Abstract
As promising biomarkers and therapy targets, microRNAs (miRNAs) are involved in various physiological and tumorigenic processes. Genetic variants in miRNA‐binding sites can lead to dysfunction of miRNAs and contribute to disease. However, systematic investigation of the miRNA‐related single nucleotide polymorphisms (SNPs) for pancreatic cancer (PC) risk remains elusive. We performed integrative bioinformatics analyses to select 31 SNPs located in miRNA‐target binding sites using the miRNASNP v2.0, a solid database providing miRNA‐related SNPs for genetic research, and investigated their associations with risk of PC in two large case‐control studies totally including 1847 cases and 5713 controls. We observed that the SNP rs3802266 is significantly associated with increased risk of PC (odds ratio (OR) = 1.21, 95% confidence intervals (CI) = 1.11‐1.31, P = 1.29E‐05). Following luciferase reporter gene assays show that rs3802266‐G creates a stronger binding site for miR‐181a‐2‐3p in 3′ untranslated region (3′UTR) of the gene ZHX2. Expression quantitative trait loci (eQTL) analysis suggests that ZHX2 expression is lower in individuals carrying rs3802266‐G with increased PC risk. In conclusion, our findings highlight the involvement of miRNA‐binding SNPs in PC susceptibility and provide new clues for PC carcinogenesis.
Highlights
Pancreatic cancer (PC) is the twelfth most common cancer around the world[1] and is one of the most lethal human cancers with a rather low 5-year survival rate of 5%.2,3 In China, there are estimated over 90 000 new cases and nearly 80 000 related deaths in 2015.4 Chronic pancreatitis, type 2 diabetes, obesity and cigarette smoking have been established as PC risk factors
The single nucleotide polymorphisms (SNPs) identified by genome-wide association studies (GWASs) and exome-wide association study (EWAS) could only explain a minor portion of heritability,[11,12] and the missing heritability of PC remains to be dissected.[13]
The SNP rs3802266 was tested in Replication Stage, and positive results were successfully replicated in the additional sample set. These findings revealed that rs3802266 A>G could influence the binding between miR-181a-2-3p and the ZHX2 3′ untranslated region (3′UTR)
Summary
Pancreatic cancer (PC) is the twelfth most common cancer around the world[1] and is one of the most lethal human cancers with a rather low 5-year survival rate of 5%.2,3 In China, there are estimated over 90 000 new cases and nearly 80 000 related deaths in 2015.4 Chronic pancreatitis, type 2 diabetes, obesity and cigarette smoking have been established as PC risk factors. Genome-wide association studies (GWASs) have identified multiple PC-associated chromosome loci.[7]. Our group performed a exome-wide association study (EWAS) and discovered three new associated regions (19p13.12, 8p21.3 and 2p24.1).[10]. We upgraded a widely used online database called miRNASNP v2.0,20 and successfully applied it to locate a functional SNP rs1062044 affecting miR-423-5p binding to the gene LAMC1 in colorectal cancer susceptibility locus 1q25.3.21 Still, systematic investigation on the miRNA-binding SNPs for PC risk is absent. Pancreatic cancer EWAS data-mining should help us to perform a genome-wide evaluation of the miRNA-binding polymorphisms. We integrated the data from the database miRNASNP v2.0 and our PC EWAS to genome-widely screen miRNA-binding polymorphisms for PC risk. Followed by an independent case-control study and corresponding functional assays, we identified a PC-associated variant rs3802266 affecting miR-181a-2-3p binding to the gene ZHX2
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
