Transaminases remain one of the most promising biocatalysts for use in chiral amine synthesis. Here, the identification, cloning and screening of novel transaminases from a drain metagenome is described, with some enzymes exhibiting outstanding properties such as 50% DMSO solvent tolerance.

A metagenomics mining strategy was used to identify new transaminases that were utilised in the synthesis of allylic amines.

Biocatalytic reactions are increasingly being used as a sustainable strategy in organic synthesis and it is recognised that there is need for new enzyme discovery. To establish the utility and versatility of a metagenomics approach, metagenomic DNA extracted from the oral cavity was sequenced and used to create an in silico contig library. This enables individual open reading frames, operons or all the enzymes of a particular family to be identified and then retrieved from the original DNA by PCR. As proof of principle a lactate dehydrogenase, a malate dehydrogenase and transketolases were identified in silico, successfully cloned and assayed. This new enzyme retrieval sequence directed method gives constructive access to metagenomic diversity and importantly improves on the low hit rate experienced when using conventional metagenomic screens.Over the last 30 years enzymes have increasingly been used in commercial chemical processes. [1] Enzyme catalysts have distinct advantages over chemical catalysts such as cost, sustainability, low toxicities and the use of moderate reaction conditions. The greatest advantage however, is the wide range of catalytic activities displayed by naturally occurring enzymes and the stereoselectivities that can be achieved. Drawbacks when using enzymes can include a low organic solvent tolerance, narrow pH working ranges, and low thermostability. However these properties along with enhanced stereoselectivities can be engineered using random or directed evolution techniques. [2] One of the key requirements for this type of engineering is multiple amino acid sequences. Having many enzymes that cat-alyze the same reaction, with variation in their primary sequence is a much more effective starting point for the engineering process than any single example of an enzyme family. One successful method for identifying large numbers of novel enzymes has been to mine fully sequenced and annotated genomes of laboratory cultivable strains. Such sequence directed genome mining lends itself to the retrieval of multiple targets, however a limitation to this method is the number of annotated available species. [3] With only 0.1-5 % of bacteria cultivable in the laboratory, much of the existing diversity of bacteria are inaccessible in this way. [4] At the same time, with the advances in high throughput sequencing, whole or partial genomes of uncultivable bacteria are being made available for study. The nascent field of metagenomics gives insight into the genomes of previously unstudied bacteria and by extension a potential wealth of new biocatalysts for the generation of small molecules and large bioactive molecules such as polyketides and glycopeptides. [5] The use of metagenomics to obtain new enzymes and biocatalysts for industrial applications is steadily growing but still in its infancy. [6] Efforts to capture enzymes from metagenomic samples have relied upon the creation of physical metagenomics libraries in more tractable bacteria. [7] Poor transcription and translation, enzymes with activ...