A novel isothiourea derivative selectively inhibits the reverse mode of Na+/Ca2+ exchange in cells expressing NCX1.

No.7943 (2-[2-[4-(4-nitrobenzyloxy)phenyl]ethyl]isothiourea methanesulfonate), a selective inhibitor of the Na+/Ca2+ exchanger (NCX1), has been newly synthesized. It dose-dependently inhibited Na+i-dependent 45Ca2+ uptake and Na+i-dependent [Ca2+]i increase in cardiomyocytes, smooth muscle cells, and NCX1-transfected fibroblasts (IC50 = 1.2-2.4 microM). Inhibition was observed without prior incubation with the agent and was completely reversed by washing cells with buffer for 1 min. Interestingly, No.7943 was much less potent in inhibiting Na+o-dependent 45Ca2+ efflux and Na+o-induced [Ca2+]i decline (IC50 = >30 microM), indicating that it selectively blocks the reverse mode of Na+/Ca2+ exchange in intact cells. In cardiac sarcolemmal preparations consisting mostly of inside-out vesicles, the agent inhibited Na+i-dependent 45Ca2+ uptake and Na+o-dependent 45Ca2+ efflux with similar, but slightly lower, potencies (IC50 = 5.4-13 microM). Inhibition was noncompetitive with respect to Ca2+ and Na+ in both cells and sarcolemmal vesicles. These results suggest that No.7943 primarily acts on external exchanger site(s) other than the transport sites in intact cells, although it is able to inhibit the exchanger from both sides of the plasma membrane. No.7943 at up to 10 microM does not affect many other ion transporters nor several cardiac action potential parameters. This agent at these concentrations also did not influence either diastolic [Ca2+]i or spontaneous beating in cardiomyocytes. Furthermore, No.7943 markedly inhibited Ca2+ overloading into cardiomyocytes under the Ca2+ paradox conditions. Thus, No.7943 is not only useful as a tool with which to study the transport mechanism and physiological role of the Na+/Ca2+ exchanger but also has therapeutic potential as a selective blocker of excessive Ca2+ influx mediated via the Na+/Ca2+ exchanger under pathological conditions.

exchanger catalyzes bidirectional electrogenic exchange of Na ϩ for Ca 2ϩ across the plasma membrane, its direction being determined by the magnitude and orientation of electrical and chemical ion gradients. The exchanger works in concert with other cellular Ca 2ϩ transporters including the sarcolemmal Ca 2ϩ pump and Ca 2ϩ channels and intracellular Ca 2ϩ sequestration and release systems. Thus the function of the exchanger under physiological or pathological conditions is often difficult to define, because the membrane potential or intracellular concentrations of Na ϩ and Ca 2ϩ may vary in different cell types and change in response to agonist or electrical stimulation.
Recent molecular cloning studies have revealed that the Na ϩ /Ca 2ϩ exchanger isoforms expressed in various cell types are highly homologous to the cardiac clone and are the product of the same gene (NCX1) (3)(4)(5). These isoforms, however, differ in a small region near the carboxyl end of the large central loop, which is due to alternative splicing (4,5). In brain and skeletal muscle, a Na ϩ /Ca 2ϩ exchanger isoform that is a product of a different gene (NCX2) is also expressed (6).
The physiological role of the Na ϩ /Ca 2ϩ exchanger has been studied most extensively in cardiac muscle. During each action potential, the exchanger rapidly extrudes the Ca 2ϩ that has entered the cardiomyocytes via the sarcolemmal L-type Ca 2ϩ channels to trigger the release of Ca 2ϩ from the SR (7,8). In addition, the exchanger has been shown to play a much greater role than the sarcolemmal Ca 2ϩ pump in the slow extrusion of Ca 2ϩ from cardiomyocytes during diastole or under resting conditions (7,9). On the other hand, the exchanger appears capable of bringing Ca 2ϩ into cardiomyocytes during cardiac depolarization, although triggering the release of Ca 2ϩ from the SR via the exchanger is much less efficient than via the L-type Ca 2ϩ channels (10). Under pathological conditions such as ischemia-associated reperfusion injury, the exchanger is thought to cause Ca 2ϩ overloading of cardiomyocytes due to an increase in [Na ϩ ] i (11,12). In other cell types, including nonexcitable cells such as kidney cells, however, the specific contribution of the exchanger to the [Ca 2ϩ ] i regulation has been difficult to determine, because of the relatively low density of the exchanger in the plasma membranes of these cells and the lack of a specific inhibitor.
A potent and selective inhibitor of the Na ϩ /Ca 2ϩ exchanger, if available, should be extremely useful to study the reaction mechanism of Na ϩ /Ca 2ϩ exchange and to clarify its physiological and pathophysiological roles. Moreover, such an inhibitor may serve as a therapeutic agent by virtue of its inotropic, cardioprotective, antiarrhythmic, or antihypertensive effects. Although a variety of natural products, synthetic organic compounds, and inorganic cations have been tested for their ability to inhibit the exchanger (13), few selective inhibitors exist. * This work was supported by Grant-in-Aid 353 for Scientific Research on Priority Areas from the Ministry of Education, Science and Culture of Japan and by Special Coordination Funds Promoting Science and Technology (Encouragement System of COE) from the Science and Technology Agency of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ To whom correspondence and reprint requests should be addressed: Dept. of Molecular Physiology, National Cardiovascular Center Research Institute, Fujishiro-dai 5, Suita, Osaka 565, Japan. Tel.: 81-6-833-5012 (ext. 2519); Fax: 81-6-872-7485. 1  Amiloride analogues such as 3Ј,4Ј-dichlorobenzamil inhibit Na ϩ /Ca 2ϩ exchange at micromolar concentrations, but they exert a cytotoxic effect by inhibiting a number of other ion transporters and cell metabolism (14 -16). On the other hand, XIP, a synthetic peptide derived from the amino acid sequence of the cardiac Na ϩ /Ca 2ϩ exchanger, inhibits Na ϩ /Ca 2ϩ exchange with high potency (IC 50 ϭ 0.1Ϫ1 M) and relatively high specificity (17) and has thus been used in previous studies (18,19). However, XIP is highly cationic and interacts with calmodulin. With the latter property, it modulates activities of calmodulin-activated enzymes such as the sarcolemmal Ca 2ϩ -ATPase (17). Since XIP does not seem to permeate through the cell membrane but acts from the cytoplasmic surface, its use is significantly limited. Therefore, development of a new potent inhibitor that is selective for the Na ϩ /Ca 2ϩ exchanger in vivo is highly desired.
We report here that a newly synthesized compound, 2-[2-[4-(4-nitrobenzyloxy)phenyl]ethyl]isothiourea methanesulfonate (designated No.7943), is a potent and selective inhibitor of the Na ϩ /Ca 2ϩ exchanger. This compound ( Fig. 1) has been identified by screening a compound library for inhibition of Na ϩ i -dependent Ca 2ϩ uptake into isolated cardiac sarcolemmal vesicles. Surprisingly, this compound selectively inhibits the reverse mode of Na ϩ /Ca 2ϩ exchange in intact cells. We also describe that this compound prevents the excess Ca 2ϩ influx evoked by the Ca 2ϩ paradox, which has been widely studied as an experimental model for Ca 2ϩ overloading in cardiomyocytes.

EXPERIMENTAL PROCEDURES
Cell Cultures-Primary cultures of neonatal rat cardiomyocytes were prepared by the method described previously (20). Briefly, hearts from 1-or 2-day-old Wistar rats were minced, and cells were dissociated with 0.1% trypsin in buffer A (20 mM Hepes/Tris (pH 7.4), 137 mM NaCl, 5.36 mM KCl, 0.81 mM MgSO 4 , 0.44 mM KH 2 PO 4 , 0.34 mM Na 2 HPO 4 , and 5.55 mM glucose) containing no added Ca 2ϩ . After centrifugation, the pellet was resuspended in DMEM (Life Technologies, Inc.) supplemented with 5% heat-inactivated FCS, 1.5 M vitamin B 12 , 1 g/ml insulin, 5 g/ml transferrin, 50 units/ml penicillin, and 50 M streptomycin. Dispersed cells were placed in 150-mm dishes (Falcon) for 1 h, and non-attached cells were seeded onto polystyrene dishes or onto sterile glass coverslips. To inhibit growth of fibroblasts, 10 M cytosine arabinoside was included in the final culture medium for 48 h. Spontaneously beating cells were used after 3-5 days of culture.
Vascular smooth muscle cells were isolated from the thoracic aorta of male Wistar rats (200 -300 g) by enzymatic dispersion as described by Chamley et al. (21). The cells were grown for 4 to 5 days in DMEM supplemented with 10% heat-inactivated FCS and antibiotics as above. After reaching confluency, cells were cultured in serum-free medium for an additional 24 -48 h to enhance redifferentiation. CCL39 cells (ATCC) were maintained in DMEM supplemented with 7.5% heatinactivated FCS and antibiotics.
Stable Expression of NCX1 in CCL39 Cells-The 3-kilobase SmaI-HindIII fragment of the dog cardiac Na ϩ /Ca 2ϩ exchanger cDNA (3,4) was inserted into the mammalian expression vector pKCRH (22). CCL39 cells were transfected with the constructed vector and then screened for high expression of Na ϩ /Ca 2ϩ exchanger as described previously (23). The selected cells stably expressed about 15-fold higher amounts of the Na ϩ /Ca 2ϩ exchanger protein as compared with nontransfected cells.
Na ϩ i -dependent 45 Ca 2ϩ Uptake into Cells-Cells in 24-well dishes were loaded with Na ϩ by incubation at 37°C for 30 min in 0.5 ml of normal BSS (10 mM Hepes/Tris (pH 7.4), 146 mM NaCl, 4 mM KCl, 2 mM MgCl 2 , 1 mM CaCl 2 , 10 mM glucose, and 0.1% BSA) containing 1 mM ouabain and 10 M monensin. 45 Ca 2ϩ uptake was initiated by switching the medium to Na ϩ -free BSS, replacing NaCl with equimolar choline chloride, or to normal BSS, both of which contained 0.1-4 mM (370 kBq) 45 CaCl 2 , 1 mM ouabain, and 10 M verapamil. After a 15-or 30-s incubation, 45 Ca 2ϩ uptake was stopped by washing cells four times with an ice-cold solution containing 10 mM Hepes/Tris (pH 7.4), 120 mM choline chloride, and 10 mM LaCl 3 . Cells were solubilized with 0.1 N NaOH, and aliquots were taken for determination of radioactivity and protein. Na ϩ i -dependent 45 Ca 2ϩ uptake was estimated by subtracting 45 Ca 2ϩ uptake in normal BSS from that in Na ϩ -free BSS. Na ϩ o -dependent 45 Ca 2ϩ Efflux from Cells-45 Ca 2ϩ efflux from cells cultured in a 35-mm dish was assayed as described previously (24). Cells were equilibrated with 45 Ca 2ϩ by incubating them at 37°C for 4 h in 1 ml of BSS containing 740 kBq of 45 Ca 2ϩ . After rinsing cells six times with Ca 2ϩ -and Na ϩ -free BSS for 1 min, 45  ] i was monitored using fura-2 as a fluorescent Ca 2ϩ indicator. Cells cultured on glass coverslips were loaded with 4 M fura-2/acetoxymethyl ester for 20 min at 37°C in buffer A containing 1 mM CaCl 2 and 0.1% BSA (for cardiomyocytes) or in BSS (for smooth muscle cells and transfected CCL39 cells). Loaded cells were then washed twice with the same medium. Glass coverslips were fixed to a mount that was diagonally inserted into a cuvette filled with 2.2 ml of the particular medium. The fluorescence signal was monitored and [Ca 2ϩ ] i calculated as described previously (25).
Assay of Na ϩ /Ca 2ϩ Exchange in Sarcolemmal Vesicles-Sarcolemmal vesicles were prepared from dog ventricular muscle according to Jones (26). Na ϩ i -dependent Ca 2ϩ uptake into vesicles was measured essentially as described previously (27). Briefly, 5 l of Na ϩ -loaded vesicles (1-2 mg/ml) was rapidly diluted into 0.25 ml of uptake medium (20 mM Mops/Tris (pH 7.4), 160 mM KCl, 5-80 M 45 CaCl 2 (10 kBq), and 0.5 M valinomycin) at 37°C. The reaction was stopped at 1.5 s by adding 4 ml of ice-cold washing medium (160 mM KCl and 1 mM LaCl 3 ). Vesicles were collected on a glass fiber filter and washed twice with the same medium. Blanks were obtained by measuring 45 Ca 2ϩ uptake in medium containing NaCl instead of KCl. These blanks were subtracted from all data points to correct for Na ϩ -independent 45 Ca 2ϩ uptake.
Na ϩ o -dependent Ca 2ϩ efflux was quantitated by measuring Na ϩ oinduced 45 Ca 2ϩ loss from vesicles that had been preloaded with 45 Ca 2ϩ by Na ϩ i -dependent Ca 2ϩ uptake (28). Briefly, Na ϩ -loaded vesicles (5 l) were diluted with 0.5 ml of the uptake medium containing 10 M 45 CaCl 2 for 2 min at 37°C. 45 Ca 2ϩ efflux was then initiated by addition of 0.5 ml of efflux medium (20 mM Mops/Tris (pH 7.4), 120 -160 mM KCl, 0.2 mM EGTA, and 0 -40 mM NaCl). The reaction was stopped 10 s later by addition of the washing medium. Blanks were obtained by measuring 45 Ca 2ϩ loss in the efflux medium containing no NaCl.
