Mechanism of the control of (Na+ + K+)-ATPase by long-chain acyl coenzyme A.

Long-chain fatty acid esters of CoA activate (Na+ + K+)-ATPase (the sodium pump) when ATP is suboptimal. To explore the nature of the interactions of these CoA derivatives with the pump, reversible effects of palmitoyl-CoA on the purified membrane-bound kidney enzyme were studied under conditions where interference from the irreversible membrane-damaging effect of the compound was ruled out. With 50 microM ATP, while saturating palmitoyl-CoA increased (Na+ + K+)-ATPase activity, it caused partial inhibition of Na+-ATPase activity without affecting the steady-state level of the phosphoenzyme. Palmitoyl-CoA did not change the K0.5 of ATP for Na+-ATPase, but it altered the complex Na+ activation curve to suggest the antagonism of the low-affinity, but not the high-affinity, Na+ sites. At a low ATP concentration (0.5 microM), K+ inhibited Na+-ATPase as expected. In the presence of palmitoyl-CoA and 0.5 microM ATP, however, K+ became an activator, as it is at high ATP concentrations. The activating effect of palmitoyl-CoA on (Na+ + K+)-ATPase activity was reduced with increasing pH (6.5-8.5), but its inhibitory effect on Na+-ATPase was not altered in this pH range. The data show two distinct actions of palmitoyl-CoA: 1) blockade of the extracellular "allosteric" Na+ sites whose exact role in the control of the pump is yet to be determined, and 2) activation of the pump through increased rate of K+ deocclusion. Since in their latter action the fatty acid esters of CoA are far more effective than ATP at a low-affinity regulatory site, we suggest that these CoA derivatives may be the physiological ligands of this regulatory site of the pump.

ATPase activity was reduced with increasing pH (6.5-8.5), but its inhibitory effect on Na+-ATPase was not altered in this pH range. The data show two distinct actions of palmitoyl-CoA: 1) blockade of the extracellular "allosteric" Na+ sites whose exact role in the control of the pump is yet to be determined, and 2) activation of the pump through increased rate of K+ deocclusion. Since in their latter action the fatty acid esters of CoA are far more effective than ATP at a lowaffinity regulatory site, we suggest that these CoA derivatives may be the physiological ligands of this regulatory site of the pump.
(Na+ + K+)-ATPase (the sodium pump) is the intrinsic enzyme of the plasma membrane that carries out the coupled active transport of Na+ and K+ in eucaryotic cells. Our studies with a purified preparation of the enzyme indicated that the CoA esters of long-chain fatty acids increased enzyme activity at suboptimal, but not optimal, ATP concentrations (1). More recently, we also showed that in the presence of suboptimal ATP, these CoA esters stimulated ATP-dependent Na+/K' exchange in cardiac sarcolemmal vesicles (2) and resealed red cell membranes (3). In the latter system, it was demonstrated that the activating effects of these compounds on the pump were exerted from the intracellular, but not the extracellular, side of the membrane (3), hence, providing further support * This work was supported by National Institutes of Health Grant HL-36573 awarded by the National Heart, Lung, and Blood Institute, United States Public Health Service/Department of Health and Human Services. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore he hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
for the proposition that intracellular CoA esters of fatty acids may be involved in the physiological control of the sodium pump (1-3). The studies presented here were done with the purified enzyme to learn more about the mechanism of this control. The findings indicate that the CoA derivatives activate the pump by increasing the rate of release of K' from the K+-occluded state of the enzyme. The data also show additional effects of CoA esters on low-affinity Na' sites whose precise role in the control of the pump is not known.

EXPERIMENTAL PROCEDURES
The purified membrane-bound enzyme of canine kidney was prepared as described above (4). The specific activities of the various preparations used here were in the range of 1000-1500 /*mol of ATP hydrolyzed per mg of protein/h. Assay of ATPase activity was done at 37 "C through the determination of the initial rate of release of ture for Na+-dependent ATPase assay contained 100 mM Na+, 5 mM M e , 1 mM EGTA,' the indicated ATP and palmitoyl-CoA concentrations, 50 mM Tris-HC1 (pH 7.4), and 2-15 pg/ml enzyme protein.
For (Na+ + K+)-dependent ATPase assay, the reaction mixture contained the same components plus 20 mM K'. Unless indicated otherwise, the duration of assay was 60 s or less, and the weight ratio of palmitoyl-CoA to enzyme protein was not greater than 6. Under these conditions, there was no irreversible inactivation of the enzyme (see "Results"). The level of phosphoenzyme obtained in the presence of ATP was determined as described before (4). In experiments where pH was varied, the buffers were 50 mM Mes (pH 6.5); 25 mM Mes, 25 mM Tris (pH 7.0), and 50 mM Tris-HC1 (pH 7.4-8.5). [y-3ZP]ATP was obtained from Du Pont-New England Nuclear; "vanadate-free" ATP and palmitoyl-CoA were purchased from Sigma. 32 Pi from [y-32P]ATP. Unless otherwise specified, the reaction mix-

