A multistep approach to structure-based drug design: studying ligand binding at the human neutrophil elastase.

In this study we show that a combination of different theoretical methods is a viable approach to calculate the binding affinities of new ligands for the human neutrophile elastase. This protease degrades elastin and likely aids neutrophils in fulfilling their immunological functions. Abnormally high human neutrophil elastase (HNE) levels are involved in several diseases; therefore, inhibitors of HNE are of interest as targets for drug design. A recent study has revealed that cinnamic acid and bornyl ester derivatives bind to HNE, but DeltaG0 values from ligand docking results exhibited no correlation with those calculated from the IC50 values. To accurately compute binding affinities, we generated possible protein ligand complex structures by ligand docking calculations. For each of the ligands, the 30 most likely placements were used as starting points of nanosecond length molecular dynamics simulations. The binding free energies for these complex structures were estimated using a continuum solvent (MM-PBSA) approach. These results, along with structural data from the molecular dynamics runs, allowed the identification of a group of similar placements that serve as a model for the natural protein ligand complex structure. This structural model was used to perform thermodynamic integration (TI) calculations to obtain the relative binding free energies of similar ligands to HNE. The TI results were in quantitative agreement with the measured binding affinities. Thus, the presented approach can be used to generate a probable complex structure for known ligands to HNE and to use such a structure to calculate the effects of small ligand modifications on ligand binding, possibly leading to new inhibitors with improved binding affinities.