To study the effects of short term low dose prednisone administration on serum lipids and lipoproteins we measured the concentration and composition of serum lipoproteins; serum apoproteins (apo) A-I, A-II, and B; and plasma lipolytic enzymes before and during prednisone administration (30 mg/day for 7 days) in eight normal men. We also measured insulin binding to adipocytes. Serum high density lipoprotein (HDL) cholesterol increased significantly after 2 days of prednisone administration; the maximal increase was 27% (P less than 0.01 after 5 days). The rise of HDL cholesterol was accounted for by that of HDL2 cholesterol. There were marked changes in the distribution of HDL particles; HDL2 increased, whereas HDL3 decreased. These changes were also apparent after 2 days of prednisone administration and were maximal at 5 days [mean, 1.58 +/- 0.12 (+/- SE) vs. 2.00 +/- 0.14 g/L (P less than 0.001) for HDL2; 1.82 +/- 0.11 vs. 1.61 +/- 0.06 g/L (P less than 0.05) for HDL3], and they were due to opposing changes in cholesterol, phospholipids, and proteins in the HDL subfractions. The change in HDL2 protein correlated inversely with that in HDL3 protein (r = -0.73; P less than 0.05). Notably, prednisone did not change the apo A-I concentration, but that of apo A-II decreased (0.32 +/- 0.02 vs. 0.27 +/- 0.01 g/L; P less than 0.05). Consequently, the lipid to protein ratio of HDL increased. Prednisone induced no significant changes in very low density or low density (LDL) lipoproteins. Adipose tissue LPL activity did not increase until after 7 days of prednisone intake (1.10 +/- 0.28 vs. 3.43 +/- 1.02 mumol FFA/g.h; P less than 0.05), and the same was true for muscle LPL (0.49 +/- 0.14 vs. 0.82 +/- 0.11 mumol FFA/g.h; n = 4; P = 0.06). Specific insulin binding was normal, but both basal and maximal insulin-stimulated glucose transport decreased significantly. In summary, prednisone induces changes in serum HDL which are characterized by redistribution of particles within HDL density toward less dense particles and a quantitative rise of lipids in the HDL2 fraction.