RESULTS
Reversible and Irreversible Effects of Palmitoyl-CoA on Enzyme-The activity of the purified membrane-bound enzyme used here can be irreversibly inhibited by high concentrations of all detergents including the amphiphilic CoA esters of fatty acids. Although our previous experiments had indicated that the stimulatory effects of these compounds on (Na' + K+)dependent ATPase activity were due to their readily reversible interactions with the extramembranous segments of the enzyme protein (1,5 ) , in order to study the mechanism of this reversible effect in more detail, it was necessary to define the conditions under which interference from the irreversible membrane-damaging effects could be ruled out. In experiments of Fig. l a , the effect of the palmitoyl-CoA:enzyme ratio on time-dependent irreversible inhibition of the enzyme was studied. Several different concentrations of the enzyme were preincubated with a fixed concentration of palmitoyl-CoA for varying periods up to 20 min; samples were then diluted 10fold into identical media containing substrates and assayed in less than 60 s for (Na+ + K')-dependent ATPase activity.
The data showed that progressive time-dependent inhibition Further experiments showed that the irreversible inhibitory effect of palmitoyl-CoA on the enzyme was antagonized by salts. When such effects of the usual salts of the assay medium (MgC12, NaC1, and KC1) were examined in more detail, it became evident that this protective effect was not a simple function of ionic strength. The results of these experiments, only some of which are presented in Fig. lb, indicated the following. 1) M F offered the most effective protection against irreversible inhibition; maximal protection was obtained in the range of 5-10 mM MgC12. 2) Whereas Na+ and K+ provided little protection when used alone, they were synergistic with suboptimal MgZ+ concentrations. There were no differences between Na+ and K+, and the maximal synergistic effect of each was obtained at 20-30 mM.
Based on the above findings, it became possible to choose the experimental conditions described under "Experimental Procedures" (i.e. 2-15 pg of enzyme/ml assay, 5 mM MgC12, palmitoyl-CoA:enzyme ratio not greater than 6, and assay periods of less than 60 s) to prevent the occurrence of significant irreversible effects of palmitoyl-CoA in all subsequent experiments. The importance of avoiding irreversible enzyme inactivation by palmitoyl-CoA becomes evident below when both activating and inibitory effects of the compound are described. Needless to say, any irreversible inactivation can only lead to the underestimation of the magnitude of an observed activating effect. On the other hand, it is essential to know whether or not an observed inhibitory effect is distinct from irreversible inactivation. That such inactivation is indeed avoided under the prescribed experimental conditions and that there are reversible inhibitory effects of palmitoyl-CoA on the enzyme will be demonstrated directly in the experiments of Fig. 3.
Comparison of Palmitoyl-CoA Effects on (Nu+ + K+)and Na+-dependent ATPase Activities-Our previous experiments showing that CoA esters of fatty acids decreased KO.5 of ATP for (Na' + K+)-dependent ATPase activity suggested that these compounds mimic the effect of ATP at a low-affinity ATP site within the reaction cycle of the enzyme (1,2). Since there have been disagreements on whether or not the highand low-affinity ATP sites are the same (6, 7) and since only a high-affinity ATP site is apparent in the expression of the Na+-dependent ATPase activity of the enzyme, it became of interest to examine the acyl-CoA effects on this activity. In the experiments of Fig. 2, the effects of varying concentrations of palmitoyl-CoA on Na+-dependent and (Na+ + K+)-dependent activities at the same substrate concentration (50 p M ) were compared. In agreement with previous results (l), (Na+ + K+)-dependent activity increased more than %fold. The Na+-dependent activity, however, was inhibited by palmitoyl-CoA to the maximal extent of 60-70%. Of particular interest was this partial nature of the inhibition (Fig. 2), suggesting that ATP and palmitoyl-CoA are not mutually exclusive.
Reversibility of Palmitoyl-CoA Inhibition of Nu+-dependent ATPase Activity-Although due to the choice of assay conditions, the inhibition of Na+-dependent activity by palmitoyl-CoA (Fig. 2) was not likely to be due to the irreversible membrane-damaging effect of the compound (see above), it was important to establish this point unambiguously. In the experiments of Fig. 3, the enzyme that was placed in the medium for the assay of Na+-dependent activity was inhibited to the extent of about 60% by the simultaneous addition of palmitoyl-CoA. After 60 s, K+ was added to the medium, and (Na+ + K+)-dependent activity was assayed for 20 s. This activity was more than 90% of the same activity of a control enzyme sample that had not been preincubated with palmitoyl-CoA (Fig. 3). Had the initial 60% inhibition of the Na+dependent activity been due to irreversible inactivation, the subsequent (Na+ + K+)-dependent activity could not have been more than 40% of the control. These data clearly indicate the reversibility of the inhibition of the Na+-dependent activity. inhibitor of Na+-dependent activity (Fig. 2). When the effect of 30 PM palmitoyl-CoA on the steady-state level of acidstable phosphoenzyme obtained from 50 PM ATP in the presence of 100 mM Na' and 5 mM M P was determined at 24 "C in four separate experiments, the control level was 2.1 f 0.2 nmol of Pi/mg of protein and that in the presence of palmitoyl-CoA was 2.3 f 0.2 nmol of Pi/mg of protein. These data, in conjunction with the partial nature of the inhibition of Na+-dependent ATPase caused by palmitoyl-CoA (Fig. 2), clearly show that palmitoyl-CoA and ATP (at the high-affinity catalytic site) may bind simultaneously and without significant interactions.
The experiments of Fig. 4 showed a dramatic effect of palmitoyl-CoA on the Na+ activation curve of the Na+-dependent ATPase activity. The curve for the control enzyme had an intermediate plateau in the range of 5-10 mM Na+, as has been noted repeatedly (8). Palmitoyl-CoA had no significant effect on activity at Na+ concentrations below this plateau, but it inhibited at Na+ concentrations above the plateau (Fig. 4). below the Koa value for the low-affinity ATP site, K+ inhibits Na+-dependent ATPase activity because the release of K+ from the enzyme is too slow without ATP being at the lowaffinity site (7). The experiments of Fig. 5 showed that palmitoyl-CoA overcame such an inhibitory effect of K+ completely, providing further evidence for an effect of the CoA ester similar to that of ATP at its low-affinity site.

