18O-exchange evidence that mutations of arginine in a signature sequence for P-type pumps affect inorganic phosphate binding.

We have proposed a model for part of the catalytic site of P-type pumps in which arginine in a signature sequence functions like lysine in P-loop-containing enzymes that catalyze adenosine 5'-triphosphate hydrolysis [Smirnova, I. N., Kasho, V. N., and Faller, L. D. (1998) FEBS Lett. 431, 309-314]. The model originated with evidence from site-directed mutagenesis that aspartic acid in the DPPR sequence of Na,K-ATPase binds Mg(2+) [Farley, R. A., et al. (1997) Biochemistry 36, 941-951]. It was developed by assuming that the catalytic domain of P-type pumps evolved from enzymes that catalyze phosphoryl group transfer. The functions of the positively charged amino group in P-loops are to bind substrate and to facilitate nucleophilic attack upon phosphorus by polarizing the gamma-phosphorus-oxygen bond. To test the prediction that the positively charged guanidinium group of R596 in human alpha(1) Na,K-ATPase participates in phosphoryl group transfer, the charge was progressively decreased by site-directed mutagenesis. Mutants R596K, -Q, -T, -M, -A, -G, and -E were expressed in yeast membranes, and their ability to catalyze phosphorylation with inorganic phosphate was evaluated by following (18)O exchange. R596K, in which the positive charge is retained, resembled the wild type. Substitution of a negative charge (R596E) resulted in complete loss of activity. The remaining mutants with uncharged side chains had both lowered affinity for inorganic phosphate and altered phosphate isotopomer distributions, consistent with increased phosphate-off rate constants compared to that of the wild type. Therefore, mutations of R596 strengthen our hypothesis that the oppositely charged side chains of the DPPR peptide in Na,K-ATPase form a quaternary complex with magnesium phosphate.

The electrophilic and leaving group phosphates in the catalytic mechanism of yeast pyrophosphatase.

Binding of pyrophosphate or two phosphate molecules to the pyrophosphatase (PPase) active site occurs at two subsites, P1 and P2. Mutations at P2 subsite residues (Y93F and K56R) caused a much greater decrease in phosphate binding affinity of yeast PPase in the presence of Mn(2+) or Co(2+) than mutations at P1 subsite residues (R78K and K193R). Phosphate binding was estimated in these experiments from the inhibition of ATP hydrolysis at a sub-K(m) concentration of ATP. Tight phosphate binding required four Mn(2+) ions/active site. These data identify P2 as the high affinity subsite and P1 as the low affinity subsite, the difference in the affinities being at least 250-fold. The time course of five "isotopomers" of phosphate that have from zero to four (18)O during [(18)O]P(i)-[(16)O]H(2)O oxygen exchange indicated that the phosphate containing added water is released after the leaving group phosphate during pyrophosphate hydrolysis. These findings provide support for the structure-based mechanism in which pyrophosphate hydrolysis involves water attack on the phosphorus atom located at the P2 subsite of PPase.

Probing essential water in yeast pyrophosphatase by directed mutagenesis and fluoride inhibition measurements.

The pattern of yeast pyrophosphatase (Y-PPase) inhibition by fluoride suggests that it replaces active site Mg(2+)-bound nucleophilic water, for which two different locations were proposed previously. To localize the bound fluoride, we investigate here the effects of mutating Tyr(93) and five dicarboxylic amino acid residues forming two metal binding sites in Y-PPase on its inhibition by fluoride and its five catalytic functions (steady-state PP(i) hydrolysis and synthesis, formation of enzyme-bound PP(i) at equilibrium, phosphate-water oxygen exchange, and Mg(2+) binding). D117E substitution had the largest effect on fluoride binding and made the P-O bond cleavage step rate-limiting in the catalytic cycle, consistent with the mechanism in which the nucleophile is coordinated by two metal ions and Asp(117). The effects of the mutations on PP(i) hydrolysis (as characterized by the catalytic constant and the net rate constant for P-O bond cleavage) were in general larger than on PP(i) synthesis (as characterized by the net rate constant for PP(i) release from active site). The effects of fluoride on the Y-PPase variants confirmed that PPase catalysis involves two enzyme.PP(i) intermediates, which bind fluoride with greatly different rates (Baykov, A. A., Fabrichniy, I. P., Pohjanjoki, P., Zyryanov, A. B., and Lahti, R. (2000) Biochemistry 39, 11939-11947). A mechanism for the structural changes underlying the interconversion of the enzyme.PP(i) intermediates is proposed.

