Genetic mapping of the spinocerebellar ataxia 2 (SCA2) locus on chromosome 12q23–q24.1

GENOMICS

25,433-435

(1995)

Genetic Mapping of the Spinocerebellar Ataxia 2 (SCA2) Locus on Chromosome 12q23-q24.1 A. LUNKES,* A. HERNANDEZ,*-’ C. MAGAR#JO, *s2 5. GISPERT,* N. SANTos,**t G. ORozco,t L. HEREDERO,’ J. BECKMANN,* AND G. AUBURGER**~ *Departmentof Neurology, University Hospital, P.0. Box 70 1007, 4000 7 D&se/don: Germany; *Department of Neurology, Hospital Lenin, Holguin, Cuba; and Kentre d’Etude du Polymorphisme Humain JCEPH),27 Rue Juliette Dodu, 75010 Paris, France Received June 14, 1994; revised November 8, 1994

chromosome 12q23-q24.1 markers DlZS58 and PLA2 (Gispert et al., 1993a) in an interval of 30 CM. Exploiting the large SCA2 collective in Holguin further for a refined genetic map, we now report a region of complete linkage disequilibrium close to the marker D12S105 together with multipoint and haplotype evidence placing the SCAB mutation in a ~-CM interval telomeric to the marker D12S84 and centromeric to the marker D12S79, in contrast to previous multipoint reports favoring a position proximal to D12S84 (Pulst et al., 1994; Lopes-Cendes et al., 1994).

A refined genetic map of the spinocerebellar ataxia 2 locus was constructed through linkage and haplotype analysis of 11 large pedigrees from the Holgufn SCA2 family collective. Three-point analysis makes a localization of the SCA2 mutation in the 6-CM interval DlZS64-DlZS79 likely. This is consistent with haplotype results indicating a crossover event between two branches of the SC32 family Rs and placing the mutation on the telomeric side of D12S84. The microsatellite D12S105 within this interval shows a peak two-point lod score of 2 = 16.14 at 6 = 0.00 recombination and complete linkage ~sequilib~um among affected individuals. These data together with the observation of a common disease haplotype among all family ancestors support the notion of an SCA2 founder effect in Hol0 1996 Academic Press, Inc. gnin province.

MATERIALS AND METHODS

INTRODUCTION

Autosomal dominant spinocerebellar ataxias (SCAs) were first described by Marie (18931, the phenotype was investigated in detail by Schut (19501, and subclassifications were attempted on clinical-neuropathological grounds by Greenfield (1954) and on clinicalgenetic grounds by Harding (1983). Spinocerebellar ataxia 2 (SCA2) was originally described in a large collective of over 500 patients probably descended from the same ancestor in Holguin province, Cuba (Orozco et al., 1990). A phenotype with early neuropathy/late slow saccadic eye movements/late myoclonus clinically (Orozco et al., 1990) and olivo-pontocerebellar atrophy with selective sparing of the dentate nucleus neuropatholo~cally (Estrada et al., submitted) was determined. Genetic characterization of 30 pedigrees (Auburger et al., 1990; Gispert et al.,1993b) led to the localization of the SCAB mutation between the 1Present address: Institute for Hematology and Immunolo~, Apto 8070, Havanna 8, Cuba. a Present address: National Center for Medical Genetics, Calle 31, No. 3102 esq. 146, Havanna, Cuba. Telephone: ’ To whom correspondence should be addressed. c492113117803. Fax: +492113117804. 433

