The trinucleotide repeat expansion on chromosome 6p (SCA1) in autosomal dominant cerebellar ataxias

Brnìn 1t994). ll1 , 645 549

The trinucleotide

repeat expansion

on chromosome

6p (SCA1) in autosomal

dominant

cerebellar

ataxias

P Ciuntj,r M. G. Sweeney,r M. Spadaro,l C. Jodice,l A. Novelletto,l P Malaspina,3 M. Frontalia and A , E , H J r O r n g

'Uni|e$ir\) Depa 'nenî oJ Clinical Neurclogr. htsîitute of Neurolog\', Landon, UK, )ts!ittto di Clinica delte Ma atîie nenose e Mentali, Universi!,- It Sapien:!', 3 Dipaftimenîo di RioloSia. Univercib^ Tor Ver7aîa and'kîirtto.li

Medicitla SpeùnentaLe, Rane, hal\

Summary

Affected nenbers of 73 Íanilies eith a raierv o.f auîasonaL domìna t laîe onseî cerebelltt atarias (ADCA|) were investigate.l .for îl1e tìturcIeotide (CAG) .epeaî e.rpansion which is Íouhd ia pe.lisrees e.rhibiitlg tinkaqe to îhe SCAI Iocus on chromosone ó. Most oj îh, families werc too snaLl for linkage analt"sis. The nutatío^ \ras o t Íound in ADCA type L in 19 out af 38 such kùt1rcds i estigarcd (50qÒ). h b'as slíghtlf nore connon ín ltalian (59Ea) than Briîish(50qa) ÍaníLies, and i,as ako faund in Malar-sian, Bangladeshi and Janaican kindreds. Averall, ADCA Npe I paîienrs tírh the

Key words: dominant ataria: trinucleodde rcpeat

Coùespo ence ta: Prafessaî A. E. Haùing, Insrlute af Neunlos:". Queen square, London wctN 3BC. UK

erponsion had a lowet incidence of hyporcfexia and faciaL Jascic lation tha rhose withouî. The trinucleotide eryansíon

was ,ot lound in eight Janilies vùh ADCA and mac lopathr or 21 kindrcds wnh a pùre îype of ADCA. confumin9 thal these ,^ndromes are geneticalll disîinct. h was also 4oî detected ìn 12 patìents wiù sporadic degenercdye aîúia'

DNA analwìs for îhe SCAI nuîation is Lsefu! díaenosticaqr in singLe patients at snalL fanìlies, and can be used Íar pre t|nptanatic testíng iÌhere appropriate.

Introduction

The autosomal dominant late ons€t cerebellar aîarias (ADCAS) _{are clinically and geneticaÌly} heterogeneoùs. Fam iìy studies suggesr tha!, compared wilh orher syndromes which arc gene!ìcally distinct, there is ìitde cìinical evidence of hete.ogeneity in famili€s with ADCA associared with variabl€ combinations of supranuclear ophthalrnoplegia, optic atrophy, mild dementìa, peripheral neuropaÌhy or eiùa pyramidai dysfuncdon (Harding, 1982). There is, neverhe-less, evidence for locus heterogenejty wirhin this broad group (ADCA type I). Linkage sludies have showî thar the disease gene maps to the short arm oichromosome 6 in several large famiìies (Zoghbi er al., 1988, 1993; Jodice e! aL., 19931 Kwia*owski er al.. 1993), and there is a locus on chromosome 12 in Cuban pedigrees (Cispen e, dl., 1993). It is difficulr to dislinguish these clinically, and rhe sa!ìe applies ro families with Machado-Joseph disease, a syndrome initially rcponed in individuals of Azorean?onuguese descenr (Courinho and Andrade. 1978). Th€ gene locus for MachadFJoseph disease has been mapped to chromosome l4q ìn both Japanese and @ OrÍord Universiry Press t994

Portuguese families (Takiyami sf dl., 1993). However. an ADCA gene has also b€en mapped to îhis locus in three French families not considered to have Machado-Jos€ph disease on clinical grounds (Stevanin et a1., 1994).

