Enantioselectivity and Enzyme-Substrate Docking Studies of a Ketoreductase from Sporobolomyces salmonicolor (SSCR) and Saccharomyces cerevisiae (YOL151w)

Models for two ketoreductases were created and used to predict the stereoselectivity of the enzymes. One was based on the crystal structure of Sporobolomyces salmonicolor. This model was used to predict the stereoselectivity for 46 ketone reductions using this enzyme; only 6 were incorrectly predicted. The stereochemistries of the products were compared to the experimental values found in the literature. The Prelog rules were also used to predict the stereoselectivity for this enzyme; however the Prelog rules seem to be highly substrate dependent. As a result, predicting stereoselectivity of KREDs is more complicated than is allowed for with just substrate size and geometry. This enzyme showed Prelog docking geometry for 13 substrates if the enzyme is assumed to prefer an anti-Prelog docking geometry. For SSCR the molecular modeling proved to be a better method for predicting stereoselectivity of the enzymes. The second model was a homology model for YOL151w based on the enzyme crystal structure of Sporobolomyces salmonicolor carbonyl reductase, SSCR. In this homology model, 14 compounds were docked and the predicted stereochemistry was compared to the literature values. Of these, 5 were incorrectly predicted.