Huntington Disease-Like 2: the First Patient with Apparent European Ancestry

Clin Genet 2008: 73: 480–485 # 2008 The Authors Printed in Singapore. All rights reserved Journal compilation # 2008 Blackwell Munksgaard CLINICAL GENETICS doi: 10.1111/j.1399-0004.2008.00981.x Original Article Huntington disease-like 2: the first patient with apparent European ancestry

Santos C, Wanderley H, Vedolin L, Pena SDJ, Jardim L, Sequeiros J. C Santosa, H Wanderleyb, Huntington disease-like 2: the first patient with apparent European L Vedolinb, SDJ Penac, ancestry. L Jardimb,d and J Sequeirosa,e Clin Genet 2008: 73: 480–485. # Blackwell Munksgaard, 2008 aInstituto de Biologia Molecular e Celular, b Huntington disease-like 2 (HDL2) is a rare autosomal dominant disorder Porto, Portugal, Servicxo de Gene´tica Me´dica do Hospital de Clı´nicas de Porto of the nervous system, apparently indistinguishable from Huntington Alegre, Porto Alegre, Brazil, disease (HD). HDL2 is caused by the expansion above 40 CTG/CAG cDepartamento de Bioquı´mica e repeats, in a variably spliced exon of the junctophilin-3 gene, on Imunologia, Universidade Federal de chromosome 16q24.3. All patients described so far have been of African Minas Gerais, Belo Horizonte, Brazil, ancestry. A clinical evaluation, including the Unified Huntington’s dDepartamento de Medicina Interna, Disease Rating Scale, and brain Magnetic resonance imaging were Universidade Federal do Rio Grande achieved in a 48-year-old Brazilian man of apparent European do Sul, Porto Alegre, Brazil, and eInstituto extraction, and presenting a picture very suggestive of HD. Gene de Cieˆncias Biome´dicas de Abel Salazar, mutation analysis (HD, HDL1, HDL2, dentatorubralpallidoluysian Universidade do Porto, Portugal atrophy and spinocerebellar ataxia 17) was performed. After exclusion of the HD mutation and other HDL disorders, we identified an expansion Key words: ancestry – chorea – of 47 CTG/CAG at the HDL2 locus. To clarify the origin of the DRPLA – haplotype – HDL1 – mutation and estimate the patient’s ancestry, we performed haplotype Huntington disease – SCA17 studies and used the insertion/deletion polymorphisms method. Despite the fact that this patient had an estimated likelihood of 97.4% of being of Corresponding author: Cla´udia Santos, European ancestry, the haplotype containing the expanded allele has PhD, Instituto de Biologia Molecular e Celular, R. Campo Alegre, 823, been found only in Africans. Thus, this is the first HDL2 case reported in 4150-180 Porto, Portugal. a patient with an apparent European ancestry, although bearing an Tel.: 1351 22 607 49 40; African HDL2 haplotype. This work stresses the importance of fax: 1351 22 609 91 57; performing the diagnosis of HDL2 in HD-like patients of various e-mail: [email protected] ethnicities, and particularly in highly mixed populations. Received 27 December 2007, revised and accepted for publication 28 January 2008

Huntington disease-like 2 (HDL2) (OMIM 606438) (DRPLA) and SCA17, may also have overlap- is a newly described disorder of autosomal domi- ping symptoms with HD, namely Parkinsonism, nant inheritance, the symptoms of which may dementia, chorea and dystonia (3). be undistinguishable from Huntington disease HDL2 has not yet been found among Europeans (HD) (OMIM 143100). HD is a disorder of the (4–7) or Asians (8). HDL2 is the result of an central nervous system, characterized by invol- expanded CTG/CAG repeat, in the junctophilin-3 untary choreic movements, progressive motor gene (JPH3), on chromosome 16q24.3 (6). Normal impairment, behavioural anomalies and cognitive individuals have 6–28 repeats, whereas the mu- decline. tated gene in patients contains 40–59 repeats (8). HDL2 belongs to a group of HDL disorders, The pathogenic mechanism underlying HDL2 also including HDL1 (OMIM 603218), an auto- remains elusive. Different JPH3 transcripts be- somal dominant disease caused by an extra octa- cause of alternative splicing and different polya- peptide repeat in the prion protein gene (PRNP), denylation sites have been described (6). The on chromosome 20p12 (1); and HDL3 (OMIM CTG/CAG tract may be located in intron 1 of 604802), an autosomal recessive, mapped to 4p15.3 the first isoform described (9) or within an alter- (2). Furthermore, two dominant spinocerebellar natively spliced exon (exon 2A), either untrans- ataxias (SCAs), dentatorubralpallidoluysian atrophy lated or translated to a polyleucine or to a

