Reprogramming androgen receptor and lysine-specific demethylase 1 transcriptome in castration-resistant prostate cancer
Rizvi, Abbas Ali
MetadataShow full item record
Patients who undergo androgen deprivation therapy (ADT) for the treatment of prostate cancer (PCa) often recur into castration-resistance prostate cancer (CRPC), which is clinically lethal. As PCa are inherently dependent upon AR activity for survival and progression ADT entails suppression of the androgen receptor (AR) signaling. While AR-directed therapies initially inhibit tumor growth, de novo resistance or development of acquired resistances to therapy is reached through restored AR function. Lysine-specific demethylase 1 (LSD1/KDM1A) is an AR coregulatory protein, such that when recruited by AR, it acts as an AR coactivator for transcription. LSD1 is overexpressed in a various types of aggressive cancers, including PCa. We hypothesize that LSD1 alters AR downstream signaling defining a transcriptional lesion that promotes malignant transformation and progression in prostate epithelial cells. Pharmacological inhibition of AR and LSD1 may offer a therapeutic advantage for patients with CRPC. Our aims are: (1) to profile the transcriptional alterations associated with AR and LSD1 interactome using publically available (TCGA) PCa patient data, (2) to determine the anti-cancer effect of LSD1 inhibition combined with AR blockade in C4-2 cells, and (3) to profile the AR- and LSD1-mediated transcriptome in C4-2 cells. Aim 1) to determine the differentially expressed genes (DEGs) between normal prostate and tumor, RNAseq data from the TCGA, comprising of 374 tumor samples, and 52 matched tumor to normal prostate tissue samples. The signature studied includes genes interacting with AR and LSD1, obtained from NCBI. The DEGs were calculated using the statistical computing language R (limma package) with a false discovery rate (FDR) of 0.05 and a log fold-change > +/- 1. Aim 2) LNCaP C4-2 (C4-2), androgen independent CRPC cell line was used. AR and LSD1 were targeted simultaneously in C4-2 cells with IC25 of the AR inhibitor bicalutamide and IC25 of the LSD1 inhibitor tranylcypromine (TCP). Results were further validated with LSD1 knock down in C4-2 cells and treatment with IC25 bicalutamide for 96 hours. Aim 3) RNA-Seq was conducted to determine differentially expressed transcripts (DETs) between four conditions in C4-2 cells: (1) shCTR + vehicle vs. shCTR + DHT, (2) shCTR + vehicle vs. shLSD1 + vehicle, (3) shLSD1 + vehicle vs. shLSD1 + DHT, and (4) shCTR + DHT vs. shLSD1 + DHT. The DETs were calculated using R edgeR with an FDR of 0.05. Subsequent enrichment analyses to understand the pathways involved between these conditions was conducted. Fifty-one total DEGs were identified from AR and LSD1 signature. EZH2 was identified as the only overlapping DEG between AR (22 DEGs) and LSD1 (29 DEGs) signatures. Simulation analysis revealed that the observed frequency of DEGs in our signatures was significantly lower than expected by chance. The combinatorial approach by treating C4-2 cells with bicalutamide and TCP showed an enhanced anti-proliferative effect, greater than the single treatment alone. Gene expression and protein levels of LSD1 were successfully reduced in the C4-2 shLSD1 model. C4-2 shLSD1 cells showed a significant (p-value < 0.05) reduction in proliferation when treated with IC25 bicalutamide and compared to vehicle treated cells. Our work has shown that AR and LSD1 form a unique signature that includes genes that are differentially expressed between normal and tumor prostate tissue samples. A combinatorial therapeutic approach appears to be more efficient in reducing PCa cells proliferation than individual agents alone, suggesting that the profiling of AR and LSD1 transcriptional networks could lead to further identification of novel drug targets for patients with aggressive PCa. Our RNA-Seq experiment revealed a balance in up- and downregulated DETs between all conditions, indicating that LSD1 has dual coregulatory functions. AR activation alone yielded pathways that were involved in carcinogenesis and tumorigenesis. The effect of LSD1 knockdown when AR was activated revealed that LSD1 affects pathways that are involved in tumor progression into a metastatic phenotype. Overall these results suggest that AR alterations could prime prostate epithelial cells for malignant transformations while LSD1 drives the progression to aggressive disease.