Article pubs.acs.org/JAFC
Serum Metabolomics Analysis of Quercetin against AcrylamideInduced Toxicity in Rats Shuang Yang,∥ Can Cao,∥ Shuai Chen, Liyan Hu, Wei Bao, Haidan Shi, Xiujuan Zhao,* and Changhao Sun Department of Nutrition and Food Hygiene, Public Health College, Harbin Medical University, 150081 Harbin, Heilongjiang, China S Supporting Information *
ABSTRACT: The current study aimed to investigate whether quercetin plays a protective role in acrylamide (AA)-induced toxicity using a metabolomics approach. Rats were randomly divided into groups as follows: control, treated with AA [5 mg/kg body weight (bw)], treated with different dosages of quercetin (10 and 50 mg/kg bw, respectively), and treated with two dosages of quercetin plus AA. After a 16 week treatment, rat serum was collected for metabolomics analysis. Biochemical tests and examination of liver histopathology were further conducted to verify metabolic alterations. Twelve metabolites were identified for which intensities were significantly changed (increased or reduced) as a result of the treatment. These metabolites included isorhamnetin, citric acid, pantothenic acid, isobutyryl-L-carnitine, eicosapentaenoic acid, docosahexaenoic acid, sphingosine 1phosphate, lysoPC(20:4), lysoPC(22:6), lysoPE(20:3), undecanedioic acid, and dodecanedioic acid. The results indicate that quercetin (50 mg/kg bw) exerts partial protective effects on AA-induced toxicity by reducing oxidative stress, protecting the mitochondria, and regulating lipid metabolism. KEYWORDS: acrylamide, metabolomics, quercetin, serum, UPLC-QTOF-MS/MS
■
INTRODUCTION Acrylamide (AA), as an important industrial compound, is widely applied in wastewater treatment, cosmetic formulations, and many other fields. In 2002, high AA content was identified for the first time in certain foods cooked at high temperatures (e.g., potato chips and French fries).1 Humans inevitably come in contact with AA via food and drinking water. As reported by the European Food Safety Authority, the mean and 95th percentile dietary AA exposures are 0.4−1.9 and 0.6−3.4 μg/kg bw per day, respectively.2 AA has been reported to show neurotoxicity, genotoxicity, reproductive toxicity, and carcinogenicity (classified as a probable human carcinogen).3 Therefore, AA-induced toxic effects have become an increasing concern. Some strategies for preventing the toxic effects of AA have been implemented. First, AA in food can be mitigated during production and processing by reducing AA precursors in raw materials or changing processing parameters.3,4 Nevertheless, AA remains present in a variety of foods.5 Furthermore, some plant extracts and phytochemicals were proven to show protective effects against AA,6,7 which prompted researchers to explore protection strategies against AA in vivo. Quercetin, a common flavonoid, is present ubiquitously in fruits and vegetables; the compound is thought to possess numerous bioactivities, including antioxidant, anticancer, antiviral, and anti-inflammatory activities. With regard to the bioactivities of quercetin, although the results of a few studies are not fully consistent and often contradictory,8,9 some epidemiological studies have shown that increasing daily intake of quercetin can reduce the risk of cardiovascular disease, diabetes, and other age-related chronic diseases.10 Recently, some researchers have paid attention to the effects of quercetin on AA-related toxicity.11,12 In these studies, some oxidative © XXXX American Chemical Society
stress markers (8-hydroxydeoxyguanosine, protein carbonyls, etc.) and enzyme activities (e.g., glutathione-S-transferase, and myeloperoxidase activities) in serum and some tissues (including the brain, liver, kidneys, and testes) were detected, indicating that quercetin plays a role in ameliorating the toxic effects of AA by reducing oxidative stress. However, such studies were carried out by conventional toxicological methods and mainly focused on certain organs or biochemical parameters rather than the global system. Metabolomics is defined as “the quantitative measurement of the dynamic multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modification”.13 All metabolites in biological samples, such as plasma and urine, can be rapidly determined and reflect the physiological status of the organism systematically. Given its high resolution, high sensitivity, and high efficiency, metabolomics has been widely employed in many fields, such as drug discovery and food safety.14,15 The influence of exogenous substances on organisms depends on many factors, such as the nature of the compounds, level, duration, and route of exposure. Humans are exposed to a complex environment where various compounds interact with one another and affect their toxicity.16 Both AA and quercetin exist widely in diets; hence, the compounds may undergo common metabolic pathways as true with some other xenobiotics.17 Moreover, both substances can influence the redox status of the organism to some extent.2,18 Therefore, we hypothesized that quercetin might affect AA-induced toxicity at Received: Revised: Accepted: Published: A
