A model for the reaction pathways of the K+-dependent phosphatase activity of the (Na+ + K+)-dependent ATPase

$\text{(Na}{}^{\mathrm{+}}\text{+ K}{}^{\mathrm{+}}\text{)-dependent}$ ATPase preparations from rat brain, dog kidney, and human red blood cells also catalyze a K⁺-dependent phosphatase reaction. K⁺ activation and Na⁺ inhibition of this reaction are described quantitatively by a model featuring isomerization between E₁ and E₂ enzyme conformations with activity proportional to E₂K concentration:

Differences between the three preparations in $\text{K}{}_{0.5}$ for K⁺ activation can then be accounted for by differences in equilibria between E₁K and E₂K with dissociation constants identical. Similarly, reductions in $\text{K}{}_{0.5}$ produced by dimethyl sulfoxide are attributable to shifts in equilibria toward E₂ conformations. Na⁺ stimulation of K⁺-dependent phosphatase activity of brain and red blood cell preparations, demonstrable with KCl under 1 mM, can be accounted for by including a supplementary pathway proportional to E₁Na but dependent also on K⁺ activation through high-affinity sites. With inside-out red blood cell vesicles, K⁺ activation in the absence of Na⁺ is mediated through sites oriented toward the cytoplasm, while in the presence of Na⁺ high-affinity K⁺-sites are oriented extracellularly, as are those of the $\text{(Na}{}^{\mathrm{+}}\text{+ K}{}^{\mathrm{+}}\text{)-dependent}$ ATPase reaction. Dimethyl sulfoxide accentuated Na⁺-stimulated K⁺-dependent phosphatase activity in all three preparations, attributable to shifts from the E₁P to E₂P conformation, with the latter bearing the high-affinity, extracellularly oriented K⁺-sites of the Na⁺-stimulated pathway.     

Effects of oligomycin and quercetin on the hydrolytic activities of the (Na+ +K+)-dependent ATPase

Quercetin inhibited a dog kidney (Na+ + K+)-ATPase preparation without affecting Km for ATP or K0.5 for cation activators, attributable to the slowly-reversible nature of its inhibition. Dimethyl sulfoxide, a selector of E2 enzyme conformations, blocked this inhibition, while the K+-phosphatase activity was at least as sensitive to quercetin as the (Na+ + K+)-ATPase activity, all consistent with quercetin favoring E1 conformations of the enzyme. Oligomycin, a rapidly-reversible inhibitor, decreased the Km for ATP and the K0.5 for cation activators, and its inhibition was also diminished by dimethyl sulfoxide. Although oligomycin did not inhibit the K+-phosphatase activity under standard assay conditions, a reaction presumably catalyzed by E2 conformations, its effects are nevertheless accommodated by a quantitative model for that reaction depicting oligomycin as favoring E1 conformations. The model also accounts quantitatively for effects of both dimethyl sulfoxide and oligomycin on Vmax, Km for substrate, and K0.5 for K+, as well as for stimulation of phosphatase activity by both these reagents at low K+ but high Na+ concentrations.     

Mechanisms by which Li+ stimulates the (Na+ and K+)-dependent ATPase.

The addition of LiCl stimulated the (Na+ + K+)-dependent ATPase activity of a rat brain enzyme preparation. Stimulation was greatest in high Na+/low K+ media and at low Mg-ATP concentrations. Apparent affinities for Li+ were estimated at the alpha-sites (moderate-affinity sites for K+ demonstrable in terms of activation of the associated K+-dependent phosphatase reaction), at the beta-sites (high-affinity sites for K+ demonstrable in terms of activation of the overall ATPase reaction), and at the Na+ sites for activation. The relative efficacy of Li+ was estimated in terms of the apparent maximal velocity of the phosphatase and ATPase reactions when Li+ was substituted for K+, and also in terms of the relative effect of Li+ on the apparent Km for Mg-ATP. With these data, and previously determined values for the apparent affinities of K+ and Na+ at these same sites, quantitative kinetic models for the stimulation were examined. A composite model is required in which Li+ stimulates by relieving inhibition due to K+ and Na+ (i) by competing with K+ for the alpha-sites on the enzyme through which K+ decreases the apparent affinity for Mg-ATP and (ii) by competing with Na+ at low-affinity inhibitory sites, which may represent the external sites at which Na+ is discharged by the membrane Na+/K pump that this enzyme represents. Both these sites of action for Li+ would thus lie, in vivo, on the cell exterior.     

