Clinical and Genetic Study of Autosomal Recessive Cerebellar Ataxia Type 1 Nicolas Dupré, MD, MSc,1,2 François Gros-Louis, PhD,2 Nicolas Chrestian, MD,1 Steve Verreault, MD, MSc,1 Denis Brunet, MD,1 Danielle de Verteuil, BSc,3 Bernard Brais, MD, PhD,3 Jean-Pierre Bouchard, MD,1 and Guy A. Rouleau, MD, PhD2 Objective: Define the phenotype and genotype of a cluster of families with a relatively pure cerebellar ataxia referred to as autosomal recessive cerebellar ataxia type 1 (ARCA-1). Methods: We ascertained 64 probands and affected members of 30 French-Canadian families all showing similar clinical features and originating from the same region of Quebec. After informed consent, we performed detailed clinical history, neurological examination, brain imaging, nerve conduction studies, and SYNE1 mutation detection of all available subjects. Results: Based on the cases examined, ARCA-1 is a cerebellar syndrome characterized by recessive transmission, middle-age onset (mean, 31.60; range, 17– 46 years), slow progression and moderate disability, significant dysarthria, mild oculomotor abnormalities, occasional brisk reflexes in the lower extremities, normal nerve conduction studies, and diffuse cerebellar atrophy on imaging. We identified a total of seven mutations in our population, thereby providing evidence of genotypic heterogeneity. Patients with different mutations did not show significant phenotypic heterogeneity. Interpretation: We identified a cluster of French-Canadian families with a new recessive ataxia of relatively pure cerebellar type caused by mutations in SYNE1. The function of SYNE1 is thus critical in the maintenance of cerebellar structure in humans. We expect that this disease will be a common cause of middle-age-onset recessive ataxia worldwide. Ann Neurol 2007;62:93–98 The hereditary ataxias can be divided based on their mode of inheritance into autosomal dominant, autosomal recessive, X-linked, and mitochondrial ataxias. These disease categories share the prototypic feature of impaired walking, although they usually present a variety of other neurological symptoms such as pyramidal signs, peripheral neuropathy, extrapyramidal signs, cognitive loss, or retinopathy. The autosomal recessive ataxias are also a heterogeneous group of disorders composed mainly of Friedreich’s ataxia, ataxia telangiectasia, ataxia with vitamin E deficiency, autosomal recessive spastic ataxia of Charlevoix–Saguenay (ARSACS), abetalipoproteinemia, and ataxia with oculomotor apraxia types 1 and 2.1 Over the past decade, we have identified a large cluster of French-Canadian families whose ancestors originate mostly from the same region of the Province of Quebec in Canada. This region, located in southeastern Quebec near the US border, is called Beauce. The affected individuals in these families all share similar clinical characteristics that define this new disease entity we named ARCA-1, also known as recessive ataxia of Beauce.2,3 Genome-wide linkage and fine-mapping analysis on selected families with ARCA-1 established a minimum candidate interval of about 0.5Mb on chromosome 6q containing a single causal gene (SYNE1).4 Spanning over 0.5Mb genomic DNA, SYNE1 is one of the biggest genes in the human genome formed of 147 exons that encodes a 27,652kb messenger RNA and an 8,797 amino-acid-long protein. To date, five different truncating mutations within SYNE1 have been described.4 In this article, we report two novel mutations, also leading to premature termination of the protein, and define the full clinical and molecular spectrum of ARCA-1. From the 1Faculty of Medicine, Laval University, Department of Neurological Sciences, Centre Hospitalier Affilié Universitaire de Québec–Enfant-Jésus, Quebec City; 2Center for the Study of Brain Diseases, Université de Montréal, Centre Hospitalier de l’Université de Montréal (Notre-Dame); and 3Laboratory of Neurogenetics of Motion, Center for the Study of Brain Diseases, Université de Montréal, Centre Hospitalier de l’Université de Montréal (Notre-Dame), Montréal, Québec, Canada. Received Dec 22, 2006, and in revised form Feb 8, 2007. Accepted for publication Mar 2, 2007. Subjects and Methods Subjects were referred to the study protocol by their treating physician based on a preliminary assessment consistent with the core features of ARCA-1 (ataxia and dysarthria of middle-age onset with cerebellar atrophy) and a recessive family history (parents