Erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA) was shown to reverse the hypoxic pressor response (HPR) in the isolated, blood-perfused rat lung model. EHNA, an adenosine deaminase inhibitor, showed reversal of the HPR in a dose-dependent manner (EC50 = 129 +/- 30 microM). We found that the reversal of HPR by EHNA was not mediated by the adenosine receptors because the EHNA effect was not blocked by the adenosine receptor antagonist, 8-p-sulfophenyl-theophylline (67 microM; n = 6). Pretreatment with a cy-clic-3',5'-adenosine monophosphate (cAMP)-dependent protein kinase inhibitor, Rp-adenosine-3',5'-cyclic monophosphorothioate (0.5 mM; n = 4), blocked EHNA reversal of the HPR. As an alternative mechanism of action, EHNA inhibition of cyclic nucleotide phosphodiesterase(s) isozymes was studied in endothelium intact and denuded pulmonary arteries. Using anion-exchange chromatography the cyclic nucleotide phosphodiesterase (PDE) separated into predominantly PDE families 2 and a mixture of 3 and 4. DEAE fractions showing cAMP hydrolysis activated by 5 microM cyclic-3',5'-guanosine monophosphate (cGMP) had a Km for cAMP of 6.3 microM and an apparent Kact for cGMP of 1.4 microM. EHNA was shown to inhibit PDE2 competitively. In intact vessels, the IC50 for EHNA was 3.3 microM using 0.03 microM [3H]-cAMP substrate assayed in the presence of 2 microM cGMP and in denuded vessels 3.7 microM at 0.03 microM [3H]-cAMP substrate in the presence of 5 microM cGMP. Fractions in which cAMP hydrolysis was inhibited or not affected by 5 microM cGMP (PDE3 and 4, respectively) showed an IC50 of > 200 microM for EHNA. We conclude that reversal of the hypoxic pressor response by EHNA in the isolated, perfused rat lung model occurs with a mechanism involving in part inhibition of smooth muscle PDE2.