The authors have investigated whether the steroid-sparing effect of methotrexate (MTX) in severe orally glucocorticoid-insensitive asthmatics may be accounted for by the ability of this drug to increase the T-cell responsiveness sensitivity to dexamethasone in vitro. In addition the authors have investigated whether low-dose MTX treatment is associated with anti-inflammatory effects in peripheral blood and the bronchial mucosa. In eight patients with severe atopic asthma, using > or =15 mg x day(-1) prednisolone, the inhibitory effect of dexamethasone on mitogen stimulated peripheral blood mononuclear cells (PBMC) in vitro was tested before and after 8 weeks of uncontrolled treatment with MTX. Endobronchial biopsies were taken before and after MTX therapy in seven subsequent patients, and analysed using immunohistochemistry. In eight patients, serum was drawn for measuring levels of free interleukin (IL)-8. The in vitro sensitivity of PBMC to dexamethasone (at 1.6 x 10(-9) and 3.2 x 10(-10) mol x L(-1)) was significantly lower in the asthmatics before treatment when compared with the control subjects (p=0.03 and =0.001) but increased significantly after MTX treatment (p=0.04 and =0.02) to normal responsiveness. This was not associated with a decrease in peripheral blood T-cell numbers or activation. Except for a significant increase in the numbers of CD3+ (p=0.04), no significant numerical changes in activated T-cells, eosinophils, or mast cells were found (p>0.05). However, MTX treatment was associated with a significant fall in serum levels of free IL-8 (p=0.03). It is hypothesized that the steroid-sparing effect of methotrexate originates from increased sensitivity of lymphocytes to the inhibitory effects of glucocorticoids. The absence of an inhibitory effect on inflammatory cells in blood and mucosa suggests that this effect is achieved by modulating cell function rather than cell number.