Ranking ligand binding affinities with avidin: a molecular dynamics-based interaction energy study

Area of Science:

• Computational chemistry
• Biophysics
• Molecular modeling

Background:

• Accurate calculation of binding free energy is crucial for drug discovery and understanding molecular interactions.
• Avidin-biotin interactions serve as a model system for studying ligand-protein binding.
• Molecular dynamics (MD) simulations offer a powerful tool for investigating these interactions at an atomic level.

Purpose of the Study:

• To evaluate the accuracy of two computational methods for calculating binding free energy of biotin analogues to avidin.
• To assess the impact of protein flexibility and ligand properties on binding free energy calculations.
• To understand the origins of discrepancies between calculated and experimental binding strengths.

Main Methods:

• Utilized molecular dynamics (MD) simulations to generate trajectories for 14 biotin analogues binding to avidin.
• Employed the linear interaction energy (LIE) approximation and interaction free energy methods for binding energy calculations.
• Applied thermodynamic integration (TI) to investigate overestimations in binding free energy.

Main Results:

• Both LIE and interaction free energy methods showed good correlation with experimental binding free energy for 10 out of 14 biotin analogues.
• Both methods overestimated binding strength by over 7 kcal/mol for the remaining four ligands.
• Protein flexibility and ligand charge were identified as significant factors affecting calculation accuracy.

Conclusions:

• MD simulations, particularly LIE and interaction free energy methods, are viable for estimating binding free energies of biotin analogues to avidin.
• Protein flexibility and ligand charge need careful consideration for accurate binding free energy predictions.
• MD simulations provide valuable insights into active site interactions and variations in binding affinity.