Metabolomics Reveals Protection of Resveratrol in Diet-Induced Metabolic Risk Factors in Abdominal Muscle Open Access

The metabolic syndrome (MetS) is a constellation of interrelated risk factors of metabolic origin, metabolic risk factors, that recognized as to accelerate the formation of atherosclerotic cardiovascular disease (ASCVD) [1]. The American National Cholesterol Education Program illuminates that abdominal obesity, one of the six components of MetS, are closely associated with heart and vascular diseases [2]. Metabolomics, an emerging ‘omics’ science of comprehensive characterization of metabolites, is applied to systematically analyze various metabolites in biological systems [3]. In addition, metabolomics is often used to explore novel drugs and potential targets, to diagnose diseases and identify critical risk factor of diseases [4]. Metabolites are the products and substrates of metabolism that can drive various types of cellular functions including energy storage and releasing, cell proliferation and apoptosis [5]. Recently, metabolomics is increasingly being used to cardiovascular diseases, by which to identify novel biomarkers and mechanisms related to common cardiovascular diseases [2].

It is well reported that there has been a collective decline during recent decades in metabolic health in economically developed countries, and a remarkable increases in obesity, chronic uremia and cardiovascular diseases. Skeletal muscle is critical to metabolic health, which comprises a large percentage of the body mass. Thus, figure out abnormal muscle metabolism or function is considered to be helpful to find out potential therapeutic targets or biomarkers. Skeletal muscle metabolism disorder is derived from various diseases, including diabetes, chronic uremia, aging, heart failure and even atherosclerosis [6-10]. In diabetes, perturbation of glucose and lipid metabolism is closely associated with insulin resistance in skeletal muscle [11, 12]. Skeletal muscle metabolic myopathy is accompanied by a dynamic increase or decrease to meet the metabolic demands of the tissue in mitochondrial content [13]. In heart failure, oxidative capacity of mitochondria, the volume density of mitochondria (VVM) and the surface density (SVMC) of mitochondrial cristae are all reduced [14]. Skeletal muscle sterol metabolism disorder produces a remarkable changing of cholesterol, HDL-C, LDL-C and triglycerides, which are critical inducers in atherosclerosis [15]. Furthermore, skeletal muscle dysfunction is always accompanied by various diseases including chronic obstructive pulmonary disease (COPD). The potential mechanism of COPD is associated with underlying risk factors including hypoxia, oxidative stress, nutritional depletion and systemic inflammation [16].

The polyphenol resveratrol (Rev), a natural antioxidant derived from grapes, has been found to exhibit various cardioprotective effects [17]. Previous findings from our lab show that Rev protects against cardiomyocytes apoptosis induced by As₂O₃ through attenuating oxidative stress [18]. In addition, we found that Rev could ameliorate cardiac hypertrophy by attenuation level of miR-155 through activation of Breast Cancer Type 1 Susceptibility Protein (BRCA1) [19]. Recently, Voloshyna et al. indicate that Rev could decrease platelet aggregation, induce vasorelaxation, impair endothelial activation, and regulate lipid and lipoprotein metabolism. The mechanism is partially associated with activation of sirtuin 1 (SIRT1) and AMP-activated protein kinase (AMPK) [20]. Importantly, a recent multiple groups study reports that Rev could influence various metabolites that are related to energy metabolism, fatty acids metabolism and lipid metabolism [21, 22]. In muscle, Rev supplementation increases lipid content and reduces hepatic lipid level, triglycerides, circulating glucose, inflammation markers in obese humans by activating AMPK, promoting protein expression of SIRT1 and PGC-1α, enhancing citrate synthesis, and muscle mitochondrial respiration [21]. However, the role of Rev on the metabolic risk factors in abdominal muscle is still unknown.

In the present study, we describe that 11 types of metabolites are remarkable changed induced by treatment with Rev in abdominal muscles, which are closely associated with amino acid, pyrimidine, glyoxylate, dicarboxylate, urea cycle, methionine and glutamic acid metabolism. And most of these changed metabolites are closely associated with the carbohydrate, amino acid and lipid metabolisms. Further exploration reveals that Rev inhibits diet-induced atherosclerosis and foam cell formation in the lesions. These results explore new therapeutic targets on basis of metabolomics analysis and potential biomarkers of atherosclerosis. The protective functions of Rev in muscle metabolism provide a mechanism of Rev on MetS and its associated diseases including atherosclerosis.

