Speaker
Description
A glycosynthase is a mutant enzyme that is made from a retaining glycosidase. A retaining glycosidase has a catalytic nucleophile and an acid/base (both, normally Asp or Glu) in the active site. When the catalytic nucleophile is mutated to non-nucleophilic residue, the mutated glycosidase cannot normally hydrolyse its cognate substrate, but still catalyses the glycosidic bond formation between an activated glycosyl donor with an opposite anomeric configuration and an appropriate nucleophile. The usual donor substrate is a glycosyl halide, which can be easily and cheaply synthesized in laboratory. Thus, it has a clear economic advantage over glycosyltransferases that use expensive nucleotide sugar substrates. Moreover, since glycosynthases supresses hydrolysis due to the removal of the catalytic nucleophile, the yield is normally high unlike glycoside hydrolases that catalyse a glycosidic bond formation under right conditions.
The current hurdle in the glycosynthase approach is the universally low kinetics and high substrate molar equivalence required to promote synthesis and supress still remaining hydrolysis. The low kinetics is inevitable since a glycosynthase is basically a mutant enzyme. The residual hydrolysis activity is a remnant of the ancestral activity. This project has attempted to address these issues of glycosynthase to make it a viable option to replace glycosyltransferase by applying a computation modelling. In this project, we avoided the long arduous laboratory screening process of directed evolution, and instead use Enzbond’s proprietary methods to design and screen novel mutants. The traditional complexity of the sequence-space relationship is intensified in reactions where there is a bifurcation of the mechanism (i.e hydrolysis and synthesis originate from the same point) as the residues surrounding and in direct contact with the substrate can play both additive and subtractive roles in observed activity. This temporal modulating behaviour results in an observed residue complexity that both directed evolution and artificial intelligence have difficulty in delineating. Utilizing Enzbond’s proprietary methods and technology, we were able to delineate the contribution of the residues and supress the hydrolysis mechanism even at stoichiometric loadings of the donor substrate. Mutations were conducted and mutants screened in silico, exploring the vast amino acid sequence space that could be achieved in laboratory. We have achieved an optimization of a glycosynthase activity based on Agrobacterium sp. glycosynthases (Abg), 2F6, which is a year-worth in terms of laboratory work and found a high-performance glycosynthase that can produce an aryl glycoside as a model product. In our optimisation, kcat/Km. of the glycosynthase 2F6 increased over 2 folds. This reaction has a potential to be applied to the synthesis of higher valued aryl glycosides such as flavonoid glycosides.
Based on our first generation data, further round of mutagenesis is expected to produce enzymes with improved activities. This study suggests a high potential of in silico mutagenesis to generate high activity enzymes that can be used in industry and may provide a framework for artificial intelligence algorithm.
References
Manuscript in preparation
2F6: Kim, Y.-W., Lee, S. S., Warren, R. A. J. & Withers, S. G. Directed Evolution of a Glycosynthase from Agrobacterium sp. Increases Its Catalytic Activity Dramatically and Expands Its Substrate Repertoire. J. Biol. Chem. 279, 42787-42793 (2004)
| Keywords | protein engineering, glycosynthase, in silico, directed evolution, enzyme kinetics |
|---|
