth, low oxygen consumption rate, high nutritional value, etc. [29]. Many adjustments of polyploid methylation patterns can induce the activation and silencing of certain genes, leading to changes of epigenetics, making the polyploid better adapted to the needs of the environment. In this experiment, based on analyzing changes of different ploidy on loach genomic DNA methylation patterns, it was found that, compared with the diploid loach, methylation patterns of tetraploid sites have changed 68.17% (hypermethylation 36.84%, demethylation 26.80%, and hypomethylation 4.53%), and the triploid sites have changed 73.05% (hypermethylation 45.34%, demethylation 23.80%, and hypomethylation 3.91%). It can be seen that the proportion of hypermethylation gene is significantly higher than the proportion of demethylation. Polyploid loach, possibly through methylation of some functional genes, does not express to mitigate the effects of genome double dose effect. In research with wheat [30], rice [26], and Stevia rebaudiana [25], conclusions are basically identical. Triploid fish have characteristics of infertility, fast growth, good meat quality, disease resistance etc.; they also have great significance in breeding [31]. An increasing number of studies show that the regulation of genetic level is conducive to the stability of hybridization and the genome doubling polyploid. Furthermore, the epigenetic modification, in which DNA methylation is the main mechanism, is very important for species formation and the successful evolution of multiple hybrid organisms. With the deepening research of DNA methylation, there are some reports about the relationship between heterosis and DNA methylation in plants. Romagnoli et al. [32] first put forward the relationship between heterosis and gene expression in crops, and found that 33% of specific expression products re-express. The study of triploid loquat and its parent by Wang [33] shows that the change of DNA methylation can promote the formation of triploid loquat heterosis. The study of hybrid larch species by Li et al. [34] shows that the formation of heterosis is related to the significant increase of DNA methylation in the offspring. The study of dandelion by Verhoeven et al. [35] compared MSAP and amplified fragment length polymorphism (AFLP) fragment inheritance in a diploid and triploid cross, and revealed de novo methylation variation between triploid F1 individuals. However, research in aquatic animals is relatively deficient. The research of DNA methylation patterns in red crucian carp and allotetraploid crucian carp through the MSAP method used by Song et al. [36] showed that allotetraploid crucian carp inherit 61.69% of their methylation pattern from both parents or one of them, indicating that the methylation level mainly follows the laws of Mendelian inheritance and maintains stable methylation patterns in tetraploid generations. The hybrid progeny of Zhi Kong scallops and Japanese scallops have 19.98% of sites in the methylation state, as reported by Yu et al. [37]. Cao et al. [6], through MSAP analysis of 20 full-sib in grass carp, found that methylation sites accounted for 75.9% of total sites. For triploid loach, whether it is a naturally occurring or synthetic, in the process of triploid formation from different individual genome reorganization and double phenomenon, to regain the balance of the gene of unpaired chromosomes, it cannot through large-scale chromosomal evolution, restore the chromosome diploidy in a short period of time. Therefore, it may regulate the extra genes and turning on and off gene expression through the changes of DNA methylation level and pattern. This study of methylation level and pattern analysis of diploid and tetraploid loach and reciprocal cross progeny shows that DNA methylation level and pattern of reciprocal cross F1 generation have obvious changes to parents. The DNA methylation level of hybrid F1 generation is between its parents, less than male parent diploid, and higher than female parent tetraploid. The hemimethylation and total methylation of reciprocal cross F1 generation is significantly higher than the orthogonal generation (p < 0.01). In individual detection, we found that all methylation rates are greater than the hemimethylation rates in most genomes of loach. Again, the loach methylation mode is given priority to full methylation. The result that all methylation rate is higher than hemimethylation rate in mammalian genomes are the same findings of Tang et al. [38]. The methylation pattern of reciprocal cross F1 generation have four types, namely monomorphism, demethylation, hypermethylation and hypomethylation, but they are mainly composed of type hypermethylation. This conclusion is basically identical to the research conclusion of Wang [33], but is not the same finding in corn, Arabidopsis thaliana [12,39], larch [34] and rice [40]. Related studies [34] have shown that methylation means that genes translate from activation to suppression; however, demethylation means that genes translate from suppression to activation. Whether methylation or demethylation is good for heterosis, different sites of enhanced or reduced methylation have a different effect on heterosis. 4. Materials and Methods 4.1. Ethics Statement This study was performed according to the Guide for the Care and Use of Laboratory Animals in Dalian Ocean University, Dalian, China. All animal experiments comply with Chinese laws, regulations and ethics. 4.2. Materials Thirty diploid loaches from farmers market were used as samples in Dalian, Liaoning Province, China. All natural triploid and tetraploid loaches were from Honghu, Hubei Province, China. Sixty individuals include 30 triploid loaches and 30 tetraploid loaches. All of the loaches were fed in aquaria (22 ± 1 °C) in the laboratory of Dalian Ocean University. 4.3. Ploidy Identification In order to ensure the ploidy of loaches used in this experiment. The flow cytometry (Partec PAS-III, PARTEC, Münster, Germany) and chromosome number counts are used to observe ploidy detection. The specific method is as followed. 4.3.1. Flow Cytometer Blood was collected from the caudal peduncle of loaches, and stained by DAPI (4′,6-diamidino-2-phenylindole). Using the blood DNA content of diploid loaches as a normal diploid standard, each blood sample was measured separately. 4.3.2. Preparation of Chromosome Samples Loach was intraperitoneally injected with phytohemagglutinin (PHA; 6 μg·g−1 body weight), followed by a second PHA treatment 18–20 h later. At 4–6 h after the second PHA treatment, 0.1% colchicine (6 μg·g−1 body weight) was administered intraperitoneally. At 2–3 h after colchicines treatment, the animals were euthanized, and branchia tissue samples were collected. The branchia samples were treated under hypotonic stress in 0.8% sodium citrate for 40–45 min. The samples were fixed for 45 min in Carnoy’s solution (methanol: acetic acid = 3:1). The chromosome samples were prepared by the cold drop method and dyed by Giemsa. Statistical chromosome numbers were observed using optical microscopy. 4.4. Artificial Induce Spawning and Insemination The parents were chosen from good development of gonad diploid and natural tetraploid loaches, injected with human chorionic gonadotropin (HCG) (injection dose: female, 20 to 25 IU·g−1; male, 10 to 12.5 IU·g−1). After 12 h, gently press female tetraploid loach abdomen, the eggs were discharged and collected in a 9 cm culture dish. Extrusion to genital pore on both sides along the male body made semen discharge. Semen was collected in centrifuge tubes by capillary (diluted 100 times with fresh water physiological saline). Using dry fertilization, hybridized combinations were obverse cross (4 × 2) and inverse cross (2 × 4). During the incubation and breeding period, the temperature was maintained at 25 ± 1 °C. Water was aerated, and dead fry were removed timely from the nursery pond. Fresh air and circulation was maintained in the breeding room. 4.5. DNA Extraction Genomic DNA from loach fin tissue (there were 20 diploid, triploid and tetraploid loaches; 20 female and male loach of hybrid F1 generation; and four parents from obverse and inverse cross generations) was extracted by the SS-Phenol extraction method. The method was as follows. Around 0.2 g tail fin tissue was put it into 400 μL urea buffer (0.1 mol·L−1 Tris-HCl, pH: 7.5) with 10 μL proteinase K in buffer. Fin was digested for 12 h at 37 °C in a thermostat water bath. Digested fin was extracted once by saturated phenol, twice by phenol, chloroform, isoamyl alcohol mixture (volume ratio: 25:24:1), and once by chloroform. DNA was precipitated 30 min by 100% ethanol and centrifuged 10 min at speed of 12,000 rpm. The precipitation was dissolved in TE buffer (pH: 8.0) of 100 μL. The quality and concentration of DNA was tested by 1% agarose gel electrophoresis and nucleic acid concentration meter (Eppendorf Bio-Photometer D30, Eppendorf AG, Hamburg, Germany). The concentration of the various samples was adjusted for consistency, and then samples were stored at −20 °C. 4.6. Methylation-Sensitive Amplified Polymorphism (MSAP) Analysis MSAP analysis of loach was by the MSAP technique reaction system [41] set up by this laboratory. 4.7. Enzyme Cleavage and Adaptor Ligation Genomic DNA was enzyme cleaved by the endonuclease EcoR I/Hpa II and EcoR I/Msp I. A total of 20 μL enzyme cleavage reaction volume included: 800 ng of genomic DNA, 4 μL of 10× buffer Tango, 10 μL of EcoR I, and 10 μL Hpa II or Msp I. The reagents were thoroughly mixed and incubated 8 h at 37 °C. The adaptor ligation reaction comprised: 17 μL of enzyme cleavage product, both 1.5 μL of EcoR I and H-M adaptor (Table 6), 2 μL T4 DNA ligase (Transgen Biotech, Beijing China), 6 μL of 5× T4 buffer, and double-distilled water to 30 μL. Thoroughly mix reagent and incubate ligation for 2 h at 22 °C. 4.8. Preselective Amplification The pre-amplification reaction consisted of: 4 μL adaptor ligation product, both 0.8 μL of primers E-A (10 mmol·L−1) and H-M (10 mmol·L−1) (Table 6), and 1.6 μL of each dNTP (2.5 mmol·L−1), 2 μL 10× PCR buffer (Mg2+ free), 1.2 μL MgCl2 (2.5 mmol·L−1), 0.2 μL Taq DNA polymerase (5 U·μL−1), and double-distilled water to 20 μL. PCR conditions were as follows: initial denaturation for 2 min at 94 °C, followed by 30 cycles of denaturation for 30 s at 94 °C, annealing for 40 s at 56 °C and extension for 60 s at 72 °C, and final extension for 60 s at 72 °C. Product was temporary stored at 4 °C. The quality and concentration of pre-amplification product was tested by 1% agarose gel electrophoresis. Pre-amplification product was diluted 20 times and used as selective amplification template. Residual product was stored at −20 °C. Adaptor and primer combinations are shown in Table 6. 4.9. Selective Amplification Pre-amplification product was diluted 20 times in double-distilled water and was used as the template for selective amplification. The selective amplification reaction consisted of: 2 μL diluted pre-amplification product, both 1.5 μL of E (10 mmol·L−1) and H (10 mmol·L−1) (Table 6), 1.5 μL of each dNTP (2.5 mmol·L−1), 3 μL of 10× PCR buffer (Mg2+ free), 1.2 μL of MgCl2 (2.5 mmol·L−1), 0.2 μL Taq DNA polymerase (5 U·μL−1), and double-distilled water to a final volume of 20 μL. PCR conditions were as follows: initial denaturation for 2 min at 94 °C, followed by 12 cycles of denaturation for 30 s at 94 °C, annealing for 40 s at temperature from 65 to 56 °C (each cycle reduces in 0.7 °C increments) and extension for 60 s at 72 °C, followed by 30 cycles of denaturation for 40 s at 94 °C, annealing for 40 s at 56 °C and extension for 60 s at 72 °C, and final extension for 60 s at 72 °C. Product was temporary stored at 4 °C. Formamide loading buffer (10 μL) was added to the selective amplification product and the mixture was denatured for 5 min at 94 °C followed by incubation in on ice to denature the product immediately. The denatured product was tested by polyacrylamide gel electrophoresis (PAGE) and silver nitrate dying gel. 4.10. MSAP Bands Statistics, Analysis Method Genomic DNA of parents and offspring of different ploidy loach were cleaved by two groups of endonuclease, EcoR I/Hpa II denotes “H”, EcoR I/Msp I denotes “M”. This study chose eight pairs of selective amplification primers, and counted DNA bands on an electropherogram. At the same fragment size between different lanes of one polyacrylamide gel, a visible band was noted as “1”, no band was noted as “0”. Results counted the number of bands of different banding patterns between 100 to 700 bp of product that was amplified by eight pairs of selective primers. According to whether the amplification product bands appeared in the track, methylation band type was classified into four types (Table 7, Figure 2). Type I, both H and M tracks have bands, indicates that the site is non-methylation. Type II, M track has band and H track is without, indicates that the site is methylated inside the DNA double-strand, also known as full methylation. Type III, H track has band and M track is without, indicates that the site is methylated outside the DNA single strand, also known as hemimethylation. Type IV, neither H nor M track have bands, indicates three cases: that site methylation is outside the DNA double-strand, or is inside and outside the DNA double-strand, or the site has no CCGG sequence. The formulas of methylation ratios were as follows: full methylation ratios (%) = full methylation bands/total bands × 100%; hemimethylation ratios (%) = hemimethylation bands/total bands × 100%; total methylation ratios = full methylation ratios + hemimethylation ratios. 