PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Biochemistry. Author manuscript; available in PMC 2010 June 23.
Published in final edited form as:
PMCID: PMC2752442
NIHMSID: NIHMS134377

Hydron Transfer Catalyzed by Triosephosphate Isomerase. Products of the Direct and Phosphite-Activated Isomerization of [1-13C]-Glycolaldehyde in D2O

Abstract

Product distributions for the reaction of glycolaldehyde labeled with carbon-13 at the carbonyl carbon ([1-13C]-GA) catalyzed by triosephosphate isomerase (TIM) in D2O at pD 7.0 in the presence of phosphite dianion and in its absence were determined by 1H NMR spectroscopy. We observe three products for the relatively fast phosphite-activated reaction (Amyes, T. L., and Richard, J. P. (2007) Biochemistry 46, 5841-5854): [2-13C]-GA from isomerization with intramolecular transfer of hydrogen (12% of products), [2-13C, 2-2H]-GA from isomerization with incorporation of deuterium from D2O at C-2 (64% of the products), and [1-13C, 2-2H]-GA from incorporation of deuterium from D2O at C-2 (23% of products). The much slower unactivated reaction in the absence of phosphite results in formation of the same three products along with the doubly deuterated product [1-13C, 2,2-di-2H]-GA. The two isomerization products ([2-13C]-GA and [2-13C, 2-2H]-GA) are formed in the same relative yields in both the unactivated and the phosphite-activated reactions. However, the additional [1-13C, 2-2H]-GA and the doubly deuterated [1-13C, 2,2-di-2H]-GA formed in the unactivated TIM-catalyzed reaction are proposed to result from a nonspecific reaction(s) at the protein surface. The data provide evidence that phosphite dianion affects the rate, but not the product distribution, of the TIM-catalyzed reaction of [1-13C]-GA at the enzyme active site. They are consistent with the conclusion that both reactions occur at an unstable loop-closed form of TIM, and that activation of the isomerization reaction by phosphite dianion results from utilization of the intrinsic binding energy of phosphite dianion to stabilize the active loop-closed enzyme.

Triosephosphate isomerase (TIM)1 catalyzes the stereospecific, reversible 1,2-hydrogen shift at dihydroxyacetone phosphate (DHAP) to give (R)-glyceraldehyde 3-phosphate (GAP) by a single-base (Glu-165) proton transfer mechanism through an enzyme-bound cis-enediol(ate) intermediate (Scheme 1) (1, 2). The enzyme's low molecular weight (dimer, 26 kDa/subunit), high cellular abundance (3), and the centrality of proton transfer at carbon in metabolic processes (4, 5) have made TIM a prominent target for studies on the mechanism of enzyme action (6-8).

We showed previously that deprotonation of the neutral substrate (R)-glyceraldehyde by TIM is ca. 109-fold slower than the partly diffusion-controlled (9) turnover of the physiological phosphorylated substrate GAP (10). More than 80% of the 4 × 1010-fold enzymatic rate acceleration for deprotonation of GAP is due to utilization of the intrinsic binding energy of the small nonreacting phosphodianion group of the substrate (10, 11). Part of this “intrinsic phosphate binding energy” is expressed in the Michaelis complex and serves the obvious function of anchoring the substrate to TIM. However, these results suggested that a large fraction of this binding energy is utilized to activate the bound carbon acid substrate towards proton transfer.

Our work is now focused on understanding the role of flexible “phosphate gripper loops”, such as loop 6 of TIM (6, 12-18), in activating enzyme-bound substrates for a variety of reactions (19, 20). We reported earlier that TIM catalyzes the slow deprotonation of glycolaldehyde (GA) in D2O, resulting in deuterium labeling of GA at C-2 that can be monitored by 1H NMR spectroscopy (Scheme 2), with a second-order rate constant kcat/Km = 0.26 M-1 s-1 (19). The addition of exogenous phosphite dianion results in a very large increase in the observed second-order rate constant (kcat/Km)obsd for turnover of GA, and the plot of (kcat/Km)obsd against [HPO32-] provides evidence for saturation of the enzyme by phosphite. The data show that the separate binding of the phosphite dianion “piece” to TIM results in a 700-fold acceleration of proton transfer from carbon (Scheme 2) (19). This demonstration that the interactions between TIM and phosphite dianion activate a bound carbon acid substrate toward deprotonation, in the absence of a covalent connection, shows that the phosphodianion group of GAP does not simply anchor the substrate to the enzyme. Rather, it also serves the additional important function of providing binding energy that is utilized specifically for transition state stabilization (19, 21). We proposed that the phosphite activation of TIM can be attributed to utilization of the intrinsic phosphite binding energy to hold TIM in a catalytically active loop-closed form (19).

The TIM-catalyzed reactions of the whole substrates GAP and DHAP in D2O give complex mixtures of products, as shown in Scheme 3 for the reaction of GAP (22, 23). This mixture results from partitioning of the enediol(ate) phosphate intermediate at two different stages of the overall reaction (1, 24). Proton transfer from GAP to the carboxylate side chain of Glu-165 labels the side chain with H. In D2O, this labeled side chain partitions between reprotonation of the enediol(ate) phosphate at C-1 to form H-labeled DHAP (49% yield), and effectively irreversible exchange with deuterium from solvent D2O (kex, Scheme 3). The D-labeled side chain then partitions between reaction of the enediol(ate) phosphate at C-2 to form D-labeled GAP (21% yield) and at C-1 to form D-labeled DHAP (31% yield) (22).

It is not known whether TIM catalyzes hydrogen transfer between the hydroxymethylene and carbonyl carbons of glycolaldehyde (GA) because this formal isomerization reaction is degenerate. Therefore, reaction of the D-labeled side chain of Glu-165 with the putative enediol(ate) intermediate of the TIM-catalyzed deprotonation of GA results in formation of only one product, [2-2H]-GA (19). We have therefore extended our comparison of TIM-catalyzed isomerization of whole substrates GAP and DHAP with that of the substrate pieces {GA + HPO32-} to an examination of the product distribution for the TIM-catalyzed reaction of [1-13C]-glycolaldehyde ([1-13C]-GA) in D2O, in both the presence and absence of phosphite dianion.

We report here that the phosphite-activated TIM-catalyzed reaction of [1-13C]-GA in D2O gives the three expected products of partitioning of the enzyme-bound enediol(ate) intermediate (see Chart 1), in yields similar to those of the corresponding products of the TIM-catalyzed reactions of the whole substrates GAP (22) and DHAP (23). Moreover, similar yields of these products are observed for the unactivated TIM-catalyzed reaction of [1-13C]-GA in D2O in the absence of phosphite. These results suggest that the active catalyst for the slow isomerization of [1-13C]-GA is an unstable loop-closed form of TIM and that the intrinsic binding energy of phosphite dianion is utilized to stabilize the enzyme in the active loop-closed form.

