Endothelin and ischaemic arrhythmias–antiarrhythmic or arrhythmogenic?

Abstract

Objective: The aim of this study was to investigate the influence of endogenously released and exogenously applied endothelin-1 (ET-1) on ischaemia-induced arrhythmias. Methods: Ischaemia was induced in pentobarbitone-anaesthetised rats by ligation of a coronary artery for 30 min. To determine the role of endogenous ET-1 in ischaemic arrhythmias, either the ET_A receptor antagonist BQ123 (50 μg/kg/min, i.v.; n=10) or the ET_B receptor antagonist PD161721 (0.1 or 1 mg/kg i.v.; n=10 per group) was administered before the onset of ischaemia. To assess the influence of exogenous ET-1 on arrhythmias, ET-1 (1.6 nmol/kg i.v.) was administered 5 min before ischaemia in the absence (n=12) or presence of BQ123 (n=10) or PD161721 (n=10). The total number of ventricular ectopic beats (VEB's) were counted and expressed as median (Q₁−Q₃) and the incidence of ventricular fibrillation (VF) and ventricular tachycardia (VT) in each group was determined. Mean arterial blood pressure (MABP) and heart rate (HR) were measured. Results: In control animals (n=20), the incidence of VF was 65% and the total VEB count was 2775 (1870–4041). Both BQ123 and the higher dose of PD161721 reduced the VEB count to 654 (348–1489; P<0.05) and 782 (432–1153; P<0.05) respectively. Only PD161721 reduced the incidence of VF (to 10%; P<0.05). Administration of ET-1 reduced VEB's to 1530 (1204–2017); P<0.05) and the incidence of VF to 17% (P<0.05). Neither PD161721 nor BQ123 modified this antiarrhythmic effect of ET-1, but rather enhanced the reduction in arrhythmias. Before occlusion, ET-1 caused a transient fall in MABP (from 107±3 to 63±3 mmHg; P<0.05). PD161721, but not BQ123, partially blocked this effect. Upon occlusion, MABP fell in control animals (from 106±4 to 67±4 mmHg at 1 min post-occlusion; P<0.05). This was significantly attenuated by ET-1, although neither of the antagonists were able to block this effect of ET-1. Conclusions: ET-1 released endogenously during ischaemia is arrhythmogenic whereas exogenous application of ET-1 may, under certain conditions, be antiarrhythmic.

Endothelin-1, Ventricular arrhythhmias, Anaesthetised rats, BQ123, PD161721

Time for primary review 29 days

1 Introduction

Endothelin-1 (ET-1) is an endothelium-derived substance for which two specific binding sites have been identified, ET_A and ET_B[1]. Both ET_A and ET_B receptors are located in the vasculature [2, 3]and cardiac muscle [4]. Elevated plasma levels of ET-1 have been reported in patients with ischaemic heart disease [5]and in animal experimental models of ischaemia [6], suggesting a potential role of ET-1 in the consequences of ischaemia/reperfusion. It has been shown that endogenous ET-1 is released from the heart under basal, hypoxic or ischaemic conditions and that a monoclonal antibody of ET-1 can reduce infarct size in an in vivo rat model of myocardial ischaemia [7], and in in vivo and in vitro rabbit models of coronary artery occlusion [8]. Studies with both ET_A-selective (BQ-123; [9]) and non-selective (TAK-044 and bosentan, [10, 11]) ET receptor antagonists, have demonstrated the ability of these drugs to reduce infarct size in a range of species. In addition to contributing to the extension of ischaemia-induced myocardial injury, there is evidence that endogenous ET-1 may be arrhythmogenic. Both BQ-123 and the non-selective ET receptor antagonist PD145065 have been shown to suppress ischaemia-induced cardiac arrhythmias in in vivo rat [12]and rabbit [13]models of coronary artery occlusion. ET-1 has been shown to have direct electrophysiological effects on cardiac muscle which may contribute to arrhythmogenesis [14], although it is not known which ET-receptor mediates these effects.

