Reduced concentration of serum growth hormone-binding protein in children with idiopathic short stature. National Cooperative Growth Study.

One possible explanation for the growth failure in children with idiopathic short stature (ISS) is reduced peripheral responsiveness to GH. In Laron syndrome, growth retardation is caused by GH resistance due to GH receptor (GH-R) defects, which are associated in most cases with absent or low serum concentrations of the GH-R-related GH-binding protein (GHBP). We tested the hypothesis that some children with ISS have reduced serum concentrations of GHBP and that this may reflect decreased sensitivity to GH. A ligand-mediated immunofunctional assay was used to measure biochemically active GHBP in serum from 1549 children, including 773 controls, 573 with ISS, 107 with GH deficiency (GHD), and 96 with Turner syndrome (TS). Ages ranged from 1-17 yr. Serum GHBP concentrations in children with GHD, ISS, and TS were converted to SD scores and compared to controls by analysis of variance. In male and female ISS subjects, approximately 90% had GHBP concentrations below the age- and sex-adjusted mean for controls, and 20% had GHBP concentrations below the normal range. The mean serum GHBP SD score was lower in both males and females with GHD (-0.6) or ISS (-1.2) than in controls (both P < 0.005). The mean for ISS males was significantly lower than that for GHD males (P < 0.0001). The mean GHBP SD score for girls with TS (-0.3) did not differ significantly from that of the control females. The decreased levels of serum GHBP in some children with idiopathic short stature suggest that these children could have a defect at the level of the GH-R.

Growth hormone-binding protein levels: studies of children with short stature.

A high-affinity growth hormone-binding protein (GHBP) in serum is derived from the extracellular domain of the GH receptor. In an attempt to investigate the differences in GHBP levels in various conditions of poor growth, we measured GHBP levels by two methods--an Ultrogel chromatographic technique and a ligand-mediated immunofunctional assay (LIFA). The following three groups of children were studied: Turner's syndrome (n = 7), idiopathic and/or familial short stature ([ISS] n = 15), and organic or idiopathic hypopituitarism (n = 19). All groups were similar in age (Turner's syndrome, 10.1 +/- 0.9 years; ISS, 10.0 +/- 0.7; hypopituitarism, 11.5 +/- 1.0) and height SEM score (Turner's syndrome, -2.9 +/- 0.3; ISS, -3.0 +/- 0.4; hypopituitarism, -2.3 +/- 0.4). Their values were compared with those values of GHBP in healthy controls of similar age. Immunofunctional assay values for GHBP were as follows: Turner's syndrome, 235.4 +/- 26.0 pmol/L; ISS, 122.4 +/- 11.0; and hypopituitarism, 157.1 +/- 23.0. These results were significantly different in subjects with ISS and hypopituitarism as compared with a group of healthy controls between the ages of 9 and 12 years (N = 255; GHBP = 287.9 +/- 10.9 pMol/L; P < .001 compared with both ISS and hypopituitarism). Similar changes were found using Ultrogel chromatography. This difference in GHBP levels is still significant even when more stringent criteria are applied to define hypopituitarism (ie, peak GH responses to stimuli < 6.0 ng/mL, instead of < or = 10 ng/mL originally).(ABSTRACT TRUNCATED AT 250 WORDS)

Growth hormone-binding proteins of human serum: developmental patterns in normal man.

The binding of GH by a low affinity binding protein (LA-BP) was measured from birth into adulthood and compared with binding of a high affinity binding protein (HA-BP) in human serum. Pooled serum samples for each year of age from birth to 16.5 yr were formed from 2500 separate samples and assayed for both binding proteins using a Sephadex chromatographic method. Individual samples in this age range and those from adults were assayed in a similar manner. Binding of [125I]GH was minimal by both binding proteins in cord blood (binding by LA-BP, 2.77 +/- 0.32%; binding by HA-BP, 2.58 +/- 0.35% mean +/- SEM). A 4-fold increase to maximal binding for LA-BP occurred by the age of 5 yr and remained relatively constant through adolescence, except for a transient decrease at puberty. From 16.5-20 yr of age, binding by LA-BP decreased to a level no greater than that seen at birth. Binding by the HA-BP, which increased 6- to 8-fold reached maximal binding between 23-25 yr of age. Binding by HA-BP did not decrease in adults between the ages of 20-30 yr. Whereas pregnancy increased LA-BP activity, GH binding by HA-BP was unaltered.

