A refined model of the sugar binding site of yeast hexokinase B.

The sugar binding site of monomeric yeast hexokinase B complexed with the competitive inhibitor o-toluoylglucosamine has been examined in the model resulting from a crystallographic refinement at 2·1 Å resolution. Difference Fourier maps calculated assuming various sugar configurations demonstrate that the o-toluoylglucosamine binds in the chair equatorial conformation with its 1-hydroxyl axial (α-anomer). The absence of a chemically derived amino acid sequence has complicated our interpretations of sugar-enzyme interactions. Nevertheless, we conclude that the carboxyl group of Asp189 is hydrogen-bonded to both the 6- and 4-hydroxyl groups. The 4-hydroxyl group is hydrogen-bonded also to Asx188 and Asx215, while the 3-hydroxyl is interacting with both Asx245 and Asx 188, consistent with the enzyme's observed sugar specificity. The carboxyl group of Asp 189 is excluded from solvent in the presence of glucose and may be acting as a general base to enhance the nucleophilicity of the 6-hydroxyl group and thereby promote its attack on the γ-phosphate of ATP.Glucose is shown to bind to the enzyme in the same orientation and conformation as the sugar moiety of o-toluoylglucosamine, so that the 6-hydroxyl group and the carboxyl of Asp 189 are in identical positions in complexes with these two sugars. The fact that o-toluoylglucosamine is not a substrate must be explained by two observations. First, the binding of glucose results in one lobe rotating by 12 ° relative to the other lobe, thereby closing off the slit into which the sugar has bound (Bennett &; Steitz, unpublished results). Second, o-toluoylglucosamine does not produce this conformational change, because the bulky toluoyl group prevents the closing of this slit between the two lobes. We conclude, therefore, that the large glucose-induced conformational change must be essential for subsequent catalytic steps.It appears unlikely from this study that thiols play any direct role in catalysis or in substrate binding. One thiol group, however, lies 5·5 Å from the 3-hydroxyl and is hydrogen-bonded to three of the Asx groups that are binding the sugar. Chemical modification of this buried thiol would disrupt the glucose binding site, which could account for the observation (Otieno et al., 1977) that cyanylation of one of the enzyme's thiols abolishes enzymatic activity.A sulfate molecule is bound to the enzyme by two serine side-chains and its sulfur atom is 5·5 Å from the 6-hydroxyl group of glucose. If the γ-phosphate of ATP binds to this sulfate binding site, it would still be a little too far from the 6-hydroxyl for direct phosphoryl transfer.