In contrast to the studies which have suggested a role for endogenous ET-1 in contributing to the pathological effects of ischaemia, it has recently been reported that exogenous administration of low doses of ET-1 (i.e. below threshold required to induce vasoconstriction) have a cardioprotective effect against ischaemia/reperfusion injury by reducing infarct size in anaesthetised rats [15], in rabbit isolated hearts [16]and in rat isolated hearts [17]. This cardioprotective effect of ET-1 can be abolished by glibenclamide and 5-hydroxydecanoate (5-HD; K⁺_ATP channel blockers [15, 16]), by chelerythrine (a PKC inhibitor; [16, 17]) and by the ET_A receptor antagonist BQ-123 and the non-selective antagonists PD156707 and bosentan [16, 17]. These studies indicate that exogenously applied ET-1 may protect against ischaemia/reperfusion injury via an ET_A receptor-mediated PKC-dependent mechanism and activation of K⁺_ATP channels. Although, the cardioprotective effect of ET-1 against ischaemia/reperfusion-induced infarct size has been demonstrated, no studies have examined whether or not ET-1 can exhibit a similar protection against ischaemia-induced cardiac arrhythmias.

The aim of the present study is to test the hypothesis that ET-1 may have dual effects in the setting of myocardial ischaemia-induced arrhythmogenesis, namely that when released endogenously during ischaemia, ET-1 has a pro-arrhythmic effect whereas when administered exogenously it may exhibit a cardioprotective (i.e. anti-arrhythmic) effect. The role of endogenous ET-1, and its receptor subtypes, in the generation of ischaemia-induced arrhythmias, was studied in anaesthetised rats by the use of the recently developed ET receptor antagonist, PD161721, [Sodium3-cyclohexylmethyl-4-(2,3-dihydro-benzo-[1,4]dioxin-6-yl)-2-(7-methoxy- benzo[1,3]dioxo-5-yl)-4-oxo-but-2-enoate] [18]and the ET_A receptor antagonist, BQ-123, [cyclo(d-Asp-l-Pro-d-Val-l-Leu-d-Trp)] [19]. The possible anti-arrhythmic effect of ET-1 was investigated by the administration of low doses of ET-1 prior to coronary artery occlusion and the involvement of ET_A and ET_B receptors in any observed effect determined by using the receptor antagonists above.

2 Methods

2.1 Surgical procedure

All experiments were performed in accordance with the United Kingdom Home Office Guide on the Operation of Animals (Scientific Procedures) Act 1986. Male Sprague-Dawley rats (220–350 g) were anaesthetized with sodium pentobarbitone (60 mg/kg intraperitoneally, i.p.) and maintained under anaesthesia by bolus injections of sodium pentobarbitone (3–6 mg intravenously, i.v.) as required. The trachea was cannulated (using a polythene cannula, Portex; OD 0.98 mm, ID 0.58 mm) for artificial respiration and systemic arterial blood pressure (BP) was recorded via a catheter inserted into the left carotid artery and attached to a blood pressure transducer (Gould, P231D, USA). The right and left jugular veins were cannulated for administration of additional anaesthetic or drugs as appropriate. A pre-calibrated steel thermistor probe was inserted into the rectum to measure core temperature which was maintained at 37–38°C with the aid of a heating lamp. A standard limb lead I electrocardiogram (ECG) was monitored continuously throughout the experimental period by insertion of subcutaneous needle electrodes. Both the ECG and BP were continuously recorded on a Grass polygraph (model 7D, Grass Instruments, Quincy, MA, USA).

