lable at ScienceDirect

Neurobiology of Aging xxx (2014) 1e7

Contents lists avai

Neurobiology of Aging

journal homepage: www.elsevier .com/locate/neuaging

Linkage analysis and whole-exome sequencing exclude extramutations responsible for the parkinsonian phenotype ofspinocerebellar ataxia-2

Chaodong Wang a,b,c,d,1, Yanming Xu e,1, Xiuli Feng f, Jinghong Ma a,b,c, Shu Xie f,Yanli Zhang a, Bei-Sha Tang g, Piu Chan a,b,c,h, i,*

aDepartment of Neurobiology, Xuanwu Hospital of Capital Medical University, Beijing, ChinabDepartment of Neurology, Beijing Institute of Geriatrics, Xuanwu Hospital of Capital Medical University, Beijing, ChinacBeijing Institute for Brain Disorders Parkinson’s Disease Center, Beijing, ChinadDepartment of Neurology, The Affiliated Sanming First Hospital of Fujian Medical University, Sanming, Fujian, ChinaeDepartment of Neurology, West China Hospital of Sichuan University, Chengdu, ChinafNational Human Genome Center in Beijing, Beijing, ChinagDepartment of Neurology, Xiangya Hospital of Central South University, Changsha, ChinahKey Laboratory on Neurodegenerative Disease of Ministry of Education, Beijing, ChinaiKey Laboratory on Parkinson’s Disease of Beijing, Beijing, China

a r t i c l e i n f o

Article history:Received 23 July 2014Accepted 27 July 2014

Keywords:Spinocerebellar ataxia-2ParkinsonismLinkage analysisWhole-exome sequencingMutation

* Corresponding author at: Department of NeurobCapital Medical University, #45 Changchun Street, 10010 83198677; fax: þ86 10 83161294.

E-mail address: pbchan90@gmail.com (P. Chan).1 These authors contributed equally to this work.

0197-4580/$ e see front matter � 2014 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.neurobiolaging.2014.07.039

a b s t r a c t

CAG expansion within the exon 1 of ataxin-2 (ATXN2) gene responsible for spinocerebellar ataxia-2(SCA2) has been reported to cause pure parkinsonism and other neurodegenerative disorders. Howev-er, it remains unclear whether CAG expansion is the only cause for SCA2 and its clinical alternatives, andwhether extra mutations exist to modify the phenotypic diversity. To address this, we have conductedfine genetic mapping and exome sequencing for a large Chinese SCA2 pedigree predominantly mani-festing parkinsonism (called SCA2-P). In addition, we compared the CAG expansions between the SCA2-Pand 16 SCA2 families presenting as pure ataxia (SCA2-A). As a result, CAG repeat expansions, rangingfrom 37 to 40 copies, were detected among 10 affected and 8 nonsymptomatic members of the SCA2-Pfamily. The CAG repeats in the diseased alleles were interrupted by CAA in the 30-end. In contrast, CAGexpansion ranging from 36 to 54 without CAA interruption was detected in all probands of the SCA2-Afamilies. Genetic mapping located the SCA2-P pedigree on 12q24.21, which spans the ATXN2 gene. Exomesequencing for 3 patients and 1 normal member revealed no extra mutations in this family. In addition,by genotyping single-nucleotide polymorphisms around SCA2 locus, we have excluded the existence ofhaplotypes predisposing different patterns of CAG expansion. These results demonstrate that the ATXN2CAG expansion is the sole causative mutation responsible for SCA2-P, and that genetic modifiers may notbe the major cause of the phenotypic diversity of SCA2.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

Mutations in several genes (SNCA, LRRK2, VPS35, EIF4G1, and soforth) have been identified in patients with autosomal dominantParkisonsim (ADP) (Chartier-Harlin et al., 2011; Paisan-Ruiz et al.,2004; Polymeropoulos et al., 1997; Zimprich et al., 2011).

iology, Xuanwu Hospital of053 Beijing, China. Tel.: þ86

ll rights reserved.

However, such mutations have been proven to be rare in Chinese(Chan et al., 2000, 2013; Lin et al., 2008; Zhang et al., 2012).Although other mutations (G2385R and K616R) in LRRK2 gene havebeen reported in Chinese ADP (Guo et al., 2013; Wang et al., 2010),these reports have not been verified extensively, raising the need toscreen other causative mutations in this population.

Over the recent years, trinucleotide expansion, (CAG)n,responsible for a number of autosomal dominant cerebellarataxias (also called spinocerebellar ataxias, SCAs), has beenrecognized as a cause of parkinsonism in some families, especiallyin those with spinocerebellar ataxia-2 (SCA2) and SCA3 (Furtadoet al., 2004; Lim et al., 2006; Wang et al., 2009; Wu et al., 2004).Genetic analyses revealed that SCA2-related Parkinsonism

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e72

(SCA2-P) is associated with relatively mild CAG repeat expansionsranging from 33 to 43 in the exon 1 of the ataxin-2 (ATXN2) gene,which are interrupted by a CAA interruption, whereas those pre-senting as pure ataxia (SCA2-A) patients harbor a high-rangeexpansion from 32 to over 200 without the CAA interruption(Charles et al., 2007). More recent studies have demonstrated thatmoderate expanded ATXN2 repeats are also a genetic risk factorfor multiple other neurodegenerative diseases such as amyo-trophic lateral sclerosis, frontotemporal lobar degeneration, Alz-heimer’s disease, progressive supranuclear palsy, and multisystematrophy, and, more importantly, the expanded repeat alleles in aselection of such patients can also show CAA interruptions (Liuet al., 2013; Ross et al., 2011; Van Langenhove et al., 2012). How-ever, it is unclear why CAG mutations in the same gene thatencode the identical mutant protein can lead to considerablyvariable clinical manifestations.

