Vorschau

PDF, Englisch - Hauptdokument Download (8MB) | Nutzungsbedingungen

Abstract

Genome-wide profiling for genetic alterations in cancer has identified mutations in genes that are associated with epigenetic programming of genomes for DNA methylation patterns, histone modifications patterns and the positioning of nucleosomes. Here a systematic evaluation of the available cancer genome profiling data established by large international consortia, in order to identify recurrently mutated genes or pathways was described. Using curated list of approximately 700 epigenetic regulators and currently available genome-wide datasets on genetic and epigenetic alterations in cancers, the distribution of alterations in epigenetic regulators was described. Epigenetic genes were classified as potential oncogenic or those with tumor-suppressor function based on the location of mutations relative to functional domains and their frequencies. A panel of 50 epigenetic genes, including: DNMTs, histones (H3F3A, HIST1H3B), histone editors (KDM5C, KDM6A) and writers (MLLs, SETD2, EZH2, ATM) that can promote epigenetic changes in cancer was identified. Using correlative analysis of publicly available methylation data with information on deregulated epigenetic driver genes, many identified subtype-specific methylation clusters were correlated with groups of up to 3 epigenetic regulators. This analysis provides a source for the identification and link between methylation groups and deregulated epigenetic genes. Major cancer specific methylation changes have been observed in promoters and gene bodies. Tissue-specific cancer methylation differences have been located in enhancers and regulatory regions of non-coding RNAs. Based on identified results, the major mechanism of non-coding RNA deregulation in cancer has been investigated on independent data cohort. Using integrative analysis of non-coding RNA in early-onset prostate cancer, non-coding RNAs were classified as tumor-suppressive and oncogenic. About 120 novel prostate cancer specific non-coding RNAs that have been epigenetically deregulated have been identified. Our study on the defects in regulators of the epigenome will help to understand mechanisms leading to distinct epigenetic patterns and will allow the molecular validation of defined correlations in experimental settings.