INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE SWELLING OF CARDIAC myocytes induced by exposure to hyposmolar superfusates stimulates the activity of the membrane Na⁺-K⁺ pump (27, 32). However, the cellular mechanisms involved in the transduction of the osmotic stress into pump stimulation have not been investigated. The swelling of myocytes activates stretch-sensitive ion channels, including channels permeable to Na⁺ and Ca²⁺ (3, 23). Pump stimulation could therefore be secondary to an influx of extracellular Na⁺ or Ca²⁺, resulting in a rise in cytosolic Na⁺ or Ca²⁺ concentrations, known determinants of pump activity (33). Cell swelling is also known to rapidly activate a variety of intracellular messengers, including protein tyrosine kinase and serine(threonine) protein kinase (24, 26; see Ref. 25 for review). Serine(threonine) protein kinases can induce the phosphorylation of the Na⁺-K⁺ pump, whereas tyrosine kinases are involved in inducing the dephosphorylation of pump molecules via a mechanism involving activation of protein serine/threonine phosphatases (21). Because the activation of kinases and phosphatases can regulate the Na⁺-K⁺ pump (2, 7), increased pump activity during exposure of cardiac myocytes to hyposmolar superfusates could be mediated by protein kinase- and/or protein phosphatase-dependent pathways.

Understanding the mechanism by which the Na⁺-K⁺ pump is regulated by osmotic stress is desirable because the pump plays a role in the maintenance of cell volume under physiological conditions and during pathophysiological states, including acute myocardial ischemia and reperfusion. Using the whole cell patch-clamp technique, we studied Na⁺-K⁺ pump current (I_p) in ventricular myocytes exposed to hyposmolar superfusates. We found that swelling-induced Na⁺-K⁺ pump stimulation is independent of the influx of extracellular Na⁺ and Ca²⁺. However, stimulation was dependent on tyrosine kinase and involved the activation of an okadaic acid-inhibitable protein phosphatase. Pump stimulation induced by exposure to hyposmolar superfusates was rapidly reversible on reexposure to an isosmolar superfusate. This reversal of pump stimulation could be blocked by inhibition of protein kinase. We propose that the Na⁺-K⁺ pump is regulated by cell volume via steps which involve the phosphorylation and dephosphorylation of the pump molecule by kinases and phosphatases, respectively.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cells and solutions. Single ventricular myocytes were isolated from male New Zealand White rabbits weighing 2.5-3.5 kg as described previously (32). They were stored at room temperature until used. For measurement of I_p, myocytes were suspended in superfusates flowing through a tissue bath mounted on an inverted microscope. The superfusate used initially in all experiments contained (in mM) 140 NaCl, 5.6 KCl, 2.16 CaCl₂, 0.44 NaH₂PO₄, 10 glucose, 10 HEPES, and 1 MgCl₂. The solution was titrated to pH 7.4 ± 0.01 at 35°C with 1 M NaOH. Hyposmolar and isosmolar Ca²⁺-free modified Tyrode solutions, used subsequently when I_p was measured, contained (in mM) 105 NaCl, 5.6 KCl, 0.44 NaH₂PO₄, 10 glucose, 10 HEPES, 2 MgCl₂, 0.2 CdCl₂, and 2 BaCl₂. BaCl₂ was added to block K⁺ channels, and CdCl₂ was added to block the Na⁺/Ca²⁺ exchanger. The solutions were titrated to pH 7.4 ± 0.01 at 35°C with 1 M NaOH. The isosmolar solution contained 70 mM sucrose, whereas the hyposmolar solutions were sucrose free. We measured solution osmolarity with a vapor pressure osmometer (model 5500 osmometer; Wescor, Logan, UT). The osmolarities of the hyposmolar and isosmolar solutions were 240 ± 3 and 305 ± 4 mosM, respectively (mean ± SD). Superfusates used at the time I_p was measured were warmed to 35 ± 0.5°C. For measurements of I_p at a fixed membrane voltage (V_m) of 40 mV, the pipette filling solution contained (in mM) 70 potassium glutamate, 1 KH₂PO₄, 5 HEPES, 5 EGTA, and 2 MgATP. Unless otherwise indicated, the solution also contained 10 mM sodium glutamate and 80 mM tetramethylammonium chloride (TMA-Cl). It was titrated to pH 7.05 ± 0.01 at 35°C with 1 M KOH. In experiments designed to examine the I_p-V_m relationship, the filling solution of the patch pipettes contained (in mM) 10 sodium glutamate, 1 KH₂PO₄, 5 HEPES, 5 EGTA, 2 MgATP, 60 TMA-Cl, 20 tetraethylammonium chloride, 70 CsOH, and 50 aspartic acid. The solution was titrated to a pH of 7.05 ± 0.01 at 35°C with HCl. The resistances of pipettes filled with either solution were 0.8-1.0 M. The offset potential was nulled by placing both the ground electrode and patch pipette in the pipette filling solution (20). We used the computer program JPCalc (P. Barry, Univ. of New South Wales) to estimate junction potentials between the ground electrode and superfusates. The differences in these junction potentials were <0.3 mV between the different superfusates used. They are not corrected for when I_p is reported.

