Transglycosylation of engineered cyclodextrin glucanotransferases as O-glycoligases

An O-glycoligase is a hydrolytically impaired mutant of a retaining α-glycosidase in which the catalytic acid/base has been removed, but which can still perform transglycosylation when incubated with activated glycosyl fluoride donor sugars. In this paper, we describe another example, wherein a cyclodextrin glucanotransferase mutant (CGT-E284A) with an alanine residue at its general acid/base catalyst position (Glu284), was constructed. This mutant was hydrolytically inactive, but exhibited significant transglycosylation activity using α-maltosyl fluoride (αG2F) as donor, and either 4-nitrophenyl glucosides or maltosides as acceptors. To improve transglycosylation activity, a site-saturation mutagenesis library at Glu284 was created. Through a thin-layer chromatography-based screening process, two mutants were identified; (1) a mutant with a glycine residue at Glu284 (CGT-E284G) exhibiting improved transglycosylation activity compared with the original alanine mutant and (2) a mutant with a serine residue with residual hydrolytic activity. Kinetic analysis revealed that 4-nitrophenyl maltosides were better acceptors than 4-nitrophenyl glucosides. Transglycosylation activities of CGT-E284A and CGT-E284G were inhibited at high concentrations (>0.8 mM) of the acceptor sugars. In contrast, typical saturation kinetic behavior was observed upon varying the donor (αG2F) concentration at a fixed acceptor concentration (0.8 mM). The catalytic efficiencies (apparent kcat/KM) of CGT-E284G were generally three- to sixfold higher than those of CGT-E284A. Due to the rate at high concentrations of the acceptors, higher transglycosylation yields were achieved at a low concentration of the acceptors (69-84% at 1 mM) compared to those at a higher concentration (22-36% at 10 mM).