Catalytically important ionizations along the reaction pathway of yeast pyrophosphatase.

Five catalytic functions of yeast inorganic pyrophosphatase were measured over wide pH ranges: steady-state PP(i) hydrolysis (pH 4. 8-10) and synthesis (6.3-9.3), phosphate-water oxygen exchange (pH 4. 8-9.3), equilibrium formation of enzyme-bound PP(i) (pH 4.8-9.3), and Mg(2+) binding (pH 5.5-9.3). These data confirmed that enzyme-PP(i) intermediate undergoes isomerization in the reaction cycle and allowed estimation of the microscopic rate constant for chemical bond breakage and the macroscopic rate constant for PP(i) release. The isomerization was found to decrease the pK(a) of the essential group in the enzyme-PP(i) intermediate, presumably nucleophilic water, from >7 to 5.85. Protonation of the isomerized enzyme-PP(i) intermediate decelerates PP(i) hydrolysis but accelerates PP(i) release by affecting the back isomerization. The binding of two Mg(2+) ions to free enzyme requires about five basic groups with a mean pK(a) of 6.3. An acidic group with a pK(a) approximately 9 is modulatory in PP(i) hydrolysis and metal ion binding, suggesting that this group maintains overall enzyme structure rather than being directly involved in catalysis.

Functional characterization of Escherichia coli inorganic pyrophosphatase in zwitterionic buffers.

Catalysis by Escherichia coli inorganic pyrophosphatase (E-PPase) was found to be strongly modulated by Tris and similar aminoalcoholic buffers used in previous studies of this enzyme. By measuring ligand-binding and catalytic properties of E-PPase in zwitterionic buffers, we found that the previous data markedly underestimate Mg(2+)-binding affinity for two of the three sites present in E-PPase (3.5- to 16-fold) and the rate constant for substrate (dimagnesium pyrophosphate) binding to monomagnesium enzyme (20- to 40-fold). By contrast, Mg(2+)-binding and substrate conversion in the enzyme-substrate complex are unaffected by buffer. These data indicate that E-PPase requires in total only three Mg2+ ions per active site for best performance, rather than four, as previously believed. As measured by equilibrium dialysis, Mg2+ binds to 2.5 sites per monomer, supporting the notion that one of the tightly binding sites is located at the trimer-trimer interface. Mg2+ binding to the subunit interface site results in increased hexamer stability with only minor consequences for catalytic activity measured in the zwitterionic buffers, whereas Mg2+ binding to this site accelerates substrate binding up to 16-fold in the presence of Tris. Structural considerations favor the notion that the aminoalcohols bind to the E-PPase active site.

Inferences about the catalytic domain of P-type ATPases from the tertiary structures of enzymes that catalyze the same elementary reaction.

The machinery to catalyze elementary reactions is conserved, and the number of solved enzyme structures is increasing exponentially. Therefore, structures of enzymes that catalyze phosphate transfer are reviewed, and a supersecondary structure connecting the Walker A sequence to another sequence containing functional amino acids is proposed as an additional signature for the active site. The new signature is used to infer the identity of the P-loop in P-type biological pumps and may be useful in predicting targets for site-directed mutagenesis in other enzymes of unknown structure like the AAA family and ABC transporters.

Trimeric inorganic pyrophosphatase of Escherichia coli obtained by directed mutagenesis.

Escherichia coli inorganic pyrophosphatase is a tight hexamer of identical subunits. Replacement of both His136 and His140 by Gln in the subunit interface results in an enzyme which is trimeric up to 26 mg/mL enzyme concentration in the presence of Mg2+, allowing direct measurements of Mg2+ binding to trimer by equilibrium dialysis. The results of such measurements, together with the results of activity measurements as a function of [Mg2+] and pH, indicate that Mg2+ binds more weakly to one of the three sites per monomer than it does to the equivalent site in the hexamer, suggesting this site to be located in the trimer:trimer interface. The otherwise unobtainable hexameric variant enzyme readily forms in the presence of magnesium phosphate, the product of the pyrophosphatase reaction, but rapidly dissociates on dilution into medium lacking magnesium phosphate or pyrophosphate. The kcat values are similar for the variant trimer and hexamer, but Km values are 3 orders of magnitude lower for the hexamer. Thus, while stabilizing hexamer, the two His residues, per se, are not absolutely required for active-site structure rearrangement upon hexamer formation. The reciprocal effect of hexamerization and product binding to the active site is explained by destabilization of alpha-helix A, contributing both to the active site and the subunit interface.