4.11. Test of Significance The test used was the Duncan’s multiple range test by Statistical Product and Service Solutions (SPSS) 19.0 (IBM, Chicago, IL, USA). p < 0.05 means significant differences and p < 0.01 means extremely significant difference. Acknowledgments This study was supported in part by a Natural Science Foundation of China (No.31272650) grant to Ya-Juan Li. Author Contributions He Zhou: Established the MSAP reaction system and wrote the paper. Tian-Yu Ma: Hybridized loaches, analyzed MSAP of different ploidy of loaches and data. Rui Zhang: Contributed to MSAP of parents and F1 offspring of 4 × 2 hybrid combination. Qi-Zheng Xu: Contributed to MSAP of parents and F1 offspring of 2 × 4 hybrid combination. Fu Shen: Fed and Hybridized Loaches. Yan-Jie Qin: Tested method and designed the primer of MSAP. Wen Xu: Contributed to MSAP analysis of different ploidy of loaches. Yuan Wang: Hybridized loaches. Ya-Juan Li: Contributed to MSAP, hybridization of loaches, ploidy identification, and modifying the thesis. Conflicts of Interest The authors declare no conflict of interest. Figure 1 DNA-content flow-cytometrical histograms (A–C) and chromosomes (D–F) of diploid (A,D), triploid (B,E), and tetraploid (C,F) loach Misgurnus anguillicaudatus. Flow-cytometrical histograms showing: a diploid cell population with 2C DNA content (A); a triploid cell population with 3C DNA content (B); and a tetraploid cell population with 4C DNA content (C); The Y-axis denotes cell numbers and X-axis denotes channel numbers in each graph. Metaphase spreads of: a diploid cell with 50 chromosomes (D); a triploid cell with 75 chromosomes (E); and a tetraploid cell with 100 chromosomes (F). Scale indicates 10 μm. Figure 2 Genomic DNA MASP patterns of loach DNA of diploid (2n), triploid (3n) and tetraploid (4n) by primer combination E-CC and HM-TC. “H2”, “H3” and “H4”, respectively, indicate that total DNA of diploid (2n), triploid (3n) and tetraploid (4n) was digested by EcoR I/Hpa II; “M2”, “M3” and “M4”, respectively, indicate that total DNA of diploid (2n), triploid (3n) and tetraploid (4n) was digested by EcoR I/Msp I. “Ma” indicates the molecular weight marker (2000-bp ladder). I: Non-methylated sites (1.1); II: Full-methylated sites (0.1); III: Hemimethylated sites (1.0). ijms-17-01299-t001_Table 1Table 1 Analysis of genomic DNA methylation level and variance in different ploidy loach. Ploidy Total Sites Non-Methylated Sites (%) Methylated Sites (%; Mean ± SD) Full Methylated Sites Hemimethylated Sites Total Methylated Sites Diploid 2196 1291 (58.79 ± 6.68) a 516 (23.50 ± 8.82) a 389 (17.71 ± 6.62) a 905 (41.21 ± 6.40) a Triploid 1684 1031 (61.22 ± 7.10) ab 354 (21.02 ± 7.38) a 299 (17.76 ± 5.66) a 653 (38.78 ± 7.10) ab Tetraploid 2012 1267 (63.97 ± 6.23) b 447 (22.22 ± 5.71) a 298 (14.81 ± 4.48) a 745 (37.03 ± 4.83) b Different small letters denote significant differences among different ploidy strains (p < 0.05). ijms-17-01299-t002_Table 2Table 2 Comparison of DNA methylation patterns between triploid and tetraploid loach with diploid loach. Patterns 3n or 4n 2n 3n-2n 4n-2n Hpa II Msp I Hpa II Msp I Number of Patterns and Frequency (%) Number of Patterns and Frequency (%) A 558 (26.94) 653 (31.82) A1 1 1 1 1 359 390 A2 1 0 1 0 67 93 A3 0 1 0 1 132 170 B 493 (23.80) 550 (26.80) B1 1 1 1 0 46 63 B2 1 1 0 1 67 95 B3 1 1 0 0 158 168 B4 1 0 0 0 104 104 B5 0 1 0 0 118 120 C 939 (45.34) 756 (36.84) C1 1 0 1 1 77 65 C2 0 1 1 1 76 102 C3 0 0 1 1 285 220 C4 0 0 1 0 228 171 C5 0 0 0 1 273 198 D 81 (3.91) 93 (4.53) D1 0 1 1 0 38 50 D2 1 0 0 1 43 43 B + C + D 1513 (73.05) 1399 (68.17) Total 2071 2052 ijms-17-01299-t003_Table 3Table 3 Analysis of genomic DNA methylation level and variance in diploid and tetraploid loach and its hybrid F1 generation. Hybrid Combinations Total Sites Non-Methylated Sites (%) Methylated Sites (%; Mean ± SD) Full Methylated Sites Hemimethylated Sites Total Methylated Sites 4 × 2 Obverse cross Tetraploid (Female) 118 86 (72.88 ± 0.00) A 19 (16.10 ± 0.00) A 13 (11.02 ± 0.00) A 32 (27.12 ± 0.00) A Diploid (Male) 103 60 (58.25 ± 0.00) C 23 (22.33 ± 0.00) B 20 (19.42 ± 0.00) B 43 (41.75 ± 0.00) C Offspring Total 2101 1424 (67.78 ± 5.38) B 445 (21.18 ± 5.400) B 232 (11.04 ± 3.89) A 677 (32.22 ± 5.38) B Hybrid Combinations Total Sites Non-Methylated Sites (%) Methylated Sites (%; Mean ± SD) Full Methylated Sites Hemi-Methylated Sites Total Methylated Sites 2 × 4 Inverse cross Diploid (Female) 71 27 (38.03 ± 0.00) aA 21 (29.58 ± 0.00) B 23 (32.39 ± 0.00) B 44 (61.97 ± 0.00) cB Tetraploid (Male) 77 49 (61.25 ± 0.00) cB 17 (21.25 ± 0.00) A 14 (17.50 ± 0.00) A 31 (38.75 ± 0.00) aA Offspring Total 1476 809 (54.81 ± 8.93) bB 355 (24.05 ± 6.04) A 312 (21.14 ± 8.29) A 667 (45.19 ± 9.06) bA Different small letters denote significant differences among different ploidy strains (p < 0.05). Different capital letters denote highly significant differences among different ploidy strains (p < 0.01). ijms-17-01299-t004_Table 4Table 4 Analysis of genomic DNA methylation level and variance in obverse and inverse cross hybrid F1 generation of diploid and tetraploid loach. Methylated Ratios Obverse Cross (4 × 2) F1 Inverse Cross (2 × 4) F1 Full methylated ratios (%) 1.18 ± 5.40 a 4.05 ± 6.04 A Hemimethylated ratios (%) 1.04 ± 3.89 a 1.14 ± 8.29 B Total methylated ratios (%) 2.22 ± 5.38 a 5.19 ± 9.06 B Different small letters denote significant differences among different ploidy strains (p < 0.05). Different capital letters denote highly significant differences among different ploidy strains (p < 0.01). ijms-17-01299-t005_Table 5Table 5 Patterns of cytosine methylation in the female loach, male loach and their hybrid. Patterns Female Male Offspring Difference Patterns Sites and Frequency (%) HpaII MspI HpaII MspI HpaII MspI Obverse Cross (4 × 2) Inverse Cross (2 × 4) A 630 (30.41) 457 (27.17) A1 + + + + + + 405 353 A2 + − + − + − 79 21 A3 − + − + − + 51 6 A4 − − − − − − 95 77 B 353 (17.04) 292 (17.36) B1 + + − + + + 91 65 B2 + + + − + + 61 29 B3 − + + + + + 62 37 B4 + − + − + + 10 27 B5 − + − − + + 3 3 B6 − − + − + + 14 15 B7 − − − + + + 4 14 B8 − − − − + + 2 6 B9 + + − − + + 90 70 B10 + − − − + + 16 26 C 815 (39.33) 672 (39.95) C1 − − − + − − 75 73 C2 − − + − − − 86 96 C3 − + − − − − 61 54 C4 − − + + − − 60 83 C5 + + − − − − 88 79 C6 + − − − − − 66 68 C7 + + + − − − 26 13 C8 − + + + − − 16 14 C9 − + − + − − 43 21 C10 + + + + − − 8 2 C11 + + − + − − 31 16 C12 + − + − − − 20 36 C13 − + + + − + 51 32 C14 + + − + − + 87 57 C15 + + + − + − 40 14 C16 + + + + + − 23 2 C17 + + − − + − 2 2 C18 + + + + − + 23 4 C19 + + − + + − 7 3 C20 + − − − − + 1 3 D 274 (13.22) 261 (15.52) D1 − − + + + + 12 10 D2 − − − − − + 29 30 D3 − − − − + − 13 15 D4 − − + − − + 4 8 D5 − − + + − + 7 13 D6 − + − − + − 1 1 D7 − + + + + − 5 2 D8 − − − + − + 81 78 D9 − − + − + − 28 23 D10 − + − − − + 67 59 D11 + − − − + − 27 22 B + C + D 1442 (69.59) 1225 (72.83) Total 2072 1682 ijms-17-01299-t006_Table 6Table 6 Sequences of adaptors and primers used for methylation-sensitive amplified polymorphism (MSAP). Primer/Adapter Primer/Adapter Sequences 5′–3′ EcoR I adapter EcoR I adapter I CTCGTAGACTGCGTACC EcoR I adapter II AATTGGTACGCAGTC Hpa II-Msp I adapter Hpa II-Msp I adapter I CGAGCAGGACTCATGA Hpa II-Msp I adapter II GATCATGAGTCCTGCT Preselective amplification primers E-A GACTGCGTACCAATTCA H-M ATCCATGAGTCCTGCTCGGC Selective amplification primers E-CT GACTGCGTACCAATTCACT E-CC GACTGCGTACCAATTCACC HM-TC ATCCATGAGTCCTGCTCGGCTC ijms-17-01299-t007_Table 7Table 7 Methylation status of CCGG loci based differential sensitivity of isoschizomers. Type E + H Bands Type E + M Bands Type Status of CCGG Sites Methylation Status of CCGG Sites I 1 1 CCGG Non-methylated GGCC II 0 1 C5mCGG Methylation sites is inside the double-strand of DNA (Full methylated) C5mC5mGG III 1 0 5mCCGG Methylation sites is outside the single-strand of DNA (Hemimethylated) C5mC5mGG IV 0 0 1. C5mC5mGG 1. Methylation site is inside and outside the double-strand of DNA GGC5mC5m 2. C5mCGG 2. Methylation site is outside the double-strand of DNA GGCC5m 3. Inexistence CCGG 3. Inexistence CCGG sequence ==== Refs References 1. Gui J.F. Development and utilization of polyploidy fish Reserv. Fish 1985 3 53 56 2. Ma G. The research progress in chromosome pattern and numerical variation of freshwater fish in China J. Gansu Sci. 1996 8 77 80 3. Comai L. Genetic and epigenetic interactions in allopolyploid plants Plant Mol. Biol. 2000 43 387 399 10.1023/A:1006480722854 10999418 4. Li Z. Lu X. Gao Y. Liu S. Tao M. Xiao H. Qiao Y. Zhang Y. Luo J. Polyploidization and epigenetics Chin. Sci. Bull. 2011 56 245 252 10.1007/s11434-010-4290-1 5. Dowen R.H. Pelizzola M. Schmitz R.J. Lister R. Dowen J.M. Nery J.R. Dixon J.E. Ecker J.R. Widespread dynamic DNA methylation in response to biotic stress Proc. Natl. Acad. Sci. USA 2012 109 E2183 E2191 10.1073/pnas.1209329109 22733782 6. Cao Z.M. Ding W.D. Yu J.H. Cao L.P. Wu T.T. Differences in methylated loci among different grass carp individuals from one pair of parents Acta Zool. Sin. 2007 53 1083 1088 7. Arai K. Polyploidy, clone and atypical reproduction in the loach J. Anim. Genet. 2009 37 59 80 10.5924/abgri.37.59 8. Li K. Li Y. Zhou D. A comparative study of the karyotypes in two species of mud loaches Zool. Res. 1983 4 75 81 9. Li Y.C. Li K. Karyotypes of Opsariichthys bidens and Misgurnusanguillicaudatu s, with a consideration of the relationship between polymorphism of chromosome number and taxonomy of fish J. Wuhan Univ. 1987 1 107 112 10. Li Y.J. Yin J. Wang J.B. Yuan X. Wei J. Sun X.W. Arai K. A study on the distribution of polyploidy loaches in China Nippon Suisan Gakk. 2008 74 177 182 10.2331/suisan.74.177 11. Li Y.-J. Tian Y. Zhang M. Tian P. Yu Z. Abe S. Arai K. Chromosome banding and FISH with rDNA in the diploid and tetraploid loach Misgurnus anguillicaudatus Ichthyol. Res. 2010 57 358 366 10.1007/s10228-010-0168-0 12. Li Y.-J. Yu Z. Zhang M. Qian C. Abe S. Arai K. The origin of natural tetraploid loach Misgurnus anguillicaudatus (Teleostei: Cobitidae) inferred from meiotic chromosome configurations Genetica 2011 139 805 811 10.1007/s10709-011-9585-x 21670989 13. Li Y.-J. Yu Z. Zhang M. Qian C. Abe S. Arai K. Induction and viable gynogenetic progeny using eggs and UV irradiated sperm from the Chinese tetraploid loach, Misgurnus anguillicaudatus Aquac. Int. 2013 21 759 768 10.1007/s10499-012-9551-3 14. Li Y.J. Gao Y.C. Zhou H. Ma H.Y. Li J.Q. Arai K. Meiotic chromosome configurations in triploid progeny from reciprocal crosses between wild-type diploid and natural tetraploid loach Misgurnus anguillicaudatus in China Genetica 2015 143 555 562 10.1007/s10709-015-9853-2 26130482 15. Bian Y. Analysis on Genetic Variation and DNA Methylation of Different Loquat Ploidies Southwest University Chongqing, China 2010 16. Chen C.B. Xue Z.Y. Hu B.Q. Wang C.G. Song W.Q. Detection of the DNA methylation difference between the triploid and diploid polar trees For. Res. 2013 26 69 75 17. Shaked H. Kashkush K. Feldman M. Levy A.A. Sequence elimination and cytosine methylation are rapid and reproducible responses of the genome to wide hybridization and allopolyploidy in wheat Plant Cell 2001 13 1749 1759 10.1105/tpc.13.8.1749 11487690 18. Han F.P. Fedak G. Ouellet T. Liu B. Rapid genomic changes in interspecific and intergeneric hybrids and allopolyploids of Triticeae Genome 2003 46 716 723 10.1139/g03-049 12897878 19. Madlung A. Masuelli R.W. Watson B. Reynolds S.H. Davison J. Comai L. Remodeling of DNA methylation and phenotypic and transcriptional changes in synthetic Arabidopsis allotetraploids Plant Physiol. 2002 129 733 746 10.1104/pp.003095 12068115 20. Cedar H. DNA methylation and gene activity Cell 1988 53 3 4 10.1016/0092-8674(88)90479-5 3280142 21. Chen Z.J. Pikaard C.S. Epigenetic silencing of RNA polymerase I transcripition: A role for DNA methylation and histone modification in nucleolar dominance Genes Dev. 1997 11 2124 2136 10.1101/gad.11.16.2124 9284051 22. Chen Z.J. Pikaard C.S. Transcriptional analysis of nucleolar dominance in polyploid plants: Biased expression/silencing of progenitor rRNA genes is developmentally regulated in Brassica Proc. Natl. Acad. Sci. USA 1997 94 3442 3447 10.1073/pnas.94.7.3442 9096413 23. Zhai X.Q. Zhang X.S. Fan G.Q. Zhao Z.L. Cao X.B. Analysis of diploid and its autotetraploid Paulownia tomentosa with AFLP and MSAP J. Cent. South Univ. For. Technol. 2014 34 89 93 24. Yang L. Li H. Xiang Z.X. Analysis on DNA methylation diversity of diploid and autotetraploid of Stevia rebaudiana Bertoni J. Nucl. Agric. Sci. 2013 27 1125 1130 25. Zhang J. Liu Y. Xia E.H. Yao Q.Y. Liu X.D. Gao L.Z. Analysis of genetic variation and DNA methylation alteration in autotetraploid and diploid rice Chin. J. Appl. Environ. Biol. 2013 19 630 636 26. Wang C.G. Gu Y. Chen C.B. Jiao D.L. Xue Z.Y. Song W.Q. Analysis of the level and pattern of genomic DNA methylation in different ploidy watermelons by MSAP (Citrullus lanatus ) J. Mol. Cell Biol. 2009 42 118 126 27. Hu B.Q. Wang C.G. Fang C.Q. Song W.Q. Lin S.H. Analyzing the levels and patterns of DNA methylation in Yali pear autopolyploid Acta Sci. Natl. Univ. Nankaiensis 2011 44 32 37 28. Li A. Hu B.Q. Xue Z.Y. Chen L. Wang W.X. Song W.Q. Chen C.B. Wang C.G. DNA methylation in genomes of several annual herbaceous and woody perennial plants of varying ploidy as detected by MSAP Plant Mol. Biol. Rep. 2011 29 784 793 10.1007/s11105-010-0280-3 24415835 29. Yin J. Zhao Z.S. Chen X.Q. Li Y.Q. Zhu L.Y. Karyotype comparison of diploid and tetraploid loach, Misgurnus anguillicanudatus Acta HydroBiol. Sin. 2005 29 469 472 30. Nie L.H. Wang Y.Z. Sun Q.X. Fang W.P. Xie D.Y. MSAP analysis of DNA methylation alteration in synthetic hexaploid wheat Acta Agric. Boreali-Sin. 2012 27 77 80 31. Aoki T. Takashima F. Hirono T. DNA Molecular Genetics of Fish Kouseisha Kouseikaku Co., Ltd. Toukyou, Japan 1997 38 41 32. Romagnoli S. Maddaloni M. Livini C. Motto M. Relationship between gene expression and hybrid vigor in primary root tips of young maize (Zea mays L.) plantlets Theor. Appl. Genet. 