Family studies. Pedigrees with relatively few untyped members make rapid multipoint calculation possible. Moreover, the existence of several completely sampled meioses per family facilitates the assignment of maternal and paternal alleles and thus the construction of haplotypes. According to these criteria, I1 families (PL, Rs, Es, Ve, LB, In, Su, Al, Mu, Pu, So) with autosomal dominant cerebellar ataxia type I from Holguin province were chosen for this multipoint and haplotype analysis. Age at onset in the respective families ranged from 10 to 57, 9 to 56, 13 to 58, 11 to 55, 18 to 40, 10 to 65, 14 to 63, 24 to 58,24 to 65, 18 to 47, and 5 to 40 years. Although in general late manifestations seemed to be followed by a milder course, a survival time of 43 years was indeed documented in one case with onset at 11 years of age. Analysis for anticipation in these families as reported by Pulst et al. (1994) was significant on formal analysis, but it must be noted that the statistical result could be due to the sample bias of recruiting large pedigrees for linkage analysis, which are invariably descended from a late and mildly affected founder with many offspring. PCR and RFLP analysis. PGR amplification of microsatellites and the PAH-VNTR was performed as described previously (Weber et al., 1991; Weissenbach et al., 1992; Polymeropoulos et al., 1990; Goltsov et al., 1992). After PCR amplification of gonomic DNA (Dworniczak et al., 1991) and appropriate restriction of the product, the BglII PAH-RFLP was analyzed by horizontal electrophoresis with 0.8% agarose. Linkage analysis. Two- and four-point linkage analyses were performed using the LINKAGE program package (Ott, 1985), version 5.1. Allele frequencies were determined from 30 spouses within the families. Since no cases of incomplete penetrance or phenocopies were apparent in the 11 families on analysis of the disease haplotypes, four-point analysis was performed with two liability classes ipenetrances of 50 and 80%). For the multipoint map, marker order and distances were used as determined by the CHLC consortium (Buetow et al., 1994). Since at the common locus of D12S84 and D12S105 (distance 0 centimorgans) the latter was uninformative or

0888-754395 $6.00 Copyright 0 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.

HERNANDEZ

434

ET AL.

TABLE 1 Two-Point

Polymorphism D12S58 PAH D12S78 D12S84 D12S105 D12S79 D12S86 PLA2

Lod Scores of Individual Microsatellites in the SCA!2 Region Calculated from the Holguin Founder Population

from 11 Pedigrees

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.15

5.80 7.81 -6.22 12.32 16.14 4.18 -2.67 -3.06

7.81 10.29 -1.38 14.46 15.81 6.98 1.00 0.68

8.37 10.61 -0.16 14.64 15.48 7.31 1.99 2.31

8.66 10.70 0.52 14.61 15.15 7.41 2.54 3.35

8.81 10.68 0.97 14.47 14.81 7.39 2.89 4.08

8.88 10.61 1.29 14.29 14.47 7.33 3.13 4.62

8.89 10.49 1.52 14.06 14.13 7.22 3.30 5.02

8.86 10.35 1.70 13.81 13.79 7.09 3.42 5.32

8.79 10.18 1.83 13.54 13.44 6.94 3.50 5.55

8.68 10.00 1.93 13.25 13.09 6.77 3.55 5.71

8.56 9.80 2.00 12.95 12.74 6.60 3.58 5.83

7.67 8.65 2.09 11.30 10.94 5.61 3.45 5.89

nonrecombinant in all meioses relevant for multipoint analysis, and since a haplotype of both markers would have increased calculation times unduly, the four-point mapping of the SC42 locus was calculated with the genotypes of D12S84 only. At the PAH locus, the information from VNTR and Bgl II RFLP typings was haplotyped under the assumption of minimal recombination. The degree of allelic association was assessed by 2 x 2 comparative analysis and evaluated according to the x2 distribution.

RESULTS

The microsatellites D12S58, D12S78, D12S84, D12S105, D12S79, D12S86, and PLA2 (Weissenbach et al., 1992; Weber et al., 1991; Polymeropoulos et al., 1990) and the PAH-VNTR (Goltsov et al., 1992) and the PAH-BgZII RFLP (Dworniczak et al., 1991) were typed in 11 large pedigrees with autosomal dominant spinocerebellar ataxia from Holguin, Cuba. The previously reported linkage with markers D12S58 and PLA2 (Gispert et al., 1993a) was confirmed, showing a maximal two-point lod score at the locus D12S105 (Table 1) of 16.14 at 8,,X = 0 with a support interval of 6 = 0.03. For D12S58 Z,, equalled 8.9 at emax = 0.06 with a support interval of e1 = 0.04, BZ = 0.08, for PAH the values were Z,,, = 10.7, Omax= 0.03, 8i = 0.02, & = 0.08, for D12S78 Z,,, = 2.1, 8,, = 0.15, 8i = 0.1, & = 0.18, for D12S84 Z,,, = 14.7, 8,,, = 0.02, & = 0.01, & = 0.05, for D12S79 Z,, = 7.41, &,, = 0.03, & = 0.02, & = 0.08, for D12S86 Z,, = 3.6, 8,,, = 0.10, 0i = 0.07, & = 0.14, and for PLA2 Z,,, = 5.9,0,,, = 0.16, 8i = 0.10, & = 0.10. In none of the 11 families were negative lod scores generated with D12S105, indicating homogeneity of the patient collective (data from Homog analysis not shown). Moreover, a common ancestral haplotype was found to cosegregate with the disease in each of the families. These results support the clinical and epidemiological notion of a common descent of all Holguin SCA families from one ancestor. Within the candidate region, significant allelic association with the disease was demonstrated for markers PAH (x2 = 16.6, P < 0.001, 1 df), D12S78 (x2 = 1.41 due to association with a frequent allele, P < 0.30), D12S84 (x2 = 33.3, P < O.OOl), D12S105 (x2 = 15.8, P < O.OOl), and D12S79 (x2 = 14.6, P < 0.001) in the founders of all 11 families. Allelic