Materials and Methods

Rabbit muscle glycerol 3-phosphate dehydrogenase (GPDH) was purchased from United States Biochemical or MP Biomedicals. Bovine serum albumin (BSA) was from Roche. DEAE Sepharose Fast Flow was from GE Healthcare. D,L-Glyceraldehyde 3-phosphate diethyl acetal (barium salt), dihydroxyacetone phosphate (lithium salt), Dowex 50WX4-200R and NADH (disodium salt) were from Sigma. Triethanolamine hydrochloride and imidazole were from Aldrich. Sodium phosphite (dibasic, pentahydrate) was from Riedel-de Haën (Fluka). [1-13C]-Glycolaldehyde (99% enriched with 13C at C-1, 0.09 M in water) was purchased from Omicron Biochemicals. D2O (99.9% D) and DCl (35% w/w, 99.9% D) were from Cambridge Isotope Laboratories. Imidazole was recrystallized from benzene. Water was from a Milli-Q Academic purification system. All other commercially available chemicals were reagent grade or better and were used without further purification.

The plasmid pBSX1cTIM containing the wild-type gene for TIM from chicken muscle (25) and E. coli strain DF502 (strepR, tpi-, and his-) whose DNA lacks the gene for TIM (26) were generous gifts from Professor Nicole Sampson. E. coli strain DF502 was transformed with pBSX1cTIM and TIM was expressed and purified according to published procedures with ion exchange chromatography performed on DEAE sepharose (15). The enzyme obtained from the final column was judged to be homogeneous by gel electrophoresis. The concentration of TIM was determined from the absorbance at 280 nm using an extinction coefficient of 3.2 × 104 M-1 cm-1 (27). The following kinetic parameters were determined for turnover of GAP in 30 mM triethanolamine buffer at pH 7.5 and 25 °C (I = 0.1, NaCl) using a coupled assay (see below): kcat = 2300 s-1 and Km = 0.45 mM (22).

Preparation of Solutions

Solution pH or pD was determined at 25 °C using an Orion model 720A pH meter equipped with a Radiometer pHC4006-9 combination electrode that was standardized at pH 4.00 and 7.00 at 25 °C. Values of pD were obtained by adding 0.40 to the observed reading of the pH meter (28). Before preparation of solutions in D2O the bulk of the water of crystallization of sodium phosphite was removed by drying in vacuo as described previously (19). Buffered solutions of 30 mM imidazole (20% free base, pD 7.0) and 80 mM phosphite (50% free base, pD 7.0) in D2O were prepared as described previously (19).

D,L-Glyceraldehyde 3-phosphate diethyl acetal (barium salt) was converted to the free aldehyde by treatment with a suspension of Dowex 50WX4-200R (H+ form) in boiling water, as described previously (22) and the resulting solution of D,L-glyceraldehyde 3-phosphate (pH ≈ 2) was stored at -20 °C. Before use in routine enzyme assays the solution was adjusted to the appropriate pH by the addition of 1 M NaOH, after which it was stored at -20 °C. The concentration of GAP in the racemic mixture was determined from the amount of NADH consumed during quantitative TIM-catalyzed isomerization of GAP to form DHAP that was coupled to the oxidation of NADH using GPDH.

[1-13C]-Glycolaldehyde (1 mL of a 90 mM solution in H2O) was reduced to a volume of ca. 100 μL by rotary evaporation. 5 mL of D2O was added and the volume was again reduced to ca. 100 μL by rotary evaporation. This procedure was repeated twice more and ca. 900 μL of D2O was added to the final solution to give a final volume of ca. 1 mL. The stock solution of [1-13C]-GA in D2O was stored at room temperature to minimize the content of glycolaldehyde dimer (19) and the concentration of [1-13C]-GA was determined by 1H NMR spectroscopy as follows. 50 μL of the stock solution of [1-13C]-GA in D2O was diluted with 700 μL of 30 mM imidazole buffer (20% free base, pD 7.0) in D2O and the concentration of [1-13C]-GA was determined from comparison of the integrated areas of the signals for the protons of [1-13C]-GA hydrate and the C(4,5)-protons of imidazole, with a correction for the presence of 6.1% of the free carbonyl form of GA that is present in equilibrium with the hydrate (19, 29).

Enzyme Assays

All enzyme assays were carried out at 25 °C. Changes in the concentration of NADH were calculated from the changes in absorbance at 340 nm using an extinction coefficient of 6220 M-1 cm-1. One unit is the amount of enzyme that converts 1 μmol of substrate to product in 1 min under the specified conditions.

GPDH was dialyzed against 20 mM triethanolamine buffer (pH 7.5) at 4 °C and was assayed by monitoring the oxidation of NADH by DHAP at 340 nm. The assay mixture contained 100 mM triethanolamine (pH 7.5, I = 0.1), 0.01% BSA, 0.2 mM NADH and 1 mM DHAP.

The activity of TIM in the stock solutions dialyzed in D2O or of the diluted reaction mixtures in D2O was assayed by coupling the isomerization of GAP to give DHAP to the oxidation of NADH by DHAP using GPDH, monitored at 340 nm as described previously (19, 30). Dilute solutions of TIM were stabilized with 0.01% BSA. The standard assay mixture (1.0 mL) contained 100 mM triethanolamine (pH 7.5, I = 0.1), 0.2 mM NADH, 5 mM D,L-glyceraldehyde 3-phosphate (2.5 mM GAP, ≈ 6 Km), 1 unit of GPDH and 0.01 - 0.05 nM TIM. Before the addition of TIM, the low background velocity that is due mainly to the isomerization of GAP catalyzed by TIM present as a contaminant in the commercial GPDH was determined over 2 - 4 min. This background velocity, vo, generally represented ≤ 2% of the total initial velocity, vi, that was determined subsequently over 5 - 10 min after the addition of TIM. Values of Vmax (M s-1) were calculated from the values of (vi - vo) using the initial concentration of GAP and Km = 0.45 mM. The concentration of TIM in the assay mixture was calculated from the values of Vmax (M s-1) using the relationship [TIM] = Vmax/kcat with kcat = 2300 s-1.

TIM-Catalyzed Reactions of [1-13C]-GA Monitored by 1H NMR

Chicken muscle TIM was dialyzed at 4 °C against 30 mM imidazole buffer (20% free base, pD 7.0) in D2O at I = 0.024 for reactions in the presence of phosphite dianion or at I = 0.1 (NaCl) for reactions in its absence.

