Quantitative Metabolomic Profiling of Plasma, Urine, and Liver Extracts

Article pubs.acs.org/jpr

Quantitative Metabolomic Profiling of Plasma, Urine, and Liver Extracts by 1H NMR Spectroscopy Characterizes Different Stages of Atherosclerosis in Hamsters Wei Guo,†,# Chunying Jiang,†,# Liu Yang,†,‡,# Tianqi Li,† Xia Liu,† Mengxia Jin,† Kai Qu,†,‡ Huili Chen,† Xiangju Jin,† Hongyue Liu,† Haibo Zhu,*,†,‡ and Yinghong Wang*,† †

State Key Laboratory for Bioactive Substances and Functions of Natural Medicines and ‡Beijing Key Laboratory of New Drug Mechanisms and Pharmacological Evaluation Study, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, No.1 Xiannongtan Street, Beijing 100050, P. R. China S Supporting Information *

ABSTRACT: Atherosclerosis (AS) is a progressive disease that contributes to cardiovascular disease and shows a complex etiology, including genetic and environmental factors. To understand systemic metabolic changes and to identify potential biomarkers correlated with the occurrence and perpetuation of diet-induced AS, we applied 1H NMR-based metabolomics to detect the time-related metabolic profiles of plasma, urine, and liver extracts from male hamsters fed a high fat and high cholesterol (HFHC) diet. Conventional biochemical assays and histopathological examinations as well as protein expression analyses were performed to provide complementary information. We found that diet treatment caused obvious aortic lesions, lipid accumulation, and inflammatory infiltration in hamsters. Downregulation of proteins related to cholesterol metabolism, including hepatic SREBP2, LDL-R, CYP7A1, SR-BI, HMGCR, LCAT, and SOAT1 was detected, which elucidated the perturbation of cholesterol homeostasis during the HFHC diet challenge. Using “targeted analysis”, we quantified 40 plasma, 80 urine, and 60 liver hydrophilic extract metabolites. Multivariate analyses of the identified metabolites elucidated sophisticated metabolic disturbances in multiple matrices, including energy homeostasis, intestinal microbiota functions, inflammation, and oxidative stress coupled with the metabolisms of cholesterol, fatty acids, saccharides, choline, amino acids, and nucleotides. For the first time, our results demonstrate a time-dependent metabolic progression of multiple biological matrices in hamsters from physiological status to early AS and further to late-stage AS, demonstrating that 1H NMR-based metabolomics is a reliable tool for early diagnosis and monitoring of the process of AS. KEYWORDS: quantitative metabolomics, NMR, hamsters, high fat and high cholesterol diet, atherosclerosis

■

INTRODUCTION Atherosclerosis (AS) is the major cause of cardiovascular disease, which is the leading cause of death in the world, responsible for approximately 29% of all deaths worldwide.1 It is a complicated and multifactorial disease associated with a variety of genetic and environmental factors.2 However, the diagnosis of AS is not always easy. Despite the power of ultrasound, angiography, and computed tomography to assess carotid arteries, most of these measurements are invasive and can only detect established plaques. Furthermore, they provide limited biochemical information on AS, which is essential for diagnosis at the molecular level. Thus, their usefulness remains limited. Biomarkers, from a clinical perspective, can assist in the monitoring, diagnosis, and prognosis of a multifactorial disease, which correspond to different stages in the development of the disease.3 In addition to validated epidemiological risk factors, such as low-density lipoprotein cholesterol (LDL-C), dietary factors, and other factors, certain potential biomarkers related to the risks of cardiovascular disease have been identified.4,5 © 2016 American Chemical Society

However, there are few biomarkers that afford ideal diagnostic properties, and analysis with limited biomarkers may result in inaccurate diagnosis of AS. Therefore, it is desirable to discover new and systematic biomarkers for the onset and development of AS at earlier stages with the help of new and effective approaches. In general, compared with other unmodified rodent animals, the hamster has been regarded as a preferable model for dietinduced AS due to several advantages.6−8 Hence, investigation of this model could contribute to research into the risk factors associated with diet-induced AS. Metabolomics, the systematic study of small molecule metabolites in multiple biochemical pathways, can enable us to obtain valuable information on dynamic metabolic responses of living systems with certain metabolic diseases, such as AS. 9−11 Although various biomarkers have been identified in these studies, the qualitative Received: February 27, 2016 Published: August 29, 2016 3500

DOI: 10.1021/acs.jproteome.6b00179 J. Proteome Res. 2016, 15, 3500−3510

Article

Journal of Proteome Research

−80 °C. After centrifugation at 278 K at 3,000g for 10 min, the supernatants of urine samples were stored at −80 °C. At the end of the feeding period for each group, the hamsters were anesthetized with sodium pentobarbital (60 mg/kg, I.P.). The abdominal aorta and liver tissues were collected and then immediately frozen at −80 °C until further analyses.

analysis used reflects only relative change trends of metabolites rather than absolute concentration quantification. As a targeted approach, quantitative metabolic profiling may significantly improve biomarker discovery in a noninvasive and unbiased manner by providing quantitative information across multiple pathways, demonstrating powerful potential for mechanism research.12 Quantitative analysis of absolute concentration is especially essential and challenging. The targeted metabolites are mathematically modeled from a pure compound spectra database, which is then interrogated to identify and quantify metabolites in complex spectra of biosamples. Nevertheless, to the best of our knowledge, this specific approach has not yet been applied to the realm of diet-induced AS in hamsters for evaluating the effects of diet intervention on metabolic phenotypes. In the current study, we investigated the time-dependent metabolic progression of high fat and high cholesterol (HFHC) diet-induced AS in a hamster model using a combination of quantitative 1H NMR-based metabolomics, conventional biochemical assays, histopathological examinations, as well as protein expression analyses. Our findings provide a system-level metabolic atlas to guide future studies on AS by probing the metabolic profiles of multiple biological matrices from hamsters during different stages of AS.

■

Clinical Chemistry and Pathological Characteristics

The serum levels of total cholesterol (TC), triglyceride (TG), LDL-C, and high-density lipoprotein cholesterol (HDL-C) were examinated in duplicate with commercial enzymatic assays (purchased from BioSino Biotechnology and Science Inc., Beijing, China) using standard routine procedures. Aortic roots were embedded in Optimal Cutting Temperature (OCT) solution on dry ice. For lipid accumulation to be evaluated, outflow tract sections were sliced into 7 μm thickness and stained with Oil red O using routine laboratory procedures. Likewise, sections were routinely stained with hematoxylin and eosin (H&E) to evaluate inflammatory cell infiltration in the aortic root. To evaluate macrophage infiltration in the aortic root, we treated sections with 3% H2O2 to inactivate endogenous peroxides. The sections were subsequently incubated with an anti-CD68 primary antibody overnight at 4 °C,15 followed by polymer helper for 10 min and poly-HRP antirabbit IgG for 20 min at 37 °C. Next, 3,3′-diaminobenzidine tetrahydrochloride (DAB) was used for visualization in microscopic analysis.

