Targeting androgen receptor (AR) has been the mainstream therapy for treating prostate cancer (PCa); however, resolving the drug resistance remains a major hurtle. Even with FDA-approved second-generation antagonists against AR-axis signaling, AR activity is eventually restored in the lethal metastatic castration-resistant PCa (mCRPC). Therefore, novel strategies targeting AR signaling remain to be developed.

One major mechanism that leads to the PCa development is reprogramming of AR cistrome by transcriptional factors and chromatin modifiers. In this process, AR is recruited to a subset of newly established enhancers that can drive the expression of proliferation genes. FOXA1 is one such transcriptional factor that determines cell lineage and is characterized as a “pioneer” factor to facilitate the access of additional transcription factors (such as AR or ER) to the regions with compact chromatin. However, the molecular basis of how FOXA1 was guided to bind to these regions remains unclear.

Lysine specific demethylase 1 (LSD1, KDM1A) is an important epigenetic modifier that is well known for its repressor function through demethylating histone 3 lysine 4. In PCa, our genome-wide studies showed that LSD1 also functions as a major activator in mediating AR-dependent enhancers and facilitates the transcription of androgen-regulated genes. Our previous studies showed that LSD1 binding significantly overlaps with FOXA1, and these overlapping sites (marked with high H3K4me1,2) are enriched for regulating androgen-stimulated genes. We also showed that LSD1 interacts with FOXA1 and this interaction enhances binding of both proteins at those sites. These findings suggest that LSD1 may facilitate FOXA1 binding to chromatin and subsequently regulate the enhancer availability to AR-mediated transcription.

In the current study, we first investigated the genomic landscape of LSD1 in PCa cells. Epigenomic profiling of LSD1 binding sites shows global enrichment of active enhancer markers with strong FOXA1 binding. Remarkably, the genome-wide FOXA1 binding was drastically diminished by inhibiting LSD1, indicating that its demethylase activity is absolutely required for FOXA1 binding to chromatin. Furthermore, LSD1 inhibition impaired the accessibility of AR-dependent enhancers. Using a LSD1 inhibitor that has entered clinical trials for lung cancer, we found that inhibiting LSD1 robustly disrupted androgen-induced AR genomic binding and transcriptional output and decreased PCa cell growth. Importantly, through this mechanism LSD1 inhibition also impairs AR-V7 chromatin binding in CRPC models and suppressed AR-V-driven tumor growth. Mechanistically, we hypothesized that FOXA1 may be a direct target of LSD1 and removing the lysing methylation stabilizes FOXA1 binding to chromatin. Mass-spec (LC/MS/MS) on immunoprecipitated FOXA1 in PCa cells resulted in the identification of methylated lys270 (located near Forkhead domain), where its methylation was increased by LSD1 inhibitor. Mutant FOXA1 (K270R) bound more tightly with chromatin and its binding was not affected by LSD1 inhibitor treatment. To further validate this mechanism, we are performing in vitro demethylase assay using synthetic peptides to confirm the LSD1-mediated demethylation on lys270. In summary, our results uncover a novel molecular mechanism of LSD1 action on FOXA1 chromatin binding, and reveal the global impact of LSD1-FOXA1 interaction on AR cistrome. This study will add another layer to the interplay between epigenetic modifiers and transcriptional factors and will provide important therapeutic implications of targeting LSD1 in PCa treatment.

Citation Format: Shuai Gao, Dong Han, Sujun Chen, Wanting Han, Housheng Hansen He, Steven P. Balk, Changmeng Cai. LSD1 inhibition disrupts FOXA1-dependent AR cistrome [abstract]. In: Proceedings of the AACR Special Conference: Prostate Cancer: Advances in Basic, Translational, and Clinical Research; 2017 Dec 2-5; Orlando, Florida. Philadelphia (PA): AACR; Cancer Res 2018;78(16 Suppl):Abstract nr A041.