Because the activity of the sodium pump (Na-K-ATPase) influences the secretion of aldosterone, we determined how extracellular potassium (Ko) and calcium affect sodium pump activity in rat adrenal glomerulosa cells. Sodium pump activity was measured as ouabain-sensitive 86Rb uptake in freshly dispersed cells containing 20 mM sodium as measured with sodium-binding benzofluran isophthalate. Increasing Ko from 4 to 10 mM in the presence of 1.8 mM extracellular calcium (Cao) stimulated sodium pump activity up to 165% and increased intracellular free calcium as measured with fura 2. Increasing Ko from 4 to 10 mM in the absence of Cao stimulated the sodium pump ∼30% and did not increase intracellular free calcium concentration ([Ca2+]i). In some experiments, addition of 1.8 mM Cao in the presence of 4 mM Ko increased [Ca2+]i above the levels observed in the absence of Cao and stimulated the sodium pump up to 100%. Ca-dependent stimulation of the sodium pump by Ko and Cao was inhibited by isradipine (10 μM), a blocker of L- and T-type calcium channels, by compound 48/80 (40 μg/ml) and calmidizolium (10 μM), which inhibits calmodulin (CaM), and by KN-62 (10 μM), which blocks some forms of Ca/CaM kinase II (CaMKII). Staurosporine (1 μM), which effectively blocks most forms of protein kinase C, had no effect. In the presence of A-23187, a calcium ionophore, the addition of 0.1 mM Cao increased [Ca2+]i to the level observed in the presence of 10 mM Ko and 1.8 mM Cao and stimulated the sodium pump 100%. Ca-dependent stimulation by A-23187 and 0.1 mM Cao was not reduced by isradipine but was blocked by KN-62. Thus, under the conditions that Ko stimulates aldosterone secretion, it stimulates the sodium pump by two mechanisms: direct binding to the pump and by increasing calcium influx, which is dependent on Cao. The resulting increase in [Ca2+]i may stimulate the sodium pump by activating CaM and/or CaMKII.
- calcium- and calmodulin-dependent kinase II
increasing extracellular potassium (Ko) has been thought to stimulate aldosterone secretion by depolarizing the membrane potential, which opens voltage-dependent calcium channels and leads to an influx of extracellular calcium (Cao; see Refs. 5,6, 9, 26, 29). The depolarizing effect of Ko is in large part due to a decrease in the equilibrium potential for potassium (27). However, increasing Ko over the range that stimulates aldosterone secretion should also stimulate the sodium pump by increasing the amount of Ko that binds to the pump (14). Because sodium pump activity is ultimately responsible for most of the resting membrane potential, the ability of Ko to regulate aldosterone secretion could in part be due to its effect on sodium pump activity. In fact, it has been clearly shown that changes in sodium pump activity can both stimulate and inhibit aldosterone secretion, depending on the species studied and the experimental conditions (4, 16, 31, 35).
The addition of physiological concentrations of Caostimulates the sodium pump even at normal levels of Ko(16). The mechanism of this stimulation is not yet known but could be due to an increase in intracellular free calcium (Cai 2+), which in some cells has been shown to activate sodium pump activity (1, 23, 24, 34). However, if this is the case, then one also might expect Ko to stimulate the sodium pump by increasing calcium influx through voltage-dependent calcium channels. Such an effect has not yet been reported.
Thus, in this study, we have examined how Cao and Ko affect the activity of the sodium pump under conditions that they stimulate the secretion of aldosterone. These studies will help determine how aldosterone secretion may be affected by changes in the activity of the sodium pump and will help define mechanisms that regulate sodium pump activity during cell activation.
MATERIALS AND METHODS
A-23187, KN-62, calmidazolium, compound 48/80, gramicidin A, and monensin were purchased from Calbiochem (San Diego, CA). 86Rb was purchased from NEN Research Products (Boston, MA). Isradipine was a gift from Sandoz Pharmaceuticals (East Hanover, NJ).
