Spinocerebellar ataxia type 2 presenting as familial levodopa-responsive parkinsonism

BRIEF COMMUNICATIONS

Molecular and Clinical Correlation in Five Indian Families with Spinocerebellar Ataxia 12 Achal K. Srivastava, DM,1 Shweta Choudhry, MSc,2 Musuwadi S. Gopinath, MD,1 Sanghamitra Roy, MSc,2 Manjari Tripathi, DM,1 Samir K. Brahmachari, PhD,2,3 and Satish Jain, DM1

Spinocerebellar ataxia 12 (SCA12) is a recently identified form of autosomal dominant cerebellar ataxia associated with the expansion of an unstable CAG repeat in the 5ⴕ untranslated region of the gene PPP2R2B. We analyzed 77 Indian families with autosomal dominant cerebellar ataxia phenotype and confirmed the diagnosis of SCA12 in 5 families, which included a total of 6 patients and 21 family members. The sizes of the expanded alleles ranged from 55 to 69 CAG repeats, and the sizes of the normal alleles ranged from 7 to 31 repeats. We believe our study is the first to demonstrate that SCA12 may not be as rare in some populations as previously thought. Ann Neurol 2001;50:796 – 800

Autosomal dominant cerebellar ataxias (ADCAs) are a genetically heterogeneous group of hereditary neurodegenerative disorders manifesting clinically in varying degrees of brainstem and cerebellar pathology or dysfunction.1,2 The discovery of distinct disease-causing genetic loci, which were previously defined on the basis of clinical1,2 and pathological criteria,3 has led to the genetic subgrouping of ADCAs into spinocerebellar ataxia 1 (SCA1),4 SCA2,5 SCA3/Machado-Joseph disease (MJD),6 SCA6,7 SCA7,8 SCA8,9 SCA10,10 SCA12,11 SCA17,12 and dentatorubral pallidoluysian atrophy (DRPLA).13 Various ADCAs are caused by the abnormal expansion of trinucleotide repeat motifs in their corresponding genes,4 –9, 11–13 except for SCA10, which is due to a pentanucleotide (ATTCT) repeat expansion.10

From the 1Department of Neurology, Neurosciences Center, All India Institute of Medical Sciences, New Delhi, India; 2Functional Genomics Unit, Center for Biochemical Technology (CSIR), Delhi, India; and 3Molecular Biophysics Unit, Indian Institute of Science, Bangalore, India. Received May 9, 2001, and in revised form Aug 24. Accepted for publication Aug 24, 2001. Published online Nov 1, 2001; DOI 10.1002/ana.10048 Address correspondence to Dr Jain, Department of Neurology, Neurosciences Center, All India Institute of Medical Sciences, New Delhi, India 110 029. E-mail: [email protected]

796

© 2001 Wiley-Liss, Inc.

Recently, an expansion of CAG repeat in the 5⬘ untranslated region of the gene PPP2R2B, encoding a brain-specific regulatory subunit of protein phosphatase 2A, has been found to be responsible for a novel form of ADCA, termed SCA12.11 The expanded alleles contain 55 to 78 repeats, compared with 7 to 32 repeats in normal chromosomes.11,14,15 SCA12 is rare,11,14,15 and after its first description in a large American pedigree of German decent by Holmes and colleagues,11 only one more family, from India, has been identified so far with SCA12 expansion.14 In this article, we report the molecular and clinical features of SCA12 mutation in five unrelated Indian families. In addition, we provide information on the occurrence of SCA12 among Indian ADCA patients and report that SCA12 is not rare in India. Subjects and Methods A total of 293 individuals from 77 ADCA families and 135 normal controls were studied. All probands and affected relatives were clinically examined by a team of neurologists and were also subjected to nerve conduction studies and imaging of the brain (computed tomography [CT] or magnetic resonance imaging [MRI]). ADCA type was diagnosed in accordance with standard criteria.2 Blood samples were collected from all patients, normal relatives (who were also evaluated clinically), and healthy controls. All participants gave informed consent and confidentiality was ensured. Parental consent was obtained for minors. Some of these pedigrees have been reported previously.16

