Direct simulation of non-adiabatic dynamics in large-scale enzymatic systems

Abstract

The simulation of general non-adiabatic processes, such as electron transfer, proton-transfer, and protoncoupled electron transfer, in enzymic systems presents a challenge for theor. methods due to the large system-sizes, the coupling between classical and quantum mech. motions, and the multiple time-scales present in these systems. To address these challenges we have developed the kinetically-constrained ring-polymer mol. dynamics (KC-RPMD) method for the direct simulation of non-adiabatic processes in large-scale biol. and chem. systems. KC-RPMD, like the conventional ring-polymer mol. dynamics (RPMD), is based on Feynman's path-integral formulation of statistical mechanics and is able to simulate quantum dynamical processes using purely classical dynamics; KC-RPMD goes beyond conventional RPMD, allowing for the treatment of general, multi-electron non-adiabatic processes. We apply KC-RPMD in fully atomistic simulations (15,000 atoms in explicit solvent) to understand the long-range electron transfer in various mutants of the blue-copper protein Azurin. Through the anal. of the KC-RPMD trajectories, we show for the first time the importance of donor-acceptor compression and the role of dynamical hydrogen-bonding networks during the electron transfer process in Azurin.