To resolve the mechanisms underlying the SCA2 phenotypicdiversity, especially parkinsonism, 2 fundamental questions need tobe but have never been answered: whether ATXN2 CAG expansionis the only cause for SCA2 and its clinical alternatives, are thereextra mutations modifying the phenotype? What is the molecularbasis of selective expression of expanded polyQ in different path-ologically involved tissues? To address these questions, we haveconducted fine genetic mapping and whole-exome sequencing in alarge Chinese pedigree with autosomal dominant parkinsonismcarrying the ATXN2 CAG expansion. We also compared the DNAsequence of the core promoter of ATXN2 between SCA2-P andSCA2-A, which usually determines the tissue-specific expression ofthe gene.

2. Methods

2.1. Study subjects

A large pedigree (Fig. 1) predominantly manifesting parkinso-nian symptoms with autosomal dominant inheritance residing inHubei Province in Central China was initially visited in 2007 andreported by another group (Sun et al., 2011) and revisited in 2011 byneurologists from Xuanwu Hospital, Capital Medical University.Clinical assessment was performed by specialists in movementdisorders. Diagnosis of parkinsonism was established according tothe UK Parkinson’s Disease Society Brain Bank criteria. In additionto the SCA2-P families, 16 SCA2-P families presenting with pureataxic symptoms and classified by molecular analysis as SCA2 werecollected by collaborators from Xiangya Hospital of Central SouthUniversity. Peripheral blood was collected, and DNA was extractedfor genetic analysis. Written informed consent was obtained fromeach participating subjects, and the study was approved by Insti-tutional Ethics Committee at both Xuanwu Hospital and XiangyaHospital.

Fig. 1. The large Chinese pedigree of SCA2-Parkinsonism. The family consists of 82 membeheritance pattern is autosomal dominant. Age at onset varied from 32 to 45 years and dparkinsonian symptoms including resting tremor, bradykinesia, rigidity, and postural instabigait and limb ataxias and ataxic dysarthria, which occurred 4 years later after parkinsonian

2.2. Analysis of trinucleotide expansions within the ATXN2 gene

Amplification of the CAG repeat region was performed with thefollowing primers selected from the flanking sequence of the repeatregion. Forward: 50-GGGCCCCTCACCATGTCG-30; reverse: 50-CGG-GCTTGCGGACATTGG-30. The amplified fragment covers both theCAG repeats and the 2 single-nucleotide polymorphisms (SNPs),rs695871 and rs695872, which were used for SNP haplotype anal-ysis. For sizing of the CAG repeats, fluorescence-polymerase chainreaction products were analyzed on an ABI-Prism 3100 automaticsequencer (PE Applied Biosystems, Foster City, CA, USA) using theGS500 size standard. For sequence analysis of the ATXN2 alleles, TAcloning strategy were used as previously reported. The rs695871-rs695872 haplotype and its linkage to the CAG repeats weredetermined according to the sequence.

2.3. Whole-genome scan and fine mapping by 2-point linkageanalysis

Whole-genome screening for the disease locus for the largeSCA2-P pedigree was performed using 382 microsatellite markersfrom the ABI PRISM Linkage Mapping Set Version 2 (Applied Bio-systems). Microsatellites for fine-mapping were chosen from theGenethon linkage map and the Marshfield sex-average linkage map(http://research.marshfieldclinic.org/genetics) and fluorescence-labeled with 6-FAM or HEX. Polymerase chain reaction (PCR) wasperformed using 20 ng of DNA, 5 pmol of each primers, 0.25 mMeach of dNTPs, 15 mM Tris-HCL (pH 8.0), 2.5 mM MgCl2, and 0.5units of DNA polymerase (AmpliTaq Gold; Applied Biosystem).Amplification conditions were as follows: preincubation at 95 �C for12 minutes, 10 cycles of denaturing at 94 �C for 15 seconds,annealing at 55 �C for 15 seconds, and extension at 72 �C for30 seconds, and final extenteion at 72 �C for 30 seconds. The PCRproducts were mixed with a gel-loading cocktail containing mo-lecular weight standards (Genescan 400HDROX) and analyzed on4% polyacrylamide denaturing gels with ABI PRISM 377 DNAsequencer (Applied Biosystems). Microsatellite alleles were deter-mined with GeneScan 3.0 and Genotyper 2.1 softwares.

Log10 of-odds scores were calculated using the MLINK programof the LINKAGE package. The parameters used in linkage analysiswere autosomal dominant inheritance, complete penetrance, amutation rate of zero, equal male-female recombination rate, equalmicrosatellite-allele frequency, and a disease-allele frequency of 1in 10,000.

2.4. Haplotype analysis of microsatellite markers around theATXN2 locus

Haplotype analysis of the SCA2-P family consisting of affectedand unaffected members was carried out using microsatellite

rs in 5 generations, with 16 affected members (arrow indicates the proband). The in-isease duration ranged from 1 to 15 years. All the patients showed predominantlylity, which were almost symmetric. Only IV-10 presented mild cerebellar signs such assymptoms.

