Sodium acts as a potassium analog on gastric H,K-ATPase.

The effects of Na+ on gastric H,K-ATPase were investigated using leaky and ion-tight H,K-ATPase vesicles. Na+ activated the total ATPase activity in the absence of K+, reaching levels of 15% relative to those in the presence of K+. The Na+ activation, which takes place at the luminal side of the membrane, depended on the ATP concentration and the type of buffer used. The steady-state ATP phosphorylation level, studied with leaky vesicles, was reduced by Na+ due to both activation of the dephosphorylation reaction and a shift to E2 in the E1<==>E2 equilibrium. By studying this equilibrium in ion-tight H,K-ATPase vesicles, it was found that Na+ drives the enzyme via a cytosolic site to the nonphosphorylating E2 conformation. No H(+)-like properties of cytosolic Na+ could be detected. We therefore conclude that Na+ behaves like K+ rather than like H+ in the H,K-ATPase reaction.

different partial reactions can be distinguished such as (i) the steady-state ATP phosphorylation reaction (steps 2, 3, and 4), (ii) the dephosphorylation reaction (steps 5 and 6), and (iii) the E 2*>E1 transition (steps 7 and 1), Due to the common characteristics the ion specificities of the two ion transporting enzymes have been studied intensively. Proton-like effects of sodium on H,K-ATPase (a "Na,Kn-ATPase activity) (9) and sodium-like effects of protons on Na,K-ATPase (a "H,KM -ATPase activity) have been claimed (10), although the latter effects were not found when ATP phosphorylation was studied (11). In Na,K-ATPase, Na+ shows, besides effects of its own, K+-like properties in the absence of K+. This Na,Na-ATPase or Na-ATPase activity is the result of activation of both the ATP phosphorylation, and the dephosphorylation reaction by Na4' (12), The data regarding the effects of Na+ on H,K-ATPase is somewhat confusing. In some H,KATPase studies an identical activation by N a+ of the ATP hydrolysis reaction, a H,Na ATPase activity, has been observed (13). Similar K+like effects of Na"1 ' have been found on the rate of ATP phos phorylation (13-15), but the dephosphorylation reaction has been claimed to be either activated (16), or insensitive toward Na"1 ' (15), Furthermore, two studies (9,17) indicate that Na4* behaves more like H + and drives the enzyme to an E 1 conformation.
In preliminary experiments De Jong (18) observed that the K0 5 for ATP in the phosphorylation reaction was considerably increased by N a+. Such an effect of NaH ' cannot easily be explained when Na+ behaves like H+. It could be explained, however, when Na* behaves as a K* analog. In that case the ion activates the dephosphorylation reaction and drives the enzyme into the E 2 conformation.
With the use of ion-tight H,KATPase vesicles, where no activation of the dephosphorylation process by extravesicular cations can occur (19), and by comparing their properties with those of leaky vesicles, where such activation does occur, the effects of N a+ on the total ATPase reaction, the steady-state ATP phosphorylation level, the dephosphorylation reaction, and the E 1 E% transition were investigated. The results show that N a+ displays K'Mike actions under those reaction condi tions, thus activating the dephosphorylation process at the luminal side of the membrane and driving the enzyme into an E 2 conformation by interacting at the cytosolic side.

MATERIALS AND METHODS
H>K'ATPase Preparations-Gastric H,K-ATPase was purified from pig gastric mucosa as reported previously (20). Fresh (ion-tight) H,K-ATPase vesicles were collected at the 0.25 M sucrose and 7% Ficoll (w/v) in 0.25 M sucrose interface and stored at 4 °C. Leaky vesicles were prepared by diluting these ion-tight vesicles in 20 mM Tris acetate (pH 7.0), followed by centrifugation (100,000 x g), resuspension, in water and freeze drying. This preparation was stored at -20 °C in 0.25 M sucrose, 50 mM Tris acetate (pH 7.0). In the experiments in which ion-tight H,K-ATPase vesicles were used the osmolarity was kept con stant with 0.25 M sucrose and 50 mM Tris acetate (pH 7.0).
Protein Determination-Protein was determined with the Bio-Rad protein assay (21) using bovine serum albumin as a standard. All data was expressed in Lowry protein values which are 1.5 times higher than the Bio-Rad values (20).
