Movement Disorders Vol. 18, No. 9, 2003, pp. 1041–1056 © 2003 Movement Disorder Society Brief Report TorsinA Immunoreactivity in Inclusion Bodies in Trinucleotide Repeat Diseases bodies in Parkinson’s disease5,6 and in intranuclear neuronal inclusions in autosomal dominant chorea-acanthocytosis (ADChAc).7 ADChAc has recently been identified as being due to the mutation causing Huntington’s disease-like 2 (HDL2)8 (Walker et al., submitted), with a trinucleotide expansion of the junctophilin-3 gene on chromosome 16.9 Intranuclear inclusions containing ubiquitin, expanded polyglutamine tracts, and other proteins are found in cortex in Huntington’s disease (HD) and in the brainstem in spinocerebellar ataxia III (SCA III). The role of inclusion bodies in these and other disorders remains to be determined. It has been hypothesized that they may protect neurons, possibly by sequestering dysfunctional proteins; alternatively, they may be an indication of neuronal dysfunction. We used double-labeling immunohistochemistry in cases of HD and SCA III to determine whether these inclusions also contained torsinA, and to investigate the relationship of torsinA to other inclusion body markers in the previously reported ADChAc/HDL2 case.7 Ruth H. Walker, MB, ChB, PhD,1,2* Paul F. Good, PhD,3 and P. Shashidharan, PhD2 1 Department of Neurology, Veterans Affairs Medical Center, Bronx, New York, USA 2 Department of Neurology, Mount Sinai School of Medicine, New York, New York, USA 3 Department of Pathology, Mount Sinai School of Medicine, New York, New York, USA Abstract: A mutation of the DYT1 gene, which codes for torsinA, has been identified as a cause of autosomal dominantly inherited dystonia. The function of torsinA is not yet known, but it is found throughout the central nervous system and has been identified in Lewy bodies in Parkinson’s disease. We examined cases of Huntington’s disease, spinocerebellar ataxia type III, and Huntington’s diseaselike 2 using antibodies to torsinA, and found that ubiquitinated, intranuclear neuronal inclusions were torsinA-immunoreactive, possibly indicating a role for torsinA in protein degradation. © 2003 Movement Disorder Society MATERIALS AND METHODS Tissue blocks were taken from 2 cases of genetically confirmed HD, 3 cases of SCA III, and from the proband reported previously with ADChAc/HDL2,7 who had 51/13 trinucleotide repeats in the junctophilin-3 gene (normal ⬍40 repeats) (Walker et al., submitted). In the HD and ADChAc/HDL2 cases, neocortex was examined; in SCA III cases, the midbrain. Five-micron paraffin sections were deparaffinized, processed for antigen retrieval and incubated at room temperature for 2 hours with rabbit polyclonal antibody to torsinA2 (1:100) and mouse monoclonal antibody to expanded polyglutamine repeats (1:1,000) (Chemicon, Temecula, CA) or to ubiquitin (1:500; Chemicon). After washing, sections were incubated with Alexa-488 anti-mouse and Alexa-594-labeled anti-rabbit secondary antibodies (1:500; Molecular Probes, Eugene, OR). Sections were then washed, mounted, coverslipped using Vectorshield (Vector Laboratories, Burlingame, CA), examined, and imaged using a Leica confocal fluorescence microscope. For controls, primary antibodies were omitted. Key words: torsinA; inclusion bodies; trinucleotide repeats Dystonia is a neurological disorder characterized by involuntary excessive muscle contractions causing twisting movements and abnormal postures. A 3-bp (GAG) deletion of the DYT1 gene on chromosome 9q34 causes one type of inherited dystonia.1 TorsinA, the protein product of the DYT1 gene, shares homology with the heat shock family of proteins1 and binds ATP,2 however, its function is not yet known. Immunoreactivity to torsinA antibodies is seen throughout the central nervous system (CNS) in all examined species2– 4 and in several peripheral organs.2 Additionally, torsinA is found in Lewy *Correspondence to: Ruth H. Walker, MB, ChB, PhD, Department of Neurology, Box 1137, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029. E-mail: ruth.walker@mountsinai.org Received 21 August 2002; Revised 29 October 2002, 28 January 2003; Accepted 12 February 2003 1041 1042 R.H. WALKER ET AL. FIG. 1. Confocal double-labeling fluorescence immunohistochemistry in HD (motor cortex) (A–L), SCA III (oculomotor nucleus) (M–O) and ADChAc/HDL2 (parahippocampal gyrus) (P–R) for expanded polyglutamine repeats (A,D,G,J,M), ubiquitin (P) and torsinA (B,E,H,K,N,Q) showing variable patterns of doublelabeling (C,F,I,L,O,R). Arrows indicate torsinA immunoreactivity in inclusion bodies. Arrowheads in (F) and (R) indicate autofluorescent lipofuscin granules. Scale bar ⫽ 20 ␮m Movement Disorders, Vol. 18, No. 9, 2003 TORSINA IN INCLUSION BODIES RESULTS In HD and SCAIII, in addition to the ADChAc/HDL2 case as reported previously,7 torsinA immunoreactivity was noted in intranuclear intraneuronal inclusion bodies (Fig. 1B,E,H,K,N,Q). This was not found in normal controls. These inclusions were immunoreactive also for expanded polyglutamine repeats (Fig. 1A,D,M) and for ubiquitin (Fig. 1P) as confirmed by superposition of the confocal images (Fig. 1C,F,O,R). The relationship between the torsinA and polyglutamine labeling was variable. In some cases, the torsinA immunoreactivity seemed to be adjacent to or surrounding the polyglutamine immunoreactivity (Fig. 1C,F), in others the torsinA immunoreactivity was surrounded by polyglutamine immunoreactivity (Fig. 1O). In some cases, inclusions were labeled intensely with one antibody but with negligible labeling by the other (Fig. 1G–I). When there were polyglutamine-immunoreactive cytoplasmic aggregations, some of these, but not all, were also torsinA immunoreactive (Fig. 1L). When primary antibodies were omitted, there was no fluorescent labeling and inclusion bodies were not visible (data not shown). DISCUSSION The function of torsinA is as yet unknown, but may be suggested by its homology to the heat shock family of proteins,1 which may play a role in neuroprotection. The distribution of torsinA immunoreactivity in neuronal nuclei, perikaryal cytoplasm and axon terminals throughout the brain2 suggests that torsinA plays an important role in neuronal function. Overexpression of torsinA with the GAG deletion in cultured mammalian cells resulted in aggregations of the protein associated with the endoplasmic reticulum (ER),10,11 although this was not seen when torsinA with the 18-bp deletion12 was transfected.13 In human brain tissue from cases of DYT1 dystonia, there was no evidence of aggregations of torsinA and no association with the ER.14 This suggests that, whatever the nature of the dysfunction of torsinA in human disease, it is not reflected neuroanatomically at the light microscopic level. Intraneuronal inclusions of different types characterize many neurodegenerative diseases. We confirm the presence of torsinA in inclusions in ADChAc/HDL2 and demonstrate variable colocalization with ubiquitin and expanded polyglutamine tracts in this and two other expanded trinucleotide repeat diseases. These inclusions have been shown to contain many proteins, including some known to be involved in proteolysis, such as ubiquitin and heat shock proteins. The variable immunoreactivity for some of these proteins in SCA III has been interpreted as implicating disruption of proteolytic func- 1043 tion as the cause of the inclusions.15 It is possible that this may also be the explanation for the variable pattern of torsinA immunoreactivity in inclusion bodies. Alternatively, it may be that aggregation of this protein is sporadic and only intermittently related to inclusion body formation. Another explanation is that the antigenic epitope may be masked by binding to other proteins, or by changes in conformation of torsinA. We have previously described torsinA immunoreactivity in Lewy bodies in Parkinson’s disease,5 in which a close relationship between torsinA and ␣-synuclein has been postulated to indicate a functional association between the two proteins.6 The presence of torsinA in ubiquitinated intranuclear inclusions in neurodegenerative disease may shed light upon the function of torsinA, suggesting that it may play a chaperone-type role in protein folding or proteolysis. An alternative possibility is that