Effects of N-acetylimidazole on human erythrocyte ATPase activity. Evidence for a tyrosyl residue at the ATP binding site of the (Na+ plus K+)-dependent ATPase.

1. Acetylation of human erythrocytes by N-acetylimidazole alters the structure of stroma prepared from these cells and the degree of alteration appears to be dependent upon the level of the initial treatment. These changes do not occur when stroma are acetylated. 2. Deacetylation by hydroxylamine or mild alkaline treatment causes a complete recovery of the (Na+ plus K+)-dependent and the Ca2+ -stimulated ATPase activities and indicates that the inhibition is due to the acetylation of a tyrosyl residue. There is only partial recovery of the Mg2+ -dependent ATPase after deacetylation. 3. ATP or Mg-ATP completely protect the (Na+ plus K+)-dependent ATPase, but not the Ca2+ -stimulated system. 4. The results indicate that the (Na+ plus K+)-dependent and the Ca2+ -stimulated ATPase activities have separate substrate binding sites and most likely are separate enzyme systems. 5. Acetylation of human erythrocytes has no effect on D-glucose transport.     

Odorous chemical perturbations of (Na+ + K+)-dependent ATPase activities. Effects on native and lipid-substituted preparations from individual turbinals from dog olfactory tissue.

Individual turbinals from the right and left sides of dog olfactory tissue were removed and nerve-ending-particle preparations were prepared. (Na+ + K+)-dependent ATPase activities of the individual preparations, and the effect of several odorous compounds [including (+)- and (-)-carvone] on the (Na+ + K+)-dependent ATPase activities, were determined. The maximally stimulatory odorant concentration in the reaction mixture for the majority of odorants was found to be 1.0 mM. Matched pairs of left/right turbinals showed a lack of bilateral symmetry of response. (Na+ + K+)-dependent ATPase activities of various dog brain nerve-ending particle preparations responded only slightly to 1.0 mM odorants. The role of phospholipids in the (Na+ + K+)-dependent ATPase activity was found to be critical. Partial replacement of endogenous lipid with either synthetic phospholipids or extracted lipids resulted in changes in stimulation obtained with endogenous lipids alone.     

Substrate sites for the (Na+ + K+)-dependent ATPase.

Kinetic studies on a rat brain (Na+ + K+)-dependent ATPase (EC 3.6.1.3) preparation demonstrated high-affinity sites for ATP, with a Km near 1 mum, and low affinity sites for ATP, with a Km near 0.5 mM. In addition, the dissociation constant for ATP at the low affinity sites was approached through the ability of ATP to modify the rate of photo-oxidation of the enzyme in the presence of methylene blue; a value of 0.4 mM was obtained. The temperature dependence of the Km values in these two concentration ranges also differed markedly, and the estimated entropy of binding was +27 cal/degree per mol at the high affinity sites, whereas it was -20 cal/degree per mol at the low affinity sites. Moreover, the relative affinities of various congeners of ATP as of the K+ -dependent phosphatase reaction of the enzyme indicated an interaction at the low-affinity sites for ATP: ATP, ADP, CTP, and the [beta-gamma] -imido analog of ATP all competed with Ki values near those for the ATPase reaction at the low affinity sites. Conversely, the Km for nitrophenyl phosphate as a substrate for the phosphatase reaction was near its Ki as a competitor at the low-affinity sites of the ATPase reaction. These observations are incorporated into a reaction scheme with two classes of substrate sites on a dimeric enzyme, manifesting idverse enzymatic and transport characteristics.     

Studies on the phosphorylated intermediates of a K+-stimulated ATPase from rabbit gastric mucosa.

A density gradient-purified microsomal membrane preparation from rabbit fundic gastric mucosa was used for a detailed study of the K+-stimulated ATPase and associated intermediate reactions. Membranes incubated with gamma-[32P]ATP show the rapid incorporation of 32P into phosphoprotein. Phosphoprotein levels were markedly reduced (1) when ATP hydrolysis went to completion or (2) upon addition of unlabeled ATP, thus suggesting the participation of a rapid turnover phosphorylated intermediate in the gastric microsomal ATPase. Addition of K+, Rb+ or Tl+ greatly reduced the level of the intermediate while stimulating ATPase activity; the observed affinities of these cations were similar for the effects on both ATPase and intermediate levels, with Tl+ greater than K+ greater than Rb+. Neither ATPase nor intermediate were stimulated by Na+, and ouabain was without effect on the reactions, thus differentiating this system from the (Na+ + K+)-ATPase. Addition of various inhibitors showed differential effects on the partial reactions of the gastric ATPase system. N-ethylmaleimide and Zn2+ showed characteristics of completely abolishing the K+-stimulated component of ATPase as well as the effects of K+ in reducing the level of intermediate, thus suggesting that these agents exert their inhibitory effect on a phosphoprotein phosphatase partial reaction. F- abolished the K+-stimulated ATPase, but its more complex effects on the intermediate suggested an additional reaction step within the domain of the phosphorylated intermediate. Results are consistent with a model system for the gastric microsomal ATPase involving a Mg2+-dependent protein kinase, a phosphorylated intermediate(s), and a K+-stimulated phosphoprotein phosphatase.     