DNA Isolation, Genotyping, and Sequencing Analysis Genomic DNA was isolated from peripheral blood leukocytes using a modification of the salting-out procedure.17 The repeat-containing regions at SCA1,4 SCA2,5 SCA3/ MJD,6 SCA6,7 SCA7,18 SCA8,9 SCA12,11 and DRPLA13 loci were amplified by polymerase chain reaction (PCR) using previously published primers. The exact size of each repeat was determined by Gene Scan analysis using an ABI Prism 377 automated DNA sequencer (Applied Biosystems, Foster City, CA). All of the expanded and a few normal SCA12 alleles were analyzed by DNA sequencing. PCR products were generated from genomic DNA using primer pair SCA12-FP (5⬘ TGGCCCTTAGCTGAGTGG 3⬘) and SCA12-RP (5⬘ TGCTGGGAAAGAGTCGTG 3⬘). Approximately 100ng of genomic DNA was amplified in a 50␮l reaction volume containing a final concentration of 5mM Tris, 25mM KCl, 0.75mM MgCl2, 0.05% gelatin, 20pmol of each primer, 200␮M dNTPs, and 0.5 units of Taq DNA polymerase. Samples were denatured at 94°C for 3 minutes followed by 35 cycles of denaturation (94°C, 30 seconds), annealing (55°C, 30 seconds), extension (72°C, 30 seconds), and a final extension of 7 minutes at 72°C in a Perkin Elmer GeneAmp PCR System 9600 (Applied Biosystems, Foster City, CA). The PCR products were purified from bands cut out of agarose gel using a QIAquick gel extraction kit (Qiagen) and were directly sequenced using the dideoxy chain terminator chemistry on an ABI Prism 377 automated DNA sequencer with the PCR primers.

Fig 1. Pedigrees of five Indian families affected by SCA12. Solid symbols represent affected status, symbols with a question mark represent presymptomatic status (for individuals with CAG expansion who have not yet developed clinical symptoms), and open symbols represent unaffected status. For the 27 individuals who were actually examined and genotyped, the SCA12 CAG repeat allele sizes and the age at disease onset are listed below their pedigree symbols.

Results Frequency of SCA12 Mutation in Indian ADCA Patients Among the 77 families studied, the SCA1 mutation was detected in 12 families (15.6%), SCA2

in 19 families (24.7%), SCA3/MJD and SCA7 in 2 families each (2.6%), and SCA12 in 5 families (6.5%). None had expansion at SCA6, SCA8, or DRPLA loci. The remaining 37 families (48%)

Srivastava et al: Indian Families with SCA12

797

Clinical Features The clinical features of the Indian SCA12 patients included in this study are summarized in the table. Age at onset of the disease ranged from 26 to 50 years (mean, 37.2 years), and the duration of illness at the time of examination varied from 3 to 13 years. All the patients had gait ataxia and dysarthria. The initial Fig 3. Gel electrophoresis of the fluorescence-labeled polymerase chain reaction (PCR) products was conducted on an ABI Prism 377 automated DNA sequencer (Applied Biosystems, Foster City, CA) to determine the SCA12 CAG repeat sizes. Normal alleles show sharp bands and no heterogeneity (bottom), whereas a range of products was observed in expanded alleles (top), suggesting somatic mosaicism in the DNA derived from peripheral blood leukocytes. The length of the expanded allele was determined by the darkest (usually central) band, which is flanked by one to three larger and smaller PCR products of lighter intensities. The sizes of the CAG repeats in all expanded and few representative normal alleles were also confirmed by sequencing. Fig 2. Distribution of CAG repeat at SCA12 locus in 270 normal chromosomes from 135 unrelated healthy controls. The proportion of heterozygotes was 100 of 135 (74%).

were negative for expansion at all the loci studied. Repeat Length Distribution of Normal and Expanded SCA12 Alleles The SCA12 CAG repeat region was amplified in 27 individuals (6 patients and 21 family members) from 5 genotypically confirmed SCA12 families (Fig 1) and compared with 135 unrelated normal individuals. The repeat size ranged from 7 to 31 repeats in normal individuals, with a heterozygosity of 74% (Fig 2). The most common allele contained 10 CAG repeats constituting 48% of the normal chromosomes. Nine of the 27 individuals of SCA12 families showed expansion, which included 6 patients and 3 asymptomatic at-risk individuals (Fig 3). The expanded allele length in patients ranged from 55 to 69 and was 55, 55, and 66 in 3 asymptomatic individuals, AT047:II6, AT047:III2, and AT089:II2 (see Fig 1). Instability of expanded CAG repeats during transmission was studied in three parent-offspring pairs (AT047:II4-III2, AT089:I1-II1, and AT089:I1-II2). We observed a change of ⫺1(56 –55), ⫹2 (67– 69), and ⫺1(67– 66) repeat units in the repeat size of the expanded allele when transmitted from the affected parent to the offspring. All the expanded alleles (n ⫽ 9) and a few normal alleles (n ⫽ 40) were sequenced to confirm the size and the internal structure of the repeat. All the alleles analyzed possessed an uninterrupted CAG repeat configuration.

