Background: DNA-methyltransferase (DNMT)-3A plays an important role in the development of embryogenesis and the generation of aberrant methylation in carcinogenesis. The aim of this study was to investigate the role of a DNMT3A promoter genetic variant on its transcriptional activity and to evaluate the association between DNMT3A gene polymorphism and the susceptibility to gastric cancer (GC) and oesophagus carcinoma (EC) in the Chinese population.
Methods: We selected one of the single nucleotide polymorphisms (SNPs) -448A>G in the DNMT3A promoter region and evaluated its effect on activity using a luciferase assay. -448A>G polymorphisms of DNMT3A were determined by polymerase chain reaction/restriction fragment length polymorphism and confirmed by sequencing. The distribution of -448A>G polymorphisms was detected in 208 GC patients and 346 healthy controls matched for age and gender. The distribution of -448A>G polymorphisms was also detected in 96 EC patients and matched 241 healthy controls. The association of -448A>G polymorphisms of DNMT3A and the risk of GC and EC was evaluated by stratified analysis according to the patient's age and gender.
Results: In a promoter assay, carriage of the -448 A allele showed a significantly higher promoter activity (> two fold) compared with the -448G allele (P < 0.001). The allele frequency of -448A among GC patients and controls was 32.9% versus 19.9%, respectively. Overall, we found that, compared with GG carriers, the DNMT3A -448AA homozygotes has a > six fold increased risk of GC. Stratification analysis showed that AA homozygotes have a more profound risk in the subgroups of individuals at the age range <or= 60 years in GC. However, individuals with -448AG and -448AA were not statistically significantly associated with an increased risk of EC compared with those carried the -448GG genotype.
Conclusions: The DNMT3A -448A>G polymorphism is a novel functional SNP and contributes to its genetic susceptibility to GC. -448A>G can be used as a stratification marker to predict an individual's susceptibility to GC, especially in the subgroups of individuals at the age range <or= 60 years. However, the relative distribution of -448A>G in EC can not be used as a prediction marker in order to evaluate an individual's susceptibility to EC.