Metabonomic Changes Associated with Atherosclerosis Progression

Article pubs.acs.org/jpr

Metabonomic Changes Associated with Atherosclerosis Progression for LDLR−/− Mice Dan Li,†,# Lulu Zhang,‡,# Fangcong Dong,‡ Yan Liu,† Ning Li,‡ Huihui Li,‡ Hehua Lei,‡ Fuhua Hao,‡ Yulan Wang,‡,∥ Yi Zhu,*,†,⊥ and Huiru Tang*,‡,§ †

Department of Physiology and Pathophysiology, Peking University Health Science Center, Beijing 100191, China CAS Key Laboratory of Magnetic Resonance in Biological Systems, State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, National Center for Magnetic Resonance in Wuhan, Wuhan Institute of Physics and Mathematics, University of Chinese Academy of Sciences, Wuhan, 430071, China § State Key Laboratory of Genetic Engineering, Collaborative Innovation Center for Genetics and Development, Metabonomics and Systems Biology Laboratory, School of Life Sciences, Fudan University, Shanghai 200433, China ∥ Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, Zhejiang University, Hangzhou 310058, China ⊥ Department of Physiology and Pathophysiology, Tianjin Medical University, Tianjin 300070, China ‡

S Supporting Information *

ABSTRACT: Atherosclerosis resulting from hyperlipidemia causes many serious cardiovascular diseases. To understand the systems changes associated with pathogenesis and progression of atherosclerosis, we comprehensively analyzed the dynamic metabonomic changes in multiple biological matrices of LDLR−/− mice using NMR and GC−FID/MS with gene expression, clinical chemistry, and histopathological data as well. We found that 12 week “Western-type” diet (WD) treatment caused obvious aortic lesions, macrophage infiltration, and collagen level elevation in LDLR −/− mice accompanied by up-regulation of inflammatory factors including aortic ICAM-1, MCP-1, iNOS, MMP2, and hepatic TNFα and IL-1β. The WD-induced atherosclerosis progression was accompanied by metabonomic changes in multiple matrices including biofluids (plasma, urine) and (liver, kidney, myocardial) tissues involving multiple metabolic pathways. These included disruption of cholesterol homeostasis, disturbance of biosynthesis of amino acids and proteins, altered gut microbiota functions together with metabolisms of vitamin-B3, choline, purines, and pyrimidines. WD treatment caused down-regulation of SCD1 and promoted oxidative stress reflected by urinary allantoin elevation and decreases in hepatic PUFA-to-MUFA ratio. When switching to normal diet, atherosclerotic LDLR−/− mice reprogrammed their metabolisms and reversed the atherosclerosis-associated metabonomic changes to a large extent, although aortic lesions, inflammation parameters, macrophage infiltration, and collagen content were only partially alleviated. We concluded that metabolisms of fatty acids and vitamin-B3 together with gut microbiota played crucially important roles in atherosclerosis development. These findings offered essential biochemistry details of the diet-induced atherosclerosis and demonstrated effectiveness of the integrated metabonomic analysis of multiple biological matrices for understanding the molecular aspects of cardiovascular diseases. KEYWORDS: atherosclerosis, LDLR−/− mice, high-fat diet, metabonomics, NMR, gut microbiota

■

INTRODUCTION Atherosclerosis is a serious disease of large arteries probably resulting from a combination of many factors such as dyslipidaemia and systematic inflammation.1−4 It is the primary cause of cardiovascular and cerebrovascular diseases becoming the leading cause of death and morbidity worldwide. In Western countries, for instance, atherosclerosis is considered as the underlying cause of ∼50% of all deaths.5,6 Therefore, extensive research efforts have been made during the last decades to understand the mechanisms of pathogenesis and progression for atherosclerosis with the interactions of both © XXXX American Chemical Society

genetic and environmental risk factors. It is now known that atherosclerosis involves the lesion formation at the branch points of arterial blood vessels starting from the fatty streaks with lipid-loaded macrophages (foam cells) in the subendothelial space followed by fibrous plaques containing smooth muscle cells and collagen.5,7 The accumulation of lipids and inflammatory components in intima causes the initial lesion formation with enhanced expression of inflammatory factors in Received: January 15, 2015

A

DOI: 10.1021/acs.jproteome.5b00032 J. Proteome Res. XXXX, XXX, XXX−XXX

Article

Journal of Proteome Research aorta;1−8 however, the involving molecular processes and biochemical interactions remain to be fully understood given the etiological complexity of such diseases. Several animal models have been employed to investigate the detailed pathophysiological aspects of atherosclerosis including primates, rabbits, and swines.5,7 Mouse models have also been developed by knocking out genes such as apolipoprotein E (ApoE−/−),9,10 low-density lipoprotein receptor (LDLR−/−),11,12 and both (ApoE−/−/LDLR−/−).12 Among them, LDLR−/− mice is one of the best animal models for studying the pathogenesis and progression of atherosclerosis because this model can mimic the growth of the hyperlipidemia-induced aortic plaques in clinical trials.11−13 Furthermore, the elevation of blood plasma cholesterol in LDLR−/− mice causes atherosclerosis,11−13 which is consistent with the epidemiological results showing hypercholesterolaemia as an independent risk factor for atherosclerosis in human.5,7 When fed with normal chow diet (CD), LDLR−/− mice exhibited mild hypercholesterolaemia (∼260 mg/dL) compared with the wild-type mice (∼113 mg/dL);11−14 however, when fed with a high-fat and high-cholesterol diet mimicking Western-type diet (WD), LDLR−/− mice exhibited a profound hyperresponsiveness with their plasma cholesterol levels reaching about an order of magnitude (∼2000 mg/dL) higher than wild-type mice, leading to plaques as well.11−14 Such hypercholestrolaemia in LDLR−/− mice was developed within 2 weeks and sustained throughout 8 weeks WD-feeding with a selective LDL elevation.11−13 Studies of both chemokines and cytokines revealed that atherosclerotic development was associated with inflammation and dysregulations of lipid metabolisms.1−4,14−16 This was further confirmed with measurements of a number of inflammatory factors such as ICAM-1, iNOS, CD68, MCP-1, and MMP-2 in lesions17−19 together with TNF-α, IL-1β, and CD68 in liver.19−22 All of these findings indicate that atherosclerotic development is probably also associated with reprogramming of metabolic network; however, the metabolic changes associated with the atherosclerosis pathogenesis and progression in the systems level for this WD-induced model remain to be fully investigated. Metabonomic analysis is a useful approach to underpin the metabolic aspects of atherosclerosis because metabonomics comprehensively detects and quantifies the metabolite composition in an integrated biological system and its dynamic responses to the changes of both endogenous and exogenous factors.23,24 Previous studies have proven metabonomics as a powerful approach for understanding the molecular aspects of cardiovascular diseases,25−28 inflammatory processes,29,30 obesity-related metabolic diseases,25,31−33 and nutritional effects on mammals.32,34−36 Recently, plasma and urinary metabonomic analyses revealed that high-fat diet-induced alterations in choline metabolism for LDLR−/− and ApoE−/− mice together with energy metabolism changes as well.37 For the time being, however, the effects of high-fat and high-cholesterol diet mimicking so-called “Western-type diet” on the mammalian metabolisms have not been fully investigated, especially in multiple organs. It is also not clear whether the WD-induced metabolic changes are reversible (and to what extent) when switching back to normal diet for a certain length of time, which is essentially important in terms of dietary management for hypercholesterolaemia. In this work, we systematically investigated the WD-induced metabolic perturbations in multiple matrices of LDLR−/− mice

including biofluids (plasma and urine) together with liver, kidney, and myocardial tissues using nuclear magnetic resonance (NMR) spectroscopy and gas chromatography coupled with flame ionization detector/mass spectrometry (GC−FID/MS). Quantitative RT-PCR analyses of the expression of some relevant genes were also conducted for such a model together with clinical chemistry and histopathological assessments. Our objectives are (1) to understand the dynamic metabonomic changes associated with the progression of WD-induced atherosclerosis and (2) to investigate the possible reversibility with diet switch (from WD to normal diet) for LDLR-null mice. To the best of our knowledge, this is the first integrated studies on the biochemistry of atherosclerosis progression with the combination of metabonomes in multiple matrices, relevant gene expression, clinical chemistry, and fatty acid composition in plasma and liver, especially with diet-switch to mimic the cases in dietary managements.

■

MATERIALS AND METHODS

Materials

Methanol, K2CO3, sodium chloride, NaH2PO4·2H2O, and K2HPO4·3H2O were purchased from Sinopharm Chemical Reagent (Shanghai, China). Sodium azide (NaN3) was obtained from Tianjin Fuchen Chemical Reagent Factory (Tianjin, China). Methanol and hexane (HPLC grade) together with 3,5-ditertbutyl-4-hydroxytoluene (BHT) and a mixed standard methyl esters of 37 fatty acids were bought from Supelco (Bellefonte, PA), whereas acetyl chloride, methyl heptadecanoate (99%), and methyl tricosanate (99%) were bought from Sigma-Aldrich. EDTA-d12 (99.9% D), D2O (99.9% D), and sodium 3-trimethylsilyl [2,2,3,3-d4] propionate (TSP) were from Cambridge Isotope Laboratories (Tewksbury, MA). Phosphate buffer used in this study was prepared with NaH2PO4 and K2HPO4, as previously reported.38 Rat horseradish peroxidase-conjugated secondary antibody and 3,3′diaminobenzidine tetrahydrochloride were purchased from Zhong-shan Golden Bridge (Beijing, China). Two different diets were purchased from Research Diet (New Brunswick, NJ) with product numbers of D12102C and D12109, with the former employed as CD and the latter (containing 40 kcal% fat, 1.25% cholesterol, 0.5% cholate) employed as high-fat and high-cholesterol diet mimicking WD. Animal Experiments and Sample Collections

All animal experimental procedures were conducted in accord with the National Guidelines for Experimental Animal Welfare (MOST of PR China, 2006). LDLR−/− male mice initially obtained from Jackson Laboratory, all 7 to 8 weeks old, were housed in an SPF animal laboratory kept at a constant temperature of 20−25 °C and relative humidity of 40−60% with a 12 h light/dark cycle. All animals had free access to water and given food. After acclimatization for 2 weeks, animals were randomly divided into three groups (control, WD, and dietswitched ones). The control (n = 15) and WD groups (n = 16) were fed with CD and WD, respectively, for 12 weeks, whereas the diet-switched group (n = 16) was fed with WD for 6 weeks, followed by CD for a further 6 weeks. Corresponding wild-type mice (C57BL/J6) were also treated with CD (n = 15) and WD (n = 15), respectively. Urine samples were collected every week. Serum samples were prepared in a standard manner and used for clinical chemistry, whereas plasma samples were collected in a standard fashion with sodium heparin as anticoagulant. All tissue samples B


Article

Journal of Proteome Research

film thickness of 0.1 μm was used with a 1:60 splitter. Sample injection volume was 1 μL. Helium gas was used as carrier and makeup gas. FID was used with air and hydrogen gas flow rate of 40 and 400 mL/min, respectively. The column temperature was set as follows: initial, 55 °C with an 1 min hold; ramp, 55 °C/min to 205 °C (30 °C/min) with a 3 min hold, 205 to 230 °C (5 °C/min). Both the injection port and detector temperatures were set at 230 °C. EI-MS spectra were acquired with the EI voltage of 70 eV and the m/z range of 45−450 with helium gas flow rate of 18.4 mL/min. Fatty acid identifications were achieved by comparing data with a mixture of 37 known standards and further confirmed with the mass spectral data from standard databases. Quantification of the methylated fatty acids was done with the FID data using methylated C17:0 and C23:0 as internal standards and quality controls. The results were calculated as μmol fatty acids per liter for plasma and μmol per gram for liver tissue, respectively. The molar percentages were calculated for total fatty acids (ToFA), saturated fatty acids (SFA), unsaturated fatty acids (UFA), polyunsaturated fatty acids (PUFA), monounsaturated fatty acids (MUFA), n6 and n3 type fatty acids, and n6-to-n3 ratios, respectively.

