Spinal and bulbar muscular atrophy: androgen receptor dysfunction caused by a trinucleotide repeat expansion

JOURNAL

OF THE

NEUROLOGICAL SCIENCES Journal of the Neurological Sciences 135 ( 1996) 149- 157

Spinal and bulbar muscular atrophy: androgen receptor dysfunction caused by a trinucleotide repeat expansion Helen E. MacLean aT*,Garry L. Warne a, Jeffrey D. Zajac b a Centre for Hormone Research, Royal Children’s Hospital, b Department of Medicine, University of Melbourne, Royal Melbourne

Parkville, Hospital,

Victoria, Parkuille,

3052, Australia Victoria, 3050, Australia

Received 29 June 1995; revised 29 August 1995; accepted 8 September 1995

Abstract Kennedy’s disease (spinal and bulbar muscular atrophy) is an X-linked form of motor neuron disease that affects adult men. The syndrome is characterized by progressive atrophy of the limb muscles, pelvic and shoulder girdles and dysphagia and dysarthria, and is caused by the degeneration of spinal and bulbar motor neurons. Kennedy’s disease is caused by a trinucleotide repeat expansion of a CAG repeat in exon A of the androgen receptor gene, and is one of a group of neurological diseases caused by trinucleotide repeat expansions in different genes. The mutation in Kennedy’s disease involves an increased number of glutamine residues in the amino-terminal domain of the receptor. Point mutations and deletions in the androgen receptor gene cause androgen insensitivity syndrome, however subjects with Kennedy’s disease have normal virilization, although progressive gynaecomastia, testicular atrophy and infertility may occur. Androgen receptors are expressed widely in the normal brain, and in the anterior horn cells of the spinal cord; however, their role in neuronal tissue is not known, nor is it known how the androgen receptor gene mutation causes neuronal degeneration. Kennedy’s disease is likely to be a ‘gain of function’ abnormality, so that the presence of the receptor with an increased number of glutamines is toxic to motor neurons. It is possible that the mutation alters interaction of the receptor with other neuronal transcription factors, or neuronotoxicity may occur because of a non-specific effect caused by the presence of a protein with a large homoglutamine domain. Studies of patients with Kennedy’s disease have shown that expression of androgen receptor mRNA and protein in spinal cord may be decreased, as can be the affinity of the mutant receptor for androgen. In vitro studies have shown impaired transcription activation ability of the mutant androgen receptor. The age at onset of Kennedy’s disease may correlate with the size of the CAG repeat, however there is a large degree of variability of age at onset between subjects with the same number of repeats. Further study of the effect of the Kennedy’s disease mutation on androgen receptor function in motor neurons will allow us to increase our understanding of the pathogenesis of this disease. Keywords:

Kennedy’s disease; Spinal and bulbar muscular atrophy; Androgen receptor; Trinucleotide

1. Introduction Kennedy’s disease (spinal and bulbar muscular atrophy) is an adult onset form of motor neuron disease that affects males only, and is caused by an abnormality of the androgen receptor (AR). It was first recognized as a discrete syndrome by Kennedy, Alter and Sung (Kennedy et al., 19681, who described a slowly progressive syndrome with muscle cramps in the 4th or 5th decade progressing to generalised fasciculations and proximal muscle weakness with bulbar involvement. The disease affects men only, and is inherited in an X-linked recessive fashion.

* Corresponding author. Tel.: +61 3 9345 6604; Fax: +61 3 9347

7763. 0022-510X/96/$15.00 SSDI 0022-5

0 1996 Elsevier Science B.V. All rights reserved 10X(95)00284-7

2. Clinical

repeat; Neurodegenerative disease; Nervous system

description

Kennedy’s disease is characterized by its slowly progressive nature, and bulbar involvement. Severe muscle cramps can occur ten to fifteen years before any other signs (Stefanis et al., 1975; Harding et al., 1982). Muscle weakness usually begins in the shoulder and pelvic girdles, with distal involvement detectable later in the course of the disease. Weakness of the bulbar muscles occurs causing dysphagia and dysartbria, and aspiration pneumonia can occur in the terminal stages (Harding et al., 1982). Gynaecomastia is present in more than half of affected individuals, and testicular atrophy, progressive infertility and impotence can occur (Arbizu et al., 1983). EMG studies show motor unit action potentials decreased in number but in-

150

H.E. MucLean

et al./Journnl

oj’rhe Neurological

creased in amplitude (Barkhaus et al., 1982), and reflexes are absent (Stefanis et al., 1975; Nagashima et al., 1988). Post-mortem findings include marked loss of neurons in the motor nuclei of the Vth nerve as well as facial, ambiguous and hypoglossal nuclei and anterior horn cells of the entire spinal cord (Kennedy et al., 1968; Nagashima et al., 1988). Histological studies of testes have shown spermatogenesis arrested at the spermatid level with occasional clustered Leydig cells (Arbizu et al., 1983) or complete hyalinization of seminiferous tubules with Leydig cells diffusely and nodularly increased (Nagashima et al., 1988). Hormonal studies on subjects with Kennedy’s disease show no clear pattern of abnormality. Serum testosterone and estradiol levels are usually normal, and some cases of increased levels of luteinizing hormone (LH) and follicle stimulating hormone (FSH) have been reported (Warner et al., 1990). There appears to be no correlation between hormone levels and the development of gynaecomastia. The age at onset and severity of Kennedy’s disease is highly variable. Individuals as young as ten years have shown symptoms of the disease, although onset usually occurs in the 30s or 40s. Severity can range from mild muscular weakness and tremor to severe disabling weakness and death by aspiration pneumonia. The disease is compatible with a normal life-span.