September 19, 2016 November 8, 2016 November 11, 2016 November 11, 2016 DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
water consumption of each rat was recorded during the treatment period. At each time point, no significant difference was observed for water consumption among all groups (P > 0.05, Table S1). All experimental procedures were conducted in conformity with the institutional guidelines for the care and use of laboratory animals of Harbin Medical University (Heilongjiang, China) and conformed to the National Research Council’s Guide for the Care and Use of Laboratory Animals. Sample Collection and Preparation. All of the animals survived until the end of the treatment period. Rats were anesthetized with chloral hydrate via intraperitoneal injection. Blood samples were collected from the abdominal aorta with pro-coagulation tubes. The serum and liver samples were collected and processed for parameter assays and liver histopathology examination. The activities of SOD and CAT as well as contents of GSH, MDA, FADS1, FADS2, apoM, and PanK were measured according to the manufacturer’s protocol. Sample preparation as well as information about quality control samples is further explained in the Supporting Information (Supplemental Methods). Chromatography and Mass Spectrometry. A BEH C18 column (100 × 2.1 mm, i.d. = 1.7 μm, Waters Corp., Milford, MA, USA) and an ACQUITY UPLC system (Waters Corp., Wexford, Ireland) were applied for chromatography analysis. Mass spectrometry was performed on a Waters Micromass Q-TOF Micro Mass Spectrometer (Manchester, UK). The detailed parameters are described under Supplemental Methods. Data Processing and Metabolite Identification. UPLCQTOF-MS data were processed using Markerlynx XS (Waters, Milford, MA, USA) within Masslynx software (version 4.1) for peak detection. Detailed collection parameters can be found in the Supporting Information. After data preprocessing, the data matrix was introduced to EZinfo 2.0 software (Umeå, Sweden) for multivariate statistical analysis. Unsupervised principal component analysis (PCA) was first used for all samples to visualize the general separation. Then supervised partial least-squares discriminant analysis (PLS-DA) and orthogonal partial least-squares discriminant analysis (OPLS-DA) were further performed to highlight the difference between groups. Moreover, to avoid overfitting of PLS-DA models, a default 7-fold cross-validation and testing with 500 random permutations were performed using SIMCA-P software (version 12.0; Umetrics AB, Umeå, Sweden). In OPLS-DA models, potential markers were selected on the basis of variable importance in the projection (VIP) values (>1.0) and statistical tests (P < 0.05). For these selected metabolites, MS/MS experiments were further conducted to obtain fragmentation patterns. Metabolites were detected and identified on the basis of accurate mass and MS/MS information by searching online databases: HMDB (http://www.hmdb.ca), METLIN (http://metlin.scripps.edu), Massbank (http://www.massbank.jp), and LIPID MAPS (http://www. lipidmaps.org). MS/MS spectra of metabolites were matched with the structure information from the databases using Mass Fragment application manager software within MassLynx software. Metabolites were finally confirmed by comparing retention times and fragmentation patterns with standards. Statistical Analysis. Receiver operating characteristic (ROC) curve analysis was performed for every confirmed biomarker to determine the area under the curve (AUC), which is a measure to compare the predictive ability of metabolites. Other data were expressed as the mean ± SD. Differences among groups were analyzed by one-way analysis of variance (one-way ANOVA) or nonparametric Kruskal−Wallis test depending on data distribution. When statistical differences were indicated, LSD test was used as a post hoc test. A two-tailed value of P < 0.05 was considered significant. The analyses mentioned above were performed using SPSS version 19.0 (SPSS, Chicago, IL, USA).
the metabolic level. In our previous study, AA was administered to rats for 16 weeks, and some adverse effects (such as hepatotoxicity, nephrotoxicity, and neurotoxicity) were observed at a dose of 5 mg/kg bw using ultraperformance liquid chromatography quadrupole time-of-flight tandem mass spectrometry (UPLC-QTOF-MS/MS).19 The current study aimed to investigate whether quercetin plays a protective role in AA-induced toxic effects at the metabolic level and to explore relevant mechanisms.
