Lack of insulin effect on its own receptors in fetal rat hepatocytes

To determine the effect of insulin on its receptor concentrations in hepatocytes of fetal and adult rats, these cells were preincubated in the presence or absence of insulin. The reduced [125I]-insulin binding observed in adult hepatocytes was dependent on the concentration of insulin and on the duration of exposure, while in fetal hepatocytes insulin did not induce any reduction in insulin binding. In contrast, glucagon receptors were unaffected by preincubation with insulin. The modifications observed in insulin binding were accounted for by changes in receptor concentrations rather than any change in receptor affinity for the hormone. Studies on the kinetic properties of the insulin receptors of fetuses and adult rats revealed that association and dissociation rates were undistinguishable. These results indicate an absence of insulin receptor down-regulation in the fetus, which could favour anabolic processes during intrauterine life.     

Insulin receptors in developing rat liver: Acquisition of down regulation in the immediate postnatal period

Alterations in the high and low affinity insulin receptor concentrations in developing rat liver were investigated. The number of high affinity receptors in partially purified plasma membranes from fetal rats increased from Days 19 through 22 of gestation, with no further increase in binding during the postnatal period. Fetuses of diabetic rats had approximately three times as many high affinity insulin receptors as age-matched fetuses of normal rats; however, by 1 day after birth the receptor number decreased to the normal level. Neither the number of low affinity receptors nor the affinity of insulin binding to high or low affinity receptors changed during development or between offspring of normal and diabetic rats. These changes in the number of high affinity hepatic insulin receptors from prenatal animals did not correlate with the concentration of plasma insulin. When suckling pups were rendered diabetic the changes in the number of high affinity insulin receptors correlated with alterations in plasma insulin concentrations. The number of high affinity sites/microgram DNA in hepatocytes from Day 18 fetal rats was not altered when cells were cultured for 48 h in medium containing 0, 250, or 5000 μU/ml of added insulin. When cultured hepatocytes derived from 1-day-old and adult rats were maintained in medium with added insulin concentrations of 250 or 5000 μU/ml the number of high affinity receptors/microgram DNA decreased as compared to the number of high affinity receptors in hepatocytes cultured in medium with no added insulin. This decrease in receptor number was accompanied by an increase in the affinity of insulin binding to its high affinity receptors. The data show that (i) only the high affinity insulin receptor number increases in rat liver during the prenatal period, (ii) fetuses of diabetic rats show a greater increase in high affinity receptors than do fetuses of normal animals, and (iii) the phenomenon of down regulation for high affinity insulin receptors is not observed in fetal rat liver, but is acquired in the immediate postnatal period.     

Asymmetric distribution of the phosphatidylinositol-linked phospho-oligosaccharide that mimics insulin action in the plasma membrane

We have investigated the topography of a glycosyl-phosphatidylinositol implicated in insulin action by a combination of two complementary methods: (a) chemical labelling with a non-permeable (isethionyl acetimidate) and a permeable (ethyl acetimidate) probe; and (b) enzymatic modifications with beta-galactosidase (EC 3.2.1.23) or phosphatidylinositol-specific phospholipase C (EC 3.1.4.3). Using the first approach the majority of the glycosyl-phosphatidylinositol is found in the outer surface of intact hepatocytes, adipocytes, fibroblasts and lymphocytes, but not in erythrocytes which presented only a 20% of the total labelled glycosyl-phosphatidylinositol to the exterior. Upon insulin addition (10 nM), about 60% of the total glycosyl-phosphatidylinositol was hydrolysed in both hepatocytes and adipocytes but not in erythrocytes. In agreement with the extracellular localization in hepatocytes and with the proposed role of this glycolipid in insulin action, treatment of rat hepatocytes with beta-galactosidase from Escherichia coli, an enzyme that hydrolyses the oligosaccharide moiety of the glycosyl-phosphatidylinositol, cleaved 65% of the total glycophospholipid and blocked the effect of insulin (but not of glucagon) on pyruvate kinase (EC 2.7.1.40). Similar treatment with phosphatidylinositol-specific phospholipase C from Bacillus cereus hydrolysed 62% of the total glycosyl-phosphatidylinositol. From the various approaches used it is concluded that the majority of this glycophospholipid is at the outer surface in a variety of insulin-sensitive cells.     

