Biochemical Characterization of the α-l-Rhamnosidase DtRha from Dictyoglomus thermophilum: Application to the Selective Derhamnosylation of Natural Flavonoids.

α-l-Rhamnosidases are catalysts of industrial tremendous interest, but their uses are still somewhat limited by their poor thermal stabilities and selectivities. The thermophilic DtRha from Dictyoglomus thermophilum was cloned, and the recombinant protein was easily purified to homogeneity to afford 4.5 mg/L culture of biocatalyst. Michaelis-Menten parameters demonstrated it to be fully specific for α-l-rhamnose. Most significantly, DtRha demonstrated to have a stronger preference for α(1 → 2) linkage rather than α(1 → 6) linkage when removing rhamnosyl moiety from natural flavonoids. This selectivity was fully explained by the difference of binding of the corresponding substrates in the active site of the protein.

Is the acid/base catalytic residue mutation in ß-d-mannosidase DtMan from Dictyoglomus thermophilum sufficient enough to provide thioglycoligase activity?

Glycoside hydrolases can be turned into thioglycoligase by mutation of the acid/base catalytic carboxylate residue. These mutants have proven valuable to generate S-glycosides, however, few examples in literature have described efficient thioglycoligase activity, and even fewer the underlying molecular mechanism. DtMan, a GH2 family ß-d-mannosidase from the thermophilic Dictyoglomus thermophilum was cloned and expressed in E. coli. The recombinant protein is highly specific for ß-d-mannosides, and exhibits efficient catalysis constants coupled to thermostability. However, seven variants bearing mutated acid/base residue could not be turned into efficient thioligases. Crystal structure of DtMan Glu425Cys mutant and molecular modeling calculations have demonstrated that unlike other GH2 thioligase reported, active site accessibility of thiol acceptor may be impaired by entrance loop rigidity. This structural feature may explain why DtMan mutants do not exhibit thioglycoligase activity.

Unraveling the substrate recognition mechanism and specificity of the unusual glycosyl hydrolase family 29 BT2192 from Bacteroides thetaiotaomicron.

Glycosyl hydrolase (GH) family 29 (CAZy database) consists of retaining α-l-fucosidases. We have identified BT2192, a protein from Bacteroides thetaiotaomicron, as the first GH29 representative exhibiting both weak α-l-fucosidase and ß-d-galactosidase activities. Determination and analysis of X-ray structures of BT2192 in complex with ß-d-galactoside competitive inhibitors showed a new binding mode different from that of known GH29 enzymes. Three point mutations, specific to BT2192, prevent the canonical GH29 substrate α-l-fucose from binding efficiently to the fucosidase-like active site relative to other GH29 enzymes. ß-d-Galactoside analogues bind and interact in a second pocket, which is not visible in other reported GH29 structures. Molecular simulations helped in the assessment of the flexibility of both substrates in their respective pocket. Hydrolysis of the fucosyl moiety from the putative natural substrates like 3-fucosyllactose or Lewis(X) antigen would be mainly due to the efficient interactions with the galactosyl moiety, in the second binding site, located more than 6-7 Å apart.