DNA Promoter Methylation in Breast Tumors: No Association with Genetic Polymorphisms in MTHFR and MTR

Introduction

Both CpG island promoter hypermethylation and global DNA hypomethylation are prominent features of breast tumors and are important in the carcinogenic process (1, 2). However, the factors that result in aberrant DNA methylation in normal and neoplastic tissues are not well known. The one-carbon metabolism, critical in the availability of methyl groups and therefore DNA methylation, may affect both hypomethylation and hypermethylation (3). We examined whether common genetic variations of one-carbon metabolism genes, specifically methylenetetrahydrofolate reductase (MTHFR) and methionine synthase (MTR), are associated with promoter methylation of the three genes selected because they are functionally important and have been found to be methylated in breast tumors and may therefore influence breast carcinogenesis: E-cadherin, involved in cell adhesion (4); p16, important in cell cycle regulation (5); and retinoic acid binding receptor-β₂ (RAR-β₂), important in receptor-mediated cell signaling (6). These associations were examined in a case-control study of primary, incident breast cancers, examining the association between MTHFR and MTR polymorphisms and breast cancer risk by tumor methylation status as well as the likelihood of methylation in tumors by genotype.

Materials and Methods

Detailed study methods have been published previously (7). In brief, the Western New York Exposures and Breast Cancer Study included 1,170 primary, histologically confirmed, incident breast cancer cases, ages 35 to 79 at diagnosis and 2,115 randomly selected population controls, frequency-matched to cases on age and race. Extensive in-person interviews and self-administered questionnaires were administered to participants including queries on demographic factors and breast cancer risk factors. The response rates were 72% for cases and 63% for controls. Information on tumor size, histologic grade, and cancer stage was abstracted from medical charts using a standard protocol. Estrogen receptor (ER) status was determined by immunohistochemistry as described previously (7). DNA was extracted from blood and mouthwash samples using the GenQuik DNA Extraction Kit (BioServe Biotechnologies, Ltd.). Archived tumor blocks were obtained from 920 (78.6%) breast cancer cases.

The allelic discrimination of the MTHFR C677T, A1298C, and MTR A2756G polymorphisms were assessed by real-time PCR with TaqMan genotyping assay with primers, probes, and conditions as described on the National Cancer Institute's Cancer Genome Anatomy Project SNP500 Cancer Database web site.⁶ Promoter methylation of E-cadherin, p16, and RAR-β₂ was determined by real-time methylation-specific PCR after bisulfite modification of genomic tumor DNA isolated from archived paraffin-embedded tissues of 803 breast cancer tumors (described in detailed previously; ref. 7).

The exact χ² goodness-of-fit test was used to test the Hardy-Weinberg equilibrium of the genotypes. Characteristics of participating cases with and without promoter methylation of specific genes and controls were compared using the ANOVA test for continuous variables and χ² test for categorical variables. For comparisons of cases with and without promoter methylation to controls, odds ratios (OR) and 95% confidence intervals (CI) were calculated using polytomous logistic regression. Unconditional logistic regression was used for case-case comparisons of those with and without promoter methylation to estimate the ORs and 95% CIs for the associations of MTHFR and MTR genotypes with promoter methylation in breast cancer. Interactions between genotype and menopause, ER status, folate intake, or alcohol intake were evaluated by evaluation of a multiplicative term in the regression model. All analyses were adjusted by age and race. For case-case comparisons, we also adjusted for ER status. Potential confounding effects of other demographic factors and known breast cancer risk factors were also examined, and the results changed by <10% (data not shown). All statistical tests were based on two-sided probability. All statistical analyses were done using SAS version 9.1 (SAS Institute).

Results

The frequencies of genotypes of MTR and MTHFR polymorphisms for cases and controls and distributions of selected characteristic factors are shown in Table 1 . All polymorphisms were in Hardy-Weinberg equilibrium for cases and controls in both the whole population and in Caucasians only. The distributions of MTR and MTHFR polymorphisms were similar for cases with or without E-cadherin, p16, and RAR-β₂ gene promoter methylation.