3. AR gene mutation in Kennedy’s disease Prior to the identification of the Kennedy’s disease mutation, chromosomal mapping studies had mapped the disease locus to the long arm of the X chromosome (Fischbeck et al., 1991). The neuronal loss in Kennedy’s disease is limited to motor nuclei that express the AR (Sar and Stumpf, 1977), and this fact combined with the genetic localisation and endocrine component to the disease pointed towards the AR gene as a candidate for the disease locus. The AR is the ligand-dependent nuclear transcription factor that mediates androgen action (Lubahn et al., 1988). The receptor consists of three functional domains: the steroid binding domain, the DNA binding domain and the amino-terminal domain (Brinkmann et al., 1989). The only mutations of the AR previously identified resulted in androgen resistance, causing androgen insensitivity syndrome (AIS). Mutations causing AIS usually affect either the DNA or steroid binding domains of the receptor (McPhaul et al., 1993). Complete abolition of AR function causes affected XY individuals to develop with female external phenotype (Quigley et al., 1992) and in partial AIS phenotypes can include ambiguous genitalia, micropenis, gynaecomastia and infertility. No neurological dysfunction has been reported in any case of AIS. La Spada and co-workers identified an expansion in the number of trinucleotide CAG repeats in exon A of the AR gene in 35 unrelated patients with Kennedy’s disease (La Spada et al., 1991). With a LOD score of 13.2 indicating a

Sciences 135 (1996)

149-157

Fig. 1. AR gene analysis in Kennedy’s disease. PCR products spanning the CAG repeat of exon A were amplified using PCR and fractionated on a 1.4% agarose gel. Lane 1 contains pGem DNA size markers; normal male DNA is in lanes 4, 5, 7, 9, 10 and 11; normal female DNA is in lane 8; heterozygote female DNA is in lane 2; affected male DNA is in lanes 3, 6 and 12; and lane 13 contains the negative control (no DNA). The arrow indicates the enlarged PCR products observed in Kennedy’s disease.

very highly significant linkage between the CAG repeat expansion and the disease, the data overwhelmingly supported the hypothesis that this mutation was the cause of the disease. This mutation has now been identified in more than 150 patients with Kennedy’s disease, and is now the most common type of AR gene mutation. Kennedy’s disease can be rapidly diagnosed using the polymerase chain reaction (PCR) (Saiki et al., 1988) to amplify a DNA fragment across the CAG repeat region of the AR gene from patient DNA. The PCR products are size fractionated on an agarose gel, and compared in size to a known DNA size standard (Fig. 1). This test can be used to identify affected individuals, heterozygote females and pre-symptomatic males. The trinucleotide CAG repeat of exon A is polymorphic in the normal population, with the size varying between 12 and 30. In Kennedy’s disease the expanded repeat ranges in number from 40 to 72 (Igarashi et al., 1992; Doyu et al., 1992; Yamamoto et al., 1992; Amato et al., 1993). The CAG repeat is in the 5’ coding region of the gene, and encodes glutamine (Gln) residues in the amino-terminal domain of the receptor, which is involved in the modulation of transcription activation (Rundlett et al., 1990; Simental et al., 1991). Many transcription factors have Glnrich regions (Courey and Tjian, 1988; Kao et al., 1990), and it is likely that these form transactivation domains (Gerber et al., 1994). Synthetic poly(Gln) stretches have been shown to form P-sheets, and these may act in vivo as polar zippers to join specific complementary proteins involved in transcription activation (Perutz et al., 1994).

4. AR in the nervous system Many parts of the brain express AR, including the brainstem, hypothalamus, cerebellum and amygdala (Clancy et al., 1992; Simerly et al., 1990). The rat brain shows no gross sexual dimorphism of receptor expression, however more subtle differences in regulation of AR ex-

H.E. MucLem et uL/Journal

of the New&&l

pression between males and females cannot be ruled out. In the brain, ARs are found predominantly in neurons, however glial cells also express AR in vivo and in vitro (Jung-Testas et al., 1992; Sheridan and Fujimoto, 1988). ARs are present in motor neurons of the Vth, VIIth, Xth and XIIth cranial nerves (Simerly et al., 1990). ARs are also found in the foetal and post-natal rat spinal cord, with immunohistochemical studies localizing AR protein to the anterior horn cells (Matsuura et al., 1993) and the ventral aspect of the lumbar spinal cord (Cain et al., 1994). Tissue-specific expression of AR may occur in the spinal cord, with one study showing 41 kDa and 95 kDa forms of the AR expressed in prostate tissue, but only the 41 kDa form of the receptor found in spinal cord (Matsuura et al., 1993). Androgens play an important role in motor neuron growth, development and regeneration (Kurz et al., 1986; Goldstein and Sengelaub, 1992). There are sexually dimorphic nuclei in the rat that are androgen dependent (Nordeen et al., 19851, with androgen acting through the AR to prevent cell death in the male nuclei during foetal development (Sengelaub et al., 1989). Androgens can also attenuate cell death in adult rats, with androgens preventing cell degeneration of axotomized motor neurons following nerve transection (Yu, 1988). It is thought that programmed cell death (apoptosis) occurs in both these cases, and that androgen treatment suppresses apoptosis. Androgens are known to suppress apoptosis in the prostate, with withdrawal of androgens following castration causing prostate regression (Isaacs et al., 1994).

5. AR function in Kennedy’s disease In subjects with Kennedy’s disease AR function in classical sex target tissues is adequate to cause normal male differentiation, virilization, and normal fertility in many cases, but with progressive gynaecomastia, azoospetmia, testicular atrophy and impotence occurring. Three studies have investigated the ability of the mutant AR to bind androgens in cultured genital skin fibroblasts, a well-characterized model for measurement of AR function in AIS. Warner et al. (1992) found a lowered total AR binding in scrotal skin fibroblasts from three subjects from one family with Kennedy’s disease, and Danek et al. (1994) described a decreased maximal binding in genital skin fibroblasts from one subject. Our group found that five out of six subjects had a significant decrease in affinity of AR for androgen in cultured supra-pubic skin fibroblasts (MacLean et al., 19951, indicating that the Kennedy’s disease mutation can affect the ability of the receptor to bind ligand. The progressive development of infertility and gynaecomastia in subjects with Kennedy’s disease could be related to the gradual decrease in testosterone levels that occurs in aging males (Nankin and Calkins, 1986), although most individuals continue to have