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e7 3

markers, D12S84, D12S1616, D12S79, D12S1330, and D12S366.These markers span a region around the ATXN2 gene in thefollowing order: telomere-D12S84-D12S1616-D12S79-D12S1330-D12S366e centromere. The frequency of these markers was alsodetermined in normal chromosomes derived from unaffectedfamily members and normal control individuals from the samegeographic origin as the patients. The phase of the markers wasdetermined in the normal chromosomes segregating in the affectedpedigrees.

2.5. Whole-exome sequencing for affected and unaffected membersof the SCA2-P pedigree

Qualified genomic DNA extracted from the 3 patients (patientIV-10, IV-11, and IV-18) and 1 unaffected family member (IV-6) wassheared by sonication and then hybridized to the SureSelect Biotinylated RNA Library (BAITS) for enrichment, according to the manu-facturer’s instructions. The enriched library targeting the exomewas sequenced on the HiSeq 2000 platform to get paired-end readswith read length of 90 bp. The average sequencing depth on targetwas 57.5�.

The sequenced reads were aligned to the human genomereference (UCSC hg 18 version) using SOAPaligner (Li et al., 2008).Those reads were then aligned in the designed target regions andcollected for SNP calling and subsequent analysis. Consensus ge-notype and quality were estimated by SOAPsnp (v 1.03) (Li et al.,2010). The low-quality variations were filtered out. For insertionsor deletions (indels) in the targeted exome regions, reads werealigned to the reference genome using bwa. The alignment resultwas used to identify the breakpoints by GATK.

Based on the hypothesis that the mutation underlying familieswith SCA2-PD should not be present in the general population,nonsynonymous/splice acceptor, and donor site/insertions or de-letions (NS/SS/Indel) variants reported in the dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/, Build 129), 8 previously exome-sequenced HapMap samples (“HapMap 8”) and 1000 GenomeProject (http://www.ncbi.nlm.nih.gov/Ftp/) were removed. Synon-ymous changes were identified and filtered from the variant listusing SIFT software (version 4.0, http://sift.jcvi.org/).

2.6. Sequence analysis of the promoter region of ATXN2 gene

The promoter region of the ATXN2 gene (�599 to þ239 nt),according to Aguiar et al. (1999), was amplified by PCR using thefollowing 2 pairs of primers which spanned 2 partially over-lapping fragments. Primer pair 1: forward: 50-TAGGTACCCA-GATGTGGGT-30; reverse: 50-GAACTGGGTTGCTTTCTCG-30. Primer

Table 1Clinical features of the affected members of the Chinese SCA2-PD pedigree

Pedigreemember/sex

Age (y) Age at onset (y) Diseaseduration (y)

Initialsymptoms

Side-dom

IV-3/M 53 38 15 B SymIV-5/F 40 36 4 B SymIV-8/M 47 33 15 B SymIV-9/F 46 36 10 B SymIV-10/F 44 36 8 B SymIV-11/M 41 36 5 B SymIV-14/M 52 48 4 B SymIV-16/F 45 44 1 B SymIV-18/F 47 32 15 B SymIV-19/F 46 45 1 B Sym

Key: B, bradykinesia; F, female; H-Y, Hoehn-Yahr stage; M, male; MMSE, mini-mental staUPDRS, unified Parkinson’s disease rating scale.

pair 2: forward: 50-ATAAAGGCGGATCCGGGTA-30; reverse: 50-CCGTT G CTACCA AAACAGTC-30. The PCR product was purifiedand sequenced by ABI 3730 DNA analyzer (Applied Biosciences,Inc, CT, USA).

3. Results

3.1. Clinical features of the SCA2-P and SCA2-A pedigrees

The SCA2-P pedigree was shown in Fig. 1. Compared with thereport by Sun et al. (2011), we have obtained much more detailedclinical and pedigree information for all the family members.Moreover, the 2 members in the pedigree (IV-10 and IV-13, as re-ported) who were unaffected in 2007 have developed typicalparkinsonism 3 years later. In addition, we have collected the in-formation for the 5th generation of the pedigree and identified 6more unaffected mutation carriers who were younger than25 years.

The family consists of 82 members in 5 generations, with 16affected members. The inheritance pattern is autosomal domi-nant. Blood samples from 61 family members were collected,including 10 affected members (all the IV generation). The clinicaldata for all the affected members were listed in Table 1. Age atonset varied from 32 to 45 years and disease duration ranged from1 to 15 years. All the patients showed predominantly parkinsoniansymptoms including resting tremor, bradykinesia, rigidity, andpostural instability, which were almost symmetric. The initialsymptom of all the affected members was bradykinesia. Restingtremors occurred in 5 of 10 patients. Only IV-10 presented mildcerebellar signs such as gait and limb ataxias and ataxic dysarthria,which occurred 4 years later after parkinsonian symptoms. Eightpatients were treated with amantadine and artan and all showedresponse in the first 2 years but the efficacy diminished withdisease progression. All the 5 patients taking levodopa showed asustained response. Magnetic resonance imaging scan was per-formed for all the 10 patients. Only patients IV-3 and IV-10 showedcerebellar atrophy, which was not present in the remaining pa-tients, even in those with disease duration longer than 10 years(IV-8, IV-9, and IV-18).

