![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
* Division of Nutritional Sciences, Cornell University, Ithaca, NY;
Department of Nutritional Sciences and Toxicology, University of California, Berkeley, CA;
** Channing Laboratory, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA;
Brigham and Women’s Hospital, Division of Aging, Boston, MA;
Department of Biological Statistics and Computational Biology, Cornell University, Ithaca, NY; and
Jean Mayer U.S. Department of Agriculture Human Nutrition Research Center on Aging at Tufts University, Boston, MA
2To whom correspondence should be addressed. E-mail: pac6{at}cornell.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
T) affects CVD risk, and that the effect depends on a linked step in the metabolic pathway catalyzed by methylenetetrahydrofolate reductase (MTHFR). A nested case-control study of incident CVD was conducted within the all-male Normative Aging Study cohort. Of the incident CVD cases, 507 had DNA samples; 2 controls/case were selected by risk set sampling (matched on age and birth year). A significant gene-gene interaction (P-values 0.0013, 0.0064) was found between MTHFR and cSHMT, and there was little or no change in the coefficients in covariate-adjusted models. The effect of MTHFR 677CT genotype on CVD risk varied by cSHMT 1420CT genotype. Among men with cSHMT 1420CT TT genotype, the odds ratios (OR) for CVD risk for MTHFR 677CT CT and TT genotypes compared with the MTHFR 677CT CC genotype were 3.6 (95% CI, 1.7–7.8) and 10.6 (95% CI, 2.5–46.0), respectively. Among men with the cSHMT 1420CT CC/CT genotype, the corresponding ORs were 1.0 (95% CI, 0.8–1.2) and 1.3 (95% CI, 0.9–1.8). Plasma total homocysteine concentrations were highest in the subgroup of men with both polymorphisms, MTHFR 677CT TT and cSHMT 1420CT TT, consistent with a higher risk of CVD in this subgroup. A more complete understanding of the molecular mechanism awaits identification of the functional effect of the polymorphism.
KEY WORDS: • homocysteine • folate • men • cardiovascular disease
The importance of folate in health is well established on the basis of its critical role in metabolism and its relevance to public health even in well-nourished Western populations (1). Folate cofactors carry 1-carbon units that are essential for nucleotide biosynthesis and numerous methylation reactions (2). Altered metabolism of folate is related to the pathogenesis of neural tube defects, cancer, and cardiovascular disease (CVD)3 (3).
Cytoplasmic serine hydroxymethyltransferase (cSHMT) is a key enzyme affecting the intracellular homeostasis among folate cofactors (4). cSHMT reversibly converts serine and tetrahydrofolate (THF) to glycine and 5,10-methylenetetrahydrofolate (5,10-methyleneTHF) (4). The 1-carbon moiety of 5,10-methyleneTHF is in turn directed to the synthesis of purines or thymidylate or to the methionine cycle where it remethylates homocysteine to synthesize methionine and S-adenosyl methionine (2). Three functions of cSHMT are indicated: isotope tracer studies suggest that cSHMT preferentially supplies 1-carbon units to thymidylate biosynthesis; cSHMT catalyzes glycine-dependent serine synthesis, hence depleting 5,10-methyleneTHF for S-adenosyl methionine synthesis; and cSHMT sequesters 5-methyltetrahydrofolate (5-methylTHF), hence decreasing S-adenosyl methionine synthesis (5).
A single nucleotide polymorphism (SNP) in the cSHMT gene was identified: the base at nucleotide 1420 is either C or T, resulting in the amino acid leucine or phenylalanine, respectively, at position 474 of the protein (6). Screening of the coding and regulatory regions of the cSHMT gene yielded only 2 other variants: cSHMT 1181GA, an extremely rare variant (6), and an E to Q substitution at amino acid position 340 (7).
Direct evidence regarding the functional consequence of the cSHMT 1420CT polymorphism is not available, but there is some evidence that the SNP is associated with both metabolic disruption and disease risk. Heil et al. (6) reported lower homocysteine concentrations among mothers whose children had neural tube defects when the mothers were homozygous for the 1420CT variant of cSHMT (TT) (compared with mothers with the CC genotype). In another group of unselected individuals, there were no effects on homocysteine concentrations; neither of these findings was stratified for methylenetetrahydrofolate reductase (MTHFR) variants (6). A recent study found both the CT and the TT genotypes of cSHMT associated with a substantially reduced risk of leukemia relative to the cSHMT 1420CT CC genotype (8). A study of colorectal cancer risk reported no association of the cSHMT SNP with either homocysteine concentration or with cancer risk (9).
