Characterization of diverse natural variants of CYP102A1 found within a species of Bacillus megaterium

Abstract

An extreme diversity of substrates and catalytic reactions of cytochrome P450 (P450) enzymes is considered to be the consequence of evolutionary adaptation driven by different metabolic or environmental demands. Here we report the presence of numerous natural variants of P450 BM3 (CYP102A1) within a species of Bacillus megaterium. Extensive amino acid substitutions (up to 5% of the total 1049 amino acid residues) were identified from the variants. Phylogenetic analyses suggest that this P450 gene evolve more rapidly than the rRNA gene locus. It was found that key catalytic residues in the substrate channel and active site are retained. Although there were no apparent variations in hydroxylation activity towards myristic acid (C14) and palmitic acid (C16), the hydroxylation rates of lauric acid (C12) by the variants varied in the range of >25-fold. Interestingly, catalytic activities of the variants are promiscuous towards non-natural substrates including human P450 substrates. It can be suggested that CYP102A1 variants can acquire new catalytic activities through site-specific mutations distal to the active site.

Figure 1

Summarized phylogeny of CYP102A1 natural variants and intergenic sequence (ITS) alleles from B. megaterium strains. (a) Phylogenetic analyses of CYP102A1 variants are based on the amino acid substitutions (Table 2 and Fig. S1) and silent mutations are excluded. Relative abundances are shown in parentheses. (b) Phylogenetic analyses of B. megaterium strains, which express CYP102A1, were based on the ITS gene sequences. The CYP102A1 variant expressed by each strain is shown as a number with an asterisk in parentheses. Numbers on tree branches show the percent bootstrap support for all branches important for interpretation. Nodes with bootstrap values of 1,000 resamplings (expressed by percentages) are indicated and the bar scales represent the substitution of amino acids (a) or nucleotides (b) per site.

Figure 2

Comparison of distinct regions of 16S rRNA gene sequences and ITS from B. megaterium. Two and seven nucleotides were variable among 1,394 and 338 nucleotides, respectively, of 16S rRNA (a) and ITS (b) genes of B. megaterium strains.

Figure 3

Biochemical properties of natural variants. (a) Dissociation constants (K_d values) of substrates (lauric acid, myristic acid, and palmitic acid) to CYP102A1 natural variants. (b) Turnover numbers of the hydroxylation of fatty acids (lauric acid, myristic acid, palmitic acid) by the variants of CYP102A1. (c) Rates of fatty acid-dependent NADPH oxidation by the variants of CYP102A1.

Figure 4

Thermal stability for each domain of CYP102A1 variants. Enzymes (2 μM) were incubated at different temperatures between 25 and 70°C for 20 min with subsequent cooling to 4°C in a PCR thermocycler. The stability of the heme domain was calculated from heat-inactivation curves of CO-binding difference spectra. The stability of the reductase domain was calculated from the reduction of ferricyanide catalyzed by reductase activity.

Figure 5

Catalytic promiscuity of natural variants of CYP102A1 towards human P450 substrates. Purified natural variants of CYP102A1 were characterized for human P450 enzyme activities using specific substrates: phenacetin O-deethylation for P450 1A2; 7-ethoxycoumarin (7-EC) O-deethylation for P450s 1A2, 2A6, and 2E1; 7-ethoxy-4-trifluoromethylcoumarin (7-EFC) O-deethylation for P450s 1A2 and 2B6; chlorzoxazone 6β-hydroxylation for P450 2E1; coumarin 7-hyroxylation for P450 2A6. Data are shown as the means ± SEM.