480 HDL2 in a patient of apparent European ancestry polyalanine stretch (6). The JPH3 gene encodes ously published (6, 13–15), with modifications to for junctophilin-3, a protein predominantly ex- permit automated fluorescent genotyping. For- pressed in the brain (9, 10). The function of ward primers were labelled at its 5# end, with the JPH3 is related to the formation of the junc- fluorochromes 6-FAM or HEX. Polymerase chain tional membrane structure in excitable cells, by reaction (PCR) was performed in a final volume interacting with the plasma membrane, through of 25 ml, containing Taq buffer 13 (Applied its cytoplasmic region and spanning the endo- Biosystems, Foster City, CA), 200 mM deoxyri- plasmic/sarcoplasmic reticulum membrane bonucleotide triphosphate (dNTPs), 1 mM through its C-terminal hydrophobic region (10). MgCl2, 400 nM each primer, 1.5 U Taq Gold The JPH3 knockout mouse shows impaired (Applied Biosystems) and 10% dimethyl sulfo- performance of motor-coordination tasks, likely xide (DMSO). Fragment separation was made because of abnormal neuronal signalling (11). using capillary electrophoresis in an ABI Despite the above evidence, the involvement of PRISM 310 Genetic Analyser; 1 ml of each the mutant JPH3 in HDL2 pathogenesis has PCR was mixed with 12 ml of deionized form- not yet been fully clarified. By contrast, a recent amide and 0.5 ml GeneScan Rox 500 labelled study showed that the CUG repeat-containing marker (Applied Biosystems). PCR products RNA foci are sufficient to cause cell toxicity in were denatured at 95°C for 2 min, prior to the HDL2 (12). run. The presence of the eight extra octapeptide In this work, we describe the first Brazilian repeats on PRNP gene was analysed by PCR, HDL2 family. Given that the Brazilian popula- followed by an agarose gel (16). tion was originally formed from three ancestral roots (Amerindians, Europeans and Africans), we decided to study the patient’s ancestry. We Haplotype study: DNA cloning and sequencing have also performed a haplotype analysis to try to determine the origin of the mutated allele. We analysed the HDL2 repetitive tract and a downstream region of 6.2 kb, using primers L237-2 5#-agatgccaccgcattcgg-3# and JPH3R 5#- # Subjects and methods gcacatggcagtgaccagcatgg-3 . From the HapMap database (www.hapmap.org), we selected four Subject single nucleotide polymorphisms (SNPs) encom- A 48-year-old male was the fourth child of non- passed within the analysed region: rs2562059, consanguineous parents. The family, which was rs1864151, rs1864152 and rs2573124 (chromo- of Brazilian origin, stated that they were not some positions 86196356, 86199494, 86199574 aware of any African ancestry, even many gener- and 86201418). These SNPs are located 920, ations before. 4059, 4140 and 5985 bp away from the end of His father, already deceased, had a similar clini- the CTG segment. In accordance to the HapMap cal picture since age 50 years, suggesting an database, these are the best SNPs in the region autosomal dominant transmission in this family. taking into account their power of discrimination All his five sibs were asymptomatic; he had three between European and African populations. The asymptomatic offspring (aged 26, 24 and rs2562059 alleles C/G have a relative frequency 22 years). The patient was neurologically exam- of 0.6/0.4 in Europeans and 0.267/0.733 in Afri- ined, and a blood sample was obtained after cans; the rs1864151 alleles A/C have frequencies informed consent. After the initial evaluation, he of 0.617/0.383 in Europeans and 0.525/0.475 in suffered a cerebral vascular accident, with sec- Africans; the rs1864152 alleles A/C have frequen- ondary infections, and died in a septic state, cies of 0.158/0.842 in Europeans and 0.208/0.792 without further neurological evaluation. in Africans; and the rs2573124 alleles C/A have frequencies of 0.602/0.398 in Europeans and 0.271/0.729 in Africans. The PCR assay was performed in a 50 mlfinal Mutation analysis volume of 13 Expand High Fidelity Buffer Genomic DNA was extracted from blood lym- (Roche Applied Science, Mannheim, Germany), phocytes by standard methods. Repeat expan- 200 mM dNTPs, 300 nM of each primer, 10% sions causing HD, as well as other related DMSO, 125 ng of genomic DNA and 2.6 U of disorders (HDL2, DRPLA and SCA17), were Expand High Fidelity enzyme mix (Roche). screened in the proband. Trinucleotide repeats in Cycling conditions were performed according the HD, JPH3, ATN1,andTBP genes were ampli- to the manufacturer’s instructions (Expand High fied using primers and cycling conditions previ- Fidelity PCR System). DNA fragments were