MATERIALS AND METHODS

Chemicals and Reagents

Protein Expression Analysis

CH3OH, CHCl3, NaCl, NaH2PO4·2H2O, and K2HPO4·3H2O were obtained from Sinopharm Chemical Reagent (Shanghai, China). D2O (99.9% D), chloroform-d (99.9% D), sodium 3trimethylsilyl-propionate-2,2,3,3,-d4 (TSP), and dimethyl-silapentane-sulfonate (DSS) were purchased from Qingdao Tenglong Microwave Co. Ltd. (Qingdao, China). Standard chow was provided by Beijing HFK Bioscience Co. Ltd. (Beijing, China). HFHC diet was prepared by mixing the standard chow with lard oil (245 g/kg HFHC diet) and cholesterol (20 g/kg HFHC diet) according to the literature.13,14 A compositional table of diets is shown in Table S1.

Liver samples (approximately 100 mg) were lysed using routine laboratory procedures, and protein concentrations were determined using the BCA assay for Western blotting. Total protein of tissue was separated on a 10% SDS-PAGE gel and then transferred onto a nitrocellulose membrane. After blocking, the target proteins were probed with 1:2,000 antiABCA1, 1:300 anti-SREBP2, 1:900 anti-SR-BI, 1:300 anti-LDLR, 1:3,000 anti-LCAT, 1:1,000 anti-SOAT1, 1:500 antiHMGCR, 1:1,000 anti-CYP7A1, and 1:5,000 anti-β-actin antibodies overnight at 4 °C and then incubated with HRPconjugated secondary antibodies for 1.5 h at room temperature. The proteins blots were visualized and quantified by the use of chemiluminescence (ECL plus Western blotting detection system; GE Healthcare UK Ltd.) and PhotoShop analysis software.

Animal Experiments and Sample Collections

All animal treatments in this study were approved by the Research Ethics Committee of the Chinese Academy of Medical Sciences and Peking Union Medical College. Syrian golden hamsters (12 weeks, male and 90−110 g) were bought from Vital River Laboratory Animal Technology Co. Ltd. (Beijing, China). All hamsters were maintained under a constant condition (a temperature of 20−24 °C, a relative humidity of 35−55%, and a 12 h day/night cycle) and provided with ad libitum access to food and water. After acclimatization for 2 w, 64 adult male Syrian golden hamsters were randomly separated into two groups according to the length of the experiment: an intermediate-term group (24 w) and a longterm group (42 w). Hamsters in each group were then randomly separated into two subgroups, namely, control (n = 16) and HFHC groups (n = 16). The control and HFHC groups were fed the standard chow and HFHC diet, respectively. At the time points of 0, 3, 6, 9, 15, 21, 24, 26, 35, and 42 w, blood samples were prepared via retro-orbital bleeds, and urinary samples were collected from metabolism cages for 12 h for each group. Blood samples were processed in standard routine procedures for serum and plasma and then stored at

Sample Preparation

For plasma samples, 60 μL of 0.9% sodium buffer (D2O:H2O = 1:9 with 0.1% TSP) was mixed with 30 μL of plasma. After centrifugation at 3,000g for 5 min at 278 K, 60 μL of the supernatant was pipetted into an NMR tube (1.7 mm). For each sample, three spectra were acquired by using NOESYPR1D (RD−90°−t 1−90°−t m −90°−acq), CPMGPR1D (Carr−Purcell−Meibom−Gill, RD−90°−(τ−180°−τ)n−acq), and LEDBPGPPR2s1d (bipolar pulse pair-longitudinal eddy current delay, RD−90°−G1−τ−180°−G2−τ−90°−Δ−90°− G3−τ−180°−G4−τ−90°−te−90°−acq) pulse sequences (from the Bruker pulse sequence library) with normal water presaturation. Additionally, 300 μL of plasma was centrifugated through a 3-kDa Nanosep microcentrifuge device at 13,000g to remove insoluble impurities and proteins. Subsequently, 90 μL of filtrate was mixed with 10 μL of 0.9% sodium buffer (100% D2O, 5 mM DSS), and 60 μL of the obtained mixture was pipetted into an NMR tube (1.7 mm). For urine samples, after centrifugation at 13,000g at 278 K for 5 min, 20 μL of 3501


Article


determining R2 for model interpretability and Q2 for model predictability. The corresponding variables with a correlation coefficient |r| greater than the cutoff value (according to the number of animals in each group) were selected as primary metabolites, whose concentrations were compared between groups to reveal metabolic shifts. Additionally, statistical analysis was also conducted with SPSS19.0 (IBM; USA) using the two-tailed Student’s t test. A p-value of less than 0.05 was considered to be statistically significant between two groups.

phosphate buffer solution (100% D2O, 0.2 M NaH2PO4/ K2HPO4, 5 mM DSS, pH 7.4) was mixed with 180 μL of the supernatant, and then 60 μL of the obtained mixture was pipetted into an NMR tube (1.7 mm). For liver tissues, samples were prepared using the previously reported method.16 In these experiments, one-dimensional spectra were recorded using NOESYPR1D as described above, and DSS was utilized to refer NMR chemical shifts and calibrate concentrations. NMR Spectra Acquisition and Processing

All NMR spectra were recorded at 298 K using a Bruker Avance 500 MHz spectrometer (1H frequency: 500.13 MHz; Bruker, Germany) with a 1.7 mm TXI probe. For NOESYPR1D spectra, the mixing time was set to 100 ms and recycle delay was set to 1 s. For CPMGPR1D spectra, the spin−spin relaxation delay was set to 320 ms. For LEDBPGPPR2s1d spectra, the diffusion delay was set to 120 ms and delay Te was set to 5 ms. All spectra were acquired with 128 scans and 12 ppm spectral width. The free induction decay (FID) was multiplied by 0.3−1 Hz line-broadening factor, zero padded to 128 k data points and Fourier transformed. For the purpose of quantitative profiling, it is crucial that the NOESYPR1D spectra are acquired using the same parameters as the metabolite standard spectra in the Chenomx database. For resonance assignment purposes, additional 2-dimensional NMR spectra were acquired using standard Bruker pulse programs, including homonuclear total correlation spectroscopy (2D 1H−1H TOCSY) and heteronuclear single quantum coherence spectroscopy (2D 1H−13C HSQC). For the CPMGPR1D and LEDBPGPPR2s1d spectra of plasma and NOESYPR1D spectra of liver lipophilic extract, 1H NMR spectra were manually phased and baseline-corrected using TOPSPIN (version 3.0, Bruker, Germany), and referenced to the chemical shift of TSP or TMS at δ 0.0 ppm. Subsequently, each spectrum over the range of δ 0.5−6.0 ppm was binned with a width of 0.04 ppm and integrated using the AMIX software package (version 3.8.3, Bruker, Germany). For plasma samples, the regions containing the water signal (δ 4.7−5.1 ppm) were discarded prior to statistical analysis. Following this step, each spectrum was normalized to the total sum of the spectral integrals to compensate for concentration differences between samples. For quantitative metabolomic profiling of filtered plasma, urine, and liver hydrophilic extract, spectra were processed with the Chenomx NMR Suite 7.5 software (Chenomx Inc., Edmonton, Canada) using the “targeted profiling” approach.17 Metabolites in each sample were identified and quantified in the order of decreasing typical concentration based on the database stored in this specific software. Spectra were profiled in random order. Metabolite concentration in each sample was then normalized by the sum of total concentration (excluding urea for urinary samples due to its excessively high concentration).

