1. The vasodilator properties of a novel S-nitrosated glyco-amino acid (RIG200) were investigated in isolated rat femoral arteries and compared with those of the parent S-nitrosothiol compound, S-nitroso-N-acetylpenicillamine (SNAP). 2. Spectrophotometric analysis revealed that 2.5 mM solutions of RIG200 decomposed more slowly (half-life (t1/2) = 216.2 +/- 26.7 min) than SNAP (t1/2 = 37.2 +/- 13.8 min) in Krebs buffer at 24 degrees C. Furthermore, the rate of decomposition of SNAP, but not of RIG200, was significantly reduced by the Cu(I) chelator, neocuproine. We concluded that the relative stability of RIG200 is due, at least in part, to its resistance to trace Cu(I)-catalyzed decomposition. Nitric oxide (NO) generation from SNAP and RIG200 was confirmed by use of an NO electrode. 3. Experiments to investigate the vasodilator effects of RIG200 were carried out on isolated femoral arteries taken from adult male Wistar rats (400-550 g). Lengths of artery (7-8 mm long) were cannulated, dissected free and perfused at constant flow rate (0.6 ml min-1) with Krebs buffer. Vessels were precontracted with phenylephrine (10.2 +/- 0.3 microM) and developed pressures of 91.8 +/- 4 mmHg, detected upstream by a differential pressure transducer. 4. Concentration-dependent vasodilator responses to bolus injections of SNAP or RIG200 (10 microliters; 10(-8)-10(-3) M) made into the perfusate of endothelium-intact vessels were transient, recovering the preinjection pressure in < 20 min. 5. Responses to equivalent bolus injections of SNAP in endothelium-denuded vessels were also transient but those in response to concentrations of RIG200 > 10(-5) M were sustained. Responses to 10(-3) M RIG200 were sustained for periods > 4 h. Sustained vasodilatation was reversed by the NO scavenger, ferrohaemoglobin (10 microM) but was unaffected by the NO synthase inhibitor, N omega-nitro-L-arginine methyl ester (200 microM), indicating involvement of NO from a source other than NO synthase. 6. We suggest that a possible explanation for the prolonged effect of RIG200 is retention of the compound by the vascular wall, facilitated by endothelial denudation. Slow decomposition of RIG200 in situ would release sufficient NO to maintain a 'vasodilator tone' which persists for more than 4 h. Selective retention by damaged vessels could have important therapeutic implications with regard to targeted delivery of NO, restoring protection to areas deprived of endogenous NO, whilst avoiding unwanted hypotension.