Control of Cardiac Sarcolemmal Adenylate Cyclase and Sodium, Potassium-Activated Adenosinetriphosphatase Activities

Transcription

1 Control of Cardiac Sarcolemmal Adenylate Cyclase and Sodium, Potassium-Activated Adenosinetriphosphatase Activities By Michihiko Tada, Madeleine A. Kirchberger, Jo-Anna M. lorio, and Arnold M. Katz ABSTRACT A plasma membrane preparation purified from guinea pig ventricles without the use of high concentrations of detergents or structure-disrupting salts was used to compare the mechanisms controlling sodium, potassium-activated adenosinetriphosphatase (Na, K-ATPase) and adenylate cyclase activities. The basal ATPase activity of 4-6 pi moles Pj/hour mg" 1 protein, measured in 120 ml! NaCl or KC1, was approximately doubled in 100 mm NaCl plus 20 mm KC1. This increment, the Na, K-ATPase, was abolished by 10~ 5 M ouabain, the K, for ouabain being approximately 3 x 10" 7 M. /-Epinephrine had no effect on Na, K- ATPase, but NaF was inhibitory. Adenylate cyclase, which had a basal activity of approximately 200 pmoles cyclic AMP/min mg" 1 protein, was increased approximately 50% by NaCl or KC1 alone at concentrations up to 0.2M. There was no additional stimulation of adenylate cyclase activity when Na + and K + were included together. Both Z-epinephrine and NaF caused significant stimulation of adenylate cyclase, but neither basal nor activated cyclic AMP production was influenced by ouabain. Half-maximal stimulation was seen at approximately 5 x 1O~"M ^-epinephrine. Both the catecholamine and NaF increased the V max of cardiac plasma membrane adenylate cyclase without significantly influencing K m. Increasing Ca 2+ in the range between 10~ 7 and 10" 3 M inhibited basal, Z-epinephrine-stimulated, and NaF-stimulated activities. Basal rates of cyclic AMP production were more sensitive to Ca 2+ than was l- epinephrine-stimulated adenylate cyclase activity, so that Z-epinephrine stimulation was increased from approximately 60% in 0.5 mm EGTA to approximately 150% in 10-7 M Ca 2+ and 400% in 10~ 5 M Ca z+. The inhibitory effect of Ca 2+ on adenylate cyclase activity may represent a negative feedback mechanism by which elevation of intracellular Ca 2+ concentration lowers cellular levels of cyclic AMP and thus reduces Ca 2+ influx into the myocardium. KEY WORDS ouabain guinea pig cardiac sarcolemma fluoride In the heart, in which contractile performance is regulated by systems intrinsic to the muscle, a number of stimuli impinge on the myocardial cell that can, ultimately, modify the interactions between the heart's contractile proteins. These stimuli, which arise from From the Division of Cardiology, the Department of Medicine, and the Department of Physiology, The Mount Sinai School of Medicine of the City University of New York, New York, New York This investigation was supported by U. S. Public Health Service Grant HL and Research Contract NIH-NHLI M from the National Heart and Lung Institute, a grant-in-aid from the New York Heart Association, and a gift from the Jack Martin Fund. Dr. Kirchberger is a New York Heart Association Research Fellow. Dr. Katz is Philip J. and Harriet L. Goodhart Professor of Medicine (Cardiology). Dr. Tada's permanent address is First Department of Medicine, Osaka University Medical School, Osaka, Japan. Received February 25, Accepted for publication August 27, cyclic adenosine-3', 5'-monophospate calcium negative feedback epinephrine physiological, pharmacological, and pathological influences, probably have their initial action on the plasma membrane. The recent availability of a plasma membrane preparation, purified from mammalian hearts without the use of high concentrations of polar (e.g., LiBr, Nal) or nonpolar (e.g., deoxycholate) structure-disrupting agents (1), permits comparison of the mechanisms influencing two such control systems: the sodium, potassium-activated adenosinetriphosphatase (Na, K-ATPase), believed to represent an activity of the sodium pump (2) that has been suggested to mediate the positive inotropic actions of cardiac glycosides (3-5) and the adenylate cyclase system that has been proposed to mediate cellular actions of agents such as catecholamines (6-9), glucagon (8, 10), and thyroxine (11). Although much information has been obtained to define the