1990 80 769 775 10.1007/BF00224190 24221107 33. Wang W.X. Genome Variation and DNA Methylation Analysis of Natural and Artificial Triploid Loquats Southwest University Chongqing, China 2008 34. Li A. Liu C. Han C. Zhou X. Chen C. Song W. Wang C. Variation in cytosine methylation patterns between reciprocal hybrids and their parental line in Larix Acta Sci. Natl. Univ. Nankaiensis 2012 45 65 71 35. Verhoeven K.J. van Dijk P.J. Biere A. Changes in genomic methylation patterns during the formation of triploid asexual dandelion lineages Mol. Ecol. 2010 19 315 324 10.1111/j.1365-294X.2009.04460.x 20015141 36. Song C. Liu S J. Xiao J. He W. Zhou Y. Qin Q. Zhang C. Liu Y. Different ploidy crucian carp related molecular biology research Sci. China Life Sci. 2012 55 300 311 37. Yu T. Yang A.G. Wu B. Zhou L.Q. Analysis of Chlamys farreri , Patinopecten yessoensis and their offspring using methylation-sensitive amplification polymorphism (MSAP) J. Fish China 2010 34 1335 1342 38. Tang S.Q. Zhang Y. Xu Q. Sun D.X. Yu Y. Analysis of methylation level of genome in various tissues of different animal species J. Agric. Biotechnol. 2007 14 507 510 39. Zhao X.X. Chai Y. Liu B. Epigenetic inheritance and variation of DNA methylation level and pattern in maize intraspecific hybrids Plant Sci. 2007 172 930 938 10.1016/j.plantsci.2007.01.002 40. Takamiya T. Hosobuchi S. Noguchi T. Asai K. Nakamura E. Habu Y. Paterson A.H. Iijima H. Murakami Y. Okuizumi H. Inheritance and alteration of genome methylation in F1 hybrid rice Electrophoresis 2008 29 4088 4095 10.1002/elps.200700784 18958879 41. Zhou H. Xu W. Li Y. Qin Y. Li X. Ma T. Xiao Y. Ma H. Establishment and optimization of MSAP reaction system for loach Misgurnus anguillicaudatus J. Dalian Ocean Univ. 2014 29 550 555

PMC005xxxxxx/PMC5000697.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081300ijms-17-01300ArticleThe Effects of Female Sexual Hormones on the Expression of Aquaporin 5 in the Late-Pregnant Rat Uterus Csányi Adrienn Bóta Judit Falkay George Gáspár Robert Ducza Eszter *Ishibashi Kenichi Academic EditorDepartment of Pharmacodynamics and Biopharmacy, Faculty of Pharmacy, University of Szeged, H-6720 Szeged, Hungary; [email protected] (A.C.); [email protected] (J.B.); [email protected] (G.F.); [email protected] (R.G.)* Correspondence: [email protected]; Tel.: +36-62-545-56722 8 2016 8 2016 17 8 130027 5 2016 03 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Thirteen mammalian aquaporin (AQP) water channels are known, and few of them play a role in the mammalian reproductive system. In our earlier study, the predominance of AQP5 in the late-pregnant rat uterus was proven. Our current aim was to investigate the effect of estrogen- and gestagen-related compounds on the expression of the AQP5 channel in the late-pregnant rat uterus. Furthermore, we examined the effect of hormonally-induced preterm delivery on the expression of AQP5 in the uterus. We treated pregnant Sprague-Dawley rats subcutaneously with 17β-estradiol, clomiphene citrate, tamoxifen citrate, progesterone, levonorgestrel, and medroxyprogesterone acetate. Preterm delivery was induced by subcutaneous mifepristone and intravaginal prostaglandin E2. Reverse-transcriptase PCR and Western blot techniques were used for the detection of the changes in AQP5 mRNA and protein expressions. The amount of AQP5 significantly increased after progesterone and progesterone analogs treatment on 18 and 22 days of pregnancy. The 17β-estradiol and estrogen receptor agonists did not influence the AQP5 mRNA level; however, estradiol induced a significant increase in the AQP5 protein level on the investigated days of gestation. Tamoxifen increased the AQP5 protein expression on day 18, while clomiphene citrate was ineffective. The hormonally-induced preterm birth significantly decreased the AQP5 level similarly to the day of delivery. We proved that AQP5 expression is influenced by both estrogen and progesterone in the late-pregnant rat uterus. The influence of progesterone on AQP5 expression is more predominant as compared with estrogen. aquaporin 5pregnancyestrogengestagen ==== Body 1. Introduction Aquaporins (AQPs) are small hydrophobic integral membrane proteins which facilitate rapid passive movement of water across the cell membrane [1]. To this day, thirteen AQP isoforms have been identified in mammals, and twelve of them can be found in female and male reproductive tissues in mice, rats, dogs, swine, and humans [2,3]. According to the coding sequence and the permeability features, AQPs are divided into three major subtypes: classical AQPs, aquaglyceroporins, and unorthodox aquaporins. AQP0, 1, 2, 4, 5, and 8 are classical AQPs which are water-selective channels, while AQP3, 7, 9, and 10 can be classified as aquaglyceroporins and are permeable to glycerol, urea, and other small solutes, as well as water [4,5]. AQP11 and 12 belong to the unorthodox aquaporin group, the function of which has not been clearly identified [6,7]. AQPs have an essential role in the female reproductive system, and in the mammalian uterus the AQP1, 2, 3, 4, 5, 7, 8, and 9 isoforms have been detected [2]. The AQP1, 2, 3, and 8 isoforms might be involved in water movement during uterine imbibition, and they may play important roles in hormone-mediated water and other small molecule transport [8]. AQP8 knockout (KO) pregnant mice had a significantly higher number of embryos, and fetal/neonatal weight and the amount of amniotic fluid was greater compared to wild type controls [9]. Contrarily, the number of embryos and the fetal weight decreased in AQP1-KOs. The AQP1-KO placenta demonstrated increased degeneration with evidence of altered blood vessel structure and increased syncytiotrophoblast nodules [10]. AQP5-KO female mice did not show any difference in conception and implantation as compared with wild-type animals [11]. He et al. [12] found that AQP2 expression in human endometrium correlates with serum 17β-estradiol and progesterone levels; therefore, it is menstrual cycle-dependent. Former studies identified the estrogen response element in the promoter region of the AQP2 and 5 genes, which provides the immediate regulation of these water channels by estrogen [13,14]. Water channel regulation was also investigated in the uterus of cycling bitches, and it was discovered that AQP5 was particularly expressed in response to high levels of progesterone [15]. Skowronska et al. [16] also investigated the changes of AQP5 expression due to progesterone and estradiol treatments in an in vitro study on porcine uterus. They found that AQP5 gene expression was down-regulated after the estrogen and progesterone treatments during the mid-luteal phase, but during luteolysis, it was increased by estrogen. A recent experiment was carried out on ovariectomized rats, and they received testosterone, estrogen, or a combination of them. This study determined that testosterone enhanced the expression of AQP5 in the uterus, and this effect was diminished by a following estrogen treatment [17]. They also observed the increase of AQP5 protein expression in ovariectomized rat uterus in response to progesterone alone or in combination with estrogen [18]. It has been postulated that AQPs take part in the processes of fertilization, blastocyst formation, and implantation [19]. AQP1 and 2 are the most concentrated in the endometrium at the time of implantation, suggesting that they may have a physiological role in uterine receptivity [12,20]. Moreover, the cellular and subcellular localizations of amniotic AQPs indicate that the AQPs play distinct functional roles, such as apoptosis for amniotic fluid homeostasis or the tissue remodeling of amniotic membranes during pregnancy. It was proven that AQP1, 3, 8, 9, and 11 play crucial roles in the transfer of water across the placenta [21]. In earlier studies, we determined the expression of AQP1, 2, 3, 5, 8, and 9 isoforms in the pregnant rat uterus. We observed that AQP5 mRNA and protein expression showed the most significant changes during pregnancy. AQP5 expression was the most remarkable on days 18–21 of pregnancy and dramatically dropped on the last day (day 22) of gestation. We determined the effect of oxytocin on the myometrial expression of AQP5. Our results led us to suppose that oxytocin selectively decreases the expression of AQP5 at the end of pregnancy and may be of importance in the initiation of delivery in rats [22]. It is well known that the expression of various AQPs can be regulated by steroid sex hormones [18,23] in female reproductive tissues, but we have limited data concerning the expression of AQP5 under the effect of female sexual hormones in the late-pregnant uterus and the delivery. The primary aim of this study was to investigate the effects of estrogen and progesterone receptors agonists on the expression of AQP5 in the late-pregnant rat uterus. We also wanted to investigate the alteration of uterine AQP5 expression during hormonally-induced preterm birth in rats. 2. Results 2.1. The Effects of Estrogen-Related Compounds on Aquaporin 5 (AQP5) Expression We investigated the effect of 17β-estradiol on AQP5 expression on day 18 and 22 of pregnancy. 17β-estradiol pretreatment did not cause any significant changes in the AQP5 mRNA levels, either on day 18 or day 22 of pregnancy compared to the control (Figure 1A). The protein expression of AQP5 significantly increased both on day 18 (p < 0.001) and day 22 (p < 0.001) of pregnancy, compared to the non-treated control uterus (Figure 1B). The AQP5 mRNA expression did not change as a result of tamoxifen citrate pretreatment on the investigated days (Figure 2A). In contrast, the level of AQP5 protein increased on day 18 (p < 0.0124), but did not change on the last day of pregnancy, compared to the control (Figure 2B). Clomiphene citrate pretreatment did not cause any significant changes in the AQP5 mRNA (Figure 3A) and protein (Figure 3B) levels on day 18 and day 22 of pregnancy, compared to the control. 2.2. The Effects of Gestagen-Related Compounds on AQP5 Expression Progesterone, levonorgestrel, and medroxyprogesterone acetate pretreatment caused a significant increase in AQP5 mRNA and protein levels both on day 18 (p < 0.001) and day 22 (p < 0.001) of pregnancy (Figure 4, Figure 5 and Figure 6). 2.3. The Effect of Hormonally-Induced Preterm Delivery on AQP5 Expression In the hormonally-induced preterm delivery model, a significant decrease of AQP5 mRNA and protein levels was caused on day 20 of pregnancy, compared to the non-treated animals on pregnancy day 20 (p < 0.001). This decrease in AQP5 expression was similar to the last day of pregnancy (Figure 7). 3. Discussion AQPs are detected in the female reproductive tissues, and it has been revealed that they are involved in embryo implantation and endometrial development during pregnancy [2]. In our earlier studies, it was determined that AQP5 expression shows the highest value on day 18 of pregnancy, and a remarkable decrease was found on the last day of pregnancy in rats [22]. It is a known fact that characteristic hormonal changes take place during pregnancy. In humans, the levels of estrogen and progesterone show a constantly growing trend during gestation. At the end of pregnancy the amount of estrogen in blood remains high, while the serum progesterone level moderately declines [24,25]. In pregnant rats, the level of estradiol is constant, and then doubles from day 18 until day 21 of gestation. On the last day of pregnancy, the amount of estradiol shows a slight decrease [26]. The progesterone level increases continuously during pregnancy in rats until day 19 of gestation, at which point it decreases dramatically [27]. It is well known that the expressions of water channels are influenced by sexual hormones. Jablonski et al. [8] proved that the AQP1 channel was slightly estrogen-regulated in the non-pregnant mouse uterus, and that the expression of AQP2 was significantly stimulated by estrogen. Another study confirmed that AQP1, 4, and 5 are excessively expressed in the periimplantation period in mouse uterus, and that AQP5 expression depends on the estrogen stimulation of the progesterone-primed uterus [19]. In the early days of pregnancy, the location of the AQP5 channel changes from the cytoplasm to the apical plasma membrane [20]. As seen in the foregoing, there is limited information about the hormonal effects on AQP5 expression in the pregnant rat uterus. In the present study, we determined a significant increase in AQP5 expression after progesterone pretreatment in the late-pregnant rat uterus. We investigated two progesterone analogs, because levonorgestrel and medroxyprogesterone acetate have a relatively good binding affinity on the progesterone receptor [28,29]. The increase in AQP5 expression was higher in the mRNA level, but the protein expression showed a similar increase after levonorgestrel and medroxyprogesterone acetate treatment, as compared with progesterone treatment. The central concept of molecular biology deals with the transfer of information from DNA via mRNA to proteins. Several biological factors were identified which influence this process, so we cannot always prove a strong correlation between mRNA and protein expression [30]. We suppose that