Insights into generalization of the rate-limiting steps of the dehalogenation by LinB and DhaA: A computational...

LinB and DhaA are well-known haloalkane dehalogenases (HLDs) capable of converting a plethora of halogenated alkanes, also those considered persistent pollutants. The dehalogenation reaction that these two enzymes catalyze has been studied to determine its rate-limiting step (rls) for the last two decades now. As a result, it has been determined that HLDs can show different rate-limiting steps for individual substrates, and at this point we do not have a basis for any generalization in this matter. Therefore, in this work we aimed at gaining insights into the enzymatic dehalogenation of selected dibromo- and bromochloro-ethanes and propanes by LinB and DhaA using computational approach to determine whether defined structural similarities of the substrates result in a unified mechanism and the same rls. By predicting halogen binding isotope effects (BIEs) as well as computing interaction energy for each HLD-ligand complex the nature of the protein-ligand interactions has been characterized. Furthermore, C and Br kinetic isotope effects (KIEs) as well as the minimum free energy paths (MFEPs) were computed to investigate the chemical reaction for the selected systems. Accuracy of the approach and robustness of the computational predictions were validated by measuring KIEs on the selected reactions. Overall results strongly indicate that any generalization with respect to the enzymatic process involving various ligands in the case of DhaA is impossible, even if the considered ligands are structurally very similar as those analyzed in the present study. Moreover, even small structural differences such as changing of one of the (non-leaving) halogen substituents may lead to significant changes in the enzymatic process and result in a different rls in the case of LinB. It has also been demonstrated that KIEs themselves cannot be used as rls indicators in the reactions catalyzed by the studied HLDs.