Computational design of new and improved antibodies and enzymes

Hestrin prize lecture of the Israel Society for Biochemistry and Molecular Biology, co-awarded with Meytal Landau

Enzymes and antibodies developed for research and clinical applications may exhibit suboptimal stability, expressibility, affinity or activity. Existing optimization strategies rely on laborious iterations of experimental diversification and screening and often end in failure. We are developing a new strategy that relies on phylogenetic analysis and Rosetta atomistic design calculations to eliminate such iterations and directly design dramatically improved proteins. We demonstrate that this strategy can automatically design variants of enzymes, antibodies, and vaccine immunogens that exhibit orders of magnitude improvements in expression levels, affinity, selectivity and catalytic rate. Thus, evolution-guided protein design enables direct optimisation of biophysical and functional aspects that are essential for the development of protein therapeutics. Current efforts focus on the design of high-affinity antibodies that target challenging conserved epitopes.

Netzer et al. Ultrahigh Specificity in a Network of Computationally Designed Protein-Interaction Pairs. Nat. Commun. 2018.

Khersonsky et al. Automated Design of Efficient and Functionally Diverse Enzyme Repertoires. Mol. Cell 2018.

Goldenzweig et al. Automated Structure- and Sequence-Based Design of Proteins for High Bacterial Expression and Stability. Mol. Cell 2016.

Campeotto et al. One-Step Design of a Stable Variant of the Malaria Invasion Protein RH5 for Use as a Vaccine Immunogen. Proc. Natl. Acad. Sci. U. S. A. 2017.