The roles of ADP2- and Mg2+ in control steps of phosphoglycerate kinase

1H-NMR measurements were made of solutions of yeast phosphoglycerate kinase containing the nucleotide, ADP, and Mg2+ in varying concentrations in order to investigate the affect that the metal ion has on the mode of ADP binding to the enzyme. A preliminary study of adenosine binding to phosphoglycerate kinase was made in order to be sure of the nature of the adenine site. From the change in chemical shifts of the 'basic patch' histidine resonances (His62, 167 and 170), the nucleotide C8-H, C2-H and C1'-H resonances and resonances 40 and 41 (assigned to Thr373 and Thr375 in the hydrophobic, i.e. catalytic, site), it is apparent that there are at least two ADP binding sites on the enzyme: one at the hydrophobic (catalytic) site and one at the electrostatic site. A comparison of the results for ADP and ATP reveals differences due to the differential binding of the phosphate groups. The presence of Mg2+ results in further differences being observed. The data suggest that the primary binding site of ADP, in the absence of Mg2+, involves electrostatic interactions between the diphosphate chain of the substrate and the 'basic patch' region of the N-terminal domain. In the presence of greater than or equal to 1:1 ratio of Mg2+/ADP, however, the primary binding site involves predominantly hydrophobic interactions between the adenosine moiety and the catalytic site, with secondary binding occurring at the electrostatic site. Addition of Mg2+, therefore, tends to reduce the affinity of the electrostatic site (presumably by competing for ADP). It is suggested that alpha-helix XII, including residues 372, 373 and 375, moves differentially on binding ADP, Mg ADP, ATP or Mg . ATP, consistent with Mg2+ assisting the transfer of the gamma-phosphate of ATP to 3-phosphoglycerate during catalysis.     

Detection of conformational changes in chloroplast coupling factor 1 by 8-anilino-1-naphthalene-sulphonate fluorescence changes

Chloroplast coupling factor 1 (CF1) contains a high-affinity binding site for 8-anilino-1-napthalene sulphonate (ANS,Kd = 5-6 microM). The binding of ANS to the enzyme is associated with a fluorescence enhancement and a blue-shift in the emission spectrum. ANS only slightly inhibits ATP hydrolysis by CF1. Adenine nucleotides and inorganic phosphate induce a fast ANS fluorescence quenching of about 50% which is due to a decrease in the affinity of the enzyme for ANS (Kd increases from 6 microM to 22 microM) and in the fluorescence quantum yield of the bound probe (by 33%) but not in the number of ANS sites (n = 1). Conversely, Mg and Ca ions induce a fluorescence enhancement of bound ANS. Inactivation of the enzyme enhances ANS fluorescence, eliminates the response to adenine nucleotides and inorganic phosphate but increases the response to divalent metals. The affinity of latent CF1 for ADP (Kd = 12 microM) is considerably higher than for ATP (Kd = 95 microM) in buffer containing EDTA. The Kd for inorganic phosphate is 140 microM. Mg increases the apparent affinity for ATP (Kd = 28 microM) but not for ADP or Pi. Binding of ATP to the tight-sites does not inhibit the ADP or Pi-induced fluorescence quenching but decreases the affinity for ADP (Kd = 34 microM) and for inorganic phosphate (Kd = 320 microM). These results suggest that the ADP and phosphate binding sites are different but not independent from the tight sites. Activation of a Mg-specific ATPase in CF1 by octyl glucoside decreases the affinity for ADP and inorganic phosphate by about threefold but increases the affinity for ATP. ATPase activation of CF1 also increases the Ki for ADP inhibition of ATP hydrolysis. ATPase activation also influences the ANS responses to Ca and Mg. Ca-ATPase activation increases the fluorescence enhancement and the apparent affinity for Ca whereas Mg-ATPase activation specifically increases the Mg-induced fluorescence enhancement. The fluorescence of CF1-bound ANS is enhanced by Dio-9 and quenched by phloridzin, quercetin, Nbf-Cl and FITC. Nbf-Cl and FITC completely inhibit the ADP-induced fluorescence quenching whereas Dio-9 inhibits the Mg-induced fluorescence enhancement. ANS does not relieve the quercetin or phloridzin inhibition of ATP hydrolysis indicating that these inhibitors do not compete with ANS for a common binding site. ANS may be used, therefore, as a sensitive probe to detect conformational changes in CF1 in response to activation or inactivation and to binding of substrates and of inhibitors.     