This study was designed to test the hypothesis that the cSHMT 1420CT variant affects the risk of CVD. Because polymorphisms in the MTHFR gene, particularly MTHFR 677CT (10,11), affect functionality of a linked step in the folate metabolic network, we hypothesized that the relation of cSHMT 1420CT to CVD risk would differ according to MTHFR genotype. This study explored whether the genotype-disease association was consistent with the effects of the genotype on plasma total homocysteine concentration, a biological marker of folate-dependent methylation, which was measured in a subset of cases and controls. The role of dietary intake of folate and other B vitamins on the genotype-disease association was also explored.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1600 men (70%) were actively participating in follow-up visits (mean age of 70 y). The rate of continued participation of NAS men over the follow-up period was excellent, with <1% annual attrition for all causes.
From November 1961 through December 1998, 749 incident cases of coronary heart disease and stroke occurred. Possible occurrences of nonfatal coronary heart disease, including angina pectoris and myocardial infarction, were evaluated based on medical records and physician examination (13). Angina was defined by Framingham Heart Study criteria as recurrent chest discomfort lasting up to 15 min distinctly related to exertion or excitement that was relieved by rest or nitroglycerin (13). Nonfatal myocardial infarction was diagnosed only when documented by unequivocal electrocardiography changes and chest discomfort consistent with myocardial infarction. Possible occurrences of nonfatal stroke were identified by report of a neurological deficit of sudden or rapid onset that persisted for 24 h or longer, and all reports were confirmed by a neurologist’s review of the medical records (14). Occurrences of fatal coronary heart disease or stroke (in men with no prior nonfatal CVD event) were confirmed by death certificates, which were coded according to the 8th revision of the International Classification of Diseases (15); 6% of cases had only stroke. DNA was available for 72% of cases; thus 535 incident CVD cases were studied. Cases without DNA samples were primarily cases that occurred early in the follow-up. The cases without DNA samples were older, had slightly higher systolic blood pressure, and had greater cumulative smoking exposure at the beginning of the cohort study.
Matched controls (n = 1048) were selected by risk-set sampling (16–18) and were matched to each case by age at onset and birth period (in 5-y intervals) of the case. Thus, as each case occurred, all other cohort members at risk (including cases that occurred at later times), under active follow-up, with the same matching conditions, and with available DNA formed a risk set for the incident case. Two men from the risk set were randomly chosen as matched controls. Cases were eligible to be a control in the time period before they were diagnosed, and some men were chosen as a control for more than one case (19). The study was approved by the Brigham and Women’s Hospital Human Subjects committee, the Veterans’ Administration R & D committee, and the Cornell University Committee on Human Subjects.
Data collection: covariates.
Extensive data are available on these men, including physical measurements, lifestyle factors, and some blood analyses. Beginning in 1987, men completed an FFQ referring to intake in the prior year (20), at every visit. Estimates of dietary intake, including B vitamins, methionine, coffee, and alcohol, were derived from the frequency and dosage information on the FFQ using software developed by the Nurses’ Health Study (20). At least 1 FFQ was available before the date of diagnosis for about half of the CVD cases and noncases. Serum HDL cholesterol was measured beginning in 1981, but the majority of cases had HDL measurements before the CVD event occurred. Lipids were assayed over the course of the follow-up as follows: serum cholesterol was assayed enzymatically (SCALVO Diagnostics); HDL cholesterol was measured in the supernatant after precipitation of the LDL cholesterol and VLDL fractions with dextran sulfate and magnesium, using the Abbott Biochromatic Analyzer 100 (Abbott Laboratories); triglyceride was measured with a Dupont ACA discrete clinical analyzer (Biomedical Products Department, Dupont). Plasma total homocysteine was assayed in an unselected subset of stored blood samples. Plasma samples were stored at –80°C, and transferred to the Jean Mayer USDA Human Nutrition Research Center on Aging, where they were analyzed. Total homocysteine in plasma was determined by an adaptation of the method described by Araki and Sako (21). The CV for this assay was 4.0%. Homocysteine data were available for 54% of cases and 72% of controls. Most (93%) of the plasma homocysteine data were obtained from blood samples collected after the CVD event.