2 CARDIAC ADENYLATE CYCLASE AND Na, K-ATPase physiological and pharmacological systems that control the activities of Na, K-ATPase and adenylate cyclase, most studies have used different preparative methods to purify these two activities. In the present study, the mechanisms which control Na, K-ATPase and adenylate cyclase activities of myocardial plasma membranes were compared to determine the extent of independence and interdependence of these two control systems, both of which are located at the cell surface. Methods Plasma membranes from guinea pig hearts were prepared as described previously (1). Briefly, the method employed homogenization, osmotic shock, and extraction of the contractile proteins with hypertonic KC1, in some cases followed by exposure to changes in ambient pressure. The preparation used in these experiments was that obtained after the extraction step unless otherwise stated. Adenylate cyclase activity was assayed under the standard conditions described previously (1) within 30 hours after preparation of the membranes, except that 1 mg/ml of creatine phosphokinase and 25 mm di-tris phosphocreatine were used as the adenosine triphosphate (ATP) regenerating system. In experiments in which the effects of Ca 2+ on adenylate cyclase activity were studied, assays were performed at 25 C for 210 minutes in a medium containing 5 mm [a- 12 P]ATP, 5 mm MgCL, 0.5 mm unlabeled cyclic adenosine-3', 5'-monophosphate (cyclic AMP), 0.8 mg/ml of plasma membranes, various concentrations of ionized Ca 2+, and 100 mm imidazole buffer at ph 6.8. In studies at low Ca 2+ concentrations (below 10" 6 M), levels of ionized Ca 2+ were maintained with Ca-ethyleneglycol bis(/?-aminoethylether)-n, N'-tetraacetic acid (EGTA) buffers (12). The amount of cyclic AMP released was determined by thin-layer chromatography in 0.3M LiCl on polyethyleneimine-impregnated plastic sheets, according to the method of Bar and Hechter (13). ATP levels, determined by thin-layer chromatography of aliquots of the reaction mixtures on polyethyleneimine-impregnated sheets in 2N HCOOH and 0.5M LiCl, remained essentially constant during the incubations; more than 90% of the radioactivity was recovered as ATP. Standard ATPase assay media contained mg/ml of plasma membranes, 5 mm MgATP, 10" 8 M Ca 2+, 40 mm imidazole buffer at ph 6.8 and 37 C, and various concentrations of KC1, NaCl, or both. The amount of inorganic phosphate (P>) produced was determined as described previously (1). The Na, K-ATPase activity was estimated by subtracting the ATPase in 0.12M NaCl from that in 0.02M KC1 plus 0.10M NaCl. Tetra (triethylammonium) [a- ;t2 P]ATP (5-10 me/ //.mole) in aqueous solution at ph 7.4 was pur- chased from New England Nuclear Corporation. Creatine phosphokinase and di-tris phosphocreatine, obtained from Boehringer Mannheim Corporation and Sigma Chemical Company, respectively, were dissolved in cold 5 mm Tris-HCl (ph 7.5) immediately prior to use. ATP, obtained as Na^ATP from the Sigma Chemical Company, was deionized by chromatography on Dowex 50 and neutralized with Tris as described earlier (14). Ouabain octahydrate, obtained from Sigma Chemical Company, was dissolved in cold distilled water and stored at 0 C in the dark. All chemicals were reagent grade, and distilled water was deionized and redistilled from glass prior to use. /-Epinephrine, obtained from the Sigma Chemical Company, was dissolved in cold distilled water and used within 2 hours. Results ATPase activity of the isolated guinea pig sarcolemma showed the pattern of activation by various mixtures of NaCl and KC1 that is typical of plasma membrane Na, K-ATPases (Fig. 1). Thus, ATPase activity was stimulated when Na + and K + were present to- CATION CONCENTRATION (M) FIGURE 1 12 NaCl 0 KCI Effects of different combinations of Na* and K* on ATPase activity of cardiac sarcolemma. Plasma membranes (0.15 mgl ml) were incubated for 10 minutes under standard conditions (see Methods) with different combinations of NaCl and KCI, as indicated on the abscissa, in the presence (solid bars) and absence (open bars) of 10 s U ouabain. Each bar represents the mean ± SE of triplicate determinations.

