Computational and experimental studies of protein kinase-inhibitor interactions

Protein kinase catalyzes the reaction that transfers phosphate groups from nucleoside triphosphates, usually adenosine triphosphate (ATP), to specific serine, threonine, or tyrosine residues in substrate proteins as a way of regulating their activities. They play fundamental roles in mediating cellular processes in eukaryotic cells: metabolism, transcription, cell cycle progression, cell motility, apoptosis and differentiation. Deregulation of kinase activities lead to a variety of human diseases including cardiovascular diseases, inflammatory diseases, neurodegenerative diseases and cancer. Therefore, they are attractive target for drug design and therapeutic intervention. Two challenges exist for drug design targeting protein kinases. First, most kinase drugs target the highly conserved ATP-binding pocket in the protein kinase, which raises potential issues of off-target interactions and the resultant side effects. Second, drug resistance due to mutations in protein kinase domain has become a serious problem in modern drug discovery because they render the existing drug ineffective. To overcome the two challenges and design drugs with high specificity and high tolerance to resistance, a fast and effective screening method to quickly determine the specificity of the drug or the effects of potential resistance-causing mutations will be highly beneficial. To this end, we explored the possibility of using molecular dynamics simulations and free energy calculations on the MAP kinase p38α to aid the drug design efforts in three different scenarios. First, we demonstrated the ability of free energy simulation methods to predict the experimentally measured thermodynamics effects of p38α mutations on the binding affinity of a small molecule inhibitor. Second, we demonstrated the ability of free energy simulation methods to predict the binding affinity of a small molecule inhibitor against a panel of p38 isozymes. Third, we demonstrated the ability of longtime (μs) molecular dynamics simulations to provide extensive sampling of the kinase configurational space which can be utilized to identify novel drug binding sites. In summary, we employed a unified approach to examine the structural and energetic properties of protein kinase-inhibitor interactions using molecular dynamics simulations and free energy calculations. We found that these methods, if treated properly have great potential in the aid of designing drugs with high specificity and high tolerance to resistance.