| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Russian Academy of Medical Sciences, Department of Pediatrics, Endocrinology Research Center (O.V.F., V.A.P., I.I.D.), Moscow 117036; DNA-Diagnostics Laboratory (O.V.E., A.V.P.), Research Center for Medical Genetics, Moscow 115478; and Russian Academy of Sciences, Laboratory of Automatic DNA-Sequencing (A.B.P.), Engelgardt Institute of Molecular Biology, Moscow 119991, Russia
Address all correspondence and requests for reprints to: Olga V. Fofanova, Department of Pediatrics, Endocrinology Research Center, Russian Academy of Medical Sciences, 11 Dmitrija Uljanov Street, 117036 Moscow, Russia. E-mail: olga-vf{at}yandex.ru.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Several mutations in the GH-1 gene (GH-1), the first identified candidate gene for IGHD, have been reported in patients with sporadic and familial forms of the disorder in different ethnic groups (5). In Russia, mutation analysis in patients with IGHD has not been previously performed.
The goals of the present investigation were to study Russian patients with familial and sporadic IGHD for GH-1 gene mutations, and to determine the possible hot spots within the GH-1 gene responsible for the disorder.
| Patients and Methods |
|---|
|
|
|---|
The studied cohort consisted of 28 children (16 boys, 12 girls) with isolated GH deficiency from 26 unrelated families, admitted for diagnosis to the Department of Pediatrics, Endocrinology Research Center (Moscow, Russia). Ten patients from 8 families were familial cases [6 families with IGHD type II (IGHD II), 2 families with IGHD IB], and 18 patients were sporadic. The cohort for molecular analysis for GH-1 mutations met the following inclusion criteria: 1) severe growth retardation of early onset (before 1 yr of life); 2) total GH deficiency, with peak GH levels after stimulation [insulin tolerance test (ITT), and clonidine] < 5 µg/liter; 3) normal basal levels of free T4 (fT4), or total T3, and total T4, cortisol, and prolactin in blood.
In the studied cohort, mean (±SD) chronological age (CA) was 8.3 ± 4.3 yr, and bone age (BA) was 4.6 ± 3.6 yr. All patients were extremely short from early months of life, with height SD score (HSDS) of -3.2 ± 1.2 at the age of 1 yr and -4.3 ± 0.9 at the age of 2 yr. Over two thirds (67.8%) of the patients showed dysmorphic features of congenital GH-deficiency, including prominent forehead, saddle nose, and mid-facial hypoplasia. According to the pre-study data, 52% of documented patients (12/23) had low morning blood glucose levels (2.23.8 mmol/liter). In our study, mean fasting glucose levels were 3.6 ± 0.9 mmol/liter. Among the group with total GH deficiency, 74% of the patients showed stimulated peak GH levels less than 2 µg/liter (ITT: Actrapid HM, Novo Nordisk, Bagsvaerd, Denmark; 0.1 IU/kg body weight, iv; Clonidine, Organika, Novokuznetsk, Russia, 0.15 mg/m2 body square, orally).
Birth length in the group was 50.3 ± 2.8 cm, and there were no perinatal complications in 82.2% of the newborns. Five of 28 (17.8%) children, all with sporadic GH deficiency, had perinatal complications, including asphyxia in one patient, asphyxia and head trauma in two patients, head trauma in one patient, and trauma of the cervical part of the vertebral column in one patient.
In the studied group, all patients except two were prepubertal. Two patients with familial IGHD II developed spontaneous puberty before the first examination. Subsequently, long-term follow-up showed the development of spontaneous puberty in nine adolescents with pubertal bone age.
The study was approved by the Ethical Clinical Research Committee of the Endocrinology Research Center.
