The COX-2 -765 G>C Polymorphism is Associated with Increased Risk of Gastric Carcinogenesis in the Chinese Hui Ethnic Population

Gastric cancer is a major health problem, being the most common malignancy in the world with an estimated one million new cases every year (Hartgrink et al., 2009). The high morbidity areas are found in Asia, especially China. It is reported that the standardized mortality rate of gastric cancer in Hui population is higher than the averag elever of China (The Data-Center of China Public Health Science). In China, Hui minority ethnic populations are descendant of Mongols and Muslims immigrants. Through interracial breeding between the Han, the Mongols, and the Uygur, they developed the Hui nationality. There are many concentrated Hui communities in Gansu province where there are high incidence areas of gastric cancer. Our previous study suggests COX-2 plays an important role in carcinogenesis and its polymorphisms, including a new single nucleotide polymorphism (SNP) are associated with susceptibility to gastric cancerin the Han population in Gansu Province (Li et al., 2006; Li et al., 2010; Zhu et al., 2010). To the best of our knowledge, there are no other reports on the SNPs of COX-2 and its association with increased gastric cancer risk in the Chinese Hui population. In this study, we investigated whether the COX-2-765G>C polymorphism contributes to gastric cancer in the Chinese Hui ethnic population.


Introduction
Gastric cancer is a major health problem, being the most common malignancy in the world with an estimated one million new cases every year (Hartgrink et al., 2009). The high morbidity areas are found in Asia, especially China. It is reported that the standardized mortality rate of gastric cancer in Hui population is higher than the averag elever of China (The Data-Center of China Public Health Science). In China, Hui minority ethnic populations are descendant of Mongols and Muslims immigrants. Through interracial breeding between the Han, the Mongols, and the Uygur, they developed the Hui nationality. There are many concentrated Hui communities in Gansu province where there are high incidence areas of gastric cancer. Our previous study suggests COX-2 plays an important role in carcinogenesis and its polymorphisms, including a new single nucleotide polymorphism (SNP) are associated with susceptibility to gastric cancerin the Han population in Gansu Province (Li et al., 2006;Li et al., 2010;Zhu et al., 2010). To the best of our knowledge, there are no other reports on the SNPs of COX-2 and its association with increased gastric cancer risk in the Chinese Hui population. In this study, we investigated whether the COX-2-765G>C polymorphism contributes to gastric cancer in the Chinese Hui ethnic population.

Study population
Subjects investigated were Muslim inhabitants, who lived in Linxia Hui Autonomous Prefecture for more than 20 years with three generations of non-ethnic marriage. The case control study was undertaken between October 2010 to November 2012, and consisted of 100 gastric cancer patients, 102 precancerous lesion patients and 105 control subjects. Participants were recruited from2A grade hospitals in Gansu Province. The diagnosis of gastric cancer and precancerous lesion were on the basis of clinical, endoscopical and histopathological examinations. The 105 control subjects were confirmed to be free of the disease by a health examination, involving endoscopy or upper gastrointestinal investigation.Subjects were informed of the detailed study protocol,and signed consent forms, and the study was approvedby local ethics committees.
A questionnaire collected information on (a) demographic factors, such as age, sex, ethnic, native place; (b) medical history of digestive disease and family history of gastric cancer. Subject with at least one firstdegree relative (a parent or sibling) or two second-degree relatives (grandparents, aunts, or uncles) with gastric cancer (GC) were considered to have a positive family history. (c)frequency of eating pickles (at least 5 times a week or more).
Each subject donated 3 mL peripheralvein blood in EDTA-K2 anticoagulative tube for DNA extraction and 2 mL serum for H. pylori infection test, and all specimens were kept frozen at -80℃.

Sample DNA extraction
Blood samples were collected with a standard venipuncture technique using EDTA containing tubes. Genomic DNA was extracted using EZ Spin Column Genetic DNA Isolation Kit (BBI, Canada) following the manufacturer's instruction. The integrity and purity were determined by using agarose gel electrophoresis and a spectrophotometer respectively. The final concentration of each DNA sample was adjusted to approximately 20 ng/mL and stored at -80℃ until use.

Detection of H. pylori infection
Presence and type of H. pylori infection were tested using a H. pylori antibody Immunoblotting Kit (Blot Biotech,Shenzhen, China) following the manufacturer's instructions. Type I H. pylori infection was defined when either CagA or VacA was positive, or both were positive. TypeⅡ H. pylori infection was defined if ureases (UreA/ UreB)were positive. Patients were defined as H. pylori negative if CagA, VacA and ureases were negative.

Genotyping of COX-2-765G>C polymorphism
COX-2-765G>C polymorphismwas determined by pyrosequencing technology. PCR primers and sequencing primer were designed by TakaraBioTechnology (Dalian) Corporation Limited. The sequences of PCR primers were as follows: forward primer was TTTATATTGGTGACCCGTGGAGC, and reverse primer was biotinylated-CCTTCACCCCCTCCTTGTTT. The sequencing primer was GGAGAATTTACCTTTCCC. All primers were validatedby basic local alignment search tool (BLAST) homology searches (http://www.ncbi.nlm.nih. gov).

Quality control
Twenty percent of samples from patients and controlswere regenotyped by other laboratory personnel, and results showed 100% similarity in both of the conditions. No discrepancy was found after sequencing randomly selecting 10% of the samples.

Statistical analysis
The Hardy-Weinberg equilibrium equation was used to determine whether the proportion of each genotype obtained was in agreement with expected values as calculated from allele frequencies. Student and Chi-square test analyses was used to compare categorical variables and genotype frequencies between GC, GPL and controls, using a 5% level of significance. Odds ratios (ORs) and 95% confidence intervals (CIs) calculated bynonconditional logistic regression analysiswas performed to analyze the association betweenthe genotypes of COX-2 and the risk factors ofGC. Multivariate logistic regression was used to analyze all risk factors with GC and GPL.
Allele frequency distribution was stratified by age, sex, GC family history, eating pickles history and H. pylori infection, and interactions were considered in the model. All data analysis was performed using the computer software SPSS 13.0 (SPSSInc., Chicago, IL, USA).

Demographic and clinical characteristics
The frequency distribution of sex, age, H. pylori infection, family history and eating pickle vegetables in subjects with gastric cancer, GPL and controls is presented in Table 1. Rates of H. pylori infection and eating pickled vegetables were significantly higher in gastric cancer and GPL participants compared to controls whereas no significant differences were seen between gastric cancer and GPL. The rates of gastric cancer family history was significantly higher in gastric cancer and GPL participants compared to controls. The was nosignificant differences in age and sex between gastric cancer and GPL participantscompared to controls.

Stratification analysis of COX-2 -765 G>C polymorphism
The risk of GPL or gastric cancer related to COX-2 -765G>C genotypes were further examined with stratificationby H. pylori infection, family history and eating pickled vegetables (Table 3). The intake of pickled vegetables was the most importantrisk factor. The presence of the -765 C allele in subjects with eating pickle vegetables conferred a 3-fold risk of gastric cancer compared with GG genotypes. However, individuals carried C allele appeared to have no association between  GPL and controls (OR=1.830,p=0.261). In H. pylori positive groups, the risk of gastric cancer was significantly higher in subjects who carried C allele compared with controls (OR=2.121, 95%CI=1.020-4.498, p=0.048). While there was no association between GPL and controls among H. pylori positive participants (OR=1.658,p=0.190). There was no increased risk amongst participants with a family history of gastric cancer in all three groups.

Discussion
Living habits of the Hui minority have a deep Islamic influence. As a result smoking and drinking, the two major risk factors for gastric cancer (Yaegashi et al., 2014), are forbidden in their daily lives. However, the Hui minority in Northwest China like to eating pickled vegetables. Our study demonstrated the intake of pickle vegetables was the highest risk factor for gastric cancer in the Hui population which has been shown in other studies (Ren et al., 2012). H. pylori are also a proven risk factor for development of gastric cancer (Uemura et al., 2001).
In this study, we found H. pylori infection increased the risk of gastric diseases, but there was no different between gastric cancer and GPL groups. This finding suggests that H. pylori infection didn't aggravate inflammation reactions, and was only an early event during of development of gastric cancer. Some studies investigating the effect of H. pylori eradication have shown contradictory results for reversibility of precursor lesions and reduction of gastric cancer rate (Kuipers and Sipponen., 2006). Therefore, H. pylori eradication was a good prophylactic method but clinical way.
Gene polymorphisms vary substantially between ethnic groups. As a results it is important to identify the susceptible genes in different ethnicity. COX-2 is a rate-limiting enzyme that catalyzes the formation of prostaglandins (PG) from arachidonic acid (AA). COX-2 -765 G>C is one of the most extensively studied polymorphisms. It has been reported that the -765 CC genotype does not exist in Han population (Liu et al., 2006;Tan et al., 2007). However, we found the frequency of genotype was about 5% -765 G>C increased the risk of gastric cancer in the Han population. Pereira et al. found similar results amongst Portuguese populations (Pereira et al., 2006). However, Sitarz et al. findings suggest the polymorphism plays a protective role in the development of gastric cancer (Sitarz et al., 2008). Interestingly, they founded there was no correlation between the presence of the C allele and a difference in COX-2 expression. In addition, this polymorphism showed no associations with gastric cancer and precancerous lesions risk in Chinese Han people.
[9] Therefore, it might indicate that the -765 G>C polymorphism plays a role in carcinogenic processes in specific populations.
The present study has some limitations. The sample size was relatively small, although it was sufficient to detect moderate associations. In addition, the risk of GC related to the COX-2 SNP might be influenced by use of non-steroidal anti-inflammatory drugs. Those factors might interact with COX-2 genotype or act as potential confounder in the analysis. Unfortunately, information on this factor in the present study was unavailable. It would be important in the future to investigate the interaction between the genotype and the expression in vitro.
In summary, this study provides evidence that the COX-2-765 G>C polymorphism increased the genetic susceptibility of gastric cancer in Hui population. We also found an interaction among H. pylori infection, eating pickle vegetablesand the COX-2 -765 C carriers.