Advancing GIST-Based Solvent Functionals through Multiobjective Optimization of Solvent Enthalpy and Entropy Scoring...

Area of Science:

• Computational chemistry and molecular modeling
• Biophysics and structural biology
• Drug discovery and medicinal chemistry

Background:

• Understanding water's role in protein-ligand binding thermodynamics is crucial for drug discovery.
• Existing models often simplify the complex enthalpy and entropy contributions of solvation.
• Isothermal titration calorimetry (ITC) provides experimental thermodynamic data for binding events.

Purpose of the Study:

• To optimize solvent scoring terms within grid inhomogeneous solvation theory (GIST) using experimental ITC data.
• To investigate the impact of water molecule enthalpy and entropy on protein-ligand binding affinity.
• To assess the transferability of optimized parameters across different protein targets.

Main Methods:

• Multiobjective optimization was employed to fit GIST-based solvent functionals to ITC data for thrombin-ligand pairs.
• Analysis of different solvent functional implementations and error assessment of the most successful model.
• Visualization of scoring regions in protein binding pockets and unbound ligands.

Main Results:

• Good agreement was achieved between calculated and experimental binding affinity data.
• Ligand desolvation scores positively correlated with binding affinity, suggesting a proxy for protein interactions.
• Limited transferability of parameters was observed between similar protein targets, necessitating target-specific refitting.

Conclusions:

• The study presents a consistent method for assigning thermodynamic quantities to water molecules, aligning with experimental data.
• The approach effectively bridges the gap between experimentally determined water positions and measured thermodynamic signatures.
• Optimized GIST functionals provide valuable insights into the energetic contributions of water in protein-ligand interactions.