Introduction

Top Abstract Introduction Methods Results Discussion References

We previously reported that ET receptors on rat adult and neonatal ventricular myocytes belong to a single population of ET_ARs and that ET_BRs are absent (Hilal-Dandan et al., 1994, 1997). Furthermore, these ET_A couple to multiple G proteins: ET induces PI hydrolysis via coupling to a PTX-insensitive G_q protein and inhibits adenylyl cyclase activity by coupling to a PTX-sensitive G_i protein (Hilal-Dandan et al., 1992, 1994). The positive inotropic response of myocytes to ET is sensitive to PTX (Kelly et al., 1990), whereas the hypertrophic responses to ET are mediated through both PTX-sensitive and -insensitive G proteins (Hilal-Dandan et al., 1997).

A curious feature of the interactions of ET with the heart is the prolonged duration of inotropic response; indeed, the response is not readily washed out (Moravec et al., 1989). Previous studies of the kinetic features of ET action have generally not assessed immediate effects of ET receptor occupancy but rather have assessed more distal and complex responses, such as contraction or, atrial natriuretic peptide secretion. We used purified adult and neonatal myocytes to characterize the relative reversibility of ET binding and ET-induced signaling and to test the hypothesis that proximal aspects of ET action, such as binding and activation of G protein-linked signaling pathways, are irreversible and contribute to the persistent effects of ET. Furthermore, we propose a model of the life cycle of the myocyte ET_AR in which the quasi-irreversible nature of binding is a prelude to refractoriness and thence to slow regeneration of responsive receptor.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. [¹²⁵I]ET-1 (2200 Ci/mmol) was obtained from DuPont-New England Nuclear Research Products (Boston, MA). The ET-1 that we used was the synthetic human-porcine ET-1. All chemicals and biochemicals were reagent grade and were obtained from Sigma Chemical Co. (St. Louis, MO).

Isolation of neonatal ventricular myocytes and cell culture. Myocytes from the ventricles of 1- to 2-day-old Sprague-Dawley rats were prepared according to a collagenase-pancreatin digestion procedure (Iwaki et al., 1990), purified on a discontinuous Percoll gradient and plated onto gelatin-coated dishes in Dulbecco's modified Eagle's medium/Medium 199 (4:1) supplemented with 10% horse serum, 5% fetal calf serum and penicillin/streptomycin (each at 100 U/ml). For biochemical studies, cells were plated at a density of 1 × 10⁶ cells/60-mm plate. After 24 hr of culture, the cells were washed and incubated overnight in serum-free medium. For experiments involving peptide hormones, the medium was supplemented with 10 µg/ml leupeptin and 1 mg/ml bovine serum albumin before the addition of agonists and antagonists.

Isolation of adult ventricular myocytes. Adult cardiac myocytes were isolated and purified from hearts of Sprague-Dawley rats (male, 300 g) according to a collagenase dissociation procedure (Hilal-Dandan and Khairallah, 1991).

ET binding studies. Neonatal myocytes were cultured in 35-mm six-well plates at a density of 10⁵ cells/well and incubated with [¹²⁵I]ET-1 (~13 pM) with or without increasing concentrations of unlabeled putative competitors at 37°C for 2 hr. Incubations were terminated by washing each well twice with 2 ml of cold (4°C) buffer consisting of 25 mM Tris-Cl, 10 mM MgCl₂ and 1 mM EDTA, pH 7.4. Cells were solubilized with 1 ml of 0.4 N NaOH; radioactivity was quantified by gamma counting. Specific binding (described in Results) was saturable and was a linear function of cellular protein.

Phosphoinositide hydrolysis. Activation of the G_q pathway was assessed by measuring IP production in response to ET. Myocytes were labeled overnight with myo-[³H]inositol (5 µCi/ml) and then washed twice; experiments were initiated by the addition of medium containing 10 mM LiCl, 10 µg/ml leupeptin, 1 mg/ml bovine serum albumin and agonists. Incubations were terminated by aspiration of the supernatant and the addition of 1 ml of 5% ice-cold TCA. The samples were extracted four times with six volumes of water-saturated ether and fractionated by anion exchange chromatography (Brown et al., 1985). Radioactivity in fractions corresponding to total IPs was quantified by liquid scintillation spectrometry. Data are expressed as cpm/plate of cells.

cAMP accumulation. Activation of the G_i pathway by ET was assessed on the basis of the effect of ET to reduce cAMP accumulation in response to isoproterenol. Neonatal myocyte cultures were incubated at 37°C in medium containing 0.5 mM isobutylmethylxanthine (a phosphodiesterase inhibitor) for 15 min before the addition of ET-1 or congeners and 1 µM isoproterenol. The experiments were terminated by removal of the medium and the addition of 1 ml of 5% ice-cold TCA. The TCA extracts were purified over Dowex AG50W × 4 columns (200-400-mesh). cAMP content was determined according to the method of Gilman (1970).