481 Santos et al. purified using a kit (Amersham Biosciences, a widebased, slow and interrupted by choreic Buckinghamshire, UK) after excision of DNA movements of his legs and trunk. There was agarose bands. dysarthria, motor impersistence, dysmetria, and By cloning the purified DNA fragments (6.2 dysdiadocokinesia; his ocular pursuit was inter- kb) into the pCR4-TOPO vector (Invitrogen, rupted. Deep tendon reflexes were normal. Cho- Carlsbad, CA), under the manufacturer’s instruc- rea was important in the tongue, trunk, upper tions, we obtained isolated bacterial colonies, car- and lower limbs, and mild-to-moderate rigidity rying either the normal or the expanded allele. was present in his arms. Sequence analysis was performed by cycle sequencing in an ABI PRISM 310 using T3 and T7 primers (to confirm insert orientation), as well Brain magnetic resonance imaging as primer L237-2 (to confirm the CTG repeat size); primers SNP3JPH3F 5#-agacctgtcctccagaggtg-3# The fluid attenuation inversion recovery, T1- and SNP3JPH3R 5#-tctgctttgctcagtgtggt-3# (to and T2-weighted images showed signs of brain detect SNP rs2562059); primers SNP4JPH3F 5#- atrophy, marked enlargement of the ventricles, ctctaaatatgacgagctctc-3# and SNP5JPH3R 5#-act- mild periventricular white matter lesions, and caagttcgagccaggac-3# (to detect SNPs rs1864151 bilateral atrophy of the caudate nucleus (Fig. 1). and rs1864152); and primers JPH3R and SNP6JPH3F 5#-gcttactcactcgccactcacctc-3# (to detect SNP rs2573124). Gene analysis The genetic study of the HD locus showed a normal allele with 23 CAGs and a large normal Patient’s ancestry by the INDELS’ method allele with 30 CAGs, excluding the clinical diagno- The patient was typed for a set of 40 biallelic short sis of HD. Repeat expansions in the TBP (37/38 insertion/deletion polymorphisms (INDELS). As CAA/CAG) and ATN1 (13/13 CAGs) genes were ancestral population references, previous data also excluded; we also could not find the octa- from the Human Genome Diversity Cell Line peptide PRNP expansion of 192 bp, characteris- Panel (HGDP-CEPH) was used on 161 Euro- tic of HDL1. Finally, we tested for HDL2 and peans, 126 Africans and 103 Amerindians (17). detected a CTG/CAG repeat expansion with 47 To estimate the ancestral contribution from these CTGs and a normal allele of 14 CTGs in the JPH3 three populations, we used the STRUCTURE pro- gene. gram version 2.1 (18) (available at http://pritch. bsd.uchicago.edu/software.html). This software uses multilocal genotypes to allocate ancestry Haplotype study proportions of individuals to different clusters Sequence analysis of 6.2 kb downstream from (populations). The software defines k clusters the CTG segment showed the presence of haplo- (where k is provided by the user), each of them types (CTG)14-CACC and (CTG)47-GACA. being characterized by a set of allelic frequencies Comparing these to haplotype frequencies pub- for each locus. lished in the HapMap database, we could con- clude that the block CACC is much more frequent in Europeans (42.5%) than in Africans Results (1.7%); but, the haplotype with the expanded Clinical features allele GACA has, thus far, been found only in persons of African ancestry (16.3%). In addition, The neurological examination of this 48-year-old the four intragenic SNPs (GACA) found within patient was consistent with the diagnosis of HD. the mutant allele corresponded to ancestral alleles The Unified Huntington’s Disease Rating Scale according to the SNPs database (www.ncbi.nlm. motor section (0 ¼ normal and 120 ¼ maximal nih.gov/SNP). severity) was applied; the patient’s score was 49. The patient turned listless, introspective, and with delayed response by the age of 44 years. Patient’s ancestry This was followed by slurring speech, difficulties in walking, progressive deterioration of mental The patient and his relatives are of Brazilian ori- abilities, and choreiform movements. On physi- gin. They did not refer any known African ances- cal examination, he was disoriented in time and tor. In view of previous cases of HDL2 having space, and recent memory was lost. His gait was been described exclusively in persons of definite

482 HDL2 in a patient of apparent European ancestry

Fig. 1. Brain magnetic resonance imaging of the patient. Coronal T2- weighted image shows severe caudate head atrophy (a); axial T2-weighted images show diffuse basal ganglia atrophy, white matter lesions and brain atrophy (b, c and d).

