вход по аккаунту


Analysis of the SCAI CAG repeat in a large number of families with dominant ataxia Clinical and molecular correlations.

код для вставкиСкачать
Analysis of the SCAl CAG Repeat in a
Large Number of Famhes with Dominant
Ataxia: Clinical and Molecular Correlations
0. Dubourg, MD,' A. Durr, MD," G. Cancel, MS," G . Stevanin, MS," H. Chneiweiss, M D , PhD,?
C. Penet, BS," Y . Agid, M D , PhD," and A. Brice, MD"
Autosomal dominantly inherited ataxias are a clinically and genetically heterogeneous group of neurodegenerative
disorders. The gene involved in one subtype, spinocerebellar ataxia 1 (SCA l), was first localized to chromosome 6p.
An unstable CAG repeat has been identified as the responsible mutation. In this study, 88 families with various types
of inherited acaxias and 16 individuals with sporadic cerebellar ataxia were investigated to determine the frequency
of this mutation, the behavior of the SCAl CAG repeat during transmission, and the clinical features specific to this
form of disease. Only 12 of the families carried the SCAl mutation; 10 of the 12 were of French origin. When
transmitted paternally. the repeat was more unstable and larger in size. Age at onset was inversely correlated with
the number of CAG repeats. Anticipation in age at onset of about 11 years was observed in offspring. Analysis of the
clinical features did not distinguish SCAl from other forms of dominantly inherited ataxias. I n the absence of distinguishing clinical characteristics, the diagnosis of SCAl in single affected patients or family members can only be
made by direct detection of the mutation, opening the way for presymptomatic testing.
Dubourg 0, Durr A, Cancel G, Stevanin G, Chneiweiss H. Penet C, Agid Y , Brice A .
Analysis of the SCAl CAG repeat in a large number of families with dominant ataxia:
clinical and molecular correlations. A n n Neurol 1995;37:176- 180
Autosomal dominant cerebellar ataxias ( ADCAs) are a
clinically and genetically heterogeneous group of neurological disorders, involving the degeneration of neurons in inferior olives, pontine nuclei, and cerebellum
[ 1J. ADCA type I combines latc-onset cerebellar
ataxia, generally detected in the third or fourth decade,
variably associated with ophthalnioplegia, optic atrophy, dementia, and extrapyramidal signs [ I ] .T h e duration of the disease is about 20 years. T h e age at onset
can vary considerably within and among families [ 3 ] ,
in part due to anticipation, resulting in earlier ages at
onset in successive generations [4]. Phenotypical expression varies as well [ S ] , but the frequency of the
main associated signs is correlated to disease duration
[ 4 ] , so that this variability is more apparent than real.
Linkage studies have identified at least four loci for
ADCA type I: spinocerebellar ataxia 1 (SCA1) on
chromosome 6pZi-p2./1 [6, 7). SCA2 on chromosome
12q23-26-1 [ 8 ] , SCA3 on chromosome 14q26.3-qter
[9f, and SCA4 on chromosome 16s [ l O J . T h e SCAl
gene was identified by positional cloning [ l l ) , after its
location had been established by linkage to the human
leukocyte antigen (HLA) locus [ 6 ] ,followed by analy-
sis of a large number of families with multiple markers
C12-18). T h e SCAl gene encodes a 10-kb messenger
R N A ( m R N A ) which contains a C A G trinucleotide
repeat that is expanded in affected individuals [ 1 I].
Unstable C A G repeats are also responsible for Huntington's disease [ IS}, dentatorubral-pallidoluysian atrophy [20, 211, and Kennedy syndrome [22]. Expanded C A G repeats seem particularly unstable when
transmitted by males [ 2 3 ] . T h e mechanism underlying
this variation is unknown, but is probably meiotic, as
in Huntington's disease [24], since somatic mosaicism
has been excluded C.231.
We collected a large series of 88 families with
ADCA type I and other types of hereditary or sporadic
cerebellar ataxias, and performed a clinical and molecular study in order to (1) evaluate the frequency of the
S C A l mutation in ADCA type I families, (2)investigate whether this mutation is also found in other types
of cerebellar ataxias, ( 3 ) analyze the instability of the
mutation during paternaUmaterna1 transmission, and
( 4 ) determine whether there are clinical features that
distinguish SCAl from ADCA type I with other genetic bases.
