Computational library design and screening: Creating elephant paths in enzyme evolution
September 24-28, 2017
Traditional directed evolution employs iterations combining mutagenesis to generate genetic diversity and phenotypic selection to find improved variants. The final result typically consists of enzyme variants carrying multiple mutations, sometimes replacing up to 15% of the amino acids, and such variants can exhibit highly attractive properties. The improvements must be based on biophysical principles, such as improved hydrophobic packing and long range electrostatic interactions that account for better stability, removal of steric clashes to broaden substrate range, or modified electrostatics to change pH optima. Since many of these biophysical principles are known and can be modeled with computational algorithms, it is challenging to pursue the replacement of laborious and time-consuming directed evolution protocols by computational workflows. This is especially attractive if the desired properties cannot be screened in a high-throughput format, e.g. due to the complexity of an expression system or the lack of miniaturized assays. Accordingly, we are exploring the use of computational protein design, docking simulations and molecular dynamics simulations of mutant enzyme libraries to develop enzyme engineering strategies in which most of the laboratory screening is replaced by in-silico methods. We improved thermostability (ΔTm,app +22-35ºC) and cosolvent (DMSO, DMF, methanol) compatibility of an epoxide hydrolase, two dehalogenases and a peptide amidase using such a computational design and screening strategy (FRESCO, framework for rapid enzyme stabilization by computational library design). We also examined the use of a computational workflow (CASCO, catalytic selectivity by computation) that included active site redesign and molecular dynamics simulations on large numbers of mutant enzymes to predict multi-site mutants with controlled substrate selectivity. These were made in a single step to give highly enantioselective and enantiocomplementary epoxide hydrolases for use in the conversion of meso substrates to enantioenriched diols as well as enantioselective hydrolases for use in biocatalytic kinetic resolutions.
Please click Additional Files below to see the full abstract.
Dick Janssen and Hein J. Wijma, "Computational library design and screening: Creating elephant paths in enzyme evolution" in "Enzyme Engineering XXIV", Pierre Monsan, Toulouse White Biotechnology, France Magali Remaud-Simeon, LISBP-INSA, University of Toulouse, France Eds, ECI Symposium Series, (2017). http://dc.engconfintl.org/enzyme_xxiv/139