INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

AFTER THE ORIGINAL DISCOVERY of endothelin-1 (ET-1) in vascular endothelial cells (43), synthesis of this 21-amino acid peptide was found to occur in endocardial endothelial and myocardial cells as well (24, 39).

Predominant abluminal release and short half-life of ET-1 in blood restricts its activity primarily to paracrine and autocrine actions (17, 24). In mammalian tissues, two G protein-coupled receptor types, ET_A and ET_B, mediate these actions differentially (1, 36). ET_A receptor stimulation elicits vasoconstriction (22), mitogenesis (27), and increased inotropism (13). ET_B receptor activation promotes vasodilatation mediated by nitric oxide (NO) (4, 40) and prostacyclin (4, 15) release and has growth-inhibitory effects associated with apoptosis (28). These receptors also mediate pulmonary clearance of circulating ET-1 (8) and reuptake of ET-1 by endothelial cells (29).

Increased ET-1 synthesis has been reported in a number of cardiovascular disease states including pulmonary hypertension (38), acute myocardial infarction (25), and congestive heart failure (CHF) (42). In CHF, plasma and salivary levels of ET-1 are elevated and correlate directly with the New York Heart Association functional classes and inversely with prognosis (17). Increased plasma levels of ET-1 may, however, contribute to the maintenance of cardiac function in failing hearts (35).

Unlike its effects on myocardial contractility, the influence of ET-1 on diastolic properties of the myocardium is still poorly understood. Previous studies addressing this issue were carried out in the intact heart, either in situ or in vitro, where, given the potent effects of ET-1 on coronary and peripheral circulations, it was not possible to differentiate between effects of ET-1 on intrinsic myocardial diastolic properties from those resulting from load changes and coronary vasoconstriction (3, 6, 12, 14, 15, 26). In isolated papillary muscles, where this limitation is overcome, we recently gathered evidence that ET-1 might influence diastolic function by accelerating relaxation rate and decreasing resting tension (20). To further clarify this issue, we conducted the present study in rabbit papillary muscles and human atrial strips with the aim of characterizing the diastolic effects of ET-1 and some of their underlying mechanisms.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996) and to the principles outlined in the Declaration of Helsinki. The Ethics Committee of Hospital S. João, Porto, Portugal, approved the study with human auricular strips.

Experimental preparation. The study was performed in isolated right papillary muscles (n = 61) from male New Zealand White rabbits (Oryctolagus cuniculus; 2.0-3.0 kg) and in human right auricular trabeculae (n = 7) from seven patients undergoing coronary artery bypass grafting surgery. Rabbits were anesthetized with intravenous pentobarbital sodium (25 mg/kg). A left thoracotomy was performed, and beating hearts were quickly excised and immersed in modified Krebs-Ringer solution (composition in mmol/l: 98 NaCl, 4.7 KCl, 2.4 MgSO₄ · 7H₂O, 1.2 KH₂PO₄, 4.5 glucose, 1.8 CaCl₂ · 2H₂O, 17 NaHCO₃, 15 C₃H₃NaO₃, 5 CH₃COONa, 0.02 atenolol) at 35°C with cardioplegic 2,3-butanedione monoxime (BDM; 3%) and 5% newborn calf serum. Right human auricles were obtained during venous cannulation preceding cardiopulmonary bypass. They were immediately immersed in a Tyrode oxygenated solution (composition in mmol/l: 119.8 NaCl, 5.4 KCl, 1.05 MgCl₂, 1.8 CaCl₂ · 2H₂O, 5.0 glucose, 22.6 NaHCO₃, 0.42 NaH₂PO₄, 0.28 ascorbic acid, 0.05 EDTA, and 0.02 atenolol) at 35°C with BDM and newborn calf serum. Atenolol was used to prevent -adrenergic-mediated effects. Solutions were in equilibrium with 95% O₂ and 5% CO₂, maintaining the pH between 7.38 and 7.42. After dissection, papillary muscles (length: 4.2 ± 0.1 mm; weight: 2.9 ± 0.2 mg; preload: 4.7 ± 0.2 mN) and auricular muscle strips (length: 3.5 ± 0.2 mm; weight: 2.3 ± 0.5 mg; preload: 4.3 ± 1.0 mN) were vertically mounted in a 10-ml Plexiglas organ bath containing the aforementioned solutions and attached to an electromagnetic length-tension transducer (University of Antwerp, Belgium). Preload was estimated according to muscle dimensions, and the electrical stimulus (0.6 Hz for rabbit papillary muscles; 0.2 Hz for human trabeculae) was set at 10% above threshold. Twenty minutes later, bathing solutions were replaced by corresponding ones without BDM. During the next 2 h, muscles were stabilized. Bathing solutions were then replaced by corresponding ones without calf serum, and maximum physiological length (L_max) was calculated. Protocols were initiated after two superimposable isotonic and isometric control twitches separated by a 10-min interval were obtained.

