The Eq+librium Constants of the Adenosine Triphosphate Hydrolysis and the Adenosine Triphosphate-Citrate Lyase Reactions

SUMMARY The observed standard free energy change (AGibs) for the hydrolysis of the terminal pyrophosphate bond of ATP has been experimentally determined under physiological conditions using an entirely new set of reactions. The observed equilibrium constant (K,,bs) for the combined reactions of acetate kinase (EC 2.7.2.1) and phosphate acetyltransferase (EC 2.3.1.8) has been determined at 38”, pH 7.0, ionic strength 0.25, and varying free [Mg”+]. The K,,bs of these combined reactions reflects the difference between AG& for the hydrolysis of acetyl-CoA and the AGibs for the hydrolysis of ATP. Using 2 and square brackets to indicate total concentration, The observed value


SUMMARY
to obtain values for the ATP hydrolysis reaction.
Alberty (3) and Phillips et al. (4), using the same original data of Benzinger et al. (1) and Levintow and Meister (a), have arrived at quite different values. Rosing and Slater (5) have redetermined the equilibrium constant of the glutamine synthetase reaction which they then combined with the equilibrium constant of the glutaminase reaction measured by Benzinger (1) to arrive at yet another set of values.
The work based on the glutaminase and glutamine synthetase reactions has therefore yielded a considerable range of values for the AGzbs of Equation 1. For example at 37", pH 7.0, and free [Mg*+] = lop3 1\2, the calculated values of the AGzbs range from -6.79 Cal per mole (-28.4 kJ per mole) (5)  The total magnesium concentrations were analyzed with a Varian Techtron atomic absorption spectrophotometer (type AA-5). pH measurements were made at 38" with a Radiometer Copenhagen microelectrode (type E5021a) and pH meter (type 27) using precision Radiometer buffers as standards.
Metabolite Assays-CoA was assayed using a-ketoglutarate dehydrogenase (8). Stock solutions of CoA were also analyzed with phosphate acetyltransferase (9) with agreement of the two methods.
Acetyl-CoA was assayed by the method of Wieland and Weiss (10) as modified by Pearson (11). Stock solutions of acetyl-CoA were also measured by the method of Srere (12) with agreement of the two methods.
ATP, ADP, AMP, and Pi were measured enzymatically as described previously (13,14). Acetatc was measured enzymatically (15). The values obtained for potassium acetate agreed well with that predicted from the weight of the dry salt.
The values obtained for magnesium acetate agreed well with the magnesium content. Contaminations-All reagents were assayed for possible contaminations of acetate, phosphate, and magnesium, and these contaminations have been taken into account. Total contaminations of phosphate and acetate were < 1 To of the total added. Total magnesium contaminations were ~0.027~ pmole per ml in the final reaction mixture, coming mainly from the ATP and ADP.
Procedure-The details of the concentrations of the reaction mixtures are given in Table II. For each reaction mixture a total of 6.0 ml of final volume was prepared.
The reaction was initiated at room temperature and allowed to stand for 10 min. Aliquots of 1.5 ml of the reaction mixture were then pipetted into each of three tubes (13 x 100 mm), and the tubes and reaction mixture remaining in the original tube were capped and placed in a 38" shaking water bath.
At 15, 30, and 45 min, 300 ~1 of 7.2 M ice-cold perchloric acid were added to one of the tubes which was then shaken and chilled rapidly to 0". After all of the samples were taken, the pH was measured in the reaction mixture remaining in the original tube.
After standing in ice for 20 min the tubes were centrifuged at 0" at 700 x g for 15 min. ,4 1.3.ml portion of the clear supernatant was transferred to fresh tubes and 150 ~1 of N 6 RI KzC03 was added slowly to the cold tube.
The concentration of the K&O3 was previously adjusted so that 150 ~1 would neutralize the extract to pH 5.5 to 6.0. After standing for 20 min in ice, the tubes were centrifuged at 700 X g at 0" for 20 min and bhe clear supernatant was assayed for ATP, ADP, acetyl-CoA, and CoA. This extract contained no detectable activity of acetate kinase or phosphate acctyltransferase.
AMP was also assayed in some tubes. The AMP found could be accounted for by the AMP contaminations in the ATP and ADP added, indicating that thcrc was no detectable myokinase activity under the conditions of incubation used. One calibrated micropipette was used to sample any given extract by guest on March 24, 2020 http://www.jbc.org/ Downloaded from for each of the assays. This eliminated potential errors in the ratios of the metabolitcs due to multiple pipettings of the reaction mixture and extract.
The final concent'rations of phosphate and acetate were taken to be equal to the initial concentrations corrected on the assumption that the change in acetate and phosphate was stoichiometric with the change in ATP and acetyl-CoA.
A potential error in a reaction mixture containing acetyl phosphate is the spontaneous hydrolysis of this compound to acetate and phosphate. However, since large and nearly equal concentrations of acetate and phosphate were initially added, it can be estimated from Table  II  Calculations hare also been carried out with constants used by Rosing and Slater (5) including corrections for potassium binding.
The calculations based on these constants IT-ere not very differeut from the calculations based on the constants of Table I  where the I& constants are the obscrred magnesium binding constants at a given pH and potassium ion concentration. which becomes, by substitution of the constants of

Equilibrium
Constant of Combined Reactions of Acetate Kinase and Phosphate Acet!/ltransjerase- Table  II shows the details of the determination of the constant at one magnesium concentration.
The reaction has been studied starting with either Cod, ADP, and acelyl phosphate (the kinetically more favored or "forward" reaction)? or with ATP and acetyl-Cog (the "reverse" 6969 reaction). Equilibrium was reached by 15 min at 38". It should be noted that once equilibrium was reached in the forward direction, the ATP and acetyl-CoA concentrations (the products) tended to slowly decrease, reflecting the spontaneous hydrolysis of the acetyl phosphate with reversal of the reactions to maintain equilibrium.
The variation of the Zio,,s of the combined reaction of acetate kinase and phosphate acetyltransferase with magnesium is summarized in Table III. Attempts to achieve equilibrium with the reverse reaction at the lowest magnesium concentrations (0.32 and 0.02 mM total) were unsuccessful even with prolonged incubations.
Failure to achieve equilibrium resulted from the fact that the rate of the acetate kinase reaction in the direction of acetyl phosphate production is slower at very low magnesium concentrations than the spontaneous hydrolysis of acetgl phosphate. In the other trials of the reverse reactionequilibrium was achieved rapidly, being complete xithin 15 min. The Ziobs varies from 0.164 =I= 0.003 at 10 mu total magnesium (2.7 mivr free) to 0.989 f 0.019 at 0.02 nlM total magnesium (0.002 mM free).
The average in the forward direction is  The procedure is the same as that described in Table II. Varying amolmts of magnesium as the chloride or acetate salt were added and the ionic strength was adjusted to 0.25 with KCl. The const,:tnt is defined by Temperature was 38", pI1 was 7.01 =t 0.01; values are given as II~C:L~~S + standard error. i 0.967 i 0.007 and in the reverse direction is 1.031 =t 0.013. An average of 0.984 =t 0.009 has been taken.
The value of AG& for the hydrolysis of acetyl-CoA under the same physiological conditions has been previously determined to be -S.54 Cal per mole (-35.75 kJ per mole) independent of the free [11g2+] (6). Thn,rfore the AG& for the hydrolysis of ATP at $1 7.0, 38", and I = 0.25 (with potassium as the monovalent ion) ran be calculated and expressed as a function of magnesium as shown in Fig. 1 The binding constants and acid dissociation constants used (Table I) to calculate the free [&If;"+] and the zero magnesium constant adequately describe the experimental values (Fig. I).
It follows that these binding constants arc probably good estimates of the actual constants under thcsc conditions. This view is supported by the finding that if other binding constants are used (5) the standard error of the mean of the calculated zrro magnesium constant increases a-fold and the csperimental points no longer fit the calculated curve. Because of the uncertainties in the arid dissociation and magnesium binding constants, no att,empt has been made to predict the effect of ionic strength on the equilibrium constant of the ATP hydrolysis reaction in the absence of experimental data. I-Ioxevcr, since the acid dissociation constants and the magnesium binding constants of Table I are   The Zcobs of the ATP-cilratc lyase reaction will be