Multiscale Enhanced Sampling Simulation and its Application to Intrinsically Disordered Protein, Sortase A

Abstract

Protein free energy landscapes derived from all-atom molecular dynamics simulation have played an important role in the elucidation of protein functional dynamics with high structural and energetic resolution. Since the characteristic time scale of biologically relevant processes such as protein structural changes far exceeds the feasible computational time, the calculation of protein free energy landscapes requires the acceleration of sampling and mapping along the reaction coordinates or the pathway of structural change. Here, a multiscale simulation method, “multiscale essential sampling(MSES)”, has been proposed for calculating free energy surface of proteins in a “sizable” dimensional space with “good scalability”. In MSES, the configurational sampling of a full-dimensional model is enhanced by coupling with the accelerated dynamics of the essential degrees of freedom. The Hamiltonian exchange method can remove the biasing potential in MSES, and allows us to derive the free energy surface of the essential degrees of freedom. The formula of MSES ensures good scalability in the Hamiltonian exchange. As an application, the MSES simulation has been performed of an intrinsically disordered protein, sortase A, a transpeptitase in Gram-positive bacteria. Sortase A cleaves a C-terminal sorting signal of the surface proteins at a conserved LPXTG motif with the help of a calcium ion, leading to adhesion to the cell wall peptideglycan. The solution structures with and without a peptide (LPAT) reveal that a disordered loop undergoes a disorder-to-order transition upon peptide binding. Comprehensive conformational sampling of the disordered loop has been performed to elucidate how flexible the disordered loop in the peptide free form is and how the peptide and calcium binding affect the flexibility. The free energy landscape thus calculated clarifies the role of the calcium ion in the enzyme activity.