Spinal And Bulbar Muscular Atrophy, X-Linked 1

A number sign (#) is used with this entry because X-linked spinal and bulbar muscular atrophy (SBMA, SMAX1), also known as Kennedy disease, is caused by a trinucleotide CAG repeat expansion in exon 1 of the gene encoding the androgen receptor (AR; 313700.0014). CAG repeat numbers range from 38 to 62 in SBMA patients, whereas healthy individuals have 10 to 36 CAG repeats.

Mutations in the AR gene also cause androgen insensitivity syndrome (AIS; 300068).

Description

Kennedy disease is an X-linked recessive form of spinal muscular atrophy. It occurs only in men. Age at onset is usually in the third to fifth decade of life, but earlier involvement has been reported. The disorder is characterized by slowly progressive limb and bulbar muscle weakness with fasciculations, muscle atrophy, and gynecomastia (Harding et al., 1982). The disorder is clinically similar to, but genetically distinct from, classic forms of autosomal spinal muscular atrophy (see, e.g., SMA1; 253300).

Clinical Features

Kennedy et al. (1968) described spinal-bulbar muscular atrophy in 9 males from 2 unrelated kindreds. Patients had onset of fasciculations followed by muscle weakness and wasting at approximately 40 years of age. Characteristic features included bulbar signs, facial fasciculations, and dysphagia. Three patients had gynecomastia. Pyramidal, sensory, and cerebellar signs were absent, suggesting lower motor neuron involvement. The disorder was compatible with a long life. Early reports of Japanese families may have referred to the same disorder (Takikawa, 1953; Murakami, 1957; Kurland, 1957; Tsukagoshi et al., 1965).

Quarfordt et al. (1970) described 4 brothers with adult-onset proximal spinal muscular atrophy. Type II hyperlipoproteinemia was present in all 4 and was absent from their 1 unaffected sib, a sister. Some children of affected males, too young to show the neurologic abnormality, also showed hyperlipoproteinemia.

Schoenen et al. (1979) listed the clinical hallmarks of Kennedy disease as onset in the third decade, slow progression, involvement of facial and bulbar muscles, and wasting of the proximal and, in some cases, the distal musculature. Clinical signs were usually asymmetric, and there were consistent and abundant fasciculations predominantly in the perioral muscles. Other features included intention tremor and gynecomastia. The disorder shows X-linked recessive inheritance. Endocrinologic studies suggested an anatomic defect in the hypothalamus leading to androgen deficiency and estrogen excess.

Punnett and Schotland (1979) studied a family with 7 affected males in 4 generations.

Pearn and Hudgson (1978) described a spinal muscular atrophy syndrome characterized by adolescent onset, gross hypertrophy of the calves, and a slowly progressive clinical course. They proposed X-linked inheritance for the family of 1 of their patients who had an affected brother and 2 affected maternal uncles. In a study of 100 patients with SMA, Bouwsma and Van Wijngaarden (1980) found 23 cases with hypertrophied calves, elevated serum creatine kinase and onset between 1 and 20 years of age. All were male and many were brothers.

Harding et al. (1982) reported 10 men with SBMA from 8 families. Proximal limb muscle weakness developed in the third to fifth decades of life, often preceded by muscle cramps on exertion and tremor of the hands. Weakness and fasciculation of the facial muscles and tongue were also prominent. All had gynecomastia and some were infertile. Plasma creatine kinase levels were increased and muscle biopsies showed neurogenic atrophy with secondary myopathic changes.

In 4 Italian males in 3 sibships related as first cousins and by history in their maternal grandfather, Guidetti et al. (1986) described an X-linked adult-onset neurogenic muscular atrophy, mainly proximal, with late distal and bulbar involvement. All patients had essential tremor, gynecomastia, and impotence, although all affected males had children. One patient had hyperlipidemia.

Hausmanowa-Petrusewicz et al. (1983) studied 8 of 12 male patients with X-linked spinal muscular atrophy. Six had had gynecomastia. First muscle symptoms began between 21 and 44 years of age, at which time the patients also noted disturbances in their sexual function. Some were reported to have sterility. Biopsy showed pronounced involutional changes in Leydig cells, and plasma testosterone level was decreased. Muscle involvement was most marked proximally in the limbs. Fasciculations in the muscles of the trunk and limbs and fibrillation and atrophy of the tongue were noted. Bulbar muscle involvement was less striking than in reports by others.

Warner et al. (1990) reported 4 affected brothers and an affected maternal uncle. In addition to typical neurologic features including early muscle cramps, visible fasciculation and weakness of oropharyngeal and limb muscles, dysarthria, and areflexia, the patients showed facial asymmetry, gynecomastia, testicular atrophy, and infertility. All 5 affected males had hypobetalipoproteinemia. Warner et al. (1990) pointed to other reported cases of lipid abnormalities, but concluded that the relationship to the neurologic disorder was unclear. They stated that Fischbeck had informed them in 1989 of linkage between the disorder in this family (Fischbeck et al., 1986) and markers on Xq13.21.

Amato et al. (1993) reported 17 SBMA patients from 7 families. One was a 71-year-old man. Weakness in all patients was usually slowly progressive, but they described 1 exception: a man who rapidly progressed from being a healthy firefighter to wheelchair-dependency in less than 1.5 years. Four carrier females had no clinical features of the disorder.

Doyu et al. (1993) reported a case of extraordinarily late onset of this disorder in an 84-year-old Japanese man, with no family history of any related disease, who noted mild difficulty in climbing stairs when he was in his mid-seventies. At the age of 83, painful muscle cramps in the calves and difficulty in walking were more apparent, although he could ride a bicycle and work in the field. At age 84, he showed diffuse muscular weakness and striking amyotrophy in the 4 limbs as well as in the truncal and facial muscles. The tongue was mildly atrophic, but there was no dysphagia. Fasciculation was striking in his face, tongue, neck, anterior chest, and arm and leg muscles, and was enhanced by mild voluntary contractions. There was no gynecomastia. The number of tandem CAG repeats in the first exon of the AR gene was 40, which was the shortest in the authors' series of patients with SBMA; 45 of their cases showed a range from 40 to 55, and the normal range was 17 to 24. No statement was made about children of the patient. This is likely one of the oldest and most mildly affected patients reported.

In studies of 34 patients with Kennedy disease, Sperfeld et al. (2002) found that onset was in adolescence, which is earlier than previously thought. Most frequently, early symptoms were gynecomastia, muscle pain, and premature muscle exhaustion. Weakness was not a typical initial symptom and was frequently found in distal limbs if present early. They found a correlation between the number of CAG repeats and the age at onset of weakness, but not to the age at onset of Kennedy disease.

Echaniz-Laguna et al. (2005) reported an Italian family in which 7 boys had early-onset, rapidly progressive SBMA. Molecular analysis detected expanded CAG repeats ranging from 50 to 54, in the intermediate range. The mean age at onset was 13 years (range 8 to 15) of proximal upper and lower limb atrophy and weakness, which was associated with postural upper limb tremor in 4 patients. All patients developed fasciculations and dysarthria in their teens, and 3 patients had loss of independent ambulation in their mid-twenties. All the boys had prepubertal gynecomastia before age 8, without being overweight. EMG tests in 4 patients showed widespread denervation in all four limbs and bulbar muscles. Echaniz-Laguna et al. (2005) noted that 3 patients had been diagnosed with nonspecific limb-girdle muscular dystrophy in childhood.

Female Homozygotes

Schmidt et al. (2002) described 2 sisters, aged 34 and 42 years, homozygous for the CAG expansion in the AR gene causing Kennedy disease, in whom symptoms were limited to occasional muscle cramps and twitches. Both sibs had mild hand tremor, and the elder had rare perioral fasciculations and mild motor axonal loss in the sternocleidomastoid muscle. Both parents were deceased, but each was thought to have contributed an X chromosome containing the CAG expansion to the sibs. This was supported by a history of muscle disease in the father and the diagnosis of Kennedy disease in a maternal cousin. Women heterozygous for the Kennedy disease gene are generally asymptomatic; however, Kennedy et al. (1968) noted in their original report that 'several female siblings of affected males had muscle cramps.' One of the remarkable features of the homozygous sisters reported by Schmidt et al. (2002) was the mildness of their manifestations. The authors suggested that the increased severity of Kennedy disease in men is due to higher levels of androgen receptor stimulation compared to women, which may produce higher levels of abnormal transcriptional regulation. Blockade of androgen receptor may thus offer some therapeutic benefit in Kennedy disease.