3 10 gether, maximal activation being attained at 0.02M KCl and 0.01M NaCl. The specific activity of Na, K-ATPase, estimated by subtracting the ATPase activity in 0.12M NaCl from that in 0.02M KCl plus 0.10M NaCl, was usually within the range of 4.5 to 6.5 /xmoles Pi/hour mg" 1 protein; that for ATPase in the presence of either NaCl or KCl alone (basal ATPase) was 4-6 /xmoles Pj/hour mg-' protein. Ouabain (10~ 5 M) completely abolished the activation by KCl plus NaCl, reducing ATPase activity to the basal level seen with 0.12M NaCl or KCl alone (Fig. 1). Ouabain had no effect on the basal ATPase activity measured in either 0.12M KCl or NaCl alone. As shown in Figure 2, half-maximal inhibition of the ATPase activity by ouabain, determined in 0.02M KCl and 0.10M NaCl, was seen at approximately 3 x 10~ 7 M ouabain. The ATPase activity in 0.02M KCl plus 0.10M NaCl was inhibited by NaF at concentrations greater than 1 mm. Slight inhibition of basal ATPase in 0.12M NaCl was also seen (Fig. 3). The Na, K-ATPase was inhibited by 50% at approximately 20 mm NaF (Fig. 3). At concentrations O 10 10' 10" 10^ 10" OUABAIN CONCENTRATION (M) FIGURE 2 Effects of ouabain on Na, K-ATPase activity of cardiac sarcolemma. Plasma membranes (0.15 mglml) were incubated for 10 minutes under standard conditions (see Methods) with 0.02M KCl plus 0.10M NaCl in the presence of various ouabain concentrations. Each point represents the mean ± SE of triplicate determinations. 9, Na+K O, X, (Na+K)-Na NaF CONCENTRATION (mm) FIGURE 3 Effect of NaF on Na, K-ATPase activity of cardiac sarcolemma. Plasma membranes (0.2 mglml) were incubated under standard conditions (see Methods) for 10 minutes in 0.02M KCl plus 0.10M NaCl (solid circles) and 0.12M NaCl alone (open circles) in the presence of various concentrations of NaF. The net Na, K-ATPase activity (crosses) was estimated as described in Methods. as high as 10~ 4 M, i-epinephrine had no effect on Na, K-ATPase activity. KCl caused marked stimulation of the adenylate cyclase activities of both the purified plasma membranes and the crude homogenate (Fig. 4). The rate of cyclic AMP production increased almost linearly with increasing KCl concentrations up to 0.2M. Similarly, NaCl at concentrations up to 0.2M caused activation of adenylate cyclase; the extent of activation by KCl and NaCl was indistinguishable (Table 1). In contrast to the ATPase activity, sarcolemmal adenylate cyclase activity was not stimulated by various combinations of NaCl and KCl as long as the total cation concentration remained constant (Fig. 5a). Furthermore, ouabain (10~ 5 M) had no effect on either basal cyclic AMP production (Fig. 5b) or on adenylate cyclase activity in the presence of different combinations of Na + and K + (Fig. 5a). This lack of a cardiac glycoside effect was apparent over a wide range of ouabain concentrations (Table 2).

4 CARDIAC ADENYLATE CYCLASE AND Na, K-ATPase TABLE 1 Effects of NaCl and KCl on Cardiac Sarcolemmal Adenylate Cyclase Activity None KCl NaCl Addition Concentration (M) Adenylate cyclase activity (pmoles cyclic AMP/mg min~') 211± ±6 350± ± ± ± 9 KCl CONCENTRATION (M) FIGURE 4 Dependence of cardiac sarcolemmal adenylate cyclase activity on KCl concentration. Plasma membranes (0.4 mglml) were incubated under standard conditions (see Methods) in the presence of various concentrations of KCl (solid circles). Studies of the "initial homogenate," prepared as described previously (1), are also shown (open circles). Each point represents the average ± SE based on six determinations. Plasma membrane adenylate cyclase activity was stimulated by NaF and Z-epinephrine, maximal stimulation under the standard assay conditions being within the range of 100% to 150% for 8 mm NaF and 30% to 50% for 10~ 4 M Z-epinephrine. Typical examples of stimulation are shown in Figure 6. NaF at 8 mm gave maximal stimulation; higher concentrations were inhibitory. Maximal stimulation by Z-epinephrine was obtained at 10-4 M, half-maximal stimulation being seen at approximately 5 x 10~ 6 M. Propranolol abolished the stimulation by Z-epinephrine, and d-epinephrine failed to activate adenylate cyclase activity (Table 3). The stimulatory effects of Z-epinephrine and NaF on the adenylate cyclase activity were examined as functions of ATP concentrations up to 0.5 mm at the fixed MgCl 2 concentration of 10 mm. Lineweaver-Burk plots showed that the maximal velocity (V max ) of the basal ac- Assays were carried out with 0.4 mg/ml of plasma membranes under standard conditions described in Methods. The values of adenylate cyclase activity are averages+ SE based on six determinations. tivity was increased 21% and 89% by Z-epinephrine and NaF, respectively, whereas the Kn, for ATP of 0.06 mm was not significantly altered by either /-epinephrine or NaF (Fig. 7). The activities of basal, Z-epinephrine-stimulated, and NaF-stimulated adenylate cyclase depended on ionized Ca 2+ concentration. The absolute levels of basal and Z-epinephrine-stimulated adenylate cyclase activities both decreased as Ca 2+ concentration was increased, but the extent of stimulation 8} -? ^ OO 200 O.02.O NaCl O2 O KCl O CATION CONCENTRATION (M) FIGURE S Effects of different combinations of Na* and K* on adenylate cyclase activity of cardiac sarcolemma. a: Plasma membranes (0.4 mglml) were incubated under standard conditions (see Methods) with (open bars) or without (solid bars) 10~ 5 \l ouabain in the presence of different combinations of NaCl and KCl, as indicated on the abscissa, b: Plasma membranes were incubated as in Figure 5a except that added NaCl or KCl was not present. Each bar represents the mean + SE of six determinations.