Adenine nucleotides affect the binding of 3-phosphoglycerate to pig muscle 3-phosphoglycerate kinase

Pig muscle 3-phosphoglycerate kinase was complexed with 1-anilino-8-naphthalenesulfonate (ANS) in order to monitor the binding of substrates to the enzyme. The enzyme-dye interaction did not influence the enzymic activity under the experimental conditions used. By measuring the substrate-dependent change in the fluorescence emission of ANS molecules tightly bound to the enzyme (Kd less than or equal to 0.05 mM), fluorimetric titrations were carried out in 0.1 M Tris/HCl buffer pH 7.5, containing 5 mM mercaptoethanol, at 20 degrees C. The dissociation constants obtained for the separate bindings of 3-phosphoglycerate, MgATP, 1,3-bisphosphoglycerate and MgADP were 0.03 +/- 0.01 mM, 0.15 +/- 0.10 mM, 0.00005 +/- 0.00001 mM and 0.15 +/- 0.10 mM respectively. binding of 3-phosphoglycerate is weakened when MgATP is also bound to the enzyme: the dissociation constant of 3-phosphoglycerate in this ternary complex (0.25 +/- 0.08 mM) is comparable to its Km value (0.38 +/- 0.10 mM). The same weakening can be observed in the non-productive ternary complexes where MgATP is replaced by MgADP (Kd = 0.20 +/- 0.10 mM) or AMP (Kd = 0.12 +/- 0.05 mM), whereas adenosine has no such effect. This indicates the importance of the negatively charged phosphate(s) of nucleotides in influencing the binding of 3-phosphoglycerate. In contrast to 3-phosphoglycerate, the binding of the substrate analogue, glycerol 3-phosphate is practically not affected by the presence of MgATP: the dissociation constant to the free enzyme (0.40 +/- 0.10 mM) is comparable to its inhibitory constant (0.70 +/- 0.20 mM). This finding and the similarity of the dissociation constant of glycerol 3-phosphate binding (0.40 +/- 0.10 mM) and the Km value of 3-phosphoglycerate (0.38 +/- 0.10 mM) suggest that, during the enzymic reaction, binding of 3-phosphoglycerate occurs probably without involvement of the carboxyl group.     

Investigating interdomain region mutants Phe194----Leu and Phe194----Trp of yeast phosphoglycerate kinase by 1H-NMR spectroscopy.

Site-directed mutagenesis has been used to produce two mutant forms of yeast phosphoglycerate kinase in which the interdomain residue Phe194 has been replaced by a leucine or tryptophan residue. Using 1H-NMR spectroscopy, it was found that the mutations at position 194 induce both local and long-range conformational changes in the protein. It was also found that 3-phosphoglycerate binding to the mutant proteins induces somewhat different conformational effects to those observed for wild-type phosphoglycerate kinase. The affinity of mutant Phe194----Trp for 3-phosphoglycerate was found by NMR studies to be unaffected, while the affinity of Phe194----Leu mutant is reduced by about threefold relative to the wild-type enzyme. The binding of ATP at the electrostatic site of the mutant proteins is also seen to be about three times weaker for the Phe194----Leu mutant when compared to wild-type or Phe194----Leu mutant. These results are discussed in the light of the kinetic studies on the mutants which show that for Phe194----Leu mutant the Km values for both 3-phosphoglycerate and ATP, as well as the Vmax, are decreased relative to the wild-type enzyme, while for mutant Phe194----Trp, the Km values for 3-phosphoglycerate and ATP are unaffected and the Vmax is decreased when compared to wild-type enzyme. Kinetic studies in the presence of sulphate reveal that the anion activation is greater for mutant Phe194----Trp and less for mutant Phe194----Leu, relative to that observed for wild-type phosphoglycerate kinase. The NMR data, taken together with the kinetic data, are consistent with the on and off rates of 3-phosphoglycerate being affected by the mutations at position 194. It is suggested that Phe194 is important for the mobility of the interdomain region and the relative movement of the 3-phosphoglycerate binding site which allows the optimum conformation for catalysis to be attained. Apparently Trp194 reduces the mobility of the interdomain region of the protein, while Leu194 increases it.     