unaffected). They were informed of the procedures entailed in the protocol and signed, before their participation, a consent form approved by the local eth- Published online May 14, 2007, in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.21143 Address correspondence to Dr Dupre, Department of Neurological Sciences, CHAUQ–Enfant-Jésus, 1401, 18th Street, Quebec City, QC, Canada, G1J 1Z4. E-mail: firstname.lastname@example.org © 2007 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 93 ics review boards (Centre Hospitalier de l’Université de Montréal and Centre Hospitalier Affilié Universitaire de Québec). All 64 available affected and unaffected members of the 30 recruited families underwent a thorough neurological examination and were examined independently by at least two neurologists. We also used an assessment scale5 to grade severity of symptoms: dysarthria (0 ⫽ no impairment; 1 ⫽ mild dysarthria but comprehensible; 2 ⫽ moderate dysarthria with interruption in flow; 3 ⫽ severe dysarthria and incomprehensible, very difficult to understand; 4 ⫽ completely unintelligible); dysmetria (0 ⫽ no impairment; 1 ⫽ mild dysmetria but reaches the target; 2 ⫽ moderate dysmetria, reaches target after several attempts; 3 ⫽ severe dysmetria, short of target after many attempts; 4 ⫽ cannot use hands); gait (0 ⫽ normal; 1 ⫽ stance width increased, mildly unstable gait but can walk without support; 2 ⫽ moderately unstable gait and needs support for walking; 3 ⫽ unable to walk, needs the assistance of two persons; 4 ⫽ wheelchair bound). Twenty-two subjects underwent electrophysiological studies including compound motor action potentials of median, ulnar, tibial, and peroneal nerves, as well as sensory nerve action potentials of median, ulnar, radial, and sural nerves, with standard values for filters, stimulus duration, and electrode positioning. Brain imaging with magnetic resonance imaging was performed on 50 affected subjects. On receipt of informed consent, blood samples were obtained from affected individuals in 30 families. DNA was extracted from peripheral blood by standard methods.6 For mutation screening, a set of 154 polymerase chain reaction (PCR) primer pairs were designed from genomic DNA to amplify each exon of the SYNE1 gene, including the flanking splice sites and the untranslated regions.4 Products were PCR-amplified, checked on agarose gels, and then sequenced using the forward primers for all of the amplicons. Each fragment containing mutations was PCR-amplified a second time and sequenced with the reverse primer to confirm that the identified mutations were not due to PCR artifact. Results Sample Case This 42-year-old woman originates from the Beauce region in Quebec. On her initial visit, she mentioned that around the age of 30 she started noticing that her speech was slurred. During the same period, she also noticed some mild walking impairment. As she advanced in her thirties, the slurred speech and walking impairment progressed. Strangers would sometimes wonder if she had been drinking alcohol, and her gait was becoming more wide based with a tendency to fall if she did not pay more attention than usual during her movements. In addition, her hands sometimes felt clumsier. On neurological examination at age 42, she had normal funduscopy, normal ocular saccades and pursuit, and no nystagmus. She had significant cerebellar dysarthria. Strength was normal throughout, with no spasticity. Reflexes were normal in the upper and lower limbs, with down-going toes. Sensory examination was normal to pain, temperature, vibration, pro- 94 Annals of Neurology Vol 62 No 1 July 2007 prioception, and light touch. Gait was broad based. There was significant dysmetria on finger-to-nose and heel-to-shin, as well as cerebellar hypotonia (Holmes’ sign) in the upper limbs. Magnetic resonance imaging demonstrated diffuse cerebellar atrophy, and nerve conduction studies were normal. Genetic testing confirmed that she was homozygous for the g.306434A⬎G mutation. Overall, the ARCA1 phenotype consists of a middleage onset disease that presents with either dysarthria, cerebellar ataxia, or both coincidentally (Table 1 and Fig 1). Over time, all patients develop significant dysarthria and ataxia, with other associated features such as dysmetria, brisk lower extremity tendon reflexes, and minor abnormalities in saccade and smooth pursuit. None of the subjects evaluated showed extrapyramidal signs, cognitive