Force Field Limitations of All-Atom Continuous Constant pH Molecular Dynamics

Area of Science:

• Computational chemistry
• Biophysics
• Molecular dynamics

Background:

• Constant pH molecular dynamics (CpHMD) simulations are vital for studying pH-dependent biological phenomena.
• Protonation equilibria (pKa values) are influenced by electrostatic interactions and solvation, making them sensitive to the chosen force field and water model.

Purpose of the Study:

• To investigate the impact of different protein force fields on pKa calculations using all-atom Particle Mesh Ewald (PME) continuous constant pH (PME-CpHMD) simulations.
• To evaluate the accuracy of Amber ff19sb and ff14sb force fields with their respective water models for pKa predictions.

Main Methods:

• Replica-exchange titration simulations were performed using PME-CpHMD.
• Two Amber force fields (ff19sb and ff14sb) paired with their standard water models (OPC and TIP3P, respectively) were tested on a mini-protein (BBL).
• The influence of atom-pair specific Lennard-Jones corrections (NBFIX) was also assessed.

Main Results:

• Both Amber force fields, ff19sb and ff14sb, showed overestimated pKa downshifts for key residues (His166, Glu141, Glu161).
• These errors stem from issues like undersolvation of neutral histidines and overstabilization of salt bridges.
• ff19sb with OPC water demonstrated higher accuracy compared to ff14sb with TIP3P water.
• NBFIX corrections partially mitigated the pKa downshifts related to salt bridges.

Conclusions:

• The accuracy of pKa calculations in constant pH simulations is highly dependent on the quality of the protein force field and water model.
• Improvements in force fields are essential for reliable predictions of protonation equilibria in biological systems.
• The ff19sb force field with OPC water shows promise for more accurate pKa calculations.