Evaluation of the components of insulin-like growth factor (IGF)-IGF binding protein (IGFBP) system in adolescents with type 1 diabetes and persistent microalbuminuria: relationship with increased urinary excretion of IGFBP-3 18 kD N-terminal fragment

OBJECTIVE IGFs and their binding proteins (IGFBPs) have an important role i n controlling glucose homeostasis and there is evidence to support their in volvement in complications related to type I diabetes. The aim of this stud y was to evaluate the components of the IGF-IGFBP system in adolescents wit h type 1 diabetes that had developed persistent microalbuminuria (MA). DESIGN AND PATIENTS A cohort of 49 adolescents with type 1 diabetes were en rolled in the study. Patients were evaluated at baseline and 1 year later ( follow-up). Twenty-six patients with persistent urinary albumin excretion ( UAE) of more than 20 mu g/min/1.73 m(2) (21.6-109.4 mu g/min/1.73 m(2)) in three different nocturnal urinary collections within 6 months were consider ed to have MA (baseline mean: 41.9+/-22.3 mu g/min/1.73 m(2); follow-up: 55 .9 +/- 24.8 mu g/min/1.73 m(2)). Twenty-three patients with UAE of less tha n 20 mu g/min/1.73 m(2) were assigned to the group without MA (baseline mea n: 8.6 +/- 3.7 mu g/min/1.73 m(2); follow-up: 11.8 +/- 4.2 mu g/min/1.73 m2 ). Pasting serum levels of IGPBP-1, IGFBP-2, IGFBP-3, IGF-1 and free-IGF-1 were determined using appropriate immunoenzymatic, radioimmuno- or immunora diometric assays. Overnight 12-h urinary collections were obtained and asse ssed for IGFBP-3 levels, determined by immunoradiometric assay. Urinary and circulating immunoreactive IGPBP-3 forms were determined by Western-immuno blotting (WIB) analysis using a specific polyclonal antibody and monoclonal antibodies directed against N-terminal and C-terminal epitopes of IGFBP-3. IGFBP-3 protease activity was determined using protease assay and by analy sis of the intact over the fragmented immunoreactive forms of IGFBP-3 deter mined by WIB analysis. RESULTS Patients with MA showed higher levels of urinary IGFBP-3 (649+/-440 ng/h/m(2)) than patients without MA (398+/-229 ng/h/m(2); P<0.05). Urinary levels of IGFBP-3 were directly correlated to UAE (P<0.001). WIB analysis, using monoclonal antibodies directed against characterized N-terminal and C -terminal IGFBP-3 epitopes, determined that the immunoreactive form of IGFB P-3 found in urine from patients with diabetes was an N-terminal 18 kD frag ment. Serum IGFBP-3 levels were lower in patients with MA (baseline: 3613+/ -598 mu g/l; one year follow-up: 3347+/-624 mu g/l) compared with patients without MA (baseline: 4701+/-1484 mu g/l; follow-up: 4177+/-703 mu g/l; P<0 .001). In serum from patients with MA, intact IGFBP-3 was decreased, as ind icated by WIB analysis. Conversely, IGPBP-3 proteolysis was increased in pa tients with MA (baseline: 131+/-21% of control; follow-up: 130+/-23% of con trol), compared to patients with normal UAE (baseline: 96+/-23% of control; follow-up: 96+/-14% of control; P<0.001). Serum IGFBP-3 protease activity was directly correlated to urinary IGFBP-3 levels (P<0.001). Serum IGPBP-1 levels were increased in patients with MA (baseline: 36+/-20 mu g/l; follow -up: 36+/-17 mu g/l) compared with patients without MA (baseline: 17 +/- 11 mu g/l; follow-up: 18+/- mu g/l; P<0.05). Serum IGFBP-2 levels were also p ersistently increased in patients with MA (baseline: 503+/-134 mu g/l; foll ow-up: 484+/-166 mu g/l) compared with patients without MA (baseline: 375+/ -83 mu g/l; follow-up: 390+/-85 mu g/l; P<0.05). On the other hand, free IG F-1 levels were decreased in patients with MA (baseline: 2.3+/-1.5 mu g/l; follow-up: 2.5+/-1.4 mu g/l) compared with those patients without MA (basel ine: 4.1+/-2.1 mu g/l; follow-up: 4.0+/-2.2 mu g/l; P<0.05). CONCLUSIONS It has been demonstrated that in adolescents with type 1 diabet es and persistent microalbuminuria, the IGF-IGFBP system is deranged. A uri nary 18kD N-terminal IGFBP-3 fragment has been characterized that, in patie nts with microalbuminuria, correlates with urinary albumin excretion and wi th serum IGFBP-3 protease activity. The mechanisms behind these changes rem ain unclear, but alterations in circulating levels of IGFBPs may alter IGF- 1 bioactivity. Determination of urinary levels of IGFBP-3 might be supporti ve to the measurement of urinary albumin as an early sign of diabetic nephr opathy.