■

RESULTS To investigate the systemic metabolic changes correlated with the occurrence and perpetuation of AS, we employed Syrian golden hamsters to model AS pathology through HFHC diet feeding for 42 w. The time-related metabolic profiles of plasma, urine, and liver exacts were detected with comparative nonquantitative and quantitative 1H NMR-based metabolomics. Complementary information was further achieved from conventional serum biochemical assays, histopathological examinations, as well as protein expression analysis. Measurement of Pathological Characteristics

Serum biochemical parameters were measured at different time points, and the data are shown in Figure S1. In comparison to the control group, the levels of TC, TG, HDL-C, and LDL-C in the HFHC group were significantly increased; these values reached a peak at 3 w post-HFHC feeding and were maintained at higher levels throughout the study period. Representative images of the aortic roots with oil-red O staining and H&E staining as well as macrophage areas with CD68 immunohistochemical staining from different groups are shown in Figure S2. From these figures, it was apparent that the control group had no aortic lesions. In contrast, compositional changes of the aortic root were found in the HFHC group. HFHC diet feeding for 24 w caused pronounced enhancement of lipids (Figure S2A) and inflammatory infiltration (Figure S2B) associated with macrophages (Figure S2C). In addition, compositional changes of the aortic root were found to be more apparent after 42 w of HFHC feeding. In general, these results showed that hamsters fed a HFHC diet exhibited typical blood lipids and pathological characteristics along with the progression of AS. Protein Expression Analysis of Hepatic Tissue

The protein response of the liver to the proatherogenic HFHC diet was assessed according to the expression of proteins related to cholesterol metabolism. These proteins included sterol regulatory element-binding protein 2 (SREBP2), scavenger receptor class B type I (SR-BI), ATP binding cassette transporter A1 (ABCA1), low-density lipoprotein receptor (LDL-R), 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGCR), cholesterol 7 alpha-hydroxylase (CYP7A1), sterol o-acyltransferase 1 (SOAT1), and lecithin-cholesterol acyltransferase (LCAT) in hepatic tissue. The results are shown in Figure S3. From these figures, we found that HFHC feeding for 24 w caused significant decreases in the expression of SREBP2, LDL-R, and CYP7A1 combined with an elevation of ABCA1. Interestingly, we found that HFHC feeding for 42 w not only intensified most of the changes in protein expression observed at 24 w but also resulted in greater variation in the expression of additional proteins, including SR-BI, HMGCR, SOAT1, and LCAT.

Multivariate Data Analysis

Output data were processed with the SIMCA-P+ 12.0 software (Umetrics, Sweden) to elucidate patterns of metabolic profiles. First, a nonsupervised principal component analysis (PCA) was conducted to screen potential outliers and observe group clustering on mean-centered data. Subsequently, a supervised partial least-squares discriminant analysis (PLS-DA) or a supervised orthogonal projection to latent structure-discriminant analysis (OPLS-DA) was conducted with 7-fold crossvalidation to maximize the variations among groups on unitvariance scaled data. The quality of model was monitored by 3502


Article


Figure 1. Trajectory derived from PCA of 1H-CPMG-NMR spectra (A) and 1H-LEDBP-NMR spectra (B) from hamster plasma in the HFHC group (n = 15, mean ± SD) normalized to the sum of the spectrum mapping of the time-related trajectory of metabotypes at 0, 3, 9, 15, 21, 26, 35, and 42w.

Figure 2. OPLS-DA scores (upper) and correlation coefficient plots (lower) derived from NMR data for hamster plasma samples at 24 (A) and 42 w (B) (red triangles, control, n = 15; blue diamonds, HFHC, n = 15). Positive bars (±SEM) of correlation coefficient plots denote metabolites significantly higher in the control group, whereas negative bars (±SEM) denote metabolites significantly increased in the HFHC group.

Metabolite Profiling of Plasma, Urine, and Liver Extract Samples

provided different but complementary metabolite information. Specifically, 40 metabolites in the filtered plasma specimens were quantified, including saccharides (glucose, maltose, and galactose), ketones, tricarboxylic acid (TCA) cycle intermediates, amino acids, and choline metabolites. Eighty metabolites in urine samples were quantified, including metabolites

Typical 1H NMR spectra of plasma, urine, and liver extracts are demonstrated in Figures S4−S6, presenting resonances from a wide range of metabolites. Different kinds of biological samples 3503


Article


Figure 3. Trajectory derived from PCA of targeted profiling of 80 measured urine metabolites in HFHC group (n = 14, mean ± SD) normalized to the total concentration of all measured metabolites mapping the time-related trajectory of metabotypes at 0, 3, 6, 9, 15, 21, 26, 35, and 42 w.