The results of case-case and case-control comparisons evaluating the associations between MTR A2756G and MTHFR C677T and C1298A genotypes and breast tumors with or without promoter methylation are presented in Table 2 . In case-control comparisons, we did not find any association between MTHFR and MTR polymorphisms with breast cancer risk stratified by methylation status. In case-case analyses, there was no association between MTR A2756G, MTHFR C677T, or C1298A polymorphisms and E-cadherin, p16, or RAR-β₂ gene methylation in breast tumors for both premenopausal and postmenopausal women. We also analyzed the above relations using dominant and recessive models of inheritance in both case-control comparisons and case-case comparisons, however, no associations were observed. In addition, results of analyses stratified by ER status, folate intake, and lifetime alcohol intake were similar; there were no interactions and all values for interaction tests were P > 0.05. For the joint effects of MTR and MTHFR genotypes on the likelihood of promoter methylation in at least one gene, compared with the MTR 2756AA, MTHFR 677CC, and 1298AA genotype, women with breast cancer with more variant alleles for either MTR 2756G, MTHFR 677T, or 1298C alleles tend to increase the likelihood of promoter methylation in at least one gene, although associations were not statistically significant and there was no trend (ORs, 1.21, 1.30, and 1.45; 95% CIs, 0.65-2.22, 0.70-2.41, and 0.72-2.91 for women with any one, any two, and three variant alleles, respectively).

Discussion

MTHFR and MTR are key enzymes in the biosynthesis of 5-methyl tetrahydrofolate and methionine, which are precursors for DNA methylation reactions; and these enzymes' activities also affect the availability of tetrahydrofolate for nucleotide biosynthesis (3). MTHFR C677T and A1298C variants are associated with a reduction of enzyme activity (8, 9) and the MTR A2756G variant is associated with lower homocysteine concentrations (10, 11), and have been investigated for their possible effect on breast carcinogenesis with inconsistent results (12-17).

Polymorphisms of MTHFR C677T and A1298C and MTR A2756G have been investigated in relation to promoter methylation of genes in breast tumors in one other study of 227 breast cancer cases. Consistent with our findings, no association was observed between MTHFR C677T and MTR A2756G polymorphisms and the frequency of promoter methylation in seven genes, including E-cadherin, p16, and RAR-β₂, in breast cancer (18). In our study, we also found that this association was not modified by menopausal status, ER status, folate intake, or total alcohol intake.

There have been studies examining these genotypes with promoter methylation of genes in tumors from other sites with inconsistent results (19-26). Curtin et al. found an increased likelihood of highly CpG-methylated phenotype in colon tumors for those with one or two variant MTHFR 1298C alleles, and the association was modified by high-risk dietary profiles (low folate and methionine intake and high alcohol use; ref. 19). In other studies, these genotypes were not associated with the likelihood of p16 promoter hypermethylation of colorectal cancers (20-22). Similarly, MTHFR C677T, A1298C, or MTR C2756G genotypes were not associated with E-cadherin and p16 promoter methylation in esophageal (23) and cervical cancers (25).

The strengths of our study include the population-based study design and a relatively large sample size, leading to more stable risk estimates. Nevertheless, the number of cases in subgroups were small and the CIs were wide. We had 80% power to detect ORs of 2.5 for the association between MTR genotype and methylation and an OR of 2.0 for the association between MTHFR polymorphisms and methylation. Additionally, the study applied the candidate gene approach and primarily focused on potential common genetic variants (>5%) and polymorphisms with amino acid changes. We cannot rule out the possibility that genetic polymorphisms of one-carbon metabolisms other than those included in our study may be related to the likelihood of methylation. Given multiple genes for proteins involved in one-carbon metabolism (3), the confounding and/or modifying effects of other genes also cannot be excluded. Further lack of response among cases has potential for selection bias. Compared with participating cases, those nonparticipants were of a somewhat lower education level and older. However, the two groups were similar in terms of tumor stage, distant metastases, and other breast cancer risk factors. Furthermore, our inability to obtain the paraffin-embedded breast tumor tissue for all cases may have led to bias. In comparisons of cases without available archived tumor tissue to those with available tissue, those with tissue were somewhat younger at diagnosis and had a higher tumor-node-metastasis stage of breast tumor. However, the two groups were similar in terms of tumor size, histologic grade, nuclear grade, ER and progesterone receptor status. With regard to the validity of the measure of methylation in formalin-fixed paraffin-embedded tissue, a recent study showed a high correlation in methylation between paraffin-embedded tumor tissues and fresh samples from the same subject, measured with a similar real-time PCR method as ours, concluding that paraffin-embedded samples are well suited for methylation assessment (27).

In summary, we found no evidence that these common polymorphisms of the MTHFR and MTR genes were associated with the prevalence of promoter methylation of E-cadherin, p16, and RAR-β₂ genes in breast tumors.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.