|
|
||||||||
CASE REPORT |
Biochemistry, Endocrinology and Metabolism Unit and London Centre for Paediatric Endocrinology, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK, 1 Departments of Child Health and 2 Endocrinology, Kings College Hospital, London, UK and 3 Division of Molecular Neuroendocrinology, National Institute for Medical Research, Mill Hill, London, UK
(Correspondence should be addressed to M T Dattani; Email: mdattani{at}ich.ucl.ac.uk)
Abstract
Background: Type II isolated GH deficiency (IGHD type II) is caused by dominant negative splicing or point mutations of the GH-1 gene. Studies have suggested that dominant mutant GH forms prevent the secretion of wild-type GH, resulting in eventual cell death; surprisingly, some patients with these GH mutations develop other hormonal deficiencies (ACTH, TSH).
Subjects: The proband presented at the age of 2.3 years with IGHD. His father, also known to have been treated for IGHD as a child, had subsequently been lost to follow-up, having remained without treatment during this time. At re-evaluation at the age of 38 years, he complained of lack of stamina and poor libido. Clinical and biochemical assessment confirmed severe GHD, borderline ACTH insufficiency, suboptimal basal and stimulated gonadotropins, and a poor prolactin response to provocation. The basal testosterone concentration was low, and he complained of secondary infertility. Magnetic resonance imaging revealed anterior pituitary hypoplasia in both patients. Genetic testing revealed a heterozygous splicing mutation in GH-1 (intervening sequence-3 + 1G>A) in both patients, known to cause IGHD type II.
Interventions: The proband showed an excellent growth response to recombinant human GH (rhGH). His father, also treated with rhGH, showed improved quality of life on rhGH, but testosterone concentrations continued to decline, necessitating treatment with testosterone with symptomatic benefit but no improvement in semen quality.
Conclusions: This case supports recent experimental and clinical observations suggesting that the cytotoxicity associated with accumulation of dominant negative mutant 17.5 kDa GH causes a form of GHD that can evolve into multiple hormone deficiencies. Hence, patients diagnosed initially with IGHD type II require continued long-term clinical follow-up.
Introduction
Four distinct familial forms of isolated growth hormone deficiency (IGHD) have been described (1, 2). Large homozygous deletions (6.7, 7.0, 7.6, and 45 kb) within the GH-1 gene associated with the complete absence of growth hormone (GH) lead to GHD type IA, and GH treatment is frequently associated with the development of antibodies. Additionally, frameshift and nonsense mutations may also be associated with type IA GHD. Type IB GHD is associated with less severe short stature and detectable concentrations of GH, and is due to recessive mutations within GH-1 and GHRHR. Type II autosomal dominant GHD is associated with heterozygous splice site and occasional mis-sense mutations (R183H, V110F, and P89L) within the GH-1 gene (1), while type III GHD is inherited as an X-linked condition.
Recent studies have revealed novel insights into the pathogenesis of autosomal dominant IGHD type II. The condition is classically associated with abnormal splicing within GH-1 and is most frequently due to mutations within the first 6 bp of intervening sequence 3 (5'IVS-3) (3). The GH-1 gene is located on the long arm of chromosome 17 (17q22) and consists of five exons encoding human GH, a 191 amino acid (22 kDa) peptide, accounting for more than 80% of circulating GH, with approximately 510% consisting of a 20 kDa form of human GH (hGH) that arises as a result of the use of a cryptic 3' splice site in exon 3. Mutations at the donor splice site of IVS-3 have been shown to generate alternatively spliced transcripts lacking sequences corresponding to exon 3 (Fig. 1
). When translated, these produce a 17.5 kDa variant of hGH, which lacks aa 3271, the entire loop that connects helices 1 and 2 in the tertiary structure of hGH (46).
|
Following these observations, additional hormone (partial thyroid-stimulating hormone (TSH) and/or adrenocorticotropic hormone (ACTH)) deficiencies were detected in 8 of 57 patients with IGHD type II (13) due to the IVS-3 + 1 (n = 5) and P89L (n = 3) mutations. Of these, two patients had overt hypothyroidism and ACTH insufficiency as measured by low thyroxine and fasting morning cortisol concentrations, whereas the remainder had ACTH insufficiency. The hypothalamopituitarygonadal endocrine axis was not investigated in these patients, but the authors reported that the gonadotrope axis appeared to be clinically normal.
