Human prostate luminal cell differentiation requires NOTCH3 induction by p38-MAPK and MYC.

Many pathways dysregulated in prostate cancer are also involved in epithelial differentiation. To better understand prostate tumor initiation, we sought to investigate specific genes and mechanisms required for normal basal to luminal cell differentiation. Utilizing human prostate basal epithelial cells and an in vitro differentiation model, we tested the hypothesis that regulation of NOTCH3 by the p38 MAPK family (hereafter p38-MAPK), via MYC, is required for luminal differentiation. Inhibition (SB202190 and BIRB796) or knockdown of p38α (also known as MAPK14) and/or p38δ (also known as MAPK13) prevented proper differentiation. Additionally, treatment with a γ-secretase inhibitor (RO4929097) or knockdown of NOTCH1 and/or NOTCH3 greatly impaired differentiation and caused luminal cell death. Constitutive p38-MAPK activation through MKK6(CA) increased NOTCH3 (but not NOTCH1) mRNA and protein levels, which was diminished upon MYC inhibition (10058-F4 and JQ1) or knockdown. Furthermore, we validated two NOTCH3 enhancer elements through a combination of enhancer (e)RNA detection (BruUV-seq) and luciferase reporter assays. Finally, we found that the NOTCH3 mRNA half-life increased during differentiation or upon acute p38-MAPK activation. These results reveal a new connection between p38-MAPK, MYC and NOTCH signaling, demonstrate two mechanisms of NOTCH3 regulation and provide evidence for NOTCH3 involvement in prostate luminal cell differentiation.

A streamlined method for the design and cloning of shRNAs into an optimized Dox-inducible lentiviral vector.

BACKGROUND: Short hairpin RNA (shRNA) is an established and effective tool for stable knock down of gene expression. Lentiviral vectors can be used to deliver shRNAs, thereby providing the ability to infect most mammalian cell types with high efficiency, regardless of proliferation state. Furthermore, the use of inducible promoters to drive shRNA expression allows for more thorough investigations into the specific timing of gene function in a variety of cellular processes. Moreover, inducible knockdown allows the investigation of genes that would be lethal or otherwise poorly tolerated if constitutively knocked down. Lentiviral inducible shRNA vectors are readily available, but unfortunately the process of cloning, screening, and testing shRNAs can be time-consuming and expensive. Therefore, we sought to refine a popular vector (Tet-pLKO-Puro) and streamline the cloning process with efficient protocols so that researchers can more efficiently utilize this powerful tool. METHODS: First, we modified the Tet-pLKO-Puro vector to make it easy ("EZ") for molecular cloning (EZ-Tet-pLKO-Puro). Our primary modification was to shrink the stuffer region, which allows vector purification via polyethylene glycol precipitation thereby avoiding the need to purify DNA through agarose. In addition, we generated EZ-Tet-pLKO vectors with hygromycin or blasticidin resistance to provide greater flexibility in cell line engineering. Furthermore, we provide a detailed guide for utilizing these vectors, including shRNA design strategy and simplified screening methods. RESULTS: Notably, we emphasize the importance of loop sequence design and demonstrate that the addition of a single mismatch in the loop stem can greatly improve shRNA efficiency. Lastly, we display the robustness of the system with a doxycycline titration and recovery time course and provide a cost/benefit analysis comparing our system with purchasing pre-designed shRNA vectors. CONCLUSIONS: Our aim was twofold: first, to take a very useful shRNA vector and make it more amenable for molecular cloning and, secondly, to provide a streamlined protocol and rationale for cost-effective design, cloning, and screening of shRNAs. With this knowledge, anyone can take advantage of this powerful tool to inducibly knockdown any gene of their choosing.

SND1 binds to ERG and promotes tumor growth in genetic mouse models of prostate cancer.

SND1 and MTDH are known to promote cancer and therapy resistance, but their mechanisms and interactions with other oncogenes remain unclear. Here, we show that oncoprotein ERG interacts with SND1/MTDH complex through SND1's Tudor domain. ERG, an ETS-domain transcription factor, is overexpressed in many prostate cancers. Knocking down SND1 in human prostate epithelial cells, especially those overexpressing ERG, negatively impacts cell proliferation. Transcriptional analysis shows substantial overlap in genes regulated by ERG and SND1. Mechanistically, we show that ERG promotes nuclear localization of SND1/MTDH. Forced nuclear localization of SND1 prominently increases its growth promoting function irrespective of ERG expression. In mice, prostate-specific Snd1 deletion reduces cancer growth and tumor burden in a prostate cancer model (PB-Cre/Ptenflox/flox/ERG mice), Moreover, we find a significant overlap between prostate transcriptional signatures of ERG and SND1. These findings highlight SND1's crucial role in prostate tumorigenesis, suggesting SND1 as a potential therapeutic target in prostate cancer.

Selective androgen receptor modulators activate the canonical prostate cancer androgen receptor program and repress cancer growth.

