Abstract 3541: Systematic assessment of the mechanism of action landscape of glue compounds

Abstract

Abstract We have recently completed a large-scale project to elucidate the Mechanism of Action (MoA) of clinically relevant compounds in over twenty tumor specific contexts. Many novel, translationally relevant findings have arisen from this work, including (a) the identification of context-specific inhibitors of many proteins considered to be poor pharmacologic targets—including transcription (co-)factors and undruggable signaling proteins; (b) graph theory-based elucidation of drug polypharmacology; and (c) identification of MoA overlap between previously unrelated drugs. However, a limitation of the study is it relies on the existing repertoire of small-molecular inhibitors that is heavily biased towards specific target classes. Compounds that inhibit master regulators (MR) of tumor state are missing or under-represented. In this new project, we explore an extra molecular space of “molecular glues”, as they might be more effective modulators of MRs. Similar to our prior oncology drug study, we will leverage the automated PLATE-Seq technology and the VIPER algorithm to characterize compound MoA in a proteome-wide fashion. Specifically, we will first generate genome-wide RNA-Seq profiles of patient-matched cell lines using a panel of ~150 glue compounds, including a subset of them named PROteolysis TArgeting Chimeras (PROTACs) that were designed to degrade therapeutic targets by recruiting them to E3 ligases. We will analyze the PLATE-seq perturbational profiles with a set of complementary algorithms including the VIPER, which, akin to a multiplexed, tissue-specific gene reporter assay, can accurately measure the differential activity of over 6,000 regulatory/signaling proteins, by assessing the enrichment of their transcriptional targets in differentially expressed genes. Current effort of this project is focused on six PDA cell lines including three MAPK-pathway dependent (M+) and three MAPK-pathway independent (M-) contexts, each one representing the gastrointestinal lineage state (GLS), morphogenic state (MOS), and acinar to ductal metaplasia-like state (ALS). Using this data, we explore similarity-based connections between PLATE-seq perturbational profiles and the Connectivity Map (CMAP) that contains transcriptional consequences of chemical and genetic perturbations. We will also compare PLATE-seq perturbational profiles to tissue-specific differential expression signatures that distinguish cell lines sensitive to compounds in the Cancer Therapeutics Response Portal (CTRP) and Broad Repurposing Hub, each of which houses results from exposing hundreds of compounds exposed to dozens to hundreds of cell lines. Taken together, the data and analyses will help understand MoAs and polypharmacologies/off-targets of molecular glues, which also serve to enable interpretation and prioritization of new candidate MoA for molecular glues by comparing them to compounds with known targets and MoAs. Citation Format: Lucas ZhongMing Hu, Charles Karan, Andrea Califano. Systematic assessment of the mechanism of action landscape of glue compounds [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3541.