Insulin receptors on erythrocytes and hepatocytes from streptozotocin induced diabetic rats

Insulin receptors on hepatocytes and erythrocytes were studied in rats two and eight weeks after the injection of streptozotocin (50 mg/kg) to see if erythrocyte insulin receptors change parallel with hepatocyte insulin receptors in response to hypoinsulinemia. Insulin binding to hepatocytes increased two (14.0 +/- 2.5% v.s. 7.7 +/- 0.7%; P less than 0.025) and eight weeks (15.9 +/- 1.9% v.s. 6.6 +/- 1.1%; P less than 0.005) after the streptozotocin injection. Scatchard analysis revealed that this increase was due to a rise in both the receptor concentration and affinity. The number of receptors was comparable in the two- and eight-week-streptozotocin rats while the increase in the affinity was more pronounced in the latter group. Insulin binding to the erythrocytes was also increased in both two- (5.0 +/- 0.7% v.s. 4.2 +/- 0.6%) and eight-week- (4.3 +/- 0.6% v.s. 2.7 +/- 1.2%) streptozotocin rats. This increase was due to a rise in the receptor concentration rather than the affinity. However, compared to hepatocytes, these changes were inconsistent and statistically not significant. Furthermore, no correlation was obtained between the binding and plasma insulin concentration. These results indicate that insulin receptors on rat erythrocytes are less sensitive to a change in the plasma insulin concentration and do not always reflect accurately the receptor state on hepatocytes.     

Insulin and glucagon binding to isolated hepatocytes of Egyptian sand rats (Psammomys obesus): evidence for an insulin receptor defect

To determine whether a defect in insulin binding could contribute to insulin resistance in Egyptian sand rats (Psammomys obesus), insulin binding to isolated hepatocytes from euglycemic sand rats was compared to that of normal Sprague-Dawley rats (Rattus norvegicus). Because of its potential importance in glucoregulation, glucagon binding to hepatocytes from these species was also measured. Hepatocytes of sand rats exhibit an almost complete lack of insulin receptors compared to hepatocytes from Sprague-Dawley rats, whereas there are numerous high affinity glucagon binding sites on sand rat hepatocytes. The lack of insulin binding to sand rat tissues is sufficient to entirely explain the insulin resistance seen in this species. Glucagon may be primarily responsible for glucose homeostasis in Psammomys obesus.     

Insulin receptors on hepatocytes from spontaneously diabetic Chinese hamsters

Insulin receptors on hepatocytes were studied in spontaneously diabetic Chinese hamsters, which are the animal models for insulin deficient diabetes. Insulin binding in diabetic animals increased mainly due to an increase in the number of receptors. Although binding affinity of diabetic animals was similar to that of control animals, a kinetic study revealed that both the association rate constant and the dissociation rate constant decreased in diabetic animals. Negatively cooperative interactions between receptors were demonstrated in control and diabetic animals, and both the magnitude and sensitivity of this effect was the same in both types of animals. A significant inverse correlation between insulin binding and the plasma insulin concentration was found in these animals. These results therefore suggest that there is an increase in the insulin binding in the insulin deficient diabetic state mainly due to an increase in the number of receptors with a decrease in both the association and dissociation rate constants, and these changes may be important in the altered metabolic state.     

TSH induces insulin receptors that mediate insulin costimulation of growth in normal human thyroid cells

The mitogenic/goitrogenic effects of thyrotropin (TSH) on human thyrocytes in vitro and in vivo depend on permissive comitogenic effects of insulin-like growth factors (IGFs), which are mimicked in vitro by the low-affinity binding of high supraphysiological concentrations of insulin to IGF-I receptors. Contrary to general assumption, we show here that very low concentrations of insulin, acting through insulin receptors but not IGF-I receptors, can also support the stimulation of DNA synthesis by TSH in primary cultures of normal human thyrocytes. Moreover, TSH through cAMP increases the content of insulin receptors demonstrated by Western blotting and the cells' responsiveness to low insulin concentrations. These observations provide the first in vitro evidence in normal human thyroid cells of a functional interaction between TSH and insulin acting through its own receptor.     