HPLC-grade dichloromethane and methanol were purchased from J&K Scientific (Beijing, China) and Honeywell Burdick & Jackson (Muskegon, MI, USA), respectively. Pyridine (anhydrous, cat. no. 270970, 99.8%), methoxyamine hydrochloride (cat. no. 226904, 98%), MSTFA (N-methyl-N-(trimethylsilyl)-trifluoroacetamide, cat. no. 69479, ≥98.5%, for GC derivation) and tridecanoic acid (cat. no. T0502, ≥98%) were obtained from Sigma-Aldrich (Munich, Germany). Resveratrol (Rev, cat. no. R5010, ≥99%) was purchased from Sigma-Aldrich (Munich, Germany). Reference standards used for identification and verification of metabolites were obtained from Sigma-Aldrich, J&K Scientific or Alfa Aesar China (Tianjin, China).

ApoE^-/- mice (8-10 weeks, 22 ± 2 g) were housed under standard animal room conditions (temperature, 21±1° C; humidity, 55-60%). Food and water were freely available throughout experiment. All procedures were approved by the Institutional Animal Care and Use Committee of Harbin Medical University [Protocol (2009)-11]. The use of animals was compliant with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). The animals were randomly divided into three groups: normal group (N); High fat diet (HFD) group (H); HFD plus Rev (10 mg/kg/day) treatment group (HR). And the mice were gavaged Rev twice daily for 24 weeks. The mice were fed with HFD diet containing 0.3% cholesterol and 21% (wt/wt) fat for 24 weeks.

After 24 weeks HFD feeding, the brachiocephalic arteries and abdominal muscle were carefully excised from the mice for further analysis. The brachiocephalic arteries were fixed O/N in 4% PFA, and then dehydrated with 30% sucrose O/N, embedded in OCT and frozen at -80°C. For morphometric analysis, serial sections were cut at 7 µm thickness using a cryostat. The sections were stained with hematoxylin & eosin for the quantification of the lesion area. Aortic lesion size was obtained by averaging the lesion areas in 4 slides (12 sections) from the same mouse. Oil Red O staining was performed to indicate the lipid content in the lesions with an Oil Red O staining kit (Nanjing Jiancheng Biology Engineering Institute, Nanjing, Jiangsu, China) according to manufacturer’s instructions. Electron microscopy technology (EMT) was used to detect the lipid deposition in the lesions.

After being sheared and accurately weighed when still frozen, fresh muscle tissue samples were moved to 2 ml centrifugal tubes. A grinding ball and 600 µl of ice-cold 80% methanol (v/v, 5 µg/ml tridecanoic acid as the internal standard) were added to the tube, and vortexed adequately for 30 s. Thereafter, tissue samples were delivered to the grinding apparatus (MM-400, Retsch, Germany). The oscillation time and frequency were set to 1.5 min and 25 times/s. After homogenization, the homogenate was confronted with certification at 12000 g, 4 °C for 30 min to precipitate tissue fragments and proteins. Subsequently, the supernatant was withdrawn for vacuum drying in Speedvac Concentrator (Termo Scientifc, USA). Thereafter, 50 µl of methoxyamine in pyridine (20 mg/ml) was added to the dried sample and vortexed thoroughly for 30 s, prior to 1.5 h oximation reaction in a water bath at 37 °C. Following oximation reaction, metabolites in the sample were silylated with 40 µl of MSTFA in a water bath at 37 °C for 1 h. Finally, the derivatized sample was sent to certification at 12000 g, 4 °C for 30 min, and the supernatant was injected for subsequent GC-MS analysis. In order to monitor the repeatability and stability, QC (Quality Control) samples were prepared by mixing equal supernatant from every sample and dividing into 480 µl per sample. Each QC sample was inserted every 5 analytical sample, and treated in the same way as other samples in the process of vacuum drying, derivatization, GC-MS analysis and data processing.