Other Features

Nagashima et al. (1988) described the autopsy findings in 2 affected brothers, both of whom died in their sixties. An unusual feature was the presence of distal sensory neuropathy. In a clinicopathologic study involving 9 cases, with autopsies on 3 and sural nerve biopsies from 6 others, Sobue et al. (1989) also concluded that a lower motor and primary sensory neuronopathy is a major feature. According to them, the main feature of Kennedy disease distinguishing it from other autosomal forms of SMA is the presence of sensory abnormalities; sensory abnormalities are not found in true SMA.

In their series of 10 patients, Harding et al. (1982) found that most patients had decreased sensory nerve action potentials in the absence of clinical sensory loss. Wilde et al. (1987) reported neurophysiologic investigations and sural nerve biopsy in patients with SBMA. The findings confirmed that both motor and sensory nerves were affected. The authors stressed the importance of recognizing this disorder as a separate entity which should not be classified with the spinal muscular atrophies. Both Harding et al. (1982) and Wilde et al. (1987) referred to the disorder as X-linked bulbospinal neuronopathy. Boylan (1991) even suggested that Kennedy disease should more accurately be characterized as a motor sensory neuronopathy rather than as an SMA.

Kachi et al. (1992) also reported prolonged sensory conduction times in patients with SBMA. These results were consistent with pathologic findings in which primary sensory neurons are involved as well as the lower motor neurons, whereas upper motor neurons are well preserved.

Dejager et al. (2002) performed detailed endocrine investigations in 22 men with Kennedy disease. Clinical signs of partial androgen resistance were present in more than 80% of the patients, with gynecomastia being the most prominent. Thirteen patients had alteration of testicular exocrine function. Hormonal profile of partial androgen resistance was present in 86% of the patients, with an elevated testosterone level in 68%. Dejager et al. (2002) noted that androgen insensitivity seems to appear later in life in Kennedy disease, similar to the development of neurologic signs. The authors stated that in clinical practice, Kennedy disease patients are often misdiagnosed as having ALS, and that careful examination of the endocrine component could avoid such a deleterious misdiagnosis.

Sumner and Fischbeck (2002) reported 2 patients who presented with 'jaw drop,' preferential weakness of the temporalis and masseter muscles, who were later found to have expanded CAG repeats in the AR gene. Both patients were given incorrect diagnoses at first, including diabetic sensory polyneuropathy, myasthenia gravis (254200), and ALS. The authors emphasized that jaw drop may be a presenting sign in patients with SBMA.

Sperfeld et al. (2005) found that 23 (47%) of 49 patients with Kennedy disease had recurrent laryngospasm compared to 3 (2%) of 147 patients with early-stage ALS.

Diagnosis

Prenatal Diagnosis

Georgiou et al. (2001) developed a single-cell PCR assay for the AR gene and described the application of this assay for preimplantation genetic diagnosis in a couple at risk, where the female partner was a carrier of 47 repeats. Diagnosis was based on the detection of both normal and expanded alleles. Allele dropout of the expanded allele was observed. Neither expansions nor contractions were observed in the blastomeres biopsied from 11 embryos. Two embryos were unaffected, 8 were female carriers, and 1 was an affected male embryo.

Mapping

Fischbeck et al. (1986) found linkage of Kennedy disease to a locus on chromosome Xq21.3-q22 (maximum lod score of 3.53 at theta = 0.04 for marker DXYS1).

Mukai and Yasuma (1987) found that colorblindness (see 303900) and bulbospinal muscular atrophy were not closely linked in a Japanese family.

Ferlini et al. (1991) mapped the SBMA locus to Xq12 by linkage with a considerable number of RFLP markers.