K +or Na*-activated. Hydrolysis of ATP-The K+-and Na+-activated ATPase activities were determined with a radiochemical method. For this purpose 0.1-22 pg of H,K-ATPase was added to 100 pi of medium, which contained 0.01 pM to 5 mM [y-32P]MgATP (specific activity 0.15-200 mCi'inmol"1), 0.1-5 mM MgCl2, 0.1 mM ouabain, 20-50 mM Tris-HC1 (pH 7.0), and varying concentrations of either KC1 or NaCl, After incubation for 1-30 min at 37 °C the reaction was stopped by adding 500 p\ of ice-cold 10% (w/v) charcoal in 6% (w/v) trichloroacetic acid and after 10 min at 0 °C the mixture was centrifuged for 10 s (10,000 X g)t To 0.15 ml of the clear supernatant containing the liberated inorganic phosphate (32Pi), 3 ml of OptiFluor (Canberra Packard, Tilburg, The Netherlands) was added and the mixture was analyzed by liquid scin tillation analysis. In general, blanks were prepared by incubating the enzyme in the absence of KC1 but in the presence of 0.1 mM SCH 28080.1 Steady'State ATP Phosphorylation Level-H,K~ ATPase (5-50 pg ml"1) was incubated at 22 °C for 3-10 s in 100 p\ of medium containing 20 ¡jM [y-32P]ATP (specific radioactivity 0.03-0.3 Ci mmol"1, Radio chemical Centre, Amersham, UK), 0.12 mM MgCl2, and 50 mM Tris acetate (pH 7.0). The reaction was stopped by adding 5 ml of 5% (w/v) trichloroacetic acid in 0.1 M phosphoric acid. After filtration over a Schleicher & Schüll filter (type ME28, 1.2-jum pore width, Dassel, Germany) and washing with stopping solution, the a2P-protein content was determined. Blanks were prepared by denaturing the enzyme prior to incubation with the phosphorylation medium (20).
The E }<->E2 Transition-Under iso-osmotic conditions ion-tight II,IC-ATPase vesicles were incubated at room temperature in 0.25 M sucrose, 20 mM Tris acetate (pH 7.0), 0.1 mM MgCl2, and NaCl, KC1, or choline chloride in the concentrations indicated. After 10 s the ATP phospho rylation capacity was determined by incubation for another 3 s in the presence of 50 nM [*y-32P]ATP. The phosphoenzyme complex was col lected as described above, Déphosphorylation Studies-After 10 s phosphorylation with 1 ¡iU [y-32PlATP (see above), 10 volumes of 20 pU non-radioactive ATP and the ligand to test were added and incubated for 3-10 s at room temper ature. The reaction was stopped by adding 5 ml of stopping solution. The amount of [32P]phosphoenzyme was determined as described above. Dephosphorylation is expressed as the decrease in acid-stable phosphoenzyme during the incubation period (% hydrolysis).
Sodium and Potassium Determinations-The N a+ and K+ contents of the enzyme preparations and media were determined flame photo metrically (FCM 6343, Eppendorf, Hamburg, Germany). When the K+ levels were determined in 125 mM NaCl solutions, the recovery of KC1 was checked by the addition of 10 pM KC1 as an internal standard. The recovery of K+ was 10.6 ± 1.6 pu (mean ± S.D., n = 6). The K concentrations of 125 mM CsCl, LiCl, NaCl, NH4C1, RbCl, and choline chloride in 20 mM Tris acetate (pH 7.0) were 6.8, 2.5, 4.8, 5.2, 320, and 3. tor giving the half-maximal activation or phosphorylation level and the I50 as the value giving 50% inhibition of the activity or level. From data on the phosphorylation level (E-P) and ATP hydrolysis rate, ut the apparent dephosphorylation rate constant, k (turnover number), was calculated using the equation: v -k' [E~P] (22).