Prostate cancer (PC) is driven by androgen receptor (AR) activity, a master regulator of prostate development and homeostasis. Frontline therapies for metastatic PC deprive the AR of the activating ligands testosterone (T) and dihydrotestosterone (DHT) by limiting their biosynthesis or blocking AR binding. Notably, AR signaling is dichotomous, inducing growth at lower activity levels, while suppressing growth at higher levels. Recent clinical studies have exploited this effect by administration of supraphysiological concentrations of T, resulting in clinical responses and improvements in quality of life. However, the use of T as a therapeutic agent in oncology is limited by poor drug-like properties as well as rapid and variable metabolism. Here, we investigated the antitumor effects of selective AR modulators (SARMs), which are small-molecule nonsteroidal AR agonists developed to treat muscle wasting and cachexia. Several orally administered SARMs activated the AR program in PC models. AR cistromes regulated by steroidal androgens and SARMs were superimposable. Coregulatory proteins including HOXB13 and GRHL2 comprised AR complexes assembled by both androgens and SARMs. At bioavailable concentrations, SARMs repressed MYC oncoprotein expression and inhibited the growth of castration-sensitive and castration-resistant PC in vitro and in vivo. These results support further clinical investigation of SARMs for treating advanced PC.

Genomic attributes of homology-directed DNA repair deficiency in metastatic prostate cancer.

Cancers with homology-directed DNA repair (HRR) deficiency exhibit high response rates to poly(ADP-ribose) polymerase inhibitors (PARPi) and platinum chemotherapy. Though mutations disrupting BRCA1 and BRCA2 associate with HRR deficiency (HRRd), patterns of genomic aberrations and mutation signatures may be more sensitive and specific indicators of compromised repair. Here, we evaluated whole-exome sequences from 418 metastatic prostate cancers (mPCs) and determined that one-fifth exhibited genomic characteristics of HRRd that included Catalogue Of Somatic Mutations In Cancer mutation signature 3. Notably, a substantial fraction of tumors with genomic features of HRRd lacked biallelic loss of a core HRR-associated gene, such as BRCA2. In this subset, HRRd associated with loss of chromodomain helicase DNA binding protein 1 but not with mutations in serine-protein kinase ATM, cyclin dependent kinase 12, or checkpoint kinase 2. HRRd genomic status was strongly correlated with responses to PARPi and platinum chemotherapy, a finding that supports evaluating biomarkers reflecting functional HRRd for treatment allocation.

Notch3 promotes prostate cancer-induced bone lesion development via MMP-3.

Prostate cancer metastases primarily localize in the bone where they induce a unique osteoblastic response. Elevated Notch activity is associated with high-grade disease and metastasis. To address how Notch affects prostate cancer bone lesions, we manipulated Notch expression in mouse tibia xenografts and monitored tumor growth, lesion phenotype, and the bone microenvironment. Prostate cancer cell lines that induce mixed osteoblastic lesions in bone expressed 5-6 times more Notch3, than tumor cells that produce osteolytic lesions. Expression of active Notch3 (NICD3) in osteolytic tumors reduced osteolytic lesion area and enhanced osteoblastogenesis, while loss of Notch3 in osteoblastic tumors enhanced osteolytic lesion area and decreased osteoblastogensis. This was accompanied by a respective decrease and increase in the number of active osteoclasts and osteoblasts at the tumor-bone interface, without any effect on tumor proliferation. Conditioned medium from NICD3-expressing cells enhanced osteoblast differentiation and proliferation in vitro, while simultaneously inhibiting osteoclastogenesis. MMP-3 was specifically elevated and secreted by NICD3-expressing tumors, and inhibition of MMP-3 rescued the NICD3-induced osteoblastic phenotypes. Clinical osteoblastic bone metastasis samples had higher levels of Notch3 and MMP-3 compared with patient matched visceral metastases or osteolytic metastasis samples. We identified a Notch3-MMP-3 axis in human prostate cancer bone metastases that contributes to osteoblastic lesion formation by blocking osteoclast differentiation, while also contributing to osteoblastogenesis. These studies define a new role for Notch3 in manipulating the tumor microenvironment in bone metastases.

Supraphysiological androgens suppress prostate cancer growth through androgen receptor-mediated DNA damage.

Prostate cancer (PC) is initially dependent on androgen receptor (AR) signaling for survival and growth. Therapeutics designed to suppress AR activity serve as the primary intervention for advanced disease. However, supraphysiological androgen (SPA) concentrations can produce paradoxical responses leading to PC growth inhibition. We sought to discern the mechanisms by which SPA inhibits PC and to determine if molecular context associates with anti-tumor activity. SPA produced an AR-mediated, dose-dependent induction of DNA double-strand breaks (DSBs), G0/G1 cell cycle arrest and cellular senescence. SPA repressed genes involved in DNA repair and delayed the restoration of damaged DNA which was augmented by PARP1 inhibition. SPA-induced DSBs were accentuated in BRCA2-deficient PCs, and combining SPA with PARP or DNA-PKcs inhibition further repressed growth. Next-generation sequencing was performed on biospecimens from PC patients receiving SPA as part of ongoing Phase II clinical trials. Patients with mutations in genes mediating homology-directed DNA repair were more likely to exhibit clinical responses to SPA. These results provide a mechanistic rationale for directing SPA therapy to PCs with AR amplification or DNA repair deficiency, and for combining SPA therapy with PARP inhibition.