loss, retinopathy, cardiomyopathy, sensory abnormalities, or autonomic disturbances. The disease progresses slowly and evolves into a moderate degree of disability. There appears to be no effect on life expectancy. Nerve conduction studies performed on 22 affected individuals were always normal, showing therefore no sign of a peripheral sensory or motor neuropathy in this disease. Single-fiber electromyography was also normal when performed on one subject, suggesting preserved function at the neuromuscular junction. Imaging findings on 50 affected individuals invariably showed marked diffuse cerebellar atrophy (Fig 2). On detailed review of imaging in 38 affected subjects, there was no cerebral cortical atrophy; no midbrain, pontine, or bulbar atrophy; no atrophy of inferior olives; and no white matter changes. Genetic analysis demonstrated seven mutations (Table 2 and Fig 3), including five previously described mutations4 and two novel truncating mutations (g.409218C⬎T and g.281100-281101delTG). The most frequent mutation, g.306434A⬎G, was present homozygously in 20 of 64 subjects (31.2%). This same mutation was present heterozygously with the g.310067A⬎G mutation in 11 of 64 (17.2%), with the g.247012A⬎T mutation in 6 of 64 (9.4%), with the g.426494C⬎T mutation in 1 of 64 (1.6%), and with the g.334338-334342delATTTG mutation in 4 of 64 (6.2%). Of these 64 subjects, there was 1 (1.6%) g.310067A⬎G homozygote, 3 (4.7%) g.247012A⬎T homozygotes, and 1 (1.6%) g.247012A⬎T/ g.409218C⬎T heterozygote. Finally, 6 of 64 (9.4%) carried a known mutation on one chromosome and an unknown mutation on the other chromosome, whereas in 9 (14.0%) the mutation was not found using the technique described earlier and in 2 (3.1%) genetic testing could not be performed. The g.306434A⬎G mutation was present on 50.8% of chromosomes tested, whereas mutations remain to be identified in close to one-fifth of carrier chromosomes. We performed genotype–phenotype correlation studies by seg- Table 1. Clinical Results of Patients with ARCA-1 (n ⴝ 64) Sex Male: 37 (58%) Female: 27 (42%) First complaint Dysarthria: 8 (12.5%) Ataxia: 40 (62.5%) Both: 16 (25%) Age of onset (dysarthria), yr Mean: 34.79 SD: 7.62 Range: 17–50 Age of onset (ataxia), yr Mean: 31.60 SD: 7.81 Range: 17–45 Age at evaluation, yr Mean: 45.36 SD: 10.71 Range: 24–69 Duration at evaluation, yr Mean: 14.33 SD: 9.77 Range: 3–40 Dysarthria Mean: 1.78 Range: 0–3 Dysdiadochokinesis Mean: 1.26 Range: 0–2 Dysmetria Mean: 1.44 Range: 0–3 Gait Mean: 1.48 Range: 0–3 Assessment scale Neurological examination Dysarthria 64 (100%) Nystagmus 6 (9.4%) Slow saccades 20 (31.2%) Abnormal pursuit 28 (43.8%) Brisk reflexes in the lower limbs 21 (32.8%) Ankle clonus and/or Babinski sign Dysmetria on finger-to-nose 4 (6.2%) 58 (90.6%) Dysmetria on heal-to-shin 58 (90.6%) Ataxia 63 (98.4%) SD ⫽ standard deviation. regating based on the most common genotypes (g.306434A⬎G homozygotes and g.306434A⬎G/ g.310067A⬎G heterozygotes) using the following parameters: age of onset (dysarthria, ataxia, overall), disease duration, and eye movement abnormalities. This analysis gave no statistically significant differences. In Fig 1. Age of onset by decade. addition, when we analyzed the clinical data of patients bearing the less common genotypes, we were unable to demonstrate that they showed any atypical clinical features. Discussion ARCA-1 is thus a new recessive relatively pure cerebellar ataxia that is caused by various mutations in SYNE1. ARCA-1 shows relative homogeneity of the phenotype, despite being caused by more than seven different mutations. The age of onset does not vary significantly in function of given mutations like in trinucleotide repeat disorders, whereas the disease shows little associated features accompanying the core symptoms of dysarthria and dysmetria. SYNE1 encodes a protein of about 8,797 amino acid residues (⬎1,000kDa).4 The protein contains two N-terminal actin-binding regions that comprise tandem paired calponin-homology domains, a transmembrane domain, multiple spectrin repeats, and a C-terminal Klarsicht domain. Although SYNE1 is expressed in multiple tissues, its greatest level in the central nervous system of mice is in the cell bodies of the Purkinje cells Dupré et al: SYNE1-Related Cerebellar Ataxia 95 Fig 2. Magnetic resonance imaging of a 43-year-old autosomal recessive cerebellar ataxia type 1 (ARCA-1) patient after 5 years of disease