Several clinical reports have shown changes in plasma Se concentration with corticosteroid treatments, but the results have been inconsistent. Few experimental studies have been done on this subject. In the present study the effect of dexamethasone (DEX) treatment on Se concentrations and activities of Se-dependent glutathione peroxidase (EC 18.104.22.168; SeGPx) were examined in adult male ICR mice. In the first experiment, DEX was given via drinking water containing 5 or 50 mg DEX/I. At 1 or 3 weeks of DEX treatment, mice were dissected and the Se concentrations as well as SeGPx activities in various tissues, including plasma, were determined. At 1 week the DEX-treated groups had significantly lower hepatic Se concentrations and significantly higher plasma and cerebral concentrations than the control group. The DEX-treated groups showed lower SeGPx activities in the hepatic cytosol and higher SeGPx activities in the plasma than the saline (9 g NaCl/l)-treated group, in parallel with the changes in Se concentrations. At 3 weeks, neither hepatic nor plasma Se concentrations showed a significant change. In the second experiment, mice were injected subcutaneously with DEX and, thereafter, mice were food-deprived. The DEX-injected groups had higher plasma Se concentrations. A similar finding was obtained also when the DEX- or saline-injected mice were not food-deprived. Thus, the difference between the DEX-treated and control groups was possibly caused by redistribution of tissue Se. These results suggested that the effects of DEX on Se concentrations were tissue dependent and that the higher plasma Se observed in DEX-treated groups might be explained by the release of tissue Se into plasma as plasma SeGPx.