Figure 4. OPLS-DA scores (upper) and correlation coefficient plots (lower) derived from NMR data for hamster urine samples at 15 (A) and 35 w (B) (red triangles, control, n = 14; blue diamonds, HFHC, n = 14). Positive bars (±SEM) of correlation coefficient plots denote metabolites significantly higher in the control group, whereas negative bars (±SEM) denote metabolites significantly increased in the HFHC group.

quantified metabolites from different matrices is shown in Table S2.

associated with energy metabolism (citrate, succinate, pyruvate, lactate, creatine, and creatinine), metabolites from gut microbiota, metabolites of cholines, amino acids, and nucleosides together with others (amines and amides). In the aqueous liver extract, we quantified 60 metabolites. The quantified metabolites from these biosamples were unambiguously assigned depending on a comparison with the Chenomx metabolite database, the literature data,18−20 as well as 2D 1 H−1H TOCSY and 1H−13C HSQC experiments. A list of

1

H-CPMG-NMR and 1H-BPP-LED-NMR of Plasma Samples

For investigating the time course of plasma metabolic variations along with the development of AS, PCA models of CPMG and LEDBP spectra from the control and HFHC groups at different time points were respectively constructed for the first two PCs. PCA score plots of both CPMG and LEDBP spectra from the control group showed no significant differences throughout the 3504


Article


Figure 5. OPLS-DA scores (upper) and correlation coefficient plots (lower) derived from NMR data for hydrophilic liver extract at 24 (A) and 42w (B) (red triangles, control, n = 16; blue diamonds, HFHC, n = 16). Positive bars (±SEM) of correlation coefficient plots denote metabolites significantly higher in the control group, whereas negative bars (±SEM) denote metabolites.

intensified. In addition to these findings, greater variation of additional metabolites, such as creatinine and urea, was also observed.

study period (Figure S7), whereas the averaged PCA scores of both CPMG and LEDBP spectra in the HFHC group showed significantly different metabolic trajectories as indicated by the arrow (Figure 1). Specifically, the samples at 0 w could be discriminated clearly from the samples at other times. Along with the extension of feeding time, the HFHC group moved away to the 3 w position along the PC1 axis onward with an increase in LDL/VLDL and N-acetyl glycoproteins (NAG) and a decrease in phosphatidylcholine (PtdCho) and the PUFA/ MUFA ratio. From 3 w, the HFHC group moved away along the PC2 axis with a maximum shift reached at the end of 42 w. These results suggested that metabolic disturbances occurred at different stages and could be considered as a result of the progression of AS.

1

H-NOESY of Urine

For investigating the time course of urinary metabolic changes in the development of AS, PCA models of the targeted profiling of 80 measured urine metabolites from the control and HFHC groups at different time points were constructed for the first two PCs. The control group showed no significant differences throughout the study period (Figure S8), whereas the HFHC group showed significantly different metabolic trajectories that were similar to those obtained from the plasma samples (Figure 3). Specifically, the HFHC group moved away from the 0 w position to the 3 w position along the PC1 axis onward. From 3 w, the HFHC group moved away along the PC2 axis with a maximum shift reached at the end of 42 w. For illustrative purposes, the cross-validated score plots and the corresponding coefficient plots of the two groups at 15 and 35 w are displayed in Figure 4. According to the correlation coefficient plots, the main changes at 15 w included increased dimethylglycine (DMG), phenylacetylglycine (PAG), urocanate, and inosine. When fed further to 35 w, long-lasting increases in the levels of urinary betaine, creatinine, hippurate, allantoin, nicotinamide N-oxide (NMNO), methylamine (MA), dimethylamine (DMA), trimethylamine (TMA), and trimethylamine N-oxide (TMAO) were accompanied by decreases in taurine, cis-aconitate, citrate, alanine, and 3-indoxylsulfate. The level of 4-hydroxyphenylacetate (4-HPA) increased at 15 w and decreased thereafter.

Quantitative Analysis of Plasma Metabolites

For the observed changes to be investigated in detail, the quantitative variations in plasma metabolites were investigated with an OPLS-DA approach by comparing the 1H NMR profiled metabolite concentrations between the two groups at 24 and 42 w (Figure 2). The quality of the model was assessed by the interpretability represented by X (R2X) and Y (R2Y) variables and the predictability represented by Q2 with 7-fold cross-validation. Clear separation at 24 and 42 w was achieved between the two groups, as evidenced by the consistently high Q2 values. The metabolites responsible for the separation of the control and model groups were summarized by the OPLS-DA regression coefficients, and only metabolites with statistically significant differences (p < 0.05) were listed. In detail, HFHC feeding for 24 w led to pronounced elevations in the levels of plasmatic glutamine, betaine, choline, valine, leucine, isoleucine, and 2-hydroxybutyrate. Concurrently, diet treatment resulted in significant declines in the levels of glucose, citrate, alanine, adenosine, acetate, and carnitine. When fed further to 42 w, most of the metabolite changes mentioned above were

1

H NMR Spectra of Liver Extracts

Metabolic changes in both liver hydrophilic extract from the two groups at 24 and 42 w were investigated with an OPLS-DA approach. Figure 5 shows an obvious classification of 3505


Article


Figure 6. HFHC feeding induced metabolic changes in the development of AS. Metabolites in red indicate a high level, whereas metabolites in blue represent a lower level in the HFHC group compared with that in the control group. The numbers 1 and 2 stand for liver samples taken at 24 and 42 w (control, n = 16; HFHC, n = 16); a and b stand for plasma samples taken at 24 and 42 w (control, n = 15; HFHC, n = 15), and I and II stand for urine samples taken at 15 and 35w (control, n = 14; HFHC, n = 14), respectively.

genesis from the physiological to pathophysiological status (Figure 6).

hydrophilic extract between the two groups. From the correlation coefficient plots, it was found that the levels of glycine, taurine, betaine, glutamate, glutamine, aspartate, leucine, isoleucine, valine, inosine, TMAO, and 2-hydroxybutyrate increased, whereas those of glucose, lactate, alanine, and carnitine decreased at 24 w. When further fed with the HFHC diet to 42 w, a greater separation was observed between the control and HFHC groups. Except for leucine, isoleucine, and valine, the metabolite changes found above remained present at 42 w. In addition, decreased levels of citrate, and adenosine and increased glycerol level were observed. The PLS-DA strategy was applied to the liver lipophilic extract (Figure S9). According to the loading plots, the HFHC group showed significantly increased levels of TG, multiple saturated fatty acids (SFA), and free and esterified cholesterol but decreased levels of PtdCho, the cholesteryl ester/total cholesterol ratio, and the PUFA/MUFA ratio.

