Mode-specific Inhibition of Sodium-Calcium Exchange during Protein Phosphatase Blockade*

The effects of the protein phosphatase inhibitors calyculin A and okadaic acid on Na+/Ca2+exchange activity were examined in transfected Chinese hamster ovary cells expressing the bovine cardiac Na+/Ca2+exchanger. Incubating the cells for 5–10 min with 100 nmcalyculin A reduced exchange-mediated 45Ca2+uptake or Ba2+ influx by 50–75%. Half-maximal inhibition of 45Ca2+ uptake was observed at 15 nm calyculin A. The nonselective protein kinase inhibitors K252a and staurosporine provided partial protection against the effects of calyculin A. Okadaic acid, another protein phosphatase inhibitor, nearly completely blocked exchange-mediated Ba2+ influx. Chinese hamster ovary cells expressing a mutant exchanger in which 420 out of 520 amino acid residues were deleted from the central hydrophilic domain of the exchanger remained sensitive to the inhibitory effects of calyculin A and okadaic acid. Surprisingly, Nao +-dependent Ca2+ efflux appeared to be only modestly inhibited, if at all, by calyculin A or okadaic acid. We conclude that protein hyperphosphorylation during protein phosphatase blockade selectively inhibits the Ca2+ influx mode of Na+/Ca2+ exchange, probably by an indirect mechanism that does not involve phosphorylation of the exchanger itself.

The effects of the protein phosphatase inhibitors calyculin A and okadaic acid on Na ؉ /Ca 2؉ exchange activity were examined in transfected Chinese hamster ovary cells expressing the bovine cardiac Na ؉ /Ca 2؉ exchanger. Incubating the cells for 5-10 min with 100 nM calyculin A reduced exchange-mediated 45 Ca 2؉ uptake or Ba 2؉ influx by 50 -75%. Half-maximal inhibition of 45 Ca 2؉ uptake was observed at 15 nM calyculin A. The nonselective protein kinase inhibitors K252a and staurosporine provided partial protection against the effects of calyculin A. Okadaic acid, another protein phosphatase inhibitor, nearly completely blocked exchange-mediated Ba 2؉ influx. Chinese hamster ovary cells expressing a mutant exchanger in which 420 out of 520 amino acid residues were deleted from the central hydrophilic domain of the exchanger remained sensitive to the inhibitory effects of calyculin A and okadaic acid. Surprisingly, Na o ؉ -dependent Ca 2؉ efflux appeared to be only modestly inhibited, if at all, by calyculin A or okadaic acid. We conclude that protein hyperphosphorylation during protein phosphatase blockade selectively inhibits the Ca 2؉ influx mode of Na ؉ /Ca 2؉ exchange, probably by an indirect mechanism that does not involve phosphorylation of the exchanger itself.
The Na ϩ /Ca 2ϩ exchange system is the primary Ca 2ϩ efflux mechanism in cardiac myocytes and plays a crucial role in regulating the cellular Ca 2ϩ content and the force of myocardial contraction (see reviews in Refs. [1][2][3]. The activity of the Na ϩ /Ca 2ϩ exchanger is regulated by ATP-and Ca 2ϩ -dependent mechanisms. The Ca 2ϩ -dependent process is thought to involve the binding of Ca 2ϩ to regulatory sites within the central hydrophilic domain of the exchanger (4 -6), but the mechanism underlying the ATP-dependent regulatory process has been less clearly defined. Investigations of exchange currents in excised sarcolemmal patches indicate that the effects of ATP in this experimental system are mediated by synthesis of phosphatidylinositol 4,5-bisphosphate (7,8), rather than protein kinase activity (9).
On the other hand, evidence from several different laboratories supports the view that the Na ϩ /Ca 2ϩ exchanger is regu-lated by a phosphorylation-dephosphorylation process. The large body of information favoring a phosphorylation mechanism in squid giant axons has recently been reviewed (10). For the exchangers of vascular smooth muscle and cardiac cells, direct evidence for phosphorylation of the exchange protein and a modest stimulation of exchange activity during protein kinase C activation have been reported (11,12). Studies with rat astrocytes, neurons, and the Xenopus expression system suggest that protein kinase A stimulates the activity of certain exchanger isoforms but not others (13). In contrast to the stimulation of exchange activity described in these studies, inhibition of exchange activity was reported upon activation of protein kinases A and C in bovine chromaffin cells (14) and human mesangial cells (15).
