| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Reproductive Endocrine Unit (S.B.S., J.S.A., N.A.A., W.F.C.) and Department of Neurology (D.H.M.), Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Stephanie Seminara, M.D., Reproductive Endocrine Unit, Bartlett Hall Extension 505, Massachusetts General Hospital, Fruit Street, Boston, Massachusetts 02114. E-mail: . seminara.stephanie{at}mgh.harvard.edu
Abstract
Although the co-occurrence of cerebellar ataxia and hypogonadism has been recognized for close to 100 yr, cases of Gordon Holmes syndrome are quite rare. This report describes the largest kindred characterized to date. The parents of the three affected siblings are first cousins, suggesting that the disease was inherited as an autosomal recessive trait. The siblings’ initial evaluation was notable for low serum levels of sex steroids and gonadotropins (consistent with hypogonadotropic hypogonadism), progressive ataxia, and dementia. Extended treatment with physiological doses of pulsatile GnRH failed to stimulate a gonadotropin response. Brain imaging revealed volume loss in the cerebellum, with extensive abnormalities in the cerebral white matter. This unique family suggests that a common genetic mechanism is responsible for the syndrome of progressive hypogonadotropism and cerebellar ataxia.
PHENOTYPIC OVERLAP AND interfamilial variability make the hereditary ataxias a notoriously difficult group of disorders to classify. Endocrine dysfunction is a common clinical feature in these patients, with hypogonadism occurring in approximately 5%. Gordon Holmes first described the association of cerebellar ataxia and hypogonadism, 100 yr ago, when he reported three brothers and one sister who, in their mid-thirties, developed cerebellar symptoms and signs of sex steroid deficiency (1). As gonadotropin assays were developed, both hypo- (2) and hypergonadotropic (3) forms were reported. However, males with cerebellar ataxia and isolated defects in sperm morphology without any clearly defined endocrinopathy have also been described, suggesting that the clinical spectrum of this disorder is broad (4).
The present report describes an extended pedigree with ataxia and hypogonadotropic hypogonadism. The report presents the detailed endocrine testing performed on multiple affected family members, fits this index pedigree into its proper biological context, and considers the genetic implications of a syndrome characterized by a combination of such disparate phenotypes.
Subjects and Methods
Case presentations
Experimental subjects XI:4, XI:5, and XI:6 are siblings. Three sisters with normal menarche are also members of the sibling group. The parents are first cousins and are also related through additional consanguineous loops (Fig. 1
). The family emigrated from Palestine in 1948, and all the siblings were born and raised in Amman, Jordan.
|
A diagnosis of idiopathic hypogonadotropic hypogonadism (IHH) was made, and the patient was treated with 1500 IU Pregnyl (CG for injection) and 75 IU Pergonal (menotropins for injection, comprised of equal quantities of FSH and LH activity), im, three times a week. After 6 months of therapy, the patient noted an increase in facial hair and more frequent shaving. His testes grew to 20 cc bilaterally. A free T level was within the normal range at 694 pM (326–1283 pM). Sperm were present in the ejaculate. A scrotal ultrasound was performed, and both testes appeared normal in size and echopattern, except for a spermatocele on the right. The patient was lost to follow-up.
In 1996, at age 31, XI:4 presented with a chief complaint of infertility. He and his second wife had been having unprotected relations for 2 yr. On examination, worsening dysarthria and increasingly clumsy movements were evident. XI:4 requested a noninjection form of fertility treatment. Clomiphene citrate was administered (50 mg, by mouth, three times a day, for 10 d), but there was no increase in LH, FSH, or T. He was again lost to follow-up.
In 1998, at age 33, XI:4 presented again for treatment of infertility. A testicular biopsy was performed that revealed early germ cells, prominent perilobular hyalinization, small foci of sclerosis, and no evidence for spermatogenesis. Mild Leydig cell hyperplasia was present. The neurological abnormalities were noted to have progressed, with unsteady gait, worsened dysarthria, and a more withdrawn personality. In 1999, GnRH stimulation testing (100 µg) was performed. The LH level rose from a basal value of 0.06 IU/liter to a peak level of 5.80 IU/liter at 90 min. The FSH level rose from a baseline value of 0.14 IU/liter to a peak of 2.0 IU/liter at 120 min.
