Evidence for the involvement of nitric oxide in A2B receptor-mediated vasorelaxation of mouse aorta.

We have investigated the role of adenosine and its analogs on vasorelaxation of mouse aorta in intact endothelium with rank order of potency as follows: 5'-N-ethylcarboxamidoadenosine (NECA) > 2-chloroadenosine > adenosine >> CGS-21680, which is consistent with the profile of A(2B)-adenosine receptor (A(2B)AR). In endothelium-intact tissues, acetylcholine produced relaxation ranging from 65 to 80% in phenylephrine (PE, 10(-7) M)-precontracted mouse aorta, whereas no relaxation was observed in endothelium-denuded tissues. The A(2B)AR antagonist alloxazine (10(-5) M) shifted concentration-response curve for NECA (EC(50) = 0.005 x 10(-5) M) to the right with an EC(50) of 2.8 x 10(-5) M, demonstrating that this relaxation is partially dependent on functional endothelium mediated predominantly via A(2B)AR in this tissue. This conclusion was further supported by the following findings: 1) in the endothelium-intact mouse aorta, the EC(50) values for NECA and adenosine were found to be 0.05 and 1.99 x 10(-4) M, respectively; however, in denuded endothelium, these values were 0.098 and 3.55 x 10(-4) M, respectively; 2) NECA-induced relaxation was significantly blocked by N(G)-nitro-l-arginine methyl ester (l-NAME; 10(-4) M) in endothelium-intact tissues, which was reversed by pretreatment with l-arginine (10(-4) M), whereas no significant inhibition was found in endothelium-denuded tissues; 3) total nitrites and nitrates (NOx) in intact endothelium with l-NAME (10(-4) M) alone and in combination with l-arginine were 59% (P < 0.05) and 96%, respectively, in comparison with control (PE + NECA); and 4) endothelial nitric oxide synthase gene expression was found to be 67% (P < 0.05) less in endothelium-denuded as opposed to endothelium-intact mouse aorta. Thus these data demonstrate that adenosine-mediated vasorelaxation is partially dependent on A(2B)AR in mouse aorta.