A proposal for the Mg2+ binding site of P-type ion motive ATPases and the mechanism of phosphoryl group transfer.

Mutations of D586 in the DPPR sequence of sodium pump decrease the enzyme's affinity for inorganic phosphate [Farley R. A., Heart, E., Kabalin, M., Putnam, D., Wang, K., Kasho, V. N., and Faller, L. D. (1997) Biochemistry 36, 941-951]. Therefore, it was proposed that D586 coordinates the Mg2+ required for catalytic activity. This hypothesis is tested (1) by determining the substrate for catalysis of 18O exchange between inorganic phosphate and water and (2) by comparing conserved amino acid sequences in P-type pumps with the primary structures of enzymes of known tertiary structure that catalyze phosphoryl group transfer. From the isotope exchange data, it is concluded that the Mg2+-dependent and Na+- and K+-stimulated ATPase binds Mg2+ before inorganic phosphate. Sequence homology is demonstrated between the conserved DPPR and MV(I,L)TGD sequences of P-type pumps and two conserved adenylate kinase sequences that coordinate Mg2+ and/or bind nucleotide in the crystal structure of the yeast enzyme. A model for the Mg2+ site of P-type pumps and the mechanism of phosphoryl group transfer is proposed and tested by demonstrating that the conserved sequences are also structurally homologous.

Structural and functional consequences of substitutions at the tyrosine 55-lysine 104 hydrogen bond in Escherichia coli inorganic pyrophosphatase.

Tyrosine 55 and lysine 104 are evolutionarily conserved residues that form a hydrogen bond in the active site of Escherichia coli inorganic pyrophosphatase (E-PPase). Here we used site-directed mutagenesis to examine their roles in structure stabilization and catalysis. Though these residues are not part of the subunit interface, Y55F and K104R (but not K104I) substitutions markedly destabilize the hexameric structure, allowing dissociation into active trimers on dilution. A K104I variant is nearly inactive while Y55F and K104R variants exhibit appreciable activity and require greater concentrations of Mg2+ and higher pH for maximal activity. The effects on activity are explained by (a) increased pK(a)s for the catalytically essential base and acid at the active site, (b) decreases in the rate constant for substrate (dimagnesium pyrophosphate) binding to enzyme-Mg2 complex vs enzyme-Mg3 complex, and (c) parallel decreases in the catalytic constant for the resulting enzyme-Mg2-substrate and enzyme-Mg3-substrate complexes. The results are consistent with the major structural roles of Tyr55 and Lys104 in the active site. The microscopic rate constant for PPi hydrolysis on either the Y55F or K104R variants increases, by a factor of 3-4 in the pH range 7.2-8.0, supporting the hypothesis that this reaction step depends on an essential base within the enzyme active site.

Site-directed mutagenesis of the sodium pump: analysis of mutations to amino acids in the proposed nucleotide binding site by stable oxygen isotope exchange.

A model for the active site of P type ATPases has been tested by site-directed mutagenesis of amino acids in two conserved sequences of Mg(2+)-dependent and Na(+)- and K(+)-stimulated ATPase. The mutants K501R, K501E, D586E, D586N, P587A, and P588A were expressed in yeast cells and compared with wild type. In addition to previously published assays of adenosine 5'-triphosphate binding and hydrolysis, measurements of 18O exchange between Pi and water have been used to identify steps in the E2 half of the reaction cycle affected by the mutations. The study supports the prediction that K501 in the KGAP sequence interacts with adenosine 5'-triphosphate. However, quantitative comparisons of the effect of mutation K501E on the activity with the effects of mutations to an enzyme of known structure that also catalyzes phosphoryl group transfer make a direct role for the positive charge on the side chain of K501 in catalysis by stabilizing the transition state unlikely. No evidence for the predicted interaction between D586 and the hydroxyl groups of ribose was found. However, the data do indicate that the spatial organization of the loop containing the DPPR sequence is critical for phosphorylation of the enzyme. A role for D586 in coordinating the Mg2+ that is required for activity is proposed.

Feasibility of analysing [13C]urea breath tests for Helicobacter pylori by gas chromatography-mass spectrometry in the selected ion monitoring mode.

