The overall goal of this research is to improve potential energy functions for molecular dynamics (MD) computer simulations of biological macromolecules, focusing on developing a realistic cytoplasmic environment that is based on a fundamental understanding of solute-solvent interactions. While much effort has been invested in developing the potential energy functions for intramolecular interactions of proteins, nucleic acids, sugars, membranes, etc., integration of new developments in water models into the biomolecular force fields has lagged. In addition, while most water model development has focused on pure water properties, our model development has also included the solvation properties of water for biological molecules, but the lack of experimental structural data for solute-water interactions to use as targets has hampered our efforts. Thus, in this collaborative project, we will determine structural data from new neutron diffraction experiments and we will develop new, structure-based solute-water potentials for water, ions, and co-solutes that can be used with existing biomolecular force fields. For this exploratory project, Na+, K+, Cl-, NO2-, and NO3- were chosen as important components in the cellular environment. The new experiments will use neutron diffraction with isotopic substitution (NDIS), a powerful method for separating the solute-solute, solute-water and water-water correlations, to generate high- resolution solute-water structural data. Although some neutron diffraction data for the solutions of interest exists, recent improvements in beam intensities, instrument stability and sensitivity, and data analysis will give the greater accuracy necessary for parameter development. In addition, the use of partial structure factors for certain pairs from the simulations using the new potentials to aid the separation of the different pairs in the neutron data will be explored. Simulated structure factors will be back-transformed and directly compared to the experimental data in Q space, thus forming a feedback loop between the simulation and experiment. The water potential will use the soft-sticky dipole-quadrupole-octupole (SSDQO) model, which consists of molecular multipoles centered on a single site for the entire molecule. SSDQO reproduces numerous properties of liquid water over a wide range of conditions, mimics the electrostatics of the quantum mechanical electron density of a water molecule better than typical models, and yet is computationally efficient. Development will be in CHARMM, a widely used biomolecular simulation program.
The specific aims of this research are: (1) Determine highly accurate experimental structures of ion-water solutions (2) Develop the ion-SSDQO force field using the new data
The development of multiscale, multipole computer simulations will make simulations of cellular processes feasible with current computer capabilities. Such simulations are important for gaining a molecular understanding of the disruption of health cell function by viruses, toxins, and other disease causing agents, thus paving the way for designing new drugs to combat disease.
|Ichiye, Toshiko (2016) What makes proteins work: exploring life in P-T-X. Phys Biol 13:063001|