Sciences I35 (1996) 149-157

151

testosterone levels within the normal range throughout their life (Warner et al., 1990). The decrease in maximal binding also suggests that AR expression may be decreased in Kennedy’s disease. A post mortem study carried out by Nakamura et al. (1994) on one patient showed using quantitative reverse transcriptase-PCR that there was decreased AR mRNA in the Kennedy’s disease spinal cord, compared to subjects with amyotrophic lateral sclerosis (ALS) and lung cancer. Westem blot analysis also showed a decrease in amount of AR protein in the spinal cord compared to ALS and lung cancer subjects. It is not known whether this decrease in amount of AR mRNA and protein occurs throughout the life of Kennedy’s disease subjects, or occurs gradually leading to the progressive motor neuron degeneration. One study using an antibody to the steroid binding domain of the AR showed no detectable staining in the upper prickle cells of the scrotum in three Kennedy’s disease patients (Matsuura et al., 1992). This lack of detectable AR protein may have arisen because of loss of AR, decrease in AR stability or change in AR conformation. These findings suggest that perturbation of AR levels may be occurring in Kennedy’s disease, however these results remain to be confirmed by further studies.

6. In vitro studies on Kennedy’s AR The ability of the Kennedy’s disease mutant AR to activate transcription of a responsive promoter has been measured in an in vitro system (Mhatre et al., 1993). AR cDNAs constructed with either 40 or 50 CAG repeats were co-transfected into monkey kidney (COS) cells with a known androgen responsive promoter linked to a growth hormone reporter gene. Expression of growth hormone correlated inversely with the length of the CAG repeat, indicating the expansion of the repeat impairs the ability of the receptor to activate transcription. This study showed that binding of androgen to the mutant receptor was normal with either 40 or 50 CAG repeats, in contrast to binding in cultured genital skin fibroblasts from affected subjects.

7. Trinucleotide

repeat diseases

Kennedy’s disease is one of an increasing number of diseases known to be caused by expansion of trinucleotide repeat sequences in different genes (Willems, 1994). These diseases can be classified into two groups based on whether the repeat expansion occurs in the coding or non-coding region of a gene (Table 1). In the first group, which comprises Kennedy’s disease, Huntington’s disease (HD) (Huntington’s Disease Collaborative Research Group, 19931, dentatorubral-pallidoluysian atrophy (DRPLA) (Nagafuchi et al., 1994) (also described as Haw River

152

H.E. MacLean et al.//ournal

of the Neurological Sciences 135 (1996) 149-157

Table 1 Summary of disorders caused by trinucleotide repeat expansions Disease

Chromosome

Trinucleotide

Location in gene

Repeat length

Correlation between repeat length vs.

Normal

Mutant

Increasing severity through parental transmission

Kennedy’s disease

X

CAG

5 translated

17-29

40-72

Age at onset +/-

Huntington’s disease Dentatorubral-pallidoluysian atrophy/ Haw river syndrome Spino-cerebellar ataxia I Machado-Joseph disease

4 12

CAG CAG

5’ translated 5’ translated

7-34 7-25

37-150 49-75

+ +

3-13

63-68

-

6 14

CAG CAG

5’ translated 5’ translated

19-36 13-36

43-81 68-79

+ +

+

paternal -

Myotonic dystrophy

19

CTG

5-30

50- > 3000

+

+

maternal

Fragile X A

X

CGG

2-65

> 200

+

Fragile X E Fragile X F

X X

GCC CGG

3’ untranslated region 5’ untranslated region ?not transcribed ?not transcribed

6-25 6-29

> 200 300-500

Fragile 16A

16

CCG

?not transcribed

16-49

- 3.0kb- 5.7kb

syndrome (Burke et al., 1994)), spino-cerebellar ataxia I (SCA I) (Orr et al., 1993) and Machado-Joseph disease (MJD) (Kawaguchi et al., 19941, the trinucleotide is a CAG repeat that encodes Gln residues in the respective proteins. In each case the size of the CAG repeat expansion appears constrained, so that the diseases usually have less than 100 repeats. It may be that the affected proteins cannot tolerate very large homo(Gln) tracts, so that the larger expansions are selected against, or disrupt protein function to such a degree that they cause a different, loss-of-function phenotype. The second group, in which the expansion of the trinucleotide repeat appears to be outside the coding region of the genes, includes myotonic dystrophy (DM) (Mahadevan et al., 1992), Fragile X syndrome (FRAX A) (Kremer et al., 19911, Fragile X E (FRAX E) (Knight et al., 19931, Fragile X F (FRAX F) (Parrish et al., 1994) and Fragile 16A (FRA16 A) (Nancarrow et al., 1994). The size of the repeat expansions within this group is much larger, with up to 6 kb expansions observed in myotonic dystrophy. Because they occur in non-coding sequence, greater size increases are likely to be able to be tolerated. Within this latter group the distinction must be made between myotonic dystrophy and FRAX A, in which the repeat expansions are known to occur in the non-coding portions of genes and are associated with distinct clinical syndromes, and the other conditions that appear to cause fragile sites in the genome that may be unrelated to any pathological condition. Interestingly, all the CAG repeat expansion disorders manifest themselves with neurological abnormalities, usually of late onset and slow progression. The pathogenesis of these diseases is not known, and in most cases the normal function of the affected genes is also unknown. It

Severity

paternal paternal

+

-

maternal +

-

is unlikely to be coincidental that all of these conditions with increased number of Gln residues result in neuronal degeneration.