evolution. Sagittal T1 shows marked diffuse cerebellar atrophy with no cerebral cortical atrophy and no midbrain, pontine, or bulbar atrophy. and in neurons of the olivary region of the brainstem, whereas in humans it is also expressed predominantly in the cerebellum; it is not expressed in glial cells. In the peripheral nervous system, SYNE1 is involved in anchoring specialized myonuclei underneath the neuromuscular junctions.4 It was found in a muscle biopsy of an ARCA-1 patient that fewer myonuclei come to lie beneath the neuromuscular junction, although this has no consequences clinically, electrophysiologically, or ultrastructurally. SYNE1 is part of the spectrin family of structural proteins that share a common function of linking the plasma membrane to the actin cytoskeleton. This family also includes dystrophin (Duchenne’s and Becker’s muscular dystrophies),7 SPTBN2 (spinocerebellar ataxia type 5),8 PLEKHG4 (16q-autosomal dominant cerebellar ataxia),9 and Spnb4.10 The closest description of ARCA-1 phenotype in the literature is Holmes’ hereditary ataxia. Holmes’ type of hereditary ataxia11 was described a century ago (1907) in a family of eight siblings, with four of them presenting with middle-age-onset dysarthria, ataxia, and hypogonadism (not present in ARCA-1). The inheritance pattern was most likely autosomal recessive, because the parents were not affected. Autopsy of one affected case showed diffuse cerebellar atrophy with no pontine or olivary involvement. To our knowledge, no linkage or gene defect responsible for this type of ataxia has been reported as yet. ARSACS is the most common of all autosomal recessive ataxias in Quebec with more than 300 affected individuals. ARSACS patients exhibit early-onset signs of spasticity in the lower limbs usually observed at gait initiation (12–18 months).12 The clinical picture noticed by parents from early childhood is always that of a gait ataxia with a tendency to fall. Nerve conduction studies demonstrate signs of progressive axonal sensorimotor neuropathy. Ataxia with oculomotor apraxia type 2 is also present in Quebec, where more than 10 families have been described. It is characterized mainly by cerebellar atrophy, axonal sensorimotor neuropathy, and increased serum ␣-fetoprotein.13 The main differences clinically between these other recessive ataxias common in Quebec and ARCA-1 is the earlier age of onset, the greater degree of disability, and the associated peripheral neuropathy. Of interest, 16q-autosomal dominant cerebellar ataxia is characterized by an age of onset older than 55 years and sensorineuronal hearing impairment,9 which Table 2. Known Mutations Causing ARCA-1 Variantsa Exons/Introns Protein Changes Total Carrier Chromosomes Tested in the Patient Population (n ⴝ 124) g.306434A⬎G Intron 81 Premature stop at position 5244 63 (50.8%) g.310067A⬎G Intron 84 Premature stop at position 5402 13 (10.5%) g.247012A⬎T Exon 56 R2906X 16 (12.9%) g.426494C⬎T Exon 126 Q7640X 2 (1.6%) g.334338-334342delATTTG Exon 93 Premature stop at position 5880 4 (3.2%) g.409218C⬎T Exon 118 Q7386X 1 (0.8%) g.281100-281101delTG Exon 71 4077X 1 (0.8%) Unknown 24 (19.3%) a Variants were named according to the genomic DNA sequence NM_033071; nucleotide “A” from the ATG initiation codon is referred as 1. Allele frequencies of variants were 0 of 380 French-Canadian control chromosomes. 96 Annals of Neurology Vol 62 No 1 July 2007 Fig 3. SYNE1 identified truncating mutations. (A) Sequence traces of healthy individual (top) and ARCA-1 patients (bottom) showing the novel detected mutations within exons 71 and 118 of SYNE1. (B) Protein structure of SYNE-1. Known ARCA-1 disease causing mutations are in black, and newly identified mutations lie within boxes. Light gray region defines areas rich in predicted spectrin repeats; gray boxes at the N-terminal part of the protein correspond to calponin-homology domains involved in actin binding; dark gray box indicates the C-terminal nuclear envelope binding domains containing sequences homologous to the Drosophila protein Klarsicht. is different from what we found in ARCA-1. However, there are similarities in that it is also a relatively pure cerebellar syndrome caused by mutations in PLEKHG4, also part of the spectrin family of structural proteins. On pathology, 16q-autosomal dominant cerebellar ataxia shows peculiar degeneration of Purkinje cells that undergo shrinkage and are surrounded by amorphous