Skip Navigation


Carcinogenesis Advance Access originally published online on March 26, 2007
Carcinogenesis 2007 28(7):1510-1519; doi:10.1093/carcin/bgm062
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
28/7/1510    most recent
bgm062v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hazra, A.
Right arrow Articles by Hunter, D. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hazra, A.
Right arrow Articles by Hunter, D. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Twenty-four non-synonymous polymorphisms in the one-carbon metabolic pathway and risk of colorectal adenoma in the Nurses' Health Study

Aditi Hazra1,*, Kana Wu2, Peter Kraft3, Charles S. Fuchs4,5, Edward L. Giovannucci2,4 and David J. Hunter1,2,4

1 Program in Molecular and Genetic Epidemiology, Department of Epidemiology
2 Department of Nutrition, Harvard School of Public Health, Boston, MA 02115, USA
3 Department of Biostatistics, Harvard School of Public Health, Boston, MA 02115, USA
4 Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, 181 Longwood Avenue, Boston, MA 02115, USA
5 Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA 02115, USA

* To whom correspondence should be addressed. Tel: +1 617 525 2035; Fax: +1 617 525 2008; Email: ahazra{at}hsph.harvard.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 References
 
Dietary folate and alcohol consumption as well as polymorphic variants in one-carbon metabolism genes may modulate risk of colorectal adenoma through aberrant DNA methylation and altered nucleotide synthesis and repair. We assessed the association of 24 non-synonymous single nucleotide polymorphisms (nsSNPs) in 13 genes in the one-carbon metabolism pathway and risk of colorectal adenoma in 556 incident cases and 557 controls nested in the Nurses' Health Study. Most of the SNPs were not associated with risk of colorectal adenoma. We did, however, observe a modest increased risk among carriers of the transcobalamin (TCN) II 259 Pro/Arg + Arg/Arg variant (odds ratio 1.48, 95% confidence interval 1.09–2.02) for colorectal adenoma. The TCN II Pro259Arg polymorphism may affect TCN binding and transport of vitamin B12 and thus warrants further investigation of its biological function. In addition, the methionine synthase reductase (MTRR) Arg415Cys and MTRR Ser284Thr variant carriers, also in the vitamin B12 pathway, have suggestive associations with advanced colorectal adenoma (defined as being larger than 1 cm, villous, tubular–villous or carcinoma in situ histology). We observed significant evidence for departure from multiplicative interaction for the betaine–homocysteine methyltransferase (BHMT) Arg239Gln with dietary methyl status (based on intake of dietary folate, methionine and alcohol intake) in relation to colorectal adenoma; no such interaction was observed for the other 23 SNPs. Further investigation is required to validate the association of the polymorphisms in the one-carbon metabolic genes and risk of colorectal adenoma.

Abbreviations: BHMT, betaine–homocysteine methyltransferase; CI, confidence interval; FOLH, Folate hydrolase; FTHFD, formyltetrahydrofolate dehydrogenase; MTHFD1, methylenetetrahydrofolate dehydrogenase; MTHFR, methylenetetrahydrofoloate reductase; MTHFS, methylenetetrahydrofolate synthase; MTRR, methionine synthase reductase; NHS, Nurses' Health Study; nsSNP, non-synonymous single nucleotide polymorphism; OR, odds ratio; SNPs, single nucleotide polymorphisms; TCN, transcobalamin II


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Supplementary material
 References
 
Low dietary folate (1–10) and high alcohol consumption (11,13) have been associated with increased risk of colorectal cancer and adenoma. Folate, antagonized by alcohol, is a B-complex nutrient involved in the transfer of single-carbon moieties essential for purine and pyrimidine synthesis necessary for maintenance of genomic integrity and DNA methylation reactions (14,15). Hypermethylation of promoter sequences influences gene expression. Global hypomethylation (16–19) and regional DNA hypermethylation (20–22) as well as altered nucleotide synthesis and repair (6) are associated with folate-related colorectal carcinogenesis. Furthermore, hypermethylation may be an early event in colorectal carcinogenesis and may be involved in the progression of adenomas to cancer (21,23–25).

Case–control studies and prospective studies have supported the role of dietary folate in relation to colorectal adenoma risk (1,26,27). Low folate intake (1,28–31), and low circulating folate levels (26,29,32) were associated with a greater incidence of colorectal adenomas, whereas higher folate intake was inversely associated with recurrence of colorectal adenoma (7).

Genetic variants in methylenetetrahydrofolate reductase (MTHFR) at codon 222 and 429 have been studied in relation to colorectal adenoma risk (27,33–38). High intake of alcohol (30+ g/day) was associated with increased risk of colorectal adenoma among men with the MTHFR 222 Val/Val variant (5,36), but not with the MTHFR Gln429Ala single nucleotide polymorphism (SNP) (36). Thus, the one-carbon metabolism genes may be associated with risk of colorectal adenoma under conditions of low dietary folate status.