Study of the binding of substrates and paramagnetic Mn2+ ions with phosphoglycerate kinase by saturating ESR spectra of a spin-labelled protein]

A single SH-group of phosphoglycerate kinase from yeast was modified by mercury-containing spin label. The saturation curves of ESR spectra of the spin-labeled enzyme were studied. The paramagnetic ions of Mn2+ bound to the centre of ion nonspecific binding or active centre in the complex with ATP can influence the saturation of the spin-labeled enzyme. The saturation curves of the ESR signal of the spin-labeled enzyme in the presence of paramagnetic complex of CrATP were studied. It has been demonstrated that the second nonspecific centre of ATP binding is located at the active site of the enzyme (3-phosphoglycerate binding centre).     

Steady-state kinetic properties of phosphoglycerate kinase of mung beans

Steady-state kinetics of the action of mung bean phosphoglycerate kinase have been investigated using 3-phosphoglycerate and ATP as substrates in the presence of Mg2+ ions. Keeping a constant and high Mg2+ concentration and varying the concentration of one of the substrates (ATP or 3-phosphoglycerates) at several fixed concentrations of the other substrate (3-phosphoglycerate or ATP), the Km values of Mg.ATP2- and 3-phosphoglycerate were found to be 0.42 and 0.60 mM, respectively. These values are independent of the concentration of the other substrate. A limiting value of Vmax, where the enzyme is saturated with both the substrates, was found to be 39.4 mumoles product formed per min per mg enzyme protein. This corresponds to a turnover number equal to 31.5 sec-1 (for molecular weight of the enzyme equal to 48,000). If [Mg2+] and [ATP4-] are held equal and varied together at several fixed concentrations of 3-phosphoglycerate, deviations from Michaelis-Menten kinetics (non-linear Lineweaver-Burk plots) are observed at lower values of ATP4- and Mg2+ (less than 0.1 mM), giving rise to apparent sigmoidicity in the rate versus [ATP4-] plots. It has been suggested that the real substrate for this enzyme is the Mg.ATP2- complex (and not free ATP4-). The complex dissociates at lower values of [Mg2+] and [ATP4-]. The resulting disproportionate decrease in the concentration of the complex brings about a steeper fall in the rate of reaction than is required by the Michaelis-Menten equation, giving rise to an apparent sigmoidicity.(ABSTRACT TRUNCATED AT 250 WORDS)     

Interaction between D-glyceraldehyde-3-phosphate dehydrogenase and 3-phosphoglycerate kinase labeled by fluorescein-5'-isothiocyanate: evidence that the dye participates in the interaction

An interaction of rabbit muscle D-glyceraldehyde-3-phosphate dehydrogenase and 3-phosphoglycerate kinase labeled with FITC was studied by following the changes in fluorescence intensity of the bound dye. The association between the two enzymes was found to be a rather slow process characterized by a second order rate constant of 1.1 +/- 0.2.10(3) M-1 s-1, the KD of the complex between apoenzymes being 3.2.10(-7) M. The stability of the complex increased upon increase of temperature and ionic strength of the medium, suggesting a hydrophobic character of association. The ligands which bind at the active centers of the two enzymes (NAD+, ATP, 3-phosphoglycerate) weakened the bienzyme association. Unlabeled 3-phosphoglycerate kinase was unable to displace the FITC-labeled enzyme from the complex. Taken together, the results indicate that interaction between D-glyceraldehyde-3-phosphate dehydrogenase and 3-phosphoglycerate kinase labeled by FITC is assisted by the dye, which may bind at nucleotide-binding sites of GPDH. No interaction was observed between the FITC-labeled 3-phosphoglycerate kinase and lactate dehydrogenase, which suggests that protein-protein interaction at specific "recognition" sites may be a prerequisite for the complex formation.     

Inhibition of pyruvate kinase of bovine adrenal cortex by ATP: the role of magnesium]

The effect of ATP on bovine adrenal cortex pyruvate kinase has been studied. ATP is a competitive inhibitor of the enzyme, the Ki being 3.2 mM. Based on the efficiency of tryptophan fluorescence quenching, it was concluded that the magnesium complex of ATP is a true enzyme inhibitor. The role of Mg2+ in the inhibition process consists in the formation of a bridge between the enzyme and ATP. The ATP-dependent mechanism of pyruvate kinase inhibition is a potential physiological regulator of the enzyme determining the lower threshold of its sensitivity in vivo.     

