Linking Genetic, Epigenetic and Signaling Mechanisms of Oncogene Addiction - PROJECT SUMMARY Our long-term goal is to build a network-level and single-cell understanding of the interplay between genetic, epigenetic and signaling mechanisms that determine the state of MAP kinase (MAPK) pathway dependency in tumor cells. Such understanding will be key to our ability to predict and ultimately improve the responsiveness of tumor cells to therapeutic inhibitors of MAPK signaling. Our focus, in this application, is on BRAFV600 mutated cancers, particularly melanomas, where hyperactivation of MAPK signaling has motivated the clinical evaluation of MAPK-targeted therapies, but they often lead to variable responses and no durable cure in most patients. While genetic alterations are associated with late resistance, epigenetic heterogeneity associated with tumor differentiation state or tissue lineage and its adaptive plasticity within a tumor influence the dynamic state of BRAF/MAPK dependency, thereby diminishing the therapeutic efficacy of MAPK inhibitors. Here, we propose a systems pharmacology approach to test the hypothesis that heterogeneity in the state of MAPK dependency may result from a subset of key epigenetic variations across tumor cells of heterogeneous differentiation states. To identify regulator of such variations, we screened 276 small-molecule epigenetic modulators individually or in combination with BRAF/MEK kinase inhibitors in melanoma cell lines that represent a wide spectrum of differentiation states. Integrating multiplexed single-cell analysis with multivariate statistical modeling and genetic experiments, we identified three classes of inhibitors that target seemingly distinct epigenetic states in melanoma cells: (1) a lysine demethylase 1A (KDM1A)-dependent state associated with undifferentiated (AXLHigh), MAPK inhibitor-resistant (p-ERKHigh) cells, (2) a Jumonji histone demethylase (Jmj-KDM)-dependent state associated with neural crest-like (NGFRHigh/AXLLow) cells, and (3) a state induced by BET bromodomain inhibitors in NGFRHigh cells, which substantially enhances their requirement for MAPK signaling. Single-cell analysis shows that these states might co-exist in different combinations and frequencies, highlighting mutual epigenetic vulnerabilities among genetically diverse melanoma cell populations. In this proposal, we aim to: (1) examine the molecular specificity of the identified small molecule inhibitors as well as mechanisms that govern each of the phenotypically consequential epigenetic states, (2) identify predictors of epigenetic switching in BRAF-mutated cell lines and patient-derived tumors, and (3) test new epigenetic strategies to overcome heterogeneous populations of MAPK inhibitor-tolerant cells in vitro and in vivo. This will be achieved through a systems pharmacology approach, combining genetic experiments, high-throughput single-cell imaging, proteomic measurements, and network-level computational modeling. We will use these tools as a means to measure, model, modulate and ultimately overcome heterogeneous populations of drug- tolerant cells. Our work is expected to provide rational approaches to improve the clinical benefit and durability of treatment response in patients with melanoma and potentially other BRAF-mutated cancers.