Although the association of additional polymorphisms in the one-carbon metabolism genes has been examined in colorectal cancer (36,39), the association with colorectal adenoma risk among women remains unclear. We conducted a prospective nested case–control study in the Nurses’ Health Study (NHS) to evaluate the association between the 24 non-synonymous polymorphisms in the one-carbon pathway, selected in 13 genes (Figure 1; Supplementary Table I is available at Carcinogenesis Online). Because the associations of the genotypes may differ by dietary factors that influence methyl status, we also evaluated the interaction of these SNPs with total folate, methionine and alcohol intake.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. The one-carbon metabolic pathway, highlighting the 13 polymorphic genes in this study. The one-carbon metabolic genes in this study are denoted as ovals in the figure and the SNP symbol, codon, amino acid change and reference number are indicated in parenthesis below. Many of the genes in the one-carbon metabolic pathway, such as 5,10-methylenetetrahydrofolate reductase (MTHFR Ala222Val, rs1801133; MTHFR Glu429Ala, rs1801131), contribute to prymidine and purine synthesis as well as DNA methylation reactions. Folate hydrolase (FOLH1 Tyr75His, rs202676; FOLH1 His475Tyr) (60), also known as Glutamate carboxypeptidase II His475Tyr, is involved in the conversion of dietary folate (polyglutamate) to folate (monoglutamate). 10-formyltetrahydrofolate dehydrogenase (FTHFD Leu254Pro, rs3796191; FTHFD Asp793Gly rs1127717; FTHFD Ile812Val, rs4646750) converts 10-formyltetrahydrofolate (10-formyl THF), a precursor for nucleotide biosynthesis, to tetrahydrofolate. Cytosolic serine hydroxymethyltransferase (cSHMT Leu474Phe, rs1979277) catalyzes the reversible cleavage of serine to form glycine and monocarbonic groups. In addition, glycinamide ribonucleotide transformylase (GART Val421Ile, rs8788; GART Asp752Gly rs8971), 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase (ATIC Cys116Ser rs2372536), methylenetetrahydrofolate dehydrogenase 1 (MTHFD1 Arg134Lys, rs1950902; Arg653Gln, rs2236225) and methylenetetrahydrofolate synthase (MTHFS Thr202Ala, rs8923) are also involved in nucleotide synthesis. Betaine–homocysteine methyltransferase (BHMT Arg239Gln, rs3733890), methionine synthase reductase (MTRR Ile22Met, rs1801394; Ser175Leu, rs1532268; Ser284Thr, rs2303080; Lys350Arg, rs162036; Arg415Cys, rs2287780; His595Tyr, rs10380), methionine synthase (MTR Asp919Gly, rs1805087) and transcobalamin II (TCN II Pro259Arg, rs1801198) are involved in the remethylation of homocysteine to methionine. DNA methyltransferase 1 (DNMT1 Ille311Val, rs2228612) has a direct role in DNA methylation. Other abbreviations in the figure: THF, tetrahydrofolate; DHF, dihydrofolate; SAM, S-adenosylmethionine; SAH, S-adenosylhomocysteine; dUMP, deoxyuridine monophosphate; dTMP, deoxythymidine monophosphate; B2, vitamin B2; B6, vitamin B6 and B12, vitamin B12. *Components of the dietary methyl status variable.

 

   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Supplementary material
 References
 
Study population
The NHS began in 1976 when 121 700 female registered nurses in the USA between the ages of 30 and 55 years completed a self-administered questionnaire on their medical history and baseline health-related exposures. Subsequently, these participants have completed a mailed, self-administered questionnaire biennially to update information on their lifestyle, medical history and diet (every 2–4 years).

In 1989 and 1990, 32 826 women, between the ages of 43 and 69 years, provided blood samples. Upon receipt, samples were immediately centrifuged and aliquoted into plasma, red blood cells and buffy coat fractions for storage in liquid nitrogen. As described previously (33,40), cases of colorectal adenoma and controls were chosen from among women who supplied a blood sample, had a sigmoidoscopy or colonoscopy (by 1998) after providing a blood sample and were free from diagnosed cancer (except non-melanoma skin cancer), ulcerative colitis and adenoma before endoscopy. Newly diagnosed polyps (in individuals who never had a polyp diagnosed before the date of blood draw) were reported on the 1990, 1992, 1994, 1996 or 1998 questionnaires. These polyps were confirmed to be adenomatous by review of histopathological reports and were classified by location (proximal or distal colon or rectum), size (<1 cm, ≥1 cm) and histology (tubular, tubulovillous, villous, carcinoma in situ) by study investigators blinded to the exposure information. Advanced adenoma was defined as large adenomas that were ≥1 cm in size and/or tubulovillous, villous or carcinoma in situ histology. Between 1989 and 1998, 557 women with colorectal adenoma were identified and one control was selected for each case, matched on year of birth, year of blood draw, time period of endoscopy, indication for endoscopy and time period of first or most recent endoscopy. One case was subsequently found to be hyperplastic and thus was excluded from the analysis. Thus, the total number of cases and controls analyzed in this study were 556 cases and 557 controls. Diet was assessed in 1986, 1990, 1994 and 1996 with a semiquantitative food frequency questionnaire.

