| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Faculty of Sciences (O.H.), Bar Ilan University, Ramat Gan 52900, Israel; Genetic Institute (O.H., Y.H., S.A.-S.), and Pediatric Endocrine Unit (Y.T.-R.), Ha Emek Medical Center, 18101 Afula, Israel; Pfizer Inc. (M.P.W.), New York, New York 10017; Technion Faculty of Medicine (S.A.-S., Y.T.-R.), Haifa 31096, Israel; Pediatric Endocrine Unit (Z.Z.), Kaplan Medical Center, Rehovot and the Hebrew University of Jerusalem, Jerusalem, 76100, Israel; and Department of Community Health and Epidemiology (I.L.), Carmel Medical Center, Haifa 34362, Israel
Address all correspondence and requests for reprints to: Yardena Tenenbaum-Rakover, Director, Pediatric Endocrine Unit, Ha Emek Medical Center, Afula 18101, Israel. E-mail: rakover_y{at}clalit.org.il.
| Abstract |
|---|
|
|
|---|
Objectives: The objective of the study was to assess the phenotype-genotype correlation of subjects affected with IGHD II caused by a G6664A mutation in 34 affected members of two large families.
Design and Patients: Sixty-six subjects from two core families were included. The G6664A mutation among family members was determined by restriction fragment length polymorphism.
Results: Twenty-four of the 52 members from family 1 and 10 of 14 from family 2 carried the same G6664A mutation in a heterozygous state. The affected subjects in family 1 were significantly shorter [–2.6 vs. –0.1 SD score (SDS), P < 0.0001] and had significantly lower IGF-I serum levels (–1.9 vs. –0.5 SDS, P < 0.0001), compared with normal-genotype family members. The affected adults exhibited great variability in their stature, ranging from –4.5 to –1.0 (mean –2.8 SDS), with five members being of normal height (>–2 SDS). Twelve children were diagnosed with IGHD. Two affected children had normal peak GH levels, although one of these subsequently demonstrated GH insufficiency (6.5 and 3.7 ng/ml). The affected children from both families exhibited large variability in their height, growth velocity, delay in bone age (chronological age – bone age), age at diagnosis, peak GH response, and IGF-I levels.
Conclusions: These detailed phenotypic analyses show the variable expressivity of patients bearing a G6664A mutation, reflecting the spectrum of GH deficiency in affected patients, even within families, and the presence of additional genes modifying height determination. Our findings raise a new dilemma in the guidelines for the diagnosis of GH deficiency and the indications for GH therapy.
| Introduction |
|---|
|
|
|---|
It has been assumed that genetic etiology accounts for 30% of the cases of GHD and that it is more likely in consanguineous families or when a second case of GHD occurs in the same family. Four distinct forms of IGHD have been described (4, 5, 6, 7, 8). Type II is autosomal dominantly inherited and caused by splice-site mutations and nucleotide substitutions in the GH-1 gene (7, 8) (OMIM 139250). IGHD type II (IGHD II) has mainly been described in patients presenting different patterns of mutations in intron 3 splice sites that cause skipping of exon 3 and thus deletion of amino acids 32–71 (del32–71 GH) (9, 10, 11). In these individuals, even though the normal GH-1 allele is present, the level of secreted GH is severely reduced. Three different missense mutations have been described as causing milder IGHD II: P89L and V110F in exon 4 and R183H in exon 5 (3, 7, 8, 11, 12, 13, 14, 15, 16, 17). The G6664A mutation results in the substitution of arginine by histidine at position 183 (R183H) in exon 5 of GH-1. Several families from different ethnic origin with this mutation have been described to date (7, 8, 14, 15, 16, 17). The fact that arginine in position 183 is highly conserved in different species (18) suggests an important role in GH function. Several mechanisms have been proposed to explain how this G6664A mutation in GH-1 expresses dominant-negative interference of the wild type. It has recently been suggested that the mutated GH interferes with the intracellular transport and storage of wild-type GH in neuroendocrine cells and thus affects the rate of GH release from the secretory granules (6, 19, 20, 21).
