CYP 2D6 binding affinity predictions using multiple ligand and protein conformations.

Area of Science:

• Pharmacology and Toxicology
• Computational Chemistry
• Biophysics

Background:

• Cytochrome P450 enzymes (CYPs) exhibit significant flexibility, complicating in silico prediction of xenobiotic binding affinities.
• Accurate prediction of ligand-CYP interactions is crucial for drug development and understanding xenobiotic metabolism.

Purpose of the Study:

• To develop and validate a computationally efficient method for predicting binding affinities of ligands to flexible Cytochrome P450 enzymes.
• To improve the accuracy of in silico binding affinity predictions by enhancing conformational sampling.

Main Methods:

• Employed an iterative linear interaction energy (LIE) approach combined with molecular dynamics (MD) simulations.
• Enhanced conformational sampling by initiating MD simulations from diverse protein-ligand geometries and orientations.
• Calculated binding free energies for thiourea compounds binding to the CYP 2D6 isoform.

Main Results:

• Achieved improved correlation between calculated and experimental binding free energies for thiourea compounds and CYP 2D6.
• Demonstrated that combining MD simulation results from distinct starting configurations enhances accuracy.
• The method proved effective even with relatively short MD simulation times.

Conclusions:

• The iterative LIE approach with enhanced sampling is a computationally attractive and accurate method for predicting ligand-binding affinities to flexible CYPs.
• This approach holds promise for automated high-throughput screening in drug discovery and toxicology.
• The study highlights the importance of adequate conformational sampling for flexible protein targets.