798

Annals of Neurology

Vol 50

No 6

December 2001

Table. Clinical Features of Spinocerebellar Ataxia 12 Patients Number of patients (families) Women:men Clinical features (n ⫽ 6) Cerebellar gait ataxia Dysarthria Tremor in hands as initial symptom Brisk reflexes Extensor plantar Eye signs Slow saccades Broken pursuit Horizontal nystagmus Facial myokymia Axial dystonia Dementia Electrophysiological features (n ⫽ 6) Subclinical sensory neuropathy (absent sural SNAP) Subclinical sensory-motor neuropathy (reduced CMAP and MNCV and absent sural SNAP) Imaging features (n ⫽ 6) Cortical and cerebellar atrophy (by CT/MRI)

6 (5) 1:5 6 (100%) 6 (100%) 4 (67%) 5 (83%) 2 (33%) 1 (17%) 2 (33%) 1 (17%) 2 (33%) 1 (17%) 0 (0%) n (%) 2 (33%) 3 (50%)

5 (83%)

SNAP ⫽ sensory nerve action potential; CMAP ⫽ compound motor action potential; MNCV ⫽ motor nerve conduction velocity; CT ⫽ computed tomography; MRI ⫽ magnetic resonance imaging.

symptom of the disease in 4 of our patients was tremor in the hands, while the remaining 2 presented with gait ataxia and dysarthria. Five of 6 had brisk reflexes and 2 patients had extensor plantar responses. Other symptoms were slow saccades, broken pursuits, horizontal nystagmus, facial myokymia, and axial dystonia. None of the patients had visual problems, ophthalmoplegia, dysphagia, or dementia (“deranged” Mini Mental Status Examination score). Nerve conductions revealed subclinical sensory neuropathy in 2 (absent sural sensory nerve action potential [SNAP]) and axonal sensory motor neuropathy in three (reduced amplitude of compound motor action potential [CMAP], reduced motor nerve conduction velocity [MNCV], and absent sural SNAP), but were normal in 1 affected individual. Results of nerve conduction studies in 2 at-risk individuals were normal. Brain imaging (CT/MRI) was performed on all 6 affected patients and showed cerebellar as well as cerebral cortical atrophy in 5, while the CT scan was normal for the sixth patient (AT034:III7). Correlation between Repeat Number and Age at Onset No apparent correlation was observed between the repeat size and the age at disease onset in SCA12 patients. Our observation is consistent with previous reports.11,14,19 However, in one parent-offspring pair, the transmission of the disease allele from the father

