Molecules that are not obtainable by natural enzymatic processes or synthetic chemistry require novel catalysts for their synthesis. The cytochromes P450 are one of the enzyme systems with the greatest potential for the development of novel, industrially useful biocatalysts. However, the existing P450s studied to date have some limitations under industrial conditions, such as low activity, insufficient thermostability, and inappropriate regioselectivity, motivating the discovery of new P450 enzymes. In recent years, with the development of next-generation sequencing platforms and analysis methods, metagenomics, which is based on the acquisition of total environmental DNA independent of culturing, has been widely used for the discovery of novel biomolecules found in nature.

Here, we aimed to discover novel bacterial P450 enzymes from extreme environments by a metagenomic approach. Firstly, soil samples were collected from five locations exhibiting thermophilic, halophilic, acidophilic, and psychrophilic characteristics. As a result of these “omics” studies, 624 novel microbial P450 enzyme-encoding genes have been identified. Notably, 64% of the P450 families identified in this study have no previously characterised members with known functions. In addition, the sequences of two novel ferredoxin-NADP⁺ reductases, namely Fpr1 and Fpr2, from previously characterized Pseudomonas mandelii KGI_MA19 were obtained and the conditions affecting recombinant protein production were optimized. Ten of the identified P450s, and also their redox partners, have been successfully expressed in Escherichia coli as recombinant enzymes and purified by immobilized metal affinity chromatography. The typical absorbance maximum at 450 nm was observed as the CO-bound reduced form for all expressed P450s. Substrate screening of CYP109J8 from thermophilic and CYP109G35 from halophilic samples confirmed vitamin D2 binding. Thermal assessments revealed Tms of 70.23°C for CYP109J8 and 77.48°C for CYP109G35, confirming the thermostable nature of CYP109J8 and the halo-thermostable characteristics of CYP109G35 in the presence of NaCl. This study shows the potential for metagenomic analysis to reveal novel P450 diversity.

Resin acids are diterpene extractive molecules found in different types of tree bark that act as a first line of defense from external attacks, such as pathogens. These acids are found mainly in two types: abietic and pimaric type, which are insoluble, recalcitrant and toxic molecules that are known to interact with and disturb biological membranes¹. Degradation of bark is a natural process, but virtually nothing was known about it on a microbial and molecular level until recently. By monitoring spruce bark degradation by a microbial consortium for six months, we discovered that resin acids seem to act as gatekeeper molecules. Specialized bacterial species needed to detoxify/metabolize the resin acids before the consortium could diversify². The main bacterium responsible for resin acid metabolism was isolated, a new species named Pseudomonas abieticivorans (eater of abietic acid), and it can grow on abietic and pimaric acids as sole carbon sources. This bacterium encodes a complete so-called dit (diterpene) gene cluster (≈33 genes), linked to abietic acid degradation³. Two of the enzymes encoded in this cluster are cytochrome P450 monooxygenases from the understudied CYP226A family: DitQ and DitU. DitQ is speculated to convert dehydroabietic acid into 7-oxo-dehydroabietic acid, based on studies of homologous enzymes. The role of DitU is more unclear, although it is suspected to bind and convert abietic and/or palustric acid into 7-hydroxy-palustric acid⁴.

Two ferredoxins (ditA3 and ditX) and four reductases (ditB, ditG, ditI and ditY) are also found within the dit cluster. While ditA3B and ditXY are likely co-transcribed, DitA3 is the presumed partner of the dioxygenase system DitA1A2. We have cloned and heterologously expressed all these P450s, ferredoxins and ferredoxin-reductases in E. coli. We have performed in-vitro assays, e.g. spin state shift and NADH oxidation, and tested all ferredoxin/ferredoxin-reductase combinations with our P450s. Interestingly, the expected spectral shift of DitQ from 420 to 390 nm was not observed upon substrate binding; instead, a shift to 416 nm was detected with both dehydroabietic and abietic acids, consistent with observations from colleagues in Adelaide (unpublished). Current work focuses on product analysis and inclusion of additional dit cluster proteins to the reaction setups.

Elizabeth M. J. Gillam

The University of Queensland, St. Lucia, Brisbane, 4072, Australia

The expansion of genome sequencing over the last two decades has led to a massive increase in the availability of extant protein sequences for cytochrome P450 enzymes and other protein families. Ancestral sequence reconstruction (ASR) is a computational approach that leverages this rich resource to infer how sequences have changed through the evolutionary history of protein families. Ancestral proteins at key evolutionary stages can then be resurrected, i.e. synthesized and characterized by biochemical and biophysical approaches to explore how proteins have changed structurally and functionally. To date we have applied ASR to several xenobiotic-metabolizing P450 families as well as families involved in secondary metabolism in animals and plants. This presentation will compare and contrast the results obtained with different families with respect to thermostability, solvent tolerance, substrate range and resistance to oxidative stress, and show how the characteristics of the ancestral proteins can be exploited for applications in biocatalysis.