In 2000, at age 35, XI:4 remained ambulatory and independent, although motor and cognitive functions had continued to decline. Physical examination revealed substantial facial and chest hair, with a male-pattern pubic escutcheon. Arm span was 181 cm, and height was 179 cm, with a crown-pubis distance of 79 cm and pubis-ground distance of 85 cm. Stretched penile length was 11.5 cm, and the testes were 10 cc each and slightly soft. Speech was dysarthric. Conjugate gaze was present without nystagmus. Motor tone was normal without spasticity. Tendon reflexes were normally active and symmetric. The gait was highly abnormal, wide based (10 cm) with posturing of the upper extremities, and attempts at tandem walking were unsuccessful.
Smell testing was performed using 12 common odorants (foods, cigarettes, alcohol, and others). XI:4 was able to recognize 10 of 12 items, but his responses were not consistent on repeat testing, so that he had only a 39% accuracy (16 correct identifications in 41 trials). Further detailed neurological testing is presented in a separate report (Margolin, D. H., I. Adam, K. S. Sims, S. B. Seminara, J. Sherman, and E. T. Hedley-Whyte, manuscript in preparation).
Magnetic resonance imaging of the brain revealed diffuse parenchymal volume loss, involving particularly the cerebellar hemispheres and vermis. Multiple punctate and confluent areas of T2 hyperintensity were noted in the periventricular and subcortical white matter bilaterally. Baseline levels of LH were 0.2 IU/liter (normal male, 2–12 IU/liter); and FSH, 0.6 IU/liter (normal male, 1–12 IU/liter). Serum T was 0.21 nM (normal range, 9.4–37.1 nM). XI:4 was started on T patch therapy, and additional serum samples were drawn and shipped to the Massachusetts General Hospital for testing. TSH was 0.36 µU/ml (normal, 0.5–5 µU/ml), but all other thyroid function studies were normal. PRL was 3.5 µg/liter (normal range, 0–15 µg/liter); GH levels were 0.84 µg/liter (normal range, 2–6 µg/liter); IGF-1 levels were 209 µg/liter (normal range, 114–492 µg/liter). Serum lipoprotein electrophoresis showed a type IIB hyperlipidemia (elevated low-density lipoprotein and very-low-density lipoprotein). Biochemical testing of serum and urine found no evidence of a major metabolic disorder and ruled out the common leukodystrophies. Electron microscopy found normal mitochondria and lysosomes in a skin biopsy.
Subject XI:5. XI:5 had a normal birth, infancy, and childhood. Although he had axillary and pubic hair, he had no facial or chest hair and, at age 16, was referred to an endocrinologist for delayed puberty. At age 18, XI:5 was treated with injections, three times a week for 3 yr, and developed a partial beard. By age 22, however, XI:5 developed dysarthria. Two to 3 yr later, he developed gait imbalance. Upper extremity incoordination followed soon after. In his late twenties, XI:5 had a brain CT, which revealed cerebellar and lesser cortical atrophy as well as hypodensities in cerebral white matter. By age 30, XI:5 required the use of a wheelchair. By age 36, he was making only limited vocalizations and, by age 38, was mute and visually unresponsive. Thereafter, he was bedridden, made no purposeful movements, and required total care. He died of aspiration pneumonia at age 43. An autopsy was not performed.
Physical examination, 1 month before death, revealed an emaciated male, making purposeless movements of the face, trunk, and all limbs. The arm span was 188 cm, and the height was 185 cm, with a crown-pubis distance of 85 cm and pubis-ground distance of 100 cm. Beard stubble, axillary hair, and pubic hair (female escutcheon) were present. The stretched penile length was 11.5 cm; the testes were not palpable.
