Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Endothelin is a 21-amino-acid polypeptide that was originally isolated from the culture medium of porcine aortic endothelial cells (Yanagisawa et al., 1988). Three structurally and pharmacologically distinct isoforms of endothelin have been identified: ET-1, ET-2, ET-3 (Haynes and Webb, 1993). Although initially identified as a potent vasoconstrictor peptide, ET-1 mediates a wide variety of pharmacological activities in various tissues (Rubanyi and Polokoff, 1994). Many reports have described the positive inotropic, positive chronotropic and hypertrophic effects of ET-1 on the myocardium (Ishikawa et al., 1988a,b; Moravec et al., 1989; Takanashi and Endoh, 1991). The cellular basis for the actions of ET-1 is highly complex; however, alterations in Ca⁺⁺ homeostasis appear to be central to the cardiac actions of this isopeptide. Although it is not clear whether ET-1 plays a physiological role, the isopeptide is implicated in many cardiovascular pathophysiological disorders, including hypertension, atherosclerosis, cardiogenic shock and myocardial infarction (Cavero et al., 1990; Kohno et al., 1990; McMurray et al., 1992; Huggins et al., 1993). Elevated plasma levels of ET-1 after myocardial infarction are likely to accompany much higher local concentrations of the peptide in the myocardium.

Two distinct subtypes of endothelin receptor, ET_A and ET_B, have been pharmacologically distinguished by the different potencies of the endothelin isopeptides toward the receptors. ET-1 has a higher affinity for the ET_A receptor subtype than do ET-2 and ET-3, whereas the isopeptides have a similar affinity for the ET_B receptor subtype (Rubanyi and Polokoff, 1994). Recently, the development and subsequent use of receptor-selective compounds have made receptor classification more complex (Sudjawaro et al., 1994; Warner, 1994). An increasing number of atypical ET responses have been identified that point to populations of receptors characterized as partially atypical or subtypes of ET_A and ET_B (Bax and Saxena, 1994). ET_A receptors have been subdivided, on the basis of their differential selectivities to the ET_A receptor-selective antagonist BQ-123, into ET_A1 (BQ-123-sensitive) and ET_A2 (BQ-123-insensitive) receptors (Sudjawara et al., 1994). In addition, the ET_B receptor subtypes located on vascular endothelium (ET_B1) and those located on smooth muscle (ET_B2) differ in antagonist selectivity (Douglas et al., 1994).

