1 In this study we compared the ability of superoxide anion to destroy the relaxant activity of basal and acetylcholine (ACh)-stimulated activity of NO in isolated rings of rat aorta. 2 Superoxide dismutase (SOD, 1-300 u ml-') induced a concentration-dependent relaxation of phenylephrine (PE)-induced tone in endothelium-containing rings which was blocked by NG-nitro-Larginine (L-NOARG, 30 gM), but had no effect on endothelium-denuded rings. It was likely therefore that the relaxant action of SOD resulted from protection of basally produced NO from destruction by superoxide anion, generated either within the tissue or in the oxygenated Krebs solution. 3 In contrast, a concentration of SOD (50 u ml-') which produced almost maximal enhancement of basal NO activity, had no effect on ACh (10 nM-3 gM)-induced relaxation.4 In the presence of catalase (3000 u ml-') to prevent the actions of hydrogen peroxide, superoxide anion generation using hypoxanthine (HX, 0.1 mM)/xanthine oxidase (XO, 16 mu ml-') produced an augmentation of PE-induced tone in endothelium-containing but not endothelium-denuded rings. This was likely to have resulted from removal of the tonic vasodilator action of basally-produced NO by superoxide anion, since it was blocked in tissues treated with SOD (250 u ml-'), N0-monomethyl-Larginine (L-NMMA, 30 gM) or L-NOARG (30 gM). Pyrogallol (0.1 mM) had a similar action to HX/XO, but produced an additional augmentation of tone by an endothelium-independent mechanism. 5 In contrast to their ability to destroy almost completely the basal activity of NO, HX (0.1 mM)/XO (16 mu ml-') and pyrogallol (0.1 mM) had no effect on ACh-induced relaxation at any concentration. An increase in the concentration of HX to 1 mM or pyrogallol to 0.3 mM did, however, lead to a profound decrease in the magnitude and time course of ACh-induced relaxation at all concentrations. 6 Treatment with diethyldithiocarbamate (DETCA, 0.1 mM, 1 h) to inhibit endogenous Cu-Zn SOD, augmented PE-induced tone in endothelium-containing rings and abolished the ability of HX (0.1 mM)/ XO (16 mu ml-') and L-NMMA (30 gM) to augment tone. It was likely that DETCA had led to the destruction of basal NO activity by increasing superoxide anion levels since its actions were reversed by exogenous SOD (10-300 u ml-'). 7 In contrast to its ability to destroy basal activity of NO completely, DETCA (0.1 mM) produced only a slight blockade of ACh-induced relaxation. However, if these tissues were subsequently treated with concentrations of HX (0.1 mM)/XO (16 mu ml-') or pyrogallol (0.1 mM), which had no effect by themselves on ACh-induced relaxation, a profound blockade was seen and this was reversed completely with SOD (250 u ml-'). 8 The data suggest that basal activity of NO is more sensitive to inactivation by superoxide anion than ACh-stimulated activity and this probably results from differential protection by endogenous Cu-Zn SOD. It is possible therefore that endogenous SOD lowers superoxide anion levels to such an extent that only small amounts of ...

