Design of a light-driven biocatalyst for biofuel production

Light driven reactions are rare in biochemistry, and only three light-dependent enzymes have been characterised to date. However, acceleration of chemical reactions by visible light offers environmentally friendly routes to chemical synthesis that may be practical and scalable for use in industrial manufacture. The discovery of an algal photoenzyme that uses blue light to convert fatty acids into hydrocarbons suggest a hitherto hidden scope for biotechnologically useful photochemistry by means of cofactor-dependent enzymes. Indeed, others have shown that non-natural reactions can occur for a range of cofactor-dependent enzymes under appropriate conditions. We seek to investigate what are the minimal requirements for a light-driven decarboxylase and will make use of the natural decarboxylase enzyme Fdc (a member of the widespread UbiD family). The latter enzyme makes use of a prenylated cofactor (prFMN) in vivo, but can be made to bind unmodified flavin (FMN) in vitro. Crucially, substrates are still able to bind adjacent to the flavin cofactor but no reaction is observed under dark conditions. Hence, two key elements are present: close juxtaposition of substrate/flavin and an active site geared towards decarboxylation. Encouragingly, upon illumination we find a range of products is formed in a light-dependent manner. Hence, we are ideally poised to develop non-natural and robust light-dependent decarboxylases using the prFMN-dependent UbiD enzyme family as a suitable template. Our initial investigations will focus on product identification and quantification under a range of conditions (varying light wavelength and intensity, under anaerobic / aerobic conditions etc). Our initial data already suggest that the presence of oxygen alters the product profile, suggesting distinct oxidative photochemistry occurs in presence of oxygen. We will then continue our investigations by probing a range of variant Fdc enzymes that affect active site composition and the flavin binding environment. The enzyme used is an ideal test subject in this case, as crystal structures to atomic resolution are routinely obtained in our laboratory. Thus, we are able to investigate variants of interest in mechanistic detail by combining crystallography with DFT calculations as we have done previously for the natural prFMN dependent reaction. The nature of the active site (in terms of the presence of aromatic side chains or acid/base residues) will be altered in a rational way to explore the affect on light-dependent catalysis. The long term goal is to generate an efficient FMN and light dependent UbiD that can readily be applied to range of substraets to support industrial application. As such, the project is directly within the industrial biotechnology remit, further demonstrated by the industrial support this receives.