8. Relationship severity

between

repeat length

and disease

The size of the CAG repeat expansion in Kennedy’s disease appears to be merely one factor that influences the age at onset of the disease. We have found a trend of increasing numbers of CAG repeats being associated with a younger age at onset for a number of symptoms however there was large variation between individuals, and none of these correlations was statistically significant (MacLean et al., 1995). Igarashi et al. (1992) reported a strong inverse correlation (r = - 0.801) between the CAG repeat number and the age at onset of muscular weakness only, and Doyu et al. (1992) found a significant inverse correlation (r = - 0.596, p < 0.001) between the CAG repeat number and the age at onset of limb weakness. An inverse correlation between the CAG repeat number and the age at onset and age of stair climbing difficulty was reported by La Spada et al. (19921, however these authors found that neither the age of wheelchair dependence nor the signs of androgen insensitivity correlated with the size of the CAG repeat expansion. This study also reported large variations in the age at onset and rate of progression of the disease within families. Similarly, other groups (Biancalana et al., 1992; Amato et al., 1993) have reported phenotypic variation within and between families, with no correlation between CAG repeat number and features of the disease including age at onset and severity of weakness. Thus although the number of CAG repeats in the AR gene may be influenc-

H.E. MacLean et al./Journal

of the Neurological

ing the age at onset, this effect appears to be quite variable, and other factors must also be controlling the onset and progression of the disease. The CAG repeat in Kennedy’s disease shows little meiotic instability, with only very small changes in repeat size occurring over generations, and genetic anticipation (increasing length and severity over generations) does not occur. Most of the other trinucleotide repeat expansion diseases show some degree of correlation between size of the repeat expansion and the age at onset or severity of the disease, although within most of the diseases there is also a great deal of variation. In HD the size of the CAG only accounts for approximately 50% of the variation in age at onset, with the association between repeat size and age at onset strongest for the juvenile onset form of the disease (Craufurd and Dodge, 1993). Similar observations have been made for DM (Buxton et al., 1992) and MJD (Kawaguchi et al., 1994). In SCA I both the age at onset and severity of the disease show an inverse correlation with repeat size (Orr et al., 1993, Chung et al., 1993). DRPLA shows an inverse correlation between age at onset and CAG repeat number, however only half of the variation in age at onset is accounted for by repeat number (Koide et al., 1994). Juvenile onset of DRPLA is usually associated with larger expansions, however there is a broad range of larger repeat numbers within the juvenile onset group (Nagafuchi et al., 1994). In Fragile X syndrome maternal transmission of expanded alleles causes an increase in repeat number, which is associated with conversion of the premutation to the full mutation, and expression of the disease phenotype (Vaisanen et al., 1994). The conclusion that can be drawn from these data is that in general increased trinucleotide repeat length appears to be associated with earlier age at onset, and sometimes severity of the disease. In particular, juvenile forms of most of these diseases appear to be associated with larger repeat alleles, which occurs with only maternal transmission in some cases and paternal transmission of the expanded allele in others (Table 1). Nevertheless for most of the diseases the amount of variation in age at onset accounted for by repeat size is only around 50%, so there are clearly other factors influencing both the onset and progression of the disease. In our study on cultured supra-pubic skin fibroblasts from subjects with Kennedy’s disease, we found a correlation between the affinity of the AR for androgen and the signs of androgen insensitivity (MacLean et al., 1995). Thus in those patients who showed a greater binding dysfunction, breast development and testicular atrophy was increased. This result is not surprising when related to patients with AIS, in whom there is a general correlation between the degree of receptor dysfunction and the severity of the syndrome (McPhaul et al., 1993). Therefore one component of the symptoms of Kennedy’s disease, the signs of androgen insensitivity, appears to be related specifically to the degree to which the affinity of receptor

Sciences 135 (1996) 149-157

153

for ligand is impaired. However, as there appeared to be no relationship between the number of CAG repeats and the AR affinity in these patients, the difficulty in relating the degree of receptor dysfunction to the size of the mutation remains.

9. Theories for mutation The pathogenic mechanism causing neuronal degeneration in Kennedy’s disease is not known. Affected men are hemizygous, with just one copy of the mutant AR gene, which is located on the X chromosome. Total deletion of the AR gene causes complete AIS but no neurological dysfunction, which means that the absence of AR function does not cause motor neuron degeneration. Subjects with Kennedy’s disease also have adequate AR function to allow normal male differentiation, indicating that the mutation does not have a major effect on the “classical” function of the AR. These findings suggest that the Kennedy’s disease mutation must have a dominant, ‘gain of function’ effect, so that the presence of the mutant AR is toxic for motor neurons. All other trinucleotide repeat disorders are inherited in an autosomal dominant fashion, with affected individuals having one normal and one expanded allele. This suggests a common gain of function mechanism causing dysfunction in each of these conditions. However in Kennedy’s disease, heterozygote females, who carry one normal and one mutant copy of the AR gene, are unaffected. If the mutant AR has a toxic effect on motor neurons it would be expected that a dominant effect would be seen in heterozygote females. One study on heterozygote females found that four out of eight individuals had high amplitude motor unit potentials on EMG, and one muscle biopsy showed mild type 2 preponderance and very mild fibre-type grouping, suggesting that sub-clinical manifestations of the disease may occur (Sobue et al., 1993). The X chromosome carrying the mutant AR gene may be preferentially inactivated in heterozygote females so that the majority of motor neurons express normal AR. Alternatively, if the neuronotoxic effect of the mutation is caused by a specific abnormality of AR function, it may only occur when the AR is activated by ligand. Therefore the lower levels of circulating androgens in females compared to males may protect them against the full manifestation of the mutation’s effect. The AR gene mutation may cause motor neuron degeneration either by the abnormality of a specific AR function in motor neurons, or through a non-specific effect caused by the presence of a Gln-rich protein in the neurons. The Kennedy’s disease mutation may alter the interaction of the AR with other transcription factors in motor neurons, leading to a toxic effect. The amino-terminal domain of the AR is likely to interact with a number of other transcription factors (Kupfer et al., 19931, and disruption of this