DNA isolation, PCR, and mutation analysis of the GH-1 gene
Genomic DNA was extracted from leukocytes using the phenol/chloroform method. Four DNA fragments of the GH-1 (Ref. 6 ; and GenBank accession no. J03071), covering exons 25 and their boundary regions were amplified using a thermal cycler (DNA Technology, Moscow, Russia), and analyzed by single-strand conformation polymorphism (SSCP) for possible DNA alterations. Exon 1 was not included in the study because of its small size and an absence of data regarding functional GH-1 mutations in this exon. The 30-µl reaction mixture consisted of 0.25 µM of each primer, 200 µM of each deoxynucleotide triphosphate, 67 mM Tris-HCl (pH 8.8), 16.6 mM (NH4)2SO4, 0.01% Tween-20, 26 mM (depending on each pair of primers) MgCl2, 11.5 U of Taq DNA polymerase (Fermentas, Vilnius, Lithuania), and 0.10.5 µg of genomic DNA. Sequences of the primers are shown in Table 1
. The PCR mixture was denaturated for 7 min at 94 C and cycled 32 times (exons 2, 4, and 5) or 34 times (exon 3): 45 sec denaturation at 94 C; 45 sec annealing at either 57 C (exons 2 and 5), 65 C (exon 3), or 60 C (exon 4); and a 45-sec extension at 72 C, followed by an 8-min extension at 72 C. DNA was visualized for the detection of SSCP by silver staining. Seven microliters of PCR product were mixed with 0.5 µl of 5 M NaOH, 0.5 µl of 0.5 M EDTA, and 4.5 µl deionized H2O. The mixture was heated for 15 min at 42 C. After 3 µl of formamide buffer (95% formamide, 0.5% xylol cyanol, 0.5% bromphenol blue) was added, the PCR product was electrophoresed on a 10% polyacrylamide gel (acrylamide: bisacrylamide ratio was 29:1) with 5% glycerin (5 V/cm, 0.5x Tris Borate Buffer, 1820 h, room temperature). Then the gel was developed using Silver Sequence DNA Staining Reagents kit (Promega Corp., Madison, WI), according to the manufacturers protocol. To confirm SSCP, direct DNA sequencing was performed. PCR products were extracted from 1% low-melting agarose and purified by Wizard PCR PREPS DNA Purification System (Promega Corp.) and then used as templates for direct sequencing. Both forward and reverse strands were sequenced using ABI Sequencing dye-terminator kit (Perkin-Elmer, Foster City, CA) on ABI PRISM 373 DNA Sequencer (Perkin-Elmer) according to the protocol recommended by the company.
|
| Results |
|---|
|
|
|---|
Splice site mutations in GH-1 were identified in all six families (7 children) with autosomal-dominant IGHD enrolled in the study (Table 2
). These mutations were as follows: 1) A>T transversion of the second base of the 3'-acceptor splice site of intron 2 (IVS2 -2A>T); 2) T>C transition of the second base of the 5'-donor splice site of intron 3 (IVS3 +2T>C); 3) G>A transition of the first base of the 5'-donor splice site of intron 3 (IVS3 +1G>A). These mutations destroy IVS2-acceptor or IVS3-donor invariant splice sites, which are crucial for normal splicing.
|
A novel heterozygous IVS2 -2A>T splicing mutation, located in the second nucleotide of the invariant ag dinucleotide of the 3'-acceptor splice site of intron 2 (tagGAA) of GH-1 was revealed in two patients from one family with IGHD II (Fig. 1
).
|
|
|
|
A novel heterozygous IVS3 +2T>C mutation, located in the second nucleotide of the invariant gt dinucleotide of the 5'-donor splice site of intron 3 (TCCgtg) of GH-1 was found in one family with IGHD II (Fig. 3
).
|
IVS3 +1G>A mutation in the GH-1 gene
A heterozygous IVS3 +1G>A mutation, located in the first nucleotide of the invariant gt dinucleotide of the 5'-donor splice site of intron 3 (TCCgtg) of GH-1 was identified in five children of four families with IGHD II (Fig. 4
).
