We previously reported that chronic inhibition of NO synthase (NOS) in dogs leads to an upregulation of the cyclooxygenase (COX) pathway in the endothelium of the coronary artery after stimulation by bradykinin (BK) in vitro. The present experiments were designed to identify the nature of the COX isoform involved in this phenomenon. Rings of circumflex (LCX) and left anterior descending (LAD) coronary arteries were isolated from six control dogs and six dogs treated with the NOS-inhibitor, N omega-nitro-L-arginine (L-NNA, 30 mg/kg/d, i.v., during 7 days). Concentration-response curves to BK in U46619-contracted rings from LCX coronary arteries were constructed in the presence and absence of another NOS inhibitor (NG-monomethyl-L-arginine, L-NMMA), of selective inhibitors of the inducible isoform of COX (NS-398 and L-745,337) and of a non selective inhibitor of the inducible and constitutive isoforms of COX (indomethacin). Finally, measurements of 6-keto-prostaglandin F1 alpha, the stable metabolite of prostacyclin, were performed in the incubation medium by enzymo-immuno-assay on rings of isolated LAD coronary arteries in the presence and absence of the same inhibitors of COX, before and after stimulation by BK. In rings taken from control dogs, BK evoked a concentration-dependent relaxation (Emax: 115 +/- 10%; EC50: 8 +/- 4 nM). In the presence of L-NMMA, the concentration-relaxation curve to BK was significantly shifted to the right (Emax: 77 +/- 8%; EC50: 43 +/- 22 nM, P < 0.05). Addition of NS-398, L-745,337 and indomethacin to L-NMMA did not further modify the concentration-relaxation curve to BK. After chronic inhibition of NOS, the concentration-relaxation curve to BK was similar to that observed in rings taken from control dogs in the presence of L-NMMA (Emax: 75 +/- 5%; EC50: 69 +/- 36 nM). Addition of L-NMMA, alone or in combination with NS-398 or L-745,337 did not significantly modify this concentration-relaxation curve to BK. In contrast, the L-NMMA-indomethacin combination blunted the BK-induced relaxation of the coronary artery (Emax: 28 +/- 10%, P < 0.01). Basal release of prostacyclin was not different in rings taken from control and L-NNA treated dogs (56 +/- 16 vs 58 +/- 15 pg.mm-2). BK significantly increased this release but the increment was twofold greater in rings taken from L-NNA treated dogs than in rings taken from control dogs (P < 0.05). In rings taken from control and L-NNA treated dogs, the BK-stimulated production of prostacyclin observed in the presence of the solvent was not significantly modified by L-NMMA or the L-NMMA-L-745,337 combination. In contrast, the L-NMMA-indomethacin combination as well as endothelium removal completely suppressed the BK-stimulated production of prostacyclin. These findings demonstrate that in dogs submitted to chronic inhibition of NO synthesis (1) the residual relaxation to BK of canine isolated coronary arteries is mainly due to production of prostacyclin of endothelial origin, and (2) the enhancement of prostacyclin production by these vessels is ...

No abstract