A stock solution of 500 μM monensin was prepared by adding monensin to medium 199 in 0.5% DMSO at 37°C while vortexing. The solution was maintained at 37°C and alternately was shaken and vortexed until the monensin was dissolved. KN-62 and isradipine were dissolved in DMSO.
Adrenal cell preparation.
Adrenal capsules, containing the zona glomerulosa, were removed from female Sprague-Dawley rats weighing 200–224 g and collagenase dispersed as previously described (7) in medium 199 (GIBCO, Grand Island, NY) containing modified Earle's salts (130 mM NaCl, 4.0 mM KCl, 1.8 mM CaCl2, and 0.8 mM MgCl2), 10 mM HEPES, sodium salt (pH 7.4), 0.2% BSA, and no bicarbonate. After dispersion, the cells were incubated for 2 h to allow for recovery. Cells were continuously gassed with 100% O2.
In experiments where Cao was either 0 or 1.8 mM (see Figs.3, 6, and 8), cells were prepared as usual except that they were divided into two equal groups after dispersion. One group was subsequently washed two times and resuspended in medium 199 without added calcium, and the other was suspended in normal medium 199 containing 1.8 mM CaCl2. Soon thereafter, the cells were added to the assay, and the experiments were performed as usual. In the experiments shown in Fig. 7, cells were washed two times in medium 199 without calcium and then were suspended in medium 199 containing 20 μM EGTA and either no added CaCl2 (0 μM free Ca2+) or 120 μM CaCl2 (100 μM free Ca2+). They were then immediately used to measure sodium pump activity.
Sodium pump activity.
Sodium pump activity was measured as the difference in the uptake of86Rb, a congener for potassium, in the presence and absence of ouabain, a specific inhibitor of the sodium pump. Measurements were carried out at 37°C over a 5-min period in medium 199 as previously described (16). Quadruplicate samples were run in the presence and absence of 1 mM ouabain. Experiments were performed in a total volume of 150 μl containing 100,000 cells and from 1 to 2.5 μCi of 86Rb, depending on the experiment. Measurements were carried out in a 96-well plate (cat. no. MADPN6550; Millipore) gently agitated at constant intervals and maintained at 37°C on top of a heated sand bath. Cells were added to medium 199, except where noted, followed shortly thereafter by the addition of ouabain or an equal volume of buffer (time 0). Monensin, a sodium ionophore, was added at a final concentration of 10 μM at 20 min;86Rb was added at 30 min. Fluxes were terminated 5 min later by rapid filtration of the cells using a Millipore Multiscreen Assay System followed by six washes (∼300 μl/wash) with ice-cold medium 199. The membranes (containing the washed cells) were then punched out and separately placed in 0.5 ml of 1% SDS to which scintillation fluid was then added. The filter blank in this assay was <0.015% of the total counts.
KN-62 and isradipine were added to the assay 15 and 30 min before86Rb, respectively. A-23187 was added 5 min before86Rb. In the experiments when Ko was increased from 4 to 10 mM, a small volume of concentrated KCl was added to the normal medium 5 min before the addition of 86Rb. In experiments where sodium pump activity was assayed at both 4 and 10 mM Ko, ouabain-sensitive 86Rb uptake at 10 mM Ko was multiplied by 2.5 to compensate for the difference in the ratio of 86Rb to Ko.
Measurement of intracellular sodium.
Glomerulosa cells were prepared as described above and then were incubated with shaking under oxygen at room temperature for 45 min in medium 199 (minus phenol red) containing 5 μM of the acetoxymethyl ester of sodium-binding benzofluran isophthalate (SBFI) and 0.03% pluronic following previously developed procedures (17). The cells were then washed two times by centrifugation and resuspended in the same medium without SBFI and pluronic.
