Targeting RNA-protein networks in cancer using chemical and global methods
Introduction
We report the discovery of small molecules capable of controlling cellular levels of membraneless subcellular compartments known as processing-bodies (PBs) and stress granules (SGs). PBs and SGs are cytoplasmic ribonucleoprotein granules that are part of a larger family of structures collectively referred to as biomolecular condensates. These structures can be induced in response to cellular stress for survival, but aberrant condensate regulation is associated with a growing number of disease states including cancer. Targeting disease-relevant condensates offers a unique opportunity for therapeutic discovery but has proven challenging because of the compositional diversity and dynamic nature of these evolutionarily conserved structures. In this study, we demonstrated that covalent binding to tyrosine sites on key condensate-forming RNA-binding proteins (RBPs), a class of proteins historically deemed undruggable, can prevent or induce these cytoplasmic structures in stressed cells. This study demonstrates the ability to effectively target RBPs using sulfur-triazole exchange (SuTEx) chemistry in combination with mass spectrometry-based chemical proteomics capabilities.
Methods
We deployed a phenotypic screen for condensate-modulating small molecules by monitoring SG and PB formation in cells using established immunofluorescence markers. Our previous chemical proteomic studies identified sulfonyl-triazoles as a cell-active electrophile for covalent targeting of RNA-binding and protein-protein interaction domains, which are known to facilitate high valency interactions for assembly of condensed RNP networks. We pursued a fragment-based ligand discovery (FBLD) approach because of the ability to survey a larger fraction of chemical space with a smaller number of fragments. SuTEx was chosen for FBLD because LG diversification with binding groups permitted integration of the sulfone into fragment design as opposed to appending this electrophile to existing ligands (e.g., using SuFEx). We performed competitive LC-MS/MS ABPP studies to establish covalent binding profiles for active SuTEx compound hits.
Results
Our screen identified SuTEx compounds that inhibited SGs (EKT166, HHS-166), PBs (EKT132), and both types of RNP granules in stressed cells (AHL-003). We also identified SuTEx ligands that enhanced stress-induced RNP granule levels in compound-treated cells. Using LC-MS/MS ABPP, we quantified >770 Tyr and Lys sites that are ligandable for developing covalent binders with RNP granule-modulating activity in cells. The proteomic reactivity of SuTEx electrophiles in cells (~3-9%) was comparable to hit rates previously reported for screening electrophile libraries. Among the list of liganded RNP-granule proteins, we identified key RBPs (hnRNPA) and chaperone proteins (HSPB1) that have demonstrated roles in phase separation or maintenance of condensates. The liganded sites mapped to functional domains of proteins that are reported sites for post-translational phosphorylation, ubiquitination, and sumoylation. By expanding the ligandable RNP granule proteome, the pharmacological tractability of individual proteins and sites within functional domains can be further explored to develop condensate-modulating compounds in future studies.
Conclusion
In summary, we present a systematic approach for discovering condensate-modulating covalent small molecules with defined protein interaction profiles to serve as chemical probes for investigating biomolecular condensate regulation and pharmacological tractability.