Incorporating protein flexibility into structure-based drug design: HIV -1 protease as a test case.

Abstract

Traditionally, structure-based drug design (SBDD) is based on a single X-ray structure of a protein target complexed with a known ligand. One limitation to this approach is that a single protein conformation provides little information on protein dynamics or the conformational changes of both ligand and protein upon binding. To address this limitation, protein flexibility can be incorporated into SBDD using ensembles of protein conformations to more accurately simulate the inherent motion of the system and the potential induced fit between ligand and protein. In 1999, Carlson and coworkers introduced a method using multiple protein structures (MPS), to account for protein flexibility.<super>1,2</super> By mapping individual protein active sites with small molecule probes, favorable interaction regions that are common to many structures in the ensemble can be identified and represented as receptor-based pharmacophore models. To develop this technique into a general tool for SBDD, it has been applied it to the unbound structures of human immunodeficiency virus-1 protease (HIV-1p). Molecular dynamics (MD) simulations were used to generate the MPS. The MD simulations were validated by comparing to NMR experiments to ensure that the conformations sampled were representative of those in solution. Different parameters of the technique have been investigated, and a general protocol for the MPS method has been developed. The influence of using MPS from independent MD simulations based on three different apo crystal structures has been evaluated, and a straightforward function to enable ranking of the potential ligands has been described. The pharmacophore models succeed in discriminating known ligands from drug-like non-inhibitors in a system that is highly dependent on protein flexibility, and demonstrate the utility of the MPS method for SBDD. A parameterization of the polyphosphate moiety is also presented, enabling the simulation of important cofactors such as adenosine triphosphate. <super>l</super>Carlson, H. A.; Masukawa, K. M.; McCammon, J. A., Method for Including the Dynamic Fluctuations of a Protein in Computer-Aided Drug Design. <italic>J Phys. Chem. A</italic> <bold>1999</bold>, <italic> 103</italic>, 10213-10219. <super>2</super>Carlson, H. A.; Masukawa, K. M.; Rubins, K.; Bushman, F. D.; Jorgensen, W. L.; Lins, R. D.; Briggs, J. M.; McCammon, J. A., Developing a Dynamic Pharmacophore Model for HIV-1 Integrase. <italic> J. Med. Chem.</italic> <bold>2000</bold>, <italic>43</italic>, 2100-2114.