Polarizable Force Field for Proteins and Lipids

University Of Chicago, Chicago IL

Investigators

Linked publications & trials

Abstract

? DESCRIPTION (provided by applicant): Computations based on atomistic models play an increasingly important role in understanding biomolecular systems as well as in drug development. Improvements in these models involve extensions of the underlying functional form of the potential energy as well as additional optimization targeting a wider range of experimental and quantum mechanical data. During the last funding period we made significant advances in the development of empirical force fields (FF) for proteins and lipids, with improvements to the CHARMM additive models and the production of polarizable models for proteins, lipids and ions based on the classical Drude oscillator model. The Drude FF has already been implemented in CHARMM, NAMD, ChemShell QM/MM and the OpenMM GPU suite, and is currently usable for MD simulations on the order of one microsecond as well as with Temperature and Hamiltonian Replica-Exchange sampling methodologies. In the proposed study we will investigate how the explicit treatment of electronic polarization contributes to the structure, dynamics and biological functions of proteins, lipids and ligand binding. In Aim 1 we will apply the polarizable FF to investigate the physical forces driving the folding and conformational properties of peptides and proteins as well as evaluate and further optimize the protein model targeting a range of properties. These will include quantum mechanical data, NMR observables, pKa shifts and aqueous solution data on ionic and polar neutral species representative of biomolecules, including osmotic pressure and density experimental data measured as part of this study. Membrane and protein-membrane complexes will be studied in Aim 2 using the polarizable FF with emphasis on the permeation of small species, translocation of cell penetrating peptides, and interpretation of experimental data from solution and solid state NMR, scattering, voltage-sensitive membrane-bound chromophores and 2D-IR spectroscopy. Information from these calculations will allow for additional optimization of the lipid FF and its extension to unsaturated and anionic lipids, cholesterol and sphingomyelin. Aim 3 will investigate protein-ligand interactions including the forces driving the binding of ions and drug-like molecules. The impact of electronic polarization on these interactions will be investigated with the goal of achieving a more accurate representation of ligand binding. The energy function will be extended to account for charge transfer in the case of ion binding if it is deemed necessary. Additional efforts will include development of an automated parameter optimization utility for drug-like molecules. Upon completion of the proposed study we will have an improved understanding of the physical forces driving protein and membrane structure and dynamics based on a highly optimized state-of-the-art polarizable empirical FF that will be available to the computational chemistry community, including the capability to apply the polarizable model in drug discovery.