it is merely being sequestered with other proteins as the result of neuronal dysfunction. Further work on the function of torsinA will determine the significance of this finding. Note Added in Proof A role for torsinA in protein folding and proteolysis is supported by recent work by Caldwell et al. (Hum Mol Genet 2003;12:317–319), who found that overexpression of torsinA suppressed polyglutamine-induced aggregates in C. elegans. This effect, which was interpreted as being neuroprotective, was lost when torsinA with the GAG deletion was overexposed. Acknowledgments: This work was supported by the Bachmann-Strauss Dystonia and Parkinson Foundation, and the National Institutes of Health (NS43038 to P.S.). Confocal laser scanning microscopy was carried out at the MSSM-Microscopy Center, supported with funding from an NSF Major Research Instrumentation grant (DBI-9724504). Tissue was obtained from the Harvard Brain Tissue Resource Center (supported in part by PHS MH/NS 31862), the National Neurological Research Specimen Bank, VAMC, Los Angeles, CA, and the Manhattan HIV Brain Bank (R24MH59724). REFERENCES 1. Ozelius LJ, Hewett JW, Page CE, et al. The early-onset torsion dystonia gene (DYT1) encodes an ATP binding protein. Nat Genet 1997;17:40 – 48. 2. Shashidharan P, Kramer BC, Walker RH, Olanow CW, Brin MF. Immunohistochemical localization and distribution of torsinA in normal human and rat brain. Brain Res 2000;853:197–206. 3. Konakova M, Huynh DP, Yong W, Pulst SM. Cellular distribution of torsin A and torsin B in normal human brain. Arch Neurol 2001;58:921–927. 4. Walker RH, Brin MF, Sandu D, Gujjari P, Olanow CW, Shashidharan P. Distribution and immunohistochemical characterization of torsinA immunoreactivity in rat brain. Brain Res 2001;900:348 – 354. Movement Disorders, Vol. 18, No. 9, 2003 1044 K.E. ZEUNER AND M. HALLETT 5. Shashidharan P, Good PF, Hsu A, Perl D, Brin MF, Olanow CW. TorsinA accumulation in Lewy Bodies in sporadic Parkinson’s disease. Brain Res 2000;877:379 –381. 6. Sharma N, Hewett J, Ozelius LJ, Ramesh V, McLean PJ, Breakefield XO, Hyman BT. A close association of torsinA and alphasynuclein in Lewy bodies: a fluorescence resonance energy transfer study. Am J Pathol 2001;159:339 –344. 7. Walker RH, Morgello S, Davidoff-Feldman B, Melnick A, Walsh MJ, Shashidharan P, Brin MF. Autosomal dominant chorea-acanthocytosis with polyglutamine-containing neuronal inclusions. Neurology 2002;58:1031–1037. 8. Margolis RL, O’Hearn E, Rosenblatt A, et al. A disorder similar to Huntington’s disease is associated with a novel CAG repeat expansion. Ann Neurol 2001;50:373–380. 9. Holmes SE, O’Hearn E, Rosenblatt A, et al. A repeat expansion in the gene encoding junctophilin-3 is associated with Huntington disease-like 2. Nat Genet 2001;29:377–378. 10. Kustedjo K, Bracey MH, Cravatt BF. Torsin A and its torsion dystonia-associated mutant form are lumenal glycoproteins that exhibit distinct subcellular localizations. J Biol Chem 2000;275: 27933–27939. 11. Hewett J, Gonzalez-Agosti C, Slater D, et al. Mutant torsinA, responsible for early-onset torsion dystonia, forms membrane inclusions in cultured neural cells. Hum Mol Genet 2000;9:1403– 1413. 12. Leung JC, Klein C, Friedman J, et al. Novel mutation in the TOR1A (DYT1) gene in atypical early onset dystonia and polymorphisms in dystonia and early-onset parkinsonism. Neurogenetics 2001;3:133–143. 13. O’Farrell C, Hernandez DG, Evey C, Singleton AB, Cookson MR. Normal localization of DeltaF323-Y328 mutant torsinA in transfected human cells. Neurosci Lett 2002;327:75– 78. 14. Walker RH, Brin MF, Sandu D, Good PF, Shashidharan P. TorsinA immunoreactivity in brains of patients with DYT1 and non-DYT1 dystonia. Neurology 2002;58:120 –124. 15. Schmidt T, Lindenberg KS, Krebs A, et al. Protein surveillance machinery in brains with spinocerebellar ataxia type 3: redistribution and differential recruitment of 26S proteasome subunits and chaperones to neuronal intranuclear inclusions. Ann Neurol 2002; 51:302–310. Sensory Training as Treatment for Focal Hand Dystonia: A 1-Year Follow-Up Kirsten E. Zeuner, MD, and Mark Hallett, MD* The Human Motor Control Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland, USA Abstract: In a prior study, 10 patients with focal hand dystonia learned braille reading as sensory training for 8 weeks. Practice time was 30 to 60 minutes daily. They improved both their spatial acuity using the Grating Orientation Discrimination Task (GOT) and their dystonia using the Fahn scale. Three patients continued training for 1 year. Patients showed further improvement in the GOT, writing a standard paragraph, and self-rating scales. Sensory training lasting longer than 8 weeks may lead to continued improvement. © 2003 Movement Disorder Society Key words: focal dystonia; sensory training; tactile spatial acuity Task-specific dystonia may originate from an impaired sensory system.1 One recent hypothesis is that stereotyped, repetitive inputs, such as those of musicians, could result in enlarged receptive fields and de-differentiation of representation in the sensory cortex in humans, which would lead to a motor disorder.2,3 Patients with focal hand dystonia have decreased performance in tactile spatial acuity and localization compared to controls.4,5 Therefore, sensory training as treatment was previously suggested, and we recently showed that braille reading for 8 weeks used as sensory training significantly improved spatial acuity in patients with focal hand dystonia and normal volunteers.6 The mean value for the Grating Orientation Discrimination Task (GOT) and standard deviation in patients at baseline was 2.38 ⫾ 1.09 mm and decreased to 1.75 ⫾ 0.69 mm after 8 weeks of practice. Clinical improvement in dystonia was significant in the Fahn dystonia scale7 (5 patients improved, 4 patients showed no change; the mean value and standard deviation increased from 41.07% ⫾ 14.25 to 45.45% ⫾ 15.84 after 8 weeks of practice). The time in writing a standard paragraph decreased nonsignificantly from 159.43 ⫾ *Correspondence to: Mark Hallett, M.D., NIH, NINDS, HMCS, Building 10, Room 5N226, 10 Center Drive MSC1428, Bethesda, Maryland 20892-1428. E-mail: hallettm@ninds.nih.gov Received 10 October 2002; Revised 10 February 2003; Accepted 17 February 2003 Movement Disorders, Vol. 18, No. 9, 2003 SENSORY TRAINING AS TREATMENT 1045 TABLE 1. Results of 3 patients after 1 year of sensory training Patient no. Age (yr) 1 54 2 45 3 48 Evaluation Baseline 8 wk 20 wk 1 yr GOT (mm) Standard paragraph min) Fahn scale (%) Visual analogue scale, % Verbal scale GOT (mm) Standard paragraph (min) Fahn scale (%) Visual analogue scale, % Verbal scale GOT (mm) Standard paragraph (min) Fahn scale (%) Visual analogue Scale, % Verbal scale 2.5 3.30 50 2.5 1.50 50 0 None 1.07 1.40 55 20 Minimal 1.5 NA 60 30 Minimal 1.0 1.24 50 10 Minimal 1.00 1.40 55 15 Mild 1.1 NA 60 20 Mild 1.75 1.18 55 10 Minimal 0.92 1.30 55 40 Moderate 1.1 4.50 60 50 Moderate 1.32 2.54 50 2.3 NA 60 GOT, Grating Orientation Discrimination Task; NA, not able. 44.06 to 129.13 ⫾ 45.98 seconds (6 patients improved, 2 patients remained unchanged). The visual analogue and verbal scale for dystonia did not improve.6 We present results of 3 patients who continued sensory training for a 1-year period. PATIENTS AND METHODS Two right-handed and one left-handed female patients self-selected from our earlier study and decided to continue practicing braille reading after the formal 8-week period was finished. The diagnosis of focal hand dystonia was made by the medical history and neurological examination. They received no botulinum toxin injections for at least 3 months before they started braille reading for our previous study. All patients mainly had difficulty writing; Patient 2 also noticed problems with other fine motor tasks. One patient was originally left-handed and developed writer’s cramp on her left side. She switched to her right hand and has had writer’s cramp bilaterally for 10 years. Spatial acuity was evaluated on the dominant hand with the GOT that uses eight hemispheric plastic JVP-Domes (Johnson-Van Boven-PhillipsDomes) with grating of equal ridge and gap widths cut into their faces with gap values of 3.0, 2.0, 1.5, 1.2, 1.0, 0.75, 0.5, and 0.35 mm (Stoelting, Wood