... understanding life in molecular detail

Research in The Berry Group

How to tailor enzyme activity

The group investigates the possibility of altering the properties of enzymes to carry out unnatural, but desirable, reactions using the methods of protein engineering and directed evolution. At the University of Leeds, our research includes the important class of enzymes, the aldolases, which catalyse the formation of carbon-carbon bonds. This type of reaction lies at the heart of synthetic organic chemistry and is crucially important both within the cell and in industry for increasing the molecular complexity of biomolecules. Rationale redesign of the enzyme lead to changes in substrate specificity. More recently, it has become apparent that Man's understanding of the subtleties of protein structure and its impact on catalysis is limited, and new, more random approaches (termed Directed Evolution) that mimic natural evolution have been developed to allow Nature to teach us how to create new enzymes. We have embraced this methodology and have had some notable successes in altering the specificity, reaction chemistry and stability of enzymes using this approach.

Having used FBP-aldolases to develop the methodology of both rational design and directed evolution of enzyme catalysis, in 2006 I changed direction to work on the synthetically important, N-acetylneuraminic acid lyase (NAL). This enzyme is responsible for the synthesis of sialic acid, a molecule involved in a myriad of important cellular recognition events such as the interaction of the influenza virus with human cells. The work reached an exciting phase when we started tackling the difficult challenge of the construction of enzymes for the specific stereochemically controlled synthesis of either of a pair of stereoisomers of a product at will. We were able to demonstrate, for the first time, the directed evolution of a pair of complementary stereospecific enzymes. Considering that this involves the alteration of the stereochemistry of the enzyme reaction itself, this is a striking result. These results highlighted a crucial role for residue 167 in determining stereoselectivity, since alteration to glycine conferred 4S-stereoselectivity on the enzyme, while alteration to valine resulted in 4R-specificity. This has been followed up by a demonstration that the same amino acid substitutions can be 'transplanted' into enzymes engineered to have other substrate specificity and that the changes result in the same alterations in stereochemistry in those enzymes as well. These results thus open the door to the construction of tailor-made enzymes for the synthesis of a wide range of complex molecules including antibiotics, antivirals and anti-cancer drugs. Such biocatalytic synthetic steps are new being seen as the way forward for the construction of the next generation of drug molecules by the pharmaceutical industries. This work is now being extended to generate fluorinated compounds and we are particularly interested in the possibility of engineering the aldol condensation to generate two stereo-centres at will during the enzyme reaction.

To fully benefit from directed evolution experiments, we need to understand the underlying fundamental principles of the structure-activity relationship in engineered enzymes. This is especially important in understanding how the alterations of substrate specificity, increases in stability and enzyme reaction stereochemistry have been brought about by the small number of mutations introduced during the directed evolution and their effect on the enzyme structure. I have therefore driven a new avenue of collaborative research initially with Prof Simon Phillips, and since his departure from Leeds, with Dr Arwen Pearson. A jointly supervised PhD student has recently solved the X-ray crystal structures of around 10 of the evolved enzymes and the secrets of how the changes have been brought about are now being revealed. For example, the change of enzyme substrate specificity from the normal glycerol-based group at C6 of sialic acid to a dipropylamide group seems to be mainly controlled by the size of the residue at position 192, and full libraries of variants at this position have been constructed and characterised. Excitingly, these structures have also shed light on the mechanism of these enzymes, a hotly debated topic, and we have developed a new strategy using disabled mutants of the enzyme (particularly at Tyr-137) to trap reaction intermediates along the catalytic pathway to reveal the details of the full mechanism.

Another of my recent developments has been to extend the work described above to the subsequent enzymes from the biosynthetic pathways of complex sialylated carbohydrates. Such carbohydrates mediate critical interactions in processes as diverse as protein folding, protein trafficking, antigenicity, in vivo half-life, cell-cell signalling, inflammation, in oncogenic transformations and tumorogenesis in vertebrates and in mechanisms of infection and immunity in microbe-host interactions. Sialylated complex carbohydrates are therefore important target molecules, but their size and stereocomplexity means that they are very difficult to synthesise. We have cloned and overexpressed a number of enzymes from the pathways of sialylated sugars from Campylobacter and Pasteurella including CMP-sialic acid synthetase and the sialyl transferases, CstII and Pst. Directed evolution of these enzymes is well underway, and new assays are being developed. These rely either on tracking pH changes associated with the enzyme reaction, on trapping fluorescent products in cells or on a novel FRET-based assay. The required molecular biology is now complete and libraries of variants have been constructed. Library screening is underway to search for variants with altered specificities. One of the new areas of research underway in this area is to alter not only the substrate specificity and stereochemistry, but also the regiochemistry of these important reactions. In the longer term, these experiments should lead to developments in metabolic engineering where simple feed-stuffs can be converted to complex carbohydrates in multistep pathways in engineered bacteria.

Finally in this area, we have (since May 2009) started an exciting new line of research. The 20 canonical amino acids cannot always provide enough functionality to allow catalysis of all the cell's required reactions, and Nature occasionally uses post-translational modifications to provide new functionality. We, as protein engineers, now have the capablility of also incorporating non-proteogenic amino acids into enzymes. We have adopted both the method of chemical modification and in vivo incorporation of unnatural amino acids using orthogonal tRNA/tRNA-RS pairs and have demonstrated the ability to remove and then reconstruct enzyme activity in this way. This opens an exciting possibility of further extending the range of enzyme catalysis to novel reaction types.

We would like to thank the following organisations for the funds to support our research over the years:
BBSRC, EPSRC, The Royal Society, The Wellcome Trust, and The Leverhulme Trust.