Modeled structure-based computational redesign of a glycosyltransferase for the synthesis of rebaudioside D from rebaudioside A

Abstract

Engineering of enzymes without crystal structures and homologous structures is a great challenge for structure-based computational enzyme design. In this study, a computational strategy that combines protein structure prediction, sequence optimization, and molecular dynamics simulation was developed to improve the catalytic efficiency of the UDP-dependent glycosyltransferase EUGT11 for the synthesis of rebaudioside D from rebaudioside A. The modeled structure of EUGT11 was constructed using a threading-based structure prediction method and the H27–D128 catalytic dyad for glycosylation was identified by bioinformatics analysis. Variant sequences of EUGT11 were created using the computational enzyme design method, and a single variant M2(F379A) was confirmed experimentally to have a catalytic efficiency 2.18-fold higher than that of the wild-type enzyme. The molecular dynamics simulation results revealed that the F379A mutation improved the binding of the sugar acceptor, which may explain the increased catalytic activity of the M2 variant. Although the catalytic efficiency of the M2 variant needs to be further improved for practical use, the developed computational strategy can be applied to improve the properties of enzymes when their structures are unknown.