The TIM-catalyzed reactions of [1-13C]-GA in D2O in the absence of phosphite were initiated by adding 720 μL of TIM (ca. 85 units/μL, 16 mg/mL) in 30 mM imidazole buffer (20% free base, pD 7.0) at I = 0.1 (NaCl) in D2O to 180 μL of a solution of [1-13C]-GA and NaCl in D2O to give final concentrations of 20 mM [1-13C]-GA and 24 mM imidazole at pD 7.0 and I = 0.1 (NaCl). 750 μL of this solution was transferred to an NMR tube and 1H NMR spectra at 25 °C (32 transients) were recorded hourly over a period of ca. 24 hours, during which time there was ca. 29% reaction of [1-13C]-GA. The remaining portion of the reaction mixture was incubated at 25 °C and the activity of TIM was determined via periodic standard assay (see above).

The TIM-catalyzed reactions of [1-13C]-GA in D2O the presence of phosphite dianion were initiated by adding (520) μL of TIM (ca. 64 units/μL, 12 mg/mL) in 30 mM imidazole buffer (20% free base, pD 7.0) at I = 0.024 in D2O to 1.0 mL of a solution containing 20 mM [1-13C]-GA, 6 mM imidazole (20% free base, pD 7.0) and 5 - 20 mM phosphite dianion at pD 7.0 and I = 0.1 (NaCl). The final reaction mixture contained 20 mM [1-13C]-GA, 6 mM imidazole and 5 - 20 mM phosphite dianion (10 - 40 mM total phosphite) at pD 7.0 and I = 0.1 (NaCl). 750 μL of this solution was transferred to an NMR tube and 1H NMR spectra at 25 °C (32 transients) were recorded hourly over a period of ca. 12 hours, during which time there was ca. 23% reaction of [1-13C]-GA. The remaining portion of the reaction mixture was incubated at 25 °C and the activity of TIM in was determined via periodic standard assay (see above).

For the reactions in both the absence and presence of phosphite dianion there was no significant decrease in enzyme activity (± 10%) during the course of the experiment. After accumulation of the final NMR spectrum the protein was removed by ultrafiltration and the pD of the filtrate was determined and found to be within 0.05 units of the initial value of 7.0.

1H NMR Analyses

1H NMR spectra (500 MHz) were recorded in D2O at 25 °C using a Varian Unity Inova 500 spectrometer that was shimmed to give a line width of ≤ 0.5 Hz for the most downfield peak of the double triplet due to the C-1 proton of [1-13C]-GA hydrate. Spectra (32 transients) were obtained using a sweep width of 6000 Hz, a pulse angle of 90°, and an acquisition time of 6 s. To ensure accurate integrals for the protons of interest, a relaxation delay of 120 s (>8T1) was used (19). Baselines were subjected to a first-order drift correction before determination of integrated peak areas. Chemical shifts are reported relative to HOD at 4.67 ppm.

Determination of Product Yields by 1H NMR

Figure 1 shows representative partial 1H NMR spectra at 500 MHz obtained during the reaction of [1-13C]-GA (20 mM) catalyzed by chicken muscle TIM in D2O in the presence of 5 mM phosphite dianion (A and B) and in its absence (C and D) at pD 7.0 and 25 °C (I = 0.1, NaCl). These reactions resulted in formation of the isotopomers of GA hydrate shown in Chart 2. Under our reaction conditions glycolaldehyde exists as 94% of the hydrate and only 6% of the free carbonyl form (19, 29). The integrated peak areas for the C-1 proton and for a single C-2 proton of the hydrates of the substrate [1-13C]-GA and of each of the product isotopomers were obtained using the following procedures.

Figure 1
Partial 1H NMR spectra at 500 MHz in the C-1 and C-2 regions glycolaldehyde hydrate obtained during the reaction of [1-13C]-GA catalyzed by chicken muscle TIM in D2O at pD 7.0 and 25 °C (I = 0.1, NaCl). The boxes indicate regions of the spectrum ...

[1-13C]-GA (Substrate)

The signal for the C-1 proton of [1-13C]-GA appears as a double triplet at 4.945 ppm (1JHC = 163 Hz, 3JHH = 5 Hz) (tall signals at edges of Figure 1A). The peak area for this proton was evaluated using eq 1, where a is the area of the most downfield peak of the double triplet. The signal for the C-2 protons of [1-13C]-GA appears as a double doublet at 3.410 ppm (2JHC = 3 Hz, 3JHH = 5 Hz) (offscale signal in the center of Figure 1B). The partly resolved small multiplet on the upfield edge of this signal (Figure 1B, inset) is the signal for the C-2 proton of [1-13C, 2-2H]-GA, the product of deuterium exchange (see below). The peak area for a single C-2 proton of [1-13C]-GA was evaluated using eq 2, where AT is the total area of the signals for the C-2 protons of both [1-13C]-GA and [1-13C, 2-2H]-GA, and 2e is the area of the small multiplet due to the C-2 proton of [1-13C, 2-2H]-GA (obtained as described below).

AC1=8a
(1)
AC2=(AT2e)/2
(2)

[2-13C]-GA (Product of Isomerization)

The signal for the C-1 proton of [2-13C]-GA appears as a broad triplet at 4.945 ppm (2JHC = 0.5 Hz, 3JHH = 5 Hz; the HC coupling was not always observed). However, only the most downfield peak of this signal is observable (Figure 1A, inset) because the upfield peaks are obscured by the broad doublet due to the C-1 proton of [2-13C, 2-2H]-GA, the product of isomerization with deuterium exchange (see below). The peak area for the C-1 proton of [2-13C]-GA was evaluated using eq 3, where b is the area of the most downfield peak of the broad triplet. The signal for the C-2 protons of [2-13C]-GA appears as a double doublet at 3.410 ppm (1JHC = 142 Hz, 3JHH = 5 Hz) (small signals at the edges of Figure 1B). The very small signals at the base of the peaks of this signal (Figure 1B, inset) are the 13C satellites of the very large double doublet for [1-13C]-GA arising from the presence of natural abundance 13C at C-2 of [1-13C]-GA. The peak area for a single C-2 proton of [2-13C]-GA was evaluated using eq 4, where c is the area of the most downfield peak of the double doublet (Figure 1B, inset). The values of AC1 and AC2 were corrected for the presence of a small amount (ca. 0.8 %) of [2-13C]-GA that was present at zero time as an impurity in the commercial sample of [1-13C]-GA.