All 16 families collected by Xiangya Hospital previouslydiagnosed as SCA2-ataxia (SCA2-A) showed gait and limb ataxia,and most patients exhibited dysarthria and decreased tendonreflexes. Oculomotor disturbances including slow saccadessupranuclear ophthalmoplegia horizontal nystagmus were alsopresent in some cases. The onset age of symptoms ranged from19 to 53 years. No parkinsonian symptom was observed in thesefamilies (Table 2).

inance Ataxia Parkinsonism Clinical scores

RT B R PI UPDRS total H-Y MMSE

� þ þþþ þþþ þþ 75 3 29� þ þþþ þþ þ 63 2.5 18� � þ þ � 33 2 20� � þ � � 13 2 21þ � þþ þþ � 38 2 29� þ þþ þþþ þ 58 2.5 20� � þ � � 4 2 27� � þþ þ þþ 44 3 29� þ þþ þþþ þþ 57 3 18� þ þþ þ þ 40 2.5 27

te examination; PI, posture instability; R, rigidity; RT, rest tremor; Sym, symmetric;

Table 2Clinical and genetic features of probands from 16 families with SCA2-ataxia

Pedigree number Sex Age at onset (y) Number of CAG repeats

1 M 31 522 F 24 423 M 41 414 F 35 425 M 21 476 M 38 437 F 37 448 M 36 379 M 40 4310 M 41 3611 M 19 4712 F 28 5413 F 38 4514 M 37 3615 M 49 4216 M 53 42

Key: F, female; M, male.

Table 3The repeat number and configuration of the CAG/CAA expansion in affected andunaffected members

Pedigreenumber/sex

Age (y) Affected(Y/N)

Repeatnumber

Configuration of theexpanded CAA/CAG repeats

IV-3/M 47 Y 39 (CAG) 30 (CAA) 1 (CAG) 8IV-5/F 53 Y 39 (CAG) 30 (CAA) 1 (CAG) 8IV-8/M 52 Y 39 (CAG) 30 (CAA) 1 (CAG) 8IV-9/F 46 Y 37 (CAG) 28 (CAA) 1 (CAG) 8IV-10/F 47 Y 37 (CAG) 28 (CAA) 1 (CAG) 8IV-11/M 46 Y 37 (CAG) 28 (CAA) 1 (CAG) 8IV-12/F 38 N 37 (CAG) 28 (CAA) 1 (CAG) 8IV-13/F 34 N 37 (CAG) 28 (CAA) 1 (CAG) 8IV-14/M 45 Y 39 (CAG) 30 (CAA) 1 (CAG) 8IV-16/F 44 Y 39 (CAG) 30 (CAA) 1 (CAG) 8IV-18/F 41 Y 40 (CAG) 31 (CAA) 1 (CAG) 8IV-19/F 40 Y 40 (CAG) 31 (CAA) 1 (CAG) 8V-3/F 20 N 39 (CAG) 30 (CAA) 1 (CAG) 8V-5/M 24 N 39 (CAG) 30 (CAA) 1 (CAG) 8V-11/M 19 N 37 (CAG) 28 (CAA) 1 (CAG) 8V-13/F 16 N 37 (CAG) 28 (CAA) 1 (CAG) 8V-25/M 26 N 40 (CAG) 31 (CAA) 1 (CAG) 8V-26/M 21 N 40 (CAG) 31 (CAA) 1 (CAG) 8

Key: F, female; M, male; N, no; Y, yes.

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e74

3.2. Different ATXN2 CAG expansion patterns identified between theSCA2-P and SCA2-A pedigrees

All the 10 affected (IV-3, IV-5, IV-8, IV-9, IV-10, IV-11, IV-14, IV-16, IV-18, and IV-19) and 8 nonsymptomatic (IV-12, IV-13, V-3, V-5,V-11, V-13, V-25, and V-26) members of the SCA2-P pedigree (Fig. 1)carried heterozygous CAG expansion with repeat numbers varyingfrom 37 to 40. The repeat number was slightly variable amongsiblings but stable between generations. In contrast, the normalalleles for all the affected members of the family contained 22 re-peats. Sequence analyses suggested that the normal allele con-tained 2 CAA interruptions and displayed a configuration of(CAG)8CAA(CAG)4CAA(CAG)8, whereas the expanded alleles onlycontained a reserved 30-CAA interruption and displayed a config-uration as (CAG)nCAA(CAG)8 (Table 3). In contrast, all the expandedalleles of higher range repeats of the SCA2-ataxia families consistedof no CAA interruption.

3.3. The SCA2-P pedigree was mapped to the ATXN2 locus(12q24.21) where a specific haplotype segregates with the ATXN2CAG expansion

To address whether SCA2 is the only locus responsible for thespecific type of SCA2-P, we mapped the disease locus by linkageanalysis for the family. Whole-genome scan was performed byparametric 2-point linkage analysis for 24 individuals from thegenerations III and IV (10 affected, 11 unaffected, and 3 spouses)of the family. As a result, D12S79 (12q24.21) gave the highestLOD score of 3.28 (0.00), and no other microsatellite gave LODscore >1.0. Additional microsatellites on chromosome 12revealed that only D12S86, 3.3 Mb telomeric to D12S79, gaveLOD score of 0.73 (0.00). Fine mapping with 11 microsatellitemarkers close to D12S79 narrowed the disease locus down to a16 Mb region on 12q24.21 (Table 4). Haplotype analysis sug-gested that all the affected members of the family shared the 3-3-6-3-3 haplotype for microsatellite markers in order ofD12S84-D12S1616-D12S1330- D12S79-D12S366, which spans9.6 Mb and contains the SCA2 locus (Fig. 2). Using thesemarkers, haplotype analysis was also performed for 29 membersof the V generation. Subjects carrying the CAG expansions allharbored the disease-linked haplotype, suggesting that thishaplotype is in complete linkage disequilibrium with the CAGexpansion in SCA2-P.

