Elevated Homocysteine Reduces Apolipoprotein A-I Expression in Hyperhomocysteinemic Mice and in Males With Coronary Artery Disease
http://www.100md.com
Leonie G. Mikael, Jacques Genest, Jr, Ri
参见附件。
the Departments of Human Genetics and Pediatrics (L.G.M., R.R.), Montreal Children’s Hospital
The Department of Medicine (J.G.), Royal Victoria Hospital, McGill University Health Center, Montreal, Canada.
Abstract
Hyperhomocysteinemia, a risk factor for cardiovascular disease, is caused by nutritional or genetic disturbances in homocysteine metabolism. A polymorphism in methylenetetrahydrofolate reductase (MTHFR) is the most common genetic cause of mild hyperhomocysteinemia. To examine mechanisms by which an elevation in plasma homocysteine leads to vascular disease, we first performed microarray analyses in livers of Mthfr-deficient mice and identified differentially expressed genes that are involved in lipid and cholesterol metabolism. Microarrays and RT-PCR showed decreased mRNA for apolipoprotein A (ApoA)-IV and for ApoA-I and increased mRNA for cholesterol 7 hydroxylase (Cyp7A1) in Mthfr+/– mice compared with Mthfr+/+ mice. Western blotting revealed that ApoA-I protein levels in liver and plasma of Mthfr+/– mice were 52% and 62% of levels in the respective tissues of Mthfr+/+ mice. We also performed Western analysis for plasma ApoA-I protein levels in 60 males with coronary artery disease and identified a significant (P<0.01) negative correlation (–0.33) between ApoA-I and plasma homocysteine levels. This cohort also displayed a negative correlation (–0.24, P=0.06) between high-density lipoprotein cholesterol and plasma homocysteine. Treatment of HepG2 cells with supraphysiological levels of 5 mmol/L homocysteine reduced peroxisome proliferator-activated receptor (PPAR) and ApoA-I protein levels and decreased ApoA-I promoter activity. Transfection with a PPAR construct upregulated ApoA-I and MTHFR. Our results suggest that hyperhomocysteinemia may increase risk of atherosclerosis by decreasing expression of ApoA-I and increasing expression of CYP7A1.
Key Words: homocysteine methylenetetrahydrofolate reductase apolipoprotein A-I high-density lipoprotein peroxisome proliferator-activated receptor
Introduction
Conventional risk factors for cardiovascular disease including hypercholesterolemia, hypertension, smoking, and diabetes account for approximately 50% of all cases.1 Evidence now indicates that hyperhomocysteinemia (HHcy), which occurs in 5% to 7% of the general population, is an important risk factor for atherosclerotic and thrombotic disease.2 The enzyme 5,10-methylenetetrahydrofolate reductase (MTHFR) plays a critical role in regulating total homocysteine (tHcy) levels by synthesizing 5-methyltetrahydrofolate, which provides the methyl group for homocysteine remethylation to methionine. Mild deficiency of MTHFR, the most common genetic cause of mild HHcy, occurs frequently in many populations (10% to 15% of individuals) and is attributable to homozygosity for a polymorphism (677CT) that reduces enzyme activity.3 To generate an animal model for HHcy, we had created mice with a heterozygous (+/–) disruption of the Mthfr gene.4 With 50% residual activity and 1.7-fold greater levels of plasma tHcy compared with their wild-type (+/+) littermates (5.4±1.0 versus 3.2±1.2, mean±SD),4 the Mthfr+/– mice are similar biochemically to humans with mild MTHFR deficiency. Plasma total folate levels are somewhat lower but not significantly reduced in Mthfr+/– mice (192±19.7 versus 214.7±12).5 In addition, these mice have lipid deposition in the aortic valvular region,4 decreased endothelial-dependent vasodilatation in mesenteric vessels,6 and stiffer small arteries with increased collagen deposition.7
The relationship between coronary artery disease and low levels of plasma high-density lipoprotein (HDL) cholesterol has been demonstrated in several prospective studies.8 Apolipoprotein A (ApoA)-I is the major protein component of HDL. Transgenic mice that overexpress human ApoA-I have increased HDL levels and reduced susceptibility to diet-induced atherosclerosis.9 Although it is not completely understood, the protective effect of ApoA-I may be related to its association with increased HDL production. HDL can be protective10 by such mechanisms as prevention of lipoprotein oxidation,11 decreased expression of adhesion molecules on vascular cells,12 and increased cholesterol efflux from cells such as macrophages that are involved in the initial development of the atherosclerotic plaque.13
A region on the long arm of chromosome 11q23-q24 encodes genes for 3 lipoproteins (apoA-I, apoC-III, and apoA-IV), which are clustered within a 17-kb segment.14 Comparative sequence analysis has identified a relatively newer apolipoprotein, apoA-V, located approximately 30-kb proximal to the apoA-I/C-III/A-IV gene cluster.15
The peroxisome proliferator-activated receptors (PPARs) are ligand-activated transcription factors that belong to the superfamily of nuclear hormone receptors; 4 isotypes are known: PPAR, PPAR, PPAR, and PPAR. PPAR, which is predominantly expressed in the liver, is involved in lipid metabolism and stimulates fatty acid oxidation and ketogenesis.16 The relevance of PPAR pathways to metabolic disease is underscored by the use of fibrates (PPAR agonists) to treat hyperlipidemia.17 PPAR decreases triglyceride concentrations by increasing the expression of lipoprotein lipase, and increases HDL cholesterol levels by regulating the production of ApoA-I,18 ApoA-II,19 and ApoA-V.20 It also controls reverse cholesterol transport by increasing cholesterol efflux from macrophages and inhibits transcription of the gene encoding cholesterol 7 hydroxylase (CYP7A1).21 Cyp7A1 is a liver-specific enzyme that catalyzes the 7 hydroxylation of cholesterol, the rate-limiting step in the conversion of cholesterol into bile acids.22 A negative correlation between high Hcy levels and PPAR expression has been suggested in humans.23 In a murine model of HHcy, PPAR was shown to ameliorate endothelial dysfunction.24
In this study, we used our Mthfr-deficient mouse model to study the potential interaction between homocysteine and lipid metabolism. Initial microarray findings of decreased apolipoprotein expression in liver of hyperhomocysteinemic mice were pursued by additional studies in murine tissues, in a human cohort of coronary artery disease, and in HepG2 cells. Our findings suggest that HHcy may contribute to cardiovascular disease by decreased expression of ApoA-I and increased expression of CYP7A1, mediated by PPAR.
Materials and Methods
Animal Model of Hyperhomocysteinemia
Mthfr-deficient mice (Mthfr+/–) and wild-type (Mthfr+/+) littermates were generated as reported.4 Mice were progeny from matings of Mthfr+/– mice backcrossed for at least 12 generations from 129/Sv-BALB/c F1 heterozygotes to BALB/cAnNCrlBR (Charles River, Saint-Constant, Quebec, Canada). Experimentation was approved by the Animal Care Committee of the Montreal Children’s Hospital. Mice were fed standard mouse chow (Harlan Teklad LM-485 Mouse, code 7012). Blood was collected from the tail vein of 8-month-old mice, after overnight fast, in EDTA-vacutainer tubes. Plasma was obtained by centrifugation (6000 rpm, 5 minutes, 4°C) and stored at –70°C until analysis. Mice were euthanized a few days later and livers were snap frozen in liquid nitrogen and stored at –70°C. Mthfr genotypes were determined as reported.4 Mice with a heterozygous disruption of the cystathionine--synthase (Cbs) gene were purchased from The Jackson Laboratory (Bar Harbor, Maine) and bred with C57Bl/6J mice to obtain heterozygous and wild-type mice. Genotypes were determined as reported.25
RNA Isolation, Microarray Analysis, and RT-PCR
Frozen mouse liver was pulverized in liquid nitrogen, and total RNA was isolated using TRIzol reagent (Invitrogen). Liver RNA from 3 pairs of Mthfr+/– mice and wild-type Mthfr+/+ littermates was analyzed by Affymetrix GeneChip Expression Arrays at the McGill University/Genome Québec Innovation Center (Montreal, Canada) using the murine chip MG-U74Av2; mice were matched for age (8 months) and gender (2 male and 1 female pair). Affymetrix software was used to assess genes with differential expression between wild-type and Mthfr-deficient mice, as described previously.26,27 For RT-PCR experiments, RNA from mouse liver was also DNase treated and reverse transcribed with random hexamers and Superscript II as recommended by the manufacturer (Invitrogen). The cDNA was used in RT-PCRs with primers for mouse ApoA-I, ApoA-IV, Cyp7A1, and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH). Quantitation of RT-PCR products was performed by measuring the intensity of each band using Quantity One 4.1.0 software and SPSS software for statistical analysis. Intensity of signal from the relevant genes was divided by the intensity of the signal from GAPDH, which served as control. To confirm RT-PCR results, quantitative real-time RT-PCR (q-RT-PCR) was performed using Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) in a MX4000 QPCR System (Stratagene) following the recommendations of the manufacturers.
Preparation of Crude Liver Extracts
Frozen mouse liver was pulverized in liquid nitrogen and suspended in ice-cold radioimmunoprecipitation assay lysis buffer (1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS [sodium dodecyl sulfate] in PBS) containing a protease inhibitor mixture (Complete Mini, Roche). After incubation on ice for 30 minutes, lysates were centrifuged (13000 rpm, 15 minutes, 4°C). Supernatants were collected and protein concentrations assessed.