Insulin action and binding in isolated hepatocytes from fasted, streptozotocin-diabetic, and older, spontaneously obese rats

Insulin binding and basal and insulin-stimulated uptake of α-aminoisobutyric acid were measured in isolated hepatocytes from young control rats as well as from older spontaneously obese, 72h-starved, and nonketotic streptozotocin-diabetic rats. Isolated hepatocytes from older spontaneously obese rats are similar to those from younger smaller rats in size, maximal insulin responsiveness, the dose–response relationship for insulin-stimulated aminoisobutyrate uptake, and the number and affinity of insulin receptors. Hepatocytes from 72h-fasted rats have similar numbers of insulin receptors per cell as cells from young control animals, but are significantly smaller, have an enhanced basal rate of aminoisobutyrate uptake, and are insulin resistant with regard to maximal insulin-stimulated uptake of aminoisobutyrate at 0.1mm-aminoisobutyrate. Because of the decreased maximal response to insulin, the concentration of insulin that elicits a half-maximal response of aminoisobutyrate uptake is decreased. Hepatocytes from diabetic animals, like those from starved rats, have significantly greater basal rates of aminoisobutyrate uptake; whereas the maximal absolute insulin response is the same as control cells, the percentage response is smaller. These cells bind significantly more insulin than do control cells. The increase in insulin binding is reflected in a shift to the left of the dose–response curve for insulin-stimulated uptake of aminoisobutyrate. These studies indicate that there is no insulin resistance with regard to uptake of aminoisobutyrate in hepatocytes from older obese rats. Furthermore, the insulin resistance observed in hepatocytes from starved rats occurs despite an increase in the number of receptors per unit surface area and cannot be explained by alterations in the interaction between insulin and its receptor. The enhanced insulin binding per unit surface area, however, is reflected in the shift to the left of the dose–response curve for insulin. This is also true for hepatocytes from diabetic animals, in which insulin binding per cell is increased.     

Time dependent effects of dexamethasone on serum insulin level and insulin receptors in rat liver and erythrocytes

The effects of glucocorticoid excess on regulation of insulin receptors were investigated in dexamethasone-treated rats. Glucocorticoid excess was produced by administration of dexamethasone (0.5 mg/100 g b.w.) 30 min, 4, 12, 18, 24, 42 or 70 h before experiments. This treatment caused time-dependent changes of glucose and insulin concentration in blood, as well as in amounts of specific insulin binding and insulin receptors of liver cells and erythrocytes. The time intervals in which dexamethasone produced the increase in insulin concentration were accompanied with decrease in insulin binding to receptors in membranes of liver cells, while significant changes in insulin binding to receptors of erythrocytes were not observed under the same experimental conditions. The effect is maximal 18 and 42 h after dexamethasone treatment that increase insulin blood level by about 85% and 60%, respectively. Receptor analysis revealed that changes in specific binding of insulin could be due to significant changes in amount of binding sites on cell surface rather than to mild alteration in receptor affinity. These findings suggest that besides the changes in insulin level, the alterations in insulin receptor number and affinity may play a major role in the states of altered insulin sensitivity which accompany glucocorticoid excess.     

Role of insulin and IGF1 receptors in proliferation of cultured renal proximal tubule cells.

We have used a murine proximal tubule cell line (MCT cells) to determine the presence and binding characteristics of insulin and IGF1 receptors and to correlate these parameters with the concentration-response relationships for ligand-induced cellular proliferation. Separate insulin and IGF1 receptors were identified by equilibrium binding assays. Half-maximal displacement of either peptide occurred at 3-10 nM; crossover binding to the alternate receptor occurred with a 10- to 100-fold lower affinity. Peptide effects on cellular proliferation were determined by measuring [3H]thymidine incorporation. Both insulin and IGF1 stimulate thymidine incorporation in a dose-dependent manner with similar increases above the basal level. The estimated half-maximal stimulation (EC50) occurred at 4 nM for IGF1 and 8 nM for insulin. A comparison of the receptor binding affinities with the dose-response relationships for [3H]thymidine incorporation reveals that each growth factor appears to be exerting its effect via binding to its own receptor. Therefore, in this cell line, physiologic concentrations of either insulin or IGF1 can modulate cellular growth. To our knowledge this is the first demonstration of a mitogenic effect which may be modulated by ligand binding to the insulin receptor in proximal tubule epithelia.     

Quantitative studies of the rate of insulin internalization in isolated rat hepatocytes.