A proton-NMR study of a site-directed mutation (His388----Glu) in the interdomain region of yeast phosphoglycerate kinase. Implications for domain movement

Proton NMR has been used to study a site-directed mutant of yeast phosphoglycerate kinase in which the interdomain residue His388 has been replaced by a glutamine residue. Using 1H-NMR spectroscopy, it was found that 3-phosphoglycerate binding to the mutant protein induces different conformational effects to those observed for the wild-type enzyme. These differences are not only located at the 3-phosphoglycerate binding site but are also seen as long-range effects at the surface of the protein. Measurements of the Kd for 3-phosphoglycerate from the NMR experiments show that the mutant enzyme has a 30-times reduced affinity for this substrate as compared with the wild-type enzyme. These data are consistent with the suggestion that an aromatic residue at position 388 plays an important role in the proposed hinge-bending mechanism.     

The phosphoglycerate kinases from Trypanosoma brucei. A comparison of the glycosomal and the cytosolic isoenzymes and their sensitivity towards suramin

Trypanosoma brucei has two phosphoglycerate kinase (PGK) isoenzymes, one is particle-bound and localized in glycosomes while the other is present in the cytosol. The cytosolic isoenzyme (cPGK) was 900-fold purified from cultured procyclic trypanosomes by hydrophobic interaction chromatography on phenyl-Sepharose followed by affinity chromatography on 2',3'-ATP-Sepharose and had a specific activity of 275 units/mg protein. cPGK was compared with the purified glycosomal isoenzyme (gPGK) from bloodstream-form trypanosomes as well as with the commercially available PGKs from yeast, rabbit muscle and Spirulina platensis, a blue-green alga. Like all other PGKs, cPGK was a monomeric protein with a molecular mass of approximately 45 kDa similar to that of the PGKs from other organisms but 2 kDa smaller than that of gPGK. Despite this difference in length and a great difference in isoelectric point, the two trypanosome isoenzymes strongly resembled each other in several respects. The kinetic parameters did not differ significantly from each other or from the PGKs of other organisms. Both trypanosome enzymes resembled the enzyme from S. platensis in that they had an almost absolute requirement for ATP, contrary to the enzymes from yeast and rabbit muscle, which were capable of utilizing GTP and ITP also. This difference in substrate specificity may be related to the amino acid substitutions, Trp 308----His and Ala 306----Glu in the adenine-binding site, which are only found in the two Trypanosoma isoenzymes. Kinetic analysis showed that these substitutions do not prevent binding of the ATP analogues, but probably prevent phosphoryl-group transfer. Both isoenzymes displayed an activity optimum at pH 6.0-9.0 similar to that for the enzyme of yeast. Both gPGK and cPGK were inhibited by the trypanocidal drug Suramin. This inhibition could be described as competitive both with ATP and 3-phosphoglycerate with two inhibitor molecules binding to one molecule of enzyme. The gPGK, however, was much more sensitive (Ki app. = 8.0 microM) to Suramin than either the cPGK (Ki app. = 20 microM) or the enzymes from rabbit muscle (Ki app. = 55 microM), yeast (Ki app. = 167 microM) or S. platensis (Ki app. = 250 microM). It is suggested that positive charges on the enzyme's surface may play an important role in the potentiation of the binding of the negatively charged Suramin molecule.     

Inhibition of glycolytic enzymes by 2-phosphotartronate.

2-Phosphotartronate has been synthesized by permanganate oxidation of glycerol 2-phosphate and has been tested as an inhibitor of five glycolytic enzymes that bind phosphoglycerate or phosphoglycollate. Competitive inhibition of rabbit muscle phosphoglycerate mutase, enolase and pyruvate kinase was observed. Triose phosphate isomerase and 3-phosphoglycerate kinase were not inhibited.     

Activation of pyridoxal kinase by metallothionein

Brain pyridoxal kinase, which uses ATP complexed to either Zn(II) or Co(II) as substrates, displays high catalytic activity in the presence of Zn-thionein and Co-thionein. Several steps intervene in the process of pyridoxal kinase activation, i.e., binding of Zn ions to ATP and interaction between Zn-ATP and the enzyme. Equilibrium binding studies show that ATP mediates the release of Zn ions from the metal-thiolate clusters of the thioneins, whereas spectroscopic measurements conducted on Co-thionein reveal that the absorption transitions corresponding to the metal-thiolate of the protein are perturbed by ATP. The binding Zn-ATP to the kinase proceeds with a delta G = -6.3 kcal/mol as demonstrated by fluorometric titrations. Direct interaction between the kinase and derivatized-metallothionein could not be detected by emission anisotropy measurements, indicating that juxtaposition of the proteins does not influence the exchange of metal ions. Since the concentration of free Zn in several mammalian tissues is lower than 1 nM, it is postulated that under in vivo conditions the concentration of metallothionein regulates the catalytic activity of pyridoxal kinase.     

