Another enzyme class that has emerged as a
Another enzyme class that has emerged as a potential alternative to P450s are the so-called ‘unspecific peroxygenases’ (UPOs). They also contain a heme but are extracellularly expressed fungal enzymes that are often more stable than P450 monooxygenases. Importantly, UPOs use H2O2 instead of O2 as the oxidant to drive numerous oxygenation reactions on a broad range of substrates without the need for an electron delivery system [35,36] (Figure 1a). These beneficial attributes have enabled industrial use such as for a UPO from Novozymes that hydroxylates alkanes and fatty acids with kcat values up to 1000 s−1 . A technical hurdle in engineering UPOs is their low functional expression in yeast, which was overcome by Molina-Espeja et al. through evolution of a secreting peptide leader sequence . This opened up the possibility for further improvements of the enzyme’s solvent and thermo stability as well as improved resistance against excess H2O2 [38,39]. More recently, directed evolution was used to improve the ratio of desired peroxygenative to peroxidative activity by using a dual colorimetric assay to discriminate between the targeted 1-naphthol formation and the undesired one-electron oxidation process, which generates reactive radical compounds . The evolved enzyme was further optimized for the regioselective hydroxylation of the β-blocker drug propranolol , demonstrating a high adaptability to different substrates.
α-Ketoglutarate-dependent dioxygenases Besides the prominent hemoproteins for C–H oxyfunctionalization, non-heme Fe(II)-dependent oxygenases, in particular α-ketoglutarate-dependent dioxygenases (α-KGDs), have received increased attention for performing desirable C–H activation chemistry [42,43]. This class of enzymes shares a jellyroll structural fold with a conserved His-X-Asp/Glu-Xn-His motif for chelating Fe(II). Similar to P450s, a high-valent Fe(IV)-oxo species abstracts a hydrogen zilpaterol from unactivated C(sp3)–H bonds but the electrons are provided by a cheap co-factor, α-ketoglutarate, through oxidative decarboxylation  (Figure 1b). So far, biocatalytic applications have mainly focused on α-KG-dependent amino acid hydroxylases for the microbial production of hydroxylated proline or leucine, important chiral building blocks for the chemical industry . To broaden the relatively narrow substrate scope of α-KGDs via enzyme engineering, Codexis, for example, used an HPLC-based high-throughput assay for the screening and evolution of proline hydroxylase variants from the bacterium Sinorhizobium meliloti (SmPH). The improved biocatalysts exhibited higher activity, regioselectivity and stability, and were also found to be active on different proline and pipecolic acid derivatives . In a different study, a homology model of SmPH was used to guide mutagenesis of residues near the active site and the regioselectivity in the hydroxylation of pipecolic acid could be improved from 60 to 95% in three rounds of evolution . Engineering of α-KGD deacetoxycephalosporin C synthase (DAOCS) via a combination of error-prone PCR (epPCR) and subsequent DNA shuffling generated a quadruple mutant with a 40-fold improved kcat/KM in the conversion of the non-native substrate penicillin G . A generally applicable strategy to screen for improved α-KGD activity could be based on the generation of succinate by the decarboxylation of α-KG . In this context, Escherichia coli strains were engineered, in which succinate-producing enzymes from the TCA cycle were deleted. Growth of these organisms depended on active α-KGD as the sole producer of succinate, which in principle enables selection of particularly active enzymes [50,51]. This concept was exploited to attempt directed evolution of an isoleucine hydroxylase but only marginal (less than two fold) improvements were obtained . Given the vast number of putative α-KGDs in the metagenome and the current research efforts in the characterization of natural product biosynthetic pathways, novel members of this versatile enzyme class with potentially larger substrate scope may be uncovered. Such promiscuous alpha-KGDs would be ideal candidates for further engineering by directed evolution [,54].