HFHC Feeding Disrupted Cholesterol Homeostasis

Intracellular cholesterol accumulation in the liver induced altered protein expression in hepatocytes, collectively promoting the progression of AS. At the early stages of AS, decreased expression levels of SREBP2, LDL-R, and CYP7A1 were observed. The accumulation of cholesterol in the liver following cholesterol overloading caused downregulated expression of SREBP-2, which acts primarily on cholesterol uptake from plasma (LDL-R) and endogenous cholesterol synthesis (HMGCR) and is regulated by the cholesterol content of cells via feedback inhibition of proteolytic processing.21 Consequently, suppression of LDL-R reduced the cellular uptake of LDL, resulting in LDL and ox-LDL accumulation in the intima and the subsequent development of AS,22 which was consistent with the dramatic LDL-C elevation in the serum of the hamsters (Figure S1). Diet-induced repression of CYP7A1, the major enzyme that regulates the conversion of cholesterol into bile acids and its subsequent fecal excretion, indicated a deficiency in the excretion of cholesterol, which is quantitatively the most important method for elimination of cholesterol from the body.23 Such a notion was further supported by an elevation in hepatic taurine and glycine resulting from a decreased demand of bile acid conjugations with taurine and glycine in the liver.24 At the late stage of AS, the liver strove for homeostasis by intensifying the changes in protein expression observed at 24 w. In addition, decreased expression levels of HMGCR, SR-BI, SOAT1, and LCAT were observed. The restrained expression of HMGCR, the limiting enzyme in the cholesterol biosynthetic pathway, caused a decrease in hepatic endogenous cholesterol synthesis. SR-BI, a hepatic HDL receptor, has been reported to be relevant for cardiovascular health through regulation of HDL-C metabolism.25 Because of the suppression of SOAT1 and LCAT, which has been correlated with the conversion of cholesterol to cholesteryl esters,26,27 the cholesterol esterifica-

■

DISCUSSION To study the effect of HFHC diet-induced AS, we selected the hamster as a small animal model, which has been documented for its susceptibility to dietary challenges. In our study, when fed with a HFHC diet for 42 w, these hamsters showed obvious aortic lesions accompanied by elevated lipids and inflammatory infiltration including macrophages (Figure S2), indicating successful generation of the AS model. In addition, expression analysis of proteins related to cholesterol metabolism elucidated the perturbation of cholesterol homeostasis during the HFHC diet challenge. Furthermore, multiple metabolites from plasma, urine, and liver extract biosamples at different stages were identified and exhibited time-dependent metabolic changes. In general, our article demonstrated metabolic disturbances involving multiple metabolic pathways, including energy homeostasis, intestinal microbiota functions, inflammation, and oxidative stress coupled with the metabolisms of cholesterol, fatty acids, saccharides, choline, amino acids, and nucleotides, indicating progressive development of athero3506


Article


citrate and cis-aconitate in the urine, were all decreased throughout the experimental period. Recent studies have also suggested that TCA intermediates are closely associated with AS.31 Taken together, these findings indicated that significant perturbations in energy metabolism could be induced by exposure to high loads of dietary fat and cholesterol and that vascular cells might respond to AS by converting the energy supply from glucose to lipid. Consequently, lipid oxidation became the main source of energy and ultimately resulted in an oxidation chain reaction and cell membrane damage.32

tion process may be suppressed and the structure of the lipid carrier may be changed. This finding was consistent with the decreased cholesteryl ester/total cholesterol ratio observed in the liver. Collectively, our combined analysis of metabolites and protein expression demonstrated that HFHC feeding exhibited a pro-atherogenic property via perturbing cholesterol metabolism and regulating proteins relevant to cholesterol homeostasis (Figure 7).

HFHC Feeding Altered Metabolites Related to Choline Metabolism

We observed a significantly increased level of plasmatic choline in the model group during the entire period of atherogenesis, which was consistent with previous results.28 Compared with normal chow diet, mice supplemented with choline exhibited increased macrophage levels of CD36 and SR-A1, which are two macrophage scavenger receptors involved in AS.33 This finding suggested that choline could be an early biomarker of atherogenesis. In addition, choline can also be oxidized to generate betaine and subsequently to generate DMG, sarcosine, glycine, and eventually creatinine.34 The concurrent increased levels of plasmatic and hepatic betaine at 24 w and the excretion of urinary DMG and creatinine at 15 w appeared to indicate that HFHC feeding promoted the conversion of choline to creatinine. As a methyl donor for the remethylation of homocysteine to methionine in the presence of betainehomocysteine methyltransferase (BHCMT), betaine has a pivotal role in S-adenosylmethionine (SAM) synthesis pathways. Indeed, betaine is the most essential factor in the prediction of hyperlipidemia and could be regarded as an early biomarker of cardiovascular disease.35 Furthermore, alterations in osmoregulation are likely implicated because these choline metabolites (choline and betaine) are also osmoprotectants.36 Creatinine is a common indicator of renal function, and an increased creatinine level was found only when remarkable damage occurred in functioning nephrons.37 Thus, the elevated level of plasmatic and urinary creatinine, particularly at 42 w, might indicate renal damage at the late stage of atherogenesis.

Figure 7. HFHC feeding induced altered protein expression in cholesterol homeostasis.

HFHC Feeding Induced Changes in Lipid and Glucose Metabolism

Metabolic perturbations are likely to play essential roles in both the onset and development of atherogenesis, particularly disorders of lipid metabolism.28 Continuously increased levels of plasmatic LDL/VLDL and hepatic TG, as well as decreased PtdCho and the PUFA/MUFA ratio in the plasma and liver, were observed during the entire feeding period, remarkably with a dramatic fluctuation from 0 to 24 w, which was consistent with the biochemical and pathological analyses performed. Because of the hepatic uptake of large amounts of fats in chylomicron via the lymphatic capillary, excess accumulation of TG in the liver can lead to fatty liver; this state can also be the result of decreased exports of TG from the liver. Supportive evidence showed a decreased level of PtdCho, and insufficiency of PtdCho or its precursors can impair VLDL synthesis and secretion, thereby decreasing the secretion of TG from the liver.16 The observed decrease in the PUFA/MUFA ratio indicated that the HFHC diet enhanced the peroxidation of polyunsaturated fatty acids and produced excess reactive oxygen species (ROS) and, thus, oxidative stress.29 Such a notion was further supported by decreased carnitine in the plasma and liver, which has been reported to be a potent antioxidant (free radical scavenger) that may protect tissues from oxidative damage.30 Disruption in glucose homeostasis was highlighted by a decreasing trend for the glucose level in both the plasma and liver throughout the study period with a sharp decrease observed at 24 w. In addition, as the main product of glycolysis by lactic dehydrogenase (LDH), the decreased levels of hepatic lactate revealed downregulation of glycolysis.10 Depletion of acetate in the plasma during the entire period was also observed, which may have been readily converted to acetylCoA entering gluconeogenesis. Furthermore, intermediates of the TCA cycle, including citrate in the plasma and liver and

HFHC Feeding Caused Alterations to Amino Acid and Nucleotide Metabolism

With the development of AS, some metabolites related to amino acid metabolism showed perturbations, including valine, leucine, isoleucine, alanine, lysine, glutamate, and glutamine. The elevations in the levels of plasmatic and hepatic branched chain amino acids (BCAAs: valine, leucine, and isoleucine) were observed at 24 w. Importantly, the metabolism of BCAAs was profoundly altered under conditions of insulin resistance and/or deficiency.38 The concurrent decreases in plasmatic and hepatic alanine levels combined with an elevation of hepatic glutamate began to occur at 24 w, and this trend continued until 42 w. This result was consistent with previous findings and most likely indicated injury of liver function.28,39 Furthermore, the continuously elevated glutamine levels in the plasma and liver during the entire treatment period may result from the inhibition of glutamine metabolism by cholesterol overloading.40 Accumulation of glutamine, a physiological inhibitor of nitric oxide (NO) synthesis in endothelial cells and intact blood vessels, may account for the inhibition of NO formation, which is of primary importance in regulating endothelial permeability and vascular tone as well as decreasing the flux of 3507