Metabolic profiling of the sample was acquired by GCMS-QP 2010 plus (Shimadzu, Japan), and the parameters of GC-MS analysis were similar to our previous work with a few alterations [23, 24] In brief, 1 µl of the derivatized sample was injected with the AOC-20i autosampler. Metabolites were separated by using DB-5 MS capillary column (30 m × 250 µm × 0.25 µm, J&W Scientifc Inc., USA). High-purity helium was chosen as carrier gas, whose constant linear velocity and split ratio were set to 40.0 cm/s and 10: 1, respectively. The oven temperature was initially kept at 70 °C for 3 min, and increased to 300 °C at a rate of 5 °C/min, finally maintained for 10 min. Temperatures of the ion source, interface and inlet were set to 230, 280 and 300 °C, respectively. The ionization mode was EI (electron impact, 70 eV), and the detector voltage was set the same as the tuning voltage. Mass signals (m/z, 33–600) were acquired in full scan mode using GC-MS solution 2.7 (Shimadzu, Japan). The even time and solvent delay time during mass acquisition were 0.2 s and 5.3 min, respectively. Prior to running analytical samples, dichloromethane was injected into the GC-MS analytical system to elute impurities from the analytical system. On the other hand, a light diesel sample was injected to gain the retention times of n-alkanes after running all samples, and then retention indexes of metabolites in the sample could be obtained by employing ChromaTOF 4.43 (LECO Corporation, USA).

Raw Mass data was firstly converted to NetCDF format employing GC-MS solution 2.7 (Shimadzu, Japan), and then utilized to create the peak table using XCMS, including ion peak filtration, identification and matching, retention time alignment, filling missing values, and so on [25]. In addition, mass data deconvolution was carried out using ChromaTOF 4.43 to obtain feature ions and mass spectrum of each metabolite. Metabolites were identified mainly based on the results of mass spectra searches of commercial libraries (such as NIST 11, Wiley and Fiehn) and the elution orders of homology and isomers. Furthermore, feature ions of identified metabolites were manually integrated and added in the peak table generated by XCMS, and then there were totally 1979 ions in the peak table. Subsequently, the raw peak area in the peak table was divided by that of tridecanoic acid and the tissue weight, and then multiplied by 1×10⁶. Then, the data were used for following statistical analysis. Principal component analysis (PCA), partial least-squares discriminant analysis (PLS-DA) and pathway analysis were performed employing MetaboAnalyst 3.0 [26]. The nonparametric test (two-tailed Mann-Whitney U test) and heat map plot were performed by using MeV 4.9.0 [27].

All data are expressed as mean ± SD. Statistical differences were measured using one-way analysis of variance (ANOVA). A value of p≤0.05 was considered statistically significant. Data analysis was performed using GraphPad Prism Software Version 7.0 (GraphPad, San Diego, CA).

To explore the potential role of Rev on abdominal muscle metabolism and atherosclerosis, the metabolic profiles of abdominal muscles in different groups (N, H and HR group) were analysis by GC-MS. Typical GC-MS chromato-grams were displayed (Fig. 1). Principal component analysis (PCA) and partial least-squares discriminant analysis (PLS-DA) were showed in Fig. 2. The result showed that QC samples had high reproducibility each other (Fig. 2A). Among the total 1979 ions acquired from QC samples, the relative standard deviation (RSD) value of 1682 ions (85% of total ions) were less than 20% (Fig. 2B). As shown in Fig. 2C, there was an obvious separation among N, H and HR groups based on PLS-DA analysis. To further delineate the metabolic differences of each group, PLS-DA was performed between every two groups (Fig. 2D-F). A clearly separation was shown between H and N group (Fig. 2D), H and HR group (Fig. 2E), and N and HR group (Fig. 2F), respectively. All PLS-DA models had no overfitting by the permutation test and were recognized to be reliable.

To definite the effect of Rev on diet-induced metabolites in the abdominal muscle, multiple of metabolites were screened. These metabolites were randomly divided into four different types including carbohydrate, amino acid, lipid and others metabolites. As shown in Fig. 3A, the heat map result indicated that Rev elevated most of amino acids, lipids and carbohydrates in abdominal muscles.

Metabolic pathway analysis was further performed between H and N or HR group (Fig. 3B and Fig. 3C). In comparison with chow diet feeding mice, HFD promoted 20 metabolic pathways in abdominal muscles including pyruvate metabolism, galactose metabolism, biosynthesis of unsaturated fatty acids and citrate cycle (Fig. 3B). In addition, Rev treatment altered 12 different metabolic pathways in HFD-fed mice including galactose metabolism (Fig. 3C). Importantly, our result uncovered that galactose metabolism, biosynthesis of unsaturated fatty acids, alanine, aspartate and glutamate metabolism, and glyoxylate and dicarboxylate metabolism were all involved in both the development of atherosclerosis and the protective effects of Rev. These findings suggested that Rev might exert the protective effects partially by altering multiple metabolic pathways in abdominal muscles.

