Vasanthi Viswanathan

Regulation of ferroptotic cancer cell death by GPX4.

Ferroptosis is a form of nonapoptotic cell death for which key regulators remain unknown. We sought a common mediator for the lethality of 12 ferroptosis-inducing small molecules. We used targeted metabolomic profiling to discover that depletion of glutathione causes inactivation of glutathione peroxidases (GPXs) in response to one class of compounds and a chemoproteomics strategy to discover that GPX4 is directly inhibited by a second class of compounds. GPX4 overexpression and knockdown modulated the lethality of 12 ferroptosis inducers, but not of 11 compounds with other lethal mechanisms. In addition, two representative ferroptosis inducers prevented tumor growth in xenograft mouse tumor models. Sensitivity profiling in 177 cancer cell lines revealed that diffuse large B cell lymphomas and renal cell carcinomas are particularly susceptible to GPX4-regulated ferroptosis. Thus, GPX4 is an essential regulator of ferroptotic cancer cell death.

Dependency of a therapy-resistant state of cancer cells on a lipid peroxidase pathway

Plasticity of the cell state has been proposed to drive resistance to multiple classes of cancer therapies, thereby limiting their effectiveness. A high-mesenchymal cell state observed in human tumours and cancer cell lines has been associated with resistance to multiple treatment modalities across diverse cancer lineages, but the mechanistic underpinning for this state has remained incompletely understood. Here we molecularly characterize this therapy-resistant high-mesenchymal cell state in human cancer cell lines and organoids and show that it depends on a druggable lipid-peroxidase pathway that protects against ferroptosis, a non-apoptotic form of cell death induced by the build-up of toxic lipid peroxides. We show that this cell state is characterized by activity of enzymes that promote the synthesis of polyunsaturated lipids. These lipids are the substrates for lipid peroxidation by lipoxygenase enzymes. This lipid metabolism creates a dependency on pathways converging on the phospholipid glutathione peroxidase (GPX4), a selenocysteine-containing enzyme that dissipates lipid peroxides and thereby prevents the iron-mediated reactions of peroxides that induce ferroptotic cell death. Dependency on GPX4 was found to exist across diverse therapy-resistant states characterized by high expression of ZEB1, including epithelial–mesenchymal transition in epithelial-derived carcinomas, TGFβ-mediated therapy-resistance in melanoma, treatment-induced neuroendocrine transdifferentiation in prostate cancer, and sarcomas, which are fixed in a mesenchymal state owing to their cells of origin. We identify vulnerability to ferroptic cell death induced by inhibition of a lipid peroxidase pathway as a feature of therapy-resistant cancer cells across diverse mesenchymal cell-state contexts.

Drug-tolerant persister cancer cells are vulnerable to GPX4 inhibition

Acquired drug resistance prevents cancer therapies from achieving stable and complete responses. Emerging evidence implicates a key role for non-mutational drug resistance mechanisms underlying the survival of residual cancer ‘persister’ cells. The persister cell pool constitutes a reservoir from which drug-resistant tumours may emerge. Targeting persister cells therefore presents a therapeutic opportunity to impede tumour relapse. We previously found that cancer cells in a high mesenchymal therapy-resistant cell state are dependent on the lipid hydroperoxidase GPX4 for survival. Here we show that a similar therapy-resistant cell state underlies the behaviour of persister cells derived from a wide range of cancers and drug treatments. Consequently, we demonstrate that persister cells acquire a dependency on GPX4. Loss of GPX4 function results in selective persister cell ferroptotic death in vitro and prevents tumour relapse in mice. These findings suggest that targeting of GPX4 may represent a therapeutic strategy to prevent acquired drug resistance.

DiSCoVERing Innovative Therapies for Rare Tumors: Combining Genetically Accurate Disease Models with In Silico Analysis to Identify Novel Therapeutic Targets.