154

H.E. MucLeun

ef al./Journal

of the

interaction would explain the reduced transactivation of the mutant AR in vitro observed by Mhatre et al. (1993). This could cause either the progressive down-regulation of cell survival factors or up-regulation of apoptotic factors, leading to neuronal death. The second theory, that Kennedy’s disease occurs because of a non-specific, toxic effect on motor neurons, is based on the fact that all trinucleotide repeat diseases in which the CAG repeat encodes Glns cause degeneration of neurons. In most of the trinucleotide repeat diseases, including Kennedy’s disease, the gene products are expressed in a wide range of tissues, but the neurons appear particularly vulnerable to the pathologic expression of the disease. It has been postulated that in these diseases, abnormal interactions between glutaminases and the mutant Gln-containing proteins may give rise to the pathological phenotype (Green, 1993). The toxic effect found in neurons may occur because these cells are less able than other cell types to degrade aggregates produced from cross-linking of Gln-rich substrates. Alternatively, proteins with increased numbers of Glns may have a high affinity for other Gln-containing proteins, with aggregation of the proteins forming insoluble precipitates. Another possibility is that in neurons the proteins with expanded Gln-rich domains may bind with higher affinity to other neuronalspecific transcription factors. Transcription factors normally required for the expression of other neuronal genes may be preferentially ‘mopped up’ by the Gln-rich proteins, leading to a general decrease in expression of other genes and eventually to cell death.

10. Transgenic mouse Kennedy’s disease will be of particular value for the study of trinucleotide repeat diseases, because unlike most of the other diseases the normal function of the AR in many tissues is known. In order to investigate how the Kennedy’s mutant AR functions in neural tissue, transgenie animal models for Kennedy’s disease will need to be constructed. Transgenic mouse lines over-expressing a human AR cDNA containing 45 repeats have been constructed, however these mice failed to show a neurological phenotype (Bingham et al., 1995). The abnormal genes were under the control of either the metallothionein or neuron specific enolase promoters, and inserted randomly into the genome. However expression of the mutant CDNAs in brain and spinal cord was very low. This fact, combined with the fact that most mice also had normal endogenous AR expression, may explain why these mice did not develop any neurological abnormalities. In order to study the mutant gene with the correct temporal and tissue-specific expression, homologous recombination would be required to replace the endogenous mouse AR gene with a gene containing an expanded CAG repeat in exon A. Until such mice are constructed the study of the

Neurological

Sciences 135 (1996)

149-157

AR in neural tissue will be limited to post-mortem human samples, in which neuronal degeneration has usually occurred many years before.

11. Treatment Until the pathogenic mechanism of AR dysfunction is known, the most appropriate treatment for Kennedy’s disease will remain uncertain. It is tempting to consider that because the AR in Kennedy’s disease shows impaired function, then the best treatment would be to give exogenous androgens to men with Kennedy’s disease in order to increase receptor activity. However it is possible that this treatment could in fact exacerbate the condition. If neuronal degeneration occurs because of a specific AR gain of function activity, then it is possible that treatment with androgen could cause further activation of the receptor and thus increase the production of neuronotoxic product. Similarly, any treatment that would increase the amount of AR protein in motor neurons could also be detrimental, whether the AR has a specific or non-specific deleterious effect. In general, androgen tends to down-regulate the expression of AR mRNA in prostate and brain, while castration increases gene expression (Quarmby et al., 1990; Wolf et al., 1993; Burgess and Handa, 1993). However one study has shown that treatment with androgenic-anabolic steroids increases AR protein in brain (Menard and Harlan, 1993). Treatment with anti-androgen might block any specific activation of the mutant AR but it would also increase AR levels, and may therefore also be harmful. Until more is known about the effect of the Kennedy’s disease mutation, the potential risks in using either androgens or anti-androgens as therapeutic agents must be carefully considered.

12. Summary Kennedy’s disease is one of a group of diseases caused by trinucleotide repeat expansions, in this case with the AR gene being affected. The major problem still to be answered in relation to Kennedy’s disease is how does the mutation, which appears to have little effect on classical AR function, cause the death of motor neurons. Until the normal role of the AR in the nervous system is more fully understood, including the knowledge of genes regulated by the AR in neural cells, this question will remain unanswered. However AR function is quite well characterized, particularly in comparison to the gene products of the other repeat diseases. This fact, combined with the research currently being undertaken to identify neurally-expressed androgen responsive genes, makes the study of Kennedy’s disease very valuable. Insights gained into the mechanism of pathogenesis in Kennedy’s disease could well be applicable to a wider range of neurodegenerative disorders.

