Scrutinizing the protein hydration shell from MD simulations against SAXS/SANS data from a worldwide round robin benchmark.

Abstract

The hydration shell of proteins plays key roles in protein stability, folding, recognition, and enzyme activity. However, it remained unclear whether the hydration shell predicted by explicit-solvent molecular dynamics (MD) simulations matches experimental conditions, while accurate experimental probes of the hydration shell structure remained limited. Small-angle scattering (SAS) in solution using X-rays (SAXS) or neutrons (SANS) in principle provides information on the hydration shell, since both the radius of gyration (Rg) and the zero-angle scattering (I0) depend on the hydration shell contrast relative to bulk solvent. Using MD simulations and explicit-solvent SAXS/SANS calculations, we computed Rg and I0 for five different proteins and a set of 18 different combinations of protein force fields and water models, and we validated the simulation results against consensus data from a recent worldwide round robin benchmark (Trewhella, Vachette et al., doi:10.1107/S2059798322009184). Overall, we find remarkable agreement between MD simulations and consensus SAS data; however the agreement is force field dependent. Certain force field/water model combinations underestimate the hydration layer contrast significantly, while water models with increased dispersion interactions may, for some proteins, overestimate the hydration layer contrast. Our study shows that explicit-solvent SAS calculation and consensus SAS data provide a novel route for scrutinizing the hydration layer of proteins.