Purpose: We used human stem and progenitor cells to develop a genetically accurate novel model of MYC-driven Group 3 medulloblastoma. We also developed a new informatics method, Disease-model Signature versus Compound-Variety Enriched Response (“DiSCoVER”), to identify novel therapeutics that target this specific disease subtype. Experimental Design: Human neural stem and progenitor cells derived from the cerebellar anlage were transduced with oncogenic elements associated with aggressive medulloblastoma. An in silico analysis method for screening drug sensitivity databases (DiSCoVER) was used in multiple drug sensitivity datasets. We validated the top hits from this analysis in vitro and in vivo. Results: Human neural stem and progenitor cells transformed with c-MYC, dominant-negative p53, constitutively active AKT and hTERT formed tumors in mice that recapitulated Group 3 medulloblastoma in terms of pathology and expression profile. DiSCoVER analysis predicted that aggressive MYC-driven Group 3 medulloblastoma would be sensitive to cyclin-dependent kinase (CDK) inhibitors. The CDK 4/6 inhibitor palbociclib decreased proliferation, increased apoptosis, and significantly extended the survival of mice with orthotopic medulloblastoma xenografts. Conclusions: We present a new method to generate genetically accurate models of rare tumors, and a companion computational methodology to find therapeutic interventions that target them. We validated our human neural stem cell model of MYC-driven Group 3 medulloblastoma and showed that CDK 4/6 inhibitors are active against this subgroup. Our results suggest that palbociclib is a potential effective treatment for poor prognosis MYC-driven Group 3 medulloblastoma tumors in carefully selected patients. Clin Cancer Res; 22(15); 3903–14. ©2016 AACR.

Inhibition of Zinc-Dependent Histone Deacetylases with a Chemically Triggered Electrophile

Unbiased binding assays involving small-molecule microarrays were used to identify compounds that display unique patterns of selectivity among members of the zinc-dependent histone deacetylase family of enzymes. A novel, hydroxyquinoline-containing compound, BRD4354, was shown to preferentially inhibit activity of HDAC5 and HDAC9 in vitro. Inhibition of deacetylase activity appears to be time-dependent and reversible. Mechanistic studies suggest that the compound undergoes zinc-catalyzed decomposition to an ortho-quinone methide, which covalently modifies nucleophilic cysteines within the proteins. The covalent nature of the compound-enzyme interaction has been demonstrated in experiments with biotinylated probe compound and with electrospray ionization-mass spectrometry.

Targeting a Therapy-Resistant Cancer Cell State Using Masked Electrophiles as GPX4 Inhibitors

We recently discovered that inhibition of the lipid peroxidase GPX4 can selectively kill cancer cells in a therapy-resistant state through induction of ferroptosis. Although GPX4 lacks a conventional druggable pocket, covalent small-molecule inhibitors are able to overcome this challenge by reacting with the GPX4 catalytic selenocysteine residue to eliminate enzymatic activity. Unfortunately, all currently-reported GPX4 inhibitors achieve their activity through a reactive chloroacetamide group; this dependence hinders their selectivity and stability and makes them unsuitable for use in vivo. Development of therapeutically useful GPX4 inhibitors may be achieved by the identification of new electrophilic chemotypes and mechanisms of action that do not suffer these shortcomings. Here, we report our discovery that nitrile oxide electrophiles, and a set of remarkable chemical transformations that generates them in cells from masked precursors, provide an effective strategy for selective targeting of GPX4. Our results, which include structural insights, target engagement assays, and diverse GPX4-inhibitor tool compounds, provide critical insights that may galvanize development of therapeutic agents for exploring the efficacy and safety of inhibiting the currently-undruggable GPX4. Our discovery that nitrile oxide electrophiles engage in highly selective cellular interactions and are bioavailable in their masked forms may also be valuable in the development of covalent inhibitors of other challenging targets.