Although several families with the G6664A mutation have been described, only one study (8) has shown a phenotype-genotype correlation. In the present study, we assessed the genetic, biochemical, and clinical data of patients affected with the G6664A mutation in the GH-1 gene in two unrelated families of disparate ethnic origin. The presence of 34 subjects affected with the G6664A mutation aged 1.1–81 yr among 66 family members enabled us to analyze the association between genotype and phenotype and evaluate the clinical outcome over the years.
| Subjects and Methods |
|---|
|
|
|---|
Core family 1. Family 1 is a large consanguineous kindred of Christian-Arab descent spanning five generations. The proband (V-13) was the male progeny of first-cousin parents, admitted at the age of 4.6 yr for evaluation of short stature. At first examination, his height was –4.0 SD score (SDS) and he exhibited typical features of GHD (frontal bossing, low nasal bridge, micrognathia, high pitched voice, truncal obesity, and acromicria). Bone age (BA) was delayed by 2.0 yr. Laboratory evaluation revealed normal thyroid and adrenal function, normal prolactin levels, and low IGF-I and IGF binding protein-3 values (–1.9 and –3.0 SDS, respectively). GHD was diagnosed based on low GH response to clonidine and arginine provocative tests (4.1 and 3.4 ng/ml, respectively). Pituitary magnetic resonance imaging (MRI) revealed a small adenohypophysis. GH treatment was begun at the age of 5.8 yr with a good response. A review of his family history revealed many short-statured members in the extended family. Fifty-two family members from five generations, 27 males, 25 females, aged 1.1–81 yr, with a high rate of consanguinity, were included in the study.
Core family 2. Family 2 is of Ashkenazi-Jewish origin with seven short children. Fourteen subjects from three nuclear families were included in the study (six males, eight females). The proband (III-3), aged 2 yr, was diagnosed with GHD based on low GH response in two provocative tests (2.8 and 3.4 ng/ml, respectively). MRI revealed hypoplastic adenohypophysis (almost invisible). GH therapy was initiated at the age of 3 yr with catch-up growth.
Methods
Height and weight were measured and expressed in SDS for age and sex (22). Blood samples were collected for hormone levels and DNA extraction. Medical history was reviewed. All affected children underwent two different GH provocative tests apart from two children from family 2 (III-1, III-2), who underwent only one provocative test. One adult subject (V-5) from family 1 underwent two provocative tests as well. GH provocative tests were performed with clonidine (0.125 mg/m2) and arginine (0.5 g/kg). Blood samples were taken to determine baseline levels for GH and IGF-I and at standard intervals during the tests for GH level. All provocative tests were performed at 0800 h after fasting. GHD was defined as a peak GH of less than 10 ng/ml in two different provocative tests according to the Israeli national guidelines (http://www.ghresearchsociety.org/bin/Consensus.asp). Data on subjects treated with GH were collected for up to 20 yr. Thyroid function, basal cortisol, prolactin, LH, and FSH were determined in all children at diagnosis and at each yearly follow-up. The data on the adult individuals were collected from their medical records and questionnaires. Children with GHD underwent brain MRI before GH therapy.
Hormonal study
GH and IGF-I were measured by using Immulite 2000 (Diagnostic Products Corp., Los Angeles, CA). IGF-I was standardized for age and sex with the normative data provided. GH assay coefficients of variation were 2.9% (intraassay) and 4.2% (interassay) at a concentration of 7.9 ng/ml and 4.6 and 5.7%, respectively, at a concentration of 3.7 ng/ml.
Molecular analysis
After obtaining informed consents, genomic DNA was extracted from peripheral blood mononuclear cells. In the probands of both families, all five GH-1-coding exons and the intronic flanking sequences of each exon were sequenced as previously described (8). Family members were screened for the presence of the specific mutation found in the index case (G6664A) by restriction fragment length polymorphism using DraIII restriction enzyme as previously described (8).
Haplotype analysis
To determine whether affected subjects carrying the G6664A mutation in the two families had a common ancestry, genotypes at three microsatellites (D17S794, D17S795, polymeropoulos) (8, 23) were determined by PCR with fluorescently labeled primers in an ABI-310 sequencer (Applied Biosystems, Victoria, Australia) and GeneMapper 3.1.2 software (Applied Biosystems). Haplotypes were constructed manually.
Statistical analysis
Data analysis was performed using the SPSS 14 statistical package (SPSS Inc., Chicago, IL). Comparison of height (SDS) and IGF-I (SDS) between the affected subjects and normal-genotype family members was performed by Student t test. Mann-Whitney test was used to compare birth length and weight in the affected vs. unaffected family members of family 1.
This study was approved by the institutional review boards and by the Israeli Ministry of Health.
| Results |
|---|
|
|
|---|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
A low stimulated GH peak was shown in most of the affected children (12 of 14), but not a total lack of response, consistent with partial GHD. The variability in stimulated GH response between individuals suggests that this type of GHD is a spectrum rather than an all-or-nothing phenomenon. This may explain, at least in part, the variability in GV in affected children and in heights among individuals, even within the same family. In one child, the maximum stimulated GH was initially normal (12.5 ng/ml), but within 1 yr was insufficient. This may indicate that GHD caused by the G6664A mutation is an evolving process and emphasizes the importance of repeating GH provocative tests over time in children that exhibit growth deceleration.
