The effect of previous nitrate therapy on vascular responses to endothelin-1 (ET-1) and NO was investigated in human internal mammary artery (IMA) in vitro. Cumulative concentration–response curves to ET-1 were constructed in rings of IMA and the data grouped into IMA from patients given nitrates prior to the bypass graft operation (nitrate group) and IMA from patients who were not prescribed nitrates (control group). No significant differences were observed between the two groups, either in EC 50 value [ P >0.05; 3.5nM (2.4–5.3nM; 95% confidence interval) and 4.8 (2.2–10nM), nitrate and control groups respectively] or E max ( P >0.05; 78±7.5% and 85±9.5%, nitrate and control group respectively). No significant differences in concentration–response curves to the NO-donor diethylamine NONOate (DEA/NO) in rings of IMA pre-constricted with 10nM ET-1 were observed between control and nitrate groups [ P >0.05; EC 50 values 0.59 (0.21–1.7) µ M and 0.17 (0.03–0.87) µ M; E max 110±5.7% and 112±4.5%, nitrate and control groups respectively]. Concentration–response curves to DEA/NO constructed in normal coronary artery were not significantly different from those in coronary artery obtained from patients with ischaemic heart disease (IHD) [ P >0.05; E max 124±11% and 138±20%; EC 50 0.08 (0.02–0.30) µ M and 0.23 (0.02–24) µ M, normal and IHD respectively]. These data indicate that nitrate therapy does not induce long-term changes in the ET signalling pathway. Furthermore, the tolerance to nitrate therapy is likely to be because of impaired bio-transformation of the drug rather than reduced sensitivity of the media to NO. The similar responses to DEA/NO in normal and atherosclerotic coronary artery suggests that the reduced vasodilator responses in IHD is because of a dysfunctional endothelium and is not mediated by changes in the NO signalling pathway of the smooth muscle.