Factors mediating a regional difference in vasorelaxing effects of atrial natriuretic polypeptide (ANP) were investigated, using the canine pulmonary and renal arteries. Isolated preparation was suspended in an organ bath, and the isometric tension was recorded. ANP relaxed both arteries dose-dependently under precontraction with methoxamine (3 x 10(-5) M). The maximum relaxation was extremely different between the two arteries: 78.5% and 8.6% of maximum relaxations caused by papaverine in the pulmonary and renal arteries, respectively. Removal of the endothelium affected neither the relaxation of pulmonary artery nor that of renal artery. Methylthionine (10(-5) M) greatly diminished the maximum relaxation in the pulmonary artery, but not in the renal artery, indicating that some free radical species may mediate the relaxation of pulmonary arteries to ANP. Methoxamine-induced contractions were inhibited by isosorbide dinitrate (10 and 30 nM) in both pulmonary and renal arteries, in a noncompetitive manner. Nifedipine (10 and 30 nM) inhibited the contractions competitively in the renal artery, but not in the pulmonary artery. Even in the Ca++-free solution, ANP diminished the methoxamine-induced contractions of the pulmonary arterial strips more effectively than those of the renal arterial strips. From these results, we concluded that the regional difference in the ANP-induced relaxation between the pulmonary and renal arteries was derived, partly, from a difference in the mechanism of intracellular Ca++ mobilization.