■
MATERIALS AND METHODS
Chemicals and Reagents. Acrylamide (99.8% purity) and quercetin (95% purity) were purchased from Sigma-Aldrich (Germany). Ultraperformance liquid chromatography (UPLC) grade methanol and acetonitrile were obtained from Dikma Science and Technology, Co. Ltd. (Canada). Standards of isobutyryl-L-carnitine (97% purity), citric acid (99.5% purity), undecanedioic acid (97% purity), dodecanedioic acid (99% purity), and isorhamnetin (95% purity) were purchased from Sigma-Aldrich (Germany). Standard of pantothenic acid (99% purity) was obtained from ABBLIS Chemicals LLC (USA). Standards of sphingosine 1-phosphate (S1P) (98% purity), eicosapentaenoic acid (EPA) (98% purity), and docosahexaenoic acid (DHA) (98% purity) were purchased from Cayman Chemical (USA). Assay kits for aspartate aminotransferase (AST) (catalog no. AH841) and alanine aminotransferase (ALT) (catalog no. AH843) were purchased from Wako Pure Chemical Industries, Ltd. (Japan), and a kit for alkaline phosphatase (ALP) (catalog no. CH0101203) was from Sichuan Maker Biotechnology Co., Ltd. (Chengdu, China). Kits for total protein (catalog no. A045-2), superoxide dismutase (SOD) (catalog no. A001-1), catalase (CAT) (catalog no. A007-1), glutathione (GSH) (catalog no. A006-1), and malondialdehyde (MDA) (catalog no. A003-1) were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Enzyme-linked immunosorbent assay kits for fatty acid desaturase 1 (FADS1) (catalog no. HX3232), fatty acid desaturase 2 (FADS2) (catalog no. HX0323), apolipoprotein M (apoM) (catalog no. HX3283), and pantothenate kinase (Pank) (catalog no. HX0328) were purchased from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China). Deionized water was purified using a Milli-Q system (Millipore, Billerica, MA, USA). Leucine encephalin was obtained from Sigma-Aldrich (St. Louis, MO, USA). Other chemicals were of analytical grade. Animal Treatment. The dose of AA was 5 mg/kg bw, which exerted significant toxic effects on rats according to our previous study.19 As for quercetin, low dose (10 mg/kg bw) and high dose (50 mg/kg bw) correspond to minimum (5.96 mg/day) and maximum quercetin intake (29.4 mg/day), respectively, in different countries.20,21 Sixty male Wistar rats (180 ± 20 g) were purchased from Vital Laboratory Animal Technology Co. Ltd. (Beijing, China). They were housed individually in metabolic cages with temperature and humidity at 22 ± 2 °C and 45 ± 15%, respectively, under a cycle of 12 h light/ 12 h dark. Food and drinking water were provided ad libitum for 1 week prior to the treatment. After acclimatization, rats were randomly divided into six groups according to body weights (10 for each group), that is, control group (C), low dose of quercetin-treated group (Q1), high dose of quercetintreated group (Q2), AA-treated group (A), low dose of quercetin plus AA-treated group (AQ1), and high dose of quercetin plus AA-treated group (AQ2). Quercetin dissolved in 0.5% carboxymethylcellulose (CMC) with corresponding concentration was administered intragastrically to rats of groups Q1, Q2, AQ1, and AQ2, whereas rats of groups C and A received 0.5% CMC in the same way. AA diluted in drinking water was given to rats of groups A, AQ1, and AQ2, whereas rats of the other groups were given unmodified drinking water instead. Both quercetin and acrylamide were administered to rats once a day (at 8:00 and 9:00 a.m., respectively). The treatment was continued for 16 weeks. Body weights of rats were measured every week, and daily B
DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
■