At a functional level, the 17.5 kDa variant generated by exon 3 skipping has been shown to disrupt the Golgi apparatus and impair trafficking of GH (14). In secretory, but not in non-secretory, cell lines co-expression of
exon3hGH inhibited wild-type (WT)-hGH secretion in a concentration-dependent manner (15). The aberrant hGH, therefore, acts in a dominant negative manner in the presence of WT-GH (9, 1517). In a transgenic animal model of IGHD type II, mice expressing
exon3hGH showed a phenotype characterized by dwarfism, reduced GH content in the anterior pituitary, marked anterior pituitary hypoplasia, and, in the most severely affected lines, the contents of other pituitary hormones (prolactin, TSH, and luteinizing hormone (LH)) were all significantly reduced in adult males (17).
We now report a pedigree in which a father and son presented with short stature due to GHD. The father had been treated with both pituitary-derived and recombinant hGH in childhood, but had been untreated for 20 years. Following the diagnosis of GHD in his son, the father was re-evaluated and reported a lack of energy and vitality and later complained of secondary sub-fertility. Investigations not only confirmed GHD, but also revealed partial ACTH, prolactin, and gonadotropin deficiencies. Thus, we now show that gonadotropin insufficiency is also associated with this GH-1 mutation in human IGHD type II.
Methods
Clinical evaluation
Pituitary function was assessed using standard dynamic tests (18). Magnetic resonance imaging (MRI) (1.5 T Siemens Magnetom Symphony, Siemens Medical Systems, Malvern, PA, USA) included T1 and T2 weighted high-resolution pituitary imaging through the hypothalamopituitary axis (T1 sagittal 3 mm slices, T1 and T2 coronal 3 mm slices).
Assays
GH was measured using the Nichols Advantage method standardized against the International Standard 80/505. insulin-like growth factor I (IGF-I) and IGF-binding protein-3 (IGFBP-3) were measured using the Nichols Advantage 2-site chemiluminescent assay. TSH, free thyroxine (fT4), prolactin, LH, follicle-stimulating hormone (FSH), cortisol, and testosterone measurements were all performed on the chemiluminescent Bayer ADVIA-Centaur analyzer.
Genetic investigation
Following Institutional Review Board Approval and with informed consent, GH-1 was amplified using primers (available on request) flanking exons and including splice sites. IVS-3 was included in overlapping amplicons of exons 3 and 4 and all primers were designed with the aid of WAVEMAKER software (Transgenomic Ltd, Elancourt, France). PCR products were sequenced using the BigDye Terminator Thermal Cycle Sequencing protocol and run out on the ABI 377 Genetic Analyzer (both Applied Biosystems, Cheshire, UK) in both the forward and the reverse orientations and analyzed using Sequencher (Gene Codes Corporation, MI, USA).
Results
Phenotype
The proband (III.2, Fig. 2A
), presented at the age of 2.3 years with short stature (height 75 cm (3.4 SDS), weight 9.35 kg (2.5 SDS), mid-parental height 164 cm (1.6 SDS)). His birth weight was 3.35 kg (0.5 SDS) with a birth length of 52 cm (0.70 SDS). His height velocity had been suboptimal since infancy (Fig. 2B
), and his dentition was delayed. On presentation, his facial features were suggestive of a diagnosis of GHD (Fig. 2C
); his voice was high-pitched, and his phallus small, with both testes in the scrotum. Glucagon provocation revealed profound GHD (peak GH concentration 2.7 mU/l, IGF-I<2 nmol/l), with normal cortisol secretion (peak cortisol concentration 916 nmol/l), and a low normal fT4 concentration (11.6 pmol/l; normal range (NR) 10.321.9 pmol/l) with a TSH concentration of 2.1 mU/l. He was commenced on recombinant hGH (rhGH) treatment at the age of 2.7 years and mounted an excellent response to this (height 0.9 SDS at 5.5 years of age). MRI revealed anterior pituitary hypoplasia, with a normal posterior pituitary and infundibulum (Fig. 2F
).
