The insulin-like growth factor (IGF) system plays an important role in cell proliferation and survival. However, more recently, a small number of studies have shown that IGFs induce apoptosis in some cells. Our initial studies showed this occurred in LIM 1215 colon cancer cells but not RD rhabdomyosarcoma cells. IGFs induced both proliferation and apoptosis in LIM 1215 cells, and the induction of apoptosis was dose-dependent. [R54, R55]IGF-II, which binds to the IGF-I receptor with normal affinity but does not bind to the IGF-II receptor, induced apoptosis to the same extent as IGF-II, whereas [L27]IGF-II, which binds to the IGF-I receptor with 1000-fold reduced affinity, had no effect on apoptosis. These results suggest that the IGF-I receptor is involved in induction of apoptosis. Western blot analyses demonstrated that Akt and Erk1/2 were constitutively activated in RD cells. In contrast, phosphorylation of Akt and Erk1/2 were transient and basal expression of Akt protein was lower in LIM 1215 cells. Analysis of apoptosis-related proteins showed that IGFs decreased pro-caspase-3 levels and increased expression of pro-apoptotic Bad in LIM 1215 cells. IGFs co-activate proliferative and apoptotic pathways in LIM 1215 cells, which may contribute to increased cell turnover. Since high turnover correlates with poor prognosis in colorectal cancer, this study provides further evidence for the role of the IGF system in its progression.

Insulin-like growth factors IGF-I and IGF -II are important mediators of growth. A family of six high affinity IGF binding proteins (IGFBPs) modulate IGF action. IGFBPs have three domains, of which the N- and C-domains are involved in high affinity IGF binding. IGFBP-6 is unique in its 20-100-fold IGF-II binding specificity over IGF-I. The aim of this study was to determine the contributions of the N- and C-domains of IGFBP-6 to its IGF binding properties. We confirmed that differential dissociation kinetics are responsible for the IGF-II binding preference of IGFBP-6. The N-domain has rapid association kinetics, similar to full-length IGFBP-6, but both IGF-I and -II dissociate rapidly from this domain, thereby reducing its binding affinity for IGF-II approximately 50-fold. However, the N-domain binds IGF-I and -II with similar affinities and it has a similar IGF-I binding affinity to full-length IGFBP-6. This suggests that the C-domain confers the IGF-II binding preference of IGFBP-6; indeed, IGF-I bound inconsistently with very low affinity to the C-domain. Coincubation studies showed that isolated N- and C-domains of IGFBP-6 do not strongly cooperate to enhance IGF binding. The results of the binding studies are supported by the effects of the IGFBP-6 domains on IGF-induced colon cancer cell proliferation; the N-domain inhibited IGF-II induced proliferation with approximately 20-fold lower potency than IGFBP-6 and it was equipotent in inhibiting IGF-I- and IGF-II-induced proliferation. Coincubation of C-domain had no additional effect on N-domain-induced inhibition of proliferation. In conclusion, both the N- and C-domains of IGFBP-6 are involved in IGF binding, the C-domain is responsible for the IGF-II binding preference of IGFBP-6 and intact IGFBP-6 is necessary for high affinity IGF binding.

Insulin-like growth factors (IGFs) stimulate proliferation and differentiation of PC12 rat pheochromocytoma cells and modulate catecholamine release in bovine adrenal medullary cells. Dexamethasone increases catecholamine synthesis in PC12 cells. We therefore studied the effects of IGFs and dexamethasone on catecholamine content in PC12 cells. Dopamine (DA) and norepinephrine (NE) content of PC12 cells were measured after incubation for 72 h with IGFs (100 ng/ml) and/or dexamethasone (500 nM). IGF-I (100 ng/ml) and IGF-II (100 ng/ml) decreased DA and NE content to approximately 35% and approximately 25% of control, respectively. [Leu27]IGF-II, which binds to the IGF-I receptor with markedly decreased affinity, did not reduce catecholamine levels, indicating that the effect is likely to be mediated by the IGF-I receptor. Dexamethasone (500 nM) increased levels of DA and NE to 173 +/- 20% and 331 +/- 48% of controls, respectively. Coincubation with IGFs did not significantly affect the stimulation of DA by dexamethasone, but abolished the rise in NE. Levels of tyrosine hydroxylase mRNA, protein and activity were increased following incubation with dexamethasone, but were unchanged by IGFs. These results indicate that IGFs decrease catecholamine content in PC12 cells via the IGF-I receptor. Complex regulation involving multiple synthetic and/or degradative steps is implicated in this process.