Genotyping methods
Genotyping was performed at the Dana-Farber/Harvard Cancer Center High-Throughput Polymorphism Core. DNA was extracted from 50 µl buffy coat fractions diluted with 150 µl of PBS by the Qiagen QIAamp Blood Kit (Qiagen, Chatsworth, CA) spin protocol. The genotypes of the one-carbon polymorphisms were determined by measuring end-point fluorescence using the 5' nuclease assay (Taqman) on the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA) (41). The 24 non-synonymous SNPs (nsSNPs) were identified through the NCI SNP500Cancer database (http://snp500cancer.nci.nih.gov), the dbSNP database (http://www.ncbi.nlm.nih.gov/SNP) and the International HapMap Project database (http://www.hapmap.org). The SNPs that were included in this study all resulted in amino acid changes and therefore are potentially functional. The FOLH1 His475Tyr SNP was not referenced in the above databases, but was investigated in relation to serum folate levels (60). Initially, 26 SNPs in 15 genes were selected with a minor allele frequency ≥2%. However, we were unable to design a Taqman assay for the rs1051266 SNP and the rs1143693 SNP was not polymorphic in our data (39). Therefore, our analysis included the known 24 nsSNPs in 13 genes. Koushik et al. (39) used the Polymorphism Phenotyping and Sorting Intolerant from Tolerant web tools to predict the potential effect of these SNPs on protein structure and function.

Quality control was ensured by including a random 10% of the samples in the 96 well plates as duplicates. The quality control samples served as internal controls to validate the genotyping methods; there was 100% concordance of the quality control samples. Laboratory personnel were blinded to the status (case, control or quality control) of samples. The median genotyping success for the 24 SNPs was 95%.

Statistical analysis
A total of 556 incident adenoma cases and 557 controls were included in the analyses, although the number of cases and controls for some of the polymorphisms was slightly lower because of missing data. Genotype distributions were evaluated for agreement with Hardy–Weinberg equilibrium by the chi-square test. We evaluated differences among cases and controls using the chi-square test for categorical variables and the paired t-test for continuous variables.

We used both conditional and unconditional logistic regression for the analyses to compute genotype odds ratios (ORs), 95% confidence intervals (CIs), allelic trend tests and interactions. In the multivariate models, we adjusted for known risk factors for adenomas, including family history of colorectal cancer, pack-years of smoking, post-menopausal hormone use, body mass index, physical activity, aspirin use, total energy intake and consumption of red meat, alcohol and folate. Unconditional logistic analyses explicitly adjusted for matching variables, including age, history of previous endoscopy (before blood draw), year of endoscopy, as well as month/year of blood collection and fasting status at time of blood draw.

The SNPs were initially examined with the codominant model, evaluating the SNPs as a three-level categorical variable, where the homozygous variant genotype and heterozygous genotype are compared with the homozygous wild-type referent category. P values (43) were obtained from the tests for linear trend of log and ORs were calculated using an ordered categorical variable by assigning scores to the genotypes: 0 (no variant allele), 1 (carrying one variant allele) and 2 (carrying two variant alleles).

In addition, we have also evaluated the SNPs using the 2 degree-of-freedom likelihood ratio test comparing models with and without the genotype variable. Overall, most of the P values for the genotypes in the one-carbon metabolism genes did not change significance (data not shown) with the 2 degree-of-freedom likelihood ratio test. The methionine synthase reductase (MTRR) 175 SNP had a P value 0.03 from the 2 degree-of-freedom likelihood ratio test that may be driven by the multivariate results for the MTRR Leu/Leu variant OR 1.59 (95% CI 1.03–2.45). However, since the majority of SNPs had a minor allele frequency of <5%, we also used the dominant model, comparing variant carriers with the referent homozygous wild-type.

The risk for advanced and small adenoma was evaluated using polytomous logistic regression with an ordinal outcome variable. This outcome variable was modeled as a three-level categorical variable: advanced adenoma, small adenoma and controls.

Haplotypes for genes with multiple SNPs were imputed using the expectation-maximization method, and ORs for haplotypes were calculated using haplotype trend regression (44), with the most common haplotype serving as the referent. We also calculated a global chi-square test evaluating the contribution of all haplotypes in a gene as risk factors for colorectal adenoma. We report d-prime (D') and r2 measures of linkage disequilibrium between SNPs within each gene (45).

We evaluated the associations between the polymorphisms and dietary methyl status that were defined by consumption of folate, methionine and alcohol. The primary analysis of dietary methyl status was based on the 1986 diet instead of the cumulative updated diet, since previous studies suggest that folate and alcohol act relatively early in the colorectal adenoma-to-carcinoma sequence (2,33,36). The variable dietary methyl status was modeled as a three-level categorical variable: high, intermediate and low dietary methyl status. High dietary methyl status was based on energy-adjusted folate and methionine consumption above the median levels among controls (folate: 334.5 mg/day; methionine: 1.74 g/day), and alcohol intake <5 g/day. Low dietary methyl status was defined as folate and methionine consumption below the median cut point among controls and alcohol intake ≥5 g/day. Individuals with neither high nor low methyl status were categorized as having intermediate dietary methyl status. We evaluated departure from multiplicative interaction on the OR scale using the P value obtained from the likelihood ratio test comparing models with and without the interaction term between genotype and dietary methyl status (as an ordered categorical variable).

We also examined the association of pairwise gene–gene combinations with risk of colorectal adenoma. We calculated a test for significance of the interaction term to determine whether a SNP in a pairwise model was associated with adenoma. All statistical tests were two sided. All statistical analyses were performed with SAS (version 9.1; SAS Institute, Cary, NC, USA) (4).


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Supplementary material
 References
 
Overall, we found no evidence of departure from Hardy–Weinberg equilibrium in controls at the 0.05 level, except for the FOLH Tyr75His polymorphism (P = 0.01, more heterozygotes were observed than expected). The minor allele frequency for all the SNPs was at least 3%. The conditional logistic regression analyses included all 24 SNPs.