SDS-PAGE and Western Blotting
Electrophoresis (50 μg of protein from liver extracts, 25 μg of protein from cultured cells, 1 μL of mouse or human plasma) was performed on SDS–polyacrylamide gels. Proteins were transferred to nitrocellulose membranes at 67 V for 2 hours at 4°C. Nonspecific binding sites were blocked with 5% skim milk powder in Tris-buffered saline/Tween buffer overnight at 4°C. Primary antibodies were: rabbit anti-mouse ApoA-I (Biodesign International), anti-human ApoA-I (Calbiochem), anti–-actin (Sigma), anti-albumin (Bethyl Laboratories), anti-PPAR (Santa Cruz Biotechnology), and anti-Mthfr (described by Frosst et al3). Secondary antibody was a peroxidase-coupled anti-rabbit IgG (Amersham Biosciences). Signal detection was achieved with ECL Plus chemiluminescence system (Amersham Biosciences) and exposure to x-ray films. Signals were quantified with Quantity One 4.1.0 software. SPSS software was used for statistical analysis.
Clinical Studies
The patient population consisted of a cohort of coronary artery disease patients from Montreal that had previously been characterized for homocysteine levels and other relevant parameters before treatment was initiated.28,29 Mean plasma tHcy concentration for the cohort is 10.8 μmol/L.29 We selected 60 plasma samples from males, aged 31 to 59 years, with a range of Hcy levels having a mean of 10.5 μmol/L and an equal distribution above and below this mean, and examined immunoreactive ApoA-I protein by Western blotting.
Plasmid Constructs for Human ApoA-I Promoter
The human ApoA-I promoter fragment was amplified from HepG2 DNA, between positions –330 and +69 relative to the transcription start site, using primers 5'-TATAGCTAGCAACACAATGGACAATGGCAACTG-3' and 5'-TATAAAGCTTGA ACCTTGAGCTGGGGAGC-3' that incorporated NheI and HindIII restriction sites, as previously reported.30 The amplified fragment, which contained the peroxisome-proliferator receptor element (PPRE), was digested with NheI and HindIII and inserted between the respective sites in pGL3-basic vector to create an ApoAI-wt plasmid. Mutations in the PPRE were generated by PCR using primers 5'-ACTGATCCCTTGTCCCCTGCCCTGCAGCCCCCGCA-3' and 5'-AGGGGACAAGG GATCAGTGGGGGCGGGAGGGGAGT-3' (mutations underlined) to create the ApoAI-mut plasmid.30 Integrity of the inserts was verified by sequencing.
Plasmid Construct for Human PPAR
The human PPAR sequence was amplified from HepG2 cDNA and cloned into expression vector pCMV-Myc (BD Biosciences). After transformation into INVF’ Competent Escherichia coli cells (Invitrogen), potential clones carrying the human PPAR were screened and inserts from the positive clones were verified by sequencing.
HepG2 Cell Culture and Transient Transfection Assays
Human HepG2 hepatoma cells (American Type Culture Collection) were propagated at 37°C, in 5% CO2, in DMEM-F12 with 10% FBS, 100 μg/mL penicillin and streptomycin. For transfections, cells were seeded at approximately 5x105 into 6-well plates and medium was changed regularly. At 90% confluence, cells were transfected with either 2 μg of vector containing PPAR or pCMV vector alone using Lipofectamine 2000 Reagent in OPTI-MEM medium (Invitrogen) following the recommendations of the manufacturer. Cells were harvested 72 hours after transfection. RNA isolation, cDNA synthesis, RT-PCR, and q-RT-PCR were performed as above. Increased expression of PPAR was confirmed by RT-PCR in transfected cells. For Western blotting, cells were washed with ice-cold PBS and treated as described above. For transfection with ApoAI promoter-reporter plasmids, HepG2 cells were cotransfected with pCMV--galactosidase. Firefly luciferase activity was detected by a Luciferase Assay Kit (Promega); transfection efficiencies were controlled by normalizing to -galactosidase activity (Applied Biosystems). Transfections were performed at least 4 times.
Homocysteine Treatment of HepG2 Cells
At 80% confluence, the medium was replaced with DMEM-F12 supplemented with 0, 1, 5, or 10 mmol/L DL-homocysteine or L-cysteine (Sigma). Cells were collected at specific time points for up to 24 hours, washed with ice-cold PBS and treated for Western blotting. In promoter experiments, Hcy was added 6 hours after transfection by replacing OPTI-MEM with 0, 1, or 5 mmol/L Hcy in DMEM-F12; cells were harvested 24 hours after transfection.
Results
Effects of Mthfr Deficiency on ApoA-IV, ApoA-I, and Cyp7A1 mRNA Levels in Mouse Liver
The Affymetrix software algorithm was used to analyze microarray results and examine differentially expressed genes in 3 pairs of livers from Mthfr+/– mice compared with Mthfr+/+ littermates. We chose genes with a greater than 1.8-fold change (increase or decrease) and a difference of 150 U of fluorescence in the raw data analysis (Table I in the online data supplement available at http://circres.ahajournals.org). We were particularly interested in genes that were implicated in lipid and cholesterol metabolism and honed in on 2 such genes, ApoA-IV and Cyp7A1. Mthfr+/– mice had reduced expression of ApoA-IV compared with Mthfr+/+ mice in all 3 pairs, with an average fold change of –2.6. Compared with their Mthfr+/+ littermates, Mthfr+/– mice showed increased expression of Cyp7A1 in all 3 pairs, with an average fold change of +2.1. Microarray analysis also revealed a decrease in ApoA-I mRNA levels in 2 pairs of Mthfr+/– mice compared with their wild-type littermates; the fold change, however, did not reach statistical significance. RT-PCR was performed to validate results of the microarray for the 3 genes of interest (Figure 1a). For quantitation, the signal intensity of each of the PCR products was measured and presented relative to GAPDH (internal control) (Figure 1b). The mean and standard deviation for mRNA levels in Mthfr+/– relative to Mthfr+/+ mice for ApoA-I were 0.71±0.07 in 4 pairs of mice (P<0.01); for ApoA-IV, 0.61±0.2 in 4 pairs of mice (P=0.06); and for Cyp7A1, 1.5±0.1 in 3 pairs of mice (P<0.01). Results for ApoA-I were also examined by q-RT-PCR for 2 pairs of liver RNA from Mthfr+/+ and Mthfr+/– mice; these findings were confirmatory (Mthfr+/– values were 45±20% of those in Mthfr+/+; data not shown).
Effects of Mthfr Deficiency on ApoA-I Protein Levels in Mouse Liver and Plasma
To determine whether differentially expressed levels of mRNA for ApoA-I translated into differential protein expression, Western blot analysis was performed on liver extracts from Mthfr+/+ and Mthfr+/– mice. Figure 2a shows reduced levels of the 27-kDa ApoA-I protein in Mthfr-deficient mice compared with wild-type mice. To quantify levels of ApoA-I, signal intensity was divided by the signal obtained from -actin (internal control). Levels of ApoA-I protein and SD in liver of Mthfr+/– mice were 0.52±0.12 of the value in Mthfr+/+ mice (4 pairs, P<0.01). Western blot analysis was also performed on plasma from Mthfr+/+ and Mthfr+/– mice using albumin as internal control (Figure 2b); this analysis revealed that ApoA-I protein levels in plasma of Mthfr+/– mice were 0.62±0.19 of the levels in Mthfr+/+ mice (7 pairs, P<0.01). To ensure that the altered expression of ApoA-I related to the hyperhomocysteinemia rather than the disturbance in folate metabolism, we also performed Western blotting on plasma from mice with a heterozygous knockout of the cystathionine--synthase gene (Cbs+/–) (Figure 2c). These mice have hyperhomocysteinemia because of a defect in homocysteine transsulfuration instead of folate-dependent homocysteine remethylation. ApoA-I levels in Cbs+/– mice were 0.71±0.16 of the levels in Cbs+/+ mice (2 pairs, 3 experiments, P<0.01).
Correlation of Plasma Homocysteine With Levels of ApoA-I Protein and HDL Cholesterol in Males with Coronary Artery Disease
To determine whether our observation of reduced ApoA-I levels in hyperhomocysteinemic Mthfr-deficient and Cbs-deficient mice could be extrapolated to humans, we assessed ApoA-I protein levels in 60 plasma samples from a Montreal cohort of men with coronary artery disease, which we had previously examined for homocysteine and relevant nutrients.28,29 The mean tHcy value for this study was 10.5 μmol/L. We selected males with homocysteine levels above and below this mean for Western analysis, using albumin as an internal control (Figure 3). We observed lower levels of the ApoA-I protein in individuals with Hcy levels >10.5 μmol/L (mean=0.44, n=29) compared with individuals with Hcy levels <10.5 μmol/L (mean=0.55, n=31), P<0.05. Furthermore, there was a negative correlation between levels of ApoA-I and plasma homocysteine with a significant r value of –0.33 (P<0.01) (Figure 4a). We also observed a negative correlation between HDL cholesterol and plasma homocysteine in this group (r=–0.24); these findings had borderline significance (P=0.06) (Figure 4b).
Effect of Homocysteine on ApoA-I and PPAR Levels
To determine whether Hcy regulated ApoA-I through an effect on PPAR, HepG2 cells were treated with 0, 1, 5, and 10 mmol/L Hcy. A Hcy dose-dependent decrease in levels of both ApoA-I and PPAR was observed at 8 hours (Figure 5a); this decrease was evident as early as 4 hours and was sustained over 24 hours at 5 and 10 mmol/L Hcy (data not shown). Quantitation of ApoA-I and PPAR protein at 8 hours for 2 experiments showed a mean decrease of 10%, 40%, and 70% for ApoA-I and 30%, 40%, and 65% for PPAR, with increasing Hcy concentrations. These results are consistent with inhibition of PPAR by Hcy. Because PPAR is a regulator of ApoA-I, we suggest that the decrease in ApoA-I levels by Hcy is caused by PPAR downregulation.
