Combining In Silico Protein Stability Calculations with Structure-Function Relationships to Explore the Effect of Polymorphic Variation on Cytochrome P450 Drug Metabolism

Abstract

We carried out an in silico structural analysis of 348 non-synonymous single nucleotide polymorphisms, found across nine of the major human drug metabolising cytochrome P450 isoforms, to determine the effects of mutations on enzyme structure and function. Previous functional studies in our group have delineated regions of the cytochrome P450 structure important for substrate recognition, substrate and product access and egress from the active site and interaction with the cytochrome P450 reductase. Here we combine the information from those studies with new in silico calculations on the effect of mutations on protein stability and we compare our results to experimental data in order to establish the likely causes of altered drug metabolism observed for cytochrome P450 variants in functional assays to date, in the process creating a cytochrome P450 polymorphic variant map. Using the computational tool Site Directed Mutator we predicted destabilising mutations that result in altered enzyme function in vitro with a specificity of 83%. We found that 75% of all cytochrome P450 mutations that show altered activity in vitro are either predicted to be destabilising to protein structure or are found within regions predicted to be important for catalytic activity. Furthermore, we found that 70% of the mutations that showed similar activity to the wild-type enzyme in in vitro studies lie outside of functional regions important for catalytic activity and are predicted to have no effect on protein stability. Our resultant cytochrome P450 polymorphic variant map should therefore find utility in predicting the likely functional effect of uncharacterised variants on drug metabolism.