H.E. MacLecrn

rr ul./Journd

of’thr

Acknowledgements This work was supported by grants from the National Health and Medical Research Council of Australia, and the Royal Children’s Hospital Research Foundation. HEM holds an NH&MRC Biomedical Postgraduate Scholarship. References Amato, A.A., T.W. Prior, R.J. Barohn, P. Snyder, A. Papp and J.R. Mendell (1993) Kennedy’s disease: a clinicopathologic correlation with mutations in the androgen receptor gene. Neurology, 43: 791794. Arbizu, T., J. Santamaria, J.M. Gomez, A. Quilez and J.P. Serra (1983) A family with adult spinal and bulbar muscular atrophy, X-linked inheritance and associated testicular failure. J. Neural. Sci., 59: 371-382. Barkhaus, P.E., W.R. Kennedy, L.Z. Stem and R.B. Harrington (1982) Hereditary proximal spinal and bulbar motor neuron disease of late onset. Arch. Neurol., 39: 112-l 16. Biancalana, V., F. Serville, J. Pommier, J. Julien, A. Hanauer and J.L. Mandel (1992) Moderate instability of the trinucleotide repeat in spino bulbar muscular atrophy. Hum. Molec. Genet., l(4): 255-258. Bingham, P.M., M.O. Scott, S. Wang, M.J. McPhaul, E.M. Wilson, J.Y. Garbem, D.E. Merry and K.H. Fischbeck (1995) Stability of an expanded trinucleotide repeat in the androgen receptor gene in transgenie mice. Nature Genet., 9: 191-196. Brinkmann, A.O., P.W. Faber, H.C.J. van Rooij, G.G.J.M. Kuiper, C. Ris, P. Klaassen, J.A.G.M. van der Korput, M.M. Voorhorst, J.H. van Laar, E. Mulder and J. Trapman (1989) The human androgen receptor: Domain structure, genomic organization and regulation of expression. J. Steroid Biochem., 34(1-6): 307-310. Burgess, L.H. and R.J. Handa (1993) Hormonal regulation of androgen receptor mRNA in the brain and anterior pituitary gland of the male rat. Mol. Brain. Res., 19: 31-38. Burke, J.R., MS. Wingfield, K.E. Lewis, A.D. Roses, J.E. Lee, C. Hulette, M.A. Pericak-Vance and J.M. Vance (I 994) The Haw River syndrome: Dentatorubropallidoluysian atrophy (DRPLA) in an African-American family. Nature Genet., 7: 521-524. Buxton, J., P. Shelboume, J. Davies, C. Jones, T. Van Tongeren, C. Aslanidis, P. de Jong, G. Jansen, M. Anvret, B. Riley, R. Williamson and K. Johnson (1992) Detection of an unstable fragment of DNA specific to individuals with myotonic dystrophy. Nature, 355: 547548. Cain, M.P., S.A. Kramer, D.J. Tindall and D.A. Husmann (1994) Expression of androgen receptor protein within the lumbar spinal cord during ontologic development and following antiandrogen induced cryptorchidism. J. Ural., 152: 766-769. Chung, M.Y., L.P.W. Ranum, L.A. Duvick, A. Servadio, H.Y. Zoghbi and H.T. Orr (1993) Evidence for a mechanism predisposing to intergenerational CAG repeat instability in spinocerebellar ataxia type I. Nat. Genet., 5: 254-258. Clancy, A.N.. R.W. Bonsall and R.P. Michael (1992) Immunohistochemical labeling of androgen receptors in the brain of rat and monkey. Life Sciences, 50: 409-417. Courey, A.J. and R. Tjian (1988) Analysis of Spl in vivo reveals multiple transcriptional domains, including a novel glutamine-rich activation motif. Cell, 55: 887-898. Craufurd, D. and A. Dodge (1993) Mutation size and age at onset in Huntington’s disease. J. Med. Genet., 30: 1008- 101 I. Danek, A., T.N. Witt, K. Mann, H.U. Schweikert, G. Romalo, A.R. La Spada and K.H. Fischbeck (1994) Decrease in androgen binding and effect of androgen treatment in a case of X-linked bulbospinal neuronopathy. Clin. Invest., 72: 892-897.

Neurological

Sciences

135 (1996)

149-157

155

Doyu, M., G. Sobue, E. Mukai. T. Kachi, T. Yasuda, T. Mitsuma and A. Takahashi (1992) Severity of X-linked recessive bulbospinal neuronopathy correlates with size of the tandem CAG repeat in androgen receptor gene. Ann. Neurol., 32(5): 707-710. Fischbeck, K.H., D. Souders and A. La Spada (1991) A candidate gene for X-linked spinal muscular atrophy. Adv. Neural., 56: 209-213. Gerber, H.P., K. Seipel, 0. Geogiev, M. Hiifferer, M. Hug, S. Rusconi and W. Schaffner (1994) Transcriptional activation modulated by homopolymeric glutamine and proline stretches. Science, 263: 80881 I. Goldstein, L.A. and D.R. Sengelaub (1992) Timing and duration of dihydrotestosterone treatment affect the development of motoneuron number and morphology in a sexually dimorphic rat spinal nucleus. J. Comp. Neural., 326: l47- 157. Green, H. (1993) Human genetic diseases due to codon reiteration: relationship to an evolutionary mechanism. Cell, 74: 955-956. Harding, A.E., P.K. Thomas, M. Baraitser, P.G. Bradbury, J.A. MorganHughes and J.R. Ponsford (1982) X-linked recessive bulbospinal neuronopathy: a report of ten cases. J. Neural. Neurosurg. Psychiatry, 45: 1012-1019. Huntington’s Disease Collaborative Research Group (1993) A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell, 72: 971-983. Igarashi, S., Y. Tanno, 0. Gnodera, M. Yamazaki, S. Sato, A. Ishikawa, N. Miyatani, M. Nagashima, Y. Ishikawa, K. Sahashi, T. Ibi, T. Miyatake and S. Tsuji (1992) Strong correlation between the number of CAG repeats in androgen receptor genes and the clinical onset of features of spinal and bulbar muscular atrophy. Neurology, 42: 23002302. Isaacs, J.T., Y. Furuya and R. Berges (1994) The role of androgen in the regulation of programmed cell death/apoptosis in normal and malignant prostatic tissue. Seminars Cancer Biol., 5: 391-400. Jung-Testas, I., M. Renoir, H. Bugnard, G.L. Greene and E.E. Baulieu (I 992) Demonstration of steroid hormone receptors and steroid action in primary cultures of rat glial cells. J. Steroid B&hem. Mol. Biol., 41(3-E):

621-631.

