Automated Design of Efficient and Functionally Diverse Enzyme Repertoires

Abstract

Summary Substantial improvements in enzyme activity demand multiple mutations at spatially proximal positions in the active site. Such mutations, however, often exhibit unpredictable epistatic (non-additive) effects on activity. Here, we describe FuncLib - an automated method for designing multipoint mutations at enzyme active sites using phylogenetic analysis and Rosetta design calculations. We applied FuncLib to two unrelated enzymes, a phosphotriesterase and an acetyl-CoA synthetase. All designs were active and most showed activity profiles that significantly differed from the wild type and from one another. Several dozen designs with only 3-6 active-site mutations exhibited 10-4,000-fold higher efficiencies with a range of alternative substrates, including the hydrolysis of the toxic organophosphate nerve agents soman and cyclosarin and the synthesis of butyryl-CoA. FuncLib is implemented as a web-server (http://FuncLib.weizmann.ac.il); it circumvents iterative, high-throughput screens and opens the way to design highly efficient and diverse catalytic repertoires.

Design of a basigin‐mimicking inhibitor targeting the malaria invasion protein RH5

“…Four mutations that resulted in substantial improvements in affinity occurred at the core of the RH5‐binding surface (Val26Ile, Gln100Ile, His102Tyr, and Ser190Gln), whereas three others were in the periphery (Thr29Ser, Val30Ala, and Gln164Phe). Two of these positions (100 and 102) form close atomic contacts with one another, suggesting that multipoint mutations at least in these positions may exhibit complex epistatic relationships …”

“…The theoretical sequence space of all combinations of mutations at nine positions is very large (10 12 different sequences), whereas the space of affinity‐enhancing variants as inferred from deep mutational scanning is much smaller (14 400 sequences; Table ), thus focusing design calculations on a small set of sequences that are likely to be enriched for improved binding affinity. In preliminary calculations, we noticed that the design procedure converged on variants that were at most three mutations relative to human basigin, reflecting the tendency of design algorithms to favor the crystallographically determined structure . To force additional mutations, we visually inspected the basigin‐RH5 structure and eliminated the wild‐type identities from the design choices at six of the nine positions that appeared permissive to mutation and used Rosetta to find an optimal combination of the remaining mutations.…”

“…Two of these positions (100 and 102) form close atomic contacts with one another, suggesting that multipoint mutations at least in these positions may exhibit complex epistatic relationships. 33 Past applications of deep mutational scanning to improve binding affinity used an iterative experimental selection of high-affinity variants from libraries comprising combinations of affinity-enhancing mutations. 19,21,34 Here, instead, we tested whether Rosetta design could circumvent such iterative experimental screening by finding a single optimal combination of affinity-enhancing mutations.…”

“…In preliminary calculations, we noticed that the design procedure converged on variants that were at most three mutations relative to human basigin, reflecting the tendency of design algorithms to favor the crystallographically determined structure. 33,35,36 To force additional mutations, we visually inspected the basigin-RH5 structure and eliminated the wild-type identities from the design choices at six of the nine positions that appeared permissive to mutation and used Rosetta to find an optimal combination of the remaining mutations. This resulted in a single design, comprising seven mutations that we called BSG des7 (Table 1).…”

The Ketosynthase Domain Constrains the Design of Polyketide Synthases

“…We therefore aimed at a more targeted approach to guide multipoint mutagenesis of KS domains, using the FuncLib server, a program that generates a variety of diverse multipoint mutants, by calculating stable networks of interacting active site residues. 32 FuncLib uses phylogenetic information to reveal possible, coupled mutations within a protein, by working on two levels: First, it uses phylogenetic information to suggest amino acid exchanges at selected positions, and second, it ranks the mutated proteins by calculating their stability via the Rosetta program suite. In this process, mutations that are predicted to destabilize the protein fold are discarded.…”