Protein determinations. Cells were solubilized in 0.4 N NaOH, and the protein content was estimated according to the method of Bradford (1976).

	Results

Top Abstract Introduction Methods Results Discussion References

[¹²⁵I]ET-1 binds specifically and saturably to intact neonatal ventricular myocytes. At 37°C, binding with 13 pM ligand reaches a steady state level by 2 hr and thereafter remains constant for 4 additional hr. Under the assay conditions described here, specific binding (i.e., total binding minus binding occurring in the presence of 100 nM ET-1) accounts for 80% to 85% of the total binding and is a linear function of the number of cells added to the binding reaction. On the basis of data from control experiments, we used 100,000 neonatal cells/dish in binding assays and 13 pM labeled ligand. Under these conditions, fractional binding of added [¹²⁵I]ET-1 did not exceed 10% to 15%. Virtually identical results were obtained with adult ventricular myocytes (Hilal-Dandan et al., 1994).

On neonatal myocytes, [¹²⁵I]ET-1 binds to a single population of receptors with an EC₅₀ value of ~0.073 ± 0.006 nM (mean ± S.E.M., n = 9). The ET_AR specific antagonist BQ123 (Ihara et al., 1992; Ohlstein et al., 1992) competes fully with ET-1 binding sites on the myocytes with a K_i value of ~6.8 ± 1.4 nM, whereas the ET_BR agonist sarafotoxin 6c (Williams et al., 1991) is ineffective (fig. 1A).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Interaction of [125I]ET-1 with myocytes. A, Order of potency of ET-1 and congeners. Myocytes were incubated with [125I]ET-1 (~13 pM) alone or in the presence of increasing concentrations of ETs, sarafotoxin 6c (S6c) or BQ123, as noted on graph. Data are mean of triplicate determinations from a representative experiment. B, Irreversibility. Cells were incubated with [125I]ET-1 (~13 pM). At 5, 20 and 60 min, excess unlabeled ET-1 (0.1 µM) was added. Specific binding of [125I]ET-1 was assessed periodically as described in Methods. The data are from a representative experiment that was repeated twice.

The association of [¹²⁵I]ET-1 with neonatal myocytes is specific but not readily reversible. The addition of 100 nM unlabeled ET-1 prevents further binding of [¹²⁵I]ET-1 but does not reduce the quantity of [¹²⁵I]ET-1 bound to myocytes (fig. 1B). Thus, no dissociation of bound [¹²⁵I]ET-1 is observed when a 7000-fold excess of unlabeled ligand is added (fig. 1B), even though simultaneous addition of labeled and unlabeled ligand produces the expected competition isotherm (fig. 1A). This apparent irreversibility develops rapidly, being observed by 5 min of exposure to the radiolabeled ligand (fig. 1B).

To determine whether the irreversibility of ET binding has functional correlates, we assessed activation of the G_q and G_i pathways. We assessed activation of the G_q pathway by measuring stimulation of IP production by ET in cells labeled with [³H]inositol. To assess activation of the G_i pathway, we measured the effect of ET to inhibit the beta adrenergic response (cAMP accumulation). As judged from the rank order of potency of agonists, both of these responses result from the activation of ET_ARs (Hilal-Dandan et al., 1994, 1997). Use of the ET_AR antagonist BQ123 permits assessment of reversibility of these functional responses. BQ123 completely and competitively abolishes the capacity of 10 nM ET to stimulate PI hydrolysis (fig. 2A) when the antagonist is added simultaneously with ET. Likewise, the simultaneous addition of BQ123 and ET competitively inhibits the effect of ET to reduce the beta adrenergic response (fig. 2B). However, if ET is permitted to interact with the cells briefly before the addition of the antagonist, neither G_q activation nor G_i activation is antagonized by the subsequent addition of BQ123 (fig. 3, A and B). In both instances, the simultaneous addition of antagonist prevents the ET response, whereas the addition of antagonist at 5 min after the addition of ET is ineffective. Thus, the functional consequences of receptor occupancy by the agonist parallel the binding of the ligand.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Competitive antagonism of the effects of ET-1 by BQ123. Myocytes were stimulated simultaneously with ET-1 (10 nM) and increasing concentrations of the ETAR antagonist BQ123, and both PI hydrolysis and cAMP responses were measured. The experiments were conducted as described in Methods. Data are from a representative experiments that was repeated twice. A, BQ123 inhibits ET-1-stimulated PI hydrolysis with a Ki value of 1.7 ± 0.3 nM. B, BQ123 reverses the effect of ET-1. To reduce cAMP accumulation in response to beta adrenergic receptor stimulation with a Ki value of 4.4 ± 1.2 nM. INE, isoproterenol.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Irreversibility of the functional effects of ET-1 on myocytes. After initiation of responses to ET-1 (1 nM), sufficient BQ123 (1 µM) was added to fully antagonize the effects of ET-1 on PI turnover and cAMP production. A, PI hydrolysis. BQ123 was added 10 min after the addition of ET-1, and sampling continued for 20 min. The experiment was otherwise conducted as described in Methods. Data are mean ± range of duplicate determinations repeated three times. B, cAMP accumulation. BQ123 was added 6 min after the addition of ET-1 and/or isoproterenol (INE), and sampling was continued for 9 min. The experiment was otherwise conducted as described in Methods. Data are mean ± range of duplicate determinations from a representative experiment repeated three times.

The apparent irreversibility of ET binding and effects could represent very tight binding or removal of the ligand/receptor complex to a compartment not accessible to dissociation and competition. We probed the basis of this irreversible interaction in two ways: acid stripping and conducting binding studies at low temperatures. Acid stripping is often effective in removing peptide and protein ligands that are tightly associated with the external surface of the plasma membrane (Haigler et al., 1979). We applied acid washing as a means of removing surface-accessible [¹²⁵I]ET-1 after equilibration. On both adult and neonatal ventricular myocytes, we find that a 10-min acid stripping (see Methods) removes approximately half of specifically bound [¹²⁵I]ET-1 (range of three experiments, 40-60%; 10-15% may be removed with an additional acid wash). At low temperatures, membrane fluidity increases, and transmembrane processes such as endocytosis may be inhibited (Haigler et al., 1979; Morrison et al., 1996). When [¹²⁵I]ET-1 binding is conducted at 4°C to minimize internalization of the bound ligand, acid stripping is still able to remove only half of the bound ligand, just as at 37°C. Specifically, we exposed neonatal cells to 13 pM iodinated ligand for 90 min at either 4°C or 37°C, removed the free ligand by washing and then performed acid stripping. In a representative experiment, total specific binding was 4781 ± 282 cpm at 37°C and 3310 ± 160 cpm at 4°C (mean ± S.E.M., n = 3). As a percentage of total bound cpm, [¹²⁵I]ET remaining after acid stripping was 46.5 ± 6.3% when binding had been performed at 4°C and 54.1 ± 5.0% when binding had been performed at 37°C. Thus, under conditions in which uptake of [¹²⁵I]ET-1 should be minimized, a large fraction of bound ligand is resistant to acid washing. These data suggest that the observed irreversibility does not necessarily reflect cellular uptake of the ligand.

Studies with isolated myocyte membranes and detergent-permeabilized cells support this notion. Using myocyte membranes, we reproduced the data of figure 1B (binding of radiolabeled ligand is not reversed when excess unlabeled ligand is subsequently added) and the acid stripping data detailed above. Thus, the irreversible interaction seems not to require the whole cell; such a result suggests that cellular uptake is not the basis for irreversible binding of [¹²⁵I]ET-1 to myocytes. Irreversibly bound ET is not readily released from the cell. For example, the addition of 0.1% Triton X-100 permeabilizes the sarcolemma of 100% of cells (as judged by uptake of Trypan blue) but does not solubilize or release any cell-associated [¹²⁵I]ET-1(fig. 4A). These data suggest that the interaction of ET-1 with its receptor is stable to detergent and is not loosely associated with the cell surface or with a readily solubilized intracellular fraction.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Irreversibility of [125I]ET-1 binding and function in the presence of reducing agents and detergent. A, Binding is not reversed by the addition of reducing agents or detergent. Myocytes were exposed to [125I]ET-1 at 37°C, and specific binding was determined as described in Methods. Then, 10 mM DTT, 15 mM -mercaptoethanol (-ME) or 0.1% Triton X-100 was added concurrently with the radioligand (with) or after the myocytes had incubated with the radioligand for 60 min (after), with the incubation then continuing for an additional 45 min. Data are from a representative experiment that was repeated three times. B, Ability of ET to inhibit beta adrenergic response is not reversed by DTT. Myocytes were stimulated with isoproterenol (1 µM) in the presence and absence of ET (1 nM). At 4 min after the addition of isoproterenol and ET, 10 mM DTT was added, and sampling was continued for 6 min as described in Methods. Data are mean ± range of duplicate determinations from a representative experiment.

We also characterized the effects of sulfhydryl reagents on the interaction of ET-1 with myocytes (fig. 4A). When DTT and mercaptoethanol are added concurrently with [¹²⁵I]ET-1, they substantially reduce binding, suggesting that the S---S bonds of the ligand or the S---S bonds at the cell surface are important to the peptide/cell interaction. However, the addition of these reagents after ET-1 binding has proceeded for 60 min does not reduce binding, indicating that the irreversibility cannot be ascribed simply to S---S bonds at the cell surface. Function of ET-1 is likewise not altered by subsequent addition of DTT (fig. 4B): the effect of ET-1 to lower the beta adrenergic response of adult ventricular myocytes is not removed by the addition of 10 mM DTT 4 min after the addition of ET. Thus, in this respect also, binding and function show parallel courses of irreversibility.

In preliminary experiments, we also observe this irreversible aspect of the interaction of ET in studies of the hypertrophic response. When myocytes are treated with ET for 24 hr, they undergo morphological changes that include a significant and visible increase in cell size. Unlike the hypertrophic response induced by alpha-1 adrenergic receptor stimulation, the hypertrophic response due to ET is not reversed by washing of the cells to remove the extracellular stimulus (data not shown).

	Discussion

Top Abstract Introduction Methods Results Discussion References

Our data indicate that the interaction of endothelin with both neonatal and adult rat ventricular myocytes is irreversible as assessed in binding studies. We find that the activation of G_i- and G_q-linked pathways by ET are, likewise, irreversible, as is the hypertrophic response of neonatal cells. This irreversibility develops rapidly and is apparent and complete after 5 min of exposure to the ligand. The lack of ready reversibility of both ligand binding and biochemical responses suggests that the observed ligand binding is not artifactual but represents functional agonist/ET_AR interactions.

The ventricular myocytes that we use offer several advantages for investigation of these sorts of issues: ET binding and the activation of G_i- and G_q-linked pathways are well correlated in these models (Hilal-Dandan et al., 1992, 1994), binding and response can be measured under similar experimental conditions in identical cell preparations that are relatively homogeneous, the responses are G protein-linked processes that are thought to be immediately distal to ET-occupied receptors in the response pathway and the receptors involved are a single class of ET_A sites (Hilal-Dandan et al., 1994, 1997). Thus, the likelihood that binding and responsiveness reflect the same ligand/receptor interaction is high in these cardiac myocyte preparations.

As a consequence of the irreversible nature of ET binding, constants derived from ET binding studies are in error. The mathematical analysis of binding data, concentration-dependence experiments and agonist-antagonist competition experiments assumes reversible mass action association and dissociation, conditions that are not fulfilled in ET binding studies. Thus, traditional analytical methods will provide constants that are approximate at best. Several researchers have given good descriptions of this difficulty in cardiac cells. As Leite et al. (1994) point out, the approach to equilibrium in ET binding studies may be slow and studies may be conducted at less-than-equilibrium conditions; this could have the practical effect of causing data from kinetic studies (k₊₁/k₁) and data from pseudoequilibrium to differ. Indeed, affinities calculated from rate constants are frequently in the picomolar range, whereas affinities derived from equilibrium binding are in the range of 0.1 to 3 nM (see Leite et al., 1994).

In general, questions of functional reversibility, refractoriness and receptor internalization and recycling have not been addressed in prior studies of ET/myocyte interactions. Our data show that both binding and responsiveness of ET are relatively irreversible. For example, the ET_AR antagonist BQ123 will compete for ligand binding and biochemical effects (activation of G_i- and G_q-linked pathways) if added simultaneously with ET; however, when added after ET, the antagonist prevents neither binding nor activation of the G_i and G_q pathways (figs. 2 and 3). Apparently, ET binds irreversibly to myocytes, and the irreversibility is reflected in the biochemical responses of the cells to ET. Leite et al. (1994) have made related observations in cultured rat atrial myocytes in which the apparent half-time for the dissociation of bound ligand is 25 hr. Waggoner et al. (1992) report that the half-time for dissociation of ET bound to membranes from several tissues including heart exceeds 30 hr at 37°C.

