Introduction

Esophageal cancer (EC) is the sixth most common cancer with an estimated 482,300 new cases and 406,800 deaths occurred in 2008 worldwide [1]. Incidence rates of EC vary internationally by almost 16-fold, with 22.14 per 10,000 in China in 2009 [1, 2]. Every year there are about 250,000 EC cases diagnosed in China, accounting for half of the global cases [3, 4]. The mortality rate for EC patients is very high, and 5-year survival rate accounts only 12.3 % [5]. Esophageal squamous cell carcinoma (ESCC) is the most frequent subtype of EC and a lethal malignancy associated with low survival rates. In the highest risk area such as Iran and China, 90 % of cases are ESCC histologically [6].

Many epidemiological studies indicated that the risk factors for ESCC involved poor nutritional status, low fruits and vegetables consumption, smoking, heavy alcohol use and drinking hot beverages [7–11]. However, only a fraction of individuals who are exposed in risk factors ultimately develop EC, suggesting that genetic factors such as single nucleotide polymorphisms (SNPs) may play an important role in developing EC [12–14]. Methylenetetrahydrofolate reductase (MTHFR), a key enzyme of methylation, is located on 1p36.3. It is a methyl donor and catalyzes reduction of 5,10-methylene-tetrahydrofolate to 5-methyltetra hydrofolate [15]. Functional polymorphisms of MTHFR may lead to attenuate of 5-methyl tetrahydrofolic acid (THFA) to induce a decrease in the conversion of homocysteine to methionine, which could result in a carcinogenesis [16]. There were more than 20 kinds of genetic polymorphisms of MTHFR; some non-synonymous SNPs were the most studies genetic polymorphisms. An impact of these polymorphisms on ECs has been performed. However, the association between the MTHFR SNPs and EC was inconsistence. According to the biological significance of MTHFR, it is possible that functional SNPs in the gene may contribute to the development of ESCC. The aim of this study was to assess the association between MTHFR rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C genotypes and ESCC susceptibility. In a hospital-based case–control study, we conducted genotyping analyses for the five SNPs with 629 ESCC cases and 686 controls in Chinese Han population.

Materials and methods

Subjects

In total, 629 from unrelated Chinese Han ESCC patients and 686 cancer-free subjects were consecutively recruited from the Affiliated People’s Hospital of Jiangsu University and Affiliated Hospital of Jiangsu University (Jiangsu Province, China) between October 2008 and December 2010. Diagnoses of all ESCC cases were confirmed by postoperative pathologic means. The ESCC patients who formerly had a history of cancer or autoimmune diseases, or had undergone radiotherapy or chemotherapy were removed. Ethnicity (Chinese), frequency of sex and average age (±5 years) of the 686 controls were group matched to cases. The majority of the control individuals were admitted to the two hospitals for the cure of trauma. At recruitment, this study was approved by the Institutional Review Board of Jiangsu University (Zhenjiang, China) and written informed consent was obtained from each subject. Two experienced research doctors were assigned to administer a structured questionnaire to each subject. The information collected included demographic data (e.g., age and gender) and related risk factors (including tobacco use and alcohol consumption). After completed the in-person interview, each participant donated 2-ml sample of peripheral venous blood. Subjects who smoked at least one cigarette per day over 1 year were defined as “smokers,” and those who drinked no less than three times a week for >6 months were considered to be “alcohol drinkers”.

DNA extraction, SNP selection and genotyping

Ethylenediamine tetraacetic acid (EDTA)-anticoagulated peripheral venous blood sample was collected from each subject, and genomic DNA was isolated using the QIAamp DNA Blood Mini Kit (Qiagen, Berlin, Germany). MTHFR tagging SNPs were chosen based on the HapMap Project (http://www.hapmap.org/, phase II Nov08, on NCBI B36 assembly, dbSNP b126; population: Chinese Han population, CHB; minor allele frequency (MAF) ≥0.05, Hardy–Weinberg equilibrium (HWE) P ≥ 0.05 and call rate ≥95 %) on the basis of pairwise linkage disequilibrium (LD) r 2 threshold of 0.8. with Haploview 4.2 software [17]. Genotypes of MTHFR at the rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C sites were analyzed by using the ligation detection reaction (LDR) method [18]. Technical support was come from the Shanghai Biowing Applied Biotechnology Company. For quality control, 160 (12.17 %) randomly selected samples were repeated analysis by LDR method with high DNA quality and the accordance rates were 100 %.

Statistical analysis

Chi-square test (χ 2) was used to examine the differences in the distributions of demographic characteristics, selected variables and genotypes between cases and controls. The associations between MTHFR rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C genotypes and the risk of ESCC were evaluated by odds ratios (ORs) and their 95 % confidence intervals (CIs) using unconditional logistic regression analyses for crude ORs and adjusted ORs when appropriate. An internet-based HWE calculator (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl) was used to assess the HWE among the control subjects. Statistical analysis was performed by SAS 9.1.3 software (SAS Institute, Cary, NC). Statistical significance was defined as P < 0.05 (two-tailed) for all statistical analyses.

