Differential effect of hydrocortisone on eosinophil and neutrophil proliferation

Abstract

Glucocorticosteroid therapy results in an increase in the number of circulating neutrophils and a decrease in the number of eosinophils. Utilizing the double layer soft agar technique, we examined the effect of physiologic to pharmacologic concentrations of hydrocortisone on the proliferation of human neutrophil progenitors and eosinophil progenitors from peripheral blood and bone marrow. When peripheral blood cultures were studied, eosinophil proliferation was inhibited in a dose-responsive fashion with 10(-8) - 10(-5) M hydrocortisone succinate, and comprised 49 +/- 4% of the colonies in control cultures and only 4 +/- 1% (P less than 0.01) at pharmacologic levels of hydrocortisone (10(-5) M). The number of neutrophil colonies, on the other hand, increased by 31% when 10(-5) M hydrocortisone was added to cultures. In order for corticosteroids to exert this effect, it was necessary to add them within 24 h of the initiation of culture. The effect of hydrocortisone on granulocyte proliferation could not be blocked by progesterone, a structurally analogous steroid. To determine whether hydrocortisone was acting directly on the progenitor cell or via an effector cell, its effect on modulating cell populations and stimulating-factor production was studied. Removal of E-rosetting cells and/or adherent cells did not affect the inhibition of eosinophil colony growth or the enhancement of neutrophil colony growth. Furthermore, addition of the potent inhibitor of T cell function, cyclosporin A, failed to affect eosinophil colony frequency, suggesting that inhibition of T cell function was an unlikely explanation for the observed hydrocortisone effect. Leukocyte conditioned media (LCM), derived from peripheral blood mononuclear cells incubated with hydrocortisone, was devoid of both neutrophil and eosinophil colony-stimulating activity, whereas a control LCM stimulated both neutrophil and eosinophil proliferation. The data suggest that the observed hydrocortisone effect on granulocyte colony formation is unlikely to be mediated by an intermediary, and that hydrocortisone acts directly on progenitor cells.