|
Family I consisted of two affected siblings, sister and brother from different marriages, who inherited the pathological allele from their mother. Molecular analysis was performed in all affected members of the family. Both the girl and the boy were born at term, by caesarian section, without any perinatal complications. Birth lengths were 48 cm and 50 cm, and birth weights were 3.3 kg and 3.0 kg, for the girl and boy, respectively. In both children, growth retardation was apparent from the birth, but they were diagnosed as having IGHD at different ages. In the girl, GH deficiency was confirmed at the age of 13 yr (-4.7 HSDS), with peak GH levels of 0.6 µg/liter in the ITT test, and 0.6 µg/liter in the clonidine test. She entered spontaneous puberty at 13.2 yr of age. The girl was irregularly treated with pituitary GH in childhood, and only near puberty did she start treatment with recombinant GH. As a consequence, her final height (-4.0 HSDS) was short. The boy was first examined at the age of 1 yr (-3.1 HSDS). He had clinical features of congenital GH deficiency, including prominent forehead, saddle nose, and blue-like scleras. GH stimulation tests were not performed on the boy because of his young age and availability of clinical and anamnesis data. IGHD was confirmed by a low basal blood level of IGF-I (<2.5 µg/liter). At the age of 1.9 yr, the boy started treatment with recombinant GH, and HV increased to 17.1 cm/yr. The mother was diagnosed as having IGHD at the age of 38 yr. She had never been treated with GH. Her parents had normal height, with her mother being 155 cm, and father being 182 cm in height.
Family II consisted of three generations of affected individuals. Apart from the proband (-3.9 HSDS), who was analyzed by molecular analysis, the pedigree included the affected father (-6.7 HSDS), who died before our study from myocardial infarction, and grandfather (-4.5 HSDS), also not available for the study. The proband was born at term by caesarian section, without perinatal complications, with a birth length of 50 cm and a birth weight of 3.0 kg. IGHD was diagnosed at the age of 5.2 yr (-3.9 HSDS). She grew well on recombinant GH. Short-statured relatives from the paternal line, including an aunt (145 cm), two uncles (140 cm each), and three of their offspring (130 cm) were also reported. An extremely short mother (-5.7 HSDS) had Seckel syndrome. The phenotype of the short-statured aunt (-5.0 HSDS) from the maternal line, who died from bone sarcoma at the age of 18 yr, also had Seckel syndrome.
Family III had an affected daughter and father. The girl was born at term with a normal delivery, and no perinatal complications. Birth length was 49 cm, and birth weight was 3.1 kg. IGHD was diagnosed at the age of 2.3 yr (-3.8 HSDS). Her short-statured father (-3.7 HSDS) was not available for the study. Her healthy mother had normal height (+1.3 HSDS).
Family IV included an affected son and mother. Mutation analysis in the boy was performed by the group of Prof. J. A. Phillips III (Vanderbilt University, Nashville, TN), who detected an IVS3 +1G>A mutation in GH-1. The single child in the family was born at term by caesarian section, with no perinatal complications. Birth length was 50 cm and birth weight was 3.35 kg. Hypopituitarism without hormonal confirmation was diagnosed at the age of 2 yr, and at the age of 5.8 yr he started irregular treatment with pituitary GH. Spontaneous puberty occurred at 15 yr. At first examination at 20 yr of age, his height was -4.5 HSDS, BA was 16 yr, and pubertal development was at Tanner stage IV. IGHD was confirmed by an extremely low peak GH-level (<0.1 µg/liter) during the ITT.
