(479f) Computational Spectroscopy of Protein Structure

Proteins harbor a number of cavities of relatively small volume. Although these packing defects are associated with the thermodynamic instability of the proteins, the cavities also play specific roles in controlling protein functions, e.g., ligand migration and binding. This issue has been studied using several simulation methods. However, these methods are computationally expensive, which prohibits studying larger and biologically more relevant systems. In this study, a new method is developed to identify surface atoms, volume, interior cavity size distribution and the identification of percolation channels in globular proteins. Myoglobin has been studied as an example. This study uses molecular dynamics and Monte Carlo simulations to provide a complete atomic level map of cavities in myoglobin. Our simulation results are consistent with and tie together previous experimental and simulation findings. Specifically, we characterize: (i) All possible trajectories in which the ligand shuttles between the heme and the exterior aqueous environment and (ii) The dynamic properties of interior cavities in myoglobin. The computations are performed both in myoglobin wild-type and V68F myoglobin mutant, which is experimentally known to slow ligand-binding kinetics. We find that the cavity size and span distributions are quite different from two independent systems.