were collected immediately after sacrificing animals. For histopathologic assessments, the aortic trees were separately fixed in paraformaldehyde solution (4%) immediately after collection. Samples for RT-PCR and all other samples were snap-frozen in liquid nitrogen immediately after collection and then stored at −80 °C until further analyses. Clinical Chemistry and Histopathological Assessments

Clinical chemistry analyses were conducted with standard routine procedures in Clinical Laboratory in the Third Affiliated Hospital of Peking University including total serum cholesterol (TC), high-density lipoprotein cholesterol (HDLC), low-density lipoprotein cholesterol (LDL-C), glucose (Glc), total protein (TP), albumin (ALB), and urea. Triglycerides (TG) in sera were measured using an automated clinical chemistry analyzer kit (Biosino Biotech, Beijing, China). These data were expressed in the form of “mean ± SEM” and analyzed with one-way ANOVA with the Student’s test or Tukey’s post hoc test where appropriate, with p < 0.05 considered as statistically significant (Table S1 in the SI). For lesion assessments, the aortic tree was dissected on wax and then fixed in 4% paraformaldehyde solution. After the lesions were stained with Oil red O for 30 min followed by assimilated with 60% isopropyl alcohol, samples were distained with 60% isopropyl alcohol and 70% ethanol, respectively. Tissues were photographed with lesion areas calculated by ImageJ. Aortic roots were embedded in Tissue-Tek OCT molds (Zhong-shan Golden Bridge, Beijing, China) and frozen at −40 °C for microscopic analysis. Cross sections of outflow tracts (7 μm in thickness) were stained in Oil red O followed by hematoxylin counterstaining to assess the lipid accumulation. For the inflammatory cells infiltration in aortic root, outflow tract sections were stained by hematoxylin and eosin. To assess the macrophage infiltration in plaques, we treated sections with 3% H2O2 to inactivate the endogenous peroxides and then incubated them with the primary antibody anti-Mac3 (1:100; Santa Cruz, CA) overnight at 4 °C, followed by horseradish peroxidase-conjugated secondary antibody for 30 min at 37 °C. 3,3′-Diaminobenzidine tetrahydrochloride was used to develop color. Collagen fibers were stained by 0.1% picric acid-sirius red (Fluka, Germany).

1

H NMR Spectroscopic Methods for Metabonomics Analysis

For plasma samples, 30 μL of plasma was mixed with 30 μL of saline solution in D2O and transferred to 1.7 mm NMR tubes. One-dimensional 1H NMR spectra of plasma were recorded at 298 K on a Bruker Avance II 500 MHz NMR spectrometer (Bruker, Germany) operating at 500.13 MHz for proton frequency with a Bruker 5 mm BBI probe. Three spectra were recorded for each sample with NOESYPR1D, CPMGPR1D, and LEDBPGPPR2s1d pulse sequences (from Bruker pulse sequence library) with normal water presaturation. For all spectra, 256 transients were acquired with 32 k data points, a spectral width of 20 ppm, and the recycle delay of 2 s; 90° pulse length was set to ∼10 μs for each sample. For NOESEY1DPR spectra, mixing time was set to 0.1s. For CPMGPR1D spectra, the total transverse relaxation delay (2nτ) was 80 ms and τ value was set to 400 μs. The chemical shift was referenced to the anomeric proton resonance of α-glucose (δ 5.233). For urinary metabonomic analysis, an optimized method38 was adopted. In brief, a 60 μL urine sample was mixed with 4.8 μL of potassium fluoride solution (5 M) and 440 μL of water containing 20% D2O. After centrifugation (16 099g) for 10 min, 450 μL of supernatant was transferred to a 5 mm NMR tube followed by the addition of 1.8 μL of EDTA-d12 solution (0.333M) and 45 μL of phosphate buffer (1.5 M, pH 7.4, 100% D2O) containing 2.9 mM TSP and 0.01% of NaN3. For (liver, kidney, heart) tissues, hydrophilic metabolites were extracted using the previously reported methods.25,27,36,40,41 In brief, each sample (∼50 mg) was extracted three times using a tissuelyzer (Qiagen Tissue-Lyser, Retsch, Germany) and methanol/H2O (2:1 v/v) as solvent. Three soobtained supernatants were pooled together and freeze-dried after the removal of methanol in vacuo. The extracts were weighed and then dissolved in 550 μL of phosphate buffer (0.15M, pH7.45 and 70% D2O) containing 0.058 mM TSP and 0.01% of NaN3. After centrifugation (16 099g, 4 °C) for 10 min, 500 μL of supernatant was transferred to a 5 mm NMR tube for analysis. 1 H NMR spectra of urine samples and tissue extracts were recorded at 298 K on a Bruker Avance III 600 MHz NMR

Real-Time PCR Analyses

Total RNA in aortic arch was isolated by the Trizol reagent method (Transgen Biotech, Beijing, China) and reversetranscribed by use of the first-strand cDNA synthesis kit (Thermo Scientific, Rockford, IL). Amplification reactions were conducted according to the manufacturer’s protocol, and the gene expression levels were normalized to β-actin. Primers employed are listed in Table S2 in the SI. Plasma and Liver Fatty Acids

Quantitative measurements of fatty acids in plasma were conducted with the methods previously reported.29,39 For liver tissues, ∼2 mg sample in 0.1 mL of methanol was first homogenized with a tissuelyzer (20 Hz, 90s) three times and then treated in the same way as for plasma samples. After acetyl chloride catalyzed methyl esterification,29,39 the methylated fatty acids were measured on a Shimadzu GCMS-QP2010Plus spectrometer (Shimadzu Scientific Instruments) having a flame ionization detector (FID) and a mass spectrometer (MS) with an electron impact (EI) source. A DB-225 capillary GC column (Agilent) with a length of 10 m, inner diameter of 0.1 mm, and C



■

spectrometer equipped with a 5 mm cryogenic TCI probe using NOESYGPPR1D sequence. The water signal was saturated with a continuous wave pulse during the recycle delay (2 s) and mixing time (80 ms). 64 transients were collected and spectra were referenced to the methyl resonance of TSP (δ 0.000). For the purposes of resonance assignments, a set of 2D NMR spectra was acquired on some representative plasma, urine, liver, heart, and kidney extract samples with the spectral acquisition and processing parameters previously reported.40,42 These experiments included 1H−1H J-resolved spectroscopy (JRES), 1H−1H total correlation spectroscopy (TOCSY), 1 H−1H correlation spectroscopy (COSY), 1H−13C heteronuclear single quantum correlation spectroscopy (HSQC), and 1 H−13C heteronuclear multiple bond correlation spectroscopy (HMBC), some of which were acquired on a Bruker AVIII 850 MHz spectrometer.

Article

RESULTS

To investigate the mammalian metabolic changes associated with the pathogenesis and progression of atherosclerosis, we employed LDLR−/− and corresponding wild-type C57BL/6J mice as the pathology model and controls, respectively, which were fed with normal CD or a WD containing high fat, high cholesterol and high cholate. A diet-switch group was also introduced here with LDLR−/− mice fed with WD for 6 weeks followed by chow for further 6 weeks. Metabonomic analyses were conducted with NMR, while fatty acid composition analyses were conducted with GC−FID/MS. Complementary information was further obtained from comprehensive histopathological assessments, quantitative RT-PCR, and serum clinical chemistry including blood cholesterol, triglycerides (TG), glucose, HDL-C, LDL-C, total proteins (TP), albumin (ALB), and urea. Animal Phenotyping with Histopathological Assessment and Clinical Chemistry Analysis

Data Processing for Metabonomic Analysis

An exponential function was applied to all free induction decays with line-broadening factor of 1 Hz prior to Fourier transformation. Phase and baseline were manually corrected for 1H NMR spectra, and the spectral region δ 0.5−9.5 was integrated into bins with an equal width of 0.004 ppm (2 Hz) using an AMIX software package (V3.9.2, Bruker Biospin, Germany). Regions for residual solvent and urea signals were discarded for all samples. These regions were δ 4.35−5.17 for plasma, δ 4.4−5.37 and δ 5.41−6.2 for urine, δ 4.56−5.17 and δ 3.35−3.37 for liver extracts, δ 4.46−5.22 and δ 3.358−3.362 for kidney tissues, and δ 4.48−5.23 and δ 3.346−3.374 for heart tissue extracts. Each bucket was then normalized to the plasma volume for plasma and to the tissue weights for tissue extracts (to reflect the absolute metabolite concentration) whereas it was normalized to the sum of total integrals for urine samples prior to statistical data analysis. Multivariate data analysis was conducted using SIMCA-P+ software (V. 11.0, Umetrics, Sweden). Principal component analysis (PCA) was done on the mean-centered NMR data to generate an overview and find potential outliners. Orthogonal projection to latent structure with discriminant analysis (OPLSDA) was then carried out for the unit-variance scaled data with seven-fold cross-validation. The model quality was monitored by model parameters, Q2 for the predictability of the model and R2 for the interpretability of the model. CV-ANOVA tests were used to further assess the model validity.43,44 Following backtransformation, the loading plots indicating the treatmentinduced alterations of metabolites were plotted with the correlation coefficient color-coded for each data point45 using an in-house-developed Matlab script (MATLAB 7.1, the Mathworks, Natwick, MA). The color-coded correlation coefficients indicate the importance of metabolites contributing to the class separation with a “warm” color (e.g., red), one being more important than a “cold” color (e.g., blue) one. A correlation coefficient |r| greater than the cutoff value (depending on the sample numbers in each group) was considered to be statistically significant (p < 0.05).45 The choline, phosphorylcholine (PC), glycerophosphorylcholine (GPC), and D-3-hydroxybutyrate (3-HB) levels in plasma were calculated with a curve-fitting routine as peak area (arbitrary unit) per liter. Each variable was assessed for normal distribution and its intergroup statistical significance using the Student’s t test or nonparametric test with p value smaller than 0.05 as statistically significant.

C57BL/6J mice had no aortic lesions after feeding with either CD or WD for 12 weeks (Figure S1). No aortic lesions were observable for LDLR−/− mice after feeding with CD for 6 or 12 weeks (Figure S1 in the SI). In contrast, LDLR−/− mice showed significant formation of aorta lesions in the forms of aortic arch, thoracic aorta, and abdominal aorta lesions after WD feeding for 6 weeks; significantly more of such lesions were observable after the further 6 weeks of WD-feeding (Figure S1 in the SI). LDLR−/− mice in the diet-switch group had fewer such lesions than the animals without diet switch but significantly more aorta lesions than those fed with WD for only 6 weeks (Figure S1 in the SI). A further analysis of the lesion in aortic root revealed the WD-caused compositional changes (Figure S2 in the SI). Twelve weeks of WD feeding caused marked elevation of lipids (Figure S2a), inflammatory infiltration (Figure S2b in the SI) together with the collagen (Figure S2c in the SI), and macrophage levels (Figure S2d in the SI). For the dietswitched group, lipids, inflammatory infiltration, macrophage, and collagen levels were all lower than the WD group (12 weeks), although these remained higher than the CD group (Figure S2 in the SI). WD also caused significant level elevations for serum cholesterol, LDL-C, and TG accompanied by significant HDL-C decrease for LDLR−/− mice (Table S1 and Figure S3 in the SI). LDLR−/− mice on WD for 6 weeks showed about 80, 500, and 700% increases in blood TG, cholesterol, and LDL-C levels, respectively, compared with these on CD. With WD for 6 more weeks, blood TG, cholesterol, and LDL-C levels were not further elevated. Furthermore, compared with chow feeding, HDL-C level was decreased for ∼50% with WD feeding for 6 weeks, while it decreased further for ∼20% with WD feeding for a further 6 weeks. For C57BL/6J mice, in contrast, 6 weeks of WD feeding caused only ∼80% increase in cholesterol level but ∼60% decrease in TG level. Moreover, after 6 weeks of WD feeding, blood cholesterol and TG levels in LDLR−/− mice were 10 and 9 times higher, respectively, than in C57BL/6J mice, whereas with 6 weeks of CD, blood cholesterol and TG levels in LDLR−/− mice were about 2 to 3 times higher than in C57BL/6J mice. The blood cholesterol level for the diet-switched LDLR−/− mice was significantly lower than that of the WD-fed ones but remained higher than those animals on CD for 12 weeks. Both LDL-C and HDL-C levels had no significant differences D