| Discussion |
|---|
|
|
|---|
Five dominant-negative mutations of splicing have been described in intron 3 of GH-1. The first one, IVS3 +6T>C, revealed in a Turkish family with IGHD II (9), inactivated the donor splice site of IVS3 and resulted in alternative splicing with skipping of exon 3. The second, IVS3 +1G>A mutation was reported in 9 patients of three IGHD II families from Sweden and North America, and a South Africa Indian family (10). Family I consisted of an affected father (-6.0 HSDS) and son (-6.5 HSDS), and family II included four generations, with a son of 3.3 yr (-6.0 HSDS), mother (-6.0 HSDS), grandmother (-6.0 HSDS), and great grandmother (-5.0 HSDS). Family III included two sisters of 3.75 yr and 1.25 yr (both-5.1 HSDS) and the mother (-7.1 HSDS). The authors demonstrated that a G>A transition in the invariant CpG dinucleotide of exon 3/IVS3 boundary skipped exon 3 with loss of amino acids 3271 and shortening the mature protein to 151 amino acids. Subsequently, this mutation was reported in one Chilean IGHD II family (11) and in Japanese patients, including one IGHD II family from Ibaraki and two sporadic cases with de novo mutation from Nagoya and Nagasaki (12). One more sporadic case from Japan with an IVS3 +1G>A mutation in a compound heterozygous state was also reported (13). A heterozygous de novo IVS3 +1G>C mutation was found in one patient from Germany by screening 10 children with sporadic severe IGHD (14) and in one Dutch family (15). An IVS3 +28G>A mutation was reported in a Thai family with IGHD II (16). Segregation and expression studies of the mutant variant confirmed its dominantnegative effect by perturbing mRNA splicing. At last, IVS3 +5G>A mutation was reported in IGHD II Chilean family (11) and Japanese family from Kyoto (17).
This report presents summary data on the molecular analysis of GH-1 mutations in a large cohort of children with total IGHD, both familial and sporadic, living in the Russian Federation. As shown in Table 2
, we found the entire spectrum of GH-1 splice site mutations associated with IGHD. Another important issue is the high incidence (100%) of GH-1 splicing mutations in the Russian families studied with IGHD II. Molecular analysis revealed that all the families (6/6) with IGHD II enrolled in the study had splice site mutations in GH-1, in contrast to sporadic cases.
Our study expanded the list of reported GH-1 defects in IGHD II. We identified two novel GH-1 splicing mutations in familial cases. The dominant-negative IVS2 -2A>T mutation in intron 2 of GH-1 was found in a family from the Middle Russia region (18). This point mutation, involving the highly conserved tagGAA sequence, destroyed the invariant ag dinucleotide of the IVS2/exon3 acceptor splice site, which is crucial for normal splicing. The second novel dominant-negative IVS3 +2T>C mutation was revealed in intron 3 of GH-1 in a family from the Black Sea region (19). The mutation is located in the invariant TCCgtg sequence of exon 3/IVS3 donor splice site, close to the IVS3 +1G>A mutation, causing perturbed splicing. The same mutation has been also found in one family from Germany with two affected siblings and father (20). According to the authors, the predicted effect of the mutation is the skipping of exon 3 of GH-1 during splicing.
Mutation IVS3 +1G>A was previously reported in many unrelated individuals of different ethnic origins (10, 11, 12). At least two cases were described as de novo mutations (17); a different origin of the mutation was shown for some other cases (20). These data, when put together, exclude the founder effect as an explanation for the high prevalence of this mutation. Taking into account that this mutation occurs in a CpG dinucleotide, which has an increased mutation rate due to spontaneous deamination of a methylated cytosine (10, 21, 22), we suggest a mutational hot spot at this particular site. The mutational hot spot hypothesis for IVS3 +1G>A was previously proposed (23). This hypothesis is further supported by our detection of this mutation in four unrelated families out of six with IGHD II (24).
We are reporting the largest number of families with IGHD II from one country. Since 1995, a total of six families with familial IGHD II due to IVS3 +1G>A mutation have been described (10, 11, 12, 20). These publications comprised geographically widely separated ethnic groups; however, each presented only one family from any single country. Taken together, our findings and data reported in other ethnic groups show that IVS3 +1G>A mutation is the most common GH-1 mutation associated with the IGHD II.
