Abstract. Insulin-like growth factor I and II (IGF I and II) were determined by five different assays in human serum, in the sera of ten mammalian species and in chicken, turtle, and frog serum. Sera of all tested mammals contain two different IGFs corresponding to human immunoreactive IGF I and receptor reactive IGF II. Receptor reactive IGF II of most animal species does not show significant cross-reactivity in the RIA for human IGF II. IGF activity was also detected in sera of non-mammals, such as chicken and turtles, but not in frog serum. The IGF values obtained with the different assay system corresponded rather well: there is a good correlation between the values obtained in the protein binding and the fat cell assay, and between the results of the latter assays and the sum of immunoreactive IGF I and receptor reactive IGF II. The results suggest that those regions in the IGF I and II molecules which are responsible for reactivity with the type I IGF and the insulin receptor have not essentially changed during evolution. Similarly, the C-region, which mainly determines the immunological properties of IGFs, appears to have remained relatively constant in the IGF I, but not in the IGF II molecule.
Ines Zangger, Jürgen Zapf and E. Rudolf Froesch
Katharina Binz, Jürgen Zapf and E. Rudolf Froesch
Insulin-deficient, streptozotocin-diabetic rats show severe metabolic disturbances and stop growing. Besides insulin, these animals also lack growth hormone and insulin-like growth factor-I. We examined whether or not growth parameters correlate with IGF-I serum levels in young rats with streptozotocin-diabetes of different severity. In the diabetic rats, blood glucose varied between 18.4 and 38.6 mmol/1 (healthy controls between 6.1 and 9.3), IGF-I serum levels between 2.6 and 15.6 nmol/1 (controls between 19.6 and 26.5), and serum insulin levels between 0.05 and 0.14 nmol/1 (controls between 0.36 and 0.55). We found a highly significant linear correlation between IGF-I serum levels and the two investigated growth parameters, tibial epiphyseal width and longitudinal tibial bone growth. The finding that these indices of growth are strongly correlated with IGF-I serum levels in young rats with diabetes of different severity, suggests that IGF-I is a major determinant of growth. This is in keeping with our earlier demonstration that exogenously infused IGF-I promotes growth in diabetic rats.
Susanne Keller, Christoph Schmid, Jürgen Zapf and E. Rudolf Froesch
IGF-I infused at pharmacological doses in healthy men markedly decreases C-peptide levels, whereas insulin levels remain within the normal range. One possible explanation is decreased insulin removal. As the liver is the major site of insulin degradation, we studied insulin degradation by HepG2 cells in the presence of IGF. We found that IGF-I at a concentration of 130 nmol/l inhibits insulin degradation by HepG2 cells when the initial insulin concentration is 0.34 nmol/l. The effect of IGF-I on insulin degradation is dose-dependent and the rate of insulin degradation is dependent on the insulin concentration. IGF-II is 6 to 10 times more potent than IGF-I in inhibiting 125I-insulin binding to HepG2 cells and in protecting insulin from being degraded. Thus, IGF-I and IGF-II inhibit insulin degradation most likely by competing for binding at insulin binding sites of liver cells.
J. Eugen Eigenmann, Jan. J. de Bruijne and E. Rudolf Froesch
Abstract. The roles of plasma insulin-like growth factor I (IGF I) and growth hormone (GH) were studied in 7 beagle dogs before and during starvation and during refeeding. IGF I levels significantly decreased from 75.2 ± 5.9 ng/ml at 7 days prior to the start of starvation to 9 ± 1.7 ng/ml at 19 days after the commencement of starvation (mean ± sem; P < 0.0001). During refeeding IGF I significantly rose from 9 ± 1.7 ng/ml to 55.5 ± 7.5 ng/ml within 9 days (mean ± sem; P < 0.002).
During starvation plasma GH levels significantly increased (P < 0.05) and these elevated levels returned to normal during refeeding. The dogs' GH secretory capacity significantly increased during starvation (P = 0.012) and became normal again during refeeding. The following conclusions can be drawn from this study: 1) starvation in the dog leads to a significant and drastic reduction of the circulating levels of IGF I, and 2) starvation in the dog, as in man, leads to increased circulating GH levels and to an increased GH-secretory capacity possibly brought about by a lack of a negative feedback normally exerted by IGF I.