5 12 TADA, KIRCHBERGER, IORIO, KATZ TABLE 2 Effect of Ouabain on Adenylate Cyclase Activity of Plasma Membranes of Guinea Pig Heart Concentration of ouabain (M) 0 io " io- 7 io- 6 io- D io- 4 Adenylate cyclase activity (pmoles cyclic AMP/mg min" 1 ) 195 ±8 207 ± ± ± ±8 198 ±8 207 ± 9 Discussion Two different control systems can be identified in purified guinea pig myocardial plasma membranes. One of these modifies the ATPase activity that is generally attributed to the sodium pump; the other modulates adenylate cyclase activity, which controls cyclic AMP production in the heart. The present findings indicate that the plasma membrane Na, K-ATPase and adenylate cyclase activities are regulated independently in that factors which modify the activity of one enzyme do not cause parallel changes in the other (Table 5). Thus, plasma membrane ATPase is stimulated by the combination of Na + and K +, inhibited by ouabain, and insensitive to epinephrine, whereas adenylate cyclase activity, although enhanced by either Measurements were carried out with 0.4 mg/ml of plasma membranes under standard conditions as described in Methods. The values of adenylate cyclase activity are averages i.?e based on six determinations. by Z-epinephrine was greater at higher Ca 2+ concentrations (Fig. 8). The extent of stimulation of adenylate cyclase activity by 8 mm NaP similarly increased with increasing Ca 2+ concentration (Table 4). The Z-epinephrine sensitivity of adenylate cyclase activity was examined in the presence of 10~ 7 M and 10~ 5 M Ca 2+ (Fig. 9). Increasing Ca 2+ concentration caused very little shift in the Z-epinephrine concentration needed for half-maximal stimulation, from 2 x 10-6 M in 10" 7 M Ca 2+ to 5 x 10~ 6 M in 10~ 5 M Ca 2+ (both of which are similar to the value of 5 x 10~ 6 M measured in 0.5 mm EGTA under slightly different conditions, Fig. 6, bottom), suggesting that Ca 2+ may inhibit only an i-epinephrine-insensitive component of adenylate cyclase. It should be noted that adenylate cyclase activity was much less when reactions were carried out at ph 6.8 and 25 C, in contrast to the other studies reported in the present paper, in which ph was 7.4 and temperature was 37 C. However, the ph dependency of the Ca- EGTA buffers made it necessary to work away from the optimal conditions. S NaF CONCENTRATION (mm) 20 /-EPINEPHRINE CONCENTRATION (M) FIGURE 6 Effects of NaF and l-epinephrine on adenylate cyclase activity of cardiac sarcolemma. Plasma membranes (0.4 mglml) were incubated under standard conditions (see Methods) in the presence of various concentrations of NaF (top) or l-epinephrine (bottom). Each point represents the percent stimulation over control and is the average of four determinations.