Molecular Genetics

In 35 unrelated patients with SBMA, La Spada and Fischbeck (1991) and La Spada et al. (1991) identified an expanded CAG repeat in the first exon of the AR gene (313700.0014). The abnormality was not observed in 263 normal persons. The AR CAG repeat is normally polymorphic, with an average repeat number of 22 +/- 3. SBMA patients had 11 different (CAG)n alleles, with repeat numbers ranging from 40 to 52, more than 6 standard deviations above the normal mean. The AR gene abnormality segregated with the disease in 15 SBMA families, with no recombination in 61 meioses (lod score = 13.2 at theta = 0.0). The CAG repeat encodes a polyglutamine tract in a portion of the AR protein that is not directly involved in hormone- or DNA-binding. There was no correlation between the size of the CAG repeat and the presence of altered in vitro androgen binding. However, there was correlation between CAG repeat length and disease severity; the mildest clinical manifestations were associated with the smallest CAG repeat.

Ferlini et al. (1995) identified a CAG repeat expansion in the AR gene in 3 of 25 sporadic patients with heterogeneous motor neuron diseases.

By in vitro examination of different tissues from 2 SBMA patients, Spiegel et al. (1996) did not find somatic instability of the CAG repeat expansion. Jedele et al. (1998) found no evidence of somatic mosaicism in multiple tissues from a fetus with SBMA.

Genotype/Phenotype Correlations

La Spada et al. (1991) observed a correlation between CAG length and disease severity. In an analysis of 26 Japanese SBMA patients from 21 families, Doyu et al. (1992) found the same results: the greater the number of CAG repeats, the lower the age of onset of limb muscular weakness and the higher the age-adjusted disability score. La Spada et al. (1992) concluded that although there was a correlation between disease severity and CAG repeat length, other factors seemed to contribute to the phenotypic variability. They found that expanded (CAG)n alleles underwent alteration in length when transmitted from parent to offspring. Of 45 meioses examined, 12 (27%) demonstrated a change in CAG repeat number. Both expansions and contractions were observed, although their magnitude was small. There was a greater rate of instability in male meiosis than in female meiosis.

The findings of Amato et al. (1993) in 17 patients from 7 families appeared to be at variance with the findings of others: no large expansion of the mutation was observed in transmission through 3 generations of 1 family, and phenotypic expression, although variable between and within families, was not related to the size of the mutation. The authors stated that anticipation had not clearly been observed in Kennedy disease.

Pathogenesis

Warner et al. (1992) studied androgen receptor function in cultured scrotal skin fibroblasts from 8 patients with Kennedy syndrome from 4 families. High-affinity dihydrotestosterone binding was decreased in 3 patients from 1 family similar to values in subjects with androgen resistance syndromes. The values were normal in 5 SBMA patients from 3 other families.

Ogata et al. (1994) used immunohistochemical staining with monoclonal antibody 5F4 directed against the androgen receptor to examine the distribution of androgen receptors in spinal cord and brainstem in 2 cases of Kennedy spinal bulbar muscular atrophy and in 4 cases of sporadic ALS. In all cases, there was dense immunoreactivity in the nuclei of anterior horn cells, as well as some neurons of the substantia gelatinosa, nucleus proprius, substantia intermedia, and central gray. In the brainstem, there was reactivity in the neurons of cranial nerves III, IV, and VI, as well as of the nucleus of Onufrowicz. Even severely affected motor neurons showed positive immunostaining.

By postmortem examination of 5 SBMA patients, Banno et al. (2006) found that the degree of mutant AR accumulation in spinal motor neurons correlated to that of scrotal skin epithelial cells. Mutant AR accumulation also correlated with CAG repeat length and disease severity. Scrotal biopsies from 13 additional patients also correlated with CAG repeat length and disease severity. In 5 additional SBMA patients who received subcutaneous leuprorelin injections, mutant AR protein was decreased in scrotal skin biopsies. After 24 weeks of treatment, there were no significant motor function changes, but 3 patients expressed subjective improvement. Banno et al. (2006) concluded that scrotal skin biopsy findings may be a biomarker of SBMA and could be used in therapeutic clinical trials.

McCampbell et al. (2000) demonstrated that CREB-binding protein (CREBBP; 600140) is incorporated into nuclear inclusions formed by polyglutamine-containing proteins in cultured cells, transgenic mice, and tissue from patients with SBMA. Soluble levels of CREBBP were reduced in cells expressing expanded polyglutamine despite increased levels of CREBBP mRNA. Overexpression of CREBBP rescued cells from polyglutamine-mediated toxicity in neuronal cell culture. The authors proposed a CREBBP-sequestration model of polyglutamine expansion disease.