Assays of Other Ion Transporters-L-type Ca 2ϩ channel activity was assayed by measuring DHP-sensitive 45 Ca 2ϩ uptake into cultured smooth muscle cells as described previously (25). Briefly, cultured smooth muscle cells in 24-well dishes were preincubated at 37°C for 30 min in 0.5 ml of normal BSS. To initiate 45 Ca 2ϩ uptake, cells were rinsed with BSS containing 1 mM (370 kBq) 45 CaCl 2 in the presence or absence of 1 M (ϩ)-PN200 -110. After 2 min, 45 Ca 2ϩ uptake was stopped, and radioactivity and protein were determined. DHP-sensitive 45 Ca 2ϩ uptake was estimated by subtracting 45 Ca 2ϩ uptake in the presence of (ϩ)-PN200 -110 from that in the absence of (ϩ)-PN200 -110.
Na ϩ /H ϩ exchange activity was assayed by measuring 5-(N-ethyl-Nisopropyl)amiloride-sensitive 22 Na ϩ uptake as described previously (29). Cultured cardiomyocytes in 24-well dishes were preincubated with BSS containing 30 mM NH 4 Cl for 30 min at 37°C and subsequently washed twice with Na ϩ -free BSS for 40 s. 22 Na ϩ uptake was then initiated by adding the Na ϩ -free BSS containing 1 mM (37 kBq) 22 NaCl, 1 mM ouabain, and either 0 or 0.1 mM 5-(N-ethyl-N-isopropyl)amiloride. After 40 s, cells were washed four times with an ice-cold phosphatebuffered saline to stop 22 Na ϩ uptake. Passive 22 Na ϩ uptake into cultured cardiomyocytes was measured for 30 min at 37°C in BSS containing 22 NaCl (740 kBq/ml) and 1 mM ouabain.
Na ϩ ,K ϩ -ATPase activity was measured by incubating cardiac sarcolemmal vesicles (100 g) for 20 min at 37°C in a 1-ml reaction medium containing 20 mM Hepes/Tris (pH 7.4), 100 mM NaCl, 10 mM KCl, 5 mM MgCl 2 , 3 mM Na 2 ATP, and 1 mM EGTA. The reaction was stopped by addition of 10% trichloroacetic acid, and P i liberated was determined (30). The difference between activities in the presence and absence of 0.2 mM ouabain was taken as Na ϩ ,K ϩ -ATPase activity.
Ca 2ϩ -ATPase activity was measured using cardiac sarcolemmal vesicles or SR vesicles that were prepared from dog ventricular muscle Inhibitor of Na ϩ /Ca 2ϩ Exchange (31). The ATPase reaction was performed at 37°C with 50 g of sarcolemmal vesicles for 20 min or with 10 g of SR vesicles for 1 min in 0.5 ml of standard medium (20 mM Hepes/Tris (pH 7.2), 100 mM KCl, 5 mM MgCl 2 , 0.1 mM CaCl 2 (or 1 mM EGTA), 1 mM ATP, 2 mM phosphoenolpyruvate, 0.2 mg/ml pyruvate kinase, and 2 M A23187), to which 5 g/ml calmodulin, 1 M thapsigargin, and 1 mM ouabain were added further when ATP hydrolysis by sarcolemmal vesicles was measured. After the reaction was terminated by adding 3 N HCl (0.1 ml) containing 2.5 mM 2,4-dinitrophenyl hydrazine, pyruvate produced was determined as described previously (32). The difference between activities in the presence and absence of CaCl 2 was taken as Ca 2ϩ -ATPase activity.
Measurement of Action Potential-Cardiac action potential was measured according to the standard method (33). Briefly, papillary muscle bundles were isolated from guinea pig right ventricle, which were mounted in a chambered organ bath and superfused with Tyrode's solution (137 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 0.5 mM NaH 2 PO 4 , 11.9 mM NaHCO 3 , and 5.5 mM glucose, gassed by 95% O 2 /5% CO 2 ) at 10 ml/min at 36°C. Muscle preparations were stimulated at a constant rate of 2 Hz through bipolar electrodes with square-wave pulses of 0.5 ms and an intensity 2 times above threshold. Transmembrane electrical activity was recorded with conventional glass microelectrodes with tip resistances of 5-15 M⍀. Transmembrane potentials were measured by a high input impedance preamplifier (MEZ-7200, Nihon Kohden), displayed on dual beam oscilloscope (VC-11, Nihon Kohden), and stored on videotape recorder. The maximum rate of rise of the action potential (V max ) was obtained by an electronic differentiator with linear differentiation.
Protein Determination-Protein was measured by the modified Lowry method (34) with BSA as a standard.
Statistical Analysis-Data are expressed as the means Ϯ S.E. Differences for multiple comparisons were analyzed by one-way analysis of variance followed by the Dunnett's test. Values of p Ͻ 0.05 were considered statistically significant.