Both ET_A and ET_B receptor subtypes have been identified, using radioligand binding assays, in cardiac tissues isolated from human (Molenaar et al., 1993), rabbit (Takanashi and Endoh, 1991), guinea pig (Ono et al., 1994) and rat (Koseki et al., 1989). Both receptor subtypes coexist in cardiomyocytes, as has been demonstrated using in situ hybridization techniques (Hori et al., 1992). Many of the intracellular actions of ET-1 in cardiac tissues are mediated through the ET_A receptor subtype. For example, ET-1 inhibits a protein kinase-A-dependent chloride current via the ET_A receptor subtype in guinea pig ventricular cardiomyocytes (James et al., 1994). Moreover, BQ-123 abolished the decrease, produced by ET-1, in the accumulation of cyclic AMP and I_Ca in atrial cells (Ono et al., 1994). It is speculated that stimulation of the ET_B receptor subtype is likely to produce electrophysiological effects very different from those described for the ET_A receptor subtype. ET-1 has been reported both to increase (Lauer et al., 1992; Bkaily et al., 1995; Tong et al., 1995) and to decrease (Tohse et al., 1990; Ono et al., 1994; Cheng et al., 1995) the I_Ca in cardiomyocytes, depending on the concentration and experimental conditions (Lauer et al., 1992; Kelso et al., 1996). The purpose of the current study was to establish, using a number of recently developed receptor-selective and receptor-non-selective antagonists, whether specific receptor subtypes are responsible for mediating the effects of ET-1 on I_Ca. An integral part of this investigation was to examine also the effects of ET-3, which has an equal affinity for the ET_A and ET_B receptor subtypes.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Isolation of cardiomyocytes. Ventricular cardiomyocytes were obtained from male New Zealand White rabbits (2.5-3 kg) after enzymatic dissociation using collagenase (Kelso et al., 1995a). All procedures were undertaken in accordance with Guidance on the Operations of the Animals (Scientific Procedures) Act 1986. Rabbits were anesthetized using sodium pentabarbitone (50 mg/kg i.v.) after heparinization (400 I.U./kg i.v.), and the chest was opened and the heart quickly removed and cannulated, through the ascending aorta, on a modified Langendorff perfusion apparatus. Blood was flushed from the coronary vasculature with a Ca⁺⁺-free modified Krebs Ringer buffer (KRB) containing 110 mM NaCl, 2.6 mM KCl, 25 mM NaHCO₃, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄ and 11 mM glucose (pH 7.4, 37°C), which had just previously been aerated with 95%O₂/5%CO₂. The perfusate was subsequently supplemented with 0.12% (w/v) collagenase and recirculated for approximately 15 min while maintained at 37°C and continuously aerated with 95%O₂/5%CO₂. After enzymatic digestion, the hearts were cut at the atrioventricular junction, sliced vertically toward the apex and chopped into cubes of 0.7 mm³ using a mechanical tissue chopper (McIlwain Chopper, Mickle Laboratory Engineering Co. Ltd., Surrey, U.K.). The minced tissue was placed in the collagenase-containing perfusate that had been supplemented with 0.2% (w/v) BSA, and the mixture was triturated using a 10-ml serological pipette for approximately 5 min. The loosened cells were filtered through a nylon mesh gauze of pore size 200 µm and washed twice. Ca⁺⁺ was restored by means of centrifugation at 25 × g twice, and the cells were resuspended in modified KRB solutions containing 250 µM and 500 µM CaCl₂, respectively. Finally, the cells were layered onto a solution of 4% (w/v) BSA containing 1 mM CaCl₂ and were left to settle by gravity at 37°C. After approximately 5 min the supernatant was aspirated and the resulting cell material resuspended at a density of 1 to 2 mg protein/ml in a storage medium (M199 with Earle's salts, containing 5 mM creatine, 5 mM taurine, 2 mM carnitine, 100 I.U./ml streptomycin, 100 µg/ml penicillin, pH 7.4) at 37°C. Suspensions of cardiomyocytes were more than 70% viable as estimated by their elongated rod-shaped morphology.

Recording techniques. An aliquot of cell suspension was placed in a transparent recording chamber and allowed to settle for 10 min before bathing with a modified Tyrode's solution containing 137 mM NaCl, 5.4 mM KCl, 3 mM CaCl₂, 1.2 mM MgCl₂, 5 mM HEPES and 10 mM glucose (pH 7.4). The I_Ca was recorded in voltage-clamp mode using an Axopatch 1D patch-clamp amplifier, and electrodes were filled with 110 mM K-gluconate, 20 mM KCl, 2 mM MgCl₂, 10 mM HEPES, 11 mM EGTA, 1 mM CaCl₂, 0.1 mM Na₂GTP and 2.5 mM creatine phosphate (pH 7.2). Patch electrodes were fabricated from thin-walled borosilicate capillaries with a filament (1.5 mm in outside diameter, Clarke Electrochemical) by means of a horizontal lazer puller (Sutter Instruments, Model P2000) and had tip resistances of 1 to 3 M when filled with the electrode solution. Access to the cell interior was achieved by applying a brief pulse of negative pressure to the electrode after a gigaseal was formed. After stabilization, Ca⁺⁺ currents were elicited by stepping the membrane voltage for 200 ms from a holding potential of 40 mV to test potentials of 30 to +60 mV at 5-s intervals and 10-mV increments; recordings were made at 90-s and 180-s intervals before and after the application of each drug combination. Current "rundown" was not significant over the time course of the experimentsthat is, over a 10-min period. All currents were stored on computer for subsequent analysis using customized software. Drugs were applied locally to the cell using a gravity-fed microperfusion system at approximately 150 µl/min, which allowed the solution bathing the cell to be changed in approximately 2 s.

Data analysis. The I_Ca was measured, using standard methodology, as the difference between the peak of the inward current and the steady-state current level at the end of the voltage pulse (Varro et al., 1991). Current-voltage relationships were constructed, and peak I_Ca values were compared at +10 mV. I_Ca values were expressed as mean ± S.E. and data were analyzed, by comparing before and after drug applications, using analysis of variance followed by a Dunnett's multiple comparison test (n > 2) or Student's t test (n = 2); P values less than .05 were taken as indicating statistical significance.

Materials. ET-1 was purchased from Bachem Inc. (Torrance, CA). RES-701 and IRL-1038 were obtained from the American Peptide Co. (Sunnyvale, CA), and BQ-788 PD155080 and PD145065 were gifts from Parke-Davis Pharmaceutical Co., Ann Arbor, MI. All antagonists were dissolved in DMSO and stored in aliquots of 10⁴ M at 20°C; the final concentration of DMSO was <0.01%. ET-1 was dissolved in dilute acetic acid (0.01%) and stored in aliquots of 10⁵ M at 20°C. Collagenase (type I) was purchased from Serva Feinbiochemica (Heidelberg, Germany). Medium 199 was obtained from Gibco Ltd. (Paisley, U.K.). All other chemicals were of analytical grade (U.K.); twice-distilled water that had been deionized through a Millipore-Q system (Millipore, Harrow) was used in all experiments.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effects of ET-1 and ET-3 on the I_Ca. Figure 1A shows original current traces at test potentials between 30 mV and +50 mV, in the absence and presence of ET-1 (1 and 10 nM). At the end of each experiment, the current was completely abolished using nifedipine (5 µM), which substantiates that the current measured was indeed that of the I_Ca (results not shown). Contamination of the I_Ca was minimized by inducing voltage pulses from 40 mV, which inactivates the Na⁺ current. The influence of K⁺ currents was minimized by subtracting the current at the end of the test pulse from the peak inward current (Varro et al., 1991). ET-1 (10 nM) decreased (P < .05) the I_Ca to 1.86 ± 0.18 nA from a control value of 2.90 ± 0.16 nA (n = 6); this inhibitory effect was maximum after 90 s, remained stable at 180 s and was partially reversed to 2.46 ± 0.30 nA (fig. 1, B and C) after a 180-s washout of the peptide. ET-1, at a concentration of 1 nM, did not significantly alter the I_Ca from control. The corresponding current-voltage relationships, shown in figure 1B, revealed that the effects of ET-1 were not associated with a shift along the voltage axis but decreased the I_Ca at all potentials. ET-1 had no effect on the background holding current.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Effects of ET-1 on ICa in ventricular cardiomyocytes isolated from rabbit myocardium (N = 5). a) Representative tracings recorded from a single cell showing the ICa before and 180 s after the application of ET-1 (10 nM), from a holding potential of 40 mV to test potentials between 30 mV and +50 mV. b) Current-voltage relationships constructed as the difference between the peak of the inward current and the steady-state current level at the end of the voltage pulse (n = 6). Recordings were made after a 180-s exposure to each concentration of the drug, and washout was assessed after perfusion in control solution, in the absence of ET-1, for a further 180 s. c) The concentration-dependent effects of ET-1 on peak ICa, at +10 mV, were illustrated as percent change relative to control. Vertical lines are S.E.M. *P < .005 represents a difference from control peak ICa. P < .01 represents a difference compared with peak ICa in the presence of ET-1 (10 nM). N represents number of animals; n represents number of cells.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effects of ET-3 on the ICa (N = 6). Because the effect of ET-3 on ICa was slow, it was necessary to examine the concentration effects of ET-3 in separate experiments. a) Current-voltage relationships constructed as the difference between the peak of the inward current and the steady-state current level at the end of the voltage pulse (n = 9). Recordings were plotted after a 270-s exposure to ET-3 (1 nM), and washout was assessed after perfusion in control solution, in the absence of ET-3, after a further 270 s. b) The time course of changes in peak ICa amplitude at +10 mV in response to ET-3 (1 nM). An initial recording was taken at time 0 in the absence of ET-3; three further recordings were taken in the presence of ET-3 at 90-s intervals, followed by three recordings after removal of the drug and perfusion in control solution (n = 9). Control readings were also obtained in the absence of drug (n = 6). c) Current-voltage relationships in the absence and presence of ET-3 (10 nM) (n = 12). d) The concentration-dependent effects of ET-3 on peak ICa were illustrated as percent change relative to control. Vertical lines are S.E.M. *P < .01 represents a difference from control peak ICa. P < .005 represents a difference compared with peak ICa in the presence of ET-3. N represents number of animals; n represents number of cells.