Population Genetics

Tanaka et al. (1996) investigated the origin of the SBMA mutations in the Japanese population by analyzing the (CAG)n and (GGC)n repeats of the AR gene locus in unrelated SBMA and normal X chromosomes in Japanese males. They found linkage disequilibrium between the (GGC)n haplotype and the (CAG)n mutation. Their results demonstrated a founder effect for SBMA in a Japanese population, indicating that de novo pathologic expansion of the CAG repeat is probably rarer in SBMA than in other diseases due to triplet repeat expansion.

Udd et al. (1998) estimated the prevalence of Kennedy disease in the Vasa region of western Finland to be 13 per 85,000 male inhabitants. Kennedy disease was the most common motor neuron disorder in the Vasa region, exceeding the prevalence of ALS by a factor of 2. None of the patients, despite previous examinations, had correct diagnoses before 1995. Nine of the 10 families belonged to the Swedish-Finnish language group, yielding a 1.75 times higher prevalence within that specific part of the population. All cases were confirmed by DNA tests. The range of CAG-repeat expansions in the patients was 41 to 47 repeats (normal up to 35). There was no clear correlation between the repeat number and the clinical severity of the disease. One heterozygous female alive in her eighties had a movement disability beginning at age 60. Her diagnosis remained unclassified progressive encephalopathy with myoclonus leading to a bedridden state. It was uncertain whether the heterozygous AR mutation had any causal connection with her CNS disorder.

Lund et al. (2000) haplotyped 13 Finnish, 10 Swedish, 12 Danish, and 2 Norwegian SBMA families with a total of 45 patients and 7 carriers for multiple microsatellite markers spanning a 25.2-cM region around the AR gene in search of a genetic founder effect. All the Scandinavian SBMA families shared the same 18-repeat microsatellite allele for the intragenic GGC repeat, which was present in only 24% of the controls. Linkage disequilibrium was also seen for the closest microsatellite markers. In addition, extended haplotypes of the Finnish, Swedish, and Danish SBMA families revealed country-specific common founder haplotypes, which over time became gradually shortened by recombinations. No common haplotype was found among the controls. The data suggested that the SBMA mutation was introduced into western Finland 20 generations earlier. Haplotype analysis implied a common ancestor for most Scandinavian SBMA patients.

Lund et al. (2001) haplotyped 123 SBMA families (from Finland, Sweden, Norway, Denmark, Germany, Belgium, Italy, Japan, Australia, and Canada) for the intragenic SNP marker ARd12, the intragenic GGC repeat, and 16 microsatellite markers spanning a 25.2 cM region around the AR gene. All the Finnish, Swedish, and Norwegian patients carried the same kind of haplotype with absent intragenic ARd12 StuI site, while the patients from all other countries, including Denmark, harbored the StuI site, suggesting that the Danish patients derived their disease chromosome from another ancestor than the previously reported common Scandinavian founder (Lund et al., 2000). The haplotype analysis showed 2 founder haplotypes in German, 3 in Italian, 2 in Japanese, and 2 in Australian patients, while no common haplotype was detected among the Canadian patients. These results implied that the CAG repeat expansion mutation in SBMA is not a unique event. No particular expansion-prone haplotype could be detected. Among 95 SBMA patients with defined ages at onset, the authors found a weak negative correlation between the CAG repeat length and the age of onset.

History

Sinnreich and Klein (2004) reviewed the early history of Kennedy disease.

Animal Model

McManamny et al. (2002) developed a transgenic mouse model of SBMA expressing a full-length human AR cDNA carrying 65 (AR-65) or 120 CAG repeats (AR-120), with widespread expression driven by the cytomegalovirus promoter. Mice carrying the AR-120 transgene displayed behavioral and motor dysfunction, while mice carrying 65 CAG repeats showed a mild phenotype. Progressive muscle weakness and atrophy was observed in AR-120 mice and was associated with the loss of alpha-motor neurons in the spinal cord. There was no evidence of neurodegeneration in other brain structures. Motor dysfunction was observed in both male and female animals, suggesting that the polyglutamine repeat expansion may cause a dominant gain-of-function mutation in AR. The male mice displayed a progressive reduction in sperm production consistent with testis defects reported in human patients.