AC1=4b
(3)
AC2=2c
(4)

[1-13C, 2-2H]-GA (Product of Deuterium Exchange)

The signal for the C-1 proton of [1-13C, 2-2H]-GA appears as a broad double doublet at 4.940 ppm shifted 0.005 ppm upfield of the double triplet due to the C-1 proton of [1-13C]-GA as a result of the β-deuterium (19). However, this signal is not resolved; it lies under the two most upfield peaks of each half of the double triplet due to the C-1 proton of [1-13C]-GA. Thus, for all intents and purposes, we will refer to the signal for the C-1 proton of [1-13C]-GA as an apparent double triplet (Figure 1A). The peak area for the C-1 proton of [1-13C, 2-2H]-GA was evaluated using eq 5, where AT is the area of downfield half of the apparent double triplet due to the C-1 proton of [1-13C]-GA, and a is the area of the most downfield peak of this apparent double triplet. This analysis is possible because a represents a signal that is due exclusively to the C-1 proton of [1-13C]-GA. The signal for the C-2 proton of [1-13C, 2-2H]-GA appears as a multiplet on the upfield edge of the signal for the C-2 protons of [1-13C]-GA and is expected to be a double double triplet as a result of a three-bond HH coupling (J = 5 Hz), a two-bond HC coupling (J = 3 Hz) and a two-bond HD coupling (J ≈ 2 Hz). The observed multiplet is composed of 8 peaks and its peak area was evaluated from the integrated area e of the 4 most upfield peaks of this signal using eq 6 (Figure 1B, inset).

AC1=2AT8a
(5)
AC2=2e
(6)

[2-13C, 2-2H]-GA (Product of Isomerization with Deuterium Exchange)

The signal for the C-1 proton of [2-13C, 2-2H]-GA appears as a broad doublet at 4.940 ppm (3JHH = 5 Hz, the three-bond HD and two-bond HC couplings are not resolved) (Figure 1A, inset). This doublet obscures the upfield peaks of the broad triplet due to the C-1 proton of [2-13C]-GA (see above). The peak area for the C-1 proton of [2-13C, 2-2H]-GA was evaluated using eq 7, where AT is the total area of the signals for the C-1 protons of both [2-13C, 2-2H]-GA and [2-13C]-GA, and b is the area of the most downfield peak of the broad triplet due to the C-1 proton of [2-13C]-GA. The signal for the C-2 proton of [2-13C, 2-2H]-GA appears as a double double triplet at 3.389 ppm (1JHC = 142 Hz, 3JHH = 5 Hz, 2JHD ≈ 2 Hz) shifted 0.021 ppm upfield of the double doublet due to the C-2 protons of [2-13C]-GA as a result of the α-deuterium (Figure 1B, inset) (19, 31-33). The peak area for the C-2 proton of [2-13C, 2-2H]-GA was evaluated using eq 8, where d is the area of the downfield half of the double double triplet (Figure 1B, inset).

AC1=AT4b
(7)
AC2=2d
(8)

The observed peak areas Aobs for these different protons at various reaction times were normalized according to eq 9 to give AP, where Astd and (Astd)o are the observed peak areas for an internal standard at time t and at t = 0, respectively. The internal standard was the upfield peak of the doublet at 6.7 ppm (J = 600 Hz) due to the P-H proton of phosphite, or the signal at 7.3 ppm due to the C-(4,5) protons of imidazole.

The disappearance of the substrate [1-13C]-GA was followed by monitoring the decrease in the peak area for the C-1 proton of [1-13C]-GA that was calculated using eq 1. The fraction of the substrate [1-13C]-GA remaining at time t, fS, was calculated using eq 10,where AS and (AS)o are the normalized peak areas for the C-1 proton of [1-13C]-GA at time t and t = 0, respectively (cf. eq 9). Observed first-order rate constants kobsd (s-1) for the disappearance of [1-13C]-GA were determined from the slopes of semi-logarithmic plots of reaction progress against time (eq 11) that covered the first 30% of the reaction.

The normalized yields of the individual products of the TIM-catalyzed reactions of [1-13C]-GA (Chart 2) were calculated using eq 12, were AP is the normalized peak area for a single C-1 or C-2 proton of the individual product calculated using eq 9, and ΣAP is the sum of the normalized peak areas for the same proton at all of the reaction products.

AP=Aobs((Astd)oAstd)
(9)
fS=AS(AS)o
(10)
InfS=kobsdt
(11)
fP=APAP
(12)

Results

TIM-Catalyzed Reactions of [1-13C]-GA Activated by Phosphite Dianion

Figures 1A and 1B show partial 1H NMR spectra at 500 MHz in the region of the C-1 and C-2 protons, respectively, obtained after the reaction of 23% of [1-13C]-GA (20 mM) catalyzed by chicken muscle TIM (7.9 μM) in D2O in the presence of 5 mM phosphite dianion at pD 7.0, 25 °C and I = 0.1 (NaCl). The disappearance of the signals due to the substrate [1-13C]-GA is accompanied by the appearance of signals due to the three isotopomeric products of the isomerization and deuterium exchange reactions shown in Scheme 4. Full details of the assignments of the signals for the C-1 and C-2 protons of the substrate and these products are given in the Materials and Methods. Normalized peak areas for a the C-1 proton and a single C-2 proton of the substrate and of the individual products were calculated using eq 9. There is good agreement between the decrease in the normalized peak area for a single C-2 proton of the substrate [1-13C]-GA and the increase in the sum of the normalized peak areas for a single C-2 proton of the products [2-13C]-GA, [2-13C, 2-2H]-GA and [1-13C, 2-2H]-GA. This shows that these are the only significant products of the TIM-catalyzed reaction of [1-13C]-GA activated by phosphite dianion.

The TIM-catalyzed incorporation of deuterium into [1-13C]-GA to give [1-13C, 2-2H]-GA and [2-13C, 2-2H]-GA results in a large upfield shift of 0.021 ppm of the signal for the remaining C-2 proton of these compounds as a result of the α-deuterium (19, 31-33). Thus, the signals for the C-2 protons of these products are well-resolved from those due to the C-2 protons of the substrate [1-13C]-GA and the isomerization product [2-13C]-GA (Figure 1B, peaks d and e). By contrast the signals for the C-1 protons of [1-13C, 2-2H]-GA and [2-13C, 2-2H]-GA are not resolved from those for [1-13C]-GA and [2-13C]-GA because there is only a small upfield shift of 0.005 ppm as a result of the β-deuterium (Figure 1A) (19). For example, the ratio of the areas of the three downfield peaks of the apparent double triplet due to the C-1 proton of [1-13C]-GA in Figure 1A is 1.00:2.18:1.17, rather than the expected 1.00:2.00:1:00. This is because the two most upfield peaks lie over those of the double doublet due to the C-1 proton of the deuterium exchange product [1-13C, 2-2H]-GA. The Materials and Methods section describes in detail the procedures that were used to calculate the relative yields of the three products of isomerization and deuterium exchange using either the poorly-resolved signals for the C-1 protons or the better-resolved signals for the C-2 protons. In all cases there was acceptable agreement between the yields of the reaction products determined using both approaches.