We had previously reported that ATP depletion partially inhibits Na ϩ /Ca 2ϩ exchange activity in transfected Chinese hamster ovary (CHO) 1 cells expressing the bovine cardiac Na ϩ / Ca 2ϩ exchanger (16). A phosphorylation mechanism did not appear to be involved, because inhibitors or activators of various protein kinases did not affect Na ϩ /Ca 2ϩ exchange activity in these cells, and phosphorylation of the immunoprecipitated exchange protein was not detectable. As part of these studies, we reported that okadaic acid and calyculin A, potent inhibitors of protein phosphatases 1 and 2A, had no effect on exchangemediated 45 Ca 2ϩ fluxes when the inhibitors were added directly to the assay medium. In the present report, we reinvestigated the effects of these agents to determine whether exposing the cells to the inhibitors for longer incubation periods would reveal an acceleration of exchange activity through a phosphorylation-dependent process. Instead, we found that calyculin A and okadaic acid strongly inhibited the Ca 2ϩ influx (reverse) mode of Na ϩ /Ca 2ϩ exchange. Curiously, these treatments had little or no effect on the Ca 2ϩ efflux (forward) mode of exchange. The mechanism involved is uncertain, but it does not appear to entail phosphorylation of the exchanger itself.
(37°C) with Tris. K-PSS and NMDG-PSS had the same composition as Na-PSS except that NaCl was replaced with KCl (total concentration, 140 mM) or NMDG. Na-and K-PSS were mixed to yield the Na ϩ and K ϩ concentrations specified as 20/120, 28/112, or 40/100 Na/K-PSS. Fura-2-AM was purchased from Molecular Probes, Inc. (Eugene, OR). Calyculin A and okadaic acid were from Alexis Biochemicals (San Diego, CA). All other biochemicals were purchased from Sigma. 45 Ca 2ϩ Uptake and Efflux Assay-These assays were conducted essentially as previously described (16). Briefly, the culture medium was removed from cells grown in 24-well fibronectin-coated plastic dishes and replaced with 1 ml of nominally Ca 2ϩ -free Na-PSS containing 0.4 mM ouabain. For calyculin-treated cells, calyculin A (100 nM) was added to the preincubation medium after 20 min. After 30 min of incubation at 37°C, the medium in each well was replaced with 0.2 ml of 40/100 Na/K-PSS containing 1 mM 45 CaCl 2 (37°C). After the desired interval, the cells were washed four times with termination fluid consisting of 100 mM MgCl 2 ϩ 10 mM LaCl 3 ϩ 5 mM Mops/Tris, pH 7.4. The radioactivity was solubilized in 0.1 N HNO 3 , counted in a scintillation counter, and normalized to the amount of protein in the wells. For the 45 Ca 2ϩ efflux assay, ouabain-treated cells were loaded with 45 Ca 2ϩ by reverse Na ϩ /Ca 2ϩ exchange in 40/100 Na/K-PSS as described above with or without 100 nM calyculin A present. After the 10-min loading period, the medium was replaced with 1 ml of either Na-PSS or NMDG-PSS, and the reaction was terminated after various intervals as described above.
Fura-2-based Assays of Ba 2ϩ Influx and Ca 2ϩ Efflux-The Ba 2ϩ influx assay for Na ϩ /Ca 2ϩ exchange activity is fully described in Ref. 21. Cells were grown to confluence in 75-cm 2 culture flasks, washed three times with Na-PSS, and incubated for 1 min at 37°C with Na-PSS containing 5 mM EDTA to detach cells from the flask. The suspended cells were centrifuged at 700 ϫ g for 1 min, resuspended in Na-PSS ϩ 1 mM CaCl 2 , centrifuged again, and resuspended in 4 -5 ml of Na-PSS ϩ 1 mM CaCl 2 containing 1% bovine serum albumin. The cells were divided into 300-l aliquots and incubated for 30 min with 3 M fura-2-AM and 0.25 mM sulfinpyrazone (to retard transport of fura-2 out of the cells). The fura-2 and sulfinpyrazone were added as 1000-fold concentrated stock solutions in dimethyl sulfoxide.
After the 30-min loading period, the cells were centrifuged for several seconds in an Eppendorf mini-centrifuge, washed, and preincubated for 5 min in 100 l of Na-PSS ϩ 1 mM CaCl 2 or with other additions and conditions as indicated in individual experiments. These cells were then added directly to fluorescence cuvettes containing 3 ml of 20/120 Na/K-PSS ϩ 0.3 mM EGTA. Gramicidin (2 g/ml) was added to the cuvettes to equilibrate transmembrane Na ϩ and K ϩ gradients. After monitoring fura-2 fluorescence for 30 s, 1 mM BaCl 2 (30 l of 0.1 M stock solution) was added to the cuvette, and fura-2 fluorescence was monitored for an additional 220 or 270 s, as indicated for individual experiments.
Fura-2 fluorescence was measured at 510 nm with alternate excitation at 350 and 390 nm using a Photon Technology International RF-M 2001 fluorometer (South Brunswick, NJ); data points were obtained at 1.8-s intervals. All fluorescence values were corrected for autofluorescence using cells that had not been loaded with fura-2. Data are presented as the ratio of fluorescence for excitation at 350/390 nm and represent the mean values (Ϯ S.E.; error bars shown in figures) for the number of experiments (n) indicated in the figure legends.