6 CARDIAC ADENYLATE CYCLASE AND Na, K-ATPase 13 TABLE 3 Effects of I- and d-epinephrine and of d, l-propranolol on Cardiac Sarcolemmal Adenylate Cyclase Activity Adenylate cyclase activity (pmoles cyclic AMP/mg Addition mirr 1 ) None 2.27 ± J M/-epinephrine 9.54 ±1.81* 10~ 4 M rf-epinephrine 2.27 ± " 4 M /-epinephrine + 5 x 10" J M d, /-propranolol 3.20 ±0.91+ The values of adenylate cyclase activity are means ± SE of six determinations. Assays were performed at 25"C and ph 6.8 as described in Methods for studies with Ca 2+ -buffers, except that ATP and protein concentrations were 2 mm and 2.5 mg/ml, respectively. Ca 2+ concentration was 10" 5 M. *t = 2.815, P < 0.01 compared with control. ft = 2.782, P < 0.01 compared with 10" 4 M/-epinephrine. K + or Na + alone, is not affected by Na + and K + together, is insensitive to ouabain, and is stimulated by epinephrine. Both enzymes are inhibited by Ca 2+, but the percent stimulation of adenylate cyclase that is produced by epinephrine is greater at higher Ca 2+ concentrations. The sensitivity of Mg 2+ -dependent ATPase to K + and Na + and that of Na, K-ATPase to ouabain exhibited by the present plasma o Basal FIGURE [ATP(mM)]" 1 Lineweaver-Burk plot of the ATP dependence ofl-epinephrineand NaF-stimulated adenylate cyclase activity of cardiac sarcolemma. Plasma membranes (0.4 mg/ml) were incubated under standard conditions at various ATP concentrations (see Methods) in the presence of 10'"M l-epinephrine (crosses) or 8 mm NaF (solid circles) and without additions (open circles). Lines were drawn with the aid of a PDPSe computer by the method of least squares. FIGURE 8 Effects ofca 2 * on adenylate cyclase activity of cardiac sarcolemma. Plasma membranes (0.8 mglml) were incubated under conditions described in the text (see Methods) at different Co 2 * concentrations and in the presence of 0.5 mm EGTA as labeled on the abscissa. Adenylate cyclase was measured under basal conditions (open bars) and in the presence ofl0 4 M l-epinephrine (striped bars). Each bar represents the mean ± SE of six determinations. The percent stimulation by epinephrine is shown by the solid bars. These studies were carried out at ph 6.8 and 25 C. membrane preparation were similar to those reported by Stam et al. (15) and Matsui and Schwartz (16). The observed kinetic parameters of ouabain-induced inhibition of Na, K- ATPase activity are in good agreement with those reported previously (15, 16). The specific activity of Na, K-ATPase, /xmoles P^hour mg~' protein is comparable to that of dog heart Na, K-ATPase (15) but is about one fifth of that of a calf heart microsomal preparation (16). In the latter preparations, however, the membranes were treated with detergent (deoxycholate) and concentrated salt (Nal), which may activate this enzyme. These substances were avoided in our preparation to preserve the labile hormone receptors of the adenylate cyclase system. The basal adenylate cyclase activity of myocardial plasma membranes in the present investigation was over 200 pmoles cyclic AMP/min mg~' protein under standard conditions in which ATP concentration was submaximal (0.2 mm). The maximal rates of basal, epinephrine-stimulated, and NaFstimulated myocardial adenylate cyclase were estimated from the Lineweaver-Burk plots to be about 330, 400, and 630 pmoles/ min mg- 1 protein, respectively (Fig. 7). These specific activities are higher than those reported previously (6-10) and may reflect the extraction of contractile proteins during the

7 14 TADA, KIRCHBERGER, IORIO, KATZ TABLE 4 Ca 2+ Concentration (M) io- 8 io- 7 io- 6 io- 5 io- 4 Effects of Ca 2+ Concentration on Stimulation of Cardiac Sar- colemmal Adenylate Cyclase Activity by NaF Adenylate cyclase activity (pmoles cyclic AMP/mg min- 1 ) Control 27.8 ± ± ± ± ±0.7 8 mm NaF 63.1 ± ± ± ± ±4.5 Stimulation (%) Measurements were carried out with Ca 2+ buffers as described in methods. The values of adenylate cyclase activity are means ± SE based on six determinations. These studies were carried out at ph 6.8 and 25 C. preparation and the inclusion of EGTA, a Ca 2+ chelator, in the assay media. Adenylate cyclase activity of the cardiac plasma membranes was stimulated by KCl or NaCl, but unlike ATPase activity adenylate cyclase activity was not stimulated further when KCl and NaCl were present together. In intact tissues there is evidence that K + may participate in the physiological control of cyclic AMP production. Thus, elevation of K+ concentration in the perfused rat heart, which leads to membrane depolarization, has been reported to abolish epinephrine-induced cyclic AMP production (17). Similarly, in slices of guinea pig cerebral cortex, incubation with elevated K + concentration or with depolarizing agents such as veratridine and batrachotoxin increases cyclic AMP levels (18, 19). Electrical stimulation of brain slices also has been found to increase cyclic AMP production (20). These observations could be explained if altered K + levels at the cell surface act to modify the adenylate cyclase activity which is located on the plasma membrane. However, the present investigation demonstrates that the adenylate cyclase activity of the isolated myocardial plasma membrane is not influenced when KCl is partially substituted for NaCl, although cyclic AMP production is enhanced when ionic strength is raised with either KCl or NaCl. These apparent discrepancies may be attributed to the fact that isolated plasma membranes are probably not polarized electrically, unlike the plasma membranes of the intact heart. Although the physiological 5O implication of the stimulation of myocardial plasma membrane adenylate cyclase activity by increasing ionic strength is not clear, observations analogous to the present findings have been reported by Dousa and Hechter (21) and Dousa (22) in a vasopressinsensitive adenylate cyclase preparation from the renal medulla. In these studies, comparison of the effects of polar (KCl and NaCl) and nonpolar (glucose and urea) solutes indicated that it was the polarity of the solute rather than its osmolarity that modified the adenylate cyclase activity. Ouabain at concentrations as high as 10~ 5 M was without effect on adenylate cyclase activity in the presence and the absence of various combinations of NaCl and KCl. Previous reports disagree as to whether cardiac glycosides inhibit adenylate cyclase activity. Cardiac glycosides, like omission of K + from the medium, have been found to inhibit lipolysis (23, 24) possibly by inhibiting the effect of catecholamines (25). Inhibition of epinephrine-induced stimulation of lipolysis and glu- 4O- /-EPINEPHRINE CONCENTRATION (M) FIGURE 9 Dependence on Co 2 * concentration of l-epinephrine stimulation of cardiac sarcolemmal adenylate cyclase activity. Plasma membranes (0.8 mglml) were incubated in the presence of various concentrations of l-epinephrine under conditions described in the text (see Methods) in Ca-EGTA buffers at 10~ 7 M Ca?* (solid circles) and 10~ 5.\l Ca*~ (open circles). These studies were carried out at ph 6.8 and 25 C. Each point represents the average ± SE based on four determinations.