Study population characteristics
Our study population included 556 colorectal adenoma cases and 557 matched controls from the NHS. The characteristics of the cases and controls have been described (40). We examined whether the baseline distributions of known risk factors for colorectal adenoma were similar among these cases and controls who had provided a blood sample compared with previous observations for the overall NHS cohort among participants who provided a blood sample. The factors that differed between cases and controls are similar to those seen in the whole cohort (8). The risk patterns were similar to those reported for the entire cohort, though not always statistically significant (Table I). The mean age of cases and their matched controls was 62.3 years. Family history of colorectal cancer, smoking >25 pack-years, body mass index >30 kg/m2 and consumption of >1 serving of red meat/day was associated with an increased risk of colorectal adenoma. In addition, cases were less probably to be current users of post-menopausal hormones or to take multivitamins and aspirin (<7 aspirin tablets/week).


View this table:
[in this window]
[in a new window]

  		Table I. Characteristics of colorectal adenoma cases and controls in the NHS

 
Among the adenoma cases were 153 proximal adenomas, 254 distal adenomas, 70 rectal adenomas and 61 individuals with multiple adenomas. Endoscopied controls that had a sigmoidoscopy may have had an undetected proximal adenoma. Therefore, we did a subanalysis restricted to women with adenomas in the distal colon and rectum. The observed point estimates of the 24 nsSNPs in the subanalysis did not differ [point estimates varied <20%, except for BHMT 239 Arg/Gln + Gln/Gln that had on OR 1.28 (95% CI 0.94–1.74)] from the overall main effects presented in Table II.


View this table:
[in this window]
[in a new window]

  		Table II. Polymorphisms in the one-carbon metabolic pathway and risk of colorectal adenoma in the NHSa

 
This analysis reflects the cumulative consumption of folate prior to fortification in 1997, since only the last cycle of dietary follow-up was ascertained after folate fortification. Nevertheless we performed a subanalysis of the data stratified by year of adenoma diagnosis (prior to fortification analysis: adenoma diagnosis in 1995 or earlier; post-fortification analysis: adenoma diagnosis after 1996) and did not find any salient differences (point estimates varied <20%) in the association of the 24 polymorphisms and colorectal adenoma risk (data not shown).

Main effects of the one-carbon polymorphisms and colorectal adenoma
The main associations observed for the 24 non-synonymous polymorphisms and risk of colorectal adenoma are presented in Table II. Overall, we did not observe strong associations for any of the 24 polymorphisms in the one-carbon metabolic pathway and risk of colorectal adenoma. A borderline increased risk for the transcobalamin (TCN) II 259 Pro/Arg + Arg/Arg variants and colorectal adenoma risk was noted (OR 1.33, 95% CI 1.01–1.75) that was slightly stronger after adjustment for covariates (OR 1.48, 95% CI 1.09–2.02). There was a suggestion of increased risk of colorectal adenoma with the 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase (ATIC) 116 Cys/Ser + Ser/Ser variant (OR 1.30, 95% CI 1.00–1.70), methylenetetrahydrofolate synthase (MTHFS) 202 Ala/Thr + Thr/Thr variants (OR 1.39, 95% CI 0.98–1.98) and the MTRR 175 Leu/Leu variant (OR 1.59, 95% CI 1.03–2.45). FOLH 475 His/Tyr + Tyr/Tyr variants had suggestive association with an increased risk of colorectal adenoma, but this association attenuated after adjustment for the known colorectal adenoma risk factors. The other nsSNPs were not significantly associated with modified risk of colorectal adenoma. In Table III, we present a polytomous logistic regression analysis examining the associations of the 24 SNPs among cases with advanced colorectal adenoma and cases with small adenoma compared with controls. The linked MTRR 284 Ser/Thr + Thr/Thr and MTRR 415 Arg/Cys + Cys/Cys variant carriers were associated with an increased risk of advanced colorectal adenoma (MTRR 284: OR 1.96, 95% CI 1.05–3.67; MTRR 415: OR 1.98, 95% CI 1.06–3.56).


View this table:
[in this window]
[in a new window]

  		Table III. Polytomous logistic regression for polymorphisms in the one-carbon metabolic pathway and colorectal adenoma status

 
Estimates of the interaction of dietary methyl status and the genotypes are presented in Table IV. Among drinkers, baseline average alcohol consumption was 6.56 g/day for cases and 7.22 g/day for controls; 38% of women reported no alcohol consumption (Table I). Compared with individuals with folate and methionine intakes above the median and <5 g/week of alcohol consumption, defined as high methyl status, individuals with intermediate or low methyl status were associated with a suggestive increased risk for colorectal adenoma, with an OR of 1.36 (95% CI 0.96–1.92). Similar point estimates were obtained in the analysis of methyl status as a three-category variable, however, to maximize power the results are presented for the collapsed variable combining the intermediate and low categories (data not shown). Generally, we did not find evidence of a departure from the multiplicative model for the SNPs and dietary methyl status, except for BHMT Arg239Gln (test for interaction P = 0.01), which may be due to chance.


