Systematic Identification Of Regulatory SNPs Associated With Cancer

Since 2005,many genome-wide association studies(GWAS)were carried out to identify SNPs associated with disease risk from the whole genome,and more than ten thousands of risk SNPs were identified.Among them,960 SNPs were associated with cancer risk.In consideration of linkage disequlibrium,there are 10,673 cancer risk assocated SNPs linked with the 960 GWAS tag SNPs.However,over 90%of the GWAS identified SNPs are located in non-coding regions and it is challenging to assess their functional impacts.To systematically identify the function of SNPs in non-coding regions and explore the potential mechanisms underlying their regulation of tumorigenesis,we adapted the self-transcribing active regulatory region sequencing(STARR-seq)strategy,a high-throughput technique to functionally quantify enhancer activities.From 10,673 SNPs linked with 996 cancer risk-associated SNPs identified in previous GWAS studies,we identified 575 SNPs in the fragments that positively regulate gene expression(PRE),and 758 SNPs in the fragments with negative regulatory activities(NRE).Then we validated the activities of the SNPs by dual-luciferase assay.After that,for those SNPs located in regulatory elements,we compared the regulatory activities of two alleles to identify those SNPs for which two alleles conferred different regulatory activities(regulatory SNPs).From the 1,333 SNP-containing regulatory elements,we identified 70 regulatory SNPs.We discovered that those regulatory SNPs enriched in transcription factor binding sites and were more likely to disrupt transcription factor binding.In addition,we mapped potential target genes of these regulatory SNPs through expression quantitative trait loci(eQTL)analysis.Finally,we chose two SNPs for further analysis.One is a breast cancer risk SNP rs11055880,which is located in an enhancer region of ATF7IP and its alternative allele can alter the enhancer activity.We used CRISPR interference to confirm the regulatory role of the SNP-containing region for ATF7IP.The other SNP was rs12142375,which was linked with childhood leukemia risk SNP rs546784,and located in the intron of PDE4B.The GWAS risk allele G conferred higher enhancer activity and correlated to a higher expression level of PDE4B,which was in accordance with clinical data that cases with GG genotype showed highest PDE4B expression level.Our research systematically identified SNPs that affecting gene expression by modulating activities of distal regulatory elements.We discovered their target gene through eQTL analysis,and for two SNPs uncovered the mechanism underlying their regulation of target gene expression.This study will help the interpretation of GWAS results and provide improved information for cancer risk assessment.Our research also provided a new approach to explore the function of non-coding variants,and identified thousands of SNPs located in regulatory elements,which could be used as a resource for future investigations.