Neurological examination was remarkable for severe dementia. There was no response to visual threat, sudden loud noises, or pain. The eyes were conjugate, with a full range of motion but a strong right-gaze preference. There was no nystagmus. The cough reflex was preserved. Motor tone was relatively normal; spasticity was not observed. Bilateral ankle clonus was present. Plantar scratch elicited flexor responses.
Subject XI:6. XI:6 had a normal birth, infancy, and childhood. Family members recall that XI:6 had an acute sense of smell. In adolescence, XI:6 had normal adrenarche and thelarche, but menarche was delayed until age 16. After two menses, XI:6 had secondary amenorrhea for 12 months, one final menstrual period, and no menses subsequently. She was evaluated by an endocrinologist, who diagnosed hypogonadotropic hypogonadism and began treatment with injections three times a week.
XI:6 married and, while on therapy, developed an ectopic (tubal) pregnancy. Her medications were discontinued, and she became amenorrheic once again. In her 20s, she experienced personality changes, becoming withdrawn and lazy. At approximately age 30, she developed dysarthria. A head CT scan revealed cerebellar and lesser cortical atrophy as well as hypodensities in cerebral white matter. In her early thirties, XI:6 developed upper extremity tremor and ataxia. By age 36, she was confined to a wheelchair; and, by age 37, she had only unintelligible vocalizations. At age 39, she was bedridden, making no purposeful movements and requiring total care. She died of aspiration pneumonia at age 40. A limited autopsy did not include the endocrine organs.
Physical examination at age 39 revealed emaciation and purposeless movements of the face, trunk, and all limbs. Her height was 168 cm. Breasts were classified as Tanner V. Axillary and pubic hair were present. Neurological examination was remarkable for profound dementia. Unlike her brother XI:5, XI:6 was able to fixate on an examiner’s face, although she showed no sign of recognition of family members. XI:6 did blink in response to visual threat and withdraw from painful stimuli. While awake, she made continuous moaning vocalizations. Conjugate gaze was present, without an evident gaze preference or nystagmus. Motor tone was relatively normal, without spasticity. A Babinski response was reliably elicited on the right.
Clinical phenotyping
XI:5 (at age 42) and XI:6 (at age 38) underwent treatment with the GnRH pulsatile pump, at 25 ng/kg, every 2 h, iv, for 7 d. Each morning, while on the GnRH pump, gonadotropin responses to a single iv GnRH bolus were monitored. LH, FSH, and free-
-subunit (FAS) levels were determined every 15 min, for 2 h, after the GnRH pulse. This study was approved by the subcommittee on human studies of the Massachusetts General Hospital; the parent of X1:5 and X1:6 gave informed consent before the study.
Assays
Serum LH and FSH concentrations were determined by the microparticle enzyme immunoassay, using the automated AxSYM system (Abbott Laboratories, Abbott Park, IL; using the Second International Reference Preparation as the reference standard. The assay sensitivity for both LH and FSH was 1.6 mIU/ml. The intraassay coefficients of variation (CV) for LH and FSH were less than 7% and 6%, respectively, with interassay CV for both hormones being less than 7.4%. For the 10 patients who previously participated in the GnRH studies, LH and FSH concentrations were determined by RIA using the Second International Reference Preparation as the reference standard. Inter- and intraassay CV were less than 10% (5).
Serum T concentrations were measured using the DPC Coat-A-Count RIA kit (Diagnostic Products, Los Angeles, CA), which had an intra- and interassay CV of less than 10%. E2 was measured by the Abbott Laboratories AxSYM system, which had an analytical sensitivity of 36.7 pM and a functional sensitivity of 73.4 pM. The intraassay CV was less than 6.4%, with an interassay CV of less than 10.6%. FAS was measured by a monoclonal antibody RIA, using a highly purified -subunit of hCG as the standard (6, 7). Secretory patterns were analyzed for pulses using a modified version of the Santen and Bardin method (8, 9).