To measure Ca 2ϩ efflux activity, the cells were placed in cuvettes containing either Na-or K-PSS ϩ 0.3 mM EGTA. Ca 2ϩ release from internal stores was initiated by adding ionomycin (0.6 l; 10 mM in dimethyl sulfoxide), and fura-2 fluorescence was monitored at the excitation wavelengths appropriate for Ca 2ϩ (340 and 380 nm). Dimethyl sulfoxide alone at the concentrations used had no effect on Ca 2ϩ release. The difference in the amplitude and duration of the ionomycin-induced [Ca 2ϩ ] i transient in K-PSS versus Na-PSS was taken as an indication of exchange-mediated Ca 2ϩ efflux activity (22,23).

RESULTS
Calyculin A Inhibits Ca 2ϩ Influx by Na ϩ /Ca 2ϩ Exchange-To measure Na ϩ /Ca 2ϩ exchange activity, CK1.4 cells were pretreated with ouabain for 30 min in Na-PSS to elevate the cytosolic Na ϩ concentration and then placed in a low [Na ϩ ] medium (40/100 Na/K-PSS) containing 1 mM 45 CaCl 2 . As shown in Fig. 1A, 45 Ca 2ϩ uptake was diminished by 55-75% when the cells were treated with 50 nM calyculin A for the final 10 min of the preincubation period. A 5-10-min exposure to calyculin A was necessary to elicit this inhibitory effect. As previously reported (16), calyculin A did not inhibit 45 Ca 2ϩ uptake when added to the assay medium without preincubation, and it had no effect on 45 Ca 2ϩ uptake in cardiac sarcolemmal vesicles (data not shown). Thus, calyculin A does not directly act on the Na ϩ /Ca 2ϩ exchanger but presumably modulates its activity through a phosphorylation-dependent mechanism.
The nonselective protein kinase inhibitor K252a (1 M) offered partial protection against the inhibitory effects of calyculin A when included in the preincubation medium ( Fig. 1). In the absence of calyculin, K252a had no effect on 45 Ca 2ϩ uptake (data not shown). A related broad spectrum protein kinase inhibitor, staurosporine (1 M), also protected against inhibition by calyculin A (data not shown). Other protein kinase inhibitors were tested but were found not to protect against the effects of calyculin. These included 1 M KN-62 (calmodulindependent protein kinase), 200 M genistein (tyrosine kinases), 0.5 M calphostin C (protein kinase C), 10 M H-89 (protein kinase A), and another relatively nonselective inhibitor, 50 M H-7. When calyculin-treated cells were subsequently incubated in a calyculin-free medium for an additional 10 min, the rates of 45 Ca 2ϩ uptake continued to decline. Including K252a in the calyculin-free medium enhanced recovery, but the rates of 45 Ca 2ϩ uptake remained less than 40% of untreated controls (data not shown).
The Na ϩ /Ca 2ϩ exchange protein contains 11 transmembrane segments with a cytosolically disposed hydrophilic domain of 520 amino acids residing between the 5th and 6th transmembrane segments (1)(2)(3). The hydrophilic domain is thought to be essential for regulation of activity, because its elimination by chymotrypsin treatment or site-directed mutagenesis abrogates normal regulatory behavior (19). CHO cells expressing a mutant exchanger in which 440 residues from this domain have been deleted (CK138 cells) were treated with calyculin A and assayed for 45 Ca 2ϩ uptake by reverse Na ϩ /Ca 2ϩ exchange. As shown in Fig. 1B, calyculin A inhibited 45 Ca 2ϩ uptake by 40 -65% in these cells. As with the CK1.4 cells, inhibition by calyculin A was partially alleviated by K252a. We conclude that the effects of calyculin A are not mediated by phosphorylation of residues within the deleted segment of this mutant. Although phosphorylation of one or more of the residual serine or threonine residues remains a possible mechanism for these effects, this seems implausible due to the absence of normal regulatory behavior in this mutant.
The concentration dependence for inhibition of the initial rate of 45 Ca 2ϩ uptake by calyculin A is shown in Fig. 2. Maximal inhibition (64 Ϯ 4%) was observed at 100 nM, and the half-maximally effective concentration was approximately 15 nM. Longer incubation periods with 100 nM calyculin did not increase the degree of inhibition.
Exchange-mediated Ba 2ϩ Uptake- 45 Ca 2ϩ entering the cells by reverse Na ϩ /Ca 2ϩ exchange is accumulated by intracellular organelles, especially mitochondria. Thus, the inhibition of 45 Ca 2ϩ uptake observed with calyculin A might have involved a reduction in organellar Ca 2ϩ sequestration rather than an inhibition of Na ϩ /Ca 2ϩ exchange activity. To examine this possibility, we measured the exchange-mediated influx of Ba 2ϩ in fura-2-loaded cells treated with 2 g/ml gramicidin as described in Ref. 21. The presence of gramicidin, a channelforming monovalent cation ionophore, eliminates transmembrane Na ϩ and K ϩ gradients and provides a continuous supply of cytosolic Na ϩ for the activation of exchange activity (21). As shown in Fig. 3, treatment of CK1.4 cells or CK138 cells for 5 min with 100 nM calyculin A markedly inhibited the initial rate of Ba 2ϩ influx in both cell types. Because Ba 2ϩ is not accumulated by the endoplasmic reticulum in these cells and does not appear to be extensively accumulated by mitochondria (21), we conclude that the effects of calyculin A reflect the inhibition of exchange activity itself rather than an impairment of organellar Ca 2ϩ sequestration.
The residual rate of Ba 2ϩ influx in the calyculin-treated CK138 cells was greater (p Ͻ 0.025) than for the calyculintreated CK1.4 cells (Fig. 3C). Moreover, the degree of inhibition by calyculin A for the CK138 cells (44 Ϯ 4.6%) was significantly less than for the CK1.4 cells (74 Ϯ 2.4%; p Ͻ 10 Ϫ5 ). Thus, the effects of calyculin A are less pronounced for the mutant than for the wild-type exchanger. A similar pattern was evident in the 45 Ca 2ϩ uptake studies (Fig. 1). We conclude that the hydrophilic domain of the exchanger mediates a portion of the inhibitory effects of protein phosphatase blockade but that substantial inhibition was still observed even in the absence of most of the hydrophilic domain.
When CK1.4 cells were treated with okadaic acid, a protein phosphatase inhibitor with greater potency toward PP-2A than PP-1, we found that exchange-mediated Ba 2ϩ influx was nearly completely blocked (87 Ϯ 2.3% inhibition of initial rate) (Fig. 4, A and C). Incubations of 60 min with okadaic acid were required to observe this inhibition, consistent with a slow entry of this inhibitor into cells (24,25). Under identical conditions, okadaic acid methyl ester, an inactive analog of okadaic acid, had no effect on Ba 2ϩ influx (Fig. 4D). The okadaic acid treatment inhibited the initial rate of Ba 2ϩ influx in CK138 cells by 65 Ϯ 3.4% (Fig. 4B). Although the percentage of inhibition was less than for the CK1.4 cells, the rates of Ba 2ϩ uptake after okadaic acid treatment for both cell types were similar to those seen for passive Ba 2ϩ leakage into the cell in the absence of exchange activity (21). Thus, the values given must be seen as lower limits for the percentage of inhibition of exchange activity by okadaic acid. Tautomycin (1 M), a third protein phosphatase inhibitor with reduced affinity for PP-2A compared with PP-1, had no effect on Ba 2ϩ influx after either a 5-or a 60-min preincubation (Fig. 4D). The results suggest that the effects of calyculin A and okadaic acid are mediated by blockade of protein phosphatase 2A (see under "Discussion").
Other experiments demonstrated that K252a, as in the experiments with calyculin A, partially protected against inhibition of Ba 2ϩ influx by okadaic acid (data not shown). We were unable to mimic the effects of the protein phosphatase inhibitors on Ba 2ϩ influx by incubating the cells with phorbol 12myristate 13-acetate (10 M, 5 min), dibutyryl cAMP (1 mM, 5 or 60 min) or 8-bromo-cGMP (1 mM, 5 or 60 min).
To determine whether the inhibition of Ba 2ϩ entry could be overcome by increasing the cytosolic Na ϩ concentration, we carried out the experiment shown in Fig. 5. CK1.4 cells were preincubated for 5 min with 2 g/ml gramicidin in either Na-PSS or 28/112 Na/K-PSS, with or without 100 nM calyculin A. Under these preincubation conditions, the cytosol was loaded with either a high concentration (ϳ140 mM) or a low concentration (ϳ28 mM) of Na ϩ due to the presence of the monovalent cation ionophore gramicidin; the K m for activation of exchange activity by cytosolic Na ϩ is 18 -28 mM (19,26). The cells were then diluted into K-PSS, and 1 mM Ba 2ϩ was added after 30 s. As expected, the cells with high cytosolic Na ϩ (Fig. 5A) showed a greater rate and extent of Ba 2ϩ uptake than cells with low cytosolic Na ϩ (Fig. 5B). Calyculin A treatment inhibited the initial rates of Ba 2ϩ influx by 57 Ϯ 3.6 versus 68 Ϯ 3.3% at 140 and 28 mM Na ϩ , respectively (Fig. 5C). The difference in the two values (p Ͻ 0.05) is quite small and suggests that calyculin A did not markedly affect the affinity of the exchanger for Na ϩ . In other experiments (data not shown), Ba 2ϩ influx was inhibited to the same extent by calyculin A at Ba 2ϩ concentrations of 1 mM and 5 mM; the K m for external Ba 2ϩ uptake by Na ϩ /Ca 2ϩ exchange in these cells is 3 mM (21). The results indicate that calyculin A inhibition is characterized by a reduction in the V max for exchange activity, rather than changes in the affinity of the exchanger for the transported ions.
