We have tested the hypothesis that the beneficial effects of corticosteroids in asthma may result from reduction in the number of inflammatory cells infiltrating the bronchial mucosa with inhibition of cytokine gene expression. A randomized parallel group study was performed in 18 moderately severe asthmatic patients in whom an elective trial of corticosteroid treatment was indicated. Fiberoptic bronchoscopy was performed and bronchial biopsies taken from segmental carinae before and after 2 wk treatment with prednisolone (0.6 mg/kg/d) or matched placebo tablets. Immunohistology was performed on 6-microns cryostat sections using monoclonal antibodies. The number of cells expressing cytokine messenger RNA (mRNA) was assessed by in situ hybridization using S35-labeled riboprobes. When prednisolone- and placebo-treated groups were compared there was a decrease in airway methacholine responsiveness (p < 0.01) and an increase in FEV1 (p < 0.05) after prednisolone. This was accompanied by a reduction in CD3+ T lymphocytes (p < 0.05), "activated" EG2+ eosinophils (p < 0.02), and tryptase-only (mucosal-type) MCT cells (p < 0.02) but not MCTC (tryptase+chymase positive) cells in prednisolone-treated patients. In prednisolone-treated patients there was also a reduction in the number of cells expressing mRNA for interleukin-4 (IL-4, p < 0.01), and interleukin-5 (IL-5, p < 0.03) and an increase in cells expressing mRNA for interferon-gamma (IFN-gamma) (p < 0.01). These results support the view that corticosteroid treatment in asthma may act by modulation of cytokine expression with consequent inhibition of the local bronchial inflammatory infiltrate and tissue eosinophilia.