To calibrate the SBFI as a function of sodium, cells were loaded with SBFI as described above. Cells from one preparation were then divided into five equal groups and centrifuged, and each group was resuspended separately in one of five solutions, each containing different concentrations of sodium. Cells were then centrifuged again and resuspended in the same solution. The five solutions contained 0, 10, 20, 30, and 40 mM NaCl and the appropriate concentration of choline chloride so that the sum of the NaCl and choline chloride equaled 82 mM. In addition, all of the solutions contained a final concentration of 50 mM KCl, 1.8 mM CaCl2, 1 mM MgSO4, 1 mM KH2PO4, 5 mM glucose, 2 mMl-glutamine, 10 mM HEPES, and 4 μM gramicidin. The gramicidin was included to equilibrate sodium and potassium concentrations across the plasma membrane (17). The fluorescence of each set of cells was then measured as shown in Fig.1 A, and the concentration of sodium was calculated as previously described (33).
Measurement of [Ca2+]i.
Glomerulosa cells were prepared as described above and then were incubated with shaking under oxygen at 37°C for 45 min in medium 199 (minus phenol red) containing 3 μM of the acetoxymethyl ester of fura 2. The cells were then centrifuged and resuspended in the same medium without fura 2 and were incubated with shaking under oxygen at 37°C for 30 min. The cells were then centrifuged and resuspended in medium 199 made to contain no added CaCl2. The cells were then centrifuged again. The next step depended on the type of experiment that was performed. For the experiment shown in Fig. 4 A, the cells were resuspended in medium 199 containing 1.8 mM CaCl2 and 10 μM monensin. Immediately thereafter, the cells were put in a cuvette, and the change in fluorescence was recorded as a function of time. For the experiment shown in Fig.4 B, the cells were suspended in medium 199 containing no added CaCl2. The suspension of cells was then added to a cuvette, and the measurement of fluorescence began. As the fluorescence was measured, small volumes of concentrated KCl, EGTA, A-23187, and CaCl2 were added to the cuvette as shown (Fig. 4B). The concentration of [Ca2+]i was calculated as previously described (15).
Measurements of fluorescence.
Fluorescence was measured using a dual excitation monochrometer spectrofluormeter (SPEX Fluorolog 1680; Spex Industries, Edison, NJ). The measurements were carried out at 37°C in a stirred cuvette containing 200,000 cells/ml. The fluorescence of the SBFI was excited at 340 and 385 nm and was measured at 500 nm (17); the fluorescence of fura 2 was excited at 340 and 380 nm and was measured at 505 nm (15).
Each data point is the mean and SE of quadruplicate samples of an individual experiment. Each experiment is representative of at least three similar experiments. The data were analyzed by one-way ANOVA. Groups were compared using a Bonferroni multiple-comparisons test. Values of P ≤ 0.05 were considered statistically significant.
To measure the concentration of intracellular sodium in adrenal glomerulosa cells, the response of incorporated SBFI was first measured as a function of sodium (Fig.1 A). This calibration indicated that the initial sodium concentration in freshly dispersed cells suspended in medium 199 was 20 ± 3.5 (SD) mM sodium (Fig.1 B). This value is close to the value of 23.9 ± 1.8 mM measured by others in freshly dispersed cells and much less than the value of 48.5 ± 5.5 mM in cells activated by ANG II (33). Cultured glomerulosa cells have a lower sodium concentration of 9.2 ± 3.5 mM, perhaps due to recovery from deleterious treatment during enzymatic dispersion (33). Adding monensin to freshly dispersed cells only modestly increased the intracellular sodium (Fig. 1 B). Thus, under these experimental conditions, intracellular sodium was well within the physiological range for adrenal glomerulosa cells, even in the presence of monensin.
Sodium pump assay conditions.
