Quasi-atomistic receptor modeling. A bridge between 3D QSAR and receptor fitting

Area of Science:

• Computational chemistry and molecular modeling.
• Drug discovery and development.
• Structure-based drug design.

Background:

• Traditional 3D quantitative structure-activity relationship (3D QSAR) and receptor modeling have limitations in capturing receptor flexibility and specific interactions.
• Accurate modeling of receptor-ligand interactions is crucial for predicting binding affinity and guiding drug design.

Purpose of the Study:

• To introduce and validate a novel quasi-atomistic receptor modeling approach.
• To enhance the prediction of ligand binding free energies by simulating flexible receptor cavities and specific interactions.
• To develop a computational tool (Quasar) for generating receptor models and deriving structure-activity relationships.

Main Methods:

• Developed quasi-atomistic receptor modeling combining surface properties with flexible receptor envelopes and induced fit.
• Incorporated H-bond flip-flop particles to mimic amino acid residue versatility in hydrogen bonding.
• Utilized a directional force field for hydrogen bonds to simulate ligand selectivity, including stereoisomers.
• Employed genetic algorithms and cross-validation protocols within the Quasar software to generate receptor models.

Main Results:

• Successfully applied the concept to six diverse receptor systems (beta 2-adrenergic, dopaminergic, aryl hydrocarbon, cannabinoid, neurokinin-1, and HIV protease).
• Achieved high accuracy in predicting relative free energies of ligand binding for independent test ligands, with errors between 0.55 to 0.94 kcal/mol.
• Demonstrated a prediction uncertainty in binding affinity of a factor of 2.5 to 5.0.

Conclusions:

• Quasi-atomistic receptor modeling provides a robust framework for bridging 3D QSAR and receptor modeling.
• The Quasar software and methodology enable accurate prediction of ligand binding affinities, facilitating rational drug design.
• This approach holds significant potential for accelerating the discovery of new drugs by improving the prediction of structure-activity relationships.