Measurement of membrane currents. We measured I_p in freshly isolated cells 2-9 h after excision of the heart. I_p was defined as the shift in holding current induced by 100 µM ouabain (Sigma Chemical, St. Louis, MO). For each cell I_p was measured during the superfusion of either isosmolar or hyposmolar Ca²⁺-free Tyrode solutions. Membrane currents were recorded by using the continuous single-electrode voltage-clamp mode of an Axoclamp 2A amplifier and AxoTape or pCLAMP software (Axon Instruments, Foster City, CA). The I_p values reported were normalized for membrane capacitance as described previously (32). For measurement of the I_p-V_m relationship, we superfused myocytes with Ca²⁺-free solutions having a composition identical to that of solutions used to measure I_p at the fixed V_m of 40 mV with the exception that we added 1 mM anthracene-9-carboxylic acid (9-AC; Sigma Chemical) to inhibit stretch-activated Cl channels (12).

Reagents. In some experiments tyrphostin A25 or its inactive analog, tyrphostin A63, okadaic acid or its inactive analog, okadaic acid methyl ester, or staurosporine was added to the patch pipette. All were purchased from Calbiochem (La Jolla, CA). Tyrphostin A25 and A63 were dissolved in DMSO. The final concentration of DMSO was 0.04%. Staurosporine and okadaic acid methyl ester were also dissolved in DMSO; the final concentrations of DMSO were 0.005 and 0.2%, respectively. In experiments designed to examine the I_p-V_m relationship, 9-AC was dissolved in DMSO with a final concentration of 0.1%. It has previously been shown that DMSO in concentrations higher than those used in the present study have no effect on the steady-state holding current or I_p in isolated cardiac myocytes (19). LY-294002, purchased from Calbiochem, was dissolved in ethanol and added to the superfusates on the day of experimentation. The final ethanol concentration in the superfusates was 0.03%. Ethanol at this concentration has previously been shown in control experiments to have no effect on basal Na⁺-K⁺ pump activity (11).

Statistical analysis. Results are expressed as means ± SE unless otherwise indicated. Comparisons are made by using Student's t-test for unpaired observations. Statistical comparisons based on the same control experiments were made with Dunnett's test. Comparisons between the voltage dependence curves were made by two-way ANOVA for the interaction of voltage and osmolarity. Differences between means were regarded as statistically significant at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Role of Ca²⁺ influx in stimulation of I_p by cell swelling. We have previously shown that exposure of rabbit cardiac myocytes to hyposmolar solutions stimulates I_p when the patch pipette Na⁺ concentration ([Na]_pip) is near physiological intracellular levels (32). In contrast, Sasaki et al. (27) did not report a significant enhancement of the I_p values for patch-clamped guinea pig cardiac myocytes under similar experimental conditions ([Na]_pip 10 mM). Our experiments were performed in the presence of extracellular Ca²⁺, whereas the experiments of Sasaki et al. (27) were performed with superfusates that were nominally Ca²⁺ free. This raises the possibility that the pump stimulation induced by exposure to hyposmolar superfusates may be dependent on the influx of Ca²⁺ via stretch-activated channels. Because it cannot be ensured that the dialysis of the pipette solution completely changes intracellular Ca²⁺ levels, such an influx might cause changes in the intracellular concentrations of this ion. We performed experiments to examine this possibility. The pipette solution we used included 10 mM Na⁺, and, to buffer intracellular Ca²⁺, 5 mM EGTA. The whole cell configuration was first established while myocytes were being superfused with isosmolar Ca²⁺-containing Tyrode solution. We then switched to a superfusate that contained 105 mM Na⁺ and that was isosmolar (310 mosM) or hyposmolar (240 mosM). These solutions were nominally Ca²⁺ free. The Ca²⁺-free Tyrode solution was superfused for 3 min before we measured I_p. We measured I_p values for nine cells superfused with isosmolar solution and for seven cells superfused with hyposmolar solution. Representative recordings of shifts in holding currents during the measurement of I_p have been published previously (32). The means of the I_p values were 0.25 ± 0.03 pA/pF in isosmolar superfusates and 0.58 ± 0.06 pA/pF in hyposmolar superfusates. The difference in mean I_p values was statistically significant.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of osmotic swelling on sarcolemmal Na+-K+ pump current (Ip). Superfusate was isosmolar (Iso) or hyposmolar (Hypo), and patch pipette Na+ concentration ([Na+]pip) was either 10 or 80 mM. Ip has been normalized for cell capacitance. No. in parentheses is no. of experiments in each group.