Sodium pump activity was measured in freshly dispersed cells in the presence of monensin, as previously described (16). Monensin was added to assure that the sodium pump activity did not change during the assay due to a reduction in intracellular sodium, which is a primary pump substrate. Under these conditions, the uptake of 86Rb was linear for at least the first 6 min in the presence and absence of ouabain (Fig.2 A). Therefore, the ouabain-sensitive uptake, which is a measure of sodium pump activity, was also linear during this period (Fig. 2 B). In most of the experiments, the ouabain-sensitive 86Rb uptake accounted for >85% of the total uptake; in all experiments, it accounted for at least 70% of the total influx.
Effects of Ko and Cao on sodium pump activity and [Ca2+]i.
In the absence of Cao, increasing Ko from 4 to 10 mM stimulated the sodium pump 30% (Fig.3). This is the degree of activation one would expect from increased binding of Ko to its extracellular binding site on Na-K-ATPase (20). In the presence of 1.8 mM Cao, increasing Ko from 4 to 10 mM stimulated sodium pump activity 165% (Fig. 3).
Increasing Ko from 4 to 10 mM under these experimental conditions also increased [Ca2+]i (Fig.4 A). Under similar conditions, others have found that elevating Ko increases [Ca2+]i from 200 to 600 nM (28). Thus Ko may stimulate the sodium pump more in the presence of Cao because it increases [Ca2+]i, which stimulates the sodium pump in some cells (1, 23, 24, 34).
Adrenal glomerulosa cells contain both L- and T-type calcium channels, which are both blocked by isradipine, a 1,4-dihdropyridine (2). Thus, to test if the increased activation of Ko observed in the presence of Cao could be due to the ability of Ko to increase [Ca2+]i through voltage-dependent calcium channels, we tested the effect of Ko in the presence and absence of isradipine. For these experiments, we chose a concentration of isradipine that fully inhibits ouabain-induced aldosterone secretion (35). Isradipine significantly reduced the ability of elevated Ko to stimulate the sodium pump in the presence of Cao (Fig. 5 A). In fact, in the presence of isradipine, increasing Ko from 4 to 10 mM stimulated the sodium pump 36% (Fig. 5 A), an amount similar to that observed in the absence of Cao (Fig.3). These results suggest that Ko could have stimulated the sodium pump in a Ca-dependent fashion by increasing Ca influx through voltage-gated calcium channels.
Increasing Cao from 0 to 1.8 mM stimulated sodium pump activity up to twofold at 4 mM Ko in some preparations of adrenal glomerulosa cells (Fig. 6). Lower levels of stimulation were sometimes seen in other experiments (data not shown). In preparations of cells where Cao stimulated activity, the effect was blocked by isradipine (Fig. 6 A), suggesting that Cao stimulated the sodium pump by increasing [Ca2+]i through voltage-dependent calcium channels.
To determine if the addition of 1.8 mM Cao in the presence of medium containing 4 mM Ko could stimulate the sodium pump by increasing [Ca2+]i, the concentration of [Ca2+]i was measured in the presence and absence of 1.8 mM Cao. Cells previously loaded with fura 2 and washed in the absence of Cao were suspended attime 0 in normal medium 199 containing 4 mM Koand 1.8 mM Cao (Fig. 4 A). [Ca2+]i immediately began to increase and then reached a steady state ∼5 min later (Fig. 4 A). At this point, [Ca2+]i was 140 nM, considerably above the level of 70 nM observed when the cells were suspended at 4 mM Ko in the absence of Cao (Fig. 4 B). Thus cells suspended in normal media containing 1.8 mM Caoand 4 mM Ko have higher levels of [Ca2+]i than cells suspended in the same medium containing no added calcium. These measurements are consistent with the idea that the addition of Cao could stimulate the sodium pump by increasing [Ca2+]i in cells suspended in medium 199 containing normal levels of 4 mM K and 1.8 mM Ca. Others have shown that freshly dispersed glomerulosa cells suspended in solutions containing normal concentrations of Ko and Caohave resting levels of ∼200 nM [Ca2+]i(28). Given the results that correlate sodium pump stimulation with small increases in [Ca2+]iabove resting levels, relatively small differences in resting [Ca2+]i could account for our observation that the addition of 1.8 mM Cao stimulates the sodium pump more in some preparations than in others.
