| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Very large (CAG)n DNA repeat expansions in the sperm of two spinocerebellar ataxia type 7 males
Introduction
Results
Discussion
Materials And Methods
SP-PCR
Computer analyses
Acknowledgements
References
Very large (CAG)n DNA repeat expansions in the sperm of two spinocerebellar ataxia type 7 males
Received July 26, 1999; Revised and Accepted September 22, 1999
Genetic anticipation, i.e. increasing disease severity and decreasing age of onset from one generation to the next, is observed in a number of diseases, including myotonic dystrophy type 1, Huntington's disease and several of the spinocerebellar ataxias. All of these disorders are associated with the expansion of a trinucleotide repeat and array length is positively correlated with disease severity and inversely correlated with the age of onset. The expanded repeat is highly unstable and continues to expand from one generation to the next, providing a molecular explanation for anticipation. Spinocerebellar ataxia type 7 (SCA7) is one of the latest additions to the list of triplet repeat diseases and is distinct from the other SCAs in that it is accompanied by retinal degeneration. Pedigree analyses have previously revealed that the SCA7 repeat is highly unstable and liable to expand, in particular when transmitted by a male. Surprisingly, though, an under-representation of male transmission has also been reported. We now demonstrate directly by single molecule analyses that the expanded repeat is extraordinarily unstable in the male germline and biased toward massive increases. Nearly all of the mutant sperm of two SCA7 males contain alleles that are so large that most of the affected offspring would at best have a severe infantile form of the disease. Indeed, the gross under-representation of such very large expanded alleles in patients suggests that a significant proportion of such alleles might be associated with embryonic lethality or dysfunctional sperm.
INTRODUCTION
Highly unstable triplet repeats are now known to account for the unusual genetics of a growing number of inherited human disorders, including fragile X syndrome, myotonic dystrophy type 1 (DM1), Huntington's disease (HD), spinal and bulbar muscular atrophy (SBMA), dentatorubral pallidoluysian atrophy (DRPLA) and several of the spinocerebellar ataxias (1). A notable feature of many of these diseases is the process of anticipation whereby the disease severity increases and the age of onset decreases from one generation to the next. Anticipation was first well documented in DM1 (2), where it is often very severe; the disease progressing from the mild late onset form, through the classic adult onset form to the most severe congenital form in as little as three generations (3). Anticipation has now been explained at the molecular level by the positive correlation of disease severity and the inverse correlation of the age of onset with the inherited repeat length and the propensity of the repeat to increase in length when transmitted through the germline (4).
Spinocerebellar ataxia type 7 (SCA7) is an inherited neurodegenerative disorder which, in addition to ataxia of both gait and limbs, is accompanied by retinal degeneration (5). The precise clinical features observed in SCA7 are extremely variable, as is the age of onset and duration of the disease, although it is genetically quite homogeneous with all but one of the families mapped having been linked to a locus on chromosome 3p (6-11). The age of onset within SCA7 families is, however, not random, but shows a marked decrease from one generation to the next, with a mean rate of anticipation of ~24 years/generation (12). The phenotype and high level of anticipation suggested that SCA7 might well belong to the growing list of inherited human diseases associated with the expansion of a trinucleotide CAG repeat in the coding region of the gene. This hypothesis was supported by data that identified CAG/CTG repeat tract expansions using the anonymous repeat expansion detection method (13) and a novel protein with an expanded polyglutamine array (14) in patients. More recently, the hypothesis was confirmed by cloning the expanded (CAG)n repeat and identification of the SCA7 gene, termed ataxin-7 (15-17). Ataxin-7 is an 892 amino acid protein with no known function. The repeat tract is only moderately polymorphic in the normal population (hetero-zygosity ~35%) with alleles in the range 4-19 repeats, but with ~75% of alleles containing 10 repeats. Patients have expansions from 38 repeats, whereas individuals with intermediate sized repeats probably have a low risk of disease themselves, but are at risk of transmitting pathogenic expansions to their offspring (18). As with the other polyglutamine expansion disorders, there is a positive correlation of repeat length with disease severity and an inverse correlation with age of onset. Most affected individuals have alleles in the range 38-60 and individuals with alleles >80 repeats in length are rare. More recently, two individuals have been described with what has been termed infantile SCA7 (19,20). In this form of the disease onset occurs at or around birth and, in addition to retinal and neuronal degeneration, is accompanied by patent ductus arteriosus associated with congestive heart failure, severe hypotonia and a failure to thrive. Both affected infants died within 7 months of birth and were subsequently shown to have the largest expansions yet described for the polyglutamine repeat diseases, with alleles in the range ~250-300 repeats (19,20).
