1. Endothelin-1 (ET-1) produces constriction of the rat mesenteric vascular bed in vivo via ETA and ETB receptor subtypes. The aim of this study was to investigate the relative roles of these receptor subtypes in rat isolated, endothelium-denuded, small mesenteric arteries, under pressure, by use of ET-1; the ETA receptor antagonist, BQ-123; the ETB receptor selective agonist, sarafotoxin S6c (SRTX S6c); the ETB receptor selective antagonist, BQ-788; and the ETA/ETB antagonist, TAK-044. 2. In 3rd generation mesenteric arteries, ET-1 (10(-13)-10(-7) M) produced concentration-dependent contractions (pD2 9.86). SRTX S6c (10(-12)-10(-7) M) also induced concentration-dependent contractions in 53% of arteries studied, although the Emax was much less than that obtained with ET-1 (10.7 +/- 2.9% vs 101.9 +/- 2.6% of the 60 mM KCl-induced contraction). 3. Neither ETB receptor desensitization, by a supra-maximal concentration of SRTX S6c (10(-7) M), nor incubation with BQ-788 (3 x 10(-8) M), had any significant effect on the ET-1 concentration-response curve, although both treatments tended to enhance rather than inhibit responses to ET-1. 4. In the presence of BQ-123 (10(-6) M), responses to low concentrations of ET-1 (up to 10(-10) M) were unaffected but responses to concentrations of ET-1 above 10(-10) M were significantly inhibited. 5. SRTX S6c desensitization followed by incubation with BQ-123 (10(-6) M) or co-incubation with BQ-788 (3 x 10(-8) M) and BQ-123 caused inhibition of responses to all concentrations of ET-1, resulting in a rightward shift of the ET-1 concentration-response curve. The same effect was obtained by incubation with TAK-044 (10(-8) M and 3 x 10(-7) M). 6. Thus, responses of rat small mesenteric arteries to ET-1 are mediated by both ETA and ETB receptors. The relative role of ETB receptors is greater than that predicted by the small responses to SRTX S6c or by resistance of ET-1-induced contraction to ETB receptor desensitization or BQ-788. The effect of ETB receptor desensitization or blockade is only revealed in the presence of ETA receptor blockade, suggesting the presence of a 'crosstalk' mechanism between the receptors. These results support the concept that dual receptor antagonists, like TAK-044, may be required to inhibit completely constrictor responses to ET-1.