Effects of pharmacological agents.
To help determine the mechanism by which increasing [Ca2+]i might stimulate the sodium pump, we tested the effects of pharmacological agents known to mediate the effects of Cai 2+ in adrenal glomerulosa cells. Ca-dependent stimulation of the sodium pump by both Ko and Cao was blocked by compound 48/80 (40 μg/ml; see Ref.13) and calmidizolium (10 μM; see Ref. 21), which inhibit calmodulin (CaM; data not shown). These results prompted us to test the effects of drugs that inhibit more specific CaM-dependent proteins. We found that KN-62, which inhibits Ca/CaM-dependent kinase II (CaMKII; see Ref. 32), blocked the Ca-dependent stimulation of the sodium pump by both Ko(Fig. 5 B) and Cao (Fig. 6 B). These data suggest that the Ca-dependent stimulation of the sodium pump could be mediated by a CaM-dependent mechanism, perhaps by a form of CaMKII.
Because KN-62 can inhibit calcium influx through voltage-dependent Ca channels (22), one explanation for the ability of KN-62 and the other CaM antagonists to inhibit Ca-dependent stimulation of the sodium pump is by limiting calcium influx. To test this hypothesis, we measured the effect of Cao in the presence of A-23187, a calcium ionophore that would increase [Ca2+]iindependent of calcium channels. In designing these experiments, we first determined how much Cao had to be added in the presence of 10 μM A-23187 to achieve the same [Ca2+]i observed in the presence of 10 mM Ko and 1.8 mM Cao. For instance, in the experiment shown in Fig. 4 B, adding 140 μM Cao(in the presence of 10 μM EGTA) gave the same [Ca2+]i as was observed in the presence of 10 mM Ko and 1.8 mM Cao (Fig. 4 A). In other experiments, we found that under the same conditions we had to add as little as 100 μM Cao (data not shown). Using this information, we tested the effect of KN-62 on the ability of Cao to stimulate the sodium pump in the presence of A-23187. Increasing Cao from 0 to 100 μM in the presence of A-23187 stimulated the sodium pump over twofold (Fig.7 A). This Ca-dependent stimulation was completely inhibited by KN-62 (Fig. 7 A) without blocking the influx of calcium (data not shown). These data suggest that KN-62 blocks the Ca-dependent stimulation of the sodium pump independent of its ability to inhibit calcium influx through voltage-dependent calcium channels.
Finally, to test if isradipine could be blocking Ca-dependent stimulation of the sodium pump by inhibiting protein kinase C (8,18), we tested if isradipine would block Ca-dependent stimulation of the sodium pump when Ca influx was mediated by A-23187. The results show that isradipine had no effect on the ability of 100 μM Cao to stimulate the sodium pump in the presence of A-23187, which is when calcium could enter the cells independent of endogenous calcium channels (Fig. 7 B). Likewise, isradipine had no effect on the ability of KN-62 to block the Ca-dependent stimulation seen in the presence of A-23187 (Fig. 7, A vs.B). Furthermore, staurosporine, which at least partially inhibits all known forms of protein kinase C (19, 25), had no effect on the ability of Cao to stimulate the sodium pump (Fig. 8). Thus it is unlikely that isradipine was blocking the ability of calcium to stimulate the sodium pump by inhibiting protein kinase C.