Results

Characteristics of the study population

Characteristic of all subjects is presented in Table 1. There were no significant differences between patients and controls in terms of age distributions and sex distributions (P = 0.155 and P = 0.185, respectively), which indicated the matching was adequate. However, significant difference was detected on drinking status and smoking rate between patients and controls (P < 0.001). The primary information of five tagging SNPs of MTHFR was included in Table 2. For these five SNPs, the genotyping success rate ranged from 96.43 to 98.48 % in all 1,315 samples. In this study, MAF of control subjects was similar to that for Chinese in database for all these five SNPs (Table 2). The genotypic frequencies for MTHFR rs3753584 A>G, rs4846048 A>G and rs9651118 T>C polymorphisms among controls were in HWE (P = 0.648, P = 0.312 and P = 0.344) except MTHFR rs1801133 C>T and rs4845882 G>A (P = 0.045 and P = 0.029) (Table 2).

Table 1 Distribution of selected demographic variables and risk factors in ESCC case subjects and control subjects
Full size table
Table 2 Primary information for MTHFR rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C polymorphisms
Full size table

MTHFR polymorphisms and the risk of ESCC

The genotype distributions of MTHFR rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C in cases and controls were presented in Table 3. In the single locus analyses, the genotype frequencies of MTHFR rs1801133 C>T were 30.57 % (CC), 46.83 % (CT) and 22.60 % (TT) in the case subjects and 28.79 % (CC), 53.29 % (CT) and 17.92 % (TT) in the control subjects, and the difference was statistically significant (P = 0.04). When the MTHFR rs1801133 CC homozygote genotype was used as the reference group, the CT genotype was not associated with the risk of ESCC (CT vs. CC: OR 0.83, 95 % CI 0.64–1.07, P = 0.147) and the TT genotype was not associated with the risk of ESCC (TT vs. CC: OR 1.19, 95 % CI 0.86–1.63, P = 0.289). In the recessive model, when the MTHFR rs1801133 CC/CT genotypes were used as the reference group, the TT homozygote genotype was associated with the risk of ESCC (TT vs. CC/CT: OR 1.34, 95 % CI 1.02–1.76, P = 0.038). In the dominant model, the MTHFR rs1801133 CT/TT variants were not associated with the risk of ESCC, compared with the MTHFR rs1801133 CC genotype (CT/TT vs. CC: OR 0.92, 95 % CI 0.72–1.17, P = 0.488) (Table 3). After adjusting for age, gender, smoking and drinking status, a statistically increased risk of ESCC was also observed in the recessive model (TT vs. CC/CT: adjusted OR 1.39, 95 % CI 1.05–1.84, P = 0.021) (Table 3).

Table 3 Logistic regression analyses of associations between MTHFR polymorphisms and the risk of ESCC
Full size table

The genotype frequencies of MTHFR rs4845882 G>A were 71.08 % (GG), 26.98 % (GA) and 1.94 % (AA) in the case subjects and 68.64 % (GG), 26.92 % (GA) and 4.44 % (AA) in the controls, and the difference was statistically significant (P = 0.039). When the MTHFR rs4845882 GG homozygote genotype was used as the reference group, the GA genotype was not associated with the risk of ESCC (GA vs. GG: OR 0.97, 95 % CI 0.76–1.24, P = 0.974). The AA genotype was associated with the decreased risk of ESCC (AA vs. GG: OR 0.42, 95 % CI 0.21–0.84, P = 0.013). In the recessive model, when the MTHFR rs4845882 GG/GA genotypes were used as the reference group, the AA homozygote genotype was associated with the decreased risk of ESCC (AA vs. GG/GA: OR 0.43, 95 % CI 0.22–0.84, P = 0.014). In the dominant model, the MTHFR rs4845882 GA/AA variants were not associated with the risk of ESCC, compared with the MTHFR rs1801133 GG genotype (GA/AA vs. GG: OR 0.89, 95 % CI 0.70–1.13, P = 0.340) (Table 3). After adjusting for age, gender, smoking and drinking status, a statistically decreased risk of ESCC was observed both in the homozygote comparing model (AA vs. GG: adjusted OR 0.41, 95 % CI 0.20–0.82, P = 0.011) and in recessive model (AA vs. GG/GA: adjusted OR 0.41, 95 % CI 0.21–0.82, P = 0.012) (Table 3).

