Computational approach to characterization of human liver drug-metabolizing enzymes.

Cytochromes P450 are the key enzymes for activating and inactivating many drugs; individual expression levels of CYPs may play a crucial role in drug safety and drug efficacy. Statistical comparison of biochemical profiles of 23 human liver microsomes have been used to characterize human liver samples. The profile included 12 parameters, namely activity of NADPH-cytochrome P450 reductase, cytochrome P450 content and cytochrome P450-dependent monooxygenase activities with marker substrates. Unsupervised statistical methods including cluster analysis and principal component analysis revealed with very high confidence the presence of two groups. Difference between the groups was explained by peculiarities of reductase activity and cytochrome P450 enzyme activities with 7-ethoxyresorufin, 7-methoxyresorufin, 7-methoxycoumarin, 7-benzyloxyresorufin and 7-benzyloxyquinoline. Results of biochemical assays coupled with multidimensional data analysis can be further used for targeted proteomic profiling of microsome oxidation mechanisms.