or possible African origin, we tried to gain further nucleus, indistinguishable from those observed in insight into the patient’s ancestry. After typing HD. A genetic test excluded an HD expansion. the 40-indel set, the software estimated a propor- At our laboratory, only about 4% of cases clini- tion of European, Amerindian and African ances- cally diagnosed as HD do not carry the HD try of 97.4% (95% CI, 74.3–100.0%), 1.9% (95% mutation (19). Other groups reported HDL CI, 0.0–23.4%) and 0.7% (95% CI, 0.0–9.6%). cases, in their series, to be between 1% (13) and 7% (5). After exclusion of other HDL disorders, we eventually found a 47 CTG/CAG repeat expansion at the JPH3 gene; affected individuals Discussion usually have 40–59 CTG/CAG repeats (6, 8). We identified the first patient with HDL2 re- The expanded allele had most probably been portedly of European ancestry, thus presenting inherited from his father, who died with similar evidence that the JPH3 mutation may also be symptoms, without being genetically tested. Age found in patients without apparent African origin. of onset was 44 years in agreement with the cor- The clinical presentation of this 48-year-old relation with repeat length of HDL2 (8). Brazilian male included choreic movements, Twenty-five HDL2 families described to this rigidity, dysarthria, dementia and behavioural date had definite or possible African ancestry anomalies, as observed before in other HDL2 (12). The pedigree in which the mutation was patients. Magnetic resonance imaging showed first detected was of African-American origin (6, signs of brain atrophy, marked enlargement of 20). Then, other African-American families were the ventricles, mild periventricular white matter described (8, 21, 22): a Moroccan woman, from lesions and bilateral atrophy of the caudate an area predominantly populated by individuals

