Nucleophilic Activation by Positioning in Phosphoryl Transfer Catalyzed by Nucleoside Diphosphate Kinase^,

Abstract

The nonenzymatic reaction of ATP with a nucleophile to generate ADP and a phosphorylated product proceeds via a dissociative transition state with little bond formation to the nucleophile. Consideration of the dissociative nature of the nonenzymatic transition state leads to the following question: To what extent can the nucleophile be activated in enzymatic phosphoryl transfer? We have addressed this question for the NDP kinase reaction. A mutant form of the enzyme lacking the nucleophilic histidine (H122G) can be chemically rescued for ATP attack by imidazole or other exogenous small nucleophiles. The ATP reaction is 50-fold faster with the wild-type enzyme, which has an imidazole nucleophile positioned for reaction by a covalent bond, than with H122G, which employs a noncovalently bound imidazole nucleophile [(kcat/KM)ATP]. Further, a 4-fold advantage for imidazole positioned in the nucleophile binding pocket created by the mutation is suggested from comparison of the reaction of H122G and ATP with an imidazole versus a water nucleophile, after correction for the intrinsic reactivities of imidazole and water toward ATP in solution. X-ray structural analysis shows no detectable rearrangement of the residues surrounding His 122 upon mutation to Gly 122. The overall rate effect of approximately 10(2)-fold for the covalent imidazole nucleophile relative to water is therefore attributed to positioning of the nucleophile with respect to the reactive phosphoryl group. This is underscored by the more deleterious effect of replacing ATP with AlphaTauPgammaS in the wild-type reaction than in the imidazole-rescued mutant reaction, as follows. For the wild-type, AlphaTauPgammaS presumably disrupts positioning between nucleophile and substrate, resulting in a large thio effect of 300-fold, whereas precise alignment is already disrupted in the mutant because there is no covalent bond to the nucleophile, resulting in a smaller thio effect of 10-fold. In summary, the results suggest a catalytic role for activation of the nucleophile by positioning in phosphoryl transfer catalyzed by NDP kinase.

Chemical Rescue of Phosphoryl Transfer in a Cavity Mutant: A Cautionary Tale for Site-Directed Mutagenesis,

“…To quantitate this difference (k cat /K M ) H 2 O ATP for H122A was measured, which required that the enzyme be saturated by Im (Scheme 2). However, at the highest Im concentrations, a modest downward curvature was observed in the Im dependence of the H122A reaction with ATP ( Figure 7), as had been seen previously in the Im dependence of the H122G reaction with ATP (11). As in the earlier work, this curvature was presumed to arise from binding of an inhibitory Im, in addition to the Im that binds and rescues in the nucleophile site (Scheme 3).…”

“…The solid line is drawn through the data solely to guide the eye and does not represent a particular model. The data are from (11). A cleft at the base of the nucleophile cavity (arrow, top center) may accommodate the aliphatic tail of long-chain alcohols, allowing them to react with ATP despite having volumes that exceed the volume of the cavity created by the mutation.…”

“…Mutant Dictyostelium NDPKs were overexpressed in Escherichia coli (XL1-Blue) using plasmid pndk as described (12), with minor modifications. Both proteins were purified as described previously for H122G and stored at -20°C (11). Protein concentration was determined using the calculated extinction coefficient at 280 nm: 9200 M -1 cm -1 (13).…”

“…It has previously been noted that properties of enzyme active sites are likely to influence the reactivity of individual nucleophiles, complicating interpretation of enzymatic Brøn-sted relationships (11,31,32). This limitation is exemplified by the H122G results depicted in Figure 2B.…”

“…As in the earlier work, this curvature was presumed to arise from binding of an inhibitory Im to the nucleotide base site, in addition to the Im that binds and rescues in the nucleophile site. An inhibitory Im would be expected to bind similarly to the base sites of wild-type NDPK, H122G, and H122A, so a binding constant of 0.45 M was assigned to the inhibitory H122A site based on the Im inhibition of wild-type NDPK (11). The saturation data for H122A were then fit to a model with two Im, one that serves as the nucleophile and one, with a K i ) 0.45 M, that is responsible for the inhibition, as diagrammed in Scheme 3.…”