We also compared effects of homocysteine and cysteine on ApoA-I expression in HepG2 cells. Western blotting (Figure 5b) revealed that ApoA-I expression was reduced by 47% with 5 mmol/L Hcy (a value similar to that in Figure 5a), whereas the same concentration of cysteine and the lower 1 mmol/L concentrations of both homocysteine and cysteine had little effect (14% or less of a decrease). These results suggest that the ApoA-I decrease is not attributable to a nonspecific effect of thiols.
Effect of Homocysteine on ApoA-I Promoter Activity
To determine whether the downregulation of ApoA-I by Hcy is caused by decreased activity of the ApoA-I promoter and if this decrease is mediated through PPAR-response element (PPRE), we constructed human apoA-I promoter-luciferase plasmids carrying wild-type PPRE (ApoAI-wt) or mutant PPRE (ApoAI-mut) (Figure 6a). HepG2 cells were transiently transfected with the constructs and after 6 hours, treated with Hcy (0, 1, or 5 mmol/L). The relative promoter activity with 0 homocysteine in cells transfected with ApoAI-mut was significantly lower than that in cells transfected with ApoAI-wt (43%, P<0.01, Figure 6b), indicating that the PPRE is a strong site driving the basal expression of the ApoA-I promoter. Similar results have been reported with an ApoA-II promoter construct in which a mutation in the PPRE (J-site) resulted in a strong decrease in baseline chloramphenicol acetyltransferase activity as well as a loss of inducibility by PPAR.31 When cells were transfected with ApoAI-wt plasmid, treatment with 1 and 5 mmol/L Hcy significantly decreased promoter activity to 89% (P<0.05) and 33% (P<0.01), respectively. With 5 mmol/L Hcy treatment, cells transfected with ApoAI-mut plasmid had decreased promoter activity (11%) compared with untreated cells (43%, P<0.01). Our results indicate that Hcy reduces ApoA-I promoter activity. However, because the mutant construct is also inhibited by Hcy, Hcy may be acting through a PPRE-independent mechanism.
Effect of PPAR on ApoA-I and MTHFR Expression
To confirm that PPAR affects ApoA-I synthesis in HepG2 cells, we cloned the human cDNA for PPAR into pCMV-Myc vector and performed transient transfection assays. In addition, we speculated that PPAR might also regulate MTHFR because sequences in the upstream region of the Mthfr promoter, identified through analyses using Visualization Tools for Alignment (VISTA software), contained a stretch of 5 nucleotides (GCACC) recognized as putative PPRE. This sequence was conserved and aligned between the mouse and human Mthfr promoters.
Western blotting (Figure 7a) showed that HepG2 cells transfected with PPAR had a 1.9-fold increase in ApoA-I protein compared with cells transfected with empty vector; total MTHFR protein also increased 1.6-fold (mean of 5 experiments, P<0.01). MTHFR has 2 protein isoforms, which differ in the length of their coding sequence and are generated by alternative splicing in the 5' region.32 The 2 bands for MTHFR in Figure 7a consist of the shorter MTHFR isoform, which is posttranslationally modified by phosphorylation.33 To confirm the identity of the MTHFR isoforms in HepG2 cells, a protein extract from bacteria expressing only the shorter human MTHFR32 isoform was run (lane 1 in Figure 7b) and this band corresponded with the MTHFR isoform with increased mobility (nonphosphorylated) in HepG2 cells (lane 2 in Figure 7b). The increase in total MTHFR protein was observed for both the phosphorylated and nonphosphorylated forms. The longer MTHFR isoform was not observed in these cells. q-RT-PCR analysis confirmed the aforementioned increases, with an 2-fold increase in mRNA levels for ApoA-I and a more modest 1.3-fold increase in MTHFR mRNA (data not shown).
Discussion
Although tHcy has become recognized as a risk factor for vascular disease, the biological mechanisms have not been completely elucidated. Several mechanisms have been proposed by which HHcy can contribute to endothelial dysfunction and atherosclerosis; these include oxidative stress, induction of proinflammatory factors, and endoplasmic reticulum stress.34 In this study, we identified a novel mechanism: reduced expression of the atheroprotective lipoprotein ApoA-I, the major protein component of HDL. We observed decreased mRNA for ApoA-I in liver and decreased ApoA-I protein in liver and plasma of Mthfr-deficient mice. These mice represent an excellent animal model for mild HHcy in humans. The mice have a 1.5- to 2-fold elevation in plasma homocysteine, lipid deposition in the valvular region of the aorta, and endothelial dysfunction; these findings are similar to those in humans with hyperhomocysteinemia with or without MTHFR deficiency. Other genetically engineered animal models have been used to study hyperhomocysteinemia, including mice with a complete knockout of cystathionine -synthase (CBS–/–)35 or mice with knockouts in CBS and ApoE (CBS–/– and ApoE–/–).36 Other models of hyperhomocysteinemia have been generated by homocysteine infusions or by high dietary intake of homocysteine or methionine.37 Although these other models have been useful, the Mthfr+/– mice are a particularly attractive model because of their modest increase in homocysteine and the high natural occurrence of MTHFR deficiency in humans.
Our finding of reduced ApoA-I expression in CBS-deficient mice, as well as in MTHFR-deficient mice, suggests that it is the elevation in homocysteine, rather than the disturbance in folate metabolism, that regulates this apolipoprotein. Consistent with this argument are the effects of homocysteine on the ApoA-I promoter and in HepG2 cells, which were conducted in folate-replete medium.
The increase in mRNA for Cyp7A1 in liver of hyperhomocysteinemic mice is also consistent with enhanced atherogenesis. Cyp7A1 catalyzes 7 hydroxylation of cholesterol, the rate-limiting step in bile acid synthesis. Bile acids are transported from the liver to the intestine, where they facilitate cholesterol absorption by promoting the formation of micellar cholesterol.
Two key proteins involved in reverse cholesterol efflux, ApoA-I and Cyp7A1, were altered in expression in liver of Mthfr-deficient mice. Both proteins are regulated at the transcriptional level by the transcription factor PPAR, which has been shown to increase ApoA-I expression30 and decrease Cyp7A1 expression.21 Our findings in HepG2 cells confirm the upregulation of human ApoA-I by PPAR, because transfection with a PPAR construct increased ApoA-I levels and a mutated PPRE markedly reduced ApoA-I promoter activity. Homocysteine reduced the levels of PPAR and ApoA-I protein in HepG2 cells, suggesting that PPAR may be involved in homocysteine regulation of ApoA-I. Homocysteine also reduced ApoAI promoter activity; however, a PPRE-independent mechanism could be involved because the ApoAI promoter with a mutagenized PPRE was still subject to homocysteine inhibition. Although the homocysteine levels in the medium were higher than the pathophysiological range of plasma tHcy, these concentrations are often used38,39 in cultured cell systems to achieve a 2- to 6-fold increase in intracellular homocysteine.38
The regulation of MTHFR by PPAR is a novel finding that provides an additional interaction between homocysteine and lipid metabolism. A recent study has demonstrated that MTHFR activity can be regulated by phosphorylation; the phosphorylated enzyme is suggested to be less active than the nonphosphorylated enzyme.33 Our experiments showed increased expression of the phosphorylated and nonphosphorylated forms of MTHFR following transfection of HepG2 cells with PPAR. These findings add more complexity to the regulation of MTHFR, which has 2 promoters, alternative splicing and extra coding sequences for the higher molecular weight isoform. Additional work is required to delineate the functional significance of this PPAR-dependent increase in MTHFR expression.
Our finding of homocysteine modulation of PPAR has also been suggested in studies of homocysteine-induced endothelial cell dysfunction. Homocysteine may antagonize the effects of PPAR and PPAR in endothelial cell remodeling.40 Interestingly, treatment of CBS-deficient mice with fibrates, which are known to induce PPAR, have been reported to reverse the Hcy-mediated endothelial dysfunction in these mice.24
The altered expression of ApoA-I in hyperhomocysteinemic mice was confirmed in a human cohort, a well-characterized group of males with coronary artery disease. In previous studies of this cohort, we showed that the MTHFR mutant genotype (677TT) was associated with increased plasma homocysteine only when folate status was low. Therefore, the association of homocysteine with reduced ApoA-I would only be expected in humans with the 677TT genotype when their folate status was low. We did not observe an association between the 677TT genotype and reduced ApoA-I levels in this study, because some of the males with this genotype had adequate folate and were therefore not hyperhomocysteinemic. Sampling of this population was performed before the introduction of folate fortification of food in Canada. It would be interesting to examine the association among folate, homocysteine, and apolipoproteins in North Americans after fortification.
In summary, we have demonstrated a significant negative relationship between levels of Hcy and ApoA-I in hyperhomocysteinemic males and in hyperhomocysteinemic Mthfr-deficient mice. Studies of large clinical populations are required to determine whether combined evaluation of both biomarkers is a better predictor of disease risk than either alone. In addition, it remains to be determined whether folate supplementation, which is known to reduce homocysteine levels, also modifies lipoprotein profiles.
Acknowledgments
This work was supported by the Canadian Institutes of Health Research (CIHR). R.R. was supported by a Senior Investigator Award from CIHR, and L.M. was the recipient of a Fellowship from the Montreal Children’s Hospital Research Institute. We gratefully acknowledge the assistance of Qing Wu for animal care, Sébastien Dubé for statistical analysis, and Andrea Lawrance for advice.