0.20 6.52 7.33 1.94 9.49 9.09 4.53 3.07 5.41

0.30 3.93 4.43 1.30 5.65 5.33 2.39 1.93 3.66

0.40 1.47 1.61 0.57 2.00 1.89 0.68 0.73 1.59

association with disease inheritance was still complete for D12S84 among the family founders, but incomplete taking into account the observation of intrafamilial crossing over; in the case of D12S105, the very same allele was found on every affected chromosome throughout the collective. Four-point linkage analysis made a localization of the SCA2 mutation in the ~-CM interval between D12S84 and D12S79 likely (Fig. l), yielding maximal multipoint lod scores of 17.6 between D12S78 and D12S84, of 19.7 between D12S84 and D12S79, and of 17.0 distal to D12S79. The positioning result is based on several recombinant events and in particular on one crossing over in the large family Rs (peak multipoint lod score of this family Z,,, = 3.49) between two family branches of five affected individuals each (Fig. 2), placing the SCAB mutation on the telomeric side of D12S84. DISCUSSION

Both the positive lod scores in each of the families and the observation of a common disease haplotype are

SCA2

FIG. 1. Four-point lod scores in the SCAP region calculated from 11 pedigrees from the Holguin population.

MAPPING

family

435

OF SCA2

Rs

REFERENCES Auburger, G., Orozco-Diaz, G., Capote, R., Gispert, S., Perez, M., Cueto, M., Meneses, M., Farrall, M., Williamson, R., Chamberlain, S., and Heredero, L. (1990). Autosomal dominant ataxia: Genetic evidence for locus heterogeneity from a Cuban founder effect population. Am. J. Hum. Genet. 46: 1163-1177.

D12S58 PAH D12S78 D12S84 D12S105 D12S79 D12S86 PLA2

57 65 22

657567 461341 522252

Buetow, K., Weber, J., Ludwigsen, S., Scherpbier-Heddema, T., Duyk, G., Sheffield, V., Wang, Z., and Murray, J. (1994). Integrated human genome-wide map constructed using the CEPH reference panel. Nature Genet. 6: 391-393. Dworniczak, B., Wedemeyer, N., and Horst, J. (1991). PCR detection of the Bgl II RFLP at the human phenylalanine hydroxylase (PAH) locus. Nucleic Acids Res. 19: 1959.

FIG. 2. Haplotype analysis in two meioses representative of two branches of the Rs family from the Holguin SCA2 population. Both the partial and the complete haplotype were typed consistently in five affected individuals of each of the two family branches.