Our study did not find pronounced interfamilial phenotypic variability among children having the IVS3 +1G>A mutation. All patients had a similar severe growth retardation at the age of 1 yr, a degree of GH deficiency, and an absence of truncal adiposity. Both affected mothers were extremely short, but the degree of truncal adiposity varied. Nevertheless, there was some intrafamilial variability in the early phenotype in two half-siblings from family I. At 1 yr of age the younger brother was less retarded in height than his older sister. Similar intrafamilial variability has been described in other families with IGHD II (20), and in families with IGHD IA (25). We can speculate that this clinical polymorphism can be explained to some degree, by the influence of other gene-modifiers, i.e. by the background genotype of the individual and by environmental variables. Our study showed that mutations of splicing, located in intron 2 (3'-acceptor splice site) or intron 3 (5'-donor splice site) of the GH-1 resulted in a similar phenotype of congenital GH deficiency. It is characterized by the absence of prenatal growth retardation, but early and progressive postnatal growth retardation; total GH deficiency (<2 µg/liter in GH-stimulation tests); lack of deficiency of other anterior pituitary hormones during long-term follow-up; spontaneous puberty with subsequent fertility; and highly effective recombinant GH treatment. None of the studied children was overweight, taking into account the age range from 1.020.0 yr.
The mechanisms responsible for the dominant-negative effect of the mutant GH-1 allele are not clearly understood. It is assumed that GH storage and secretion are perturbed due to interactions between wild-type and mutant proteins in secretory granules of somatotropes (22). All previously identified dominant-negative mutations were found in intron 3, and affected splicing through direct modification of splicing sites, or mutation in an intron splice enhancer (26). We found three different kinds of mutations, which cause autosomal dominant segregation of the disease, including the only one in intron 2. However, all of them most probably result in the skipping of exon 3. Thus, we can speculate that only skipping of exon 3 results in the particular modification of GH, which leads to interaction of this mutated protein with normal GH protein molecules and inactivation of them.
In conclusion, based on the presented data, the GH-1 gene can be considered to be the main candidate gene for IGHD II in patients, living in the Russian Federation. Splice site mutations in the GH-1 gene invariantly appear to be responsible for familial autosomal-dominant IGHD. The variety of splice mutations described here further indicates allelic genetic heterogeneity of IGHD II in the Russian population.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: BA, Bone age; CA, chronological age; fT4, free T4; HSDS, height SD score; HV, height velocity; IGHD, isolated GH deficiency; IGHD II, IGHD type II; ITT, insulin tolerance test; SSCP, single-strand conformation polymorphism.
Received February 20, 2002.
Accepted November 18, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
V. Petkovic, D. Lochmatter, J. Turton, P. E. Clayton, P. J. Trainer, M. T. Dattani, A. Eble, I. C. Robinson, C. E. Fluck, and P. E. Mullis Exon Splice Enhancer Mutation (GH-E32A) Causes Autosomal Dominant Growth Hormone Deficiency J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4427 - 4435. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Salemi, S. Yousefi, D. Lochmatter, A. Eble, J. Deladoey, I. C. A. F. Robinson, H.-U. Simon, and P. E. Mullis Isolated Autosomal Dominant Growth Hormone Deficiency: Stimulating Mutant GH-1 Gene Expression Drives GH-1 Splice-Site Selection, Cell Proliferation, and Apoptosis Endocrinology, January 1, 2007; 148(1): 45 - 53. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Vivenza, L. Guazzarotti, M. Godi, D. Frasca, B. di Natale, P. Momigliano-Richiardi, G. Bona, and M. Giordano A Novel Deletion in the GH1 Gene Including the IVS3 Branch Site Responsible for Autosomal Dominant Isolated Growth Hormone Deficiency J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 980 - 986. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Mullis, I. C. A. F. Robinson, S. Salemi, A. Eble, A. Besson, J.-M. Vuissoz, J. Deladoey, D. Simon, P. Czernichow, and G. Binder Isolated Autosomal Dominant Growth Hormone Deficiency: An Evolving Pituitary Deficit? A Multicenter Follow-Up Study J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2089 - 2096. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E Mullis Genetic control of growth Eur. J. Endocrinol., January 1, 2005; 152(1): 11 - 31. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. C. C. Ryther, A. S. Flynt, B. D. Harris, J. A. Phillips III, and J. G. Patton GH1 Splicing Is Regulated by Multiple Enhancers Whose Mutation Produces a Dominant-Negative GH Isoform That Can Be Degraded by Allele-Specific Small Interfering RNA (siRNA) Endocrinology, June 1, 2004; 145(6): 2988 - 2996. [Abstract] [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 |