1. Exposure of smooth muscle cells (SMC) to glyceryl trinitrate (GTN, 75-600 microM) for 30 min led to a concentration-dependent increase in nitrite (NO2-), one of the breakdown products of nitric oxide (NO). This was not affected by 30 min pretreatment of the cells with 0.5 mM of sulphobromophthalein (SBP) an inhibitor of glutathione-S-transferase (GST), by metyrapone or SKF-525A inhibitors of cytochrome P450. These experiments were confirmed by organ bath studies using rabbit aortic strips denuded of endothelium and contracted with phenylephrine. Thus, a 30 min incubation of the strips with 0.5 mM SPB, metyrapone or SKF-525A did not affect the relaxations in response to GTN (10(-10)-10(-6) M). 2. Potentiation of the anti-platelet effect of GTN (44 microM) by endothelial cells (EC, 40 x 10(3) cells) was not affected by prior incubation of EC with SBP, metyrapone or SKF-525A (all at 0.5 mM). 3. Potentiation of the antiplatelet activity of GTN (11-352 microM) by small numbers of SMC (24 x 10(3) cells) or EC (40 x 10(3) cells) treated with indomethacin (10 microM) was attenuated when the SMC or EC were treated in culture with a high concentration of GTN (600 microM) for 18 h beforehand (referred to as 'tolerant' cells). In addition, tolerant SMC produced far less NO2- than non-tolerant SMC. 4. Exposure of non-tolerant SMC or EC (10(5) cells) to GTN (200 microM) for 3 min increased (3-4 fold) the levels of guanosine 3':5'-cyclic monophosphate (cyclic GMP). This increase was much less (< I fold) in the tolerant SMC or EC (105 cells). The basal levels of cyclic GMP were similar in normal or tolerant SMC or EC. Sodium nitroprusside (80 JAM) or atrial natriuretic factor (ANF, I0- M) increased the levels of cyclic GMP in normal or tolerant SMC or EC to the same extent.5 The anti-platelet effects of GTN (44 JM) were potentiated by the sulphydryl donor N-acetylcysteine(NAC, 0.5mM). Incubation of GTN (150-1200fJM) for 30min with NAC (0.1-1mM) led to aconcentration-dependent increase in N02- formation. The reduced ability of tolerant SMC or EC to potentiate the anti-platelet activity of GTN was restored by NAC (0.5 mM). These anti-aggregatory effects were abolished by concurrent co-incubation with oxyhaemoglobin (10 JM) indicating that they were due to NO release.6 Thus, in SMC or EC, metabolism of GTN to NO does not depend on glutathione-S-transferase or the cytochrome P450 system. Furthermore, when compared to normal cells, tolerant SMC or EC metabolize GTN to NO less effectively as assessed by the reduced capacity to potentiate the antiplatelet effects of GTN, to release NO2- and to increase the level of cyclic GMP. This decrease in NO formation shows that tolerance to GTN is mainly due to impaired biotransformation of GTN to NO. NAC, by directly forming NO from GTN, compensates for this failing mechanism.