3.4. Lack of haplotypes predisposing CAG expansion in SCA2-P orSCA2-A

Choudhry et al. (2001) reported that 2 SNPs, rs695871 andrs695872, upstream the exon 1 of ATXN2 was in complete linkagedisequilibrium with the CAG expansion. The 2 SNPs distinguished2 haplotypes, GT and CC, which were associated with normal andexpanded SCA2 alleles. To test whether these biallelic haplotypeswere associated with different patterns of CAG expansion betweenSCA2-P and SCA2-A patients, we genotyped the 2 SNPs for all thesepatients. Strikingly, all the tested patients displayed a universal C-C monotype for rs695871-rs695872, and the G-T haplotype wasnot present. These results suggest that this SNP haplotype is notassociated with a specific CAG expansion pattern. We also geno-typed all the 7 SNPs (rs3809274, rs1544396, rs9300319, rs593226,rs616513, rs695871, and rs653178) around the SCA2 locus thatwere reported to form a haplotype that genetically select the pre-expanded allele in Europeans (Yu et al., 2005). As a result, none ofthese variants was detected in our SCA2-P, SCA2-A patients, ornormal controls, indicating the lack of such haplotypes associatedwith SCA2 expansion in Chinese.

3.5. Whole-exome sequencing excluded extra mutations responsiblefor SCA2-P

To exclude the existence of other possible causative mutationsfor SCA2-PD, exome sequencing was performed on DNA samplesobtained from 3 affected (IV-10, IV-11, andIV-18) and 1 unaffectedsibling (IV-6) of the SCA2-P family. We generated an average of6.75 Gb of sequence from each affected individual as paired-end,90-bp reads; 6.41 Gb (94.96%) passed the quality assessment andwere aligned to the human reference sequence; and an average of3.69 Gb of sequence data was mapped to the target region. AfterSNPs and Indels calling, we identified a mean of 90, 634 variantsfrom the reference sequence per subject, in which an average of13,200 NS/SS/Indel variants were detected in each of the SCA2-Ppatients sequenced. Given that this is a rare disorder, it is un-likely that causative variants will be present in the general pop-ulation. We therefore removed the NS/SS/Indel variants reportedin the dbSNP129, “HapMap 8,” and the SNP release of the 1000Genome Project. After this initial filter, we generated an average of443 NS/SS/Indel variants from each patient. We then focused on

Table 4Fine mapping of the disease locus by 2-point linkage disequilibrium analyses

Marker Physicaldistance (Mb)

LOD score

0 0.1 0.2 0.3 0.4 0.5

D12S84 107.5 �1.3 1.8 1.45 0.93 0.36 0.00D12S1343 110.2 0.91 0.67 0.43 0.21 0.05 0.00D12S1616 111.8 0.74 0.52 0.32 0.15 0.04 0.00D12S1330 113.6 2.92 2.3 1.66 0.99 0.36 0.00D12S79 114.4 3.28 2.62 1.9 1.15 0.41 0.00D12S2079 115.7 3.04 2.41 1.74 1.04 0.37 0.00D12S2082 116.3 2.6 2.03 1.45 0.86 0.3 0.00D12S366 117.1 2.87 2.25 1.61 0.95 0.33 0.00D12S86 117.7 0.73 0.52 0.32 0.15 0.04 0.00D12S1349 120.8 �0.4 1.89 1.53 1.01 0.41 0.00D12S1612 123.5 �0.3 1.97 1.6 1.06 0.43 0.00

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e7 5

the genes with NS/SS/Indel variants occurring in the linkage re-gion (12q12), as well as outside the region. We identified noconsensus SNP or Indel variant in the 3 patients (patient IV-10, IV-11, andIV-18) that was absent in the unaffected member (IV-6), ineither the disease-linked or nonlinked region. Although patientIV-10 displayed much more severe ataxic phenotype and cere-bellar atrophy than the other 2 patients, there was no extra mu-tation detected in this patient.

3.6. No sequence difference between SCA2-P and SCA2-A wasidentified in the promoter region of ATXN2 gene

To identify possible sequence difference within the promoterregion of the ATXN2 gene that may influence the tissue-specificexpression of the mutant gene, we compared the promotersequence (�599 to þ239) of the SCA2 gene between all the carriersfor CAG expansion from the SCA2-P and SCA2-A patients. As aresult, except for 1 polymorphism (C/G) in the �207 nt position, nosequence variation was identified in the promoter sequence.Moreover, the frequency of the minor allele “G” for this poly-morphism in the CAG expansion carriers in the SCA2-P family

Fig. 2. Haplotype analysis of the nuclear pedigree of the large Chinese SCA2-Parkinsonism fthe 7 microsatellites as shown in Table 4. In the cases not determined explicitly, most provabdisease-associated haplotypes, including the deduced ones (boxed), 2 recombination eventdefined by these recombinations was shared by all the patients and some unaffected carrie

(35.3%, 6/17) was not different from those in the SCA2-A probands(31.3%, 5/16) (p > 0.05).