Data collection: genotyping methods.
In 1999, DNA was extracted from stored frozen buffy coat of 7 mL of whole blood, using the QiAmp DNA blood kits (QIAGEN). Genotypes of 3 polymorphisms (cSHMT 1420CT, MTHFR 677CT, and MTHFR 1298AC) were determined by the TaqMan procedure using the allelic discrimination technique (ABI Prism 7900 Sequence Detection System, Applied Biosystems). The DNA samples for cases and controls were plated in random order with a mixture of TaqMan Universal PCR Master Mix, primers, and probes. PCR cycling conditions consisted of one 2-min cycle at 50°C, one 10-min cycle at 95°C, followed by 40–46 cycles at 95°C for 15 s and at 60°C for 1 min. The methods for the analysis of the cSHMT 1420CT polymorphism were described in detail elsewhere (8), and the primers and probes for the 2 MTHFR polymorphisms were created according to standard methods for the TaqMan procedure. Genotypes were successfully determined for 96% of the study group.
Statistical analysis.
Genotype frequencies among controls were compared with those expected in Hardy-Weinberg equilibrium (22) and tested with the 
2 statistic. Linkage disequilibrium among the 3 polymorphic sites was tested using the Likelihood Ratio Test and Fisher’s Exact Test (22). Ordinary least-squares regression models were used to estimate the relation of genotype to plasma total homocysteine, considering homocysteine as the outcome variable. Univariate procedures were used to estimate the covariate means for cases and their matched controls to identify potentially confounding factors. Conditional logistic regression analysis (SAS PHREG; SAS Institute) was used to analyze the risk set sampled cases and controls (19). To test whether the cSHMT-CVD association varied by other genotypes and to assess effect modification of the cSHMT-CVD association by other risk factors, product terms were included in the regression model.
The cSHMT-CVD association was estimated in unadjusted models and in models adjusted for major risk factors for CVD (covariates available on majority of cases and controls). In addition, the cSHMT-CVD association was estimated in the subset of cases and controls with dietary data, and in the subset with homocysteine data. Once again, further models were considered to assess potential confounding factors, mediating effects, and effect modification. Data are reported as regression coefficients or means and SE, and as odds ratios (ORs) with 95% CI. Differences were considered significant at P < 0.05, but exact P-values are shown for all comparisons.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
T (11% TT) (6,8), MTHFR 677CT (11% TT) (23), and MTHFR 1298AC (10% CC) (24) in Caucasians. All 3 polymorphisms were in Hardy-Weinberg equilibrium, MTHFR 677CT and MTHFR 1298AC were in linkage disequilibrium, and neither MTHFR variant was in linkage disequilibrium with cSHMT 1420CT.
|
T CT genotype with CVD risk (vs. CC), both overall and stratified by the MTHFR 677CT genotype. All further analyses therefore considered men with the cSHMT 1420CT CC and CT genotype as a single group. The MTHFR 1298AC genotype had little or no association with CVD risk, and there was no evidence of an interaction with the other genotypes; thus, this genotype was not considered further.
In unadjusted models to assess the main effect of cSHMT before considering other genotypes, there was a protective effect of small magnitude for the cSHMT 1420CT TT genotype (OR 0.9; 95% CI 0.6–1.3). In models testing whether the MTHFR-CVD association differed by the cSHMT 1420CT polymorphism, the gene-gene interaction was significant (P-values ranged from 0.0013 to 0.03 for model coefficients for interaction in Table 3). In the cSHMT 1420CT TT genotype stratum, men with the MTHFR 677CT TT genotype had 10 times the rate of CVD vs. MTHFR 677CT CC, and men with the MTHFR 677CT CT genotype (vs. CC) had 4 times the rate of CVD (Table 4). In the cSHMT 1420CT CC/CT genotype strata, men with the MTHFR 677CT TT genotype had an increase of 30% in the rate of CVD vs. MTHFR 677CT CC, and men with the MTHFR 677CT CT genotype (vs. CC) had little or no elevation in the rate of CVD (Table 4). Similarly, the cSHMT-CVD association varied according to the MTHFR genotype; i.e., the risk of CVD in men with the cSHMT 1420CT TT genotype vs. men with the cSHMT 1420CT CC/CT genotype was 0.4 (95% CI, 0.2–0.8), 1.6 (95% CI, 1.0–2.7), and 3.5 (95% CI, 0.9–14.1) among men with 0, 1, and 2 variant alleles at MTHFR 677CT, respectively.