Influence of the ET_A receptor subtype on the ET-1-induced decrease in I_Ca. PD155080, at concentrations of 1 and 10 µM, had no effect on the I_Ca (n = 6) (peak I_Ca values of 2.72 ± 0.26 nA and 2.76 ± 0.28 nA, respectively, compared with a control value of 2.90 ± 0.33 nA) (graph not shown). In order to minimize current rundown, the effects of ET-1 in combination with PD155080 were examined in a separate group of cells from those in which the effects of the antagonist alone were examined, or from those where the effects of ET-1 alone were examined. ET-1 (10 nM) in the presence of PD155080 (1 and 10 µM) did not alter the peak I_Ca amplitude, as observed from the original current trace in figure 3A, or over the voltage range of 30 to +60 mV (fig. 3B). The peak I_Ca amplitudes recorded in the presence of ET-1 in combination with PD155080 (1 and 10 µM) were 2.53 ± 0.13 nA and 2.58 ± 0.14 nA, respectively, which was not different from the control value of 2.65 ± 0.15 nA (n = 6). Washout of ET-1 in combination with PD155080 resulted in a small but significant increase in the peak I_Ca amplitude (fig. 3C) to 2.88 ± 0.13 nA. However ET-1 in combination with PD155080 did not affect the voltage dependence of activation on I_Ca.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Influence of the ETA receptor-selective antagonist PD155080 on the ET-1-induced decrease of ICa (N = 5) a) Representative tracings showing the inhibitory effect of the ET-1 (10 nM)-induced decrease in peak ICa at +10 mV, by PD155080 (1 µM), in a single cell. b) Current-voltage relationships in the absence and presence of ET-1 (10 nM) in combination with PD155080 (1-10 µM). Recordings were obtained at 180-s intervals before and after exposure to ET-1 (10 nM) in combination with PD155080 (1 and 10 µM) and after washout of drug in control solution. c) Comparison of percent changes, relative to the control, in peak ICa by ET-1, in the absence and presence of PD155080. The blank bar is included for comparative purposes (fig. 1), whereas the shaded bars represent the change in peak ICa values from another group of cells (n = 6). Vertical lines are S.E.M. *P < .005 compared with peak ICa in the absence of drugs. PD155 represents PD155080.

Influence of the ET_B receptor subtype on the ET-1-induced decrease in I_Ca. At concentrations of 1 and 10 µM, neither RES-701 (2.50 ± 0.23 nA and 2.53 ± 0.29 nA, respectively, compared with a control value of 2.64 ± 0.24 nA) nor BQ-788 (2.20 ± 0.23 nA and 2.28 ± 0.27 nA, respectively, compared with a control value of 2.35 ± 0.22 nA) alone had any effect on the I_Ca (n = 6-8) (graphs not shown). ET-1 (10 nM) in combination with RES-701 (1 and 10 µM) decreased (P < .05) the peak I_Ca amplitude (n = 10) to 1.63 ± 0.13 nA and 1.59 ± 0.23 nA, respectively, from a control value of 2.65 ± 0.24 nA. This effect was partially reversed upon washout of the peptides to a peak I_Ca amplitude of 2.26 ± 0.24 nA (fig. 4A). However, the decrease in current amplitude seen at all potentials was not associated with a shift along the voltage axis (fig. 4B). The magnitude of the decrease in peak I_Ca amplitude produced by ET-1 in combination with RES-701 (34 ± 4% to 36 ± 4%) was similar to the decrease produced by ET-1 alone (34 ± 4%; fig. 4C). Moreover, ET-1 (10 nM) in combination with IRL-1038 (1 µM) decreased (P < .05) the peak I_Ca amplitude (n = 4) to 1.77 ± 0.46 nA from a control value of 2.37 ± 0.50 nA (graphs not shown).