AT089:I1 (disease onset at 50 years; repeat length n ⫽ 67) to his affected son AT089:II1 (onset at 27 years; repeat length n ⫽ 69) did accompany a decrease in age at onset with an increase in the number of repeat units. Contraction in the length of the expanded allele from 56 to 55 and from 67 to 66 was observed in the transmission of the disease allele from affected fathers (AT047:II4 and AT089:I1, respectively) to their asymptomatic children (AT047:III2 and AT089:II2). As the latter are young (20 and 32 years) and still not clinically affected, it is not possible to comment on the consequence of this contraction on the age at onset at this time. Discussion Our study demonstrates that SCA12 is not rare in India and accounts for approximately 7% of the ADCA cases seen in our centre. The molecular study of SCA12 families revealed a pathological CAG expansion in all 6 affected individuals and in 3 asymptomatic individuals. The repeat sizes of the expanded alleles ranged from 55 to 69 repeat units and were unstable during transmission from parents to offspring. However, the instability was restricted to few repeat unit changes (⫺1 to ⫹2 repeat units). In contrast to other CAG repeat disorders20 in which expansions are more common than contractions, in our SCA12 families contractions (in two of the three vertical transmissions) were observed more frequently. Furthermore, an inverse correlation between increasing repeat size and earlier age at disease onset, a characteristic of triplet repeat disorders,20 was also not observed in our SCA12 patients. The clinical features of our SCA12 patients are comparable to those described earlier11,14,19 and span the full spectrum of ADCA type I except for optic atrophy. Even in the presence of overt ataxia in all cases, 67% of the patients reported hand tremor as the initial symptom. Oculomotor disturbances (in 3 of 6 patients), facial myokymia (in 2 of 6), and axial dystonia (in 1 of 6) were the other features we observed. In contrast to previous reports,11,14,19 none of our patients had clinical evidence of dementia (2 patients aged 57 and 62 years had ataxia for over a decade), while 2 had facial myokymia. Additionally, we found evidence of subclinical neuropathy in 5 of 6 patients. A clinical comparison of the SCA12 phenotype with other SCAs revealed no specific signs that could consistently identify this subtype, making genotyping essential for its accurate diagnosis. In retrospect, we conclude that those ADCA patients presenting with hand tremor in a setting of milder ataxia, with brisk reflexes, evidence of subclinical peripheral neuropathy, and cerebral cortical and cerebellar atrophy, can be suspected of having SCA12.

Srivastava et al: Indian Families with SCA12

799

Financial support from the Department of Biotechnology, Government of India, to S. K. Brahmachari of the Programme on Functional Genomics, is duly acknowledged. We thank Dr M. Mukerji for scientific assistance and Ms R. Jaya, Ms Sakshi, Ms Manju, Ms Gurjit, and Mrs Anuradha for technical support.

References 1. Harding AE. The clinical features and classification of the late onset autosomal dominant cerebellar ataxias: a study of 11 families, including descendants of ‘the Drew family of Walworth.’ Brain 1982;105:1–28. 2. Harding AE. Clinical features and classification of inherited ataxias. Adv Neurol 1993;61:1–14. 3. Konigsmark BW, Weiner LP. The olivopontocerebellar atrophies: a review. Medicine (Baltimore) 1970;49:227–241. 4. Orr HT, Chung MY, Banfi S, et al. Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. Nat Genet 1993;4:221–226. 5. Pulst SM, Nechiporuk A, Nechiporuk T, et al. Moderate expansion of a normally biallelic trinucleotide repeat in spinocerebellar ataxia type 2. Nat Genet 1996;4:269 –276. 6. Kawaguchi Y, Okamoto T, Taniwaki M, et al. CAG expansion in a novel gene for Machado-Joseph disease at chromosome 14q32.1. Nat Genet 1994;8:221–227. 7. Zhuchenko O, Bailey J, Bonnen P, et al. Autosomal dominant cerebellar ataxia (SCA6) associated with small polyglutamine expansions in the alpha 1A-voltage-dependent calcium channel. Nat Genet 1997;15:62– 69. 8. David G, Abbas N, Stevanin G, et al. Cloning of the SCA7 gene reveals a highly unstable CAG repeat expansion. Nat Genet 1997;17:65–70. 9. Koob MD, Moseley ML, Schut LJ, et al. An untranslated CTG expansion causes a novel form of spinocerebellar ataxia (SCA8). Nat Genet 1999;21:379 –384. 10. Matasuura T, Yamagata T, Burgess DL, et al. Large expansion of the ATTCT pentanucleotide repeat in spinocerebellar ataxia type 10. Nat Genet 2000;26:191–194. 11. Holmes SE, O’Hearn EE, McInnis MG, et al. Expansion of a novel CAG trinucleotide repeat in the 5⬘ region of PPP2R2B is associated with SCA12. Nat Genet 1999;23:391–392. 12. Nakamura K, Jeong SY, Uchihara T, et al. SCA17, a novel autosomal dominant cerebellar ataxia caused by an expanded polyglutamine in TATA-binding protein. Hum Mol Genet 2001;10:1441–1448. 13. Koide R, Ikeuchi T, Onodera O, et al. Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian atrophy (DRPLA). Nat Genet 1994;6:9 –13. 14. Fujigasaki H, Verma IC, Camuzat A, et al. SCA12 is a rare locus for autosomal dominant cerebellar ataxia: a study of an Indian family. Ann Neurol 2001;49:117–21. 15. Worth PF, Wood NW. Spinocerebellar ataxia type 12 is rare in the United Kingdom. Neurology 2001;56:419 – 420. 16. Saleem Q, Choudhry S, Mukerji M, et al. Molecular analysis of autosomal dominant hereditary ataxias in the Indian population: high frequency of SCA2 and evidence for a common founder mutation. Hum Genet 2000;106:179 –187. 17. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 18. Gouw LG, Castaneda MA, McKenna CK, et al. Analysis of the dynamic mutation in the SCA7 gene shows marked parental effects on CAG repeat transmission. Hum Mol Genet 1998;7: 525–532. 19. O’Hearn E, Holmes SE, Calvert PC, et al. SCA-12: tremor with cerebellar and cortical atrophy is associated with a CAG repeat expansion. Neurology 2001;56:299 –303.