Dale, IL).8 The orientation of the gaps was presented in decreasing order randomly either vertically or horizontally 20 times on the tip of the finger, and patients had to identify the direction. The Fahn dystonia scale and verbal and visual analogue self-rating scales were used for clinical evaluation. The Fahn dystonia scale includes questions about the severity of dystonia. Scores between 90 and 100% indi- cate no limitations of activities. If there were any limitations of functional activities, the patient had to check items on a list of seven questions. The lower the score the more severe was the hand dystonia.7 In our previous study, the average disability score at baseline was 41.1%. The scores of the three patients are presented in Table 1. The time needed to write a standard paragraph was recorded. The verbal scale had four levels ranging from 0 ⫽ no improvement to 3 ⫽ major improvement; the visual analogue scale ranged from 0% ⫽ no improvement to 100% ⫽ full recovery. All tests were done at baseline, after 20 weeks, and 1 year. Patients trained in braille reading with standardized books from the American Printing House for the Blind.9 Practice time was between 30 and 60 minutes daily, and patients were blindfolded while practicing. The lefthanded patient was trained with her left hand only. Patients were asked to practice with their finger clinically most affected and then continue with the fingers on either side of it. Fingers were always trained individually. Patients were supervised daily in the first week, twice a week in the second and third weeks, weekly until week 20, monthly until 6 months, and then every 3 months. They practiced at home with specific exercises that were given to them with printed copies from the Braille Connection Teacher’s Edition so that every patient could monitor her own mistakes. RESULTS Patients further improved in the GOT, writing a standard paragraph, and self-rating scales (Table 1). Dystonia measured clinically with the Fahn-Scale remained at the same level in all patients during follow-up. Patient 1 Movement Disorders, Vol. 18, No. 9, 2003 1046 K.E. ZEUNER AND M. HALLETT FIG. 1. This is a writing example of Patient 2. At baseline, she needed 2:54 minutes, and after 1 year, 1:30 min for the same standard paragraph. A: A section of the standard paragraph at baseline. B: Writing after 1 year. It is apparent from this figure that, after 1 year of sensory training, writing is more fluent (see letters M, s, and h), and less pressure was used. showed fluctuations in her performance in the GOT. Her threshold was lower after week 20 than after 1 year but was clearly better compared to her baseline threshold. She thought there was only minimal improvement and no further improvement after week 20. It should be noted that she stopped working several years ago and, therefore, did not use her hand for writing as much as the other two patients. Her results differed from Patients 2 and 3, who continuously improved in writing and spatial acuity. Patient 2 decided twice to go back to practice after a hiatus of several weeks, because she clearly noted improvement after practice. Of interest, she noticed that initially the effect diminished 2 weeks after she stopped practicing. After practicing for 1 year, the effect lasted for approximately 4 to 6 weeks without practice, and then her writing worsened. Her writing examples are given in Figure 1. Patient 3 could only write a few words initially. She was not able to hold the pen and was unable to keep the pen on the paper. She could write the standard paragraph after 6 months of practice. Patient 3 also noticed worsening of her writing after she paused her practice for several weeks. Movement Disorders, Vol. 18, No. 9, 2003 DISCUSSION Recently, we showed that spatial acuity improves in patients with focal hand dystonia after sensory training for 8 weeks.6 It is known that blind braille readers have a better threshold in the GOT than normal controls.10 Therefore, we concluded that sensory training such as braille reading is a method to overcome the sensory dysfunction, which possibly results from de-differentiation in the sensory cortex after repetitive inputs. The improvement in dystonia was mainly a statistical improvement not noticed by most of the patients.6 Apparently spatial acuity improves before the writing does. Three patients continued to practice for 1 year. Patient 1 showed no further improvement in the GOT, but her writing was better, and she graded her overall improvement as being minimal. Because she did not test her writing daily and as regularly as the two other patients did, it was more difficult for her to observe changes in her writing performance. Patients 2 and 3 noticed further improvement after 6 months to 1 year of regular practice. They both improved GENETIC HETEROGENEITY IN MYOCLONUS–DYSTONIA further or remained at the same level as after week 20 in the GOT; the writing speed improved in Patient 2. Patient 3 could write a standard paragraph after 6 months of practice; the effect lasted up to 1 year. Both patients have to write frequently in their daily jobs. The weaknesses of this study are the small number of patients and there was no control group. That the improvement in writing diminished after practice was temporarily stopped is only based on the patients’ reports; we cannot provide data for these changes. A placebo effect cannot be ruled out, although the GOT and the time recorded for writing a standard paragraph are objective measurements. It appears that practice needs to last longer than 8 weeks for clinical improvement to be noticed by the patients. In addition, dystonia patients have to practice regularly because the effect diminishes after they stop. However, according to Patient 2, the benefit lasts longer even without practice after regular practice for at least 6 months. In conclusion, the 1-year follow-up indicated that sensory training longer than 8 weeks may produce continuing improvement in writing in some patients. Length of regular practice may be needed to result in changes such as re-differentiation in the sensory cortex. Braille reading is a practical method for treating focal dystonia in some patients. However, the patient needs to be motivated to practice regularly everyday over a long period of time. Acknowledgments: We thank the patients for participating in this study. We also thank Devera G. Schoenberg, MSc, for skillful editing. REFERENCES 1. Hallett M. Is dystonia a sensory disorder? Ann Neurol 1995;38: 139 –140. 2. Byl NN, Merzenich MM, Cheung S, Bedenbaugh P, Nagarajan SS, Jenkins WM. A primate model for studying focal dystonia and repetitive strain injury: effects on the primary somatosensory cortex. Phys Ther 1997;77:269 –284. 3. Byl NN, Melnick M. The neural consequences of repetition: clinical implications of a learning hypothesis. J Hand Ther 1997;10: 160 –174. 4. Bara-Jimenez W, Shelton, Hallett M. Spatial discrimination is abnormal in focal hand dystonia. Neurology 2000;55:1869 – 1873. 5. Sanger TD, Tarsy D, Pascual-Leone A. Abnormalities of spatial and temporal sensory discrimination in writer’s cramp. Mov Disord 2001;16:94 –99. 6. Zeuner KE, Bara- Jimenez W, Noguchi PS, Goldstein SR, Dambrosia JM, Hallett M. Sensory training for patients with focal hand dystonia. Ann Neurol 2002;51:593–598. 7. Fahn S. Assessment of the primary dystonias. In: Munsat T, editor. The quantification of neurologic deficit. Boston: Butterworths; 1989. p 241–270. 1047 8. Van Boven RW, Johnson KO. The limit of tactile spatial resolution in humans: grating orientation discrimination at the lip, tongue, and finger. Neurology 1994;44:2361–2366. 9. Caton H. The braille connection. Louisville: The American Printing House for the Blind, Inc.; 1997. 