BACKGROUND: The [13C]urea breath test for Helicobacter pylori is nonradioactive, as well as noninvasive, but few clinical laboratories have the expensive isotope ratio mass spectrometer used for analysis. METHODS: To demonstrate the feasibility of analysing [13C]urea breath tests with a gas chromatograph-mass spectrometer routinely used for drug testing, 13CO2 standards for breath tests and breath samples from patients in a multiple-blind study were analysed. The breath samples were also analysed by isotope ratio mass spectrometry, and the diagnoses were compared with biopsy results. RESULTS: The precision of the enrichment measurements by gas chromatography-mass spectrometry was 1.1 parts per thousand, and the calculated differences in enrichment between standard gases equaled the certified values. The sensitivity (94%), specificity (94%), and percentage agreement (94%) for diagnosis of Helicobacter pylori (n = 34) were as high or higher than for analysis of replicate breath samples by isotope ratio mass spectrometry and comparable to the values reported for diagnosis of the bacterium by other currently accepted tests. CONCLUSIONS: The study demonstrates that a gas chromatograph-mass spectrometer can be used to analyse [13C]urea breath tests, thus potentially lowering the cost of the test and increasing the number of laboratories that can perform the test.

Catalysis by Escherichia coli inorganic pyrophosphatase: pH and Mg2+ dependence.

Steady-state rates of PPi hydrolysis by Escherichia coli inorganic pyrophosphatase (E-PPase) were measured as a function of magnesium pyrophosphatase (substrate) and free Mg2+ ion (activator) in the pH range 6.0-10.0. Computer fitting of hydrolysis data in combination with direct measures of Mg2+ binding to enzyme has resulted in a model that quantitatively accounts for our results. The major features of this model are the following: (a) E-PPase catalysis proceeds both with three and with four (and possibly with five) Mg2+ ions per active site; (b) catalysis requires both an essential base and an essential acid, and the pKas of these groups are modulated by the stoichiometry of bound Mg2+; and (c) the four-metal route predominates for concentrations of free Mg2+>0.2mM. The model straightforwardly accounts for the apparent linkage between increased pKa of an essential base and activity requirements for higher Mg2+ concentration observed for several active site variants. Microscopic rate constants for overall catalysis of PPi-Pi equilibration were determined at pH 6.5-9.3 by combined analysis of enzyme-bound PPi formation and rates of PPi hydrolysis, PPi synthesis, and Pi-H2O oxygen exchange. The catalytic activity of E-PPase at saturating substrate increases toward PPi hydrolysis and decreases toward PPi synthesis and Pi-H2O oxygen exchange with increasing pH. These changes are mainly due to an increased rate of dissociation of the second released Pi and a decreased rate of enzyme-bound PPi synthesis from enzyme-bound Pi, respectively, as the pH is raised .

Effect of E20D substitution in the active site of Escherichia coli inorganic pyrophosphatase on its quaternary structure and catalytic properties.

Glutamic acid 20 is an evolutionarily conserved residue found within the active site of the inorganic pyrophosphatase of Escherichia coli (E-PPase). Here we determine the effect of E20D substitution on the quaternary structure and catalytic properties of E-PPase. In contrast to wild-type enzyme, which is hexameric under a variety of conditions, E20D-PPase can be dissociated by dilution into nearly inactive trimers, as shown by electrophoresis of cross-linked enzyme, analytical ultracentrifugation, and measurement of catalytic activity as a function of enzyme concentration. Hexamer stability is increased in the presence of both substrate and Mg2+, is maximal at pH 6.5, and falls off sharply as the pH is lowered or raised from this value. Measured at saturating substrate, 20 mM Mg2+ and pH 7.2, E20D substitution (a) decreases activity towards inorganic pyrophosphate (PPi) hydrolysis and oxygen exchange between water and inorganic phosphate (P1), (b) increases the rate of net PPi synthesis, and (c) decreases the amount of enzyme-bound PPi in equilibrium with Pi in solution. Measurements of PPi hydrolysis rate as a function of both Mg2+ concentration and pH for the E20D variant show that its decreased activity is largely accounted for on the basis of an increased pKa of the catalytically essential base at the active site, and the need for a Mg2+ stoichiometry of 5 in the enzyme-substrate complex, similar to what is seen for the D97E variant. By contrast, wild-type PPase catalysis over a wide range of Mg2+ concentration and pH is dominated by an enzyme-substrate complex having a total of four Mg2+ ions. These results are consistent with a supporting role for Glu20 in PPase catalysis and demostrate that even conservative mutation at the active site can perturb the quaternary structure of the enzyme.

Dissociation of hexameric Escherichia coli inorganic pyrophosphatase into trimers on His-136-->Gln or His-140-->Gln substitution and its effect on enzyme catalytic properties.

Each of the five histidines in Escherichia coli inorganic pyrophosphatase (PPase) was replaced in turn by glutamine. Significant changes in protein structure and activity were observed in the H136Q and H140Q variants only. In contrast to wild-type PPase, which is hexameric, these variants can be dissociated into trimers by dilution, as shown by analytical ultracentrifugation and cross-linking. Mg2+ and substrate stabilize the hexameric forms of both variants. The hexameric H136Q- and H140Q-PPases have the same binding affinities for magnesium ion as wild-type, but their hydrolytic activities under optimal conditions are, respectively, 225 and 110% of wild-type PPase, and their synthetic activities, 340 and 140%. The increased activity of hexameric H136Q-PPase results from an increase in the rate constants governing most of the catalytic steps in both directions. Dissociation of the hexameric H136Q and H140Q variants into trimers does not affect the catalytic constants for PPi hydrolysis between pH 6 and 9 but drastically decreases their affinities for Mg2PPi and Mg2+. These results prove that His-136 and His-140 are key residues in the dimer interface and show that hexamer formation improves the substrate binding characteristics of the active site.

Rates of elementary steps catalyzed by rat liver cytosolic and mitochondrial inorganic pyrophosphatases in both directions.

We have investigated kinetics of pyrophosphate synthesis and phosphate-water oxygen exchange catalyzed by rat liver cytosolic and mitochondrial pyrophosphatases in the presence of Mg2+ as cofactor. A common kinetic model derived for these reactions implies that they involve formation of enzyme-bound pyrophosphate and proceed through two parallel pathways: pathway I, utilizing two magnesium phosphate molecules, and pathway II, utilizing both magnesium phosphate and free phosphate. Pyrophosphate formation is greatly facilitated in the active sites of both pyrophosphatases ([E.PPi]/[E.2Pi] = 0.11-0.24) compared to solution. The rate constants for PPi binding/release, bound PPi hydrolysis/synthesis, and two Pi binding/release steps catalyzed by cytosolic and mitochondrial pyrophosphatases were enumerated for pathway I. There is no unique rate-limiting step for pathway I for both enzymes in either direction. A modulating effect of magnesium phosphate on the oxygen exchange is observed with the cytosolic pyrophosphatase, explicable in terms of an allosteric phosphate-binding site or random-order release of two phosphate molecules from the active site. A remarkable feature of these mammalian pyrophosphatases versus their microbial counterparts is their high efficiency in pyrophosphate synthesis. The turnover numbers in the direction of synthesis are 14 and 9.3 s-1 for the cytosolic and mitochondrial enzymes, respectively (9 and 16% relative to hydrolysis turnover numbers). The results demonstrate that the enzyme-catalyzed synthesis of pyrophosphate, the simplest high-energy polyphosphate, can proceed at a high rate in the absence of an external energy input, such as that provided by protonmotive force in membrane systems.

Oxygen exchange reactions catalyzed by vacuolar H(+)-translocating pyrophosphatase. Evidence for reversible formation of enzyme-bound pyrophosphate.

Vacuolar membrane-derived vesicles isolated from Vigna radiata catalyze oxygen exchange between medium phosphate and water. On the basis of the inhibitor sensitivity and cation requirements of the exchange activity, it is almost exclusively attributable to the vacuolar H(+)-pyrophosphatase (V-PPase). The invariance of the partition coefficient and the results of kinetic modeling indicate that exchange proceeds via a single reaction pathway and results from the reversal of enzyme-bound pyrophosphate synthesis. Comparison of the exchange reactions catalyzed by V-PPase and soluble PPases suggests that the two classes of enzyme mediate P(i)-HOH exchange by the same mechanism and that the intrinsic reversibility of the V-PPase is no greater than that of soluble PPases.

Study of the mechanism of MF1 ATPase inhibition by fluorosulfonylbenzoyl inosine, quinacrine mustard, and efrapeptin using intermediate 18O exchange as a probe.