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Influence of the ETB receptor-selective antagonist RES-701 on the ET-1-induced decrease in ICa (N = 5). a) Representative tracings of peak ICa, at +10 mV, obtained in a single cell before and after the application of ET-1 (10 nM) in combination with RES-701 (1 µM) and after washout of the drugs. b) Current-voltage relationships in the absence and presence of ET-1 (10 nM) in combination with RES-701 (1 µM). Recordings were obtained at 180-s intervals. c) Comparison of percent changes, relative to the control, in peak ICa, for ET-1 in the absence and presence of RES-701 (1 and 10 µM). The blank bar is included for comparative purposes (fig. 1), whereas the shaded bars represent the change in peak ICa values from another group of cells (n = 10). Vertical lines are S.E.M. *P < .005 compared with peak ICa in the absence of drugs. RES represents RES-701.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Influence of the ETB receptor-selective antagonist BQ-788 on the ET-1-induced decrease in ICa (N = 6) a) Representative tracings of peak ICa, at +10 mV, obtained in a single cell before and after the application of ET-1 (10 nM) in combination with BQ-788 at a concentration of 1 µM and after washout of the drugs. b) Current-voltage relationships in the absence and presence of ET-1 (10 nM) in combination with BQ-788 (1 and 10 µM). Recordings were obtained at 180-s intervals. c) Comparison of percent changes, relative to the control, in peak ICa, for ET-1 (10 nM) in the absence and presence of BQ-788 (1 and 10 µM). The blank bar is included for comparative purposes (fig. 1), whereas the shaded bars represent the change in peak ICa values from another group of cells (n = 8). Vertical lines are S.E.M. *P < .05 compared with peak ICa in the absence of drugs. P < .005 compared with peak ICa in the presence of ET-1.

ET_A/B receptor-non-selective antagonist in the presence of ET-1. ET-1 (10 nM), in combination with the ET_A/B receptor-non-selective antagonist PD145065 (1 and 10 µM), had no effect on the I_Ca (n = 8). Values of peak current amplitude in the presence of ET-1, in combination with PD145065, were 1.98 ± 0.26 nA and 2.05 ± 0.30 nA at concentrations of 1 and 10 µM, respectively (fig. 6A) and were not changed from a control value of 2.11 ± 0.29 nA. Hence, PD145065 completely inhibited the effect of ET-1 alone on the I_Ca (fig. 6B).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   The inhibitory effect of the ETA/B receptor-non-selective compound PD145065 on the negative ICa response of ET-1 (N = 5) a) Current-voltage relationships in the absence and presence of PD145065 (1 and 10 µM) in combination with ET-1 (10 nM). Recordings were obtained at 180-s intervals. b) Comparison of percent changes, relative to the control, in peak ICa, for ET-1 (10 nM) in the absence and presence of PD145065 (1 and 10 µM). The blank bar is included for comparative purposes (fig. 1), whereas the shaded bars represent the change in peak ICa values from another group of cells (n = 8). Vertical lines are S.E.M. *P < .005 compared with peak ICa in the absence of drugs. PD145 represents PD145080.

Influence of the ET_B receptor subtype on the ET-3-induced increase in I_Ca. ET-3 (1 nM) in the presence of either BQ-788 or IRL-1038 (1 µM) did not alter the I_Ca from control values of 2.07 ± 0.11 nA and 2.11 ± 0.15 nA, respectively (fig. 7). Both ET_B receptor-selective antagonists abolished the increase (26 ± 7%) produced by ET-3 (1 nM) (fig. 2D). The combination of ET-1 and ET-3, at equimolar concentrations of 10 nM, did not alter the I_Ca from a control value of 2.20 ± 0.18 nA; that is, both the decrease produced by ET-1 (fig. 1) and the increase produced by ET-3 (fig. 2) were completely abolished.