View this table:
[in this window]
[in a new window]

  		Table IV. Association of the polymorphisms in the one-carbon metabolic pathway among colorectal adenoma cases and dietary methyl statusa

 
We also assessed the association between the haplotypes of the folate hydrolase (FOLH) (D' = 0.89), formyltetrahydrofolate dehydrogenase (FTHFD), phosphoribosylglycinamide formyltransferase (D' = 1.0), methylenetetrahydrofolate dehydrogenase (MTHFD) (D' = 1.0), MTHFR (D' = 0.28) and MTRR (please see Supplementary Table II is available at Carcinogenesis Online) polymorphisms and the risk of colorectal adenoma (data not shown). There was a suggestion of an association with the FOLH1 (global test for differences in risk: {chi}2 = 5.99, df = 2, P = 0.05) haplotypes and colorectal adenoma risk. Three haplotypes constructed from the 2 FOLH1 SNPs, FOLH Tyr75His and FOLH1 His475Tyr, were estimated with a frequency ≥5%. Haplotypes that carried the variant alleles were associated with an increased colorectal adenoma risk, with conditional adjusted ORs ranging from 1.19 (95% CI 0.94–1.51) for the 1-0 haplotype to 1.42 (95% CI: 0.92–2.19) for the 0-1 haplotype, compared with the wild-type haplotypes (0-0).

In addition, we performed exploratory analysis of pairwise gene–gene combinations to determine whether any two SNPs in the one-carbon pathway were jointly associated with colorectal adenoma risk (please see Supplementary Table III available at Carcinogenesis Online for gene–gene interactions with P value <0.05). Slightly higher than expected by chance, 22 interactions, out of 281 tests performed, had a P value of <0.05 (P value calculated using the likelihood ratio test comparing a model with no genetic effects with a model containing the main effect of each SNP and the product term between the pairwise SNPs). Although six tests had a P value <0.01, no P values were statistically significant after the multiple comparisons (46) Bonferroni correction (P < 0.0002).


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Supplementary material
 References
 
Genes in the one-carbon metabolic pathway may modulate risk of colorectal adenoma by influencing methyl group availability for DNA methylation reactions or nucleotide synthesis (4,22,27,35,47). A recent study showed that 60% of colorectal adenomas are abnormally methylated on at least one locus. Furthermore, CpG island methylation phenotype has been demonstrated in large colorectal adenomas and in adenomas with tubular or villous histology (24,25).

We examined the relationship of 24 nsSNPs in 13 genes in the one-carbon metabolic pathway and risk of colorectal adenoma. The numerous statistical tests we performed may be interpreted as hypothesis generating, since these SNPs have not been collectively examined for adenoma risk. In summary, most of the 24 nsSNPs in the one-carbon metabolic pathway did not significantly alter the risk of colorectal adenoma in our study. In our study, carrier status for the TCN II Pro259Arg polymorphism was associated with an increased risk of colorectal adenoma. TCN II is a transport protein essential for the cellular uptake of vitamin B12 (48). Vitamin B12 is a critical cofactor in the remethylation of homocysteine to methionine (48,49). The TCN II Pro259Arg (C776G) polymorphism may affect transcobalamin binding activity and transport of cellular vitamin B12 (50–54) but further studies are needed to confirm these results (55).

In a study of these 24 nsSNPs and colorectal cancer in the NHS and Health Professionals Follow-Up Study, the B12-related SNPs were also significantly associated with risk of colorectal cancer (39). In a previous study of colorectal cancer, the TCN II Pro259Arg variants were not significantly associated with overall risk (among men and women). However, the TCN II Pro259Arg variants may have an inverse relationship with colorectal cancer among women (OR 0.70, 95% CI 0.49–1.00) (39). In summary, the TCN II Pro259Arg variants were not significantly associated with advanced colorectal adenoma or cancer, but were associated with overall colorectal adenoma risk in this study. Although this finding may be due to chance, this result suggests that TCN II Pro259Arg may be relevant to early adenoma development. This previous study also observed an association with the variant carrier genotypes of MTRR Ser284Thr (OR 1.85, 95% CI 1.05–3.27) and MTRR Arg415Cys (OR 2.03, 95% CI 1.14–3.56) with an increased risk of colorectal cancer (39). In our study, the linked MTRR Ser284Thr and MTRR Arg415Cys variant carriers were associated with a 2-fold increased risk for advanced colorectal adenoma. These results suggest that B12-related genes, methionine synthesis and DNA methylation might be important in the colon adenoma-to-carcinoma sequence.

We also examined the influence of common haplotypes in the FOLH1, FTHFD, MTHFD, MTHFR and MTRR genes on risk of colorectal adenoma. The analysis provides additional information about the susceptibility to colorectal adenoma associated with variants in FOLH (global test P = 0.05). However, haplotype analysis did not reveal any additional information about susceptibility to colorectal adenoma in relation to the FTHFD, MTHFD, MTHFR and MTRR SNPs.

MTHFR reduces 5,10-methyleneTHF to 5-methylTHF, required for the production of methionine and S-adenosylmethionine, the methyl donor for DNA. A decrease in MTHFR activity associated with the 222 Val/Val variant may lead to an elevation in 5,10-methyleneTHF, required for DNA synthesis. The relationship of the MTHFR Ala222Val polymorphism and risk of colorectal cancer and adenoma has been studied (27,33,36,39,49,56,57). A meta-analysis of seven colorectal adenoma studies did not find a significant association with the MTHFR Ala222Val SNP (37). However, when folate status is low, the MTHFR Val/Val variant is associated with an increase in homocysteine concentration and DNA hypomethylation (18). Alcohol consumption may alter the folate pool toward serine, impeding the function of MTRR and MTHFR (11,12). Hence, studies have shown that the risk associated of the MTHFR 222 Val/Val variant is more evident in the presence of a low-methyl diet (39) that is mostly driven by high alcohol consumption at levels of 30 g/day (36). However, an analysis of relation of dietary methyl status and the MTHFR SNPs with risk of colorectal adenoma, in this study among women, was not consistent with findings among men in the Health Professionals Follow-Up Study (36). The modest consumption of alcohol, the main component of the dietary methyl status, coupled with the high mean intake of dietary folate at baseline (higher than the recommended intake of 400 µg/day) in our study population may partially explain the null findings.