Genetic testing
Genomic DNA was isolated from whole blood. The DNA from patient XI:4 was analyzed by PCR assay for the trinucleotide expansions associated with a number of neurodegenerative diseases. These analyses included an examination of: 1) the polymorphic CAG repeat tract in the 5' end of the Huntington disease gene, huntingtin, (10, 11, 12); 2) the polymorphic CAG repeat in the coding region of the gene for spinocerebellar ataxia type 1 (ataxin-1) (13, 14); 3) the CAG repeat in the gene for spinocerebellar ataxia type 2 (ataxin-2) (15, 16); 4) the CAG tract in the gene for spinocerebellar ataxia type 3 (Machado-Joseph Disease) (17, 18); 5) the CAG repeat in the gene for spinocerebellar ataxia type 6 (1A-voltage-dependent calcium channel) (19, 20); 6) the CAG repeat in the aminoterminal region of the gene encoding spinocerebellar ataxia type 7 (ataxin-7) (21); 7) the CAG expansion in the gene for dentatorubral pallidoluysian atrophy (atrophin-1) (22); and 8) the GAA repeat in the first intron of the gene for Friedreich ataxia (frataxin) (23). Repeat expansions were amplified by PCR using primers and conditions as described. The PCR-amplified products were sized by denaturing gel electrophoresis with a sequencing ladder, and the number of triplet repeats was calculated.
Primers were designed to amplify each of the three exons of the GnRH receptor gene, GNRHR, including acceptor-donor splice sites. Exon 1 was divided into two fragments: 1AF: GTGCTACAGTTACATGTGGC, 1AR: GCAGCTTCATTCTTGAGAGC, 1BF: GGAAAGATCCGAGTGACGG, 1BR: CTGACTTCCAGAACCCAAGC, 2F: GTAACATGTTAGAAGGCTAGC, 2R: GTATCTGTCACATAGTTCATGC, 3F: GAATTAGTGATGCTGTCTTCC, 3R: GAGGCTCTGAAGACTGAG. DNA from patients XI:4, XI:5, and XI:6 was analyzed by PCR using an ABI Prism Big Dye Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City, CA) and an ABI Model 377 DNA sequencing system (PE Applied Biosystems). In addition, polymorphic markers were chosen that flank the GNRHR to perform haplotype analysis and exclude the possibility of mutations in noncoding sequence (i.e. promoter, regulatory regions, introns). Markers (D4S1568, D4S409, D4S3018, D4S2387) were determined from the Human Genome Browser (http://genome.ucsc.edu/goldenPath/hgTracks.html), and haplotypes were constructed manually by visually inspecting the allele patterns across the region. Haplotypes were then confirmed using the haplotype feature of the GENEHUNTER genetic analysis program. These genetic studies were also approved by the subcommittee on human studies of the Massachusetts General Hospital; all subjects (or their legal guardian) gave their informed consent.
Results
Endocrine profiles
Table 1 shows the gonadotropin and sex steroid data from multiple affected family members (see Fig. 1
). Individuals XI:4, XI:5, and XI:6 all had low sex steroid levels and inappropriately low gonadotropin levels consistent with a diagnosis of IHH. Other family members all had normal sex steroid and gonadotropin levels for their age. The only exception was individual XI:10, who had low E2, LH, and FSH levels and was postpartum and nursing at the time these laboratory samples were drawn.
|
Both patients XI:5 and XI:6 received pulsatile GnRH sc by infusion pump for 7 d, and gonadotropin responses to a single iv pulse were monitored on d 1, 5, and 7. Their responses are graphed against the responses of 10 patients with IHH and normal neurological function who underwent an identical protocol (Fig. 2) (5). In these 10 patients with IHH, an LH response was seen immediately after their first exposure to GnRH, with a mean LH of 7.81 IU/liter in the sample drawn immediately after the GnRH bolus. However, in both patients XI:5 and XI:6, LH levels failed to exhibit any response, from d 1 up to and including d 7. In patient XI:5, LH levels failed to rise above the minimal detection level of the assay. Because FAS levels provide a particularly sensitive marker of GnRH secretion (6), FAS levels were also determined during the 7 d of exposure to GnRH. Although patient XI:5 had FAS levels on d 1 that were actually above those typically observed in patients with IHH, there was a decrease, rather than an increase, in levels over the course of the study.