Footnotes
Original received August 26, 2005; resubmission received October 24, 2005; revised resubmission received December 28, 2005; accepted January 12, 2006.
References
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. [Order article via Infotrieve]
Refsum U, Ueland PM, Nygrd O, Vollset SE. Homocysteine and vascular disease. Ann Rev Med. 1998; 49: 31–62. [Order article via Infotrieve]
Frosst P, Blom H, Milos R, Goyette P, Sheppard C, Matthews R, Boers G, den Heijer M, Kluijtmans L, van den Heuvel L, Rozen R. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet. 1995; 10: 111–113. [Order article via Infotrieve]
Chen Z, Karaplis A, Ackerman S, Pogribny I, Melnyk S, Lussier-Cacan S, Chen MF, Pai A, John S, Smith R, Bottiglieri T, Bagley P, Selhub J, Rudnicki M, James SJ, Rozen R. Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet. 2001; 10: 433–443.
Ghandour H, Chen Z, Selhub J, Rozen R. Mice deficient in methylenetetrahydrofolate reductase exhibit tissue-specific distribution of folates. J Nutr. 2004; 134: 2975–2978.
Virdis A, Iglarz M, Neves M, Amiri F, Touyz R, Rozen R, Schiffrin E. Effect of hyperhomocystinemia and hypertension on endothelial function in methylenetetrahydrofolate reductase-deficient mice. Arterioscler Thromb Vasc Biol. 2003; 23: 1352–1357.
Neves M, Endermann D, Amiri F, Virdis A, Pu Q, Rozen R, Schiffrin E. Small artery mechanics in hyperhomocysteinemic mice: effects of angiotensin II. J Hypertens. 2004; 959–966.
Gordon T, Castelli W, Hjortland M, Kannel W, Dawber T. High density lipoprotein as a protective factor against coronary heart disease: the Framingham study. Am J Med. 1977; 62: 707–714. [Order article via Infotrieve]
Rubin E, Krauss R, Spangler E, Verstyuft J, Clift S. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein AI. Nature. 1991; 353: 265–267. [Order article via Infotrieve]
Assmann G, Nofer J. Atheroprotective effects of high-density lipoproteins. Annu Rev Med. 2003; 54: 321–341. [Order article via Infotrieve]
Suc I, Escargueil-Blanc I, Troly M, Salvayre R, Negre-Salvayre A. HDL and apoA prevent cell death of endothelial cells induced by oxidized LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 2158–2166.
Cockerill G, Rye K, Gamble J, Vadas M, Barter P. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol. 1995; 15: 1987–1994.
Ho Y, Brown M, Goldstein J. Hydrolysis and excretion of cytoplasmic cholesteryl esters by macrophages: stimulation by high density lipoprotein and other agents. J Lipid Res. 1980; 21: 391–398.
Groenendijk M, Cantor R, de Bruin T, Dallinga-Thie G. The apoAI-CIII-AIV gene cluster. Atherosclerosis. 2001; 157: 1–11. [Order article via Infotrieve]
Pennachio L, Olivier M, Hubacek J, Cohen J, Cox D, Fruchart J-C, Krauss R, Rubin E. An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science. 2001; 194: 169–172.
Duval C, Fruchart JC, Staels B. PPARa, fibrates, lipid metabolism and inflammation. Arch Mal Coeur Vaiss. 2004; 97: 665–672. [Order article via Infotrieve]
Castrillo A, Tontonoz P. PPARs in atherosclerosis: the clot thickens. J Clin Invest. 2004; 114: 1538–1540.
Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E, Martin G, Fruchart J-C, Laudet V, Staels B. The nuclear receptors peroxisome proliferator-activated receptor-a and Reverba mediate the species specific regulation of apolipoprotein A-I expression by fibrates. J Biol Chem. 1998; 273: 25713–25720.
Fruchart J-C, Staels B, Duriez P. PPARs, metabolic disease and atherosclerosis. Pharmacol Res. 2001; 44: 345–352. [Order article via Infotrieve]
Prieur X, Coste H, Rodríguez J. The human apolipoprotein AV gene is regulated by peroxisome proliferator-activated receptor a and contains a novel farnesoid X-activated receptor response element. J Biol Chem. 2003; 278: 25468–25480.
Marrapodi M, Chiang Y. Peroxisome proliferators-activated receptor a (PPARa) and agonist inhibit cholesterol 7a-hydroxylase gene (CYP7A1) transcription. Lipid Res. 2000; 41: 514–520.
Agellon L. Metabolism and function of bile acids. In: Vance DE, JE V, eds. Biochemistry of Lipids, Lipoproteins and Membranes. Amsterdam: Elsevier; 2002, 433–447.
Brude IR, Finstad HS, Seljeflot I, Drevon CA, Solvoll K, Sandstad B, Hjermann I, Arnesen H, Nenseter MS. Plasma homocysteine concentration related to diet, endothelial function and mononuclear cell gene expression among male hyperlipidaemic smokers. Eur J Clin Invest. 1999; 29: 100–108. [Order article via Infotrieve]
Sood H, Hunt M, Tyagi S. Peroxisome proliferator ameliorates endothelial dysfunction in a murine model of hyperhomocysteinemia. Am J Physiol Lung Cell Mol Phsyiol. 2003; 284: L333–L341.
Schwahn B, Wendel U, Lussier-Cacan S, Mar M, Zeisel S, Leclerc D, Castro C, Garrow T, Rozen R. Effects of betaine in a murine model of mild cystathionine-beta-synthase deficiency. Metabolism. 2004; 53: 594–599. [Order article via Infotrieve]
Chen Z, Ge B, Hudson T, Rozen R. Microarray analysis of brain RNA in mice with methylenetetrahydrofolate reductase deficiency and hyperhomocysteinemia. Brain Res Gene Expr Patterns. 2002; 1: 89–93. [Order article via Infotrieve]
Leclerc D, Deng L, Trasler J, Rozen R. ApcMin/+ mouse model of colon cancer: gene expression profiling in tumors. J Cell Biochem. 2001; 93: 1242–1254.
Christensen B, Frosst P, Lussier-Cacan S, Selhub J, Goyette P, Rosenblatt D, Genest JJ, Rozen R. Correlation of a common mutation in the methylenetetrahydrofolate reductase gene with plasma homocysteine in patients with premature coronary artery disease. Arterioscler Thromb Vasc Biol. 1997; 17: 569–573.
Schwahn B, Chen Z, Laryea M, Wendel U, Lussier-Cacan S, Genest JJ, Mar M-H, Zeisel S, Castro C, Garrow T, Rozen R. Homocysteine-betaine interactions in a murine model of 5,10-methylenetetrahydrofolate reductase deficiency. FASEB J. 2003; 17: 512–514.
Morishima A, Ohkubo N, Maeda N, Miki T, Mitsuda N. NFkB regulates plasma apolipoprotein A-I and high density lipoprotein cholesterol through inhibition of peroxisome proliferator-activated receptor a. J Biol Chem. 2003; 278: 38188–38193.
Vu-Dac N, Schoonjans K, Kosykh V, Dallongeville J, Fruchart J-C, Staels B. Fibrates increase human Apolipoprotein A-II expression through activation of the peroxisome proliferator-activated receptor. J Clin Inv. 1995; 96: 741–750. [Order article via Infotrieve]
Tran P, Leclerc D, Chan M, Pai A, Hiou-Tim F, Wu Q, Goyette P, Artigas C, Milos R, Rozen R. Multiple transcription start sites and alternative splicing in the methylenetetrahydrofolate reductase gene result in two enzyme isoforms. Mamm Genome. 2002; 13: 483–492. [Order article via Infotrieve]
Yamada K, Strahler J, Andrews P, Matthews R. Regulation of human methylenetetrahydrofolate reductase by phosphorylation. Proc Natl Acad Sci U S A. 2005; 102: 10454–10459.
de Koning A, Werstuck G, Zhou J, Austin R. Hyperhomocysteinemia and its role in the development of atherosclerosis. Clin Biochem. 2003; 2003: 431–441.
Namekata K, Enokido Y, Ishii I, Nagai Y, Harada T, Kimura H. Abnormal lipid metabolism in cystathionine beta-synthase-deficient mice, an animal model for hyperhomocysteinemia. J Biol Chem. 2004; 279: 52961–52969.
Wang H, Jiang X, Gaubatz J, Ma L, Magera M, Yang X, Berger P, Durante W, Pownall H, Schafer A. Hyperhomocysteinemia accelerates atherosclerosis in cystathionine beta-synthase and apolipoprotein E double knock-out mice with and without dietary perturbation. Blood. 2003; 101: 3901–3907.
Troen A, Lutgens E, Smith D, Rosenberg I, Selhub J. The atherogenic effect of excess methionine intake. Proc Natl Acad Sci U S A. 2003; 100: 15089–15094.
Werstuck G, Lentz S, Dayal S, Hossain GS, Sood S, Shi Y, Zhou J, Maeda N, Krisans S, Malinow M, Austin R. Homocysteine-induced endoplasmic reticulum stress causes dysregulation of the cholesterol and triglyceride biosynthetic pathways. J Clin Invest. 2001; 107: 1263–1273.
Kokame K, Agarwala K, Kato H, Miyata Y. Herp, a new ubiquitin-like membrane protein induced by endoplasmic reticulum stress. J Biol Chem. 2000; 275: 32846–32853.