8 CARDIAC ADENYLATE CYCLASE AND Na, K-ATPase 15 TABLE S Control of Plasma Membrane Sodium, Potassium-Activated Adenosinetriphosphatase and Adenylate Cyclase Activities Modifier Na* + K* together Cardiac glycosides NaF Epinephrine Ca 2 " Na, K-ATPase Increase Decrease Decrease No effect Decrease Adenylate cyclase No effect No Effect Increase Increase Decrease in basal activity increase in epinephrine and NaF stimulation coneogenesis of rat epididymal fat pads by ouabain has been suggested to be due to interference with activation of adenylate cyclase (26). However, in brown fat cells (27) and in guinea pig brain (19) ouabain has been reported to enhance cyclic AMP formation. Cardiac plasma membrane Na, K-ATPase activity was inhibited by NaF, whereas adenylate cyclase activity was stimulated by low concentrations of NaF. At concentrations above 10 mm, NaF caused inhibition of adenylate cyclase. Stimulation of adenylate cyclase activity by NaF is not, however, attributable to the ability of NaF to inhibit ATPase activity, because the ATP regenerating system in the adenylate cyclase assay medium kept ATP concentrations constant throughout the incubation period, as monitored by thin-layer chromatography (see Methods). Epinephrine, which stimulated adenylate cyclase activity, is without effect on Na, K ATPase activity (28). The role of Ca 2+ in the control of adenylate cyclase activity remains controversial. An inhibitory effect of Ca 2+ or a stimulatory effect of EGTA on the basal and the total level of hormone-stimulated cyclic AMP production similar to that found in the present study has also been seen in other tissues (29-32). Similarly, inhibition of basal adenylate cyclase has been found in previous studies of cardiac muscle preparations (33, 34). Sulakhe and Dhalla (34), who studied the effects of increasing Ca 2+ concentration in the millimolar range, did not demonstrate an increase in the stimulation by epinephrine such as was seen in the present study when Ca 2+ concentrations were increased in the micromolar range (Fig. 8). In contrast, other studies with noncardiac preparations indicate that EGTA can inhibit basal or polypeptide hormone-stimulated adenylate cyclase activity, or both (35-39); however, Lefkowitz et al. (35), Bockaert et al. (38), and Bar and Hechter (39) have presented evidence that Ca 2+ has a stimulatory action at low concentration, although higher Ca 2+ concentrations inhibit these activities. These discrepancies may be partly due to the reported Ca 2+ dependence of phosphodiesterase activity (40, 41), since this enzyme, which degrades cyclic AMP, is activated when Ca 2+ is raised from 10" 6 M to 10-5 M. Thus, the inhibition of basal adenylate cyclase activity by Ca 2+ shown in Figure 8 may be partly explained by enhanced cyclic AMP breakdown at the higher Ca 2+ concentrations. Under the present assay conditions, in which 0.5 mm unlabeled cyclic AMP was included, effects of phosphodiesterase would be minimized. The present finding that Ca 2+ inhibits cardiac sarcolemmal adenylate cyclase activity indicates that Ca 2+ may participate in the control of cyclic AMP levels in the myocardial cell. This putative control mechanism appears to operate within the range of Ca 2+ concentrations that is found intracellularly. At 10-7 M Ca 2+, which approximates the Ca 2+ concentration in the myocardium during diastole (42), basal levels of adenylate cyclase activity are almost twice the level measured in 10~ 5 M Ca 2+, the systolic level of this cation. Inhibition of cyclic AMP production when Ca 2+ levels are increased may represent part of a control mechanism by which increasing intracellular Ca 2+ lowers cyclic AMP concenj3-adrenergic agonists ATP FIGURE 10 Proposed interrelationships between Ca*~ and cyclic AMP levels in the myocardium.