483 Santos et al. of sub-Saharan origin, was found in an Euro- a recipient of a scholarship from IBMC – Instituto de Biologia pean collection (5); the Mexican pedigree (8, 22) Molecular e Celular. originated from a region colonized by Africans; in an European group of 252 HDL patients, one HDL2 case was a black person from the French References West Indies (23); and at least 15 families were 1. Moore RC, Xiang F, Monaghan J et al. Huntington disease from South Africa, where the prevalence of phenocopy is a familial prion disease. Am J Hum Genet HDL2 has been described to be higher than in 2001: 69: 1385–1388. 2. Kambouris M, Bohlega S, Al-Tahan A, Meyer BF. any other country (24). HDL2 accounts only for Localization of the gene for a novel autosomal recessive 1% of the HDL patients in North America and neurodegenerative Huntington-like disorder to 4p15.3. Am has not been found in Japan (8). Thus, it has J Hum Genet 2000: 66: 445–452. been proposed that the diagnosis of HDL2 3. The American College of Medical Genetics ASHG, should be considered preferentially in persons of Huntington Disease Genetic Testing Working Group. African origin. The higher frequency of the Laboratory guidelines for Huntington disease genetic testing. Am J Hum Genet 1998: 62: 1243–1247. HDL2 mutation in African people could be ex- 4. Bauer I, Gencik M, Laccone F et al. Trinucleotide repeat plained by a common ancestor or a tendency for expansions in the junctophilin-3 gene are not found in further expansion in a specific genetic back- Caucasian patients with a Huntington’s disease like pheno- ground. Very recently, the first person with type. Ann Neurol 2002: 51: 662. HDL2 and middle-eastern origin (from Kuwait) 5. Stevanin G, Camuzat A, Holmes E et al. CAG/CTG repeat expansions at the Huntington’s disease-like 2 locus are was described, suggesting a possible second rare in Huntington’s disease patients. Neurology 2002: 58: founder event (25); however, the African ances- 965–967. try of this patient was not completely excluded. 6. Holmes SE, O’Hearn E, Rosenblatt A et al. A repeat In our patient, the haplotype carrying the nor- expansion in the gene encoding junctophilin-3 is associated with Huntington disease-like 2. Nat Genet 2001: 29: 377– mal allele (CTG)14-CACC is much more frequent in Europeans (42.5%) than in Africans (1.7%); 378. 7. Costa MC, Teixeira-Castro A, Constante M et al. Exclu- nevertheless, the haplotype with the expanded sion of mutation in the PRNP, JPH3, TBP, ATN1, allele (CTG)47-(GACA) has only been found in CREBBP, POU3F2 and FTL genes as a cause of disease persons of African ancestry (16.3%). in Portuguese patients with Huntington-like phenotype. J Because our patient was Brazilian, it was diffi- Hum Genet 2006: 51: 645–651. cult to determine the exact family ancestry. The 8. Margolis RL, Holmes SE, Rosenblatt A et al. Huntington’s patient and his relatives, however, considered disease-like 2 (HDL2) in North America and Japan. Ann Neurol 2004: 56: 670–674. themselves as Caucasians and did not refer any 9. Nishi M, Mizushima A, Nakagawara K-I, Takeshima H. African ancestor. Confirming this assumption, Characterization of human junctophilin subtype genes. the INDELS method showed that the likelihood Biochem Biophys Res Commun 2000: 273: 920–927. of his ancestry was 97.4% European, 1.8% 10. Takeshima H, Komazaki S, Nishi M, Lino M, Kangawa K. Amerindian and 0.7% African. There is, how- Junctophilin: a novel family of junctional membrane ever, an important caveat: these estimates were complex proteins. Mol Cell 2000: 6: 11–22. 11. Nishi M, Hashimoto K, Kurimyama K et al. Motor based on a sample of only 40 loci, and the 95% discoordination in mutant mice lacking junctophilin type 3. confidence limits for African ancestry extended Biochem Biophys Res Commun 2002: 292: 318–324. from 0% to 9.6%. The fact that he carries a hap- 12. Rudnicki D, Holmes SE, Lin MW, Thornton CA, Ross C, lotype exclusively described in persons of African Margolis RL. Huntington’s disease-like 2 is associated with origin shows, however, he should have at least CUG repeat-containing RNA foci. Ann Neurol 2007: 61: 272–282. one African ancestor. 13. Andrew SE, Goldberg YP, Kremer B et al. Huntington This case shows that the diagnosis of HDL2 disease without CAG expansion: phenocopies or errors in should always be considered, whenever HD has assignment? Am J Hum Genet 1994: 54: 852–863. been excluded and the clinical presentation may 14. Koide R, Ikeuchi T, Onodera O et al. Unstable expansion suggest it, regardless of the apparent origin of of CAG repeat in hereditary dentatorubral-pallidoluysian the family, and particularly so in populations of atrophy (DRPLA). Nat Genet 1994: 6: 9–13. 15. Nakamura K, Jeong SY, Uchihara T et al. SCA17, a novel mixed ancestry. It would be interesting now to autosomal dominant cerebellar ataxia caused by an compare ancestral haplotypes among families expanded polyglutamine in TATA-binding protein. Hum with HDL2. Mol Genet 2001: 10: 1441–1448. 16. Beck JA, Mead S, Campbell TA et al. Two-octapeptide repeat deletion of prion protein associated with rapidly Acknowledgements progressive dementia. Neurology 2001: 57: 354–356. 17. Bastos-Rodrigues L, Pimenta JR, Pena SDJ. The genetic The authors wish to thank Paula Magalha˜es for her technical structure of human populations studied through short support. We also wish to thank Maria do Carmo Costa and insertion-deletion polymorphisms. Ann Hum Genet 2006: Sandra Martins for their valuable comments. Cla´udia Santos is 70: 658–665.

484 HDL2 in a patient of apparent European ancestry

18. Pritchard JK, Stephens M, Donnelly P. Inference of 22. Walker RH, Rasmusssen A, Rudnicki D et al. Hunting- population structure using multilocus genotype data. ton’s disease-like 2 can present as chorea-acanthocytosis. Genetics 2000: 155: 945–959. Neurology 2003: 61: 1002–1004. 19. Costa MC, Magalha˜es P, Ferreirinha F et al. Molecular 23. Stevanin G, Fujigasaki H, Lebre AS et al. Huntington’s diagnosis of Huntington disease in Portugal: implications disease-like phenotype due to trinucleotide repeat expansions for genetic counselling and clinical practice. Eur J Hum in the TBP and JPH3 genes. Brain 2003: 126: 1599–1603. Genet 2003: 11: 872–878. 24. Krause A, Hetem C, Holmes E, Margolis RL. Genetic he- 20. Margolis RL, O’Hearn E, Rosenblatt A et al. A disorder terogeneity in South African patients with the Huntington di- similar to Huntington’s disease is associated with a novel sease phenotype. Am J Hum Genet 2004: 161: 797/F3 (Abstract). CAG repeat expansion. Ann Neurol 2001: 50: 373–380. 25. Wild EJ, Mudanohwo EE, SweeneyMG et al. Huntington’s 21. Walker RH, Jankovic J, O’Hearn E, Margolis RL. disease phenocopies are clinically and genetically hetero- Phenotypic features of Huntington’s disease-like 2. Mov geneous. Mov Disord Published Online 7 January 2008 Disord 2003: 18: 1527–1530. doi: 10.1002/mds.21915.

485