10. Van Boven RW, Hamilton RH, Kauffman T, Keenan JP, PascualLeone A. Tactile spatial resolution in blind braille readers. Neurology 2000;54:2230 –2236. Analysis of the ⑀-Sarcoglycan Gene in Familial and Sporadic Myoclonus-Dystonia: Evidence for Genetic Heterogeneity Enza-Maria Valente, MD,1 Anjum Misbahuddin, MRCP,2 Francesco Brancati, MD,1,3 Mark R. Placzek, BSc,2 Barbara Garavaglia, PhD,4 Sergio Salvi, BSc,1 Andrea Nemeth, PhD, MRCP,5 Charles Shaw-Smith, FRCP,5 Nardo Nardocci, MD,6 Anna-Rita Bentivoglio, MD, PhD,7 Alfredo Berardelli, MD,8 Roberto Eleopra, MD,9 Bruno Dallapiccola, MD,1,3 and Thomas T. Warner, PhD, FRCP2* 1 C.S.S. Mendel Institute, San Giovanni Rotondo and Rome, Italy; 2Department of Clinical Neurosciences, Royal Free and University College Medical School, London, United Kingdom; 3Department of Experimental Medicine and Pathology, University La Sapienza, Rome, Italy; 4Department of Biochemistry and Genetics, National Neurologic Institute Carlo Besta, Milan, Italy; 5Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom; 6Department of Child Neurology, National Neurologic Institute Carlo Besta, Milan, Italy; 7Department of Neurology, Catholic University, Rome, Italy; 8Department of Neurological Sciences and NEUROMED Institute, University La Sapienza, Rome, Italy; 9 Department of Clinical Neurosciences, University Hospital of Ferrara, Italy Abstract: The ⑀-sarcoglycan gene (SGCE) on human chromosome 7q21 has been reported to be a major locus for inherited myoclonus– dystonia. Linkage to the SGCE locus has been detected in the majority of families tested, and Drs. Valente and Misbahuddin contributed equally to this study. *Correspondence to: Thomas T. Warner, Ph.D., F.R.C.P., Department of Clinical Neurosciences, Royal Free and University College Medical School, Rowland Hill Street, London NW3 2PF United Kingdom. E-mail: twarner@rfc.ucl.ac.uk Received 28 June 2002; Revised 9 October 2002; Accepted 10 February 2003 Movement Disorders, Vol. 18, No. 9, 2003 1048 E.-M. VALENTE ET AL. mutations in the coding region have been found recently in families with autosomal dominant myoclonus– dystonia. To evaluate the relevance of SGCE in myoclonus– dystonia, we sequenced the entire coding region of the ⑀-sarcoglycan gene in 16 patients with either sporadic or familial myoclonus– dystonia. No mutations were found. This study suggests that ⑀-sarcoglycan does not play an important role in sporadic myoclonus– dystonia and supports genetic heterogeneity in familial cases. © 2003 Movement Disorder Society Key words: myoclonus-dystonia syndrome; myoclonus, dystonia; ⑀-sarcoglycan; genetic heterogeneity The nosology of conditions incorporating both myoclonic jerks and torsion dystonia has been the subject of considerable debate over the years.1,2 Classic myoclonus– dystonia is characterized by myoclonus usually involving the arms and axial muscles, often accompanied by dystonia of the neck and arms. Onset of the disorder is mostly in childhood or the teenage years and many patients find their symptoms improve with alcohol. However, cases with onset in adult life have been described, and alcohol does not always have a beneficial effect.1,3 Myoclonus– dystonia is inherited as an autosomal dominant trait with incomplete penetrance that appears to occur predominantly if the disease allele is passed on by the mother, suggesting maternal imprinting. Obsessive– compulsive behavior and panic attacks are also reported to occur in some families.4 Sporadic cases have also been reported. A locus for myoclonus– dystonia has been mapped to chromosome 7q21. Seventeen families linked to the chromosome 7q21 locus have been reported so far. The phenotype was variable with age of onset up to 38 years and with different combinations of myoclonic jerks and dystonic movements and postures, usually confined to the upper body but occasionally involving the lower limbs as well. Alcohol responsiveness and psychiatric disturbances were observed only in a subset of patients.5– 8 Subsequently, mutations within the ⑀-sarcoglycan (SGCE) gene in this region have been identified. Six families with mutations in the SGCE gene were reported initially who have a more uniform phenotype, with onset in the first two decades of life, more marked myoclonus than dystonia, upper body involvement, and, in most cases, good response to alcohol consumption.9 ⑀-Sarcoglycan was initially discovered due to its 50% sequence homology to ␣-sarcoglycan, a major constituent of the dystrophin– glycoprotein complex (DGC) in striated muscle along with ␤-, ␥-, and ␦-sarcoglycan.10 Mutations in these four genes cause autosomal recessive limb girdle muscular dystrophies. Movement Disorders, Vol. 18, No. 9, 2003 ⑀-Sarcoglycan forms part of the DGC in smooth muscle where ␣-sarcoglycan is not found11 and is part of the sarcoglycan complex in the Schwann cell membrane.12 SGCE is also widely expressed in several areas of the brain.9 Mouse embryonic studies show that SGCE expression can be detected in a wide distribution early in development.10,11 A previous study identified a single family with a Val154Ile substitution in the DRD2 gene.13 However, mutations in the DRD2 genes have subsequently been excluded in a large number of familial and sporadic cases.14 –16 In addition, the original family has been reassessed recently and a mutation in the SGCE gene identified.17 Although the SGCE gene seems to be the major gene responsible for autosomal dominant myoclonus– dystonia, its role in sporadic cases and genotype–phenotype correlation have not been studied. The purpose of this study was to assess the role of SGCE mutations in a cohort of 16 sporadic and familial cases presenting with both dystonia and myoclonus by sequencing the coding region of the SGCE gene. PATIENTS AND METHODS Sixteen patients with myoclonus– dystonia were selected for this study. Each patient was assessed by a neurologist with a specialist interest in movement disorders, and diagnosis was made according to published criteria.3 Personal and family history was collected, and, for familial cases, a detailed family tree was constructed. Neurological examination was performed, and the anatomical involvement of both myoclonus and dystonia was recorded. After obtaining informed consent, each patient donated a blood sample. DNA was extracted using the Nucleon BACC1 kit (Amersham Life Science). All patients previously had tested negative for the GAG or 18-bp deletion in the DYT1 gene.18,19 Polymerase chain reaction (PCR) reactions in a volume of 25 ␮l were performed for each of the 11 exons of SGCE using 0.625 units of Extensor Hifidelity DNA polymerase mix (Abgene) and 40 pmol of each primer (primer sequences and PCR conditions given in Table 1). Reactions were each of 30 cycles. PCR products were cleaned using the QIAquick PCR purification kit (QIAgen), and sequencing was performed in 10-␮l reactions using the Big Dye Terminator cycle sequencer reagent (Applied Biosystems) and 1.6 pmol of primer. Sequencing reactions were analysed on an ABI 3100 using dedicated software. Intron/exon boundaries were sequenced in all cases. Twenty to thirty base pairs 5⬘ and 3⬘ of each exon were sequenced. GENETIC HETEROGENEITY IN MYOCLONUS–DYSTONIA 1049 TABLE 1. Sequences and polymerase chain reaction conditions Annealing temperature Exon Primer sequences 1 F 5⬘ ACACCAACCGAAGTTGAAGCGCGTGTAC 3⬘ R 5⬘ CGCTCAGGCGCCCGGAACC 3⬘ 65 2 F 5⬘ TGAATTATCAAGGGCGTATCTCA 3⬘ R 5⬘ TTAGACCATTTGAAATAATGTTAATGA 3⬘ 50 3 F 5⬘ TTGATTGAAACTACCAAAGCAAC 3⬘ R 5⬘ GTATAGTTTTGCTCTTTCTAGGTG 3⬘ 58 4 F 5⬘ TCATTGCCCAGAGAAGGAAT 3⬘ R 5⬘ CATCAGTTATATTAGGTATGTGGCATT 3⬘ 59 5 F 5⬘ ATGCCCTTTTTCACCAAAATTAG 3⬘ R 5⬘ GCAATAGGCCATCTTCCATCTAT 3⬘ 61 6 F 5⬘ TCCTGCTTTTAAGGTGGATTGTT 3⬘ R 5⬘ CAAACGTTAACTCCAGCCACAT 3⬘ 61 7 F 5⬘ TTTGCAACGATTAATTTGTTGTGT 3⬘ R 5⬘ TGAAACTTTCGTTTTAATGGAATC 3⬘ 59 8 F 5⬘ GACAATGTCAGCATTTCCACATC 3⬘ R 5⬘ AGTTTTAGTTTCTACCCCTCCTAAA 3⬘ 58 9 F 5⬘ AAATTGATGACCCATCAGGCTAA 3⬘ R 5⬘ CAACAGAAAGCTCTGTTCTTTACA 3⬘ 58 10 F 5⬘ CATGACTGGGGTCATAGTTTACC 3⬘ R 5⬘ CACAAGTGTTTTGCCTTATTTGG 3⬘ 60 11 F 5⬘ GAAGATGGAAACTTTCTCCTT 3⬘ R 5⬘ GGAATGAGAATGAACACATA 3⬘ 47 RESULTS The clinical features of the 16 patients are shown in Table 2. Nine patients were Italian and seven were British. Ten cases were apparently sporadic. Three had a definite positive family history with autosomal dominant transmission but no evidence of imprinting. Family trees of these familial cases are shown in Figure 1. Three patients had a possibly affected relative. Age of onset varied from 1 to 41 years (mean ⫾ SD, 15.2 ⫾ 13.1 years), and symptoms started in the arm or TABLE 2. Clinical features of patients Patient no. Sex/age (yr) Age of onset (yr) Alcohol sensitivity Family history 1 M/20 10 Upper limbs Limbs and cranial-cervical Unknown All limbs Trunk, all limbs Upper limbs All limbs Upper limbs Upper limbs and trunk Cranial-cervical Upper limbs Cranial-cervical and upper limbs Cervical and writer’s cramp Bilateral writer’s cramp Upper limbs and trunk Yes Unknown Yes No Yes Unknown 41 8 13 14 10 5 11 Cervical and trunk Cervical and right arm Cervical and right leg Right arm Cervical and upper limbs Cervical, trunk and upper limbs Cervical and upper limbs Cervical and trunk Cervical and right arm Writer’s cramp, right leg Writer’s cramp Neck, writer’s cramp All limbs and trunk Cervical and upper limbs Yes Yes Yes No Yes Unknown Yes 4 10 Cervical, upper limbs, left leg Cervical and upper limbs Cervical, upper limbs, left leg Cervical and upper limbs Unknown Unknown Possible (brother) No No No Yes No Possible (father) No No Yes No No No Possible (father) No Yes 2 3 4 5 6 7 F/53 M/16 M/53 M/38 F/32 F/23 41 9 38 22 1 7 8 9 10 11 12 13 14 F/68 M/36 F/33 F/23 M/60 M/23 M/27 15 16 M/13 M/36 Distribution of myoclonus Distribution of dystonia Movement Disorders, Vol. 18, No. 9, 2003 1050 E.-M. VALENTE ET AL. FIG. 1. Simplified pedigrees of the three families (index case nos. 5, 10, and 16 in Table 2) with autosomal dominant myoclonus– dystonia. Black symbols denote definitely affected individuals, deceased members are marked with a diagonal bar. A thin horizontal bar above symbols indicates members of the family who were clinically examined and whose blood was sampled. neck in almost all cases. All patients showed a combination of dystonia and myoclonus. Eight of 10 patients who had tried alcohol showed marked benefit. Psychiatric disturbances were not reported, with the exception of mild depression in two cases. Patient 5 developed jerks of the limbs in adult life plus dystonic posturing of the upper limbs. The striking feature was the intermittent nature of the symptoms with episodes of normality lasting hours. Sequencing of the coding region of SGCE in all patients did not identify any mutation or significant change from the published sequence. DISCUSSION Although the pathogenetic mechanism explaining the myoclonus– dystonia phenotype in patients with SGCE mutations is unclear, the present study suggests that there is additional genetic heterogeneity and that SGCE mutations do not play a major role in sporadic cases. Several families have been reported to show linkage to the SGCE region,5– 8 and to date, mutations within the gene have been found in 13 families and in 2 apparently sporadic cases.9,20 In the current study, we describe 16 patients, none of whom had mutations in the coding region of the ⑀-sarcoglycan gene. This study did not exclude mutations in the noncoding sequence of the SGCE gene (introns and Movement Disorders, Vol. 18, No. 9, 2003 promoter region) and would not have picked up exonic duplications or deletions. However, it is unlikely that these would account for disease in all of the patients. This report demonstrates genetic heterogeneity for familial myoclonus– dystonia syndrome. It is likely that one or more unmapped genes are responsible for autosomal dominant myoclonus– dystonia in the three families described here. Unfortunately, these families are not currently suitable for linkage analysis due to the limited number of family members available for blood sampling. The present study suggests that ⑀-sarcoglycan does not play a major role in sporadic myoclonus– dystonia. The recent report by Asmus and colleagues20 found SGCE mutations in two apparently sporadic cases, although it is not clear how many sporadic cases were examined. It remains to be understood whether sporadic myoclonus– dystonia is caused by a novel gene with reduced penetrance or represents a multifactorial disorder, determined by the interaction of modifier genes and environmental factors. The clinical phenotype in the families with SGCE mutations reported to date is relatively pure. Almost all of our cases had both prominent dystonia (only 1 patient had purely focal dystonia) as well as myoclonus, and involvement of the lower limbs with either dystonia or myoclonus was more common. In addition, the age of onset of 4 of the cases was over 20 years of age, although one of these (Patient 5) had other family members affected in early childhood. The proband of this family, however, had an unusual phenotype with almost paroxysmal symptoms. This phenotypic variability may be explained by genetic heterogeneity, suggesting that SGCE mutations lead to a narrow clinical spectrum of myoclonus– dystonia with early onset, predominant involvement of the upper limbs and neck, and alcohol responsiveness. In addition, the presence of maternal imprinting may also suggest that the SGCE gene is involved, as this was not evident in our kindreds. In conclusion, our data strongly suggest that ⑀-sarcoglycan mutations do not play an important role in sporadic myoclonus– dystonia and support genetic heterogeneity in familial patients. In addition, it supports the view that SGCE mutations produce a characteristic phenotype. Clinical and genetic analysis of additional familial and sporadic cases is needed to better understand the role of ⑀-sarcoglycan and possibly to identify novel myoclonus– dystonia genes. Acknowledgments: We thank Dr. V. Caputo, Dr. F. Invernizzi, and Dr. G. Zorzi for their useful contribution to this work and Prof. A. Lees and Dr. M. Husain for allowing study of their patients. This work was partly supported by The Dystonia Society (UK) and the ALDEI Foundation. MOYAMOYA-INDUCED PAROXYSMAL DYSKINESIA REFERENCES 1. Quinn NP. Essential myoclonus and myoclonic dystonia. Mov Disord 1996;11:119 –124. 2. Obeso JA, Rothwell JC, Lang AE, Marsden CD. Myoclonic dystonia. Neurology 1983;33:825– 830. 3. Gasser T. Inherited myoclonus-dystonia syndrome. In: Fahn S, Marsden CD, DeLong MR, editors. Advances in neurology. Philadelphia: Lippincott-Raven Publishers; 1998. p 325–334. 4. Saunders-Pullman R, Shriberg J, Heiman G, et al. Myoclonus dystonia: possible association with obsessive-compulsive disorder and alcohol dependence. Neurology 2002;58:242–245. 5. Nygaard TG, Raymond D, Chen C, et al. Localization of a gene for Myoclonus-Dystonia to chromosome 7q21-q31. Ann Neurol 1999; 46:794 –798. 6. Klein C, Schilling K, Saunders-Pullman RJ, et al. A major locus for myoclonus-dystonia maps to chromosome 7q in eight families. Am J Hum Genet 2000;67:1314 –1319. 7. Vidailhet M, Tassin J, Durif F, Nivelon-Chevallier A, Agid Y, Brice A, Dürr A. A major locus for several phenotypes of myoclonus-dystonia on chromosome 7q. Neurology 2001;56:1213– 1216. 8. Asmus F, Zimprich A, Naumann M, et al. Inherited myoclonusdystonia syndrome: narrowing the 7q21-q31 locus in German families. Ann Neurol 2001;49:121–124. 9. Zimprich A, Grabowski M, Asmus F, et al. Mutations in the gene encoding ⑀-sarcoglycan cause myoclonus-dystonia syndrome. Nat Genet 2001;29:66 – 69. 10. Ettinger AJ, Feng G, Sanes JR. ⑀-sarcoglycan, a broadly expressed homologue of the gene mutated in Limb-Girdle Muscular Dystrophy 2D. J Biol Chem 1997;272:32534 –32538. 11. Straub V, Ettinger AJ, Durbeej M, Venzke DP, Cutshall S, Sanes JR, Campbell KP. ⑀-sarcoglycan replaces ␣-sarcoglycan in smooth muscle to form a unique dystrophin-glycoprotein complex. J Biol Chem 1999;274:27989 –27996. 12. Imamura M, Araishi K, Noguchi S, Ozawa E. A sarcoglycandystroglycan complex anchors Dp116 and utrophin in the peripheral nervous system. Hum Mol Genet 2000;9:3091–3100. 13. Klein C, Brin MF, Kramer P, et al. Association of a missense change in the D2 dopamine receptor with myoclonus dystonia. Proc Natl Acad Sci U S A 1999; 96:5173–5176. 14. Klein C, Gurvich N, Sena-Esteves M, et al. Evaluation of the role of the D2 dopamine receptor in myoclonus dystonia. Ann Neurol 2000;47:369 –373. 15. Dürr A, Tassin J, Vidailhet M, Durif F, Jedynak P, Agid Y, Brice A. D2 dopamine receptor gene in myoclonic dystonia and essential myoclonus. Ann Neurol 2000;48:127–128. 16. Grimes DA, Bulman D, St George-Hyslop P, Lang AE. Inherited myoclonus-dystonia: evidence supporting genetic heterogeneity. Mov Disord 2001;16:106 –110. 17. Klein C, Wu L, Doheny D, et al. E-Sarcoglycan mutations found in combination with other dystonia gene mutations. Ann Neurol 2002;52:675– 679. 18. Ozelius LJ, Hewett JW, Page CE, et al. The early-onset torsion dystonia gene encodes an ATP-binding protein. Nat Genet 1997; 17:40 – 48. 