From 'INSERM LI 289, HOpital de la SalpkrriCre, and tINSERM
U 114, College de France, Paris. France.
Address correspondence r o Dr Brice, INSERM U 289, Hbpicdl de
la SalpCtrii-re, lr7 boulevnrcl de I'Hbpital, 7 5 0 5 1 Paris Cedex 13.
Received May 18. 1994. Accepted for puhlication Aug 3, 199i.
176 Copyright GI 1095 by t h e American Neurological Association
Material and Methods
Pa tients
Eighty-eight families with progressive cerebellar ataxia were
analyzed. Families with autosomal dominant inheritance were
classified according to Harding [ S ] . The diagnostic criteria
for ADCA type I were fulfilled in 78 families. Previous analyses in 16 of these families indicated that 4 were linked to
the SCAl gene on chromosome 6p 131 and 3 to the SCA3
locus on chromosome 14q [9]. ADCA type 11 was diagnosed
in 3 families in whom cerebellar ataxia was associated with
pigmentary retinopathy. A pure progressive cerebellar syndrome. defined as ADCA type Ill, was observed in 2 families. Sixteen patients with sporadic disease and 3 families in
whom autosomal recessive inheritance was suspected were
also investigated. I n addition, a family with autosomal dominant spastic paraplegia associated with cerebellar signs and a
family with cerebellar ataxia, deafness, and visual loss in
which a mitochondriopathy had been previously excluded,
were included in the study.
Seventy-one families were of French origin. The others
were from the West Indies ( 4 ) ,Belgium ( 3 ) , Portugal (2),
Italy (2),Spain (1). Algeria ( I ) , Yugoslavia (11, Poland (1).
Guyana (1 ), and India (1). The families with non-ADCA type
I or the sporadic form originated exclusively from France.
Statistical analysis was performed using the nonparametric
Mann-Whitney U test and the Yates corrected x L test. The
ages at onset of 14 deceased affected individuals were included in the anticipation calculation. Intraclass variance in
CAG triplet numbers was compared using the F test.
D N A Ami‘ysis
Blood samples were taken from at least one consenting individual of each family, and D N A was extracted using standardized procedures. Polymerase chain reaction (PCR) was carried out in a 25-pl reaction volume using 200 ng of DNA,
1 mM magnesium chloride, 50 mM potassium chloride, 10
mM Tris hydrochloric acid, p H 8.3, 100 FM deoxyadenosine
tnphosphate, 100 pM deoxycytidine triphosphate. 100 p M
deoxythymidine triphosphate, 25 p M deoxyguanosine triphosphate (dGTP), 75 p M 7 deaza dGTP, 10 pmol of each
primer, 2% formamide, and 1 unit of Taq polymerase
(Cetus). Primers Rep1 and Rep2 were as described [ l l ) .
Samples were initially denatured at 96°C for 5 minutes, and
then at 94°C for 10 minutes. during which Taq polymerase
was added using a hot-start procedure. Initial denaturation
was followed by 30 cycles of denaturation for 30 seconds at
94°C. annealing for 30 seconds at 57°C. extension for 30
seconds at 72”C, and a final elongation step at 72°C for 10
minutes. Ten microliters of each PCR product were mixed
with 10 pI of formamide loading buffer, denatured at 96°C
for 15 minutes, run on a 6% acrylamide sequencing gel,
blotted on nylon membrane, and hybridized with a [y-”P) 5’
end-labeled (CAG): oligonucleotide. Autoradiographs were
interpreted after overnight exposure at - 80°C. Allele sizes
were estimated by comparison to a M I 3 sequencing ladder.
Frequency of the SCAl Mutation
The SCAl mutation was detected in 12 of the 78 families classified as ADCA type I, including the 4 previ-
ously linked to the SCAl locus and 8 other families
that were not sufficiently informative for linkage analysis. Ten of the 12 families were of French origin. The
other 2 were from Italy and India. The mutation was
not found in any of the ADCA type I families of other
origin (Belgium, Portugal, Spain, Algeria, Poland, Yugoslavia, West Indies, and Guyana) or in any of the
other hereditary or sporadic forms of cerebellar ataxia.