View larger version (27K): [in this window] [in a new window]   Fig. 7.   Effects of ET-1 (10 nM), BQ-788 (1 µM) and IRL-1038 (1 µM) on the increase in peak ICa amplitude produced by ET-3 (10 nM). Peak ICa values at +10 mV were recorded before and after a 270-s exposure of the cells to ET-3, in combination with ET-1 (n = 10), BQ-788 (n = 5) or IRL-1038 (n = 8). Vertical lines are S.E.M. *P < .005 is the difference from control peak ICa. IRL-10 represents IRL-1038.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Both ET-1 and ET-3 have potent positive inotropic effects on cardiac tissues (Rubanyi and Polokoff, 1994). Although Ishikawa et al. (1988b) originally speculated that the positive inotropic effect of ET-1 could be attributed to an increase in the I_Ca, subsequent studies have yielded conflicting findings. ET-1 has been reported to increase (Lauer et al., 1992; Tong et al., 1995; Bkaily et al., 1995), to decrease (Tohse et al., 1990; Ono et al., 1994; Cheng et al., 1995) and even to have no effect on the I_Ca (Furukawa et al., 1992; Habuchi et al., 1992) in cardiac tissues. The decrease in I_Ca produced by ET-1 could be reversed by altering the experimental conditions to maintain elevated intracellular GTP (Lauer et al., 1992). Other investigators, however, found that ET-1 decreased the I_Ca even when GTP was added to the dialyzing pipette solution (Cheng et al., 1995). Recently, using the perforated whole-cell patch-clamp technique, we demonstrated that ET-1 increases the I_Ca at a concentration of 1 nM (Kelso et al., 1996). This effect was reversed, in the same cells, at greater than nanomolar concentrations, producing a decrease in current amplitude at concentrations of 10 to 100 nM. This finding prompted us to investigate whether multiple endothelin receptor subtypes were responsible for the effects of ET isopeptides on the I_Ca.

Using the ruptured whole-cell patch-clamp technique, we found that ET-1, at a concentration of 10 nM, decreased the I_Ca by 34 ± 4% from control values (fig. 1), which is in contrast to an increase of similar magnitude (32 ± 5%) produced by ET-3 (fig. 2). Because ET-3 has been referred to as an ET_B receptor-selective agonist (Ono et al., 1994) it appeared that the ET_B receptor subtype might be involved in the positive effect on the I_Ca (an increase in current amplitude), whereas the ET_A receptor subtype was likely to be involved in the negative effect of ET-1 on the I_Ca (the decrease in current amplitude).

PD155080 is a non-peptide receptor-selective antagonist that has high potency and selectivity similar to those of BQ-123 for the ET_A receptor subtype (Doherty et al., 1995). Unlike BQ-123, the novel antagonist PD155080 has no effect on the I_Ca per se (Kelso et al., 1995b). PD155080 completely inhibited the decrease in I_Ca produced by ET-1 (fig. 3), which confirms that the ET_A receptor subtype is involved in this response. Interestingly, washout of ET-1, in the presence of the antagonist, resulted in a consistent increase (~10%) in the I_Ca. Because the effects of ET-1 were only partially reversed, it is likely that the positive effect of the agonist, which is apparent at lower concentrations (Kelso et al., 1995b), is unmasked, possibly through its action at a different receptor subtype. No such effect was observed using the ET_A/B receptor-non-selective antagonist PD145065. ET-1, in combination with PD145065, did not alter the I_Ca from control values, nor was there any change in current amplitude after washout of the compounds (fig. 6). Although these differences are inconclusive, it is possible that the peptide antagonist PD145065, like ET-1, is less readily removed from the receptors than PD155080 and that, therefore, incomplete washout of PD145065 would prevent the increase in current amplitude observed using PD155080.

