MiR-146a Rs2910164G/C Polymorphism Is Associated With Gastric Cancer Risk Through Aberrant Exprssion Of MiR-146a

Background and objectGastric cancer (GC) is the fourth most common cancer in the world with morethan70%of cases occur in the developing world. More than50%of cases occur inEastern Asia. GC is the second leading cause of cancer death in both sexes worldwide.In Asia, GC is the third most common cancer after breast and lung and is the secondmost common cause of cancer death after lung cancer. Although the incidence andmortality rates are slowly declining in many countries of Asia, GC still remains asignificant public health problem. Therefore, gastric cancer is becoming one of thebiggest economic burden among diseases in the nearly future. Despite the advances incurrent therapeutic approaches of gastric cancer, the development of adjuvantchemotherapy, which improve the clinical therapeutic effect in some way. On thecontrary, the prognosis of gastric cancer still remains poor in those advanced caseswith lymph node metastasis. Therefore, some reporters believed that some genes seemto contribute to the malignant potential of gastric cancer, but it remains extremelyimportant that the identification of the precise factors which predict the prognosis and recurrence of gastric cancer.MicroRNAs (miRNAs) are a family of mature,21-to25-nucleotide-long smallnoncoding and highly conserved RNA genes products that regulate gene expressionby base pairing with target mRNAs at the3’-untranslated region, leading to mRNAcleavage or translational repression. It has been suggested that miRNAs are involvedin various biological processes, including the development of disease, celldifferentiation, cell proliferation and cell death. Moreover, recent evidence hasrevealed that the change of miRNAs expression play critical roles duringtumorigenesis, development and prognosis of solid cancers, including gastric cancer,which indicated that miRNAs could act as tumor suppressors or oncogenes.Single-nucleotide polymorphisms (SNPs) is the most common type of geneticvariation in the human genome and contribute to human phenotypic differences.Polymorphisms in human miRNA genome lesions could modify the biologicalprocesses by influencing the processing and/or target selection of miRNAs, which areimplicated in cell cycle regulation, and thereby play critical roles in carcinogenesis.Currently, some researchers believed that the single nucleotide polymorphisms (SNPs)in miRNA or it’s binding sites were the abnormal source of genetic variation, whichcontribute to cancer susceptibility. There were two different nucleotide on specificposition in the genome, whose frequency was no less than1%in the population. Thedifference of a single nucleotide of this genomic was known as SNPs, which has aspecific biology, can regulate the expression of miRNA and effect its regulatoryfunctions, thus has significant association of the tumor development. SNPs are theminor differences in gene sequence which present in each individual humanpopulation and fairly well distributed throughout the human genome. Otherwise,most SNPs do not affect the function of the gene, but the SNPs in some specific areacan make people susceptible to disease or affect the sensitivity of the drug. Recently,the association between SNP in miRNA sequence and cancer risk opened up anavenue to explore novel molecular mechanism of cancer development.MiR-146a has been found aberrantly expressed in various types of cancer duringthe past decades, it probably acted as a novel regulator in cancer development and progression. The level of miR-146a is regulated by a single nucleotide polymorphism(SNP). Recently, some researchers reported that the G/C polymorphism (rs2910164)in miR-146a, was the reason which affecting the expression of miR-146a in papillarythyroid carcinoma. Likewise, it was also associated with the risk of hepatocellularcarcinoma, prostate carcinoma, papillary thyroid carcinoma and breast cancer.However, the relation between miR-146a and gastric cancer is still controversial, asexpression of miR-146a has been found both up-and down-regulated in gastriccancer. Therefore, we focus on confirming the expression of miR-146a in gastriccancer tissues and cell lines, and characterized the association between rs2910164polymorphism and gastric cancer risk, which may be useful for identifying