The fraction of the [1-13C]-GA substrate remaining, fs, and the fractional yields of each of the products, fP, were determined from the normalized peak areas for the much better-resolved C-2 protons of the reactants and products using eqs 10 and eq 12, respectively. Figure 2A shows the decrease with time in the fraction of remaining substrate, fs, during the reaction of [1-13C]-GA (20 mM) catalyzed by chicken muscle TIM (7.9 μM) in D2O in the presence of 5 mM phosphite dianion at pD 7.0, 25 ° and I = 0.1 (NaCl). Figure 2B shows the fractional yields of the products [2-13C]-GA, [2-13C, 2-2H]-GA and [1-13C, 2-2H]-GA, fP, determined during the first ca. 20% of this reaction. The initial normalized product yields, fE, obtained from short extrapolations of the observed yields to zero time, are reported in Table 1. The same procedure was used to determine the product yields for the TIM-catalyzed reactions of [1-13C]-GA activated by 10 mM and 20 mM phosphite dianion (Table 1).

Figure 2
Dependence of the fraction of remaining substrate and the normalized fractions of the products during the reaction of [1-13C]-GA (20 mM) catalyzed by chicken muscle TIM (7.9 μM) in the presence of 5 mM HPO32- in D2O at pD 7.0 and 25 °C ...
Table 1
Product Distributions for the Reaction of [1-13C]-Glycolaldehyde Catalyzed by Chicken Muscle Triosephosphate Isomerase in D2O in the Absence and Presence of Phosphite Dianion.a

First-order rate constants, kobsd (s-1), for the TIM-catalyzed reactions of [1-13C]-GA were determined by monitoring the disappearance of [1-13C]-GA by 1H NMR (eq 11). Table 1 reports the apparent second-order rate constants (kcat/Km)obsd (M-1 s-1) for the TIM-catalyzed reactions of the minor free carbonyl form of [1-13C]-GA that were calculated from the values of kobsd using eq 13, where fhyd = 0.94 is the fraction of GA present in the hydrated form (19). These apparent second-order rate constants for the reaction catalyzed by the chicken muscle enzyme are ca. 2-fold smaller than those determined earlier for the enzyme from rabbit muscle under the same reaction conditions (19).

(kcat/Km)obsd=kobsd(1fhyd)[TIM]
(13)

Unactivated TIM-Catalyzed Reaction of [1-13C]-GA in the Absence of Phosphite

In the absence of phosphite dianion we observe only a very sluggish chicken muscle TIM-catalyzed reaction of [1-13C]-GA in D2O at pD 7.0 and 25 °C. The value of (kcat/Km)obsd = 0.19 M-1 s-1 for this reaction (Table 1) is similar to (kcat/Km)obsd = 0.26 M-1 s-1 that we reported for the deuterium exchange reaction of unlabeled GA catalyzed by rabbit muscle TIM under the same reaction conditions (19).

Figures 1C and 1D show partial 1H NMR spectra at 500 MHz in the region of the C-1 and C-2 protons, respectively, obtained after the reaction of 29% of [1-13C]-GA (20 mM) catalyzed by chicken muscle TIM (340 μM) in D2O at pD 7.0, 25 °C and I = 0.1 (NaCl). The spectrum in the region of the C-2 protons (Figure 1D) shows that the unactivated TIM-catalyzed reaction results in formation of the same three isomerization and deuterium exchange products that are observed for the phosphite-activated reaction (Scheme 4). However, the total concentration of these products, calculated from the normalized peak areas of a single C-2 proton of each, is only ca. 50% of the calculated decrease in the concentration of the reactant [1-13C]-GA. We reported previously that there are slow nonspecific protein-catalyzed reactions of unlabeled GA (19) and of (R)-glyceraldehyde (10) that occur outside the active site of TIM. By analogy we conclude that the three products that are observed for the TIM-catalyzed reaction of [1-13C]-GA in D2O both in the presence and the absence of phosphite dianion result from the specific reaction of GA bound in the active site of TIM. However, slow nonspecific reaction(s) that occur outside the active site compete with the specific reaction and result in formation of other products.

There is good evidence that a major product of the nonspecific TIM-catalyzed reaction of [1-13C]-GA in D2O is [1-13C, 2,2-di-2H]-GA, resulting from deuterium exchange of both C-2 protons. After disappearance of 29% of [1-13C]-GA in the unactivated TIM-catalyzed reaction the ratio of the areas of the three downfield peaks of the apparent double triplet due to the C-1 proton is 1.00:2.10:1.36 (Figure 1C, inset). By contrast, after disappearance of 23% of [1-13C]-GA in the phosphite-activated TIM-catalyzed reaction this ratio is 1.00:2.18:1.17 (Figure 1A). The anomalous peak ratio observed for the unactivated TIM-catalyzed reaction can be attributed to a large shoulder on the downfield side of the most upfield peak of each half of the apparent double triplet (Figure 1C, inset). We assign this shoulder to the C-1 proton of [1-13C, 2,2-di-2H]-GA which lacks a C-2 proton. The signal for [2,2-di-2H]-GA formed during the rabbit muscle TIM-catalyzed reaction of GA in D2O is broad singlet shifted 0.009 ppm upfield of the triplet due to GA, and it appears as a larger shoulder on the downfield side of the most upfield peak of this triplet (19). The presence of this shoulder is reflected in a higher the yield of the deuterium-labeled product(s) of the unactivated TIM-catalyzed reactions of [1-13C]-GA calculated using the peak area for the C-1 proton (eq 7) than that calculated using the peak area of the C-2 proton (eq 8). These 1H NMR spectral data were used to calculate an absolute yield of 17% [1-13C, 2,2-di-2H]-GA which brings the total absolute yield of the products of the unactivated TIM-catalyzed reaction in the absence of phosphite dianion ([2-13C]-GA, [2-13C, 2-2H]-GA, [1-13C, 2-2H]-GA and [1-13C, 2, 2-2H]-GA) to 70%. We have not characterized the other product(s) that are formed under these reaction conditions.

The fractional yields of the products of the specific reaction of [1-13C]-GA in the active site of TIM, [2-13C]-GA, [2-13C, 2-2H]-GA and [1-13C, 2-2H]-GA, fP, were calculated using the normalized peak areas for the better-resolved C-2 protons of the reactants and products using eq 12, which normalizes the yields to a total of 100%. Table 1 gives the initial normalized product yields, fE, that were obtained from short extrapolations of the observed yields to zero time.