HIF-2α drives an intrinsic vulnerability to ferroptosis in clear cell renal cell carcinoma

Kidney cancers are characterized by extensive metabolic reprogramming and resistance to a broad range of anti-cancer therapies. By interrogating the Cancer Therapeutics Response Portal compound sensitivity dataset, we show that cells of clear-cell renal cell carcinoma (ccRCC) possess a lineage-specific vulnerability to ferroptosis that can be exploited by inhibiting glutathione peroxidase 4 (GPX4). Using genome-wide CRISPR screening and lipidomic profiling, we reveal that this vulnerability is driven by the HIF-2α–HILPDA pathway by inducing a polyunsaturated fatty acyl (PUFA)-lipid-enriched cell state that is dependent on GPX4 for survival and susceptible to ferroptosis. This cell state is developmentally primed by the HNF-1β–1-Acylglycerol-3-Phosphate O-Acyltransferase 3 (AGPAT3) axis in the renal lineage. In addition to PUFA metabolism, ferroptosis is facilitated by a phospholipid flippase TMEM30A involved in membrane topology. Our study uncovers an oncogenesis-associated vulnerability, delineates the underlying mechanisms and suggests targeting GPX4 to induce ferroptosis as a therapeutic opportunity in ccRCC. HIGHLIGHTS ccRCC cells exhibit strong susceptibility to GPX4 inhibition-induced ferroptosis The GPX4-dependent and ferroptosis-susceptible state in ccRCC is associated with PUFA-lipid abundance The HIF-2α–HILPDA axis promotes the selective deposition of PUFA-lipids and ferroptosis susceptibility AGPAT3 selectively synthesizes PUFA-phospholipids and primes renal cells for ferroptosis

Abstract 3026: Targeting GPX4 in tumor-associated stromal cells increases inflammatory-cell infiltration

The lack of a T-cell inflamed microenvironment in tumors limits responsiveness to many immunotherapies. T-cell exclusion is often mediated by a dense infiltration of fibroblast-like stromal cells. Up to 55% of triple-negative breast cancers have ‘stroma-rich’ tumors with markedly lower T-cell inflammation. Here we report a therapeutic strategy that can potentially convert stroma-rich tumors into T-cell inflamed tumors by forcing stromal cells to secrete 5-lipoxygenase products which are powerful chemo-attractants for T cells. We were initially interested in identifying selective inhibitors of stromal-cell function. To achieve this, we used phenotype-based small-molecule screening in which the phenotype of stroma-induced cancer cell migration in vitro was a surrogate for stroma-induced metastasis in vivo. We identified a compound, RSL3 that inhibited this migration. RSL3 was selectively cytotoxic to stromal cells over cancer cells, in comparisons of immortalized cell lines as well as comparisons of patient-derived primary breast cancer cells to cancer-associated fibroblasts (CAFs). We therefore undertook studies to identify its mechanism of action. RSL3 was recently reported to target the redox enzyme glutathione peroxidase 4 (GPX4). GPX4 metabolizes lipid peroxides so we performed metabolomic profiling of RSL3-treated stroma-cancer co-cultures and found elevated arachidonic acid products of lipoxygenase enzymes. Stromal cells were found to contain >10-fold higher levels of lipoxygenase products than carcinoma cells. Blocking either 5-lipoxygenase (5-LO) or 15-lipoxygenase (15-LO) with selective inhibitors abrogated RSL3’s cytotoxicity to stromal cells. Thus, high lipoxygenase activity in stromal cells increases their susceptibility to GPX4 inhibition. Because 5-LO products like leukotriene B4 are powerful chemo-attractants for myeloid cells and T cells, we studied the impact of GPX4 knockdown in vivo using xenografts of cancer cells co-injected with stromal cells. GPX4 knockdown resulted in a large increase in myeloid-cell infiltration into tumors but, surprisingly, T-cell infiltration was suppressed. We reasoned that 15-LO products are immunosuppressive based on recent findings that 15-LO gene amplifications are inversely correlated with T-cell infiltration in breast cancers in The Cancer Genome Atlas. Consistent with this, GPX4 inhibition of stroma-cancer co-cultures suppressed T-cell chemotaxis but combined inhibition of GPX4 and 15-LO significantly enhanced T-cell chemotaxis over untreated controls in vitro. We are undertaking in vivo testing of this combination. In summary, our unbiased chemical biology approach has revealed a therapeutic strategy to promote T-cell inflammation. We envision this to be a priming strategy for stroma-rich cancers to become responsive to a variety of different immunotherapies thereby unleashing their full curative potential. Citation Format: Shrikanta Chattopadhyay, Cherrie Huang, Ninib Baryawno, Nicolas Severe, Vasanthi Viswanathan, Carlotta Costa, David Scadden, Stuart Schreiber. Targeting GPX4 in tumor-associated stromal cells increases inflammatory-cell infiltration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3026. doi:10.1158/1538-7445.AM2017-3026