Hormone and receptor studies: relationship to linear growth in childhood and puberty.

Preliminary data suggested different patterns of hormonal control of linear growth in males and females. To better define these patterns, serum samples were collected from 75-125 boys and a similar number from girls for each year of age between 3-16 yr (n = 2416). Fewer samples were collected from 2-yr-olds, newborns, and adults (n = 151). Samples for each age were aliquoted, combined, and assayed for GH, GH-binding protein (GHBP), insulin-like growth factor-I, and testosterone. GHBP, expressed as a percentage of the [125I]GH bound, increased yearly in males and females, with no relationship to the secretion of sex hormones. The increase in binding of [125I]GH and, by inference, GH receptors occurred at a greater rate between the ages of 2-10 yr than between 10-16 yr (in terms of absolute binding, 1.2 +/- 0.11% vs. 0.38 +/- 0.04% yearly; P less than 0.001). In each age group, however, the increase in GHBP exhibited a strong positive correlation with linear height (r = 0.96-0.98 in males; r = 0.92-0.99 in females). Before puberty, GH and insulin-like growth factor-I concentrations were consistently greater in females. Between 10-16 yr of age, height velocity (centimeters of growth per yr) correlated strongly with GH in girls (r = 0.86), but did not correlate with GH in boys of a similar age (r = -0.13). The major pubertal growth spurt in males strongly correlated with a rise in serum testosterone concentration beginning at age 11 yr (r = 0.92). Small peaks of GH secretion before and after the major period of accelerated growth in males possibly prolonged the major growth phase, but did not initiate it.

Diabetes mellitus. A perspective on management.

Diabetes mellitus results in multiple complications including over 10,000 new cases of blindness each year in the United States. The evidence that consistent good control significantly reduces these complications is overwhelming, i.e., a return to normal or close to normal glycemic state. This evidence is reviewed. Particularly noteworthy are data showing that when hemoglobin A1c values are consistently less than 8.4% (with 8.0% being the upper limit of normal), only 2.9% of subjects with existing retinopathy progress to a more severe retinopathy, whereas with higher hemoglobin A1cs, there is a progressive increase of severe retinopathy. Multiple biochemical hemodynamic and endocrine processes which appear abnormal in diabetes, return to normal when euglycemia is reestablished. Despite such evidence, physicians often resist conclusions regarding proper management of the diabetic state. The reasons are briefly reviewed.

The growth hormone-insulin-like growth factor I axis: studies in man during growth.

Several major differences are noted between males and females in their patterns of growth at puberty. Accelerated pubertal growth in both males and females depends upon the integrity of the GH-receptor system. In males, acceleration of growth results primarily from enhanced sensitivity of the GH-receptor-IGF I system to GH brought about by testosterone. Whether testosterone itself is responsible for this observation is still unclear. Perhaps the initial GH, IGF I peak present in males and absent in females occurs at the time when sleep-related rises of gonadotropins and testosterone begin just prior to puberty. Though the pygmy data certainly supports a relationship between testosterone and the GH-receptor-IGF I axis, the undisputed tall stature of eunuchs remains a puzzle. It is possible that the maturing male gonad secretes another growth factor and/or growth inhibitor in conjunction with testosterone and that it is this unidentified factor which modulates growth. At any rate, acceleration of growth in males results from sensitization or the GH-receptor-IGF I system while growth acceleration in females results almost solely from increased secretion of GH and not sensitization of the system.

Growth hormone-binding protein: II. Studies in pygmies and normal statured subjects.

The serum concentrations of a specific GH-binding protein, derived from the GH receptor, were assayed in sera from 62 African pygmies and 101 normal statured controls. Samples were assayed in the absence and presence of excess GH using 2 separatory procedures. Interassay variability for samples was corrected by a standard reference pool of sera from adults assayed with all unknown samples. Results were expressed as specific binding relative to this standard. The mean percent relative specific binding for GH increased with age in normal-statured controls throughout childhood and adolescence. Relative specific binding for GH was 37.0 +/- 2.0% (mean +/- SEM) in control subjects between the ages of 1-5 yr (mean age, 2.9 yr) and increased progressively to 93.0 +/- 7.0% in young adults (mean age, 23 yr). The relative specific binding of GH by serum from pygmies did not exceed 30.1 +/- 3.4% of the control adult standard at any age period (P less than 0.001), and there was no progressive age-related increase in binding. The decrease from normal binding was minimal in pygmies during childhood (29%), but the decrease from normal was 60-70% in adolescents and adults. Thus, short stature in pygmies probably results not from an absolute deficiency of GH receptors per se, as in Laron dwarfism, but from a failure of cellular GH receptors to increase in a normal manner. This is most compatible with a change in regulating expression of the GH receptor gene, rather than a structural defect in the coding sequence of the GH receptor gene.

Insulin-like growth factor-II in nonislet cell tumors associated with hypoglycemia: increased levels of messenger ribonucleic acid.

The role of insulin-like growth factor-II (IGF-II) in the hypoglycemia associated with nonislet cell tumors is controversial. In this study we have addressed this question by measuring the IGF-II mRNA levels in extracts of these tumors. Hybridization of a 32P-labeled IGF-II cDNA to a Northern blot of RNA from three nonislet cell tumors associated with hypoglycemia (a hemangiopericytoma, fibrosarcoma, and malignant mesenchymal tumor) demonstrated six hybridizing bands, 6.8, 5.6, 4.7, 3.6, 2.6, and 2.1 kilobases in length. These bands were similar to those described by others in a range of tumors and normal tissues. Tissue IGF-II mRNA levels were quantitated using a solution hybridization/RNase protection assay. IGF-II mRNA levels in the tumors were similar to the level present in one line of human hepatoblastoma-derived Hep G2 cells, 5- to 6-fold higher than that in another line of Hep G2 cells, and 2- to 3-fold higher than that in term placenta. In contrast, little or no IGF-II mRNA was detected in a nonfunctioning islet cell adenoma or normal spleen. There was no evidence for amplification of the IGF-II gene in the one tumor in which it was sought. These data suggest that nonislet cell tumors associated with hypoglycemia produce large amounts of IGF-II mRNA and that this IGF-II mRNA appears to be the product of an IGF-II gene, which is apparently normal in the region encoding mature IGF-II peptide.

Molecular forms of serum insulin-like growth factor (IGF)-binding proteins in man: relationships with growth hormone and IGFs and physiological significance.

Insulin-like growth factor-I (IGF-I) and IGF-II are associated in the blood with specific binding proteins (BPs), forming complexes that elute in gel filtration with estimated mol wt around 40 and 150 kD. The latter appears to be under GH control. Five molecular forms of BP (41.5, 38.5, 34, 30, and 24 kD) have been identified by Western blotting using 125I-labeled IGF. All five forms are present in the smaller complexes, but only the 41.5- and 38.5-kD forms are found in the larger complexes. In this study immunoblotting showed that the 41.5- and 38.5-kD forms were recognized by antibodies directed against the GH-dependent BP purified from human plasma, and the 30-kD form was recognized by antibodies directed against the BP purified from amniotic fluid. The 34- and 24-kD forms proved to be immunologically unrelated to the other three. In sera with large quantities of the 41.5- and 38.5-kD forms, an additional band was often observed immediately ahead of the migration front of the 30 kD band. This was recognized by the anti-GH-dependent BP antibody and probably corresponds to a degradation product of the 41.5- and 38.5-kD BPs. Serum 41.5- and 38.5-kD BPs have been found to be elevated in acromegaly, where GH hypersecretion causes increased IGF-I levels, and diminished in cases of genetic or idiopathic GH deficiency and defects of the GH receptor (Laron's syndrome), where both IGF-I and IGF-II are decreased, as well as in Pygmy adults and children who have isolated IGF-I deficiency. In all of these conditions, the proportions of the 34- and 30-kD forms were inversely related to those of the 41.5- and 38.5-forms. Under treatment, the BP profiles tended to return to normal. In cases of GH deficiency caused by a tumor, the BP profiles resembled those of hypopituitary or normal serum, depending on whether IGF levels were diminished or normal. It, therefore, seems that BP synthesis is coordinated with IGF-I synthesis and may not be directly GH dependent. The results of neutral pH gel filtration analysis of hypopituitary (idiopathic and tumoral) and normal sera point to a relationship between the levels of circulating IGFs and those of the 150-kD IGF-BP complex whose binding units are the 41.5- and 38.5-kD BPs. It, therefore, seems that the 150-kD complex controls the bioavailability of IGF-I and IGF-II.(ABSTRACT TRUNCATED AT 400 WORDS)

Insulin-like growth factor secretion by human B-lymphocytes: a comparison of cells from normal and pygmy subjects.

