Genetic polymorphisms of DNA double strand break gene Ku70 and gastric cancer in Taiwan

Gastric cancer is the fourth most common cancer over the world and affects approximately 900,000 individuals every year [1]. Although the identification of Helicobacter pylori has revolutionized the understanding of its epidemiology and pathogenesis, the initiation etiology and genomic contributing factors of gastric cancer are still largely unknown [2]. Human DNA repair mechanisms protect the genome from various insults caused by endogenous and environmental agents [3], and mutations or defects in the DNA repairing system are thought to be essential for tumorigenesis [4, 5]. Therefore, it is logical to suspect that genetic variants of DNA repair genes might contribute to gastric cancer pathogenesis.

DNA double strand breaks (DSBs) are repaired by the DNA DSB repair system [6, 7], which consists of two subpathways, homologous recombination (HR) and nonhomologous end-joining (NHEJ) [8]. In humans, NHEJ is the predominant repair system. At present, several proteins involved in the NHEJ pathway have been identified; namely, the ligase IV and its associated protein XRCC4, the three components of the DNA dependent protein kinase (DNA-PK) complex, Ku70, Ku80, and the catalytic subunit PKcs [9]. Genetic variation in DNA repair genes has been postulated as an important contributor to the aetiology of gastric cancer [10, 11]. However, there is seldom information regarding gastric cancer and DNA repair gene polymorphisms. As for NHEJ, genetic polymorphisms influence DNA repair capacity and confer predisposition to several cancers, including skin [12], breast [13‐15], bladder [16, 17], and oral cancers [18, 19].

In our previous study, we have found that one polymorphism of XRCC4 DSB repair gene is associated with gastric cancer susceptibility in Taiwan [20]. In this study, we assumed that the upstream gene Ku70 in NHEJ, like XRCC4, its polymorphisms may also contribute to gastric carcinogenesis. To test this hypothesis, we determined the genotypic frequency of four polymorphisms of the Ku70 gene at promoter T-991C (rs5751129), promoter G-57C (rs2267437), promoter A-31G (rs132770), and intron 3 (rs132774), using a polymerase chain reaction (PCR)-based restriction fragment length polymorphism (RFLP) method. To the best of our knowledge, this is the first study carried out to evaluate the contribution of Ku70 polymorphisms in gastric oncology all over the world.

There were no significant differences between both groups in their age, sex, and other cancer prevalence. The frequencies of the genotypes of the Ku70 promoter T-991C, promoter C-57G, promoter A-31G, and intron 3 polymorphisms in the gastric cancer and control groups are shown in Table 2. The ORs after adjusted by age and gender of the people carrying TC and CC genotypes were 2.41 (95% CI = 1.53-3.88) and 3.21 (95% CI = 0.96-9.41) respectively, compared to those carrying TT wild-type genotype. The P for trend was significant (P < 0.0001). In the dominant model (TC plus CC versus TT), the association between Ku70 promoter T-991C polymorphism and the risk for gastric cancer was also significant (adjusted OR = 2.48, 95% CI = 1.74-3.92). On the contrary, as for the Ku70 promoter C-57G, promoter A-31G, and intron 3 polymorphisms, the distributions of these polymorphisms were in Hardy-Weinberg equilibrium but there was no difference between gastric cancer and control groups in the distribution in the genotype frequency at these SNP sites (Table 2).

We have further evaluated the genetic association between gastric cancer in different age range and gender by performing the stratification analysis of the Ku70 promoter T-991C genotypes (Table 3). As for those people under the age of 55, the distribution of the Ku70 promoter T-991C genotypes was not significantly different between case and control groups. Interestingly, an increased risk was found in those people at the age of 55 or elder who carried TC genotype, compared with TT genotype (adjusted OR = 2.52, 95% CI = 1.37-4.68, P = 0.0139) (Table 3). When data were stratified by gender, the significantly increased risk was found in those male carrying TC genotype, compared with TT genotype (adjusted OR = 2.58, 95% CI = 1.33-4.47, P = 0.0085). There was no difference found in female (Table 3).