Article

Journal of Proteome Research lipoproteins into the vessel wall.41 Moreover, HFHC feeding clearly caused significant changes in many nucleotide derivatives, as highlighted by the depletion of plasmatic and hepatic adenosine along with the elevation of hepatic and urinary inosine. Importantly, the breakdown products of adenosine include xanthine and hypoxanthine, both substrates for xanthine oxidase, which is a pivotal enzyme to generate ROS.42 However, because of the complexity of amino acid and nucleotide metabolism pathways, much remains to be fully investigated together with the underlying mechanisms.

but also caused greater variations in additional metabolites, including NAG, BCAAs, NMNO, MA, DMA, TMA, glutamine, and taurine. At the last stage, significantly increased TMAO and creatinine levels, as well as decreased levels of alanine and adenosine, may serve as an indicator of liver and kidney damage. Thus, our study demonstrates that 1H NMR-based metabolomics can not only provide comprehensive biochemical information but also offer a noninvasive approach to identify potential biomarkers for the initiation and progression of AS.

■

HFHC Feeding Altered Gut Microbiota Functions

ASSOCIATED CONTENT

S Supporting Information *

Our observation of the urinary MA, DMA, and TMA elevations during the entire experimental period indicated diet-mediated modification of the gut microbiota metabolic activity in the development and progression of AS, as these are choline metabolites of gut microbiota.34 TMAO, an oxidation product of TMA, was elevated in both the liver and urine during the entire period with dramatic elevation observed at 42 w. It is well-known that TMAO is commonly generated from DMA and TMA during choline or L-carnitine metabolism by gut microbiota.33 One of the main roles of TMAO is to function as a renal osmolyte,43 and this compound was previously connected with renal medullary damage and chronic renal failure.43,44 Furthermore, HFHC feeding also caused significant changes of urinary 3-indoxylsulfate, benzoate, 4-HPA, PAG, urocanate, and hippurate from 15 to 35 w, which further supported HFHC diet-induced changes in the gut microbiota.45,46 Some observations have also been highlighted in previous investigations.9 Nevertheless, further studies will be required to obtain a more detailed characterization of the roles that the gut microbiota plays in the progression of AS.

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jproteome.6b00179. Figure S1, TC, TG, LDL-C, and HDL-C levels for hamsters treated with control and HFHC diets at different time points; Figure S2, histopathological assessments of aortic roots with oil-red O staining, H&E staining, and macrophage areas with CD68 immunohistochemical staining from control and HFHC groups at 24 and 42 w; Figure S3, expression analysis of proteins related to cholesterol metabolism for hamsters treated with control and HFHC diets at 24 and 42 w; Figure S4, 500 MHz 1H NMR spectra of plasma samples; Figure S5, 500 MHz 1H NMR spectra of urine samples; Figure S6, 500 MHz 1H NMR spectra of liver extract samples; Figure S7, PCA score plots of 1H-CPMG-NMR spectra and 1H-LEDBP-NMR spectra from hamster plasma in the control group at different time points; Figure S8, PCA score plots of 1H NMR spectra from hamster urine in the control group at different time points; Figure S9, PLS-DA score plots, corresponding loading plots, and validation plots of 1H NMR spectra from liver lipophilic extract at 24 and 42 w; Table S1, formulas of diets used in the current study; Table S2, quantitative metabolites of hamsters in filtered plasma, urine, liver hydrophilic extract and corresponding 1H chemical shift assignment used for quantification (PDF)

HFHC Feeding Induced Inflammation and Oxidative Stress

Our observation of the elevated plasma NAG at the early stage of atherogenesis (15 w) reflected atherogenesis-induced inflammation, as these proteins are well-established as late markers for the acute-phase protein response to inflammation.47 The level of 2-hydroxybutyrate in the both plasma and liver of the HFHC group showed a significant increase from 24 to 42 w; this compound was mainly found in the liver and highly expressed during oxidative stress, where it is required for the synthesis of the cellular antioxidant glutathione.48 Such a notion was further supported by consistently elevated urinary excretion of allantoin and NMNO in the HFHC group from 15 to 35 w. As previously reported, allantoin is considered an indicator of oxidative stress.49 Nicotinamide-adenine dinucleotide (NAD) is associated with intracellular respiration to oxidize fuel substrates, whereas NMNO is a precursor for NAD that can be converted to NAD by xanthine oxidase in the liver,50 and its elevated excretion has also been found to indicate diet-mediated oxidative stress. Thus, these findings suggested that inflammation and oxidative stress caused by HFHC feeding might play central roles in the onset and development of AS in hamsters.

■

AUTHOR INFORMATION

Corresponding Authors

*(H.Z.) E-mail: [email protected]. Tel./Fax: 86-1063188132. *(Y.W.) E-mail: [email protected]. Tel./Fax: 86-10-63165216. Author Contributions #

W.G., C.J., and L.Y. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by National High Technology Research and Development Program of China (Grant No: 2014AA021101) and National Natural Sciences Foundation of China (Grant No: 81273514, 91539126).

■

CONCLUSIONS In this study, distinct metabolic profiles of the control and HFHC groups were observed during different stages of AS. At the early stages of AS, distinct metabolic profiles were highlighted by elevated TC, TG, LDL/VLDL, betaine, and choline levels accompanied by declined levels of citrate, lactate, and glucose. With the progression of AS, a dietary challenge not only exacerbated the variations of metabolites mentioned above

■

REFERENCES

(1) Ramsey, S. A.; Gold, E. S.; Aderem, A. A systems biology approach to understanding atherosclerosis. EMBO Mol. Med. 2010, 2, 79−89. 3508