Limited knowledge exists regarding the efficacy of insulin-like growth factor I (IGF-I) administration as a therapeutic intervention for muscular dystrophies, although findings from other muscle pathology models suggest clinical potential. The diaphragm muscles of mdx mice (a model for Duchenne muscular dystrophy) were examined after 8 weeks of IGF-I administration (1 mg/kg s.c.) to test the hypothesis that IGF-I would improve the functional properties of dystrophic skeletal muscles. Force per cross-sectional area was similar to49% greater in the muscles of treated mdx mice (149.6 +/- 9.6 Kn/m(2)) compared with untreated mice (100.1 +/- 4.6 kN/m(2), P < 0.05), and maintenance of force over repeated maximal contraction was enhanced similar to30% In muscles of treated mice (P < 0.05). Diaphragm muscles from treated mice comprised fibers with similar to36% elevated activity of the oxidative enzyme succinate dehydrogenase, and similar to23% reduction in the proportion of fast IId/x muscle fibers with concomitant increase in the proportion of type Ha fibers compared with untreated mice (P < 0.05). The data demonstrate that IGF-I administration can enhance the fatigue resistance of respiratory muscles in an animal model of dystrophin deficiency, in conjunction with enhancing energenic enzyme activity. As respiratory function is a mortality predictor in Duchenne muscular dystrophy patients, further evaluation of IGF-I intervention is recommended.

Butyrate has potent anti-tumorigenic effects on many colon cancer cell lines, including inhibition of growth and promotion of apoptosis in vitro. Nevertheless, despite the butyrate concentration in the colonic lumen being sufficient to result in the death of almost all cells in vitro, colon cancers still develop and grow in vivo, suggesting that cancer cells must develop mechanisms by which they escape the effects of butyrate observed in vitro. Insulin-like growth factor-II (IGF-II) is an autocrine growth factor in many colon cancer cells. The aim of this study was to determine whether IGF-II influences butyrate-mediated apoptosis in LIM 2405 human colon cancer cells. Butyrate and trichostatin A, both of which are histone deacetylase inhibitors although the latter is more specific, induced apoptosis as determined by floating cell counting, Hoechst 33258 staining, DNA laddering and a cell death detection ELISA. IGF-II inhibited the effects of both agents. Butyrate but not trichostatin A also induced LIM 2405 cell migration. In contrast to the above results, IGF-II enhanced butyrate-induced cell migration. Levels of IGF binding protein-3 (IGFBP-3), which may induce apoptosis by IGF-dependent or -independent mechanisms, were increased by butyrate and trichostatin A; IGF-II augmented this effect. It is therefore unlikely that IGFBP-3 mediates butyrate-induced apoptosis. We suggest that IGF-II inhibits the pro-apoptotic effect of butyrate downstream of histone deacetylase inhibition. In contrast, IGF-II promotes histone deacetylase-dependent IGFBP-3 expression and histone deacetylase-independent migration. IGF-II may promote tumour growth by mediating the development of resistance to the pro-apoptotic effects of butyrate.

IGF-II is an autocrine growth factor for many colon cancer cells. This study aimed to determine the role of IGF-II in proliferation and adhesion of LIM 1215 colon cancer cells. RT-PCR demonstrated expression of IGF-I and IGF-II mRNA. Addition of IGF-I or -II increased monolayer proliferation in a dose-dependent manner. Although addition of IGFBP-6 had no effect on basal proliferation, coincubation of IGFBP-6 decreased IGF-II but not IGF-I-induced proliferation. Colony formation in agar was increased by IGF-II, an effect inhibited by coincubation with IGFBP-6. IGFBP-6 alone significantly decreased colony formation. Preincubation of cells with IGF-II increased adhesion to type IV collagen, fibronectin and laminin. IGFBP-6 had no effect on basal cell adhesion but completely inhibited the effects of IGF-II. LIM 1215 colon cancer cells are therefore IGF-responsive but IGF-II is not a major autocrine factor for these cells in monolayer, suggesting heterogeneity between colon carcinoma cell lines with respect to the role of the IGF system.

