Engineering Modular Protein Inhibitors for Selective Targeting of Pathological Biomolecules

PROJECT SUMMARY/ABSTRACT Multifunctional proteins are excellent candidates for developing novel therapeutics, however, natural proteins have a limited span for binding affinity, selectivity for a specific target, or propensity to control binding to proteins other than their intended therapeutic target. My research lab aims to enhance the protein engineering and design toolbox by exploring the relationships between protein sequence, structure, and function in multifunctional proteins. We aim to engineer designer protein scaffolds that can bind selectively to one or more specific targets, minimizing off-target effects. Multifunctional metalloproteinases (MPs) and their natural inhibitors, tissue inhibitors of metalloproteinases (TIMPs), interact with each other and other cell signaling molecules in a complex network to orchestrate key cellular functions, including remodeling of the extracellular matrix (ECM). TIMPs have unique features that make them ideal scaffolds for protein engineering and design, including five flexible loops that provide a vast protein-protein interacting interface for improving or switching multifunctional proteins. TIMPs share structure and function similarities with a single chain fragment of monoclonal antibody (scFv mAb) in binding to the target. mAbs are well-established protein therapeutics with numerous FDA-approved examples, however, mAbs are not natural inhibitors of enzymes and many other ECM targets. We aim to harness the full potential of mAbs and the multifunctional capability of TIMPs by generating engineered hybrid mAb-TIMP protein scaffolds using a combination of experimental and computational approaches. To reach this goal, we need to overcome the limitations of directed evolution for protein engineering and design multifunctional enzyme inhibitors with high selective binding to specific targets using computational approaches like machine learning. This MIRA grant will address these limitations through a five-year strategy composed of two primary research strategies. In research strategy 1, we aim to develop new protein engineering and design platforms for engineering multifunctional enzyme inhibitors by modifying yeast surface display and high throughput screening methods. Using machine learning, we will generate a feedback loop between cycles of directed evolution and rational protein design that will allow us to engineer and design novel TIMP variants with absolute selectivity for specific MPs or other cell signaling proteins. In research strategy 2, we aim to understand the molecular mechanism of engineered proteins --- those driving selective binding to MPs and cell receptors, and growth factors using protein structural studies. Further, we will understand the effect of these modular multifunctional protein inhibitors in cell-based assays. The scope of this work is beyond the interaction of mAbs or TIMPs with target proteins in ECM—the strategies and tools developed here can be used to engineer and design any multifunctional protein, enzyme, or inhibitor to regulate complex biological systems and develop new and improved protein therapeutics with higher selectivity for multiple pathological protein targets with low side effects.