Article

Journal of Proteome Research Table 1. Metabolites of LDLR−/− Mice in Plasma, Urine, and Liver, Kidney, and Heart Tissues no.

metabolites

1 2 3

bile acids cholesterol lipids

4

valine

5

leucine

6

isoleucine

7

D-3-hydroxybutyrate

8

lactate

9

threonine

10

alanine

11

lysine

12

acetate

13

NAGd

14

OAGd

15

arginine

moieties CH3 CH3 CH3 CH2 CH2CH2CO CH2CC CH2CO CCCH2CC CHCH γCH3 γ′CH3 βCH αCH COOH δCH3 δ′CH3 βCH2 γCH αCH COOH δCH3 β′CH3 γCH2 γCH2′ βCH αCH COOH γCH3 αCH2 αCH2′ βCH COOH βCH3 αCH COOH γCH3 αCH βCH COOH βCH3 αCH COOH γCH2 δCH2 βCH2 εCH2 αCH COOH CH3 COOH CH3 CH3′ CO CH3 CO γCH2 βCH2 δCH2 αCH CNH COOH

δ1H (multiplicity)a

δ13C

biomatricesb

0.65(brs) 0.84(s) 0.80(b) 1.27(m) 1.57(m) 2.01(m) 2.23(m) 2.76(m) 5.30(m) 1.00(d) 1.05(d) 2.27(m) 3.62(m)

14.0 20.9 16.6 32.3 27.6 29.5 35.6 28.4 131.2 19.7 20.8 31.8 63.3 177.5 24.2 24.6 44.3 27.1 52.3 178.2 17.5 13.9 27.6 27.6 39.2 62.5 176.7 24.6 49.7 49.7 69.1 183.3 23.1 71.4 185.2 22.5 63.5 68.7 175.8 19.2 53.1 178.5 24.5 29.6 32.8 42.6 55.5 177.4 26.2 184.1 25.1 25.1 176.4 23.2 177.3 26.6 30.5 43.2 57.2 159.8 177.5

P, L P, L P, H, K, L

0.96(d) 0.97(d) 1.70(m) 1.69(m) 3.73(m) 0.95(t) 1.01(d) 1.27(m) 1.47(m) 1.99(m) 3.68(m) 1.20(d) 2.31(dd) 2.41(dd) 4.17(m) 1.33(d) 4.11(q) 1.33(d) 3.60(d) 4.26(m) 1.48(d) 3.78(q) 1.49(m) 1.75(m) 1.91(m) 3.03(t) 3.76(t) 1.93(s) 2.04(s) 2.07(s) 2.14 (s) 1.70(m) 1.93(m) 3.25(m) 3.78(m)

E

P, H, K, L

P, H, K, L

P, H, K, L

P, H, K, L

P, H, K, L, U

P, H, K, L

P, H, K, L

P, H, K, L

P, H, K, L, U P, U

P P, H, K, L


Article

Journal of Proteome Research Table 1. continued no.

metabolites

16

aspartate

17

glutamate

18

glutamine

19

GSSGd

20

choline

21

PCd

22

GPCd

23

β-glucose

24

triglycerides

25

α-glucose

26

sarcosine

27

phenylalanine

28

tyrosine

moieties βCH2 βCH2′ αCH α-COOH COOH βCH2 γCH2 αCH α-COOH COOH βCH2 γCH2 αCH α-COOH COOH Glu-αCH Glu-γCH2 Glu-βCH2 Cys-CH Cys-CH′ Cys-αCH N(CH3)3 NCH2 OCH2 N(CH3)3 NCH2 OCH2 N(CH3)3 NCH2 OCH2 1-CH 2-CH 3-CH 4-CH 5-CH 6-CH 6-CH′ OCH2 OCH2′ OCH 1-CH 2-CH 3-CH 4-CH 5-CH 6-CH2 CH3 CH2 COOH 2,6-CH 4-CH 3,5-CH βCH2 βCH2′ αCH 1-C(ring) COOH 3,5-CH 2,6-CH βCH2 βCH2′


δ13C

biomatricesb

2.68(dd) 2.83(dd) 3.92(m)

39.3 39.9 54.5 176.4 180.2 29.8 36.2 57.1 177.3 183.8 29.8 36.2 57.9 176.8 180.5 56.9 34.3 29.1 41.4 41.4 55.5 57.2 70.1 58.2 57.2 77.5 61.1 56.9 69.1 62.5 99.3 77.5 79.3 72.9 79.2 63.7 63.9 62.9 62.9 72.2 95.4 74.9 76.2 72.7 74.4 63.7 39.4 50.1 185.8 132.3 130.3 131.9 39.5 39.5 59.3 138.1 176.9 118.8 133.6 38.2 38.2

P, H, K, L

2.06(m) 2.34(m) 3.75(t)

2.13(m) 2.46(m) 3.78(t)

3.71(m) 2.55(m) 2.17(m) 2.98(m) 3.31(m) 4.76(t) 3.20(s) 3.52(m) 4.08(m) 3.22(s) 3.60(m) 4.17(m) 3.24(s) 3.69(m) 4.33(m) 4.63(d) 3.26(dd) 3.46(dd) 3.40(dd) 3.47(dd) 3.74(dd) 3.90(dd) 4.06(m) 4.26(m) 5.20(m) 5.23(d) 3.54(dd) 3.73(dd) 3.40(dd) 3.83(dd) 3.83(dd) 2.76(s) 3.65(s) 7.33(m) 7.38(m) 7.43(m) 3.12(dd) 3.29(dd) 4.00(m)

6.91(d) 7.20(d) 3.06(dd) 3.17(dd) F

P, H, K, L

P, H, K, L

H, L

P, H, K, L

P, H, K, L

P, H, K, L

P, K, L, U

P

P, K, L, U

H, K, L, U

P, H, K, L

P, H, K, L


Article


metabolites

29

glycine

30

2-keto-isovalerate

31

3-methyl-2-ketovalerate

32

2-hydroxyisobutyrate

33

valerate

34

2-ketoisocaproate

35

α-L-fucose

36

β-L-fucose

37

sucrose

38

glycogen

39

2-ketoglutarate

40

asparagine

41

malate

moieties αCH 1-C(ring) 4-C(ring) COOH CH2 COOH CH3 CH2 CO COOH γCH3 CH3 βCH2 αCH CO COOH CH3 C COOH δCH3 γCH2 βCH2 αCH2 COOH CH3 CH CH2 CO COOH CH3 5-CH 4-CH 2-CH 1-CH CH3 5-CH 4-CH 2-CH 1-CH Glc-1-CH Glc-2-CH Glc-3-CH Frc-1-CH Frc-2-CH 1-CH 4-CH 2-CH 3, or 5-CH 6-CH2 CH2COOH CH2CO COOH CO αCH βCH2 βCH2′ α-COOH COOH CH2 CH2′ CH−OH


δ13C

3.93(m)

58.1 129.7 157.8 177.1 44.5 175.1 18.9 39.5 220.8 175.2 13.3 16.8 27.0 46.9 214.0 175.0 29.3 76.3 186.6 16.0 24.7 33.2 40.0 179.9 24.4 26.7 50.8 212.0 175.0 18.4 69.3 NDc 70.9 94.9 18.7 73.6 NDc NDc 99.8 95.2 75.6 NDc 79.3 76.7 102.6 NDc 74.7 NDc 65.5 33.3 40.4 184.5 208.5 54.1 37.5 37.5 176.2 177.2 45.4 45.4 73.2

3.57(s) 1.12(d) 3.02(m)

0.88(t) 1.05(d) 1.69(m) 2.92(m)

1.36(s)

0.88(t) 1.31(m) 1.60(m) 2.28(t) 0.93(dd) 2.09(m) 2.59(m)

1.21(d) 4.20(m) 3.80(m) 3.76(m) 5.21(d) 1.25(d) 3.80(m) 3.73(m) 3.62(m) 4.54(d) 5.40(d) 3.56(m) 3.75(m) 4.23(m) 4.05(m) 5.41(m) 3.42(m) 3.60(m) 3.72(m) 3.86(m) 2.45(t) 3.01(t)

4.00(dd) 2.96(dd) 2.86(dd)

2.37(dd) 2.67(dd) 4.29(dd) G

biomatricesb

P, H, K, L, U U

U

U

U

U

U

U

U

L

P, U

K

H, K, L


Article


metabolites

42

citrate

43 44

pyruvate 3-ureidopropionate

45

succinate

46

carnitine

47 48 49 50 51

trimethylamine methylamine dimethylamine succinimide putrescine

52

creatine

53

creatinine

54

tyrosine-O-β-glucuronide

55

methionine

56

β-alanine

57

ethanolamine

58 59

trimethylamine N-oxide betaine

60 61

methanol taurine

62

phenylacetate

63

guanidoacetate

64

tartrate

65 66

allantoin cis-aconitate


moieties αCOOH βCOOH CH2 CH2′ C−OH CH2COOH C-COOH CH3 CH2−COOH CH2−NH CH2 COOH N(CH3)3 COOH−CH2 CH2−N CH−OH CH3 CH3 CH3 CH2 CH2 CH2−NH2 CH3−N CH2 CH3 CH2 2, 6-CH 3, 5-CH 1−CH(glucuronide) 2−CH(glucuronide) 3−CH(glucuronide) 2−CH(glucuronide) 4-C(ring) 1-C(ring) αCH γCH2 SCH3 βCH2 COOH NCH2 CH2COOH COOH CH2−NH2 CH2−OH CH3 N(CH3)3 CH2 CH3 CH2−S CH2−N CH2 2, 6−CH(ring) 4−CH(ring) 3, 5-CH (ring) CH2 CNH COOH CH COOH CH CH2

2.56(d) 2.65(d)

2.36(s) 2.38(t) 3.31(q) 2.41(s) 3.22(s) 2.45(m) 3.43(m) 4.55(m) 2.88(s) 2.61(s) 2.72(s) 2.78(s) 1.77(m) 3.06(t) 3.03(s) 3.92(s) 3.04(s) 4.06(s) 7.13(d) 7.29(d) 5.13(d) 3.63(m) 3.90(m) 4.43(m)

3.86(t) 2.65(t) 2.14(s) 2.16(t) 3.18(t) 2.56(t) 3.14(t) 3.83(t) 3.28(s) 3.27(s) 3.91(s) 3.36(s) 3.27(t) 3.43(t) 3.55(s) 7.28(m) 7.29(m) 7.32(m) 3.80(s)

4.35(s) 5.37(s) 3.12(s) H

δ13C 183.6 182.4 48.3 48.3 78.3 181.8 184.6 29.4 40.4 39.9 36.7 184.8 57.2 45.5 73.5 67.3 47.8 27.6 37.5 41.6 27.3 40.3 39.2 56.7 33.1 59.2 120.1 133.1 103.2 90.9 73.5 73.4 130.3 159.0 56.7 31.6 16.5 32.7 177.3 40.0 36.7 181.8 44.3 60.8 62.3 56.1 68.7 51.6 50.4 38.4 46.4 131.8 132.9 133.2 47.5 159.9 178.1 77.1 181.3 66.3 46.5

biomatricesb

P, U

P, K, L, U U P, K, L, U H, L

U U U U U P, H, K, L, U U K

K

K

K, L, U U K, L, U H, K, L, H, K, L, U U

U

U U U


Article


metabolites

67 68 69

urea orotate fumarate

70

trans-aconitate

71

4-PYd

72

2-PYd

73

4-HPPAd

74

2-HPPAd

75

methylguanidine

76

hippurate

77

PAGd

78

4-CGd

79

indoxyl sulfate

80

inosine


moieties CH quaternary C CH2−COOH CH−COOH C−COOH NH2 CH CH COOH CH2 CH N−CH3 3-CH 2-CH 6-CH C NH2−CO CO N−CH3 3-CH 4-CH 6-CH C NH2−CO CO CH2 3, 5-CH (ring) 2, 6-CH (ring) CH3 CH 3, 5-CH (ring) 2, 6-CH (ring) CH3 CO NH−CH2 3, 5-CH 4-CH 2, 6-CH NH C CO COOH CH2(glycine) 2 or 6-CH 4-CH 3 or 5-CH CH3 2 or 6-CH 3 or 5-CH 5−CH(glucuronide) 4−CH(glucuronide) 3−CH(glucuronide) 4-C(ring) 1-C(ring) 8-CH 7-CH 2-CH 6-CH 9-CH 8-CH 2-CH