Abstract B11: Targeting mesenchymal cells in the tumor stroma by GPX4 inhibition

Fibroblast-like mesenchymal cells are the most abundant component of tumor stroma and have been shown to promote tumor progression. However, there are no therapeutic agents that target these cells. Developing therapeutics against mesenchymal cells has been challenging because they lack clear “druggable” targets. Here we used a phenotypic screening approach to identify therapeutic targets in bone-marrow mesenchymal stem cells (MSCs), precursors of cancer-associated fibroblasts, that are reported to induce a metastatic phenotype in breast cancer cells (Karnoub et al, Nature 2007). Using the in vitro phenotype of enhanced migration of breast cancer cells (GFP-labeled MDA-MB-231; MDA) induced by MSCs (3-fold faster than MDA alone), we identified inflammatory signaling and glutathione peroxidase 4 as potential targets in MSCs. To validate this screening model, we first confirmed that MSCs can enhance metastasis using orthotopic xenografts of luciferase-labeled MDA cells co-injected with primary human MSCs in NOD-SCID mice. Mice with MSC+MDA tumors had 5-fold greater thoracic bioluminescence (corresponding to lung metastasis) than mice with MDA tumors alone. We then performed gene-expression profiling of MSC+MDA co-cultures in vitro compared to cells grown alone. The top pathway upregulated in co-cultures was the interferon pathway, suggesting that an inflammatory response follows MSC-MDA interactions. Using publicly available gene-expression datasets, we prioritized transcripts that are expressed in patient stroma and correlate with poor survival in a meta-analysis of 20 whole-tumor datasets. To determine if these transcripts are necessary for MSC-induced metastatic behavior, we performed shRNA knockdown and measured effects on in vitro migration of MSC+MDA co-cultures compared to normal endothelial cells. Knockdown of 9 genes was specific for MSC+MDA migration but the effects were weak. This suggested functional redundancy and indicated that targeting individual upregulated genes is insufficient to block the phenotype. We then performed a small-molecule screen on the MSC+MDA migration phenotype, counter-screening hits on endothelial cell migration. Only one compound, RSL3, showed a large and selective inhibition of MSC+MDA migration. RSL3 was selectively toxic to MSCs with no effect on MDA cells. The target of RSL3 was recently identified to be glutathione peroxidase 4 (GPX4) that metabolizes lipid peroxides. To identify lipid mediators of RSL3 toxicity, we profiled lipid levels in RSL3-treated MSC+MDA co-cultures and found that the top changes were lipoxygenase products of arachidonic acid. Pre-treatment with zileuton or PD146176, that inhibit 5- and 15-lipoxygenase respectively, abrogated RSL39s toxic effects on MSCs. We found that MSCs contain >10-fold higher lipoxygenase products than MDA cells, highlighting an inflammatory role of MSCs. Fibroblasts from different tissues (lungs, spleen, breast, skin) were similarly sensitive to RSL3, identifying a previously unrecognized vulnerability in these mesenchymal cells. Since RSL3 is a tool compound with poor plasma stability, we used GPX4 shRNA knockdown (KD) to determine in vivo effects of depleting MSCs in MSC+MDA xenografts. Unexpectedly, we found accelerated tumor growth in GPX4 KD tumors compared with controls. However, this is consistent with previous reports showing that mesenchymal cells restrain tumor growth and can have both pro- and anti-tumor effects. Histological examination revealed increased myeloid cell infiltration into GPX4 KD tumors reflecting increased inflammation mediated by lipoxygenase products like leukotriene B4. In summary, phenotypic screening has identified GPX4 inhibition as a novel approach to target mesenchymal cells. This approach may be particularly effective at recruiting immune/ inflammatory cells into tumors with the exciting possibility of synergizing with cancer immunotherapy. Citation Format: Shrikanta Chattopadhyay, Cherrie Huang, Ninib Baryawno, Nicolas Severe, Vasanthi Viswanathan, Zarko Boskovic, Siddhartha Mukherjee, Jeff Gentry, Ben Wittner, Sridhar Ramaswamy, Alykhan Shamji, David Scadden, Stuart Schreiber. Targeting mesenchymal cells in the tumor stroma by GPX4 inhibition. [abstract]. In: Proceedings of the AACR Special Conference: Function of Tumor Microenvironment in Cancer Progression; 2016 Jan 7–10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2016;76(15 Suppl):Abstract nr B11.