4. Discussion

Although SCA2 has been extensively reported as a cause for ADPin Chinese and other populations, none of the previous studies haveprovided evidences addressing whether ATXN2 CAG expansion isthe “only” causative mutation for this specific subtype of ADP. Bygenetic mapping, CAG repeat analysis and whole-exomesequencing in a large Chinese SCA2-P pedigree, we for the firsttime confirm that ATXN2 is the sole locus for SCA2-P, and no extramutations in the coding regions throughout the genome areresponsible for the parkinsonian phenotype. Moreover, bycomparing promoter sequence between SCA2-P and SCA2-A, weexclude a sequence variation in the ATXN2 promoter that maydetermine expression of the expanded polyQ in different pathologicsites (basal ganglia vs. cerebellum). Thus, it seems unlikely that thephenotypic diversity of SCA2 results mainly frommodifying geneticfactors.

Our study has also confirmed the importance of low-range CAGexpansion and CAA interruption in SCA2-P, as reported previously.All expanded alleles in the SCA2-P patients displayed a loss of the50-CAA interrupt, compared with loss of both of the CAA interruptsin most of the SCA2-A patients. However, it remains unclear whyparkinsonism appears especially in SCA2 with low-range CAGexpansion and interruption. Sobczak and Krzyzosiak (2005) foundthat CAG repeats containing CAA interruptions form branchedhairpin structures in SCA2 transcripts, and, thus, may result in anRNA toxicity-mediated pathogenesis. More recent studies sug-gested that intermediate-length polyglutamine expansionsenhance TDP-43- and FUS-related pathology in amyotrophic lateralsclerosis (Elden et al., 2010; Farg et al., 2013) via promoting theiraggregation or mislocalization. However, role of ataxin-2 in eitherRNA toxicity or protein-protein interaction has not been tested inthe pathology of SCA2-P. Future studies are critical to determinewhether moderately expanded ataxin-2 polyQ repeats could

amily. Haplotypes of 22 family members were constructed based on the genotyping ofle haplotypes were deduced with the GENEHUNTER program (thin italic letters). In thes were observed: one at D12S84 and the other at D12S366. The 9.6 cM interval regionrs.

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e76

interact with a-synuclein or other PD-related proteins and promoteprotein aggregation and degeneration of dopaminergic neurons.

Another puzzle of SCA2mutation is howare different patterns ofCAG expansion formed? Studies have demonstrated that in-terruptions of CAG repeats by 1e2 CAAs are required for most of thenormal alleles to maintain a stable structure, without which leadsto high-range expansion consists of pure CAGs. A study showed that(CAG)8CAA(CAG)4CAA(CAG)8, the most common form of the pre-expanded allele in Europeans, is a result of genetically positive se-lection by a haplotype composed of 7 SNPs around the SCA2 locus(Yu et al., 2005). In addition, Choudhry et al. (2001) reported thepresence of an “ancestral” or “at risk” haplotype that predispose tosubsequent expansion of the repeat allele in Indians (Choudhryet al., 2001). Most of normal alleles segregated with this haplo-type are either pure or lacked the most proximal 50-CAA interrup-tion and thus prone to expansions leading to ataxic SCA2. However,none of such haplotypes was identified in the current SCA2-P orSCA2-A pedigrees, suggesting that there was no “ancestral” or “atrisk” haplotype predisposing SCA2-P or SCA2-A expansions inChinese.

In summary, we have confirmed that the CAG expansion withCAA interruption in ATXN2 genewas the only genetic cause of SCA2-P, and exclude a second genetic mutation that modify the parkin-sonian phenotype of SCA2. We also exclude a specific haplotypepredisposing different patterns (low or high range) of the CAGexpansion. These results indicated that geneticmodifiersmaynot bethe major cause of the phenotypic diversity of SCA2. Future studiesare needed to investigate how intermediate length of expandedataxin-2 polyQ interacts with alpha-synuclein or other PD-associated proteins to influences the pathophysiology of SCA2-P.

Disclosure statement

None of the authors have conflicts of interest to declare inrelation to the present research.

Acknowledgements

This study was supported by grants from the National BasicResearch Development Program of China (2011CB504101) and theMinistry of Science and Technology of China (2012AA02A514) to DrPiu Chan; from National Natural Science Foundation of China(No.81371320) to Dr Chaodong Wang and Dr Yanming Xu(No.30700243).

References

Aguiar, J., Santurlidis, S., Novok, J., Alexander, C., Rudnicki, D., Gispert, S., Schulz, W.,Aburger, G., 1999. Identification of the physiological promoter for spino-cerebellar ataxia 2 gene reveals a CpG island for promoter activity situatedinto the exon 1 of this gene and provides data about the origin of the non-methylated state of these types of islands. Biochem. Biophys. Res. Commun. 254,315e318.

Chan, D.K., Mellick, G., Cai, H., Wang, X.L., Ng, P.W., Pang, C.P., Woo, J., Kay, R., 2000.The alpha-synuclein gene and Parkinson disease in a Chinese population. Arch.Neurol. 57, 501e503.

Charles, P., Camuzat, A., Benammar, N., Sellal, F., Destée, A., Bonnet, A.M., Lesage, S.,Le Ber, I., Stevanin, G., Dürr, A., Brice, A., French Parkinson’s Disease GeneticStudy Group, 2007. Are interrupted SCA2 CAG repeat expansions responsible forparkinsonism? Neurology 69, 1970e1975.

