Despite recent advances in treatment modalities for advanced prostate cancer (PCa) and castration-resistant prostate cancer (CRPC), PCa continues to be a major cause of morbidity and mortality in American men. An improved understanding of the tumor microenvironment and immune surveillance has opened new avenues for targeted therapeutic strategies for the treatment of PCa. The primary cause of treatment failure is the development of cellular resistance to the cytotoxic effects of cancer therapeutics. Both cell intrinsic and extrinsic mechanisms can underlie this acquired therapeutic resistance. More recently, the tumor microenvironment, and more importantly tumor-related inflammation, have emerged as a crucial players in promoting cancer initiation and progression, invasion and metastasis, and resistance to both conventional and targeted therapies. In particular, there have been fundamental leaps in our understanding of the complex roles of tumor-infiltrating myeloid cells (TIMs). TIMs are a heterogeneous population of monocytic and granulocytic myeloid derived suppressor cells (MDSCs) and tumor-associated macrophages (TAMs) that have a profound impact on tumor development. TIMs have been reported to promote tumor growth. In PCa, first-line radiotherapy (RT) and blockade of androgen signaling have potent anti-cancer effects, but PCa patients inevitably develop resistance mechanisms to evade therapy. Recently, elevated inflammation has been demonstrated after RT and androgen inhibition, yet the impact of this aberrant inflammation and the potential therapeutic options for preventing resistance in PCa have been largely unexplored. Therefore, we sought to further understand the impact of inflammation in acquired resistance to these two important therapeutic strategies, and whether combinatorial therapies may prevent resistance and improve therapeutic effectiveness. Our laboratory and others have shown a critical role for macrophage colony-stimulating factor-1 (M-CSF-1, CSF-1) signaling through its receptor, CSF1R, in the recruitment of TIMs to tumors, which promote cancer progression. Using murine models of prostate cancer, we found that RT resulted in pronounced increases in CSF-1 expression in PCa cells, which was transcriptionally regulated by the non-receptor tyrosine kinase, ABL1 (c-Abl), highly correlating with increased TIM recruitment. Importantly, utilizing a CSF1R inhibitor, PLX3397, in combination with RT resulted in significant depletion of TIMs and delayed RT tumor recurrence. In parallel, the potential impact of TIMs in therapeutic resistance to androgen blockade therapy (ABT) was also interrogated. Similarly to RT, we found that inhibiting androgen signaling using the novel androgen receptor inhibitor MDV3100 (Enzalutamide) resulted in increased expression anti-inflammatory cytokine IL-10 and Th2 cytokine IL-13 in PCa cells, as well as CSF-1 expression in PCa cells and in mice sera. Moreover, cellular interactions between macrophages and PCa cells treated with MDV3100 skewed macrophages to a protumorigenic "alternative" activation state. In addition, tumors grown in surgically castrated mice showed increased CSF-1 expression in serum and in whole tumors compared to sham surgery mice, resulting in a considerable increase in TAM infiltration. More importantly, TAM blockade using PLX3397 in combination with castration resulted in a significant delay in the onset of CRPC. Lastly, we sought to further investigate potential mechanisms underlying TIM-mediated resistance to androgen depletion. Cytotoxic effects of androgen depletion are primarily due to induction of apoptosis after inhibition of androgen receptor (AR) signaling. Insulin-like growth factor 1 receptor (IGF1R) signaling is a well-established mode of CRPC development due to its anti-apoptotic and mitogenic functions. Furthermore, the Th2 cytokine IL-13, is known to induce IGF-1 expression in macrophages. Interestingly, we found that the increased IL-13 expression from MDV3100 treated Myc-CaP cells, correlated with increased macrophage IGF-1 expression. Subsequently, macrophage-derived IGF-1 levels correlated with increased p-IGF1R staining in tumors from castrated mice versus sham surgery controls. Additionally, tumors grown in castrated mice showed a significant increase in IGF-1 expression, and a subsequent reduction of IGF-1 levels upon macrophage depletion with PLX3397. Macrophage depletion also correlated with decreases in Ki67 proliferation index, as well as delayed onset of CPRC. Collectively, these data suggest that a strong interplay between TIMs and tumor cells generate a pro-tumor microenvironment that favors therapeutic resistance in PCa.

Overall, these studies increase our understanding of the complex role of TIMs in the prostatic microenvironment, and argue in favor of their contribution to PCa progression and therapeutic resistance to RT and therapies targeting androgen signaling. The rational targeted therapies pursued here and other combinatorial approaches hold promise to improve the long-term efficacy of new therapeutics and ultimately improve patient lives.

