1-Alkyl-d-gluconamidines: Are they ‘perfect’ mimics of the first transition state of glucosidase action?

The inhibition of four β-glucosidases of plant, fungal, and mammalian origin by N1-butyl- and N1-dodecyl-d-gluconamidine was determined. Comparison with the inhibition by the corresponding N-alkyl-d-glucosylamines revealed that the strongly basic amidines (pKa 10.8) were at the most 10-times more inhibitory than the weakly basic glucosylamines (pKa 6.5). The small enhancement of inhibitory potency, resulting from transforming the tetrahedral C-1 geometry of the glucosylamines to the planar sp2-geometry of the amidines, was ascribed to the inability of the fully protonated amidines to function as hydrogen bond acceptors with the catalytic acid of the enzyme. Additional evidence for the importance of a hydrogen bond for strong inhibition came from the comparison of K1-values of the weakly basic 5-amino-5-deoxyhexopyranoses and 1,5-iminohexitols with those of the corresponding glyconamidrazones (pKa 8.4), which also have a planar C-1 geometry but are largely protonated under the assay conditions and which had similar or up to 104-times larger K1-values than the former. Transition state resemblance was judged from the ratio KS(alkyl β-glucoside)/KI(alkyl gluconamidine) relative to the rate acceleration factor kcat/kuncat (Wolfenden, Acc. Chem. Res., 5 (1972) 10–16). Compared to ratios of kcat/kuncat from ≥ 1011 to > 1013, the ratios for KS/KI were only from 103 to 2 × 104 except for β-glucosidase A3 from Asp. wentii which had KS/KI 2.8 × 106. This enzyme differs from the others by being strongly inhibited by cationic glycon and substrate analogues rather than by basic ones. The pH-dependence of 1/K1 and the ‘slow’ approach to the inhibition is discussed with respect to transition state resemblance.