Zhiyu Li, PhD

My research centers on transforming human serum albumin into both a target-binding protein and a versatile drug delivery vehicle for long-acting, intracellular therapeutics. Albumin is uniquely suited for this role because of its exceptional stability, long circulation half-life, and natural tumor accumulation. By rationally engineering albumin fusion proteins, I develop “natural” delivery systems that can carry peptides, proteins, and small molecules into cells and even into subcellular compartments. This strategy overcomes long-standing barriers in biologics, such as poor intracellular delivery and rapid clearance, while enabling the co-delivery of multiple therapeutic cargos for synergistic efficacy.

A major focus of my work is the discovery and development of a p53-derived therapeutic peptide that simultaneously targets four key intracellular regulators (MDM2, MDM4, BCL-XL, and MCL-1) involved in apoptosis and survival. By restoring p53 functions through both transcription-dependent and transcription-independent mechanisms, this peptide induces robust cancer cell death. When fused to albumin, the resulting multifunctional biologic delivers efficiently to the intracellular and mitochondrial compartments, exhibits enhanced stability, and reduces off-target toxicity. This work highlights how albumin fusion proteins can access essential intracellular pathways that are traditionally considered difficult to target with drugs.

My research is strongly driven by AI-assisted and computational protein design. I integrate molecular modeling, structure prediction, and rational engineering to guide fusion architecture, linker design, and payload optimization. AI-based tools accelerate the identification of stable, functional protein designs and enable rapid transition from concept to validated biotherapeutics. This computational-experimental loop forms the foundation of a long-term pipeline for discovering next-generation protein drugs, including new biologics aimed at intracellular oncogenic targets such as mutant KRAS.

In parallel, I develop antibody-mimetic systems that function as bi-specific and multi-specific therapeutics without the complexity of traditional antibodies. Using modular protein engineering approaches, these platforms enable precise assembly, programmable targeting, and flexible payload integration. Together, my research integrates drug delivery, DNA damage and apoptosis biology, and protein engineering to create innovative biotherapeutics for cancer and aging-related diseases, bridging rational design with AI-driven discovery to accelerate clinical translation.