Evidence for communality in the primary determinants of CYP74 catalysis and of structural similarities between CYP74 and classical mammalian P450 enzymes.

In silico structural analysis of CYP74C3, a membrane-associated P450 enzyme from the plant Medicago truncatula (barrel medic) with hydroperoxide lyase (HPL) specificity, showed that it had strong similarities to the structural folds of the classical microsomal P450 enzyme from rabbits (CYP2C5). It was not only the secondary structure predictions that supported the analysis but site directed mutagenesis of the substrate interacting residues was also consistent with it. This led us to develop a substrate-binding model of CYP74C3 which predicted three amino acid residues, N285, F287, and G288 located in the putative I-helix and distal haem pocket of CYP74C3 to be in close proximity to the preferred substrate 13-HPOTE. These residues were judged to be in equivalent positions to those identified in SRS-4 of CYP2C5. Significance of the residues and their relevance to the model were further assessed by site directed mutagenesis of the three residues followed by EPR spectroscopic and detailed kinetic investigations of the mutated proteins in the presence and absence of detergent. Although point mutation of the residues had no effect on the haem content of the mutated proteins, significant effects on the spin state equilibrium of the haem iron were noted. Kinetic effects of the mutations, which were investigated using three different substrates, were dramatic in nature. In the presence of detergent with the preferred substrate (13-HPOTE), the catalytic center activities and substrate binding affinities of the mutant proteins were reduced by a factor of 8-32 and 4-12, respectively, compared with wild-type--a two orders of magnitude reduction in catalytic efficiencies. We believe this is the first report where primary determinants of catalysis for any CYP74 enzyme, which are fully consistent with our model, have been identified. Our working model predicts that N285 is close enough to suggest that a hydrogen bond with the peroxy group of the enzyme substrate 13-HPOTE is warranted, whereas significance of F287 may arise from a strong hydrophobic interaction between the alkyl group(s) of the substrate and the phenyl ring of F287. We believe that G288 is crucial because of its size. Any other residue with a relatively bulky side chain will hinder the access of substrate to the active site. The effects of the mutations suggests that subtle protein conformational changes in the putative substrate-binding pocket regulate the formation of a fully active monomer-micelle complex with low-spin haem iron and that structural communication exists between the substrate- and micelle-binding sites of CYP74C3. Conservation in CYP74 sequence alignments suggests that N285, F287, and G288 in CYP74C3 and the equivalent residues at positions in other CYP74 enzymes are likely to be critical to catalysis. To support this we show that G324 in CYP74D4 (Arabidopsis AOS), equivalent to G288 in CYP74C3, is a primary determinant of positional specificity. We suggest that the overall structure of CYP74 enzymes is likely to be very similar to those described for classical P450 monooxygenase enzymes.