Yeast 3-phosphoglycerate kinase. Essential arginyl residues at the 3-phosphoglycerate binding site.

Yeast 3-phosphoglycerate kinase (ATP:3-phospho-D-glycerate 1-phospho-transferase, EC 2.7.2.3) is inactivated by phenylglyoxal. Loss of activity correlates with the modification of two arginyl residues, both of which are protected by all of the substrates. The modification is not accompanied by any significant conformational change as determined by optical rotatory dispersion. Ultraviolet difference spectrophotometry indicates that the inactivated enzyme retains its capacity for binding the nucleotide substrates whereas the spectral perturbation characteristic of 3-phosphoglycerate binding is abolished in the modified enzyme. The data suggest that at least one of the two essential arginyl residues is located at or near the 3-phosphoglycerate binding site. A likely role of this residue could be its interaction with the negatively charged phosphate or carboxylate groups of 3-phosphoglycerate.     

An investigation of large inhibitors binding to phosphoglycerate kinase and their effect on anion activation.

This study extends, to a series of larger anions, our earlier investigation of the interaction of the trypanocidal drug suramin and other small negatively charged molecules with yeast phosphoglycerate kinase. 1H-NMR structural studies of phosphoglycerate kinase in the presence of varying concentrations of these large molecules (designed to mimic, at one end, the anionic charge distribution in the substrate 3-phosphoglycerate, while possibly being able to interact across the cleft of the enzyme) including inositol 1,4,5-triphosphate, 4-amino-6-trichloroethenyl-1,3- benzenedisulphonamide, gallic acid and sulphasalazine are described. The anion activation and/or inhibition of the enzyme by these molecules are also reported. Evidence that binding to the general anion site in the 'basic patch' region of the protein may be responsible for either the activating or inhibiting effects, while binding at the hydrophobic (catalytic) site leads to inhibition only is presented. A reaction scheme which explains these observations is given.     

Characterization of the binding of the fluorescent ATP analog TNP-ATP to insulysin

It has recently been reported that insulin-degrading enzyme (IDE) contains an allosteric site which binds polyanions such as ATP and PPPi. This site is distinct from the catalytic site where homotrophic allosteric effects are produced. In this study, we have characterized the binding of ATP to this anion binding site using the fluorescent ATP analog 2',3'-O-(2,4,6-trinitrophenyl)-adenosine triphosphate (TNP-ATP), which exhibits a higher affinity to the enzyme than ATP itself. TNP-ATP binding to IDE was accompanied by a more than 4-fold increase in fluorescence. The dissociation constant (K(D)) of TNP-ATP was determined as 1.15 microM, while the activation constant (K(A)) was determined to be 1.6 microM. Competition experiments were used to show that ATP (Ki = 1.3 mM) and PPPi (Ki = 0.9mM) bind with a higher affinity than ADP (2.2 mM) and AMP (4.0 mM). Adenosine did not bind to the anion binding site.     

Frequency domain fluorescence studies of yeast phosphoglycerate kinase and its ternary complex

A frequency domain fluorescence study of yeast phosphoglycerate kinase has been performed to observe the effect of substrates on the structure and dynamics of the enzyme. At 20 degrees C and pH 7.2, a biexponential decay is observed for tryptophanyl emission. The short fluorescence lifetime (0.4 ns) component is associated with a spectrum having a 329-nm maximum and a 18.4-kJ/mol activation energy, Ea, for thermal quenching. The long-lifetime (3.5 ns) component has a 338-nm maximum and an Ea of only 7.9 kJ/mol. Tentatively we assign the short and long-lifetime components to Trp-333 and Trp-308. Binding of the substrates ATP and 3-phosphoglycerate leads to a significant increase in the fluorescence lifetime, the red shift of the emission spectrum and in the decrease in the Ea for both components. Acrylamide-quenching studies indicate that the two tryptophan residues have about the same degree of kinetic exposure to the quencher and that the binding of the substrates causes a very slight change in the quenching pattern. These fluorescence studies indicate that the binding of the substrates to phosphoglycerate kinase may influence the conformational dynamics around the two tryptophan residues located on one of the protein's domains.     