Experimental protocol. Effects of ET-1 on contraction, relaxation, and diastolic properties of cardiac muscle were studied in rabbit papillary muscles (n = 9; 0.1, 1, and 10 nM) and in human muscle strips (n = 7; 10 nM). In rabbit papillary muscles, ET-1 (1 nM) was also added to the bathing solution in the presence of 1) the nonselective ET_A/ET_B receptor antagonist PD-145065 (0.1 µM, n = 9); 2) the selective ET_A receptor antagonist BQ-123 (0.1 µM, n = 9); 3) the selective ET_B receptor antagonist BQ-788 (0.1 µM, n = 6); and 4) the Na⁺/H⁺ exchanger inhibitor 5-(N-methyl-N-isobutyl)amiloride (MIA; 1 µM, n = 6). Effects of 6.3 mM CaCl₂, which increased inotropism to a similar extent to ET-1 (1 nM), were also studied in rabbit papillary muscles (n = 7). Because the selective ET_A receptor antagonist BQ-123 (0.1 µM) blocked most of the effects of ET-1 on the contractile parameters, in a subset of papillary muscles (n = 7) the reproducibility of ET-1 effects was evaluated by analyzing the response to ET-1 before and after the addition of BQ-123 (0.1 µM) as well as after this blocker was washed out. More precisely, the contractile performance of these muscles was assessed in the following conditions: 1) baseline, 2) after addition of ET-1 (1 nM), 3) after washout of ET-1, 4) after addition of BQ-123, 5) after addition of ET-1 (1 nM) in the presence of BQ-123, 6) after washout of ET-1 and BQ-123, 7) after addition of ET-1 (1 nM), and 8) after addition of ET-1 (10 nM).

Data acquisition and analysis. Isotonic, afterloaded, and isometric twitches were recorded and analyzed. Selected parameters include the following: resting tension (RT) at the beginning (RT_beg; mN/mm²) and at the end (RT_end; mN/mm²) of the twitch; active tension (AT; mN/mm²); maximum velocity of tension rise (dT/dt_max; mN · mm² · s¹); maximum velocity of tension decline (dT/dt_min; mN · mm² · s¹); peak isotonic shortening (PS; %L_max); maximum velocity of shortening (dL/dt_max; L_max/s); maximum velocity of lengthening (dL/dt_min; L_max/s); and time to half-relaxation (tHR, ms).

Statistical methods. Values are means ± SE. Effects on the various contractile parameters of a single dose of a given drug were analyzed by a paired t-test, whereas the effects of increasing doses of a given drug or of a single dose of various drugs added successively were analyzed by one-way repeated-measures ANOVA. Evaluation of the effects of ET-1 in the absence or presence of a blocker in the same group of muscles was performed with a repeated-measures two-way ANOVA. When significant differences were detected with any of the ANOVA tests, the Student-Newman-Keuls test was selected to perform pairwise multiple comparisons. P < 0.05 was accepted as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline performance of rabbit papillary muscles was similar in all experimental protocols. Mean values of the contractile parameters from the 61 papillary muscles were as follows: RT_beg, 8.7 ± 0.5 and RT_end, 8.4 ± 0.4 mN/mm² [P = not significant (NS), RT_end vs. RT_beg]; AT, 22.0 ± 1.6 mN/mm²; dT/dt_max, 149 ± 11 mN · mm² · s¹; dT/dt_min, 116 ± 8 mN · mm² · s¹; PS, 10.0 ± 0.5% of L_max; dL/dt_max, 0.70 ± 0.04 L_max/s; dL/dt_min, 2.7 ± 0.18 L_max/s; tHR, 399 ± 9 ms.