5.71(s)

5.78(brs) 6.19(s) 6.52(s) 3.44(s) 6.60(s) 3.89(s) 6.69(d) 7.83(dd) 8.56(d)

3.64(s) 6.66(d) 7.96(dd) 8.32(d)

4.01(s) 6.88(m) 7.18(m) 1.37(d) 3.58(q) 6.89(m) 7.20(m) 2.82(s) 3.97(s) 7.56(m) 7.62(m) 7.83(m) 8.55(brs)

3.68(s) 7.36(m) 7.37(m) 7.41(m) 2.30(s) 7.05(d) 7.23(d) 5.08(d) 3.61(m) 3.89(m)

7.20(dd) 7.27(dd) 7.36(s) 7.51(d) 7.72(d) 8.24(s) 8.35(s) I

δ13C

biomatricesb

127.2 146.6 181.1 177.9 179.3 104.4 138.0 177.7 39.2 133.7 47.4 123.1 146.3 150.0 120.5 170.9 180.7 41.5 121.1 142.4 145.4 117.0 167.6 172.1 47.1 118.8 133.6 21.4 50.2 118 135 30.2 158.7 46.7 131.9 134.9 130.4 ND 173.3 172.6 179.3 45.0 131.6 NDc 131.4 22.6 119.5 136.1 103.2 90.9 NDc 133.1 157.1 122.5 124.9 119.2 115.2 120.4 149.1 143.2

U U H, K, L, U U U

U

U

U

U U

U

U

U

H, K, L


Article


metabolites

81

xanthosine

82

uracil

83

AMPd

84

nicotinurate

85

nicotinamide N-oxide

86

acetamide

87

N-methylnicotinamide

88

glycolate

89

guanine

90

histidine

91

adipate

92

CMPd

93

hypoxanthine

94 95

formate guanosine

moieties 1−CH(ribose) 2−CH(ribose) 3−CH(ribose) 4-CH′(ribose) CH2 NH-CO 1−CH(ribose) 2−CH(ribose) CH NH−CH 5-CH2(ribose) 4−CH(ribose) 3−CH(ribose) 2−CH(ribose) 1−CH(ribose) 2−CH(adenosine) 8−CH(adenosine) NH−CH2 2-CH 6-CH 4-CH 5-CH 5-CH 6-CH 2-CH 4-CH CH3 CO CH3 5-CH 4-CH 6-CH 2-CH CH2 COOH CH C-NH2 CC CC 5-CH 2-CH COOH−CH2 CH2 C−CH(cytosine) N−CH(cytosine) 1-CH(ribose) 2-CH(ribose) 3-CH(ribose) 4-CH(ribose) 5-CH2(ribose) 5-CH2′(ribose) 2-CH 8-CH H-COOH CH(guanine) 1-CH(ribose) 2-CH(ribose) 3-CH(ribose) 4-CH(ribose) 5-CH(ribose) 5-CH′(ribose)


δ13C

6.11(d) 4.79(m) 4.43(dd) 4.28(m) 3.87(dd)

91.3 76.8 73.4 88.5 64.5 161.8 90.8 NDc 103.4 146.5 66.7 87.1 73.3 77.2 89.7 155.6 142.6 45.1 150.5 154.7 139.8 127.2 130.8 133.9 141.4 144.5 25.1 177.0 51.7 131.1 146.3 149.9 148.1 64.4 182.7 144.5 81.6 155.8 168.7 119.9 138.4 40.2 28.4 99.3 144.4 91.8 72.4 77.1 86.1 65.6 65.6 148.4 144.8 172 139.1 92.3 76.1 73.3 88.1 65.1 65.1

5.86 (d) 4.73 (m) 5.80(d) 7.54(d) 4.04(m) 4.37(m) 4.51(m) 4.81(m) 6.14(d) 8.27(s) 8.61(s) 4.01(s) 8.94(s) 8.71(d) 8.26(d) 7.60(dd) 7.74(m) 8.13(m) 8.75(m) 8.49(m) 2.00(s) 4.48(s) 8.19(m) 8.91(m) 8.97(d) 9.28(s) 3.96(s) 7.68(s)

7.10(s) 7.88(s) 2.18(t) 1.54(m) 6.13(d) 8.11(d) 5.99(m) 4.33(m) 4.33(m) 4.23(m) 4.05(m) 3.98(m) 8.19(s) 8.21(s) 8.46(s) 7.89(s) 5.92(d) 4.35(m) 4.39(m) 4.22(m) 3.87(d) 3.82(d) J

biomatricesb

L H, K, L H, K, L

H, K, L

U

U U

U U

H, K, L U K, L

H H, K, L K, L


Article

Journal of Proteome Research Table 1. continued a

s, singlet; brs, broad singlet; d, double; t, triple; m, multiplet; q, quartet; dd, double doublet. bP, plasma; H, heart; K, kidney; L, liver; U, urine. cND, signals not determined. dNAG, N-acetyl-glycoproteins; OAG; O-acetyl-glycoproteins; GSSG: oxidized glutathione; PC: phosphorylcholine; GPC: glycerophosphorylcholine; 4-PY, N1-methyl-4-pyridone-5-carboxamide; 2-PY, N1-methyl-2-pyridone-5-carboxamide; 4-HPPA, 4-hydroxyphenylpyruvate; 2-HPPA, 2-(4-hydroxyphenyl)propanoate; PAG: phenylacetylglycine; 4-CG: 4-cresol glucuronide; AMP, adenosine 5′- monophosphate; CMP, cytosine 5′-monophosphate.

contrast, the levels for bile acids, HDL, unsaturated fatty acids (UFA), polyunsaturated fatty acids (PUFA) (Figure 1a), 3hydroxybutyrate (Figure 1e), choline, and PC (Figure 1c) showed concurrent decreases. After diet switch to control diet for 6 weeks, the levels for LDL, VLDL, 3-hydroxybutyrate, and choline metabolites (i.e., choline, PC, GPC) and their ratios all recovered to levels for chow-feeding LDLR−/− mice (Figure 1b−e); however, the levels of HDL, bile acids, PUFA, and UFA remained higher than those on CD (Figure 1b,d). WD-fed LDLR−/− mice had significantly higher plasma levels in C16:0, C16:1n7, C18:0, C18:1, C18:2n6, C20:0, C20:1n9, C20:3n6, and C20:4n6 than the chow-fed animals (Table 2). The n6 PUFA level and n6-to-n3 ratio were 70−80% higher in WD-fed mice plasma than in chow-fed ones (Table 2). The ratios for PUFA-to-ToFA (total fatty acids), PUFA-to-UFA, and PUFA-to-MUFA in the WD-fed mice were significantly lower than those in chow-fed animals. The diet switch completely reversed these WD-caused changes for all plasma fatty acids and the aforementioned PUFA-related ratios (Table 2); however, the level of total PUFA in the diet-switched mice remained significantly higher than that in the chow-fed mice. Furthermore, WD feeding caused significant level elevations in urinary fucose, acetylglycoproteins, creatinine, 3-methyl-2ketovalerate (3M2KV), nicotinamide N-oxide (NMNO), phenylacetylglycine (PAG), phenylacetate, methylamine (MA), dimethylamine (DMA), guanine, 3-ureidopropionate (3-UP) and allantoin (Figure 2). Concurrently, WD feeding caused significant level declines for urinary citrate, succinate, 2ketoglutarate, 2-ketoisovalerate (2KIV), 2-ketoisocaproate (2KIC), N-methyl-2-pyridone-5-carboxamide (2-PY), N-methyl-4-pyridone-3-carboxamide (4-PY), 4-hydroxyphenylpyruvate (4-HPPA), hippurate, trimethylamine (TMA), taurine, and putrescine. Diet switch clearly but gradually alleviated these WD-induced urinary metabonomic changes to some extent (Figure 2). Switching to CD gradually alleviated the WDcaused urinary metabonomic changes, which were completely eliminated 3 weeks after such diet switch (Figure 2). Moreover, WD treatment caused significant level elevations for methionine (Met), valine (Val), leucine (Leu), isoleucine (Ile), glutamate (Glu), aspartate (Asp), arginine (Arg), phenylalanine (Phe), betaine, ethanolamine, fumarate, and xanthosine with concurrent level decreases for GSSG, taurine, carnitine, nicotinurate, hypoxanthine, and CMP (Figure 3) compared with chow-feeding. Diet switch reversed most of the WD-induced metabonomic changes in liver tissue. This is reflected by little compositional differences for hydrophilic metabolites in liver between the chow-fed and diet-switched mice. The only exception was observed for taurine, which had higher levels in the diet-switched mice. For hepatic fatty acids, nevertheless, WD-fed mice had significantly lower levels in C14:0, C16:0, C16:1n7, C18:2n6, C18:3n6, C20:4n6, C20:5n3, and C22:6n3 than chow-fed mice (Table 2). Compared with chow-feeding, WD treatment also caused significant decreases in both hepatic n6 and n3 PUFA, PUFA-to-MUFA, and PUFA-to-ToFA ratios accompanied by

between the diet-switched group and the chow-fed one (Figure S3 and Table S1 in the SI). WD-Caused Changes in Expression of the Inflammation-Related Genes in Aorta and Liver

Inflammatory infiltration was further assessed with the expression of inflammation-related genes. These included intercellular adhesion molecule 1 (ICAM-1) and monocyte chemoattractant protein-1 (MCP-1), inducible nitric oxide synthetase (iNOS), matrix metalloproteinase-2 (MMP2), and the markers of macrophage, CD68, and F4/80 in plaques together with hepatic CD68, interleukin-1β (IL-1β), and tumor necrosis factor α (TNFα). Results showed that WD feeding for 6 and 12 weeks caused significant expressional elevations for all of these genes (Figure S4 in the SI). After switching the diet to chow for 6 weeks, the mRNA levels for all of those genes were lower than such levels without diet-switch (WD group), although they remained significantly higher than those in LDLR−/− mice fed with chow for 12 weeks, with the only exception for hepatic IL-1β (Figure S4 in the SI). WD-Caused Metabonomic Changes in Plasma, Urine, and Liver, Kidney, and Myocardial Tissues 1

H NMR spectra of plasma, urine, and hydrophilic extracts of liver, kidney, and myocardial tissues (Figure S5 in the SI) were measured as the metabonomic profiles for mice on CD, WD, and WD-to-CD (diet switch), respectively. Metabolites in these biological matrices were unambiguously assigned (Table 1) based on the literature data40,46−48 and in-house databases and were further confirmed individually, as previously reported39,49 using 1H−1H COSY, 1H−1H TOCSY, 1H JRES, 1H−13C HSQC, and 1H−13C HMBC 2D NMR spectral data. These matrices had different but complementary metabolite information. For instance, the plasma samples contained 32 identified metabolites including acetylglycoproteins, lipoproteins together with amino acids, glucose, ketones, TCA cycle intermediates, and choline metabolites. In urine, we detected 55 metabolites including TCA cycle intermediates and cometabolites from mice and their gut microbiota together with metabolites of amino acids, choline, vitamin B3, purines, and pyrimidines. Methyl signals for acetylglycoproteins were also observed in urine samples with whose large-molecule features further confirmed with a diffusion-edited experiment (data not shown). Among tissues examined here, more than 40 metabolites were assigned for both liver and kidney whereas 35 metabolites were assigned for myocardium. Furthermore, 15 fatty acids were measured quantitatively (with GC−FID/MS) for blood plasma and liver tissues including C16:0, C18:0, C20:0, C22:0, and their corresponding unsaturated forms (Table 2). To the best of our knowledge, no data have been reported on fatty acid composition so far for both plasma and liver of LDLR−/− mice fed with WD or treated with diet-switch. Plasma results showed that WD feeding for 12 weeks caused significant elevation of plasma LDL and VLDL together with the choline-to-PC, choline-to-GPC, and GPC-to-PC ratios for LDLR−/− mice compared with chow-feeding (Figure 1). In K


