The long-term aim of the research described in this proposal is to understand the structural basis underlying the unique substrate specificities of the human CYP2C proteins. This is important for the avoidance of inhibitory drug-drug interactions involving this important P450 sub-family, and in particular CYP2C9, during the development of new therapeutic agents. Our preliminary data derived from CoMFA modeling, homology modeling and site-directed mutagenesis studies, suggest that the binding of coumarin anticoagulants and anticonvulsant hydantoins to CYP2C9 is governed by an aromatic binding interaction within the B'-helix, and two electrostatic interactions (E1 and E2) within elements of the F, G and I-helices of the enzyme. The primary goal of this research is to test this hypothesis in order to refine our active-site model for CYP2C9. CYP2C19, which share greater than 90% sequence homology with CYP2C9, also metabolizes the coumarins and hydantoins but generates metabolite profiles quite distinct from CYP2C9. Therefore, the active-sites of CYP2C9 and CYP2C19 likely share some common binding determinants which should facilitate the related goal of developing a three-dimensional active-site model for CYP2C19.
The Specific Aims of this proposal are; 1. Construction of new CoMFA models for CYP2C9 and CYP2C19 based on Ki values for the inhibition of both isoforms by Type I and Type II ligands. 2. Identification of active-sites residues in CYP2C9 and CYP2C19 through photo-affinity labeling and analysis of adducted residues by electrospray and MALDI mass spectrometry. 3. Identification of electrophilic binding site determinants in CYP2C9 through the construction of hybrid CYP2C9/CYP2C19 proteins and point mutants, and analysis of their interaction with valproic acid, phenytoin and phenprocoumon. This three-tiered approach involves the use of complementary techniques which will permit the iterative refinement of three-dimensional structural representations of CYP2C9 and CYP2C19. This approach provides a powerful internal control for the procedures which we are using, by ensuring that discrete CYP2C9 and CYP2C19 structural representations are created rather than a low resolution """"""""global"""""""" CYP2C model. Finally, if we can develop discrete models for the closely related CYP2C9 and CYP2C19 isoforms, there should be little impediment to the future generation of active-site models for the other human CYP2C isoforms.
Showing the most recent 10 out of 361 publications