Refractoriness seems to occur concomitantly with or as an immediate consequence of irreversible binding. In adult and neonatal myocytes, the IP response to ET desensitizes only slightly over a period of min (Hilal-Dandan et al., 1994, 1997) but unquestionably desensitizes over somewhat longer periods in neonatal cells (McDonough et al., 1993). In atrial myocytes, the IP response and atrial natriuretic peptide production exhibit marked refractoriness after 10 to 20 min of exposure to ET; the desensitization to ET seems to be homologous, and recovery from the refractory state occurs with a half-time similar to that for dissociation of bound ligand (in the range of 20-25 hr) (Leite et al., 1994). Van Heugten et al. (1993) also investigated the phenomenon of refractoriness of the IP response in neonatal cardiac myocytes and concluded that ET induces an homologous refractoriness that is not due to substrate depletion or inactivation of the total pool of phospholipase C, since alpha adrenergic stimulation still yields a maximal response. Indeed, these authors concluded that the likely site of desensitization is "upstream of the GTP-binding proteins."

Whether agonist-occupied or "desensitized" ET_ARs are internalized is unclear. We find that at 4°C, at which internalization of the ligand should be minimized, the interaction of [¹²⁵I]ET with adult ventricular myocytes is still irreversible and binding experiments with membranes will reproduce results obtained in studies on intact cells. These data suggest that the basis of the irreversibility is not internalization of bound ET by the myocyte.

Consideration of the time courses of function and ligand binding suggests to us that the issues of irreversible binding, refractoriness and functional recovery of ET-treated cells, are related. Indeed, the irreversible interaction of ET with the ET_AR could be a prelude to refractoriness. Our observations and those of several other groups (Hilal-Dandan et al., 1994; Leite et al., 1994; Van Heugten et al., 1993) on the issues of binding and responsiveness suggest a kinetic scheme for the functional life cycle of an ET_AR in cardiac myocytes (fig. 5). In this model, ET is the ligand, ET_AR is the type A ET receptor, the forward rate constants (k₊₁, k₊₂, k₊₃) are large and the reverse rate constants (k₁, k₂) are small or essentially zero. The rate of development of refractory ligand/receptor complex is k₊₄, and the rate of restoration of functional receptor to the cell surface is k₆ (slow). We observed negligible effects of GTP on agonist binding to membranes rather than the expected reduction in agonist affinity that is characteristic of G protein-linked systems (Hilal-Dandan et al., 1994); thus, we propose that ET/ET_AR*, in which the ligand is already irreversibly bound, is the long-lived complex that interacts with G proteins to produce biochemical effects. The persistence of the responses also supports this interpretation. The antagonist BQ123, when added simultaneously with ET, competes for the initial interaction of ET for its receptor, but k₊₂ is so rapid and k₂ is so small that the addition of BQ123 subsequent to the addition of ET is without effect.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Proposed life cycle of the myocyte ETAR. ET interacts (k+1) with the ETAR to form a transitory complex that rapidly (k+2) forms an essentially irreversible complex (ET/ETAR*). The rates (k1, k2) of the reverse reactions are negligible. The relatively stable complex ET/ETAR* interacts with G proteins to form the active G-associated complex (ET/ETAR*/G/GTP). The rate at which the active complex dissociates to give G protein and ET/ETAR* may vary with the type of G involved. The refractory complex ET/ETAR** may arise by an aging of ET/ETAR* (k+4) or by the action of GRKs on the active complex (k5). Regeneration of free receptor (ETAR) occurs not by dissociation of the initial transitory and stable complexes but rather by slow dissociation of the refractory complex ET/ETAR** (k6, small, perhaps ~0.03/hr; see Leite et al., 1994), perhaps aided by the synthesis and insertion of new receptor or by receptor recycling. AC, adenylyl cyclase; PLC, phospholipase C.

In experiments lasting 72 hr, McDonough et al. (1993) and Leite et al. (1994) have defined a refractory state that develops as functions of duration of exposure and concentration of ET. These observations, plus the finding that removal of the free ET neither reverses the response nor restores the responsiveness of refractory cells very quickly (Leite et al., 1994; McDonough et al., 1993; present study), suggest that functional reversal occurs via a refractory or desensitized state, ET/ET_AR**, an inaccessible ligand/receptor complex that appears at a rate (k₊₄) that is less than k₊₁, k₊₂ and k₊₃ and that dissociates slowly to regenerate free ET_AR. GRKs (specifically, GRK2 and GRK5; see Ament et al., 1995) may be activated when ET-1 interacts with its receptor and may contribute to refractoriness and dissociation of the activated complex (depicted as k₅). It is also possible that recovery of responsiveness involves synthesis and/or insertion of new receptor.