estrogen may increase the stability of AQP5 protein without alteration of mRNA expression, and that this process can lead to the increase in protein expression. In our study, 17β-estradiol pretreatment did not cause any changes in the AQP5 mRNA expression, but the amount of AQP5 protein was significantly increased, both on day 18 and day 22 of pregnancy. In view of this, it appears that estrogen has a significant effect on AQP5 protein expression in the late-pregnant rat uterus. Tamoxifen citrate is a nonsteroidal triphenylethylene derivative with a selective estrogen receptor modulator effect. This means that tamoxifen acts like antiestrogen in breast but estrogen agonist in bone and uterus, and it is used in the treatment of certain estrogen-dependent breast cancer [31]. In the case of tamoxifen citrate, there was a significant increase in the amount of AQP5 protein on day 18 of pregnancy. Clomiphene citrate is a nonsteroidal drug which also has selective estrogen receptor modulator effects. It consists of two isomers which have mixed estrogenic and antiestrogenic effects, and these features depend on the target tissue [32]. Clomiphene citrate did not cause any changes either in the AQP5 mRNA expression or in the AQP5 protein amount. We suppose that clomiphene could not exert its effect on the hypothalamic-pituitary axis during the four-day-long pretreatment. There are different types of animal models for preterm birth, since preterm delivery could have various pathophysiological backgrounds. We used a hormonally-induced model to prove the progesterone effect on AQP5 expression. Preterm delivery was induced by antigestagen mifepristone and prostaglandin E2. Mifepristone induces preterm delivery by blocking the progesterone receptor [33]. Prostaglandin E2 contributes to cervical ripening and the induction of labor [34]. In this model, day 20 was the starting day of preterm birth. We found a marked decrease of AQP5 mRNA and protein levels on day 20, and this change in AQP5 expression was similar to the last day (day 22) of gestation. The explanation for this phenomenon could be that in the preterm delivery, the progesterone level decreases significantly, followed by the reduction of AQP5 expression. In summary, we can conclude that AQP5 expression is influenced by both female hormones with progesterone predominance in the late-pregnant rat uterus. Our preterm birth model studies suggest that the lack of progesterone effect leads to reduced AQP5 expression and may contribute to the initiation of preterm labor. 4. Materials and Methods 4.1. Experimental Animals 4.1.1. Housing and Handling of the Animals The animals were treated in accordance with the European Communities Council Directive (86/609/ECC) and the Hungarian Act for the Protection of Animals in Research (Article 32 of Act XXVIII), and all experiments involving animal subjects were carried out with the approval of the Hungarian Ethical Committee for Animal Research (permission number: IV/198/2013). Sprague-Dawley rats (INNOVO Ltd., Gödöllő, Hungary) were kept at a controlled temperature of 20–23 °C, in relative humidity of 40%–60% and under a 12 h light/dark cycle. The animals were fed a standard rodent pellet diet (INNOVO Ltd., Isaszeg, Hungary), with tap water available ad libitum. 4.1.2. Mating of the Animals Mature female Sprague-Dawley rats (180–200 g) in estrus were collected. The vaginal impedance was measured by an Estrus Cycle Monitor EC40 (Fine Science Tools, Foster City, CA, USA). The appropriate female and male rats (240–260 g) were mated in a special mating cage. In this cage there was a time-controlled movable metal door separating the rats of different sex. Since rats are usually active at night, the separating door was opened in the early morning hours. Four or five hours after the potential copulation, vaginal smears were taken from the female rats and were examined by a microscope at a 1200× magnification. The presence of a copulation plug or the presence of sperm in the native vaginal smear was accepted as proof of the mating. These female animals were separated and regarded as first-day pregnant rats [22,35]. 4.2. In Vivo Treatments of the Rats The 17β-estradiol valerate, tamoxifen citrate, and clomiphene citrate (Sigma-Aldrich, Budapest, Hungary) pretreatment of the pregnant animals was started on day 14 and day 18 of pregnancy. The compounds were dissolved in olive oil. The animals were injected subcutaneously with 1 µg/0.1 mL of 17β-estradiol [35,36], 5 mg/0.1 mL of tamoxifen citrate [37], and 1 mg/0.1 mL of clomiphene citrate [38,39] once a day for four days. On day 18 and 22, the uterine samples were collected and the molecular studies were carried out. The progesterone, levonorgestrel, and medroxyprogesterone acetate (Sigma-Aldrich) pretreatment of the pregnant animals was started on day 11 and 15 of pregnancy. These hormones were dissolved in olive oil and injected subcutaneously every day in a dose of 0.5 mg/0.1 mL of progesterone [35,36] and levonorgestrel [40], and 5 mg/0.1 mL of medroxyprogesterone acetate [41] for seven days. On day 18 and 22, the uterine samples were collected, and the molecular studies were carried out. The duration of the hormonal treatments were determined based on our earlier studies [35,42,43]. The preterm delivery group was treated with mifepristone (Sigma-Aldrich), which was dissolved in olive oil and given subcutaneously in a dosage of 3 mg/0.1 mL on day 19 of pregnancy at 9 A.M. On the same day at 4 P.M., the animals received intravaginal prostaglandin E2 at a dose of 0.5 mg/mL [44]. The preterm birth occurred on the following day (day 20 of pregnancy). Uterine samples were collected after the beginning of the preterm birth, and molecular studies were carried out. 4.3. RT-PCR Studies 4.3.1. Tissue Isolation The rats (250–300 g) were sacrificed by CO2 inhalation. Newborn rats were sacrificed by immediate cervical dislocation. The uterine tissues from pregnant animals (n = 6) (tissue between two implantation sites) were rapidly removed and placed into RNAlater Solution (Sigma-Aldrich). The tissues were frozen in liquid nitrogen and stored at −75 °C until the extraction of total RNA. 4.3.2. Total RNA Preparation Total cellular RNA was isolated by extraction with guanidinium thiocyanate-acid-phenol-chloroform according to the procedure of Chomczynski and Sacchi [45]. After precipitation with isopropanol, the RNA was washed with 75% ethanol and then re-suspended in diethyl pyrocarbonate-treated water. RNA purity was controlled at an optical density of 260/280 nm with BioSpec Nano (Shimadzu, Kyoto, Japan); all samples exhibited an absorbance ratio in the range of 1.6–2.0. RNA quality and integrity were assessed by agarose gel electrophoresis. 4.3.3. Real-Time Quantitative Reverse-Transcriptase PCR Reverse transcription and amplification of the PCR products were performed by using the TaqMan RNA-to-CT-Step One Kit (Life Technologies, Budapest, Hungary) and an ABI StepOne Real-Time cycler. Reverse-transcriptase PCR amplifications were performed as follows: 48 °C for 15 min and 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. The samples of qPCR experiments contained “no-template” control, “absolute” control or RNA samples from non-treated and treated uterus. The generation of specific PCR products was confirmed by melting curve analysis. Table 1 contains the assay IDs for the primers used and the reaction parameters. All samples were run in triplicate. The fluorescence intensities of the probes were plotted against PCR cycle number. The amplification cycle displaying the first significant increase of the fluorescence signal was defined as the threshold cycle (Ct). 4.4. Western Blot Analysis The uterine tissues from pregnant animals (tissue between two implantation sites) were homogenized using a Micro-Dismembrator (Sartorius AG, Goettingen, Germany) and centrifuged at 5000× g for 15 min at 4 °C in RIPA Lysis Buffer System (Santa Cruz Biotechnology, Inc., Dallas, TX, USA), which contains phenylmethylsulfonyl fluoride (PMSF), sodium orthovanadate, and protease inhibitor cocktail. Total protein amounts from supernatant were determined with spectrophotometry (BioSpec-nano, Shimadzu, Japan). Twenty-five micrograms of sample protein per well was subjected to electrophoresis on 4%–12% NuPAGE Bis-Tris Gel in XCell SureLock Mini-Cell Units (Life Technologies). Proteins were transferred from gels to nitrocellulose membranes using the iBlot Gel Transfer System (Life Technologies). The Ponceau S (Sigma-Aldrich) was used to check the standard running and transfer conditions. The blots were incubated overnight on a shaker with AQP5 (35 kDa) and β-actin (43 kDa) polyclonal antibodies (Santa Cruz Biotechnology, Inc., Dallas, TX, USA, diluted 1:200, host: rabbit, specificity: mouse, rat and human) in blocking buffer. Antibody binding was detected with the Western Breeze® Chromogenic immunodetection kit (ThermoFisher Scientific. The AQP5 antibody reacted to several bands, including a band at the size of AQP5 (43 kDa). The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by hormonal or drug treatments. Images were captured with the EDAS290 imaging system (Csertex Ltd., Budapest, Hungary), and the optical density of each immunoreactive band was determined with Kodak 1D Images analysis software. The β-actin was used for protein normalization for this semi-quantitative methods. Optical densities were calculated as arbitrary units after local area background subtraction. 4.5. Statistical Analysis All experiments were carried out on six animals, and the molecular biology studies were repeated three times. Statistical analyses were performed using Prism 5.0 software (Graph Pad Software, Inc., San Diego, CA, USA). ANOVA Dunnett test or two-tailed unpaired t test were used. p < 0.05 was considered as a level of significance. Author Contributions Adrienn Csányi: wrote the manuscript and participated in the experiments. Judit Bóta: has participated in the in vivo experiments and statistical analysis of the results. George Falkay and Robert Gáspár: have participated in the design of the experiments and the writing of the manuscript. Eszter Ducza: has supervised and organized the whole study and manuscript writing as corresponding author. Conflicts of Interest The authors declare no conflict of interest. Abbreviations AQP Aquaporin Figure 1 Results of PCR and Western immunoblotting experiments after 17β-estradiol treatment. The changes of mRNA (A) and protein (B) expression of aquaporin 5 (AQP5) after 17β-estradiol (E) pretreatment in pregnant rat uterus on days 18 and 22. ns > 0.05, *** p < 0.001 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by 17β-estradiol treatment. Figure 2 Results of PCR and Western immunoblotting experiments after tamoxifen treatment. The changes of mRNA (A) and protein (B) expression of AQP5 after tamoxifen citrate (T) pretreatment in pregnant rat uterus on days 18 and 22. ns > 0.05, * p < 0.05 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by tamoxifen treatment. Figure 3 Results of PCR and Western immunoblotting experiments after clomiphene citrate treatment. The changes of mRNA (A) and protein (B) expression of AQP5 after clomiphene citrate (C) pretreatment in pregnant rat uterus on days 18 and 22. ns > 0.05 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by clomiphene citrate treatment. Figure 4 Results of PCR and Western immunoblotting experiments after progesterone treatment. The changes of mRNA (A) and protein (B) expression of AQP5 after progesterone (P) pretreatment in pregnant rat uterus on days 18 and 22. *** p < 0.001 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by progesterone treatments. Figure 5 Results of PCR and Western immunoblotting experiments after levonorgestrel treatment. The changes of mRNA (A) and protein (B) expression of AQP5 after levonorgestrel (L) pretreatment in pregnant rat uterus on days 18 and 22. *** p < 0.001 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by levonorgestrel treatments. Figure 6 Results of PCR and Western immunoblotting experiments after medroxyprogesterone acetate treatment. The changes of mRNA (A) and protein (B) expression of AQP5 after medroxyprogesterone acetate (MPA) pretreatment in pregnant rat uterus on days 18 and 22. *** p < 0.001 as compared with the data on non-treated control uterus. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by MPA treatments. Figure 7 Results of PCR and Western immunoblotting experiments after hormonally-induced preterm delivery. The changes of mRNA (A) and protein (B) expression of AQP5 after the hormonally-induced preterm birth (PB) in pregnant rat uterus and non-treated pregnant rat uterus. *** p < 0.001 as compared with the data on non-treated pregnancy day 20. Each bar denotes the mean ± S.D. n = 6. The bands resulting from non-specific bindings of the polyclonal AQP5-antibody were not changed by hormonal treatments. ijms-17-01300-t001_Table 1Table 1 Parameters of the applied primers and PCR reactions. The real-time reverse transcription polymerase chain reactions were used to determine the changes in mRNA expression. In our studies, the parameters of inventoried TaqMan assays were defined by Life Technologies (ThermoFisher Scientific, Budapest, Hungary). TaqMan Assays Assay ID (ThermoFisher Scientific) Accession Number Assay Location Amplicon Length Annealing Temp. (°C) Reaction Volume (µL) AQP5 Rn00562837_m1 NM_012779.1 473 69 60 20 β-Actin Rn00667869_m1 NM_031144.3 881 91 60 20 ==== Refs References 1. Denker B.M. Smith B.L. Kuhajda F.P. Agre P. Identification, purification, and partial characterization of a novel Mr 28,000 integral membrane protein from erythrocytes and renal tubules J. Biol. Chem. 1988 263 15634 15642 3049610 2. Zhang D. Tan Y.