The selection of controls from a defined cohort of women who had given a blood sample in the NHS reduces the risk of selection bias. However, there are several limitations to this study. Although these 24 nsSNPs were chosen with prior biological relevance, there is a concern of multiple testing. Furthermore, these findings are limited by the incomplete functional characterization of several of these nsSNPs. Therefore, the suggestive associations of the 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase and MTHFS SNPs warrant further investigation. At this time, we do not have plasma B12 levels for all the study participants and therefore were unable to evaluate interactions with the B12-related genes in this study. The US Food and Drug Administration mandated the fortification of grain products with folic acid in 1996 (58), but the adenomas in this study were diagnosed from 1989 to 1998 and thus mostly in the pre-fortification era.

Although our study included 556 cases of colorectal adenoma, we had limited numbers for analysis of advanced adenoma and interaction analysis with dietary methyl status. The low consumption of alcohol coupled with high dietary folate intake among the women in our study population may mask associations between these SNPs and dietary methyl status. The results could be different in a population with more heavy alcohol consumption and marginal folate status. Therefore, we may have underestimated the role of reduced methyl group availability in this pathway. Moreover, given the numerous statistical tests we performed, our positive findings need to be replicated.

There are limited data on the role of genes, beyond MTHFR, in the one-carbon pathway in the etiology of colorectal adenoma and the data on women are particularly sparse (9,33,34,37). The findings from Polymorphism Phenotyping and Sorting Intolerant from Tolerant did not support strong functional differences for these SNPs (39). Our results extend the current knowledge of the genetic variation in the one-carbon metabolic pathway and risk of colorectal adenoma. The findings from our prospective examination suggest that these non-synonymous polymorphisms in the one-carbon metabolic genes have limited influence on the risk of colorectal adenoma. Similar to previous studies, we did not find a significant association between MTHFR Ala222Val (9,33,34,37) or methionine synthase (MTR) Asp919Gly (59) polymorphism and risk of colorectal adenoma. We did observe a main effect of the TCN II Pro259Arg polymorphism, suggesting that additional investigation of the biological function of the TCN II Pro259Arg genetic variants may help elucidate the relationship of folate, B12, and alcohol consumption with risk of colorectal adenoma. Studies on adenoma among individuals consuming more alcohol (i.e. 20 g) and less folate may be more probably to be informative. These results suggest the need for future studies on B12 levels and their association with SNPs in the one-carbon metabolic pathway. Further investigation of these findings coupled with functional knowledge of these SNPs will offer insight into the role of the polymorphisms in the one-carbon metabolic network in the etiology of colorectal adenoma. Although analysis of two-way combinations of SNPs did not identify more statistically significant combination than expected by chance, integrated pathway analysis (9) may afford us the opportunity to elucidate how genetic variations in multiple genes collaborate in the etiology of colorectal neoplasia.


   Supplementary material
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
References
 
Supplementary tables I and II can be found at http://carcin.oxfordjournals.org/.


   Acknowledgments
 
We thank Pati Soule, Hardeep Ranu and Craig Labadie for their laboratory assistance. We are indebted to the dedicated participants of the NHS for their ongoing commitment. This research is supported by the National Institutes of Health Research Grants CA87969, U54 CA100971, P01-CA087969 and The Entertainment Industry Foundation's National Colorectal Cancer Research Alliance. A.H. is supported in part by training grant National Institutes of Health T-32 CA 09001-30.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 References
 

