Treatment-induced neuroendocrine transdifferentiation (NEtD) complicates therapies for metastatic prostate cancer (PCa). Based on evidence that PCa cells can transdifferentiate to other neuroectodermally-derived cell lineages in vitro, we proposed that NEtD requires first an intermediary reprogramming to metastable cancer stem-like cells (CSCs) of a neural class and we demonstrate that several different AR+/PSA+ PCa cell lines were efficiently reprogrammed to, maintained and propagated as CSCs by growth in androgen-free neural/neural crest (N/NC) stem medium. Such reprogrammed cells lost features of prostate differentiation; gained features of N/NC stem cells and tumor-initiating potential; were resistant to androgen signaling inhibition; and acquired an invasive phenotype in vitro and in vivo. When placed back into serum-containing mediums, reprogrammed cells could be re-differentiated to N-/NC-derived cell lineages or return back to an AR+ prostate-like state. Once returned, the AR+ cells were resistant to androgen signaling inhibition. Acute androgen deprivation or anti-androgen treatment in serum-containing medium led to the transient appearance of a sub-population of cells with similar characteristics. Finally, a 132 gene signature derived from reprogrammed PCa cell lines distinguished tumors from PCa patients with adverse outcomes. This model may explain neural manifestations of PCa associated with lethal disease. The metastable nature of the reprogrammed stem-like PCa cells suggests that cycles of PCa cell reprogramming followed by re-differentiation may support disease progression and therapeutic resistance. The ability of a gene signature from reprogrammed PCa cells to identify tumors from patients with metastasis or PCa-specific mortality implies that developmental reprogramming is linked to aggressive tumor behaviors.

Purpose: Patients with metastatic prostate cancer are increasingly presenting with treatment-resistant, androgen receptor-negative/ low (AR À/Low ) tumors, with or without neuroendocrine characteristics, in processes attributed to tumor cell plasticity. This plasticity has been modeled by Rb1/p53 knockdown/knockout and is accompanied by overexpression of the pluripotency factor, Sox2. Here, we explore the role of the developmental transcription factor Sox9 in the process of prostate cancer therapy response and tumor progression.Experimental Design: Unique prostate cancer cell models that capture AR À/Low stem cell-like intermediates were analyzed for features of plasticity and the functional role of Sox9. Human prostate cancer xenografts and tissue microarrays were evaluated for temporal alterations in Sox9 expression. The role of NF-kB pathway activity in Sox9 overexpression was explored.Results: Prostate cancer stem cell-like intermediates have reduced Rb1 and p53 protein expression and overexpress Sox2 as well as Sox9.Sox9 was required for spheroid growth, and overexpression increased invasiveness and neural features of prostate cancer cells. Sox9 was transiently upregulated in castration-induced progression of prostate cancer xenografts and was specifically overexpressed in neoadjuvant hormone therapy (NHT)-treated patient tumors. High Sox9 expression in NHT-treated patients predicts biochemical recurrence. Finally, we link Sox9 induction to NF-kB dimer activation in prostate cancer cells.Conclusions: Developmentally reprogrammed prostate cancer cell models recapitulate features of clinically advanced prostate tumors, including downregulated Rb1/p53 and overexpression of Sox2 with Sox9. Sox9 is a marker of a transitional state that identifies prostate cancer cells under the stress of therapeutic assault and facilitates progression to therapy resistance. Its expression may index the relative activity of the NF-kB pathway.

Despite the substantial benefit of androgen deprivation therapy (ADT) for metastatic prostate cancer, patients often progress to castration-resistant disease (CRPC) that is more difficult to treat. CRPC is associated with renewed androgen receptor activity in tumor cells and restoration of tumor androgen levels through acquired intratumoral steroidogenesis (AIS). Although prostate cancer (PCa) cells have been shown to have steroidogenic capability in vitro, we previously found that benign prostate stromal cells (PrSCs) can also synthesize testosterone (T) from an adrenal precursor, DHEA, when stimulated with a hedgehog (Hh) pathway agonist, SAG. Here, we show exposure of PrSCs to a different Smoothened (Smo) agonist, Ag1.5, or to conditioned medium from sonic hedgehog overexpressing LNCaP cells induces steroidogenic enzyme expression in PrSCs and significantly increases production of T and its precursor steroids in a Smo-dependent manner from 22-OH-cholesterol substrate. Hh agonist-/ligand-treated PrSCs produced androgens at a rate similar to or greater than that of PCa cell lines. Likewise, primary bone marrow stromal cells became more steroidogenic and produced T under the influence of Smo agonist. Treatment of mice bearing LNCaP xenografts with a Smo antagonist, TAK-441, delayed the onset of CRPC after castration and substantially reduced androgen levels in residual tumors. These outcomes support the idea that stromal cells in ADT-treated primary or metastatic prostate tumors can contribute to AIS as a consequence of a paracrine Hh signaling microenvironment. As such, Smo antagonists may be useful for targeting prostate tumor stromal cell-derived AIS and delaying the onset of CRPC after ADT.

Our findings indicate that Gli2 overexpression in PCa cells might support development of castration resistant PCa through AR co-activation and suggests that AR might modulate transcription from Gli2.

Hedgehog (Hh) is an oncogenic signaling pathway that regulates the activity of Gli transcription factors. Canonical Hh is a Smoothened- (Smo-) driven process that alters the post-translational processing of Gli2/Gli3 proteins. Though evidence supports a role for Gli action in prostate cancer (PCa) cell growth and progression, there is little indication that Smo is involved. Here we describe a non-canonical means for activation of Gli transcription in PCa cells mediated by the binding of transcriptionally-active androgen receptors (ARs) to Gli3. Androgens stimulated reporter expression from a Gli-dependent promoter in a variety of AR + PCa cells and this activity was suppressed by an anti-androgen, Enz, or by AR knockdown. Androgens also upregulated expression of endogenous Gli-dependent genes. This activity was associated with increased intranuclear binding of Gli3 to AR that was antagonized by Enz. Fine mapping of the AR binding domain on Gli2 showed that AR recognizes the Gli protein processing domain (PPD) in the C-terminus. Mutations in the arginine-/serine repeat elements of the Gli2 PPD involved in phosphorylation and ubiquitinylation blocked the binding to AR. β-TrCP, a ubiquitin ligase that recognizes the Gli PPD, competed with AR for binding to this site. AR binding to Gli3 suppressed its proteolytic processing to the Gli3 repressor form (Gli3R) whereas AR knockdown increased Gli3R. Both full-length and truncated ARs were able to activate Gli transcription. Finally, we found that an ARbinding decoy polypeptide derived from the Gli2 C-terminus can compete with Gli3 for binding to AR. Exogenous overexpression of this decoy suppressed Gli transcriptional activity in PCa cells. Collectively, this work identifies a novel pathway for non-canonical activation of Hh signaling in PCa cells and identifies a means for interference that may have clinical relevance for PCa patients.