Role of Na⁺ influx in pump stimulation. Cell swelling and mechanical stretch open nonselective stretch-activated cation channels (23). An influx of Na⁺ through such channels might stimulate pump activity in the present study by increasing the Na⁺ concentration at intracellular binding sites in a restricted subsarcolemmal space to levels above [Na]_pip. To examine this possibility, we measured I_p while myocytes were being superfused with Na⁺-free isosmolar or hyposmolar superfusates. Both superfusates were Ca²⁺ free. We replaced Na⁺ in the superfusates with 105 mM N-methyl-D-glucamine. Isosmolar superfusates contained 70 mM sucrose, whereas hyposmolar solutions were sucrose free. [Na]_pip was 10 mM. In the absence of extracellular Na⁺, the mean I_p of eight myocytes superfused with hyposmolar solutions was 0.59 ± 0.04 pA/pF, whereas the mean I_p of six myocytes superfused with isosmolar solutions was 0.31 ± 0.03 pA/pF. The difference was statistically significant (P < 0.001). This indicates that the stimulatory effect of cell swelling on I_p is not due to an increased Na⁺ influx. A direct effect of cell swelling on the Na⁺-K⁺ pump is suggested.

Effect of cell swelling on pump I_p-V_m relationship. The experiments described above indicate that myocyte swelling stimulates the Na⁺-K⁺ pump at a fixed membrane voltage (V_m) of 40 mV. Because pump function is voltage dependent, we examined if stimulation also occurs at other voltages. After establishing the whole cell configuration, we switched to a Ca²⁺-free superfusate and voltage-clamped myocytes at a holding potential of 40 mV. We then applied 320-ms voltage steps to test potentials (V_m) from 100 to +60 mV in 20-mV increments. The command voltage was returned to 40 mV for 2 s between each step to a test potential. To derive I_p at each test potential, the membrane currents recorded during the superfusion of ouabain were subtracted from the membrane currents at corresponding voltages before the superfusion of ouabain. Details of the protocol and representative current traces have been described previously (10). Figure 2A summarizes the I_p-V_m relationships for six myocytes superfused with isosmolar solution and six myocytes superfused with hyposmolar solution. Pump stimulation induced by cell swelling was most prominent at the most negative values of V_m and decreased with a shift of V_m toward positive levels. I_p values measured in these experiments using a [Na]_pip of 10 mM are larger during the superfusion of both isosmolar and hyposmolar solutions than those measured in experiments performed at a constant V_m of 40 mV with 10 mM [Na]_pip (see Fig. 1). This difference can be attributed to differences in the compositions of the pipette solutions used in the two experimental protocols. In experiments examining the I_p-V_m relationship, we used a pipette solution containing Cs⁺ as a substitute for K⁺ to abolish time- and voltage-dependent K⁺ conductances. We have previously shown that the inhibitory effect of intracellular Cs⁺ on I_p is significantly less than the inhibitory effect of intracellular K⁺. Such a reduction in inhibitory effect is expected to cause an increase in I_p (14). The I_p-V_m relationships have been normalized to the I_p recorded at 0 mV in Fig. 2B to facilitate the comparison of their slopes. The I_p-V_m relationship in isosmolar solutions was nearly linear and had a positive slope throughout the entire range of test potentials. The I_p-V_m relationship for six myocytes superfused with hyposmolar solution appeared less steep at negative potentials than that for myocytes superfused with isosmolar solutions. It had a negative slope at potentials more positive than 40 mV. The difference in these curves was statistically significant (P < 0.001; 2-way ANOVA). We conclude that pump stimulation induced by myocyte swelling is voltage dependent.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   A: voltage dependence of Ip values for myocytes superfused with isosmolar () and hyposmolar superfusates (). Ip-membrane voltage (Vm) relationships of 6 myocytes in each group have been summarized by plotting mean ± SE Ip measured at each test potential. B: normalized voltage dependence of Ip. Plots were obtained by first normalizing Ip-Vm relationship for each myocyte to Ip recorded at 0 mV and then averaging normalized Ip values at each test potential.

