The 11th International Symposium on 3D Genomics

44 / 2024-10-08 11:38:20

Prostate cancer transcriptomic regulation by the interplay of germline risk alleles, somatic mutations and 3D-genomic architecture

ChIAPET,GWAS,risk allele,somatic mutation,prostate cancer

Prostate cancer (PCa) is one of the most heritable human cancers. Genome-wide association studies (GWAS) have identified at least 185 PCa germline risk alleles. Many PCa germline risk alleles are enriched in the non-coding regions of the genome and are hypothesized to regulate transcription.

Identification of the transcriptional targets of risk alleles has been difficult for myriad reasons. First, in many cases it is unclear which of multiple alleles in linkage-disequilibrium (LD) is functional. Second, functional interrogation of risk alleles is complicated by their relatively small effect-sizes and which can compound over decades to subtly influence tumorigenesis. Third, individual risk alleles may regulate multiple genes via chromatin interactions, leading to effects in genes distal in the linear genome. Fourth, germline risk alleles occur in the context of an evolving somatic genome and transcriptome, leading to joint germline-somatic influences in regulation of target gene transcription.  

To help resolve these problems, we used integrative 3D spatial genomics to identify the chromatin interaction targets of 45 PCa risk alleles, 31 of which were associated with transcriptional regulation of target genes in 565 localized prostate tumors. To supplement these 31, we verified transcriptional targets for 56 additional risk alleles using linear proximity and linkage disequilibrium (LD) analysis in localized prostate tumors. Some individual risk alleles influenced multiple target genes; others specifically influenced only distal genes while leaving proximal ones unaffected. Several risk alleles exhibited wide-spread germline-somatic interactions in transcriptional regulation, having different effects in tumors with loss of PTEN or RB1 relative to those without. These data clarify functional PCa risk alleles in large linkage blocks and outline a strategy to model multi-dimensional transcriptional regulation.