Effects of increasing concentrations of ET-1 (0.1, 1, and 10 nM) on contraction, relaxation and diastolic properties of rabbit papillary muscles are summarized and illustrated in Fig. 1, where it can be easily appreciated that such concentrations progressively increased AT, dT/dt_max, and dT/dt_min. For instance, 1 nM ET-1 increased 64 ± 18% AT, 59 ± 20% dT/dt_max, and 39 ± 13% dT/dt_min. Note that, for parameters expressed as negative values, such as dT/dt_min, the terms increase and decrease refer to their absolute values. At all concentrations, the effects of ET-1 on dT/dt_max were always of greater magnitude than its effects on dT/dt_min. Effects on tHR were not statistically significant. The novel finding of this study refers, however, to the effects of ET-1 on RT after an acute increase in afterload. As shown in Fig. 1, left, RT decreases significantly after an isometric twitch in the presence of ET-1. Such a decrease was not significant at baseline and became progressively larger with increasing doses of ET-1. In fact, compared with RT_beg, RT_end decreased 3.4 ± 1.4, 6.9 ± 1.5, and 12.5 ± 3.1% in the presence of 0.1, 1 and 10 nM of ET-1, respectively (Fig. 1, right). Such a decrease in RT reflects an increase in myocardial distensibility, as restoring the value of RT to its initial value results in an increase in the resting length of the muscle. This aspect is well illustrated in Fig. 2, where, in presence of 10 nM ET-1, the effects of increasing afterloads on muscle length (top) and tension (bottom) are shown. It can be seen that even the smaller afterload elevation displayed, which does not exceed 20% of AT of the isometric twitch, induces a decrease in RT after the twitch. Restoring RT to its initial value resulted in a distension of the muscle to a resting length larger than L_max. Decrease in RT augmented as afterload increased and was maximal after isometric twitches. After all afterload elevations, an increase of 5% of L_max was observed. Nonetheless, no significant decrease in contractile performance, namely in AT and dT/dt_max, was observed.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Effects of increasing concentrations of endothelin-1 (ET-1; 0.1, 1, and 10 nM) on isometric twitches of rabbit papillary muscles. Right: representative isometric twitches; middle: effects of ET-1 on active tension (AT) and peak rates of tension rise and decline (dT/dtmax and dT/dtmin, respectively). Left: effects of ET-1 on resting tension (RT). Compared with baseline, ET-1 dose dependently increased AT, dT/dtmax, and dT/dtmin, and decreased RT. The latter effect reflects an increase in myocardial distensibility. Data are means ± SE; %baseline. P < 0.05: avs. baseline, bvs. 0.1 nM ET-1, cvs. 1 nM ET-1.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effects of increasing afterloads on muscle length (top) and tension (bottom) of rabbit papillary muscles in the presence of ET-1 (10 nM). Afterload elevation induces a decrease in RT after the twitch. When the value of RT was restored to its initial value, resting length of the muscle increased. Decrease in RT augmented as afterload increased and was maximal in the isometric twitch.

Of note is that, in the eight rabbit papillary muscles that were never in contact with BDM, 10 nM ET-1 induced similar effects on muscle performance and RT to what was observed in the muscles that were stabilized in the presence of BDM, increasing 140 ± 36% AT, 154 ± 39% dT/dt_max, and 145 ± 38% dT/dt_min, whereas RT decreased 15.3 ± 3.2% after an isometric twitch.