The mitochondrial F1-ATPase (MF1) is known to be largely or totally inhibited by combination or reaction with one fluorosulfonylbenzoyl inosine (FSBI), quinacrine mustard, or efrapeptin per enzyme. Measurements were made with 18O in attempt to ascertain if the weak catalytic activity remaining after exposure to excess of these reagents was due to retention of some activity by the enzyme modified by these inhibitors. Any such activity could have different characteristics that might be revealed by the distribution of [18O]Pi isotopomers formed from [gamma-18O]ATP. The MF1 inhibited by FSBI showed progressive appearance of two new catalytic pathways as inhibition proceeded. Both pathways appeared to be operative in the enzyme after one beta subunit per enzyme had been modified by FSBI. A high exchange pathway showed no detectable change as ATP concentration was lowered. The lower exchange pathway showed an increase in the amount of exchange with lowering of the ATP concentration, similar to the cooperative behavior observed with the unmodified enzyme. With excess ATP more product was formed by the low exchange pathway, showing that compulsory alternation between two catalytic sites was not retained. The behavior can be explained by the ability of the modified beta subunit to undergo binding changes similar to those occurring in catalysis, with the other two beta subunits catalyzing sluggish hydrolysis by different pathways because of the asymmetry introduced by the modification. Inhibition by quinacrine mustard also resulted in the appearance of two new pathways, somewhat similar to those from FSBI inhibition. In contrast, activity remaining with excess efrapeptin present showed only one pathway like that of the native enzyme. This can be attributed to a low equilibrium concentration of free enzyme and total inhibition of MF1 combined with efrapeptin.

[Superoxide dismutase in liver ischemia]. / Superoksiddismutaza pri ishemii pecheni.

The properties of Cu,Zn-superoxide dismutase (SOD) from rat liver after 2-hour total ischemia or after ischemia with subsequent 24-hour reperfusion were studied. Two hours after ischemia the specific activity of SOD decreases drastically (about 3-fold) - from 510 +/- 11 u./mg in normal tissue and 196 +/- 33 u./mg after ischemia showing a further increase after reperfusion (276 +/- 40 u./mg). Using competitive immunoenzymatic analysis, the relative contents of SOD in the cytosol were determined. After ischemia the SOD content in the cytosolic fraction decreased (approximately 3-fold) but returned to the initial level after reperfusion. Polyacrylamide gel electrophoresis revealed that in control samples active SOD is heterogeneous and produces 3-4 bands, similar to the purified SOD from rat liver. After the ischemia the intensity of minor fast band IV increased and a new band V of a still higher mobility appeared. After the reperfusion the electrophoretic patterns were similar to control. Two or three times more SOD antigen from ischemia liver cytosol was absorbed to the surface of polystyrol plate in a direct sorption enzyme immunoassay procedure as compared to that from intact liver cytosol. It is suggested that the decreases of amount and the activity as well as changes of properties of SOD could be due to its oxidative modification and degradation of the modified enzyme.

Kinetics and thermodynamics of catalysis by the inorganic pyrophosphatase of Escherichia coli in both directions.

Combined evidence obtained from the measurements of pyrophosphate hydrolysis and synthesis, oxygen exchange between phosphate and water, enzyme-bound pyrophosphate formation and Mg2+ binding enabled us to deduce the overall scheme of catalysis by Escherichia coli inorganic pyrophosphatase in the presence of Mg2+. We determined the equilibrium constants for Mg2+ binding to various enzyme species and forward and reverse rate constants for the four steps of the catalytic reaction, namely, binding/release of PPi, hydrolysis/synthesis of PPi and successive binding/release of two Pi molecules. Catalysis by the E. coli enzyme in both directions, in contrast to baker's yeast pyrophosphatase, occurs via a single pathway, which requires the binding of Mg2+ to the sites of four types. Three of them can be filled in the absence of the substrates, and the affinity of one of them to Mg2+ is increased by two orders of magnitude in the enzyme-substrate complexes. The distribution of 18O-labelled phosphate isotopomers during the exchange indicated that hydrolysis of pyrophosphate in the active site is appreciably reversible. The equilibrium constant for this process estimated from direct measurements is 5.0. The ratio of the maximal velocities of pyrophosphate hydrolysis and synthesis is 69. The rate of the synthesis is almost entirely determined by the rate of the release of pyrophosphate from the enzyme. In the hydrolytic reaction, enzyme-bound pyrophosphate hydrolysis and successive release of two phosphate molecules proceed with nearly equal rate constants.