Mujumdar V, Tummalapalli C, Aru G, Tyagi S. Mechanism of constrictive vascular remodeling by homocysteine: role of PPAR. Am J Physiol Cell Physiol. 2002; 282: C1009–C1015.
the Departments of Human Genetics and Pediatrics (L.G.M., R.R.), Montreal Children’s Hospital
The Department of Medicine (J.G.), Royal Victoria Hospital, McGill University Health Center, Montreal, Canada.
Abstract
Hyperhomocysteinemia, a risk factor for cardiovascular disease, is caused by nutritional or genetic disturbances in homocysteine metabolism. A polymorphism in methylenetetrahydrofolate reductase (MTHFR) is the most common genetic cause of mild hyperhomocysteinemia. To examine mechanisms by which an elevation in plasma homocysteine leads to vascular disease, we first performed microarray analyses in livers of Mthfr-deficient mice and identified differentially expressed genes that are involved in lipid and cholesterol metabolism. Microarrays and RT-PCR showed decreased mRNA for apolipoprotein A (ApoA)-IV and for ApoA-I and increased mRNA for cholesterol 7 hydroxylase (Cyp7A1) in Mthfr+/– mice compared with Mthfr+/+ mice. Western blotting revealed that ApoA-I protein levels in liver and plasma of Mthfr+/– mice were 52% and 62% of levels in the respective tissues of Mthfr+/+ mice. We also performed Western analysis for plasma ApoA-I protein levels in 60 males with coronary artery disease and identified a significant (P<0.01) negative correlation (–0.33) between ApoA-I and plasma homocysteine levels. This cohort also displayed a negative correlation (–0.24, P=0.06) between high-density lipoprotein cholesterol and plasma homocysteine. Treatment of HepG2 cells with supraphysiological levels of 5 mmol/L homocysteine reduced peroxisome proliferator-activated receptor (PPAR) and ApoA-I protein levels and decreased ApoA-I promoter activity. Transfection with a PPAR construct upregulated ApoA-I and MTHFR. Our results suggest that hyperhomocysteinemia may increase risk of atherosclerosis by decreasing expression of ApoA-I and increasing expression of CYP7A1.
Key Words: homocysteine methylenetetrahydrofolate reductase apolipoprotein A-I high-density lipoprotein peroxisome proliferator-activated receptor
Introduction
Conventional risk factors for cardiovascular disease including hypercholesterolemia, hypertension, smoking, and diabetes account for approximately 50% of all cases.1 Evidence now indicates that hyperhomocysteinemia (HHcy), which occurs in 5% to 7% of the general population, is an important risk factor for atherosclerotic and thrombotic disease.2 The enzyme 5,10-methylenetetrahydrofolate reductase (MTHFR) plays a critical role in regulating total homocysteine (tHcy) levels by synthesizing 5-methyltetrahydrofolate, which provides the methyl group for homocysteine remethylation to methionine. Mild deficiency of MTHFR, the most common genetic cause of mild HHcy, occurs frequently in many populations (10% to 15% of individuals) and is attributable to homozygosity for a polymorphism (677CT) that reduces enzyme activity.3 To generate an animal model for HHcy, we had created mice with a heterozygous (+/–) disruption of the Mthfr gene.4 With 50% residual activity and 1.7-fold greater levels of plasma tHcy compared with their wild-type (+/+) littermates (5.4±1.0 versus 3.2±1.2, mean±SD),4 the Mthfr+/– mice are similar biochemically to humans with mild MTHFR deficiency. Plasma total folate levels are somewhat lower but not significantly reduced in Mthfr+/– mice (192±19.7 versus 214.7±12).5 In addition, these mice have lipid deposition in the aortic valvular region,4 decreased endothelial-dependent vasodilatation in mesenteric vessels,6 and stiffer small arteries with increased collagen deposition.7
The relationship between coronary artery disease and low levels of plasma high-density lipoprotein (HDL) cholesterol has been demonstrated in several prospective studies.8 Apolipoprotein A (ApoA)-I is the major protein component of HDL. Transgenic mice that overexpress human ApoA-I have increased HDL levels and reduced susceptibility to diet-induced atherosclerosis.9 Although it is not completely understood, the protective effect of ApoA-I may be related to its association with increased HDL production. HDL can be protective10 by such mechanisms as prevention of lipoprotein oxidation,11 decreased expression of adhesion molecules on vascular cells,12 and increased cholesterol efflux from cells such as macrophages that are involved in the initial development of the atherosclerotic plaque.13
A region on the long arm of chromosome 11q23-q24 encodes genes for 3 lipoproteins (apoA-I, apoC-III, and apoA-IV), which are clustered within a 17-kb segment.14 Comparative sequence analysis has identified a relatively newer apolipoprotein, apoA-V, located approximately 30-kb proximal to the apoA-I/C-III/A-IV gene cluster.15
The peroxisome proliferator-activated receptors (PPARs) are ligand-activated transcription factors that belong to the superfamily of nuclear hormone receptors; 4 isotypes are known: PPAR, PPAR, PPAR, and PPAR. PPAR, which is predominantly expressed in the liver, is involved in lipid metabolism and stimulates fatty acid oxidation and ketogenesis.16 The relevance of PPAR pathways to metabolic disease is underscored by the use of fibrates (PPAR agonists) to treat hyperlipidemia.17 PPAR decreases triglyceride concentrations by increasing the expression of lipoprotein lipase, and increases HDL cholesterol levels by regulating the production of ApoA-I,18 ApoA-II,19 and ApoA-V.20 It also controls reverse cholesterol transport by increasing cholesterol efflux from macrophages and inhibits transcription of the gene encoding cholesterol 7 hydroxylase (CYP7A1).21 Cyp7A1 is a liver-specific enzyme that catalyzes the 7 hydroxylation of cholesterol, the rate-limiting step in the conversion of cholesterol into bile acids.22 A negative correlation between high Hcy levels and PPAR expression has been suggested in humans.23 In a murine model of HHcy, PPAR was shown to ameliorate endothelial dysfunction.24
In this study, we used our Mthfr-deficient mouse model to study the potential interaction between homocysteine and lipid metabolism. Initial microarray findings of decreased apolipoprotein expression in liver of hyperhomocysteinemic mice were pursued by additional studies in murine tissues, in a human cohort of coronary artery disease, and in HepG2 cells. Our findings suggest that HHcy may contribute to cardiovascular disease by decreased expression of ApoA-I and increased expression of CYP7A1, mediated by PPAR.
Materials and Methods
Animal Model of Hyperhomocysteinemia
Mthfr-deficient mice (Mthfr+/–) and wild-type (Mthfr+/+) littermates were generated as reported.4 Mice were progeny from matings of Mthfr+/– mice backcrossed for at least 12 generations from 129/Sv-BALB/c F1 heterozygotes to BALB/cAnNCrlBR (Charles River, Saint-Constant, Quebec, Canada). Experimentation was approved by the Animal Care Committee of the Montreal Children’s Hospital. Mice were fed standard mouse chow (Harlan Teklad LM-485 Mouse, code 7012). Blood was collected from the tail vein of 8-month-old mice, after overnight fast, in EDTA-vacutainer tubes. Plasma was obtained by centrifugation (6000 rpm, 5 minutes, 4°C) and stored at –70°C until analysis. Mice were euthanized a few days later and livers were snap frozen in liquid nitrogen and stored at –70°C. Mthfr genotypes were determined as reported.4 Mice with a heterozygous disruption of the cystathionine--synthase (Cbs) gene were purchased from The Jackson Laboratory (Bar Harbor, Maine) and bred with C57Bl/6J mice to obtain heterozygous and wild-type mice. Genotypes were determined as reported.25
RNA Isolation, Microarray Analysis, and RT-PCR
Frozen mouse liver was pulverized in liquid nitrogen, and total RNA was isolated using TRIzol reagent (Invitrogen). Liver RNA from 3 pairs of Mthfr+/– mice and wild-type Mthfr+/+ littermates was analyzed by Affymetrix GeneChip Expression Arrays at the McGill University/Genome Québec Innovation Center (Montreal, Canada) using the murine chip MG-U74Av2; mice were matched for age (8 months) and gender (2 male and 1 female pair). Affymetrix software was used to assess genes with differential expression between wild-type and Mthfr-deficient mice, as described previously.26,27 For RT-PCR experiments, RNA from mouse liver was also DNase treated and reverse transcribed with random hexamers and Superscript II as recommended by the manufacturer (Invitrogen). The cDNA was used in RT-PCRs with primers for mouse ApoA-I, ApoA-IV, Cyp7A1, and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH). Quantitation of RT-PCR products was performed by measuring the intensity of each band using Quantity One 4.1.0 software and SPSS software for statistical analysis. Intensity of signal from the relevant genes was divided by the intensity of the signal from GAPDH, which served as control. To confirm RT-PCR results, quantitative real-time RT-PCR (q-RT-PCR) was performed using Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen) in a MX4000 QPCR System (Stratagene) following the recommendations of the manufacturers.
Preparation of Crude Liver Extracts
Frozen mouse liver was pulverized in liquid nitrogen and suspended in ice-cold radioimmunoprecipitation assay lysis buffer (1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS [sodium dodecyl sulfate] in PBS) containing a protease inhibitor mixture (Complete Mini, Roche). After incubation on ice for 30 minutes, lysates were centrifuged (13000 rpm, 15 minutes, 4°C). Supernatants were collected and protein concentrations assessed.
