Quantum mechanical methods for enzyme modeling: Accurate computation of kinetic isotope effects

A number of recently developed quantum mechanical approaches for enzyme kinetics modeling have been described, including the treatment of the potential energy surface for reactive systems and the incorporation of nuclear quantum effects in dynamics simulations. Two aspects are emphasized: (1) the quantum mechanical representation of reactive potential energy surfaces for enzyme reactions, and (2) the incorporation of nuclear quantum effects in enzyme dynamics. A free energy perturbation approach that transforms one isotope into another was developed in centroid path integral simulations for accurate computation of primary and, importantly, secondary kinetic isotope effects in enzymes. The latter method is called the PI-FEP/UM approach, which couples with a novel application of the bisection sampling technique to achieve robust and converged results in these calculations. For constructing reactive potential energy surfaces, a novel mixed molecular orbital and valence bond (MOVB) theory has been developed, which can be used in context of combined QM/MM potential as the QM representation, and fully coupled with the quantal X-Pol force field as an integral part of the block-localization that includes non-adiabatic coupling. These methods have been illustrated by a number of examples, ranging from gas-phase potential energy surface to chemical reactions in water to enzymatic processes. These examples show that the incorporation of quantum mechanical effects is essential for enzyme kinetics simulations and the methods described here offer a great opportunity to more accurately and reliably model the mechanism and free energies of enzymatic reactions.