Monks et al. (2007) found that transgenic mice overexpressing wildtype human AR exclusively in skeletal muscle displayed androgen-dependent muscle weakness and early death. Transgenic mice also showed changes in muscle morphology and gene expression consistent with neurogenic atrophy and exhibited motor axon loss. These features reproduced those seen in models of Kennedy disease. Monks et al. (2007) concluded that toxicity in skeletal muscle is sufficient to cause motoneuron disease and that overexpression of AR can exert toxicity comparable with that of the polyglutamine-expanded protein.

Montie et al. (2009) genetically manipulated the nuclear localization signal of polyglutamine-expanded AR. Transgenic mice expressing this mutant AR displayed inefficient nuclear translocation and substantially improved motor function compared with Sbma mice. Analysis of cell models of SBMA indicated that nuclear localization of polyglutamine-expanded AR was necessary but not sufficient for aggregation and toxicity and that androgen binding by AR was required for these disease features. Studies of cultured motor neurons showed that the autophagic pathway was able to degrade cytoplasmically retained polyglutamine-expanded AR and represented an endogenous neuroprotective mechanism. Pharmacologic induction of autophagy rescued motor neurons from the toxic effects of even nuclear-residing mutant AR, suggesting a therapeutic role for autophagy in this nucleus-centric disease. Montie et al. (2009) concluded that polyglutamine-expanded AR must reside within nuclei in the presence of its ligand to cause SBMA.

Therapeutic Strategies

In Drosophila and human tissue culture models of SBMA, Caplen et al. (2002) found that sequence-specific small dsRNAs of 22 nucleotides rescued the toxicity and caspase-3 activation induced by plasmids expressing a transcript encoding an expanded AR polyglutamine tract.

In a mouse model of SBMA (AR-97Q), Katsuno et al. (2003) showed that leuprorelin, a luteinizing hormone-releasing hormone (LHRH) agonist that prevents testicular testosterone production, reversed both the motor dysfunction and nuclear accumulation of mutant androgen receptors in male transgenic mice. Flutamide, an androgen antagonist that promotes nuclear translocation of androgen receptors, yielded no therapeutic effect. The authors concluded that leuprorelin exhibited a therapeutic effect by preventing mutant androgen receptor translocation and nuclear accumulation.

Minamiyama et al. (2004) studied the therapeutic effects of sodium butyrate (SB), a histone deacetylase (see HDAC1; 601241) inhibitor, in a transgenic mouse model of SBMA. Oral administration of SB ameliorated neurologic phenotypes and increased acetylation of nuclear histone in neural tissues. These therapeutic effects, however, were seen only within a narrow range of SB dosage. The authors suggested that SB may be a possible therapeutic agent for SBMA and other polyQ diseases.

In cultured rat cells transfected with AR-112Q, Yang et al. (2007) found that treatment with ASC-J9, a synthetic analog of curcumin, disrupted the interaction between AR-112Q and its coregulator and also increased cell survival by decreasing AR-polyQ aggregation and increasing AR-polyQ degradation. Intraperitoneal injection of ASC-J9 into AR-97Q mice resulted in improved disease symptoms and a decrease in muscular atrophy. Treated mice retained normal sexual function and fertility.

In a transgenic mouse model of SBMA, Tokui et al. (2009) showed that ubiquitin-proteasomal function was well preserved and was even increased during advanced stages when the mice developed severe phenotypes. Oral administration of the Hsp90 (140571) inhibitor 17-(dimethylaminoethylamino)-17-demethoxygeldanamycin (17-DMAG) markedly ameliorated motor impairments in SBMA mice without detectable toxicity and reduced amounts of monomeric and nuclear-accumulated mutant AR. Mutant AR was preferentially degraded in the presence of 17-DMAG in both SBMA cell and mouse models when compared with wildtype AR, and 17-DMAG also significantly induced Hsp70 (see 140550) and Hsp40 (see 602837). Tokui et al. (2009) concluded that 17-DMAG would exert a therapeutic effect on SBMA via preserved proteasome function.