Once in the expanded range, the SCA7 repeat becomes very unstable and is prone to expansion from one generation to the next, accounting for the high level of anticipation observed. Indeed, the SCA7 repeat appears to be the most expansion-prone of the polyglutamine expansion disorders. The repeat is particularly unstable on transmission from males with a mean increase of 22 repeats/transmission, compared with only six repeats in females (15-17,19-22), consistent with the greater level of anticipation observed in paternal transmission. Given that SCA7 is an autosomal dominant disease that affects both sexes equally (23), we would expect transmitting parents to be equally divided between males and females. Indeed, it might even be expected that an excess of male transmission would be observed, since they transmit the largest expansions and therefore would parent more clinically identifiable offspring. It is thus surprising that a recent study of several SCA7 families revealed that most of the transmitting parents of affected individuals were females (22). This excess was observed not only in the families studied, but also in an analysis of previously published SCA7 families. It was proposed that a possible reason for this apparent under-representation of paternal transmissions could be increased neonatal or fetal wastage associated with the transmission of very large expansions in the male germline (22).
RESULTS
To determine whether the SCA7 repeat is particularly unstable and prone to large expansions in the male germline we have obtained blood, muscle and sperm DNA from a 33-year-old SCA7 male [individual II.8 in an Ecuadorian SCA family (ESCA) (19)]. This man had an expanded allele of 53 repeats associated with onset of visual problems at age 15 years and ataxia at age 20. To analyse the level of repeat variability within each sample we have used sensitive single molecule-based small pool PCR (SP-PCR) procedures (24,25). Analysis of the blood DNA of this man revealed a significant level of somatic instability for the expanded allele with a sharp lower boundary and a clear bias toward expansion (Fig. 1A), similar to the pattern of somatic repeat variability observed in DM1 patients (25-27). Small gains of up to ~10 repeats were frequent, whereas expanded alleles up to 130 repeats in length were seen in a small subset of cells. A similar level of repeat length variation was also observed in muscle (data not shown). In contrast, repeat variability in this man's sperm was extreme, with a massive bias toward large expansions (Fig. 1A). Individual sizing of 379 expanded sperm alleles amplified from single molecules revealed an essentially 100% mutation rate, 99% biased toward further expansion (Fig. 2A). The mean allele size was 189 repeats, representing a mean gain of 136 repeats relative to the 53 repeat progenitor allele.
Figure 1. SP-PCR analysis of CAG repeat variation at the SCA7 locus. (A) SP-PCR analysis of the blood and sperm DNA of individual ESCA II.8. (B) SP-PCR analysis of the buccal cell and sperm DNA of individual ESCA II.5. Five representative SP-PCRs for two DNA dilutions each containing the indicated cellular equivalents of DNA are shown. The scale shows the position of the molecular weight markers converted into the number of repeats.
Figure 2. Repeat length variation in SCA7. Repeat length variation in the sperm of individual ESCA II.8 (A) and ESCA II.5 (B). The graphs demonstrate the percentage of expanded sperm alleles falling into the mutant size groups shown as determined by SP-PCR (n = 379 for ESCA II.8 and n = 199 for ESCA II.5). (C) Repeat length versus age of onset correlations in SCA7. The relationship between allele length (number of repeats) and age of onset of symptoms is shown. These data (n = 161) were collated from Benton et al. (19), Johansson et al. (20), David et al. (21) and Gouw et al. (22). For allele lengths of [le]80 repeats, regression analysis revealed a coefficient of correlation of -0.74 (n = 153).
