| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Laboratoire d’Explorations Fonctionnelles Endocriniennes, Hôpital d’Enfants Armand Trousseau, Assistance Publique des Hôpitaux de Paris (AP-HP), and INSERM, U-515, Hôpital Saint-Antoine, 75571 Paris, France
Address all correspondence and requests for reprints to: Prof. Yves Le Bouc, Laboratoire d’Explorations Fonctionnelles Endocriniennes, Hôpital Armand Trousseau, 26 avenue du Dr. Netter, 75571 Paris Cedex 12, France. E-mail: lebouc{at}st-antoine.inserm.fr
Abstract
The IGFs, IGF-I and IGF-II, regulate fetal growth by activating IGF type 1 receptors (IGF-IR). We aimed to quantify the binding of IGF-I to its cognate receptors in intrauterine growth-retarded children (IUGR). We measured the affinity of the erythrocyte IGF-IR and the number of IGF-IR receptors in 17 children with retarded growth (mean height, -2.7 SD), normal levels of GH, and a history of idiopathic intrauterine growth retardation (height at birth, -10 to -2 SD; mean, -3.1 SD). These children had reduced receptor affinity (Kd = 0.47 nM; P < 0.01) and more receptors per cell [binding capacity (Bmax) = 11.7 binding sites/cell; P < 0.05)] compared with control children (Kd = 0.32 nM; Bmax = 7.8 binding sites/cell). Moreover, the distributions of Kd and Bmax suggested that there were two groups of IUGR children. Group 1 included subjects with normal receptor binding function (Kd = 0.36 nM; Bmax = 8.2 sites/cell) and normal levels of circulating IGF-I. Group 2 comprised children with low receptor affinity (Kd = 0.56 nM) and increased receptor number (Bmax = 14.7 sites/cell). This group showed significantly decreased IGF-I levels (-2.1 SD; P < 0.01). We investigated these IGF-IR binding parameters in two additional groups of growth-retarded children (Turner syndrome and patients with chronic renal failure), in whom the IGF-I axis was not believed to be the primary cause, and found that Kd and Bmax were normal or nearly normal. We also measured IGF-IR binding parameters in 4 Seckel syndrome patients with IUGR and severely retarded growth (mean height, -7.9 SD). Their receptor affinity was reduced, but not statistically different, from that in controls, and their receptor number was normal, whereas IGF-I levels were elevated. Our results suggest heterogeneous alterations in IGF-IR binding function in IUGR patients.
NEONATES WHOSE BIRTH weight and height are more than 2 SD below the mean of their corresponding gestational age group are defined as intrauterine growth-retarded (IUGR) children (1). These subjects have an increased risk of perinatal mortality and postnatal growth failure. Furthermore, recent epidemiological studies reported that IUGR individuals have an increased risk of cardiovascular and metabolic diseases (2, 3, 4, 5, 6). Most of the numerous children born after IUGR, catch up during the first year of life (7), although about 10–20% reach a final height that is far below their genetic target height (8). There are various causes of IUGR, and they are often not well understood, although evidence shows that the nutritional and endocrine relationships between the fetus and the mother are important (9, 10, 11).
The role of the IGF system [IGF-I, IGF-II, and IGF type 1 receptor (IGF-IR)] in fetal maturation and growth promotion has been clearly demonstrated in transgenic mice (12, 13, 14). Null mutations in the IGF system components, alone or in combination, lead to profound embryonic and postnatal growth deficiency. The most marked effects on growth occur after IGF-IR is knocked out or after the double knockout of IGF-IR (or IGF-I) and IGF-II (45% and 30% of normal body weight, respectively) (13, 14, 15). These and other studies showed that IGF-I, IGF-II, and their common receptor have key roles in the regulation of fetal growth.