SDS-PAGE and Western Blotting
Electrophoresis (50 μg of protein from liver extracts, 25 μg of protein from cultured cells, 1 μL of mouse or human plasma) was performed on SDS–polyacrylamide gels. Proteins were transferred to nitrocellulose membranes at 67 V for 2 hours at 4°C. Nonspecific binding sites were blocked with 5% skim milk powder in Tris-buffered saline/Tween buffer overnight at 4°C. Primary antibodies were: rabbit anti-mouse ApoA-I (Biodesign International), anti-human ApoA-I (Calbiochem), anti–-actin (Sigma), anti-albumin (Bethyl Laboratories), anti-PPAR (Santa Cruz Biotechnology), and anti-Mthfr (described by Frosst et al3). Secondary antibody was a peroxidase-coupled anti-rabbit IgG (Amersham Biosciences). Signal detection was achieved with ECL Plus chemiluminescence system (Amersham Biosciences) and exposure to x-ray films. Signals were quantified with Quantity One 4.1.0 software. SPSS software was used for statistical analysis.
Clinical Studies
The patient population consisted of a cohort of coronary artery disease patients from Montreal that had previously been characterized for homocysteine levels and other relevant parameters before treatment was initiated.28,29 Mean plasma tHcy concentration for the cohort is 10.8 μmol/L.29 We selected 60 plasma samples from males, aged 31 to 59 years, with a range of Hcy levels having a mean of 10.5 μmol/L and an equal distribution above and below this mean, and examined immunoreactive ApoA-I protein by Western blotting.
Plasmid Constructs for Human ApoA-I Promoter
The human ApoA-I promoter fragment was amplified from HepG2 DNA, between positions –330 and +69 relative to the transcription start site, using primers 5'-TATAGCTAGCAACACAATGGACAATGGCAACTG-3' and 5'-TATAAAGCTTGA ACCTTGAGCTGGGGAGC-3' that incorporated NheI and HindIII restriction sites, as previously reported.30 The amplified fragment, which contained the peroxisome-proliferator receptor element (PPRE), was digested with NheI and HindIII and inserted between the respective sites in pGL3-basic vector to create an ApoAI-wt plasmid. Mutations in the PPRE were generated by PCR using primers 5'-ACTGATCCCTTGTCCCCTGCCCTGCAGCCCCCGCA-3' and 5'-AGGGGACAAGG GATCAGTGGGGGCGGGAGGGGAGT-3' (mutations underlined) to create the ApoAI-mut plasmid.30 Integrity of the inserts was verified by sequencing.
Plasmid Construct for Human PPAR
The human PPAR sequence was amplified from HepG2 cDNA and cloned into expression vector pCMV-Myc (BD Biosciences). After transformation into INVF’ Competent Escherichia coli cells (Invitrogen), potential clones carrying the human PPAR were screened and inserts from the positive clones were verified by sequencing.
HepG2 Cell Culture and Transient Transfection Assays
Human HepG2 hepatoma cells (American Type Culture Collection) were propagated at 37°C, in 5% CO2, in DMEM-F12 with 10% FBS, 100 μg/mL penicillin and streptomycin. For transfections, cells were seeded at approximately 5x105 into 6-well plates and medium was changed regularly. At 90% confluence, cells were transfected with either 2 μg of vector containing PPAR or pCMV vector alone using Lipofectamine 2000 Reagent in OPTI-MEM medium (Invitrogen) following the recommendations of the manufacturer. Cells were harvested 72 hours after transfection. RNA isolation, cDNA synthesis, RT-PCR, and q-RT-PCR were performed as above. Increased expression of PPAR was confirmed by RT-PCR in transfected cells. For Western blotting, cells were washed with ice-cold PBS and treated as described above. For transfection with ApoAI promoter-reporter plasmids, HepG2 cells were cotransfected with pCMV--galactosidase. Firefly luciferase activity was detected by a Luciferase Assay Kit (Promega); transfection efficiencies were controlled by normalizing to -galactosidase activity (Applied Biosystems). Transfections were performed at least 4 times.
Homocysteine Treatment of HepG2 Cells
At 80% confluence, the medium was replaced with DMEM-F12 supplemented with 0, 1, 5, or 10 mmol/L DL-homocysteine or L-cysteine (Sigma). Cells were collected at specific time points for up to 24 hours, washed with ice-cold PBS and treated for Western blotting. In promoter experiments, Hcy was added 6 hours after transfection by replacing OPTI-MEM with 0, 1, or 5 mmol/L Hcy in DMEM-F12; cells were harvested 24 hours after transfection.
Results
Effects of Mthfr Deficiency on ApoA-IV, ApoA-I, and Cyp7A1 mRNA Levels in Mouse Liver
The Affymetrix software algorithm was used to analyze microarray results and examine differentially expressed genes in 3 pairs of livers from Mthfr+/– mice compared with Mthfr+/+ littermates. We chose genes with a greater than 1.8-fold change (increase or decrease) and a difference of 150 U of fluorescence in the raw data analysis (Table I in the online data supplement available at http://circres.ahajournals.org). We were particularly interested in genes that were implicated in lipid and cholesterol metabolism and honed in on 2 such genes, ApoA-IV and Cyp7A1. Mthfr+/– mice had reduced expression of ApoA-IV compared with Mthfr+/+ mice in all 3 pairs, with an average fold change of –2.6. Compared with their Mthfr+/+ littermates, Mthfr+/– mice showed increased expression of Cyp7A1 in all 3 pairs, with an average fold change of +2.1. Microarray analysis also revealed a decrease in ApoA-I mRNA levels in 2 pairs of Mthfr+/– mice compared with their wild-type littermates; the fold change, however, did not reach statistical significance. RT-PCR was performed to validate results of the microarray for the 3 genes of interest (Figure 1a). For quantitation, the signal intensity of each of the PCR products was measured and presented relative to GAPDH (internal control) (Figure 1b). The mean and standard deviation for mRNA levels in Mthfr+/– relative to Mthfr+/+ mice for ApoA-I were 0.71±0.07 in 4 pairs of mice (P<0.01); for ApoA-IV, 0.61±0.2 in 4 pairs of mice (P=0.06); and for Cyp7A1, 1.5±0.1 in 3 pairs of mice (P<0.01). Results for ApoA-I were also examined by q-RT-PCR for 2 pairs of liver RNA from Mthfr+/+ and Mthfr+/– mice; these findings were confirmatory (Mthfr+/– values were 45±20% of those in Mthfr+/+; data not shown).
Effects of Mthfr Deficiency on ApoA-I Protein Levels in Mouse Liver and Plasma
To determine whether differentially expressed levels of mRNA for ApoA-I translated into differential protein expression, Western blot analysis was performed on liver extracts from Mthfr+/+ and Mthfr+/– mice. Figure 2a shows reduced levels of the 27-kDa ApoA-I protein in Mthfr-deficient mice compared with wild-type mice. To quantify levels of ApoA-I, signal intensity was divided by the signal obtained from -actin (internal control). Levels of ApoA-I protein and SD in liver of Mthfr+/– mice were 0.52±0.12 of the value in Mthfr+/+ mice (4 pairs, P<0.01). Western blot analysis was also performed on plasma from Mthfr+/+ and Mthfr+/– mice using albumin as internal control (Figure 2b); this analysis revealed that ApoA-I protein levels in plasma of Mthfr+/– mice were 0.62±0.19 of the levels in Mthfr+/+ mice (7 pairs, P<0.01). To ensure that the altered expression of ApoA-I related to the hyperhomocysteinemia rather than the disturbance in folate metabolism, we also performed Western blotting on plasma from mice with a heterozygous knockout of the cystathionine--synthase gene (Cbs+/–) (Figure 2c). These mice have hyperhomocysteinemia because of a defect in homocysteine transsulfuration instead of folate-dependent homocysteine remethylation. ApoA-I levels in Cbs+/– mice were 0.71±0.16 of the levels in Cbs+/+ mice (2 pairs, 3 experiments, P<0.01).
Correlation of Plasma Homocysteine With Levels of ApoA-I Protein and HDL Cholesterol in Males with Coronary Artery Disease
To determine whether our observation of reduced ApoA-I levels in hyperhomocysteinemic Mthfr-deficient and Cbs-deficient mice could be extrapolated to humans, we assessed ApoA-I protein levels in 60 plasma samples from a Montreal cohort of men with coronary artery disease, which we had previously examined for homocysteine and relevant nutrients.28,29 The mean tHcy value for this study was 10.5 μmol/L. We selected males with homocysteine levels above and below this mean for Western analysis, using albumin as an internal control (Figure 3). We observed lower levels of the ApoA-I protein in individuals with Hcy levels >10.5 μmol/L (mean=0.44, n=29) compared with individuals with Hcy levels <10.5 μmol/L (mean=0.55, n=31), P<0.05. Furthermore, there was a negative correlation between levels of ApoA-I and plasma homocysteine with a significant r value of –0.33 (P<0.01) (Figure 4a). We also observed a negative correlation between HDL cholesterol and plasma homocysteine in this group (r=–0.24); these findings had borderline significance (P=0.06) (Figure 4b).
Effect of Homocysteine on ApoA-I and PPAR Levels
To determine whether Hcy regulated ApoA-I through an effect on PPAR, HepG2 cells were treated with 0, 1, 5, and 10 mmol/L Hcy. A Hcy dose-dependent decrease in levels of both ApoA-I and PPAR was observed at 8 hours (Figure 5a); this decrease was evident as early as 4 hours and was sustained over 24 hours at 5 and 10 mmol/L Hcy (data not shown). Quantitation of ApoA-I and PPAR protein at 8 hours for 2 experiments showed a mean decrease of 10%, 40%, and 70% for ApoA-I and 30%, 40%, and 65% for PPAR, with increasing Hcy concentrations. These results are consistent with inhibition of PPAR by Hcy. Because PPAR is a regulator of ApoA-I, we suggest that the decrease in ApoA-I levels by Hcy is caused by PPAR downregulation.