-J. Qu F. Sheng J.-Z. Huang H.-F. Functions of water channels in male and female reproductive systems Mol. Aspects Med. 2012 33 676 690 10.1016/j.mam.2012.02.002 22343019 3. Zhu C. Jiang Z. Bazer F.W. Johnson G.A. Burghardt R.C. Wu G. Aquaporins in the female reproductive system of mammals Front. Biosci. Landmark Ed. 2015 20 838 871 25553483 4. Hara-Chikuma M. Verkman A.S. Physiological roles of glycerol-transporting aquaporins: The aquaglyceroporins Cell. Mol. Life Sci. 2006 63 1386 1392 10.1007/s00018-006-6028-4 16715408 5. Ishibashi K. Kondo S. Hara S. Morishita Y. The evolutionary aspects of aquaporin family AJP Regul. Integr. Comp. Physiol. 2011 300 R566 R576 10.1152/ajpregu.90464.2008 21148473 6. Yakata K. Hiroaki Y. Ishibashi K. Sohara E. Sasaki S. Mitsuoka K. Fujiyoshi Y. Aquaporin-11 containing a divergent NPA motif has normal water channel activity Biochim. Biophys. Acta BBA Biomembr. 2007 1768 688 693 10.1016/j.bbamem.2006.11.005 17178102 7. Itoh T. Rai T. Kuwahara M. Ko S.B.H. Uchida S. Sasaki S. Ishibashi K. Identification of a novel aquaporin, AQP12, expressed in pancreatic acinar cells Biochem. Biophys. Res. Commun. 2005 330 832 838 10.1016/j.bbrc.2005.03.046 15809071 8. Jablonski E.M. McConnell N.A. Hughes F.M. Huet-Hudson Y.M. Estrogen regulation of aquaporins in the mouse uterus: Potential roles in uterine water movement Biol. Reprod. 2003 69 1481 1487 10.1095/biolreprod.103.019927 12855592 9. Sha X.Y. Xiong Z.F. Liu H.S. Zheng Z. Ma T.H. Pregnant phenotype in aquaporin 8-deficient mice Acta Pharmacol. Sin. 2011 32 840 844 10.1038/aps.2011.45 21602842 10. Zheng Z. Liu H. Beall M. Ma T. Hao R. Ross M.G. Role of aquaporin 1 in fetal fluid homeostasis J. Matern. Fetal Neonatal Med. 2014 27 505 510 10.3109/14767058.2013.820697 23808411 11. Zhang Y. Chen Q. Zhang H. Wang H. Ma T. Qiao J. Duan E. Aquaporin-dependent excessive intrauterine fluid accumulation is a major contributor in hyper-estrogen induced aberrant embryo implantation Cell Res. 2015 25 139 142 25342561 12. He R.-H. Sheng J.-Z. Luo Q. Jin F. Wang B. Qian Y.-L. Zhou C.-Y. Sheng X. Huang H.-F. Aquaporin-2 expression in human endometrium correlates with serum ovarian steroid hormones Life Sci. 2006 79 423 429 16483614 13. Zou L.-B. Zhang R.-J. Tan Y.-J. Ding G.-L. Shi S. Zhang D. He R.-H. Liu A.-X. Wang T.-T. Leung P.C.K. Identification of estrogen response element in the aquaporin-2 gene that mediates estrogen-induced cell migration and invasion in human endometrial carcinoma J. Clin. Endocrinol. Metab. 2011 96 E1399 E1408 10.1210/jc.2011-0426 21715543 14. Kobayashi M. Takahashi E. Miyagawa S. Watanabe H. Iguchi T. Chromatin immunoprecipitation-mediated target identification proved aquaporin 5 is regulated directly by estrogen in the uterus Genes Cells 2006 11 1133 1143 10.1111/j.1365-2443.2006.01009.x 16999734 15. Aralla M. Borromeo V. Groppetti D. Secchi C. Cremonesi F. Arrighi S. A collaboration of aquaporins handles water transport in relation to the estrous cycle in the bitch uterus Theriogenology 2009 72 310 321 10.1016/j.theriogenology.2009.01.023 19395011 16. Skowronska A. Młotkowska P. Wojciechowicz B. Okrasa S. Nielsen S. Skowronski M.T. Progesterone, estradiol, arachidonic acid, oxytocin, forskolin and cAMP influence on aquaporin 1 and 5 expression in porcine uterine explants during the mid-luteal phase of the estrous cycle and luteolysis: An in vitro study Reprod. Biol. Endocrinol. 2015 13 7 10.1186/s12958-015-0004-5 25884220 17. Salleh N. Mokhtar H.M. Kassim N.M. Giribabu N. Testosterone induces increase in aquaporin (AQP)-1, 5, and 7 expressions in the uteri of ovariectomized rats J. Membr. Biol. 2015 248 1097 1105 10.1007/s00232-015-9823-8 26198330 18. Lindsay L.A. Murphy C.R. Redistribution of aquaporins 1 and 5 in the rat uterus is dependent on progesterone: A study with light and electron microscopy Reproduction 2006 131 369 378 10.1530/rep.1.00914 16452730 19. Richard C. Gao J. Brown N. Reese J. Aquaporin water channel genes are differentially expressed and regulated by ovarian steroids during the periimplantation period in the mouse Endocrinology 2003 144 1533 1541 10.1210/en.2002-0033 12639938 20. Lindsay L.A. Murphy C.R. Redistribution of aquaporins in uterine epithelial cells at the time of implantation in the rat Acta Histochem. 2004 106 299 307 10.1016/j.acthis.2004.05.004 15350812 21. Kobayashi K. Yasui M. Cellular and subcellular localization of aquaporins 1, 3, 8, and 9 in amniotic membranes during pregnancy in mice Cell Tissue Res. 2010 342 307 316 10.1007/s00441-010-1065-6 20957496 22. Ducza E. Seres A.B. Hajagos-Tóth J. Falkay G. Gáspár R. Oxytocin regulates the expression of aquaporin 5 in the late-pregnant rat uterus: Alterations in AQPs in the late-pregnant rat uterus Mol. Reprod. Dev. 2014 81 524 530 10.1002/mrd.22320 24644013 23. Lindsay L.A. Murphy C.R. Aquaporins are upregulated in glandular epithelium at the time of implantation in the rat J. Mol. Histol. 2007 38 87 95 10.1007/s10735-007-9083-8 17342345 24. Boroditsky R.S. Reyes F.I. Winter J.S. Faiman C. Maternal serum estrogen and progesterone concentrations preceding normal labor Obstet. Gynecol. 1978 51 686 691 566408 25. Tulchinsky D. Hobel C.J. Yeager E. Marshall J.R. Plasma estrone, estradiol, estriol, progesterone, and 17-hydroxyprogesterone in human pregnancy. I. Normal pregnancy Am. J. Obstet. Gynecol. 1972 112 1095 1100 10.1016/0002-9378(72)90185-8 5025870 26. Shaikh A.A. Estrone and estradiol levels in the ovarian venous blood from rats during the estrous cycle and pregnancy Biol. Reprod. 1971 5 297 307 5163522 27. Morishige W.K. Pepe G.J. Rothchild I. Serum luteinizing hormone, prolactin and progesterone levels during pregnancy in the rat Endocrinology 1973 92 1527 1530 10.1210/endo-92-5-1527 4695432 28. Schindler A.E. Campagnoli C. Druckmann R. Huber J. Pasqualini J.R. Schweppe K.W. Thijssen J.H. Classification and pharmacology of progestins Maturitas 2003 46 7 16 10.1016/j.maturitas.2003.09.014 12963166 29. Philibert D. Bouchoux F. Degryse M. Lecaque D. Petit F. Gaillard M. The pharmacological profile of a novel norpregnane progestin (trimegestone) Gynecol. Endocrinol. 1999 13 316 326 10.3109/09513599909167574 10599548 30. Maier T. Güell M. Serrano L. Correlation of mRNA and protein in complex biological samples FEBS Lett. 2009 583 3966 3973 10.1016/j.febslet.2009.10.036 19850042 31. Dutertre M. Smith C.L. Molecular mechanisms of selective estrogen receptor modulator (SERM) action J. Pharmacol. Exp. Ther. 2000 295 431 437 11046073 32. Dickey R.P. Clomiphene citrate for women with unexplained infertility Unexplained Infertility Schattman G.L. Esteves S.C. Agarwal A. Springer New York New York, NY, USA 2015 261 271 33. Telleria C.M. Deis R.P. Reproductive function in rats after mifepristone-induced termination of pregnancy Contraception 1996 53 185 190 10.1016/0010-7824(96)00009-1 8689885 34. Thomas J. Fairclough A. Kavanagh J. Kelly A.J. Vaginal prostaglandin (PGE2 and PGF2a) for induction of labour at term Cochrane Database of Systematic Reviews Cochrane Pregnancy and Childbirth Group John Wiley & Sons, Ltd. Chichester, UK 2014 35. Bóta J. Hajagos-Tóth J. Ducza E. Samavati R. Borsodi A. Benyhe S. Gáspár R. The effects of female sexual hormones on the expression and function of α1A- and α1D-adrenoceptor subtypes in the late-pregnant rat myometrium Eur. J. Pharmacol. 2015 769 177 184 10.1016/j.ejphar.2015.11.015 26593425 36. Hódi Á. Földesi I. Ducza E. Hajagos-Tóth J. Seres A.B. Klukovits A. Gáspár R. Tocopherol inhibits the relaxing effect of terbutaline in the respiratory and reproductive tracts of the rat: The role of the oxidative stress index Life Sci. 2014 105 48 55 24780318 37. Cyr M. Modulation by estrogen-receptor directed drugs of 5-hydroxytryptamine-2A receptors in rat brain Neuropsychopharmacology 2000 23 69 78 10.1016/S0893-133X(00)00085-3 10869887 38. Nutu M. Feng Y. Egecioglu E. Weijdegård B. Stener-Victorin E. Shao R. Stromal cell-specific apoptotic and antiestrogenic mechanisms may explain uterine defects in humans after clomiphene citrate therapy Am. J. Obstet. Gynecol. 2010 203 65.e1 65.e10 10.1016/j.ajog.2010.03.039 20494331 39. Bharti S. Misro M.M. Rai U. Clomiphene citrate potentiates the adverse effects of estrogen on rat testis and down-regulates the expression of steroidogenic enzyme genes Fertil. Steril. 2013 99 140 148 10.1016/j.fertnstert.2012.08.050 22999797 40. Graham B.M. Milad M.R. Blockade of estrogen by hormonal contraceptives impairs fear extinction in female rats and women Biol. Psychiatry 2013 73 371 378 10.1016/j.biopsych.2012.09.018 23158459 41. Brandon J.M. Raval P.J. Interaction of estrogen and histamine during ovum implantation in the rat Eur. J. Pharmacol. 1979 57 171 177 10.1016/0014-2999(79)90362-5 39766 42. Hajagos-Toth J. Bota J. Ducza E. Csanyi A. Tiszai Z. Borsodi A. Samavati R. Benyhe S. Gaspar R. The effects of estrogen on the α2-adrenergic receptor subtypes in rat uterine function in late pregnancy in vitro Croat. Med. J. 2016 57 100 109 10.3325/cmj.2016.57.100 27106352 43. Hajagos-Tóth J. Bóta J. Ducza E. Samavati R. Borsodi A. Benyhe S. Gáspár R. The effects of progesterone on the α2-adrenergic receptor subtypes in late-pregnant uterine contractions in vitro Reprod. Biol. Endocrinol. 2016 14 33 10.1186/s12958-016-0166-9 27301276 44. Elovitz M.A. Mrinalini C. Animal models of preterm birth Trends Endocrinol. Metab. 2004 15 479 487 10.1016/j.tem.2004.10.009 15541647 45. Chomczynski P. Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction Anal. Biochem. 1987 162 156 159 10.1016/0003-2697(87)90021-2 2440339

PMC005xxxxxx/PMC5000698.txt

==== Front Int J Mol SciInt J Mol SciijmsInternational Journal of Molecular Sciences1422-0067MDPI 10.3390/ijms17081301ijms-17-01301ArticleCharacterizing the Effects of Washing by Different Detergents on the Wavelength-Scale Microstructures of Silk Samples Using Mueller Matrix Polarimetry Dong Yang 12†He Honghui 1†He Chao 12Zhou Jialing 12Zeng Nan 1Ma Hui 13*Hardy John G. Academic EditorHolland Chris Academic Editor1 Shenzhen Key Laboratory for Minimal Invasive Medical Technologies, Institute of Optical Imaging and Sensing, Graduate School at Shenzhen, Tsinghua University, Shenzhen 518055, China; [email protected] (Y.D.); [email protected] (H.H.); [email protected] (C.H.); [email protected] (J.Z.); [email protected] (N.Z.)2 Department of Biomedical Engineering, Tsinghua University, Beijing 100084, China3 Department of Physics, Tsinghua University, Beijing 100084, China* Correspondence: [email protected]; Tel./Fax: +86-755-2603-6238† These authors contributed equally to this work. 10 8 2016 8 2016 17 8 130127 6 2016 04 8 2016 © 2016 by the authors; licensee MDPI, Basel, Switzerland.2016This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).Silk fibers suffer from microstructural changes due to various external environmental conditions including daily washings. In this paper, we take the backscattering Mueller matrix images of silk samples for non-destructive and real-time quantitative characterization of the wavelength-scale microstructure and examination of the effects of washing by different detergents. The 2D images of the 16 Mueller matrix elements are reduced to the frequency distribution histograms (FDHs) whose central moments reveal the dominant structural features of the silk fibers. A group of new parameters are also proposed to characterize the wavelength-scale microstructural changes of the silk samples during the washing processes. Monte Carlo (MC) simulations are carried out to better understand how the Mueller matrix parameters are related to the wavelength-scale microstructure of silk fibers. The good agreement between experiments and simulations indicates that the Mueller matrix polarimetry and FDH based parameters can be used to quantitatively detect the wavelength-scale microstructural features of silk fibers. Mueller matrix polarimetry may be used as a powerful tool for non-destructive and in situ characterization of the wavelength-scale microstructures of silk based materials. silkmicrostructurepolarizationMueller matrix ==== Body 1. Introduction Silk based materials have been applied to many areas such as medicine [1], biotechnology [2,3], biomaterials [4,5], fine chemical industry [6] and so on. Silk consisting of fibroin and sericin may suffer from microstructural changes due to hostile external environmental conditions such as PH, temperature, or exposure to ultraviolet light [7,8,9,10]. In our daily life, silk fabrics are often washed using different detergents, which may have a different influence on the microstructure of silk. Recently, many techniques have been used to study the physics and chemistry properties of silk fibers, such as X-ray diffraction, scanning electron microscopy, nuclear magnetic resonance, infrared spectroscopy, Raman spectroscopy and so on. It will be useful if we have non-destructive and real-time techniques to quantitatively monitor the structural variations of silk based materials under different conditions. Mueller matrix polarimetry has been recognized as a potentially powerful technique for probing the wavelength-scale microstructural and optical properties of complex scattering samples [11,12]. Mueller matrix imaging techniques have many distinctive advantages for non-invasive and in-situ applications on biological specimens or other delicate samples [13,14,15,16]. Most of the existing non-polarization optical instrumentations can be upgraded for Mueller matrix measurements by adding proper polarization state generator (PSG) and polarization state