Effect of kinase inhibition on swelling-induced pump stimulation. Both mechanical deformation of the cardiac sarcolemmal membrane and osmotic swelling in a hyposmolar solution activate protein kinase C (PKC) (24). Because PKC and other kinases, in turn, modulate Na⁺-K⁺ pump activity in a variety of noncardiac tissues (2, 7), we examined the effect of the potent but nonselective protein kinase inhibitor staurosporine and the effect of bisindolylmaleimide I, an inhibitor believed to be specific for PKC. We added 10 nM staurosporine to the patch pipette solution. [Na]_pip was 10 mM in all experiments. After the whole cell configuration had been established, we waited for 5 min to allow intracellular dialysis of the staurosporine before myocytes were exposed to isosmolar or hyposmolar superfusate, and then I_p was determined. The mean I_p of seven cells exposed to isosmolar solution was 0.24 ± 0.02 pA/pF, and the mean I_p of six cells exposed to hyposmolar solution was 0.40 ± 0.04 pA/pF. The difference was statistically significant. This indicates that nonselective kinase inhibition does not prevent the stimulation of I_p induced by cell swelling. A similar result was obtained when we used 1 µM bisindolylmaleimide I.

Effect of tyrosine kinase inhibition on pump stimulation in hyposmolar solutions. Swelling and mechanical stretch of cardiac myocytes rapidly activate tyrosine kinase (24, 26). Because activation of the insulin receptor tyrosine kinase has an effect on the I_p-V_m relationship (13) similar to the effect of swelling found in this study, we examined the effect of the tyrosine kinase inhibitor tyrphostin A25. In the first series of experiments, we examined the effect of tyrphostin A25 on pump stimulation induced by myocyte swelling at a fixed V_m of 40 mV. We included 100 µM tyrphostin A25 in the patch pipette solution. [Na]_pip was 10 mM. After the whole cell configuration was established, myocytes were superfused with either isosmolar or hyposmolar Ca²⁺-free solutions and I_p was measured. The mean I_p of six myocytes superfused with isosmolar solution was 0.26 ± 0.03 pA/pF. This is similar to the mean I_p measured under similar conditions with tyrphostin A25-free pipette solutions (Fig. 1), suggesting that there is no tonic effect of tyrosine kinase on the I_p of myocytes in isosmolar solutions. When tyrphostin A25 was included in pipette solutions, the mean I_p of six myocytes exposed to hyposmolar solutions was 0.29 ± 0.04 pA/pF. This is similar to mean I_p values of the myocytes exposed to isosmolar solutions, indicating that tyrphostin A25 completely blocked the pump stimulation induced by cell swelling. To exclude a nonspecific effect of tyrphostin A25 on the Na⁺-K⁺ pump, experiments were repeated with 100 µM tyrphostin A63, the inactive analog of tyrphostin A25, in the patch pipette. In the presence of tyrphostin A63, the mean I_p of five myocytes superfused with isosmolar solution and the mean I_p of five myocytes superfused with hyposmolar solutions were 0.20 ± 0.03 and 0.48 ± 0.07 pA/pF, respectively. The difference was statistically significant. Results obtained with tyrphostin A25 and A63 in pipette solutions have been summarized in Fig. 3A. We conclude that the tyrosine kinase inhibitor tyrphostin A25 can abolish the pump stimulation induced by cell swelling and that this is a specific effect of the drug.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   A: effect of tyrosine kinase and phosphatidylinositol 3-kinase (PI3K) inhibition. Tyrosine kinase inhibitor tyrphostin A25 (Tyr A25) and its inactive analog, tyrphostin A63 (Tyr A63), were included in patch pipette solutions. In a separate group of experiments PI3K inhibitor LY-294002 was included in superfusates. Inhibition of tyrosine kinase or PI3K abolished swelling-induced pump stimulation. No. in parentheses is no. of experiments in each group. B: effect of tyrphostin A25 on Ip-Vm relationship during exposure to isosmolar () and hyposmolar () solutions.

