Bronchial hyperresponsiveness can be produced experimentally after inhalation of numerous nonimmunospecific stimuli; our objective was to determine whether acrolein, a component of cigarette smoke, could increase bronchial reactivity to intravenously administered acetylcholine in guinea pigs. Bronchial responsiveness was assessed twice before and 1, 2, 6, and 24 h after exposures to less than or equal to 0.01 (sham), 0.31, 0.67, 0.94, or 1.26 parts per million for 2 h (5-7 guinea pigs/group). To examine the possible relationships of responsiveness to inflammatory mediator release and cellular infiltration, bronchoalveolar lavage was performed in another 30 guinea pigs before (control) and 0, 1, 2, 6, or 24 h after exposures. Pulmonary resistance was increased immediately after exposure (5 min) and returned to control values within 30-60 min. Increased bronchial responsiveness was evident within 1 h and became maximal 2-4 h after exposure. The acetylcholine dose necessary to double resistance decreased from 104.2 +/- 7.3 to 79.6 +/- 15.9 at 1 h and was 32.5 +/- 7.9 at 2 h and 32.8 +/- 7.6 micrograms.kg-1 at 6 h. Increases in two eicosanoids, thromboxane B2 (from 167 +/- 21 to 314 +/- 77 pg/ml) and prostaglandin F2 alpha (from 98 +/- 20 to 285 +/- 62 pg/ml) occurred immediately after exposure, whereas an influx of neutrophils occurred 24 h later (from 2.2 +/- 1.2 to 11.3 +/- 3.6%). These temporal relationships suggest that neutrophil infiltration may be a sufficient but not a necessary condition for the onset of bronchial hyperresponsiveness and that injury to cells normally present in the lung are responsible for the mediators thought to influence bronchial responsiveness.