Antioxidant Activities and MDA Levels in the Liver. SOD and CAT activities as well as GSH and MDA levels in the liver tissue are summarized in Table 1. MDA level was significantly elevated in group A relative to that in group C (P < 0.01). In group AQ2, the MDA level was significantly lower than that in group A (P < 0.01) but higher than that in group C (P < 0.05). GSH level and SOD and CAT activities were significantly reduced in group A relative to those in group C (P < 0.01). In group AQ2, such values were significantly higher than those in group A (P < 0.05) but lower than those in group C (P < 0.05). Moreover, no significant difference was noted among groups C, Q1, and Q2 (P > 0.05) or between groups A and AQ1 (P > 0.05). Biochemical Parameters in the Serum and Liver Tissue. The liver enzyme activities in the serum are shown in Table 2. The rats treated with AA showed markedly enhanced activities of ALT, AST, and ALP compared with those of the control group (P < 0.01). In group AQ2, the activities of these enzymes significantly decreased compared with those in group A (P < 0.05) but increased compared with those in group C (P < 0.05). The levels of FADS1, FADS2, and PanK in the liver and the content of apoM in the serum are shown in Figure 3. In group A, the levels of FADS1, FADS2, and apoM significantly decreased relative to those in group C (P < 0.01), but increased significantly in group AQ2 relative to those in group A (P < 0.05 or 0.01) and significantly decreased with respect to those in group C (P < 0.05 or 0.01). The PanK level was markedly higher in group A than in group C (P < 0.01). In group AQ2, the PanK level was significantly lower than that in group A (P < 0.05) but higher than that in group C (P < 0.05). No significant difference was observed between groups A and AQ1 (P > 0.05) or among groups C, Q1, and Q2 (P > 0.05). Metabolic Profiling. Serum samples were analyzed by UPLC-QTOF-MS/MS in positive and negative modes. After peak alignment, a total of 5703 variables (3556 for positive mode and 2147 for negative mode) were obtained and subjected to further multivariate analyses. As the most common unsupervised method in metabolomics research, PCA was applied to obtain an overview of metabolic changes. As shown in score plots of PCA (Figure 4), a clear separation between groups A and C (red triangles and blue circles, respectively) was noted with partial overlap among other groups; quality control samples (green triangles) clustered tightly in the center of the plots. Six ions extracted from the chromatographic peaks (m/z 120.0393, 203.0939, 380.2269, 496.2656, 524.3128, and 568.3279 in the positive mode) were selected to evaluate the method reliability. The relative standard deviations of peak intensity, retention time, and m/z were estimated as 6.135− 15.325, 0.028−1.262, and 0.000−0.003%, respectively. These results collectively indicated good stability of the run. PLS-DA was further performed for clearer separation (Figure S1). To estimate the robustness and predictive ability of the models, cross-validation and permutation tests were applied. The intercepts of Q2 regression lines were negative, and all of the R2Ys and Q2s were much lower than the original points to the right (Figure S2), suggesting a low risk of overfitting. After validation, the PLS models were converted into OPLS. Given the VIP values (>1.0) and statistical tests (P < 0.05), 66 metabolites in the positive mode and 108 in the negative were selected for UPLC-MS/MS experiments. A total of 12 metabolites (4 in the positive mode and 8 in the negative) were confirmed by online databases (Table 3 and Table S2).
RESULTS Relative Liver Weight. Relative liver weight was calculated by using the formula (liver weight/body weight) × 100. No significant difference was found among all groups (Figure 1, P > 0.05).
Figure 1. Relative liver weights of rats from all groups. C, control group; Q1, low-dose quercetin-treated group; Q2, high-dose quercetin-treated group; A, AA-treated group; AQ1, low-dose quercetin plus AA-treated group; AQ2, high-dose quercetin plus AAtreated group. Data represent the mean ± SD (n = 10).