Carbohydrate metabolites were uncovered to be closely associated with cardiovascular disease, diabetes and malignant tumors [28]. Carbohydrate metabolites were further analyzed in the abdominal muscles. As shown in Fig. 4, HFD feeding resulted in significant change of 10 types of carbohydrate metabolites compared with N group. Furthermore, Rev treatment dramatically elevated 6 carbohydrate metabolites compared with that in H group, while 14 types of carbohydrate metabolites were identified not dramatically changed between H group and HR group.

Besides the effects on the production of carbohydrate metabolites, Rev also regulated multiple metabolites involved in amino acid and lipid types. HFD feeding increased the production of serine, glutamic acid and aspartic acid and decreased citrulline production in abdominal muscles. However, Rev treatment led to elevated pyroglutamic acid and 2-hydroxybutyric acid production compared to HFD alone group (Fig. 5A).

Abnormal lipid metabolites could be closely associated with development of atherosclerosis. As shown in Fig. 5B, three types of lipid metabolites including 11, 14-eicosadienoic acid, cis-7, 10, 13, 16-docosatetraenoic acid and dodecanoic acid were obviously upregulated. And four types of lipid metabolites including palmitelaidic acid, heptadecanoic acid, nonadecanoic acid and 1, 3-propanediol were substantially reduced after HFD feeding for 24 weeks. In addition, we further explored that Rev treatment boosted the concentration of cis-5, 8,11, 14, 17-eicosapentaenoic acid (EPA, C20-5, n-3) in abdominal muscles, which exerted anti-inflammatory, lipid lowering and atheroprotective effects [29-31].

In view of metabolomics are the dynamics of endogenous small molecules, the changes in pyrimidine metabolism, glyoxylate and dicarboxylate metabolism, and redox in the abdominal muscles were analyzed. As shown in Fig. 5C, Rev dominantly elevated the level of glycolic acid and alpha-tocopherol compared with H group.

To investigate the effects of Rev on atherogenesis, ApoE^-/- mice were fed with HFD diet for 24 weeks in the presence/absence of Rev (10mg/kg/day). Mice fed with chow diet were used as normal control. The prevalence of atherosclerotic lesions was then analyzed in the branchiocephalic arteries from these three groups. It’s not surprising that HFD feeding displayed big atherosclerotic plaques in ApoE^-/- mice after 24 weeks while chow diet feeding mice did not develop significant atherosclerosis. However, Rev treatment inhibited the progression of diet-induced atherosclerosis (Fig. 6A). Oil Red O staining in the cross sections indicated a decreased lipid accumulation in the lesions from Rev treatment group (Fig. 6B and 6C). Electron microscopy (EM) was further employed to detect the lipid deposition in the lesions. As shown in Fig. 6D, the foam cells characterized by abundant, large lipid droplets were observed in HFD feeding group by EM. Indicating that Rev treatment obviously decreased the intracellular lipid accumulation in the lesions.

Metabolic syndromes (MetS) are a condition that includes diabetes mellitus, pathogenic obesity, dyslipidemia, and hypertension, all of which can induce heart and vascular damage, and even chronic kidney disease [1]. The purposes of our study are to clarify the relationship between atheroprotective effects of Rev and abdominal muscles metabolic components during the development of atherosclerosis. First, our results demonstrated that Rev could alter multiple differential metabolites, thereby influencing critical pathways including carbohydrate metabolism, amino acid metabolism, lipid and other metabolisms that were closely associated with development of atherosclerosis (Fig. 3). Second, we found that Rev treatment dramatically elevated 6 carbohydrate metabolites (Fig. 4). Third, it demonstrated that Rev could elevate 5 different types of metabolites in amino acid and lipid metabolism (Fig. 5). Finally, we further demonstrated that Rev could arrest progression of diet-induced atherosclerosis (Fig. 6). Findings from our lab and others have demonstrated that Rev possesses various beneficial effects on cardiovascular disease, pulmonary disease, and cancers [19, 32-37]. Jager et al. verified that Rev increased the production of serotonin, kynurenine and spermidine, and repressed the level of prostaglandin E2 (PGE2) in the human breast cancer cell lines MCF-7 and MDA-mb-231 [38], indicating that Rev might regulate some small molecules acting as potential markers for disease treatment.

In our present studies, Rev treatment reversed diet-induced lowering of glycolysis in abdominal muscles (Fig. 4). Glycolysis and muscle phosphagens are considered as critical primary manner to acquire energy [6], which might be partially attributed to increased fatty acid oxidation mediated by reducing activity of pyruvate dehydrogenase complex [39]. In addition, we also found that Rev could increase metabolites in PP pathway, which could accelerate production of NADPH. It has been reported that NADPH oxidase proteins could elevate amounts of free radical in redox signaling, thereby attenuating intracellular harmful factors or substances [40]. Based on the above results, we speculated that Rev played a critical beneficial role in development of atherosclerosis.