To ascertain whether such gross germline instability might be typical, we purified buccal cell and sperm DNA from the only other SCA7 male donor available to us [individual II.5 in the ESCA family (19)]. This man, the older brother of the first SCA7 male doner, had a shorter expansion of only 46 repeats and a later age of onset, ~26 years. The buccal cells showed a high level of variability clearly biased toward expansion and greater in extent than in the blood or muscle of his brother (Fig. 1B). Sizing of 199 individually amplified expanded sperm alleles revealed that variation was again extensive (Figs 1B and 2B), although the mean length change was significantly smaller than in his brother (P < 0.001), probably due to the shorter progenitor allele length. Nonetheless, the mutation rate again approached 100%, with a 99% expansion bias and a mean allele length of 150 repeats, representing a mean gain of 104 repeats/transmission. Interestingly, the sperm distribution appeared to be bimodal (Figs 1B and 2B), with peak I centred around ~73 repeats and including ~20% of the expanded sperm and a second larger peak II centred around ~170 repeats and incorporating ~80% of expanded sperm. To confirm the bimodal nature of the distribution, more SP-PCRs were performed with higher concentrations of DNA, ~200 sperm/reaction (Fig. 3). Although at such high DNA concentrations individual sperm alleles could no longer be resolved, the bimodal nature of the expanded allele distribution was clearly observed. SP-PCR using very low amounts of sperm DNA (<1 molecule/reaction) revealed similar numbers of reactions to contain expanded and non-expanded alleles; 53:64 and 24:18 for ESCA II.5 and ESCA II.8, respectively. These data indicate that in neither male was there a significant deviation from the expected Mendelian ratio of 1:1 normal versus expanded repeat-containing sperm.
Figure 3. Bimodality in repeat length variation in the sperm of individual ESCA II.5. Five representative SP-PCRs each containing ~200 sperm equivalents of DNA are shown. The scale shows the position of the molecular weight markers converted into the number of repeats. The two peaks of variability as identified in the single molecule analyses (Fig. 2B) are clearly reproduced.
In order to predict the likely phenotypic consequences to the next generation if these men were to father children, we compared the allele size distributions in the sperm with the data relating allele length and age of onset of symptoms. Collation of the published SCA7 genotype-phenotype data (19-22) reveals that the vast majority (~95%) of patients have allele lengths in the range 38-80 repeats (Fig. 2C). In this range, repeat length shows a strong inverse correlation with age of onset (coefficient of correlation -0.74), accounting for ~54% of the variation in age of onset. Alleles >80 repeats in length are always associated with juvenile onset of symptoms, whereas alleles >150 repeats in length have only been observed in two very severely affected offspring with the infantile form of the disease. In stark contrast to the range of alleles observed in SCA7 patients, 97% of the expanded alleles in the sperm of ESCA II.8 were >80 repeats in length, a size range that includes <5% of the patient population. Moreover, 84% of the sperm of ESCA II.8 were >150 repeats in length, a range that includes <2% of the patient population. Similarly, the majority of mutant sperm from ESCA II.5 contained alleles rarely observed in the patient population, with 87% of alleles >80 repeats and 53% >150 repeats. We predict that virtually all of the large expansions >150 repeats present in these men's sperm would at best give rise to a juvenile form of the disease in the next generation and that the majority would give rise to the severe infantile form with onset in the first year of life. Indeed, the gross under-representation of alleles >150 repeats in length in the patient population coupled with the under-representation of male transmission in SCA7 families suggests that sperm carrying such large expansions may be dysfunctional or are embryonic lethals.
