Prostate Cancer Risk SNPs enriched in Androgen Receptor Binding Sites. Homo sapiens

Genome-wide association studies (GWAS) have identified dozens of genomic loci, whose single nucleotide polymorphisms (SNPs) predispose to prostate cancer (PCa). However, the biological functions of these common genetic variants and the mechanisms to increase disease risk are largely unknown. We integrated chromatin-IP coupled sequencing (ChIP-seq) and microarray expression profiling in the TMPRSS2-ERG gene rearrangement positive DuCaP cell model with the NHGRI GWAS PCa risk SNPs catalog, in an attempt to identify disease susceptibility SNPs localized within functional androgen receptor binding sites (ARBSs). Among the 48 GWAS index SNPs and 2,702 linked SNPs defined by the 1000G project 104 were found to be localized in the AR ChIP-seq peaks. Of these risk SNPs, rs11891426 T/G in the 7th intron of its host gene melanophilin (MLPH) was found located within a putative auxiliary ARE motif, which we found enriched in the neighborhood of canonical ARE motifs. Exchange of T to G attenuated the transcriptional activity of the MLPH-ARBS in a reporter gene assay. The expression of MLPH protein in tissue samples from prostate cancer patients was significantly lower in those with the G compared to the T allele. Moreover, a significant positive correlation of AR and MLPH protein expression levels was also confirmed in tissue samples. These results unravel a hidden link between AR and a functional PCa risk SNP rs11891426, whose allele alteration affects androgen regulation of its host gene MLPH. This study shows the power of integrative studies to pin down functional risk SNPs and justifies further investigations. Overall design: Prostate cancer cell lines LNCaP and DUCaP cells were cultured in the absence or presence of R1881, an androgen in three independent experiments. Differential gene expression was determined by comparing R1881 treated samples with the corresponding controls (EtOH treated samples).