Role of phosphatidylinositol 3-kinase in mediating the stimulation of I_p. Activation of phosphatidylinositol 3-kinase (PI3K) is an essential step in the transduction of cellular signals mediated via tyrosine kinase-dependent processes (4, 15, 31). We investigated the role of PI3K in the stimulation of I_p by hyposmolar solutions. Myocytes were superfused before and after establishment of the whole cell configuration with solutions containing the specific PI3K inhibitor LY-294002 at a concentration of 5 µM. The mean I_p of six myocytes superfused with LY-294002-containing isosmolar solutions was 0.29 ± 0.02 pA/pF, and the mean I_p of six myocytes superfused with LY-294002-containing hyposmolar solutions was 0.27 ± 0.02 pA/pF. The difference was not statistically significant. We have included the results of experiments performed with LY-294002 in Fig. 3A. Because PI3K is strongly linked to tyrosine kinase-mediated messenger pathways, the abolition of swelling-induced pump stimulation by LY-294002 strongly supports the conclusion derived from experiments using tyrphostin A25, i.e., that pump stimulation is mediated by the activation of tyrosine kinase.

Role of serine/threonine protein phosphatase in I_p stimulation. Tyrosine kinase has been demonstrated to activate serine/threonine protein phosphatase 1 (PP1) (6). PP1 and another protein phosphatase, PP2A, may play a role in the regulation of the Na⁺-K⁺ pump in the kidney (16). We examined the effect of okadaic acid, an inhibitor of both PP1 and PP2A, on the stimulation of I_p by hyposmolar solutions. In initial experiments we added 1 µM okadaic acid to the patch pipette solution to completely inhibit both PP1 and PP2A (5). We measured I_p in isosmolar and hyposmolar superfusates. The mean I_p in six myocytes measured during exposure to isosmolar superfusate was 0.33 ± 0.03 pA/pF, whereas the mean I_p in eight myocytes measured during exposure to hyposmolar superfusate was 0.37 ± 0.05 pA/pF. This difference was not statistically significant. This suggests that Na⁺-K⁺ pump stimulation during exposure to hyposmolar solutions involves the activation of PP1 and/or PP2A. To exclude a nonspecific effect of okadaic acid on the pump, we repeated the experiments with the inactive analog, okadaic acid methyl ester (1 µM) in pipette solutions. The mean I_p in seven myocytes measured during exposure to isosmolar superfusate was 0.33 ± 0.02 pA/pF, whereas the mean I_p of seven myocytes exposed to hyposmolar superfusates was 0.49 ± 0.03 pA/pF. The difference was statistically significant.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effect of serine/threonine phosphatase inhibition on swelling-induced Ip stimulation. Experiments were performed during exposure to isosmolar or hyposmolar solution in presence of 1 µM or 10 nM okadaic acid. No. in parentheses is no. of experiments in each group.

Reversibility of swelling-induced pump stimulation. We have previously found that changes in the intracellular Na⁺ activity of intact papillary muscles induced by exposure to hyposmolar superfusates are rapidly reversible (32). We performed a series of experiments to examine if this is reflected by reversible changes in the I_p of patch-clamped myocytes. After establishing the whole cell configuration, we superfused isosmolar Ca²⁺-free solution for 3 min. We then switched to a hyposmolar Ca²⁺-free superfusate to induce cell swelling and pump stimulation. After 3 min we switched back to the isosmolar superfusate for 5 min and measured I_p. Figure 5 illustrates the changes in cell size during the experimental protocol. Exposure to the hyposmolar superfusate induced a myocyte swelling that was completely reversible on switching back to the isosmolar superfusate for 5 min. I_p was determined after switching back to isosmolar superfusate at the end of the 5-min period. To control for a possible time-dependent Na⁺-K⁺ pump run-down, control myocytes were superfused with isosmolar solution for 11 min after the whole cell configuration was established before I_p was determined. The protocols for exposure to superfusates and measurement of I_p are illustrated in Fig. 6A. In eight myocytes briefly exposed to hyposmolar solution, the mean I_p measured after switching back to isosmolar solution was 0.33 ± 0.02 pA/pF. The mean I_p of nine myocytes exposed to isosmolar solution only was 0.27 ± 0.03 pA/pF. The difference was not statistically significant. These results suggest that stimulation of I_p by cell swelling is rapidly reversible with restoration of extracellular osmolarity and cell size.