19. Leung JC, Klein C, Friedman J, et al. Novel mutation in the TOR1A (DYT1) gene in atypical, early onset dystonia and polymorphisms in dystonia and early onset parkinsonism. Neurogenetics 2001;3:133–143. 20. Asmus A, Zimprich A, Tezenas du Montcel S, et al. Myoclonusdystonia syndrome: ⑀-sarcoglycan mutations and phenotype. Ann Neurol 2002;52:289 –292. 1051 Moyamoya-Induced Paroxysmal Dyskinesia Pedro Gonzalez-Alegre, MD, Zakaria Ammache, MD, Patricia H. Davis, MD, and Robert L. Rodnitzky, MD* Department of Neurology, University of Iowa College of Medicine, Iowa City, Iowa, USA Abstract: Moyamoya disease (MMD) is an uncommon intracranial vasculopathy that typically presents with ischemic or hemorrhagic stroke. Persistent choreoathetosis has been identified as a rare early manifestation of MMD. We present 2 patients with paroxysmal dyskinesia as the initial symptom of MMD, one resembling paroxysmal kinesigenic dyskinesia (PKD) and the other paroxysmal non-kinesigenic dyskinesia (PNKD). We also review the cases of moyamoya-induced chorea reported previously, none of which resembled PKD or PNKD. We hypothesize that both hormonal and ischemic factors may be implicated in the pathogenesis of these abnormal involuntary movements. These cases suggest that MMD should be included in the differential diagnosis of PKD and PNKD. © 2003 Movement Disorder Society Key words: paroxysmal dyskinesia; moyamoya Moyamoya disease (MMD) is an idiopathic vasculopathy affecting mainly children and young adults, and characterized by progressive occlusion of the arteries of the circle of Willis with development of characteristic collateral circulation. The most common presenting events are ischemic and hemorrhagic strokes. Movement disorders are a rare manifestation of this condition, with an estimated frequency of 3 to 6%.1,2 The few detailed reports of abnormal involuntary movements as the presenting symptom of MMD (Table 1) consist of persistent choreoathetosis or brief unilateral paroxysmal dyskinesias (PDys), none of which meet the clinical criteria of PKD or PNKD.2–10 PDys are characterized by recurrent appearance of choreoathetotic or dystonic movements with a normal or near normal state between episodes. They are classified as kinesigenic, exercise-induced, non-kinesigenic or hypnogenic depending on their clinical characteristics. PDys are usually dominantly inherited and suspected to *Correspondence to: Robert L. Rodnitzky, MD, Department of Neurology, Director, Division of Movement Disorders, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, IA 52242. E-mail: robert-rodnitzky@uiowa.edu Received 24 April 2002; Revised 9 August 2002, 10 January 2003; Accepted 12 February 2003 Movement Disorders, Vol. 18, No. 9, 2003 Gender F M F M F F F F M F F F F F F Reference Watanabe et al.3 Watanabe et al.3 Watanabe et al.3 Pavlakis et al.2 Takanashi et al.4 Pelletier et al.5 Pelletier et al.5 Lyoo et al.6 Parmar et al.7 Parmar et al.7 Han et al.8 Shanahan et al.9 Hong et al.10 Present study Present study Patient no. 1 2 Movement Disorders, Vol. 18, No. 9, 2003 3 4 5 6 7 8 9 10 11 12 13 14 15 20 yr 11 yr 22 yr 18 yr 29 yr 9 yr 10 yr 22 yr 17 yr 17 yr 10 yr 12 yr 8 yr 46 mo 44 mo Age of onset 2 mo 11 yr 2 yr 1 mo Unknown 0 1.5 yr 9 mo 0 0 1 yr 1 yr 3 yr 3 yr 7 mo Diagnostic delay Unilateral, constant Bilateral, paroxysmal Unilateral, paroxysmal Unilateral, paroxysmal Unilateral, paroxysmal Bilateral, constant Bilateral, constant Bilateral, constant Unilateral, paroxysmal Unilateral, paroxysmal Bilateral, constant Bilateral, constant Bilateral, constant Bilateral, constant Bilateral, constant Characteristics Other findings Clumsiness right hand and leg Personality changes Dysarthria, paresthesias No Preceded by paresthesia Spasticity, hypereflexia Acalculia, right agraphestesia, right hypesthesia, right Babinski Mild left hemiparesis Right hemiparesis Unsteady gait, dysarthria, clumsiness Left hemiparesis, left hypesthesia, unable to walk Impaired cognition, uttering of words No Unsteady gait, dysarthria, LUE paresis Unsteady gait, dysarthria, decreased IQ Chorea Effective therapy Surgical Resolved after stroke Phenobarbital, spontaneous resolution? Tetrabenazine Surgical Not specified Surgical Discontinuation of oral contraceptives, spontaneous Discontinuation of oral contraceptives, spontaneous Spontaneous resolution Not specified Steroids Surgical Surgical Surgical TABLE 1. Characteristics of chorea secondary to moyamoya disease Right parietal, centrum semiovale Bilateral basal ganglia, centrum semiovale, corona radiata Bilateral frontal subcortical Bilateral basal ganglia, vascular ectasia Left centrum semiovale Periventricular Initially normal head CT scan Right putamen, right frontal, left frontotemporo parietal Left centrum semiovale Bilateral frontal Right anterior horn area Bilateral caudate, periventricular Left caudate, bilateral frontal, left parietooccipital Bilateral deep frontal and parieto-occipital Right MCA subcortical right basal ganglia Imaging (MRI-T2 abnormal signal) 1052 P. GONZALEZ-ALEGRE ET AL. MOYAMOYA-INDUCED PAROXYSMAL DYSKINESIA be due to mutations in ion channels, although there are also symptomatic forms due to a variety of pathological conditions, which commonly show some atypical features.11,12 We present 2 cases of moyamoya manifesting PDys, one resembling PKD and the other PNKD. In addition, in 1 patient a single episode recurred during pregnancy. We suggest that these observations expand the differential diagnosis of PDys. Case Histories We reviewed the charts of all patients diagnosed with MMD seen at University of Iowa hospitals and clinics as part of another study.13 From a total of 44 patients (37 adults and 7 children), 2 presented with a movement disorder (both PDys) as the first identifiable symptom, and an additional patient developed generalized dystonia years after the diagnosis of MMD as a consequence of bilateral basal ganglia strokes. The following are the characteristics of the 2 patients presenting with PDys. Patient 1 was initially evaluated by one of the authors (RLR), and her movements witnessed and described as choreoathetotic by the referring neurologist. The information from Patient 2 was obtained from the medical record. A detailed description of the abnormal involuntary movements was provided by the patient and her husband. Patient 1 A 16-year-old female presented with a 5-year history of paroxysmal choreoathetotic movements of either arm or leg lasting up to 1 minute. These episodes occurred almost on a daily basis, up to 10 per day, in clusters during a 2 to 3 month period, then recurring at 2- to 5-month intervals. They typically appeared during activity such as walking or swimming, and occasionally at rest, but never during sleep and were not related to startle. When these symptoms were at their worst, there was also mild dysarthria. She also reported occasional transient left upper extremity paresthesias and hypesthesia lasting for approximately 2 minutes, at times spreading to the right upper extremity and both legs. These episodes recurred approximately every other year but not in relation to the previously described abnormal movements. She did not have a history of rheumatic fever. She neither smoked nor used alcohol or illicit drugs. A maternal cousin had multiple sclerosis but there was no other family history of neurological disorders. The patient’s general physical examination was unremarkable. Neurological examination was remarkable for slightly slowed alternate motion rate in the left lower extremity and hyperactive patellar reflexes. The initial 1053 laboratory evaluation included normal ceruloplasmin, ANA, glucose, ESR, lipoprotein electrophoresis, urinary aminoacid screen and CSF examination including oligoclonal bands and IgG index. An EEG was normal. An MRI of the brain revealed periventricular white matter T2 hyperintensity, felt to be suspicious for demyelinating disease. Due the intermittent neurological symptoms as well as the MRI findings, multiple sclerosis was a diagnostic consideration, but a definite diagnosis could not be established. A trial of phenytoin was suggested for the possibility of an unusual form of PDys, but she never followed the suggestion, and