Liver Histopathology. Liver sections were photographed using a fluorescence microscope (Nikon Corp., Japan). As shown in Figure 2A−C, no pathologic change was detected in
Figure 2. Histopathological examination of rat liver after administration of AA and/or quercetin for 16 weeks (hematoxylin and eosin staining, 200×): (A) control; (B) 10 mg/kg bw and (C) 50 mg/kg bw quercetin-treated groups, normal portal triad, hepatocytes, and centrilobular region; (D) 5 mg/kg bw AA-treated group, swollen hepatocytes (yellow arrows), vacuolization (arrowheads), dilatated sinusoids, and pyknotic nuclei (black arrows); (E) AA plus 10 mg/kg bw quercetin-treated group, inflammatory cells (dashed circle), vacuolization (arrowheads), and pyknotic nuclei (black arrows); (F) AA plus 50 mg/kg bw quercetin-treated group, minimally dilated sinusoids and cytoplasmic vacuolization (arrowheads). Bars = 20 μm.
the liver sections from group C, Q1, or Q2, whereas those from the other three groups showed pathological injuries to various extents. For groups A and AQ1 (Figure 2D,E, respectively), swollen hepatocytes and vacuolization were observed; additionally, small amounts of inflammatory and apoptotic cells were noted. The degeneration in group AQ2 (Figure 2F) was milder than that in groups A and AQ1. C
DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 1. Effects of Acylamide and Quercetin on Antioxidant Activities and Malondialdehyde Levels in Rat Liversa groupb
GSH (mgGSH/g prot)
C Q1 Q2 A AQ1 AQ2
6.34 5.96 6.68 3.51 4.11 4.92
± ± ± ± ± ±
SOD (U/mg prot)
1.56 1.58 2.06 1.12** 1.12** 1.54*#
71.94 69.06 65.55 46.60 50.83 62.15
± ± ± ± ± ±
11.98 11.44 10.30 15.42** 14.52** 15.29*#
CAT (U/mg prot) 45.36 48.41 39.28 24.05 27.63 34.78
± ± ± ± ± ±
14.46 10.23 12.91 9.88** 8.82** 5.79*#
MDA (nmol/mg prot) 0.44 0.48 0.49 0.83 0.90 0.59
± ± ± ± ± ±
0.09 0.14 0.13 0.25** 0.23** 0.19*##
Values are the mean ± SD (n = 10). (∗) P < 0.05 and (∗∗) P < 0.01, significantly different from the control; (#) P < 0.05 and (##) P < 0.01, significantly different from the AA-treated group. bC, control group; Q1, low-dose quercetin-treated group; Q2, high-dose quercetin-treated group; A, AA-treated group; AQ1, low-dose quercetin plus AA-treated group; AQ2, high-dose quercetin plus AA-treated group. a
demonstrated how quercetin inhibited the elevation of liver enzyme levels by protecting the structural integrity of liver cells.23 Oxidative stress has been reported to play an important role in AA-induced toxic effects.24 Generally, AA and its epoxide glycidamide are detoxified in vivo by conjugating with GSH,3 which is the most probable cause for the GSH reduction in group A (Table 1). Decreased GSH can further enhance hepatic susceptibility to oxidative stress and lead to liver injuries.12,25 In the present study, AA induced decreases in antioxidant (GSH, SOD, and CAT) levels and elevation in MDA levels in the liver (Table 1). This result indicates an increased lipid peroxidation and oxidative injury. However, high-dose quercetin reduced oxidative damage, which is related to its antioxidant property by modulating antioxidant enzymes.18 In PCA score plots (Figure 4), an apparent separation between groups A and C suggests that toxic effects of AA can be detected at the metabolic level, whereas the separation between groups A and AQ2 indicates the protective effect of high-dose quercetin. With the analysis of metabolite intensities and metabolic pathways (Figure 5), AA and quercetin were found to mainly influence hepatic lipid metabolism. In Figure 5, the first main branch is concerned with the mitochondria, whereas the tricarboxylic acid (TCA) cycle and fatty acid βoxidation were affected because of the AA and quercetin treatments. Citric acid, an important component of the TCA cycle, is closely related to the metabolism of carbohydrates, lipids, and proteins. In the current study, citric acid decreased in group A compared with group C, suggesting disturbance of the TCA cycle. The TCA cycle occurs in the mitochondria; hence, the disturbance indicates mitochondrial injury. AA has been