We have previously observed that arginine vasopressin (AVP)-induced pulmonary vasodilation is attenuated by nitric oxide (NO) synthesis inhibition; however, blockade of the response is incomplete even at very high doses of the inhibitor. Thus it was hypothesized that the remaining vasodilation might be due to release of an endothelium-derived hyperpolarizing factor acting to open vascular smooth muscle K+ channels. Lungs were isolated from male Sprague-Dawley rats and perfused at constant flow with physiological saline solution containing 4% albumin. After equilibration, lungs were treated with either glibenclamide (50 mu M), Ba2+ (100 mu M), tetraethylammonium (10 mM), or the respective vehicle and were then constricted with the thromboxane mimetic U-46619. Upon development of a stable degree of vasoconstriction, AVP (2.5 x 10(-9) M) was administered and its vasodilator action noted. AVP caused an similar to 60% reversal of U-46619 vasoconstriction in control lungs, and this response was not affected by any of the K+ channel blockers. In contrast, administration of the NO synthesis inhibitor N-omega-nitro-L-arginine (L-NNA; 300 mu M) significantly attenuated AVP-induced dilation to similar to 25%. The addition of K+ channel blockers did not further diminish the vasodilatory response in L-NNA-treated lungs. In conclusion, these results suggest that ATP- and Ca2+-sensitive K+ channels are not involved in the pulmonary vasodilatory response to AVP.