9 16 TADA, KIRCHBERGER, IORIO, KATZ tration. Such a mechanism may also be manifest by the reported stimulatory effects of Ca 2+ on phosphodiesterase, the enzyme responsible for cyclic AMP breakdown. The levels of Ca 2+ needed for activation of phosphodiesterase, like those that inhibit adenylate cyclase, are extremely low, half-maximal activation of the former being seen in the micromolar range of Ca 2+ concentration (40, 43). Lowering of the cyclic AMP level, the net result of these effects of Ca 2+ on adenylate cyclase and phosphodiesterase, appears to represent one limb of a negative feedback control mechanism. The primary limb of this control mechanism involves the effect of both adrenergic agonists (44) and dibutyryl cyclic AMP (45) to increase the influx of Ca 2+ into the myocardium, thereby raising intracellular Ca 2+ concentration (Fig. 10). Acknowledgment The authors gratefully acknowledge the assistance of Mr. David Reiss in obtaining the data presented in Table 3. References 1. TADA M, FINNEV JO, SWARTZ MH, KATZ AM: Preparation and properties of plasma membranes from guinea pig hearts. J Mol Cell Cardiol 4: , SKOU JC: Enzymatic basis for active transport of Na + + K + across cell membrane. Physiol Rev 45: , REPKE KRW: Uber den biochemischen Wirkungsmodus von Digitalis. Klin Wochenschr 42: , LANGER GA: Ion fluxes in cardiac excitation and contraction and their relation to myocardial contractility. Physiol Rev 48: , BESCH HR JR, SCHWARTZ A: On a mechanism of action of digitalis. J Mol Cell Cardiol 1: , MURAD F, CHI YM, RALL TW, SUTHERLAND EW: Adenyl cyclase: III. Effect of catecholamines and choline esters on the formation of adenosine 3',5'-phosphate by preparations from cardiac muscle and liver. J Biol Chem 237: , DRUMMOND GI, SEVERSON DL, DUNCAN L: Adenylate cyclase: Kinetic properties and nature of fluoride and hormone stimulation. J Biol Chem 246: , EPSTEIN SE, LEVEY GS, SKELTON CL: Adenyl cyclase and cyclic AMP: Biochemical links in the regulation of myocardial contractility. Circulation 43: , MAYER SE: Effects of adrenergic agonists and antagonists on adenylate cyclase activity of dog heart and liver. J Pharmacol Exp Ther 181: , MURAD F, VAUGHAN M: Effect of glucagon on rat heart adenyl cyclase. Biochem Pharmacol 18: , LEVEY GS, EPSTEIN SE: Myocardial adenyl cyclase: Activation by thyroid hormones and evidence for two adenyl cyclase systems. J Clin Invest 48: , KATZ AM, REPKE DI, UPSHAW JE, POLASCIK MA: Characterization of dog cardiac microsomes: Use of zonal centrifugation to fractionate fragmented sarcoplasmis reticulum, (Na + + K + )-activated ATPase and mitochondrial fragments. Biochim Biophys Acta 205: , BAR H, HECHTER O: Adenyl cyclase in fat cell ghosts. Anal Biochem 29: , KATZ AM, REPKE DI: Quantitative aspects of dog cardiac microsomal calcium binding and calicum uptake. Circ Res 21: , STAM AC JR, WEGLICKI WB JR, FELDMAN D, SHELBURNE JC, SONNENBLICK EH: Canine myocardial sarcolemma: Its preparation and enzymatic activity. J Mol Cell Cardiol 1: , MATSUI H, SCHWARTZ A: Purification and properties of a highly active ouabain sensitive Na +, K + -dependent adenosinetriphosphatase from cardiac tissue. Biochim Biophys Acta 128: , NAMM DH, MAYER SE, MALTBIE M: Role of potassium and calcium ions in the effect of epinephrine on cardiac cyclic adenosine 3',5'-monophosphate, phosphorylase kinase, and phosphorylase. Mol Pharmacol 4: , SHIMIZU H, CREVELING CR, DALY JW: Stimulated formation of adenosine 3',5'-cyclic phosphate in cerebral cortex: Synergism between electrical activity and biogenic amines. Proc Natl Acad Sci USA 65: , SHIMIZU H, CREVELING CR, DALY JW: Cyclic adenosine 3',5'-monophosphate formation in brain