|
|
T genotype was associated with homocysteine; i.e., TT men had lower homocysteine than CT/CC men (9.9 vs. 10.7 µmol/L, respectively, P = 0.09). The MTHFR 677CT TT genotype had a strong association (P = 0.0014) with plasma homocysteine; i.e., men with the TT genotype had higher homocysteine concentration than men with the CT and CC genotypes, respectively (11.8 vs. 10.4 µmol/L (P = 0.0038) and 10.4 µmol/L (P = 0.0022), respectively). Further models included both genotypes simultaneously and allowed for the gene-gene interaction; the difference in homocysteine concentration across the MTHFR 677CT genotypes was much greater in men with the homozygote variant cSHMT 1420CT genotype. In the cSHMT 1420CT TT genotype stratum, men with the MTHFR 677CT TT genotype had homocysteine concentrations 5.3 µmol/L greater than those of men with the MTHFR 677CT CC genotype (14.8 vs. 9.5 µmol/L, respectively, P = 0.02). In the cSHMT 1420CT CT/CC genotype stratum, men with the MTHFR 677CT TT genotype had homocysteine concentrations 1.1 µmol/L greater than those of men with the MTHFR 677CT CC genotype (11.7 vs. 10.6 µmol/L, respectively, P = 0.13). To investigate whether the genotype-CVD association was mediated by plasma homocysteine, preliminary models confirmed that the same approximate results (before adjusting for homocysteine) were obtained in the reduced set of cases and controls with homocysteine data. Next, models were adjusted for homocysteine, and the findings did not differ from the unadjusted models (Tables 3, and 4). The effect of MTHFR CT and TT (vs. CC) genotype on CVD risk showed a strong dependency on cSHMT genotype, with effects of greatest magnitude observed among men with the cSHMT TT genotype. The larger SEs and wider CI in the adjusted models reflect the reduction in numbers, due to missing data on plasma homocysteine.
Other models explored the influence of potentially confounding variables. When risk factors for CVD were included in multivariate models (resulting in a reduction in the total number studied, from 1514 to 1418 in matched analyses), there was little or no change in the genotype-CVD model coefficients. Adjusted covariates included BMI, diastolic and systolic blood pressure, serum total cholesterol and triglycerides, current consumption of 
2 drinks/d (yes/no), current smoking (yes/no), number of years smoking, and cumulative smoking dose (pack-years).
Lipid subfractions and dietary intake were considered as covariates; due to missing data, the inclusion of these variables in statistical models resulted in a moderate to substantial reduction in the number of observations. In the subset of cases and controls with available data, further adjusting for dietary folate; vitamins B-2 (i.e., riboflavin), B-6, and B-12; methionine; total energy intake; alcohol and coffee consumption; physical activity (energy spent in a week on average); and serum HDL cholesterol concentration had little or no effect on the regression coefficients for the genotype-CVD association.
Given that dietary folate is expected to partially reverse the metabolic effects of the MTHFR polymorphism, we explored folate models further. The dietary supply of folate from food and supplements (assessed by the first available FFQ) had a weak inverse association with CVD risk. In analyses considering whether the gene-gene interaction (cSHMT by MTHFR) was modified by dietary folate, the 3-way interaction among cSHMT, MTHFR, and folate was not significant.
The NAS participants were white U.S. residents primarily of European descent, and 70% of the men were European whites with identical paternal and maternal ethnic origins (721 of 1034 men); 85% of these 721 men were further categorized into 5 ethnic groups, and the prevalence of the cSHMT 1420CT TT genotype ranged from 0 to 20% in the 5 groups. In models adjusted for ethnic group, there was little or no change in the magnitude of the genotype-CVD association.