|
He was lost to follow-up for a number of years, and re-presented with his son III.2 at the age of 38 years, having been untreated for 20 years. He complained of lack of stamina with extreme fatigue, poor libido, and recognized difficulties in coping with work. During the subsequent evaluation, he also reported secondary infertility. Examination revealed centripetal obesity and the typical facial features of an adult with GHD (Fig. 2E
). His testicular volumes were 15 ml bilaterally. A combined pituitary function test was performed (Table 1
), revealing severe GHD (peak GH <0.6 mU/l), a borderline cortisol response (peak cortisol to insulin-induced hypoglycemia 502.3 nmol/l), a suboptimal prolactin response (peak prolactin 262 mU/l) to TRH, and low basal gonadotropins with a poor LH response to LHRH (basal LH 1.5 U/l, FSH 1.0 U/l; peak LH 11.7 U/l (NR 1229.7 U/l), peak FSH 4.0 U/l (NR 2.97.8 U/l)) with a low testosterone (9.7 nmol/l). He was also noted to have hypercholesterolemia (fasting cholesterol 7.3 mmol/l (NR 15 mmol/l); triglyceride 4.4 mmol/l (NR 0.52.0 mmol/l)). MR scanning revealed a pituitary fossa of relatively small volume, with only a small rim of pituitary tissue in its floor (Fig. 2G
), a normal infundibulum and a eutopic posterior pituitary.
|
Genomic analysis of GH-1
Analysis of the GH-1 gene identified a heterozygous change at IVS-3 + 1 that changes the G of the invariant GT dinucleotide splice donor site to an A nucleotide (Fig. 3
), first described by Cogan et al. (20) (Online Mendelian Inheritance in Man #139250.0009). No other sequence variants were identified in the proband or his father.
|
Human IGHD type II also shows a marked degree of variability in onset, severity and progression, even within the same family (21), with only a third of the affected individuals manifesting severe short stature at diagnosis. Overall, the incidence of these mutations is low; we have now screened 88 IGHD patients for mutations in GH-1, and identified five patients with IVS 3 mutations, giving a frequency of 5.7% (our unpublished data). Mullis et al. have recently shown a co-existing variable partial ACTH and TSH insufficiency in some of these individuals (13). Additionally, the pituitary heights on MRI were lower in these patients, suggesting a degree of hypoplasia. None of the individuals in this study was reported as showing evidence of gonadotropin deficiency, although detailed assessment of the hypothalamopituitarygonadotrope axis was not performed, and the diagnosis of ACTH deficiency was made on the basis of a low fasting cortisol as measured at 0800 h, as opposed to a provocative test of cortisol secretion.
We have now shown that IGHD type II mutations can be associated with an evolving gonadotropin deficiency. We report a pedigree whereby a father and son presented with severe short stature due to profound GHD characterized by a poor GH response to provocation with low concentrations of IGF-I and IGFBP-3. The father, II.3, was initially treated with pituitary-derived and recombinant hGH, but was later untreated over a 20-year period. On reassessment, he showed evidence of profound clinical and biochemical GHD in addition to partial prolactin, ACTH (after insulin-induced hypoglycemia) and gonadotropin deficiencies with a markedly hypoplastic anterior pituitary gland. The gonadotropin deficiency had evolved, and was characterized by low basal gonadotropins, a suboptimal LH response to GnRH (22), a persistently low testosterone concentration (nadir 4.0 nmol/l) and secondary infertility.