Protein Phosphatase Inhibitors and Ca 2ϩ Efflux-We expected the protein phosphatase inhibitors to reduce the Ca 2ϩ efflux mode of exchange activity to a similar extent as for the Ca 2ϩ influx mode. Surprisingly, this was not the case. We adopted three different protocols to assess Ca 2ϩ efflux mediated by Na ϩ /Ca 2ϩ exchange. In the first approach, ouabaintreated CK1.4 cells were loaded with 45 Ca 2ϩ by a 10 min incubation in 40/100 Na/K-PSS containing 1 mM 45 CaCl 2 , with or without 100 nM calyculin A. The medium was then replaced with either NMDG-PSS or Na-PSS, and the rate of loss of 45 Ca 2ϩ from the cells was monitored. As shown in Fig. 6, 45 Ca 2ϩ efflux was more rapid in Na-PSS than in NMDG-PSS, presumably reflecting the acceleration of Ca 2ϩ efflux by Na ϩ / Ca 2ϩ exchange. Remarkably, calyculin A had no effect on Na ϩ - dependent 45 Ca 2ϩ efflux. In other experiments, the cells were treated with calyculin A in Na-PSS for 10 min before beginning the 10 min 45 Ca 2ϩ loading period in 40/100 Na/K-PSS. Although the initial level of 45 Ca 2ϩ uptake was lower under these conditions than in Fig. 6, the results were essentially the same; i.e. the presence of Na ϩ stimulated 45 Ca 2ϩ efflux, indicating that exchange activity had not been blocked by the calyculin A treatment (data not shown).

FIG. 3. Effect of calyculin A on Ba 2؉ influx in CK1.4 (A) and CK138 (B) cells.
Cells were loaded with fura-2 and assayed for Ba 2ϩ influx in 20/120 Na/K-PSS containing 2 g/ml gramicidin (see under "Experimental Procedures"). Prior to the assay, the cells were preincubated for 5 min in Na-PSS ϩ 1 mM CaCl 2 with (CA) or without 100 nM calyculin A. C, initial rates of Ba 2ϩ influx were measured as the slopes of the individual traces over the interval 88 -120 s (n ϭ 6 -9).  (21), ionomycin does not act as an ionophore toward Ba 2ϩ (n ϭ 5-6).
Another approach to measuring Ca 2ϩ efflux was to assess the effect of extracellular Na ϩ on the [Ca 2ϩ ] i transient elicited by the Ca 2ϩ ionophore ionomycin, which releases Ca 2ϩ from intracellular stores. For the data shown in Fig. 7, A and B, ionomycin (2 M) was added to suspensions of CK1.4 cells that had been preincubated for 5 min with or without 100 nM calyculin A. The suspension medium was either Na-PSS ϩ 0.3 mM EGTA or K-PSS ϩ 0.3 mM EGTA, as indicated in the figures. As shown for untreated cells in Fig. 7A, the [Ca 2ϩ ] i transient was reduced in amplitude and shorter in duration in Na-PSS compared with K-PSS, a result that reflects the activity of the Na ϩ /Ca 2ϩ exchanger in mediating Ca 2ϩ efflux (22,23). Halftimes for the decline in [Ca 2ϩ ] i from the peak value were 8.5 and 16 s in Na-and K-PSS, respectively, as determined from first-order plots of the data (see legend to Fig. 7

for details.)
With calyculin-treated cells (Fig. 7B), the half-times were 13 and 26 s in Na-and K-PSS, respectively, indicating that calyculin A prolonged the [Ca 2ϩ ] i transient in both Na-and K-PSS. The reasons for this behavior are not known. As discussed in detail elsewhere (23), the time course of the Ca 2ϩ transient reflects a complex interplay between the Ca 2ϩ content of intracellular stores, rates of Ca 2ϩ efflux from the cell, and the time dependence of Ca 2ϩ release from internal stores, which occurs over several tens of seconds in these cells. Calyculin treatment could have affected any of these contributing processes. We assumed that the rate constant for the decline in [Ca 2ϩ ] i in Na-PSS was the sum of the Na-dependent component (due to Na ϩ /Ca 2ϩ exchange) and the Na-independent component, as measured in K-PSS. Subtracting the rate constants in K-PSS from those obtained in Na-PSS (cf. legend to Fig. 7), the Na ϩdependent component was found to be reduced by 37% in the calyculin-treated cells. Although this value suggests that exchange activity may have been inhibited by the calyculin A treatment, the results with okadaic acid do not support such a conclusion, as discussed below.