Kao, CC., P.M. Lieberman, M.C. Schmidt, Q. Zhou, R. Pei and A.J. Berk (1990) Cloning of a transcriptionally active human TAT binding factor. Science, 248: l646- 1650. Kawaguchi, Y., T. Okamoto, M. Taniwaka, M. Aizawa, M. Inoue, S. Katayama, H. Kawakami, S. Nakamura, M. Nishimura, I. Akiguchi, J. Kimura, S. Narumiya. and A. Kakizuka (1994) CAG expansions in a novel gene for Machado-Joseph disease at chromosome l4q32. I. Nature Genet., 8: 221-228. Kennedy, W.R., M. Alter and J.H. Sung (1968) Progressive proximal spinal and bulbar muscular atrophy of late onset. Neurology, 18: 67 l-680. Knight, S.J.L., A.V. Flannery, M.C. Hirst, L. Campbell, Z. Christodoulou, S.R. Phelps, J. Pointon, H.R. Middleton-Price, A. Bamicoat, M.E. Pembrey, J. Holland, B.A. Oostra, M. Bobrow and K.E. Davies (1993) Trinucleotide repeat amplifcation and hypermethylation of a CpG island in FRAXE mental retardation. Cell, 74(l): I27- 134. Koide, R., T. Ikeuchi, 0. Onodera, H. Tanaka, S. Igarashi, K. Endo, H. Takahashi, R. Kondo, A. Ishikawa, T. Hayashi, M. Saito, A. Tomoda, T. Mike, H. Naito, F. Ikuta and S. Tsuji (1994) Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian atrophy (DRPLA). Nature Genet., 6: 9- 13. Kremer, E.J., M. Pritchard, M. Lynch, S. Yu, K. Holman, E. Baker, S.T. Warren, D. Schlessinger, G.R. Sutherland and R.I. Richards (1991) Mapping of DNA instability at the Fragile X to a trinucleotide repeat sequence p(CCG)n. Science, 252: I71 l-1714. Kupfer, S.R., K.B. Marschke, E.M. Wilson and F.S. French (1993) Receptor accessory factor enhances specific DNA binding of androgen and glucocorticoid receptors. J. Biol. Chem., 268(23): l751917527. Kurz, E.M., D.R. Sengelaub and A.P. Arnold (1986) Androgens regulate

156

H.E. MacLeun et ul./Journul

of the Neurologicd

the dendritic length of mammalian motoneurons in adulthood. Science, 232: 395-398. La Spada, A.R., E.M. Wilson, D.B. Lubahn, A.E. Harding and K.H. Fischbeck (1991) Androgen receptor gene mutations in X-linked spinal and bulbar muscular atrophy. Nature, 352: 77-79. La Spada, A.R., D.B. Roling, A.E. Harding, C.L. Warner, R. Spiegel, I. Hausmanowa-Petrusewicz, W.C. Yee and K.H. Fischbeck (1992) Meiotic stability and genotype-phenotype correlation of the trinucleotide repeat in X-linked spinal and bulbar muscular atrophy. Nature Genet., 2: 301-304. Lubahn, D.B., D.R. Joseph, M. Sar, J. Tan, H.N. Higgs, R.E. Larson, F.S. French and E.M. Wilson (1988) The human androgen receptor: complementary deoxyribonucleic acid cloning, sequence analysis and gene expression in prostate. Mol. Endocrinol., 2(12): 1265-I 275. MacLean, H.E., W.T. Choi, Cl. Rekaris, G.L. Wame and J.D. Zajac (1995) Abnormal androgen receptor binding affinity in subjects with Kennedy’s disease (spinal and bulbar muscular atrophy). J. Clin. Endocrinol. Metab., 80(2): 508-5 16. Mahadevan, M., C. Tsilfidis, L. Sabourin, G. Shutler, C. Amemiya, G. Jansen, C. Neville, M. Narang, J. Barcelo, K. O’Hoy, S. Leblond, J. Earle-Macdonald. P.J. de Jon& B. Wieringa and R.G. Komeluk (1992) Myotonic dystrophy mutation: An unstable CTG repeat in the 3’ untranslated region of the gene. Science, 255: 1253- 1258. Matsuura, T., T. Demura, Y. Aimoto, T. Mizuni, F. Moriwaka and K. Tashiro (1992) Androgen receptor abnormality in X-linked spinal and bulbar muscular atrophy. Neurology, 42: 1724-1726. Matsuura, T., A. Ogata, T. Demura, F. Moriwaka, K. Tashiro, T. Koyanagi and K. Nagashima (1993) Identification of androgen receptor in the rat spinal motoneurons. Immunohistochemical and immunoblotting analyses with monoclonal antibody. Neurosci. Lett., 158: 5-8. McPhaul, M.J., M. Marcelli, S. Zoppi, J.E. Griffin and J.D. Wilson (1993) Genetic basis of endocrine disease 4: The spectrum of mutations in the androgen receptor gene that causes androgen resistance. J. Clin. Endocrinol. Metab., 7til): 17-23. Menard, C.S. and R.E. Harlan (1993) Up-regulation of androgen receptor immunoreactivity in the rat brain by androgenic-anabolic steroids. Brain Res., 622: 226-236. Mhatre. A.N., M.A. Trifiro, M. Kaufman, P. Kazemi-Esfarjani, D. Figlewicz, G. Rouleau and L. Pinsky (1993) Reduced transcriptional regulatory competence of the androgen receptor in X-linked spinal and bulbar muscular atrophy. Nat. Genet., 5: 184- 188. Nagafuchi, S., H. Yanagisawa, K. Sato, T. Shirayama, E. Ohsaki, M. Bundo, T. Takeda, K. Tadokoro, I. Kondo, N. Murayama, Y. Tanaka, H. Kikushima, K. Umino, H. Kurosawa, T. Furukawa, K. Nihei, T. Inoue, A. Sano, 0. Komure, M. Takahashi, T. Yoshizawa, I. Kanazawa. and M. Yamada (1994) Dentatorubral and pallidoluysian atrophy expansion of an unstable CAG trinucleotide on chromosome 12~. Nature Genet., 6: 14-18. Nagashima, T., K. Seko, K. Hirose, T. Mannen, S. Yoshimura, R. Arima, K. Nagashima and Y. Morimatsu (1988) Familial bulbo-spinal muscular atrophy associated with testicular atrophy and sensory neuropathy (Kennedy-Alter-Sung syndrome). J. Neurol. Sci., 87: 141- 152. Nakamura, M., S. Mita, T. Murakami, M. Uchino, S. Watanabe, M. Tokunaga, T. Kumamoto and M. Ando (1994) Exonic trinucleotide repeats and expression of androgen receptor gene in spinal cord from X-linked spinal and bulbar muscular atrophy. J. Neural. Sci., 122: 74-79. Nancarrow, J.K., E. Kremer, K. Holman, H. Eyre, N.A. Doggett, D. Le Paslier, D.F. Callen, G.R. Sutherland and R.I. Richards (1994) Implications of FRAl6A structure for the mechanism of chromosomal fragile site genesis. Science, 264: 1938-1941. Nankin, H.R. and J.H. Calkins (1986) Decreased bioavailable testosterone in aging normal and impotent men. J. Clin. Endocrinol. Metab., 63: 1418-1420. Nordeen, E.J., K.W. Nordeen, D.R. Sengelaub and A.P. Arnold (1985)

