Abstract
Cytochrome P450 enzymes play a vital role in the safe elimination of foreign compounds, such as drugs and toxins, from the human body. They are responsible for increasing the hydrophilicity of a compound to allow for effective excretion through the kidneys, a process known as Phase I metabolism. Given the importance of this metabolism for the safe excretion of a foreign compound, cytochrome P450 enzymes have long been of interest to the pharmaceutical drug industry. Currently, animal testing provides one of the most effective ways of determining the safe metabolism of a drug compound, however there are still problems. The cytochrome P450 2D family has only one form in humans, CYP2D6, responsible for the metabolism of ~20% of drugs in current clinical use, but six forms in rats, CYPs 2D1-2D5 and CYP2D18. This makes it difficult to determine if a drug will be safely eliminated in humans based on rat studies alone, as shown by the isomers quinine and quinidine. Despite having similar chemical structures, these isomers show a variation in potency. By identifying the underlying cause of this species selectivity it may be possible to earlier identify which compounds show the most promise in humans saving the pharmaceutical industry both time and money. Using computational techniques, models of the rat cytochrome P450 2D enzymes have been produced based on the recent x-ray crystal structures of the human cytochrome P450 2D6 enzyme both with and without a ligand bound. The differences in species selectivity for isomers, such as quinidine and quinine, are rationalised using these models and the results are discussed with regard to previous studies. Models were generated under the CHARMM27 force field to more accurately model the haem iron. Molecular docking studies are performed for quinidine and quinine into both the haem and compound 1 forms of the rat and human CYP2D family. In this case, no one amino acid residue is identified as responsible for the variation in inhibition observed for these compounds but rather a change in the number of binding interactions occuring. This is consistent with molecular recognition in cytochrome P450 enzymes being the result of a number of non-specific interactions in a binding site. A preliminary investigation into the effect that docking topology has on ligand function in rat and human CYP2D enzymes suggests a relationship between the orientation angle of the docked compound relative to the haem iron. In the majority of cases inhibitors are found to dock parallel to the haem group and substrates normal to the haem. In a virtual screen of seven substrates and eight inhibitors into both the haem and compound 1 forms, an overall agreement with this hypothesis of 58% and 66% was found for models based on the crystal structure of human CYP2D6 with a ligand bound and without respectively. Substrates were correctly predicted 38% of the time for models based on the crystal structure with ligand bound and 42% of the time for models based on the crystal structure without. Inhibitors were predicted correctly in 78% of cases for models based on the ligand bound structure and for 85% of cases based on the structure without. In a virtual screen of a validation set of forty substrates and inhibitors, ligands were correctly predicted 58% and 55% of the time for models based on the structure with and without a ligand bound respectively. This suggests that for cytochrome P450 enzymes, potent substrates and inhibitors can be identified at an early stage of drug development through placement analysis within the active site regardless of species or stereochemistry. The more potent isomer from quinidine and quinine is defined as the compound with an orientation angle closest to an ideal inhibitor angle of 90. Quinidine is predicted to be the more potent isomer for rat CYP2D5 for which no experimental information currently exists, while quinine is predicted to be more potent in rat CYP2D18. Using sequence analysis, three amino acid residues (residues 209, 210 and 388) are identified which are predicted to impact the species selectivity of isomers as observed for quinidine and quinine in rat and human CYP2D enzymes. These residues are located both within and at some distance from the active site and are located near substrate recognition sites and access channels, suggesting that their mutation will have an affect on substrate and inhibitor access and product egress from the active site.