  1. Giovannucci E, et al. Folate, methionine, and alcohol intake and risk of colorectal adenoma. J. Natl Cancer Inst. (1993) 85:875–884.[Abstract/Free Full Text]
  2. Giovannucci E, et al. Multivitamin use, folate, and colon cancer in women in the Nurses' Health Study. Ann Intern. Med. (1998) 129:517–524.[Abstract/Free Full Text]
  3. Song J, et al. Chemopreventive effects of dietary folate on instestinal polyps in Apc+/Msh2 -/- mice. Cancer Res. (2000) 60:5434–5440.[Abstract/Free Full Text]
  4. Giovannucci E. Epidemiologic studies of folate and colorectal neoplasia, a review. J. Nutr. (2002) 132:2350S–2355S.[Abstract/Free Full Text]
  5. Giovannucci E. Alcohol, one-carbon metabolism, and colorectal cancer, recent insights from molecular studies. J. Nutr. (2004) 134:2475S–2418S.[Abstract/Free Full Text]
  6. Kim YI. Role of folate in colon cancer development and progression. J. Nutr. (2003) 133:3731–3739S.
  7. Martinez MS, et al. Folate and colorectal neoplasia, relation between plasma and dietary markers of folate and adenoma recurrence. Am. J. Clin. Nutr. (2004) 79:691–697.[Abstract/Free Full Text]
  8. Wei E, et al. Plasma vitamin B6 and the risk of colorectal cancer and adenoma in women. J. Natl Cancer Inst. (2005) 97:684–692.[Abstract/Free Full Text]
  9. Ulrich CM. Nutrigenetics in cancer research—folate metabolism and colorectal cancer. J. Nutr. (2005) 135:2698–2702.[Abstract/Free Full Text]
  10. Ulrich CM, et al. Folate supplementation, too much of a good thing. Cancer Epidemiol. Biomarkers Prev. (2006) 15:189–193.[Free Full Text]
  11. Baron JA, et al. Folate intake, alcohol consumption, cigarette smoking, and risk of colorectal adenoma. J. Natl Cancer Inst. (1998) 90:57–62.[Abstract/Free Full Text]
  12. Halsted CH, et al. Metabolic interactions of alcohol and folate. J. Nutr. (2002) 132:2367S–2372S.[Abstract/Free Full Text]
  13. Cho E, et al. Alcohol intake and colorectal cancer, a pooled analysis of 8 cohort studies. Ann. Intern. Med. (2004) 140:603–613.[Abstract/Free Full Text]
  14. Jones PA, et al. The fundamental role of epigenetic events in cancer. Nat. Rev. Genet. (2002) 3:415–428.[ISI][Medline]
  15. Stover PJ. Physiology of folate and vitamin B12 in health and disease. Nutr. Rev. (2004) 62:S3–S12.[CrossRef][ISI][Medline]
  16. Pufelete M, et al. Folate status, genomic DNA hypomethylation, and risk of colorectal adenoma and cancer, a case-control study. Gastroenterology (2003) 124:1240–1248.[CrossRef][ISI][Medline]
  17. Choi SW, et al. Folate status, effects on pathways of colorectal carcinogenesis. J. Nutr (2002) 2413S–2418S.
  18. Friso S, et al. A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc. Natl Acad. Sci. USA. (2002) 99:5606–5611.[Abstract/Free Full Text]
  19. Novakovic P, et al. Effects of folate deficiency on gene expression in the apoptosis and cancer pathways in colon cancer cells. Carcinogenesis. (2006) 27:916–924.[Abstract/Free Full Text]
  20. Wainfan E, et al. Methyl groups in carcinogenesis, effects of DNA methylation and gene expression. Cancer Res. (1992) (suppl.):2071S–2077S.
  21. Pufulete M, et al. Effect of folic acid supplementation on genomic methylation in patients with colorectal adenoma. Gut (2005) 54:648–653.[Abstract/Free Full Text]
  22. Kim YI. Nutritional epigenetics, impact of folate deficiency on DNA methylation and colon cancer susceptibility. J. Nutr. (2005) 135:2703–2709.[Abstract/Free Full Text]
  23. Bai AHC, et al. Promoter hypermethylation of tumor-related genes in the progression of colorectal neoplasia. Int. J. Cancer (2004) 112:846–853.[CrossRef][ISI][Medline]
  24. Kim HC, et al. CpG island methylation as an early event during adenoma progression in carcinogenesis of sporadic colorectal cancer. J. Gastroenterol Hepatol (2005) 20:1920–1926.[CrossRef][ISI][Medline]
  25. Kim HY, et al. CpG island methylation of genes accumulates during the adenoma progression step of the multistep pathogenesis of colorectal cancer. Genes Chromosomes Cancer (2006) 45:781–789.[CrossRef][ISI][Medline]
  26. Paspatis GA, et al. Folate status and adenomatous colonic polyps. A colonscopically controlled study. Dis. Colon Rectum (1995) 38:64–67.[CrossRef][ISI][Medline]
  27. Ulrich CM, et al. Colorectal adenomas and the C677T MTHFR polymorphism, evidence for gene-environment interaction? Cancer Epidemiol. Biomarkers Prev. (1999) 8:659–668.[Abstract/Free Full Text]
  28. Benito E, et al. Diet and colorectal adenomas: a case-control study in Majorca. Int. J. Cancer (1991) 53:213–219.
  29. Bird CL, et al. Red cell and plasma folate, folate consumption, and the risk of colorectal adenomatous polyps. Cancer Epidemiol. Biomarkers Prev. (1995) 4:709–714.[Abstract]
  30. Tseng M, et al. Micronutrients and the risk of colorectal adenomas. Am. J. Epidemiol. (1996) 144:1005–1014.[Abstract/Free Full Text]
  31. Boutron-Ruault MC, et al. Folate and alcohol intakes: related or independent roles in the adenoma-carcinoma sequence? Nutr. Cancer (1996) 26:337–346.[ISI][Medline]
  32. Paspatis GA, et al. Folate supplementation and adenomatous colonic polyps. Dis. Colon Rectum. (1994) 37:1340–1341.[ISI][Medline]
