Quantitative prediction of enantioselectivity of Candida antarctica lipase B by combining docking simulations and quantitative structure–activity relationship (QSAR) analysis

Prediction of enzyme enantioselectivity in silico could be of major utility for avoiding the expensive and time-consuming experiments. Herein, we aimed to develop a new approach to construct a quantitative enantioselectivity prediction model with high accuracy for Candida antarctica   lipase B (CALB). In the work, Autodock was used to generate substrate conformations for improving the calculation efficiency, followed by the quantitative structure–activity relationship (QSAR) analysis. The effects of acyl donors and 5 molecular interaction fields (steric, electrostatic, hydrophobic, hydrogen bond donor and acceptor fields) on model construction were investigated. The results indicated that the application of actual acyl donors was indispensible for model construction. Inclusion of the relevant molecular interaction fields could significantly improve the predictive accuracy which suggested that enantioselectivity was a consequence of multiple molecular interactions. The final model was derived based on four molecular interaction fields (steric, electrostatic, hydrophobic, hydrogen bond acceptor fields) with actual acyl donors owning higher predictive accuracy (Rpred2=0.92) than previous report (Rpred2=0.79). Furthermore, the contour map produced by the model facilitated us to better elucidate the molecular basis of enzyme enantioselectivtiy, and was potential for the application of rational design of the enzyme.

Figure optionsDownload as PowerPoint slideHighlights
• Quantitative enantioselectivity prediction model combining QSAR with Autodock.
• Explore the effects of acyl donors and relevant force fields on model construction.
• The model based on actual acyl donors and core force fields predicts more accurate.
• The model helps us to better elucidate the origin of enzyme enantioselectivtiy.
• The contour map produced was potential for the rational design of enzyme.