Aùrosomal dominaÍr late onser cerebellar araxia and pigmentary macular dystrophy (ADCA ll) is clinically and genetically distinct from ADCA type I (Harding, 1982; Erevoldson et al., 1994). No positive linkage data have been reported in this disorder, allhough the disease locus is not finked to úe HLA locus on chromosome 6p (K\trnar eî al., 1986). The pathological findings in ADCA types I and II are usually those of olivopontocerebellar alrophy, bur nearÌy always with abnormaliries €lsewherc, inclùding Ìhe basal ganglia, spinal cord and peripheral newes. There is also a laler onset (usually over 50 yea.s) relatively'pure'cerebellar syndrome in which ocular or extmpyramidal features do not occur (ADCA type III), with autopsy findings of cerebellar corical atrophy with loss ofcells in úe denta@ and,/or olivary nuclei (Hofftnan er al., l97l). No linkage studies have been

646

î^ble I Clinical clussifL\uion.tnd reruhs ofSCA| núttatioù a,úl|sis

Number oi tamilies wirh mùrariot/roral number of families r%r B r i r i s h

A D C A I ADCA II A D C A II I

Other dominant iìtaxiirs with demenia,hyocìonus

cùebellar araxia (cases)

ó/r2 (50) 016 Dt2l r0/r7 (59) all D t l 0, 3t9 l33t 0 / t OE r 9 / 1 8 ( 5 0 ) 0/8 a/24 0/l o t l a t l a t 1 2

on

reponed in thìs disorder, apar from exclusion of linkage to 6p in a family wirh rather e.rlier onset and a faúìy indolenr coorse (Fronral; cr .r/., 1992).

R€cen(ly, the motation in the ADCA gene on chromosome 6p (ar rh€ locus t€nÌed SCAI) hîs been identified {Chung eî al., 1993, On i/ d1., 1993). lt is an unstable expanded ùìnucleotide (CAC) repeat sequence, with simrÌaf charrcter i s r i c s ! o r h € H u n t i n g r o n \ d i s e a s e m u t a t i o n ( H u n t i n g t o n s D i s e a . e C o l l $ o r i î r ! e C r o u p l o o l r . T h e r e . - n n v e r s e co.relation between repeat length and age of onset of the disease. This discovery makes it possible ro determine ii aifecrcd members of families too small fo. linkage analysis h a v € t h e SCAI mutation. T h e results o f s u c h a s t u d y i n 7 3 famiiies with varìous forms of ADCA and I2 patients with late onset degenerative ataxias who had no hisrory of affected relrlives are reporîed in this pap€r.

Patients and methods

Patients

DNA was analysed from at least one member of 73 families wilh a varìety of dominant ataxias (Table l). Familìes were considered to have ADCA type I ìf the age of onset was <50 y€ars in îhe majority ofrffected iamily members, ànd at least one patient per family had features such as sùpranuclear ophthalmoplegia, optic atrophy. pe.ipheral neuropathy. dementìa and fascicùlations of the limbs and,/or face during ùeir illness. Also included ìn ùis group is one family where the only palienl examined had a pure c€r€bellar synd.ome at an early srage of th€ diseas€ (<5 years duration), as ìs c o m n o n in A D C A t y p e I (Hardins, 1 9 8 2 ) . l n c l u s i o n i n this category wils warranted by consistendy early onset in the famìly (<45 years) and rhe p.esence of cerebell.r rnd br3lnnem atrophy seen on rmaging srudies. ruggesring r paúological rubstrate of olivopontocerebellar atophy rather than cerebello-olivary arophy which is usuallyseen in ADCA type IIl. Thre€ of ùe 12 Brjdsh ianilies wiù ADCA rype I were in the study of Hardine (1982), including îamilies I and 2: one of these (iamily 2) was known to have a locus on chromosome ó on rhe basis oflink,rge analysis (Shrìmplon et al., 1993). Of the l7 ltaìian famìlies wiîh ADCA lype I.

seven were known to have iì locus on chrcmosome 6p (Conighi

"r a1.. 1984r Spada.o et al.. 1992. Jodice er a1-. 1993). and in two this had been excluded (unpublished data).