Abstract 181: Therapeutic approaches to metastasis induced by mesenchymal stem cells in the tumor microenvironment

Metastasis is the primary cause of death in non-hematological cancers yet there are no specific therapeutics against it because of a lack of validated targets. The Weinberg lab demonstrated that bone-marrow derived mesenchymal stem cells (MSCs) home to the stroma of breast tumors and induce metastasis. We verified these findings in MDA-MB-231 (MDA) xenografts in which co-injection of human MSCs increased MDA thoracic metastasis by 5-fold. In vitro, MSC co-culture induced GFP-labeled MDA cells to migrate 3-fold faster in a modified ‘wound-healing’ assay, mirroring metastasis in vivo. To identify therapeutic targets within MSC-induced metastasis, we performed gene-expression analysis of MSC-MDA co-cultures separated by flow cytometry compared with cells grown alone. The interferon pathway was found to be the most activated pathway upon co-culture, increasing mainly in MSCs. We studied the relevance of these genes in human cancers by first analyzing 3 different gene-expression datasets comparing human breast cancer stroma with normal stroma. Genes upregulated in breast cancer stroma were then studied in a meta-analysis of 19 whole tumor gene-expression datasets correlating gene-expression changes with survival. We identified 103 genes that are upregulated in MSCs by MSC-MDA interactions, are increased in human breast cancer stroma and are significantly associated with poor survival. To determine if they are necessary for MSC-induced metastatic behavior, we performed shRNA knockdown in MSC-MDA co-cultures measuring effects on in vitro migration. Knockdown of a number of interferon-associated genes significantly reduced migration supporting an unexpected functional role of interferons in metastasis. The top interferon gene, ISG15, is an attractive candidate for therapeutic targeting because it is a secreted ubiquitin-like factor that conjugates a number of cytoskeletal proteins involved in motility. In parallel, we conducted a small-molecule screen with 1600 compounds on the migration of MSC-MDA co-cultures to identify small-molecule inhibitors of metastasis. Counter screens on highly motile endothelial cells excluded compounds that non-specifically inhibit normal cell migration. Only 1 compound, RSL3, specifically blocked MSC-induced MDA migration with cytotoxicity at >10-fold higher concentrations. RSL3 inhibits the glutathione peroxidase 4 (GPX4) enzyme that metabolizes lipid peroxides including arachidonic acid metabolites participating in inflammatory cascades like interferon gamma signaling. RSL3 activity was completely abrogated by co-treatment with the 5-lipoxygenase inhibitor zileuton consistent with the role of arachidonic acid metabolites in MSC-MDA migration. In summary, targeting components of interferon and arachidonic acid pathways have been discovered as novel therapeutic approaches against microenvironment-induced breast cancer metastasis. Citation Format: Shrikanta Chattopadhyay, Cherrie Huang, Siddhartha Mukherjee, Rushdia Z. Yusuf, Vasanthi Viswanathan, Ben S. Wittner, Jeff Gentry, Alykhan Shamji, Sridhar Ramaswamy, David T. Scadden, Stuart L. Schreiber. Therapeutic approaches to metastasis induced by mesenchymal stem cells in the tumor microenvironment. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 181. doi:10.1158/1538-7445.AM2014-181