Rationalizing protein-ligand interactions via the effective fragment potential method and structural data from...

Area of Science:

• Computational Chemistry
• Structural Biology
• Drug Design

Background:

• Accurate prediction of protein-ligand interactions is crucial for structure-based drug design.
• Non-covalent interactions play a key role in molecular recognition and binding affinity.
• Polarizable quantum mechanics-based force fields offer a promising avenue for modeling these interactions.

Purpose of the Study:

• To evaluate the efficacy of the Effective Fragment Potential (EFP) method in calculating protein-ligand interactions.
• To assess the influence of dynamic and solvent effects on binding affinity predictions.
• To explore the application of EFP in structure-based drug design for cyclin-dependent kinase 2.

Main Methods:

• Utilized the Effective Fragment Potential (EFP) method, a polarizable quantum mechanics-based force field.
• Calculated protein-ligand interactions in seven inactive cyclin-dependent kinase 2-ligand complexes.
• Employed molecular dynamics simulations and clustering analysis to obtain representative structures, considering and excluding solvent effects.

Main Results:

• High correlations (R2 up to 0.95) were observed between experimental binding affinities and EFP interaction energies.
• Excluding water molecules and using representative structures from clustering analysis yielded the highest correlation.
• EFP pairwise interaction energy decomposition successfully identified critical ligand-residue interactions and their nature.

Conclusions:

• The Effective Fragment Potential (EFP) method demonstrates high accuracy in predicting protein-ligand binding affinities.
• Dynamic and solvent effects can be effectively managed using representative structures and appropriate exclusion criteria.
• EFP shows significant potential for application in structure-based drug design, particularly for kinase inhibitors.