Clear cell renal cell carcinoma (ccRCC) is responsible for tens of thousands of deaths annually. Immune-based therapies have yielded the only durable responses and some encouraging recent results. There are few immunocompetent murine models for this disease. Attempts to generate a ccRCC model through von Hippel Lindau (VHL) inactivation have produced modest results. None of these studies developed metastatic carcinoma. The RENCA cell line is the major murine immunocompetent model used, but it has wild type VHL. In order to create an improved immunocompetent murine model for ccRCC, we used the clustered regularly interspaced short palindromic repeat (CRISPR) technology to knock out VHL in the RENCA cell line. VHL knockout led to morphological and molecular changes indicative of an epithelial-mesenchymal transition (EMT). We went on to show that this leads to enhanced migratory and invasive potential in vitro as well as increased metastasis to the lungs in vivo. Double knockout studies using CRISPR revealed that these changes were hypoxia-inducible factor-1α (HIF-1α)-dependent. In performing our animal studies, we discovered that stable integration of our Cas9-containing expression cassette produced an immune rejection. Use of integrase deficient lentivirus followed by clonal selection allowed us to produce a VHL knockout cell line that was not expressing Cas9. We submitted our cell lines for RNA sequencing (RNA-Seq) and evaluated the genes and pathways that were upregulated upon VHL knockout. We found that the ten most upregulated genes in our cell lines corresponded to poor survival in The Cancer Genome Atlas (TCGA) data set. Further analysis reveals potential genes and pathways worthy of further study. Taken together, the studies presented herein demonstrate CRISPR-mediated disruption of VHL and the generation of a more aggressive, clinically relevant ccRCC model based on the RENCA cell line.

Metastasis occurs in one-third of clear cell renal cell carcinoma (ccRCC) patients and accounts for the majority of ccRCC-related deaths. Unfortunately, no therapy is available for this lethal disease because the molecular mechanism of metastasis is not fully understood. Although it has been known for 25 years that the loss of function of the VHL tumor suppressor gene is pathognomonic of RCC, its involvement in malignant progression remains elusive. Using a CRISPR/ Cas9 gene editing system and a mouse orthotopic implantation model, we previously found that the intratumoral heterogeneity of VHL expression is essential in ccRCC metastasis. Specifically, the VHL(-) cell population serves as the metastatic driver. Therefore, in this study, we sought to develop 1) a novel animal model of cooperative ccRCC metastasis and 2) small molecules with selective toxicity against the VHL(-) cells. For the first aim, we have utilized the chicken chorioallantoic membrane (CAM) model to establish a novel, cost- and labor-efficient animal model of cooperative metastasis, which almost perfectly recapitulated the characteristic phenomena observed in the mouse orthotopic implantation model. Using this model, we have also further explored exosome’s involvement in cooperative metastasis.For the second part of the project, we hypothesize that targeting the metastatic driver population, VHL(-), will effectively inhibit metastases. We have utilized high-throughput screening (HTS) for this purpose. A ratiometric black-box approach was employed to identify selective modulators of cell growth and survival of murine VHL(-) ccRCC cell lines with little to no effect on VHL(+) cells. Statins, the commonly used cholesterol-lowering drugs, showed up as hits. Dosage curves confirmed this result, and animal studies suggest its efficacy against ccRCC metastasis. The mechanism of action (MOA) is currently under investigation, but data implies that it may be distinct from statin's primary role as cholesterol-lowering. Strikingly, other metastatic cancer subtypes, such as triple-negative breast cancer (TNBC) and neuroendocrine prostate cancer (NEPC), were found to be more sensitive to statin treatment compared to their less aggressive counterparts. By investigating this further, we can potentially develop a pan-cancer small-molecule inhibitor of metastasis based on the FDA-approved, cheap, and widely available statins.