Following tracheal intubation, a longitudinal skin incision was made to the left of the midline and the muscle layers covering the chest opened by blunt dissection. The chest was opened by a left thoracotomy performed between the fourth and the fifth ribs approximately 3 mm from the sternum. Artificial respiration with room air was immediately initiated using a respirator (C.F. Palmer, UK, volume 1.5 ml per 100 g, rate 54 strokes per min). This is sufficient to maintain PCO₂, 18–24 mmHg, PO₂, 100–130 mmHg, and pH within normal limits, 7.4 units [20]. The two ribs above the thoracotomy were sectioned and the pericardium incised to allow access to the heart. The heart was then exteriorised by application of gentle pressure to either side of the ribcage. A 6/0 braided silk suture (attached to a 10-mm micropoint reverse cutting needle) was placed around the left main coronary artery [20]. The heart was replaced into the thoracic cavity and the animal was allowed to stabilise for 15 min before the onset of the experimental protocol. Any animal with a mean arterial BP (MABP) <70 mmHg was discarded.

2.2 Parameters measured and arrhythmias analysis

In this study, systolic and diastolic arterial BP (SBP, DBP) were measured from the arterial BP trace and mean arterial blood pressure (MABP) was calculated. Heart rate (HR, beat per min, bpm) was calculated from the ECG. Ventricular arrhythmias were analysed according to the guidelines of the Lambeth Conventions for determination of experimental arrhythmias [21]. These were identified as single ventricular ectopic beats (defined as discrete and recognisable premature QRS complexes in relation to the P wave, with downward T wave), salvos (defined as a run of two or three ventricular ectopic beats in a row), and ventricular tachycardia (VT, a run of four or more consecutive ectopic beats, occurring with clearly characterised R waves and at a rate faster than sinus rhythm). These ventricular arrhythmias were associated with a decrease in blood pressure, which was most pronounced during VT. The total number of ventricular ectopic beats (VEBs) was calculated as the sum of the individual arrhythmias in animals that survived throughout the period of coronary artery occlusion. Ventricular fibrillation (VF) was defined when individual QRS complexes could no longer be distinguished, successive waves were inconsistent both in amplitude and in rhythm, and was accompanied by a sharp fall in blood pressure to zero mmHg with no pulse pressure. The incidence of ventricular tachycardia was recorded for each group and, as the rat can spontaneously recover from VF, the incidence of reversible and irreversible VF was noted. Mortality from irreversible VF, bradycardia and atrio-ventricular (A-V) block was also recorded, and these animals were excluded from the statistical analysis of the number of VEBs.

2.3 Experimental protocols

All animals were allowed a 15-min stabilisation period prior to drug treatment or coronary artery occlusion. The left main coronary artery was occluded for 30 min. In all experimental protocols time-matched control experiments were carried out.

2.3.1 The effect of the ET_A receptor antagonist BQ-123 and the ET_B receptor antagonist PD161721 on ischaemic arrhythmias

By employing a similar method to that used in this study, we have reported previously that an infusion of BQ-123 (50 μg/kg/min) inhibited both the ischaemic arrhythmias and the pressor responses to cumulative doses of ET-1 in anaesthetised rats [15]. Therefore, this dose of BQ-123 was chosen in these experiments. Using an infusion apparatus (Harvard, USA), the i.v. infusion (via the right jugular vein) of BQ-123 (50 μg/kg/min) was commenced 10 min prior to and during 30 min of occlusion at a rate of 0.02 ml/min (total infusion volume of 0.8 ml/40 min; n=10). Control rats (n=20) were administered an equal volume of saline with the same infusion rate and duration.

In an initial set of experiments carried out in anaesthetised closed chest rats it was shown that 0.1 mg/kg PD161721 had no effect on the haemodynamic responses to a submaximum intravenous dose of ET-1 (1.6 nmol/kg) while 1 mg/kg showed a maximum attenuation of the hypotensive effect of ET-1 without affecting the pressor response. Thus at this dose it is acting as a selective ET_B-receptor antagonist. Accordingly, rats were allocated to one of three groups – control (bolus saline, n=20); PD161721, 0.1 mg/kg, (n=7) and PD161721, 1 mg/kg (n=10). The bolus doses of PD161721 were administered i.v., 10 min before coronary artery occlusion.