Type II ADCA was defined by the consìrent findin-s of maculopathy in atf€cted lamily m€mb€rs (Harding. 19821 Enevoldson er dl., 1994). Type III ADCA was defined as a pure cerebella. syndronre. occrìsionally with distaÌ loss of vibrarion sense or pyramidal signs in rhe ìegs. bur wirhout th€ additional features s€en in ADCA ryp€ I r€f€rr€d lo abov€. after disease dura(ions >5 years in alì aff€cted family nembers. This syndrcme is probably hererogeneous- In ll families, the age of onset was consistendy late, over 40 in aìl patients and 50 or over in all but three. In th€ other 13 (one Iulian; Frontali €/ rìl.. I992). onset was earlier, ranging between the second ro iou.rh decade of life. In all these 13 families ihe disease course was only slowly progressive. with patients rìot losing the ability to walk until 30 or 40 yeals afi€r onset. Thr€e families wi(h ra.e or less wellcharacterized dominanr araxias were also stùdied. of which d€tails are given in Table L

Fiiry-six of lhe families were studied clinically in th€ UK by PC. or A.E.H. Th€y origjnated from the UK (42), Ind;a, Pakistan or Bangladesh (thrce), lhe West lndies (thre€), Iraly (two). Malaysia (one), the Philippines (one). Brazil (one), China (one) Denrnark (one) and Belgium (one). The 17 families studied in ltaly (all Italian) were assessed clinically by M.S. or PG. We also investigated 12 patients (ll in the UK, one ìn ltaly) with'idiopathic'lare onset cerebellar ata-(ia (Harding, l98l), in whom there was no history of affecÈd reladves. Nine of these were Brirish, one Afro-Caribbern, one Indian and one ltalian.

For clinical comparisons. age ofonse! data were compaÌed using the Mann-WhitÍey test. ,nd the presence or absence o[ clinical features _{using Xr analys;s} of raw data with Yates'

DNA d.nallsis

In London, analysis oi the trinucleotide repea! in úe SCAI gene was perfo.med in UK-based patienb and 44 healthy

SCAI n îaîion it1 do'rinant atatias 647 î^hle 2 Ca|Voríson oJ L liilicul Jè atLxe s of ADCA I Jatti n'es $ ith utd virhau! îhe SCA I nt|tltion

S C A l N o ( S C A I Median age aÌ onwÌ (yeùt i!

Supranucle!r ophÌhalmoplegia Sraring gaze Dysphagia ExtrupyÉmidal fu!rurcs DepÈssedhbsenr !eîexcs 3 4 ( 1 7 - 5 3 ) ( , = 101) ól 5 3 34 (9-5.r) 6 1 t 0 t 0 48 14

llr

+Frcm àvailable clinicel dara on .16 SCAI patierts did 30 non-SCAI parieirs.

control sub.jecrs, using a modification ofth€ meúod desc.ibed (Off er al.. 1993). The polymerase chain reaction (PCR) was carried _{out in a 25 Él reaction} volume using 100 ng DNA,

L 5 m M M g c l r , i0 m M K C l , l 0 m M T r i s . p H 8 . 1 0 0 j r M dATP, l0 !M dCTe I00 UM dTTP 100 !M dGTP l0 pmol each pnmer. 0.1 ICi lo-]rlPdcTP and 2.5 unris Taq polymerase. Primers were as descrìb€d (OÍ er al. 1993). In;lial DNA denrturarion al 94'C for I min wr5 followed by 25 cycles of 30 s €ach ar 94, 57 and 72'C, respecdvely. and a 6nal elongarion srep of l0 min at 72oC. Products of the PCR w€re fractionated on 67, denaturing acrylanide gels. Allele sìzes were estimated using an Ml3 sequencing l:dder, w h i c h w e conside. g i v e s a n a c c u r a c y o f + 2 r e p e a t u n i s i th e SCAI mutarion was defined by the presence of >42 repears in one alÌele (On et al., 1993). The patients based in Italy were investigared in Rome using the merhod repoded by Orr eî at. (t993).

a n d m o r e m a r k e d i n s t à b i l i t y d u r i n g p a t e m a l r r a n s m i s s i o n (OÍ e! ul., 1993).

A compdrison of the clinical feàtù.es of ADCA type I in p r r e n r \ w r r h . ì n J w i r h o u r r h e S C A I e r p r n 5 i o n i s g i v e n i n Table 2. The nedian and ranges of age at onset of symproms were vÉry similar Overall, rhe non SCAI patierts had a higher incidcnce of depressed or absent, as opposed to incrcrsed, reRexes. Relìex loss increa5es wirh disease durarion in ADCA lype I (Harding, 1982), but there was a higher proponion of non-SCA I cases with hyporeaexia (four our of

l l c o m p a r e d w i t h three _{o u t o f 3 l S C A I , X'z= I0.7. P =} 0.004) in patients with disease durarions of 7 years or less. In longer durarion cases there was no signìfrcant difference between the two groups in €lms ofincìdence of hyporeflexia ( f o ù r o u r o f l 5 S C A l , s e v e n o u l o f l 7 n o n ' S C A I ) . T b e o t h e l s i g n i f l c a n t f l n d i n g w a s t h a t m o r c n o n SCAI patients h a d facial or lingual fascicularion. This w$ nol.elated to disease d'rration (dara not shown).

