Urinary insulin-like growth factors (IGF) and IGF-binding proteins in normal subjects, growth hormone deficiency, and renal disease

Recent studies in our laboratories have shown that urine from healthy adults contains immunoreactive and intact insulin-like growth factor-binding protein-3 (IGFBP-3). The aim of this study was to assess urinary IGF-I, IGF-II, and IGFBP-3 in a cross-sectional study of healthy subjects, as well as characterize urinary IGFBPs (uIGFBPs) in patients with GH deficiency (GHD) and renal disease, such as, Alport syndrome, immunoglobulin A nephropathy, focal segmental glomerulosclerosis, and systemic lupus erythematosus. Urinary concentrations of IGF-I and IGF-II in pooled spot morning urines of healthy subjects, measured by RIA, were low and relatively unaltered throughout age, when expressed as either nanograms per milliliter or nanograms per milligram creatinine. To determine the complement of IGFBPs in urine of healthy subjects, spot morning urine samples were subjected to Western ligand blot and immunoblot analysis. IGFBP-3 was detected at 40-50 kDa, possibly due to variable glycosylation of uIGFBP-3. In addition, a 32-kDa IGFBP-2 and smaller unclassified IGFBPs were detected. Unlike uIGFs, urinary concentrations of IGFBP-3 (uIGFBP-3; nanograms per milligram creatinine) were age-, but not sex-related. Levels of uIGFBP-3 ranged from 40-60 ng/mL in children between 4 and 10 yr of age. After 11 yr, immunoreactive uIGFBP-3 progressively declined, attaining a plateau after 26 yr of age to approximately 18 ng/ mg creatinine. uIGFBP-3 did not correlate with uIGF levels. Regulation of IGFBP-3 in the urine of normal subjects and of renal disorders was examined by RIA, Western ligand blot (WLB), and protease assay. Intact uIGFBP-3 was consistently found in normal urine and little urinary protease was identified. In GHD patients, IGFBP-3 by WLB was low or undetectable, whereas RIA levels of uIGFBP-3 were normal or high, consistent with the presence of IGFBP-3 proteolytic activity. In Alport syndrome, both RIA measures and WLB analysis were high, as was the IGFBP-3 proteolytic activity. Patients with immunoglobulin A nephropathy, focal segmental glomerulosclerosis and systemic lupus erythematosus measured low-normal levels of IGFBP-3 by WLB and RIA, and displayed little protease activity. This study provides normative data concerning radioimmunoassay-able levels of IGFBP-3 in urine. The presence of normal-elevated levels of uIGFBP-3 by RIA in GHD indicates that uIGFBP-3 levels are not under GH control and are unlikely to represent filtered serum IGFBP-3. The differences in uIGFBP-3 levels determined by WLB and RIA are consistent with the presence of urinary protease activity, and may be a clinical marker for GHD or renal disease.