2.3.2 The effect of pretreatment with bolus doses of ET-1 (0.4 and 1.6 nmol/kg) on ischaemia-induced arrhythmias

From a cumulative dose response study of the effects of ET-1 on mean arterial blood pressure, two doses of ET-1 (0.4 and 1.6 nmol/kg) were chosen for these experiments. The lower dose exhibited a subthreshold effect on mean arterial blood pressure while the higher dose had a submaximal hypertensive effect. Animals were allocated into one of three groups and given an i.v. bolus dose of either saline (control, n=20); ET-1 (0.4 nmol/kg, n=9); or ET-1 (1.6 nmol/kg, n=12) 5 min before coronary artery occlusion.

2.3.3 The effects of ET-1 in the presence of either the ET_A or ET_B receptor antagonists

On the basis of results obtained using the above protocols, the following doses of drug were chosen: ET-1 (1.6 nmol/kg, n=12; BQ-123 (50 μg/kg/min, n=10); PD161721 (1 mg/kg, n=10). The protocol used is illustrated in Fig. 1. The infusion of BQ-123 was commenced 10 min prior to and maintained during 30 min of ischaemia (at a rate of 0.02 ml/min and a total infusion of 0.8 ml per 40 min). PD161721 and ET-1 were administered as an i.v. bolus 10 and 5 min prior to ischaemia, respectively.

Fig. 1

Experimental protocol to study the effect of BQ-123 or PD161721 on the antiarrhythmic effects of ET-1.

2.4 Statistical analysis

The incidence of VT, VF and mortality are expressed as a percentage incidence within a group and statistical significance assessed by using Fisher's exact test. The number of VEBs are expressed as median (Q1−Q₃) and compared by Mann-Whitney nonparametric test. MABP and HR are expressed as mean±SEM and assessed within the group by one-way analysis of variance and significant differences by Dunnett multiple comparison test. However, MABP and HR between groups were compared by two-tailed unpaired Student's t-test. Differences between groups were considered significant if P<0.05.

2.5 Drugs

Phenobarbitone sodium and heparin were purchased from Rhône Mérieux (Ireland) and Leo Laboratory (Ireland), respectively. ET-1 (human/porcine) and BQ-123 (ET_A receptor antagonist) were obtained from Sigma Chemicals and Alexis, respectively. PD161721 (ET_B receptor antagonist) was a gift from Dr A. Doherty, Parke-Davis, USA. Each drug was dissolved in distilled water to yield stock solutions of 10 μg per 0.1 ml (for ET-1), 2 mg per 0.1 ml (for BQ-123 and PD161271). All stock solutions were stored at −25°C and diluted in heparinised saline (20 units of heparin per 1 ml of normal saline) immediately before use.

3 Results

3.1 Effects of ET_A and ET_B receptor antagonists on ischaemia-induced arrhythmias

Coronary artery occlusion resulted in severe ventricular arrhythmic activity in control animals. In most animals, ectopic beats began to appear ∼6–7 min after occlusion, reached a peak of activity at ∼11–15 min, and then declined by 16–20 min after occlusion. As shown in Fig. 2, the majority of ectopic activity occurred as VT. All control animals experienced VT while VF, the most severe form of ventricular arrhythmias, occurred in 65% (13 out of 20) of the animals. Of the rats which developed VF, more than half of the animals reverted spontaneously to sinus rhythm. The remaining animals died from VF, resulting in a group mortality of 30% (Fig. 3).

Fig. 3

The incidence of total ventricular fibrillation (VF) and mortality during 30 min of coronary artery occlusion in rats which received a bolus dose of saline (control, open bars; n=20), PD161721 (0.1 mg/kg, cross-hatched bars; n=7), PD161721 (1 mg/kg, solid bars; n=10) or BQ-123 (50 μg/kg/min, hatched bars; n=10). *P≤0.05 indicates significantly different from the saline group.