Prostate cancer (PCa) is the most commonly diagnosed non-cutaneous malignancy in men and the second leading cause of cancer-related deaths in the US. Men with PCa that has recurred after surgery or radiation therapy usually respond to androgen deprivation therapy (ADT); however despite initial responses rate of 80 to 90 percent, nearly all men eventually develop progressive disease following ADT; this is referred to as castration-resistant prostate cancer (CRPC), which remains an incurable disease. It has been reported that neuroendocrine (NE) cell type in prostate cancer is highly associated with castration-resistant disease. However, the exact role of NE cells in promoting the transition from castration-sensitive cancer to castration-resistant disease is not fully understood. The goal of this study was to investigate the gene expression profile, cellular function and therapeutic resistant properties of NE cells purified through CXCR2 surface marker from prostate adenocarcinoma. We further investigated the role of CXCR2 in mediating CRPC carcinogenesis and explore the use of CXCR2 inhibitor combine with Enzalutamide as a potential therapy for advanced prostate cancer. In this research, we have identified a novel cross-talk between neuroendocrine cells in hormone na�ve prostate cancer with both prostate cancer progression and hormonal therapy resistance. We have demonstrated that C-X-C motif chemokine receptor 2 (CXCR2) is a unique surface marker of neuroendocrine cells in prostate cancer and it is over-expressed in metastatic castration resistant, high grade, and NEPC patient cases. Importantly, we discovered that CXCR2 plays a critical role in facilitating this intercellular communication and NEPC transformation from typical prostate adenocarcinoma. CXCR2 overexpression led to enzalutamide resistance, loss of AR expression, and lineage plasticity with acquisition of stem cell-like properties. We further demonstrated that CXCR2 inhibition can prevent/delay enzalutamide resistance and restore AR function, leading to reductions in tumor burden. This research provides the mechanism for the first time of therapeutic resistance mediated by NE cells and raises the possibility to the final cure of lethal prostate cancer through CXCR2 blocker.

The advanced stage of prostate cancer is the second leading cause of cancer-related death for American men. Novel, effective treatment options and more cancer-specific diagnostic tools are urgently needed to facilitate patient management. Here, we explored the construction and application of an array of gene-based molecular imaging and therapeutic vectors in a variety of clinically relevant settings. These vectors exploit prostate cancer-specific promoters to control the transcription of imaging reporter genes or therapeutic genes, and thus can achieve stringent cancer selectivity. We showed that gene-based imaging vectors can detect prostate cancer metastasis earlier than conventional imaging modalities; we also developed a dual-reporter imaging system that can specifically interrogate the activation/inhibition status of androgen receptor, and thus monitor the efficacy of androgen deprivation and androgen receptor blockade therapies. Further, we explored the applicability of gene-based vectors in detecting circulating tumor cells and have obtained promising results. Moreover, we employed two strategies to improve the clinical usability of viral vectors in immunocompetent hosts (such as humans). In the first strategy, we used a broad-acting immunosuppressant, rapamycin, to reduce the scale of adenovirus-induced inflammatory responses and also to decrease the eliminating effects of host adaptive immune system on viral vectors and viral-infected, transgene-expressing cells. Rapamycin significantly augmented the magnitude and prolonged the duration of virus-delivered transgene expression, and thus enhanced the diagnostic capability of these imaging vectors. The second approach took advantage of the versatile surface modifying ability of non-viral reagents. We designed a polypeptide coating that can non-covalently bind to adenoviral particles and confer protections against blood components such as neutralizing antibodies and coagulation factors. The coating significantly decreased the inhibitory effects of neutralizing antibodies on adenoviral transduction and markedly abrogated coagulation factors-mediated hepatocyte infection. Furthermore, this polypeptide coating reduced the antigenicity of adenoviral vectors in vivo, greatly diminished anti-adenovirus antibody production, and therefore could facilitate repeated administration of viral vectors in immunocompetent hosts. Overall, we herein present a substantial amount of evidence supporting the pre-clinical and translational use of gene-based imaging and therapeutic vectors to assist the management of patients with advanced, treatment-refractory prostate cancer.

Prostate is cancer is the most commonly diagnosed cancer in men worldwide and the second biggest killer among cancer related deaths in men. Due to advances in both detection as well as treatment, the disease can often be caught early when it is still confined within the prostate gland. The prognosis of patients when the disease is still confined within the organ is generally good. However, the prognosis in patients when the disease has spread to regional lymph nodes is very poor. Metastatic prostate cancer accounts for nearly all the mortality associated with prostate cancer. Numerous clinical studies have documented the importance of lymph node involvement in predicting patient survival. These data suggest that not only is lymph node involvement an important indicator of systemic metastases, they may be an active participant in the process. In the present report, we provide insight into mechanisms that are involved in prostate cancer metastasis to lymph nodes. First, we examine the relationship between the lymphangiogenic mechanisms and prostate dissemination to lymph nodes. Second we, offer insight into lymph node specific homing mechanisms used by prostate cancer cells. Finally, we describe a strategy to interrogate the significance of lymph nodes as a reservoir of malignant cells that can fuel future metastatic dissemination. A better understanding of the mechanisms involved in lymph node metastasis will allow us to better treat the metastatic disease.

