A Polymorphism At The MiR-502Binding Site In The3’-untranslated Region Of The SET8Gene Is Associated With Clinical Characteristics And The Prognosis Of Esophageal Cancer

Objective: MicroRNA(miRNA) is a highly conservated small no-codingRNA, which functions in post-transcriptional regulation of mRNA expression.It usually results in gene-silencing via translational repression or targetdegradation by base-pairing with complementary sequences of the3’untranslated region(3’UTR) of a target mRNA. MiRNA plays importantroles in a broad range of biological processes, such as cell proliferation,differentiation, hormone secretion, tumorigenesis and development,apoptosis. Single nucleotide polymorphisms(SNP) is a variation that occurswhen a single nucleotide sequence in the genome differs, which is widelydistributed in the entire human genome, and it contributes to the geneticvariations between individuals in term of gene sequence. There are somefunctional SNPs located in the miRNA binding sites of target gene, suchvariation of the single base can influence the complementarity between themiRNA and its target mRNA, subsequently can alter the target gene’sexpression. SNPs in miRNA binding sites was identified for their associationwith cancer risk and prognosis of the tumors in recent studies, but fews hadfocused on the relationship for SNPs in miRNA binding sites andesophageal cancer. rs16917496, located in the3’UTR of SET8gene, ispolymorphic binding site for miR-502. This will be the first study access theassociation of binding site polymorphism for miR-502located in the SET83’UTR with the clinical characteristics and the prognosis of esophagealcancer patients. It will also investigate the relationship between clinicalcharacteristics and prognosis in esophageal cancer patients. It contributes toscientific basis for the reasonable treatment and prognostic evaluation ofesophageal cancer patients. Methods:1Sample collection and patients following-up: Blood samples forexperimental group were collected from65esophageal cancer patients whowere diagnosed of esophageal carcinoma by gastroscopy and histopathologicalmethod at Fourth Hospital of Hebei Medical University. All patientsunderwent tumor resection in the Department of Esophageal Surgery fromJuly2002to September2007. The patients’ clinical characteristics (gender,age, et al) were accurate recorded. We followed up for5-years survivalstatus of post-operative patients by outpatient clinic, phones and letters. Bloodsamples for control group were collected from60healthy controls at FourthHospital of Hebei Medical University from October2011to May2012.Patients in control group have complete medical history, with no associateddiseases with esophageal cancer.2DNA extraction, amplification and sequencing: The genomic DNA wasextracted with Wizard Genomic DNA extraction kit from blood samples. Thetarget gene of qualified DNA samples were then amplified by PCR, theforward primer is5’-TCACGACGGT GCTACCTAAG-3’, and the reverseprimer is5’-CATGCTGGTGTGACACAGTC-3’.The products were confirmedby agarose gel electrophoresis and sequencing. Polymorphisms wereconfirmed by repeating the analysis on both strands.3Statistical analysis: Student’s t-test was used to analysis measurementdata, and χ2test (Pearson Chi-Square test) was used to analysis enumerationdata. The univariate analysis was performed by Kaplan-Meier method andLog-Rank method, the multifactor analysis was done by COX regressionmodel. P-value, relative risk(RR) and confidence interval(CI) were calculated.All analyses were performed by SPSS software17.0, P＜0.05was consideredstatistically significant.Results:1There is no statistical difference between esophageal cancer patientsand healthy controls with distribution frequency of T, C allelotypes (χ2=0.305,P=0.581). The distribution frequency for each genotype (T/T Vs. C/C, T/T Vs. C/T and T/T Vs. C/T+C/C) were no statistical difference between esophagealcancer patients and controls (P>0.05). Thus, rs16917496is not a cancer riskmarker for esophageal cancer patients.2No association exist for the genotype distribution of rs16917496andclinical characteristics in esophageal cancer patients (P>0.05).3In those60esophageal cancer patients who have accurate survival data,univariate analysis showed that patients with C/C, C/T, T/T genotypes had a5-year survival rate of87.5%,34.8%,27.6%, respectively. There wassignificant difference of5-year survival rates for C/C Vs. T/T (P=0.010), C/CVs. C/T (P=0.026). But no statistical significant difference in the survival ratesof patients with C/T and T/T (P=0.560). Patients with C/C genotype haslonger survival time than that of C/T or T/T patients. Clinical characteristicssuch as gender, age, tumor location, tumor size, pathological classification,intravascular tumor emboli were not associated with the post-operativesurvival rate of esophageal cancer patients, but tumor stage was associatedwith esophageal cancer outcome (P=0.011) with5-year survival rates of stage0+Ⅰgroup and stageⅡ+Ⅲ group were80.0%and30.0%.The factors which have significant difference in univariate analysis werethen analyzed by Cox regression modle. It showed that the SNP rs16917496located in SET83’UTR were independent predictor for esophageal cancerpatients, the death risk of the C/T and T/T genotypes group was7.801times ofthat for the C/C genotype group. The clinical stages were correlated with theprognosis of esophageal cancer patients too, the mortality risk of stage Ⅱ~Ⅲgroup was4.953times of that for the stage0~Ⅰgroup.Conclusions:1SET8genotype can be used as the independent risk factor which wascorrelated with the prognosis of esophageal cancer patients.2The polymorphism at the miR-502binding site in the3’UTR of theSET8genotype was not related to the risk of developing esophageal cancer.The different genotypes of rs16917496targeting in SET8are not significantlycorrelated to the different clinical characteristics of esophageal cancer patients.