Buffer, pH, and ionic strength effects on the (Na+ + K+)-ATPase

A dog kidney (Na+ + K+)-ATPase preparation also catalyzes K+-independent and K+-activated phosphatase reactions with p-nitrophenyl phosphate as substrate. K+-independent activity increases with declining pH over the range 7.5 to 5.8, whereas the other two activities decrease. The increased K+-independent activity is similar with imidazole, histidine, and several Good buffers, and is thus attributable to free H+, probably by affecting enzyme conformations rather than by changing affinity for Mg2+ or substrate or by H+ occupying specific K+-sites. The decrease in K+-phosphatase and (Na+ + K+)-ATPase activities with pH also occurs similarly with those buffers, and is not due to changes in apparent affinity for substrate or for cation activators. However, the Good buffers Pipes and ADA inhibit the K+-independent phosphatase reaction strongly, the K+-activated reaction moderately, and the (Na+ + K+)-ATPase reaction little; both contain two acidic groups, unlike the other buffers tested. Inhibition of the phosphatase reaction by Pipes is associated with a decreased apparent affinity for K+ and an increased sensitivity to inhibition by Na+ and ADP, consistent with Pipes hindering conformational transitions to the E2 enzyme forms required for phosphatase hydrolytic activity.     

Characterization of the phosphorylated intermediate of the K+-translocating Kdp-ATPase from Escherichia coli

During ATP hydrolysis the K+-translocating Kdp-ATPase from Escherichia coli forms a phosphorylated intermediate as part of the catalytic cycle. The influence of effectors (K+, Na+, Mg2+, ATP, ADP) and inhibitors (vanadate, N-ethylmaleimide, bafilomycin A1) on the phosphointermediate level and on the ATPase activity was analyzed in purified wild-type enzyme (apparent Km = 10 microM) and a KdpA mutant ATPase exhibiting a lower affinity for K+ (Km = 6 mM). Based on these data we propose a minimum reaction scheme consisting of (i) a Mg2+-dependent protein kinase, (ii) a Mg2+-dependent and K+-stimulated phosphoprotein phosphatase, and (iii) a K+-independent basal phosphoprotein phosphatase. The findings of a K+-uncoupled basal activity, inhibition by high K+ concentrations, lower ATP saturation values for the phosphorylation than for the overall ATPase reaction, and presumed reversibility of the phosphoprotein formation by excess ADP indicated similarities in fundamental principles of the reaction cycle between the Kdp-ATPase and eukaryotic E1E2-ATPases. The phosphoprotein was tentatively characterized as an acylphosphate on the basis of its alkali-lability and its sensitivity to hydroxylamine. The KdpB polypeptide was identified as the phosphorylated subunit after electrophoretic separation at pH 2.4, 4 degrees C of cytoplasmic membranes or of purified ATPase labeled with [gamma-32P]ATP.     

Microsomal (Na- +K+)-activated ATPase from frog skin epithelium. Cation activations and some effects of inhibitors.

A method is described for the extraction of microsomal ouabain-sensitive (a- + K+)-activated ATPase from separated frog skin epithelium. The method yields a microsomal fraction containing (Na+ K+)-stimulated activity in the range of 30- 40 nmol - mg -1 - min -1 at 26 degrees C. This portion which is also ouabain sensitive, is about half of the total activity in media containing Mg2+, Na+ and K+. These preparations also contain Mg2+-dependent or Ca2+-dependent activities which are not additive and which are not significantly affected by ouabain, Na+, K+ or Li+. The activations of the ouabain-sensitive ATPase activity by Mg2+, Na+, and K+ are similar to those described in other tissues. It is found that Li+ does not substitute for Na+ as an activator but in high concentrations does produce partial activation in the presence of Na+ with no K+. These results are pertinent to the reported observations of ouabain-sensitive Li+ flux across frog skin. It is concluded that this flux is not apparently due to a direct activating effect of Li+ on the sodium pump.     