|
). Genetic testing
Triplet repeat expansions were not detected in the genes for Huntington disease; spinocerebellar ataxias types 1, 2, 3, 6, and 7; DRPLA; and Freidreich ataxia. In addition, no coding sequence or splice site mutations were identified in the GNRHR. Because of the lack of affected phenotypes in previous generations, as well as the high degree of consanguinity, the haplotype analysis was performed under a recessive mode of inheritance. Affected individuals, XI:4 and XI:5, have different paternal chromosomes across the region defined by the four markers (Fig. 1). Additionally, individual XI:8 (unaffected) is identical for both paternal and maternal chromosomes. This data excludes the GNRHR as a causative factor in this family’s phenotype.
Discussion
The hereditary ataxias are a heterogeneous group of disorders that are often accompanied by both neurological and nonneurological signs. The association of cerebellar ataxia and hypogonadism [Gordon Holmes syndrome (GHS)] was first recognized as a distinct syndrome in 1907. Despite almost 100 yr of clinical recognition, there is little understanding of the pathophysiological mechanisms that bind together the disparate phenotypes of ataxia and hypogonadism. This case report provides detailed biochemical phenotyping of an extended kindred with this disorder and presents hypotheses for future genetic studies.
The hypogonadism of most patients with GHS is hypogonadotropic, with a defect in the production or release of gonadotropins by the pituitary gland. In the vast majority of case reports, plasma concentrations of LH and FSH have failed to rise after repetitive stimulation with GnRH, consistent with a primary pituitary defect (24, 25, 26, 27, 28, 29). However, in a small subgroup of patients, LH and FSH levels have demonstrated a normal response after GnRH stimulation, suggesting a primary hypothalamic disturbance (30, 31). Therefore, both hypothalamic and pituitary hypogonadotropism can be associated with cerebellar ataxia. This duality has been observed in other hypogonadal syndromes. For example, in adrenal hypoplasia congenita (hypogonadotropic hypogonadism and adrenal insufficiency), investigators have observed patterns consistent with both hypothalamic (32) and pituitary (33, 34, 35) defects.
In the family presented here, the pituitary response to exogenous GnRH was also assessed. Because the use of single bolus GnRH stimulation testing has long been considered inadequate to differentiate between hypothalamic and pituitary hypogonadotropism (36), we administered 7 d of therapy with exogenous pulsatile GnRH. Our group has previously shown that Kallmann syndrome patients given pulsatile GnRH (25 ng/kg sc, every 2 h, over a 1-wk period) demonstrate a robust gonadotropin response, with marked increases in both LH and FSH (5). After 7 d of an identical regimen of pulsatile GnRH, both patients XI:5 and XI:6 failed to elevate their gonadotropins, consistent with a primary pituitary abnormality.
Although the standard iv 100-µg GnRH stimulation test does not facilitate the classification of patients as having hypothalamic vs. pituitary defects, the data obtained from the single-bolus GnRH stimulation performed in patient XI:4 may nevertheless have relevance in this case. XI:4’s clinical symptoms began, in the early 1990s, with erectile dysfunction. Yet, almost 10 yr later, he mounted a normal gonadotropin response to 100 µg GnRH. This suggests that, despite a lengthy interval of hypogonadism, a population of gonadotropes remained capable of responding to a pharmacological GnRH stimulus. Therefore, the inability to respond to GnRH stimulation may be an end-stage marker for patients with this degenerative disease process.