The genotype frequencies of MTHFR rs4846048 A>G were 84.12 % (AA), 15.38 % (AG) and 0.49 % (GG) in the cases and 80.44 % (AA), 18.09 % (AG) and 1.47 % (GG) in the controls, and the difference was not statistically significant (P = 0.082). Logistic regression analyses revealed that the MTHFR rs4846048 A>G polymorphisms was not associated with the risk of ESCC. After adjusting for age, gender, smoking and drinking status, the results showed that the MTHFR rs4846048 A>G polymorphisms were associated with the decreased risk of ESCC in homozygote comparing model (GG vs. AA: adjusted OR 0.26, 95 % CI 0.07–0.96, P = 0.044) and in recessive model (GG vs. AA/AG: adjusted OR 0.27, 95 % CI 0.07–0.99, P = 0.048) (Table 3).

MTHFR rs3753584 A>G and MTHFR rs9651118 T>C SNPs did not achieved significant differences in the genotype distributions between patients and controls (P = 0.871 and P = 0.694) (Table 3). Logistic regression analyses revealed that the MTHFR rs3753584 A>G and MTHFR rs9651118 T>C polymorphisms were not associated with the risk of ESCC (Table 3).

Discussion

In this hospital-based case–control study, we investigated the associations of MTHFR rs1801133 C>T, rs3753584 A>G, rs4845882 G>A, rs4846048 A>G and rs9651118 T>C SNPs with the risk of ESCC in Chinese Han population. Our results revealed that MTHFR rs1801133 C>T was associated with the increased risk of ESCC, while MTHFR rs4845882 G>A and rs4846048 A>G were associated with the decreased risk of ESCC.

ESCC is one of the most general cancers worldwide. Of late, more and more evidence has demonstrated that genetic components, environmental factors, gene–gene and gene–environment interactions play pivotal roles in ESCC development and progression [19, 20]. Recent studies indicated that susceptibility of ESCC could be modulated by MTHFR SNPs; however, the results were inconsistent. In view of these investigations, we chose five tagging sites to evaluate their roles in ESCC. The metabolic pathway is considered to be very important in keeping normal DNA methylation, DNA synthesis and DNA repair [21]. The MTHFR genetic variation resulted in 5-methyltetrahydrofolate reduction and homocysteine amassing in the body, which made the methyl donor of the methionine dys-synthesis, eventually led to hypomethylation of DNA, decreasing the activity of the enzyme and increasing a number of cancers susceptibility [22].

MTHFR rs1801133 C>T (MTHFR C667T) mutation results in an alanine to valine substitution and a reduction in enzyme activity [23]. Recently, several investigations indicated that MTHFR rs1801133 C>T was associated with EC in Chinese population [24, 25]. In combination with our study, our outcomes showed that the mutation in MTHFR rs1801133, causing reduction in enzyme activity, DNA methylation and diminished DNA synthesis/repair, might dramatically increase the susceptibility of ESCC.

To the best of our knowledge, it was the first case–control study to assess the association between MTHFR rs4845882 G>A genotype and the susceptibility of cancer. Rs4845882 G>A and MTHFR tagging SNP rs1801131 (1298 A>C) are almost complete LD. A meta-analysis demonstrated that MTHFR rs1801131 acted as a protective role in the carcinogenesis of hepatocellular carcinoma [26]. Another study indicated that MTHFR rs1801131 A>C was not associated with cervical cancer risk [27]. In this study, we found MTHFR rs4845882 G>A played a protective role in the carcinogenesis of ESCC.

MTHFR rs4846048 is located at 463 bp upper stream of a polyadenylation signal position [28]. There is one polyadenylation signal sequence in most eukaryotic genes; however, sometimes multiple such sequences are existed, and the alternatively polyadenylated mRNAs are usually conditioned by translation efficiency and tissue-specific expression [29]. Therefore, the significant association of MTHFR rs4846048 with ESCC may indicate that the SNP of this polyadenylation signal site acts an crucial role in the occur and development progress of ESCC.

Several limitations should be acknowledged. First, all subjects were recruited from two hospitals and might not fully represent the general Chinese population. It might result in unavoidable selection bias. Second, the moderate sample sizes in our study restricted statistical power to indicate a more reliable effect. Third, MTHFR rs1801133 C>T and rs4845882 G>A genetic distribution of controls were deviated from HWE. Fourth, in current study, we do not have the data for the level of folate intake in individuals to further conduct examination of the gene–nutrient interaction. Further, better designed studies should be carried out to verify these results. Finally, since the detailed dataset on cancer metastasis and survival information of each subject was not available till now, the role of MTHFR polymorphisms in ESCC progression and prognosis could not be conducted further analyses.

In conclusion, this study indicates a significant association between the MTHFR rs1801133 C>T, rs4845882 G>A and rs4846048 A>G SNPs and risk of ESCC in Han Chinese population. Future, larger sample size studies on the role of the MTHFR SNPs—nutrient (the level of folate intake) interaction—are needed to verify these results.