Phenobarbital modulates the (Na+, K+)-stimulated ATPase and Ca2+-stimulated ATPase activities by increasing the bilayer fluidity of dog brain synaptosomal plasma membranes

The binding of [14C]phenobarbital into synaptosomal plasma membranes of dog brain follows a sigmoid path. The "best fit" curve of this binding is the one described by the Hill equation (r2 less than 0.93 and Hill coefficient, n = 1.32). (Na+, K+)-stimulated ATPase and Ca2+-stimulated ATPase activities are modulated by phenobarbital. Arrhenius plots of (Na+, K+, Mg2+)-dependent ATPase revealed that phenobarbital (2 mM) lowered the transition temperature and altered the Arrhenius activation energies of this enzyme. The allosteric inhibition by F- of the (Na+, K+)-stimulated ATPase was studied in control and phenobarbital-treated membranes. The lowering of the transition temperature and changes in Arrhenius activation energy about the transition temperature in combination with changes observed in the allosteric properties of the (Na+, K+)-stimulated ATPase by F-, produced by phenobarbital, would be expected if it is assumed that phenobarbital "fluidizes" synaptosomal plasma membranes.     

Ouabain-insensitive Na+ stimulation of a microsomal Mg2+ -ATPase in gills of sea bass (Dicentrarchus labrax L.)

Bass gill microsomal preparations contain a Mg2+-dependent Na+-stimulated ATPase activity in the absence of K+, whose characteristics are compared with those of the (Na+ + K+)-ATPase of the same preparations. The activity at 30 degrees C is 11.3 mumol Pi X mg-1 protein X hr-1 under optimal conditions (5 mM MgATP, 75 mM Na+, 75 mM HEPES, pH 6.0) and exhibits a lower pH optimum than the (Na+ + K+)-ATPase. The Na+ stimulation of ATPase is only 17% inhibited by 10-3M ouabain and completely abolished by 2.5 mM ethacrinic acid which on the contrary cause, respectively, 100% and 34% inhibition of the (Na+ + K+)-ATPase. Both Na+-and (Na+ + K+)-stimulated activities can hydrolyze nucleotides other than ATP in the efficiency order ATP greater than CTP greater than UTP greater than GTP and ATP greater than CTP greater than GPT greater than UTP, respectively. In the presence of 10(-3)M ouabain millimolar concentrations of K+ ion lower the Na+ activation (90% inhibition at 40 mM K+). The Na+-ATPase is less sensitive than (Na+ + K+)-ATPase to the Ca2+ induced inhibition as the former is only 57.5% inhibited by a concentration of 1 X 10(-2)M which completely suppresses the latter. The thermosensitivity follows the order Mg2+--greater than (Na+ + K+)--greater than Na+-ATPase. A similar break of the Arrhenius plot of the three enzymes is found. Only some of these characteristics do coincide with those of a Na+-ATPase described elsewhere. A presumptive physiological role of Na+-ATPase activity in seawater adapted teleost gills is suggested.     

Activation of (Na+ + K+)-dependent ATPase by lipid vesicles of negative phospholipids.

1. Kidney (Na+ + K+)-stimulated ATPase was depleted of phospholipids by extraction with lubrol and inserted in lipid structures of known composition. Both ouabain-sensitive ATPase and phosphatase reactions could be partially restored by lipid replacement. 2. Lipid vesicles of natural and synthetic negative phospholipids proved to be effective. The low activity of uncharged liposomes was increased when negative charges were included into the bilayer structure. 3. Reactivation by negative phospholipids was accompanied by spontaneous re-assembly of a stable lipid-protein complex. By contrast, the interaction of lipid deficient ATPase complex with uncharged lamellae was possible only after sonication of lipid-protein suspension. Reactivation did not ensue. 4. The ouabain-sensitive ATPase reactivated by synthetic dioleoylphosphatidylglycerol yielded curvilinear Arrhenius plots. The same pattern was seen with the original undepleted microsomal preparation. A discontinuity close to the temperature of fluid-order transition was found with dimyristoyl phosphatidylglycerol. 5. It is concluded that reassembly of lipid-deficient (Na+ + K+)-stimulated ATPase requires the addition of diacylphospholipids with fluid acyl-chains and negatively charged polar heads able to assemble in an expanded lamellar configuration.     

Effects of polyamines on partial reactions of membrane (Na+ + K+)-ATPase.

Spermine and spermidine inhibit the (Na+ + K+)-ATPase (ATP phosphohydrolase, EC 3.6.1.3) reaction so that the effect increases as the ionic content due to Na+ and K+ in the reaction is reduced. Several other amines inhibit (Na+ + K+)-ATPase to varying degress and methylglyoxal-bis-(guanylhydrazone) was the most potent inhibitor among those tested. The inhibition by polyamines of the ATPase is uncompetitive with respect to Mg2+ and ATP activation of the reaction. Various naturally occurring polyamines and other amines inhibited Na+ activation of (Na+ + K+)-ATPase as well as Na+-dependent phosphoenzyme formation in an apparently competitive manner with respect to Na+. Likewise, K+-activation of (Na+ + K+)-ATPase as well as K+-p-nitrophenyl phosphatase was inhibited in an apparently competitive manner with respect to K+. Both the cation charge and structure (e.g., aliphatic chain length) may contribute to the inhibitory effects of the amines; however, Na+ sites appear to be more sensitive to cation charge than the aliphatic chain length of the amine, whereas the opposite appears to be true for K+ sites. The results do not indicate a specific effect of polyamines on (Na+ + K+)-ATPase or its partial reactions.     

Role of sulfatide on phosphoenzyme formation and ouabain binding of the (Na+ + K+)ATPase

A microsomal fraction rich in (Na+ + K+)ATPase activity has been isolated from the outer medulla of pig kidney. The ability of this preparation to form phosphoenzyme on incubation with [gamma-32P]ATP and to bind [3H]ouabain was studied when its sulfatide was hydrolyzed by arylsulfatase treatment. The K+-dependent hydrolysis of the Na+-dependent phosphorylated intermediate as well as the ouabain binding were inactivated in direct relation to the breakdown of sulfatide. Both characteristics of the (Na+ + K+)ATPase preparation, lost by arylsulfatase treatment, were partially restored by the sole addition of sulfatide. These experiments indicate that sulfatide may play a role in sodium ion transport either in the conformational transition of the K+-insensitive phosphointermediate, E1P, to the K+-sensitive intermediate, E2P, or in the configuration of the high-affinity binding site for K+ of the E2P form. In addition, this glycolipid may have a specific role in the proteolipidic subunit that binds ouabain.     

Functionally distinct classes of K+ sites on the (Na+ + K+)-dependent ATPase.

K+ interactions with a rat brain (Na+ + K+)-dependent ATPase and the associated K+-dependent nitrophenyl phosphatase activity were examined. Classes of sites for K+ were distinguished, initially, on the basis of affinity estimated by kinetic analysis in terms of KO.5 (the concentration for half-maximal activation), and by K+-accelerated enzyme inactivation by F-minus, which permits evaluation of a dissociation constant for K+, KD. Moderate-affinity sites ("alpha sites"), with a KD near 1 mM, were demonstrable for the phosphatase activity and for the "free" enzyme. High-affinity sites ("beta sites"), with a KD near 0.1 mM, were seen for the overall ATPase activity and under conditions in which enzyme phosphorylation by substrate also occurs. Further differentiation between alpha and beta sites was made in terms of (i) the characteristic changes in affinity with pH, and (ii) the efficacy of Li+ relative to K+, Rb+, Cs+, and Tl+ at these two classes of sites. Low-affinity sites ("gamma sites") through which K+ inhibits enzymatic activity were also detectable, with a KD around 140 mM. These data are incorporated into a model for the reaction sequence to accommodate both transport processes and certain K+/ATP antagonisms.     

ATPase stimulated by Na+ or K+ in gills of the freshwater mussel Anodonta.

1. Microsomal preparations from the gills of the freshwater mussel anodonta cygnea cellensis show Mg2+ -dependent Na+ - or K+ -stimulated ATPase activity, which is not inhibited by ouabain. 2. Na+ - or Ka+ -ATPase activity is decreased by Ca2+, acetylcholine, choline, and tetramethylammonium, but slightly increased by ethyl alcohol. 3. It is tentatively suggested that Na+ - or K+ -ATPase is involved in the mechanism of active monovalent cation uptake through the gills of freshwater mussels.     

Studies on (Na+ + K+)-activated ATPase. XLII. Evidence for two classes of essential sulfhydryl groups.

