1. We have developed a novel model of allergen-induced eosinophil into mouse air-pouches following sensitization and challenge with ovalbumin (Ova). This model was used to investigate the mechanism(s) underlying the anti-inflammatory action of the glucocorticoid hormone dexamethasone (Dex). 2. Injection of 10 micrograms Ova into 6-day-old dorsal air-pouches of mice sensitized to the same antigen provoked an intense cell accumulation as early as 6 h post-challenge (0.08 +/- 0.03 and 4.0 +/- 1.0 x 10(5) leucocytes in saline and Ova-treated air-pouches, respectively), maximal at 24 h (0.02 +/- 0.01 and 6.0 +/- 0.8 x 10(5) leucocytes in saline and Ova-treated air-pouches, respectively) and persisted up to 48 h. At the 24 h time-point, the cellular infiltrate consisted of 37% eosinophils, 18% neutrophils and 45% mononuclear cells, as assessed by histological examination. The same ratio of eosinophil/neutrophil was obtained by fluorescence-activated cell sorting (FACS) analysis, since 72% of the polymorphonuclear (PMN) population was positive for very-late antigen-4 (VLA-4) expression. 3. Subcutaneous (s.c.) administration of Dex (50 or 100 micrograms per mouse, -1 h) inhibited eosinophil accumulation into Ova challenged air-pouches by about 70% (P < 0.05) and 75% (P < 0.05), respectively, when compared to controls. Cell accumulation measured at 48 h after Ova injection was also significantly reduced (-75%) by Dex administration at the 24 h time-point (n = 12, P < 0.05). 4. Eosinophil numbers in the bone marrow and blood were quantitated. We found that the sensitization protocol induced a 3 fold increase in eosinophil numbers in the bone marrow (naive mice: 2.7 +/- 0.3 x 10(5); sensitized mice: 8.7 +/- 1.7 x 10(5), P < 0.05) and blood (naive mice: 0.5 +/- 0.2 x 10(5); sensitized mice: 1.5 +/- 0.3 x 10(5), P < 0.01). However, 24 h following Ova challenge, the eosinophil numbers in the bone marrow had dropped (3.7 +/- 0.8 x 10(5) with no change in the circulating pool, suggesting an equilibrium within the eosinophil pools had been reached. 5. Dex administration provoked a profound eosinopaenia in the blood of naive (5.2 +/- 1.5 to 0.9 +/- 0.6 x 10(4)) and sensitized mice (1.5 +/- 0.3 to 0.08 +/- 0.02 x 10(5)) at 4 h. This effect was reversed within 24 h. Dex also inhibited the release of eosinophils from the bone marrow in response to Ova challenge. 6. We show for the first time that express the steroid-inducible protein lipocortin 1 (LC1). FACS analysis of eosinophils emigrated into the Ova challenged air-pouches revealed detectable LC1-like immunoreactivity (373 x 10(3)). These data were also substantiated by LC1 detection in circulating eosinophils of interleukin-5 transgenic mice (strain: CBA/Ca). However, s.c. injection of Dex (50 micrograms) did not alter LC1 levels in blood eosinophils, such that 235 +/- 21 x 10(3) LC1-like molecules per cell were measured after vehicle treatment (n = 5), and 224 +/- 8 x 10(3) LC1-like molecules per cell were associated with this cell type 1 h after steroid treatment (n = 5, not significant). Finally, resident eosinophils (in the pleural cavity) were found to have much higher LC1 levels than that found in the blood circulation (2 fold increase, P < 0.05). 7. Passive immunization of mice against LC1 with a validated antiserum (termed LCS3) and protocol failed to modify the anti-migratory activity exerted by Dex towards eosinophil extravasation into Ova-challenged air-pouches. The steroid (50 micrograms s.c., -1 h) produced a similar degree of inhibition of eosinophil accumulation both in control animals (treated with a non-immune sheep serum) the LCS3-treated mice (-56% and 59%, respectively, n = 15-21, not significant). 8. In conclusion, the air-pouch provides a novel and convenient cavity to study allergen-induced cell recruitment which is sensitive to glucocorticoid hormone treatment. The effect of Dex on eosinophil distribution in these experimental conditions has been studied in detail and