1. Zymosan is a wall component of the yeast Saccharomyces Cerevisiae. Injection of zymosan into experimental animals is known to produce an intense inflammatory response. Recent studies demonstrated that the zymosan-induced inflammatory response in vivo can be ameliorated by inhibitors of nitric oxide (NO) biosynthesis. The cytotoxic effects of NO are, in part, mediated by the oxidant preoxynitrite and subsequent activation of the nuclear enzyme poly (ADP-ribose) synthetase (PARS). In the present in vitro study, we have investigated the cellular mechanisms of vascular failure elicited by zymosan-activated plasma and the contribution of peroxynitrite production and activation of PARS to the changes. 2. Incubation of rat aortic smooth muscle cells with zymosan-activated plasma (ZAP) induced the production of nitrite, the breakdown product of NO, due to the expression of the inducible isoform of NO synthase (iNOS) over 6 24 h. In addition, ZAP triggered the production of peroxynitrite in these cells, as measured by the oxidation of the fluorescent dye dihydrorhodamine 123 and by nitrotyrosine Western blotting. 3. Incubation of the smooth muscle cells with ZAP induced DNA single strand breakage and PARS activation. These effects were reduced by inhibition of NOS with NG-methyl-L-arginine (L-NMA, 3 mM), and by glutathione (3 mM), a scavenger of peroxynitrite. The PARS inhibitor 3-aminobenzamide (1 mM) inhibited the ZAP-induced activation of PARS. 4. Incubation of thoracic aortae with ZAP in vitro caused a reduction of the contractions of the blood vessels to noradrenaline (vascular hyporeactivity) and elicited a reduced responsiveness to the endothelium-dependent vasodilator acetylcholine (endothelial dysfunction). 5. Preincubation of the thoracic aortae with L-NMA (1 mM), glutathione (3 mM) or by the PARS inhibitor 3-aminobenzamide (1 mM) prevented the development of vascular hyporeactivity in response to ZAP. Moreover, glutathione and 3-aminobenzamide treatment protected against the ZAP-induced development of endothelial dysfunction. The PARS-related loss of the vascular contractility was evident at 30 min after incubation in endothelium-intact, but not in endothelium-denuded vessels and also manifested at 6 h after incubation with ZAP in endothelium-denuded rings. The acute response is probably related, therefore, to peroxynitrite formation (involving the endothelial NO synthase), whereas the delayed response may be related to the expression of iNOS in the smooth muscle. 6. The data obtained suggest that zymosan-activated plasma causes vascular dysfunction by inducing the simultaneous formation of superoxide and NO. These radicals combine to form peroxynitrite, which, in turn causes DNA injury and PARS activation. The protective effect of 3-aminobenzamide demonstrates that PARS activation contributes both to the development of vascular hyporeactivity and endothelial dysfunction during the vascular failure induced by ZAP.