Effect of No.7943 on Na ϩ
i -dependent Ca 2ϩ Influx-No.7943 dose-dependently inhibited Na ϩ i -dependent 45 Ca 2ϩ uptake into rat cardiomyocytes, rat aortic smooth muscle cells, and cardiac NCX1-transfected CCL39 cells ( Fig. 2A). The IC 50 values for individual cell types were 2.4 Ϯ 0.3, 2.0 Ϯ 0.1, and 1.6 Ϯ 0.2 M (n ϭ 3), respectively, the complete inhibition occurring at Ն30 M of this agent. Thus, the inhibitory potency of No.7943 was very similar among these cell types. Under identical conditions, 10 M 3Ј,4Ј-dichlorobenzamil inhibited Na ϩ idependent 45 Ca 2ϩ uptake into rat cardiomyocytes by 30 Ϯ 5% (n ϭ 3). At 30 M, however, 3Ј,4Ј-dichlorobenzamil exhibited cytotoxicity, causing cell rounding and detachment from dishes, whereas the same concentration of No.7943 did not exert such cytotoxicity. In contrast, XIP did not affect the Na ϩ i -dependent 45 Ca 2ϩ uptake into cardiomyocytes at concentrations up to 100 M. Of note, the inhibitory potency of No.7943 was identical whether 45 Ca 2ϩ uptake was measured for 15 s in cardiac NCX1-transfected CCL39 cells after preincubation with this agent for 5 min ( Fig. 2A) or without such preincubation (IC 50 ϭ 2.1 Ϯ 0.4 M (n ϭ 3)). No.7943 at up to 30 M did not affect 45 Ca 2ϩ uptake into nontransfected CCL39 cells (data not shown), which is consistent with the lack of detectable Na ϩ i -dependent 45 Ca 2ϩ uptake in these cells (23). We examined the effect of No.7943 on the rate of Na ϩ i -dependent 45 Ca 2ϩ uptake into NCX1-transfected cells measured as a function of Ca 2ϩ o concentration (Fig. 2B). inhibition by this agent completely disappeared after washing cells with fresh medium for 1 min (Fig. 3), indicating that the effect is fully reversible. The transient nature of [Ca 2ϩ ] i rise observed here might have arisen, at least in part, from a time-dependent decrease in Na ϩ /Ca 2ϩ exchange activity, which was caused by the exchange-induced reduction in [Na ϩ ] i and development of deeper negative membrane potential, as well as by rapid removal of cytoplasmic Ca 2ϩ by the sarcolemmal and SR Ca 2ϩ pumps. In cardiomyocytes, a similar IC 50 value (1. Effect of No.7943 on Na ϩ o -dependent Ca 2ϩ Extrusion-We examined the effect of No.7943 on 45 Ca 2ϩ efflux from 45 Ca 2ϩlabeled NCX1-transfected cells in a Ca 2ϩ -and Na ϩ -free medium or in a Ca 2ϩ -free medium containing 146 mM Na ϩ (Fig. 4) (24,35). Under these conditions, [Ca 2ϩ ] i increased to a relatively high level (700 -800 nM), and we observed spontaneous decline of [Ca 2ϩ ] i (30 Ϯ 3 nM/10 s, n ϭ 4) (Fig. 5). Addition of 50 mM Na ϩ o accelerated the [Ca 2ϩ ] i decline with a resultant initial rate of 157 Ϯ 7 nM/10 s (n ϭ 4), whereas addition of 50 mM choline chloride had no effect. No.7943 at 10 and 30 M decreased the Na ϩ o -dependent portion of [Ca 2ϩ ] i decline by 12 Ϯ 5 and 37 Ϯ 2% (n ϭ 4), respectively, although the same concentrations of this agent did not affect the background [Ca 2ϩ ] i decline (Fig. 5).
