We previously reported that NO2 and acrolein administered ex vivo to the lung altered the subsequent responsiveness of airway smooth muscle. The aim of this study was to determine the dose-response relationship for O-3 in both human isolated bronchi and rat tracheae and to investigate the mechanisms underlying O-3-induced airway responsiveness. Exposure to 1 ppm O-3 for 15 min significantly increased the maximal response to carbachol of rat tracheal rings to 149.6 +/- 5.4% of the reference response to acetylcholine (ACh) compared with that of unexposed rings (131.3 +/- 2.4%, n = 6, P < 0.05). The change in maximal airway responsiveness to carbachol, when plotted against the product of exposure concentration and exposure time to O-3, a surrogate for the dose, formed a bell-shaped curve. The peak of this dose-response curve was shifted to the right for human bronchi (50 ppm x min, n = 5) compared with that of rat tracheae (15 ppm x min, n = 6). In the rat trachea, responses to KCl were not altered by O-3, whereas those to 5- hydroxytryptamine hydrochloride (5-HT) were significantly increased. Finally, in the absence of external Ca2+, O-3 exposure still potentiated the maximal response to carbachol from 73.6 +/- 13.9 to 137.0 +/- 6.0% and that to 5-HT from 21.5 +/- 5.5 to 38.7 +/- 2.2% of the reference ACh response. These results indicate that O-3 alters the subsequent in vitro airway responsiveness depending on 1) the dose, 2) the nature of the agonist, and 3) the species investigated. Because in vitro exposure to O-3 increases responses to agonists that release intracellular Ca2+ and since this effect is maintained in Ca2+-free solution, the mechanism of O-3-induced increase in airway smooth muscle responsiveness is likely to involve an enhancement in intracellular Ca2+ release.