Programmable Protein Stabilization with Language Model-Derived Peptide Guides

Lauren Hong, Tianzheng Ye, Tian Zi Wang, Divya Srijay, Lin Zhao, Sophia Vincoff, Tianlai Chen, Matthew DeLisa, Pranam Chatterjee

Track: Biology: datasets and/or experimental results

Keywords: Language Models, Peptide Design, Protein Engineering

TL;DR: We use language models to generate target-binding peptides and fuse them to a deubiquitinase to stabilize diverse target proteins.

Abstract: Dysregulated protein degradation via the ubiquitin-proteasomal pathway can induce numerous disease phenotypes, including cancer, neurodegeneration, and diabetes. Stabilizing improperly ubiquitinated proteins via target-specific deubiquitination is thus a critical therapeutic goal. Building off the incredible advances in targeted protein degradation (TPD), recently-described targeted protein stabilization (TPS) modalities rely on a limited set of chemical linkers and small molecule warheads, which do not exist for classically “undruggable” targets. Previously, we engineered ubiquibodies (uAbs) by fusing computationally-designed “guide” peptides to E3 ubiquitin ligase domains for modular, CRISPR-analogous TPD. Here, we engineer “deubiquibodies” (duAbs), via fusion of generated peptides to the catalytic domain of the potent OTUB1 deubiquitinase. In human cells, duAbs effectively stabilize exogenous and endogenous proteins in a DUB-dependent manner. To demonstrate duAb modularity, we swap in new target-binding peptides designed via our generative language models to stabilize unique target proteins, including a transcription factor and a kinase. In total, our duAb system represents a simple, programmable, genetically-encoded strategy for TPS.

Submission Number: 44