1. Preincubation of purified (Na+ + K+)-ATPase (ATP phosphohydrolase, EC 3.6.1.3) preparations from rabbit kidney outer medulla with 5,5'-dithiobis-(2-nitrobenzoic acid) inhibits the (Na+ + 5+)-ATPase and K+-stimulated 4-nitro-phenylphosphatase activities. Phosphorylation of the enzyme by ATP and the Na+-stimulated ATPase activity are inhibited to the same extent as the (Na+ + K+)-ATPase activity, whereas the K+-stimulated 4-nitrophenylphosphatase activity is inhibited much less. 2. Titration with 5,5'-dithiobis-(2-nitrobenzoic acid) in sodium dodecyl sulphate shows the presence of 36 reactive sulfhydryl groups per molecule (Na+ + K+)-ATPase (Mr = 250 000). 3. Treatment with N-ethylmaleimide, resulting in complete inhibition of (Na+ + K+)-ATPase activity, leads to modification of 26 sulfhydryl groups, whereas treatment with 5,5'-dithiobis-(2-nitrobenzoic acid) results in modification of 12 sulfhydryl groups under the same conditions. 4. The reaction of N-ethylmaleimide with an essential SH-group is not prevented by previous blocking of sulfhydryl groups with 5,5'-dithiobis-(2-nitrobenzoic acid). 5. These findings indicate the existence of at least two classes of sulfhydryl groups on the enzyme, each containing at least one vital group. The difference between these classes consists in their different reactivity towards 5,5'-dithiobis-(2-nitrobenzoic acid) and N-ethylmaleimide.     

Studies on (Na+ +K+) activated ATPase. XLI. Effects of N-ethylmaleimide on overall and partial reactions.

1. Preincubation with N-ethylmaleimide inhibits the overall activity of highly purified (Na+ +K+)-ATPase (ATP phosphohydrolase, EC 3.6.1.3) preparations of rabbit kidney outer medulla. 2. This inhibition is decreased by addition of ATP or 4-nitrophenylphosphate under non-phosphorylating conditions, and also by addition of ADP or adenylylimidodiphosphate. 3. N-ethylmaleimide treatment leads to inhibition of K+-stimulated 4-nitrophenylphosphatase activity, Na+-stimulated ATPase activity, and phosphorylation by ATP as well as by inorganic phosphate. These inhibitions strictly parallel that of the overal (Na+ +K+)-ATPase reaction. 4. N-ethylmaleimide lowers the number of sites which are phosphorylated by inorganic phosphate, without affecting the dissociation constant of the enzyme-phosphate complex. 5. N-ethylmaleimide does not affect the relative stimulation by ATP of the K+-stimulated 4-nitrophenylphosphatase activity. 6. These effects of N-ethylmaleimide can be explained as a complete loss of active enzyme, either by reaction of N-ethylmaleimide inside the active center, or by alterations in the quaternary structure through reactions outside the active center.     

Interactions between K+ and ATP binding to the (Na+ + K+)-dependent ATPase.

K+ appears to decrease the affinity of the (Na+ + K+)-dependent ATPase (ATP phosphohydrolase, EC 3.6.1.3) for its substrate, Mg2+ - ATP, and Mg2+ - ATP, in turn, appears to decrease the affinity of the enzyme for K+. These antagonisms have been investigated in terms of a quantitative model defining the magnitude of the effects as well as identifying the class of K+ sites on the enzyme involved. K+ increased the apparent Km for Mg2+ - ATP, an effect that was antagonized competitively by Na+. The data can be fitted to a model in which Mg2+ - ATP binding is prevented by occupancy of alpha-sites on the enzyme by K+ (i.e. sites of moderate affinity for K+ accessible on the "free" non-phosphorylated enzyme, in situ on the external membrane surface). By contrast, occupancy of these alpha-sites by Na+ has no effect on Mg2+ - ATP binding to the enzyme. On the other hand, Mg2+ - ATP decreased the apparent affinity of the enzyme for K+ at the alpha-sites, in terms of (i) the KD for K+ measured by K+-accelerated inactivation of the enzyme by F-, and (ii) the concentration of K+ for half-maximal activation of the K+-dependent phosphatase reaction (which reflects the terminal hydrolytic steps of the overall ATPase reaction). These data fit the same quantitative model. Although this formulation does not support schemes in which ATP binding effects the release of transported K+ from discharge sites, it is consistent with observations that K+ can inhibit the enzyme at low substrate concentrations, and that Li+, which has poor efficacy when occupying these alpha-sites, can stimulate enzymatic activity at high K+ concentrations by displacing the inhibitory K+.