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
T genotype with the risk of CVD. This finding was supported by a similar pattern in plasma concentration of homocysteine. To the best of our knowledge, this is the first report of this gene-gene interaction in relation to CVD risk, and the first report of the relation between joint genotype status for MTHFR 677CT and cSHMT 1420CT and homocysteine, an established biological marker of 1-carbon metabolism.
The findings may be affected by other genetic variants that are in linkage disequilibrium with the cSHMT 1420CT SNP. However, no other common polymorphisms were reported in the coding region of the cSHMT gene. The findings may be affected by population stratification; that is, the cSHMT polymorphism may be associated with other genetic and/or nongenetic causal factors that vary by ethnic group. However, including ethnicity as a covariate had little or no effect on the cSHMT-CVD association in these data. Moreover, the variation in allele frequency by ethnicity was within limits considered unlikely to cause confounding by population stratification (25).
The selection of the cases and/or case definition may influence the study findings. One-third of the incident CVD cases occurred before DNA sampling started, and the loss of these cases could bias the estimation of the gene-disease association. Among the cases without DNA samples, a greater proportion had early-onset CVD (age < 60 y); in analyses limited to early-onset cases with DNA, the cSHMT 1420CT by MTHFR 677CT interaction was even stronger. Observed effect estimates in the full set of cases and controls may therefore underestimate the true association. An additional consideration is the definition of a case: the case definition was a mixture of endpoints (CVD only, stroke only, CVD and stroke). However, when stroke-only cases were excluded, the revised effect estimates were essentially unchanged.
The cSHMT polymorphism does not affect the catalytic activity or thermostability of the purified human cSHMT protein in vitro (unpublished data, Oppenheim, Szebenyi and Stover, Cornell University). Results from molecular modeling demonstrate that the 1420CT polymorphism is not positioned to influence the enzyme active site, but rather is located on the exterior of the protein on a side chain that faces the bulk solvent (unpublished data, Oppenheim, Szebenyi, and Stover, Cornell University). The polymorphism results in a phenylalanine at amino acid position 474, which is located in a region of the protein shown to be involved in protein-protein interactions (unpublished data, Woller, Szebenyi, and Stover, Cornell University). Of the 3 postulated functions of cSHMT, the polymorphism is hypothesized to increase the amount of cSHMT available to bind with 5-methylTHF, thus increasing the sequestering of 5-methylTHF and decreasing S-adenosyl methionine synthesis. This explanation is consistent with our observation of a greater effect of the MTHFR polymorphism in men with the cSHMT polymorphism, and highest levels of homocysteine in men homozygous for both polymorphisms.
The MTHFR 677CT variant disrupts the riboflavin-binding site of the enzyme, and the diminished enzyme activity associated with the polymorphism is stabilized by an adequate supply of intracellular folate (26,27). Intracellular folate status is affected by the dietary intake of folate (1). The net effect of the sequestering of 5-methylTHF by cSHMT may also be affected by folate supply. There was no statistical evidence for a 3-way interaction; however, the elevated risk associated with having both cSHMT 1420CT TT and MTHFR 677CT CT/TT genotypes was attenuated when the analysis was limited to men with intakes of folate above the median. There was no evidence of further effect modification by other methylation-related dietary factors including vitamins B-2 (riboflavin), B-6, and B-12 (cofactors of folate-regulating enzymes, MTHFR, cSHMT, and methionine synthase, respectively).
Although the functional consequences of the cSHMT 1420CT polymorphism are still under investigation, studies demonstrated effects on homocysteine (6), and leukemia risk (8), but no association with colorectal cancer (9). The study reported herein investigated the association between a novel polymorphism in cSHMT and the risk of CVD. We observed a gene-gene interaction such that the association of MTHFR 677CT and CVD varied by the cSHMT genotype. The increased risk of CVD associated with MTHFR 677CT CT and TT genotypes was of greater magnitude among men with the cSHMT 1420CT TT genotype. The genotype-CVD association was confirmed by the genotype-homocysteine association, although the association of genotype was not mediated by homocysteine in these data. The exact role of homocysteine as a causal factor vs. risk marker remains to be determined. Further studies are required to confirm these findings, and to better understand the relation of genotype to biological markers of folate-dependent methylation, including homocysteine, SAM, and SAH; this will advance our understanding of the biological mechanism underlying these findings.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 Abbreviations used: CC, genotype refers to either MTHFR 677 CC or cSHMT 1420 CC; cSHMT, cytoplasmic serine hydroxymethyltransferase; CVD, cardiovascular disease; 5-methylTHF, 5-methyltetrahydrofolate; 5,10-methyleneTHF, 5,10-methylenetetrahydrofolate; MTHFR, methylenetetrahydrofolate reductase; NAS, Normative Aging Study; OR, odds ratio; SNP, single nucleotide polymorphism; THF, tetrahydrofolate; TT, genotype refers to either MTHFR 677 TT or cSHMT 1420 TT.