The SCAl mutation was therefore uniquely associated with ADCA type I, but was found in only 15f4
(12/78) of such families: 16% of those of French descent, 12% of other origins. The difference between
the French and non-French populations was not statistically significant.
Molei-dar Analysis of the SCAI Mutntion
The clinically affected members of the 12 SCAl families and 20 of 41 unaffected at-risk subjects were heterozygous for the mutation: The number of CAG repeats was in the normal range o n one allele of the
gene, but was increased to the pathological range on
the other allele. The distribution of the number of
repeats on the normal and mutated alleles of 61 patients is shown on Figure l . There was no overlap
between the normal and pathological range. Normal
alleles contained 27 to 35 repeats, with a peak at 31
in 18 of the patients (27q’).The number of repeats in
the expanded allele varied over a wider range, 42 to
Paternal transmission affected the size of the expansion. Firstly, although the smallest expansions ( 5 5 4
repeats) were transmitted both paternally ( 4 6 g ) and
maternally (54%), the larger expansions (>.54 repeats)
were transmitted mainly by affected fathers ( 6 4 2).
Secondly, the number of repeats on the pathological
allele of a child was not necessarily the same as in the
affected parent, particularly when it was a father (Fig
2). I n the 9 cases where the size of the expansion in
an affected father and child could be verified, the number of repeats in the child compared to the father varied from - 1 to + 3 , with a mean of + 0.8. In the only
2 mother-child pairs, the number of repeats on the
pathological allele was the same in both mother and
child. Due to the small number of cases, the difference
was not statistically significant, but the tendency toward
greater instability during paternal transmission was
clear. This tendency was confirmed by analysis of variance of the number of repeats in 8 sibships with 23
individuals who received the mutated gene from their
father compared to 5 sibships with 14 individuals who
received the expanded allele from their mother. The
difference was statistically significant (p < 0.05).
Clinical Correlations
The mean age at onset of the 42 affected members of
the 12 SCAl families was 33 k 9 years (range: 21Dubourg et al: SCAl Mutation in ADCA Type I
Fig 1 . Distribution of the nrttnber of C A G repeats in normal
and pathologirzl alleles of 61 SCA 1 patients. Nontzal ulleles
(gray bars) range betueen 2 7 t o 35 repeat unitJ-. and expanded
aNeleJ. (black bars), beticleeti 42 to 67 repeat unitj-.
h( I
F i g 3. Correlation between age at onset (yean) and the number
of C A G repeat3 (r = 0.81).
Clinical Charucteristics of SCAl and SCA3 Patients
+ I
Difference i n numhcr
Fig 2. Di//erenr.eb e t ~ ~ e ethe
n number of repecrt units O N the
pathological alleles of parents and childrm. Maternal transmissions are represented by open bars and paternal transmissions.
b.y hatched bars.
52). There was a significant inverse correlation between the age at onset and the number of repeats
( r = 0.81) (Fig 3). The patient with the largest allele
(67 repeats), who had received the SCAl mutation
through three generations of affected males, had onset
at the earliest age (2 1 years). The course of the disease
was also severe, since he was already markedly affected
when first examined, 1 year after onset. The existence
of anticipation was confirmed in the 12 families. The
mean age at onset was 43 -t 9 (25-64) for parents
and 32 ? 8 (20-50) for children. The difference (11
years) was statistically significant (j< 0.001), and was
the same in 16 father-child pairs (9.25 +- 7 ) and 13
mother-child pairs (9 2 5).
The clinical features of the SCA 1 patients were compared to those of another genetically homogeneous
group, SCA3, in order to determine whether genotype-phenotype correlations could be found (Table).