Hans-Peter Guler, Jürgen Zapf, Christoph Schmid and E. Rudolf Froesch
IGF-I and -II share specific serum carrier proteins which elute on neutral Sephadex G-200 gel permeation chromatography at apparent molecular masses of 50 and 200 kD. The half-lives of free and carrier protein-bound 125I-IGF-I and -II were determined after bolus injections of the tracers into two normal adults. Labelled IGF-I and -II migrated first with the 50-kD and later with the 200-kD complex. In these complexes their apparent half-lives were 20–30 min and 12–15 h, respectively. The apparent half-life of free 125I-IGF-I and -II was 10–12 min. In a second set of experiments, recombinant human insulin-like growth factor I was infused during 6 days in two healthy adults at a dose of 20 μg · kg−1 · h−1 (corresponding to around 30 mg/day). Serum obtained before and during the infusion was subjected to neutral Sephadex G-200 gel permeation chromatography and fractions were pooled according to the apparent molecular masses at which the carrier protein complexes elute. IGF-I and -II in these pools were determined by RIA. Before the IGF-I infusion, 92 and 272 μg/l of IGF-I and -II were found in the 200-kD complex, 45 and 91 μg/l in the 50-kD complex, and 15 and 5 μg/l were present in the free form. Corresponding figures during the IGF-I infusion were 389 and 18 μg/l for the 200-Kd complex, 201 and 54 μg/l for the 50-kD complex, and 80 and < 1 μg/l for free IGF-I and -II. Using the half-lives of the tracer studies and the levels of the different molecular weight forms of IGF in serum, the production rates for IGF-I and -II were calculated to be 10 mg and 13 mg per day.
Peter D. Zenobi, Hans-Peter Guler, Jürgen Zapf and E. Rudolf Froesch
Abstract. IGF I was determined by a radioimmunoassay and IGF II by a radioreceptorassay in 20 Göttinger miniature (mini)-pigs and 13 domestic pigs of different weight and age. Immunoreactive IGF I serum levels of mini-pigs were similar to those of domestic pigs in corresponding age-classes (150–250 and 100–270 μg/l, respectively). No differences were detectable between receptor-reactive IGF II serum levels in mini-pigs (150–200 μg/l) and domestic pigs (110–270 μg/l) nor did the biological insulin-like activites (measured in the rat fat cell assay) differ in mini- and domestic pigs (81–100 and 71–98 mU insulin/l, respectively). IGF I and IGF II decreased drastically after hypophysectomy in one of the mini-pigs. Intravenous bolus injections of 30 μg/kg of recombinant human IGF I in 4 mini-pigs caused a similar degree of hypoglycemia (nadir of blood glucose 1.33 ± 0.61 mmol/l) as 0.15 IU insulin/kg, followed by a sharp growth hormone peak. We conclude that the marked difference between mini- and domestic pigs regarding body size is unrelated to serum levels of IGF I and II, a lack of response of tissues to IGF I or a reduced growth hormone secretory capacity in the mini-pig.
Hans-Peter Guler, Katharina Binz, Eugen Eigenmann, Silvia Jäggi, Daniel Zimmermann, Jürgen Zapf and E. Rudolf Froesch
The short stature of mini-poodles is associated with low serum levels of IGF-I. Standard poodles are taller and have considerably higher serum levels of IGF-I. Low IGF-I serum levels may be a symptom or the cause of small stature. We, therefore, undertook a study in which serum IGF-I levels of mini-poodles were elevated over a prolonged period of time by a constant infusion of rhIGF-I and the growth rate of the mini-poodles was followed. We infused four mini-poodles from day 91 to day 221 of age with 6 mg/day of recombinant human insulin-like growth factor I (rhIGF-I). Serum levels of IGF-I rose from about 160 to about 500 μg/l. Blood glucose remained within normal limits. Stimulation tests with clonidine and with GHRH revealed suppression of endogenous GH secretion during the IGF-I infusion. Serum levels of IGF-II and of creatinine were lower in the IGF-I-infused animals. Radial length and body weight did not increase to a greater extent in the IGF-I infused dogs than in controls. However, 'adapted body mass index' (aBMI = gram body weight/(mm radial length)2) decreased in each of the IGF-I infused animals, whereas it increased in each of the control dogs (p <0.05). We conclude that long-term infusion of IGF-I does not stimulate growth in young minipoodles, but may change body composition.
