Title: Computational Design of Tailor-made Enzymes

Abstract

Enzymes are powerful and versatile protein-based catalysts with diverse applications:  from environmentally-friendly synthesis of pharmaceuticals to the development of novel “catalytic drugs”. Their effective use is contingent on our ability to control their properties including stability, substrate selectivity, and spatiotemporal regulatability. We have developed a bottom-up design approach combining computational molecular modeling, high-throughput experiments and machine learning to obtain enzymes with tailor-made properties for diverse applications. In this talk, I will describe our efforts in developing generalizable methodology for enhancing enzyme stability (1,2), dialing in selectivity (3,4) and endowing control over enzymes by chosen external stimuli (5,6). In each case, a first-principles understanding of enzymes and protein structure forms the foundation for design, and characterization of designed proteins provides further insight into the interplay between enzyme structure and mechanism.

References:

1. Proc. Natl. Acad. Sci. (2017) 14: 12472-12477 2. Biochemistry (2022) 61:1041–1054.

3. PLoS Comp. Biol. (2017) 13:e1005614 4. Proc. Natl. Acad. Sci. (2019) 116:168-176

5. J. Am. Chem. Soc. (2018) 140:14-17 6. Proc. Natl. Acad. Sci. (2022) 119: e2116097119