shown to cause mitochondrial morphological distortion and inhibit respiration,25,26 thereby providing support for the results of the present study. Moreover, citric acid protects the liver by decreasing lipid peroxidation and inflammation.27 Thus, the diminished citric acid in group A may have partially aggravated the oxidative stress and liver damage caused by AA, a result consistent with the elevated MDA level and liver enzyme activities (Tables 1 and 2, respectively). In group AQ2, citric acid levels increased relative to those in group A, which is related with the ability of quercetin to normalize the TCA cycle by modulating enzyme activities and inhibiting lipid peroxidation.28 Pantothenic acid is essential for the synthesis of coenzyme A (CoA), and the latter is a critical cofactor involved in multiple pathways (e.g., the TCA cycle and fatty acid biosynthesis) (Figure 6).29 An increased demand on CoA, for example, in mitochondrial β-oxidation or the TCA cycle, could reduce
Table 2. Effects of Acrylamide and Quercetin on Liver Enzyme Activitiesa groupb C Q1 Q2 A AQ1 AQ2
ALT (U/L) 45.00 47.70 53.20 69.70 62.30 56.90
± ± ± ± ± ±
8.45 7.56 11.83 11.86** 13.47** 13.61*#
AST (U/L) 109.10 101.50 107.40 151.00 137.60 130.30
± ± ± ± ± ±
20.21 19.79 22.93 24.74** 21.77** 17.98*#
ALP (U/L) 55.50 54.20 60.40 78.70 70.50 67.60
± ± ± ± ± ±
10.51 12.07 12.72 10.77** 9.28** 12.12*#
a Values are the mean ± SD (n = 10). (∗) P < 0.05 and (∗∗) P < 0.01, significantly different from the control; (#) P < 0.05, significantly different from the AA-treated group. bC, control group; Q1, low-dose quercetin-treated group; Q2, high-dose quercetin-treated group; A, AA-treated group; AQ1, low-dose quercetin plus AA-treated group; AQ2, high-dose quercetin plus AA-treated group.
The metabolite intensity values are shown in Tables S3 and S4. The intensities of nine biomarkers decreased significantly in group A relative to those in group C (P < 0.01), which include pantothenic acid, isobutyryl-L-carnitine, citric acid, EPA, DHA, S1P, lysophosphatidylcholine (LPC) (20:4), LPC(22:6), and lysophosphatidylethanolamine (LPE) (20:3). By contrast, the intensities of these markers increased markedly in group AQ2 relative to those in group A (P < 0.05 or 0.01). Meanwhile, the intensities of undecanedioic acid and dodecanedioic acid were significantly enhanced in group A relative to those in group C (P < 0.01) but significantly depressed in group AQ2 relative to those in group A (P < 0.01). For the above-mentioned metabolites, no significant difference was observed among groups C, Q1, and Q2 (P > 0.05) or between groups A and AQ1 (P > 0.05). Isorhamnetin (a metabolite of quercetin) was detected only in the groups treated with quercetin and quercetin plus AA, in which the intensities increased with the quercetin doses. The results of ROC curve analysis are shown in Table S5. All of the AUCs were >0.8, indicating at least “good” predictive abilities.22
■
DISCUSSION To verify the metabolic alterations caused by AA and quercetin, some parameters (PanK, apoM, and so on) that play crucial roles in corresponding metabolic pathways were further measured (Figure 3). Although AA did not show any effect on the relative liver weight (Figure 1), liver injuries were demonstrated by histopathologic examination and liver enzyme assays (Figure 2 and Table 2, respectively). Meanwhile, these changes were attenuated by high-dose quercetin treatment (50 mg/kg bw). This finding is consistent with those of earlier studies that D
DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Figure 3. Levels of FADS1 (A), FADS2 (B), and PanK (C) in the liver and level of apoM in the serum (D). C, control group; Q1, low-dose quercetin-treated group; Q2, high-dose quercetin-treated group; A, AA-treated group; AQ1, low-dose quercetin plus AA-treated group; AQ2, highdose quercetin plus AA-treated group. Data are expressed as the mean ± SD (n = 10). (∗) P < 0.05 and (∗∗) P < 0.01, significantly different from the control; (#) P < 0.05 and (##) P < 0.01, significantly different from the AA-treated group.
