Aim 1: Synergistic invasion in type 1 papillary RCC driven by HGF and hypoxia signaling. HGF signaling contributes to disease progression, tumor invasiveness and metastasis in kidney cancers other than the clear cell type; in particular, MET kinase domain mutations in a hereditary form of papillary renal carcinoma (PRC) are primary drivers of that disease, and Met overabundance is common in both hereditary and sporadic forms. VHL loss of function is rare in PRC, but tumor hypoxia is not; hypoxia strongly enhances HGF-mediated invasiveness and metastasis in a variety of model systems, through largely undefined molecular mechanisms. Nonetheless, HGF-driven beta-catenin transcriptional activity is suppressed when VHL is functional, suggesting that the integration of hypoxia and HGF driven cell invasiveness in PRC involves other primary signaling routes. Indeed, we found that convergence of the mitogen-activated protein kinase (MAPK), phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) and reactive oxygen species (ROS) signaling pathways mediate synergistic invasiveness triggered by HGF and hypoxia in several PRC cell lines, which further entailed induction of autophagy and growth suppression. Hypoxia induced ROS accumulation suppressed DUSP2 expression, thereby enhancing ERK activation. ROS also triggered a cascade of phospholipase C-gamma activation, diacylglycerol production and protein kinase C-mediated phosphatase 2A activation which suppressed HGF-induced Akt activation and cell cycle progression while promoting autophagy. Akt suppression and the consequent shift from HGF-enhanced, proliferation-oriented metabolism to autophagy proved to be required for synergistic invasion. This tripartite signaling integration was not unique to RCC or HGF: in RCC cells, invasive synergy induced by the combination of hypoxia and EGF occurred through the same mechanism, and in ER positive breast cancer cells, this mechanism was suppressed in the absence of estrogen. These results define the molecular basis of growth factor and hypoxia invasive synergy in VHL-competent papillary RCC cells, illustrate the plasticity of invasive and proliferative tumor cell states and provide signaling profiles by which they may be predicted.
Aim 2 : The experimental therapeutic AR-12 targets proteostasis machinery in clear cell RCC. Various evidence suggests that PI3K/AKT1 and mTOR pathways may be aberrantly activated in larger subpopulations of all RCC subtypes than the reported frequencies of mutation or epigenetic suppression of key components such as PIK3CA and PTEN in TCGA datasets suggest. This, evidence of positive feedback loops from mTOR inhibition to PI3K/AKT1 signaling, and evidence of AKT1-independent signaling downstream of oncogenic PI3K activation, provide a rationale for identifying additional critical nodes in survival pathways for therapeutic targeting in RCC. Since PDPK1 (PDK-1, protein kinase B) is a key regulator of both AKT1-dependant and -independent survival pathways, PDPK1 inhibition is a compelling potential therapeutic strategy. We studied the potential efficacy of PDPK1 inhibition in ccRCC using both siRNA knockdown and pharmacological suppression by AR-12 (OSU-03012), a novel PDPK1 agent developed by Dr. Ching-Shih Chen (The Ohio State University) and then in early clinical development through Arno Therapeutics. AR-12 was obtained through an MTA. While studying ccRCC cell sensitivity to AR-12, we uncovered evidence that the drug was acting independently of PDPK1. The findings were extraordinary given the effort by various research groups that had brought the compound to phase I clinical trials. Nonetheless, we followed a logical route to identify its actual mechanism(s) of action using kinase assays, DTP COMPARE software to infer activity, and with the help of Dr. Chen's lab to immobilize AR-12 on a solid phase, by capturing putative targets from cell lysates that were then identified by mass spectrometry. Among the latter list of putative targets were a striking collection of heat shock proteins: GRP78, GRP75, HSPA9, HSP27, HSP70, HSP75, HSP90, BAG2, HSPB1 as well as ER resident proteins PRDX1 and TXN. These studies led us to Dr. Paul Dent (Virginia Commonwealth University) who had independently found indirect evidence that AR-12 disrupts ER resident and heat shock proteins. We began a collaboration with Dr. Dent that includes cytotoxic effects on tumor cells and has uncovered anti-viral effects of AR-12 in normal cells.
Aim 3 : Identify and develop selective natural product inhibitors of HIF2-alpha as therapeutic candidates for the treatment of clear cell RCC. Normoxic stabilization of HIF1-alpha alone, while capable of mimicking some aspects of VHL loss, are not sufficient to reproduce tumorigenesis and in fact, there is evidence that it acts as a tumor suppressor. HIF2-alpha, in contrast, is consistently cast as an oncoprotein. To isolate compounds that selectively modulate HIF2-alpha for use as research tools and drug development leads, a cell-based high throughput screening assay of the NCI Natural Products Repository was developed in collaboration with Drs. Tawnya McKee and James McMahon of CCR's Molecular Targets Laboratory (MTL). A series of plasmids were engineered containing tandem copies of the minimal hypoxia response element of the VEGFA gene promoter upstream of a Luc reporter. These plasmids were transfected into the VHL- and HIF1-alpha-negative ccRCC cell line 786-0 and derived cell lines were optimized for use in high-throughput detection systems in the MTL. Secondary counter screens were included for global transcriptional repression and cell toxicity. Leads from Natural Products Repository screening were chromatographically separated into component structures yielding 40 pure compounds with micromolar or submicromolar IC50 values, 80% inhibition and 10% cell toxicity. Three compounds, a new 2-pyrrolinone alkaloid langkamide and the known compounds piplartine and 3,4,5-trimethoxycinnamic acid had HIF2-alpha IC50 values of 14, 4.8, and 60 uM, respectively. The latter 2 compounds had toxicity IC50 values of 61 and 78 uM whereas langamide was not toxic at the highest dose tested (66 uM). Among a group of 10 soft coral diterpenes, 7 cembranes showed poor HIF2-alpha IC50 values in follow up and were not studied further. Three xenicin-type diterpenes had IC50 values of 3-12 uM with low cytotoxicity and remained in the study group. Of 12 marine sponge extracts, 5 with HIF2-alpha IC50 values from 0.7-15 uM were re-screened for VEGF-A protein secretion: two compound effects were due to cytotoxicity, two suppressed VEGF-A secretion with modest toxicity and one compound stimulated both VEGF-A secretion and cell growth. HIF selectivity of the 30 remaining leads was analyzed for VEGF-A secretion and PCR-based assays for a panel of five HIF-regulated genes and two HIFs in ccRCC-derived cell lines expressing either HIF2-alpha only, equal content HIF1-alpha and 2-alpha, or HIF1-alpha only. Of these, 12 compounds had IC50 values for VEGF-A secretion of 50 uM in one or more cell lines and exhibited mixed patterns of HIF selectivity. Three compounds in this group stood out for non-toxicity, potency and HIF2-alpha selectivity and had IC50 values for VEGF-A secretion of 16 uM in HIF2-alpha only expressing cells. Thus these compounds enter intact cells with minimal toxicity and selectively suppress the expression of several HIF target genes, primarily by acting through HIF2-alpha. Strong structural similarities between the 11 of the 12 compounds in the most recent focus group provide solid leads for structure-activity relationship studies as well as insight for synthesis and future medicinal chemistry refinement efforts.
