Endothelins are a family of potent endogenous mediators that have been implicated in a number of airway and other diseases. Recently, the non-peptide mixed ET(A)/ET(B) endothelin receptor antagonist bosentan has been successfully tested in the treatment of cardiovascular diseases. It was the aim of the present study to characterize the effects of bosentan on the pulmonary actions of endothelin- (ET-1), endothelin-3 (ET-3) and the ET(B)-receptor agonist IRL1620 in the isolated perfused and ventilated rat lung (IPL) and in precision-cut lung slices (PCLS). In the IPL, bosentan completely prevented the IRL1620-induced vasoconstriction (IC50 3 microM). The inhibition by bosentan of ET-1-elicited vasoconstriction showed a biphasic course, reflecting the inhibition of ET(A)-and ET(B)-mediated vasoconstriction (IC50 0.2 microM and 19 microM, respectively). In addition, bosentan prevented the ET-1- (IC50 6 microM) and IRL1620-induced (IC50 3 microM) prostacyclin release. Bosentan also completely prevented the bronchoconstriction induced by IRL1620 in the IPL (IC50 20 microM) and in PCLS (IC50 13 microM). In PCLS, the pD2-values were ET-1 7.20+/-0.23, ET-3 7.51+/-0.27 and IRL1620 7.33+/-0.29. Bosentan at 100 microM caused a rightward shift of the concentration-response curve of ET-1, ET-3 and IRL1620 by a factor of 5, 46 and 64, respectively. In all cases the slope of the Schild regression was lower than unity, disregarding a simple interaction of bosentan with one receptor. With respect to ET-1-induced bronchoconstriction, in the IPL bosentan in concentrations of up to 10 microM aggravated ET-1-induced bronchoconstriction probably due to the blockade of bronchodilatory ET(A)-receptors (IC50 0.3 microM) and even at 100 microM showed only very little protection from ET- -induced bronchoconstriction in the IPL and in the PCLS. The similar IC50-values for ET-1-induced vasoconstriction and bronchodilation suggest that only one type of ET(A)-receptor is involved. The differing IC50-values between IRL1620-induced bronchoconstriction and prostacyclin release, the slope of the Schild regression and the failure of bosentan to prevent the ET-1-induced bronchoconstriction suggest a complex interaction between the known ET-receptors or the existence of unknown ET(B)-receptor subtypes.