Our Research

Cellular plasticity is an increasing concern as a mechanism of resistance in prostate tumors progressing on second generation AR targeted therapy. With the advent of treating prostate cancer (PCa) patients upfront with these agents, it is critical for the identification of mechanisms and actionable therapy targets that drive cellular plasticity. We have reported that loss of tristetraprolin (TTP, gene ZFP36), an RNA binding protein that regulates mRNA stability increases NFkB activation and is associated with higher rates of aggressive disease and early recurrence in primary PCa. Given ZFP36 loss occurs in early PCa we sought to examine the clinical and biological impact of ZFP36 loss combined with PTEN loss, a known driver of PCa initiation. Analysis of independent primary PCa cohorts demonstrated that combined loss of PTEN and ZFP36 expression coincided with increased risk of recurrence. Engineering ZFP36 deletion specific to prostate luminal epithelial cells in vivo induced high-grade prostatic intraepithelial neoplasia (HG-PIN), and when combined with PTEN deletion resulted in rapid progression to invasive adenocarcinoma associated with increased proliferation and development of micro-metastases. Combined loss of ZFP36and PTEN in mice significantly reduced overall survival as well as time to progression when mice were surgically castrated. Gene expression analysis revealed that loss of ZFP36 expression alters the cell state that is driven by PTEN loss, demonstrated by positive enrichment of gene sets including EMT, Inflammation, TNFa/NFkB, IL6-JAK/STAT3. Surprisingly, ZFP36 loss also induced enrichment of multiple gene sets involved in mononuclear cell migration, chemotaxis, and proliferation. Use of the NFkB inhibitor, dimethylaminoparthenolide (DMAPT) induced significant therapeutic responses in tumorswith PTEN and ZFP36 co-loss and reversed resistance to castration. To date, this work identifies a novel molecular mechanism which drives resistance to castration through loss of ZFP36 expression driving the induction of cellular plasticity, that can be reversed by inhibition of NFkB activity using DMAPT.

Second-generation androgen deprivation therapies (ADT) have provided significant life-extension for patients with metastatic castration resistant prostate cancer (mCRPC), but unfortunately tumors progress via therapy resistance and currently no therapies provide durable response. Approximately between 15-20% of these mCRPC are independent of AR activity via a mechanism termed phenotypic plasticity, which activates alternative transcriptomic programs associated with loss of function of the tumor suppressor genes RB1 and/or TP53. Genomic loss of RB1 is predictor of poor survival, whereas alterations in RB1 and TP53 are associated with shorter response to ADT. These tumors often display altered kinase signaling and multilineage states including neuroendocrine, stem, and basal-like gene signatures. Using genetically engineered mouse models (GEMMs) devoid of Pten and Rb1 (DKO mice), we previously demonstrated that the reprogramming factor enhancer of zeste homolog 2 (EZH2) regulates alternative transcription programs promoting phenotypic plasticity. Here, our overall goal was to better understand the role of EZH2 in reprogramming and how this could be exploited therapeutically.

Prostate cancer (PCa) is the second leading cause of cancer-related deaths in men in the US. PCa initiation and progression is largely dependent on androgen receptor (AR) expression and function. Androgen-deprivation therapy (ADT) which can target both AR and the androgen biosynthesis pathways is the first-line therapy for PCa. However, ADT is not curative of metastatic disease and approximately 15 to 20% of patients exhibit therapy resistance involving independence of AR signaling (CRPC-AI). This phenotype referred to as phenotypic plasticity is associated with tumors displaying neuroendocrine-like features, stem or basal cell-like phenotype, altered kinase signaling, and characteristic epigenetic alterations. Preclinical and clinical data demonstrate that CRPC-AI is predominantly driven by the combinatorial loss-of-function (LOF) mutations of PTEN, TP53, and RB tumor suppressor genes. About 70% to 90% of CRPC-AI/small cell prostate cancer tumors with phenotypic plasticity acquire RB loss of function mutations and is currently the strongest prognostic marker for worst overall survival in patients; however, there are no therapeutic options to provide durable response in patients. To discover genetic vulnerabilities that can be translated to therapeutics, we performed a genome-wide CRISPR/Cas9 knockout screen to identify genes essential for proliferation or survival using a Rb-deficient prostate cancer cell line derived from the DKO GEMM (PbCre;Ptenlox/lox;Rblox/lox). We found that Rb-deficiency induced a genetic dependency on DNA damage repair (DDR) kinases including ATM, ATR, CHK1 (HR signaling). ATR inhibition in Rb-deficient cells was associated with prolonged DNA damage, increased double-strand DNA structures, and an interferon gene response indictive of viral mimicry. We validated the expression of two well-known interferon response genes by flow cytometry (PD-L1 and MHC-I) and demonstrated that their expression was dependent on Stimulator of Interferon Genes (STING). In addition, this response to ATR inhibition was dependent on expression and function of wild-type Tp53. Suspecting epigenetic reprograming, we focused on two major epigenetic programs mediating chromatin remodeling in phenotypic plasticity in prostate cancer that were highlighted from our CRISPR/Cas9 screen – polycomb gene repression and DNA methylation. Inhibition of EZH2 (polycomb) failed to rescue the viral mimicry phenotype (data not shown), whereas inhibition of DNA methyltransferase-1 (DNMT1) successfully rescued the interferon gene response associated with viral mimicry. Further, combination of ATRi and DNMT1i in models with Tp53 LOF demonstrated drug synergy. Currently, our data implies that targeting prostate cancers with RB1 LOF could be responsive to DDR targeted therapies, however this sensitivity is lost with prostate cancer progression involving TP53 LOF. Loss of TP53 promotes further tumor evolution via chromatin remodeling involving DNA methylation. Inhibition of DDR pathways and DNMT1 induces viral mimicry interferon response genes and provides proof of concept that this combination approach can promote tumor response to immunotherapy.