Effects of Mg2+ Ions on the Plasma Membrane [H+]-ATPase of Neurospora crassa

The rate of ATP hydrolysis by the Neurospora plasma membrane [H+]-ATPase has been measured over a wide range of Mg2+ and ATP concentrations, and on the basis of the results, a kinetic model for the enzyme has been developed. The model includes the following three binding sites: 1) a catalytic site at which MgATP serves as the true substrate, with free ATP as a weak competitive inhibitor; 2) a high affinity site for free Mg2+, which serves to activate the enzyme with an apparent KIl2 (termed L g A ) of about 15 WM; and 3) a separate low affinity site at which Mg2+ causes mixed type inhibition, lowering the V,,, while raising the Ks for MgATP at the catalytic site. The Ki for Mg2+ at the low affinity site (termed K M ~ ) is about 3.5 mM. The model satisfactorily explains the activity of the enzyme as Mg2+ and ATP are varied, separately and together, over a wide range. It can also account for the previously reported effects of Mg2+ and ATP on the inhibition of the Neurospora [H+]-ATPase by N-ethyl-maleimide The plasma membrane Neurospora the ,

the ATPase, but inactivation studies with NEM are not capable of revealing such a role. Therefore, we have carried out a series of kinetic experiments to measure the effects of free Mg2+, free ATP, and MgATP, varied over a wide concentration range, on ATPase activity. Particular emphasis has been directed towards sorting out which ligands are required as substrates or activators and which serve as inhibitors, and towards determining the corresponding kinetic constants. The results confirm that all three ligand binding sites are involved in the reaction cycle of the Neurospora [H+]-ATPase, and make it possible to derive a rate equation that satisfactorily accounts for enzyme activity as a function of ligand concentrations.

MATERIALS AND METHODS
For kinetic studies, the standard assay mixture contained 50.0 mM PIPES, adjusted to pH 6.7 with Tris base; 0.0025% asolectin; 5.0 m M KN,; and combinations of MgC1, and ATP in a total volume of 1.5 ml. Enzyme was added and the assay was carried out at 30 "C for 5-10 min. The reaction was terminated by the addition of trichloroacetic acid to a final concentration of 1.1%. Under these conditions, hydrolysis was a linear function of time for at least 10 min. Inorganic phosphate production was determined by the method of Dryer et al. (4).
All other methods were as described in the preceding paper (3).

Effects of Mg2+ and ATP on A T P Hydrolysis
In order to investigate the kinetic behavior of an enzyme, it is customary to vary the concentration of each ligand at several fixed concentrations of the others. With enzymes that use MgATP, however, such an analysis becomes more difficult. Since Mg" and ATP combine to form the MgATP complex, the concentrations of Mg*+free, ATPfr,,, and MgATP cannot be varied independently of each other. Instead, as discussed by London and Steck (5), the desired information must be extracted from experiments in which the total concentration of one ligand (e.g. Mg2+) is fixed and enzyme activity is observed to vary with changes in the concentration of the second (e.g. ATP).
For the experiment of Fig. 1A, the activity of the Neurospora [H+]-ATPase was measured a t several fixed MgC12 concentrations (0.5,1.0, and 2.0 mM) while ATP was increased from 0 to 7.5 mM. At each MgCl, concentration, ATPase activity rose to a maximum and then fell again. The maximal rate of ATP hydrolysis occurred a t a point where the MgATP concentration was high (most of the MgZ' in the MgATP form) but where the excess ATPrree concentration was relatively low.
Analogously, for Fig. 1B  MgCL was varied from 0 to 20 mM. The results were parallel to those just described; once again, the maximal rate was reached where the MgATP concentration was high (most of the ATP in the MgATP form) but where the excess Mg2+ concentration was relatively low. In a separate experiment, it was found that the addition of choline choloride up to a concentration of 50 mM had no significant effect on ATPase activity. Therefore, the inhibitory effects of high concentrations of MgC12 or ATP do not appear to be the result of an alteration in ionic strength.
From the results of Fig. 1, it is evident that maximal ATP hydrolysis depends upon the relative concentrations of both M F and ATP. One possible mechanism to account for this behavior would be sequential binding of Mg2+ and ATPfre, to the enzyme. Alternatively, the true substrate may be the MgA'I'P complex which binds directly to the enzyme in a single step. These two mechanisms can be differentiated on the basis of kinetic studies (6). If it is found that the apparent Ks for ATPf,, (KATp) a t a given concentration of Mg2+,, is equal to the apparent Ks for Mg2+free (KhJ at the same Concentration of ATPf,, then the mechanism of the reaction involves the binding of the MgATP complex to the enzyme.
It must be noted that the apparent Ks values for Mg2+fre, and where K,, = dissociation constant of MgATP, and KM~ATP = dissociation constant for MgATP-enzyme complex. Table I    was considerably higher than that of MgATP (1.5 mM). Thus, ATPfree might be expected to act as a weak competitive inhibitor of enzyme activity. At the same time, NEM studies have given evidence for a separate high affinity Mg2"binding site, which might be involved in activation of the enzyme.
In order to gain further information, the ATP concentration was varied in excess above several fixed MgATP concentrations, and enzyme activity was measured. In Fig. 2, values of l / u have been plotted as a function of the concentration of ATPfr,,. In sharp contrast to previous studies with ADP, where simple competitive inhibition was seen (l), inhibition by ATPfre, was markedly biphasic. Thus, although competitive inhibition at the catalytic site almost certainly takes place, it is not sufficient to explain the observed inhibition of enzyme activity by excess ATPf,,,.
T o see whether activation by Mg2+ might also be involved, the data from Fig. 2 were replotted as a function of Mg2+free. At each of the four substrate concentrations, ATPase activity rose along a saturation curve reaching half-maximal values at about 0,015 mM M$+. This result supports the idea that Mg2+ serves as an essential activator, and suggests that the effect of excess ATP can be explained largely by the removal of Mg2+ from solution. Indeed, when the curves of Fig. 3 were corrected for the probable contribution of ATP as a competitive inhibitor (by assuming a K, of 11.4 mM, equal to the measured KO; Ref. l), they were changed in only a minor way, shifting upward slightly and becoming a bit more hyperbolic (see inset to Fig. 3).

