Sucrose 6-.alpha.-D-glucosyltransferase from Streptococcus sobrinus: characterization of a glucosyl-enzyme complex

Abstract

A covalent glucosyl-enzyme was isolated from a quenched reaction of Streptococcus sobrinus sucrose 6-.alpha.-D-glucosyltransferase and radiolabeled sucrose. No complex was observed with heat-inactivated enzyme or when sucrose was replaced with radiolabeled maltose or glucose. The complex was stable at pH 2 in 1% sodium dodecyl sulfate, 6.0 M urea, and 4.0 M guanidine hydrochloride, but became increasingly labile with increased pH (32-min half-life at pH 7.0). D-Glucose was the exclusive radiolabeled compound identified when all radioactivity was released under mild alkaline conditions. Glucosyl-enzyme hydrolysis rates were linearly dependent on hydroxide ion concentration, giving a second-order rate constant of 2.15 .times. 105 M-1 min-1. When compared to the base lability of known glycosyl amino acid derivatives, the pH dependency of the glucosyl-enzyme most closely paralled a glocosyl linkage to a carboxyl group. A novel application of a carbohydrate high-performance liquid chromatography column in aqueous solution was used to identify the anomeric form of D-glucose released on (i) alkaline hydrolysis of denatured glucosyl-enzyme and (ii) native enzyme hydrolysis of sucrose. The .beta.-anomer was identified in the former case and the .alpha.-anomer in the latter. The results with the denatured glucosyl-enzyme are consistent with a .beta.-glucosyl ester linkage to an aspartic or glutamic acid that hydrolyzes at the ester carbon with retention of anomeric configuration; for native glucosyltransferase catalysis, the data are consistent with a .beta.-glucosyl covalent intermediate as well, where deglucosylation occurs by attack at the acetal carbon with anomeric inversion. However, the glucosyl-enzyme complex could not be renatured to demonstrate catalytic competence, leaving open the possibly that the covalent bond formed during collapse of the enzyme active site. While the glucosyl bond was extremely base labile, it was found to be sufficiently stable at low pH to survive pepsin proteolytic cleavage. Active-site labels for this family of oral bacterial glycosyltransferases are virtually unknown, and the complex may be a useful probe to study the structure of the enzyme active site.