Manuscript received 26 February 2005. Initial review completed 31 March 2005. Revision accepted 23 May 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Molloy, A. M. & Scott, J. M. (2001) Folates and prevention of disease. Public Health Nutr. 4:601-609.[Medline]
2. Shane, B. (2000) Folic acid, vitamin B12, and vitamin B6. Stipanuk, M. H. eds. Biochemical and Physiological Aspects of Human Nutrition :483-518 W. B. Saunders Company Philadelphia, PA.
3. McNulty, H. (1995) Folate requirements for health in different population groups. Br. J. Biomed. Sci. 52:110-119.[Medline]
4. Girgis, S., Nasrallah, I. M., Suh, J. R., Oppenheim, E., Zanetti, K. A., Mastri, M. G. & Stover, P. J. (1998) Molecular cloning, characterization and alternative splicing of the human cytoplasmic serine hydroxymethyltransferase gene. Gene 210:315-324.[Medline]
5. Herbig, A. K., Chiang, E. P., Lee, L. R., Hills, J., Shane, B. & Stover, P. J. (2002) Cytoplasmic serine hydroxymethyltransferase mediates competition between folate-dependent deoxyribonucleotide and S-adenosylmethionine biosyntheses. J. Biol. Chem. 277:38381-38389.
6. Heil, S., Van der Put, N. M., Waas, E., Heijer, M., Trijbels, F.J.M. & Blom, H. J. (2001) Is mutated serine hydroxymethyltransferase (SHMT) involved in the etiology of neural tube defects?. Mol. Genet. Metab. 73:164-172.[Medline]
7. www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=snpSearch SNP for rs7215148 National Center for Biotechnology Information [accessed May 2005].
8. Skibola, C. F., Smith, M. T., Hubbard, A., Shane, B., Roberts, A. C., Law, G. R., Rollinson, S., Roman, E., Cartwright, R. A. & Morgan, G. J. (2002) Polymorphisms in the thymidylate synthase and serine hydroxymethyltransferase genes and risk of adult acute lymphocytic leukemia. Blood 99:3786-3791.
9. Chen, J., Kyte, C., Valcin, M., Chan, W., Wetmur, J. G., Selhub, J., Hunter, D. J. & Ma, J. (2004) Polymorphisms in the one-carbon metabolic pathway, plasma folate levels and colorectal cancer in a prospective study. Int. J. Cancer 110:617-620.[Medline]
10. Verhoef, P., Rimm, E. B., Hunter, D. J., Chen, J. & Kelsey, K. (1998) A common mutation in the methylenetetrahydrofolate reductase gene and risk of coronary heart disease: results among US men. J. Am. Coll. Cardiol. 32:353-359.
11. Klerk, M., Verhoef, P., Clarke, R., Blom, H. J., Kok, F. J. & Schouten, E. G. (2002) MTHFR 677C polymorphism and risk of coronary heart disease: a meta-analysis. J. Am. Med. Assoc. 288:2023-2031.
12. Bell, B., Rose, C. L. & Damon, A. (1972) The Normative Aging Study: an interdisciplinary and longitudinal study of health and aging. Aging Hum. Dev. 3:5-17.
13. Kubzansky, L. D., Sparrow, D., Vokonas, P. & Kawachi, I. (2001) Is the glass half empty or half full? A prospective study of optimism and coronary heart disease in the Normative Aging Study. Psychosom. Med. 63:910-916.
14. Djousse, L., Ellison, R. C., Beiser, A., Scaramucci, A., D’Agostino, R. B. & Wolf, P. A. (2002) Alcohol consumption and risk of ischemic stroke: The Framingham Study. Stroke 33:907-912.