Freshly isolated human B-lymphocytes from eight subjects and Epstein-Barr virus-transformed human B-lymphocytes from seven subjects were examined for their capacity to secrete insulin-like growth factor-I (IGF-I) and IGF-II and for their capacity to respond to human GH. Similar studies were conducted with Epstein-Barr virus-transformed lymphocytes collected from six African pygmies. When transformed B-lymphocytes from normal stature subjects were cultured for 3 weeks in RPMI-1640 medium (6 x 10(3) cells/75-cm2 flask at seeding), significant amounts of IGF-I, but no IGF-II, were produced. GH (150 ng/mL) significantly increased for control cells the amount of IGF-I produced at each sampling interval compared to that by unstimulated cultures (P less than 0.05 at 1 week; P = 0.005 at 3 weeks). At 3 weeks, cell counts of cultures compared were 4.13 +/- 0.39 X 10(6)/mL for unstimulated cells and 4.23 +/- 0.87 X 10(6)/mL for GH-stimulated cells. IGF-I production at this time interval by unstimulated cells was 2.8 +/- 2.3 ng/mL, and that by GH-stimulated cells was 12.3 +/- 2.5 ng/mL (P = 0.005). Cell multiplication rates of control cultures were increased in 1 week by GH stimulation [GH stimulated, [16.7 +/- 22.0 X 10(4) cells, unstimulated, 5.73 +/- 4.1 X 10(4) cells; (mean +/- SD); n = 14; P less than 0.01]. Similar results occurred with GH studied at a lower concentration of 10 ng/mL for 3 weeks. Freshly isolated B-lymphocytes did not secrete IGF-I and II after 5 days of culture with GH. Cultures established from cells derived from pygmies produced significantly less IGF-I (4.24 +/- 2.62 ng/mL) when stimulated with 150 ng/mL GH than cultures of cells from normal stature subjects (12.3 +/- 2.5 ng/mL; 0.005 less than P less than 0.01). The cultures compared had a similar cell density. A similar significant difference in IGF-I secretion occurred between cultures of pygmy and control cells stimulated with 10 ng/mL GH. These data are consistent with previous in vivo studies in which pygmies failed to increase IGF-I and exhibit metabolic responses to exogenous GH.

Low levels of high-affinity growth hormone-binding protein in African pygmies.

The cause of growth hormone resistance and short stature in African Pygmies in unknown. Low levels of insulin-like growth factor 1 that fail to respond to growth hormone suggest a possible deficiency of growth hormone receptors. The high-affinity growth hormone-binding protein is a fragment of the growth hormone receptor and may be an indicator of the number of receptors in tissues. We measured growth hormone-binding activity in plasma from 20 pygmies and 12 control subjects (7 white Americans and 5 non-Pygmy black Africans of normal stature). Growth hormone binding to the high-affinity binding protein was significantly reduced in the Pygmies as compared with the controls (mean +/- SD, 6.50 +/- 2.33 percent vs. 12.95 +/- 3.95 percent bound per 160 microliters of plasma; P less than 0.001); however, there was substantial overlap between the values for Pygmies and controls. Growth hormone binding to the low-affinity binding protein was similar in Pygmies and control subjects. We conclude that Pygmies have low levels of high-affinity growth hormone-binding protein in their plasma, which may indicate a reduced number of growth hormone receptors in their tissues. The finding may help explain the resistance to growth in Pygmies, but there may be additional reasons, related to the receptors or not, for their short stature.

The growth hormone receptor gene in the African pygmy.

