New androgen receptor binding sites acquired with metastasis overlapped with those in fetal tissue.
Major Finding: New androgen receptor binding sites acquired with metastasis overlapped with those in fetal tissue.
Concept: These new sites were preloaded with the developmental transcription factors FOXA1 and HOXB13.
Impact: This study showcases the need to understand prostate cancer as an epigenetically driven disease.
Prostate cancer has been established to be an epigenetic disease driven in large part by the master transcription factor androgen receptor (AR). However, there remains a dearth of information concerning the epigenetic dynamics that occur during the transitions from normal prostate epithelial cell to prostate cancer cell to metastatic cell. To address this, Pomerantz, Qiu, Zhu, and colleagues analyzed the epigenomes of 268 human tissue specimens, including histologically normal prostate epithelium, primary prostate tumor tissue, patient-derived xenografts of AR+ metastatic castration-resistant prostate cancer (mCRPC), fresh-frozen mCRPC samples, fetal tissue, and cell lines grown from the human urogenital sinus (UGS; the fetal structure from which the prostate is derived). Comparing primary prostate epithelium with primary prostate cancer and mCRPC revealed reprogramming of the cistrome (the set of cis-acting elements acted on by a trans-acting factor such as AR) of AR across these transitions, with 17,655 new AR binding sites (ARBS) appearing in metastatic tissue but not primary tumor tissue. Notably, these novel ARBS were already enriched with the development-associated transcription factors FOXA1 and HOXB13 in normal tissue and primary prostate tumors. Further analysis revealed that the many aspects of the MCRPC epigenomic program were also present during fetal development, with the specifically activated genes having become quiescent in normal prostate epithelium and primary prostate tumors and reactivated by the time of metastasis. Metastasis-specific enhancers, including at HOXB13, FOXA1, and NKX3-1, were also identified, and their functional relevance was established using CRISPR-based interference screen. Importantly, an investigation of somatic and germline mutations influencing prostate cancer risk using the sites identified in this study showed that much of the heritable risk loci were in lineage-specific enhancers and had not previously been identified in studies using prostate cell lines. This work highlights the need to study prostate cancer as an epigenetic process from its origins to metastasis.
Note: Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.