Polarizable force fields represent the state of the art method for theoretical studies of biological macromolecules, including proteins. In the proposed study, we will directly test the accuracy of a polarizable force field based on the classical Drude oscillator via calculations of the vibrational Stark effect for probe molecules designed to map the electric field of proteins. The proper description of electrostatics and treatment of molecular polarizability in macromolecular force fields is critical to the development of computational methodologies which can accurately describe interactions between chemical functionalities in proteins. New polarizable force fields include polarizability terms frequently derived from quantum mechanical computations on small model systems in the gas phase. However, in a number of cases the gas phase polarizabilities have been shown to not be applicable for condensed phase simulations, such that scaled polarizability values must be used for selected classes of functional groups. Such scaling factors, which may be determined via the reproduction of dielectric constants of representative pure solvents, are then applied directly to macromolecular systems. Thus, when a macromolecular force field is designed, it contains different scaling factors corresponding to different functionalities, with the combined model assumed to yield an overall correct description of the electronic environment of the macromolecule. To date, a number of polarizable force fields have been applied for molecular simulations of proteins. However, none of these studies has directly validated the electrostatic model of the force field, or optimized the polarizability scaling parameters used in protein simulations. We will address these questions by directly computing the vibrational Stark effect for a probe molecule in a protein environment. The Stark effect is a measure of the shift in vibrations of selected functionalities as a function of chemical environment, information that may be directly related to the electric field surrounding the functionality. This information therefore may be used as a direct test of the ability of a force field to reproduce the electric field around those functional groups.
Information from these calculations will validate assumptions on polarizability scaling as applied to proteins and act as the basis for additional optimization of the force field to more accurately represent the electric fields in proteins. The resulting improved polarizable force field will provide new tools for computational studies of proteins, including drug discovery and optimization, thereby aiding in the design of protein inhbitiors, including novel theraupetic agents.
|Ringer, Ashley L; Mackerell, Alexander D (2011) Calculation of the Vibrational Stark Effect Using a First-Principles QM/MM Approach. J Phys Chem Lett 2011:553-556|