Polyglutamine diseases are hereditary neurodegenerative disorders caused by an abnormal expansion of a trinucleotide CAG repeat, which encodes a polyglutamine tract. To date, nine polyglutamine diseases are known: Huntington’s disease (HD), spinal and bulbar muscular atrophy (SBMA), dentatorubral-pallidoluysian atrophy (DRPLA) and six forms of spinocerebellar ataxia (SCA) (Table 1). SBMA, also known as Kennedy’s disease, is the first of the neurodegenerative diseases, for which the molecular basis was discovered to be the expansion of a trinucleotide CAG repeat in the gene of the causative gene.

In general, symptoms of polyglutamine diseases typically appear in mid-life and progressively deteriorate before death from fatal complications. Clinical features vary for each disorder, corresponding to the pathological distribution of neurodegeneration (Table 1).

Major symptoms of SBMA are weakness, atrophy and fasciculations of bulbar, facial and limb muscles [1]. Patients with SBMA occasionally demonstrate signs of androgen insensitivity such as gynecomastia, testicular atrophy, impaired erection and decreased fertility, some of which are detected before the onset of motor impairment. Female carriers are usually asymptomatic, but some express subclinical phenotypes including high amplitude motor unit potentials on electromyography. The progression of SBMA is usually slow, but life-threatening respiratory tract infection often occurs in the advanced stages of the disease, resulting in early death in some patients. The cardinal cause of death is aspiration pneumonia [2].

The molecular basis of SBMA is the expansion of a trinucleotide CAG repeat, which encodes the polyglutamine tract, in the first exon of the androgen receptor (AR) gene [3]. The CAG repeat within AR ranges in size from 11 to 35 in normal subjects, but from 40 to 62 in SBMA patients [3–5]. There is an inverse correlation between the CAG repeat size and the age at onset of motor impairments or the disease severity adjusted by the age at examination in SBMA [6,7] as well as in other polyglutamine diseases [8]. In a nerve conduction study of SBMA, the CAG repeat size and the age at onset were significantly different among the patients with motor- and sensory-dominant phenotypes, indicating that a longer CAG repeat is more closely linked to the motor-dominant phenotype and a shorter CAG repeat is more closely linked to the sensory-dominant phenotype [9].

Although NIs are a disease-specific histopathological finding, their role in pathogenesis has been heavily debated. Several studies have suggested that NIs may indicate a cellular response coping with the toxicity of abnormal polyglutamine protein [23]. Instead, the diffuse nuclear accumulation of the mutant protein has been considered essential for inducing neurodegeneration in polyglutamine diseases including SBMA (Figure 1). Recent data suggest that the toxic species of protein in polyglutamine diseases may be soluble mutant conformers, which can exist as oligomers or monomers containing beta-sheet conformation [24–26].

Although it is difficult to determine the toxic protein species in human histopathology, diffuse accumulation of the causative gene products has been construed as an important finding. An immunohistochemical study on autopsied SBMA patients using an anti-polyglutamine antibody demonstrated that diffuse nuclear accumulation of the pathogenic AR is more frequently observed than NIs in the anterior horn of the spinal cord [28]. Intriguingly, the frequency of diffuse nuclear accumulation of the pathogenic AR in spinal motor neurons strongly correlates with the length of the CAG repeat in the AR gene. No such correlation has been found between NI occurrence and the CAG repeat length. A similar observation has been reported in DRPLA [29]. Taken together, it appears that the pathogenic AR containing an elongated polyglutamine tract principally accumulates within the nuclei of motor neurons in a diffusible form, leading to neuronal dysfunction and eventual cell death in SBMA. In support of this hypothesis, neuronal dysfunction is halted by genetic modulation preventing nuclear import of the pathogenic polyglutamine-containing protein in cellular and animal models of polyglutamine diseases [8].

Since the human AR is widely expressed in various organs, nuclear accumulation of the pathogenic AR protein is detected not only in the central nervous system, but also in non-neuronal tissues such as scrotal skin. The degree of pathogenic AR accumulation in scrotal skin epithelial cells tends to be correlated with that in the spinal motor neurons in autopsy specimens, and it is well correlated with CAG repeat length and inversely correlated with the motor functional scale [30]. These findings indicate that scrotal skin biopsy with anti-polyglutamine immunostaining is a good biomarker with which to monitor SBMA pathogenic processes (Figure 2).