RESULTS
The Effect o f NaCl on the Steady-state ATP Phosphorylation Level- Fig. 2A shows the combined effects of ATP and NaCl on the steady-state ATP phosphorylation level of gastric H,K-ATPase. At 22 °C and pH 7.0, in the absence of NaCl the ATP affinity was very high, the K0 5 being about 0,01 p,m (19). Upon increasing the [Na+], the ATP affinity decreased. The maximal phosphorylation level tended to be slightly reduced at higher [Na+], indicating that the inhibition is not simply a competition between N a+ and ATP. If the effect of Na+ on the steady-state ATP phosphorylation was compared with its effect on the over all ATP hydrolysis rate, at 37 °C, and in the presence of 20 pu ATP ( The Effect o f Ouabain and SC H 28080 on the Na* Activa tion-The N a+-activated ATP hydrolysis rate, measured at 37 °C in the presence of 50 mM NaCl, 20 /xm ATP, 0.12 mM MgCl2, and 20 mM Tris acetate (pH 7.0), was totally insensitive toward the specific Na,K-ATPase inhibitor ouabain, indicating that the activation of the ATP hydrolysis is not due to contam ination with Na,K-ATPase. Moreover, the specific H,KATPase inhibitor SCH 28080 inhibited under these conditions the ATP hydrolysis by nearly 95% (I50 ~ 0.08 p M ). Either in the presence of 1 mM KC1 instead of 50 mM NaCl, or in the presence of both NaCl and KC1 the I50 value for SCH 28080 increased to 0.2 /xm, probably due the antagonism between SCH 28080 and K+. Although ouabain (1 mM) did not change the inhibition profile of SCH 28080, it was included in most experiments to ensure that any contaminating Na,K-ATPase activity was blocked. Corrections for the basal Mg-ATPase activity, which is the activity in the absence of K* or Na* and in the presence of 0.1 mM SCH 28080, were also made.
Comparison o f the N a*-and K + -activated ATPase Activity of H,K-ATPase-The properties of the overall H,K-ATPase activ ity depend on the conditions in which the assay is performed (19).  The extra addition of K+ to NaCl media was completely recov ered, indicating that high [Na+] did not disturb the K+ deter mination. Since K+ at these concentrations had hardly any effect, this finding already suggests that the N a+ effect is not due to contaminating K+. Second, with the use of the K * " ionophores, valinomycin and nigericin, the K* activation of ATPase activity was studied in ion-tight vesicles. In this type of H,K-ATPase vesicles the K+ activation site is located intravesicularly (20). Fig. 4A shows that in these vesicles the basal (Mg-ATPase) activity was very low and that activation by extravesicular (cytosolic) K* was not possible. In the presence of nigericin, a for H + exchanger, the K+ activation profile was nearly identical to that of a leaky H,K-ATPase preparation, in which the K* activation site is freely accessible (Fig. 3A). In the presence of the specific K+ ionophore valinomycin, however, there was only a slight acti vation, probably due to the ionophore-induced voltage differ ence across the vesicle membrane (23), The lack of activation could partially be overcome by the extra addition of the protonophore CCCP. When similar experiments (Fig. 4 B) were carried out in the presence of N a+, activation of the ATP hydrolysis was only observed in the presence of nigericin, which ionophore can also, but to a lesser extent, exchange N a+ for H+ (23). Valinomycin either alone or in combination with CCCP could not induce a N a+-activated production due to the absolute selectivity of this ionophore for K+.
These observations indicate that the activation of the ATP hydrolysis by Na+ is not due to a contamination by K4-, but that Na+ itself activates the dephosphorylation process at the lumi nal (intravesicular) side of the membrane.
The Effects ofNa* and K + on the Dephosphorylation Reac tion-The dephosphorylation reaction was studied in leaky H,K-ATPase preparations. Fig. 5A shows that both N a+ and K+ enhance its rate, with an apparent K Q B of 10 and 0.02 mM, respectively (ratio Na+/K+ = 500). Choline chloride had no effect on the dephosphorylation process, excluding effects of ionic strength.
In ion-tight vesicles Na+ was unable to activate the de phos phorylation reaction. In the presence of nigericin the dephos phorylation reaction was again activated. This shows directly that the activating Na+ site is located intravesicularly. Ex travesicular (cytosolic) Na+ did not change the kinetics of the K+-activated dephosphorylation reaction (studied with valino mycin and CCCP, data not shown).
Helmich- De Jong et al (24), showed that ATP inhibits the basal and the K *-activated dephosphorylation reaction. With the use of ion-tight vesicles, in combination with valinomycin and CCCP it was possible to show that the site of inhibition is located at the extravesicular (cytosolic) side of the membrane. Parallel to K*, the N a+~activated dephosphorylation rate was also reduced in the presence of ATP (data not shown). (19), as only the E t enzyme can be phosphorylated by ATP. Fig. 5B shows that both Na+ and K * t but not choline chloride, reduce the phosphorylation level at suboptimal ATP concentrations. In the presence of either 0.4 mM K* or 35 mM N a+ (ratio NaH 7K+ = 88) the amount of phosphoenzyme obtained was reduced by 50%. Other related monovalent cations such as Tl+, Rb+, NHJ, Cs+, and Li+ had I50 values of 0.005, 0.4, 0.5, 20, and 65 mM, respectively. This result indicates that N a+, like K+ and the other monovalent cations, drives the enzyme to the E% conformation.

The Effect o f N a* and K* on the E 1<r*E2 Transition-In closed vesicles no activation of the luminal K+ (or Na+)-site by extravesicular ligands can occur, see above and Ref. 20. At low ATP concentrations it is feasible to study the effects of these ligands on the E x<r>E2 transition by determining the steadystate ATP phosphorylation level
E^E q Studies at p H 8.0-When the extravesicular proton concentration in these experiments was reduced 10 times, by ;ing the pH to 8.0, the affinity for ATP in the steady-state ATP phosphorylation reaction decreased, K Q 5 = 0.2 jam (Fig. 6) compared to 0.01 ¡j m at pH 7.0 (see Fig. 2A, Ref. 19). Fig. 6 shows, in addition, that N a+ did not increase the phospho en zyme level at suboptimal ATP concentrations, demonstrating the absence of any H*-like properties of N a+ under these conditions. N a+ only led to a decrease in the steady-state phosphorylation level (I50 values for N a+ were 6, 18, and 40 mM in the presence of 0.2, 2, and 20 ¡j m ATP, respectively), under lining once more the ATP/Na"h antagonism.

DISCUSSION
In this study data is presented which clearly shows that Na'1 ' ions behave like K+ ions in the H,K-ATPase reaction cycle, Na'1 * activates the dephosphorylation reaction (steps 5 and 6, Fig. 1 be phosphorylated and the dephosphorylation reaction cannot be activated by the ligands Na+ or K+. It was found that Na*, like K* (19), reduces the phosphorylation capacity of the en zyme, meaning that Na* shifts the equilibrium to the E 2 con formation (reactions -1 and -7 ), Even at low H* concentra tions no increase in the steady-state ATP phosphorylation level, measured at suboptimal conditions, could be observed, excluding H*-like properties of Na+. The observation that cy tosolic Na+ has K*-like properties, driving the enzyme to the E 2 form, is in line with the inhibition of the ATP phosphoryl ation rate (13-15) and the reduction of the H*-transport rate (15,25,26) by this ion. Variable effects of N a+, in the absence of K*, on the overall H,K-ATPase activity (steps 1-7) have been reported (13, 15). These variations might be due to differences in assay condi tions, since we demonstrate that high concentrations of ATP (Fig. 3B), Mg2*, and imidazole have marked effects on the Na (and the K* (19)) affinity in the overall ATPase reaction. More over, high [Na*] inhibits the latter activity. In the overall ATPase experiments we were able to show that Na+, like K*, at relative low concentrations activated the hydrolysis of ATP, via the dephosphorylation reaction (steps 5 and 6), and inhib ited the ATPase reaction at high concentrations by driving the E 1^E Z equilibrium to the right (steps -1 and -7). The com bination of both effects explains the increasing effect of Na* on the K 0 5 for ATP in the phosphorylation reaction (Fig. 1A).
In both Na,K-ATPase and H,K-ATPase, K* activates the dephosphorylation reaction. The role of K* can be performed in both enzymes by Na* (Refs. 27 and 28 and this study), al though the affinity for Na* is much lower than that of K*. In both enzymes K* also drives the equilibrium E 1«*E2 to the E 2 form, whereas Na4" (for Na,K-ATPase) and H* (for H,K~ ATPase) shifts the equilibrium to the E 1 form. The present study shows that with H,K-ATPase Na'1 " can perform the latter role of K+, but not that of H*. With Na,K-ATPase there is no indication for an E 2 promoting effect of Na4* in the absence of K*. The ion specificity of Na* and H* as E 1 promoters in Na,K-ATPase and H,K-ATPase, respectively, is much more prominent. Neither an effect of H* on the steady-state phos phorylation level of Na,K-ATPase (11) nor of Na*1 ' on this pa rameter of H,K-ATPase (this study) was observed.
The data seems to conflict with studies by Rabon et aL (17), who used a fluorescein isothio cyan ate-labeled H,K-ATPase preparation to test the effects of Na*. The fluorescence of this modified enzyme, incapable of being phosphorylated by ATP, increased in the presence of Na*. Although an antagonism between H* and Na* was observed, the increase in fluores cence was interpreted as an increase in the E j form of the enzyme, analogous to that with Na,K-ATPase. It has not been proven, however, that an increase in fluorescence under these