Investigation of methylenetetrahydrofolate reductase tagging polymorphisms with colorectal cancer in Chinese Han population

The aim of this case-control study was to assess the relationship between the tagging polymorphisms in methylenetetrahydrofolate reductase (MTHFR) gene and the susceptibility to colorectal cancer (CRC) in a Chinese Han population. A custom-by-design 48-Plex SNPscan Kit was used to determine the genotypes of MTHFR rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A polymorphisms in 387 CRC patients and 1,536 non-cancer controls. The results revealed that MTHFR rs1801133 G>A polymorphism was associated with a decreased risk of overall CRC. While MTHFR rs4845882 G>A polymorphism conferred an increased risk to overall CRC. In a stratified analysis by CRC region, we found MTHFR rs3753584 T>C and rs9651118 T>C polymorphisms were associated with the increased risk of colon cancer. In addition, a significantly increased risk of rectum cancer associated with MTHFR rs3753584 T>C polymorphism was overt. However, MTHFR rs1801133 G>A polymorphism conferred a decreased risk to colon cancer. In conclusion, findings of the present study reveal that the tagging polymorphisms in MTHFR gene (rs3753584 T>C, rs9651118 T>C and rs4845882 G>A) are associated with the increased risk of CRC. However, MTHFR rs1801133 G>A polymorphism confers a decreased risk to CRC. Additional studies with larger sample size are needed to confirm these findings.


INTRODUCTION
Colorectal cancer (CRC) is the fourth most frequent type of malignancy among females and the fifth most frequent type among males in China, accounting for 160,600 and 215,700 cases in 2015, respectively [1]. The CRC morbidity is shooting up in developing countries including China [1,2]; nevertheless, CRC etiology remains unknown. Risk factors, such as advanced age, family history of CRC, benign adenomatous polyp, www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 38), pp: 63518-63527 Research Paper inflammatory bowel disease, drinking, smoking, being physically inactive, low intake of fruits and vegetables and high intake of dietary fat, may play important roles in the occurrence and the development of CRC [3][4][5][6][7][8][9]. Accumulating evidences indicated that besides environmental factors and individual lifestyle, some genetic factors might be related to CRC etiology.
Several studies have found association between circulating or dietary folate level and the risk of CRC [10][11][12]. A long induction period deserves close attention to the study of CRC because the course from normal cells of the rectum and colon, to microadenomas, to macroadenomas, and eventually to adenocarcinomas, experiences over a long duration, approximately 30-40 years [13]. The presence of the long induction period between reduced risk of CRC and adequate folate status is evident in both epidemiologic and molecular mechanistic studies [14]. The lower folate levels lead to an increasing rate of DNA hypomethylation and uracil misincorporation even in non-neoplastic, normal-appearing tissue of cases [14]. Folate is reduced to tetrahydrofolate (THF) to participate in one-carbon metabolism [15]. Folate metabolism is accommodated by several enzymes. Methylenetetrahydrofolate reductase (MTHFR) is the key enzyme [16][17][18]. MTHFR is involved in DNA methylation, repair and synthesis [19]. DNA strand break and repair, and impaired DNA methylation have been associated with folate deficiency and CRC.
MTHFR gene locates on 1p36. 22. The gene encodes a 74.6-kDa protein containing 656 amino acids. MTHFR is also linked to purine synthesis; therefore, plays a vital role in DNA synthesis by the provision of essential nucleotides [18]. Single nucleotide polymorphisms (SNPs) of MTHFR are considered as a potential biomarker which may influence CRC risk. Therefore, we carried out this case-control study in a Chinese Han population to determine whether MTHFR tagging SNPs (rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A) were associated with the risk of CRC.

