单加氧酶
化学
催化作用
细胞色素P450
氧气
细胞色素
加氧酶
组合化学
立体化学
生物化学
有机化学
酶
作者
Matthew N. Podgorski,Jinia Akter,Luke R. Churchman,J.B. Bruning,James J. De Voss,Stephen G. Bell
标识
DOI:10.1021/acscatal.4c01326
摘要
Cytochrome P450 enzymes (CYPs) are biocatalysts for the generation of fine chemicals including natural products, drug metabolites, and flavor and fragrance compounds. However, both the high cost of the required nicotinamide cofactors and their need for additional electron transfer proteins limit their use. Here, we investigate whether CYPs can be converted into more efficient peroxygenases through protein engineering of the enzyme's oxygen activation machinery. We improve the peroxygenase activity by modifying selected residues within the I-helix to more closely resemble those of a natural peroxygenase. We produced mutants containing two, four, and six mutations, within this region of the I-helix. In our model CYP system, the double mutant in which glutamine and glutamate residues replaced aspartate and threonine, respectively, was found to have significantly higher peroxygenase activity for the O-demethylation of 4-methoxybenzoic acid than a single glutamate mutant prototype. Importantly, it functioned better at lower H2O2 concentrations and could convert all the added substrate to product. All the mutants maintained the stereoselectivity of the CYP enzyme for the epoxidation of 4-vinylbenzoic acid. The X-ray crystal structures of these enzymes showed significant structural changes at the oxygen-binding groove in the I-helix. In crystallo reactions with 4-methylbenzoic acid exhibit electron density corresponding to the 4-(hydroxymethyl)benzoic acid metabolite. We extended this mutagenesis strategy to a bacterial steroid-hydroxylating CYP and an uncharacterized CYP from a thermophilic bacterium. In these instances, we generate peroxygenases, which catalyze the regio- and stereoselective hydroxylation of progesterone and the hydroxylation of fatty acids at low hydrogen peroxide concentrations.
科研通智能强力驱动
Strongly Powered by AbleSci AI