Abstract
A recent theory for the long time dynamics of flexible chain molecules is applied for the first time to a peptide of biological importance, the neurotransmitter met-enkephalin. The dynamics of met-enkephalin is considerably more complicated than that of the previously studied glycine oligomers; met-enkephalin contains the interesting motions of phenyl groups and of side chains relative to the backbone, motions that are present in general flexible peptides. The theory extends the generalized Rouse (GR) model used to study the dynamics of polymers by providing a systematic procedure for including the contributions from the memory function matrices neglected in the GR theory. The new method describes the dynamics by time correlation functions instead of individual trajectories. These correlation functions are analytically expressed in terms of a set of equilibrium averages and the eigenvalues and eigenfunctions of the diffusion operator. The predictions of the theory are compared with Brownian dynamics (BD) simulations, so that both theory and simulation use identical potential functions and solvent models. The theory thus contains no adjustable parameters. Inclusion of the memory function contributions profoundly affects the dynamics. The theory produces very good agreement with the BD simulations for the global motions of met-enkephalin. It also correctly predicts the long-time relaxation rate for local motions.