The effects of peroxynitrite on rat aorta: interaction with glucose and related substances.

Peroxynitrite (1-100 microM) induced a concentration-dependent relaxation of rat aortic rings; the logEC50 and maximum relaxation on endothelium-denuded rings were -5.31 +/- 0.03 and 105 +/- 5%, n = 6, respectively. The presence of the endothelium significantly impaired this relaxation (logEC50, -4.41 +/- 0.04; maximum relaxation, 71 +/- 4%; n = 6); an effect which was reversed by the inhibitor of nitric oxide synthase, N(G)-nitro-L-arginine methyl ester (L-NAME; 100 microM). Incubation with a high concentration of peroxynitrite (1 mM, 10 min followed by washout) had no effect on subsequent relaxation to acetylcholine (0.01-1 microM). It did, however, significantly depress subsequent contraction to phenylephrine (1-300 nM). This depression was dependent upon the presence of D-glucose in the Krebs solution, could be reversed by the inhibitor of soluble guanylate cyclase, methylene blue (10 microM) and reversed spontaneously after 2 h. When peroxynitrite (1 mM) was mixed with D-glucose (11 mM) and subsequently neutralised to remove unreacted peroxynitrite, a new more potent relaxant was formed. Despite this, the ability of peroxynitrite (1-100 microM) to produce relaxation of endothelium-denuded rings was similar in normal and glucose-free Krebs. Glycerol (22 mM), which like D-glucose is membrane permeant, also reacted with peroxynitrite (1 mM) to form a new more potent relaxant. L-cysteine (1 mM) had no effect by itself on the tone of aortic rings and when present in the tissue bath had no effect on the ability of peroxynitrite or neutralised peroxynitrite (1-100 microM) to produce relaxation. It did, however, potentiate the relaxant actions of the products formed from the reaction of peroxynitrite with D-glucose or glycerol. The membrane impermeant sugars, mannitol and sorbitol (each 11 mM) also reacted with peroxynitrite (1 mM), but expression of the vasorelaxant properties of their respective derivatives was seen only in the presence of L-cysteine (1 mM). Membrane permeance cannot, however, explain why peroxynitrite reacts with D-glucose and glycerol, but not mannitol or sorbitol to form products with intrinsic relaxant activity, as the product formed from the impermeant sugar, L-glucose (11 mM), also has intrinsic activity. The relaxant potency of this product was equipotent to that formed from D-glucose and was also potentiated by L-cysteine (1 mM). These result confirm that peroxynitrite can react with glucose and other compounds with alcohol functional groups to form vasorelaxant species. The relaxation induced when peroxynitrite is added to rat aortic rings is not, however, dependent upon this reaction since it occurs in glucose-free Krebs.