We also compared effects of homocysteine and cysteine on ApoA-I expression in HepG2 cells. Western blotting (Figure 5b) revealed that ApoA-I expression was reduced by 47% with 5 mmol/L Hcy (a value similar to that in Figure 5a), whereas the same concentration of cysteine and the lower 1 mmol/L concentrations of both homocysteine and cysteine had little effect (14% or less of a decrease). These results suggest that the ApoA-I decrease is not attributable to a nonspecific effect of thiols.
Effect of Homocysteine on ApoA-I Promoter Activity
To determine whether the downregulation of ApoA-I by Hcy is caused by decreased activity of the ApoA-I promoter and if this decrease is mediated through PPAR-response element (PPRE), we constructed human apoA-I promoter-luciferase plasmids carrying wild-type PPRE (ApoAI-wt) or mutant PPRE (ApoAI-mut) (Figure 6a). HepG2 cells were transiently transfected with the constructs and after 6 hours, treated with Hcy (0, 1, or 5 mmol/L). The relative promoter activity with 0 homocysteine in cells transfected with ApoAI-mut was significantly lower than that in cells transfected with ApoAI-wt (43%, P<0.01, Figure 6b), indicating that the PPRE is a strong site driving the basal expression of the ApoA-I promoter. Similar results have been reported with an ApoA-II promoter construct in which a mutation in the PPRE (J-site) resulted in a strong decrease in baseline chloramphenicol acetyltransferase activity as well as a loss of inducibility by PPAR.31 When cells were transfected with ApoAI-wt plasmid, treatment with 1 and 5 mmol/L Hcy significantly decreased promoter activity to 89% (P<0.05) and 33% (P<0.01), respectively. With 5 mmol/L Hcy treatment, cells transfected with ApoAI-mut plasmid had decreased promoter activity (11%) compared with untreated cells (43%, P<0.01). Our results indicate that Hcy reduces ApoA-I promoter activity. However, because the mutant construct is also inhibited by Hcy, Hcy may be acting through a PPRE-independent mechanism.
Effect of PPAR on ApoA-I and MTHFR Expression
To confirm that PPAR affects ApoA-I synthesis in HepG2 cells, we cloned the human cDNA for PPAR into pCMV-Myc vector and performed transient transfection assays. In addition, we speculated that PPAR might also regulate MTHFR because sequences in the upstream region of the Mthfr promoter, identified through analyses using Visualization Tools for Alignment (VISTA software), contained a stretch of 5 nucleotides (GCACC) recognized as putative PPRE. This sequence was conserved and aligned between the mouse and human Mthfr promoters.
Western blotting (Figure 7a) showed that HepG2 cells transfected with PPAR had a 1.9-fold increase in ApoA-I protein compared with cells transfected with empty vector; total MTHFR protein also increased 1.6-fold (mean of 5 experiments, P<0.01). MTHFR has 2 protein isoforms, which differ in the length of their coding sequence and are generated by alternative splicing in the 5' region.32 The 2 bands for MTHFR in Figure 7a consist of the shorter MTHFR isoform, which is posttranslationally modified by phosphorylation.33 To confirm the identity of the MTHFR isoforms in HepG2 cells, a protein extract from bacteria expressing only the shorter human MTHFR32 isoform was run (lane 1 in Figure 7b) and this band corresponded with the MTHFR isoform with increased mobility (nonphosphorylated) in HepG2 cells (lane 2 in Figure 7b). The increase in total MTHFR protein was observed for both the phosphorylated and nonphosphorylated forms. The longer MTHFR isoform was not observed in these cells. q-RT-PCR analysis confirmed the aforementioned increases, with an 2-fold increase in mRNA levels for ApoA-I and a more modest 1.3-fold increase in MTHFR mRNA (data not shown).
Discussion
Although tHcy has become recognized as a risk factor for vascular disease, the biological mechanisms have not been completely elucidated. Several mechanisms have been proposed by which HHcy can contribute to endothelial dysfunction and atherosclerosis; these include oxidative stress, induction of proinflammatory factors, and endoplasmic reticulum stress.34 In this study, we identified a novel mechanism: reduced expression of the atheroprotective lipoprotein ApoA-I, the major protein component of HDL. We observed decreased mRNA for ApoA-I in liver and decreased ApoA-I protein in liver and plasma of Mthfr-deficient mice. These mice represent an excellent animal model for mild HHcy in humans. The mice have a 1.5- to 2-fold elevation in plasma homocysteine, lipid deposition in the valvular region of the aorta, and endothelial dysfunction; these findings are similar to those in humans with hyperhomocysteinemia with or without MTHFR deficiency. Other genetically engineered animal models have been used to study hyperhomocysteinemia, including mice with a complete knockout of cystathionine -synthase (CBS–/–)35 or mice with knockouts in CBS and ApoE (CBS–/– and ApoE–/–).36 Other models of hyperhomocysteinemia have been generated by homocysteine infusions or by high dietary intake of homocysteine or methionine.37 Although these other models have been useful, the Mthfr+/– mice are a particularly attractive model because of their modest increase in homocysteine and the high natural occurrence of MTHFR deficiency in humans.
Our finding of reduced ApoA-I expression in CBS-deficient mice, as well as in MTHFR-deficient mice, suggests that it is the elevation in homocysteine, rather than the disturbance in folate metabolism, that regulates this apolipoprotein. Consistent with this argument are the effects of homocysteine on the ApoA-I promoter and in HepG2 cells, which were conducted in folate-replete medium.
The increase in mRNA for Cyp7A1 in liver of hyperhomocysteinemic mice is also consistent with enhanced atherogenesis. Cyp7A1 catalyzes 7 hydroxylation of cholesterol, the rate-limiting step in bile acid synthesis. Bile acids are transported from the liver to the intestine, where they facilitate cholesterol absorption by promoting the formation of micellar cholesterol.
Two key proteins involved in reverse cholesterol efflux, ApoA-I and Cyp7A1, were altered in expression in liver of Mthfr-deficient mice. Both proteins are regulated at the transcriptional level by the transcription factor PPAR, which has been shown to increase ApoA-I expression30 and decrease Cyp7A1 expression.21 Our findings in HepG2 cells confirm the upregulation of human ApoA-I by PPAR, because transfection with a PPAR construct increased ApoA-I levels and a mutated PPRE markedly reduced ApoA-I promoter activity. Homocysteine reduced the levels of PPAR and ApoA-I protein in HepG2 cells, suggesting that PPAR may be involved in homocysteine regulation of ApoA-I. Homocysteine also reduced ApoAI promoter activity; however, a PPRE-independent mechanism could be involved because the ApoAI promoter with a mutagenized PPRE was still subject to homocysteine inhibition. Although the homocysteine levels in the medium were higher than the pathophysiological range of plasma tHcy, these concentrations are often used38,39 in cultured cell systems to achieve a 2- to 6-fold increase in intracellular homocysteine.38
The regulation of MTHFR by PPAR is a novel finding that provides an additional interaction between homocysteine and lipid metabolism. A recent study has demonstrated that MTHFR activity can be regulated by phosphorylation; the phosphorylated enzyme is suggested to be less active than the nonphosphorylated enzyme.33 Our experiments showed increased expression of the phosphorylated and nonphosphorylated forms of MTHFR following transfection of HepG2 cells with PPAR. These findings add more complexity to the regulation of MTHFR, which has 2 promoters, alternative splicing and extra coding sequences for the higher molecular weight isoform. Additional work is required to delineate the functional significance of this PPAR-dependent increase in MTHFR expression.
Our finding of homocysteine modulation of PPAR has also been suggested in studies of homocysteine-induced endothelial cell dysfunction. Homocysteine may antagonize the effects of PPAR and PPAR in endothelial cell remodeling.40 Interestingly, treatment of CBS-deficient mice with fibrates, which are known to induce PPAR, have been reported to reverse the Hcy-mediated endothelial dysfunction in these mice.24
The altered expression of ApoA-I in hyperhomocysteinemic mice was confirmed in a human cohort, a well-characterized group of males with coronary artery disease. In previous studies of this cohort, we showed that the MTHFR mutant genotype (677TT) was associated with increased plasma homocysteine only when folate status was low. Therefore, the association of homocysteine with reduced ApoA-I would only be expected in humans with the 677TT genotype when their folate status was low. We did not observe an association between the 677TT genotype and reduced ApoA-I levels in this study, because some of the males with this genotype had adequate folate and were therefore not hyperhomocysteinemic. Sampling of this population was performed before the introduction of folate fortification of food in Canada. It would be interesting to examine the association among folate, homocysteine, and apolipoproteins in North Americans after fortification.
In summary, we have demonstrated a significant negative relationship between levels of Hcy and ApoA-I in hyperhomocysteinemic males and in hyperhomocysteinemic Mthfr-deficient mice. Studies of large clinical populations are required to determine whether combined evaluation of both biomarkers is a better predictor of disease risk than either alone. In addition, it remains to be determined whether folate supplementation, which is known to reduce homocysteine levels, also modifies lipoprotein profiles.
Acknowledgments
This work was supported by the Canadian Institutes of Health Research (CIHR). R.R. was supported by a Senior Investigator Award from CIHR, and L.M. was the recipient of a Fellowship from the Montreal Children’s Hospital Research Institute. We gratefully acknowledge the assistance of Qing Wu for animal care, Sébastien Dubé for statistical analysis, and Andrea Lawrance for advice.
Footnotes
Original received August 26, 2005; resubmission received October 24, 2005; revised resubmission received December 28, 2005; accepted January 12, 2006.