Effect of Palmitoyl-CoA on K+ Inhibition of Nu+-dependent ATPase Activity-When ATP concentration is considerably
Effects of Palmitoyl-CoA at Different pH Values-The rate of release of K+ from the enzyme (or the rate of deocclusion of K+) is increased not only by ATP at the low-affinity site, but also by an increase in pH (9,12). It was of interest, therefore, to see how pH affected the palmitoyl-CoA activation of (Na' + K+)-dependent activity at suboptimal ATP concentrations. Experiments of Fig. 6, in which this activity was measured at 50 PM ATP, showed that ( a ) the activity of the control enzyme increased with pH in the range of 6.5-8.5; ( 6 ) whereas palmitoyl-CoA increased activity at all pH values, its effect was greater at lower pH values; and ( c ) in the presence of saturating palmitoyl-CoA, an increase in pH had no activating effect. Evidently, palmitoyl-CoA, ATP at the low affinity site, and deprotonation of the enzyme have similar effects on the rate of Kt deocclusion. The maximal inhibitory effect of a saturating palmitoyl-CoA concentration on Na+dependent ATPase activity (about 60% inhibition) was not significantly affected by a change of pH within the above range (data not shown).

DISCUSSION
These results (Figs. 5 and 6) provide further support for our previous suggestion (1, 2) that the activating effects of a fatty acyl-CoA on (Na+ + K+)-dependent ATPase and ATPdependent Na+/K+ exchange are due to its ATP-like effect on the rate of release of K+ from the K+-occluded conformation of the enzyme within the Albers-Post cycle. Because this effect of the CoA derivative is observed in the presence of low concentrations of ATP, on the basis of our experiments alone, we cannot say whether the acyl-CoA is binding to the lowaffinity site or whether it binds elsewhere but increases the affinity of ATP at the low-affinity ATP site. In either case, however, it is difficult to reconcile our results with a common notion that high-and low-affinity ATP sites are expressions of ATP binding to the same site on two mutually exclusive conformational states because palmitoyl-CoA seems to have no effect on ATP at the high-affinity site (see Fig. 2 and "Results"). Accumulated evidence in favor of the multiplicity of the ATP sites and the possible existence of distinct catalytic and regulatory sites have been discussed before (6). These data, as well as the apparent duplication within the a-subunit of a region that binds ATP analogues (10, Fig. 2), respectively, one may argue legitimately that the physiological ligand for the regulatory site of the pump may be a long-chain fatty acyl-CoA rather than ATP.
The Na+-dependent ATPase activity of a purified and unsided enzyme preparation is related to the ATP-dependent Na+/Na+ exchange by a sided preparation in the absence of K+ and under conditions where extracellular Na+ has replaced extracellular K' (13). Therefore, the inhibitory effect of palmitoyl-CoA on Na+-dependent ATPase activity (Figs. [2][3][4] clearly indicates that the regulator has additional effects on the pump beyond those on K+ deocclusion. The nature of these effects, however, is less clear. There is ample evidence (8) to indicate that the complex Na' activation curve of the Na+-dependent ATPase of the purified enzyme involves multiple intracellular and extracellular Na+ sites of the pump and that the low-affinity Na+ sites whose effects are blocked by palmitoyl-CoA (Fig. 4) are of extracellular nature. It is not clear, however, if these low-affinity Na' sites are purely allosteric (14) or if they consist of both regulatory and transport sites (15). There are also disagreements on how Na'/ Na' and Na+/K' exchange may be controlled by the regulatory Na' sites (14)(15)(16), and there are apparent discrepancies concerning whether the low-affinity ATP site has a role in the operation of ATP-dependent Na+/Na+ exchange (13,(17)(18)(19). Future studies on the effects of CoA derivatives on Na+dependent ATPase and ATP-dependent Na+/Na+ exchange in sided preparations and on the properties of the phosphoenzyme intermediates of these activities will be helpful in clarifying the above uncertainties.