C14:0 C16:0 C18:0 C20:0

L

28.6 288.5 16147.5 9.965 0.739 0.240

4.7 81.0 2206.9 2.199 0.081 0.014

± ± ± ± ± ±

6.5 60.2 1168.5 0.409 0.137 0.011

± ± ± ± ± ±

29.8 269.6 5822.8 5.783 1.387 0.333

151.8 ± 32.6 878.5 ± 161.9

1135.0 6.4 725.6 452.0 12.4 937.4 91.6 899.9 13.6 721.8 455.4

81.5 ± 26.0 629.7 ± 141.3

6227.1 25.5 4085.4 2154.6 29.7 4865.0 279.4 4373.3 49.0 3685.3 2104.8

± ± ± ± ± ± ± ± ± ± ±

452.7 6.3 293.4 176.7 3.0 432.1 66.6 247.1 4.4 415.3 194.7

± ± ± ± ± ± ± ± ± ± ±

2526.1 24.2 1760.5 789.5 11.2 1568.1 230.1 1229.0 20.4 1987.4 1016.2

± ± ± ± ± ± ± ± ± ± ± 242.5 5.3 211.6 141.7 3.1 404.2 46.4 355.7 7.0 586.4 211.9

32.9 311.6 7367.1 5.830 1.379 0.327

± ± ± ± ± ± 9.7 80.0 2277.4 0.531 0.276 0.018

96.3 ± 47.2 720.2 ± 138.3

2700.2 24.4 1898.8 875.3 12.1 1623.9 201.9 1425.3 24.8 2486.0 1139.4

PC (n = 16)a

2.92 × 10−1 8.69 × 10−2 10−1 10−1 10−2 10−1 10−1 10−1

3.04 × 10−07 8.93 × 10−5 6.87 × 10−1 4.70 × 10−1 6.40 × 10−9 2.22 × 10−03 2.40 × 10−8 7.34 × 10−10 3.87 1.11 9.14 8.22 9.32 4.17

10−1 10−1 10−1 10−1 10−1 10−1 10−1 10−1 10−2 10−2 10−1

2.85 9.51 1.59 1.54 3.96 7.27 2.00 3.00 5.91 3.23 1.08

2.16 × 10−9 6.05 × 10−1 2.30 × 10−12 5.33 × 10−10 4.03 × 10−5 4.59 × 10−12 9.87 × 10−2 5.10 × 10−6 3.55 × 10−6 9.16 × 10−6 2.26 × 10−8

× × × × × ×

× × × × × × × × × × ×

p PC vs PApb

p PB vs PAb 62.2 1.3 47.8 12.9 0.3 89.5 13.2 74.7 1.6 58.7 29.6 0.8 0.2 1.5 15.8 1.4 0.4 9.4 210.5 4.279 0.705 0.289

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 8.6 0.3 7.2 2.0 0.1 16.1 3.2 14.5 0.6 12.3 8.4 0.4 0.1 0.2 1.2 0.5 0.1 2.0 29.9 0.372 0.091 0.019

LA (n = 12)a 47.2 0.7 32.9 13.3 0.3 90.8 6.2 82.6 2.1 38.2 18.7 0.4 0.2 1.7 11.4 0.7 0.2 5.7 180.1 4.656 0.475 0.236

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 8.9 0.4 8.4 1.5 0.1 26.4 3.5 24.2 0.7 4.3 4.6 0.3 0.1 0.4 0.8 0.2 0.1 0.7 35.8 0.278 0.162 0.045

LB (n = 14)a 63.2 1.1 47.9 13.1 0.3 85.0 10.6 72.5 1.8 57.7 29.4 0.7 0.2 1.4 14.3 1.2 0.3 8.8 205.9 4.339 0.709 0.281

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 9.2 0.2 7.4 2.3 0.1 18.0 3.5 14.8 0.8 10.5 7.2 0.3 0.0 0.3 0.7 0.3 0.1 1.7 28.8 0.288 0.208 0.041

LC (n = 14)a 2.18 1.50 7.21 5.20 9.71 8.84 2.02 3.21 6.96 9.47 9.55 2.23 6.77 3.04 8.74 5.42 6.67 3.74 2.87 8.31 3.98 2.60

× × × × × × × × × × × × × × × × × × × × × ×

10−4 10−4 10−5 10−1 10−1 10−1 10−5 10−1 10−2 10−5 10−4 10−2 10−1 10−1 10−10 10−4 10−5 10−5 10−2 10−3 10−4 10−3

p LB vs LAb 7.86 7.77 9.74 7.58 2.27 5.05 6.68 7.07 5.43 8.28 9.45 3.43 2.76 3.92 1.71 3.50 8.14 2.53 6.94 6.52 9.43 5.47

× × × × × × × × × × × × × × × × × × × × × ×

10−1 10−2 10−1 10−1 10−2 10−1 10−2 10−1 10−1 10−1 10−1 10−1 10−1 10−1 10−3 10−1 10−3 10−1 10−1 10−1 10−1 10−1

p LC vs LAb

a

Data are presented as mean ± SD. P, plasma (μmol/L); L, liver (μmol/g). A−C denote treatment with (A) chow, (B) Western-type diets for 12 weeks, and (C) Western-type diet for 6 weeks followed by chow diet for a further 6 weeks. bp values obtained from the Student’s t test or nonparametric analysis as approriate. cSFA, saturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids; ToFA, total fatty acids; n6, n6 PUFA; n3, n3 PUFA; n6/n3, n6-to-n3 ratios; PUFA/MUFA, PUFA-to-MUFA ratio; PUFA/ToFA, PUFA-to-ToFA ratio; SFA, the sum of C14:0, C16:0, C18:0, and C20:0; MUFA, the sum of C16:1n7, C18:1n7/9, and C20:1n9; PUFA, the sum of n6 and n3 types of PUFA; n6, the sum of C18:2n6, C18:3n6, C20:2n6, C20:3n6, and C20:4n6; n3, the sum of C18:3n3, C20:5n3, and C22:6n3. −, not detected.

MUFAc C16:1n7 C18:1n7/9 C20:1n9 PUFAc C18:2n6 C18:3n6 C20:2n6 C20:3n6 C20:4n6 C18:3n3 C20:5n3 C22:6n3 ToFAc n6/n3c PUFA/MUFAc PUFA/ToFAc

SFAc

PB (n = 16)a

PA (n = 15)a

Table 2. Fatty Acid Composition for Plasma and Liver Samples from LDLR−/− Mice

Journal of Proteome Research Article


Article


Figure 1. Plasma metabonomic changes for the LDLR−/− mice treated with Western-type diets for 12 weeks (B12) and Western-type diet for 6 weeks followed by chow diet for further 6 weeks (C) compared with treatment with chow diet for 12 weeks (A12). (a) OPLS-DA results for B12 and A12 groups with one PLS component and an orthogonal component (left: scores plots; right: loadings plots). (b) Intergroup metabonomic differences between C and A12 groups derived from univariate statistics of all variables by treating each variable (metabolite) as independent using the Student’s t test (when two strict criteria were met) or nonparametric test (when two strict criteria were not met). (c,d) Levels for choline, phosphorylcholine, glycerophosphorylcholine (arbitrary unit), and their ratios. (e) Levels for D-3-hydroxybutyrate (3-HB) (arbitrary unit). Data in panels c−e were obtained from curve-fitting the 1H NMR spectra. * and # indicated significant differences in groups C and B12, respectively, compared with A12. *, p < 0.05; **/##, p < 0.01; ###, p < 0.001. Cho, choline; PC, phosphorylcholine; GPC, glycerophosphorylcholine.

Figure 2. Dynamic urinary metabonomic changes for the LDLR−/− mice fed with (a) Western-type diets for 12 weeks and (b) Westerntype diet for 6 weeks followed by chow diet for a further 6 weeks compared with chow diet feeding. The heatmaps were generated from correlation coefficients of metabolite changes as a function of feeding period (i.e., weeks in horizontal axis) with warm (red) colored bars denoting statistical significant elevation, cold colored (blue) ones indicating decreases compared with chow diet feeding, and white ones indicating no statistical significance. 2-KG, 2-ketoglutarate; NMNO, nicotinamide N-oxide; 4-PY, N1-methyl-4-pyridone-5-carboxamide; 2PY, N1-methyl-2-pyridone-5-carboxamide; 3-UP, 3-ureido-propionate; PAC, phenylacetate; PAG, phenylacetylglycine; DMA, dimethylamine; 3M2KV, 3-methyl-2-ketovalerate; TMA, trimethylamine; 4-HPPA, 4hydroxyphenylpyruvate; 2KIV, 2-ketoisovalerate; IS, indoxyl sulfate; 4CG, 4-cresol-O-β-glucuronide;.

∼10% increases for the n6-to-n3 PUFA ratio (Table 2). The diet switch completely reversed these WD-induced changes in hepatic fatty acids to large extent. The only exception was observed for 20:4n6 and 20:5n3 levels that remained lower than their levels in the chow-fed mice (Table 2). In renal tissue, WD feeding led to significant elevation of Ile, Leu, Met, Lys, Phe, Asp, tyrosine-O-β-glucuronide (Tyr-O-βglu), and β-alanine but decreases of taurine and lactate levels compared with chow feeding (Figure 3). The diet switch completely reversed most of these changes with an exception for inosine, Asp, and β-alanine, whose levels were higher than those in the chow-fed mice. In myocardium, the WD treatment only caused a significant decrease in taurine level compared with chow feeding, and such change was completely reversed after diet switch for 6 weeks (Figure 3).

WD treatment, although the C16:1-to-C16:0 ratio showed significant changes, both of which were dietary fatty acids (Figure 4). After switching the diet to CD for 6 weeks, the expression levels of scd-1 became comparable to the chow-diet group and d5d expression was significantly higher than both the CD and WD groups.

■

DISCUSSION LDLR−/− mice fed with CD for 12 weeks did not develop atherosclerosis (Figure S1 in the SI) with the plasma cholesterol level of ∼7 mM (Table S1 in the SI); however, when they were fed a high-fat and high-cholesterol diet (mimicking WD) for 12 weeks, these mice had obvious aorta lesions showing elevated lipids, inflammatory cells (macrophage), and collagen levels (Figures S1 and S2 in the SI) together with over 600% increase in plasma cholesterol level (Table S1 in the SI). This is fully consistent with previous observations,11−14 implying the success of developing the atherosclerotic model. WD feeding also promoted macrophage infiltration and the gene expressions for ICAM-1, iNOS, MCP1, and MMP-2 in aorta together with TNF-α, IL-1β, and CD68 in liver (Figure S4 in the SI). This agreed well with the literature observations for mice17,19,22,50 and human athero-

Enzyme Variations in Hepatic Tissue

Because the changes of hepatic fatty acids were responsive to WD treatment, the expression of genes related to fatty acid metabolism was measured with quantitative RT-PCR (Figure 4). These include scd1 encoding stearoyl-CoA desaturase 1 (SCD-1) for Δ9-desaturation of fatty acids and d5d and d6d (data not shown) encoding Δ5-desaturase (D5D) and Δ6desaturase (D6D), respectively, for desaturation of polyunsaturated fatty acids. The results showed that WD caused significant down-regulation of hepatic scd-1 (compared with CD), whereas d5d and d6d were not significantly responsive to M


Article


Figure 3. Metabonomic changes for liver, kidney, and myocardial tissues from the LDLR−/− mice treated with Western-type diet for 12 weeks (B12) and Western-type diet for 6 weeks followed by chow diet for further 6 weeks (C) compared with chow diet feeding for 12 weeks (A12). Red bars indicate significant increases whereas the blue ones denote decreases; black bars mean no significant differences. Horizontal axis denotes the ratios for metabolite changes, (Ctreated − Cchow)/Cchow, with empty bars representing (CB12 − CA12)/CA12 and the shaded bars representing (Cc − CA12)/CA12. AMP, adenosine 5′monophosphate; CMP, cytosine 5′-monophosphate; Tyr-O-β-Glu, tyrosine-O-β-glucuronide; GSSG, oxidized glutathione.