The ionomycin-induced [Ca 2ϩ ] i transients for untreated cells and cells incubated for 60 min with 1 M okadaic acid are shown in Fig. 7, C and D. The half-times for the decline in [Ca 2ϩ ] i in untreated cells (Fig. 7C) were 14 and 18 s in Na-and K-PSS, respectively. With the batch of cells used for this experiment, the effect of Na ϩ was somewhat reduced compared with the results described for the cells in Fig. 7A. For cells treated with okadaic acid (Fig. 7D), the corresponding halftimes were 26 and 46 s. In this case, the Na ϩ -dependent component was not reduced but increased by 16% following the okadaic acid treatment, suggesting that there was no impairment of exchange activity. This analysis must be viewed cautiously because the Na-dependent component of the decline in [Ca 2ϩ ] i provides only a rough estimate of exchange activity. As discussed elsewhere (23), ionomycin-induced Ca 2ϩ release is not instantaneous in these cells but appears to continue at a declining rate over several tens of seconds. Thus, the duration of the [Ca 2ϩ ] i transient will be determined to a significant degree by the underlying rate of Ca 2ϩ release. Both calyculin A and okadaic acid induce major cytoskeletal alterations (cf. "Discussion"), which could influence the quantity of stored Ca 2ϩ and/or the rate of Ca 2ϩ release. These factors make it difficult to quantitate exchange activity with certainty in these experiments. Nevertheless, the results clearly indicate that the exchanger remains capable of carrying out Ca 2ϩ efflux following treatment with the phosphatase inhibitors. The results with okadaic acid are especially noteworthy because this agent almost completely abrogated exchange-mediated Ba 2ϩ influx (cf. Fig. 4) but did not appear to markedly impair Ca 2ϩ efflux.
A final index of exchange-mediated Ca 2ϩ efflux is shown in Fig. 8. In this experiment, CK1.4 cells were pretreated for 10 min, with or without 100 nM calyculin A, in the presence of ionomycin to deplete internal stores. The cells were then added to cuvettes containing either Na-PSS or K-PSS (final ionomycin concentration, 1.3 M), and the rate of Ca 2ϩ entry upon addition of 1 mM CaCl 2 was monitored. In these experiments, Ca 2ϩ entry occurs both through the ionophoretic behavior of ionomycin and through store-dependent Ca 2ϩ entry channels. For the CK1.4 cells in K-PSS, reverse Na ϩ /Ca 2ϩ exchange also contributes to Ca 2ϩ entry (22). As shown in Fig. 8, the initial rate of Ca 2ϩ influx was reduced by 30 Ϯ 9% in Na-PSS (Fig. 8D), whereas the amplitude of the rise in [Ca 2ϩ ] i was reduced by 53 Ϯ 6% (Fig. 8A). The reduced rise in [Ca 2ϩ ] i in Na-PSS reflects the effect of Na ϩ in inhibiting Ca 2ϩ influx and promoting Ca 2ϩ efflux via the exchanger (22). This interpretation is supported by the data for vector-transfected control cells, which do not express an endogenous Na ϩ /Ca 2ϩ exchanger; for these cells, both the rate (Fig. 8D) and the extent (Fig. 8C) of the rise in [Ca 2ϩ ] i were identical in Na-PSS and K-PSS.
With calyculin-treated CK1.4 cells (Fig. 8, B and D), the rate of Ca 2ϩ entry in K-PSS was reduced by 42 Ϯ 8% compared with untreated cells, probably because calyculin A-treatment inhibits both store-dependent Ca 2ϩ entry (see under "Discussion") and reverse Na ϩ /Ca 2ϩ exchange. In Na-PSS, the initial rate of Ca 2ϩ influx (Fig. 8D) was reduced by 62 Ϯ 6% in Na-PSS compared with K-PSS, whereas the amplitude of the rise in [Ca 2ϩ ] i was reduced by 59 Ϯ 4% (Fig. 8B). Note that the effect of Na ϩ on the initial rate of Ca 2ϩ entry was proportionally greater for the calyculin-treated cells (62% inhibition) than for the untreated cells (30% inhibition) (p Ͻ 0.001) (Fig. 8D). If we assume that the differences between the initial rates of Ca 2ϩ influx in K-PSS and Na-PSS were entirely due to Ca 2ϩ efflux by the exchanger, the results would indicate that the rate of Na o ϩdependent Ca 2ϩ efflux had increased by 21 Ϯ 15% in the calyculin-treated cells. We conclude that the calyculin A treatment did not inhibit exchange-mediated Ca 2ϩ efflux in these experiments.