We studied internalization of 125I-labelled insulin in isolated rat hepatocytes. Using the acidification technique, we were able to dissociate the ligand from its cell-surface receptors, and thus to separate internalized from surface-bound insulin. Because during the first 5 min of incubation of 125I-labelled insulin with freshly isolated hepatocytes there is no loss of internalized label, the ratio of the amount of internalized ligand to the amount of cell-surface-bound ligand may serve as an index of insulin internalization. Within the first 10 min of insulin's interaction with hepatocytes, the plot of the above ratio as a function of time yields a straight line. The slope of this line is referred to as the endocytic rate constant (Ke) for insulin and denotes the probability with which the insulin-receptor complex is internalized in 1 min. At the insulin concentration of 0.295 ng/ml, the Ke is 0.049 min-1. It is independent of insulin concentration until the latter exceeds 1 ng/ml. At the insulin concentration of 3.2 ng/ml, the Ke accelerates to 0.131 min-1. With the Ke being the probability of insulin-receptor-complex internalization, 4.9% of occupied insulin receptors will be internalized in 1 min at an insulin concentration of 0.295 ng/ml, and 13.1% of occupied insulin receptors will be internalized in 1 min at 3.2 ng/ml. When the insulin concentration decreases from 3.2 to 0.3 ng/ml, the Ke decreases accordingly. The half-time of occupied receptor internalization is 15.4 min at the lower insulin concentration and 5.3 min at the higher insulin concentration.     

Studies on the regulation of insulin receptors in cultured BALB/3T3 fibroblasts

The level of [¹²⁵I]insulin binding to BALB/ 3T3 fibroblasts was low in growing cells and high in stationary cells. Since frequent changes of medium (every 2 h) did not modify the hormone binding of the stationary cells, it is unlikely that serum factors directly regulate the number of insulin receptors. Cells were grown to different densities by plating them in different concentrations of serum. Insulin binding was low in dense cultures maintained actively growing by high serum concentration, while binding was high in sparse cultures which were growth-arrested due to serum depletion. Thus, cell density does not directly regulate the insulin receptors. The growth status of the cells is the only factor that explains consistently the variations of insulin binding in these and previous [1, 2] experiments. Synchronization of the cells by two different methods did not show a reproducible cellcycle dependence for the insulin receptors.     

Insulin but not IGF-I is required for the maintenance of the adipose phenotype in the adipogenic cell line 1246

Summary The mouse adipogenic cell line 1246 which possesses both insulin and insulin-like growth factor I (IGF-I) receptors was used to investigate the role of IGF-I and insulin on the proliferation of adipocyte precursors and their differentiation into mature adipocytes. Results indicate that both insulin and IGF-I stimulate the proliferation of the 1246 adipocyte precursors with IGF-I being slightly more potent than insulin. Dose-response studies indicated that both polypeptides acted at physiological concentrations corresponding to binding to their own receptors. In contrast, comparison of insulin and IGF-I capacity to stimulate terminal adipose differentiation indicated that only insulin was active when added at physiological concentrations. IGF-I could not stimulate adipocyte differentiation except at supraphysiological concentrations (100 ng/ml and above) permitting its binding to the insulin receptors on 1246 cells. Time course study of expression of early and late markers of adipose differentiation indicated that the induction of markers such as adipose differentiation-related protein (ADRP), lipoprotein lipase (LPL) and fatty acid binding protein (FAB) took place even in the absence of insulin. However, the level of early and late differentiation markers decreased to a level below the one found in undifferentiated cells when cells had been maintained in the absence of insulin after differentiation had been initiated. These data indicate that although insulin is not necessary for the early onset of the adipose differentiation program, it is stringently required for the maintenance of the adipocyte phenotype and cannot be substituted by IGF-I.     

