The Influence of Different Treatments of Electrostatic Interactions on the Thermodynamics of Folding of Peptides^†

Replica exchange molecular dynamics simulations were performed to investigate the effects of different electrostatic treatments on the structure and thermodynamics of a small β-hairpin forming peptide. Three different electrostatic schemes were considered:  regular cutoffs, generalized reaction field (GRF), and particle mesh Ewald (PME), with the peptide modeled using OPLS/AA all-atom force field with explicit TIP3P water. Both the GRF and PME methods yielded results consistent with experiment, with free energy surfaces displaying a single minimum corresponding to the native β-hairpin structure. In contrast, use of straight cutoffs led to the population of an additional local minimum corresponding to nonhairpin conformations that compete with the formation of the native β-hairpin at low temperatures. This extra minimum would not be apparent in conventional constant-temperature molecular dynamics simulations run for a few nanoseconds. This result points to the critical need of careful sampling of conformational space to assess the quality of different numerical treatments of long-range forces. While differences emerged in the nature of the unfolded states populated using PME and GRF approaches, simulations on the β-hairpin forming peptide and on two additional control peptides indicate that the GRF treatment of electrostatics offers a satisfactory compromise between accuracy and computational speed for the identification of low-energy conformations. A GRF-based approach emerges as a viable means for treating larger biological systems that would be prohibitively costly to simulate using PME methods.