In humans, very few cases of IUGR with postnatal growth failure have been attributed to genetic abnormalities of the IGF system. One case involving a 15-yr-old boy with severe growth failure (-6.9 SD) and mental retardation associated with a homozygous partial deletion of the IGF-I gene has been reported (16). The chromosomal location of the IGF-IR gene is 15q26.3, and there have been descriptions of about 10 cases of deletions in the distal long arm of chromosome 15, and 33 patients with ring chromosome 15, which often results in terminal 15q deletion (17, 18). Those studies, however, do not prove the role of the IGF-IR gene defects in IUGR, as deletions of the long arm of chromosome 15 result in losses of other genes as well. Consequently, these chromosomal defects are often associated with other abnormalities, including developmental delay, organ hypoplasia, triangular-shaped head, low set ears, and brachydactylia.
To assess the role of IGF-IR in the pathogeny of human growth, studies have been carried out on African Efe Pygmies, the population with the smallest adult height. In comparison with a North American or an African rural control population, Pygmy children are small at birth and growth retarded at 6 months (-2.7 SD), and their growth retardation is progressive (-4.2 SD at 5 yr) (19). Hattori et al. (20) found a marked reduction in the number of IGF-IR on the cell surface of immortalized T lymphocytes from Pygmies. They also found decreased IGF-I receptor gene transcription and reduced receptor signaling, but normal affinity for IGF. This suggested that human stature is partly genetically controlled by the expression of IGF-IR (20).
IGF-IR is present on the surface of red blood cells (21, 22, 23) and conserves its kinase properties on circulating erythrocytes (24). However, the number of receptors decreases dramatically during red blood cell maturation [from 500 binding sites/cell on erythroid colony-forming units (25) to less than 10 functional binding sites/cell on mature erythrocytes]. Nevertheless, IGF-I binding can be accurately quantified in erythrocytes, and binding studies have suggested that IGF-I binding on circulating erythrocytes reflects ligand-receptor interactions in cells of other organs (26, 27, 28, 29, 30). Consequently, blood samples can be used to evaluate variations in IGF-IR binding function. We aimed to quantify aspects of IGF-I binding to its cognate receptor on erythrocytes in a population of IUGR children with normal GH levels but short stature.
Experimental Subjects
Table 1
shows the characteristics
of the experimental subjects. We studied 17 children between 4–16 yr
of age [1 nonpubertal and 1 pubertal girl (stage 4); 12 nonpubertal
and 3 pubertal boys (stage 5)] with short postnatal stature (below -2
SD) associated with idiopathic IUGR. Their mean birth
height was -3.1 SD (range, -10.0 to -2.0), and their
mean birth weight was -2.1 SD (range, -4.2 to -0.8)
(31). All of the patients had normal circulating
concentrations of GH (>10 ng/ml in the ornithine stimulation test).
Fifteen of the 17 idiopathic IUGR patients have been treated with GH
[Genotonorm (Pharmacia Biotech, Piscataway, NJ), Maxomat
(Sanofi-Synthelabo, Paris, France), and Umatrope
(Eli Lilly & Co., Inc., Indianapolis, IN); 1.2–1.4
U/kg·wk]. The treatment duration was adapted to individual
circumstances. Seven had received GH treatment before this study, but
treatment had been stopped at least 18 months before receptor analysis.
None of the subjects was undergoing GH treatment at the time of this
study. Two of the GH-treated children responded very well and finally
reached a height inside the normal range (-0.9 and -1.0
SD). We also analyzed 5 severely growth retarded children
(-7.1 ± 2.5 SD) after IUGR, whose additional
clinical characteristics revealed that 1 patient was suffering from
Russell-Silver syndrome (severe intrauterine and postnatal growth
retardation, skeletal asymmetry of limbs, and clinodactyly of the fifth
finger) (32), 2 were suffering from Seckel syndrome (IUGR
with severe postnatal short stature, mental deficiency, clinodactyly of
the fifth finger, microcephaly, prominent nose, and other craniofacial
abnormalities) (33), and 2 others had clinical features of
Seckel syndrome, but their clinical picture was incomplete (1 patient
was apparently not mentally retarded, and the other did not present
clinodactyly). Three patients of the syndromic group had received GH
treatment before study, but treatment had been stopped at least 18
months before receptor analysis. These 2 groups of patients
(nonsyndromic and syndromic IUGR) were analyzed separately because of
the supposed heterogeneity of causes of growth retardation. We compared
the IUGR group with 2 other pathology-associated growth abnormalities:
Turner syndrome (n = 6), leading to growth failure but normal
levels of GH, and medically treated chronic renal failure (CRF; n
= 5), leading to growth retardation with normal or elevated levels of
IGF-I. Control data were collected from a group of 8 children, aged
6–17 yr (3 nonpubertal and 2 pubertal girls; 3 pubertal boys), and
from 12 normal adults (18–67 yr old). All subjects, or their parents
if they were less than 18 yr of age, signed an informed consent form
before participation, according to the recommendations of the
institutional ethics committee (Direction de la Recherche Clinique de
l’Assistance Publique des Hôpitaux de Paris).