For the vector-transfected control cells, calyculin A treatment inhibited the initial rate of Ca 2ϩ entry by 66 Ϯ 9% (Fig.  8, C and D), probably due to inhibition of store-dependent Ca 2ϩ entry (see under "Discussion"). The reasons for the greater inhibition of Ca 2ϩ entry by calyculin A in control cells com-pared with CK1.4 cells are not known. In any event, the rates of Ca 2ϩ entry in Na-and K-PSS were essentially identical in the calyculin-treated control cells, indicating that the effects of Na ϩ seen in the CK1.4 cells were indeed due to Na ϩ /Ca 2ϩ exchange activity. DISCUSSION Calyculin A is a potent inhibitor of PP-1 and PP-2A, with K i values of 1 and 0.12 nM, respectively, as determined from recent binding and dose inhibition measurements (25). Okadaic acid exhibits K i values of 0.032 nM for PP-2A and 147 nM for PP-1 (25). In intact cells, phosphatase inhibitors must be applied at concentrations much higher than their K i values to be effective, presumably because of poor entry into the cells. Thus, in a recent study with MCF7 cells, Favre et al. (27) reported that 100 nM calyculin A inhibited both PP-1 and PP-2A activity by 60 -70% within 20 min, the earliest time point examined; a lower concentration of calyculin A (10 nM) required nearly 2 h of incubation to produce the same degree of inhibition. Longer incubations of intact cells with 100 nM calyculin A did not produce more than 70% inhibition of either phosphatase, although 100% inhibition could be observed in cell-free extracts. In the same study, okadaic acid (1 M) inhibited nearly 100% of the PP-2A activity within 60 min but did not inhibit PP-1 activity at all during a 3-h incubation period. Thus, the response of protein phosphatases to inhibitors in intact cells is often slow and incomplete, and this behavior complicates the interpretation of results.
In the present study, calyculin A inhibited "reverse-mode" exchange activity by 60 -75% after incubation periods of 5-10 min, and okadaic acid inhibited Ba 2ϩ influx by nearly 100% after incubation periods of 60 min. The relative effectiveness of calyculin A and okadaic acid is similar to that reported by Favre et al. (27) for PP-2A inhibition, and we therefore suggest that the effects of these inhibitors are probably mediated by blockade of PP-2A. We also tested tautomycin (1 M), which has a reduced affinity for PP-2A (30 nM) compared with PP-1 (0.5 nM) (25), and did not observe inhibition of exchange activity (Fig. 4C). The significance of this finding is uncertain, however, because different cell types vary greatly in their response to tautomycin. In the experiments of Favre et al. (27), for example, exposure to 10 M tautomycin for more than 2 h was required for 50% inhibition of PP-1 activity. On the other hand, studies in our laboratory with L6 cells revealed significant effects of 1 M tautomycin within 5-10 min of incubation. 2 The effects of the phosphatase inhibitors probably result from hyperphosphorylation of cellular proteins, because protein kinase inhibitors such as K252a (Fig. 1) and staurosporine partially protected against the effects of calyculin A and okadaic acid. More selective inhibitors of protein kinase A, protein kinase C, or calmodulin-dependent protein kinase did not influence the effects of calyculin A. Moreover, we were unable to mimic the effects of the protein phosphatase inhibitors by incubating the cells with agents known to activate protein kinases A, C, and G. Thus, the protein kinase(s) responsible for the phosphorylation-dependent inhibition of exchange activity remain unidentified.
Direct phosphorylation of the exchanger itself is probably not involved in these effects, although this possibility cannot be completely eliminated. Previously published immunoprecipitation experiments with 32 P-labeled CK1.4 cells failed to disclose detectable amounts of phosphorylated exchanger, even in calyculin-treated cells (16). (Exchanger phosphorylation has been detected in other cell types, however (11,12).) Moreover, cells expressing a deletion mutant (CK138 cells), which is missing much of the central hydrophilic domain of the exchanger, remained susceptible to the effects of protein phosphatase blockade (Figs. 1, 3, and 4). Many residual serines and threonines remain in this mutant, including a potential casein kinase 2 phosphorylation site at 720-SAGEDDDD. However, it is unlikely that the deletion mutant could mediate a regulatory response to putative phosphorylation, because previous studies have demonstrated that all well characterized modes of exchanger regulation are absent in this mutant (19). Calyculin A and okadaic acid were somewhat less effective in blocking Ba 2ϩ influx in the mutant than the wild-type exchanger (Figs. 1 and  3). These results suggest that the hydrophilic domain mediates a portion of the inhibitory effect of these agents on exchange activity. Although it is possible, but unlikely, that this portion of the response is due to exchanger phosphorylation, other aspects of exchanger function mediated by the hydrophilic domain, e.g. possible interactions with cytoskeletal proteins (16), could equally well be involved.
It is remarkable that exchange-mediated Ca 2ϩ efflux activity appeared to be resistant to the effects of protein phosphatase blockade (Figs. 6 -8). Na o ϩ -dependent 45 Ca 2ϩ efflux was unaffected by calyculin A treatment (Fig. 6). Na ϩ also remained effective in attenuating the ionomycin-induced increase in [Ca 2ϩ ] i in cells treated with calyculin A or okadaic acid (Fig. 7). Although the results with calyculin A were consistent with partial inhibition of exchange-mediated Ca 2ϩ efflux, the results with okadaic acid were not. The okadaic acid data are particularly noteworthy because this agent blocked exchange-mediated Ba 2ϩ influx nearly completely (Fig. 4, A and C). Finally, the presence of extracellular Na ϩ blunted the rise in [Ca 2ϩ ] i produced by the addition of extracellular Ca 2ϩ in ionomycintreated CK1.4 cells, whether or not the cells had been pretreated with calyculin A (Fig. 8). The results of each of these approaches are consistent with the conclusion that the protein phosphatase blockade did not inhibit exchange activity operating in the Ca 2ϩ efflux mode.
