Endothelin-1 (ET-1) causes long-lasting vasoconstriction resulting from tight binding to ETA-receptors. This hinders effects of competitive antagonists. We hypothesized that ET-receptor function can be modulated by non-competitive, allosteric mechanisms. We studied isolated rat mesenteric arteries after exposure to capsaicin in the continuous presence of L-NAME and indomethacin. Endothelins caused contractions (potency ET-1 >= ET-2 >> ET-3). 4AlaET-1 (ETB-agonist) was ineffective. BQ788 (ETB-antagonist) did not modify ET-1-induced responses. Presence of BQ123 (ETA-antagonist) reduced the sensitivity to ETs to a markedly different extent (pA2 of preventive effect: ET-3 > ET-1 > ET-2). BQ123 also caused relaxation of contractions in the presence of ETs and of contractions persisting after exposure to ETs. These relaxing effects were agonist-dependent; smaller and larger than predicted by the preventive effect of BQ123 in the case of ET-1 and ET-2, respectively. ACT-062724 (ACT), a novel non-peptidergic inhibitor of the binding of 125I-ET-1 to human ETA-receptors was similarly investigated. Presence of ACT reduced the sensitivity to ET-1 and ET-2 to the same extent. ACT reversibly relaxed contractile responses to ET-1 and ET-2. For both agonists, relaxing effects of ACT were significantly larger than predicted by preventive effects. In conclusion, preventive and relaxing effects of putative ETA-receptor antagonists differ and differ between agonists. Differential effects of antagonists on receptor-affinity and -efficacy and agonist-dependence rank among the criteria of allosteric receptor modulation which thus seems to apply to ETA-receptors. Future efforts may focus on candidate drugs that work best on activated receptors.