Pharmacological characterization of guanidinoethyldisulphide (GED), a novel inhibitor of nitric oxide synthase with selectivity towards the inducible isoform.

1. Guanidines, amidines, S-alkylisothioureas, and recently, mercaptoalkylguanidines have been described as inhibitors of the generation of nitric oxide (NO) from L-arginine by NO synthases (NOS). We have recently demonstrated that guanidinoethyldisulphide (GED), formed from the dimerisation of mercaptoethylguanidine (MEG), is a novel inhibitor of nitric oxide synthases. Here we describe the pharmacological properties of GED on purified NOS isoforms, various cultured cell types, vascular ring preparations, and in endotoxin shock. 2. GED potently inhibited NOS activity of purified inducible NOS (iNOS), endothelial NOS (ecNOS), and brain NOS (bNOS) enzymes with Ki values of 4.3, 18 and 25 microM, respectively. Thus, GED has a 4 fold selectivity for iNOS over ecNOS at the enzyme level. The inhibitory effect of GED on ecNOS and iNOS was competitive vs. L-arginine and non-competitive vs. tetrahydrobiopterin. 3. Murine J774 macrophages, rat aortic smooth muscle cells, murine lung epithelial cells, and human intestinal DLD-1 cells were stimulated with appropriate mixtures of pro-inflammatory cytokines or bacterial lipopolysaccharide to express iNOS. In these cells, GED potently inhibited nitrite formation (EC50 values: 11, 9, 1 and 30 microM, respectively). This suggests that uptake of GED may be cell type and species-dependent. The inhibitory effect of GED on nitrite production was independent of whether GED was given together with immunostimulation or 6 h afterwards, indicating that GED does not interfere with the process of iNOS induction. 4. GED caused relaxations in the precontracted vascular ring preparations (EC50: 20 microM). Part of this relaxation was endothelium-dependent, but was not blocked by methylene blue (100 microM), an inhibitor of soluble guanylyl cyclase. In precontracted rings, GED enhanced the acetylcholine-induced, endothelium-dependent relaxations at 10 microM and caused a slight inhibition of the relaxations at 100 microM. The vascular studies demonstrate that the inhibitory potency of GED on ecNOS in the ring preparations is considerably lower than its potency against iNOS in the cultured cells. These data suggest that the selectivity of GED towards iNOS may lie, in part, at the enzyme level, as well as differential uptake by cells expressing the various isoforms of NOS. 5. In a rat model of endotoxin shock in vivo, administration of GED, at 3 mg kg-1 bolus followed by 10 mg kg-1 h-1 infusion, starting at 90 min after bacterial lipopolysaccharide (LPS, 15 mg kg-1, i.v.), prevented the delayed fall in mean arterial blood pressure, prevented the development of the vascular hyporeactivity to noradrenaline of the thoracic aorta ex vivo and protected against the impairment of the endothelium-dependent relaxations associated with this model of endotoxaemia. The same bolus and infusion of the inhibitor did not alter blood pressure or ex vivo vascular reactivity in normal animals over 90 min. 6. Administration of GED (10 mg kg-1, i.p.) given at 2 h after LPS (120 mg kg-1, i.p.) and every 6 h thereafter caused a significant improvement in the survival rate in a lethal model of endotoxin shock in mice between 12 and 42 h. 7. In conclusion, we found that GED is a competitive inhibitor of iNOS activity. Its selectivity towards iNOS may lie both at the enzyme level and at the level of cell uptake. GED has beneficial effects in models of endotoxin shock that are driven by iNOS. GED or its derivatives may be useful tools in the experimental therapy of inflammatory conditions associated with NO overproduction due to iNOS expression.