1 The role of nitric oxide (NO) derived from constitutive and inducible nitric oxide synthase (cNOS and iNOS) and its relationship to oxygen-derived free radicals and prostaglandins (PG) was investigated in a carrageenan-induced model of acute hindpaw inflammation. 2 The intraplantar injection of carrageenan elicited an inflammatory response that was characterized by a time-dependent increase in paw oedema, neutrophil infiltration, and increased levels of nitrite/nitrate (NO2-/NO3-) and prostaglandin E2(PGE2) in the paw exudate. 3 Paw oedema was maximal by 6 h and remained elevated for 10 h following carrageenan administration. The non-selective cNOS/iNOS inhibitors, NG-monomethyl-L-arginine (L-NMMA) and NG-nitro-L-arginine methyl ester (L-NAME) given intravenously (30-300 mg kg-1) 1 h before or after carrageenan administration, inhibited paw oedema at all time points. 4 The selective iNOS inhibitors, N-iminoethyl-L-lysine (L-NIL) or aminoguanidine (AG), failed to inhibit carrageenan-induced paw oedema during the first 4 h following carrageenan administration, but inhibited paw oedema at subsequent time points (from 5-10 h). iNOS mRNA was detected between 3 to 10 h following carrageenan administration using ribonuclease protection assays. iNOS protein was first detected 6 h and was maximal 10 h following carrageenan administration as shown by Western blot analysis. Administration of the iNOS inhibitors 5 h after carrageenan (a time point where iNOS was expressed) inhibited paw oedema at all subsequent time points. Infiltrating neutrophils were not the source of iNOS since pretreatment with colchicine (2 mg kg-1) suppressed neutrophil infiltration, but did not inhibit the iNOS mRNA expression or the elevated NO2-/NO3- levels in the paw exudate. 5 Inhibition of paw oedema by the NOS inhibitors was associated with attenuation of both the NO2-/NO3- and PGE2 levels in the paw exudate. These inhibitors also reduced the neutrophil infiltration at the site of inflammation. 6 Recombinant human Cu/Zn superoxide dismutase coupled to polyethyleneglycol (PEGrhSOD; 12 x 10(3) u kg-1), administered intravenously either 30 min prior to or 1 h after carrageenan injection, inhibited paw oedema and neutrophil infiltration, but had no effect on NO2-/NO3- or PGE2 production in the paw exudate. The administration of catalase (40 x 10(3) u kg-1), given intraperitoneally 30 min before carrageenan administration, had no effect on paw oedema. Treatment with desferrioxamine (300 mg kg-1), given subcutaneously 1 h before carrageenan, inhibited paw oedema during the first 2 h after carrageenan administration, but not at later times. 7 These results suggest that the NO produced by cNOS is involved in the development of inflammation at early time points following carrageenan administration and that NO produced by iNOS is involved in the maintenance of the inflammatory response at later time points. The potential interactions of NO with superoxide anion and PG is discussed.