Insulin-like growth factor binding protein-6 (IGFBP-6) is an O-linked glycoprotein which specifically inhibits insulin-like growth factor (IGF)-II actions. The effects of O-glycosylation of IGFBP-6 on binding to glycosaminoglycans and proteolysis, both of which reduce the IGF binding affinity of other IGFBPs were studied. Binding of recombinant human nonglycosylated (n-g) IGFBP-6 to a range of glycosaminoglycans in vitro was approximately threefold greater than that of glycosylated (g) IGFBP-6. When bound to glycosaminoglycans, IGFBP-6 had approximately 10-fold reduced binding affinity for IGF-II. Exogenously added n-gIGFBP-6 but not gIGFBP-6 also bound to partially purified rat PC12 phaeochromocytoma membranes. Binding of n-gIGFBP-6 was inhibited by increasing salt concentrations, which is typical of glycosaminoglycan interactions. O-glycosylation also protected human IGFBP-6 from proteolysis by chymotrypsin and trypsin. Proteolysis decreased the binding affinity of IGFBP-6 for IGF-II, even with a relatively small reduction in apparent molecular mass as observed with chymotrypsin. Analysis by ESI-MS of IGFBP-6 following limited chymotryptic digestion showed that a 4.5-kDa C-terminal peptide was removed and peptide bonds involved in the putative high affinity IGF binding site were cleaved. The truncated, multiply cleaved IGFBP-6 remained held together by disulphide bonds. In contrast, trypsin cleaved IGFBP-6 in the mid-region of the molecule, resulting in a 16-kDa C-terminal peptide which did not bind IGF-II. These results indicate that O-glycosylation inhibits binding of IGFBP-6 to glycosaminoglycans and cell membranes and inhibits its proteolysis, thereby maintaining IGFBP-6 in a high-affinity, soluble form and so contributing to its inhibition of IGF-II actions.

The insulin-like growth factor (IGF) system plays an important role in skin. HaCaT human keratinocytes proliferate in response to IGFs and synthesize IGF-binding protein-3 (IGFBP-3). Recently, IGFBP-6 was also identified by NH2-terminal sequencing, but it has not been identified by Western ligand blotting. In the present study, IGFBP-6 was detected in HaCaT-conditioned medium by use of immunoblotting and Western ligand blotting with125I-labeled IGF-II. Proteolytic activity against IGFBPs, an important mechanism for regulation of their activity, was then studied. An acid-activated, cathepsin D-like protease that cleaved both IGFBP-6 and IGFBP-3 was detected. Although proteolysis did not substantially reduce the size of immunoreactive IGFBP-6, it greatly reduced the ability of IGFBP-6 to bind125I-IGF-II as determined by Western ligand blotting and solution assay. HaCaT keratinocytes do not express IGF-I mRNA, but IGF-II mRNA and protein expression was detected. These observations suggest the possibility of an autocrine IGF-II loop that is regulated by the relative expression of IGF-II, IGFBP-3, and IGFBP-6, and IGFBP proteases in these keratinocytes, although demonstration of this loop requires further study.

Earlier studies in our laboratory demonstrated that the steroidal antiestrogen ICI 182,780 is very effective in abolishing the tamoxifen‐resistant proliferation of MCF 7/5‐23 cells [1]. In addition, preliminary binding studies showed that ICI 182,780 increased the binding of insulin‐like growth factor (IGF)‐I to the MCF 7/5‐23 cells, although this finding was not the result of an increase in the expression of the insulin‐like growth factor‐I receptor (IGF‐IR). Hence, we reasoned that the inhibition of tamoxifen‐resistant cell growth by ICI 182,780 might have been due to increased expression of insulin‐like growth factor binding proteins (IGFBPs).

The predominant insulin-like growth factor binding protein (IGFBP) synthesized by PC12 rat phaeochromocytoma cells is IGFBP-6. Since cAMP agonists regulate IGFBP-6 in other cells, and they may increase neurite outgrowth and catecholaminergic enzyme expression in PC12 cells, we studied regulation of IGFBP-6 by these agents. After 72 h incubation, forskolin and 8-(4-chlorophenylthio)-cAMP (CPT-cAMP) both increased IGFBP-6 protein levels in conditioned media to maximum levels of 231 +/- 40 and 275 +/- 30%, respectively. Incubation with forskolin resulted in a small, transient rise in IGFBP-6 mRNA levels which was insufficient to account for the increased protein levels. The increased protein levels also could not be attributed to increased cell number, protection of IGFBP-6 from proteolysis or release of IGFBP-6 from a cell-associated reservoir. These findings suggest that increased protein levels may have been due to increased translation of mRNA. Co-incubation of forskolin with dexamethasone (which decreases IGFBP-6 protein and mRNA) demonstrated that the effects of the latter were dominant. The effects of cAMP agonists and IGF-II, which increases IGFBP-6 protein but not mRNA levels, were not inhibited by rapamycin, suggesting that p70 S6 kinase is not involved. The effects of cAMP agonists on IGFBP-6 levels were not directly correlated with neurite outgrowth. These findings extend our knowledge of the molecular basis of the regulation of IGFBP-6 by cAMP agonists, and indicate a novel action of these agents in PC12 cells.

