1 IgGI-mediated anaphylactic bronchoconstriction was elicited by intravenous administration of antigen to guinea-pig 2 days after passive sensitization with IgGl-rich serum, and this response was not affected by heating the serum (at 56°C, for 4 h). IgE-mediated bronchoconstriction, provoked 14 days after passive sensitization with IgE-rich serum, was completely abolished by the heating of the serum. 2 S-1452 (10 mg kg-', p.o.), a selective thromboxane (Tx) A2 antagonist, significantly but incompletely suppressed the IgGl-mediated bronchoconstriction, but did not affect the IgE-mediated one, while diphenhydramine (5 mg kg'-, i.v.), a histamine antagonist, almost completely inhibited both IgGI-and IgE-mediated bronchoconstriction.3 Pretreatment with propranolol (1 mg kg-', i.v.), a P-adrenergic blocker, in addition to diphenhydramine, caused a long-lasting bronchoconstriction following antigen challenge in both animal models. This histamine-independent bronchoconstriction was markedly suppressed by S-1452 at a low dose of 0.1 mg kg-'. 4 A significant increase in bronchial responsiveness to i.v. acetylcholine (ACh), compared to the prechallenge value, occurred as early as 3 min and persisted for 24 h after antigen challenge in the IgGI model, but was not observed in the IgE model. S-1452 (10mg kg-', p.o.) inhibited the IgGI-mediated bronchial hyperresponsiveness, as assessed 60 min after antigen challenge. 5 A marked elevation of TxB2 levels was observed in bronchoalveolar lavage fluid (BALF) 3 min after antigen challenge in the IgGl model, while levels were not changed in the IgE model. In contrast, the plasma TxB2 level assessed 1 min after antigen challenge was increased in both the IgGI and IgE models. 6 The results indicate that the inhibition of IgGl-but not IgE-mediated bronchoconstriction by higher doses of S-1452 may result from the suppression of increased bronchial responsiveness to allergic mediators such as histamine, which is probably due to TxA2 generated in the airway lumen rather than in plasma. In both the IgGI and IgE models, plasma TxA2 appeared to contribute directly to the bronchoconstriction, its action being almost completely masked by histamine-mediated bronchoconstriction.