Inhibition of Na ϩ /Ca 2ϩ Exchange in Sarcolemmal Vesicles by No.7943-The inhibitory effect of No.7943 on Na ϩ /Ca 2ϩ exchange was studied using cardiac sarcolemmal vesicles. As in cells, No.7943 completely and dose-dependently inhibited the initial rate of Na ϩ i -dependent Ca 2ϩ uptake into Na ϩ -loaded sarcolemmal vesicles (IC 50 ϭ 5.4 Ϯ 0.3 M (n ϭ 3)) (Fig. 6A). Eadie-Hofstee plots of the rate of Na ϩ i -dependent Ca 2ϩ uptake versus extravesicular [Ca 2ϩ ] i (Fig. 6B) revealed that 5 M No.7943 decreased the V max to 11 Ϯ 0.2 nmol/mg/1.5 s (n ϭ 3) from the control value of 22 Ϯ 0.6 nmol/mg/1.5 s (n ϭ 3), whereas it did not affect the K m for Ca 2ϩ (35 Ϯ 4.3 M for control and 34 Ϯ 2.8 M for the presence of the agent). The results indicate that inhibition is noncompetitive with respect to Ca 2ϩ . In contrast, inhibition by XIP was incomplete (about 70%) with an IC 50 value of 1.0 Ϯ 0.1 M (n ϭ 3) (Fig. 6B). 3Ј,4Ј-Dichlorobenzamil, on the other hand, completely inhibited Na ϩ i -dependent Ca 2ϩ uptake with an IC 50 of 19 Ϯ 0.7 M (n ϭ 3). Absence of the effect on the K m for extravesicular Ca 2ϩ was also seen with 1 M XIP (Fig. 6B)  ] i increase in Na ؉loaded aortic smooth muscle cells by No.7943. Fura-2-loaded smooth muscle cells, which had been preincubated with BSS containing 1 mM ouabain and 10 M verapamil for 1 h at 37°C, were exposed to a Na ϩ -free medium (NaCl replaced with equimolar choline chloride) for 2 min and then to the original BSS containing Na ϩ , ouabain, and verapamil for another 2 min. [Ca 2ϩ ] i was monitored throughout this procedure, during which [Ca 2ϩ ] i transiently increased and then returned to a basal level. This procedure was repeated four times to examine the effects of 1, 3, and 10 M No.7943 on [Ca 2ϩ ] i as shown in the figure.
No.7943 was cumulatively added 2 min before each exposure to the Na ϩ -free medium. In the fifth trace, No.7943 was removed by washing cells for 1 min with BSS not containing this agent and then exposed to the Na ϩ -free medium.  Table I, 10 M No.7943, which inhibited Na ϩ i -dependent 45 Ca 2ϩ uptake by about 90% (Fig. 2A), did not significantly influence Na ϩ /H ϩ exchange, DHP-sensitive 45 Ca 2ϩ uptake, passive 22 Na ϩ uptake, sarcolemmal and SR Ca 2ϩ -ATPases, and Na ϩ ,K ϩ -ATPase. On the other hand, 30 M No.7943 inhibited only the DHP-sensitive 45 Ca 2ϩ uptake by 35 Ϯ 2% (n ϭ 3, p Ͻ 0.05). We also tested the effect of No.7943 on the action potential parameters of guinea pig papillary muscle. Treatment of muscle preparations with up to 10 M No.7943 for 30 min did not significantly affect the resting membrane potential (Ϫ88 Ϯ 1.6 mV in control versus Ϫ87 Ϯ 3.3 mV (n ϭ 6) with No.7943), action potential amplitude (107 Ϯ 2.2 mV in control versus 106 Ϯ 2.6 mV (n ϭ 6) with No.7943), the maximum rate of rise of action potential (V max ) (174 Ϯ 8.5 V/s in control versus 148 Ϯ 3.8 V/s (n ϭ 6) with No.7943), or the action potential duration at 90% repolarization (144 Ϯ 8.2 ms in control versus 156 Ϯ 9.1 ms (n ϭ 6) with No.7943). However, 30 M No.7943 decreased the V max by 28 Ϯ 4% (n ϭ 6, p Ͻ 0.05). Since V max reflects activity of cardiac voltage-gated Na ϩ channels (33), No.7943 at Ն30 M appears to inhibit the Na ϩ channels. All these results support the notion that the inhibition by low concentrations (0.3-10 M) of No.7943 is selective.

Effects of No.7943 on [Ca 2ϩ ] i in Cardiomyocytes under
Normal and "Ca 2ϩ Paradox" Conditions-In untreated fura-2loaded neonatal cardiomyocytes, small transient increases in [Ca 2ϩ ] i were induced by spontaneous beating (Fig. 7A). Diastolic [Ca 2ϩ ] i and the rate of spontaneous beating in these cells were 128 Ϯ 11 nM and 15 Ϯ 5 beats/min (n ϭ 3), respectively. No.7943 at up to 10 M did not significantly change either diastolic [Ca 2ϩ ] i or the rate of spontaneous beating. However, 30 M No.7943 increased diastolic [Ca 2ϩ ] i by 31 Ϯ 4% (n ϭ 4) and caused cessation of spontaneous beating (Fig. 7A). Under identical conditions, application of 10 M verapamil or 1 M nicardipine induced cessation of spontaneous beating without affecting diastolic [Ca 2ϩ ] i (data not shown). Of note, after the treatment with verapamil, No.7943 (30 M) failed to induce an increase in diastolic [Ca 2ϩ ] i , suggesting that the [Ca 2ϩ ] i increase may be due to both the continued Ca 2ϩ influx via verapamil-sensitive Ca 2ϩ channels and inhibition of Ca 2ϩ extrusion via Na ϩ /Ca 2ϩ exchanger.
