1. Male Sprague-Dawley or Wistar rats were injected with bacterial lipopolysaccharide (LPS; 5 mg kg-1, i.p.) and killed after 1, 3, 6, 15, and 24 h. The brains, mesenteries, spleens, lungs, livers, kidneys, hearts, aortae and diaphragms were removed and frozen immediately. Control rats were injected with sterile saline and killed after 6 h. 2. The organs were homogenized in a semi-frozen state and NO synthase (NOS) activity measured in tissues from both LPS-treated and saline-treated groups by the ability of homogenates to convert [3H]-L-arginine to [3H]-L-citrulline in a NADPH-dependent manner. 3. The NOS activity in all organs taken from control animals was found to be calcium-dependent, with the highest activity being in the brain. After LPS-treatment an induced calcium-independent NOS was detected in all tissues tested, with the exception of the brain. The spleen, lung, mesentery and liver had the highest amounts of LPS-induced NOS activity. No induction of calcium-dependent NOS was detected. 4. Induction of NOS was maximum 6 h after administration of LPS and had returned to control levels in 24 h. 5. The constitutive NOS in brain and mesentery and the LPS-induced activities in the spleen, lung, liver and mesentery were inhibited by NG-monomethyl-L-arginine (L-NMMA) or NG-nitro-L-arginine methyl ester (L-NAME) according to concentration. The IC50 for L-NAME was 2.5 microM against the constitutive NOS from brain, and 20-25 microM against the inducible NOS. For L-NMMA the IC50 was 20-25 microM against either NOS isoform. 7. The vascular responses to endothelin-I (ET-1), the thromboxane A2-mimetic 11 alpha,9 alpha-epoxymethanoprostaglandin F2alpha (U46619), phenylephrine (PE) or 5-hydroxytryptamine (5-HT) were measured in the simultaneously perfused arterial and venous mesenteric vascular beds from both control and LPS-treated(6 h) rats. Vasoconstrictor responses to all agonists tested were unaffected by LPS treatment. In the presence of L-NAME (100 microM) vasoconstrictor responses were potentiated in both the arterial and venous portion of the mesenteric beds from both control and LPS-treated rats. The potentiation of responses to U46619 was significantly greater in beds from LPS-treated rats.8. Injection of LPS i.p. is associated with induction of NOS in all organs tested, except for the brain. In the mesentery this is not accompanied by a hyporesponsiveness to constrictor agents suggesting an increased sensitivity, particularly to U46619. This may explain the poor perfusion and tissue damage in the splanchnic circulation associated with sepsis.