|
Hormones and chemical products
Recombinant human IGF-I (rhIGF-I), rhIGF-II, and des-(1–3)IGF-I [rhdes-(1–3)IGF-I] were purchased from GroPep Pty. Ltd. (Adelaide, Australia), and insulin was purchased from Sigma-Aldrich Corp. (St. Louis, MO). IGF-I was used as both unlabeled ligand and tracer. 125I was obtained from Amersham Pharmacia Biotech (Little Chalfont, UK). All chemicals were purchased from Sigma-Aldrich Corp., except MgCl2, EDTA triplex, dihydrate trisodium citrate, chloramine-T, and NaCl, which were obtained from Merck & Co., Inc. (Darmstadt, Germany), Ficoll-Paque (Pharmacia Biotech), and BSA (Miles, Inc., Elkhart, IN).
Preparation of erythrocytes
Patients fasted for 1 night, and then 3 ml blood were collected into EDTA tubes for blood cell counts (CELL-DYN 4000) and plasma IGF-I determination, and 10–15 ml blood were collected into tubes containing 3 ml ice-cold citrate-phosphate-D-glucose buffer (26.3 g/liter dihydrate trisodium citrate, 3.3 g/liter monohydrate citric acid, 2.2 g/liter monohydrate monosodium phosphate, and 23.2 g/liter anhydrous D-glucose in sterile water). The stability of IGF-IR binding parameters during storage of the blood samples in ice-cold citrate-phosphate-D-glucose was tested until 13 d after sampling in three independent samples and stayed unchanged for at least the first 7 d. Blood samples were stored for a maximum of 5 d before analysis. As previously described (34, 35), blood was centrifuged (10 min, 400 x g, 4 C), the plasma-containing supernatant was discarded, and the erythrocyte pellet was diluted in saline buffer to a final volume of 20 ml and layered on 15 ml of a Ficoll-Paque mixture. After centrifugation (20 min, 400 x g, 20 C), the supernatant was removed. The erythrocyte pellet was resuspended in 20 ml saline buffer, and this procedure was repeated once more. The resulting erythrocyte pellet was washed twice in a large excess of ice-cold incubation buffer (50 mM HEPES, 50 mM Tris, 10 mM MgCl2, 2 mM EDTA, 10 mM D-glucose anhydrous, 50 mM NaCl, 5 mM KCl, and 0.1% BSA, pH 8.0, at room temperature). The final erythrocyte suspension contained 7.3 ± 0.1 x 1012 cells/liter, with 1.2 ± 0.1% (0.2–3.3%) reticulocytes and 0.1% leukocytes. The trypan blue dye exclusion method showed that more than 98% of the cells were viable after 24 h at 4 C.
IGF-I iodination and receptor binding assays
A modified chloramine-T method (36) was used to
iodinate rhIGF-I, which was then purified on Sephadex G-25 and Ultrogel
AcA54 columns. The specific activities ranged from 163–225 µCi/µg.
The binding assays were carried out in the above incubation buffer.
