A germline variant N375S in MET and gastric cancer susceptibility in a Chinese population

MET tyrosine kinase and its ligand, hepatocyte growth factor (HGF), play a pivotal role in the activties of tumor cells. A germline missense variant in exon 2 of the MET gene, N375S (rs33917957 A>G), may alter the binding affinity of MET for HGF and thus modify the risk of tumorigenesis. In this study, we performed a case-control study to assess the association between N375S and gastric cancer risk in 1,681 gastric cancer cases and 1,858 cancer-free controls. Logistic regression analysis was applied to estimate crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the associations between genotypes and gastric cancer risk. We found that MET N375S variant genotypes (NS/SS) were associated with a significantly decreased risk of gastric cancer (OR = 0.78, 95% CI = 0.63-0.96, P = 0.021) compared with the wildtype homozygote (NN). The finding indicates that this germline variant in MET may decrease gastric cancer susceptibility in Han Chinese.


INTRODUCTION
Although the incidence of gastric cancer decreased in the past decades, gastric cancer remains the fourth most common cancer and the second leading cause of cancer related death worldwide. About half of global gastric cancer cases occur in China, and the incidence and mortality rates (per 100,000 persons) were 42.4 and 28.1 for men and 18.3 and 13.0 for women, respectively. [1] Helicobacter pylori (H. pylori) infection has been established as a risk factor in stomach carcinogenesis [2] . Genetic factors may also play a critical role in the development of gastric cancer [3] .
MET, a tyrosine kinase, and its ligand, hepatocyte growth factor (HGF), play a pivotal role in tumor cell proliferation, survival, and metastasis [4,5] . Kunlyasu et al. [6] reported that MET was overexpressed in about half of gastric cancer tissues. Further investigations revealed that MET was more frequently overexpressed in diffuse gastric cancer [7] . MET overexpression was correlated with gastric tumor invasion and lymph node and liver metastasis [8,9] . In addition to aberrant upregulation, mutations of MET were also detected in gastric cancer and they may result in the activation of cell signaling transduction pathways. Lee et al. [10] reported that a germline juxtamembrane mutation, P1009S, showed an increased and sustained tyrosine phosphorylation of the protein compared with the wildtype. There has not yet been any study on the relationship between germline variations of MET and gastric cancer risk. We surveyed nonsynonymous variations in MET exons with the criteria of minor allele frequency (MAF) > 0.05 in Han Chinese and found only one variation in codon 375, N375S (rs33917957). Thus, we performed a case-control study including 1,681 gastric cancer cases and 1,858 controls to test the association between N375S and risk of gastric cancer.

Study subjects
This case-control study was approved by the Institutional Review Board of Nanjing Medical University. As described previously [11] , newly-diagnosed gastric cancer patients were recruited from the cities of Yangzhong and Yixing between January 2004 and July 2005 and the cities of Yangzhou and Nanjing between October 2006 and June 2010 in Jiangsu province, China. The criteria for recruitment of gastric cancer patients were as follows: 1) if these subjects were self-reported Han Chinese; 2) if they had resided in these above mentioned cities for at least 5 years; 3) if they were newly histopathologically diagnosed as primary gastric cancer; 4) if they had no previous malignancy in any other organs; 5) if they received no previous anti-tumor therapy before recruitment, including chemotherapy, and radiotherapy. As a result, a total of 1,681 gastric cancer patients were included in this study with a response rate of 87.3%. Based on the Lauren classification [12] , most gastric cancer cases were divided into two subgroups, the intestinal type and diffuse type, and those with the mixed type or whose cancer type was difficult to be determined were denoted as the unclassified type.
Additionally, 1,858 controls were randomly selected from a pool of more than 40,000 cancer-free individuals who participated in the communitybased screening program for non-infectious diseases conducted in Jiangsu province, with an overall response rate of 83.8%. All controls were frequencymatched to gastric cancer cases by age (5-year interval), gender and residential areas (city).
After written informed consent was obtained, each participant was surveyed by a structured questionnaire to collect information on demographic data and environmental exposure history. Individuals who smoked one cigarette per day for over one year were defined as smokers, and those who consumed three or more alcohol drinks a week for over six months were categorized as alcohol drinkers [12][13][14] . After the interview, about 5-mL venous blood sample was collected from each participant.

Genotyping
Genomic DNA was extracted from a leukocyte pellet by the traditional proteinase K digestion method and followed by phenol-chloroform extraction and ethanol precipitation. The germline variation, N375S (rs33917957 A>G), was genotyped using the TaqMan allelic discrimination assay on an ABI 7900 system (Applied Biosystems, CA, USA). The primers and probes were as follows. Primers: sense, 5'-TGCA TTCCCTATCAAATATGTCAAC-3', antisense,

' -G C T G G A G A C A T C T C A C A T T G T T T T -3 ' ;
Probes: allele A, FAM-CTTCTTCAACAAGATC-MGB, allele G, HEX-TTCTTCAGCAAGATC-MGB. Genotyping was performed blind without knowing the subjects' case or control status. Two negative controls in each 384-well plate were used for quality control and 5% samples were randomly selected to repeat the procedure, yielding a 100% concordance rate.

Statistical analysis
We used the Student's t-test (for continuous variables) and χ 2 test (for categorical variables) to detect the differences in demographic characteristics and genotype frequencies of N375S between the cases and controls. The goodness-of-fit χ 2 test was used to evaluate Hardy-Weinberg equilibrium (HWE) in the control subjects. Logistic regression analysis was applied to estimate crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the associations between genotypes and gastric cancer risk. Heterogeneity of associations between subgroups was analyzed by the Chi-square-based Q test. All of the statistical analyses were performed with Statistical Analysis System software (Version 9.1.3; SAS Institute, Cary, NC, USA). All tests were two-sided and the α was set at 0.05.

RESULTS
The demographic characteristics of 1,681 gastric cancer cases and 1,858 controls were summarized in our previous study [13] , The genotyping call rate for N375S was 97.85% and the successfully genotyped cases and controls were 1,633 and 1,830, respectively. The genotype distributions in gastric cancer cases and cancer-free controls are shown in Table 1. Briefly, among the 1,681 gastric cancer cases, 1,070 (63.6%) were classified as the intestinal type, 312 (18.6%) as the diffuse type, and 299 (17.8%) as unclassified. Among them, 822 (48.9%) had cardiac gastric cancer, and 725 (43.1%) had non-cardiac gastric cancer and gastric cancer in 134 (8.0%) patients was unclassified. There were no significant differences in the distribution of age, gender, smoking and drinking status between cases and controls (P = 0.884, 0.245, 0.198 and 0.072, respectively).
In the stratified analysis, the variant genotypes NS/ SS were associated with significantly decreased risk in non-cardiac or diffuse gastric cancer ( Table 2). However, no significant heterogeneity was detected for the associations among the subgroups, implying a Adjusted for age, gender, smoking and drinking status. CI: confidence interval; OR: odds ratio.  Individuals smoking one cigaretts per day for over one year were defined as smokers. Individuals consumed three or more alcohol drinks per week for over six months were defined as drinkers.
a Adjusted for age, sex, smoking status and drinking status (excluded the stratified factor in the first four strata). b P for heterogeneity. CI: confidence interval; OR: odds ratio.

DISCUSSION
Genetic variations may modify individuals' susceptibility to gastric cancer. It is of great importance to identify genetic variations that are associated with gastric cancer risk and potentially useful in gastric cancer risk prediction. The MET signaling pathway plays a critical role in various cancers, and the disruption of the balance of MET and HGF has been implicated in the etiology and progression of multiple cancers, including gastric cancer [6,7] . In this study, we found that a germline variation in MET, N375S, was associated with a significantly decreased gastric cancer risk.
N375S is located at exon 2 of MET, which corresponds to the semaphorin domain of the MET protein, which is the ligand binding region, and is highly conserved in mammalians based on the UCSC data (http:// genome.ucsc.edu/). Structure analysis indicated that N375 forms two potential hydrogen bonds, whereas S375 modeling structure retains only one hydrogen bond, which therefore weakens ligand binding affinity [15] . Given that the HGF/MET pathway is highly activated in gastric cancer [6][7][8][9] , it is biologically plausible that the MET S375 variant with a low affinity for HGF may reduce gastric cancer risk, which is consistent with our findings. Wu et al. [7] reported that MET was more frequently overexpressed in diffuse gastric cancer, indicating that the HGF/MET pathway was hyper-activated in diffuse gastric cancer. Our study showed that the protective effect of the variant genotype of N375S was more evident in diffuse gastric cancer.
Although we have a large sample size to detect the effect of MET N375S on gastric cancer susceptibility, limitation does exist in this study. The data of H. pylori infection was not available in this study; thus, it is difficult for us to adjust the potential confounding bias from H. pylori infection or to evaluate the potential gene-environment interaction.
In summary, this study, with a relatively large population, provides the first evidence that a germline missense variation in MET, N375S, may decrease susceptibility to gastric cancer in Han Chinese. Further studies incorporating diverse populations and functional assays are warranted to validate and expand our findings.