BMI1, a polycomb group (PcG) protein, plays a critical role in epigenetic regulation of cell differentiation and proliferation, and cancer stem cell self-renewal. BMI1 is upregulated in multiple types of cancer, including prostate cancer. As a key component of polycomb repressive complex 1 (PRC1), BMI1 exerts its oncogenic functions by enhancing the enzymatic activities of RING1B to ubiquitinate histone H2A at lysine 119 and repress gene transcription. Here, we report a PRC1-independent role of BMI1 that is critical for castration-resistant prostate cancer (CRPC) progression. BMI1 binds the androgen receptor (AR) and prevents MDM2-mediated AR protein degradation, resulting in sustained AR signaling in prostate cancer cells. More importantly, we demonstrate that targeting BMI1 effectively inhibits tumor growth of xenografts that have developed resistance to surgical castration and enzalutamide treatment. These results suggest that blocking BMI1 alone or in combination with anti-AR therapy can be more efficient to suppress prostate tumor growth.

Whole-genome bisulfite sequencing (WGBS) is the gold standard for studying landscape DNA methylation. Current computational methods for WGBS are mainly designed for gene regulatory regions with multiple under-methylated CpGs (UMCs), such as promoters and enhancers.

Numerous studies of relationship between epigenomic features have focused on their strong correlation across the genome, likely because such relationship can be easily identified by many established methods for correlation analysis. However, two features with little correlation may still colocalize at many genomic sites to implement important functions. There is no bioinformatic tool for researchers to specifically identify such feature pairs. Here, we develop a method to identify feature pairs in which two features have maximal colocalization minimal correlation (MACMIC) across the genome. By MACMIC analysis of 3306 feature pairs in 16 human cell types, we reveal a dual role of CCCTC-binding factor (CTCF) in epigenetic regulation of cell identity genes. Although super-enhancers are associated with activation of target genes, only a subset of super-enhancers colocalized with CTCF regulate cell identity genes. At super-enhancers colocalized with CTCF, CTCF is required for the active marker H3K27ac in cell types requiring the activation, and also required for the repressive marker H3K27me3 in other cell types requiring repression. Our work demonstrates the biological utility of the MACMIC analysis and reveals a key role for CTCF in epigenetic regulation of cell identity. The code for MACMIC is available at https://github.com/bxia888/MACMIC.

Cancers result from a set of genetic and epigenetic alterations. Most known oncogenes were identified by gain-of-function mutations in cancer, yet little is known about their epigenetic features. Through integrative analysis of 11,596 epigenomic profiles and mutations from >8200 tumor-normal pairs, we discover broad genic repression domains (BGRD) on chromatin as an epigenetic signature for oncogenes. A BGRD is a widespread enrichment domain of the repressive histone modification H3K27me3 and is further enriched with multiple other repressive marks including H3K9me3, H3K9me2, and H3K27me2. Further, BGRD displays widespread enrichment of repressed cis-regulatory elements. Shortening of BGRDs is linked to derepression of transcription. BGRDs at oncogenes tend to be conserved across normal cell types. Putative tumor-promoting genes and lncRNAs defined using BGRDs are experimentally verified as required for cancer phenotypes. Therefore, BGRDs play key roles in epigenetic regulation of cancer and provide a direction for mutation-independent discovery of oncogenes.

Conversion between cell types, e.g., by induced expression of master transcription factors, holds great promise for cellular therapy. Our ability to manipulate cell identity is constrained by incomplete information on cell identity genes (CIGs) and their expression regulation. Here, we develop CEFCIG, an artificial intelligent framework to uncover CIGs and further define their master regulators. On the basis of machine learning, CEFCIG reveals unique histone codes for transcriptional regulation of reported CIGs, and utilizes these codes to predict CIGs and their master regulators with high accuracy. Applying CEFCIG to 1,005 epigenetic profiles, our analysis uncovers the landscape of regulation network for identity genes in individual cell or tissue types. Together, this work provides insights into cell identity regulation, and delivers a powerful technique to facilitate regenerative medicine.