None of the seven families with ADCA rype II (at (ia and macular dystrophy), or ùe 24 wirh the 'pure ADCA type III. had the SCAI mutation, and ùe same appli€d lo those with less common dominanr disorders. including a family with intermittenr araxia, and the 12 patients without

Discussion

This study connrms the p.evìous clinical observctions which separated the pùre type of ADCA (type lll) and that with maculopathy (type II) from the mo.e common and variable m u l t i s y s î € m s y n d r o m e A D C A t y p e I ( H a r d i n g , 1 9 8 2 ) . T h e S C A I e x p a n s i o n w a s o n l y o b s e r v e d i n p a t i e n t s w i t h ADCA typ€ I, in half of the families o.ìginating from the UK and slighrly mo.e of those of Italian origin. Thìs may reflec! a degree of ascertaìnment bias in the Italian families as these w e r e c o l l e c t e d i n i r i a l l y f o r c h r o m o s o m e 6 l i n k a g e s t u d i € s ,

Results

The classificatioo oi th€ famili€s is shown in Table ì. a5 is rhe number of kindreds wirh ùe SCAI mumdon. Ten of 17 families origjflating from ltaly with ADCA type I (59%) had the mutation. compa.ed with six of 12 families of British ong'n (50%), and th.€e of nine other kindreds. The lnsr \rere Malaysian, Jamaican and Bangladeshìi other ADCA rype I lamilies from lndia. China. th€ Wes! Indies, B.azil. Belgium and Denmà.k did not have lhe SCAI mutation. Nor surpfisingly. all of th€ eìght tumilies who had shown linkage lo chromosome 6p markers had lhe SCAI expansion and |his was not present in the two ADCA type I families in which linkage io this rcgion had been exciuded. The cbaracÈrisrics of îhe SCAI rnutation in rhese families have been reponed elsewhe.e (Jodice at al., 1994). con6ming the reladonshìp between r€p€ar lengrh, age of ons€r and severity

ú8 P Giunti er ^1. :

Acknowledgements

We wish to thunk many colleagues _{for r€fe.ring famities and} helping ìn obraining blood samples or clinìcal informarion. The work was supported by the Friedrcich s Araxia crcllp (UK). th€ Glyn Worsnip/Raymond Bames anxia fund. CNR PF Ingegneria Ceneri€a and Telerhon Ilaìia (gram no. 358).

Bryer A, Manell RW. du Toir ED. BeighÌon P Adulr onser spinocerebellar ataxir linked ro HLA in a South African kindred of n r i x e d a n c e s r . y . T i s s u e A n t i g e n s 1 9 9 ? : 4 0 r 1 l l - 1 5 .

Chung M, Rarum LPW Duvick LA, Servadio A. Zoghbì HY. Or Hî Evidence for , mechànnn predisposin8 ro iitereeneratìonal CAG repelr insmbiìiry in spinùerebellc. c(.x i! Ìype L Nlrue Cener

19931 5: 25.L8.

Conighi C, Grunieri E. Toh R. Sensi A, N4aÌliuz PL. HLA and spìoocerebellar dúxiar study of ! famiÌy. Iral J Neu.ol Sci 198,{: 5 Suppl 2: 154-E.

Courinho e Andrade C. Aùtosomrl dominant syren degenerarion in Pódùguese families of rhe AzoBs islands. Neurology l9?8: 28: 103-9.

Enevoldson PC. SrndeA MD. Harding AE. Autosomal doóinanr cùebeUr ùàxìà wlrh maculm _{dyrrophyr a clinicdi xrd Seneric} s r u d y Ò l e i g h r f l m i l i e s . B m ì n 1 9 9 4 : l l 7 r i l 4 5 - 6 0 .

Fronrali M. Spada.o M. Ciùntì P. Birnco F, Jodice C. Pe6ichetÌi F. €r al. AurÒsoúal dominanr puÉ cerebellar ntaria. Neurclogicd and generic rudy. Brain I992r l15: 1647-54.

