POSTOPERATIVE INDUCTION OF INSULIN-LIKE-GROWTH-FACTOR BINDING PROTEIN-3 PROTEOLYTIC ACTIVITY - RELATION TO INSULIN AND INSULIN SENSITIVITY

Increased serum insulin-like growth factor (IGF)-binding protein-3 (IG FBP-3) proteolytic activity (IGFBP-3-PA) has been demonstrated in a nu mber of clinical states of insulin resistance, including severe illnes s, after surgery, and in noninsulin-dependent diabetes mellitus. In th e present study we assessed the role of insulin sensitivity in express ion of IGFBP-3-PA in serum. In 18 patients studied, a significant incr ease in IGFBP-3-PA (P < 0.005) was demonstrated after cole-rectal surg ery. Eight patients receiving an oral glucose load before surgery demo nstrated a significant greater relative increase in IGFBP-3-PA compare d with 10 patients not receiving glucose (32.9 +/- 7.1% vs. 8.6 +/- 6. 7%, respectively; P < 0.05). Both groups had reduced insulin sensitivi ty after surgery(-58 +/- 4%; P < 0.0001; n = 18), as determined by hyp erinsulinemic, normoglycemic clamps; however, the group not receiving glucose displayed 18% less insulin sensitivity than the oral glucose l oad group (P < 0.05). Multiple regression analysis demonstrated that t he relative changes in IGFBP-3-PA and C peptide levels were inversely correlated (P < 0.05), suggesting that increased IGFBP-3-PA, presumabl y increasing IGF bioavailability, may be associated with decreased ins ulin demands. Interestingly, insulin infusion during the 4-h hyperinsu linemic, normoglycemic clamp performed 24 h after surgery (post-op) re sulted in a further increase in IGFBP-3-PA in both groups (P < 0.005), whereas no significant responses could be demonstrated during the pre -op clamp. The expression of increased IGFBP-3-PA was accompanied by c onversion of endogenous intact 39/42-kDa IGFBP-3 into its 30-kDa fragm ented form as determined by Western immunoblotting, and this conversio n was virtually complete after the 4-h post-op clamp in patients displ aying marked increases in IGFBP-3-PA. Characterization of the IGFBP-3- PA demonstrated that it was specific for IGFBP-3, as no degradation of IGFBP-1 and -2 was detected, and the use of various protease inhibito rs demonstrated that serine proteases and possibly matrix metalloprote inases contribute to the increased IGFBP-3-PA level after surgery. We propose that IGF bioavailability may be increased by the induction of IGFBP-3-PA in insulin-resistant subjects, and that insulin regulates I GFBP-3-PA in this state.