Fig. 2

The number of ventricular ectopic beats (VEBs) which occurred as single beats, salvos and ventricular tachycardia (VT) and counted during 30 min of coronary artery occlusion in rats which survived and which received bolus doses of saline (control, open bars; n=14), PD161721 (0.1 mg/kg, cross-hatched bars; n=4), PD161721 (1 mg/kg, solid bars; n=10) or BQ-123 (50 μg/kg/min, hatched bars; n=7). The total number of VEBs (Tot.VEBs) is the sum of these types of arrhythmias. *P≤0.05 indicates significantly different from the saline group.

Figs. 2 and 3 also illustrate the effects of blocking ET_A and ET_B receptors on ischaemia-induced arrhythmias. Infusion of the ET_A receptor antagonist BQ-123 (50 μg/kg/min) suppressed the number of VEBs during the peak of activity and significantly reduced the total number of VEBs, primarily as a result of decreasing VT (Fig. 2). BQ-123, however, had no effect on the incidence of VF or mortality (Fig. 3). PD161721, at a dose of 1 mg/kg, but not 0.1 mg/kg, suppressed the number of VEBs during the peak of activity and significantly reduced the total number of VEBs, again mainly as a result of decreasing VT (Fig. 2). Unlike BQ-123, 1 mg/kg PD161721 also significantly reduced both the total incidence of VF and mortality (Fig. 3).

Neither BQ-123 nor PD161721 had any significant effect on MABP or HR before occlusion. The only effect observed upon coronary artery occlusion was that the higher dose of PD161721 significantly attenuated the fall in MABP at 3, 10 and 15 min after occlusion compared to control rats. In control rats, MABP fell from 106±4 to 70±4, 76±5 and 80±4 mmHg at these time points, whereas in PD161721 treated rats the corresponding values were from 105±4 to 83±5, 92±2 and 98±5 mmHg (P<0.05).

Fig. 4 demonstrates that 1 mg/kg, but not 0.1 mg/kg⁻¹ of PD161721 blocked the hypotensive effect of a sub-maximal dose of ET-1 (1.6 nmol/kg) on mean arterial blood pressure in a separate group of anaesthetised closed-chest rats. The hypertensive effect of this dose of ET-1 was potentiated by the higher, but not the lower, dose of PD161721.

Fig. 4

The effect of a submaximum dose of ET-1 (1.6 nmol/kg) on mean arterial blood pressure (MABP) in the presence of a bolus dose of saline (open squares; n=6), PD161721 (0.1 mg/kg, solid triangle; n=4) or PD161721 (1 mg/kg, open circle; n=4) in anaesthetised closed chest rats. *P<0.05 indicates significantly different from the saline group.

3.2 Effect of exogenously applied ET-1, alone and in the presence of ET_A and ET_B receptor antagonists, during myocardial ischaemia

A bolus dose of 1.6 nmol/kg, but not 0.4 nmol/kg, ET-1 significantly reduced the number of VEBs during 30-min occlusion compared to controls. This anti-arrhythmic effect of ET-1 was associated with a reduction in the number of arrhythmias occurring as VT (Fig. 5). The higher dose of ET-1 also significantly reduced the incidence of VF and mortality (Fig. 6). Figs. 5 and 6 also demonstrate that neither the ET_A nor the ET_B receptor antagonist prevented this antiarrhythmic effect of ET-1. In contrast, there was a further reduction in the total number of VEBs in the presence of both PD161721 and BQ-123.

Fig. 6

The incidence of total ventricular fibrillation (VF) and mortality during 30 min of coronary artery occlusion in rats receiving i.v. bolus doses of saline (control, open bars; n=20), ET-1 (1.6 nmol/kg) in the absence (cross-hatched bars; n=12) or the presence of PD161721 (1 mg/kg, solid bars; n=10) or BQ-123 (50 μg/kg/min, hatched bars; n=10). *P<0.05 indicates significantly different from saline group.

Fig. 5

The number of ventricular ectopic beats (VEBs) which occurred as single beats, salvos and ventricular tachycardia (VT) and counted during 30 min of coronary artery occlusion in surviving rats receiving i.v. bolus doses of saline (control, open bars; n=14), ET-1 (1.6 nmol/kg) in the absence (cross-hatched bars; n=12) or the presence of PD161721 (1 mg/kg, solid bars; n=10) or BQ (50 μg/kg/min, hatched bars; n=10). *P<0.05 indicates significantly different from saline group.

Injection of ET-1 (1.6 nmol/kg), but not 0.4 nmol/kg, induced a significant transient depressor effect (from 107±3 to 63±3 mmHg) which returned to pre-injection levels before occlusion. Furthermore, the fall in MABP seen in control animals during ischaemia was attenuated by the higher, but not the lower, dose of ET-1. These effects of the higher dose of ET-1 are illustrated in Fig. 7A. The initial depressor effect of ET-1 was partially blocked by PD161721 (Fig. 7B) but not BQ-123 (Fig. 7C) resulting in a significant increase in MABP just before occlusion in this group of rats. During ischaemia, in PD161721 treated rats, MABP did fall upon occlusion but because of the marked increase in MABP before occlusion, it did not fall below the baseline value (Fig. 7B). The ET-1-induced attenuation of the fall in MABP was not affected by BQ-123 (Fig. 7C). No significant changes in HR were observed in any of the groups either before or during coronary artery occlusion.

Fig. 7

Mean arterial blood pressure (MABP) recorded before and during 30 min of coronary artery occlusion in rats receiving i.v. bolus doses of saline (open squares) or (A) ET-1 (1.6 nmol/kg, solid squares) or (B) ET-1 in the presence of PD161721 (solid triangles) or (C) ET-1 plus BQ-123 (solid circles). *P<0.05 indicates significantly different from saline group. #P<0.05 indicates significantly different from pre-occlusion value. @P<0.05 indicates significantly different from before ET-1 injection value.

4 Discussion

While the potential importance of ET-1 as a contributory factor in the extension of myocardial injury arising from ischaemia and reperfusion has been well investigated, the arrhythmogenic potential of endogenous ET-1, under conditions of ischaemia, has received much less attention. A recent study from this laboratory was the first to demonstrate that the ET_A-receptor antagonist BQ123 was antiarrhythmic, although this effect was evident only within a very narrow dose range, leading to the suggestion that ET_B-receptors may also contribute to ventricular arrhythmias [12]. In this study, we have shown that both selective ET_A- and ET_B-receptor antagonists possess antiarrhythmic properties, albeit with different activity against the different types of arrhythmias. The total number of ventricular ectopic beats was significantly reduced by the ET_A antagonist BQ-123, without affecting the incidence of VF, confirming our previous findings [12]. In contrast, the ET_B receptor antagonist PD161721 (1 mg/kg) reduced significantly both the arrhythmia count and the incidences of reversible and irreversible VF, and mortality. This data supports the hypothesis that endogenous ET-1 released during myocardial ischaemia is arrhythmogenic and suggests that both ET_A and ET_B receptors may be involved in this response.

In contrast to the apparent pro-arrhythmic effects of endogenously released ET-1, we observed that a low bolus dose of exogenously administered ET-1 was anti-arrhythmic and caused a marked reduction in both the incidence and severity of ischaemia-induced cardiac arrhythmias. These results are in line with previous studies which showed a cardioprotective effect of exogenous bolus doses of ET-1 against ischaemic/reperfusion injury by reducing infarct size in several different animal models [15–17]. However, the results obtained in this study do differ from those obtained in our previous study when ET-1 was given to rats as a continuous, low-dose infusion throughout the period of coronary occlusion, and in which a pro- arrhythmic effect was observed [12]. The mechanism(s) underlying the observed anti-arrhythmic effect of ET-1 has not been established. It is clear from the present experiments that this anti-arrhythmic effect was observed with 1.6 but not with 0.4 nmol/kg of ET-1. Thus, the concentration of ET-1 used is important. The higher, but not the lower, dose of ET-1 had marked haemodynamic effects which may be responsible for the observed anti-arrhythmic effect. The dose of ET-1, which was anti-arrhythmic, caused a marked fall in mean arterial blood pressure prior to ischaemia but also attenuated the fall in arterial blood pressure observed following coronary artery occlusion. A bolus dose of ET-1 does cause a transient vasodilatation followed by a prolonged vasoconstriction and this would explain the observed effects on mean arterial blood pressure. If the haemodynamic effects of ET-1 do underly its anti-arrhythmic effect, this would explain why the method of administration (i.e. bolus vs infusion) may also be of importance.

Neither of the ET-1 antagonists (BQ-123 or 161721) had any haemodynamic effect before ischaemia. This suggests that, prior to myocardial ischaemia, basally released ET-1 makes little or no contribution to systemic haemodynamics in anaesthetised rats, as has been previously reported [22]. The higher dose of PD161721, but not BQ-123, attenuated the fall in mean arterial blood pressure seen following coronary artery occlusion in control rats. Until now, we have been unable to speculate on the mechanism underlying the occlusion-induced hypotension which is characteristic of this model, but these findings suggest that ET-1 released during myocardial ischaemia may contribute to the fall in mean arterial blood pressure seen upon occlusion by an action on endothelial ET_B receptors. The improvement in arterial blood pressure during ischaemia by the ET_B receptor antagonist, PD161721, is one possible mechanism for its anti-fibrillatory effect. Furthermore, the failure of the lower dose of PD161721 (0.1 mg/kg) to affect the initial depressor response to a single submaximum dose of ET-1 (1.6 nmol/kg), supports the view that the lack of effect of this dose on ischaemia-induced cardiac arrhythmias is a consequence of its lack of ability to block ET_B receptors. Therefore, this study provides, for the first time, evidence for the involvement of ET_B receptors in the role of endogenous ET-1 in worsening the electrophysiological consequences of myocardial ischaemia. However, the precise arrhythmogenic mechanisms underlying the effect of stimulation of both ET_A and ET_B receptors remains to be elucidated.

The receptor subtype(s) mediating the anti-arrhythmic effect of exogenously administered ET-1 has also not been elucidated. In the present study, neither the ET_A (BQ123) nor the ET_B selective antagonists (PD161721) prevented the anti-arrhythmic effect of ET-1. Indeed, in the presence of exogenously applied ET-1, both antagonists caused a further reduction in VEBs and the incidence of VF and mortality. Since these antagonists were anti-arrhythmic on their own in the setting of myocardial ischaemia (by preventing the pro-arrhythmic effect of endogenous ET-1) it is probable that what we have observed is the summation of two anti-arrhythmic effects. The ET_B receptor antagonist PD161721 significantly blocked the ET-1- induced vasodepressor effect prior to ischaemia, showing that this is ET_B-receptor mediated. However, neither antagonist blocked the attenuation of the fall in arterial blood pressure during ischaemia seen following bolus administration of ET-1. The reason for this is not known. It is possible that use of a higher concentration of either antagonist or combined use of both antagonists may be necessary to block all of the receptors via which exogenously applied ET-1 can produce its haemodynamic effect. It is also not known if the mechanism(s) responsible for the ability of ET-1 to protect against ischaemia-induced arrhythmogenesis and infarct size are the same. Further work is required to address these questions.

In summary, our results do support the hypothesis that ET-1 can have dual effects on cardiac arrhythmias associated with myocardial ischaemia. Both ET_A and ET_B receptor blockade had an anti-arrhythmic effect suggesting that endogenously released ET-1 may act via both receptor subtypes to produce a pro-arrhythmic effect. Our finding that a very low dose of endothelin, given as a bolus injection prior to ischaemia, can protect against ischaemic arrhythmias, suggests that, as with infarct size, exogenously applied ET-1 is capable of mimicking the anti-arrhythmic effect of preconditioning. The mechanism(s) underlying both the pro- and anti-arrhythmic effects of ET-1 remain to be determined.