To determine the effects of exogenous insulin-like growth factor-I (IGF-I) and GH on IGF-binding proteins (IGFBP)-1, -2, and -3, six healthy nonobese adult volunteers underwent two 2-week periods of diet restriction (20 Cal/kg.day), and during the last 6 days of the first period received either IGF-I (12 micrograms/kg.h by iv infusion over 16 h) or GH (0.05 mg/kg.day by sc injection). During the second 2-week study period, the alternate hormone was given. IGFBP-1 and -2 concentrations were determined by specific RIA, and changes in IGFBP-3 were assessed by ligand blotting. Free IGF-I concentrations were measured by size-exclusion high pressure liquid chromatography, followed by RIA. Diet restriction alone did not affect either IGFBP-1 or -2 significantly. IGF-I treatment increased IGFBP-1 from 78 +/- 46 ng/mL (mean pretreatment) to 137 +/- 64 ng/mL (P less than 0.001; mean for the last 4 days of IGF-I). IGF-I also caused an increase in IGFBP-2 from 315 +/- 136 to 675 +/- 304 ng/mL (P less than 0.001). GH injections caused a modest decline in IGFBP-1 concentrations but had no effect on IGFBP-2 concentrations. By ligand blotting, both IGF-I and GH caused a modest increase in IGFBP-3 band intensity. In three subjects diet restriction alone caused a small decrease in IGFBP-3 hand intensity, and this was reversed by hormone treatment. Free IGF-I concentrations in serum were increased from 1.6% to 4.4% of the total IGF-I during IGF-I infusions. GH injections caused a smaller increase in free IGF-I concentrations. The results show significant increases in IGFBP-1 and -2 during IGF-I infusion. The change in IGFBP-3, while significant, is quantitatively less than that in experimental animals that have been given IGF-I while undergoing dietary restriction. The net effect of the changes in these three forms of IGFBPs is not sufficient to maintain a normal IGF-I-binding capacity in serum, because free IGF-I levels were increased disproportionately during the IGF-I infusions. Because hypoglycemia was noted in these subjects despite insulin suppression, these alterations in IGFBPs might have changed the tissue bioavailability of IGF-I and facilitated its hypoglycemic effects.