pantothenic acid levels.30 PanK catalyzes the first committed step in CoA biosynthesis.29 We hypothesized that the decrease in pantothenic acid in group A may have contributed to the upregulation of PanK and activation of the corresponding pathway induced by AA (Figure 6). This hypothesis can be confirmed by the enhanced level of PanK (Figure 3C), as well as the altered intensity of citric acid, as mentioned above. Furthermore, pantothenic acid is a necessary coenzyme in GSH synthesis;31 hence, the decrease in pantothenic acid levels further reduces GSH as confirmed in this study (Table 1). Undecanedioic acid and dodecanedioic acid are dicarboxylic acids. Normally, these acids can be rapidly oxidized in the peroxisomes and then transferred to the mitochondria for further degradation, so that they are rarely detected in the serum.32,33 These two metabolites markedly increased in group A, indicating peroxisomal dysfunction and reduced β-oxidation. This result is in agreement with previous studies showing the influence of AA on fatty acid β-oxidation.34 The accumulation of serum dicarboxylic acids can further damage hepatocytes and mitochondria.35 Isobutyryl-L-carnitine, a short-chain acylcarnitine species, is formed within the mitochondria29 and participates in fatty acid metabolism.36 Similar to that in pantothenic acid, the decrease in isobutyryl-L-carnitine levels in group A may be associated with dysfunction in the mitochondria and fatty acid oxidation, which is consistent with some earlier studies.37
In group AQ2, undecanedioic acid and dodecanedioic acid levels decreased, whereas pantothenic acid and isobutyryl-Lcarnitine levels increased relative to those in group A. This finding suggests that quercetin promotes fatty acid oxidation and protects the mitochondria. Quercetin, as a lipophilic compound, can freely cross the inner mitochondrial membrane and accumulate in the mitochondria in its biologically active form.38 As such, quercetin maintains a stable mitochondrial structure and function. Quercetin also induces mitochondrial biogenesis in hepatocytes, which further promotes cellular recovery from damage.39 In Figure 5, another main branch is related to the synthesis and transportation of protein and lipid in the liver, which are performed mainly in the endoplasmic reticulum. S1P, a kind of sphingolipid, is a vasoprotective lipid mediator with pleiotropic biological effects.40 Circulating S1P can be generated by the liver besides the hematopoietic cells and endothelium; the liver plays an important role in the distribution and metabolism of S1P throughout the body.41 ApoM is the major carrier for circulating S1P and rate-limiting in its secretion from hepatocytes; thus, the expression level of apoM directly affects the S1P level in the blood.41 Accordingly, the decreased S1P level in group A was presumably related to the alteration in apoM, which was verified by testing the serum apoM level (Figure 3D). Plasma apoM is also derived from the liver.25 Therefore, the reduced S1P and apoM both indicate hepatic dysfunction. Besides, alteration in the metabolism of E
DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Figure 4. PCA score plots resulting from UPLC-QTOF-MS spectra of rat serum in positive (A) and negative modes (B). Black circles, AA-treated group; red triangles, low-dose quercetin plus AA-treated group; squares, high-dose quercetin plus AA-treated group; blue circles, control group; stars, low-dose quercetin-treated group; crosses, high-dose quercetin-treated group; green triangles, quality control group. N = 10.
Table 3. Potential Biomarkers in Positive and Negative Modes mass (Da)
a
retention time (min)
measured
calculated
error
elemental composition
scan mode
metabolite
2.12 2.25 7.43 6.76 0.92 4.72 4.79 7.96 4.69 8.80 5.23 5.29
220.124 232.1659 544.3396 568.3376 191.0202 215.126 229.1423 301.2192 315.0896 327.232 378.2442 502.2923
220.1179 232.1543 544.3398 568.3398 191.0197 215.1289 229.1445 301.2173 315.051 327.233 378.2415 502.2939
0.0061 0.0116 −0.0002 −0.0022 0.0005 −0.0029 −0.0022 0.0019 0.0386 −0.001 0.0027 −0.0016
C9H17NO5 C11H21NO4 C28H50NO7P C30H50NO7P C6H8O7 C11H20O4 C12H22O4 C20H30O2 C16H12O7 C22H32O2 C18H38NO5P C25H46NO7P
+ + + + − − − − − − − −
pantothenic acida,b isobutyryl-L-carnitinea,b LPC(20:4)a LPC(22:6)a citric acida,b undecanedioic acida,b dodecanedioic acida,b EPAa,b isorhamnetina,b DHAa,b S1Pa,b LPE(20:3)a
The metabolite was confirmed by databases. bThe metabolite was identified by reference compound.