DISCUSSION
We have analysed the level of repeat variability in somatic and germline DNA at the SCA7 locus in two brothers with pathogenic CAG expansions. These analyses have revealed levels of male germline instability far in excess of, and biased toward much larger expansions than, those observed in any of the other polyglutamine repeat disorders to date, including those previously analysed by direct sperm analysis such as the HD (28), DRPLA (29), SBMA (30), SCA1 (31) and SCA3 (32) repeats. Indeed, considering the relatively small progenitor allele lengths in these two men, this locus may be even more unstable than the DM1 locus (25). Why the SCA7 locus should be so unstable remains unknown, but appears to be associated with the very high GC content of the flanking DNA and its location within a CpG island (33).
The allele lengths present in these men's sperm are much larger than commonly observed in the polyglutamine expansion disorders. The precise phenotypic consequences of such large alleles remains unknown. However, comparison of the patient genotype with the phenotype data suggests that a high proportion of sperm from such men would give rise to the recently identified infantile form of the disease. It seems likely that this form of the disease with onset of cardiac abnormalities at or around birth is more common than formerly realized in what has traditionally been considered to be a late onset neurodegenerative disorder. Nonetheless, the under-representation of male transmission in SCA7 families and the gross under-representation of such very large expanded alleles in the patient population suggest that a significant proportion of such alleles does not get transmitted to the next generation. Although both men's sperm counts and morphology appeared normal, indicating no major defects in the process of spermatogenesis, it remains possible that such very large expansions might be associated with dysfunctional sperm, i.e. sperm that are unable to fertilize an egg correctly. Alternatively, such large alleles might be associated with embryonic lethality. Although an increased frequency of spontaneous abortions has not been reported as a general feature in SCA7 families, examination of the literature revealed one family with a severely affected child and four early spontaneous abortions associated with an affected father (34), and other families with unexplained still-born offspring (35). The normal role of ataxin-7 is unknown; however, it is widely expressed in adult tissues and there is some evidence to suggest that it may be a transcription factor (22). The retinal degeneration observed in the adult form of the disease and the cardiac phenotype observed in the infantile cases indicate that the pathogenic effect of this expansion is not limited to neuronal tissues in the brain. Analysis of the dbEST database indicates that in addition to a variety of adult tissues, including colon, brain, testes and thymus, and various tumours, ataxin-7 ESTs have been detected in cDNA libraries constructed from human fetal (20 weeks) liver/spleen and placenta (8-9 weeks). Moreover, the murine homologue of ataxin-7 has been detected in mouse two-cell embryos and fetal heart (data not shown). It therefore seems likely that ataxin-7 is expressed widely during embryonic development and we speculate that expanded alleles may have a toxic effect in very early embryogenesis, perhaps before the stage at which overt pregnancy is apparent.
Given the recent demonstration of de novo expansion of a 28 repeat allele with no associated phenotype to a 47 repeat allele in the next generation (18) and our demonstration of gross expansion from 46 and 53 repeat alleles, we predict the existence of SCA7 families in which only one affected generation would exist (i.e. a normal individual transmitting a moderately expanded allele to a mildly affected individual who would be incapable of transmitting an expansion to live-born offspring due to germ cell dysfunction/embryonic lethality). The phenomenon of a mildly affected parent transmitting only embryonic lethal alleles would represent the most extreme form of genetic anticipation yet identified. Where the production of dysfunctional germ cells can be best accommodated within an anticipatory continuum is subject to discussion, but in molecular terms and evolutionary significance at least germ cell dysfunction represents a similar genetic end-point. The anticipatory scenario described above for SCA7 raises the possibility that large normal alleles at an even more unstable repeat locus could expand from within the normal range to embryonic lethals in a single step presenting a disease whose only phenotype might be an increased rate of spontaneous abortions in a `normal' parent.