The mean age at onset was similar in both groups, but
rhe disease duration at the time of examination was
significantly longer in SCA3 than in SCAl patients ( p
< 0.05). The frequencies of most of the clinical signs
associated with cerebellar ataxia were similar in both
178 Annals ot Neurology
Vol 37
No 2 February 1995
Affected examined (n)
Medwomen (n)
Age at onset (yr)
Age at examination (yr)
Disease duration (yr)"
Associated signs (F)
Tendon reflexes increased
Tendon reflexes decreasedb
Extensor plantar response
Decreased vibration sense
Dy sp hagia
Sphincter disturbances
Extrapyramidal signs
38 f 11
48 2 12
10 5 8
6 2 6
ap < 0.05.
b p < 0.01.
populations, except for loss of tendon reflexes, which
was significantly more frequent in the SCA3 group but
probably related to disease duration.
Direct detection of the SCAl mutation in a large series
of hereditary cerebellar ataxias and sporadic forms
demonstrated that the SCAl mutation was associated
only with the ADCA type I form of the disease. This
subgroup, with 78 members, was by far the most numerous. Although the other hereditary ataxias and sporadic forms were less represented, the result is concordant with that in a previous report [25].
The frequency of the SCAl mutation in the ADCA
type I families studied here was approximately 1696,
strikingly lower than the 42% reported in families of
mainly British and Italian descent [25}, but similar to
the 14% observed by Ranum and colleagues in families
from various origins (261. Ten of the 12 SCAl families
were of French origin, probably because the large majority of ADCA type I families in this study were
French (61/78).In any case, the genetic heterogeneity
of ADCA type I is great, as indicated by the number
of loci already identified. According to the present
study, the SCAl mutation is responsible for less than
one sixth of the cases, and may also be unevenly distributed, even among European populations.
In accordance with previous reports, the mutated
alleles of the SCAl gene in our patients contained at
least seven more repeats than their normal counterparts. The size range was similar to that observed in
the initial study of Orr and coworkers { 1I}, but “intermediate” alleles with 39 repeats, reported by other authors (271, were not detected.
The mutation was unstable, especially during male
transmission. Although the number of parent-offspring
pairs was limited for statistical analysis, three observations support this conclusion: The mean number of
repeats was higher in paternal transmissions than in
maternal ones, the largest alleles with more than 54
repeats were predominantly transmitted by males, and
repeat length varied more when transmitted by males.
Instability was limited, however, to at most three additional repeats in the progeny of male transmitters. The
difference in repeat numbers did not exceed 10 in paternally transmitted sibships. Larger expansions may
not have been observed in our families, however, since
there were no juvenile cases among the patients. It
cannot be excluded that very large expansions were
lethal during embryonic de-Telopment. Since such lethal expansions would preferentially occur during male
transmission, the ratio of affected to total offspring
would be lower for affected male transmitters. This is
not the case however: 42 of 67 offspring of 20 male
transmitters (0.63) and 5 1 of 77 offspring of 20 female
transmitters (0.66) were affected.
The mechanism underlying the expansion of the trinucleotide repeat in the SCAl gene is not known. It
was observed that the pathological alleles in the lymphocytes of affected individuals contain the same number of repeats, indicating that there is no somatic mosaicism 1231. The expansion probably occurs during
meiosis, as has been demonstrated in a study of the
expanded CAG repeat in spermatozoa of patients with
Huntington’s disease [24}.
What are the clinical consequences of the SCA 1 mutation! The significant inverse correlation between the
number of CAG repeats in the pathological allele and
the age at onset of the disease suggests that even the
moderate expansions, observed particularly during
male transmission, should result in anticipation, which
is effectively observed. Disease onset was approximately 11 years earlier in offspring than in their parents. Anticipation has already been reported in both
type I and type I1 ADCA ( 4 , 281. As the unstable
CAG repeat is responsible for the SCAl form of
ADCA type I, a similar mechanism may be involved
in the other forms as well.
Can ADCA type I patients with the SCAl mutation
be distinguished from those with other mutations!
Since it was possible to identify the SCAl patients
by direct molecular analysis, they were compared to
another genetically homogeneous group, SCA3 patients, in an attempt to identify phenotype-genotype
correlations. There were no detectable differences, except for more frequent decreased or absent tendon
reflexes in the latter, probably a consequence of the
longer disease duration in these patients rather than a
real phenotypical difference. This result underlines the
difficulty of distinguishing subgroups in patients with
ADCA type I using a purely clinical classification. Molecular analysis only will allow a pathogenic classification and give new insights to the physiopathology of
this heterogeneous group of disorders.