  33. Chen J, et al. MTHFR polymorphism, methyl-replete diets and the risk of colorectal carcinoma and adenoma among U.S. men and women: an example of gene-environment interactions in colorectal tumorigenesis. J. Nutr. (1999) 129(suppl.):560S–564S.[ISI][Medline]
  34. Chen J, et al. Linkage disequilibrium between the 677C>T and 1298A>C polymorphisms in human methylenetetrahydrofolate reductase gene and their contributions to risk of colorectal cancer. Pharmacogenetics (2002) 12:339–342.[CrossRef][ISI][Medline]
  35. Chen J, et al. Polymorphisms in the one-carbon metabolic pathway, plasma folate levels, and colorectal cancer in a prospective study. Int. J. Cancer (2004) 110:617–620.[CrossRef][ISI][Medline]
  36. Giovannucci E, et al. Methylenetetrahydrofolate reductase, alcohol dehydrogenase, diet, and risk of colorectal adenomas. Cancer Epidemiol. Biomarkers Prev. (2003) 12:970–979.[Abstract/Free Full Text]
  37. Kono S, et al. Genetic polymorphism of methylenetetrahydrofolate reductase and colorectal cancer and adenoma. Cancer Sci (2005) 96:535–542.[CrossRef][Medline]
  38. Ulrich CM, et al. Polymorphisms in the reduced folate carrier, thymidylate synthase or methionine synthase and risk of colon cancer. Cancer Epidemiol. Biomarkers Prev. (2005) 14:2509–2516.[Abstract/Free Full Text]
  39. Koushik A, et al. Polymorphisms in one-carbon metabolism genes and associations with colorectal cancer. Cancer Epidemiol. Biomarkers Prev (2006) 15:2408–2417.[Abstract/Free Full Text]
  40. Tranah GJ, et al. APC Asp1822Val and Gly2502Ser polymorphisms and risk of colorectal cancer and adenoma. Cancer Epidemiol. Biomarkers Prev (2005) 5:863–870.
  41. Livak KJ. Allelic discrimination using fluorogenic probes and 5’ nuclease assay. Genet Anal. (1999) 14:143–149.[Medline]
  42. SAS User's Guide (2002) Cary, N.C: SAS Institute Inc.
  43. Dudbridge F, et al. Rank truncated product of P-values, with application to genomewide association scans. Genet. Epidemiol. (2003) 25:360–366.[CrossRef][ISI][Medline]
  44. Zaykin DV, et al. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum. Hered (2002) 53:79–91.[ISI][Medline]
  45. Kraft P, et al. Accounting for haplotype uncertainty in matched association studies: a comparison of simple and flexible techniques. Genetic Epidemiol. (2005) 28:261–272.[CrossRef][ISI][Medline]
  46. Daly MJ, et al. Partners in crime. Nat. Genet. (2005) 37:337–338.[CrossRef][ISI][Medline]
  47. Friso S, et al. Gene-nutrient interactions in one-carbon metabolism. Curr. Drug Metab. (2005) 6:37–46.[CrossRef][ISI][Medline]
  48. Wuerges J, et al. Structural basis for mammalian vitamin B12 transport by transcobalamin. Proc. Natl Acad. Sci. USA (2006) 103:4386–4391.[Abstract/Free Full Text]
  49. Sharp L, et al. Polymorphisms in genes involved in folate metabolism and colorectal neoplasia, a HuGE review. Am. J. Epidemiol. (2004) 159:423–443.[Abstract/Free Full Text]
  50. Afman LA, et al. Single nucleotide polymorphisms in the transcobalamin gene: relationship with transcobalamin concentrations and risk for neural tube defects. Eur. J. Hum. Genet (2002) 10:433–438.[CrossRef][ISI][Medline]
  51. Miller JW, et al. Transcobalamin II 775G>C polymorphism and indices of vitamin B12 status in healthy older adults. Blood (2002) 100:718–720.[Abstract/Free Full Text]
  52. Fodinger M, et al. Effect of TCN2 776C>G on vitamin B12 cellular availability in end-stage renal disease patients. Kidney Int. (2003) 64:1095–1100.[CrossRef][ISI][Medline]
  53. Winkelmayer WC, et al. Effects of TCN2 776C>G on vitamin B, folate, and total homocysteine levels in kidney transplant patients. Kidney Int. (2004) 65:1877–81.[CrossRef][ISI][Medline]
  54. Von Castel-Dunwoody KM, et al. Transcobalamin 776 C G polymorphism negatively affects vitamin B-12 metabolism. Am. J. Clin. Nutr. (2005) 81:1436–1441.[Abstract/Free Full Text]
  55. Gueant JL, et al. The association between plasma homocysteine and holo-transcobalamin and the transcobalamin 776C–>G polymorphism is influenced by folate in the absence of supplementation and fortified diet. Am. J. Clin. Nutr. (2006) 83:171–172.[ISI][Medline]
  56. Ma J, et al. Methylenetetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer. Cancer Res. (1997) 57:1098–1102.[Abstract/Free Full Text]
  57. Sanjoaquin MA, et al. Folate intake and colorectal cancer risk, a meta analytical approach. Int. J. Cancer. (2005) 113:825–828.[CrossRef][ISI][Medline]
  58. Matsuo K, et al. One-carbon metabolism related gene polymorphisms interact with alcohol drinking to influence the risk of colorectal cancer in Japan. Carcinogenesis. (2005) 26:2164–2171.[Abstract/Free Full Text]
  59. Foulke J. Folic Acid to Fortify US Food Products to Prevent Birth Defects (1996) Washington, D.C. US Food and Drug Administration.
  60. Devlin AM, et al. Glutamate carboxypeptidase II: a polymorphism associated with lower levels of serum folate and hyperhomocystemia. Hum. Mol. Genet. (2000) 9:2837–2844.[Abstract/Free Full Text]
Received December 5, 2006; revised March 6, 2007; accepted March 10, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
28/7/1510    most recent
bgm062v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hazra, A.
Right arrow Articles by Hunter, D. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hazra, A.
Right arrow Articles by Hunter, D. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?