analyzer (PSA) to the existing optical path [17]. Comparing with non-polarization measurements, a Mueller matrix contains much richer wavelength-scale microstructural information, which can be used to characterize the structural features of the scattering media, such as polymer materials, and biomedical samples [14,15,16,17]. Mueller matrix polarimetry may also serve as a powerful tool to distinguish the characteristic features of different textile samples, such as acetate, cotton, silk and ramie [18]. In recent studies, Mueller matrix imaging parameters have been used to provide comprehensive characterization of the polarization features to assist various pathological detections, such as skin cancer [19], cervical cancer [20], colon cancer [21], liver fibrosis [22], and so on [23,24]. In this paper, we use Mueller matrix polarimetry for quantitative characterization of wavelength-scale microstructures of silk samples. The silk samples are washed several times using different detergents, such as fabric softener, laundry powder, toilet soap and color stain net. The backscattering Mueller matrix images are taken after each washing process. The 2D images of the 16 Mueller matrix elements are then reduced to their frequency distribution histograms (FDHs) and central moments. According to the features of silk fibers, we also propose new parameters based on the central moments of FDHs to quantitatively characterize the wavelength-scale microstructural changes of the silk samples washed by different detergents. For a deeper understanding of the relationship between the Mueller matrix parameters and the microstructural variations of silk, we also perform the X-ray diffraction and scanning electron microscope measurements then carry out Monte Carlo (MC) simulations based on the sphere-cylinder scattering model (SCSM) [25]. The good agreement between experiments and simulations indicates that the Mueller matrix imaging and FDH based parameters can be used to obtain quantitatively the wavelength-scale microstructural features of silk samples. Mueller matrix polarimetry may be used as a powerful tool for non-destructive and real-time characterization of the wavelength-scale microstructures of silk based materials. 2. Results and Discussion 2.1. 2D Images of Mueller Matrix Elements Figure 1 and Figure 2 show the experimental results of 2D Mueller matrix images for the silk samples washed by fabric softener and color stain net, respectively. All the matrix elements are normalized by the m11. As we have learned from previous studies [20,26], the following structural and optical information of the silk samples before and after washing can be obtained from the Mueller matrix elements: First, in Figure 1 and Figure 2, the differences between the diagonal m22 and m33 elements, and the magnitudes of the off-diagonal elements indicate that the silk sample is highly anisotropic. Experimental and simulated results have demonstrated that a larger difference between the m22 and m33 means a more prominent anisotropy [20,26]. It can be observed from Figure 1 that with more washing time using fabric softener, the Mueller matrix of the silk sample remains almost the same, while Figure 2 shows that the washing process using color stain net significantly changes the Mueller matrix: the decreasing difference between the m22 and m33 indicates that the well-ordered anisotropic silk fibers may be destroyed. Second, in Figure 1a and Figure 2a the m12, m21 elements are both positive, while the m13, m31 elements are approximately zero, which demonstrates that the orientation of the silk fibers is mostly along the 0 degree direction [12]. Figure 2b,c show that the orientation of the silk fibers washed by color stain net becomes disordered. Third, the diagonal elements m22, m33 and m44 represent the depolarization ability [26]. Figure 1 shows that when washed by fabric softener, the depolarization ability of the silk samples almost remains constant, while in Figure 2 the decrease of the diagonal elements indicates that the depolarization ability of the silk samples increases as the washing times using color stain net rise. 2.2. Frequency Distribution Histogram (FDH) of Mueller Matrix Elements From the analysis of the 2D Mueller matrix images in Section 3.1, we can obtain a lot of information on how the wavelength-scale microstructure of the silk samples varies during the washing processes. However, these features revealed directly by the 2D images are often qualitative and sometimes too detailed. In this section, we adopt the FDH technique to transform the 2D Mueller matrix images into quantitative parameters, which can characterize the dominant wavelength-scale microstructural features of silk samples. Figure 3 and Figure 4 show the FDH curves of the 2D Mueller matrix images represented in Figure 1 and Figure 2, respectively. For more information, we also calculate the central moment parameters of the FDHs of the silk samples washed by fabric softener (Table 1), laundry powder (Table 2), toilet soap (Table 3) and color stain net (Table 4). Compared to the 2D images, the FDHs and their corresponding central moments of the Mueller matrix elements reveal the main wavelength-scale structural features of the silk samples more clearly and quantitatively [26]. The FDHs of the Mueller matrices shown in Figure 3 and Figure 4 are non-diagonal, and the curves of diagonal elements m22 and m33 are different, meaning that the silk sample is anisotropic. As the washing times increase, the difference between the FDHs of the m22 and m33 changes, indicating the anisotropy degree of the silk sample varies. For a closer observation of the anisotropy variations, we then analyze the central moment parameters shown in Table 1, Table 2, Table 3 and Table 4. It can be seen that the difference of mean values (P1) between the m22 and m33 varies slightly for the silk samples washed by fabric softener and laundry powder (the difference variations are 0.497 to 0.443 for Table 1 and 0.505 to 0.438 for Table 2, respectively), indicating the anisotropy degree stays relatively stable. Conversely, the difference between the P1 values of the m22 and m33 becomes more and more obvious as the washing times increase when using toilet soap and color stain net (the difference variations are 0.521 to 0.246 for Table 3 and 0.594 to 0.259 for Table 4, respectively), showing that washing by toilet soap and color stain net causes prominent reductions in the anisotropy degree of the silk fibers. Therefore, we can divide the four types of detergents into two groups: fabric softener and laundry powder denoted as group A, toilet soap and color stain net denoted as group B. Figure 3 and Table 1 and Table 2 also reveal that, for the silk samples washed by group A detergents, the central moment parameters of the off-center elements stay almost the same (the maximum variation is less than 0.006). It confirms that during the washing processes, the group A detergents have limited influence on the fibrous structures of the silk samples. However, Figure 4, Table 3 and Table 4 indicate that after washing by the group B detergents, the FDHs and central moment parameters of the off-diagonal Mueller matrix elements changed more prominently. For instance, the maximum change of the P1 value for the m12 element shown in Table 4 is larger than 0.02, demonstrating that the anisotropy degree of the silk samples has been changed by the group B detergents. It can also be seen from Figure 4 and Table 4 that, after washing by color stain net the variance of the FDHs, or the values of parameter P2 increase significantly. For example, in Table 4 after the sixth wash the value of P2 for the m22 element increases more than 4 times compared to the unwashed silk sample, whereas in Table 1, Table 2 and Table 3 the values of P2 keep relatively constant. It means the color stain net may seriously damage the well-ordered fibrous structures, distributing the fiber orientation over a wider range. We can see from Tables S1–S4 that the values of P3 and P4 also represent some changes, but more studies are still needed to reveal their structural meanings. 2.3. Mueller Matrix Parameters of Silk Sample Based on FDHs From the figures and tables above we can see that there is abundant structural information encoded in the FDHs of Mueller matrix elements. To obtain more quantitative information on the silk from the Mueller matrix elements, we propose three new parameters based on the FDHs and their central moments. As shown in Section 2.2, the parameter c1p1 is an indicator for the difference between the values of diagonal m22 and m33 elements, which is positively related to the anisotropy degree of the silk sample [26]. The parameter d22p1 is an indicator for the changes of the mean value of the m22 elements, which is negatively related to the depolarization ability of the silk sample [26]. The parameter d23p2 is an indicator for the distribution width of the values of the m23 element, which is related to the degree of order of the silk sample. Figure 5 shows how the new FDH parameters c1p1, d22p1 and d23p2 vary with the washing time for the groups A (Figure 5a–c) and B (Figure 5d,e) detergents. What we can learn from Figure 5a is that, as the washing time increases, the parameter c1p1 almost remains the same around 0.9 when the silk samples are washed by the group A detergents, demonstrating that the silk samples maintain their large anisotropy degree. Moreover, Figure 5b,c show that the values of parameters d22p1 and d23p2 change mildly and fluctuate within a narrow range, indicating that the depolarization ability and degree of order of the silk samples change slightly during the washings. On the other hand, Figure 5d–f reveal that compared with the samples washed by the group A detergents, the group B detergents have more influence on the silk samples: the parameters c1p1 decreases for all samples in group B, which corresponds to reductions in the anisotropy of the silk samples. However, we also see that the other two parameters of the silk samples washed by toilet soap and color stain net have different variation features. For toilet soap, d22p1 is always negative and d23p2 decreases slightly. For color stain net, d22p1 drops from positive to negative and d23p2 increases significantly. The results shown in Figure 5 indicate that for the silk samples washed by the group B detergents, their wavelength-scale microstructures may have different characteristic variations. In summary we can get the following information from the analysis above: (1) The change trends of the parameters c1p1, d22p1 and d23p2 are different for the groups A and B, so they may be used to assess if a detergent affects the wavelength-scale microstructure of silk fibers; (2) In the group B, the different change trends of d22p1 and d23p2 show that the parameters may provide the details of the wavelength-scale microstructural variations of the silk samples. 2.4. Wavelength-Scale Microstructural Variations of Silk Samples In order to study the relationship between the parameters used in Section 3.3 and the characteristic wavelength-scale microstructural variations of the silk samples, we observed the silk strands, which are consisted of tens of silk fibers, under an optical microscope and a scanning electron microscope. Figure 6 shows the optical and scanning electron microscopic images of the unwashed silk strand (Figure 6a,d) and the strands washed six times by group A detergents: fabric softener (Figure 6b,e) and laundry powder (Figure 6c,f). Figure 7 shows the optical and scanning electron microscopic images of silk strands before washing (Figure 7a,d) and washed six times by group B detergents: toilet soap (Figure 8b,e) and color stain net (Figure 7c,f). From Figure 6 and Figure 7, we can see that the wavelength-scale microstructures of the silk strands almost stay the same in group A, whereas they change prominently in group B. It confirms that the three new parameters based on FDHs can be used to assess the influences of the detergents on the wavelength-scale microstructures of silk samples. Moreover, in our previous discussion, we have found that for the silk samples in group B, the parameters d22p1 and d23p2 represent different change trends. This indicates that the parameters may be not only a kind of indicator to reveal whether the wavelength-scale microstructural changes occurred, but also a tool to characterize what kind of structural changes happened. It can be seen from Figure 7 that the structural changes are not the same for the silk fibers washed by toilet soap and color stain net, which demonstrates the assumption. 