As a final consideration, it is now possible to diagnose the SCAl form of ADCA type I by direct molecular analysis. This makes predictive testing possible,
raising important ethical, legal, and psychological issues. The international guidelines developed for genetic counseling of patients with Huntington’s disease
[29} could be usefully adapted for predictive testing in
SCA 1 families.
This study was supported by the Association Franqaise conme les
Myopathies, the Groupement de Recherche5 et d’Erudes sur les Getiomes, the Verum Foundation, and the Association pour le Developpemenr d e la Recherche sur les Maladies Genetiques Neurologiques et Psychiatriques. Dr Dubourg received a fellowship from
the Fondation pour la Recherche Mbdicale.
T h e authors wish to thank all the families who participated; Professors G. Rancurel, D. Laplane, C . Pierrot-Deseillignyq and H. M.
Cann; and Drs M. Serdaru and J. Melki for referring some of the
index patients. We are grateful to Professors J. Feingold, A. Harding,
an11 H. Zoghbi for helpful discussions; to Y. Pothiii, F. Rogeriew,
and 1. Lagroua for technical anJ medical assistance; and to Dr M.
Ruberg for critical reading of the manuscript.
I . Berciano J. Olivopontocerebellar atrophy. A review of 117
cases. J Neurol Sci 1%32;53:253-272
2. Harding AE. The clinical features and classification of the late
onset autosomal dominant cerebellar ataxia. Brain 1082;105:
.I - L 8
Zoghbl HY, Po,lack MS. Lyons LA, er al Splnocercbellar ataxtd
variable age of onset and leukocyte antigen in a l u g e kindred
Ann Neurol 1988,23 580-584
Dubourg et al: SCAl Mutation in ADCA Type I
4. Durr A. Chneiweiss H. Khati C. et al. Phenotypic variability in
autosomal dominant cerebellar ataxia type I is unrelated to genetic heterogeneity. Brain 1993;l 10: 1497- 1508
5. Harding AE. Clinical features and classihcation of inherited ataxias. Adv Neurol l093;6 I: I- 14
6. Yakura H , Wakisaka A, Fujimoto S. cc al. Hereditary ataxia and
HLA genotypes. N Engl J Med 19’4;291:154-155
7. Volz A, Fonarsch C, Ziegler A, et al. Regional mapping of the
gene for autosomal dominant spinocerebellar ataxia (SCA 1 ) by
localizing thc closely linked D6S8‘1 locus to 6~23.05-p24.1.Cytogenet Cell Genet 1992;60:i7-39
8. Gispert S, Twells R. Orozco Diaz G , et al. Chromosomal assignment of the second (Cuban) locus for autosomal dominant cerebellar ataxia (SCA2) to human chromosome 1 2 ~ 2 3 - 2 4I . Nature Genet 1993,4:295-299
9. Stevanin G, Le Guern E, Ravisi- N , et al. A third locus for
autosomal dominant cerehellar ataxia (ADCA) type I maps to
chromosome 14q26.i-qter: evidence for the existence of a
fourth locus. Am J H u m Genet 1994;54:11-20
10. Gardner K, Alderson K. Galster B, et al. Autosomal dominant
spinocerebellar ataxia: clinical description of a distinct hereditary
ataxia and genetic localization to chromosome 16 (SCA4) in a
lJtah kindred. Neurology 1994
11 O r r HT. Chung M, Banfi S, et al. Expansion of an unstable
trinucleotide C A G repeat in spinocerebellar ataxla type 1. Narure Genet 1003;4:22 1-226
12. Rich SS, Wilkie P, Schut L, et al. Spinocerebellar ataxia: localizamarkers o n
rlon of an autosomal dominant locus hetween r~.~o
human chromosome 6. Am J H u m Genet 1987;4 1:524-531
13. Khati C. Stevanin G, Durr A, er al. Genetic heterogeneity of
aurosomal dominant cerebellar ataxia type 1; clinical and genetic
analysis of 10 French families. Neurology 1993;d3:1131-1137
14. Zoghbi H Y . Jodice C, Lodewilk A, et al. T h e gene for autosonu1 dominant spinocerebellar ataxia (SCAI ) maps telomeric to
the HLA complex and is closely linked to the DOS89 locus in
three large kindreds. Am J H u m Genet 1991~%9.23-30
15. Keats BJB, Pollack MS, McCall A, et al. Tight linkage of the
gene for spinocerebellar ataxia to D6S89 on the short arm of
chromosome 6 in a kindred for which close linkage to both HLA
and F13Al is excluded. Am J H u m Genet 1991;49:972-9’7
16. Ranuni LPW. Duvick LA, Rich SS, et al. Localization of the
aurosomal dominant HLA-linked spinocerebellar ataxia (SCA 1)
locus. in two kindreds, within an 8-cM subregion of chromosunie 6p. Am J H u m Genet 1991;49:.31-41
17. Kwidtkowski TJ. Orr HT, Banfi S, et al. T h e gene for autosomal
180 Annals of Neurology
Vol 37
No 2 February 1395
dominant spinocerebellar ataxia (SCA I ) maps centromeric to
D6S89 and shows no recombination. in nine large klndreds.
