Effects of glucocorticoids on eosinophil colony growth.

The in vivo and in vitro effects of glucocorticoids on eosinophilopoiesis were examined with soft agar cultures of bone marrow and peripheral blood cells. Prednisone, 10 mg four times daily for three days, administered to normal volunteers, caused a significant drop in circulating eosinophil numbers (p less than 0.005) but did not decrease eosinophil colony numbers in cultures of bone marrow or peripheral blood. A patient with peripheral blood eosinophilia demonstrated a larger percentage decrease in mean eosinophil colony numbers (50%) after prednisone than did any normal volunteer. Incubation of bone marrow cells for 1 hour with either high concentrations of hydrocortisone, up to 3.3 X 10(-4) mol/L, or with postinfusion plasma from volunteers administered intravenous hydrocortisone, plasma levels to 1.6 X 10(-5) mol/L, did not decrease the numbers of eosinophil colonies. At a concentration of 3.3 X 10(-6) mol/L of hydrocortisone, there was a slight but statistically significant stimulation of eosinophil colony numbers. In contrast, incubation with high concentrations of dexamethasone, 3.3 X 10(-4) mol/L or 3.3 X 10(-6) mol/L, significantly reduced eosinophil colony numbers. Prednisone caused a significant reduction in plasma levels of Charcot-Leyden crystal protein but not of eosinophil granule major basic protein. The results indicate that soft agar assay of eosinophil colony growth by blood or bone marrow cells cannot be used to model the in vivo eosinopenic effect of glucocorticoids, levels of dexamethasone in excess of levels commonly administered in clinical practice are required to inhibit eosinophilopoiesis in vitro, and patients with peripheral blood eosinophilia may be more susceptible to the eosinopenic effects of glucocorticoids than normal subjects.