2.5. Monte Carlo Simulations For a better confirmation of the assumption in Section 3.3, we carry out MC simulations based on the SCSM to interpret the relationships between the characteristic wavelength-scale microstructures and the Mueller matrix parameters. Since the simulated results have no obvious variance and the parameters almost do not change in group A, we focus on the analysis of the simulated parameters c1p1 and d22p1 in group B. According to the microscopic images in Figure 7 we build two models to study the contrast mechanisms of the experimental results. It can be observed from Figure 7b,e that compared to the unwashed silk fibers shown as Figure 7a,d, after six washings by toilet soap there are more bumps and burrs on the surface of the silk fibers. Therefore, we raised the sphere/cylinder ratio in the SCSM to mimic the increasing percentage of the small fragments and particles during the washing process. The parameters in the MC simulation are set as follows for the silk samples: The thickness of the medium is 0.01 cm. The diameters of the cylindrical and spherical scatterers are 1.5 µm and 0.2 µm to mimic the single silk fibers and the roughness on the silk surfaces or particles embedded among the fibers [25,27]. The refractive indices of the scatterers and interstitial medium are 1.56 and 1, respectively [25]. The cylindrical scatterers are distributed along the x-axis, and the standard deviation of their angular distribution is 15 degree. The toilet soap is alkaline, and has dehydration condensation reaction with −COOH group of amino acid. Silk is made of fibroin and sericin which are composed of amino acids. Therefore, the reactions between the silk fibers and toilet soap solution generate peptide bonds to make the microstructure of silk compact and may produce some small particles attached to the surface of silk. Meanwhile, these reactions could alter the surface structures to generate burrs and bulges. To simulate the changes in the wavelength-scale microstructures for the silk fibers after washing by toilet soap, we vary the sphere/cylinder ratio from 10:70 to 13:67, 15:65, 21:59, 22:58, 27:53 and 28:52 in the MC simulation. As shown in Figure 8a, the MC simulated parameters c1p1 and d22p1 both decrease as the washing times increase, and the values of d22p1 are always negative. The simulated results are consistent with the experimental observations shown in Figure 5. It is shown that the increasing surface roughness can reduce the anisotropy of the silk fibers, which can be reflected by the Mueller matrix parameters. On the other hand, we can see from the microscopic images in Figure 7c,f that after washing by color stain net the large silk strand is divided into separated disordered silk fibers. The color stain net, whose primary ingredient is hydrogen peroxide H2O2 with strong oxidizing property, can oxidize protein to change the properties of the silk strands. The oxidize reactions dissolve the components of the silk, making the wavelength-scale microstructures of the silk fibers loose and disordered. In the MC simulations we reduce the diameters and increase the standard deviations of angular distribution of the cylinders at the same time, from 1.5 µm 15 degree to 1 µm 15 degree, 0.9 µm 19 degree, 0.86 µm 19 degree, 0.76 µm 20 degree, 0.68 µm 21 degree and 0.68 µm 22 degree. It can be observed that there is also a good agreement between the simulated result shown in Figure 8b and the experimental observations shown in Figure 5, meaning that during the washing process the silk strand breaks into finer fibers distributed in a wider orientation range. Based on the MC simulations using different wavelength-scale microstructural models, we can conclude that the contrast mechanism of the FDH parameters is the rising number of small particles and increasing surface roughness for the silk samples after washing by toilet soap, or the increasing disorder of the silk fibers with reduced diameters after washing by color stain net, which can be distinguished by the parameters d22p1 and d23p2 shown in Figure 5. The discovery in this study provides a potential method for characterizing the wavelength-scale microstructure of silk based materials. 2.6. Results of X-ray Diffraction and Discussions The experimental and Monte Carlo simulated results shown above indicate that the Mueller matrix imaging parameters are sensitive to the wavelength-scale structural changes, which may be induced by the molecular microstructural variations resulted from the detergents with different chemical components. For confirmation, we perform the X-ray diffraction (XRD) measurements on the silk samples. As shown in Figure 9a, there are no obvious differences in the diffraction peaks for the silk samples washed by group A detergents, indicating that their macromolecular microstructures almost do not change. On the other hand, Figure 9b reveals that after washings by group B detergents, the macromolecular microstructural features of the silk samples vary: (1) For the XRD curve 2, the diffraction peak near 9.1 degree is enhanced compared to the XRD curve 1. It can also be observed that there is a new diffraction peak near 5.7 degree, indicating a new conformation of the silk protein may be produced after washing by toilet soap [28]; (2) For the XRD curve 3, the diffraction peaks of 20.48 degree and 9.1 degree are obviously weakened compared to the XRD curve 1, meaning that the crystallinity of the silk protein may decrease after washing by color stain net [29]. In summary, the XRD measurements show that the group A detergents have very little influence on the macromolecular microstructures of the silk samples, while the group B detergents may alter the conformation and crystallinity of the silk protein, leading to increasing disorder and surface roughness of the silk fibers, which can be reflected by the Mueller matrix imaging parameters shown as Figure 5. 3. Materials and Methods 3.1. Experimental Setup and Silk Samples We adopt a typical experimental setup based on the dual rotating retarder configuration for backscattering Mueller matrix measurements [30,31]. As shown in Figure 10a, the polarization states of the incident light from the LED (633 nm, 3 W) are controlled by a linear polarizer (P1, extinction ratio > 1000:1, Daheng Optic, Beijing, China) and quarter-wave plate (R1, Daheng Optic). Then the collimated light of different polarization states illuminates the sample, and the backscattered photons pass through another quarter-wave plate (R2, Daheng Optic) and linear polarizer (P2, extinction ratio > 1000:1, Daheng Optic). Finally the photons are collected by a CCD camera (QImaging 32-0122A, 12 bit, Surrey, BC, Canada). There is an oblique angle (θ = 15°) between the illuminating and detection arms to avoid the surface reflection from the sample. During the measurements, the polarizers (P1, P2) are fixed in the horizontal direction, and the two retarders (R1, R2) rotate with a fixed rate θ1 = 5θ2. The Fourier series intensities are given by Equation (1). (1) I=α0+∑n=112(αncos 2nθ1+βnsin 2nθ1) In this paper, after 30 rotations of both retarders, the Mueller matrix elements can be calculated by using the Fourier coefficients αn and βn shown as Equation (1) [31]. Before applying to the samples, we calibrate the experimental setup by measuring the Mueller matrices of standard samples such as air. The maximum errors for the absolute values of all the Mueller matrix elements are smaller than 0.01 after calibration. In order to analyze the influence of different detergents on the wavelength-scale microstructures of silk based materials, we prepared a simpler silk sample by wrapping the silk fibers (provided by Guangxi Institute of Supervision and Testing on Product Quality, Nanning, Guangxi, China) around a glass slide as shown in Figure 10b. Then the well ordered silk fibers are washed six times using four common detergents: fabric softener (Comfort, Q/YQXA306, Unilever, Hefei, China), laundry powder (Diaopai, WL-AGB/T13171.2-2009, Nice Group China, Lishui, China), toilet soap (Safeguard, Q/GZBJ8II, Procter & Gamble, Tianjin, China) and color stain net (Bluemoon, Q/LYLZG20II, Bluemoon China, Guangzhou, China). The silk fibers around the glass slide are soaked and washed in the detergent solution for 30 min. Then they are completely dried out. During the washing processes other factors are kept the same, such as the water temperature, soaking time, drying time and so on. There are no evident contractions in the silk sample. After each washing, we take the 2D backscattering Mueller matrix images of the silk samples and examine the relationship between the Mueller matrix parameters and the wavelength-scale microstructural variations of washed silk fibers. 3.2. Frequency Distribution Histograms (FDHs) and Quantitative Parameters As a comprehensive description of polarization properties, Mueller matrix contains abundant structural information of samples. However, the relationships between the features of certain wavelength-scale microstructures and the Mueller matrix elements are not always clear, increasing the difficulties in applying Mueller matrix polarimetry to practical applications. To deal with this problem, one may transform the Mueller matrix elements into quantitative parameters with clearer physical meanings [32]. In this work, we use the frequency distribution histograms (FDHs) to characterize the wavelength-scale microstructural changes of silk samples [26]. Previous studies have shown that the 2D images of Mueller matrix elements of homogeneous samples can be transformed into a group of orientation insensitive parameters using FDH and its central moments, which can characterize the dominant wavelength-scale microstructural features of the samples [26]. By analyzing the peak positions, widths and shapes of the FDHs of the Mueller matrix elements, we may learn abundant information of the silk samples, such as the degree of order, depolarization power, orientation direction of the anisotropic silk fibers and so on. In our previous studies, four central moment parameters have been adopted for the statistical analysis: mean (P1), variance (P2), skewness (P3) and kurtosis (P4) defined as Equation (2). More detailed information about the central moment parameters can be obtained according to [26]. (2) μ=P1=E(X)σ2=P2=Var(X)skewness=P3=E(X−μ)3σ3kurtosis=P4=E(X−μ)4σ4 Suppose we have a random variable X, whose central moments are defined as Equation (2): mean (µ), variance (σ2), skewness and kurtosis. E and Var are the notations for calculating the mean (P1 or µ) and variance (P2 or σ2) values of the random variable X, respectively. Then, using µ and σ2 the skewness (P3) and kurtosis (P4) can be calculated [26]. Considering that for the backscattering Mueller matrix imaging, the central elements m22, m33, m23 and m32 always represent more prominent changes than other elements [20], we propose a new set of parameters based on their mean (P1) and variance (P2) values of the FDHs to characterize the wavelength-scale microstructure variations of the silk samples. (3) c1P1=|m22P1−m33P1|m22P1d22P1=m22iP1−m220P1m220P1d23P2=m23iP2−m230P2m230P2 As shown in Equation (3), we define the parameter c1p1 to characterize the difference between the mean values (P1) of the FDHs of diagonal m22 and m33. Meanwhile, the parameters d22p1 and d23p2 indicate the mean (P1) and variance (P2) differences between the unwashed and washed silk samples of the m22 and m23 elements. In Equation (3), m22ip1, m23ip2 are the mean and variance values of the FDH curves of the silk sample after the ith washing, while m220p1 and m230p2 are the corresponding values of unwashed sample. Here we choose the diagonal m22 element and off-diagonal m23 element for calculations. It should be noted that other elements can also provide similar information of the silk samples. 3.3. Monte Carlo (MC) Simulation For better explanations of the relationships between the Mueller matrix parameters and the wavelength-scale microstructural changes of the silk samples, we carry out MC simulations based on the SCSM to study the behavior of polarized photons as they propagate in the silk samples [27]. There are two key components in the SCSM: spherical scatterers and infinitely long cylindrical scatterers, representing different wavelength-scale microscopic structures of the silk samples. The silk fibers are approximated as cylindrical scatterers immersed in interstitial medium, meanwhile, the small particles attached to the surface of the silk fibers or embedded in the interstitial medium can be represented by spherical scatterers with different sizes and scattering coefficients. In this paper, we set the MC simulation parameters according to the silk samples, which will be introduced in the following sections. 