Article


their spectroscopic counterparts in 1H-NMR metabonomic data. BMC Bioinf. 2007, 8, S8. (21) Eberle, D.; Hegarty, B.; Bossard, P.; Ferre, P.; Foufelle, F. SREBP transcription factors: master regulators of lipid homeostasis. Biochimie 2004, 86, 839−848. (22) Tang, J. J.; Li, J. G.; Qi, W.; Qiu, W. W.; Li, P. S.; Li, B. L.; Song, B. L. Inhibition of SREBP by a small molecule, betulin, improves hyperlipidemia and insulin resistance and reduces atherosclerotic plaques. Cell Metab. 2011, 13, 44−56. (23) Princen, H. M.; Post, S. M.; Twisk, J. Regulation of bile acid synthesis. Curr. Pharm. Design 1997, 3, 59−84. (24) Tastesen, H. S.; Keenan, A. H.; Madsen, L.; Kristiansen, K.; Liaset, B. Scallop protein with endogenous high taurine and glycine content prevents high-fat, high-sucrose-induced obesity and improves plasma lipid profile in male C57BL/6J mice. Amino Acids 2014, 46, 1659−1671. (25) Leiva, A.; Verdejo, H.; Benítez, M. L.; Martínez, A.; Busso, D.; Rigotti, A. Mechanisms regulating hepatic SR-BI expression and their impact on HDL metabolism. Atherosclerosis 2011, 217, 299−307. (26) Rogers, M. A.; Liu, J.; Song, B. L.; Li, B. L.; Chang, C. C.; Chang, T. Y. Acyl-CoA:cholesterol acyltransferases (ACATs/SOATs): Enzymes with multiple sterols as substrates and as activators. J. Steroid Biochem. Mol. Biol. 2015, 151, 102−107. (27) Calabresi, L.; Simonelli, S.; Gomaraschi, M.; Franceschini, G. Genetic lecithin:cholesterol acyltransferase deficiency and cardiovascular disease. Atherosclerosis 2012, 222, 299−306. (28) Mayr, M.; Chung, Y. L.; Mayr, U.; Yin, X.; Ly, L.; Troy, H.; Fredericks, S.; Hu, Y.; Griffiths, J. R.; Xu, Q. Proteomic and metabolomic analyses of atherosclerotic vessels from apolipoprotein E-deficient mice reveal alterations in inflammation, oxidative Stress, and energy metabolism. Arterioscler. Thromb. Vasc. Biol. 2005, 25, 2135−2142. (29) Browning, J. D.; Horton, J. D. Molecular mediators of hepatic steatosis and liver injury. J. Clin. Invest. 2004, 114, 147−152. (30) Ribas, G. S.; Vargas, C. R.; Wajner, M. L-carnitine supplementation as a potential antioxidant therapy for inherited neurometabolic disorders. Gene 2014, 533, 469−476. (31) He, W.; Miao, F. J.; Lin, D. C.; Schwandner, R. T.; Wang, Z.; Gao, J.; Chen, J. L.; Tian, H.; Ling, L. Citric acid cycle intermediates as ligands for orphan G-protein-coupled receptors. Nature 2004, 429, 188−193. (32) Catala, A. A synopsis of the process of lipid peroxidation since the discovery of the essential fatty acids. Biochem. Biophys. Res. Commun. 2010, 399, 318−323. (33) 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.; Wu, Y.; Schauer, P.; Smith, J. D.; Allayee, H.; Tang, W. H.; DiDonato, J. A.; Lusis, A. J.; Hazen, S. L. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011, 472, 57−63. (34) Dumas, M. E.; Barton, R. H.; Toye, A.; Cloarec, O.; Blancher, C.; Rothwell, A.; Fearnside, J.; Tatoud, R.; Blanc, V.; Lindon, J. C.; Mitchell, S. C.; Holmes, E.; McCarthy, M. I.; Scott, J.; Gauguier, D.; Nicholson, J. K. Metabolic profiling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 12511−12516. (35) Peng, J. B.; Jia, H. M.; Xu, T.; Liu, Y. T.; Zhang, H. W.; Yu, L. L.; Cai, D. Y.; Zou, Z. M. A 1H NMR based metabonomics approach to progression of coronary atherosclerosis in a rabbit model. Process Biochem. 2011, 46, 2240−2247. (36) Mestdagh, R.; Dumas, M. E.; Rezzi, S.; Kochhar, S.; Holmes, E.; Claus, S. P.; Nicholson, J. K. Gut microbiota modulate the metabolism of brown adipose tissue in mice. J. Proteome Res. 2012, 11, 620−630. (37) Zhang, Q.; Wang, G. J.; A, J.; Ma, B.; Dua, Y.; Zhu, L.; Wu, D. Metabonomic profiling of diet-induced hyperlipidaemia in a rat model. Biomarkers 2010, 15, 205−216. (38) Lu, J.; Xie, G.; Jia, W.; Jia, W. Insulin resistance and the metabolism of branched-chain amino acids. Front. Med. 2013, 7, 53− 59.