Each reaction mixture contained between 2.5 and 3.5 x
1012 red blood cells/liter and the iodinated
IGF-I in a final volume of 0.5 ml. To analyze the effect of cell
concentration on [125I]IGF-I binding to
erythrocytes, the red cell concentration was varied from 0.5–4 x
1012 cells/liter. Competition assays were carried
out by adding between 10 and 60 pM
[125I]IGF-I (
10,000 cpm/tube) to 10
different concentrations of the unlabeled growth factor. The amount of
nonspecific binding was determined in the presence of 0.2
µM IGF-I. After 18 h at 4 C (time necessary to reach
steady state conditions; see Results for details), 200 µl
of each sample were transferred to prechilled microfuge tubes
containing 200 µl incubation buffer and 200 µl dibutylphtalate and
centrifuged (3 min, 8000 x g, 4 C). The supernatant
was discarded. The tip of the microfuge tube, containing the cell
pellet, was counted in a 
-counter (LKB 1209). For the
association experiments, 30 pM
[125I]IGF-I was added, the reaction was stopped
by centrifugation of 200 µl aliquots at given time points, and the
tubes were treated as described above. For dissociation experiments,
the radioligand (30 pM) was incubated for 24
h with red blood cells at 4 C. Dissociation was initiated by the
addition of unlabeled rhIGF-I (final concentration, 0.2
µM), and residual binding was measured at given
time points after centrifugation and counting.
Data analysis and statistics
The dissociation constant (Kd) and the maximal number of binding sites (Bmax) for the radioligand and the unlabeled ligand were determined by Scatchard analysis of saturation isotherms and competition experiments, respectively. In competition experiments, the concentration of unlabeled hormone that decreased the binding of [125I]IGF-I by 50% (IC50) was obtained according to the Hill equation (37). The inhibitory constant or apparent affinity (Ki) value was calculated from the Cheng-Prusoff equation (38), assuming a competitive inhibition Ki = IC50/(1 + L/Kd), where L represents the radioligand concentration. All values for the characterization of [125I]IGF-I binding to its erythrocyte receptor are the mean ± SEM of at least three experiments carried out in triplicate on blood samples from a control subject. The binding parameters were determined once in triplicate for both the patients and the control group and are expressed as the mean ± SD. Data from competition experiments were analyzed using LIGAND software (39), to calculate the Ki, the Kd, and the Bmax. The number of binding sites per erythrocyte was calculated from the binding capacity, the red blood cell count, and Avogadro’s number: binding sites/cell = [receptor concentration (mol/liter)/erythrocyte count (cell/liter)] x 6.023 x 1023. StatView (Abacus Concepts, Inc., Berkeley, CA) was used for statistical analysis. Comparative analysis between the different groups was performed using the Mann-Whitney test, with 95% confidence intervals. P < 0.05 was considered statistically significant.
RIA for IGF-I
Plasma IGF-I concentrations were measured by the method previously described (40). To separate IGFs from their binding proteins, plasma samples were gel-filtered on Ultrogel AcA54 columns in acetic acid. RIA was used to assay IGF-I using the antihuman IGF-I antibody, which was provided by J. Closset (Liege, Belgium). Recombinant human IGF-I (Ciba-Geigy, Basel, Switzerland) was used as both standard and tracer. Samples were studied at three concentrations and in duplicate. Intra- and interassay coefficients of variation were 4.8% and 10%, respectively. Plasma IGF-I values are expressed as the SD with regard to age and Tanner pubertal stage (41) or age only for adults. Reference IGF-I concentrations were previously established in our laboratory and are expressed as nanograms per ml: children less than 2 yr, 80 ± 20 (n = 24); 2–5 yr, 160 ± 50 (n = 17); pubertal stage 1 (P1) more than 5 yr, 210 ± 65 (n = 29); P2 girls, 360 ± 65 (n = 9); P3 to P4 girls, 425 ± 90 (n = 21); P2 boys, 330 ± 50 (n = 13); P3 to P4 boys, 600 ± 130 (n = 16); P5 less than 20 yr, 405 ± 70 (n = 14); adults 20–29 yr, 310 ± 55 (n = 28); 30–39 yr, 275 ± 50 (n = 28); 40–49 yr, 245 ± 55 (n = 28); 50–59 yr, 215 ± 50 (n = 10); and 60–69 yr, 185 ± 45 (n = 76).
Determination of GH levels
The immunoradiometric assay kit from Pharmacia Biotech (St. Quentin en Yvelines, France) was used to measure plasma GH levels under ornithine stimulation test. A value of 10 ng/ml was considered normal.
Results
Characterization of erythrocyte IGF-IR
We first analyzed the relationship between
[125I]IGF-I binding and red blood cell
concentration. Specific binding increased linearly with cell
concentrations up to 4 x 1012 cells/liter
(not shown). We found that a red blood cell concentration between 2.0
and 3.5 x 1012 cells/liter allowed 6–10%
total specific binding of the iodinated ligand. Steady state binding
conditions at 4 C were determined from the association kinetics of 30
pM [125I]IGF-I (Fig. 1A). Equilibrium was reached after
approximately 6 h of incubation and lasted until at least 24
h. For all subsequent binding experiments the incubation time was,
therefore, fixed at 18–20 h. In dissociation experiments half of the
30 pM [125I]IGF-I had dissociated
after 4 h of incubation at 4 C (not shown).
|
). Specific binding of
[125I]IGF-I to human erythrocytes was inhibited
by unlabeled rhIGF-I in a dose-dependent manner (Fig. 1C). rhIGF-I
showed the highest inhibition (Ki = 0.31 ±
0.05 nM) compared with rhdes(1, 2, 3)IGF-I (0.67 ±
0.13), rhIGF-II (2.9 ± 1.0), and insulin (297 ± 75). These
pharmacological profiles confirmed the specificity of
[125I]IGF-I binding to the IGF-IR. To ensure
that red cell membrane-associated IGF-binding proteins (IGFBPs) did not
interfere, we carried out saturation experiments in the presence of 0.5
nM rhdes(1, 2, 3)IGF-I (not shown). Due to the deletion of the
three N-terminus amino acids, this analog has an affinity for the
IGF-IR similar to that of rhIGF-I, but binds poorly to IGFBPs
(42). The results obtained in saturation experiments using
rhdes(1, 2, 3)IGF-I and rhIGF-I as competitive inhibitors were identical.
These experiments confirmed the absence of any significant amount of
IGFBPs or other binding sites from the erythrocyte preparations. In all competition assays the total binding was between 6–17%, with a specific binding of approximately 80%. It is noteworthy that the total specific binding did not depend on the concentration of reticulocytes in the erythrocyte preparation over the range of 0.2–3.3% (not shown). Scatchard analysis of competition experiments revealed that Kd values were very similar to the Ki calculated from competition curves (r2 = 0.73; P < 0.001; n = 55). Therefore, we analyzed the affinity (Kd) and the binding site number (Bmax) of IGF-IR binding. The intra- and interassay coefficients of variation were determined from six independent experiments in one control subject and were 4.2% and 15%, respectively.
Analysis of IGF-I binding parameters
The binding parameters Kd and
Bmax (Table 2 and
Fig. 2) did not differ significantly
between the children and adult control groups. In contrast to the
control groups, the Kd and
Bmax values of individuals from the IUGR group
were scattered and twice as widely spread as those in the control
groups. Receptor affinity was significantly reduced (P
< 0.01), and receptor number was significantly increased
(P < 0.05) compared with the control values. In fact,
some of the IUGR patients overlapped with the control group, whereas
the others had Kd and Bmax
values above +2 SD, such that among 17 IUGR
children, 8 exhibited an elevated Kd and 9 an
increased Bmax, whereas the remaining subjects
were within the respective normal range. In comparison, CRF children
did not show any difference in terms of affinity compared with controls
and only a slight, but not significant, increase in binding capacity.
In patients with Turner syndrome, who fail to grow but have normal GH
levels, receptor number was normal, and affinity was normal or close to
normal (Table 2 and Fig. 2). In IUGR children, when we tested the
correlation between Kd or
Bmax (or Kd x
Bmax, see below) and weight, birth weight, or
body mass index, only IGF-IR affinity and birth weight were
significantly correlated (r2 = 0.18;
P = 0.031, using Spearman’s test), indicating that
patients with elevated Kd showed a tendency to
low birth weight.
|
|
). To
increase the discriminative strength of receptor characterization, we
calculated the Kd x Bmax
index. In the resulting Kd x
Bmax distribution, nine IUGR patients were well
above the normal range (Fig. 3B). In comparison, the CRF patients had
normal Kd x Bmax indexes,
and only two patients with Turner syndrome had indexes above +2
SD. Together, these results suggested that IUGR
group 2 (Table 3) included an elevated
number of individuals with impaired IGF-IR function. Interestingly,
growth retardation was identical in both groups (Table 3). Moreover, we
did not find significant differences between IUGR groups 1 and 2
concerning age, pubertal stage, birth weight, current weight, or BMI.
To verify whether this possibly impaired IGF type 1 receptor function
had any feedback on the GH-IGF-I axis we measured circulating IGF-I
levels in all patients and controls.
|
|
Fasting values for plasma IGF-I in controls and patients (Table 3
and Fig. 4) are expressed as
SD from reference IGF-I levels (see Materials
and Methods). The IGF-I values for the control groups
were similar to reference values. In IUGR group 1, IGF-I
levels were only slightly lower than those in the control children
(IGF-I, -0.5 ± 1.7 SD). Whereas in IUGR
group 2 IGF-I levels were significantly lower (IGF-I, -2.1 ± 1.5
SD; P < 0.01) than in the
control children. IGF-I levels in IUGR group 2 were also significantly
lower than those in group 1 (P = 0.05). We found a
similar difference for the IGF-I levels measured before GH treatment
(-0.7 SD for group 1 and -1.8
SD for group 2). IGF-I levels were not
statistically different in subjects with CRF or Turner syndrome
compared with controls.
|
Plasma IGF-I values and binding parameters in syndromic IUGR
children are summarized in Table 4. The
patient with Russell-Silver syndrome exhibited binding parameters that
increased the Kd x Bmax
index. The four Seckel and Seckel-like syndromic patients had elevated
IGF-I levels, except patient 2, who suffered from malnutrition and had
very low IGF-I levels. These four children had a reduced IGF type 1
receptor affinity (Kd = 0.46 ± 0.16
nM), which was close to statistical significance
(P = 0.08), but they had no increase of their receptor
number (Bmax = 8.1 ± 1.9 sites/cell). Their
Kd x Bmax indexes were
within the normal range. However, their
Kd/Bmax ratio was
significantly higher
(Kd/Bmax = 55.8 ±
10.5) than that of the control children
(Kd/Bmax = 40.0 ±
3.1; P = 0.01) or the 17 idiopathic IUGR children
(Kd/Bmax = 41.9 ±
9.8; P < 0.05).
|
Fifteen of the 17 IUGR patients underwent GH treatment (Table 5). We separated them according to their
binding characteristics into GH-treated IUGR group 1 and group 2 and
analyzed the evolution of their body height under GH treatment.
Although the average 1-yr growth responses in groups 1 and 2 were not
significantly different, the variability in the individual growth
responses was much greater in group 2. This is compatible with the idea
that group 2 is composed of patients heterogeneous with respect to
their IGF-I receptor function. Three of the 5 syndromic IUGR patients
underwent GH treatment. They displayed the lowest responses to GH with
regard to growth velocity or height gain.
|
Although IUGR with postnatal growth retardation is relatively frequent, its etiology is often unknown. Surprisingly, abnormalities are not yet systematically sought in the IGF-IR, although this is one of the principal regulators of somatic growth (13, 14). Moreover, there are a number of mutations in its closest structural homolog, the insulin receptor, some of which are clinically relevant (43). The expected symptoms of hypothetical IGF-IR insufficiency include growth retardation during embryonic and postnatal development (44). As IGF-I receptor binding is the first step of IGF-I action, we assessed the prevalence of impaired IGF-IR binding function in idiopathic intrauterine growth failure. Firstly, we studied the binding parameters of the erythrocyte IGF-IR in a group of 17 growth-retarded children after idiopathic IUGR. We determined the affinity and number of IGF-IR-binding sites on the surface of red blood cells. The combined IUGR children showed significantly lower receptor affinity and greater receptor number per erythrocyte than control populations. The distribution of the values suggested that IUGR children formed two groups, with either normal or impaired receptor binding.
It was necessary to show that the IGF-I binding assay was technically sound, because previous characterizations of human erythrocyte IGF type 1 receptors revealed differences in the number of binding sites and their affinity for IGF-I. Our control values for the number of binding sites (5–12 sites/red blood cell) were similar to those published by Izumi et al. and Catanese et al. (5–15 sites/cell) (23, 24). Other studies found between 20–70 sites (22, 45, 46, 47) or between 1–4 sites/red blood cell (26, 30). Our affinity control values (between 0.21–0.50 nM) were similar to several published reports (0.14–0.30 nM) (26, 30), but slightly weaker affinities have also been reported (0.70–0.80 nM) (22, 23). Interestingly, studies that found high numbers of binding sites reported that these sites had low affinities (45, 46, 47). These differences are mainly due to systematic differences in data analysis and to the use of different IGF-I radioligands. We showed that membrane-associated IGFBPs did not interfere with receptor ligand binding. Moreover, the age of the red blood cell did not have any effect on the results, which is consistent with the study by Morris et al. (45), who did not find increased ligand binding in normal adults even when reticulocyte counts were above 3%. We conclude that the IGF-I receptor binding parameters determined in this study reliably reflect the binding of IGF-I to mature erythrocyte IGF-IR.
A total of 22 patients with IUGR were studied. Five subjects, however, had severe growth retardation syndromes and were analyzed separately. The remaining 17 subjects with idiopathic IUGR were divided into 2 groups according to their receptor binding parameters. Subjects of IUGR group 1 exhibited normal Kd and Bmax and normal circulating IGF-I levels. Our analysis suggested that their growth retardation was a priori not linked to abnormal IGF-I/IGF-IR binding or to receptor levels. In IUGR group 2, in contrast, we observed reduced IGF-IR affinity, suggesting that at least half of the 17 patients potentially have some underlying molecular alteration of the receptor. Interestingly, IUGR group 2 children had elevated receptor number and significantly reduced serum IGF-I levels. Reduced IGF-I levels alone may explain the up-regulation of the IGF-IR, as shown by Eshet et al. (26, 27) in growth deficiencies associated with low circulating IGF-I (GH deficiency or Laron dwarfism). Treatment of these conditions with exogenous IGF-I (28) or GH leads to increased IGF-I concentration (29) and normalization of the number of IGF-IR, suggesting that the IGF-I-binding sites are regulated in vivo according to plasma levels of IGF-I. Moreover, our findings agree with the work of Hizuka et al. (22), who also reported an inverse correlation between IGF-I levels and IGF-IR number in normal adults and acromegalic and hypopituitaric subjects. Low levels of circulating IGF-I may also be the outcome of insufficient nutrition. However, although IUGR patients are often reported to have a poor appetite, the body mass index was normal in IUGR groups 1 and 2 (data not shown). Alternatively, increased Bmax may also result from decreased ligand binding to the receptor, which would, therefore, be internalized less frequently and could eventually accumulate on the surface of the red blood cells. Additional experiments, including the exploration of receptor binding kinetics, tyrosine kinase activity, and signaling events downstream of the IGF-IR, possibly using biopsy-derived materials, are necessary to answer these questions. Another question is why the observed up-regulation of receptor number (twice the normal value) was not sufficient to produce efficient postnatal growth. The analysis of the growth velocity of these children during and height gain after exogenous GH treatment provided some information on the heterogeneity of IUGR group 2. The fact that the average response to GH was not significantly different from that of group 1, as could be expected, may also reflect that GH does not only act via the IGF-I/IGF-IR axis to promote growth and that individuals may behave differently in this regard.
We also studied a third group of IUGR patients, the syndromic patients,
including four patients with Seckel or Seckel-like syndrome. Except for
patient 2 (see Table 4) who was receiving enteral tube feeding, all of
the patients had very high circulating IGF-I levels. Despite this, they
showed the strongest and most progressive postnatal growth retardation
of all subjects in this study. Our results suggest that IGF-I receptor
binding was altered, but different from the IUGR group 2 patients, as
syndromic patients showed a very low response to GH treatment compared
with either IUGR group 1 or 2. To conclude, one needs a larger group of
syndromic IUGR patients. Interestingly, we observed two Turner syndrome
patients with elevated Kd x
Bmax product and normal IGF-I levels. Decreased
responses to IGF-I have been observed in lymphocytes, monocytes, and
fibroblasts from Turner syndrome patients, suggesting the possibility
of partial end-organ resistance to IGF-I (48, 49).
Recently, point mutations have been detected in exon 2 of the IGF-IR gene in IUGR patients with postnatal growth failure and increased IGF-I (50). This preliminary report of abnormalities in the IGF-IR did not, however, include the necessary functional demonstration of abnormal signaling through the receptor. Such studies should be extended to confirm the implication of IGF-IR abnormalities in the pathogeny of IUGR. Mutations in the insulin receptor have shown that single amino acid modifications reducing receptor affinity for insulin may have severe physiological consequences (43). Reduced receptor ligand binding was also described in IGF-IR after in vitro directed mutagenesis of its ligand binding determinants (51). Thus, it is reasonable to expect that the changes that we observed in the IGF-IR binding parameters may reflect structural alterations of the receptor and explain IUGR and subsequent growth failure. Knowledge of genetic mutations could ultimately refine the diagnosis of growth retardation.
Acknowledgments
We are indebted to Prof. J. Battin, Prof. J. J. Baudon, Prof. B. Leheup, Dr. M.-C. Raux-Demay, and Dr. B. Esteva for their support during this study. We thank Rémy Christol, Solange Quiniot, and Catherine Daux for their expert technical assistance.
Footnotes
This work was supported by University Paris VI Pierre et Marie Curie (UPRES-EA 1531) and INSERM.
Abbreviations: Bmax, Binding capacity; CRF, chronic renal failure; IGF-IR, IGF type 1 receptor; IGFBP, IGF-binding protein; IUGR, intrauterine growth-retarded; P1, pubertal stage 1; rhIGF-I, recombinant human IGF-I.
Received February 15, 2001.
Accepted July 18, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  F. Galland, P. Kamenicky, H. Affres, Y. Reznik, D. Pontvert, Y. Le Bouc, J. Young, and P. Chanson McCune-Albright Syndrome and Acromegaly: Effects of Hypothalamopituitary Radiotherapy and/or Pegvisomant in Somatostatin Analog-Resistant Patients J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4957 - 4961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Raile, J. Klammt, A. Schneider, A. Keller, S. Laue, R. Smith, R. Pfaffle, J. Kratzsch, E. Keller, and W. Kiess Clinical and Functional Characteristics of the Human Arg59Ter Insulin-Like Growth Factor I Receptor (IGF1R) Mutation: Implications for a Gene Dosage Effect of the Human IGF1R J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2264 - 2271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kawashima, S. Kanzaki, F. Yang, T. Kinoshita, K. Hanaki, J.-i. Nagaishi, Y. Ohtsuka, I. Hisatome, H. Ninomoya, E. Nanba, et al. Mutation at Cleavage Site of Insulin-Like Growth Factor Receptor in a Short-Stature Child Born with Intrauterine Growth Retardation J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4679 - 4687. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Lee, J. W. Kendig, and J. R. Kerrigan Persistent Short Stature, Other Potential Outcomes, and the Effect of Growth Hormone Treatment in Children Who Are Born Small for Gestational Age Pediatrics, July 1, 2003; 112(1): 150 - 162. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
). IUGR children
with Kd or Bmax above the normal range were
defined as group 2 (
).