Effect of DNMT3A polymorphisms on CpG island hypermethylation in gastric mucosa.

BACKGROUND: CpG methylation of tumor suppressor genes occurs in the early stage of carcinogenesis. Detecting risk factors for aberrant CpG methylation is clinically important for predicting cancer development. DNA methyltransferase (DNMT) 3a is considered to play critical roles in the DNA methylation process during pathogenesis. In this study, we evaluated the association between DNMT3A polymorphisms (rs6733868 and rs13428812) and CpG methylation status in non-cancerous gastric mucosa. METHODS: We determined the DNMT3A genotype and CpG methylation status of 4 genes (p14ARF, p16INK4a, DAPK, and CDH1) in 510 subjects without gastric cancer. Helicobacter pylori (HP) infection status was determined by the rapid urease test, urea breath test, speculum examination, or serum antibody test. We determined the DNMT3A genotype using polymerase chain reaction single-strand conformation polymorphism (PCR-SSCP). CpG methylation status was determined by methylation-specific polymerase chain reaction (MSP). When the methylated band was stronger than 10 ng/µL according to the DNA marker, we judged CpG island hypermethylation (CIHM) to be present. Associations between genotypes and susceptibilities were assessed by logistic regression analysis. RESULTS: The minor allele frequencies of both polymorphisms (rs6733868 and rs13428812) were lower in the CpG methylated groups of each of the 4 genes (p14ARF, p16INK4a, DAPK, and CDH1). Using a dominant genetic model, rs6733868 was significantly associated with the hypermethylation of each gene, whereas rs13428812 was associated with the methylation of 3 genes (all except p14ARF). When low-CIHM was defined as 1 or 2 CpG islands methylated and high-CIHM was defined as 3 or more CpG islands methylated, carrying the minor allele of rs6733868 was associated with both decreased low- and high-CIHM, and that of rs13428812 also was associated with a decrease. Comparing low-CIHM with high-CIHM, carrying the minor alleles of rs6733868 or rs13428812 was related to decreased susceptibility to high-CIHM. In HP-infected subjects, carrying the minor alleles of rs6733868 or rs13428812 had a significantly greater association with decreased susceptibility to high-CIHM. CONCLUSIONS: Our study indicates that polymorphisms of DNMT3A are associated with the accumulation of gene methylation in gastric mucosa. Carrying the minor alleles of rs6733868 or rs13428812 inhibits aberrant gene methylations, which are typically enhanced by HP infection.

Influence of MIF polymorphisms on CpG island hyper-methylation of CDKN2A in the patients with ulcerative colitis.

BACKGROUND: CDKN2A hypermethylation is among the major events associated with carcinogenesis and is also observed in non-neoplastic colonic mucosa in patients with ulcerative colitis (UC). Macrophage migration inhibitory factor (MIF) plays a crucial role in promoting gastrointestinal inflammation characteristic of UC. The aim of this study is to explore associations between CDKN2A methylation status and MIF polymorphisms (rs755622 and rs5844572). METHODS: One hundred and fifty-nine patients diagnosed with UC were enrolled in this study. The methylation status of p14ARF and p16INK4a was determined by MSP; MIF genotypes were identified by PCR-SSCP. RESULTS: We found no differences with respect to mean age, gender, clinical type (chronic continuous or relapse/remitting), or extent of disease among the patients with methylated and unmethylated p14ARF or p16INK4a. Carrying the rs755622 C allele indicated a significantly higher risk for p14ARF methylation (odds ratio (OR), 2.16; 95% confidence interval (CI), 1.08-4.32; p = 0.030); similarly, carrying the rs5844572 7-repeat allele indicated a significantly higher risk for p16INK4a methylation (OR, 2.57; 95% CI, 1.26-5.24; p = 0.0094) after an adjusted regression analysis. The carriers of the rs755662 C allele or the rs5844572 7-repeat allele were both at a significantly higher risk for methylation of both p14ARF and p16INK4a when compared to the cohort in which neither of the genes were methylated (OR, 2.70; 95% CI, 1.22-6.01; p = 0.015 and OR, 2.87; 95% CI, 1.25-6.62; p = 0.013, respectively). Additionally, carrying rs755622 C allele was significantly associated with CIHM in chronic continuous of clinical type and total colitis (OR, 25.9; 95% CI, 2.55-262.6; p = 0.0059 and OR, 4.38; 95% CI, 1.12-17.2; p = 0.034, respectively), and carrying 7-repeat allele of rs5844572 was significantly associated in chronic continuous type (OR, 14.5; 95%CI, 1.46-144.3; p = 0.022). CONCLUSIONS: Taken together, our findings suggest that MIF genotypes associated with inflammation may also be involved in promoting carcinogenesis via CDKN2A hypermethylation in patients diagnosed with UC.

MAFK Polymorphisms Located in 3'-UTR are Associated with Severity of Atrophy and CDKN2A Methylation Status in the Gastric Mucosa.

Objective: This study aimed to clarify the association of MAFK polymorphisms (rs4268033, rs3735656, and rs10226620) with the degree of gastric mucosal atrophy and CDKN2A CpG methylation status. Methods: A total of 491 subjects were enrolled in this study. Genotypes and methylation status were determined by polymerase chain reaction (PCR)-single-stranded conformation polymorphism and methylation-specific PCR (Fujita Health University, HM18-094). Methods: A total of 491 subjects were enrolled in this study. Genotypes and methylation status were determined by polymerase chain reaction (PCR)-single-stranded conformation polymorphism and methylation-specific PCR (Fujita Health University, HM18-094). Results: Either rs3735656 or rs10226620, located in the 3'-UTR of MAFK, was significantly associated with the severity of gastric mucosal atrophy using a dominant genetic model (odds ratio [OR], 2.10; p = 0.0012, and OR, 1.98; p = 0.0027, respectively). However, using a recessive genetic model, no significant association was found between three polymorphisms and gastric mucosal atrophy. The serum pepsinogen I/II ratio was significantly lower in subjects with minor alleles of rs3735656 and rs10226620 than in subjects with the wild homozygous allele (p = 0.018 and 0.013, respectively). In a subgroup including 400 of the 491 subjects, the CpG of p14ARF and p16 INK4a were methylated in 132 and 112 subjects, respectively. Fifty subjects had both CpG methylations and 206 subjects had neither methylation. When comparing the groups with both and neither methylations, there were no significant associations between three polymorphisms and CDKN2A methylation using a dominant genetic model. However, all polymorphisms investigated in this study (rs4268033, rs3735656, and rs10226620) were significantly associated with CDKN2A methylation in a recessive genetic model (OR, 3.58; p = 0.0071, OR, 4.49; p = 0.0004, and OR, 3.45; p = 0.0027, respectively). Conclusions: Our results indicate that carrying the minor allele of the MAFK polymorphisms, particularly when they are located in the 3'-UTR, has a high risk for the severity of gastric mucosal atrophy; furthermore, CDKN2A CpG methylation may develop in subjects with homozygous minor allele of these polymorphisms.