Receptor tyrosine kinase Met promotes cell survival via kinase-independent maintenance of integrin α3ß1.

Matrix adhesion via integrins is required for cell survival. Adhesion of epithelial cells to laminin via integrin α3ß1 was previously shown to activate at least two independent survival pathways. First, integrin α3ß1 is required for autophagy-induced cell survival after growth factor deprivation. Second, integrin α3ß1 independently activates two receptor tyrosine kinases, EGFR and Met, in the absence of ligands. EGFR signaling to Erk promotes survival independently of autophagy. To determine how Met promotes cell survival, we inhibited Met kinase activity or blocked its expression with RNA interference. Loss of Met expression, but not inhibition of Met kinase activity, induced apoptosis by reducing integrin α3ß1 levels, activating anoikis, and blocking autophagy. Met was specifically required for the assembly of autophagosomes downstream of LC3II processing. Reexpression of wild-type Met, kinase-dead Met, or integrin α3 was sufficient to rescue death upon removal of endogenous Met. Integrin α3ß1 coprecipitated and colocalized with Met in cells. The extracellular and transmembrane domain of Met was required to fully rescue cell death and restore integrin α3 expression. Thus Met promotes survival of laminin-adherent cells by maintaining integrin α3ß1 via a kinase-independent mechanism.

Transient induction of ING4 by Myc drives prostate epithelial cell differentiation and its disruption drives prostate tumorigenesis.

The mechanisms by which Myc overexpression or Pten loss promotes prostate cancer development are poorly understood. We identified the chromatin remodeling protein, ING4, as a crucial switch downstream of Myc and Pten that is required for human prostate epithelial differentiation. Myc-induced transient expression of ING4 is required for the differentiation of basal epithelial cells into luminal cells, while sustained ING4 expression induces apoptosis. ING4 expression is lost in >60% of human primary prostate tumors. ING4 or Pten loss prevents epithelial cell differentiation, which was necessary for tumorigenesis. Pten loss prevents differentiation by blocking ING4 expression, which is rescued by ING4 re-expression. Pten or ING4 loss generates tumor cells that co-express basal and luminal markers, indicating prostate oncogenesis occurs through disruption of an intermediate step in the prostate epithelial differentiation program. Thus, we identified a new epithelial cell differentiation switch involving Myc, Pten, and ING4, which when disrupted leads to prostate tumorigenesis. Myc overexpression and Pten loss are common genetic abnormalities in prostate cancer, whereas loss of the tumor suppressor ING4 has not been reported. This is the first demonstration that transient ING4 expression is absolutely required for epithelial differentiation, its expression is dependent on Myc and Pten, and it is lost in the majority of human prostate cancers. This is the first demonstration that loss of ING4, either directly or indirectly through loss of Pten, promotes Myc-driven oncogenesis by deregulating differentiation. The clinical implication is that Pten/ING4 negative and ING4-only negative tumors may reflect two distinct subtypes of prostate cancer.

Disruption of prostate epithelial differentiation pathways and prostate cancer development.

One of the foremost problems in the prostate cancer (PCa) field is the inability to distinguish aggressive from indolent disease, which leads to difficult prognoses and thousands of unnecessary surgeries. This limitation stems from the fact that the mechanisms of tumorigenesis in the prostate are poorly understood. Some genetic alterations are commonly reported in prostate tumors, including upregulation of Myc, fusion of Ets genes to androgen-regulated promoters, and loss of Pten. However, the specific roles of these aberrations in tumor initiation and progression are poorly understood. Likewise, the cell of origin for PCa remains controversial and may be linked to the aggressive potential of the tumor. One important clue is that prostate tumors co-express basal and luminal protein markers that are restricted to their distinct cell types in normal tissue. Prostate epithelium contains layer-specific stem cells as well as rare bipotent cells, which can differentiate into basal or luminal cells. We hypothesize that the primary oncogenic cell of origin is a transient-differentiating bipotent cell. Such a cell must maintain tight temporal and spatial control of differentiation pathways, thus increasing its susceptibility for oncogenic disruption. In support of this hypothesis, many of the pathways known to be involved in prostate differentiation can be linked to genes commonly altered in PCa. In this article, we review what is known about important differentiation pathways (Myc, p38MAPK, Notch, PI3K/Pten) in the prostate and how their misregulation could lead to oncogenesis. Better understanding of normal differentiation will offer new insights into tumor initiation and may help explain the functional significance of common genetic alterations seen in PCa. Additionally, this understanding could lead to new methods for classifying prostate tumors based on their differentiation status and may aid in identifying more aggressive tumors.