Chartier-Harlin, M.C., Dachsel, J.C., Vilariño-Güell, C., Lincoln, S.J., Leprêtre, F.,Hulihan, M.M., Kachergus, J., Milnerwood, A.J., Tapia, L., Song, M.S., Le Rhun, E.,Mutez, E., Larvor, L., Duflot, A., Vanbesien-Mailliot, C., Kreisler, A., Ross, O.A.,Nishioka, K., Soto-Ortolaza, A.I., Cobb, S.A., Melrose, H.L., Behrouz, B.,Keeling, B.H., Bacon, J.A., Hentati, E., Williams, L., Yanagiya, A., Sonenberg, N.,Lockhart, P.J., Zubair, A.C., Uitti, R.J., Aasly, J.O., Krygowska-Wajs, A., Opala, G.,Wszolek, Z.K., Frigerio, R., Maraganore, D.M., Gosal, D., Lynch, T., Hutchinson, M.,Bentivoglio, A.R., Valente, E.M., Nichols, W.C., Pankratz, N., Foroud, T.,Gibson, R.A., Hentati, F., Dickson, D.W., Destée, A., Farrer, M.J., 2011. Translation

initiator EIF4G1 mutations in familial Parkinson disease. Am. J. Hum. Genet. 89,398e406.

Chen, Y., Chen, K., Song, W., Chen, X., Cao, B., Huang, R., Zhao, B., Guo, X.,Burgunder, J., Li, J., Shang, H.F., 2013. VPS35 Asp620Asn and EIF4G1 Arg1205Hismutations are rare in Parkinson disease from southwest China. Neurobiol. Aging34, 1709.e7e1709.e8.

Choudhry, S., Mukerji, M., Srivastava, A.K., Jain, S., Brahmachari, S.K., 2001. CAGrepeat instability at SCA2 locus: anchoring CAA interruptions and linked singlenucleotide polymorphisms. Hum. Mol. Genet. 10, 2437e2446.

Elden, A.C., Kim, H.J., Hart, M.P., Chen-Plotkin, A.S., Johnson, B.S., Fang, X.,Armakola, M., Geser, F., Greene, R., Lu, M.M., Padmanabhan, A., Clay-Falcone, D.,McCluskey, L., Elman, L., Juhr, D., Gruber, P.J., Rüb, U., Auburger, G.,Trojanowski, J.Q., Lee, V.M., Van Deerlin, V.M., Bonini, N.M., Gitler, A.D., 2010.Ataxin-2 intermediate-length polyglutamine expansions are associated withincreased risk for ALS. Nature 466, 1069e1075.

Farg, M.A., Soo, K.Y., Warraich, S.T., Sundaramoorthy, V., Blair, I.P., Atkin, J.D., 2013.Ataxin-2 interacts with FUS and intermediate-length polyglutamine expansionsenhance FUS-related pathology in amyotrophic lateral sclerosis. Hum. Mol.Genet. 22, 717e728.

Furtado, S., Payami, H., Lockhart, P.J., Hanson, M., Nutt, J.G., Singleton, A.A., 2004.Profile of families with parkinsonism-predominant spinocerebellar ataxia type2 (SCA2). Mov. Disord. 19, 622e629.

Guo, R., Hu, X., Chen, Q., Zhang, Y., Zhang, Y., Sun, Y., Hu, G., 2013. The LRRK2 gene ismutated in a Chinese autosomal-dominant Parkinson’s disease family. Genet.Test. Mol. Biomarkers 17, 131e134.

Li, R., Li, Y., Kristiansen, K., Wang, J., 2008. SOAP: short oligonucleotide alignmentprogram. Bioinformatics 24, 713e714.

Li, R., Zhu, H., Ruan, J., Qian, W., Fang, X., Shi, Z., Li, Y., Li, S., Shan, G.,Kristiansen, K., Li, S., Yang, H., Wang, J., Wang, J., 2010. De novo assembly ofhuman genomes with massively parallel short read sequencing. Genome Res.20, 265e272.

Lim, S.W., Zhao, Y., Chua, E., Law, H.Y., Yuen, Y., Pavanni, R., Wong, M.C., Ng, I.S.,Yoon, C.S., Puong, K.Y., Lim, S.H., Tan, E.K., 2006. Genetic analysis of SCA2, 3 and17 in idiopathic Parkinson’s disease. Neurosci. Lett. 403, 11e14.

Lin, C.H., Tzen, K.Y., Yu, C.Y., Tai, C.H., Farrer, M.J., Wu, R.M., 2008. LRRK2 mutation infamilial Parkinson’s disease in a Taiwanese population: clinical, PET, and func-tional studies. J. Biomed. Sci. 15, 661e667.

Liu, X., Lu, M., Tang, L., Zhang, N., Chui, D., Fan, D., 2013. ATXN2 CAG repeat ex-pansions increase the risk for Chinese patients with amyotrophic lateral scle-rosis. Neurobiol. Aging 34, 2236.e5e2236.e8.

Paisán-Ruíz, C., Jain, S., Evans, E.W., Gilks, W.P., Simón, J., van der Brug, M., Lópezde Munain, A., Aparicio, S., Gil, A.M., Khan, N., Johnson, J., Martinez, J.R.,Nicholl, D., Carrera, I.M., Pena, A.S., de Silva, R., Lees, A., Martí-Massó, J.F.,Pérez-Tur, J., Wood, N.W., Singleton, A.B., 2004. Cloning of the gene con-taining mutations that cause PARK8-linked Parkinson’s disease. Neuron 18,595e600.

