Computational Design of Zinc Binding Sites at Protein Interfaces and Enzyme Active Sites Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 22, 2019
  • Der, Bryan S.
    • Affiliation: School of Medicine, Department of Biochemistry and Biophysics
  • Engineered proteins will continue to expand the molecular toolkit for applications in medicine, biotechnology, and basic research. While protein engineering efforts often use a parts list limited to the twenty amino acids, metal ions expand the parts list and are critical for the folding and function of 30-40% of known proteins. In particular, zinc ions are common as structural metal sites and catalytic metal sites. Thus, the work described here uses and develops computational methods to engineer structural zinc sites at protein interfaces and catalytic zinc sites at potential active sites. The first chapter discusses the design of a de novo zinc-mediated heterodimeric interaction that targets wild-type ubiquitin. Although zinc binding was successful, a lack of cooperativity resulted in a modest effect of zinc on ubiquitin binding affinity. The second chapter presents a de novo zinc-mediated homodimer as an alternative protein interface design strategy with more cooperative metal binding. Zinc binding improved the homodimer binding affinity by >100-fold, and crystal structures demonstrate moderate accuracy in the design of the zinc sites and the protein-protein interaction. The third chapter reveals the serendipitous discovery of de novo catalysis by this designed zinc-mediated homodimer. This discovery emphasizes the usefulness of protein interfaces for active site formation, the power of zinc for catalysis, and the modest rates achieved thus far in the field of de novo enzyme design. The fourth chapter introduces our efforts to purposefully design a new catalytic motif in a deeper protein cleft. Our approach differs from most enzyme design studies that instead rely on existing catalytic motifs and modify substrate-binding residues. A conformational change shown in the crystal structure of a designed zinc site in a TIM-barrel scaffold emphasizes the importance of second-shell hydrogen bonds to support the primary coordination shell for robust metal binding in deeper protein clefts. In summary, we have endeavored to better understand and more reliably engineer protein structure and function using a predictive computational approach, and as we improve our ability to design zinc sites in proteins, more sophisticated protein functions can be engineered for applied purposes.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Kuhlman, Brian
  • Doctor of Philosophy
Graduation year
  • 2013

This work has no parents.