The ET_B receptor-selective antagonist RES-701 (Tanaka et al., 1994) did not prevent the decrease in I_Ca produced by ET-1 (fig. 4). However, the ET_B receptor-selective antagonist BQ-788 (Ishikawa et al., 1994) partially inhibited the ET-1-induced decrease in I_Ca in a concentration-dependent manner (fig. 5), which suggests that the agonist may be mediated, in part, through a BQ-788-sensitive ET_B receptor subtype insensitive to RES-701. Such a receptor subtype has been described in vascular tissues (Sudjawaro et al., 1994). The ET_B receptor subtype that mediates vasorelaxation in vascular endothelium was observed to be pharmacologically different from the subtype that mediates vasoconstriction in smooth muscle, and these subtypes have been tentatively termed ET_B1 and ET_B2, respectively (Douglas et al., 1994). Whereas BQ-788 appeared to inhibit both ET_B1 and ET_B2 with similar affinity, RES-701 inhibited only the ET_B1 receptor-mediated response (Warner, 1994; Sudjawaro et al., 1994). It is possible that the I_Ca response in ventricular cardiomyocytes is mediated by a ET_B2 receptor subtype similar to that found in smooth muscle. However, it is more likely that BQ-788 partially inhibited the decrease in I_Ca produced by ET-1 as a result of a lack of selectivity of the antagonist for the receptor subtypes. Although BQ-788 has been identified as a potent and selective ET_B receptor antagonist (Ishikawa et al., 1994; Fukuroda et al., 1994), a recent report (Peter and Davenport, 1996) indicates that BQ-788 has low affinity and little selectivity for the ET_B receptor subtype in human ventricular tissue.

In contrast to ET-1, however, the positive effect of ET-3 appears to be mediated through the ET_B receptor subtype. BQ-788 completely abolished the increase in I_Ca produced by ET-3, but this response was also completely inhibited by IRL-1038 (fig. 7). Although these opposing responses, produced by ET-1 and ET-3, are coupled to different receptor subtypes (ET_A and ET_B, respectively), it is clear that multiple mechanisms are involved in the actions of the isopeptides on the I_Ca in ventricular cardiomyocytes. It is interesting that the magnitude of the decrease and increase in I_Ca produced by ET-1 and ET-3, respectively, are similar. Cardiac tissues have a dense population of endothelin receptors (Takai et al., 1992); however, ventricular cardiomyocytes have a much greater proportion of ET_A receptor subtypes than ET_B receptor subtypes.

The mechanisms responsible for these effects are unclear. Both ET_A and ET_B receptor subtypes can couple to phospholipase C (Vogelsang et al., 1994), resulting in the hydrolysis of phosphatidylinositol to produce inositol-1,4,5-triphosphate and 1,2-diacyl glycerol. The latter may play a role in activating voltage-dependent Ca⁺⁺ channels via activation of protein kinase C. However, inositol-1,4,5-triphosphate stimulates release of Ca⁺⁺ from the intracellular stores, resulting in an increase in intracellular Ca⁺⁺. Elevated intracellular Ca⁺⁺ levels can inhibit ET-1-induced Ca⁺⁺ influx and therefore may explain the heterogeneity at different concentrations. ET-1 is also reported to inhibit the accumulation of cyclic AMP in adult cardiomyocytes (Jones, 1996), which would decrease the I_Ca as a consequence of reduced protein kinase A-dependent phosphorylation of sarcolemmal proteins. However, in other cell systems ET-1 can stimulate cyclic AMP generation (Sokolovsky et al., 1994). ET-1 stimulates the formation of cyclic AMP in Chinese hamster ovary cells that express the ET_A receptor subtype alone but decreases cyclic AMP levels in cells that express the ET_B receptor subtype alone (Aramori and Nakanishi, 1992). It is possible that in cardiac cells, the negative effect on accumulation of cyclic AMP is coupled to an ET_B2 receptor subtype, resulting in a decrease in the I_Ca. However, the positive effect on the I_Ca is unlikely to be mediated by such a mechanism, because ET-1 has been found to increase cyclic AMP levels only at picomolar concentrations (Sokolovsky et al., 1994).

In summary, the ET_A receptor subtype is coupled to a decrease in I_Ca, whereas the ET_B receptor subtype is responsible for an increase in I_Ca. A role for endothelin has been proposed in a variety of cardiovascular disorders, including myocardial infarction and cardiac ischemia (Rubanyi and Polokoff, 1994; Warner, 1994). The ability to alter calcium homeostasis will influence such pathogenic states. Indeed, suppression of the I_Ca because of elevated levels of ET-1 is likely to be important in protecting against ventricular arrhythmias and myocardial Ca⁺⁺ overload. However, evidence for multiple receptor subtypes that have opposing actions may have important implications for future therapeutic intervention.