Polymeropoulos, M.H., Lavedan, C., Leroy, E., Ide, S.E., Dehejia, A., Dutra, A., Pike, B.,Root, H., Rubenstein, J., Boyer, R., Stenroos, E.S., Chandrasekharappa, S.,Athanassiadou, A., Papapetropoulos, T., Johnson, W.G., Lazzarini, A.M.,Duvoisin, R.C., Di Iorio, G., Golbe, L.I., Nussbaum, R.L., 1997. Mutation in thealpha-synuclein gene identified in families with Parkinson’s disease. Science276, 2045e2047.

Ross, O.A., Rutherford, N.J., Baker, M., Soto-Ortolaza, A.I., Carrasquillo, M.M.,DeJesuseHernandez, M., Adamson, J., Li, M., Volkening, K., Finger, E.,Seeley, W.W., Hatanpaa, K.J., Lomen-Hoerth, C., Kertesz, A., Bigio, E.H., Lippa, C.,Woodruff, B.K., Knopman, D.S., White 3rd, C.L., Van Gerpen, J.A., Meschia, J.F.,Mackenzie, I.R., Boylan, K., Boeve, B.F., Miller, B.L., Strong, M.J., Uitti, R.J.,Younkin, S.G., Graff-Radford, N.R., Petersen, R.C., Wszolek, Z.K., Dickson, D.W.,Rademakers, R., 2011. Ataxin-2 repeat-length variation and neurodegeneration.Hum. Mol. Genet. 20, 3207e3212.

Sobczak, K., Krzyzosiak, W.J., 2005. CAG repeats containing CAA interruptions formbranched hairpin structures in spinocerebellar ataxia type 2 transcripts. J. Biol.Chem. 280, 3898e3910.

Sun, H., Satake, W., Zhang, C., Nagai, Y., Tian, Y., Fu, S., Yu, J., Qian, Y., Qian, Y.,Chu, J., Toda, T., 2011. Genetic and clinical analysis in a Chinese parkinsonismpredominant spinocerebellar ataxia type 2 family. J. Hum. Genet. 56,330e334.

Van Langenhove, T., van der, Zee, J., Engelborghs, S., Vandenberghe, R., Santens, P.,Van den, Broeck, M., Mattheijssens, M., Peeters, K., Nuytten, D., Cras, P., DeDeyn, P.P., De Jonghe, P., Cruts, M., Van Broeckhoven, C., 2012. Ataxin-2 polyQexpansions in FTLD-ALS spectrum disorders in Flanders-Belgian cohorts. Neu-robiol. Aging 33, 1004.e17e1004.e20.

Wang, L., Guo, J.F., Nie, L.L., Xu, Q., Zuo, X., Sun, Q.Y., Yan, X.X., Tang, B.S., 2010.A novel LRRK2 mutation in a mainland Chinese patient with familial Parkinson’sdisease. Neurosci. Lett. 468, 198e201.

Wang, J.L., Xiao, B., Cui, X.X., Guo, J.F., Lei, L.F., Song, X.W., Shen, L., Jiang, H., Yan, X.X.,Pan, Q., Long, Z.G., Xia, K., Tang, B.S., 2009. Analysis of SCA2 and SCA3/MJDrepeats in Parkinson’s disease in mainland China: genetic, clinical, and positronemission tomography findings. Mov. Disord. 24, 2007e2011.

Wu, Y.R., Lin, H.Y., Chen, C.M., Gwinn-Hardy, K., Ro, L.S., Wang, Y.C., Li, S.H.,Hwang, J.C., Fang, K., Hsieh-Li, H.M., Li, M.L., Tung, L.C., Su, M.T., Lu, K.T., Lee-Chen, G.J., 2004. Genetic testing in spinocerebellar ataxia in Taiwan: expansionsof trinucleotide repeats in SCA8 and SCA17 are associated with typical Parkin-son’s disease. Clin. Genet. 65, 209e214.

C. Wang et al. / Neurobiology of Aging xxx (2014) 1e7 7

Yu, F., Sabeti, P.C., Hardenbol, P., Fu, Q., Fry, B., Lu, X., Ghose, S., Vega, R., Perez, A.,Pasternak, S., Leal, S.M., Willis, T.D., Nelson, D.L., Belmont, J., Gibbs, R.A., 2005.Positive selection of a pre-expansion CAG repeat of the human SCA2 gene. PLoSGenet. 1, e41.

Zhang, Y., Chen, S., Xiao, Q., Cao, L., Liu, J., Rong, T.Y., Ma, J.F., Wang, G., Wang, Y.,Chen, S.D., 2012. Vacuolar protein sorting 35 Asp620Asn mutation is rare in theethnic Chinese population with Parkinson’s disease. Parkinsonism Relat. Disord.18, 638e640.

Zimprich, A., Benet-Pagès, A., Struhal, W., Graf, E., Eck, S.H., Offman, M.N.,Haubenberger, D., Spielberger, S., Schulte, E.C., Lichtner, P., Rossle, S.C., Klopp, N.,Wolf, E., Seppi, K., Pirker, W., Presslauer, S., Mollenhauer, B., Katzenschlager, R.,Foki, T., Hotzy, C., Reinthaler, E., Harutyunyan, A., Kralovics, R., Peters, A.,Zimprich, F., Brücke, T., Poewe, W., Auff, E., Trenkwalder, C., Rost, B.,Ransmayr, G., Winkelmann, J., Meitinger, T., Strom, T.M., 2011. A mutation inVPS35, encoding a subunit of the retromer complex, causes late-onset Parkin-son disease. Am. J. Hum. Genet. 89, 168e175.