We recently demonstrated that adenosine increases the synthesis and release of nitric oxide (NO) from aortic endothelium via A₁ and A_2A receptors, that A₁– but not A_2A-evoked release requires a product of the cyclo-oxygenase (COX) pathway, and that prostacyclin (PGI₂) can also evoke NO release. Further we provided direct evidence that adenosine increases the synthesis of PGI₂ by aortic endothelium (Ray et al. 2002). These results suggest an interaction between the adenosine, NO and COX pathways for vasodilatation.

In the present study on aortae taken from male Wistar rats killed by halothane overdose and cervical dislocation, NO release evoked by 1 mM adenosine and measured with an NO-sensitive electrode (see Ray et al. 2002), was attenuated by the adenyl cyclase (AC) inhibitor 5Ô-dideoxyadenosine (5 µM) in the presence of either the A₁ receptor antagonist DPCPX (100 nM), from 6.75 ± 1.04 nM (mean ± S.E.M., n = 8) to 1.31 ± 0.71 nM NO*, or the A_2A adenosine receptor antagonist ZM241385, from 14.31 ± 2.92 nM to 0 nM NO* (*P < 0.05, n = 6). The phospholipase C (PLC) inhibitor U73122 (1 µM) had no effect on NO release to adenosine after 10 or 30 min (n = 7). However, the phospholipase A₂ (PLA₂) inhibitor, AACOCF₃ (10 µM), reduced the NO response to 1 mM adenosine, from 66.38 ± 10.16 nM to 34.16 ± 6.47 nM NO*** (***P < 0.0001) after 10 min (n = 6). The K_ATP channel inhibitor glibenclamide attenuated the NO response to 1 mM adenosine from 129.76 ± 9.23 nM to 76.82 ± 5.80 nM NO*** (n = 8). However, in the presence of DPCPX, glibenclamide had no further effect on the response to adenosine (n = 10), whereas in the presence of ZM241385, glibenclamide attenuated the response to adenosine from 48.85 ± 13.67 nM to 11.16 ± 5.43 nM NO*** (n = 7). All statistical analysis was carried out by ANOVA for repeated measures with Fisher’s post-hoc test.

These results suggest that (i) adenosine-stimulated release of NO evoked via either A₁ or A_2A receptors requires an increase in intracellular cyclic AMP (cAMP), (ii) stimulation of PLA₂, but not PLC, is required and (iii) the ability of A₁ receptor stimulation to evoke NO release depends upon activation of K_ATP channels. We propose that adenosine acting at A₁ receptors stimulates K_ATP channels and that the resulting K⁺ efflux leads to a Ca²⁺ influx which stimulates PLA₂. This leads to synthesis and release of PGI₂ which acts on the endothelial cells, stimulating AC, increasing intracellular cAMP and triggering the cascade of phosphorylation, which leads to the activation of NO synthase and NO release.