consistent with the hypothesis of a founder effect of spinocerebellar ataxia 2 in Holguidn province, Cuba. The detection of allelic association over a region of 14 CM favors the argument that relatively few generations have passed since this population was founded by a common ancestor. If allelic association is to be of further use for the precise sublocalization of SCAB, distantly related families from other Carribbean islands or the Iberian peninsula should therefore be recruited. The analysis of haplotypes with markers in allelic association will at the same time be beneficial for presymptomatic counseling within the probable founder population. In contrast to previous results by Pulst et al., (1994) in a North American family and Lopes-Cendes et al., (1994) in an Austrian-Canadian family, where a position of the SCA2 mutation centromeric to D12S105/84 was favored on multipoint analysis only of one North American and one Austrian-Canadian family, multipoint plus haplotype analyses of 11 families in Holguin indicate a position of the SCA2 mutation telomeric to D12S84. Further data are needed to evaluate whether these conflicting data are sufficient to postulate two independent SCA gene loci on chromosome 12q23-q24.1. ACKNOWLEDGMENTS The authors stress the contribution made by the microsatellite maps of J. Weissenbach and J. Weber and acknowledge the excellent technical assistance by Dorothea Bergmann and Nicola Humbert. We thank Drs. Freund and Wechsler (University Hospital Dtisseldorf) and the Holguin Ministry of Health for continuing help. We are grateful for the collaboration of all family members. A. Hernandez and C. Magariiio were fellows of the Deutsche Heredo-Ataxie-Gesellschaft, and S. Gispert was a fellow of the Alexander von Humboldt Gesellschaft. The work was supported by grants from DFG (N.S. and A.L.), Fritz Thyssen Stiftung, and BMFT.

Gispert, S., Twells, R., Orozco, G., Brice, A., Weber, J., Heredero, L., Scheufler, K., Riley, B., Allotey, R., Nothers, C., Hillermann, R., Lunkes, A., Khati, C., Stevanin, G., Hernandez, A., Magarino, C., Klockgether, T., Durr, A., Chneiweiss, H., Enczmann, J., Farrall, M., Beckmann, J., Mullan, M., Wernet, P., Agid, Y., Freund, H.J., Williamson, R., Auburger, G., and Chamberlain, S. (1993a). Chromosomal assignment of a second locus of autosomal dominant cerebellar ataxia @CAP) to chromosome 12q23-24.1. Nature Genet. 4: 295-299. Gispert, S., Nothers, C., Orozco, G., and Auburger, G. (1993b). Search for the chromosomal location of autosomal dominant cerebellar ataxia from Holguin, Cuba: Exclusion from candidate regions on chromosome 4 and llq. Hum. Hered. 43: 12-20. Goltsov, A. A., Eisensmith, R. C., Konecki, D. S., Lichter-Konecki, U., and Woo, S. L. (1992). Associations between mutations and a VNTR in the human phenylalanine hydroxylase gene. Am. J. Hum. Genet. 51: 627-636. Greenfield, J. G. (1954). “The Spino-Cerebellar C. Thomas, Springfield, IL. Harding, A. E. (1983). Classification paraplegias. Lancet 1: 1151-1155.

Degenerations,”

of the hereditary

C.

ataxias and

Lopes-Cendes, I., Andermann, E., and Rouleau, G. (1994). Evidence for the existence of a fourth dominantly inherited spinocerebellar ataxia locus. Genomics 21: 270-274. Marie, P. (1893). 444-447.

Sur l’heredo-ataxie

cerebelleuse.

Sem. Med.

13:

Orozco, G., Fleites, A., Sagaz, R., and Auburger, G. (1990). Autosomal dominant cerebellar ataxia: Clinical analysis of 263 patients from a homogeneous population in Holguin, Cuba. Neurology 40: 13691375. Ott, J. (1985). “Analysis of Human Genetic Linkage,” Johns Hopkins Univ. Press, Baltimore, MD. Polymeropoulos, M. H., Rath, D. S., Xiao, H., and Merrill, C. R. (1990). Trinucleotide repeat polymorphism at the human pancreatic phospholipase A-2 gene (PLA21. Nucleic Acids Res. 18: 7468. Pulst, S. M., Nechiporuk, A., and Starkman, S. (1994). Anticipation in spinocerebellar ataxia type 2. Nature Genet. 5: 8-10. Schut, J. W. (1950). Hereditary ataxia: Clinical study through six generations. Arch. Neurol. Psychiatr. 63: 535-568. Weber, J. L., Kwitek, A. E., May, P. E., Wilkie, P. J., and Decker, R. A. (1991). Human Gene Mapping 11: Report of the committee on the genetic constitution of chromosome 12. Cytogenet. Cell Genet. 58: 555-579 (A26934). Weissenbach, J., Gyapay, G., Dib, C., Vignal, A., Morisette, J., Millasseau, P., Vaysseix, G., and Lathrop, M. (1992). A second generation linkage map of the human genome. Nature 359: 794-801.