Demographic characteristics
The frequency distributions of age, sex, smoking status and drinking habit information for the 387 patients with CRC (mean ± SD age, 60.21 ± 12.48 years) and 1,536 controls (mean ± SD age, 60.82± 8.85 years) are shown in Table 1. The age and sex of the CRC patients and controls were well-matched. The differences of smoking status and drinking habit between CRC and non-cancer controls were not statistically significant (P ≥ 0.05) ( Table  1). Tumor grade was defined as high (well differentiated), medium (moderately differentiated); and low (poorly differentiated). Two experienced doctors independently assessed disease stage according to the AJCC criteria (2010). Finally, 196 cases with stage I/II and 191 with stage III/IV were included. Among the CRC patients, 218 were rectum cancer and 169 were colon cancer. And the primary information for MTHFR polymorphisms is summarized in Table 2.
Association of MTHFR rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A polymorphisms with CRC patients The genotypes of MTHFR rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A polymorphisms in CRC patients and non-cancer controls are summarized in Table 3. The observed genotype frequencies for the five MTHFR tagging SNPs in control group were all in HWE (P = 0.134, 0.082, 0.733, 0.164 and 0.747 for MTHFR rs9651118 T>C, rs4846048 A>G, rs4845882 G>A, rs3753584 T>C and rs1801133 G>A, respectively), which suggested good homogeneity within the present study participants. Four CRC patients (1.03%) and three controls (0.19%) could not be genotyped for poor DNA quantity and/or quality. Overall, we found no statistically significant difference in genotype distribution of MTHFR rs3753584 T>C, rs9651118 T>C and rs4846048 A>G polymorphisms among CRC patients and controls (Table 4).

DISCUSSION
Recently, CRC incidence and related mortality are increasing rapidly worldwide. The individual's susceptibility to CRC may be influenced by some environmental exposure and genetic factors. Recently, many case-control studies have been directed towards the association between MTHFR polymorphisms and CRC risk. However, the sample size of most studies was relatively small. Here, we attempt to assess the association between MTHFR tagging SNPs (rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A) and susceptibility of CRC. Our results indicated several MTHFR tagging polymorphisms could affect the risk of CRC.
MTHFR rs1801133 G>A polymorphism leads to an amino acid transformation (alanine→valine at 226 position of MTHFR protein). The role of MTHFR enzyme is remethylation of homocysteine to methionine. Polymorphisms in MTHFR gene are corrected with the deficiency of MTHFR enzyme activity. Compared with rs1801133 GG homozygote, MTHFR rs1801133 AA homozygote decreases 70% of the enzyme activity and MTHFR rs1801133 GA heterozygote loss 35% of enzymatic function. This transformation may increase the plasma homocysteine (Hcy) concentration and decreases the plasma folic acid concentration [26]. Recently, several meta-analyses indicated that MTHFR rs1801133 G>A polymorphism decreased the risk of CRC in Asians [27,28]. MTHFR rs1801133 G>A polymorphism locates on the NH2-terminal catalytic domain. In addition, MTHFR rs1801133 G>A polymorphism increases the availability of 5, 10-methylenetetrahydrofolate for DNA synthesis [29,30], which may partially explain the protective factor of CRC. In combination with this case-control study, our results evidence that G→A mutation in MTHFR rs1801133 increases the availability of 5, 10-methylenetetrahydrofolate for DNA synthesis; thus, this SNP may be a protective factor of CRC.   To the best of our knowledge, it was the first epidemiological study to explore the relationship of MTHFR rs4845882 G>A polymorphism with CRC risk. MTHFR rs4845882 G>A is linkage disequilibrium (LD) with rs1801131 (1298 A>C). Several metaanalyses suggested that MTHFR rs1801131 A>C polymorphism affected risk of CRC in Asians [31,32]. In the present study, we found MTHFR rs4845882 G>A polymorphism may be associated with the development of CRC. Since MTHFR rs4845882 G>A and rs1801131 A>C are in strong LD (r 2 = 0.935, http://gvs. gs.washington.edu/GVS147/LDpairwiseR2.jsp?GET_ TAYLORGRAM=1491876387794), the function of rs4845882 G>A may be affected by rs1801131 A>C.
A study reported that CC genotype of MTHFR rs9651118 conferred a reduced risk of breast cancer compared to TT genotype in a Chinese population [33]. Swartz et al. found MTHFR rs9651118 T>C polymorphism may be correlated with the decreased risk of lung cancer in Caucasians [34]. In our previous study, we found there was null association between MTHFR rs9651118 T>C polymorphism and esophageal squamous cell carcinoma in a Chinese Han population [21]. However, in this study, we found that rs9651118 CC genotype was relevant to increased risk of colon cancer. Rs9651118 T>C polymorphism is located on the intron of MTHFR gene. The function of this polymorphism is not well known. It was reported that MTHFR rs9651118 TT genotype elevated the level of Hcy compared with CC genotype [35]. In the future, more functional studies are required to identify the real biological effect of MTHFR rs9651118 T>C polymorphism on the etiology of CRC.
In a subgroup analysis by the region of CRC, MTHFR rs3753584 T>C polymorphism was associated with the risk of colon and rectum cancer. Although the function of MTHFR rs3753584 T>C polymorphism was not identified, a significantly increased risk of lung cancer was found for the variant allele carriers of this polymorphism, compared with individuals with wild homozygote [36]. In this study, we also found that C allele of MTHFR rs3753584 was probably correlated with an increased risk of colon and rectum cancer, which was consistent with the findings of those previous study.
Like all case-control studies, this study has some limitations. First, demographic variables and risk factor information only focused on age, sex, smoking and alcohol consumption. And other lifestyles were not collected, which might increase the possibility of confounding from environmental risk factors. Second, the source of non-cancer controls was hospital-based; which might not well represent the whole Chinese population. Third, in a stratified analysis by the region of CRC, the relatively small sample size may decrease the power of the results. Finally, these findings should be interpreted with very caution because the participants were only enrolled from Chinese Han population. Thus, the results may not permit extrapolation to other ethnicities.
In summary, the tagging polymorphisms in MTHFR gene (rs3753584 T>C, rs9651118 T>C and rs4845882 G>A) are associated with an increased risk of CRC. However, MTHFR rs1801133 G>A polymorphism confers a decreased risk to CRC. Our findings suggest that further validation studies are needed.