Similar effects were observed in human auricular strips (Fig. 3). In this preparation, 10 nM ET-1 increased AT from 5.5 ± 1.9 to 17.5 ± 1.9 mN/mm², dT/dt_max from 64 ± 18 to 190 ± 19 mN · mm² · s¹, dT/dt_min from 36 ± 11 to 90 ± 12 mN · mm² · s¹, and PS from 4.6 ± 1.6 to 15.8 ± 5.6% of L_max, while not altering tHR (539 ± 48 vs. 540 ± 54 ms; P = NS). With regard to RT_end, it decreased, after an isometric twitch, 8.8 ± 2.2% compared with RT_beg (6.6 ± 0.9 vs. 7.2 ± 0.8 mN/mm², P < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effects of ET-1 (10 nM) on isometric twitches of rabbit and human myocardia. Similar effects were observed in both representative examples: ET-1 increased AT and decreased RT after the isometric twitch.

At each dose of ET-1, no significant relation was found between AT and the decrease in RT after isometric twitches.

Figures 4 and 5 show the effects of 1 nM ET-1 either alone or in presence of various substances, along with the effects of 6.3 mM CaCl₂. All of the inhibitors used in these protocols were added to the bath before addition of ET-1 (1 nM). None of them altered RT per se at any afterload level, even after isometric twitches. On the remaining contractile parameters, the effects of each inhibitor per se were as follows: 1) BQ-123 decreased significantly 11 ± 3% AT, 12 ± 5% dT/dt_max, and 9 ± 5% dT/dt_min, while not altering tHR (2 ± 3%); 2) BQ-788 did not significantly change AT (+0.5 ± 3%), dT/dt_max (0.2 ± 3%), dT/dt_min (3 ± 2%), or tHR (+0.4 ± 0.8%); 3) PD-145065 decreased significantly AT (7 ± 2%), dT/dt_max (8 ± 2%), dT/dt_min (7 ± 2%), while not altering tHR (+0.4 ± 1%); 4) MIA did not change significantly AT (1.3 ± 0.9%), dT/dt_max (+0.7 ± 1.2%), dT/dt_min (0.9 ± 1.2%) or tHR (+0.2 ± 0.2%).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effects of ET-1 (1 nM) on isometric twitches of representative rabbit papillary muscles, in the various experimental conditions, along with effects of CaCl2 (6.3 mM). See text for details.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Effects of ET-1 (1 nM) on contractile parameters of rabbit myocardium in the various experimental conditions. Effects of CaCl2 (6.3 mM) are also represented. RT, change in RT after an isometric twitch. Data are means ± SE; %change induced by ET-1 or CaCl2. P < 0.05: avs. before ET-1 or CaCl2, bvs. ET-1, cvs. ET-1 + ETA/ETB receptor antagonist PD-145065, dvs. ET-1 + ETA receptor antagonist BQ-123, evs. ET-1 + ETB receptor antagonist BQ-788.

With regard to the effects of 1 nM ET-1 in the presence of the inhibitors, we observed that all the effects described above for ET-1 alone, including the afterload-induced increase in myocardial distensibility, were completely blocked by 0.1 µM of the nonselective ET_A/ET_B antagonist PD-145065 or of the selective ET_A antagonist BQ-123. However, the muscles pretreated with BQ-123 showed small but significant negative inotropic and lusitropic effects in response to 1 nM ET-1: AT decreased 4.3 ± 2.4%, dT/dt_max 5.0 ± 2.6%, and dT/dt_min 6.5 ± 3.4%, whereas tHR did not change significantly. On the contrary, in the presence of BQ-788 (0.1 µM), the effects of ET-1 (1 nM) on myocardial distensibility were preserved, whereas its positive inotropic and lusitropic effects were slightly, although not significantly, enhanced. AT increased 82 ± 17%, dT/dt_max 123 ± 28%, and dT/dt_min 66 ± 15%. In the presence of the Na⁺/H⁺ exchanger inhibitor MIA (1 µM), the effects of ET-1 (1 nM) on RT were completely abolished, whereas the positive inotropic and lusitropic effects were only partially, and even not significantly, inhibited: AT increased 37 ± 12%, dT/dt_max 58 ± 19%, and dT/dt_min 23 ± 10%.