Figure 5. Schematic representation of metabolic changes in the LDLR−/− mice associated with atherosclerosis induced by the Western-type diet.

Western-Type Diet Disrupts Cholesterol and Bile Acids Homeostasis in LDLR−/− Mice

WD-caused drastic cholesterol and LDL-C elevation in LDLR−/− mice plasma (Table S1 in the SI) agreed well with literature observations11−14 due to dysfunction of LDL receptors. Consequently, cellular LDL uptakes cannot be fulfilled,leading to the LDL and ox-LDL accumulation in intima thus atherosclerosis.11−14 Such dysfunction can also cause fatty liver and gallstone observed in this study (data not shown) due to hepatic uptakes of a large amount of fats and cholesterol in chylomicron via lymphatic capillary. Our observation of the significant taurine decrease in urine, liver, kidney, and heart of the WD-fed mice (Figure 5) was probably due to the demands for conjugating the added cholate to meet the increased emulsification requirement of fatty acids in WD (i.e., high-fat and high-cholesterol diet). WD-caused decrease in plasma bile acids (Figure 1a) is also explainable by the increased demand of bile acids for fatty acid absorption. These taurine changes could also result from the increased excretion of taurine-conjugated bile acids caused by cholesterol accumulation in the liver.37,53 Supportive evidence is available with taurine-enhanced conversion of cholesterol into bile acids via stimulation of cholesterol 7α-hydroxylase activity.53,54 With diet switch for 6 weeks, all of these WD-induced changes were partially alleviated, indicating the usefulness of diet management even after pathogenesis of atherosclerosis.

Figure 4. Liver fatty acid desaturation in the LDLR−/− mice fed with chow diet for 12 weeks (A12), Western-type diet for 12 weeks (B12), and Western-type diet for 6 weeks followed by chow diet for further 6 weeks (C). The C16:1-to-C16:0 ratio (C16:1/C16:0) was from GCFID results, whereas the gene expression results were from quantitative RT-PCR analysis of relative mRNA levels of stearoylCoA desaturase 1 (SCD1) and Δ5-desaturase (D5D). * and # indicate significant differences in groups B12 and C, respectively, compared with group A12. *, p < 0.05; **/##, p < 0.01; ###, p < 0.001.

sclerotic lesions,51 indicating the occurrence of the proatherosclerotic oxidative stress and inflammation.52 In this study, we paid particular attention to whether diet switch (to normal chow after 6 weeks of WD feeding) reversed the WD effects and to what degree, which was not reported for the time being although closely relevant to the effects of diet management on atherosclerosis development. We found that when switching diet to chow (for 6 weeks), all of these parameters for atherosclerotic lesions and inflammatory makers were substantially decreased, although not to the levels of chow feeding (Figures S1−S4 in the SI). Furthermore, we found that the diet switch largely but incompletely reversed the WD-induced systematic metabolic alterations in multiple biological matrices including biofluids (plasma, urine) and tissues (liver, kidney, and myocardium) of LDLR−/− mice (Figure 5).

WD-Caused Fatty Acid Metabolism Disorders

WD-induced significant elevations of TG (Table S1 in the SI), LDL, VLDL, and decline of HDL in blood plasma (Figure 1a) together with the fatty acid changes in both plasma and liver (Table 2) indicated that WD caused disorders in fatty acid metabolism (Figure 5). Even after diet-switch for 6 weeks, such disorders did not reverse completely (Table S1 in the SI, Table 2). Close inspection indicated that the plasma levels of all SFA, MUFA, and n6-PUFA in the WD group were higher than in control group (Table 2). In contrast, the levels of liver fatty acids were lower in the WD group than in the chow group, probably due to the combined effects of direct absorption from the high-fat diet and LDL receptor knockout. WD feeding did not cause any statistical differences for the plasma n3-PUFA N


Article


sulfate, 4-HPPA, and hippurate63(Figures 2 and 5), indicated that gut microbiota probably played important roles in atherosclerosis development and progression. Such notion is further supported by our observation of the WD-induced elevation of urinary DMA and MA, both of which are choline metabolites of gut microbiota. This is in broad agreement with the findings that gut microbiota degradation products of choline modulate lipid metabolism and glucose homeostasis in nonalcoholic fatty liver disease with implications in the dietinduced obesity,64 cardiovascular diseases,65 and intestinal inflammation;66 however, the roles that gut microbiota plays in the atherosclerosis development remain to be fully understood together with mechanistic details. Diet switch gradually enabled these WD-induced changes to return toward the normal diet situation and thus obviously indicated dietary intervention as an effective way to prevent from further development of atherosclerosis-caused metabolic alterations.

levels, probably due to anti-inflammation function of their metabolites, and these PUFAs were essential fatty acids. After diet switch, the levels of all fatty acids returned to that of the chow group, indicating that such changes of liver fatty acid profiles resulted from the combined effects of the WD and LDL receptor knockout. Total PUFA levels in WD group collectively remained higher than those in chow-fed animals (Figure 5), and diet switch only partially rebalanced the WD-induced changes. This suggests diet switch as an effective (but not curable) way for alleviating atherosclerosis development. The WD-caused decrease in C16:1n7-to-C16:0 ratio in both plasma and liver probably indicates the WD-induced downregulation of SCD1 because C16:1n7 is chiefly derived from endogenous pathways55 and SCD1 as the enzyme catalyzing conversion of C16:0 into C16:1n7 is the rate-limiting step of unsaturated fatty acid synthesis. Quantitative RT-PCR results for scd1 (Figure 4) agreed with this note nicely. This WDinduced SCD1 change was probably related to the WD-caused alterations in cholesterol and triglyceride metabolisms previously discussed because the scd1 knockout mouse showed impaired biosynthesis of hepatic cholesterol esters and triglycerides.55 Furthermore, WD feeding led to declines for the PUFA-to-MUFA ratio in both plasma and liver (Figure 5), suggesting WD-induced enhancement in oxidation of fatty acids, which was consistent with the consequence of downregulation of SCD1.56,57 Diet switch reverted most of the WD-induced changes in the levels of plasma and liver fatty acids (Table 2). The levels of C20:4n6 and SCD1 expression in liver remained significantly lower than those in the control group (Table 2, Figure 4). This probably indicates that the WD-induced inflammation is only partially recoverable with limited dietary intervention and clearly warrants further detailed studies.

WD-Caused Changes in Metabolism of Choline, Vitamin B3, Amino Acids, Purines, and Pyrimidines

Significant decreases in choline, PC, GPC, and their ratios in WD-fed mice plasma illustrated that cell membrane metabolism was associated with atherosclerosis progression (Figure 5). This notion is further supported by the WD-induced ethanolamine elevation in LDLR−/− mice liver (Figure 3a). Nevertheless, alterations in osmoregulations are probably also implicated with the WD-induced elevation of hepatic betaine because these choline metabolite are also osmoprotectants.67 Furthermore, the elevation of urinary nicotinamide N-oxide (NMNO) together with the depletion of N1-methyl-4pyridone-5-carboxamide (4-PY) and N1-methyl-2-pyridone-5carboxamide (2-PY) in WD-fed mice urine samples (Figure 2) suggested that the metabolism of nicotinate and nicotinamide (vitamin B3) was probably associated with the progression of atherosclerosis (Figure 5). Because nicotinamide is a component of NAD involved in intracellular respiration to oxidize fuel substrates29 while NMNO, 4-PY, and 2-PY are oxidation metabolites of nicotinamide in liver, the changes of these nicotinamide metabolites are probably also indications for the WD-induced oxidative stress, as observed for treatment with just high-fat diets.33,68 Our observations for level changes of urinary 3-methyl-2-ketovalerate, 2-ketoisocaproate, and 2ketoisovalerate in the WD group further indicated that metabolism of the branched-chain amino acids (Ile, Leu, Val) catalyzed by branched-chain aminotransferase 1 (BCAT1) played some roles in the atherosclerosis development. The elevation of both hepatic and renal amino acids (Figure 3a,b) together with plasma total protein (Table S1 in the SI) indicated the WD-caused promotions of protein biosynthesis and biosynthesis of amino acids via pyruvate and acetyl-CoA (via TCA cycle in the case of glutamate). Elevation of urinary allantoin, guanine, 3-UP, renal β-alanine, and hepatic xanthosine together with decreased hepatic hypoxanthine in WD-fed mice illustrated that purine and pyrimidine metabolisms were vitally associated with atherosclerosis development (Figure 5). The diet switch also enabled these WD-induced metabolic changes to reverse back to normal, even though incompletely within 6 weeks.

WD-Caused Dysfunction of Energy Metabolism, Enhanced Oxidative Stress, and Inflammation

WD feeding further caused comprehensive changes in energy metabolisms (Figure 5) highlighted by WD-induced disruptions in glucose homeostasis including decreases in plasma glucose (Table S1 in the SI), renal lactate (Figure 3), and urinary excretion of TCA intermediates (Figure 2), being consistent with the previous results.37 This and the elevation of hepatic fumarate (Figure 3a) were also consistent with the previously reported up-regulation of hepatic genes encoding aconitate hydratase (aco2), oxoglutarate dehydrogenase (ogdh), succinate dehydrogenase (sdhb), and malate dehydrogenase (mdh2) for atherosclerosis mice.58 Significantly lower PUFA-to-MUFA (and PUFA-to-UFA) ratios in both the plasma and liver of WD-fed mice (Table 2) indicated the WD-induced enhancement of oxidative stress and inflammation59 (Figure 5). This agreed with the WD-enhanced expressions of inflammatory biomarkers (Figure S4 in the SI) and the findings that atherosclerosis caused oxidative stress in ApoE-deficient mice.60 This notion is further supported by elevation of urinary allantoin, which is an indicator for oxidative stress.61 WD-induced significant increases in the n6-to-n3 ratios for both plasma and liver PUFA (Table 2) indicated the potential risk of WD feeding to heart diseases because the elevation of n6-to-n3 PUFA ratios was found to be a risk factor for cardiovascular and other chronic diseases.62

■

CONCLUSIONS LDLR−/− mice developed atherosclerosis within 6 weeks when fed with WD. Such atherosclerotic progression was accompanied by drastic alterations of metabolic network involving

WD-Caused Disturbance of Gut Microbiota Functions

WD-caused changes in the gut microbiota related urinary metabolites, including phenylacetate, PAG, 4-CG, indoxyl O



■

ACKNOWLEDGMENTS We acknowledge the Ministry of Science and Technology of China (2010CB912500 and 2012CB934000) and NSFC (21175149, 81130002, and 91439102) for financial support.

multiple pathways at the systems level (Figure 5). Such changes included WD-induced disruptions to homeostasis of cholesterol, bile acids, and fatty acids, disturbance of biosynthesis of amino acids and proteins, and alterations in gut microbiota metabolic functions together with changes for choline and nucleotide metabolisms. Such WD intake also caused inflammation and oxidative stress in multiple tissues. Gut microbiota functions are clearly implicated in the WD-induced development of atherosclerosis (Figure 5). Switching to normal diet even after the occurrence of atherosclerosis enabled LDLR−/− mice to reprogram their metabolism and reverse the atherosclerosis-associated metabolic alterations to a large extent but inflammation to less extent. These findings provided vital metabolic information associated with atherosclerosis development and demonstrated that the integrated analysis by combining the metabonomic analysis (using NMR and GC− FID/MS) and molecular biology was a powerful approach for understanding the biochemical aspects of cardiovascular conditions. Further work is ongoing to test the generality of these findings for the atherosclerosis development in normal animals, especially the importance of gut microbiota.