Competitive binding of ATP and the fluorescent substrate analogue 2',3'-O-(2,4,6-trinitrophenylcyclohexadienylidine) adenosine 5'-triphosphate to the gastric H+,K+-ATPase: evidence for two classes of nucleotide sites

ATP and the fluorescent substrate analogue TNP-ATP bind competitively to the gastric H,-K-ATPase. Substrate and product completely reverse the fluorescence enhancement caused by TNP-ATP binding to the enzyme. The fluorophore is displaced monophasically from apoenzyme. However, ATP displaces TNP-ATP from the Mg2+-quenched state in two steps of equal amplitude. The midpoints of the titrations differ by more than 2 orders of magnitude. The estimated substrate constants are in reasonable agreement with published Michaelis constants. TNP-ATP is not a substrate for the H,K-ATPase. The fluorophore prevents phosphorylation by ATP and competitively inhibits the K+-stimulated pNPPase and ATPase activities of the enzyme. Ki is approximately the same for both hydrolytic activities and consistent with the Kd of TNP-ATP measured directly. Km for pNPP is 1.48 +/- 0.15 mM. Two Michaelis constants are required to fit the ATPase data: Km1 = 0.10 +/- 0.01 mM and Km2 = 0.26 +/- 0.05 mM.     

Effect of Cibacron Blue F3GA on phosphoglycerate kinase of Lactobacillus plantarum and phosphoglycerate mutase of Leuconostoc dextranicum

Blue dextran or Cibacron Blue F3GA has been shown to inhibit yeast phosphoglycerate kinase [EC 2.7.2.3] competitively with respect to ATP (Thompson et al. (1975) Proc. Natl. Acad. Sci. U.S. 72, 663--667; Beissner and Rudolph (1979) J. Biol. Chem. 254, 6273--6277). However, we have found that phosphoglycerate kinase of Lactobacillus plantarum was inhibited by Cibacron Blue F3GA, the blue chromophore of blue dextran, noncompetitively with respect to ATP, but competitively with respect to 3-phosphoglycerate. Further inhibition studies with Cibacron Blue F3GA suggest that one molecule of the dye was bound per molecule of phosphoglycerate kinase at a saturated level of either substrate, but two molecules of the dye were bound per molecule of the kinase with an unsaturated level of either substrate used as a fixed substrate. Furthermore, phosphoglycerate mutase [EC 2.7.5.3] of Leuconostoc dextranicum was also inhibited by Cibacron Blue F3GA competitively with respect to 3-phosphoglycerate and noncompetitively with respect to 2,3-bisphosphoglycerate. These results suggest that the 3-phosphoglycerate-binding site on both phosphoglycerate kinase and phosphoglycerate mutase can interact with Cibacron Blue F3GA.     

Effect of adenosine on phosphofructokinase and phosphoglycerate kinase of rat liver]

Effects of adenosine on purified rat liver phosphofructokinase and phosphoglycerate kinase activity were investigated in vitro. Stimulation by adenosine of the phosphofructokinase has been observed at low concentrations, but the activity was markedly inhibited at high concentrations. Adenosine was an inhibitor of the phosphoglycerate kinase : Lineweaver-Burk analysis indicated that adenosine inhibition was competitive with ATP and non competitive with 3-phosphoglycerate. An interpretation of these results is proposed.     

Binding of substrates and other anions to yeast phosphoglycerate kinase.

1. The binding of all four substrates to yeast phosphoglycerate kinase has been studied using a gel filtration technique. The binding of phosphate and sulphate anions has also been investigated. 2. Two sites for each adenine nucleotide were found, one site being weaker than the other by between 30 and 50-fold. Only one binding site for the phosphoglycerate substrates was found. 3. 1,3-Bisphosphoglycerate (1,3-P2-glycerate) bound to the enzyme approximately 1000 times tighter than the other three substrates, its dissociation constant being 0.06 micrometer at ionic strength 0.15 M. 4. Sulphate and phosphate were mutually competitive and sulphate competed with the binding of all substrates except MgADP. MgADP bound to the enzyme more weakly in the presence of sulphate. The dissociation constant for sulphate binding was 1.6 mM at ionic strength of 0.15 M, and 0.05 mM at ionic strength 0.015 M. 5. These results are consistent with sulphate acting as a competitive inhibitor, as found by kinetic studies at high sulphate concentrations. The activatory effect of sulphate at lower concentrations and the substrate activation phenomea displayed by this enzyme, are interpreted in terms of a two-step dissociation of 1, 3-P2-glycerate. The presence of moderate concentrations of MgATP, 3-phosphoglycerate or sulphate causes acceleration of the rate of dissociation of the product, 1, 3-P2-glycerate, this being the rate-limiting step in the overall enzyme reaction.