Mechanism of epigenetic activation during disease progression

Division Of Basic Sciences - Nci

Investigators

Abstract

The overall goal of the project is to understand epigenetic mechanisms that drive disease progression and treatment resistance in advanced prostate cancer, with the aim of guiding development of novel therapeutics. In FY25 we published our results of a CRISPR screen designed to identify transcription factors that bind to a somatically acquired enhancer of the androgen receptor, which is activated exclusively in treatment resistance disease. We showed that three transcription factors, HOXB13, GATA2, and TFAP2C bind to the AR enhancer and promote chromatin features associated with enhancer activation such as post-translational histone modification, chromatin accessibility, and enhancer-promoter looping. Interestingly we discovered that HOXB13 in particularly was critical for reprogramming the global enhancer landscape, suggesting it may be a promising therapeutic target. In an effort to target HOXB13 we conducted a CRISPR screen to mutagenize endogenous HOXB13 to identify domains essential for its activity. We discovered point mutations outside the DNA binding domain of HOXB13 that are required for its function and prostate cancer cell survival. We are currently investigating the mechanism of these mutations and whether they are amenable to targeted therapies. In parallel, we performed base editor CRISPR screens of the AR enhancer to identify cis-regulatory elements required for activity. We used a novel proteomics approach to complement the CRISPR screen to identify factors that bind to the AR enhancer in the absence and presence of loss of function mutations. We propose that this unbiased approach can be applied to any noncoding regulatory element. In FY25, we also continued our investigations into the mechanism by which the histone arginine methyltransferase inhibitors of PRMT1/5 synergize with antiandrogen therapy in preclinical models. Results from epigenomic profiling of patient derived xenograft models suggest reprogramming of the AR landscape.