In amino acid metabolism (Fig. 5A), 2-Hydroxybutyric acid (2-HB) and pyroglutamic acid were markedly increased in HR group. There were two kinds of the concentrations of 2-HB would be elevated. One facet was deficient in energy metabolism, the other facet was in inherited metabolic diseases such as “cerebral” lactic acidosis, dihydrolipoyl dehydrogenase (E3) deficiency. 2-HB could damage mitochondrial energy metabolism and secondarily lead to reduction of lipid synthesis [41]. Increased NADH2/NAD ratio was identified as a critical factor to mediate 2-HB production. The increased NADH-oxidase could promote superoxide production, thereby aggravating the pathogenesis of atherosclerosis [42, 43]. Based on the above analysis, we speculated that it might be due to Rev could protect against production of aerobic oxidation and reduce the level of NADH2/NAD through elevating the level of pyroglutamic acid and 2-HB, thereby attenuating atherosclerosis.

Disorders of lipid metabolism are one of the most important characteristics of MetS [1], which is strongly associated with atherosclerosis and cardiovascular damage [2]. Intracellular lipid composition, distribution and abnormal lipid metabolism have been deemed as potential mechanisms in development of atherosclerosis [44, 45]. In this study polyunsaturated fatty acid (PUFA), was dominantly increased in HR group (Fig. 5B). Recently, PUFA decreased several risk factors of atherosclerosis, such as blood pressure (BP) and heart rate (HR), flow-mediated dilation (FMD), circulating adhesion molecules and intima-media thickness (IMT) [46]. Saito et al. illuminated that EPA was beneficial to attenuate high level of TG and HDL-C, thereby attenuating the incidence of CAD patient [47]. EPA also could increase sympathetic activation, reduce abdominal adiposity and body fat, attenuate insulin sensitivity, hypertension and left ventricular stiffness, thereby improving symptoms of high-fat diet-induced metabolic syndrome [48, 49]. In addition, EPA exerted beneficial effects on the pathophysiologic cascade from onset of plaque formation to rupture by incorporating into membrane phospholipids and atherosclerotic plaques. EPA also improved atherogenic dyslipidemia characterized by reduction of triglycerides without raising low-density lipoprotein cholesterol [50]. Based on the result of Fig. 5B, we deduced that Rev could relieve atherosclerosis, partially but not at the least, by enhancing the expression of EPA.

It has been manifested that glycolic acid could strongly potentiate the antioxidant action of alpha-tocopherol [51], which possessed better antioxidant properties, and involved in regulating multiple of cell signaling molecules and their target proteins [52]. Moreover, antioxidant supplementation may be an effective therapeutic strategy for metabolic syndrome [43]. In our experiment, we found that the level of glycolic acid and alpha-tocopherol were substantially increased in HR group (Fig. 5C). Thus, we extrapolated that Rev could exert an anti-atherosclerotic effect, partially but not at the least, by activating glycolic acid and alpha-tocopherol.

In summary, we found that 22 types of metabolites were involved in the diet-induced metabolic changes in abdominal muscles. Rev regulated 11 different types of metabolites and various of pathways. Rev improves these metabolic risk factors in the abdominal muscles, which might be a potential mechanism for atheroprotection. Considering the protective role of Rev in the inflammation, oxidative stress and proliferation in progression of atherosclerosis [13-16], the potential mechanism and function of these metabolites by Rev in the MetS and atherosclerosis are required to be further explored.

This work was supported, in part, by the National Natural Science Foundation of China (81503069, 81671746, 81401457, 21507128 and 21477124), the Special Financial Grant from the China Postdoctoral Science Foundation (2016T90316 and 2016T90313), the China Postdoctoral Science Foundation (2015M571449), the Natural Science Foundation of Heilongjiang Province (QC2016112), the funds for Wu Liande Foundation of HMU (WLD–QN1702), Wu Liande Youth Scientific Research Fund of Harbin Medical University-Daqing (DQWLD201602), the University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province (UNPYSCT-2016040), the Postdoctoral Science Foundation of Heilongjiang Province (LBH-Z15160).

All authors declare no competing financial interests.

G. Chen and G. Ye contributed equally to this work.