1 In this study we investigated the role of catalase in relaxation induced by hydroxylamine, sodium azide, glyceryl trinitrate and hydrogen peroxide in isolated rings of rat aorta. 2 Hydrogen peroxide (1 gM-1 mM)-induced concentration-dependent relaxation of phenylephrine (PE)-induced tone in endothelium-containing rings. In endothelium-denuded rings, however, higher concentrations (30 gM -1 mM) of hydrogen peroxide were required to produce relaxation. The endothelium-dependent component of hydrogen peroxide-induced relaxation was abolished following pretreatment with NG-nitro-L-arginine methyl ester (L-NAME, 30 MM). L-NAME (30 gM) had no effect, however, on hydrogen peroxide-induced relaxation in endothelium-denuded rings. 3 Pretreatment of endothelium-denuded rings with catalase (1000 u ml-') blocked relaxation induced by hydrogen peroxide (10 gM -1 mM). The ability of catalase to inhibit hydrogen peroxide-induced relaxation was partially blocked following incubation with 3-amino-1,2,4-triazole (AT, 50 mM) for 30 min and completely blocked at 90 min. 4 Pretreatment of endothelium-denuded rings with methylene blue (MeB, 30 Mm) inhibited relaxation induced by hydrogen peroxide (10 MM-I mM), sodium azide (1 -300 nM), hydroxylamine (1 -300 nM) and glyceryl trinitrate (1-100 nM) suggesting that each acted by stimulation of soluble guanylate cyclase. 5 Pretreatment of endothelium-denuded rings with AT (1-50 mM, 90 min) to inhibit endogenous catalase blocked relaxation induced by sodium azide (1-300 nM) and hydroxylamine (1-300 nM) but had no effect on relaxation induced by hydrogen peroxide (10 MM-I mM) or glyceryl trinitrate (1 -100 nM).6 In a cell-free system, incubation of sodium azide (10 MM-3 mM) and hydroxylamine (10 MM-30 mM) but not glyceryl trinitrate (10 MM-I mM) with catalase (1000 u ml-') in the presence of hydrogen peroxide (1 mM) led to production of nitrite, a major breakdown product of nitric oxide. AT (1-100 mM) inhibited, in a concentration-dependent manner, the formation of nitrite from azide in the presence of hydrogen peroxide. 7 These data suggest that metabolism by catalase plays an important role in the relaxation induced by hydroxylamine and sodium azide in isolated rings of rat aorta. Relaxation appears to be due to formation of nitric oxide and activation of soluble guanylate cyclase. In contrast, metabolism by catalase does not appear to be involved in the relaxant actions of hydrogen peroxide or glyceryl trinitrate.

1 In this study the impairment induced by hydrogen peroxide of vascular reactivity and the role of endogenous catalase in protection against this impairment was assessed in isolated rings of rat aorta. 2 Incubation with hydrogen peroxide at 1 mM, but not at 0.1 mM, for 15, 30 or 60 min followed by washout depressed, in a time-dependent manner, the subsequent ability of endothelium-containing and endothelium-denuded rings to contract to phenylephrine. 3 Incubation with 3-amino-1,2,4-triazole (50 mM, 90 min, followed by washout) to inhibit endogenous catalase had no e ect by itself on subsequent phenylephrine-induced contraction. However, pretreatment with 3-amino-1,2,4-triazole did lead to a profound enhancement of the ability of hydrogen peroxide (1 mM, present for the ®nal 30 min of the 90 min incubation, followed by washout) to depress phenylephrine-induced contraction in both endothelium-containing and endothelium-denuded rings. 4 Incubation with hydrogen peroxide at 1 mM, but not at 0.1 mM, for 15, 30 or 60 min followed by washout inhibited, in a time-dependent manner, the subsequent ability of acetylcholine (10 nM ± 3 mM) to induce endothelium-dependent relaxation. Furthermore, incubation with hydrogen peroxide 1 mM (30 min, followed by washout) also inhibited relaxation induced by glyceryl trinitrate (1 ± 100 nM) or isoprenaline (10 nM ± 3 mM) in endothelium-denuded rings. 5 Incubation with 3-amino-1,2,4-triazole (50 mM, 90 min, followed by washout) had no e ect by itself on relaxation induced by acetylcholine, glyceryl trinitrate or isoprenaline. In contrast, pretreatment with 3-amino-1,2,4-triazole led to profound enhancement of the ability of hydrogen peroxide (1 mM, present for ®nal 30 min of the 90 min incubation) to block relaxation to acetylcholine, glyceryl trinitrate or isoprenaline. 6 On the basis of the actions of 3-amino-1,2,4-triazole, it is likely that endogenous catalase plays an important role in the protection of vascular reactivity of rat aorta against oxidant damage by high (1 mM) but not lower (0.1 mM) concentrations of hydrogen peroxide. The data are consistent with the promotion of non-selective damage to the vascular smooth muscle cells by hydrogen peroxide, but endothelial damage may also be sustained.