Hans-Peter Guler, Kai-U. Eckardt, Jürgen Zapf, Christian Bauer and E. Rudolf Froesch
Recombinant IGF-I was infused sc at a dose of 20 μg · kg−1 · h−1 to 2 healthy subjects during a total of 79 h. Serum levels of IGF-I rose from 93 and 177 to 502 and 616 μg/l, respectively. Fasting blood glucose remained normal. During the infusion, glomerular filtration rate increased by 31% in subject No. 1 and by 32% in subject No.2. Concomitantly, renal plasma flow increased by 26% and 22%, respectively. Proximal and distal tubular reabsorption of fluid and sodium as determined by lithium clearance was elevated to a similar extent. When determined again one week after the end of the IGF-I infusion, all parameters of renal function had returned to baseline. Sodium excretion, body weight and blood pressure did not change. We conclude that IGF-I infused at pharmacological doses has marked effects on kidney function. Future studies will be necessary to define the clinical potential of recombinant IGF-I in the treatment of diseases characterized by impaired renal perfusion and filtration.
Wolfgang Moritz, Marianne Böni-Schnetzler, Wayne Stevens, E Rudolf Froesch and James R Levy
Moritz W, Böni-Schnetzler M, Stevens W, Froesch ER, Levy JR. In-frame exon 2 deletion in insulin receptor RNA in a family with extreme insulin resistance in association with defective insulin binding. Eur J Endocrinol 1996;135:357–63. ISSN 0804–4643
The phenotype and allelic expression of the insulin receptor gene is presented in a family with a patient with type A insulin resistance. Compared to controls, insulin receptor binding in transformed lymphocytes was 100%, 33% and 13% in the father, mother and proband, respectively. Reduced insulin receptor binding co-segregated with altered insulin receptor mRNA expression; the mother and daughter expressed eight insulin receptor mRNA species, including a set of four normal sized and a set of four shorter mRNA transcripts. In the proband the levels of the normal sized mRNA transcripts were suppressed relative to the shorter transcripts. Reverse polymerase chain reaction (PCR) revealed that the shorter transcripts contained an in-frame deletion of exon 2. Sequencing of the entire insulin receptor coding region revealed a paternally inherited A to T substitution in nucleotide 3205, converting isoleucine 996 to phenylalanine. which does not co-segregate with reduced binding. Therefore, we hypothesize that two findings are necessary for the presentation of type A insulin resistance in this patient: an in-frame deletion of the insulin receptor exon 2 that codes for amino acids crucial for insulin binding; and an inhibition of expression of the paternal insulin receptor allele.
Marianne Böni-Schnetzler, Division of Endocrinology and Metabolism, Department of Internal Medicine, University Hospital, 8091 Zurich, Switzerland
Jürgen Zapf, Beat Morell, Helmut Walter, Zvi Laron and E. Rudolf Froesch
The levels of insulin-like growth factors (IGF), two somatomedin-like polypeptides of human serum and of their carrier protein were determined in sera of patients with various metabolic disorders. IGF was measured by 4 different methods (fat pad and fat cell assay and competitive protein binding assay measuring total IGF, and a radioimmunoassay for IGF I) after extraction by acidic gel filtration on Sephadex G-50. This procedure is necessary to separate IGF from the carrier protein, which interferes with all of these assays.
1) In normal serum, immunoreactive IGF I accounts for one third of total IGF determined by the fat pad assay, but only for one fifth to one sixth of immunoreactive IGF I + II.
2) In acromegalics total IGF was increased 1.5- (protein binding and fat cell assay) to 2-fold (fat pad assay), but the increase was solely due to immunoreactive IGF I, which was ∼ 5-times above normal. The IGF binding activity was not elevated. Total IGF and IGF binding were decreased in hypopituitarism, Laron-type dwarfism and in liver cirrhosis. Immunoreactive IGF I was more drastically reduced in these diseases than total IGF. Apparently, only IGF I is under growth hormone control. The liver seems to be involved in the production of IGF.
3) No elevation of total IGF was found in patients with extrapancreatic tumour hypoglycaemia, but IGF binding was reduced. Immunoreactive IGF I was decreased in 5 of 10 patients. These results suggest that tumour hypoglycaemia in our patients is unlikely to be caused by increased IGF levels.
4) In patients with hyperprolactinaemia neither total IGF nor immunoreactive IGF I were elevated, and IGF binding was unchanged.
5) In newly detected insulin-deficient juvenile diabetics total IGF and immunoreactive IGF I levels were within the normal range, although the variation was greater than in normal subjects. However, IGF binding was markedly decreased.