Adaptive QM/MM Methods for Proton Transfer in Complex Environments

University Of Colorado At Denver-Downtown Campus, Denver CO

Investigators

Abstract

Hai Lin of the University of Colorado at Denver is supported by an award from the Chemical Theory, Models and Computational Methods program to develop and apply computational methods to study proton transfer in biological environments. Just as every day we walk through doors of houses, offices, schools, and shops; every minute, ions that are essential to life pass through doors of biological cells. These doors on the surface of cells are called ion channels and transporters. Situated in cellular membranes, these special proteins control which ions move across the membrane and when. Among these proteins are the CLC proton/chloride antiporters, which regulate the influx of protons and the outflow of chloride ions in a coupled manner. Details of how the CLC proton/Chloride antiporters work in transferring protons remain elusive. Moreover, it is still under debate if (and how) the proton interacts with chloride ions during its journey through the translocation pore. Professor Lin and his research team, which includes two undergraduate students, seek answers to these questions through computer simulations. They are developing advanced computational techniques that combine quantum- and classical-mechanical models. They then apply these methods to study the CLC transporters. As malfunctioning doors can block us from reaching our destination, malfunctioning ion channels and transporters can disrupt normal life processes and cause diseases. The research by the Lin group seeks to deepen our understanding of how the CLC transporters work at a fundamental, molecular level, which may, in turn, assist the development of novel therapies for diseases related to these protein malfunctions. The Grotthuss-shuttling mechanism for proton transfer involves dynamical reorganizations of covalent and hydrogen bonds. While reactive force fields (also called molecular-mechanics, or MM) and multistate empirical valence bond models are efficient in treating water dissociation and proton hopping. The extension of these models in complex environments is challenging because of the rapidly increasing number of potential parameters needed to account for more valence configurations. Moreover, electronic-structure information is missing for the bond breaking/forming processes, which are described ideally by quantum mechanics (QM). QM simulations are limited to small model systems (~hundreds of atoms) due to high computational costs. The Lin group is developing novel adaptive-partitioning QM/MM methods for proton transfer in complex environments. The adaptive-partitioning schemes feature an on-the-fly smoothly updated mobile QM region that follows the proton wherever it goes. As such, a finite-size QM region becomes infinite and can, in principle, sustain simulations as long as needed. The new algorithms are applied to the study of a prototype CLC antiporter from E. coli. With on-the-fly computed QM/MM potentials and an explicit treatment of protons, the dynamics simulations explore the hydrophobic pore hydration, water wire formation, proton translocation, and chloride ion protonation in the CLC antiporter.