Ca 2ϩ o repletion after a period of Ca 2ϩ o depletion is known to cause Ca 2ϩ overloading of cardiomyocytes and finally cell death, a process called the Ca 2ϩ paradox (11). Ca 2ϩ overloading in the Ca 2ϩ paradox is considered to be due to Ca 2ϩ influx via the reverse mode of Na ϩ /Ca 2ϩ exchange (11). We tested the effect of No.7943 on this experimental model. When cardiomyocytes were treated with a Ca 2ϩ -and Mg 2ϩ -free medium for 10 min and then placed in a normal medium containing 1 mM Ca 2ϩ , [Ca 2ϩ ] i increased rapidly and markedly in the absence of No.7943, reaching a level of Ͼ3 M (Fig. 7B). In contrast, this increase in [Ca 2ϩ ] i was remarkably suppressed by 72 Ϯ 15 and 92 Ϯ 7% (n ϭ 3) in the presence of 3 and 10 M No.7943, respectively (Fig. 7B). Verapamil (10 M), however, did not prevent this Ca 2ϩ overloading (data not shown).   (23). No.7943 inhibited the Na ϩ i -dependent 45 Ca 2ϩ uptake into NCX1-transfected CCL39 cells ( Fig. 2A) but not 45 Ca 2ϩ uptake into nontransfected CCL39 cells (see "Results"). (iii) No.7943 potently inhibited both Na ϩ i -dependent Ca 2ϩ uptake into and Na ϩ o -dependent Ca 2ϩ efflux from sarcolemmal vesicles (IC 50 ; 5.4 and 11 to 13 M, respectively) (Fig. 6, A and  C). In intact cells, therefore, No.7943 exerts a much greater inhibitory effect on the Na ϩ /Ca 2ϩ exchanger operating in the reverse mode as compared with the exchanger operating in the forward mode. Furthermore, there is no difference in the inhibitory potency of No.7943 in different exchanger isoforms from cardiac and smooth muscle cells.
No.7943 is an amphiphilic molecule with an isothiourea group whose pK a is about 10. Thus this agent is protonated and cationic in most conditions (see Fig. 1). This positive charge on the isothiourea moiety seems to be essential for inhibitory activity, as its deletion renders this agent much less active. 3 No.7943 at up to 100 M is soluble in aqueous buffers. Interestingly, inhibition by No.7943 was easily abolished by washing cells with fresh medium for 1 min (Fig. 3), indicating that the effect is completely reversible and that removal of the agent is relatively rapid. Furthermore, the inhibitory potency of No.7943 was almost identical with or without prior preincubation (see "Results"). We therefore conclude that the agent primarily acts from the extracellular side of the plasma membrane in intact cells under the conditions used in this study. It should be noted, however, that No.7943 potently inhibited Na ϩ idependent Ca 2ϩ uptake into cardiac sarcolemmal vesicles, when the latter was exposed to the agent for only 1.5 s (Fig.  6A). The agent thus appears able to inhibit exchange activity also from the cytoplasmic side of the membrane, because a majority of sarcolemmal vesicles used in this study seem to have had inside-out orientation as inferred from the extent of inhibition by XIP. XIP at 10 M, which presumably is membrane-impermeable (17), inhibited Na ϩ i -dependent Ca 2ϩ uptake into sarcolemmal vesicles by 70% (Fig. 6A).
We examined the kinetics of inhibition by No.7943 of the reverse (Ca 2ϩ influx) mode of Na ϩ /Ca 2ϩ exchange with respect to either Ca 2ϩ o or Na ϩ o concentration using intact cells (Fig. 2 (Fig. 6B). It also did not influence [Na ϩ ] o dependence of Na ϩ o -induced 45 Ca 2ϩ efflux from vesicles that had been preloaded with 45 Ca 2ϩ by Na ϩ i -dependent Ca 2ϩ uptake (Fig. 6C). These vesicular exchange reactions (Na ϩ i -dependent 45 Ca 2ϩ uptake and Na ϩ o -dependent 45 Ca 2ϩ efflux) are mostly due to inside-out vesicles (17) and presumably equivalent to the forward and reverse modes of the exchange in intact cells. From all these results, we conclude that No.7943 does not affect the interaction of transport sites on the exchanger with Ca 2ϩ or Na ϩ on either side of the plasma membrane and thus probably acts at site(s) distinct from these sites.