Recent studies have suggested that IGHD type II may be associated with an evolving endocrinopathy in mice as well as humans. McGuinness et al. overexpressed a 5' IVS-3 + 1G->A mutant construct in somatotropes in a normal mouse background (17) (i.e. the same mutation described in our pedigree). Three different lines of transgenic mice were produced with varying numbers of copies of mutant gene, on the background of two normal copies of the mouse GH gene. The onset and severityof IGHD type II was dependent on transgene copy number. The high copy line was most severely affected, with rapidly developing severe GHD and dwarfism, profound pituitary hypoplasia and almost complete loss of somatotropes, whereas the low copy mice showed milder adult-onset IGHD (17). Furthermore, the most severely affected line showed significant reduction in pituitary prolactin, TSH, and LH, while the females showed a reduction in prolactin and TSH, but not in LH. In these mice, pituitary sections revealed very few somatotropes of abnormal appearance, as well as reduced numbers of corticotropes, lactotropes, and gonadotropes, which, together with the reduced gonadotropins, could explain the sub-fertility observed in the most severely affected line (17). In many respects, the phenotype of II.3 reflects that of the more severely affected male transgenic mice that manifested late LH deficiency (17).
The 17.5 kDa isoform produced from GH-1 transcripts lacking exon 3 sequences is retained in the endoplasmic reticulum, disrupts the Golgi apparatus, and reduces the stability of the 22 kDa isoform (14, 16). In support of this idea, McGuinness et al. showed that the
exon3hGH expression prevented or destabilized secretory vesicle formation in a dominant negative manner in rat GH-producing cells (17). Rather, the endogenous rodent GH accumulated in amorphous aggregates in the cytosol, endoplasmic reticulum and Golgi, and was unavailable for exocytic release (17). Similar morphological findings were observed in somatotropes in the
exon3hGH transgenic mice, but in vivo the process may be accelerated by additional factors. One is an increased tropic drive from GHRH to increase expression of both WT-hGH and
exon3hGH, secondary to evolving GHD. Another factor is a macrophage response to the dying defective GH cells (17). It is clear from other in vitro work (9) that the type of mutation has an important bearing on the relative amounts of 17.522 kDa produced, with the IVS-3 + 1 mutation being highly effective. However, while the dependence of the onset, severity, and rate of progression of IGHD on the relative amounts of
exon3hGH and WT-hGH is easy to understand, it is still unclear how a mutation in the GH-1 gene, restricted in expression to somatotropes, could lead to other hormonal deficiencies. Small amounts of GH have been reportedly expressed ectopically in other pituitary cell types in rodents, notably gonadotropes (23), but for an effective dominant effect, this would have to be at a significant protein level. We believe this is unlikely, especially since small amounts of 17.5 kDa transcript and product are detectable endogenously in normal pituitaries (24). It seems more likely that a phenomenon observed in transgenic mice (17), namely an invasion by activated macrophages, may lead to significant bystander endocrine cell killing, which in time compromises cellular repletion of the other lineages, and ultimately to additional endocrine deficits.
In theory, removal of the GHRH drive to generate greater quantities of both wild-type and mutant GH could potentially ameliorate the phenotype and prevent the evolution of other endocrinopathies. Indeed, it is interesting to note that patient II.3 had been untreated for 20 years when he re-presented with hypopituitarism. However, Mullis et al. concluded that treatment of the patients with rhGH made no difference to the evolution of the hypopituitarism (13).
This family study illustrates the additional insights gained by precise genetic diagnosis in subjects with apparently isolated GHD. From a clinical perspective, it is evident that evaluation of pituitary function in those formerly diagnosed with childhood GHD should be repeated not only after the cessation of vertical growth, but also at intervals throughout adulthood, since such individuals are at risk of evolving pituitary endocrinopathy. We suggest that all individuals with AD IGHD should be screened at annual intervals (0800 h cortisol, free thyroxine, LH, FSH, and sex steroids (estradiol or testosterone)). Additionally, although the evolution of the phenotype has only been described in association with the IVS-3 + 1 and P89L mutations, it would, at this stage, be reasonable to monitor all patients with AD IGHD. Only further long-term studies in these rare but interesting patients will be able to shed further light on the mechanisms underlying disease evolution.
Acknowledgements
Mehul T Dattani and James PG Turton were supported by an MRC Career Establishment Grant.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |