Computational design and evaluation of peptide-guided protein degraders

PROJECT SUMMARY Pharmacologically targeting intracellular proteins is a key challenge of modern drug development, requiring innovation and the development of new technologies. This challenge is made more difficult by the fact that many protein targets remain beyond the reach of established drug discovery technologies because they lack easy-to- find and unique binding pockets, possess large, and flat contact areas. Indeed, ~85-90% of the human genome is considered “undruggable,” encoding proteins that are deemed too challenging to bind with conventional molecules. Hence, new approaches are needed for drugging intracellular protein targets. To address this unmet need, this project seeks to develop a new class of computationally-designed mRNA therapeutics that encode peptide-guided protein degraders, known as ubiquibodies (uAbs), for potent and selective degradation of historically undruggable targets not addressable by conventional drugs. Specifically, uAbs are modular, programmable proteins consisting of a genetically engineered fusion between an E3 ubiquitin ligase, linker, and protein/peptide guide. Following ectopic expression in cells, these heterobifunctional chimeras direct the activity of an E3 to a protein of interest (POI), leading to polyubiquitination and subsequent degradation of the POI by the endogenous ubiquitin-proteasome pathway (UPP). The objective of this Phase I STTR is to design customized uAbs against β-catenin and CCAAT/enhancer-binding protein homologous protein (CHOP), two intracellular transcription factors that hold promise as drug targets for hepatocellular carcinoma (HCC), alpha-1 antitrypsin deficiency (AATD), and other liver diseases. The hypothesis of this project is that uAbs can be designed to selectively remove cytosolic/nuclear β-catenin and CHOP, with the potential to inhibit the tumorigenic and proteotoxic potential, respectively, of these drug targets while also limiting toxicity. The plan to address these hypotheses includes first leveraging an artificial intelligence/machine learning (AI/ML)-powered platform to create designer uAbs that selectively degrade cytosolic/nuclear β-catenin and CHOP (Aim 1) and then to develop and evaluate a lipid nanoparticle (LNP)-based strategy for systemically delivering synthetic uAb-encoding mRNAs in cultured cells and mice. The best performing mRNA-LNP formulations will then be evaluated in mice to assess biodistribution, efficiency and duration of target degradation, and biological impacts. Overall, the proposed studies will demonstrate a new paradigm for drugging the proteome based on computational design of peptide-guided uAbs, with proof-of-concept studies in this Phase I proposal focused on accelerating the removal of two key intracellular disease drivers through LNP-mediated delivery of mRNA encoding customized uAbs. Successful completion of this project will lead to a future Phase II application that will explore the therapeutic potential of uAbs following systemic delivery of optimized mRNA-LNP formulations in relevant animal models of HCC and AATD. These studies will serve to advance the development and commercialization of UbiquiTx’s proprietary therapeutic candidates towards an IND.