References
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. [Order article via Infotrieve]
Refsum U, Ueland PM, Nygrd O, Vollset SE. Homocysteine and vascular disease. Ann Rev Med. 1998; 49: 31–62. [Order article via Infotrieve]
Frosst P, Blom H, Milos R, Goyette P, Sheppard C, Matthews R, Boers G, den Heijer M, Kluijtmans L, van den Heuvel L, Rozen R. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet. 1995; 10: 111–113. [Order article via Infotrieve]
Chen Z, Karaplis A, Ackerman S, Pogribny I, Melnyk S, Lussier-Cacan S, Chen MF, Pai A, John S, Smith R, Bottiglieri T, Bagley P, Selhub J, Rudnicki M, James SJ, Rozen R. Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet. 2001; 10: 433–443.
Ghandour H, Chen Z, Selhub J, Rozen R. Mice deficient in methylenetetrahydrofolate reductase exhibit tissue-specific distribution of folates. J Nutr. 2004; 134: 2975–2978.
Virdis A, Iglarz M, Neves M, Amiri F, Touyz R, Rozen R, Schiffrin E. Effect of hyperhomocystinemia and hypertension on endothelial function in methylenetetrahydrofolate reductase-deficient mice. Arterioscler Thromb Vasc Biol. 2003; 23: 1352–1357.
Neves M, Endermann D, Amiri F, Virdis A, Pu Q, Rozen R, Schiffrin E. Small artery mechanics in hyperhomocysteinemic mice: effects of angiotensin II. J Hypertens. 2004; 959–966.
Gordon T, Castelli W, Hjortland M, Kannel W, Dawber T. High density lipoprotein as a protective factor against coronary heart disease: the Framingham study. Am J Med. 1977; 62: 707–714. [Order article via Infotrieve]
Rubin E, Krauss R, Spangler E, Verstyuft J, Clift S. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein AI. Nature. 1991; 353: 265–267. [Order article via Infotrieve]
Assmann G, Nofer J. Atheroprotective effects of high-density lipoproteins. Annu Rev Med. 2003; 54: 321–341. [Order article via Infotrieve]
Suc I, Escargueil-Blanc I, Troly M, Salvayre R, Negre-Salvayre A. HDL and apoA prevent cell death of endothelial cells induced by oxidized LDL. Arterioscler Thromb Vasc Biol. 1997; 17: 2158–2166.
Cockerill G, Rye K, Gamble J, Vadas M, Barter P. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol. 1995; 15: 1987–1994.
Ho Y, Brown M, Goldstein J. Hydrolysis and excretion of cytoplasmic cholesteryl esters by macrophages: stimulation by high density lipoprotein and other agents. J Lipid Res. 1980; 21: 391–398.
Groenendijk M, Cantor R, de Bruin T, Dallinga-Thie G. The apoAI-CIII-AIV gene cluster. Atherosclerosis. 2001; 157: 1–11. [Order article via Infotrieve]
Pennachio L, Olivier M, Hubacek J, Cohen J, Cox D, Fruchart J-C, Krauss R, Rubin E. An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science. 2001; 194: 169–172.
Duval C, Fruchart JC, Staels B. PPARa, fibrates, lipid metabolism and inflammation. Arch Mal Coeur Vaiss. 2004; 97: 665–672. [Order article via Infotrieve]
Castrillo A, Tontonoz P. PPARs in atherosclerosis: the clot thickens. J Clin Invest. 2004; 114: 1538–1540.
Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E, Martin G, Fruchart J-C, Laudet V, Staels B. The nuclear receptors peroxisome proliferator-activated receptor-a and Reverba mediate the species specific regulation of apolipoprotein A-I expression by fibrates. J Biol Chem. 1998; 273: 25713–25720.
Fruchart J-C, Staels B, Duriez P. PPARs, metabolic disease and atherosclerosis. Pharmacol Res. 2001; 44: 345–352. [Order article via Infotrieve]
Prieur X, Coste H, Rodríguez J. The human apolipoprotein AV gene is regulated by peroxisome proliferator-activated receptor a and contains a novel farnesoid X-activated receptor response element. J Biol Chem. 2003; 278: 25468–25480.
Marrapodi M, Chiang Y. Peroxisome proliferators-activated receptor a (PPARa) and agonist inhibit cholesterol 7a-hydroxylase gene (CYP7A1) transcription. Lipid Res. 2000; 41: 514–520.
Agellon L. Metabolism and function of bile acids. In: Vance DE, JE V, eds. Biochemistry of Lipids, Lipoproteins and Membranes. Amsterdam: Elsevier; 2002, 433–447.
Brude IR, Finstad HS, Seljeflot I, Drevon CA, Solvoll K, Sandstad B, Hjermann I, Arnesen H, Nenseter MS. Plasma homocysteine concentration related to diet, endothelial function and mononuclear cell gene expression among male hyperlipidaemic smokers. Eur J Clin Invest. 1999; 29: 100–108. [Order article via Infotrieve]
Sood H, Hunt M, Tyagi S. Peroxisome proliferator ameliorates endothelial dysfunction in a murine model of hyperhomocysteinemia. Am J Physiol Lung Cell Mol Phsyiol. 2003; 284: L333–L341.
Schwahn B, Wendel U, Lussier-Cacan S, Mar M, Zeisel S, Leclerc D, Castro C, Garrow T, Rozen R. Effects of betaine in a murine model of mild cystathionine-beta-synthase deficiency. Metabolism. 2004; 53: 594–599. [Order article via Infotrieve]
Chen Z, Ge B, Hudson T, Rozen R. Microarray analysis of brain RNA in mice with methylenetetrahydrofolate reductase deficiency and hyperhomocysteinemia. Brain Res Gene Expr Patterns. 2002; 1: 89–93. [Order article via Infotrieve]
Leclerc D, Deng L, Trasler J, Rozen R. ApcMin/+ mouse model of colon cancer: gene expression profiling in tumors. J Cell Biochem. 2001; 93: 1242–1254.
Christensen B, Frosst P, Lussier-Cacan S, Selhub J, Goyette P, Rosenblatt D, Genest JJ, Rozen R. Correlation of a common mutation in the methylenetetrahydrofolate reductase gene with plasma homocysteine in patients with premature coronary artery disease. Arterioscler Thromb Vasc Biol. 1997; 17: 569–573.
Schwahn B, Chen Z, Laryea M, Wendel U, Lussier-Cacan S, Genest JJ, Mar M-H, Zeisel S, Castro C, Garrow T, Rozen R. Homocysteine-betaine interactions in a murine model of 5,10-methylenetetrahydrofolate reductase deficiency. FASEB J. 2003; 17: 512–514.
Morishima A, Ohkubo N, Maeda N, Miki T, Mitsuda N. NFkB regulates plasma apolipoprotein A-I and high density lipoprotein cholesterol through inhibition of peroxisome proliferator-activated receptor a. J Biol Chem. 2003; 278: 38188–38193.
Vu-Dac N, Schoonjans K, Kosykh V, Dallongeville J, Fruchart J-C, Staels B. Fibrates increase human Apolipoprotein A-II expression through activation of the peroxisome proliferator-activated receptor. J Clin Inv. 1995; 96: 741–750. [Order article via Infotrieve]
Tran P, Leclerc D, Chan M, Pai A, Hiou-Tim F, Wu Q, Goyette P, Artigas C, Milos R, Rozen R. Multiple transcription start sites and alternative splicing in the methylenetetrahydrofolate reductase gene result in two enzyme isoforms. Mamm Genome. 2002; 13: 483–492. [Order article via Infotrieve]
Yamada K, Strahler J, Andrews P, Matthews R. Regulation of human methylenetetrahydrofolate reductase by phosphorylation. Proc Natl Acad Sci U S A. 2005; 102: 10454–10459.
de Koning A, Werstuck G, Zhou J, Austin R. Hyperhomocysteinemia and its role in the development of atherosclerosis. Clin Biochem. 2003; 2003: 431–441.
Namekata K, Enokido Y, Ishii I, Nagai Y, Harada T, Kimura H. Abnormal lipid metabolism in cystathionine beta-synthase-deficient mice, an animal model for hyperhomocysteinemia. J Biol Chem. 2004; 279: 52961–52969.
Wang H, Jiang X, Gaubatz J, Ma L, Magera M, Yang X, Berger P, Durante W, Pownall H, Schafer A. Hyperhomocysteinemia accelerates atherosclerosis in cystathionine beta-synthase and apolipoprotein E double knock-out mice with and without dietary perturbation. Blood. 2003; 101: 3901–3907.
Troen A, Lutgens E, Smith D, Rosenberg I, Selhub J. The atherogenic effect of excess methionine intake. Proc Natl Acad Sci U S A. 2003; 100: 15089–15094.
Werstuck G, Lentz S, Dayal S, Hossain GS, Sood S, Shi Y, Zhou J, Maeda N, Krisans S, Malinow M, Austin R. Homocysteine-induced endoplasmic reticulum stress causes dysregulation of the cholesterol and triglyceride biosynthetic pathways. J Clin Invest. 2001; 107: 1263–1273.
Kokame K, Agarwala K, Kato H, Miyata Y. Herp, a new ubiquitin-like membrane protein induced by endoplasmic reticulum stress. J Biol Chem. 2000; 275: 32846–32853.
Mujumdar V, Tummalapalli C, Aru G, Tyagi S. Mechanism of constrictive vascular remodeling by homocysteine: role of PPAR. Am J Physiol Cell Physiol. 2002; 282: C1009–C1015.
您现在查看是摘要介绍页，详见ORG附件。