The present study is the first one to investigate the role of Ku70 gene polymorphisms, which has never been reported to be associated with gastric cancer risk. Our study revealed that the Ku70 promoter T-991C polymorphism was associated with the susceptibility to gastric cancer, but the C-57G, A-31G or intron 3 polymorphisms were not (Table 2), especially in those at the age of 55 or elder, and in male (Table 3). In 2003, Fu et al have reported that, Ku70 rs2267437 polymorphism (which was called C-61G in their paper) was associated with breast cancer risk [21]. It is reasonable that we did not find a significant association of this SNP with gastric cancer as they did in this SNP with breast cancer in Taiwan population. Instead, we found that the genotype of Ku70 rs2267437 was associated with gastric cacner. Since Ku70 rs2267437 located in the promoter region, its different genotypes may determine the differential expression level of Ku70 protein in the gastric tissue. Thus, the genotypes of Ku70 promoter T-991C (rs5751129) may determine the expression levels of Ku70 in different stages, regulate the DNA repair capacity and overall genome stability, and play a role in gastric carcinogenesis. Supporting the idea above, Hu et al reported that the expression of TERT and Ku70 proteins was significantly up-regulated in both precancerous lesion and gastric cancer tissues and was significantly higher in gastric cancer tissues than that in precancerous lesions [22]. In the future, the correlation between Ku70 rs5751129 genotype and phenotype still needs further investigation. In addition to be associated with oral cancer [18], Ku70 rs5751129 polymorphism was also found to be associated with two-side pterygium pathology, which are caused by an uncontrolled cell proliferation like that of a tumor [23].

We have taken into consideration and tried our best to conquer the limitations of this study. First, for the possibility that some findings of further sub-grouping are attributed to chance because of the limited numbers of patients, we just provided the main finding with these 136 gastric patients and 560 (more than four-fold of the number of the patients) non-cancer controls, which is very convincing and reliable with multiple-checked significances (P < 0.05 both in genotype and allelic frequencies, and those obvious odds ratios and 95% CIs). Second, the false positive findings are statistically explained by hidden population stratification. But in this study, it seems unlikely that population stratification is a major concern, for all the cases and the controls recruited in this study were drawn from the same Taiwanese ethnic group and the Taiwanese population has relatively homogenous genetic background [24]. Therefore, the potential confounding effect of population stratification for genotyping data is not a major concern. Third, the possible selection bias was taken into consideration and reduced to a lowest level by frequency matching on age and gender between the cases and controls. Forth, from another angle, the frequencies of Ku70 polymorphisms variant alleles were similar to those reported in the NCBI website in the Asian population studies, for example C allele frequencies of Ku70 promoter T-991C are 5.7% in our control group and 4.2~8.9% for Asian populations in NCBI, which suggest no selection bias for the subject's enrolments in terms of genotypes. Therefore, the need for the present results of Ku70 promoter T-991C to be verified in further larger gastric patient population is not so urgent. Of course, further studies in larger population and other functional repair assays should be performed in the future for further findings, particularly those among subgroups by clinical and environmental factors, and clarify the mechanisms involved.

In this study, the genotype distribution of the C allele at Ku70 promoter T-991C was significantly higher in the gastric cancer group (12.5%) than in the control group (5.7%). It was also found that patients carrying heterozygous TC for Ku70 promoter T-991C had a 2.41-fold higher risk of gastric cancer (Table 2). Combination of heterozygous and homozygous (TC or CC) was almost at the same level (OR = 2.48) for gastric cancer risk (Table 2). All these data suggested that the C allele at Ku70 promoter T-991C was a novel and important biomarker for gastric carcinogenesis, and as long as -991C was detected, the carriers were more susceptible to gastric cancer. As for the Ku70 promoter C-57G, A-31G or intron 3 polymorphisms, our results indicated that their genetic differences were not associated with gastric cancer risk in Taiwanese population (Table 2).

In conclusion, this is the first report to investigate the association between Ku70 gene polymorphisms and gastric cancer. Our findings suggested that Ku70 promoter T-991C, but not Ku70 promoter C-57G, promoter A-31G or intron 3 polymorphisms, was associated with higher susceptibility to gastric cancer. The Ku70 promoter T-991C polymorphism might become a potential biomarker for the gastric oncology prediction and this paper may also provide a valuable insight into the gastric carcinogenesis.