UC Irvine

Proteins, large biological molecules synthesized by living organisms, serve a wide array of functions. Many proteins operate as molecular scaffolds for binding to other proteins; in fact, the vast majority of molecular interactions in biology are made possible by protein-protein interactions. Over the past 15 years, powerful techniques have been developed to generate protein-based ligands in vitro to virtually any protein target. The fundamental steps to any protein engineering effort remain the same. The protein target of interest is modified to create a mutant with preferable biochemical characteristics. The protein is then expressed and purified in large quantities for use in therapeutic or diagnostic applications. In this thesis, I describe a multifaceted approach to address the diversification, identification, and production steps of the protein engineering process.

Currently, the vast majority of binding molecules that have been developed for use in biomedical research are either antibodies or antibody derivatives. Immunoglobulins, such as antibodies, are notoriously difficult to produce. These large, multi-domain proteins require complex cloning steps for recombinant expression in mammalian cell lines. Alternatively, researchers have begun to engineer binding proteins outside the immunoglobulin family by using protein scaffolds with structurally rigid cores accompanied by solvent-accessible surface loops. We characterize such a protein scaffold, the major tropism determinant of Bordetella bronchiseptica bacteriophage, which is under the control of a diversity generating genetic element. This approach allows automated generation of a large phage-displayed protein receptor library for use in selection experiments to identify binders to any protein target of choice.

Following the identification of the protein ligand, successful recombinant expression of the protein is crucial to commercial viability. In bacterial expression hosts, proteins will often misfold and clump into inclusion bodies. This protein aggregation results in drastically decreased expression yields. Returning the misfolded protein to the native state conformation by traditional methods requires laborious and time-consuming processing steps. We investigate a novel method for applying shear stress to rapidly refold proteins from inclusion bodies. This research, combined with further studies of non-immunoglobulin protein scaffolds, could drastically lower the cost of protein engineering efforts and enable new biological applications in the future.