For any given polyglutamine disease, more than one mechanism likely contributes to neuronal dysfunction and eventual cell death. They include: (i) misfolding of the disease protein resulting in altered function; (ii) deleterious protein interactions engaged in by the mutant protein; (iii) formation of toxic oligomeric complexes; (iv) transcriptional dysregulation; (v) mitochondrial dysfunction resulting in impaired bioenergetics and oxidative stress; (vi) impaired axonal transport; (vii) aberrant neuronal signaling including excitotoxicity; (viii) cellular protein homeostasis impairment; and (ix) RNA toxicity [31]. Although each of these molecular mechanisms could be subject to therapeutic interventions, upstream events are more plausible targets than secondary cellular changes.

There is no well-established disease-modifying therapy for SBMA. Potential therapeutics, however, have emerged from basic research using animal models. Among these therapeutic approaches, androgen deprivation has been translated into clinic [27]. Anti-androgen therapies have been developed taking advantage of the fact that the accumulation of the pathogenic AR proteins is dependent on the circulating level of testosterone [32,33]. Surgical castration has been shown to reverse motor dysfunction in mouse models of SBMA [34]. The luteinizing hormone-releasing hormone analogue, leuprorelin, prevents nuclear translocation of aberrant AR proteins, resulting in a significant improvement of disease phenotype in a mouse model of SBMA [35]. These results of animal studies were verified in a phase 2 clinical trial of leuprorelin, in which the patients treated with this drug exhibited decreased mutant AR accumulation in scrotal skin biopsy, significantly higher functional scores and better swallowing parameters than those receiving placebo (Figure 3). Autopsy of one patient who received leuprorelin suggested that androgen deprivation inhibits the nuclear accumulation and/or stabilization of mutant AR in the motor neurons of the spinal cord and brainstem (Figure 4). These observations suggest that administration of leuprorelin suppresses the deterioration of neuromuscular impairment in SBMA by inhibiting the toxic accumulation of mutant AR [36].

Activation of the cellular defense machinery is another promising therapeutic approach for SBMA. Over-expression of heat shock proteins (HSPs), stress-inducible molecular chaperones, inhibits toxic accumulation of abnormal AR protein and suppresses neurodegeneration in a mouse model of SBMA [37]. Similar beneficial effects have also been achieved by the pharmacological induction of HSPs [38]. On the other hand, inhibition of Hsp90 has been demonstrated to arrest neurodegeneration by activating the ubiquitin-proteasome system in SBMA. Treatment with 17-allylamino geldanamycin (17-AAG), a potent Hsp90 inhibitor, dissociated p23 from the Hsp90-AR complex, and thus facilitated proteasomal degradation of the pathogenic AR in cellular and mouse models of SBMA [39, 40]. Similar effects were observed in the SBMA mice being treated with an oral Hsp 90 inhibitor, 17-(dimethylaminoethylamino)-17-demethoxygeldanamycin (17-DMAG) [41].

Transcriptional dysregulation is another target for therapeutic intervention. Because suppression of histone deacetylase (HDAC) activities results in an augmentation of histone acetylation and a subsequent restoration of gene transcription, HDAC inhibitors have been considered to be of therapeutic benefit in polyglutamine diseases [42]. Butyrate was the first HDAC inhibitor to be discovered, and the related compound, phenylbutyrate, has been successfully employed in experimental cancer therapy. Oral administration of sodium butyrate ameliorates the symptomatic and histopathological phenotypes of a mouse model of SBMA through upregulation of histone acetylation in nervous tissues [18]. This compound has also been shown to alleviate neurodegeneration in a mouse model of DRPLA [43]. In mouse models of HD, the administration of HDAC inhibitors (sodium butyrate, suberoylanilide hydroxamic acid and phenylbutyrate) has been shown to alleviate polyglutamine toxicity and improve neuronal dysfunction [44–46].

Although the genetics of polyglutamine diseases were discovered as an abnormal expansion of a trinucleotide CAG repeat, detailed mechanisms of the diseases including SBMA have not been fully elucidated. The clinical trial of leuprorelin acetate suggests that androgen deprivation inhibits the nuclear accumulation and/or stabilization of mutant AR in the motor neurons, and thereby stabilizes the disease progression in SBMA patients. Future research approaches have to determine the main mechanisms which contribute to neuronal dysfunction and eventual cell death in SBMA.