1. To evaluate the possible contribution of endothelin-1 (ET-1) to the pathophysiology of porcine septic shock, the non-peptide, mixed ET-receptor antagonist, bosentan (RO 47-0203) was administered (5 mg kg-1, i.v.) 30 min before infusion of lipopolysaccharide (LPS) (E. coli., serotype 0111:B4) (15 micrograms kg-1 h-1) and at 3.5 h of endotoxaemia in six anaesthetized and mechanically ventilated pigs. Six other pigs served as controls and received only LPS infusion. Pulmonary and systemic haemodynamics as well as splenic, renal and intestinal blood flows were measured continuously. Release and synthesis of ET-1 and Big ET-1 were also measured. 2. Only three of the six pigs in the control group survived 3 h of LPS infusion while in the bosentantreated group all six pigs were alive at that time. A biphasic increase in mean pulmonary arterial pressure (MPAP) and pulmonary vascular resistance (PVR) was seen in control pigs. Pretreatment with bosentan did not influence the first peak but markedly attenuated the second, more prolonged increase in MPAP and PVR. The second dose of bosentan completely restored these parameters to pre-LPS levels. The LPS-induced changes in mean arterial blood pressure, heart rate and systemic vascular resistance were similar in both groups, while cardiac output (CO) was significantly higher in the bosentan-treated group. The second bosentan dose increased CO and splenic and intestinal blood flow without further lowering of blood pressure. 3. Bosentan caused an increase of the basal arterial plasma levels of ET-1-like immunoreactivity (LI), from 16.8 +/- 1.3 pM to 49.6 +/- 10.0 pM (n = 6, P < 0.01). However, the rate of the increase of ET-1 levels during the LPS infusion was not affected by bosentan. Repeated administration of bosentan during LPS infusion caused an additional increase of ET-1-LI levels. Neither the basal levels of Big ET-LI nor the LPS induced 8 fold increase in Big ET-LI were changed by bosentan. The level of preproET-1 mRNA in the lung was increased about 3 fold after 4.5 h of LPS treatment. This elevation was not influenced by bosentan. 4. From these studies using bosentan, a non-peptide, selective and mixed ET-receptor antagonist, we conclude that during LPS-induced shock bosentan can abolish the late phase pulmonary hypertension and improve cardiac output as well as increase blood flow to the splenic and intestinal vascular beds without causing a further decrease in mean arterial blood pressure. Further investigations in the clinical setting are needed to evaluate the use of ET-receptor antagonists, such as bosentan, in treatment of septic shock.