Chinese University of Hong Kong

Covalent inhibitors have re-emerged as powerful therapeutic agents, offering the ability to modulate "undruggable" proteins including oncogenic drivers that have eluded traditional drug discovery efforts. Expanding the repertoire of electrophilic moieties remains a critical frontier in covalent drug discovery. Here, we report a Rh(I)-catalyzed cycloisomerization of N-allyl bicyclo[1.1.0]butane amides (BCB amides) to access the 5-methylene-3-azabicyclo[5.1.0]octan-4-one (MABO) scaffold via strain-release reactivity. Both N-allyl BCB amides and MABOs can react with nucleophiles, highlighting their potential as electrophilic moieties for covalent drug design. Kinetic assays against glutathione revealed that selected N-allyl BCB amides and MABOs exhibit half-lives comparable to that of (±)-sotorasib, whose enantiomerically pure form received FDA approval as an anticancer drug. Biological evaluation in human-derived head and neck squamous cell carcinoma (HNSCC) models identified specific BCB amides and MABO compounds capable of modulating cancer cell growth. The toxicity of the compounds was evaluated by flow cytometry in human peripheral blood mononuclear cells (PBMCs) and by monitoring LDH release in the monocytic cell line THP-1, both under basal and activated conditions. Together, these findings suggest that BCB amides and MABOs represent promising classes of electrophilic scaffolds for covalent drug discovery.

Selective recognition of cancer-associated proteins (CAPs) by antibodies, followed by their delivery into the intracellular organelle, the lysosome, results in targeted degradation of CAPs and suppresses the growth of cancer cells. Translocating the antibody-CAP complex across the plasma membrane is, however, nontrivial. Phase-separating molecules are known to form membrane-translocating coacervates that can encapsulate proteins and transport them into the cytoplasm. Nevertheless, these coacervates generally lack the ability to guide the cargo to the lysosome. Here, we seal this gap and develop lysosome-targeting coacervates by tailoring a tetrapeptide into a phase-separating, coacervate-forming peptide. In the aqueous solution, the peptide derivative forms microdroplets, or coacervates, through liquid-liquid phase separation (LLPS), which spontaneously enter cells and colocalize with the lysosome; hence, these coacervates are referred to as Lysosome-Sorting Peptide Coacervates or LSP-Coa. We show that LSP-Coa can encapsulate proteins, facilitate the translocation of antibody-CAP complexes to the lysosome, and enable the degradation of membrane-bound CAPs - a mechanism we call Coacervate-mediated Lysosome-targeting Protein Degradation, or CoaLPD. Using the CoaLPD technology, we successfully degraded HER2 and EGFR in cancer cells and in tumor-bearing mice, showcasing its potential use as an anticancer treatment. The LSP-Coa system also increases the efficacy of PROTAC degradation through enhanced lysosomal uptake. Taken together, we present the design of lysosomal-targeting coacervates and demonstrate their use as vehicles for lysosome-specific antibody delivery and for the selective degradation of CAPs, thereby validating the CoaLPD strategy as a potential anticancer treatment.