Data collection
Every participant was personally questioned and answered a questionnaire regarding age, sex, the status of cigarette smoking, and alcohol consumption. Participants who drink more than thrice/week for >6 months and smoke at least one cigarette/day over 1 year were considered positive. Clinical characteristics, such as pathological stage and tumor site, were obtained from the medical records (Table 1).

Selection of tagging SNPs
The tagging SNPs of MTHFR gene [30.4 Kbp spanning from 11780730 to 11811103 in chromosome 1 (upstream and downstream of the gene extending 5000 bases, respectively)] were analyzed and selected from the data of Chinese Han individuals in Beijing (CHB) via the HapMap Project (http://hapmap.ncbi.nlm.nih.gov/index. html.en) [20]. The detailed process and criterion were described previously [21]. The information of selected MTHFR tagging SNPs is presented in Table 2.

DNA extraction and genotyping
Ethylenediamine tetraacetic acid-anticoagulated intravenous blood was donated by every participant. Using the Promega DNA Blood Mini Kit (Promega, Madison, USA), genomic DNA was extracted from peripheral lymphocytes by the standard experimental protocol.
A custom-by-design 48-Plex SNPscan Kit (Genesky Biotechnologies Inc., Shanghai, China), double ligation and multiplex fluorescence PCR [22], was performed to identify the genotypes of MTHFR rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A polymorphisms as described in previous studies [23,24]. A total of 77 samples were randomly selected and were tested again for quality control. Based on 4% of duplicated samples in this study, the accordance rates were 100%.

Statistical analysis
We used SAS statistical software, version 9.4 (SAS Institute, Cary, NC) for data analysis and a P < 0.05 (two-tailed) was considered to be a statistical significance. The quantitative variables were expressed as means ± standard deviation (SD). Student's t-test was harnessed to evaluate the difference of age between CRC patients and controls. Additionally, we used χ 2 test to examine the differences in age, sex, smoking status, alcohol consumption and the frequencies of genotype between patients with CRC and controls. The Hardy-Weinberg equilibrium (HWE) equation was used to assess whether the proportion of MTHFR tagging SNPs genotypes obtained was in accordance with the expected value. An online calculator (http://ihg.gsf.de/cgi-bin/ hw/hwa1.pl) was harnessed to calculate the P value of HWE [25]. The relationship of MTHFR rs3753584 T>C, rs9651118 T>C, rs1801133 G>A, rs4846048 A>G and rs4845882 G>A genotypes with CRC risk was estimated by crude/adjusted odds ratios (ORs) and 95% confidence intervals (CIs). The power of the present study (α= 0.05) was evaluated by the Power and Sample Size Calculator (http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/ PowerSampleSize).