material. Spinocerebellar ataxia type 5 is characterized by a slowly progressive cerebellar syndrome beginning mostly in the third decade.14,15 The most consistent clinical feature is downbeat nystagmus, whereas other common features included gait, stance, and limb ataxia; dysarthria; intention tremor and resting tremor; impaired smooth pursuit; and gaze-evoked nystagmus. Symptom progression is slow, and all patients remain ambulatory despite disease duration of up to 30 years. Magnetic resonance imaging shows atrophy of the cerebellar vermis and hemispheres. Again, this other ataxia caused by mutations in a spectrin family protein shows striking similarities with ARCA-1 for the predominant cerebellar involvement, middle-age onset, relatively slow progression, and moderate degree of disability. In conclusion, ARCA-1, spinocerebellar ataxia type 5, and 16q-autosomal dominant cerebellar ataxia taken together allow us to define a new category of hereditary ataxias related to the spectrin family of structural proteins. Despite significant genetic heterogeneity, this category of ataxia shares many common clinical features. We expect that this new category of inherited ataxias may be more frequent than previously thought, mainly through the contribution of ARCA-1, because we have encountered important genetic heterogeneity even within a homogeneous founder population. We speculate that a significant proportion of yet undiagnosed recessive or “sporadic” ataxias may be due to SYNE1 mutations, which would have great repercussions on our ability to diagnose more precisely these ataxia types in specialized clinics worldwide. This work was supported by the National Ataxia Foundation, (G.A.R.), the Canadian Genetic Disease Network, (G.A.R.), the Canadian Institute of Health Research (F.G.L., N.D.) and the Association des Ataxies Familiales (D.V.). F. Gosselin and M. Plante performed blood collection of patients and obtained their consent. Finally, we also thank the family members who participated in this study. Dupré et al: SYNE1-Related Cerebellar Ataxia 97 References 1. Di Donato S, Gellera C, Mariotti C. The complex clinical and genetic classification of inherited ataxias. II. Autosomal recessive ataxias. Neurol Sci 2001;22:219 –228. 2. Dupré N, Bouchard JP, Verreault S, et al. Recessive ataxia of the Beauce, a new form of hereditary ataxia of pure cerebellar type. Neurology 2002;58:A35. 3. Dupre N, Gros-Louis F, Verreault S, et al. Recessive ataxia of the Beauce, a new form of hereditary ataxia, maps to chromosome 6. Neurology 2006;66:A274. 4. Gros-Louis F, Dupre N, Dion P, et al. Mutations in SYNE1 lead to a newly discovered form of autosomal recessive cerebellar ataxia. Nat Genet 2007;39:80 – 85. 5. Alusi SH, Worthington J, Glickman S, Bain PG. A study of tremor in multiple sclerosis. Brain 2001;124:720 –730. 6. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 7. Koenig M, Hoffman EP, Bertelson CJ, et al. Complete cloning of the Duchenne muscular dystrophy (DMD) cDNA and preliminary genomic organization of the DMD gene in normal and affected individuals. Cell 1987;50:509 –517. 8. Ikeda Y, Dick KA, Weatherspoon MR, et al. Spectrin mutations cause spinocerebellar ataxia type 5. Nat Genet 2006;38:184 –190. 98 Annals of Neurology Vol 62 No 1 July 2007 9. Ishikawa K, Toru S, Tsunemi T, et al. An autosomal dominant cerebellar ataxia linked to chromosome 16q22.1 is associated with a single-nucleotide substitution in the 5⬘ untranslated region of the gene encoding a protein with spectrin repeat and Rho guanine-nucleotide exchange-factor domains. Am J Hum Genet 2005;77:280 –296. 10. Parkinson NJ, Olsson CL, Hallows JL, et al. Mutant betaspectrin 4 causes auditory and motor neuropathies in quivering mice. Nat Genet 2001;29:61– 65. 11. Holmes G. A form of cerebellar degeneration of the cerebellum. Brain 1907;30:466 – 489. 12. Bouchard JP, Richter A, Mathieu J, et al. Autosomal recessive spastic ataxia of Charlevoix-Saguenay. Neuromuscul Disord 1998;8:474 – 479. 13. Duquette A, Roddier K, McNabb-Baltar J, et al. Mutations in senataxin responsible for Quebec cluster of ataxia with neuropathy. Ann Neurol 2005;57:408 – 414. 14. Stevanin G, Herman A, Brice A, Durr A. Clinical and MRI findings in spinocerebellar ataxia type 5. Neurology 1999;53: 1355–1357. 15. Burk K, Zuhlke C, Konig IR, et al. Spinocerebellar ataxia type 5: clinical and molecular genetic features of a German kindred. Neurology 2004;62:327–329.