Computational modelling of inhibitor binding to human thrombin

Area of Science:

• Biochemistry
• Computational Chemistry
• Pharmacology

Background:

• Thrombin is a key protein in blood clot formation.
• Dysregulation of coagulation leads to cardiovascular diseases like thrombosis.
• Thrombin inhibitors are crucial therapeutic targets.

Purpose of the Study:

• To evaluate a molecular dynamics (MD) based method for predicting human thrombin inhibitor binding affinities.
• To assess the method's ability to rank relative ligand affinities and reproduce binding free energy differences.
• To explore the utility of the method for understanding structure-activity relationships.

Main Methods:

• Utilized a linear interaction energy (LIE) method based on molecular dynamics simulations.
• Applied the method to a series of eight different human thrombin inhibitors.
• Involved a specific parameterization requiring an added constant energy term for thrombin.

Main Results:

• The MD-based method reasonably ranked the relative binding affinities of the tested ligands.
• Quantitative reproduction of binding free energy differences between high and low affinity inhibitors was achieved.
• Stereospecificity for a chiral inhibitor was successfully predicted.
• A mean unsigned error of 0.68 kcal/mol was obtained for absolute binding free energies.
• The method provided insights into ligand-enzyme interactions at a microscopic level.

Conclusions:

• The evaluated MD-based method is a reliable tool for predicting thrombin inhibitor binding affinities.
• The approach aids in understanding three-dimensional structure-activity relationships.
• This computational strategy can accelerate the development of novel antithrombotic agents.