Modeling loop backbone flexibility in receptor-ligand docking simulations.

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Receptor conformational changes upon ligand binding are vital for drug action but often simplified in computational docking.
• Sampling backbone flexibility in receptor models presents a significant computational challenge due to vast conformational space.

Purpose of the Study:

• To develop and validate an efficient algorithm for reconstructing backbone flexibility in large receptor loop regions during ligand binding.
• To assess the impact of backbone flexibility on the accuracy of binding mode prediction in docking simulations.

Main Methods:

• An efficient multistage backbone reconstruction algorithm was developed for flexible loop regions.
• The algorithm was tested on three kinase receptors, including simulations starting from apo structures and cross-docking scenarios.
• The DFG-motif in p38 kinase was used as a specific example to model allosteric binding.

Main Results:

• Treatment of backbone receptor flexibility significantly improved binding mode prediction accuracy.
• Accurate models of ligand-receptor complexes were achieved starting from apo structures for tested kinases.
• The algorithm successfully modeled allosteric binding at the DFG-motif of p38 kinase.

Conclusions:

• The developed algorithm effectively addresses backbone flexibility in receptor-ligand docking simulations.
• Incorporating backbone flexibility enhances the accuracy of predicting binding modes and identifying allosteric sites.
• This approach facilitates the design of highly specific ligands by leveraging allosteric mechanisms.