Increased glucagon receptors in chronically hypersomatotrophic and hyperglucagonemic rats

The effect of increased levels of growth hormone on glucagon binding by isolated hepatocytes and on the cellular cyclic AMP response to glucagon was evaluated in rats bearing growth hormone-secreting tumor (Mt-T-W15) and in rats treated with rat growth hormone. An increased binding, due to an increased number of receptors, was observed in both groups of animals. Glucagon binding did not correlate with plasma glucagon levels, suggesting a failure of down regulation, possibly due to an effect of growth hormone and insulin on the number of receptors. Tumor-bearing and growth hormone-treated rats had larger hepatocytes so that, when hormone binding was expressed in terms of square micrometer of membrane surface, it appeared decreased. When the tumor was removed the increase in the number of glucagon receptors per cell persisted, even though the average cell size returned toward normal. It is suggested that this retention of the receptors may have been the result of continuing hyperinsulinism. Basal cAMP levels were elevated in hepatocytes of tumor-bearing and growth hormone-treated animals, possibly due to cell hypertrophy. On the other hand, the maximum cAMP response to glucagon was not altered by the experimental procedures. A negative effect of insulin on cAMP accumulation may explain this apparent paradox. Indeed, hepatocytes isolated from rats following tumor removal, but with continuing hyperinsulinemia, had a lower maximum cAMP response, even though the glucagon binding per cell or per unit of cell surface was increased.     

Hormone receptors. 7. Characteristics of insulin receptors in a new line of cloned neonatal rat hepatocytes

1. A new line of cloned, differentiated rat hepatocytes (RL-PR-C) was evaluated for its usefulness as an in vitro system for studying the regulation of the insulin receptor. 2. Insulin rapidly reversibly and specifically bound to RL-PR-C hepatocytes. Binding of tracer 125I-labeled insulin, which was competitively inhibited by native insulin as well as by proinsulin and analogs of insulin and proinsulin in proportion to their biological activity, was not influenced by glucagon, corticotropin, or human growth hormone. Anti-insulin receptor serum from a patient with Acanthosis Nigricans Type B competed with 125I-labeled insulin for binding to cell surface sites. 3. Trypsinization destroyed insulin binding sites, but these were restored by incubation under growth conditions; a 75% restoration of binding sites was achieved by one cell population doubling. 4. RL-PR-C hepatocytes responded to insulin binding by an increase in glycogen synthesis from glucose. The insulin effect was maximal at 85 nM, but was detectable at lower, more physiological, concentrations. 5. Chronic exposure (for at least 3h) of hepatocytes to insulin (10(-10)--(10(-8) M) reduced by up to 60% the number of binding sites for insulin (down-regulation). Down-regulation was prevented by cycloheximide at concentration (10 micron) sufficient to inhibit markedly protein synthesis from tracer isoleucine. Recovery from down-regulation induced by native insulin at 10(-7 M or lower concentrations was complete by 18 h under growth conditions. 6. Although RL-PR-C hepatocytes spontaneously transform after about 90 population doublings, no significant differences between normal and transformed cells were observed in insulin binding characteristics and in interaction of cells with anti-insulin receptor serum. However, transformed cells exhibited a substantially reduced (maximum of 20%) down-regulation response to insulin. 7. RL-PR-C rat hepatocytes appear, for these reasons, to be a useful model system for studying the regulation of the insulin receptor.     

Insulin receptor processing as a function of erythrocyte age. A kinetic model for down-regulation

The effects of cell aging on insulin binding and on insulin receptor processing in human erythrocytes were studied. Erythrocytes were found to exponentially lose equal proportions of both high and low affinity receptors as a function of age. The affinities of remaining surface receptors did not change significantly. The maximum extent of insulin receptor down-regulation that could be induced decreased linearly with age over the range studied. Together with dose-response and time course studies, these age-related changes in insulin binding and receptor down-regulation were used to develop a kinetic model in which receptor internalization is a function of surface receptor concentration. The ability of the model to predict the behavior of a heterogeneous population suggests that changes in receptor processing with age may be attributed to changes in the surface receptor concentration.     

Reciprocal changes of insulin and glucagon receptors in primary cultured hepatocytes

The specific [125I]insulin binding to primary cultured hepatocytes was significantly greater than that to freshly isolated hepatocytes. Low affinity insulin binding sites in cultured cells were 6-fold greater in number than those of freshly isolated cells without a significant change in high affinity sites. However, both sensitivity (insulin concentration for half maximum stimulation) and responsiveness (% of increase above the basal level) to insulin for the stimulation of ODC activity were similar for isolated and cultured cells indicating an important role of high affinity sites in the insulin action. On the other hand, the specific [125I]glucagon binding to cultured cells was significantly decreased. Low affinity glucagon binding sites in cultured cells decreased by about 50% in cultured cells without a significant change in high affinity sites. Both sensitivity and responsiveness to glucagon for the stimulation of ketogenesis from palmitate also decreased as compared with those of isolated cells, indicating an important role of low affinity sites in the glucagon action. These results indicate that insulin and glucagon receptors were reciprocally changed in cultured cells, as compared with isolated cells.     

Growth hormone and insulin binding to isolated hepatocytes in the genetically dwarf mouse

The interaction of growth hormone with its specific receptors in dwarf mice was investigated. (1) The interaction of 125I-labeled human growth hormone with isolated mouse liver cells is a specific, time-dependent and saturable process. Hepataocytes of male and female dw/dw mice bound only 10-20% as much growth hormone per unit of cell surface area as those of their litter mates. Scatchard analysis suggested that this decrease in binding was due to a decreased number of receptor sites in th liver cell of the dwarf mouse. (2) In contrast to the marked decrease in growth hormone receptors, the binding of insulin is higher in dwarf mice than in litter mates, at low hormone concentration. (3) Competition and stoichiometric studies indicate that growth hormone and prolactin bind to the same type of binding site in female and male mouse hepatocytes. These results indicate that dwarfism in this animal was associated with a loss in the number of growth hormone binding sites. The decrease in growth hormone receptors and the increase in insulin receptors correlate well with the respective biological activity of these two hormones.     

Insulin receptors and bioresponses in a human liver cell line (Hep G-2)

A newly developed human hepatoma cell line, designated Hep G-2, expresses high-affinity insulin receptors meeting all the expected criteria for classic insulin receptors. 125I-insulin binding is time-dependent and temperature-dependent and unlabeled insulin competes for the labeled hormone with a half-maximal displacement of 1-3 ng/ml. This indicates a Kd of about 10(-10) M. Since Scatchard analysis of the binding data results in a curvilinear plot and unlabeled insulin accelerates the dissociation of bound hormone, these receptors exhibit the negative cooperative interactions characteristic of insulin receptors in many other cell and tissue types. Proinsulin and des(Ala, Asp)-insulin compete for 125I-insulin binding with 4% and 2%, respectively, of the potency of insulin. Anti-(insulin receptor) antibody competes fully for insulin binding. The two insulin-like growth factors, multiplication-stimulating activity and IGF-I are 2% as potent as insulin against the Hep G-2 insulin receptor. Furthermore, Hep G-2 cells respond to insulin in several bioassays. Glucose uptake, glycogen synthase, uridine incorporation into RNA and acetate incorporation into lipid are all stimulated to varying degrees by physiological concentrations of insulin. In addition, these cells 'down-regulate' their insulin receptor, internalize 125I-insulin and degrade insulin in a manner similar to freshly isolated rodent hepatocytes. This is the first available human liver cell line in permanent culture in which both insulin receptors and biological responses have been carefully examined.     

Insulin regulation of insulin-like growth factor action in rat hepatoma cells

We have reported previously that insulin causes a complete but reversible desensitization to insulin action in rat hepatoma HTC cells in tissue culture, and that this insulin resistance is mediated by postbinding mechanisms rather than receptor down-regulation (Heaton, J. H., and Gelehrter, T. D. (1981) J. Biol. Chem. 256, 12257-12262). We report here that insulin causes a similar desensitization to the induction of tyrosine aminotransferase by the insulin-like growth factors IGF-I and IGF-II isolated from human plasma, and by multiplication-stimulating activity, the rat homologue of IGF-II. The results of both competition-binding studies and affinity cross-linking experiments indicate that insulin-like growth factors (IGFs) bind primarily to IGF receptors rather than to insulin receptors. The low concentrations at which these factors induce transaminase is consistent with their acting primarily via IGF receptors. This is confirmed by experiments utilizing anti-insulin receptor antibody which both inhibits 125I-insulin binding and shifts the concentration dependence of insulin induction of tyrosine aminotransferase to the right. This same immunoglobulin does not inhibit 125I-multiplication-stimulating activity binding and only minimally inhibits 125I-IGF-I binding. Anti-insulin receptor antibody also does not significantly shift the concentration dependence for the IGFs, suggesting that IGFs induce transaminase by acting via IGF receptors. Although insulin down regulates insulin receptors, it does not decrease IGF-I or IGF-II binding. We conclude that insulin causes desensitization of HTC cells to IGFs by affecting a postbinding step in IGF action, which may be common to the actions of both insulin and insulin-like growth factors.