4. Conclusions In this paper, we take the backscattering 2D Mueller matrix images of silk fibers washed by different detergents: fabric softener, laundry powder, toilet soap and color stain net. The analysis of the 2D images and frequency distribution histograms (FDHs) of the Mueller matrix elements reveals abundant qualitative structural information of the silk fibers, such as their anisotropy degree, depolarization ability and orientation direction. Moreover, we propose a group of new parameters based on the central moments of FDHs to obtain the wavelength-scale microstructural variation features of the silk samples during the washing processes quantitatively. The experimental results demonstrate that the Mueller matrix parameters have the potential to be used as indicators to decide if a detergent can change the structure of the silk fibers. Also, the different change trends show that the parameters may provide details on the wavelength-scale microstructural variations of the silk fibers. For deeper understanding of the relationship between the Mueller matrix parameters and the microstructural variations of silk, we also performed X-ray diffraction and scanning electron microscope measurements, then carried out Monte Carlo simulations based on the sphere-cylinder scattering model. The good agreement between experimental and simulated results confirms that the Mueller matrix polarimetry and FDH based parameters can be used as tools to monitor the wavelength-scale microstructural variations of silk. The information is valuable not only for the quantitative characterization of structural changes of silk fibers, but also for non-destructive and real-time detection of wavelength-scale microstructures of silk based materials. Acknowledgments This work has been supported by National Natural Science Foundation of China (NSFC) Grant No. 11174178, 11374179, 61205199, 61405102, and Science and Technology Project of Shenzhen Grant No. CXZZ20140509172959978, GJHZ20150316160614844. Supplementary Materials Supplementary materials can be found at www.mdpi.com/1422-0067/17/8/1301/s1. Click here for additional data file. Author Contributions Yang Dong and Honghui He conceived the research framework; Yang Dong and Jialing Zhou performed the experiments; Yang Dong and Honghui He analyzed the data; Chao He and Nan Zeng contributed analysis tools; Yang Dong and Honghui He wrote the paper; Hui Ma supervised the project and revised the manuscript. Conflicts of Interest The authors declare no conflict of interest. Figure 1 2D images of Mueller matrices of silk sample washed by fabric softener: (a) before washing; (b) after the third washing; (c) after the sixth washing. The color bar is from −1 to 1 for the m11, m22, m33, and m44, and from −0.1 to 0.1 for other elements. Figure 2 2D images of Mueller matrices of silk sample washed by color stain net: (a) before washing; (b) after the third washing; (c) after the sixth washing. The color bar is from −1 to 1 for the m11, m22, m33, and m44, and from −0.1 to 0.1 for other elements. Figure 3 Frequency distribution histograms (FDHs) of Mueller matrix elements of silk sample washed by the fabric softener: before washing (black lines), after the third washing time (red lines) and the sixth washing time (blue lines). The areas under the FDH curves are normalized to 1, and the horizontal axis is divided into 400 parts. Figure 4 Frequency distribution histograms (FDHs) of Mueller matrix elements of silk sample washed by the color stain net: before washing (black lines), after the third wash (red lines) and the sixth wash (blue lines). The areas under the FDH curves are normalized to 1, and the horizontal axis is divided into 400 parts. Figure 5 FDH parameters c1p1 (left), d22p1 (middle) and d23p2 (right) of silk samples washed by different detergents. (a–c) are the values of three parameters for the silk samples washed by the group A detergents: fabric softener in black square lines and laundry powder in red dot lines; (d–f) are the values of three parameters for the silk samples washed by the group B detergents: toilet soap in green diamond lines and color stain net in blue triangle lines. The horizontal axis represents the washing times. Figure 6 Optical microscopic images of silk samples in group A: (a) before washing; (b) after the sixth washing by fabric softener; (c) after the sixth washing by laundry powder. Scanning electron microscopic images (10,000×) of silk samples in group A: (d) before washing; (e) after the sixth washing by fabric softener; (f) after the sixth washing by laundry powder. Figure 7 Optical microscopic images of silk samples in group B: (a) before washing; (b) after the sixth washing by toilet soap; (c) after the sixth washing by color stain net. Scanning electron microscopic images (10,000×) of silk samples in group B: (d) before washing; (e) after the sixth washing by toilet soap; (f) after the sixth washing by color stain net. Figure 8 Monte Carlo simulation results of the parameters c1p1 and d22p1 using the SCSM. (a) The x-axis represents different values of sphere/cylinder ratio from (1) 10:70 to (2) 13:67; (3) 15:65; (4) 21:59; (5) 22:58; (6) 27:53 and (7) 28:52; (b) The x-axis represents different values of the diameter and standard deviation of angular distribution of the cylinders from (1) 1.5 µm 15 degree to (2) 1 µm 15 degree; (3) 0.9 µm 19 degree; (4) 0.86 µm 19 degree; (5) 0.76 µm 20 degree; (6) 0.68 µm 21 degree; and (7) 0.68 µm, 22 degree. Figure 9 X-ray diffraction curves of silk samples. (a) 1. The silk before washing; 2. The silk after the sixth washing by fabric softener; 3. The silk after the sixth washing by laundry powder; (b) 1. The silk before washing; 2. The silk after the sixth washing by toilet soap; 3. The silk after the sixth washing by color stain net. Figure 10 (a) Schematic of experimental setup for the backscattering Mueller matrix measurement. P1, P2: polarizer; R1, R2: quarter-wave plate; L1, L2: lens. The oblique incident angle θ is about 15 degree to avoid the surface reflection from the sample. The diameter of the illumination area is about 1.8 cm; (b) Silk sample used in this study. ijms-17-01301-t001_Table 1Table 1 Central moments P1 and P2 of the Mueller matrix elements for silk sample washed by fabric softener. Detergent/Parameter m12 m22 m23 m32 m33 F/P1_0 0.030 0.536 0.013 0.008 0.039 F/P1_1 0.029 0.504 0.036 0.030 0.043 F/P1_2 0.028 0.492 −0.003 −0.005 0.042 F/P1_3 0.029 0.526 0.009 0.002 0.048 F/P1_4 0.030 0.510 0.029 0.023 0.045 F/P1_5 0.029 0.524 0.015 0.009 0.048 F/P1_6 0.024 0.485 −0.012 −0.030 0.042 F/P2_0 0.011 0.022 0.018 0.022 0.016 F/P2_1 0.012 0.021 0.019 0.023 0.016 F/P2_2 0.011 0.022 0.019 0.022 0.016 F/P2_3 0.012 0.023 0.020 0.023 0.017 F/P2_4 0.011 0.022 0.019 0.023 0.017 F/P2_5 0.012 0.022 0.019 0.024 0.017 F/P2_6 0.011 0.022 0.018 0.022 0.016 ijms-17-01301-t002_Table 2Table 2 Central moments P1 and P2 of the Mueller matrix elements for silk sample washed by laundry powder. Detergent/Parameter m12 m22 m23 m32 m33 L/P1_0 0.032 0.528 0.075 0.071 0.049 L/P1_1 0.029 0.552 0.019 0.010 0.046 L/P1_2 0.027 0.534 0.037 0.036 0.051 L/P1_3 0.030 0.515 0.044 0.038 0.053 L/P1_4 0.032 0.512 0.056 0.050 0.056 L/P1_5 0.034 0.522 0.019 0.013 0.049 L/P1_6 0.032 0.482 0.004 −0.002 0.044 L/P2_0 0.012 0.019 0.023 0.027 0.018 L/P2_1 0.012 0.018 0.023 0.027 0.018 L/P2_2 0.012 0.018 0.022 0.026 0.018 L/P2_3 0.012 0.018 0.022 0.026 0.019 L/P2_4 0.012 0.019 0.022 0.026 0.019 L/P2_5 0.013 0.019 0.023 0.027 0.019 L/P2_6 0.012 0.019 0.022 0.026 0.018 ijms-17-01301-t003_Table 3Table 3 Central moments P1 and P2 of the Mueller matrix elements for silk sample washed by toilet soap. Detergent/Parameter m12 m22 m23 m32 m33 T/P1_0 0.033 0.560 0.001 −0.006 0.039 T/P1_1 0.039 0.492 0.006 −0.001 0.047 T/P1_2 0.039 0.453 −0.016 −0.023 0.052 T/P1_3 0.037 0.418 −0.004 −0.006 0.061 T/P1_4 0.033 0.379 0.003 −0.002 0.057 T/P1_5 0.031 0.367 0.004 −0.003 0.061 T/P1_6 0.026 0.307 −0.016 −0.021 0.061 T/P2_0 0.013 0.024 0.020 0.023 0.019 T/P2_1 0.013 0.025 0.021 0.023 0.019 T/P2_2 0.013 0.029 0.019 0.021 0.019 T/P2_3 0.013 0.028 0.020 0.021 0.019 T/P2_4 0.012 0.031 0.019 0.020 0.019 T/P2_5 0.012 0.027 0.019 0.020 0.019 T/P2_6 0.011 0.027 0.018 0.018 0.018 ijms-17-01301-t004_Table 4Table 4 Central moments P1 and P2 of the Mueller matrix elements for silk sample washed by color stain net. Detergent/Parameter m12 m22 m23 m32 m33 C/P1_0 0.034 0.555 0.009 0.002 0.034 C/P1_1 0.036 0.582 0.134 0.131 0.060 C/P1_2 0.039 0.636 0.073 0.067 0.042 C/P1_3 0.032 0.607 0.074 0.067 0.047 C/P1_4 0.030 0.585 0.017 0.006 0.061 C/P1_5 0.025 0.560 −0.010 −0.019 0.113 C/P1_6 0.010 0.441 0.003 −0.009 0.182 C/P2_0 0.013 0.024 0.020 0.022 0.018 C/P2_1 0.014 0.024 0.021 0.025 0.022 C/P2_2 0.017 0.029 0.027 0.029 0.027 C/P2_3 0.018 0.032 0.029 0.030 0.030 C/P2_4 0.019 0.040 0.031 0.034 0.036 C/P2_5 0.029 0.073 0.061 0.066 0.079 C/P2_6 0.032 0.107 0.073 0.078 0.119 F, L, T and C represent fabric softener, laundry powder, toilet soap and color stain net respectively. F/P1_0 and F/P2_0 represent the parameters P1 and P2 of silk sample before washing under the condition of fabric softener. The full Tables of P1–P4 values of all the Mueller matrix elements are provided as supporting material. ==== Refs References 1. Santin M. Motta A. Freddi G. Cannas M. In vitro evaluation of the inflammatory potential of the silk fibroin J. Biomed. Mater. Res. 1999 46 382 389 10.1002/(SICI)1097-4636(19990905)46:3<382::AID-JBM11>3.0.CO;2-R 10397996 2. Acharya C. Kumar V. Sen R. Kundu S.C. Performance evaluation of a silk protein based matrix for the enzymatic conversion of tyrosine to L-DOPA Biotechnol. J. 2008 3 226 233 10.1002/biot.200700120 18034433 3. Meinel L. Fajardo R. Hofmann S. Langer R. Silk implants for the healing of critical size bone defects Bone 2005 37 688 698 10.1016/j.bone.2005.06.010 16140599 4. Chitrangada A. Boris H. Subhas C.K. The effect of lactose-conjugated silk biomaterials on the development of fibrogenic fibroblasts Biomaterials 2008 29 4665 4675 18804860 5. Cao Y. Wang B.C. Biodegradation of silk biomaterials Int. J. Mol. Sci. 2009 10 1514 1524 10.3390/ijms10041514 19468322 6. Heslot H. Arfificial fibrous proteins: A review Biochimie 1998 80 9 13 10.1016/S0300-9084(98)80053-9 7. Chitrangada A. Sudip K.G. Kundu S.C. Silk fibroin film from non-mulberry tropical tasar silkworms: A novel substrate for in vitro fibroblast culture Acta Biomater. 2009 5 429 437 18676188 8. Zhou C.Z. Confalonieri F. Medina N. Zivanovic Y. Esnault C. Yang T. Jacquet M. Janin J. Duguet M. Perasso R. Fine organization of B. mori fibroin heavy chain gene Nucleic Acids Res. 2000 28 2413 2419 10.1093/nar/28.12.2413 10871375 9. Zhou P. Xie X. Effects of pH and calcium ions on the conformational transitions in silk fibroin using 2D raman correlation spectroscopy and 13C solid-state NMR Biochemistry 2004 43 11302 11311 10.1021/bi049344i 15366940 10. Hino T. Tanimoto M. Shimabayashi S. Change in secondary structure of silk fibroin during preparation of its microspheres by spray-drying and exposure to humid atmosphere J. Colloid Interface Sci. 2003 266 68 73 10.1016/S0021-9797(03)00584-8 12957583 11. Alali S. Vitkin I.A. Polarized light imaging in biomedicine: Emerging Mueller matrix methodologies for bulk tissue assessment J. Biomed. Opt. 2015 10.1117/1.JBO.20.6.061104 25793658 12. He H. Sun M. Zeng N. Du E. Liu S. Guo Y. Wu J. He Y. Ma H. Mapping local orientation of aligned fibrous scatterers for cancerous tissues using backscattering Mueller matrix imaging J. Biomed. Opt. 2014 10.1117/1.JBO.19.10.106007 25321399 13. Ghosh N. Vitkin I.A. Tissue polarimetry: Concepts, challenges, applications, and outlook J. Biomed. Opt. 2011 10.1117/1.3652896 22112102 14. Ye H. Xu J. Freudenthal J. Kahr B. On the circular birefringence of polycrystalline polymers: Polylactide J. Am. Chem. Soc. 2011 133 13848 13851 10.1021/ja205159u 21838269 15. Liu S. Du W. Chen X. Jiang H. Zhang C. Mueller matrix imaging ellipsometry for nanostructure metrology Opt. Express 2015 23 17316 17329 10.1364/OE.23.017316 26191741 16. Snik F. Jones J.C. Escuti M. Fineschi S. Harrington D. Martino A.D. Mawet D. Riedi J. Tyo J.S. An overview of polarimetric sensing techniques and technology with applications to different research fields Proc. SPIE 2014 10.1117/12.2053245 17. Wang Y. He H. Chang J. Zeng N. Liu S. Li M. Ma H. Differentiating characteristic microstructural features of canceroustissues using Mueller matrix microscope Micron 2015 79 8 15 10.1016/j.micron.2015.07.014 26280279 18. Sun M. He H. Zeng N. Du E. Guo Y. Peng C. He Y. Ma H. Probing microstructural information of anisotropic scattering media using rotation-independent polarization parameters Appl. Opt. 2014 53 2949 2955 10.1364/AO.53.002949 24922012 19. Du E. He H. Zeng N. Sun M. Guo Y. Wu J. Liu S. Ma H. Mueller matrix polarimetry for differentiating characteristic features of cancerous tissues J. Biomed. Opt. 2014 10.1117/1.JBO.19.7.076013 25027001 20. Sun M. He H. Zeng N. Du E. Guo Y. Liu S. Wu J. He Y. Ma H. Characterizing the microstructures of biological tissues using Mueller matrix and transformed polarization parameters Biomed. Opt. Express 2014 5 4223 4234 10.1364/BOE.5.004223 25574434 21. Pierangelo A. Manhas S. Benali A. Fallet C. Totobenazara J. Antonelli M. Novikova T. Gayet B. Martino A.D. Validire P. Multispectral Mueller polarimetric imaging detecting residual cancer and cancer regression after neoadjuvant treatment for colorectal carcinomas J. Biomed. Opt. 2013 10.1117/1.JBO.18.4.046014 23612875 22. Dubreuil M. Babilotte P. Martin L. Sevrain D. Rivet S. Grand Y.L. Brun G.L. Turlin B. Jeune B.L. Mueller matrix polarimetry for improved liver fibrosis diagnosis Opt. Lett. 2012 37 1061 1063 10.1364/OL.37.001061 22446225 23. Wang W. Lim L.G. Srivastava S. Yan J.S. Shabbir A. Liu Q. Roles of linear and circular polarization properties and effect of wavelength choice on differentiation between ex vivo normal and cancerous gastric samples J. Biomed. Opt. 2014 10.1117/1.JBO.19.4.046020 24781588 24. Jagtap J. Chandel S. Das N. Soni J. Chatterjee S. Pradhan A. Ghosh N. Quantitative Mueller matrix fluorescence spectroscopy for precancer detection Opt. Lett. 2014 39 243 246 10.1364/OL.39.000243 24562117 25. He H. Zeng N. Liao R. Yun T. Li W. He Y. Ma H. Application of sphere-cylinder scattering model to skeletal muscle Opt. Express 2010 18 15104 15112 10.1364/OE.18.015104 20639995 26. He C. He H. Li X. Chang J. Wang Y. Liu S. Zeng N. He Y. Ma H. Quantitatively differentiating microstructures of tissues by frequency distributions of Mueller matrix images J. Biomed. Opt. 2015 10.1117/1.JBO.20.10.105009 26502227 27. Yun T. Zeng N. Li W. Li D. Jiang X. Ma H. Monte Carlo simulation of polarized photon scattering in anisotropic media Opt. Express 2009 17 16591 16602 10.1364/OE.17.016590 19770874 28. Harris J.I. Cole R.D. Pon N.G. The kinetics of acid hydrolysis of dipeptides J. Biochem. 1956 62 154 159 10.1042/bj0620154 29. Byler D.M. Susi H. Examination of the secondary structure of proteins by deconvolved Biopolymers 1986 25 469 487 10.1002/bip.360250307 3697478 30. Azzam R.M.A. Photopolarimetric measurement of the Mueller matrix by Fourier analysis of a single detected signal Opt. Lett. 1978 2 148 150 10.1364/OL.2.000148 19684668 31. Chenuault D.B. Pezzaniti J.L. Chipman R.A. Mueller matrix algorithms Proc. SPIE 1992 1746 231 246 32. He H. Chang J. He C. Ma H. Transformation of full 4 × 4 Mueller matrices: A quantitative technique for biomedical diagnosis Proc. SPIE 2016 10.1117/12.2210878