Phenylalanine 87 of Bacillus megaterium cytochrome P450 BM3, a residue close to the heme in the substrate binding pocket, has been replaced by alanine by site-directed mutagenesis. The substitution had no effect on the rate of hydroxylation of laurate and increased the affinity for laurate of both the intact enzyme and its heme domain by 2.6-6-fold in the ferric state. NMR paramagnetic relaxation measurements showed that in the initial ferric enzyme-substrate complex, where the substrate binds relatively far from the heme, the substitution had no effect on the position or orientation of the bound substrate. However, in the next intermediate in the catalytic cycle, the reduced enzyme, the position of the bound substrate was altered so that the terminal methyl group was 3.1 A from the iron in the mutant, compared to 5.1 A in the wild-type enzyme. Analysis of the products of the action of the enzyme on laurate and myristate showed that the mutant catalyzed hydroxylation almost exclusively at the omega position, in marked contrast to the wild-type enzyme, with which no hydroxylation at this position was observed.

Oligonucleotide-directed mutagenesis has been used to replace arginine-47 with glutamate in cytochrome P-450 BM3 from Bacillus megaterium and in its haem domain. The mutant has been characterized by sequencing, mass spectrometry, steady-state kinetics and by optical and NMR measurements of substrate binding. The mutant retains significant catalytic activity towards C12-C16 fatty acids, catalysing hydroxylation in the same (ω-1, ω-2, ω-3) positions with kcat/Km values a factor of 14-21 lower. C12-C16 alkyl trimethylammonium compounds are relatively poor substrates for the wild-type enzyme, but are efficiently hydroxylated by the arginine-47 → glutamate mutant at the ω-1, ω-2 and ω-3 positions, with kcat values of up to 19 s-1. Optical spectroscopy shows that the binding of the C14 and C16 alkyl trimethylammonium compounds to the mutant is similar to that of the corresponding fatty acids to the wild-type enzyme. Paramagnetic relaxation measurements show that laurate binds to the ferric state of the mutant in a significantly different position, 1.5 Å closer to the iron, than seen in the wild-type, although this difference is much smaller (~ 0.2 Å) in the ferrous state of the complex. The binding of a substrate having the same charge as residue 47 to the ferric state of the enzyme is roughly ten times weaker than that of a substrate having the opposite charge (and thus is able to make an ion-pair interaction with this residue). The results are discussed in the light of the three-dimensional structure of the enzyme.

No abstract