Discussion

We reported earlier that the reaction of unlabeled GA catalyzed by TIM in D2O is only partially inhibited by a saturating concentration of the competitive inhibitor 2-phosphoglycolate (19). We concluded that there is a nonspecific protein-catalyzed reaction of GA that occurs outside the active site of TIM at a rate similar to that of the specific reaction of GA within the active site. This “external” reaction does not make a significant contribution to either the observed rate or the product distribution for the TIM-catalyzed reaction of [1-13C]-GA in the presence of ≥ 5 mM phosphite dianion, because this phosphite-activated reaction is at least 60-fold faster than the unactivated reaction (Table 1) (19). The observation that the unactivated TIM-catalyzed isomerization of GA within the enzyme active site occurs at a similar rate to that of the competing reaction(s) of GA on the protein surface emphasizes the crippling effect of removal of the substrate phosphodianion group on enzyme activity. Comparison of the product distributions for the TIM-catalyzed reaction of [1-13C]-GA in the presence and absence of phosphite dianion provides considerable insight into the reaction mechanism.

Phosphite-Activated TIM-Catalyzed Reaction of [1-13C]-GA

Table 1 gives the yields of the phosphite-activated TIM-catalyzed reaction of [1-13C]-GA in D2O in the presence of several concentrations of phosphite dianion, determined by 1H NMR analysis (Figure 1). Figure 3 compares the product distribution for this reaction with those from our earlier work for the TIM-catalyzed reactions of the natural phosphorylated substrates GAP (22) and DHAP (23) in D2O. These data provide strong evidence that the TIM-catalyzed reactions of the phosphodianion substrates and the phosphite-activated reaction of [1-13C]-GA each occur within the confines of the active site of the loop-closed enzyme, where the substrate is sequestered from bulk solvent (22, 23, 34). However, there are differences between these product distributions that we cannot rationalize fully. One problem is that we are unable to rigorously define the pathways for the phosphite-activated reaction of [1-13C]-GA because this substrate can bind in the active site in two different configurations relative to the co-bound phosphite dianion, resembling the binding of the whole substrate GAP or DHAP (Scheme 5).2 There is nothing known about the relative reactivity of [1-13C]-GA bound in these different configurations, but reaction in either configuration can lead to formation of the same three reaction products shown in Scheme 4 and Figure 3.

Figure 3
Product distributions for the turnover of carbon acid substrates by TIM in D2O at 25 °C. (A) Product distribution for the reaction of [1-13C]-GA catalyzed by chicken muscle TIM in the presence of 5 mM phosphite dianion at pD 7.0 (this work). (B) ...

The yield of the isomerization product [2-13C]-GA from the TIM-catalyzed reaction of [1-13C]-GA with intramolecular transfer of hydrogen (12%) is much smaller than the yield of the corresponding product of the TIM-catalyzed reaction of GAP (DHAP, 49%) but similar to the yield of the isomerization product of the TIM-catalyzed reaction of DHAP (GAP, 18%). We suggest: (a) that [1-13C]-GA reacts preferentially in the DHAP configuration (Scheme 5), and (b) that the greater conformational flexibility of the TIM-bound substrate pieces phosphite and [1-13C]-GA, compared with TIM-bound triose phosphates, favors exchange of the H-labeled carboxylate side-chain of Glu-165 with deuterium from solvent within the restricted confines of the active site at the loop-closed enzyme.

Figure 3 shows that the phosphite-activated TIM-catalyzed reaction of [1-13C]-GA in D2O gives a ca. 3-fold greater yield of the product of isomerization with deuterium incorporation ([2-13C, 2-2H]-GA, 64%) than of the product representing deuterium incorporation into the remaining substrate ([1-13C, 2-2H]-GA, 23%). A similar, but smaller, preference for incorporation of deuterium into the isomerization product, compared to incorporation into the remaining substrate, is observed for the reactions of both GAP (31% [1-2H]-DHAP and 21% [2-2H-GAP]) (22) and DHAP (43% [2-2H]-GAP and 40% [1-2H]-DHAP) (23). The H95Q mutation at TIM removes the imidazole side-chain that stabilizes the negative charge at O-1 and/or O-2 of the enediol(ate) phosphate intermediate for the wildtype enzyme (35-37) and results in a even more striking difference in the partitioning of the intermediates of the reactions of DHAP and GAP. The reaction of DHAP catalyzed by H95Q mutant yeast TIM in 3H2O results in the incorporation of tritium label exclusively into the product GAP, while the reaction of GAP results in the incorporation of tritium label exclusively into the product DHAP (37). A “criss-cross” pathway for hydron exchange between O-1 and O-2 of the enediol(ate) phosphate intermediate was suggested to explain these results (37-39). If the imidazole side chain of His-95 at wildtype TIM provides a second competing pathway for hydron exchange between O-1 and O-2 (23, 34) that allows for the exchange of deuterium from solvent into unreacted substrate shown in Figure 3, then the partial operation of this “criss-cross” pathway could explain the observed product distributions for the wildtype enzyme-catalyzed reactions of the both whole substrates GAP and DHAP and of the substrate pieces {[1-13C]-GA + HPO32-} (Figure 3).

Unactivated TIM-Catalyzed Reaction of [1-13C]-GA

We have not attempted to characterize all of the products of the slow nonspecific reaction of [1-13C]-GA catalyzed by TIM in D2O. Table 1 gives the yields of the unactivated TIM-catalyzed reaction of [1-13C]-GA in D2O in the absence of phosphite, determined by 1H NMR analysis, that were normalized to a total of 100%. There is a somewhat higher yield of [1-13C, 2-2H]-GA, and correspondingly lower yields of [2-13C]-GA and [2-13C, 2-2H]-GA, than for the activated reactions in the presence of phosphite dianion. These results are consistent with the formation of additional [1-13C, 2-2H]-GA (ca. 10% of the total of the three products shown in Scheme 4) from slow competing nonspecific protein-catalyzed exchange of the C-2 proton of [1-13C]-GA for deuterium from solvent. The ratio of the yields of the isomerization products [2-13C]-GA and [2-13C, 2-2H]-GA from the unactivated TIM-catalyzed reaction of [1-13C]-GA (0.10/0.60) is the same, within experimental error, as that for the phosphite-activated reaction (0.12/0.64). The formation of additional [1-13C, 2-2H]-GA in a nonspecific protein-catalyzed reaction by a independent pathway should result in the same proportional decrease in the yields of [2-13C]-GA and [2-13C, 2-2H]-GA but no change in the ratio of these yields, as is observed.

We also observe the formation of the doubly deuterated product [1-13C, 2,2-di-2H]-GA (17 % yield, based on total products calculated from the disappearance of [1-13C]-GA) which we propose is a major product of the nonspecific protein-catalyzed reaction of [1-13C]-GA. We suggest that this nonspecific reaction involves the formation of a Schiff's base(s) between the alkylamino side chain(s) of lysine at TIM and [1-13C]-GA to give a reactive iminium ion that undergoes rapid base-catalyzed deuterium exchange of the C-2 proton through an enamine intermediate (Scheme 6). We observe no lag in the formation of the doubly deuterated product [1-13C, 2,2-2H]-GA, which suggests that exchange of second deuterium at the Schiff's base adduct is significantly faster larger than its hydrolysis to regenerate the free enzyme and the singly-labeled [1-13C, 2-2H]-GA (Scheme 6). Model studies of the de-deuteration reaction of d6-acetone in H2O show that diamines act as bifunctional catalysts of hydron exchange and that formation of the iminium ion adduct results in multiple intramolecular amine-catalyzed hydron exchange reactions that are much faster than hydrolysis of the iminium ion to regenerate the ketone and diamine (40, 41).