■

■

ABBREVIATIONS LDLR, low-density lipoprotein receptor; NMR, nuclear magnetic resonance; PCA, principal component analysis; OPLS-DA, orthogonal projection to latent structure with discriminant analysis; CPMG, Carr−Purcell−Meiboom−Gill; SCD1, stearoyl-coenzyme A desaturase 1; D5D, delta-5 desaturase; D6D, delta-6 desaturase; GPC, glycerophosphorylcholine; PC, phosphorylcholine; BAs, bile acids; 2-KG, 2ketoglutaric acid; 3-HB, D-3-hydroxybutyrate; NAG, Nacetylglycoproteins; OAG, O-acetylglycoproteins; PUFA, polyunsaturated fatty acid; MUFA, monounsaturated fatty acid; SFA, saturated fatty acid; ToFA, total fatty acid; UFA, unsaturated fatty acid; TG, triglyceride; TCA cycle, tricarboxylic acid cycle; 4-CG, 4-cresol glucuronide; PAG, phenylacetylglycine; PAC, phenylacetate; 4-HPPA, 4-hydroxyphenylpyruvic acid; DMA, dimethylamine; TMA, trimethylamine; DMG, dimethylglycine; TMAO, trimethylamine N-oxide; 3M2KV, 3-methyl-2-ketovalerate; 2KIC, 2-ketoisocaproate; 2KIV, 2-ketoisovalerate; 3-HIV, 3-hydroxy-isovalerate; EDTA, ethylenediaminetetraacetic acid; GC-FID, gas chromatographyflame ionization detector; TSP-d4, sodium 3-trimethylsilyl [2,2,3,3-d4] propionate; ICAM-1, intercellular adhesion molecule-1; IL-1β, interleukin-1β; iNOS, inducible nitric oxide synthetase; MCP-1, monocyte chemoattractant protein-1; MMP2, matrix metalloproteinase-2; TNFα, hepatic tumor necrosis factor α

ASSOCIATED CONTENT

S Supporting Information *

Figure S1. Aortic lesions for C57BL/6J mice treated with CD for 12 weeks and WD for 12 weeks together with the LDLR−/− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 12 weeks, and WD for 6 weeks followed by CD for further 6 weeks. Figure S2. Histopathological assessments for the LDLR−/− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 12 weeks, and WD for 6 weeks followed by CD for further 6 weeks. Figure S3. Total plasma cholesterol, HDLC and LDL-C levels for LDLR−/− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 6 weeks, WD for 12 weeks, and WD for 6 weeks followed by CD for further 6 weeks. Figure S4. Results of inflammation factors for the LDLR−/−− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 12 weeks, and WD for 6 weeks followed by CD for further 6 weeks. Figure S5. Typical 600 MHz 1H NMR spectra of plasma, urine, and extracts of liver, kidney, heart from the LDLR−/− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 12 weeks, and WD for 6 weeks followed by CD for further 6 weeks. Table S1. Serum clinical chemistry data for the LDLR−/− mice treated with CD for 6 weeks, CD for 12 weeks, WD for 6 weeks and 12 weeks, WD for 6 weeks followed by CD for further 6 weeks. Table S2. Primers used for RT-PCR analysis of expressions of some genes. This material is available free of charge via the Internet at http://pubs.acs.org.

■

Article

■

REFERENCES

(1) Libby, P. Inflammation in atherosclerosis. Nature 2002, 420 (6917), 868−874. (2) Weber, C.; Noels, H. Atherosclerosis: current pathogenesis and therapeutic options. Nat. Med. 2011, 17 (11), 1410−1422. (3) Spann, N. J.; Garmire, L. X.; McDonald, J. G.; Myers, D. S.; Milne, S. B.; Shibata, N.; Reichart, D.; Fox, J. N.; Shaked, I.; Heudobler, D.; Raetz, C. R. H.; Wang, E. W.; Kelly, S. L.; Sullards, M. C.; Murphy, R. C.; Merrill, A. H.; Brown, H. A.; Dennis, E. A.; Li, A. C.; Ley, K.; Tsimikas, S.; Fahy, E.; Subramaniam, S.; Quehenberger, O.; Russell, D. W.; Glass, C. K. Regulated accumulation of desmosterol integrates macrophage lipid metabolism and inflammatory responses. Cell 2012, 151 (1), 138−152. (4) Freigang, S.; Ampenberger, F.; Weiss, A.; Kanneganti, T. D.; Iwakura, Y.; Hersberger, M.; Kopf, M. Fatty acid-induced mitochondrial uncoupling elicits inflammasome-independent IL-1 alpha and sterile vascular inflammation in atherosclerosis. Nat. Immunol. 2013, 14 (10), 1045−1053. (5) Lusis, A. J. Atherosclerosis. Nature 2000, 407 (6801), 233−241. (6) Potteaux, S.; Ait-Oufella, H.; Mallat, Z. Mouse models of atherosclerosis. Drug Discovery Today 2007, 4 (4), 165−170. (7) Breslow, J. L. Mouse models of atherosclerosis. Science 1996, 272 (5262), 685−688. (8) Feig, J. E.; Parathath, S.; Rong, J. X.; Mick, S. L.; Vengrenyuk, Y.; Grauer, L.; Young, S. G.; Fisher, E. A. Reversal of hyperlipidemia with a genetic switch favorably affects the content and inflammatory state of macrophages in atherosclerotic plaques. Circulation 2011, 123 (9), 989−998. (9) Zhang, S. H.; Reddick, R. L.; Piedrahita, J. A.; Maeda, N. Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein-E. Science 1992, 258 (5081), 468−471. (10) Plump, A. S.; Smith, J. D.; Hayek, T.; Aaltosetala, K.; Walsh, A.; Verstuyft, J. G.; Rubin, E. M.; Breslow, J. L. Severe hyper-

AUTHOR INFORMATION

Corresponding Authors

*H.T.: Tel: +86-(0)21-51630725. Fax: +86-(0)27-87199291. E-mail: [email protected]. *Y.Z.: E-mail: [email protected]. Author Contributions #

D.L. and L.Z. contributed equally to this work.

Notes

The authors declare no competing financial interest. P


Article


testicles in the preobesity state. J. Proteome Res. 2013, 12 (7), 3480− 3495. (26) Brindle, J. T.; Antti, H.; Holmes, E.; Tranter, G.; Nicholson, J. K.; Bethell, H. W. L.; Clarke, S.; Schofield, P. M.; McKilligin, E.; Mosedale, D. E.; Grainger, D. J. Rapid and noninvasive diagnosis of the presence and severity of coronary heart disease using H-1-NMRbased metabonomics. Nat. Med. 2002, 8 (12), 1439−1444. (27) Liu, Y.; Yan, X.; Mao, G. M.; Fang, L.; Zhao, B. L.; Liu, Y. H.; Tang, H. R.; Wang, N. P. Metabonomic profiling revealed an alteration in purine nucleotide metabolism associated with cardiac hypertrophy in rats treated with thiazolidinediones. J. Proteome Res. 2013, 12 (12), 5634−5641. (28) Kirschenlohr, H. L.; Griffin, J. L.; Clarke, S. C.; Rhydwen, R.; Grace, A. A.; Schofield, P. M.; Brindle, K. M.; Metcalfe, J. C. Proton NMR analysis of plasma is a weak predictor of coronary artery disease. Nat. Med. 2006, 12 (7), 862−862. (29) Li, H. H.; An, Y. P.; Zhang, L. L.; Lei, H. H.; Zhang, L. M.; Wang, Y. L.; Tang, H. R. Combined NMR and GC-MS analyses revealed dynamic metabolic changes associated with the carrageenaninduced rat pleurisy. J. Proteome Res. 2013, 12 (12), 5520−5534. (30) Dong, F. C.; Wang, B.; Zhang, L. L.; Tang, H. R.; Li, J. S.; Wang, Y. L. Metabolic response to klebsiella pneumoniae infection in an experimental rat model. PLoS One 2012, 7 (11), e51060. (31) Xu, W. X.; Wu, J. F.; An, Y. P.; Xiao, C. N.; Hao, F. H.; Liu, H. B.; Wang, Y. L.; Tang, H. R. Streptozotocin-induced dynamic metabonomic changes in rat biofluids. J. Proteome Res. 2012, 11 (6), 3423−3435. (32) He, Q. H.; Ren, P. P.; Kong, X. F.; Wu, Y. N.; Wu, G. Y.; Li, P.; Hao, F. H.; Tang, H. R.; Blachier, F.; Yin, Y. L. Comparison of serum metabolite compositions between obese and lean growing pigs using an NMR-based metabonomic approach. J. Nutr. Biochem. 2012, 23 (2), 133−139. (33) An, Y. P.; Xu, W. X.; Li, H. H.; Lei, H. H.; Zhang, L. M.; Hao, F. H.; Duan, Y. X.; Yan, X.; Zhao, Y.; Wu, J. F.; Wang, Y. L.; Tang, H. R. High-fat diet induces dynamic metabolic alterations in multiple biological matrices of rats. J. Proteome Res. 2013, 12 (8), 3755−3768. (34) He, Q. H.; Tang, H. R.; Ren, P. P.; Kong, X. F.; Wu, G. Y.; Yin, Y. L.; Wang, Y. L. Dietary supplementation with L-arginine partially counteracts serum metabonome induced by weaning stress in piglets. J. Proteome Res. 2011, 10 (11), 5214−5221. (35) Teague, C.; Holmes, E.; Maibaum, E.; Nicholson, J.; Tang, H. R.; Chan, Q. N.; Elliott, P.; Wilson, I. Ethyl glucoside in human urine following dietary exposure: detection by H-1 NMR spectroscopy as a result of metabonomic screening of humans. Analyst 2004, 129 (3), 259−264. (36) Shi, X. H.; Xiao, C. N.; Wang, Y. L.; Tang, H. R. Gallic acid intake induces alterations to systems metabolism in rats. J. Proteome Res. 2013, 12 (2), 991−1006. (37) Cheng, K. K.; Benson, G. M.; Grimsditch, D. C.; Reid, D. G.; Connor, S. C.; Griffin, J. L. Metabolomic study of the LDL receptor null mouse fed a high-fat diet reveals profound perturbations in choline metabolism that are shared with ApoE null mice. Physiol. Genomics 2010, 41 (3), 224−231. (38) Jiang, L. M.; Huang, J.; Wang, Y. L.; Tang, H. R. Eliminating the dication-induced intersample chemical-shift variations for NMR-based biofluid metabonomic analysis. Analyst 2012, 137 (18), 4209−4219. (39) Zhang, L. M.; Wang, Y. L.; Xu, Y. X.; Lei, H. H.; Zhao, Y.; Li, H. H.; Lin, X. S.; Chen, G. Z.; Tang, H. R. Metabonomic analysis reveals efficient ameliorating effects of acupoint stimulations on the menopause-caused alterations in mammalian metabolism. Sci. Rep. 2014, 4, 3641. (40) Zhang, L. M.; Ye, Y. F.; An, Y. P.; Tian, Y. A.; Wang, Y. L.; Tang, H. R. Systems responses of rats to aflatoxin B1 exposure revealed with metabonomic changes in multiple biological matrices. J. Proteome Res. 2011, 10 (2), 614−623. (41) Jiang, L. M.; Huang, J.; Wang, Y. L.; Tang, H. R. Metabonomic analysis reveals the CCl4-induced systems alterations for multiple rat organs. J. Proteome Res. 2012, 11 (7), 3848−3859.