slices: Stimulation by batrachotoxin, ouabain, veratridine, and potassium ions. Mol Pharmacol 6: , KAKIUCHI S, RALL TW, MCILWAIN H: Effect of electrical stimulation upon the accumulation of adenosine 3',5'- phosphate in isolated cerebral tissue. J Neurochem 16: , DOUSA TP, HECHTER O: Effect of NaCl and LiCl on vasopressin-sensitive adenyl cyclase. Life Sci [I] 9: , DOUSA TP: Effect of renal medullary solutes on vasopressin-sensitive adenyl cyclase. Am J Physiol 222: , Ho RJ, JEANRENAUD B, POSTERNAK T, RENOLD AE: Insulin-like action of ouabain: II. Primary antilipolytic effect through inhibition of adenyl cyclase. Biochim Biophys Acta 144:74-82, MOSINGER B, VAUGHAN M: Effect of electrolytes on epinephrine stimulated lipolysis in adipose tissue in vitro. Biochim Biophys Acta 144: , HIMMS-HAGER J: Adrenergic receptors for metabolic responses in adipose tissue. Fed Proc 29: , KYPSON J, TRINER L, NAHAS GG: Effects of ouabain and K + -free medium on activated lipolysis and epinephrine-stimulated glyconeogenesis. J Pharmacol Exp Ther 159:8-17, FAIN JM, JACOBS MD, CLEMENT-CORMIER YC: Interrelationship of cyclic AMP, lipolysis, and respiration in brown fat cells. Am J Physiol 224: , SCHWARTZ A, LINDENMEYER GE, ALLEN JC: Na +, K + - Adenosinetriphosphatase: Pharmacological or biochemical aspects. Pharmacol Rev, in press 29. MATSUZAKI S, DUMONT JE: Effect of calcium ion on horse parathyroid gland adenylate cyclase. Biochim Biophys Acta 284: , 1972

10 CARDIAC ADENYLATE CYCLASE AND Na, K-ATPase 30. BIRNBAUMER L, POHL SL, RODBELL M: Adenyl cyclase in fat cells: I. Properties and the effects of adrenocorticotropin and fluoride. J Biol Chem 244: , MARUMO F, EDELMAN IS: Effects of Ca ++ and prostaglandin E, on vasopressin activation or renal adenyl cyclase. J Clin Invest 50: , BAR HP, HECHTER O: Adenyl cyclase and hormone action: I. Effects of adrenocorticotropic hormone, glucagon, and epinephrine on the plasma membrane of rat fat cells. Proc Natl Acad Sci USA 63: , DRUMMOND GI, DUNCAN L: Adenyl cyclase in cardiac tissue. J Biol Chem 245: , SULAKHE PV, DHALLA NS: Adenylate cyclase of heart sarcotubular membranes. Biochim Biophys Acta 293: , LEFKOWITZ RJ, ROTH J, PASTAN I: Effects of calcium on ACTH stimulation of the adrenal: Separation of hormone binding from adenyl cyclase activation. Nature (Lond) 228: , BRADHAM LS, HOLT DA, SIMS M: Effect of Ca^ on the adenyl cyclase of calf brain. Biochim Biophys Acta 201: , CAMPBELL BJ, WOODWARD G, BORBERG V: Calcium-mediated interactions between the antidiuretic hormone and renal plasma membranes. J Biol Chem 247: , BOCKAERT J, ROY C, JARD S: Oxytocin-sensitive adenylate cyclase in frog bladder epithelial cells: Role of calcium, nucleotides, and other factors in hormonal stimulation. J Biol Chem 247: , BAR HP, HECHTER O: Adenyl cyclase and hormone action: III. Calcium requirement of ACTH stimulation of adenyl cyclase. Biochem Biophys Res Commun 35: , TEO TS, WANG JH: Mechanism of activation of a cyclic adenosine 3':5'-monophosphate phosphodiesterase from bovine heart by calcium ions. J Biol Chem 248: , MIKI N, YOSHIDA H: Purification and properties of cyclic AMP phosphodiesterase from rat brain. Biochim Biophys Act 268: , KATZ AM: Contractile proteins of the heart. Physiol Rev 50:63-158, KAKIUCHI S, YAMAZAKI R, TESHIMA Y, UENISHI K: Regulation of nucleoside cyclic 3':5'-monophosphate phosphodiesterase activity from rat brain by a modulator and Ca 2 *. Proc Natl Acad Sci USA 70: , REUTER H: Divalent cations as charge carriers in excitable membranes. Prog Biophys Mol Biol 26:1^43, TSIEN RW, GILES W, GREENGARD P: Cyclic AMP mediates the effects of adrenaline on cardiac Purkinje fibres. Nature [New Biol] 240: ,