the abnormal movements subsided spontaneously. At age 22 she experienced a single episode of choreoathetosis of both upper extremities during the third month of her first pregnancy. The episode lasted 10 minutes, and spontaneously remitted. She went on to have an uncomplicated vaginal delivery of a full term infant, but in the first postpartum day she developed acute left central facial paresis, left upper extremity paresis, and mild to moderate dysarthria. A brain MRI showed ischemic changes in the right frontal lobe extending into the right parietal region, and an angiogram revealed bilateral supraclinoid internal carotid artery occlusions with moyamoya phenomenon. She was treated with aspirin and then underwent a left external carotidinternal carotid artery bypass without complication 6 weeks after the ischemic event. Six weeks later she successfully underwent the same procedure on the right side. She continued taking daily aspirin, and later developed symptomatic partial epilepsy, that was controlled with phenytoin. Patient 2 A 22-year-old female developed transient dyskinesias of the right upper and lower extremities, lasting anywhere between 30 minutes to several hours, without alteration in consciousness and not induced by movement. They occurred almost daily, and up to 5 times every day, without any identifiable precipitant. Initial evaluation included normal glucose, thyroid function, blood count, liver and renal function, electrolytes, ESR, head CT scan and EEG. She was treated with phenobarbital, which she took for 2 or 3 months. At that time the movements subsided and she discontinued the treatment without recurrence of the abnormal movements. At age 24, she delivered her second child, after an uncomplicated pregnancy. A few hours later she underwent tubal ligation without complications. Twelve hours after this procedure she developed sudden right sided weakness and was then transferred to our hospital. Her past history was only remarkable for a normal pregnancy and vaginal Movement Disorders, Vol. 18, No. 9, 2003 1054 P. GONZALEZ-ALEGRE ET AL. delivery 6 years before this admission. She had a 10 pack/year history of cigarette smoking, but did not drink alcohol or use recreational drugs. Her only medication on transfer was chlorpromazine for a presumed diagnosis of postpartum psychosis. Her family history revealed a brother with idiopathic cardiomyopathy, but no neurological disorders. On neurological exam she was alert but mute and did not follow any commands. She attended to the examiner with her eyes. She had a right homonymous hemianopsia by confrontation testing, and she had a left gaze preference. She had a right arm monoparesis with normal tone. There was diffuse hyperreflexia without clonus. There were no abnormal involuntary movements. She localized nociceptive stimuli in all four extremities. A head CT scan showed multiple hypodensity in bilateral frontal subcortical areas, and two small hypodensity in the head of the right caudate nucleus. Complete blood count, electrolytes, coagulation profile, ESR, liver function tests, fibrinogen, chest X-ray, and EKG were normal. She underwent four-vessel cerebral angiography, which showed bilateral severe stenosis of the supraclinoid portion of the internal carotid artery with moyamoya phenomenon. During the next few weeks she underwent physical and speech therapy, her hemiparesis and language improved to a residual mild hemiparesis and impaired naming. She developed pseudobulbar affect requiring psychiatric intervention. Over the next 15 years she did not have any recurrence of her movement disorder or cerebrovascular symptoms and the only persistent deficit was mild cognitive impairment and mild right hemiparesis. DISCUSSION Lyoo and associates6 found that movement disorders appear as a manifestation of MMD with an estimated frequency of 3 to 6%. In our population we found 2 of 43 cases (4.6%) presenting with PDys as the initial manifestation of the disease. An additional patient developed post-stroke dystonia, raising the overall incidence of movement disorder in our MMD population to 7%. We found 15 reported cases of MMD manifesting choreoathetosis in Western literature, including the 2 patients reported here (Table 1). In the patients other than ours, the abnormal involuntary movements were transient and resolved after either corticosteroid therapy, tetrabenazine, surgical treatment of the disease, or an ischemic stroke, and were the initial symptom of the disease in most patients. As shown in Table 1, in most patients the abnormal movements preceded the diagnosis of MMD by months to years. We presume that the Movement Disorders, Vol. 18, No. 9, 2003 movement disorder was an early manifestation of the vasculopathy. The identification of MMD at early stage could lead to a prompt initiation of the appropriate therapy. PDys are characterized by recurrent paroxysmal choreoathetosis or dystonic movements. They are self-limited and are usually not associated with symptoms or signs of disease in the interval between episodes. PDys can be classified as kinesigenic, exercised induced, nonkinesigenic or hypnogenic.11,12 Paroxysmal kinesigenic dyskinesias (PKD) are characterized by very frequent, short-lasting attacks precipitated by startle or sudden movements, improving in adulthood and with a good response to anticonvulsants. Paroxysmal non-kinesigenic dyskinesias (PNKD) are precipitated by fatigue, alcohol, caffeine or emotional excitement, and are less frequent but usually last for a few hours.11 PDys are commonly idiopathic (familial or sporadic) but can also be secondary to a variety of disorders of the central nervous system.11,12,14 In the secondary PDys there are frequent atypical findings that are seldom seen in the idiopathic disorders. The precise nature of these uncommon manifestations depends on the underlying pathological process, as seen in most reported patients with MMDinduced chorea (Table 1). Four of the patients reported previously with moyamoya-induced chorea were paroxysmal and include 1 patient with singing-induced chorea,8 2 patients who developed brief episodes of unilateral chorea when started on oral contraceptives,5 and 1 patient with brief episodes of unilateral PDys associated with ulcerative colitis.9 Although paroxysmal, the symptoms in these cases were more atypical for PKD or PNKD than our patients, either by virtue of the associated precipitating events or the duration of individual attacks. In none of these patients was there spontaneous resolution of the PDys, but there was an improvement after surgical therapy, discontinuation of oral contraceptives or treatment with tetrabenazine. The rest of the patients reported previously presented with persistent involuntary movements, commonly worsened by physical or emotional stress. To our knowledge, our patients are the first cases of MMD closely resembling PKD and PNKD, and Patient 1 is the first instance of bilateral PDys of this etiology. The syndrome manifested by our first patient had features of both PKD (multiple daily episodes lasting minutes) and paroxysmal exercise-induced dyskinesia (episodes occurring during prolonged activity), although with atypical symptoms, likely as a result of underlying cerebral ischemia. A few of the episodes were not associated with activity and there were other neurological symp- MOYAMOYA-INDUCED PAROXYSMAL DYSKINESIA toms such as dysarthria and transient sensory symptoms. Some of these somewhat atypical findings have been previously described in patients with idiopathic PKD.15 Our second patient closely resembles PNKD because the involuntary movements were less frequent, of longer duration, and not associated with exercise or activity. In Patient 2 they also resolved spontaneously after a short course of phenobarbital and did not recur, possibly reflecting a period of basal ganglia hypoperfusion that was subsequently improved by collateral circulation. Although the Moyamoya-induced PDys patients described previously were not associated with activity, they were of very short duration, differentiating them from typical PNKD. In our patient the attacks lasted up to a few hours, as is common in idiopathic PNKD. In each of our patients, PDys was the presenting symptom of MMD. A potential explanation for this phenomenon is that they presented with abnormal movements very early in the disease, when the vasculopathy and ischemic episodes were less severe, and less likely to result in a fixed neurological deficit. Our patients had resolution of the abnormal involuntary movements, and both were diagnosed as having MMD years later after a postpartum ischemic stroke. The early recognition of MMD as the etiology for the initial paroxysmal chorea could have altered the subsequent management in these cases, including the strategy for delivery, perhaps suggesting cesarean section in an effort to prevent ischemic complications related to the high vascular and circulatory stress associated with pregnancy. Patient 1 had a single episode of chorea at the end of the first trimester of her first pregnancy that could not be categorized as typical chorea gravidarum because it consisted of a single short-lasting episode. Both of our patients were diagnosed with MMD after a postpartum ischemic stroke. MMD has been reported to worsen or first manifest during pregnancy.17 There is some evidence to suggest that both vascular and hormonal factors contribute to this worsening of MMD, and specifically to the development of chorea. It is known that ischemia of the basal ganglia can lead to choreoathetosis,18 and stroke and transient ischemic attacks have been reported to result in PDys.19,20 In the 15 reported cases of MMD presenting with choreoathetosis, evidence of ischemic disease in frontal and subcortical structures was common (Table 1). The fact that surgical treatment in several of these cases of MMD resulted in resolution of the abnormal involuntary movements also supports the link between dyskinesias and ischemia. In the patient reported with paroxysmal singing-induced chorea, the involuntary movements may have been generated by ischemia to the 1055 basal ganglia while opera-singing, as hyperventilation has been implicated in the development of symptoms in moyamoya patients due to vasoconstriction.8 In Patient 1 there were already ischemic lesions on the initial MRI, mistaken for possible demyelination, and in Patient 2 the dyskinesia occurred unilaterally, related to the side where she ultimately developed an infarction 2 years later. All these observations support the role of basal ganglia hypoperfusion in the genesis of choreodystonic movements in the setting of MMD, a hypothesis further supported by a recent report of a patient with persistent hemichorea secondary to MMD-related hypoperfusion in the basal ganglia, both of which resolved after a revascularization procedure.10 The influence of sexual hormones on the basal ganglia has been studied, although it is not yet understood completely.21–23 Hormonal factors are known to play a role in the pathophysiology of chorea, as exemplified by chorea gravidarum and oral contraceptive or estrogen-induced chorea.24 –26 In our reviewed cases there was a female predominance. Pelletier and associates5 reported 2 patients with MMD who developed unilateral PDys with mild ipsilateral motor deficits after starting to take oral contraceptives. In both cases, the abnormal movements, but not the motor deficits, resolved after discontinuation of the oral contraceptives. Our patient represents the first reported incidence of MMD-related chorea occurring during pregnancy. In addition to the link between estrogen and chorea, hormonal factors may play a role in the pathogenesis of MMD itself. Oral contraceptives are a risk factor for the development of MMD and the incidence of the disease is significantly higher in females.27 In our series of adult MMD we found a high incidence of menstrual irregularities and other endocrinopathies.13 Komiyama and associates17 reported that MMD can present or worsen during pregnancy. We propose that MMD should be included in the differential diagnosis of PDys. This diagnosis should be suspected even more in the presence of other neurological symptoms or signs of cerebral ischemia, and in young females with risk factors for MMD. The importance of this consideration resides in the fact that when present, chorea, whether persistent or paroxysmal, is consistently the initial manifestation of MMD, and its recognition can greatly aid the diagnosis of the disease in its earlier stages, potentially leading to therapy that can prevent future ischemic complications. REFERENCES 1. Maki Y, Enomoto T. Moyamoya disease. Child’s Nerv Syst 1988; 4:204 –212. Movement Disorders, Vol. 18, No. 9, 2003 1056 P. GONZALEZ-ALEGRE ET AL. 2. Pavlakis SG, Schneider S, Black K, Gould RJ. Steroid-responsive chorea in moyamoya disease. Mov Disord 1991;6:347–349. 3. Watanabe K, Negoro T, Maehara M, Takahashi I, Nomura K, Miura K. Moyamoya disease presenting with chorea. Pediatr Neurol 1990;6:40 – 42. 4. Takanashi J, Sugita K, Honda A, Niimi H. Moyamoya syndrome in a patient with Down syndrome presenting with chorea. Pediatr Neurol 1993;9:396 –398. 5. Pelletier J, Cabanot C, Levrier O, Thuillier JN, Cherif AA. Angiodysplasia of moya-moya type disclosed by choreic involuntary abnormal movements during oral contraception. Apropos of 2 cases. Rev Neurol (Paris) 1997;153:393–397. 6. Lyoo CH, Oh SH, Joo JY, Chung TS, Lee MS. Hemidystonia and hemichoreoathetosis as an initial manifestation of moyamoya disease. Arch Neurol 2000;57:1510 –1512. 7. Parmar RC, Bavdekar SB, Muranjan MN, Limaye U. Chorea: an unusual presenting feature in pediatric moyamoya disease. Indian Pediatr 2000;37:1005–1009. 8. Han SH, Kim YG, Cha SH, Chung SY. Moyamoya disease presenting with singing induced chorea. J Neurol Neurosurg Psychiatry 2000;69:833– 834. 9. Shanahan P, Hutchinson M, Bohan A, O’Donoghue D, Sheahan K, Owens A. Hemichorea, moya-moya, and ulcerative colitis. Mov Disord 2001;16:570 –572. 10. Hong YH, Ahn TB, Oh CW, Jeon BS. Hemichorea as an initial manifestation of moyamoya disease: reversible striatal hypoperfusion demonstrated on single photon emission computed tomography. Mov Disord 2002;17:1380 –1383. 11. Demirkiran M, Jankovic J. Paroxysmal dyskinesias: clinical features and classification. Ann Neurol 1995;38:571–579. 12. Bhatia KP. Familial (idiopathic) paroxysmal dyskinesias: an update. Semin Neurol 2001;21:69 –74. 13. Gonzalez-Alegre P, Ammache Z, Davis PH. Clinical characteristics and outcome of adult moyamoya disease in an American population [abstract]. 2002(Suppl. 3);58:A117. Movement Disorders, Vol. 18, No. 9, 2003 14. Sethy KD. Paroxysmal dyskinesias. In: Gilman S, editor. MedLink neurology. San Diego: MedLink Corporation. Available at www.medlink.com. Accessed February 12th, 2002. 15. Houser MK, Soland VL, Bhatia KP, Quinn NP, Marsden CD. Paroxysmal kinesigenic choreoathetosis: a report of 26 patients. J Neurol 1999;246:120 –126. 16. Golbe LI. Pregnancy and movement disorders. Neurol Clin 1994; 12:497–508. 17. Komiyama M, Yasui T, Kitano S, Sakamoto H, Fujitani K, Matsuo S. Moyamoya disease and pregnancy: case report and review of the literature. Neurosurgery 1998;43:360 –368. 18. Bhatia KP, Marsden CD. The behavioural and motor consequences of focal lesions of the basal ganglia in man. Brain 1994;117:859 – 876. 19. Camac A, Greene P, Khandji A. Paroxysmal kinesigenic dystonic choreoathetosis associated with a thalamic infarct. Mov Disord 1990;5:235–238. 20. Hess DC, Sethi KD, Nichols FT. Transient cerebral ischemia masquerading as paroxysmal dyskinesia. Cerebrovasc Dis 1991;1: 54 –57. 21. Kuppers E, Ivanova T, Karolczak M, Beyer C. Estrogen: a multifunctional messenger to nigrostriatal dopaminergic neurons. J Neurocytol 2000;29:375–385. 22. Schipper HM. Sex hormones in stroke, chorea and anticonvulsant therapy. Seminars in Neurology 1988;8:181–186. 23. Dluzen DE. Neuroprotective effects of estrogen upon the nigrostriatal dopaminergic system. J Neurocytol 2000;29:387–399. 24. Barber PV, Arnold AG, Evans G. Recurrent hormone dependent chorea: effects of oestrogens and progestogens. Clin Endocrinol 1976;5:291–293. 25. Leys D, Destee A, Petit H, Warot P. Chorea associated with oral contraception. J Neurol 1987;235:46 – 48. 26. Caviness JN, Muenter MD. An unusual cause of recurrent chorea. Mov Disord 1991;6:355–357. 27. Chiu D, Shedden P, Bratina P, Grotta JC. Clinical features of moyamoya disease in the United States. Stroke 1998;29:1347– 1351.