Multiple studies indicate that the pygmy responds poorly to GH. A GH-binding protein, which is a fragment of the GHR is markedly diminished in pygmy serum, suggesting a defect in the GHR. Full analysis of the molecular nature and function of the GHR gene is incomplete, but with one enzyme BstNI restriction fragment length, polymorphism was noted in pygmies located in an intron. This was an uncommon finding in other non-pygmy populations. The alteration noted may be in linkage disequilibrium with an effective mutation responsible for changes in the GHR of the pygmy. This is most likely in the regulatory regions of either the 3' or 5' extensions.

Insulin-like growth factor-I modulates endothelial cell chemotaxis.

Insulin-like growth factor-I (IGF-I) significantly promoted chemotaxis in three separate endothelial cell lines at concentrations ranging from 4 to 150 ng/ml. When added to minimum essential medium (MEM) supplemented with 1% fetal calf serum, IGF-I augmented chemotaxis to a level equalling the maximal chemotactic effect obtainable with MEM supplemented with 10% fetal calf serum (FCS 10%). The data are compatible with a potential role for IGF-I in neovascularization.

Separation of the insulin-like growth factor-binding proteins in plasma and purification of the larger molecular weight species.

Methods were developed for purification of the high mol wt (150K) insulin-like growth factor (IGF)-binding protein and its acid stable (70K) component from human plasma. High mol wt IGF-binding protein was highly purified by chromatography of Cohn IV-1 fraction of human plasma on Concanavalin A-Sepharose followed by chromatography on IGF-I-Sepharose. The acid-stable component of the high mol wt IGF-binding protein was purified to near homogeneity by chromatography of Cohn IV-1 fraction of human plasma on Con-A Sepharose, followed by chromatography on Sephadex G-50 at pH 2.5 and subsequent chromatography on IGF-I-Sepharose. Both fractions obtained after IGF-I-Sepharose chromatography were capable of binding [125I]IGF-I and gave a single protein band of 79,000 mol wt, when subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Coomassie blue staining), suggesting that the large mol wt species may be a dimer of identical or iso-mol wt subunits. Three minor contaminants of less than 5% each were detected upon subsequent silver staining. These methods represent important tools that should aid in furthering our understanding of the role of the IGF carrier proteins in the actions of IGF-I and IGF-II.

Insulin-like growth factors in pygmies. The role of puberty in determining final stature.

We measured the serum concentrations of insulin-like growth factors (IGF) I and II and testosterone in pygmy children, adolescents, and adults, as well as in controls, to determine more precisely the role of these factors in controlling growth. We had previously shown that growth hormone levels were normal in pygmies. Prepubertal pygmy children and controls did not differ in linear growth or in serum concentrations of IGF I and II. In pygmy adolescent boys, the mean (+/- SEM) serum concentration of IGF I was only one third that in control adolescents, who were similar to the pygmies in age and Tanner stage of development (154 +/- 22 vs. 435 +/- 37 ng per milliliter; P less than 0.01). A similar difference in IGF I concentration was observed in girls (278 +/- 18 vs. 570 +/- 25 ng per milliliter; P less than 0.01). IGF II and testosterone levels were normal in all groups. There was a significant difference in growth between controls and pygmies only during puberty. There was a marked acceleration of growth in the controls during adolescence, but such an acceleration was absent or blunted in the pygmies. These findings suggest that the short stature of adult pygmies is due primarily to a failure of growth to accelerate during puberty. We postulate that IGF I is the principal factor responsible for normal pubertal growth and that testosterone does not accelerate growth appreciably in the absence of an increase in the level of IGF I.

Changes in insulin-like growth factors I and II and their binding protein after a single intramuscular injection of growth hormone.

The concentrations of insulin-like growth factors I and II (IGF-I and IGF-II) and their binding proteins in serum were measured in 10 GH-deficient patients before and after a single 6-IU injection of GH. Serum IGF-I concentrations were initially low, increased significantly by 8 and 24 h, and decreased to pretreatment levels 48 and 72 h after GH administration. Serum IGF-II concentrations also were low initially and did not increase by 8 and 24 h, but were, however, significantly higher 48 and 72 h after GH administration. In GH-deficient patients before GH administration, binding of IGF-I or IGF-II to serum proteins was restricted primarily to proteins of 50K mol wt. Little or no binding to proteins of 150,000 mol wt was found. By 8 and 24 h after GH injection, IGF-I, but not IGF-II, bound primarily to a protein(s) of 150K mol wt, as in normal subjects. IGF-II remained bound to a 50K mol wt protein. By 48 and 72 h after administering GH, however, the binding pattern was reversed, and IGF-II, but not IGF-I, bound predominantly to a protein(s) of 150K mol wt. Our data demonstrate both a temporal dissociation in the responses of IGF-I and IGF-II to GH and a similar temporal dissociation in the binding of IGF-I and IGF-II to the large mol wt (150K) binding protein. This dissociation, particularly the latter, may provide a means for better characterization of protein fractions in binding IGF, particularly in terms of specificity.

Insulin-like growth factors in vitreous. Studies in control and diabetic subjects with neovascularization.

Vitreous and serum were obtained at the time of vitrectomy from 23 diabetic subjects with proliferative retinopathy and from 8 nondiabetic subjects. The mean concentration of IGF-I in vitreous from diabetic patients with neovascularization was 6.3 +/- 0.93 versus 2.7 +/- 0.96 ng/ml. Chi-square and rank analysis indicated that higher concentrations of IGF-I occurred in diabetic vitreous (P less than 0.01 by both analyses). IGF-II concentrations in vitreous of control and diabetic subjects were not significantly different. A positive correlation existed between the concentrations of IGF-I and IGF-II in vitreous and their concentrations in serum in diabetic subjects, but not in control subjects. When vitreous concentrations of IGF-I were calculated for diabetic subjects studied previously with rapid acceleration of retinal disease, these concentrations varied from 20 to 30 ng/ml. The concentrations of IGF-I in the vitreous of most diabetic subjects with severe neovascularization are thus in the range known to stimulate cellular differentiation and growth in several systems. Whether they do so in the eye, and thus contribute to the development of retinopathy, remains to be determined.

Insulin-like growth factors in patients with nonislet cell tumors and hypoglycemia.

Insulin-like growth factors (IGF) in serum from 3 patients with nonislet cell tumors and hypoglycemia were measured by radioimmunoassay and by means of a rat liver membrane assay (specific for IGF II). The concentration of IGF II by receptor assay was greater than normal when the samples were initially assayed, but subsequently decreased 48% to 80% over the next 8 to 15 weeks. In the same serum samples, concentrations of both IGF I and IGF II by radioimmunoassay were consistently less than normal. In all 3 patients growth hormone (GH) responses to intravenous arginine were depressed and in 2 of the 3 patients, a GH-dependent 150 K serum protein carrier of IGF was absent. None of these abnormalities were seen in 7 patients with insulinomas and chronic hypoglycemia. The data suggest that some patients with nonislet cell tumors and hypoglycemia produce a receptor-active, nonimmunoreactive IGF-II-like material. This material appears (a) more labile than normal IGF II and (b) capable of inhibiting GH secretion. The latter effect may decrease immunoreactive IGF I and II, as well as decrease the GH-dependent 150 K protein carrier of these factors. The extreme lability of the IGF-II-like material produced by tumors probably explains the previous contradictory reports on this topic.

Effect of psychosocial factors on success in a program of self-glucose monitoring.

We examined prospectively the relationship of psychosocial factors to glycemic control in a program of self-glucose monitoring (SGM). Measured intelligence (IQ), educational level, and socioeconomic status (assessed by the Two-Factor Index of Social Position) were determined in 25 patients who were followed during 6 months of self-glucose monitoring. Personality categories, reflecting degrees of psychological disturbance, were assigned using the Minnesota Multiphasic Personality Inventory (MMPI). None of the measured psychosocial variables correlated significantly with initial Hgb A1 values. In contrast, after 6 months of SGM, Hgb A1 levels correlated significantly with both socioeconomic status (r = 0.42, p less than 0.05) and educational levels (r = -0.42, p less than 0.05). Hemoglobin A1 levels also correlated significantly with the recorded frequency of SGM (r = -0.65, p less than 0.01), a measure of patient compliance. No significant correlation between IQ and Hgb A1 levels was seen, either initially or during follow-up. High A1 values differed significantly among groups classified by MMPI testing. Patients with severe psychological abnormalities had higher (p less than 0.05) mean Hgb A1 levels. We conclude that psychosocial factors, but not measured intelligence, have an important bearing on patient success in a program of SGM.