Our results show for the first time how sodium pump activity is affected during Ko-induced aldosterone secretion. This is significant, because the activity of the sodium pump has been repeatedly shown to alter agonist-stimulated aldosterone secretion. We have shown that increasing Ko and Cao stimulate the sodium pump and increase [Ca2+]i in the range that stimulates aldosterone secretion. From our measurements, we deduce that increasing [Ca2+]i from ∼70 to ∼300 nM stimulated the sodium pump approximately twofold. Ko also stimulated the sodium pump ∼30% in the absence of Cao, probably due to increased binding of Koto the sodium pump. Thus increasing Ko over the range that stimulates aldosterone secretion activates the sodium pump by two mechanisms: one that is Ca-dependent and the other that is not. It is not yet known how the stimulation of the sodium pump by Koand [Ca2+]i affects the membrane potential and alters the secretion of aldosterone. Nevertheless, it is interesting to note that both of the physiological agents that stimulate aldosterone secretion alter sodium pump activity: ANG II inhibits (16) and Ko stimulates. Perhaps these opposing effects may one day help explain the apparently conflicting reports that inhibiting sodium pump activity can both stimulate and inhibit aldosterone secretion, depending on the preparation and the experimental conditions (4, 16, 31, 35).
Stimulation of the sodium pump by Cai2+.
Our results confirm that Cao stimulates the sodium pump in rat glomerulosa cells (16). Hajnoczky et al. (16), however, concluded that Cao did not stimulate by increasing [Ca2+]i because the calcium ionophore ionomycin failed to stimulate sodium pump activity. However, they added ionomycin in the presence of 1.2 mM CaCl2 (16), which may have increased [Ca2+]i beyond its stimulatory range. In contrast, we observed stimulation when A-23187 was added at much lower concentrations of Cao (Fig. 4 B).
Mechanism of stimulation.
Because the stimulatory effects of Cao, Ko, and Cai 2+ were blocked with 48/80 and calmidazolium, which inhibit CaM, and by KN-62, which inhibits CaMKII, we suggest that the stimulatory effects of [Ca2+]i could be mediated by CaMKII or a related kinase. CaMKII is present in bovine glomerulosa cells and is activated by Cai 2+ under conditions that stimulate sodium pump activity (10). In principle, Ca-CaM or Ca-CaMKII could enhance sodium pump activity directly via changes in phosphorylation or by altering how the pump interacts with other proteins. For example, in guinea pig ventricular myocytes, the relative level of [Ca2+]idetermines if stimulation of protein kinase A by isoproterenol either inhibits or stimulates sodium pump activity (11, 12).
Ca-dependent stimulation of the sodium pump in adrenal glomerulosa cells could be due to an increase in the affinity for sodium, as previously observed in the rat proximal tubules (1), because the measured concentration of intracellular sodium under our experimental conditions (Fig. 1) was close to the concentration that half-maximally stimulates the α1-isoform of Na-K- ATPase (3, 30). In this case, we have assumed that α1 is the predominant isoform in these cells based on the concentration of ouabain required to inhibit activity (16). It is unlikely that calcium is stimulating the sodium pump by increasing its affinity for Ko, as observed in HeLa cells (24), because both 4 and 10 mM Ko are well above the half-maximal constant for Ko to bind to the sodium pump (20). In other cells, changes in physiological levels of [Ca2+]i can either stimulate or inhibit sodium pump activity, depending on the relative expression of regulatory proteins (23, 24, 34).
In conclusion, increasing Cao and/or Ko in the presence of Cao stimulated the sodium pump by increasing Cai 2+. The mechanism by which Cai 2+stimulated the sodium pump may involve CaM and either CaMKII or a related kinase. The concentrations of Ko and Cai 2+ that stimulated the sodium pump are in the same range that stimulate aldosterone secretion. Therefore, it is likely that the sodium pump is stimulated by Cai 2+ during Ko-stimulated aldosterone secretion. How Ca-dependent stimulation of the sodium pump alters aldosterone secretion is not yet known but could involve changes in the membrane potential.
This work was supported by a grant from the American Heart Association of Michigan, National Heart, Lung, and Blood Institute Grant HL-48885, Veterans Association Research Funds, and by the Vascular Biology Program of the Department of Internal Medicine at Wayne State University.
Address for reprint requests and other correspondence: D. R. Yingst, Dept. of Physiology, Wayne State Univ. School of Medicine, 540 E. Canfield Ave., Detroit, MI 48201 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society