the exactrole of miR-146a in gastric cancer.In our study, we found miR-146a is down-regulated in human gastric cancertissues as well as in gastric cancer cells. Individuals with variant GC/CC genotype ofmiR-146a were in decreased risk for gastric cancer compared with those carrying wildGG genotype. In addition, the G-to-C change resulted in increased expression ofmature miR-146a in gastric cancer patients. These results indicated a role formiR-146a as a tumor suppressor in gastric cancer and provided evidence thatmiR-146a SNP rs2910164might play an important role in the prognosis of gastriccancer.Methods1. Collected gastric specimens: Our study recruited417patients whose clinical datasand histopathologically confirmed to be gastric cancer from July2009to March2013in the First Affiliated Hospital of Nanchang University (Nanchang, China)and Cancer Center of Guangzhou Medical University (Guangzhou, China).Cancerous and paired non-cancerous tissues were obtained from underwent gastriccancer surgery patients and immediately snap frozen in liquid nitrogen.2. Collected non-cancer samples:420healthy people were recruited from NanchangCDC over the same area, simultaneously, collecting their blood samples.3. Cell culture: All cells were maintained at37℃in a humidified incubator containing5%CO2and treated with celastrol in complete medium, includinghuman gastric epithelial cell line (GES-1) and six gastric cancer cell lines (AGS,BGC-823, HGC-27, MKN-28, MKN-45and SGC-7901).4. RNA extraction and the detection of miR-146a: Total RNA from tissue samplesand cell lines was obtained with the TRIzol isolation reagent, and the total RNAwere reverse transcribed to synthesize cDNA. qRT-PCR method was used toquantify mature miR-146a and pri-miR-146a.5. Transient miRNA-146a transfection: AGS cell was chosen to do the transfectionexperiments which has the lowest expression of miR-146a. AGS cells in12-wellplates were transfected with miR146a precursor (miR-146a mimic) or scrambledpre-miR negative control (miR-146a mimic NC), then measured the transfectionefficiency.6. Cell proliferation assay: Transfected AGS cells were plated in96-well plates (0.5×104cells per well) at24h after transfection and cultured in normal conditions.After incubation for24h, CCK-8reagent was added to detect cell proliferation.7. Colony formation assay: After transfected for24h, AGS cells were trypsinized,counted and placed onto the fresh60mm plates, density of1000cells per plate. Thecolony formation rate was calculated after culturing for10days.8. Cell cycle analysis：AGS cells were harvested at24h after transfection and fixed in70%ice-cold ethanol, treated with RNAse A, stained with50mg/ml propidiumiodide and0.1mg/ml RNase A for DNA content analysis by flow cytometry with aFACS Calibur system (Becton Dickinson, Franklin, NJ). The percentage of cellpopulation in each phase was calculated with FlowJo FACS analysis software (TreeStar, Inc., Ashland, OR).9. DNA isolation and genotyping：Genomic DNA was extracted by QuickGene DNAwhole blood kit (Fujifilm, Tokyo, Japan) from the whole blood sample according tothe manufacturer’s recommended protocol by QuickGene-810isolation system(Fujifilm). For miR-146a SNP (rs2910164, G/C), genotyping work was performedusing Taqman Allelic Discrimination (Applied Biosystems, Foster City, CA, USA)with a commercially available primer probe set and an Applied Biosystems7500 real-time PCR system (Applied Biosystems). In order to ensure the quality ofgenotyping, sequencing was done for10%of randomly selected samples with highDNA quality, and the results were>98%concordant.10. Statistical analysis：Biostatistical analyses were performed using SPSS17.0.Student’s t-test, one-way ANOVA, or χ2-test was using to test the difference amonggroups. Pearson’s χ2-test was used to compare the the basic situation between thecase group and the control group. Goodness-of-fit χ2analysis was used to test theHardy-Weinberg equilibrium, to analysis the observed and expected values betweendemographic characteristics and genotype frequencies in the control group, whichto assess their representativeness. To assess the association of the SNP with the riskof gastric cancer, we estimated odds ratios (OR) and95%confidence intervals (CI)using the unconditional multivariate logistic regression model. P <0.05wasconsidered statistically significant.Results1. No