The ubiquitous superfamily of enzymes, the cytochrome P450s, continue to be the focus of many diverse research efforts. Interest in these heme containing monooxygenases stem from their ability to catalyze the oxidation of a wide variety of lipophilic endogenous (i.e., steroids, prostaglandins and fatty acids) and exogenous (drugs and environmental contaminants) compounds. Because of their importance in drug metabolism and toxicity, predictive models for cytochrome P450 oxidations could be extremely useful. We have previously used molecular modeling techniques to develop a predictive model for cytochrome P450 hydrogen abstraction reactions. Using the p-nitrosophenoxy radical as a model for the P450 active oxygenating species, a linear correlation was observed and a combination of the modified Swain-Lupton resonance parameter or the ionization potential of the radical formed. The latter relationship gave an estimated standard deviation of the predicted 0.8 kcal/mol, suggesting that it may be possible to obtain an estimate of the relative ability for any carbon-hydrogen bond to undergo P450 mediated hydrogen atom abstraction by calculating the relative stability and ionization potential of the resulting radical. We are now in the process of testing this model as well as expanding it to include aromatic and olefinic oxidations. In addition, reactants and products for 54 hydroxylation and desaturation reactions were modeled and used to predict the relative tendency for each reaction to occur. Finally, a model for the aromatase catalyzed formation of estrogen has been developed.
