The aims of this study were to determine if endothelin-1 (ET-1) under normal and ischaemic conditions exhibits a direct arrhythmogenic effect that is independent of its ability to cause coronary vasoconstriction, and to determine the contribution of the ETA and ETB receptor subtype. ETA/B (with ET-1) and ETA (ET-1 in the presence of BQ-788) receptor activation resulted in a significant reduction in both epi- and endocardial monophasic action potential duration (MAPD90). ETA receptor activation reduced both epi- and endocardial effective refractory period (ERP). This MAPD90 and ERP shortening were associated with a reduction in coronary flow, myocardial contractility and induction of ventricular fibrillation (VF) during ERP measurement. The ETB agonist sarafotoxin (S6c) had no marked, or concentration-dependent, effect on MAPD90, ERP, myocardial contractility or induction of arrhythmias. Neither ET-1 nor S6c, given prior to coronary artery occlusion, significantly changed the ischaemia-induced dispersion of MAPD90, ERP or the % incidence of VF. In conclusion, neither ETA nor ETB receptor stimulation has a direct arrhythmogenic effect in isolated rabbit hearts under normal or ischaemic conditions. The ET-1-induced arrhythmogenic effect observed in nonischaemic hearts is likely to be the result of the associated coronary vasoconstriction caused by ETA receptor stimulation resulting in myocardial ischaemia. Keywords: Endothelin-1, monophasic action potential duration, working heart, rabbit, sarafotoxin, BQ-788 Introduction Endothelin-1 (ET-1) levels have been shown to be increased during myocardial ischaemia in humans (Miyauchi et al., 1989; Yasuda et al., 1990), pigs (Wang et al., 1995) and rabbits (Vitola et al., 1996). Exogenous administration of ET-1 results in myocardial ischaemia in dogs through its potent coronary vasoconstrictor ability (Salvati et al., 1991) and causes ventricular arrhythmias (Merkely et al., 1998). Antagonists at ET-1 receptors protect against ischaemia-induced arrhythmias (Horkay et al., 2000; Kiss et al., 2000), supporting the view that ET-1 may cause myocardial ischaemia and play a role in the resultant arrhythmogenesis. ET-1 has also been shown to exhibit a direct arrhythmogenic effect, which is independent of its ability to cause myocardial ischaemia (Yorikane et al., 1991; Toth et al., 1995). However, the mechanism(s) that may underlie any such direct effect have not been clarified. Becker et al. (2000) showed that intracoronary administration of ET-1 induced ventricular arrhythmias that were not associated with changes in refractory periods or conduction delay, such changes usually being associated with myocardial ischaemia (Fozzard & Makielski, 1985). On the other hand, Szabo et al. (2000) reported that ET-1 induced arrhythmias and prolonged left ventricular (LV) monophasic action potential duration (MAPD) in dogs. It has been suggested that a mechanism, which may contribute to ET-1's direct arrhythmogenic effect, is regional heterogeneity in the electrophysiological effects of ET-1, increased dispersion of action potential duration and refractoriness and the generation of early after-depolarisations (EADs) (Geller et al., 1998; Duru et al., 2001). Thus, there exists controversy in the literature about the extent to which the arrhythmogenic effect of ET-1 is a consequence of the induction of myocardial ischaemia or a direct electrophysiological effect, and, if direct, what the mechanism is. To date, there have been no studies that have looked at the effect of ET-1 on dispersion of action potential duration and refractoriness; therefore, one of the aims of the current study is to determine if ET-1 modifies dispersion in such a way as to explain the arrhythmogenic effects of ET-1. ET-1 acts as an agonist at both ETA and ETB receptors. The arrhythmogenic action of exogenous ET-1, the MAPD prolongation and EAD formation has been shown to be blocked by the ETA selective antagonist, Lu 135.252, suggesting that these effects are mediated via ETA receptors (Kiss et al., 2000). In contrast, stimulation of ETB receptors with the selective agonist, sarafotoxin (S6c), exerted an antiarrhythmic action on ischaemia-induced arrhythmias in vivo (Crockett et al., 2000). Although ETB receptors are found within the heart predominantly in the conducting system and endocardium (Molenaar et al., 1993), no studies appear to have examined what electrophysiological effects they produce in whole hearts, if any such effects are more pronounced in the endo- vs the epicardium and if these could explain an antiarrhythmic action of ETB stimulation. The aims of this study were to use an isolated working rabbit heart model in which MAPD90, the effective refractory period (ERP) and conduction delay could be measured in both the endocardium and epicardium and correlated with haemodynamic variables to answer the following questions. Does ET-1, under normal and ischaemic conditions, have a direct arrhythmogenic effect that is independent of its ability to cause coronary vasoconstriction, and what is the contribution of ETA and ETB receptors to any observed effect? To address the latter question, studies were carried out using the selective ETB receptor agonist, S6c, and the mixed agonist, ET-1. Since there is no potent selective ETA agonist, the effects of ET-1 on ETA receptors were determined in the presence of the ETB antagonist, BQ-788 (Ishikawa et al., 1994). Regional myocardial ischaemia results in dispersion of action potential duration and refractoriness between the ischaemic and nonischaemic areas and it is thought that this dispersion contributes to the increased arrhythmogenicity. Therefore, in the current study, we used a regional model of ischaemia in order to determine if ET-1 or S6c modified this ischaemia-induced dispersion.