Because IHH is often associated with anosmia (Kallmann syndrome), attempts were made to ascertain the olfactory status of the affected siblings in this pedigree. Neither XI:5 or XI:6 could participate in olfactory testing because of their profound dementia, although XI:6 had a keen sense of smell as a young woman, according to family members. Bedside testing was performed on patient XI:4. He was able to smell, but his accuracy and reproducibility were both low. Drawing conclusions from this data are difficult because: 1) the testing was not quantitative; and 2) anosmia is known to be a feature of dementing syndromes (37, 38, 39), raising the possibility that a smell defect, if real, might be a secondary feature of the dementing process and not a primary feature associated with hypogonadism. However, the olfactory testing of patient XI:4, albeit crude, raises the issue of olfactory function in GHS in particular. Although the vast majority of patients with GHS are not reported to have smell defects, one patient has been reported with IHH, anosmia, and saccadic dysmetria (40). Even more notable is a mouse model of Purkinje cell degeneration (pcd) in which mutant mice have rapid degeneration of their Purkinje cells, progressive degeneration in the olfactory bulb and retina, and marked abnormalities in sperm count, motility, and morphology (41). Although congenital, complete anosmia seems unlikely to be a feature of GHS, the possibility that there are subtle or progressive olfactory defects in some patients requires further exploration.
Accepting these ambiguities in phenotype, a number of potential gene defects were evaluated in this study. Our group has recently reported that 40% of patients with IHH and presumed autosomal recessive inheritance have coding defects in the GNRHR (42). However, direct sequencing and haplotype analysis convincingly eliminated a role for the GNRHR in this family. Because this family did not have morbid obesity, defects in the genes encoding prohormone convertase 1, leptin, and the leptin receptor were not explored.
Molecular genetics has enabled the identification of several loci for the ataxias. Several of the autosomal dominant cerebellar ataxias have been attributed to excessive CAG trinucleotide repeats [SCA-1, SCA-2, SCA-3, SCA-6, SCA-7, and DRPLA (dentatorubropallidoluysian atrophy)]. Although the proteins encoded by these genes are often ubiquitously expressed, only a subset of neurons is vulnerable to the degenerative effects of the trinucleotide repeats (43). Because knockout mice do not manifest neurological phenotypes, the triplet repeat expansions seem to cause a toxic gain of function. Mutant proteins aggregate as intranuclear inclusions, disrupting nuclear function. Although triplet repeat expansions occur most commonly in the dominant ataxias, the most common hereditary ataxia (the recessively inherited Friedreich ataxia) is caused by an unstable GAA expansion. Unlike its dominant counterparts, the triplet repeat is thought to cause a loss of function, with a decrease in transcription of frataxin, a protein thought to be involved in mitochondrial DNA repair or replication. Although no data are yet available, it is possible that expansion of an unstable trinucleotide repeat may play a pathogenic role in the index family reported here as well.
Further clinical investigation involving additional families with GHS is necessary. Such studies can lead to important genetic discoveries in the development and functioning of the hypothalamic-pituitary-gonadal axis in the human.
Acknowledgments
We thank Julia Glaser for all of her meticulous work in pedigree construction.
Footnotes
D.H.M. was supported by Public Health Service Grant AI-01587 from the National Institute of Allergy and Infectious Diseases.
Abbreviations: CT, Computed tomography; CV, coefficient(s) of variation; FAS, free--subunit; GHS, Gordon Homes syndrome; IHH, idiopathic hypogonadotropic hypogonadism.
Received August 27, 2001.
Accepted December 27, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  M. Timmons, M. Tsokos, M. A. Asab, S. B. Seminara, G. C. Zirzow, C. R. Kaneski, J. D. Heiss, M. S. van der Knaap, M. T. Vanier, R. Schiffmann, et al. Peripheral and central hypomyelination with hypogonadotropic hypogonadism and hypodontia Neurology, December 12, 2006; 67(11): 2066 - 2069. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G J Breedveld, B van Wetten, G D te Raa, E Brusse, J C van Swieten, B A Oostra, and J A Maat-Kievit A new locus for a childhood onset, slowly progressive autosomal recessive spinocerebellar ataxia maps to chromosome 11p15 J. Med. Genet., November 1, 2004; 41(11): 858 - 866. [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 |