Two hypotheses may potentially explain the variability in height among our families affected subjects and the fact that some affected individuals were of normal height: 1) the GH-IGF-I axis is not the sole determinant of height, and thus, GH insufficiency may be compensated for by additional genes modifying an individuals final height; and 2) because the dominant-negative effect of R18H-GH on the wt GH is relatively mild, the presumed genetic modifications may occur in the somatotropic cell itself. In the presence of such an incompletely understood genetic modification of the detrimental IGHD II pathway, normal GH secretion may be possible for years or the entire growth period. The latter hypothesis is supported by our observation that GH insufficiency is not necessarily congenital and may develop over time.
Although fetal growth is independent of GH secretion and is mediated mainly by maternal nutrition and growth factors such as IGF-I and IGF-II (2), low birth length and weight have been reported in patients with GHD (24, 25). Postnatal growth during infancy is believed to be principally dependent on nutrition (2); however, some studies have shown impaired growth in the first year of life in GHD patients (24, 26, 27). Here we were able to show that affected subjects with the G6664A mutation in a same family have lower birth length and weight and a decline in height (SDS) in infancy, compared with unaffected siblings. Our findings of lower birth length and weight as well as growth deceleration during the first year of life in the affected family members indicate that the intact GH-IGF-I axis does indeed have some role in growth of the fetus and during infancy, supporting previous studies (24, 25, 26, 27).
Our unique observation in a pair of twins, one normal and one affected, demonstrated a difference of 2.8 SDS in height and 3.5 in weight at as early as 1.9 yr of age, emphasizing the influence of GHD on the growth process during infancy. GV was only mildly decreased in affected children of family 1, compared with a significant decline in GV in family 2, suggesting that other factors, in addition to GH, play a role in growth during childhood. Initiation of GH therapy in some of the affected children, despite normal GV (SDS), was based on our previous observation that height and GV may decline further with age, as was remarkably demonstrated in the affected twin (Fig. 3
).
In the present study, we found a hypoplastic adenohypophysis in four of nine children studied (44%). Another group reported 38% hypoplasia of the adenohypophysis in a cohort of patients with IGHD II with several GH-1 gene mutations, including a few subjects with the G6664A mutation (28). The exact mechanism by which the mutated GH affects the somatotrophs and reduces the volume of the anterior pituitary has not been established. In vitro studies have shown that the mutated G6664A-GH does not cause protein misfolding as was previously hypothesized (8) and does not show diminished bioactivity (19, 20) as has been shown in other mutations of the GH-1 gene (29, 30). It has been proposed that the G6664A-mutated GH exhibits its effect on the wild type by prolonged retention of GH molecules in the secretory granules, affecting the rate of GH release from the somatotrophs (20). Our finding of no sharing of common alleles among individuals from the two families by haplotype analysis suggests that this mutation occurred independently in each family and supports the previous assertion that G6664 is a hot spot in the GH-1 gene (8). In family 1, despite a high rate of consanguinity, none of the affected family members was homozygous for the G6664A mutation. In vitro studies in which G6664A-GH was transfected into AtT-20 cells have shown that the mutated GH does not impair its own regulated secretion, whereas AtT-20 cells coexpressing both the G6664A-mutant and normal GH show dramatically decreased, but not absent, GH-regulated secretion (19). This suggests that homozygous subjects may exhibit normal GH secretion and would therefore be expected to be of normal height.
No additional pituitary hormone deficiency was found in our patients, even in adulthood. Previous studies have found that pituitary hormone deficiencies may develop over time in patients with IGHD II caused by an exon-III-skipping mutation (31) or P89L-mutated GH (13). In contrast, our findings in G6664A-GH-mutated patients do not support development of secondary pituitary hormone deficiencies over time, suggesting that the progression to multi-hormone deficiency may be mutation specific. If this is indeed the case, genetic studies of GH-1 will have significant prognostic and therapeutic implications, rather than being merely of diagnostic or academic importance.
In summary, the large phenotypic variability observed, even in the same family, and the finding of normal stature affected members, can be explained by the fact that the GHD in G6664A mutation of GH-1 represents a spectrum rather than an all-or-nothing phenomenon and by the variable penetration of this heterozygous mutation among different individuals, reflecting the presence of additional genes modifying height determination. The finding of GHD in children that initially had a normal peak GH response suggests that repeating GH provocative tests is indicated in children that exhibit growth deceleration. These findings raise a new dilemma in the guidelines for the diagnosis of GHD and the indications for GH therapy.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online September 4, 2007
1 O.H. and Y.H. contributed equally to this publication. ![]()
Abbreviations: BA, Bone age; GHD, GH deficiency; GV, growth velocity; IGHD, isolated GH deficiency; IGHD II, isolated GH deficiency type II; MRI, magnetic resonance imaging; SDS, SD score.
Received March 27, 2007.
Accepted August 23, 2007.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |