Chorein Detection for the Diagnosis of Chorea-Acanthocytosis Carol Dobson-Stone, PhD,1 Antonio Velayos-Baeza, PhD,1 Lea A. Filippone,2 Sarah Westbury,3 Alexander Storch, MD,4 Torsten Erdmann, MD,5 Stephen J. Wroe, MD, FRCP,6 Klaus L. Leenders, MD,7 Anthony E. Lang, MD, FRCP,8 Maria Teresa Dotti, MD,9 Antonio Federico, MD,9 Saidi A. Mohiddin, MD, MRCP,10 Lameh Fananapazir, MD, FRCP,10 Geoff Daniels, PhD,11 Adrian Danek, MD,12 and Anthony P. Monaco, MD1 Chorea-acanthocytosis (ChAc) is a severe, neurodegenerative disorder that shares clinical features with Huntington’s disease and McLeod syndrome. It is caused by mutations in VPS13A, which encodes a large protein called chorein. Using antichorein antisera, we found expression of chorein in all human cells analyzed. However, chorein expression was absent or noticeably reduced in ChAc patient cells, but not McLeod syndrome and Huntington’s disease cells. This suggests that loss of chorein expression is a diagnostic feature of ChAc. Ann Neurol 2004;56:299 –302 Chorea-acanthocytosis (ChAc, OMIM 200150) is an autosomal recessive neurodegenerative disorder characterized by progressive onset of hyperkinetic movements and unusual spiny erythrocyte morphology (acanthocytosis).1 ChAc diagnosis can be challenging, due to interlaboratory variability in acanthocyte detection efficiency,2 and the clinical similarity of ChAc to Huntington’s disease (HD) and McLeod syndrome (MLS, OMIM *314850), an X-linked neuroacanthocytosis.3 CHAC, encoding the large (⬎3,000 amino acids) protein chorein, is the gene mutated in choreaacanthocytosis.4,5 CHAC, now renamed VPS13A,6 is organized into 73 exons and has two main splice forms: isoform 1A (exons 1– 68, 70 –73) and isoform 1B (exons 1– 69).4 Because of the size of the gene and the allelic heterogeneity of ChAc,7 mutation screening of VPS13A is a cumbersome process, impeding diagnostic progress. Herein, we report the generation of antichorein polyclonal antisera. We note the apparently ubiquitous distribution of chorein and demonstrate that loss of chorein expression in erythrocyte membranes and other cells is a diagnostic feature of ChAc. Subjects and Methods Subjects Informed patient consent for molecular analysis of blood samples was obtained according to local guidelines. Samples from 14 patients with a clinical diagnosis of choreaacanthocytosis (10 men, 4 women) were studied. Patients 1 to 4 correspond to CHAC2IV5, CHAC6II1 and a newly diagnosed brother, and CHAC11II2 in previous studies4,8; Patients 5, 6, 7, and 14 correspond to probands 24, 11, 2, and 23, respectively, in a previous study.7 All displayed characteristic clinical features of chorea-acanthocytosis1 that included elevation of serum creatine kinase (14 of 14 patients), ankle areflexia (10/14), cognitive or neuropsychiatric changes (12/14), limb chorea (13/14), orofacial dyskinesia (including tongue dystonia and lip biting, 12/14), dysarthria (11/14), involuntary vocalizations (7/14), and parkinsonian features (2/14). Nine of the 14 patients had experienced seizures. Blood samples from two men with MLS were also studied (Patients 3 and 13 from Danek and colleagues3). Mutation Detection From the 1Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK; 2Molecular Biology and Genetics, Cornell University, Ithaca, NY; 3St. John’s College, University of Oxford, Oxford, UK; 4Department of Neurology, Technical University of Dresden, Dresden, Germany; 5Department of Neurology, Vinzenz von Paul Hospital–Rottenmünster, Rottweil, Germany; 6Department of Clinical Neurology, Ipswich Hospital, Ipswich, UK; 7Groningen University Hospital, Groningen, The Netherlands; 8Division of Neurology, University of Toronto, Toronto, Canada; 9Department of Neurological and Behavioural Sciences, Section of Neurology and Neurometabolic Diseases, Medical School, University of Siena, Siena, Italy; 10Cardiovascular Branch, National Heart Lung and Blood Institute, Bethesda, MD; 11Bristol Institute for Transfusion Sciences, Bristol, UK; and 12Neurologische Klinik, LudwigMaximilians-Universität, Munich, Germany. Received Mar 17, 2004, and in revised form May 24. Accepted for publication May 25, 2004. Published online Jul 27, 2004, in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20200 Address correspondence to Prof Monaco, The Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Headington, Oxford, OX3 7BN, UK. E-mail: email@example.com ChAc patient DNA was screened for VPS13A mutations as described previously.7 At least one heterozygous VPS13A mutation likely to cause disease was found in each ChAc patient, as shown in the Table. Production of Polyclonal Antiserum Anti-chor1 A fragment of chorein comprising amino acids 27 to 326 was expressed as a fusion protein with glutathione S-transferase (GST-chor1). Bacterial expression and purification of GST-chor1 was performed as described by Frangioni and Neel,9 using glutathione Sepharose 4B beads (Amersham Biosciences UK Ltd, Chalfont St. Giles, UK). Immunization of rabbits with the chor1 moiety was performed by Eurogentec Bel SA (Herstal, Belgium); the serum collected 87 days after immunization was designated anti-chor1. Western Blotting of Cell Lysates Cells were harvested by trypsinization and centrifugation at 3,400g for 5 minutes. Cell pellets were washed with ice-cold phosphate-buffered saline then lysed in 10 ⫻ volume of ice- © 2004 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 299 Table. ChAc and McLeod Syndrome Patients Analyzed in This Study Patienta VPSI3A mutations 1 (4) 2, 3 (4) 4 (4) 5 (7) 6 (7) 7 (7) 8 9 10 11 12 13 14 (7) XK mutations M1 (13) M2 (3) DNA Changeb [1592del] [4354T⬎C] [237del] [6419C⬎G] [1125_1128del] [2288 ⫹ 2T⬎C] [1596 ⫺ 2A⬎C]d [1596 ⫺ 1G⬎C]d [EX46_EX50del] [5909_5910del] [188-5T⬎G] [495 ⫹ 1G⬎A] [1208_1211del] Protein Changeb [1592del] [4354T⬎C] [9429_9432del] [9190del] [?] [8472-1G⬎C] [8907 ⫹ 2T⬎A]d [4355C⬎G]d [EX46_EX50del] [?] [?] [?] [7867C⬎T] 508 ⫹ 1G⬎A EX1del [1531fsX6] [S1452P] [E80fsX10] [S2140X] [S375fsX22] [SD] [SA] [SA] [unknown] [E1970fsX3] [SA] [SD] [Q403fsX5] SD unknown [1531fsX6] [S1452P] [R3143fsX4] [V3064fsX16] [?] [SA] [SD] [S1452X] [unknown] [?] [?] [?] [R2623X] Type of Mutation Frameshift Missense Frameshift Nonsense Frameshift Splicing Splicing Splicing Deletion Frameshift Splicing Splicing Frameshift Splicing Deletion Frameshift Missense Frameshift Frameshift ? Splicing Splicing Nonsense Deletion ? ? ? ? Tissue Chorein analyzed Expressionc Lb Lb Lb Fb Fb Fb Ec Ec Ec Ec Ec Ec Ec ⫺ ⫹ ⫹ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫹ ⫺ Lb Ec ⫹⫹⫹ ⫹⫹⫹ a References for patients who were screened for mutations in a previous study are given in parentheses. Nucleotides and amino acids are numbered according to the cDNA sequence of VPS13A/CHAC isoform A reported by Rampoldi and colleagues4 (GenBank accession no. NM_033305) or that of XK13 (GenBank accession no. Z32684). Mutation nomenclature is as recommended by the Human Genome Variation Society (http://www.genomic.unimelb.edu.au/mdi/mutnomen/index.html). c Chorein expression levels are denoted as follows: - ⫽ undetectable; ⫹ ⫽ markedly reduced compared with control; ⫹⫹⫹ ⫽ similar to control levels. d Mutations detected in this study that were identified in another family previously7 b MI ⫽ McLeod syndrome patient 1; SD ⫽ splice donor; SA ⫽ splice acceptor; ? ⫽ second heterozygous mutation not found; Lb ⫽ lymphoblastoid cell line; Fb ⫽ primary skin fibroblast; Ec ⫽ erythrocyte membrane. cold lysis buffer (50mM Tris-HCL, pH 8.0, 150mM NaCl, 1% Triton X-100, and 1 ⫻ protease inhibitor cocktail; Sigma, Poole, UK). After 20-minute incubation on ice, lysates were clarified by centrifugation at 16,000g at 4°C for 20 minutes. Samples containing equal quantities of protein (20g) were denatured and loaded on a NuPAGE 3 to 8% Tris-acetate gel (Invitrogen, Paisley, UK), followed by electrophoretic transfer on to polyvinylidene difluoride membranes. Immunodetection was performed according to standard protocols,10 using anti-chor1 antiserum (1:5,000 dilution), preimmune serum (1:5,000) or mouse anti–huntingtin antibody (1:10,000; Serotec Ltd, Oxford, UK) followed by peroxidase-conjugated goat anti–rabbit or anti– mouse antibody (1:5,000; Bio-Rad Laboratories Ltd, Hemel Hempstead, UK). Proteins were visualized using the ECL plus Western Blotting Detection System (Amersham). For antibody depletion experiments, anti-chor1 antiserum was diluted 1:2,500 in 10mM Tris-HCL, pH 7.4, 0.8% NaCl. Half was passed three times through an affinity column containing chor1-conjugated agarose, and half was left untreated. Both serum preparations were subsequently adjusted to the appropriate dilution for use in Western blotting. utes. The loose pellet of membranes was washed multiple times in erythrocyte lysis solution until the supernatant remained colorless. Membranes (2.5l thereof) of were subjected to electrophoresis and blotted as above. Western Blotting of Erythrocyte Membranes Results Detection of Endogenous Chorein Western blot analysis of ChAc Patient 1 and control lymphoblastoid cell lysates shows that anti-chor1 serum recognizes chorein (see Fig 1A). A band similar in size to huntingtin (350kDa, lanes 7 and 8) was detected in control cells (lane 5); this is consistent with the predicted molecular weight of chorein (360kDa). This band was absent in cells from Patient 1 (lane 6). The signal was not detected using preimmune serum (lanes 1 and 2) or when using serum depleted in chor1binding antibodies (lanes 3 and 4), thereby confirming the detection specificity. To investigate chorein distribution, we analyzed lysates from a variety of cell lines routinely used in tissue culture techniques by Western blot (see Fig 1B). Anti-chor1 antiserum detected a high-molecular-weight signal in all cell lines analyzed. Erythrocyte membranes were prepared according to Dodge and colleagues,11 with minor modifications. In brief, 10ml of ice-cold erythrocyte wash solution (5mM Na2HPO4, pH 8.0, 0.9% NaCl, 1 ⫻ protease inhibitor cocktail) were added to 2.5ml thawed whole blood and subjected to centrifugation at 3,400g, 4°C for 10 minutes. Any intact erythrocytes were lysed by resuspension of the pellet in 2ml erythrocyte lysis solution (5mM Na2HPO4, pH 8.0, 1 ⫻ protease inhibitor cocktail) followed by centrifugation at 16,000g for 5 min- Expression of Mutant Chorein in Chorea-Acanthocytosis Cell Lines Western blot analysis of lymphoblastoid cell lysates shows that ChAc Patients 2 to 4 have markedly reduced chorein levels compared with the control (Fig 2A, see Table), in contrast with unaffected family members (lanes 3, 5, and 6) and MLS Patient 1 (lane 300 Annals of Neurology Vol 56 No 2 August 2004 isoform 1B (345kDa), which is theoretically unaffected by the exon 72 mutation. Chorein expression in primary skin fibroblasts from ChAc Patients 5 to 7 was undetectable or markedly reduced (see Table). Expression of Chorein in Erythrocyte Membranes Western blot analysis of erythrocyte membrane fractions from ChAc Patients 8 to 14, MLS Patient 2, and two healthy individuals is shown in Figure 2B. Chorein expression was undetectable or markedly reduced in all ChAc patients analyzed. In contrast, a sig- Fig 1. Detection of chorein expression. (A) Anti-chor1 antiserum detects endogenous chorein. Twenty-microgram protein samples from lymphoblastoid cell lines (C ⫽ healthy control, P1 ⫽ ChAc Patient 1) were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and analyzed by Western blot. Blots were analyzed with preimmune serum (lanes 1 and 2) or anti-chor1 antiserum (lanes 3–6). Anti-chor1 antiserum was either run through a column bound with chor1 antigen to remove chor1-binding antibodies (column-treated ⫹, lanes 3, 4) or left untreated (column-treated ⫺, lanes 5, 6). An anti-huntingtin blot (anti-Htt, lanes 7, 8) is provided for size comparison (Htt ⫽ 350kDa). Because of their similar size, it is not possible to distinguish between chorein isoforms 1A (360kDa) and 1B (345kDa). (B) Expression of chorein in different cell types. Human cell lines were derived from cervical carcinoma (HeLa), fetal lung (MRC5), embryonic kidney (293T), lymphoblasts (lymph), neuroglioma (H4), hepatocarcinoma (Hep3B), myelogenous leukemia (K562), and rhabdomyosarcoma (RD). Nonhuman cell lines were derived from African green monkey kidney (COS-7) and Chinese hamster ovary (CHO-KI). The much weaker signal in nonhuman cell lines is presumably caused either by a reduction in chorein ortholog expression or by limited cross-reaction with the antichor1 antiserum. Blot stripping and immunodetection of early endosome antigen 1 (EEA1) demonstrated equal loading of samples (data not shown). 8). The affected son in Family CHAC11 (Patient 4) has inherited a maternal 237del mutation in exon 4 and a paternal 9429_9432del mutation in exon 72. Any protein that is produced from the 237del allele will be severely truncated (predicted size 10kDa). The weak signal seen for Patient 4 therefore must derive from truncated protein generated by the 9429_9432del allele (357kDa), and/or other chorein isoforms, such as Fig 2. Analysis of chorein expression in ChAc and MLS patient cells. (A) Analysis of lymphoblastoid cell lines. Twentymicrogram protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and analyzed by Western blot. Cells were derived from ChAc patients (P1–7, lanes 1, 2, 4, 7), unaffected members of ChAc families (lanes 3, 5, 6), a McLeod syndrome patient (M1, lane 8), and a healthy control (C, lane 9). The VPS13A genotype is indicated above CHAC6 and CHAC11 family members (⫹, wild type; ⫺, mutant allele). Blot stripping and immunodetection of EEA1 demonstrated equal loading of samples (data not shown). (B) Analysis of erythrocyte membranes. Membranes were prepared from ChAc Patients 8 to 14 (P8 –14, lanes 1–7); MLS Patient 2 (M2, lane 8); and healthy controls (C1, C2, lanes 9, 10). Blood had been stored at ⫺20°C for different lengths of time, ranging from 8 weeks 4 days (C2) to 159 weeks 5 days (P14). M2 blood had been stored at ⫺192°C for 654 weeks 3 days. An arrow points to the band corresponding to chorein; a lower, cross-reacting band (asterisk) possibly corresponds to spectrin, the most abundant protein associated with the erythrocyte membrane.14 Immunodetection with anti–spectrin antibody supports this hypothesis (data not shown). Total protein staining of the blot with 0.1% Ponceau S solution before immunodetection demonstrated equal loading of samples (data not shown). Dobson-Stone et al: Chorein Detection in ChAc 301 nal comparable in strength with the controls was observed in MLS Patient 2 (lane 8), even though blood from this patient had been stored for over 12 years. Membrane fractions from two HD patients also showed normal levels of chorein expression (data not shown). Discussion Mutations in the VPS13A gene are associated with the severe neurodegenerative disorder chorea-acanthocytosis. In this study, we report the first cellular detection of chorein, the VPS13A gene product. We found that chorein is expressed in cell lines derived from a wide variety of human tissues, as well as primary skin fibroblasts and erythrocytes. This is supported by previous analyses that showed ubiquitous expression of VPS13A mRNA.4 We have demonstrated that 19 different VPS13A mutations, including one missense mutation, lead to absence or marked reduction of chorein expression in ChAc patients. Even in a patient with an isoform 1A– specific mutation (9429_9432del), very little chorein expression was observed. We observed previously that several patients with ChAc harbored isoform 1A–specific mutations and speculated that exons 70 to 73 were essential for some functions of chorein.7 The relatively poor expression of isoforms lacking these exons also may explain their inability to compensate for fulllength chorein in ChAc. We have shown that chorein can be detected in association with the erythrocyte membrane (see Fig 2B). MLS and HD patient samples show normal chorein expression levels, in contrast with ChAc patients. This has obvious implications for ChAc diagnosis. Currently, if the VPS13A gene has not been screened, ChAc can be diagnosed only by excluding other clinically similar disorders. As VPS13A is a large gene with many exons, screening is costly and time consuming. Western blotting of patient erythrocyte membranes perhaps could give an early indication of the disorder before a precise diagnosis is given by a VPS13A gene screen. Although chorein was absent or markedly reduced in all patient erythrocytes analyzed so far, one cannot exclude ChAc as a diagnosis if chorein is present in a sample. The missense mutation S1452P appears simply to affect chorein dosage (see Fig 2A); however, some substitutions may lead to apparently normal levels of chorein that is nevertheless functionally defective. Also, some mutations may allow almost normal expression levels of mutant chorein lacking only a few exons, which might not be resolved from wild-type chorein because of its large size. However, we anticipate that this would not be a significant problem, because missense changes compose less than 7% (5/75) of VPS13A 302 Annals of Neurology Vol 56 No 2 August 2004 mutations described so far (previous studies4,5,7 and this study), and most transcripts containing premature stop codons are believed to be rapidly degraded.12 In any case, if chorein is absent in a patient sample, the most likely diagnosis will be chorea-acanthocytosis. This work was supported by the Wellcome Trust (060886/Z/00/ Z,C.D.-S., 045093/Z195, A.P.M.) and the Marie Curie postdoctoral fellowship (QLGA-CT-2001-51850, A.V.-B.). We thank the patients and their families for their participation, L. Lonie for denaturing high-performance liquid chromatography analysis, and all colleagues who contributed patient samples, notably S. Carrè, X. Ferrer, A. Lees, A. Németh, J. Palace, G. Rudolf, and C. Verellen. References 1. Rampoldi L, Danek A, Monaco AP. Clinical features and molecular bases of neuroacanthocytosis. J Mol Med 2002;80: 475– 491 2. Feinberg TE, Cianci CD, Morrow JS, et al. Diagnostic tests for choreoacanthocytosis. Neurology 1991;41:1000 –1006. 3. Danek A, Rubio JP, Rampoldi L, et al. McLeod neuroacanthocytosis: genotype and phenotype. Ann Neurol 2001;50:755–764. 4. Rampoldi L, Dobson-Stone C, Rubio JP, et al. A conserved sorting-associated protein is mutant in chorea-acanthocytosis. Nat Genet 2001;28:119 –120. 5. Ueno S, Maruki Y, Nakamura M, et al. The gene encoding a newly discovered protein, chorein, is mutated in choreaacanthocytosis. Nat Genet 2001;28:121–122. 6. Velayos-Baeza A, Vettori A, Copley RR, et al. Analysis of the human VPS13 gene family. Genomics (in press). 7. Dobson-Stone C, Danek A, Rampoldi L, et al. Mutational spectrum of the CHAC gene in patients with choreaacanthocytosis. Eur J Hum Genet 2002;10:773–781. 8. Rubio JP, Danek A, Stone C, et al. Chorea-acanthocytosis: genetic linkage to chromosome 9q21. Am J Hum Genet 1997; 61:899 –908. 9. Frangioni JV, Neel BG. Solubilization and purification of enzymatically active glutathione S-transferase (pGEX) fusion proteins. Anal Biochem 1993;210:179 –187. 10. Ausubel FM, Brent R, Kingston RE, et al, eds. Current protocols in molecular biology. Vol 2. New York: Wiley, 1997. 11. Dodge JT, Mitchell C, Hanahan DJ. The preparation and chemical characteristics of hemoglobin-free ghosts of human erythrocytes. Arch Biochem Biophys 1963;100:119 –130. 12. Hentze MW, Kulozik AE. A perfect message: RNA surveillance and nonsense-mediated decay. Cell 1999;96:307–310. 13. Ho M, Chelly J, Carter N, et al. Isolation of the gene for McLeod syndrome that encodes a novel membrane transport protein. Cell 1994;77:869 – 880. 14. Marchesi VT. The cytoskeletal system of red blood cells. Hosp Pract 1985;20:113–118, 120, 125–131.