In the subset of papillary muscles where the reproducibility of ET-1 effects was evaluated, the first addition of ET-1 (1 nM) induced effects in all contractile parameters, including RT, similar to the ones described in the beginning of RESULTS. These effects of ET-1 were completely blocked in the presence of BQ-123 but reappeared in the last experimental protocols where ET-1 was added after BQ-123 was washed out. For instance, for RT, an isometric twitch decreased it by 7.2 ± 1.2% after the first dose of ET-1 (1 nM) but did not alter it in the presence of BQ-123 or BQ-123 plus ET-1 (1 nM). After the blocker was washed out, RT was again decreased after an isometric twitch by 5.4 ± 1.1% in the presence of ET-1 (1 nM) and by 10.7 ± 2.4% in the presence of ET-1 (10 nM). Although not statistically significant, effects of ET-1 in all contractile parameters showed a trend to be slightly smaller after washout of BQ-123 than what was observed after the first addition.

Finally, with regard to the effects of CaCl₂ (6.3 mM), although it showed positive inotropic and lusitropic effects similar to those of 1 nM ET-1 (AT increased 59 ± 8%, dT/dt_max 97 ± 10%, and dT/dt_min 38 ± 6%), it did not significantly affect RT.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In addition to the well-known positive inotropic and lusitropic effects of ET-1, the present study describes a novel effect of this agent on passive diastolic properties of the myocardium. It shows that ET-1 dose dependently increases myocardial distensibility of human and nonhuman myocardium; an effect that requires ET_A and Na⁺/H⁺ exchanger activation.

Positive inotropic and lusitropic effects of ET-1 and their dependence on ET_A receptor activation have been previously described by several authors in various experimental preparations, although the magnitude of effects varied among distinct animal species (see Ref. 5 for review). Rabbits are one of the most sensitive animals to ET-1 (21), which was one of the reasons for carrying out the experiments in this species. The magnitude of positive inotropic and lusitropic effects obtained in the present study is consonant with previously published data in rabbit papillary muscles (21).

On the other hand, effects of ET-1 on passive diastolic properties have received much less attention, and the scarce references to these issues have reported elevations of resting tension, ventricular end-diastolic pressure, or diastolic stiffness following administration of ET-1 (3, 6, 12, 14, 15, 26), which is the opposite of what we describe in the present study. Those studies, however, were carried out in the intact heart, where it is not possible to differentiate the effects of ET-1 on intrinsic myocardial diastolic properties from those resulting from peripheral and coronary vasoconstriction. Either load elevations or ischemia, induced by peripheral or coronary vasoconstriction, respectively, might decrease myocardial distensibility (2, 18, 19, 30) and therefore explain on its own the observations of those studies. Because the present study was performed in isolated muscle strips, this limitation was overcome, and the observed increase in myocardial distensibility presumably reflects a true change in intrinsic myocardial properties, confirming preliminary evidence previously gathered by our group (20).

This increase in myocardial distensibility induced by ET-1 followed an acute afterload elevation and became progressively larger as afterload or ET-1 concentration, and therefore developed tension, increased. These results might suggest a relation between developed tension and the magnitude of the decrease in RT. Such a relation could indicate that stress relaxation and creep, and not a specific effect of ET-1, would explain our findings. This hypothesis does not seem plausible because, if it held true, then a similar decrease in RT would be observed with 6.3 mM CaCl₂, which increased AT to a similar extent to 1 nM ET-1. Additionally, although at each concentration of ET-1 some within-group variability of AT was observed, no significant relation between the decrease in RT and AT could be obtained.

Although muscle length was stretched beyond L_max, following afterload-induced decrease in RT in presence of ET-1, no impairment in contractile performance was observed.

Because BDM was present during the stabilization period in all experimental protocols and has, on its own, important effects on myocardial performance due to activation of several phosphatases (44), it was important to clarify how it interacted with the effects of ET-1 on RT. First, it should be taken into account that the effects of BDM are generally considered to be reversible (44), and it was removed before the experimental protocol was started. Second, whereas the increase in myocardial distensibility was observed only in some of the protocols, this indicates that BDM presumably is not mainly responsible for that effect. This argument, however, cannot exclude the possibiity that such an effect would result from an interaction between ET-1 and some residual effects of BDM. To test this hypothesis, we investigated, in eight additional rabbit papillary muscles to which no BDM was added in the beginning of the experiments, the effects of 10 nM ET-1. In these muscles, the results were not significantly different from the results obtained previously for all contractile parameters, including RT.

