Explicit Ligand Hydration Shells Improve the Correlation between MM-PB/GBSA Binding Energies and Experimental...

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Drug Discovery

Background:

• Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) and Molecular Mechanics Generalized Born Surface Area (MM-GBSA) are standard computational methods for drug design.
• The impact of explicitly including water molecules in these calculations on binding affinity prediction accuracy remains debated in scientific literature.

Purpose of the Study:

• To investigate the effect of explicitly considering variably populated hydration shells on the correlation between MM-PB/GBSA binding energies and experimental biological activities.
• To determine an optimal range of water molecules for inclusion in MM-PB/GBSA calculations to enhance predictive accuracy.

Main Methods:

• Systematic evaluation of ligand hydration shells containing 10-70 water molecules across four diverse biological systems (DNA-topoisomerase, α-thrombin, penicillopepsin, avidin).
• Calculation of binding energies using MM-PB/GBSA methods with varying numbers of explicit water molecules.
• Correlation analysis between computed binding affinities and experimental data (IC50, ΔGbind).

Main Results:

• Explicitly considering hydration shells with 30-70 water molecules consistently improved the correlation between MM-PB/GBSA calculated binding affinities and experimental activities across all tested systems.
• The observed improvement in correlation was achieved with a negligible increase in computational expense.
• This suggests an optimal range for incorporating explicit water molecules in MM-PB/GBSA calculations.

Conclusions:

• Explicitly including a specific number of water molecules (30-70) in MM-PB/GBSA calculations enhances the accuracy of binding affinity predictions.
• This refined approach offers a valuable strategy for improving drug design and discovery pipelines.
• The findings provide practical guidance for optimizing MM-PB/GBSA methodology in computational drug development.