Sciences 135 (1996) 149-157

Androgens prevent normally occurring cell death in a sexually dimorphic spinal nucleus. Science, 229: 671-673. Orr, H.T., M.Y. Chung, S. Bank T.J. Kwiatkowski Jr., A Servadio, A.L. Beaudet, A.E. McCall, L.A. Duvick, L.P.W. Ranum and H.Y. Zoghbi (1993) Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. Nature Genet., 4: 221-225. Parrish, J.E., B.A. Oostra, A.J.M.H. Vekerlc, C.S. Richards, J. Reynolds, A.S. Spikes, L.G. Shaffer and D.L. Nelson (1994) Isolation of a GCC repeat showing expansion in FRAXF, a fragile site distal to FRAXA and FRAXE. Nature Genet., 8: 229-235. Peru&, M.F., T. Johnson, M. Suzuki and J.T. Finch (1994) Glutamine repeats as polar zippers: Their possible role in inherited neurodegenerative diseases. Proc. Nat]. Acad. Sci. USA, 91: 5355-5358. Quarmby, VE, W.G. Yarbrough, D.B. Lubahn, F.S. French and E.M. Wilson (1990) Autologous down-regulation of androgen receptor messenger ribonucleic acid. Mol. Endocrinol., 4(l): 22-28. Quigley, C.A., K.J. Friedman, A. Johnson, R.G. Lafreniere, L.M. Silverman, D.B. Lubahn, T.R. Brown, E.M. Wilson, H.F. Willard and F.S. French (1992) Complete deletion of the androgen receptor gene: Definition of the null phenotype of the androgen insensitivity syndrome and determination of carrier status. J. Clin. Endocrinol. Metab., 74(4): 927-933. Rundlett, S.E., X.P. Wu and R.L. Miesfeld (1990) Functional characterizations of the androgen receptor confirm that the molecular basis of androgen action is transcriptional regulation. Mol. Endocrinol., 45): 708-7 14. Saiki, R.F., D.H. Gelfand, S. Stoffel, S.J. Scharf, R. Higuchi, G.T. Horn, K.B. Mullis and H.A. Erlich (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science, 239: 487-491. Sar, M. and W.E. Stumpf (1977) Androgen concentration in motor neurons of cranial nerves and spinal cord. Science, 197: 77-78. Sengelaub, D.R., C.L Jordan, E.M. Kurz and A.P. Arnold (1989) Hormonal control of neuron number in sexually dimorphic spinal nuclei of the rat: II. Development of the spinal nucleus of the bulbocavernosus in androgen-insensitive (Tfm) rats. J. Comp. Neural., 280(4): 630-636. Sheridan, P.J. and M. Fujimoto (1988) The leptomenix in normal baboon brain contain receptors for dihydrotestosterone but not estradiol. Anatom. Rec., 220(4): 442-445. Simental, J.A., M. Sar, M.V. Lane, F.S. French and E.M. Wilson (1991) Transcriptional activation and nuclear targeting signals of the human androgen receptor. J. Biol. Chem., 266(l): 510-518. Simerly, R.B., C. Chang, M. Muramatsu and L.W. Swanson (1990) Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: An in situ hybridization study. J. Comp. Neural., 294: 76-95. Sobue, G., M. Doyu, T. Kachi, T. Yasuda, E. Mukai, T. Kumagai and T. Mitsuma (1993) Subclinical phenotypic expressions in heterozygous females of X-linked recessive bulbospinal neuronopathy. J. Neurol. Sci., 117: 74-78. Stefanis, C., T. Papapetropoulos, S. Scarpalezos, G. Lygidakis and C.P. Panayiotopoulos (1975) X-linked spinal and bulbar muscular atrophy of late onset. J. Neurol. Sci., 24: 493-503. Vaisanen, M.L., M. Kahlconen and J. Leisti (1994) Diagnosis of fragile X syndrome by direct mutation analysis. Hum. Genet., 93(2): 143-147. Warner, CL., J.E. Griffin, J.D. Wilson, L.D. Jacobs, K.R. Murray, K.H. Fischbeck, D. Dickoff and R.C. Griggs (1992) X-linked spinomuscular atrophy: A kindred with associated abnormal androgen receptor binding. Neurology, 42: 2 18 l-2 184. Warner, C.L., S. Servidei, D.J. Lange, E. Miller, R.E Lovelace and L.P. Rowland (1990) X-linked spinal muscular atrophy (Kennedy’s syndrome): a kindred with hypobetalipoproteinemia. Arch. Neurol., 47: 1117-1120. Willems, P.J. (1994) Dynamic mutations hit double figures. Nature Genet., 8: 213-215.

H.E. MacLeun et uL/Journol

of’the Neurological

Wolf, D.A., T. Herzinger, H. Hermeking, D. Blaschke and W. Hdrz (1993) Transcriptional and posttranscriptional regulation of human androgen receptor expression by androgen. Mol. Endocrinol., 7(7): 924-936. Yamamoto, Y., H. Kawai, K. Nakahara, M. Osame, Y. Nakatsuji, T. Kishimoto and S. Sakoda (1992) A novel primer extension method to

Sciences 135 (1996) 149-157

157

detect the number of CAG repeats in the androgen receptor gene in families with X-linked spinal and bulbar muscular atrophy. B&hem. Biophys. Res. Commun., 182(2): 507-513. Yu, W.H.A. (I 988) Sex differences in neuronal loss induced by axotomy in the rat brain stem motor nuclei. Exp. Neurol., 102: 230-235.