A substrate in pieces: allosteric activation of glycerol 3-phosphate dehydrogenase (NAD+) by phosphite dianion

Abstract

The ratio of the second-order rate constants for reduction of dihydroxyacetone phosphate (DHAP) and of the neutral truncated substrate glycolaldehyde (GLY) by glycerol 3-phosphate dehydrogenase (NAD (+), GPDH) saturated with NADH is (1.0 x 10 (6) M (-1) s (-1))/(8.7 x 10 (-3) M (-1) s (-1)) = 1.1 x 10 (8), which was used to calculate an intrinsic phosphate binding energy of at least 11.0 kcal/mol. Phosphite dianion binds very weakly to GPDH ( K d > 0.1 M), but the bound dianion strongly activates GLY toward enzyme-catalyzed reduction by NADH. Thus, the large intrinsic phosphite binding energy is expressed only at the transition state for the GPDH-catalyzed reaction. The ratio of rate constants for the phosphite-activated and the unactivated GPDH-catalyzed reduction of GLY by NADH is (4300 M (-2) s (-1))/(8.7 x 10 (-3) M (-1) s (-1)) = 5 x 10 (5) M (-1), which was used to calculate an intrinsic phosphite binding energy of -7.7 kcal/mol for the association of phosphite dianion with the transition state complex for the GPDH-catalyzed reduction of GLY. Phosphite dianion has now been shown to activate bound substrates for enzyme-catalyzed proton transfer, decarboxylation, hydride transfer, and phosphoryl transfer reactions. Structural data provide strong evidence that enzymic activation by the binding of phosphite dianion occurs at a modular active site featuring (1) a binding pocket complementary to the reactive substrate fragment which contains all the active site residues needed to catalyze the reaction of the substrate piece or of the whole substrate and (2) a phosphate/phosphite dianion binding pocket that is completed by the movement of flexible protein loop(s) to surround the nonreacting oxydianion. We propose that loop motion and associated protein conformational changes that accompany the binding of phosphite dianion and/or phosphodianion substrates lead to encapsulation of the substrate and/or its pieces in the protein interior, and to placement of the active site residues in positions where they provide optimal stabilization of the transition state for the catalyzed reaction.

Figure 1

A. Dependence of the initial velocity, ν_i, of the reduction of DHAP by NADH catalyzed by GPDH (1.1 nM) on the total concentration of DHAP in the presence of 0.10 mM (▲) or 0.20 mM (●) NADH at pH 7.5, 25 °C and I = 0.12 (NaCl). B. Dependence of the initial velocity, ν_i, of the reduction of GLY (20 mM total, 1.2 mM free carbonyl form) by NADH (0.20 mM) on the concentration of GPDH at pH 7.5, 25 °C and I = 0.12 (NaCl).

Figure 2

A. Dependence of ν_i/[E] for the reduction of GLY by NADH (0.20 mM) catalyzed by GPDH (0.6 µM) on the concentration of exogenous phosphite dianion at pH 7.5, 25 °C and I = 0.12 (NaCl). Concentrations of GLY (free carbonyl from) were: ▽ [GLY] = 0.12 mM; △ [GLY] = 0.30 mM; □ [GLY] = 0.60 mM; ▼ [GLY] = 1.2 mM; ● [GLY] = 1.8 mM; ▲ [GLY] = 2.4 mM; ◆ [GLY] = 3.0 mM; ■ [GLY] = 3.6 mM. B. Dependence of the apparent second-order rate constant (k_cat/K_m)_app for the phosphite-activated GPDH-catalyzed reduction of GLY by NADH (0.20 mM) on the concentration of GLY at pH 7.5, 25 °C and I = 0.12 (NaCl), calculated as the slopes of the correlations in Figure 2A.

Figure 3

A. Free energy diagram for turnover of the truncated neutral substrate GLY (S) by the binary GPDH•NADH complex (E') and for the same reaction activated by the binding of exogenous phosphite dianion (HPO₃²⁻). Activation free energies were calculated using the Eyring equation at 298 K. The weakly bound Michaelis complex E'•HPO₃²⁻ is not shown, because the stability of this complex is not defined by our experiments. B. Free energy diagram illustrating our proposed model for activation of E' towards reduction of GLY by the binding of HPO₃²⁻. Binding energies are defined to be negative when the binding is thermodynamically favorable. The difference between the total intrinsic phosphite binding energy of −7.7 kcal/mol and the observed binding energy× > −1.4 kcal/mol for binding of HPO₃²⁻ to the inactive open enzyme E'_O to give the active closed enzyme E'_C•HPO₃²⁻ is the binding energy that is used to convert E'_O to E'_C. The observed value of ΔG^‡ = +20.2 kcal/mol for turnover of GLY (Figure 3A) has been partitioned into ΔG^‡ for reduction of GLY by E'_C and ΔG_conf for the unfavorable conformational change that converts E'_O to E'_C.

Scheme 1

Scheme 2

Scheme 3

Scheme 4

Scheme 5