Water structure in solution and crystal molecular dynamics simulations compared to protein crystal structures

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• Protein function is dictated by its atomic structure and surrounding solvent.
• Accurate modeling of protein hydration is crucial for computational studies of protein structure.

Purpose of the Study:

• To compare water structure from molecular dynamics (MD) simulations with crystallographic water molecules.
• To evaluate MD simulations in both solution and crystal environments.
• To assess the accuracy of MD in reproducing protein hydration in crystal structures.

Main Methods:

• MD simulations of galectin-3 in complex with ligands were performed in a water box and a crystallographic unit cell.
• Comparison of simulated water positions to crystallographic water molecules.
• Development and application of a local clustering approach to identify and track water sites.

Main Results:

• Direct comparison of water positions showed poor agreement between MD and crystallography.
• The local clustering approach significantly improved the agreement, indicating MD reasonably reproduces crystallographic water structure.
• Crystal MD simulations identified potential unmodeled water molecules in crystal structures.

Conclusions:

• The challenge in reproducing crystallographic water structure lies in identifying dynamic water sites around flexible proteins.
• The local clustering method effectively addresses this challenge, validating MD simulations.
• Crystal MD simulations offer a complementary approach for water identification and modeling in protein crystallography.