The mechanism of action employed by a glycosidase is dictated, in part, by the distance between the two catalytic carboxylic acids. In the retaining endo-beta-1,4-xylanase from Bacillus circulans, this critical distance (approximately 5.5 A) has been altered by mutagenesis of the putative acid/base catalyst Glu172. An increase in the separation (Glu172Asp) resulted in a 400-fold decrease in the k(cat) value for xylan hydrolysis. By contrast, a decrease in the separation, achieved by the selective carboxymethylation of the Glu172Cys mutant, caused only a 25-fold reduction in the rate of xylan hydrolysis. Altering the length of the acid/base catalyst had a less detrimental effect on the hydrolysis of aryl xylobiosides, with k(cat)/Km values being reduced only 3-23-fold relative to the wild-type enzyme. Complete removal of the carboxyl group had a more dramatic effect. The Glu172Cys and Glu172Gln mutants exhibited no measurable activity on xylan or phenyl xylobioside, substrates which require acid catalysis. However, these mutants were capable of hydrolyzing aryl xylobiosides with relatively good leaving groups (pKa < 5.5), which need little protonic assistance. The addition of sodium azide caused significant rate increases for the hydrolysis of 2,5-dinitrophenyl beta-xylobioside (pKa = 5.15) by Glu172Cys and Glu172Gln. Thus, the absence of an acid/base catalyst can be partially compensated for by the addition of an anionic nucleophile. These results are consistent with Glu172 functioning as the acid/base catalyst in B. circulans xylanase and emphasize the functional importance of the carboxyl group found at this position.

The relative positioning of the two carboxyl groups at the active site of glycosidases is crucial to their function and the mechanism followed. The distance between these two groups in Bacillus circulans xylanase has been modified by mutagenesis of the catalytic nucleophile Glu78. An increase in the separation (Glu78Asp) results in a large (1600-5000-fold) reduction in the rate of the glycosylation step, but little change in the extent of bond cleavage or proton donation at the transition state. A decrease in the separation was achieved by selective carboxymethylation of the Glu78Cys mutant. This modified mutant was only 16-100-fold less active than wild-type enzyme, and its transition state structure was similarly unchanged. Complete removal of the carboxyl group (Glu78Cys) resulted in a mutant with no measurable catalytic activity. Furthermore, it did not even undergo the first step, glycosylation of the active site thiol. These results confirm the importance of precise positioning of the nucleophile at the active site of these enzymes.

Tetrachlorohydroquinone dehalogenase catalyzes the reductive dehalogenation of tetrachloro- and trichlorohydroquinone to give 2,6-dichlorohydroquinone in the pathway for degradation of pentachlorophenol by Sphingobium chlorophenolicum. Previous work has suggested that this enzyme may have originated from a glutathione-dependent double bond isomerase such as maleylacetoacetate isomerase or maleylpyruvate isomerase. While some of the elementary steps in these two reactions may be similar, the final step in the dehalogenation reaction, a thiol-disulfide exchange reaction that removes glutathione covalently bound to Cys13, certainly has no counterpart in the isomerization reaction. The thiol-disulfide exchange reaction does not appear to have been evolutionarily optimized. There is little specificity for the thiol; many thiols react at high rates. TCHQ dehalogenase binds the glutathione involved in the thiol-disulfide exchange reaction very poorly and does not alter its pK(a) in order to improve its nucleophilicity. Remarkably, single-turnover kinetic studies show that the enzyme catalyzes this step by approximately 10000-fold. This high reactivity requires an as yet unidentified protonated group in the active site.