Inhibition by Excess Mg2+
From Fig. 1B, it is also apparent that excess M$+ has an inhibitory effect on ATP hydrolysis. To examine this effect in more detail, ATP hydrolysis was measured as a function of MgATP a t several fixed MPr,, concentrations (1, 2, 5 and 10 mM). Under these conditions, the ATPfr,, concentration was very low and its inhibitory effects could be neglected.

ES][Z]/[ESI]
, the dissociation constant of I from the ternary ESI complex. These equations assume that ESI does not break down to yield products and that all the binding reactions can be treated as equilibria (8).
In Table 11, the values for K; (termed KMa) and aKi (termed aKMgl) were determined at several M e f r e e concentrations, using the data from Fig. 4 (3), dissociation constants for MgZ+ cannot be calculated reliably. It is worth noting, however, that KMga (measured kinetically) falls within the range of M e concentrations that protect against NEM inhibition, and likewise that KM, falls within the range of Mg2+ concentrations that enhance NEM inhibition. As a working assumption, therefore, rapid equilibrium seems justified at these sites as well.
All Ligands Bind Randomly-The model is illustrated in account for the behavior of the ATPase under some of the conditions examined above, Equation IX was used to calculate enzyme activity as a function of ATP at fixed MgCIz (analogous to Fig. 1A) and as a function of MgCI2 at fixed ATP (analogous to Fig. 1B). The following values were assumed for the equilibrium constants: KM~ATP = 0.72 mM (extrapolated from data in Table 11); KATp = 11.4 mM (Ref. 1); KMgA = 0.015 mM (Fig. 3); KM, , = 3.55 mM (Table 11); and LYKM~, = 15.4 mM (Table 11). Once the calculated enzyme activities had been scaled appropriately to a V,,, of 30.9 pmol of Pi/min.mg of protein, they could be seen to follow the same general pattern as the experimentally measured activities from Fig. 1 (see Fig.  6). Considering the complexity of the rate equation and the possibility of error in determining the equilibrium constants, the two sets of values are in good agreement.
Equation IX is also relevant to the behavior of the enzyme under a quite different set of conditions. When assayed over a range of equimolar ATP and Mg2+ concentrations, the Neurospora [H+]-ATPase has previously been observed to display sigmoid kinetics with a Hill number of about 1.6-2.0 (Refs. 9 and 10). The sigmoidicity can be explained by pos-  tulating that the ATPase possesses two (or more) sites for ATP, which display positive cooperativity. As illustrated in Fig. 7 (Fig. 7), much better agreement can be obtained if one allows these parameters to vary by a relatively modest factor (1.5 to 2-fold, not shown). We emphasize, however, that such agreement does not rule out the possibility of a multimeric ATPase which displays true cooperativity between subunits. Further studies, particularly of the physical structure of the enzyme, will be required to address this question directly.

DISCUSSION
The kinetic model postulated in this paper for the Neurospora plasma membrane ATPase includes three ligand-binding sites, i.e. a catalytic site for Mg nucleotides, a high affinity activating site for M e , and a low affinity inhibitory site for Mg2' . The model has been derived principally from measurements of enzyme activity over a wide range of ligand concentrations, but it can also account for the effects of Mg2+ and ATP on NEM inhibition (1,3) and tryptic degradation ( 16), and the plasma membrane [H+]-ATPase from Saccharomyces cerevisiae (17,18) and Schizosaccharomyces pombe (19). These other ATPases also appear to have high affinity activating sites for M e , and it may be significant that the K , for activation (KMgA) is similar for the following enzymes: It is more difficult to make generalizations concerning the low affinity inhibitory site for M$+. All of the above-mentioned ATPases are sensitive to inhibition by high M$+ concentrations, but the mode of inhibition appears to vary. For the [H+]-ATPase of S. cereuisiae, inhibition has been reported to be "pseudocompetitive," suggesting that the binding of Mg2+ at the inhibitory site reduces the affinity of the catalytic site for substrate, but that ES and ESZ complexes are split with equal velocity (20). For the [Na+,K+]-ATPase, Mg2+ has been variously described to act as an uncompetitive (11) or noncompetitive (12, 14) inhibitor. And for the [Ca"]-ATPase of sarcoplasmic reticulum, inhibition by high Mg2+ concentrations is partially reversible by Ca2+, making it difficult to distinguish effects at the putative inhibitory site from effects at the specific Ca2+ transport sites on the enzyme (21). Further work will be required to establish the extent to which inhibition by high Mg2+ concentrations shares a common mechanism in these various cases.