Mechanism of Activation of TIM by Phosphite Dianion

X-ray crystallographic data show that the binding interactions between the flexible “phosphate gripper” loop (loop 6) at TIM and the phosphodianion group of bound DHAP (7, 42) or the intermediate analogs 2-phosphoglycolate (43) or phosphoglycolohydroxamate (44) draw loop 6 over the bound ligand to close the active site. Figure 4 shows this conversion of TIM from the major inactive loop-open form (Eo) to the rare active loop-closed form (Ec) (KC [double less-than sign] 1) (19). The observation of similar relative yields of the isomerization products [2-13C]-GA and [2-13C, 2-2H]-GA for the unactivated (0.10/0.60) and phosphite-activated (0.12/0.64) TIM-catalyzed reactions of [1-13C]-GA in D2O shows that the dianion activator affects the rate, but not the product distribution of these two TIM-catalyzed reactions at the enzyme active site. Our data are consistent with the conclusion that both reactions involve partitioning of the enzyme-bound enediol(ate) intermediate (X, Figure 4) at the same loop-closed enzyme. We therefore suggest that the reaction of [1-13C]-GA proceeds by the (conceptual) binding of the carbon acid substrate to this rare loop-closed enzyme form. In this model phosphite dianion activation then results from utilization of its intrinsic binding energy (21) specifically to stabilize the catalytically active closed enzyme form (Ec•HPO32-) and increase its concentration in water (Figure 4) (19).

Figure 4
Proposed mechanisms for the unactivated and phosphite-activated deprotonation of glycolaldehyde by TIM to give an2 enzyme-bound enediol(ate) intermediate X. The direct unactivated of GA, (kcat/Km)E, is slow because free TIM exists mainly in the inactive ...

Acknowledgments

This work was supported by Grant GM39754 from the National Institutes of Health.

Footnotes

1Abbreviations: TIM, triosephosphate isomerase; DHAP, dihydroxyacetone phosphate; GAP, (R)-glyceraldehyde 3-phosphate; GA, glycolaldehyde; GPDH, glycerol 3-phosphate dehydrogenase; BSA, bovine serum albumin; NADH, nicotinamide adenine dinucleotide, reduced form; NMR, nuclear magnetic resonance.

2The data in Table 1 show that TIM catalyzes the isomerization of [1-13C]-GA in the presence of phosphite dianion to give [2-13C]-GA, a reaction that exchanges the positions of the carbonyl and hydroxymethylene carbons of the substrate relative to the enzyme-bound phosphite dianion piece (Scheme 5). The principal of microscopic reversibility then requires that TIM also catalyze the isomerization of [1-13C]-GA cobound with phosphite dianion in either the reactant or the isomerized product configuration. We have also observed that TIM catalyzes the exchange of both the pro-R and the pro-S hydrogens of unlabeled GA for deuterium from solvent D2O (unpublished results). This reflects the differing stereoselectivities for deprotonation of the bound GA piece by TIM when it is bound in the GAP (abstraction of the pro-S hydrogen) and DHAP (abstraction of the pro-R hydrogen) configurations (Scheme 5).

