Ligand Docking to Intermediate and Close-To-Bound Conformers Generated by an Elastic Network Model Based Algorithm...

Area of Science:

• Computational Biology
• Structural Biology
• Drug Discovery

Background:

• Accurately modeling protein flexibility is essential for small ligand-protein docking, especially for proteins with large conformational changes.
• Current methods struggle with proteins that undergo significant structural transitions, hindering drug design efforts.
• Sampling accessible protein conformers from the unbound (apo) state can aid docking and drug design.

Purpose of the Study:

• To develop an unbiased algorithm for sampling protein conformational states from the apo structure.
• To assess the algorithm's effectiveness in improving ligand docking accuracy for proteins with large conformational changes.
• To investigate the role of intermediate protein states in ligand binding.

Main Methods:

• An iterative algorithm integrating elastic network models, clustering, and energy minimization with implicit solvation was developed.
• The algorithm was applied to apo structures of five diverse proteins with large conformational changes (4.7-15 Å RMSD).
• Docking was performed on generated conformers, including apo, bound-like, and intermediate states, to evaluate binding pose accuracy.

Main Results:

• The algorithm successfully generated conformers close to the bound state (1.4-3.8 Å RMSD) and relevant intermediate states.
• For proteins with hinge-type closures, the best docking poses were achieved with conformers closest to the bound structure.
• Optimal ligand docking for adenylate kinase and calmodulin was achieved using intermediate or specifically generated conformational states, with final ligand RMSDs of 1.5-2.9 Å.

Conclusions:

• The developed algorithm effectively generates relevant protein conformational states starting from the apo structure.
• Intermediate protein conformational states are critical for accurate ligand docking, particularly for proteins undergoing large transitions.
• This approach offers a valuable tool for drug design and understanding ligand-protein interactions in flexible systems.