The selective inhibition of the reverse mode of exchange activity did not appear to involve changes in the affinity for the transported ions, or alterations in normal regulatory behavior. Thus, the inhibition by calyculin A was nearly the same at concentrations of cytosolic Na ϩ (Fig. 5) or external Ba 2ϩ (data not shown) that were above the K m value compared to concentrations at or below the K m value. Moreover, the CK138 cells, which express an unregulated exchanger (19), remain susceptible to inhibition by protein phosphatase blockade, albeit with reduced sensitivity compared with the CK1.4 cells (Figs. 1, 3,  and 4). Thus, an explanation for the results in terms of the conventional regulatory or kinetic properties of the exchanger remains elusive. Recently, the novel exchange antagonist KB-R7943 was shown to be considerably more potent in inhibiting the reverse, Ca 2ϩ influx mode of exchange activity than the Ca 2ϩ efflux mode (28,29). The mechanism underlying this differential inhibition of activity has not yet been clarified. Perhaps there is an intrinsic difference in susceptibility of the two modes of exchange activity to inhibitors or other modulatory agents. The situation might be analogous to Na i -dependent inactivation (30,31), in which exchange carriers enter an inactive state when they interact with high concentrations of Na ϩ at the cytosolic membrane surface. It is possible that in this configuration, the exchanger is preferentially sensitive to inhibition by phosphatase blockade or the presence of KB-R7943. This property could be an important factor in protecting cells against Ca 2ϩ overload during periods of ischemia, when the cystosolic Na ϩ concentration rises to levels that would bring about net Ca 2ϩ influx by reverse Na ϩ /Ca 2ϩ exchange.
Dramatic cytoskeletal changes occur in many cells upon in-hibition of protein phosphatase activity (32)(33)(34)(35). These cytoskeletal changes might influence exchanger location or its interactions with associated proteins, thereby inhibiting its activity. Preliminary microscopic observations using fluorescently labeled probes or antibodies have verified that CK1.4 cells exposed to calyculin A exhibit extensive disruption of intermediate filaments, microtubules and stress fibers (data not shown). The relationship, if any, between these cytoskeletal alterations and exchange activity is unclear. Treatment of CK1.4 cells with nocodazole and/or cytochalasin D, agents that disrupt microtubules and F-actin filaments, respectively, did not inhibit exchange activity in preliminary experiments (data not shown). Disruption of the intermediate filament system could conceivably affect exchange activity, because the cytoskeletal protein ankyrin interacts with both intermediate filaments (36,37) and the Na ϩ /Ca 2ϩ exchanger (38). This suggestion is entirely speculative, however. Protein phosphatase inhibitors often cause extensive vacuolization and membrane internalization (39,40), and this process could lead to loss of exchanger protein from the external membrane surface. However, this possibility is difficult to reconcile with the Ca 2ϩ efflux results, which suggest little or no impairment of exchanger function. Moreover, wholesale internalization of plasma membrane transporters seems unlikely because calyculin A treatment did not inhibit Na ϩ /H ϩ exchange or bumetanide-sensitive 86 Rb ϩ uptake activities (data not shown). At present, therefore, it is uncertain whether the extensive cytoskeletal alterations produced during protein phosphatase blockade are an important factor in the inhibition of exchange activity. A final consideration involves the effects of the protein phosphatase inhibitors on other Ca 2ϩ influx mechanisms. Calyculin A and okadaic acid strongly inhibit store-dependent Ca 2ϩ entry in several different cell types (41)(42)(43)(44)(45), although there have also been reports that okadaic acid stimulates store-dependent Ca 2ϩ entry under some conditions (46,47). We have found that calyculin A reduces store-dependent entry of Ca 2ϩ , Ba 2ϩ , and Mn 2ϩ in CHO cells. 3 We have also found that calyculin A and okadaic acid treatments block Ba 2ϩ entry through L-type Ca 2ϩ channels in the L6 rat skeletal muscle cell line. 2 These findings raise the possibility that protein phosphatase blockade elicits a general inhibition of divalent cation entry, perhaps involving cytoskeletal changes as a common mechanistic element. Studies examining this possibility are currently in progress.
In summary, the results have revealed a profound inhibition of the Ca 2ϩ influx mode of Na ϩ /Ca 2ϩ exchange activity during protein phosphatase blockade. Surprisingly, the Ca 2ϩ efflux mode of exchange activity appeared to be relatively unaffected. The mechanism involved is unclear, but it probably does not involve phosphorylation of the exchanger itself. The selective inhibition of Ca 2ϩ influx could involve an instrinsic difference in the susceptibility of the two modes of exchange activity to protein hyperphosphorylation or to cytoskeletal changes, or possibly a more general mechanism that suppresses divalent cation entry.