GìspeÍ S. Twelh R, Orozco G, B.ice A, Weber J, Heredero L. €t al. Chromosomal lssignmenr of lhe second locùs tor autosomal dodinrnr ceÈbellar ataxia (5CA2) to chromosome 12q23-24.1. Nrrúre Cener 1993r 4: 195-9.

Hardin-q AE- ldìoparhic làte Ònset cereb.llùr ataxia. A clinical ind g € l e r i c r ù d y o f 3 6 c a s e s . J N e ú r o l S c i l 9 8 l : 5 1 : 2 5 9 - 7 1 . Hùding AE. The dinical features and classification ofùe ìate onset aurosomal doninant cerebellar aîaxiasr a srudy ol ll faúilies. incÌudine descendanK ol th€ DÉw famiìy oi walwonh'. BÉin 1 9 8 2 ; 1 0 5 : 1 2 8 .

Hoffman PM, Sluar wH, Eùle K-lvl, Brcdy .lA. íeÉdnary late_ onser cecbellar degeneratior. Neurclogy 197 I i 2 I : ?71 7.

Hùntington\ Dúease Collaborative Croùp. A tovel genecothining a rdnucl€oride repeùt rhat ìs expanded and unsrable on Huntington\ disease chronosomes. Ceìì 1993i 121 97l-83

Jodice C, Frontali M. P€Kìchetti F. Noveìl€tto A. Pùdolfo M. Sp.daro M, er at. The ge.e for spinal cercbellat atdia I (SCAI) is flarked by rwo closely lirked highly polymorphic mic.osateìla€ Ioci. Hum Mol Cener 1993;2: l38l-7.

Jodice C, Malaspina P, Pe6ischetti F, Noveìletto A, SPaddo M. Ciunti q et al. Eifect of trirucleoride repeat len8th and Pù€rtal sex on phenorypic va.iaîion in spinocercbellar ltaxia l. Am I Hum Ccnet. 1994i 54: 959-ó5.

Khlri C, Slevúin O, Dur A, Chneiweiss H, B€lal S Seck A er al.

rnd cena;nly one was known to be 6plinked in I 984 (Conighi et al., 1984). Also, four of the SCAI families are from Calabria and probably reflecr a founder effecl in thìs part of Ìîaly. The British iamilies were rsce(ained solely by sysÌematic DNA storage irom all families wilh ADCA seen by A.E.H. over a 7 year period, and a mo.e recenl intensive coìlection of British families wih any type ol dominant araxia. In the French population. six of the l0 fam;lies for whom linkage analysis infomation is available showed evidence for a locus on chromosom€ ó (Khali et al.. 1993), a sim ilar proportion !o that observed in our European patients. The pres€nce of the SCAI mutatior in Malaysian, lamaican and Bangladeshi kindreds suggers that ùis genetic defect ìs not confined to Caucasian popularions, and may weil have multiple origins io differenr racial groups. ai is the case in Hunlinglon\ disease. SCAI has previously been described in Black American and Cape coloured (Sourh African) famili€s (Zoghbi et al.. 1988t Brye( et al., 1992). However, SCAI as defined by linkage analysis has been reported most commonly in families of European o.igin (Zoghbi

"r al., 1993), and other types of ADCA may b€ more frequenr in certain popuìations. This clerrly applics to Azorean or Poruguese fàmilìes, and possibly the Japanese (Takiyama e t a L . , 1 9 9 3 1 Z o g h b ì e t a l . , 1 9 9 3 ) .

A s i n t h e F r e n c h s l u d y ( K h r d .r dl.. 1993), i n w h i c h 6 p -linked ADCA was ìndistinguishable irom non-6plinked ADCA, we couìd identify reladvely f€w cìinical differeoces belween ADCA type I patienls with the SCAI mutarion and those without. The larcr rend to exhibil dep.essed or absent iendon.eflexes, asopposed to hypeneflexia, in lhe first7 years of the disease- The hìgher incidence of fìcial fasciculation in the non-SCAI families is of interest as this has been repofied frequently in patients wilh Machado-Joseph disease {Courinho and Andrade, 1978) and in the Cùban families whh a disease locus on chromosome 12 (Orozco Diaz er d1., 1990). h should be slressed that these were group difierences derived from a relatively large number of patients and do

not se|ve !o distjnguish SCAI from other types ol ADCA type I in indivìdual patients or families. Our non-SCAI families are likely to be genetically hetercgeneous and prcsumably have dis€ase genes on chromosones 12, 14 or elsewhere (Slevanin er a/., 1994). lt seems highly probable ùat all the mùtations underlying ADCA typ€ l. and ADCA type II which exhibils strikìng anricipaîion and a patemal effect (Enevoldson et aL., 1994), will be unstable llinucleo!ìde repeats.

