Biomolecular Chemical Simulations on Enantioselectivity and Reactivity of Lipase Enzymes to Azulene Derivatives

Biomolecular chemical simulations have recently become a useful research method in the fields of organic chemistry and bioscience. In the last few years, we have been focusing on the biomolecular computational simulation on lipase enzyme and ligand complexes to predict the enantioselectivity and reactivity of lipases toward non-natural organic compounds. In this paper, we describe the molecular simulations including molecular dynamics (MD) and fragment molecular orbital (FMO) calculations for the complexes of Candida antarctica lipase type A (CALA) and trifluoromethylazulene alcohol derivatives. From the MD calculations, we found that the fast-reacting enantiomer of esters with high enantioselectivity stays in the vicinity of the active site of CALA, while the slow-reacting enantiomer leaves the active site of CALA. On the other hand, both (R)- and (S)-enantiomers of ester with low ensntioselectivity were found to keep near to near the active site of CALA. Further, for the esters that do not react with lipase enzyme, we found that both (R)- and (S)-enantiomers move away from the active site of lipase enzyme. From the FMO calculations, we found that each fast-reacting enantiomer of esters with high enantioselectivity strongly interacts with certain particular amino acid residues in CALA containing Asp95, while both (R)- and (S)-enantiomers of ester with low enantioselectivity interact with same amino acid residues in CALA including Asp95. These results suggest that it is possible to predict not only the enantioselectivity but also the reactivity of CALA and to identify the amino acid residues important to the enzymatic reaction. Therefore, we consider that our computational simulations would be a useful method for predicting and understanding the reactivity and the enantioselectivity of lipase-catalyzed biotransformations.