Dna Methylation Sequencing Of Prostate Tumors: Possible Molecular Mechanisms For Cancer Aggressiveness

A major challenge in the clinical management of prostate cancer is the inability to definitively diagnose indolent versus aggressive cases. Contributing to this challenge is a lack of basic science understanding of the molecular basis behind aggressiveness subtypes in prostate cancer. DNA methylation is the epigenetic addition of a methyl group to the DNA base cytosine and has been found to regulate cell proliferation and environmental adaptation. We hypothesized that DNA methylation changes are a mechanism by which an aggressive cancer attains phenotypes that distinguish it from indolent cases via disruption of regulatory networks. This hypothesis was tested by comparing DNA methylation between benign prostate and both low grade (Gleason score 6) and high grade (Gleason score 8 to 10) groups. Methylome-wide next generation sequencing was performed on formalin-fixed paraffin embedded (FFPE) samples from radical prostatectomy cases using MBD-isolated genome sequencing (MiGS). Global clustering analyses showed some separation between the three groups, but also indicated some heterogeneity. Specific differentially methylated regions (DMRs) were detected between the three groups. The two most prevalent were hypermethylation specific to high grade (491 DMRs at 4% false discovery rate) and hypermethylation shared between both cancer subtypes (1334 DMRs at 1% false discovery rate). Statistical and computational analysis of this data set was non-trivial and the software developed to perform the analysis has been made available. High grade specific DMRs are abundant in intergenic and gene body contexts, and occur less frequently in promoters and CpG islands than DMRs shared between both subtypes. Intergenic DMRs are enriched for putative functional elements (ChIP-seq peaks and DNaseI hypersensitive sites from the ENCODE project) suggesting that these changes may have functional impacts as local or regional enhancers regulating multiple genes. To analyze this in more detail, DNA-protein binding sites for specific factors were compared between 5 spatial contexts. Differential patterns of enrichment were observed for regulatory binding factors between promoters, intergenic regions, and gene 39 ends. A de novo motif search revealed 8 frequent sequence motifs in intergenic DMRs that are not similar to any known binding motifs, indicating possible binding sites for as yet unknown factors. For DMRs that are proximal to genes, an interaction network based on curated pathway data was built using the GeneMANIA Cytoscape plugin. Gene ontology enrichment was observed for “positive regulation of cell motility” (Q-value=1.4e-07, 22 of 296 genes in term) and “regulatory region DNA binding” (Q-value=6.7e-07, 18 of 255 genes in term) within this network. Interestingly, each GO term appears as a distinct cluster on the interaction network. While the specific regulatory action of these DMRs and exact target genes in aggressive cancer remains to be determined experimentally, this work has established the presence of focal hypermethylation that distinguishes indolent and aggressive prostate cancer. The high frequency of intergenic but functionally enriched genomic contexts and functional enrichment of these DMRs at genic sites for transcription factors suggests that they regulate aggressiveness in a complex manner through secondary and tertiary effects. This study provides a first glimpse at molecular differences between indolent and aggressive prostate and future work may lead to the development of biomarkers and treatments based on this knowledge. Citation Format: Jeffrey M. Bhasin, Lars Matkin, Margaret G. Taylor, Byron H. Lee, Cristina Magi-Galluzi, Eric A. Klein, Bo Hu, Yaomin Xu, Angela H. Ting. DNA methylation sequencing of prostate tumors: Possible molecular mechanisms for cancer aggressiveness. [abstract]. In: Proceedings of the AACR Special Conference on Translation of the Cancer Genome; Feb 7-9, 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 1):Abstract nr A1-24.