Dataset Information

An unorthodox AR addiction drives enzalutamide resistance in prostate cancer (RNA-Seq)

ABSTRACT: Acquisition of resistance to the next-generation anti-androgen enzalutamide (ENZ) constitutes a major challenge for the treatment of castrate-resistant prostate cancer (CRPC). Most ENZ-resistant (ENZ-R) CRPCs are androgen receptor (AR)-positive (CRPCAR+), but often negative for prostate-specific antigen (PSA), a well-established surrogate of AR activity, implying the emergence of AR-indifferent disease. Through genome-wide ChIP-seq and RNA-seq profiling in disease-relevant ENZ-R CRPCAR+ cells, we identified a unique set of gained AR binding sites (ARBS-G) lacking the canonical DNA binding elements of AR and the pioneer factor FOXA1. ARBS-G loci are highly enriched with CpG islands and significantly overlapped with the binding sites of the unmethylated CpG dinucleotide-binding protein CXXC5 and the partner TET2. Expression of CXXC5 and its downstream targets including inhibitor of differentiation-1 (ID1) was upregulated in ENZ-R CRPCAR+ cell lines, patient-derived xenografts (PDXs) and patient specimens. Genetic depletion of AR, CXXC5 or ID1 restored ENZ sensitivity in human ENZ-R CRPCAR+ cells. Most importantly, ENZ-R CRPCAR+ cells, organoids, xenografts and PDXs were invariably hypersensitive to the novel BET-CBP/p300 dual inhibitor NEO2734. These results reveal that ENZ-R CRPCAR+ cancers are only indifferent to the canonical AR (cAR) activity, but remain addictive to the noncanonical AR (ncAR) function which is selectively vulnerable upon dual inhibition of CBP/p300 and BET family proteins.

ORGANISM(S): Homo sapiens

PROVIDER: GSE136129 | GEO | 2020/08/20

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Enzalutamide (ENZ) is a potent androgen receptor (AR) antagonist with activity in castration-resistant prostate cancer (CRPC); however, progression to ENZ-resistant (ENZ-R) CRPC frequently occurs with rising serum PSA levels, implicating AR full length (ARFL) or variants (AR-Vs) in disease progression. To define functional roles of ARFL and AR-Vs in ENZ-R CRPC, we designed 3 antisense oligonucleotides (ASO) targeting exon 1, intron 1, and exon 8 in AR pre-mRNA to knockdown either ARFL alone, or ARFL plus AR-Vs, and examined their respective effects in LNCaP-derived ENZ-R, as well as M12 and 22Rv1, cells. ENZ-R LNCaP xenografts express high levels of both ARFL and AR-V7 compared to CRPC LNCaP xenografts. In particular, ARFL levels were ~20-fold higher than AR-V7. ENZ-R LNCaP sub-lines, derived by selection from the ENZ xenografts, also expressed uniformly high levels of ARFL and AR-V7 compared to CRPC LNCaP cells. In addition, both ARFL and AR-V7 are highly expressed in the nuclear fractions of ENZ-R-LNCaP. In ENZ-R LNCaP cells, knockdown of ARFL alone, or ARFL plus AR-Vs, similarly induced apoptosis, suppressed cell growth and AR-regulated gene expression in vitro, and delayed tumour growth in vivo. In 22Rv1 cells that are inherently resistant to ENZ, knockdown of both ARFL and AR-Vs more potently suppressed cell growth, AR transcriptional activity and AR-regulated gene expression than knockdown of ARFL alone. These data indicate the AR is an important driver of ENZ resistance, and while the contributions of ARFL and AR-Vs can vary across cell systems, ARFL is the key driver in the ENZ-R LNCaP model. AR targeting strategies against both ARFL and AR-Vs is a rational approach for AR-dependent CRPC. CRPC and ENZ-R LNCaP xenografts were generated as described in: Gleave, M.E., et al., Serum prostate specific antigen levels in mice bearing human prostate LNCaP tumors are determined by tumor volume and endocrine and growth factors. Cancer Res, 1992. 52(6): p. 1598-605.

Project description:Prostate organogenesis is regulated by interactions between mesenchymal and epithelial cells, and via androgen signalling though androgen receptors (AR) expressed in mesenchymal cells. However, there is little knowledge of AR target genes and we have used high resolution genomic methods to examine AR function in subsets of mesenchyme during prostate development. Using ChIP-seq, we defined genomic AR binding sites (ARBS) in microdissected mesenchymal tissues of neonatal male and female rats during prostate growth. By comparing female tissue subsets (ventral mesenchymal pad and urethral smooth muscle) with male tissues (ventral prostate and dorsolateral prostate), we were able to identify sexually dimorphic ARBS. Females had a distinct AR binding profile from males with enrichment of ARBS proximal to gene transcriptional start sites as well as binding to non-classical AR binding sequence motifs. Using both tissue RNA-sequencing and single-cell RNA-sequencing of the same mesenchymal subsets, we identified 128 differentially expressed transcripts between males and females. Comparison of these to ChIP-seq ARBS allowed us to define sexually dimorphic AR target genes. Selected candidates were validated as sexually dimorphic at both the mRNA and protein level in both rat and human developing prostate tissues by qPCR and western blot. Response to testosterone stimulation was examined using ex vivo rat tissue organ cultures and qPCR. In single cell RNAseq data we observed subpopulations of mesenchymal cells in males and females as well as subpopulations common to both. The distribution of AR and target genes did not follow the subpopulation distribution, suggesting that AR action is not restricted by subset identity.Our study is the first to apply multiple transcriptomic approaches to prostate mesenchyme, and we have identified unique AR targets. We have identified sexually dimorphic, androgen responsive mesenchymal target genes that may underlie sexually dimorphic development of the prostate. We suggest that the analysis of transcription factor binding and transcriptomes in cell subsets will be informative when applied to tumour cells and stroma – since these show similar cellular heterogeneity and subset specific activity.

Dataset Information

An unorthodox AR addiction drives enzalutamide resistance in prostate cancer (RNA-Seq)

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure