UC Berkeley

Covalent drugs incorporate a mildly electrophilic functional group that reacts with protein targets to confer additional affinity beyond the non-covalent interactions involved in drug binding. In the past, concerns about these reactive molecules’ interference with biological assays and lack of selectivity often discouraged further investigation. However, the emergence of intentionally designed covalent drugs against major disease targets over the last ten years showcases the strengths of this field. The first part of this thesis will review the historical and modern milestones in covalent drug discovery with emphasis on chemoproteomics techniques that enable success, and the remaining chapters will address my contributions to the field through the development of tool covalent compounds as novel inhibitors and new induced proximity platforms.Chemoproteomics-enabled covalent ligand screening platforms can be used to identify novel ligands against undruggable protein targets like MYC. MYC is a major oncogenic transcriptional driver of most human cancers that has remained intractable to direct targeting because much of MYC is intrinsically disordered. I performed a cysteine-reactive covalent ligand screen to identify compounds that could disrupt the binding of MYC to its DNA consensus sequence in vitro and also impair MYC transcriptional activity in situ in cells. I identified a covalent ligand EN4 that targets cysteine 171 (C171) of MYC within a predicted intrinsically disordered region of the protein. I show that EN4 directly targets MYC in cells, reduces MYC and MAX thermal stability, inhibits MYC transcriptional activity, downregulates multiple MYC transcriptional targets, and impairs tumorigenesis. I also show initial structure-activity relationships of EN4 and identify compounds that show improved potency. Overall, I identified a novel ligandable site within an intrinsically disordered region of MYC that leads to inhibition of MYC transcriptional activity. In addition to facilitating the discovery of covalent small-molecule allosteric modulators of disease relevant proteins and pathways, chemoproteomics-enabled covalent drug discovery also proves useful in identifying inhibitors of critical disease proteins that contain catalytic cysteines. Among the various genes and proteins encoded by all coronaviruses, one particularly “druggable” or relatively easy-to-drug target is the coronavirus Main Protease (3CLpro or Mpro), a cysteine-protease that is involved in cleaving a long peptide translated by the viral genome into its individual protein components that are then assembled into the virus to enable viral replication in the cell. Inhibiting Mpro’s catalytic cysteine with a small-molecule antiviral would effectively stop the ability of the virus to replicate, providing therapeutic benefit across coronaviruses. As part of a collaborative effort, I utilized activity-based protein profiling (ABPP)- chemoproteomic approaches to discover and further optimize cysteine-reactive pyrazoline-based covalent inhibitors for the SARS-CoV-2 Mpro. Structure-guided medicinal chemistry and modular synthesis of di- and tri-substituted pyrazolines bearing either chloroacetamide or vinyl sulfonamide cysteine-reactive warheads enabled the expedient exploration of structure-activity relationships (SAR), yielding nanomolar potency inhibitors against Mpro from not only SARS-CoV-2, but across many previous coronaviruses. Our studies highlight promising chemical scaffolds that may contribute to future pan-coronavirus inhibitors. Finally, chemoproteomics-enabled drug discovery platforms facilitate the expansion of targeted protein degradation and related fields. Many diseases are driven by proteins that are aberrantly ubiquitinated and degraded. These diseases would be therapeutically benefited by targeted protein stabilization (TPS). We designed deubiquitinase-targeting chimeras (DUBTACs), heterobifunctional small molecules consisting of a deubiquitinase recruiter linked to a protein-targeting ligand, to stabilize the levels of specific proteins degraded in a ubiquitin-dependent manner. Using chemoproteomic approaches, we discovered the covalent ligand EN523 that targets a non-catalytic allosteric cysteine C23 in the K48-ubiquitin-specific deubiquitinase OTUB1. We showed that a DUBTAC consisting of our EN523 OTUB1 covalent recruiter linked to lumacaftor, a drug used to treat cystic fibrosis that binds ΔF508-cystic fibrosis transmembrane conductance regulator (CFTR), robustly stabilized ΔF508-CFTR protein levels, leading to improved chloride channel conductance in human cystic fibrosis bronchial epithelial cells. We also demonstrated stabilization of the tumor suppressor kinase WEE1 in hepatoma cells. Our study showcases covalent chemoproteomic approaches to develop new induced proximity-based therapeutic modalities and introduces the DUBTAC platform for TPS.