It is striking that in intact cells the potency of No.7943 as a blocker of the reverse mode of Na ϩ /Ca 2ϩ exchange is 15-25-fold greater compared with that for the forward mode. In contrast, a minimum difference was observed for the effect of No.7943 on the corresponding reactions in sarcolemmal vesicles (Fig. 6, A and C, and see above). Thus, the mode-specific inhibition by No.7943 was observed only in intact cells. Such a difference in the inhibitory pattern may be consistent with our view that the agent acts on the exchanger primarily from the extracellular side in intact cells, whereas it acts mainly from the cytoplasmic side in sarcolemmal vesicles. At present, however, we have no information about the mechanism by which this agent causes such a mode-specific inhibition of Na ϩ /Ca 2ϩ exchange in intact cells.
No.7943 at 0.3-10 M, while blocking the Na ϩ /Ca 2ϩ exchanger-mediated Ca 2ϩ influx into cells, did not significantly affect activities of other ion transporters such as Na ϩ /H ϩ exchanger, DHP-sensitive Ca 2ϩ channels, sarcolemmal and SR Ca 2ϩ -ATPases, and Na ϩ ,K ϩ -ATPase, as well as passive Na ϩ permeability (Table I and Fig. 4). In addition, the same concentration range of No.7943 did not significantly alter the action potential parameters such as resting membrane potential, action potential amplitude, the maximum rate of rise of action potential (V max ), and action potential duration at 90% repolarization (see "Results"). However, No.7943 at a high concentration (30 M) reduced the activities of voltage-dependent Na ϩ channels (measured as V max ) and DHP-sensitive Ca 2ϩ channels, as well as the forward mode of the Na ϩ /Ca 2ϩ exchange. In cultured 2 T. Iwamoto, S. Wakabayashi, and M. Shigekawa, unpublished observations. 3 T. Iwamoto and M. Shigekawa, unpublished results.

FIG. 7. Effects of No.7943 on diastolic [Ca 2؉ ] i (A) and on Ca 2؉ overloading evoked by the Ca 2؉ paradox (B) in cardiomyocytes.
Fura-2-loaded cardiomyocytes were equilibrated with buffer A containing 1 mM CaCl 2 and 0.1% BSA for 20 min at 37°C. A, No.7943 was then cumulatively added at 100-s intervals as indicated. B, cardiomyocytes were exposed to a Ca 2ϩ -, Mg 2ϩ -free buffer A containing 0.2 mM EGTA for 10 min and then to buffer A containing 1 mM Ca 2ϩ and 0 (control), 3, or 10 (Fig. 7A). Thus, inhibition of Ca 2ϩ influx via the Na ϩ / Ca 2ϩ exchanger by low concentrations of No.7943 has virtually no effect on Ca 2ϩ mobilization and spontaneous beating in cultured cardiomyocytes. On the other hand, inhibition of Ca 2ϩ extrusion via the Na ϩ /Ca 2ϩ exchanger by a high concentration of this agent causes an increase in resting [Ca 2ϩ ] i , probably due to the continued influx of Ca 2ϩ via verapamil-sensitive Ca 2ϩ channels. All these results indicate that No.7943 at relatively low concentrations is a selective inhibitor of the Na ϩ /Ca 2ϩ exchanger that only minimally affects cell ion metabolism. In this sense, No.7943 clearly is much superior to 3Ј,4Ј-dichlorobenzamil whose specificity is low (see the Introduction). We explored the therapeutic potential of No.7943 by using the Ca 2ϩ paradox model (Fig. 7B). The Ca 2ϩ paradox has been studied as an experimental model for Ca 2ϩ overloading in cardiomyocytes during the ischemia-associated reperfusion. The Na ϩ /Ca 2ϩ exchanger operating in the Ca 2ϩ influx mode has been implicated in this mechanism. The same mode of the Na ϩ /Ca 2ϩ exchange is also considered to be responsible for Ca 2ϩ overloading during reoxygenation after cardiac hypoxia (11,12). We found that low concentrations of No.7943 were very effective in preventing Ca 2ϩ influx into cardiomyocytes and the resultant structural change under Ca 2ϩ paradox conditions ( Fig. 7B and see "Results"). Importantly, this agent at the same low concentrations (up to 10 M) also effectively blocks mechanical dysfunction of isolated perfused rat hearts that is caused by the ischemia/reperfusion or by hypoxia/reoxygenation insult, whereas it had no effect on mechanical function of normal rat hearts 4 (36). Thus, No.7943 could have a therapeutic potential as a selective blocker of excessive Ca 2ϩ influx mediated via the Na ϩ /Ca 2ϩ exchanger under pathological conditions, which include cardiac ischemia/reperfusion, hypoxia/reoxygenation, and possibly some forms of essential hypertension.