Are mixed explicit/implicit solvation models reliable for studying phosphate hydrolysis? A comparative study of...

Area of Science:

• Computational Chemistry
• Biochemistry
• Physical Chemistry

Background:

• Phosphate hydrolysis is a fundamental biological process with a controversial reaction mechanism despite extensive research.
• Accurate simulation of solvation effects is crucial for understanding phosphate hydrolysis in solution.

Purpose of the Study:

• To evaluate the reliability of different solvation models for simulating phosphate diester hydrolysis.
• To compare implicit solvation (COSMO), hybrid quantum mechanical/molecular mechanical free energy perturbation (QM/MM-FEP), and mixed solvation models.

Main Methods:

• Simulated solvation free energy for three homologous phosphate diesters using COSMO, QM/MM-FEP, and mixed solvation models.
• Mixed solvation models incorporated explicit water molecules (N=1-3) within an ab initio description, with the remaining solvent treated implicitly.
• Careful selection of explicit water molecules was performed using short QM/MM runs.

Main Results:

• Both COSMO and QM/MM-FEP models reproduced observed Gibbs free energy of hydrolysis (DeltaG(obs)) within approximately 1 kcal mol(-1).
• Mixed solvation models demonstrated instability and missed significant entropic contributions as the number of explicit water molecules increased.
• The reliability of mixed models was found to be highly dependent on the careful selection of explicit water molecules.

Conclusions:

• Implicit solvation models (COSMO) and QM/MM-FEP are reliable computational approaches for studying phosphate hydrolysis.
• Mixed solvation models, as implemented, are unreliable for accurately modeling phosphate hydrolysis in solution due to instability and entropic inaccuracies.
• Further development is needed to improve the accuracy and stability of mixed solvation approaches for biological reaction simulations.