(2) Yusuf, S.; Hawken, S.; Ounpuu, S.; Dans, T.; Avezum, A.; Lanas, F.; McQueen, M.; Budaj, A.; Pais, P.; Varigos, J.; Lisheng, L. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004, 364, 937−952. (3) Gerszten, R. E.; Wang, T. J. The search for new cardiovascular biomarkers. Nature 2008, 451, 949−952. (4) Vasan, R. S. Biomarkers of cardiovascular disease: molecular basis and practical considerations. Circulation 2006, 113, 2335−2362. (5) Moe, K. T.; Wong, P. Current trends in diagnostic biomarkers of acute coronary syndrome. Ann. Acad. Med. Singapore 2010, 39, 210− 215. (6) Auger, C.; Gerain, P.; Laurent, B. F.; Portet, K.; Bornet, A.; Caporiccio, B.; Cros, G.; Teissédre, P. L.; Rouanet, J. M. Phenolics from commercialized grape extracts prevent early atherosclerotic lesions in hamsters by mechanisms other than antioxidant effect. J. Agric. Food Chem. 2004, 52, 5297−5302. (7) Jové, M.; Pamplona, R.; Prat, J.; Arola, L.; Portero-Otin, O. M. Atherosclerosis prevention by nutritional factors: a meta-analysis in small animal models. Nutr., Metab. Cardiovasc. Dis. 2013, 23, 84−93. (8) Uehara, Y.; Urata, H.; Ideishi, M.; Arakawa, K.; Saku, K. Chymase inhibition suppresses highcholesterol diet-induced lipid accumulation in the hamster aorta. Cardiovasc. Res. 2002, 55, 870−876. (9) Li, D.; Zhang, L. L.; Dong, F. C.; Liu, Y.; Li, N.; Li, H. H.; Lei, H. H.; Hao, F. H.; Wang, Y. L.; Zhu, Y.; Tang, H. R. Metabonomic changes associated with atherosclerosis progression for LDLR−/− mice. J. Proteome Res. 2015, 14, 2237−2254. (10) Yang, Y. X.; Liu, Y.; Zheng, L. Y.; Wu, T.; Li, J. C.; Zhang, Q. Q.; Li, X. Q.; Yuan, F. Y.; Wang, L. J.; Guo, J. Serum metabonomic analysis of apoE−/− mice reveals progression axes for atherosclerosis based on NMR spectroscopy. Mol. BioSyst. 2014, 10, 3170−3178. (11) Jové, M.; Ayala, V.; Ramírez, N. O.; Serrano, J. C.; Cassanyé, A.; Arola, L.; Caimari, A.; Del Bas, J. M.; Crescenti, A.; Pamplona, R.; Portero-Otín, M. Lipidomic and metabolomic analyses reveal potential plasma biomarkers of early atheromatous plaque formation in hamsters. Cardiovasc. Res. 2013, 97, 642−652. (12) Serkova, N. J.; Niemann, C. U. Pattern recognition and biomarker validation using quantitative 1HNMR-based metabolomics. Expert Rev. Mol. Diagn. 2006, 6, 717−731. (13) Guo, T.; Chen, W. Q.; Zhang, C.; Zhao, Y. X.; Zhang, Y. Chymase activity is closely related with plaque vulnerability in a hamster model of atherosclerosis. Atherosclerosis 2009, 207, 59−67. (14) Dillard, A.; Matthan, N. R.; Lichtenstein, A. H. Use of hamster as a model to study diet-induced atherosclerosis. Nutr. Metab. 2010, 7, 89. (15) Li, L.; Wang, Y.; Xu, Y.; Chen, L. F. Atorvastatin inhibits CD68 expression in aortic root through a GRP78-involved pathway. Cardiovasc. Drugs Ther. 2014, 28, 523−532. (16) Jiang, C. Y.; Yang, K. M.; Yang, L.; Miao, Z. X.; Wang, Y. H.; Zhu, H. B. A 1H NMR-based metabonomic investigation of timerelated metabolic trajectories of the plasma, urine and liver extracts of hyperlipidemic hamsters. PLoS One 2013, 8, e66786. (17) Weljie, A. M.; Newton, J.; Mercier, P.; Carlson, E.; Slupsky, C. M. Targeted profiling: quantitative analysis of 1H NMR metabolomics data. Anal. Chem. 2006, 78, 4430−4442. (18) Suna, T.; Salminen, A.; Soininen, P.; Laatikainen, R.; Ingman, P.; Makela, S.; Savolainen, M. J.; Hannuksela, M. L.; Jauhiainen, M.; Taskinen, M. R.; Kaski, K.; Ala-Korpela, M. 1H NMR metabonomics of plasma lipoprotein subclasses: elucidation of metabolic clustering by self-organising maps. NMR Biomed. 2007, 20, 658−672. (19) Vinaixa, M.; Rodríguez, M. A.; Rull, A.; Beltrán, R.; Bladé, C.; Brezmes, J.; Cañ ellas, N.; Joven, J.; Correig, X. Metabolomic assessment of the effect of dietary cholesterol in the progressive development of fatty liver disease. J. Proteome Res. 2010, 9, 2527− 2538. (20) Vehtari, A.; Makinen, V. P.; Soininen, P.; Ingman, P.; Makela, S. M.; Savolainen, M. J.; Hannuksela, M. L.; Kaski, K.; Ala-Korpela, M. A novel bayesian approach to quantify clinical variables and to determine 3509


Article

Journal of Proteome Research (39) Martin, J. C.; Canlet, C.; Delplanque, B.; Agnani, G.; Lairon, D.; Bencharif, K.; Gripois, D.; Thaminy, A.; Paris, A. 1H NMR metabonomics can differentiate the early atherogenic effect of dairy products in hyperlipidemic hamsters. Atherosclerosis 2009, 206, 127− 133. (40) Lescano-De-Souza, A. J.; Curi, R. Cholesterol inhibits glutamine metabolism in LLC WRC256 tumour cells but does not affect it in lymphocytes: possible implications for tumour cell proliferation. Cell Biochem. Funct. 1999, 17, 223−238. (41) Swierkosz, T. A.; Mitchell, J. A.; Sessa, W. C.; Hecker, M.; Vane, J. R. L-Glutamine inhibits the release of endothelium-derived relaxing factor from the rabbits aorta. Biochem. Biophys. Res. Commun. 1990, 172, 143−148. (42) Cai, H.; Harrison, D. G. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ. Res. 2000, 87, 840−844. (43) Bell, J. D.; Lee, J. A.; Lee, H. A.; Sadler, P. J.; Wilkie, D. R.; Woodham, R. H. Nuclear magnetic resonance studies of blood plasma and urine from subjects with chronic renal failure: identification of trimethylamine-N-oxide. Biochim. Biophys. Acta, Mol. Basis Dis. 1991, 1096, 101−107. (44) Hauet, T.; Baumert, H.; Gibelin, H.; Hameury, F.; Goujon, J. M.; Carretier, M.; Eugene, M. Noninvasive monitoring of citrate, acetate, lactate, and renal medullary osmolyte excretion in urine as biomarkers of exposure to ischemic reperfusion injury. Cryobiology 2000, 41, 280−291. (45) Swann, J. R.; Tuohy, K. M.; Lindfors, P.; Brown, D. T.; Gibson, G. R.; Wilson, I. D.; Sidaway, J.; Nicholson, J. K.; Holmes, E. Variation in antibiotic-induced microbial recolonization impacts on the host metabolic phenotypes of rats. J. Proteome Res. 2011, 10, 3590−3603. (46) Williams, H. R.; Cox, I. J.; Walker, D. G.; North, B. V.; Patel, V. M.; Marshall, S. E.; Jewell, D. P.; Ghosh, S.; Thomas, H. J.; Teare, J. P.; Jakobovits, S.; Zeki, S.; Welsh, K. I.; Taylor-Robinson, S. D.; Orchard, T. R. Characterization of inflammatory bowel disease with urinary metabolic profiling. Am. J. Gastroenterol. 2009, 104, 1435−1444. (47) Bell, J. D.; Brown, J. C.; Nicholson, J. K.; Sadler, P. J. Assignment of resonances for ‘acutephase’ glycoproteins in high resolution proton NMR spectra of human blood plasma. FEBS Lett. 1987, 215, 311−315. (48) Lord, R. S.; Bralley, J. A. Clinical applications of urinary organic acids. Part I: Detoxification markers. Altern. Med. Rev. 2008, 13, 205− 215. (49) Grootveld, M.; Halliwell, B. Measurement of allantoin and uric acid in human body fluids. A potential index of free-radical reactions in vivo? Biochem. J. 1987, 243, 803−808. (50) Chaykin, S.; Bloch, K. The metabolism of nicotinamide-N-oxide. Biochim. Biophys. Acta 1959, 31, 213−216.

3510


Quantitative Metabolomic Profiling of Plasma, Urine, and Liver Extracts

Recommend Documents