Nonselective ET_A/ET_B receptor blockade completely abolished the positive inotropic and lusitropic effects of ET-1 and the increase in myocardial distensibility. Selective ET_A antagonism also blocked the effects of ET-1 on myocardial distensibility but induced small, although significant, negative inotropic and lusitropic effects. On the contrary, selective ET_B inhibition showed a tendency to enhance ET-1 effects on contractility and lusitropism while not altering its effects on myocardial distensibility. Taken together, these results suggest that ET_A receptors mediate the increase in inotropy, lusitropy, and myocardial distensibility elicited by ET-1, whereas ET_B activation seems to induce negative inotropic and lusitropic effects while not affecting distensibility. NO release has been detected in the sequence of ET_B stimulation and seems to mediate its vasodilating properties (4, 40). Because most authors consider NO a negative inotrope (see Ref. 31 for review), its release might explain the observed myocardial effects of ET_B stimulation. Interestingly, although NO has been shown to increase myocardial distensibility (31), it does not seem to explain the effects observed in the present study of ET-1 on this property.

ET_A stimulation activates phospholipase C (PLC), through a G protein-mediated pathway (11), leading to the production of inositol 1,4,5-trisphosphate and 1,2-diacylglycerol (DAG) (9). In cardiomyocytes, IP₃ increases cytosolic Ca²⁺, namely by promoting its release from the sarcoplasmic reticulum (34). DAG activates protein kinase C, thereby increasing systolic Ca²⁺ and enhancing myofilament Ca²⁺ responsiveness (10, 33). It has been proposed that these effects of PKC activation are, at least partially, due to stimulation of sarcolemmal Na⁺/H⁺ exchanger and consequent rise of intracellular pH (16, 33). Cytosolic alkalinization is associated with an increase in Ca²⁺ sensitivity of contractile proteins (23), which has been shown to contribute to the inotropic and lusitropic effects of ET-1 (32, 33, 41). The present study shows that inhibition of the sarcolemmal Na⁺/H⁺ exchanger by MIA completely abolished effects of ET-1 on myocardial distensibility, whereas it only moderately and nonsignificantly reduced its positive inotropic and lusitropic effects. This result provides an additional argument against the relation between augmentation of AT and increase in myocardial distensibility induced by ET-1, further ruling out the hypothesis of creep and stress relaxation.

With regard to the reproducibility of ET-1 effects, it was tested with an experimental design of the type baseline experiment, blocker experiment, baseline experiment. This protocol showed that all the effects of ET-1, including the increase in myocardial distensibility, could be elicited before addition and after washout of the blocker (BQ-123). The slight and not statistically significant smaller magnitude of ET-1 effects in all contractile parameters after washout than before addition of BQ-123 might be due to some residual effects of this blocker that presumably could not be totally removed from the experimental preparation.

Finally, concerning the pathophysiological relevance of our findings, we must point out that even a 5% increase in muscle length, as was obtained in response to the decrease in RT observed in present study, will determine a 15.6% increase in ventricular volume (a given change in length determines the cube of that change in volume). If we take into account that such change occurs in the first twitch following the increase in afterload, this indicates that this effect might allow the ventricle to accommodate, on a single beat, a 15.6% larger volume at similar diastolic pressures. This represents an additional mechanism through which, besides load (18) and NO (31), the heart can acutely modulate the diastolic pressure-volume relation.

On the basis of this property, it is tempting to speculate that, in the presence of cardiac overload, ET-1 might increase myocardial distensibility and lead to acute ventricular dilatation without compromising contractile function. This would allow a more efficient recruitment of preload reserve, with the ventricle being able to reach higher volumes with smaller increases of filling pressures.

In conclusion, this study has shown that ET-1 increases diastolic distensibility of human and nonhuman myocardium. This effect is mediated by ET_A receptors and is dependent on the Na⁺/H⁺ exchanger activation. It is not due to creep or stress relaxation, as it showed no relation to active tension and was not elicited by Ca²⁺. Such effect indicates that, besides load and NO, ET-1 can acutely modulate the diastolic pressure-volume relation.