View larger version (57K): [in this window] [in a new window]   Fig. 5.   Reversibility of osmotically induced cell swelling in patch-clamped cardiac myocyte. Shown is a photomicrograph of a patch-clamped myocyte during sequence of exposure to isosmolar (A) and hyposmolar solution (B) and during reexposure to isosmolar solution (C). Calibration bars indicate myocyte width under basal isosmolar conditions.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Reversibility of osmotically induced Ip stimulation. A: protocol used to examine role of protein kinase C in reversing swelling-induced Ip stimulation. B: effect of staurosporine (Stauro) on reversibility of swelling-induced Ip stimulation. Open bars and solid bars, results of protocol without and with staurosporine, respectively, in patch pipette. Hypo, hyposmolar. No. in parentheses is no. of experiments in each group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Role of transmembrane ion fluxes during myocyte swelling. We have previously reported that I_p in rabbit cardiac myocytes is stimulated by exposure to hyposmolar superfusates (32). This effect was observed when [Na]_pip was near the physiological level of intracellular Na⁺ concentration, whereas there was no effect when [Na]_pip was high. We concluded that the stimulation was due to an increase in the apparent affinity of the pump for intracellular Na⁺ rather than to an influx of extracellular Na⁺ and secondary pump stimulation. In the present study, we have directly ruled out any effect of Na⁺ influx during myocyte swelling by demonstrating that swelling activates I_p even in the absence of extracellular Na⁺.

Site of effect of myocyte swelling on the Na⁺-K⁺ pump. Exposure of myocytes to hyposmolar solutions shifted the I_p-V_m relationship in a negative direction (Fig. 2B), suggesting that myocyte swelling affects an electrogenic step in the pump cycle. Voltage dependence is thought to arise, at least in part, from the interaction of the transported ligands with the pump in access channels within the electrical field of the membrane (see Ref. 22 for review). At extracellular sites, the interaction of Na⁺ and K⁺ with the pump within access channels contributes to the slope of the I_p-V_m relationship in a positive and a negative direction, respectively. Experimentally, the voltage dependence of the pump arising from the interaction of Na⁺ with extracellular sites can be eliminated by using Na⁺-free superfusates. Unless the K⁺ concentration is very low, the voltage dependence arising from the interaction of K⁺ with extracellular sites is also eliminated in such Na⁺-free solutions (22). In the present study, swelling-induced pump stimulation persisted when myocytes were superfused with Na⁺-free superfusates containing 5.6 mM K⁺, suggesting that it is not one of the voltage-dependent steps at the extracellular sites that is stimulated by cell swelling.

Messenger cascade linking cell volume to pump activity. We examined the effect of including staurosporine and bisindolylmaleimide I in the pipette solution because a previous study had indicated that the mechanical stretch of cardiac myocytes activates PKC (24). The drugs did not abolish the increase in I_p induced by cell swelling. This might be taken to indicate that PKC does not regulate sarcolemmal Na⁺-K⁺ pump activity. However, it should be noted that total PKC activity, rather than the activity of specific isoforms, was measured in the study of the effect of mechanical stretch of myocytes (24). Because the effects of PKC are believed to be isoform specific (see Ref. 18 for review), it is reasonable to think that the isoforms activated by stretch have no effect on I_p. Previous studies have indicated that PKC is activated during osmotically induced cell shrinkage (29). We found that staurosporine and bisindolylmaleimide I prevented the return of I_p to control values after reexposure of the osmotically swollen myocyte to the isosmolar solution (Fig. 6B). These findings suggest that the PKC isoform(s) activated by cell shrinkage regulates the Na⁺-K⁺ pump in cardiac tissue.