References

1. Knowles JR, Albery WJ. Perfection in enzyme catalysis: the energetics of triosephosphate isomerase. Acc Chem Res. 1977;10:105–111.
2. Rieder SV, Rose IA. Mechanism of the triose phosphate isomerase reaction. J Biol Chem. 1959;234:1007–1010. [PubMed]
3. Shonk CE, Boxer GE. Enzyme Patterns in Human Tissues. I. Methods for the Determination of Glycolytic Enzymes. Cancer Res. 1964;24:709–721. [PubMed]
4. Gerlt JA, Gassman PG. Understanding the Rates of Certain Enzyme-Catalyzed Reactions: Proton Abstraction from Carbon Acids, Acyl-Transfer Reactions, and Displacement Reactions of Phosphodiesters. Biochemistry. 1993;32:11943–11952. [PubMed]
5. Amyes TL, Richard JP. Proton Transfer to and from Carbon in Model Systems. In: Hynes JT, Klinman JP, Limbach HH, Schowen RL, editors. Hydrogen-Transfer Reactions. Vol. 3. Wiley-VCH; Weinheim: 2007. pp. 949–973. Biological Aspects I-II.
6. Xiang J, Jung Jy, Sampson NS. Entropy effects on protein hinges: The reaction catalyzed by triosephosphate isomerase. Biochemistry. 2004;43:11436–11445. [PubMed]
7. Jogl G, Rozovsky S, McDermott AE, Tong L. Optimal alignment for enzymatic proton transfer: Structure of the Michaelis Complex of triosephosphate isomerase at 1.2 Å resolution. Proc Natl Acad Sci U S A. 2003;100:50–55. [PubMed]
8. Loria JP, Berlow RB, Watt ED. Characterization of enzyme motions by solution NMR relaxation dispersion. Acc Chem Res. 2008;41:214–221. [PubMed]
9. Blacklow SC, Raines RT, Lim WA, Zamore PD, Knowles JR. Triosephosphate isomerase catalysis is diffusion controlled. Biochemistry. 1988;27:1158–1165. [PubMed]
10. Amyes TL, O'Donoghue AC, Richard JP. Contribution of phosphate intrinsic binding energy to the enzymatic rate acceleration for triosephosphate isomerase. J Am Chem Soc. 2001;123:11325–11326. [PubMed]
11. Richard JP. Acid-base catalysis of the elimination and isomerization reactions of triose phosphates. J Am Chem Soc. 1984;106:4926–4936.
12. Pompliano DL, Peyman A, Knowles JR. Stabilization of a reaction intermediate as a catalytic device: definition of the functional role of the flexible loop in triosephosphate isomerase. Biochemistry. 1990;29:3186–3194. [PubMed]
13. Sampson NS, Knowles JR. Segmental movement: definition of the structural requirements for loop closure in catalysis by triosephosphate isomerase. Biochemistry. 1992;31:8482–8487. [PubMed]
14. Sampson NS, Knowles JR. Segmental motion in catalysis: investigation of a hydrogen bond critical for loop closure in the reaction of triosephosphate isomerase. Biochemistry. 1992;31:8488–8494. [PubMed]
15. Sun J, Sampson NS. Understanding Protein Lids: Kinetic Analysis of Active Hinge Mutants in Triosephosphate Isomerase. Biochemistry. 1999;38:11474–11481. [PubMed]
16. Xiang J, Sun J, Sampson NS. The Importance of Hinge Sequence for Loop Function and Catalytic Activity in the Reaction Catalyzed by Triosephosphate Isomerase. J Mol Biol. 2001;307:1103–1112. [PubMed]
17. Kursula I, Salin M, Sun J, Norledge BV, Haapalainen AM, Sampson NS, Wierenga RK. Understanding protein lids: structural analysis of active hinge mutants in triosephosphate isomerase. Protein Eng Des Sel. 2004;17:375–382. [PubMed]
18. Sun J, Sampson NS. Determination of the amino acid requirements for a protein hinge in triosephosphate isomerase. Protein Sci. 1998;7:1495–1505. [PubMed]
19. Amyes TL, Richard JP. Enzymatic catalysis of proton transfer at carbon: activation of triosephosphate isomerase by phosphite dianion. Biochemistry. 2007;46:5841–5854. [PMC free article] [PubMed]
20. Tsang WY, Amyes TL, Richard JP. A Substrate in Pieces: Allosteric Activation of Glycerol 3-Phosphate Dehydrogenase (NAD+) by Phosphite Dianion. Biochemistry. 2008;47:4575–4582. [PMC free article] [PubMed]
21. Jencks WP. Binding Energy, Specificity and Enzymic Catalysis: The Circe Effect. Adv Enzymol Relat Areas Mol Biol. 1975;43:219–410. [PubMed]
22. O'Donoghue AC, Amyes TL, Richard JP. Hydron Transfer Catalyzed by Triosephosphate Isomerase. Products of Isomerization of (R)-Glyceraldehyde 3-Phosphate in D2O. Biochemistry. 2005;44:2610–2621. [PubMed]
23. O'Donoghue AC, Amyes TL, Richard JP. Hydron Transfer Catalyzed by Triosephosphate Isomerase. Products of Isomerization of Dihydroxyacetone Phosphate in D2O. Biochemistry. 2005;44:2622–2631. [PubMed]
24. Albery WJ, Knowles JR. Free-energy profile for the reaction catalyzed by triosephosphate isomerase. Biochemistry. 1976;15:5627–5631. [PubMed]
25. Hermes JD, Parekh SM, Blacklow SC, Koster H, Knowles JR. A reliable method for random mutagenesis: the generation of mutant libraries using spiked oligodeoxyribonucleotide primers. Gene. 1989;84:143–151. [PubMed]
26. Straus D, Gilbert W. Chicken triosephosphate isomerase complements an Escherichia coli deficiency. Proc Natl Acad Sci U S A. 1985;82:2014–2018. [PubMed]
27. Putman SJ, Coulson AFW, Farley IRT, Riddleston B, Knowles JR. Specificity and kinetics of triose phosphate isomerase from chicken muscle. Biochem J. 1972;129:301–310. [PubMed]
28. Glasoe PK, Long FA. Use of glass electrodes to measure acidities in deuterium oxide. J Phys Chem. 1960;64:188–190.
29. Collins GCS, George WO. Nuclear Magnetic Resonance Spectra of Glycolaldehyde. J Chem Soc (B) 1971:1352–1355.
30. Plaut B, Knowles JR. pH-Dependence of the triose phosphate isomerase reaction. Biochem J. 1972;129:311–320. [PubMed]
31. Amyes TL, Richard JP. Generation and stability of a simple thiol ester enolate in aqueous solution. J Am Chem Soc. 1992;114:10297–10302.
32. Amyes TL, Richard JP. Determination of the pKa of ethyl acetate: Brønsted correlation for deprotonation of a simple oxygen ester in aqueous solution. J Am Chem Soc. 1996;118:3129–3141.
33. Richard JP, Williams G, O'Donoghue AC, Amyes TL. Formation and Stability of Enolates of Acetamide and Acetate Anion: An Eigen Plot for Proton Transfer at α-Carbonyl Carbon. J Am Chem Soc. 2002;124:2957–2968. [PubMed]
34. O'Donoghue AC, Amyes TL, Richard JP. Slow proton transfer from the hydrogen-labelled carboxylic acid side chain (Glu-165) of triosephosphate isomerase to imidazole buffer in D2O. Org Biomol Chem. 2008;6:391–396. [PubMed]
35. Komives EA, Chang LC, Lolis E, Tilton RF, Petsko GA, Knowles JR. Electrophilic catalysis in triosephosphate isomerase: the role of histidine-95. Biochemistry. 1991;30:3011–3019. [PubMed]
36. Lodi PJ, Knowles JR. Neutral imidazole is the electrophile in the reaction catalyzed by triosephosphate isomerase: structural origins and catalytic implications. Biochemistry. 1991;30:6948–6956. [PubMed]
37. Nickbarg EB, Davenport RC, Petsko GA, Knowles JR. Triosephosphate isomerase: removal of a putatively electrophilic histidine residue results in a subtle change in catalytic mechanism. Biochemistry. 1988;27:5948–5960. [PubMed]
38. Harris TK, Abeygunawardana C, Mildvan AS. NMR Studies of the Role of Hydrogen Bonding in the Mechanism of Triosephosphate Isomerase. Biochemistry. 1997;36:14661–14675. [PubMed]
39. Harris TK, Cole RN, Comer FI, Mildvan AS. Proton Transfer in the Mechanism of Triosephosphate Isomerase. Biochemistry. 1998;37:16828–16838. [PubMed]
40. Hine J. Bifunctional catalysis of α-hydrogen exchange of aldehydes and ketones. Acc Chem Res. 1978;11:1–7.
41. Hine J, Li W. Catalysis of α-hydrogen exchange. XIX. Bifunctional catalysis of the dedeuteration of acetone-d6 by conformationally constrained derivatives of N,N-dimethyl-1,3-propanediamine. J Am Chem Soc. 1976;98:3287–3294.
42. Alber T, Banner DW, Bloomer AC, Petsko GA, Phillips D, Rivers PS, Wilson IA. On the three-dimensional structure and catalytic mechanism of triose phosphate isomerase. Philos Trans R Soc London Ser B. 1981;293:159–171. [PubMed]
43. Lolis E, Petsko GA. Crystallographic analysis of the complex between triosephosphate isomerase and 2-phosphoglycolate at 2.5-Å resolution: implications for catalysis. Biochemistry. 1990;29:6619–6625. [PubMed]
44. Davenport RC, Bash PA, Seaton BA, Karplus M, Petsko GA, Ringe D. Structure of the triosephosphate isomerase-phosphoglycolohydroxamate complex: an analog of the intermediate on the reaction pathway. Biochemistry. 1991;30:5821–5826. [PubMed]