Malignant ascites is a common complication in the late stages of epithelial ovarian cancer (EOC), which greatly diminishes the quality of life of patients. Although malignant ascites is known to be the result of vascular dysfunction, current approved treatments are not effective in stopping its accumulation. This study took on an alternative strategy of targeting the macrophage functions to reverse the vascular pathology of malignant ascites. We analyzed the content of immune cells and macrophages in ascites fluid of human patients and in a murine immunocompetent model (ID8) of EOC. This model developed progressive vascular disorganization and leakiness that culminated in massive ascites, mirroring the human disease. The macrophage content in ascites fluid from human patients and ID8 model directly correlates with vascular permeability. To further substantiate macrophages’ role in the pathogenesis of malignant ascites, a colony-stimulating factor 1 receptor (CSF1R) kinase inhibitor (GW2580) was used to block the functions of macrophages in the ID8 model. GW2580, administered in the late stages, was able to reduce the number of protumorigenic (M2) macrophages infiltrating the ascites and lower ascites volume dramatically. Disorganized peritoneal vasculature became normalized and vascular permeability assays showed that GW2580-treated ascites sera protected against endothelium permeability. Because these results were seen with just ascites sera, the conclusion was that soluble factors in the ascites are responsible, at least in part, for the regulation of the peritoneal vasculature in late-stage EOC. Preliminary mass spectrometry analysis revealed some interesting cadidates including apolipoproteins, CD5 antigen-like, glutathione peroxidase, and inter-alpha-trypsin inhibitor heavy chains. Future directions focus on identifying these factors. This macrophage-targeted blocking treatment may be a promising strategy towards a safe and effective means to control malignant ascites of EOC.

Metastatic Renal Cell Carcinoma(RCC) has limited safe and effective treatments available. This is due to a lack of understanding relating to the mechanism of distant cancer metastasis. Metastatic crosstalk occurs between pseudo-hypoxic von Hippel-Lindau(VHL) knockout tumor cells and VHL wildtype tumor cells. VHL knockout cells have increased production of exosomes due to their pseudo-hypoxic state to help cells adapt to hypoxic conditions. VHL knockout-derived exosomes increase the metastatic properties of VHL wildtype recipient cells. Metastatic properties such as migration and metastasis occur in response between isolated exosomes from the CRISPR modified VHL(-) Renca murine line and VHL(+) cells, generating a model of RCC metastasis and the potential to develop targeted diagnostics and therapeutics for metastatic disease.

PTEN is one of the most commonly deleted tumor suppressor genes in human prostate cancer. Our group previously demonstrated that Pten deletion in the murine prostate epithelium recapitulates the disease progression seen in human prostate cancer, culminating in invasive adenocarcinoma. In addition to Pten loss endowing prostate cells with enhanced proliferative capacity, we found that Pten loss also led to the upregulation of inflammatory pathways, including Csf-1 and Il1b expression, within the prostate epithelium. These inflammatory cytokines recruit myeloid-derived suppressor cells (MDSCs) into the prostate, which subsequently promote an immune-suppressive tumor microenvironment and thereby facilitate tumor progression. Targeting immune-responsive pathways with the CSF-1R inhibitor GW2580 successfully inhibits MDSC infiltration and delays tumor progression.

As Pten deletion alone does not produce distant macrometastasis, we surveyed additional pathways altered in human metastatic prostate cancer, and found that the RAS/MAPK pathway was significantly elevated in metastatic lesions. Indeed, when we combined Pten deletion with Kras activation in the prostate epithelium (Pb-Cre+/-;PtenL/L;KrasG12D/+) (CPK), we observed macrometastasis to the lungs and liver. Interestingly, within the prostate, we observed an epithelial-mesenchymal transition (EMT) phenotype, accompanied by significant upregulation of the EMT transcription factor Snail. Importantly, genetic deletion of Snail in CPK mice prevented distant macrometastasis, providing a mechanistic link between EMT and metastasis.

To study the dynamic regulation of the EMT process, we crossed CPK mice with Vimentin-GFP reporter mice (CPKV), and were able to isolate populations of epithelial, EMT, and mesenchymal-like prostate tumor cells. We demonstrate that EMT and mesenchymal-like tumor cells have enhanced stem-like and tumor-initiating capacities and exhibit cellular plasticity in vivo. HMGA2, a chromatin remodeling protein, is significantly upregulated in EMT and mesenchymal-like tumor cells, as well as in human metastatic castration-resistant prostate cancer (mCRPC). Knockdown of Hmga2, or suppressing Hmga2 expression with the HDAC inhibitor LBH589, inhibits epithelial-mesenchymal plasticity and stemness activities in vitro and dramatically reduces prostate tumor burden and distant metastasis in vivo. Importantly, LBH589 in combination with castration significantly prolongs survival by targeting castration-resistant mesenchymal-like tumor cells and preventing mCRPC. LBH589 treatment in combination with androgen deprivation therapy may therefore be a promising treatment for patients with mCRPC.