with adinucleotidc repeat ar the AM10 locus. Am J H u m Genet
1093; 5 3:39 1-400
18. Jodice C, Frontali M, Pcrsichetri F, et al. T h e gene for spinal
cerebellar ataxia 1 (SCA I ) is flanked by two closely linked highly
polymorphic microsatellite loci. Hum Mol Genet 1093;2: 1383I387
19. T h e Huntington’s Disease Collaborative Research Group. A
novel gene containing a trinucleutide repeat that is expanded
and unstable on Huntington’s disease chromosomes. Cell 1993;
20. Koide R, Ikeuchi T, Onudera 0, et al. Unstable expansion of
C A G repeat i n hereditary dentatorubral-pallidoluysian atrophy
IDRPLA). Nature Genet 1994;6.9-13
2 1. Ndgafuchi S, Yanagisawa H , Sato K, et al. Denratorubral and
pallidoluysian atrophy expansion of an unstable C A G trinucleotide o n chromosome 1Zp. Nature Genet 1994;6:14-18
22. La Spada AR, Wilson EM, Luhahn DB. et al. Androgenic receptor gene mutations in X-linked spinal and bulbar muscular arrophy. Nature 1991;352:7’-79
23. Chung M, Ranum LPW, Duvick LA, et al. Evidence for a mechanism predisposing to intergenerational C A G repeat instability
in spinocerebellar ataxia type I. Nature Genet 1993;5.254-258
24. Duryao M, Ambrose C. Myers R, et al. Trinucleotide repeat
length instability and age of onset in Huntington’s disease. Nature Genet 1903;4:387-392
25. Giunri P, Swteney MG, Spadaro M, et al. T h e trinucleotidr
repeat expansion on chromosome 6 p (SCA 1 ) in autosomal dominant cerebellar ataxias. Brain 1994;117:045-649
2 0 . Ranum LPW, Chung M, Banh S. et al. Molecular and clinical
correlations in spinocerebellar ataxia 1 t SCA I ): evidence for
familial effects on the age of onset. A m J H u m Genet 1994;55:
2 7 . Matilla T, Volpini V, Genis D, et al. Presymptomatic analysis
of spinocerebellar ataxia type 1 (SCA1) via the expansion ofthe
S C A l CAG-repeat in a large pedigree displaying anticipation
and parental male bias. H u m Mol Genet 1993;2:2123-2128
28. Benomar A, Lt. Guern E , Diirr A, et d. Autosomal dominant
cerebellar ataxia with retinal degeneration (ADCA type 11) is
genetically Jiffecent from ADCA type 1. Ann Neurol 19‘J4;35.
29. World Federation of Neurology: Research Committee. Rrsearch group on Huntington’s chorea. J Neurol Sci 1989;94:
Без категории
Размер файла
494 Кб
repeat, correlation, scai, ataxia, cag, large, molecular, clinical, dominantly, analysis, number, familie
Пожаловаться на содержимое документа