Network-driven identification of indisulam neo-substrates for targeted protein degradation

Abstract

Indisulam, a DCAF15-based molecular glue degrader, induces widespread proteome changes with implications for cell division and chromosome segregation. While RBM39 and RBM23 are two well-characterized indisulam neo-substrates, additional targets likely exist. To identify those degradation targets, we applied a network-based approach to prioritize novel neo-substrates from large-scale omics data. Our approach integrates proteome-wide expression measurements with information from publicly accessible databases into a multilayer heterogeneous network. Utilizing a Random Walk with Restart algorithm, we identified a preliminary list of 30 neo-substrates. These proteins are likely interactors with DCAF15 in the presence of indisulam and are subject to subsequent degradation. Experimental validation of hits from the shortlisted candidates confirmed their degradation in a proteasome-dependent manner, supporting their identification as potential novel indisulam neo-substrates. Our work employs established network resources and analytical methods to effectively identify direct targets of the indisulam molecular glue degrader. This approach is readily adaptable for exploring novel targets across other molecular glue systems, enhancing its applicability and value to the drug discovery community.