PC12 rat phaeochromocytoma cells are widely used as a model of neuronal differentiation. They express IGF receptors and are responsive to IGFs. The main IGF-binding protein synthesized by these cells is IGFBP-6. Glucocorticoids induce differentiation of PC12 cells towards a chromaffin phenotype. The effect of dexamethasone on IGFBP-6 levels was therefore studied. Dexamethasone (500 nM) decreased IGFBP-6 protein in conditioned media and mRNA levels to 61 +/- 5% (P < 0.0001) and 34 +/- 14% (P = 0.03) of control levels respectively. Incubation of PC12 cells with IGF-II (100 ng/ml) for 72 h increased IGFBP-6 protein levels in media to 217 +/- 19% of control (P < 0.0001). IGFBP-6 mRNA levels, however, were unchanged. IGF-I had similar effects on IGFBP-6 protein and mRNA levels. IGFs increased cell number by 50-60%, but this was insufficient to explain the increases in protein levels. IGFBP-6 was not released from a cell-associated reservoir or protected from proteolysis by IGFs, excluding these post-translational mechanisms as explanations for the IGF effects on IGFBP-6 levels. The effects of IGF-II and dexamethasone on IGFBP-6 levels were independent. These results indicated that (1) dexamethasone decrease IGFBP-6 at the mRNA level, and (2) IGFs stimulate IGFBP-6 levels by a post-transcriptional mechanism.

Insulin-like growth factor II (IGF-II) stimulates the proliferation and differentiation of rat myoblasts. Previous studies suggest that these response are mediated by the IGF-I receptor, but the IGF-II/mannose 6-phosphate receptor was recently implicated in differentiation of mouse myoblasts. L6A1 myoblasts synthesize IGF-binding protein-4 (IGFBP-4), IGFBP-5, and IGFBP-6, which modulate IGF action. We studied the roles of IGF receptors and IGFBPs in L6A1 myoblast proliferation and differentiation by comparing the effects of IGF-II and a number of IGF-II mutants with decreased affinities for IGF receptors and/or IGFBPs. IGF-II induced concentration-dependent proliferation with a maximum increase of 47%; half-maximal proliferation was seen with approximately 50 ng/ml. [Arg54, Arg55]IGF-II bound to the IGF-I receptor with slightly lower affinity than IGF-II, did not bind to the IGF-II/mannose 6-phosphate receptor, and bound to IGFBPs secreted by myoblasts with approximately 16-fold decreased affinity. It induced proliferation with equal potency to IGF-II. [Leu27]IGF-II, which did not bind to the IGF-I receptor but bound to the IGF-II/mannose 6-phosphate receptor and IGFBPs with slightly lower affinity than IGF-II, had a markedly impaired proliferative effect, inducing proliferation only at high concentrations. [Thr48, Ser49, Ile50]IGF-II, which bound to the IGF-I receptor with slightly lower affinity than IGF-II but did not substantially bind to the IGF-II/mannose 6-phosphate receptor or IGFBPs, induced proliferation with approximately 5-fold greater potency than IGF-II. The order of potency in inducing myoblast differentiation was the same, although there was less difference in the relative potencies of IGF-II and mutants. Coincubation of recombinant human (rh) IGFBP-6 in molar excess with IGF-II inhibited myoblast proliferation and differentiation. rhIGFBP-6 was slightly less potent did not inhibit proliferation or proliferation or differentiation induced by [Thr48,Ser49,Ile50]IGF-II. These results suggest that 1) IGF-II-induced proliferation and differentiation of L6A1 myoblasts are predominantly mediated by the IGF-I receptor; 2) the IGF-II/mannose 6-phosphate receptor is not required for these actions of IGF-II; 3) nevertheless, the IGF-II/mannose 6-phosphate receptor may be capable of mediating these actions; and 4) IGFBPs secreted by myoblasts inhibit IGF actions.