Skip to main content
eScholarship
Open Access Publications from the University of California

UC Merced

UC Merced Electronic Theses and Dissertations bannerUC Merced

Identifying Heterogeneity in Protein Crystal Structures for IRF3 Drug Discovery and Enzyme Design

Abstract

The transcription factor IRF3 is central to the innate antiviral immune response and is a therapeutically attractive target in inflammation, antiviral therapy, and cancer. Yet, no experimentally validated small-molecule ligands have been reported for IRF3. Additionally, while many viral antagonists have been reported to interact with IRF3 and disrupt its function, most lack biophysical confirmation of direct binding. Here, we apply crystallographic fragment screening and complementary biophysical methods to address both gaps. First, we present a crystallographic fragment screen of the IRF3 regulatory domain. Pan-dataset analysis across hundreds of ligand-soaked crystals leverages the statistical power of an averaged data set in identifying outliers in single data sets that represent ligands bound to IRF3. We identify 25 fragments bound across eight sites on the IRF3 regulatory domain. We then test a proposed interaction between IRF3 and the respiratory syncytial virus protein NS1 using biophysical assays. Specific binding was not detected, underscoring the need for biophysical validation of IRF3-viral antagonist interactions. We then extend our methodological approach, treating crystallographic data as a statistical ensemble rather than a route to a single structure, to computational protein design. In a pair of collaborative enzyme-design projects, crystal structures that capture conformational heterogeneity served as inputs to multistate and ensemble-based design strategies. The resulting designed proteins showed tuned conformational equilibria and improved catalytic efficiencies. Together, these projects highlight the value of recovering structural information about heterogeneity that conventional crystallographic methods miss.