15. Mendez, M. V., Scott, T., LaMorte, W., Vokonas, P., Menzoian, J. O. & Garcia, P. (1998) An association between periodontal disease and peripheral vascular disease. Am. J. Surg. 176:153-157.[Medline]
16. Lubin, J. H. (1986) Extensions of analytic methods for nested and population-based incident case-control studies. J. Chron. Dis. 39:379-388.[Medline]
17. Robins, J. & Pike, M. (1990) The validity of case-control studies with nonrandom selection of controls. Epidemiology 1:273-284.[Medline]
18. Langholz, B. & Goldstein, L. (1996) Risk set sampling in epidemiologic cohort studies. Stat. Sci. 11:35-53.
19. Pearce, N. (1989) Incidence density matching with a simple SAS computer program. Int. J. Epidemiol. 18:981-984.
20. Willett, W. C., Sampson, L., Stampfer, M. J., Rosner, B., Bain, C., Witschi, J., Hennekens, C. H. & Speizer, F. E. (1985) Reproducibility and validity of a semiquantitative food frequency questionnaire. Am. J. Epidemiol. 122:51-65.
21. Araki, A. & Sako, Y. (1987) Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. J. Chromatogr. 422:43-52.[Medline]
22. Weir, B. S. (1996) Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. Genetic Data Analysis: Methods for Discrete Population Genetic Data Sinauer Associates Sunderland, MA.
23. Fletcher, O. & Kessling, A. M. (1998) MTHFR association with arteriosclerotic vascular disease?. Hum. Genet. 103:11-21.[Medline]
24. Weisberg, I., Tran, P., Christensen, B., Sibani, S. & Rozen, R. (1998) A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol. Genet. Metab. 64:169-172.[Medline]
25. Wacholder, S., Rothman, N. & Caporaso, N. (2000) Population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias. J. Nat. Cancer Inst. 92:1151-1158.
26. Frosst, P., Blom, H. J., Goyette, P., Sheppard, C. A., Matthews, R. G., Boers, G.H.J., den Heijer, M., Kluijtmans, L. A., van den Heuvel, L.P.W.J. & Rozen, R. (1995) A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat. Genet. 10:111-113.[Medline]
27. Guttormsen, A. B., Ueland, P. M., Nesthus, I., Nygard, O., Schneede, J. & Vollset, S. E. (1996) Determinants and Vitamin Responsiveness of intermediate hyperhomocysteinemia. J. Clin. Investig. 98:2174-2183.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Baccarelli, P. A. Cassano, A. Litonjua, S. K. Park, H. Suh, D. Sparrow, P. Vokonas, and J. Schwartz Cardiac Autonomic Dysfunction: Effects From Particulate Air Pollution and Protection by Dietary Methyl Nutrients and Metabolic Polymorphisms Circulation, April 8, 2008; 117(14): 1802 - 1809. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Kelemen, T. A. Sellers, J. M. Schildkraut, J. M. Cunningham, R. A. Vierkant, V. S. Pankratz, Z. S. Fredericksen, M. K. Gadre, D. N. Rider, M. Liebow, et al. Genetic Variation in the One-Carbon Transfer Pathway and Ovarian Cancer Risk Cancer Res., April 1, 2008; 68(7): 2498 - 2506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. F. Woeller, D. D. Anderson, D. M. E. Szebenyi, and P. J. Stover Evidence for Small Ubiquitin-like Modifier-dependent Nuclear Import of the Thymidylate Biosynthesis Pathway J. Biol. Chem., June 15, 2007; 282(24): 17623 - 17631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Lim, S. S. Wang, P. Hartge, W. Cozen, L. E. Kelemen, S. Chanock, S. Davis, A. Blair, M. Schenk, N. Rothman, et al. Gene-nutrient interactions among determinants of folate and one-carbon metabolism on the risk of non-Hodgkin lymphoma: NCI-SEER Case-Control Study Blood, April 1, 2007; 109(7): 3050 - 3059. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Reed, H. F. Nijhout, M. L. Neuhouser, J. F. Gregory III, B. Shane, S. J. James, A. Boynton, and C. M. Ulrich A Mathematical Model Gives Insights into Nutritional and Genetic Aspects of Folate-Mediated One-Carbon Metabolism J. Nutr., October 1, 2006; 136(10): 2653 - 2661. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