Although we observed gross instability in the sperm of the two brothers whom we have analysed, pedigree analysis in other SCA7 families has revealed several examples of male germline transmission of expanded alleles showing only moderate length change mutations (20). Moreover, David et al. (21) detected only a relatively small proportion of sperm carrying large alleles (>80 repeats) in a bulk PCR analysis of germline variation in another SCA7 male with a 45 repeat progenitor allele. However, our own studies at other triplet repeat loci have suggested that the electrophoretic profiles generated by bulk PCR analysis with fluorescently labelled primers and detection on automated DNA sequencing type apparatus greatly underestimates the true level of variability within a sample, in particular the presence of large expanded alleles (M.T. Fortune, C. Vassilopoulos, M.I. Coolbaugh, M.J. Siciliano and D.G. Monckton, manuscript in preparation). Nonetheless, the pedigree data indicate that the SCA7 repeat is markedly more stable in some men. Bimodality of the degree of clinical anticipation has also been previously observed in paternal SCA7 transmission (23). These data suggest the existence of a polymorphism(s) within either cis- or trans-acting genetic modifiers of repeat stability. Putative intralocus cis-acting sequence modifiers of instability have previously been identified at other human tandem repeats (32,36) and it is possible that the variation in repeat stability observed between SCA7 men/families reflect flanking sequence differences. Alternatively, such variation might result from polymorphic variation, possibly even a simple dimorphism, in trans-acting genetic modifiers of repeat stability. Consistent with the existence of such a polymorphism(s) in human populations, similar allele length-independent variation in relative mutability has been observed through pedigree analyses in DM1 families (37,38) and in direct sperm analysis in DM1 (25; L. Martorell, D.G. Monckton, J. Gamez and M. Baiget, manuscript in preparation), HD (28) and DRPLA males (29).
In addition to the apparent bimodality observed between individuals, we also observed bimodality within the expanded sperm distribution of ESCA II.5. This bimodality suggests that the germline mutational pathway includes two different mechanisms, one involving small length changes and one involving large length changes. The small length change mutation events are not dissimilar from those observed in somatic tissues, whereas the large length change mutations appear to be largely germline specific. It is attractive to speculate that the small length change mutations arise in the germline during the premeiotic stages of spermatogenesis in a probably age-dependent manner (25,28) and that the large length change events take place in a single step during meiosis. The large length change effect may not operate at 100% efficiency for smaller alleles (e.g. ESCA II.5 in this study) or may not operate at all in some men. These hypotheses currently remain highly speculative, but detailed studies such as these and others (25,28-32) into the germline dynamics of repeat instability are starting to yield interesting insights into the complex mechanisms and pathways involved.
In summary, we have observed gross germline instability in two SCA7 males that gives rise to the presence, at high frequency, of very large expanded alleles in their sperm. These very large alleles would be predicted to result in the severe infantile form of the disease in the next generation or may even be associated with dysfunctional sperm or embryonic lethality. The levels of germline instability observed in this study are unprecedented in the polyglutamine repeat disorders so far characterized.
MATERIALS AND METHODS
SP-PCR
Sperm DNA was prepared from semen using a differential lysis procedure as previously described (24). SP-PCR at the SCA7 locus was performed as previously described (25) using PCR primers SCA7-A (5[prime]-GCGACTCTTTCCCCCTTTTTTTTG-3[prime]) and SCA7-BR (5[prime]-GGAAGCCTCAACCCACAGATTC-3[prime]) with the addition of 8% DMSO to the PCR reactions and an annealing temperature of 62°C. The SCA7-specific probe used was the amplification product of a PCR reaction using the internal nested primers SCA7-C (5[prime]-AGGAGCGGAAAGA-ATGTCGGAG-3[prime]) and SCA7-DR (5[prime]-ACGACTGTCCCAGCATCACTTC-3[prime]) from the DNA of a normal individual. Allele length distributions in sperm were derived by sizing the amplification products derived from SP-PCRs containing 1-10 haploid genome equivalents of DNA per reaction. No contaminating alleles were observed in any of >200 zero DNA controls.
Computer analyses
The SCA7 cDNA sequence [AJ000517 (14,15)] was used to search the dbEST database using the BLAST (39) facility at NCBI (http://www.ncbi.nlm.nih.gov/cgi-bin/BLAST/ ) for matching EST sequences. Statistical analyses were performed with Microsoft Excel.