Reported data on myocytes support such a model. For example, Leite et al. (1994) found that the effect of ET-1 to stimulate atrial natriuretic peptide secretion from rat atrial myocytes desensitizes relatively rapidly (t_1/2 ~20 min with 1 nM ET-1, corresponding to the reaction characterized by k₊₄), that binding of [¹²⁵I]ET reverses very slowly (t_1/2 ~25 hr) and that recovery of the secretory response parallels the recovery of [¹²⁵I]ET binding sites (t_1/2 ~20-24 hr, corresponding to the reaction characterized by k₆). The termination of the effects of ET is hypothesized to occur via desensitization followed by slow regeneration of surface receptors. Available studies have concentrated on the phosphoinositide response. It will be interesting to see whether the G_i-and G_q-linked responses show identical temporal patterns of desensitization and recovery. It seems plausible that the rates will differ and that, as a consequence, the integrated signaling in response to ET in a single cell will change over time. For example, if the half-life of the "active complex" is longer for G_i than for G_q and receptor regeneration is slow by comparison, then the response to ET will at first consist of contributions of both G_i and G_q but will be dominated subsequently by the G_i pathway. Such temporal coordination of several signals may be crucial to complex responses such as the growth response to ET.

Although data on myocytes generally support the model shown in figure 5, this model should not be considered exclusive but simply a convenient way of thinking about the data that provides an interpretive framework and provokes additional experiments. Furthermore, data from vascular smooth muscle present a different picture. Marsault et al. (1993) argued that receptor recycling (agonist-induced internalization of ligand/receptor complex followed by removal of ligand and externalization of free receptor) contributes to the sustained contractile response of rat aortic strips to ET. The contractile and binding data used by these authors were derived from different cellular preparations and thus may not be strictly comparable. In addition, contraction of smooth muscle may represent the response of multiple cell types or multiple ET receptor types, and the reported rate of externalization (35% in 40-60 min) may not account for the observed functional effects of BQ123. Warner et al. (1994) confirmed that BQ123 reverses ET contracture in rat aortic rings (subsequent to 3 nM ET, 10 µM BQ123 reduces contraction by 85% in 40 min). Together, these studies provide evidence that functional reversal of the effects of ET does occur slowly in rat aortic smooth muscle. In comparison, [¹²⁵I]ET-1 dissociation and reversal of response are not apparent in cardiac myocytes over similar time periods. Perhaps the rate constant for reversal in cardiac cells is smaller than that in smooth muscle.

On the issue of receptor internalization, cardiac and smooth muscle also appear to differ. We observe irreversible binding when internalization is minimized (4°C). In addition, irreversible binding of ET to myocytes does not require the intact cell; myocytes membranes produce virtually identical results. Thus, it seems possible that irreversible binding can occur in intact cardiac myocyte preparations without significant internalization.

The irreversible binding of ET and the consequent prolonged intracellular responses support the proposition that ET contributes to long-term physiological or pathophysiological regulation. Likewise, the irreversible nature of the effects of ET in myocytes suggests that ET is not a useful regulator of the heart on a beat-to-beat basis. In the context of the present report, we conclude that the quasi-irreversible nature of ET binding is appropriate in terms of the observed signal transduction. We are conducting experiments to test our hypothetical life cycle of the ET_AR (fig. 5) in adult and neonatal rat myocytes and to obtain estimates of the rates at which the G_i- and G_q-linked pathways desensitize and recover.

From a clinical viewpoint, the rate at which bound ET dissociates and whether the effects of ET are long lasting may dictate the time course of effectiveness of ET receptor antagonists. Based on comparison of our data on myocytes with data on vascular smooth muscle (Marsault et al., 1993; Warner et al., 1994), we predict that the behavior of ET/receptor complexes and of ET-stimulated signaling may vary from tissue to tissue. It seems likely that synthetic agonist and antagonist congeners of ET will show different binding properties in different species and among different organs of a species. Indeed, Nambi and Pullen (1995) have shown that truncated analogs of ET lacking disulfide bonds bind with varying degrees of reversibility in several tissue systems. Thus, the clinical pharmacokinetics of ET receptor-directed agents may prove to be a complex issue.