Exploring biocatalysts for organic synthesis: Mechanism, promiscuity and engineering
Hennessy, Alexis James Antoine
In recent years, enzymes, and their uses as biocatalysts for synthetic organic chemistry, have been positively disrupting the pharmaceutical and chemical industries. With the development of the third wave of biocatalysis and directed evolution, enzyme engineering has become easier, cheaper and quicker. However, while random enzyme evolution is possible in the absence of structural data, the development of efficient biocatalysts is accelerated when 3D structures and mechanistic information of the target are available. In this work, two enzymes, each catalysing different chemical transformations will be studied, with the aim of gaining structural and mechanistic data which shall be used to guide future engineering strategies. The first enzyme studied is the pimeloyl-ACP methyl ester esterase, BioH, from E. coli (EcBioH). Previously it was found that EcBioH is able to catalyse the Morita-Baylis-Hillman (MBH) coupling of 4-nitrobenzaldehyde (4-NBA) and methyl vinyl ketone (MVK), with less than 50% conversion. The MBH reaction forms a C-C bond by coupling an aldehyde and an electron-poor alkene, and leads to the formation of a chiral alcohol. However, the mechanism of the enzyme-catalysed biotransformation is unclear. Specifically the identities of the active residues involved in substrate binding and catalysis are not known. Here we describe the characterisation of recombinant EcBioH and study its role in catalysing the MBH reaction. We used site directed mutagenesis to probe the role of a proposed catalytic triad of conserved residues (Ser-Asp-His) in the mechanism. We also investigate the involvement of a cysteine residue by covalent modification and mass spectrometry. Since none of these residues were found to be essential for MBH catalysis, this suggested that the protein hexahistidine (also known as 6His) affinity tag used to purify the enzyme displays inherent MBH activity. This hypothesis was confirmed by screening other hexahistidine tagged proteins available in the laboratory and discovering that they also displayed MBH activity. Furthermore, an engineered EcBioH “double-tagged” at both the N- and C- termini displayed the highest MBH activity. No enzyme displayed enantioselectivity in the MBH product distribution. These results provide strong evidence that this commonly-used tag plays an important role in MBH biocatalysis. These observations suggested that the intrinsic MBH activity of the tag arises from a synergistic acid/base relationship between the histidine residues and oxo-anion intermediates.In a second, independent project, a small family of enzymes, the nitrile synthetases (NS) was investigated. Interestingly, members of this family catalyse the conversion of a carboxylic acid to a nitrile in an ATP-dependent process. The reaction proceeds in two steps; the first involves the formation of an amide intermediate, and the second step results in amide to nitrile conversion. There is still a lack of clarity of the mechanistic role of ATP in the two steps catalysed by NS. Most notably, no attempts at separating/decoupling the two steps catalysed by the NSs have been successful so far. In this work, two members of the NS family, QueC from Bacillus subtilis (BsQueC) and ToyM from Streptomyces rimosus (SrToyM) were purified from E. coli and characterised. Both NSs were assayed for their natural reaction, the transformation of 7-carboxydeazaguanine (CDG) into 7-cyanodeazaguanine (PreQ0), via the corresponding primary amide (ADG). Their substrate scope was assessed and it was found that both enzymes had a very restricted acid substrate selectivity but were able to form a small range of secondary and tertiary amides with primary and secondary amines. Crystal trials resulted in the preparation of diffraction-quality crystals and the determination of the X-ray structure of the BsQueC:ADP complex at a resolution of 2.1 Å. This structure, was combined with the structure of a previously published BsQueC:phosphate complex and a sequence alignment of various members of the NS family. This analysis located well conserved residues and guided rational engineering of BsQueC. Well-conserved hydrophilic residues in close proximity to the phosphate moiety of the ADP molecule identified three residues for mutagenesis. A large decrease of the amide to nitrile conversion activity was observed for two of the single-point mutants generated (BsQueC K163A and BsQueC R204A). Importantly, a double-mutant enzyme which combines both these mutations (BsQueC K163A R204A) was not able to catalyse nitrile formation. However, this rationally-designed biocatalyst displayed bone-fide amide synthetase activity and allowed kinetic parameters for the CDG substrate to be determined for the first time. Finally, the first large scale biocatalysed transformation of CDG to ADG was carried out and the amide product was isolated with >95% purity. These studies provide the necessary insight and strategy required to carry out further directed evolution of the NS family to expand their substrate scope and allow synthesis of commercially-relevant amide- and nitrile-containing compounds.