ACKNOWLEDGEMENTS
This paper is dedicated to the memory of Claudia Benton. We are especially grateful to Claudia and Huda Zoghbi for the release, prior to publication, of data that greatly facilitated these studies. We should also like to thank the University of Glasgow Dynamic Mutation Group and Richard Wilson for helpful discussions on this work. We are likewise indebted to the SCA7 patients and families for their continued cooperation and help. M.L.C. was supported by a fellowship from the Ministerio de Educación y Cultura (Spain). D.G.M. is a Lister Institute Research Fellow. This work was also supported by awards to D.G.M. from the Muscular Dystrophy Association (USA) and the Medical Research Council (UK) and to T.A. from the Oxnard and National Ataxia Foundations.
REFERENCES
+To whom correspondence should be addressed. Tel: +44 141 3306213; Fax: +44 141 3306871; Email: dmonck{at}molgen.gla.ac.uk
§Present address: Departamento de Genética y Microbiología, Universidad de Murcia, Apartado 4021, 30071 Murcia, Spain
¶Present address: Department of Neurology, Oldchurch Hospital, Romford, Essex RM7 0BE, UK
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Gomes-Pereira and D. G. Monckton Chemically induced increases and decreases in the rate of expansion of a CAG{middle dot}CTG triplet repeat Nucleic Acids Res., May 20, 2004; 32(9): 2865 - 2872. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Martorell, J. Gamez, M. L. Cayuela, F. K. Gould, J. P. McAbney, T. Ashizawa, D. G. Monckton, and M. Baiget Germline mutational dynamics in myotonic dystrophy type 1 males: Allele length and age effects Neurology, January 27, 2004; 62(2): 269 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Savouret, C. Garcia-Cordier, J. Megret, H. te Riele, C. Junien, and G. Gourdon MSH2-Dependent Germinal CTG Repeat Expansions Are Produced Continuously in Spermatogonia from DM1 Transgenic Mice Mol. Cell. Biol., January 15, 2004; 24(2): 629 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Nenguke, M. I. Aladjem, J. F. Gusella, N. S. Wexler, The Venezuela HD Project, and N. Arnheim Candidate DNA replication initiation regions at human trinucleotide repeat disease loci Hum. Mol. Genet., May 1, 2003; 12(9): 1021 - 1028. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. T. Libby, D. G. Monckton, Y.-H. Fu, R. A. Martinez, J. P. McAbney, R. Lau, D. D. Einum, K. Nichol, C. B. Ware, L. J. Ptacek, et al. Genomic context drives SCA7 CAG repeat instability, while expressed SCA7 cDNAs are intergenerationally and somatically stable in transgenic mice Hum. Mol. Genet., January 1, 2003; 12(1): 41 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sharma, S. Bhatti, M. Gomez, R. M. Clark, C. Murray, T. Ashizawa, and S. I. Bidichandani The GAA triplet-repeat sequence in Friedreich ataxia shows a high level of somatic instability in vivo, with a significant predilection for large contractions Hum. Mol. Genet., September 1, 2002; 11(18): 2175 - 2187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Bacon, M. G. Dunlop, and S. M. Farrington Hypermutability at a poly(A/T) tract in the human germline Nucleic Acids Res., November 1, 2001; 29(21): 4405 - 4413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Rolfsmeier, M. J. Dixon, L. Pessoa-Brandão, R. Pelletier, J. J. Miret, and R. S. Lahue Cis-Elements Governing Trinucleotide Repeat Instability in Saccharomyces cerevisiae Genetics, April 1, 2001; 157(4): 1569 - 1579. [Abstract] [Full Text]
|
|
|
|
 |
|
 E.-K. Tan and T. Ashizawa Genetic Testing in Spinocerebellar Ataxias: Defining a Clinical Role Arch Neurol, February 1, 2001; 58(2): 191 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Sobrido and D. H. Geschwind Molecular Genetics and Inherited Ataxias: Redefining Phenotypes and Pathogenesis Neuroscientist, December 1, 2000; 6(6): 465 - 474. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||