statistically difference in general datas between case group and controlgroup: There were no statistically differences between two groups, such as casegroup (417cases) and control group (420cases). The investigate factors includingsex (P=0.261), age (P=0.292), the infection status of H. pylori (P=0.217), theresidence (P=0.103), the family history of gastric cancer (P=0.082), thesmoking history (P=0.135).2. Low expression of miR-146a in gastric cancer tissue samples: Specimens wererandomly selected in53gastric cancer patients, in which there were significantdifferences of miR-146a expression between gastric cancer tissues and pairingadjacent normal tissue. Moreover,50gastric cancer tissue samples in this studywere compared with pairing adjacent normal tissue, which the former showedsignificantly lower expression (P <0.01).3. Low expression of miR-146a in gastric cancer cells: It showed a significantlylower expression of miR-146a in six cell lines deriving from gastric cancer, such as AGS, BGC-823, HGC-27, MKN-28, MKN-45and SGC-7901, whichcomparing with normal human gastric epithelial cell line (GES-1).4. Transient miR-146a transfection: AGS cells was selected for the nexttransfection studies because of its lowest miR-146a expression levels among thesesix human gastric cancer cell lines. Significantly up-regulated effect on miR-146aexpression level in AGS cells by transfection of miR-146a mimic was observed,compared with mimic NC-transfected or untransfected (mock) AGS cells.5. AGS cells proliferation were changed by miR-146a: After transfected for24hours, the proliferation rate of AGS cells with miR-146a over-expression wasdetermined via CCK-8assay. Compared to mimic NC group, the cell proliferationin miR-146a mimic transfected group was at a significantly lower rate (P <0.01).The colony formation assay yielded similar results. miR-146a mimic transfectedAGS cells showed lower colony-forming efficiency compared with controls (P <0.05).6. Cell cycle analysis: After transfected for24hours, we measured cell cycledistribution by flow cytometry. AGS transfected with miR-146a mimic resulted ina higher percentage of cells in G0/G1phase (P <0.05) and lower fraction in Sphase (P <0.01), compared with NC group.7. The relationship between different rs2910164genotype and the risk of gastriccancer: After adjustments were made for potential covariates, carriers of thers2910164GC (OR=0.719;95%CI,0.461-1.003; P=0.031) or CC genotype(OR=0.185;95%CI,0.111-0.508; P=0.005) exhibited a statistically significantdecreased risk of gastric cancer compared with the carriers of the rs2910164GGgenotype. Subjects carried rs2910164GC and CC genotypes exhibited a0.605-fold decreased risk (95%CI,0.368-0.912; P=0.011) compared to GGhomozygotes carriers. Furthermore, the C allele displayed a significantly lowcorrelation with the occurrence of gastric cancer compared to the G allele (OR=0.611;95%CI,0.504-0.922; P=0.008).8. The relationship between different rs2910164genotype and gastric cancer prognosis：After the stratified analyses, significant decreased risk of lymph nodemetastasis (OR=0.527;95%CI,0.360-0.791; P=0.005) was found in carriers ofthe rs2910164GC/CC genotype compared with the GG homozygotes carriers. Wealso found that the reduction of gastric risk was associated with GC/CC genotypetended to be more pronounced in patients in stage Ⅲ/Ⅳ (OR=0.622,95%CI,0.423-0.917, P=0.023) compared to those in stage Ⅰ/Ⅱ.9. Impact of rs2910164G/C polymorphism variant on miR-146a expression: Indifferent genotypes of the miR-146a polymorphism rs2910164, the distribution ofthe GG, GC, and CC genotypes was30(46.15%),26(40.00%), and9(13.85%),respectively. Compared to carriers of rs2910164GG genotype, GC, CC and GC/CC genotype carriers exhibited higher expression levels of mature miR-146a (P <0.01). However, the expression levels of pri-miR-146a didn’t have significantdifference in the groups.Conclusions1. The cell proliferation and colony formation in AGS cells had been inhibited byincreasing the expression of miR-146a, which proved that miR-146a might serveas a tumor suppressor gene in gastric carcinogenesis.2. Carriers of the rs2910164C allele should have less risk in suffering from gastriccancer and lymph node metastasis than G allele carriers.3. The mature miR-146a expression in the rs2910164C allele carriers were higherthan those of the G allele carriers.