sphingolipid, such as S1P, can broadly disturb lipid metabolism.42 In group AQ2, the elevated levels of S1P and apoM compared with those in group A were due to the ability of quercetin to reduce oxidative damage and promote protein synthesis in the liver, which further improved the S1P metabolism.43
LPC(20:4), LPC(22:6), and LPE(20:3) belong to lysophospholipids, which regulate various biological processes, including inflammation and tumor cell invasiveness.44 Disturbance in lipid metabolism leads to changes in LPC and LPE levels.45 Decreased LPC levels in the serum/plasma are common in patients with multiple liver diseases.46 In the current study, AA F
DOI: 10.1021/acs.jafc.6b04149 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
0.05 or 0.01), indicating that quercetin partly attenuated such changes. To our knowledge, this work is the first serum metabolomics study on AA and the first to explore the effects of quercetin on AA-induced toxicity at the metabolic level. Our results demonstrated that quercetin (50 mg/kg bw) exerts partial protective effects on AA-induced toxic effects by reducing oxidative stress, protecting the mitochondria, and regulating lipid metabolism. Further studies based on urinary metabolomics and other omics techniques are necessary to achieve a more integrative understanding of related mechanisms.43
■
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.6b04149. Figure S1, PLS-DA score plots; Figure S2, model quality for PLS-DA models; Table S1, water consumption of rats; Table S2, mass fragment information on some potential biomarkers; Tables S3 and S4, intensities of metabolites identified in positive and negative modes, respectively; Table S5, performance of potential biomarkers; supplemental methods (PDF)
Figure 5. Changed metabolic pathways in response to AA and/or quercetin treatment.
■
Figure 6. Pantothenic acid is involved in the tricarboxylic acid cycle.
AUTHOR INFORMATION
Corresponding Author
*(X.Z.) Phone: +86 451 8730 2653. E-mail: xiujuan_zhao@ sina.com. ORCID
reduced the LPC and LPE levels, which may be correlated with liver dysfunction and abnormal lipid metabolism. Meanwhile, high-dose quercetin increased such levels, suggesting that quercetin protects the liver by regulating lipid metabolism.47 DHA and EPA are two main omega-3 essential polyunsaturated fatty acids, which are obtained from food and first taken up by the liver or synthesized from other essential fatty acids.48 Studies have reported that DHA and EPA can reduce oxidative stress and liver injury.49 FADS1 and FADS2 catalyze the ratelimiting formation of long-chain polyunsaturated fatty acids in the endoplasmic reticulum.50 In the present study, DHA and EPA levels decreased in group A, which is presumed to be related to the effects of FADS1 and FADS2. This result was further confirmed by the decreased FADS1 and FADS2 levels in group A (Figure 3A,B). DHA and EPA serve critical roles in normal brain structure and function.51 DHA deficiency has been reported to be associated with motor and sensory functional impairments in rats,52 which is similar to the neurological manifestations induced by AA.2 Therefore, the decreased DHA may be a cause of AA-related neurotoxicity. Besides, LPC is the major carrier of circulating DHA to the brain and renders DHA more efficient.53 Thus, the altered DHA levels may be correlated to LPC. Compared with the levels in group A, those of DHA, EPA, FADS1, and FADS2 were elevated in group AQ2, indicating that quercetin promotes essential fatty acid synthesis in the endoplasmic reticulum by regulating related enzymes. This finding is in agreement with those of earlier studies, in which quercetin increased hepatic lipid metabolism.43,47 With regard to biomarker intensities in all groups (Tables S2 and S3), the differences between groups A and C (P < 0.01) indicate the toxic effects of AA. Although some differences were noted between groups AQ2 and A (P < 0.05 or 0.01), differences were still observed between groups AQ2 and C (P