Most of our families wirh the expansion were ioo small lor linkage analysis. and direcr muration detection is clearìy very useful in clinical practice in making a diagnosis of SCAI in such kìnd.eds. If applicrble, rhis analysis can be used for presymptomaric diagnosis, and we have already perfomed rhis in lwo at risk subjects. using the same prc-and post-test counselling protocols as in Hunrington s disease famìlies (World Federîrion of Neurology Research Croup on Hunlington's disease, 1989).

SCAI nuuion in doninant atarias 649

cererjc hcrerogeneìty of.urosomal dominaút cercb€lìar rtaria rype l:clinical lnd genetic analysis of l0 Frcnch f.milies. Neurology 1 9 9 3 i 4 3 : l l l I 7 .

Kumù D. BIa.k CE. Celsthorpe K. H€rcditary cercbelln. ah(ia aDd gerctjc linrige with HLA.. Hum Genet 198ó1 72: 327-32. KwìarkowskiTl, Or HT. Banfi S, McCùlì AE, Jodice C, Pe6ich€ui F, er !1. The gene ior aurosomal dominant spinoccrehellù araria (SCAI) maps cenrrcm.ric to D6S89 and shows no rccombinaùon, in nine lafge kindr€ds. wilh a dinucleotide repeat ar rhe AMl0 locus. Am J Hun Oener 1993; 53: 391--100.

Orozco Di.z C, Noda6e Fbnes A, Cordoves Sagaz R. Aùburger G- Auiosomal dominanr cercbellar ataria clinical nialysh of 263 paricrrs froú a homoBeneoús populaior ir Holeùin. Cub.. Neu.ology 1990; 40: 13ó9-75.

Or Hî Chure MY, Ban6 S, Kwiatkowski TJ, Sedadio A, Beaudet A. er il. Expansior of aD unstable dnucleoúde CAC repe.r in spinocerebellar alriii rype L Nrtrre Cener l99l: 41 221 ó-Shrinìpton AE, Davidson R, MacDonald N, Brock DJH. P.esymptom.rìc Ésring for autosonal dominant spinocer€bellîr ntaria typc L J M€d 6enet 1993; l0: óló-l7.

Spadaro M, Ciunri q Lulli P, Fronoli M. Jodice C, Cappeuacci S, et al. HLAlinked spinocerebellù ataria; a clinìcal a.d gen€tic srudy oflurgc ltàlian kindreds. AcraNeùrclScand 1991i 85:257-ó5.

Srev;;in C, rc Cu.m E, Ravisé Ni Chneiveiss H, Dùr A. cancet G, er-al. A rhid locus for aurosomal do.ninant cerebellar Àraxia type I maps ro chronosome l4q?4.3'qler: evìdence for the exiu€Dc€ of r lounh locus- Am J Hùm Cener ì9941 54:

ll-20-Takiyama Y, Nishizawa M, Tanlka H, Kawashima S, SakaDoto H, Kùube Y, er al. The Eene lor MachldoJoseph disease maps ro human chromosomc l4q. Nature Ccnet 1993i 4: 300-4.

World Fededion ol Neurology: Rcsedch Commitieè- Rescarch Group on Hurrington\ disease. Ethicùl issùes policy statemenl on Húntii8ton s disease moìecular geierics predictive lests. J Neurol Sci 1989i 94: 327-32.

Zoghbi HY Pollack MS, Lyons LA. Ferell RE, Daiger SP, B.aldet AL. Spinocecbellar araxia: variable nge of onsct and linkag. ro human leuko.yte antiCen in a large kindred. Ann Neurol 1988i zli 580-4.

Zoghbi HY Frontali M. Or Hî Sandkuùì L. Cnnn H- Sasd\i H. er al. Linkage studies in dominantly inhern€d ataxìas. Adv Ne!rcl

1 9 9 3 ; 6 1 : 1 3 3 - 7 .

R..eiw.l Decenber JI, 1993 Raile.l F.bnxtry 23, 1991. Ac.epQd lttarch lE, 1991