cholesterolemia and atherosclerosis in apolipoprotein-E deficient mice created by homologous recombination in es cells. Cell 1992, 71 (2), 343−353. (11) Ishibashi, S.; Brown, M. S.; Goldstein, J. L.; Gerard, R. D.; Hammer, R. E.; Herz, J. Hypercholesterolemia in low-densitylipoprotein receptor knockout mice and its reversal by adenovirusmediated gene delivery. J. Clin. Invest. 1993, 92 (2), 883−893. (12) Ishibashi, S.; Herz, J.; Maeda, N.; Goldstein, J. L.; Brown, M. S. The 2-receptor model of lipoprotein clearance- tests of the hypothesis in knockout mice lacking the low-density-lipoprotein receptor, apolipoprotein-E, or both proteins. Proc. Natl. Acad. Sci. U.S.A. 1994, 91 (10), 4431−4435. (13) Ishibashi, S.; Goldstein, J. L.; Brown, M. S.; Herz, J.; Burns, D. K. Massive xanthomatosis and atherosclerosis in cholesterol-fed lowdensity-lipoprotein receptor-negative mice. J. Clin. Invest. 1994, 93 (5), 1885−1893. (14) 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 (1), 44−56. (15) Robbins, C. S.; Hilgendorf, I.; Weber, G. F.; Theurl, I.; Iwamoto, Y.; Figueiredo, J. L.; Gorbatov, R.; Sukhova, G. K.; Gerhardt, L. M. S.; Smyth, D.; Zavitz, C. C. J.; Shikatani, E. A.; Parsons, M.; van Rooijen, N.; Lin, H. Y.; Husain, M.; Libby, P.; Nahrendorf, M.; Weissleder, R.; Swirski, F. K. Local proliferation dominates lesional macrophage accumulation in atherosclerosis. Nat. Med. 2013, 19 (9), 1166−1172. (16) Ait-Oufella, H.; Taleb, S.; Mallat, Z.; Tedgui, A. Recent advances on the role of cytokines in atherosclerosis. Arterioscler., Thromb., Vasc. Biol. 2011, 31 (5), 969−979. (17) Iiyama, K.; Hajra, L.; Iiyama, M.; Li, H. M.; DiChiara, M.; Medoff, B. D.; Cybulsky, M. I. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ. Res. 1999, 85 (2), 199−207. (18) Kowala, M. C.; Recce, R.; Beyer, S.; Gu, C.; Valentine, M. Characterization of atherosclerosis in LDL receptor knockout mice: macrophage accumulation correlates with rapid and sustained expression of aortic MCP-1/JE. Atherosclerosis 2000, 149 (2), 323− 330. (19) Graham, T. L.; Mookherjee, C.; Suckling, K. E.; Palmer, C. N. A.; Patel, L. The PPAR delta agonist GW0742X reduces atherosclerosis in LDLR(−/−) mice. Atherosclerosis 2005, 181 (1), 29−37. (20) Han, J. M.; Kim, M. J.; Baek, S. H.; An, S.; Jin, Y. Y.; Chung, H. G.; Baek, N. I.; Choi, M. S.; Lee, K. T.; Jeong, T. S. Antiatherosclerotic effects of artemisia princeps pampanini cv. sajabal in LDL receptor deficient mice. J. Agric. Food Chem. 2009, 57 (4), 1267−1274. (21) Nozue, T.; Yamamoto, S.; Tohyama, S.; Fukui, K.; Umezawa, S.; Onishi, Y.; Kunishima, T.; Sato, A.; Nozato, T.; Miyake, S.; Takeyama, Y.; Morino, Y.; Yamauchi, T.; Muramatsu, T.; Hibi, K.; Terashima, M.; Michishita, I. Low serum docosahexaenoic acid is associated with progression of coronary atherosclerosis in statin-treated patients with diabetes mellitus: results of the treatment with statin on atheroma regression evaluated by intravascular ultrasound with virtual histology (TRUTH) study. Cardiovasc. Diabetol. 2014, 13, 13. (22) Lim, H. J.; Park, J. H.; Lee, S.; Choi, H. E.; Lee, K. S.; Park, H. Y. PPAR delta ligand L-165041 ameliorates Western diet-induced hepatic lipid accumulation and inflammation in LDLR−/− mice. Eur. J. Pharmacol. 2009, 622 (1−3), 45−51. (23) Tang, H. R.; Wang, Y. L. Metabonomics: a revolution in progress. Prog. Biochem. Biophys. 2006, 33 (5), 401−417. (24) Nicholson, J. K.; Lindon, J. C.; Holmes, E. ’Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica 1999, 29 (11), 1181−1189. (25) Duan, Y. X.; An, Y. P.; Li, N.; Liu, B. F.; Wang, Y. L.; Tang, H. R. Multiple univariate data analysis reveals the inulin effects on the high-fat-diet induced metabolic alterations in rat myocardium and Q


Article

Journal of Proteome Research (42) Tian, Y.; Zhang, L.; Wang, Y.; Tang, H. Age-Related Topographical Metabolic Signatures for the Rat Gastrointestinal Contents. J. Proteome Res. 2012, 11 (2), 1397−1411. (43) Eriksson, L.; Trygg, J.; Wold, S. CV-ANOVA for significance testing of PLS and OPLS (R) models. J. Chemometr. 2008, 22 (11− 12), 594−600. (44) Trygg, J.; Wold, S. Orthogonal projections to latent structures (O-PLS). J. Chemometr. 2002, 16 (3), 119−128. (45) Cloarec, O.; Dumas, M. E.; Trygg, J.; Craig, A.; Barton, R. H.; Lindon, J. C.; Nicholson, J. K.; Holmes, E. Evaluation of the orthogonal projection on latent structure model limitations caused by chemical shift variability and improved visualization of biomarker changes in H-1 NMR spectroscopic metabonomic studies. Anal. Chem. 2005, 77 (2), 517−526. (46) 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. H-1 NMR metabonomics of plasma lipoprotein subclasses: elucidation of metabolic clustering by self-organising maps. NMR Biomed. 2007, 20 (7), 658−672. (47) Nicholson, J. K.; Foxall, P. J. D.; Spraul, M.; Farrant, R. D.; Lindon, J. C. 750-MHz H-1 and H-1-C-13 NMR-spectroscopy of human blood-plasma. Anal. Chem. 1995, 67 (5), 793−811. (48) 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 their spectroscopic counterparts in H-1 NMR metabonomic data. BMC Bioinf. 2007, 8, S8. (49) Dai, H.; Xiao, C. N.; Liu, H. B.; Tang, H. R. Combined NMR and LC-MS analysis reveals the metabonomic changes in salvia miltiorrhiza bunge induced by water depletion. J. Proteome Res. 2010, 9 (3), 1460−1475. (50) Bieghs, V.; Hendrikx, T.; van Gorp, P. J.; Verheyen, F.; Guichot, Y. D.; Walenbergh, S. M. A.; Jeurissen, M. L. J.; Gijbels, M.; Rensen, S. S.; Bast, A.; Plat, J.; Kalhan, S. C.; Koek, G. H.; Leitersdorf, E.; Hofker, M. H.; Lutjohann, D.; Shiri-Sverdlov, R. The cholesterol derivative 27hydroxycholesterol reduces steatohepatitis in mice. Gastroenterology 2013, 144 (1), 167−178. (51) Perrotta, I.; Brunelli, E.; Sciangula, A.; Conforti, F.; Perrotta, E.; Tripepi, S.; Donato, G.; Cassese, M. iNOS induction and PARP-1 activation in human atherosclerotic lesions: an immunohistochemical and ultrastructural approach. Cardiovasc. Pathol. 2011, 20 (4), 195− 203. (52) Ponnuswamy, P.; Ostermeier, E.; Schrottle, A.; Chen, J.; Huang, P. L.; Ertl, G.; Nieswandt, B.; Kuhlencordt, P. J. Oxidative stress and compartment of gene expression determine proatherosclerotic effects of inducible nitric oxide synthase. Am. J. Pathol. 2009, 174 (6), 2400− 2410. (53) Rull, A.; Vinaixa, M.; Angel Rodriguez, M.; Beltran, R.; Brezmes, J.; Canellas, N.; Correig, X.; Joven, J. Metabolic phenotyping of genetically modified mice: an NMR metabonomic approach. Biochimie 2009, 91 (8), 1053−1057. (54) Murakami, S.; Kondo-Ohta, Y.; Tomisawa, K. Improvement in cholesterol metabolism in mice given chronic treatment of taurine and fed a high-fat diet. Life Sci. 1998, 64 (1), 83−91. (55) Zhou, Y. E.; Egeland, G. M.; Meltzer, S. J.; Kubow, S. The association of desaturase 9 and plasma fatty acid composition with insulin resistance-associated factors in female adolescents. Metabolism 2009, 58 (2), 158−166. (56) Hulver, M. W.; Berggren, J. R.; Carper, M. J.; Miyazaki, M.; Ntambi, J. M.; Hoffman, E. P.; Thyfault, J. P.; Stevens, R.; Dohm, G. L.; Houmard, J. A.; Muoio, D. M. Elevated stearoyl-CoA desaturase-1 expression in skeletal muscle contributes to abnormal fatty acid partitioning in obese humans. Cell Metab. 2005, 2 (4), 251−261. (57) Browning, J. D.; Horton, J. D. Molecular mediators of hepatic steatosis and liver injury. J. Clin. Invest. 2004, 114 (2), 147−152. (58) Zhou, C. C.; Pridgen, B.; King, N.; Xu, J. X.; Breslow, J. L. Hyperglycemic Ins2(Akita)Ldlr(−/−) mice show severely elevated lipid levels and increased atherosclerosis: a model of type 1 diabetic macrovascular disease. J. Lipid Res. 2011, 52 (8), 1483−1493.

(59) Serkova, N. J.; Jackman, M.; Brown, J. L.; Liu, T.; Hirose, R.; Roberts, J. P.; Maher, J. J.; Niemann, C. U. Metabolic profiling of livers and blood from obese Zucker rats. J. Hepatol. 2006, 44 (5), 956−962. (60) Mayr, M.; Chung, Y. L.; Mayr, U.; Yin, X. K.; Ly, L.; Troy, H.; Fredericks, S.; Hu, Y. H.; Griffiths, J. R.; Xu, Q. B. Proteomic and metabolomic analyses of atherosclerotic vessels from apolipoprotein Edeficient mice reveal alterations in inflammation, oxidative stress, and energy metabolism. Arterioscler., Thromb., Vasc. Biol. 2005, 25 (10), 2135−2142. (61) Grootveld, M.; Halliwell, B. Measurement of allantoin and uric acid in human-body fluids- a potential index of free-radical reactions invivo. Biochem. J. 1987, 243 (3), 803−808. (62) Simopoulos, A. P. The importance of the omega-6/omega-3 fatty acid ratio in cardiovascular disease and other chronic diseases. Exp. Biol. Med. 2008, 233 (6), 674−688. (63) Dong, F. C.; Zhang, L. L.; Hao, F. H.; Tang, H. R.; Wang, Y. L. Systemic responses of mice to dextran sulfate sodium-induced acute ulcerative colitis using H-1 NMR spectroscopy. J. Proteome Res. 2013, 12 (6), 2958−2966. (64) Sun, Y.; Wang, Y. H.; Qu, K.; Zhu, H. B. Beneficial effects of cordycepin on metabolic profiles of liver and plasma from hyperlipidemic hamsters. J. Asian Nat. Prod. Res. 2011, 13 (6), 534−546. (65) Nicholson, J. K.; Holmes, E.; Kinross, J.; Burcelin, R.; Gibson, G.; Jia, W.; Pettersson, S. Host-gut microbiota metabolic interactions. Science 2012, 336 (6086), 1262−1267. (66) Murdoch, T. B.; Fu, H.; MacFarlane, S.; Sydora, B. C.; Fedorak, R. N.; Slupsky, C. M. Urinary metabolic profiles of inflammatory bowel disease in interleukin-10 gene-deficient mice. Anal. Chem. 2008, 80 (14), 5524−5531. (67) 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 (2), 620− 630. (68) Jung, J. Y.; Kim, I. Y.; Kim, Y. N.; Kim, J. S.; Shin, J. H.; Jang, Z. H.; Lee, H. S.; Hwang, G. S.; Seong, J. K. H-1 NMR-based metabolite profiling of diet-induced obesity in a mouse mode. BMB Rep. 2012, 45 (7), 419−424.

R


Metabonomic Changes Associated with Atherosclerosis Progression

Recommend Documents