800

© 2001 Wiley-Liss, Inc.

20. Cummings CJ, Zoghbi HY. Fourteen and counting: unraveling trinucleotide repeat diseases. Hum Mol Genet 2000;9: 909 –916.

Reorganization of Somatic Sensory Function in the Human Thalamus After Stroke Shinji Ohara, MD, PhD, and Frederick A. Lenz, MD, PhD

A patient with a thalamic stroke underwent a microelectrode-guided stereotactic thalamic exploration during surgery for control of tremor. The results of somatic sensory mapping in this patient were compared with explorations carried out during stereotactic surgery for the control of essential tremor (70 patients). There was evidence both of somatotopic reorganization and of anatomic reorganization of the representation of deep structures in the principal somatic sensory nucleus of the thalamus and the nuclei located anterior to it. This case demonstrates that thalamic reorganization can occur after a thalamic stroke and may play a role in recovery from such a stroke. Ann Neurol 2001;50:800 – 803

Reorganization of cortical function has been shown to take place after a cortical stroke, and may be related to functional recovery from strokes.1– 4 Subcortical lesions also cause functional deficits, followed by substantial recovery, perhaps due to changes in cortical function.5,6 However, the subcortical structure itself may reorganize. We present results of a microelectrode exploration during thalamic surgery for tremor after an ischemic thalamic stroke that demonstrated clear thalamic reorganization after the stroke. Case Report The patient was 59-year-old Caucasian man who experienced sudden onset of a left hemiparesis and hemianesthesia due to

From the Department of Neurosurgery, Johns Hopkins Hospital, Baltimore, MD Received May 11, 2001, and in revised form Aug 28, 2001. Accepted for publication Aug 28, 2001. Published online Nov 1, 2001; DOI: 10.1002/ana.10041 Address correspondence to Dr Lenz, Department of Neurosurgery, Johns Hopkins Hospital, Meyer 7-113, 600 North Wolfe Street, Baltimore, MD 21287-7713. E-mail: [email protected]

an ischemic stroke. Magnetic resonance imaging and computed tomography showed an infarction in the right medial temporo-occipital lobes and right posterior thalamus (Fig, A). His initial symptoms, including anesthesia, abated, but within a few weeks he developed tremor in his left hand that did not respond to medical therapy.7–11 On examination, the patient had a left hemianopia, mild left hemiparesis, and disabling tremor in the left upper extremity. His tremor occurred at rest as well as with posture and action at a frequency of approximately 4Hz. Given the failure of medical therapy, the patient was taken to the operating room for implantation of a thalamic stimulating electrode for control of tremor. The procedure was terminated after the exploration and before implantation of the electrode because it became clear that the area in which the electrode would be implanted was involved with the stroke.

Operative Procedures The coordinates of thalamic nuclei were interpolated from the coordinates of the anterior commissure–posterior commissure (AC–PC) line, as determined from a computed tomography scan. The target region was explored with the microelectrode, which was advanced along trajectories made through a burr hole located 2.5cm lateral to the midline at the level of the coronal suture. The first trajectory was toward the principal somatic sensory nucleus (ventral caudal, Vc), because the response of cells in this area to somatic stimulation was the most reliable physiological landmark for the exploration.12 Next, the regions anterior to Vc, presumed thalamic nuclei ventral intermediate and ventral oral posterior, were explored to identify the optimal site for implantation of the electrode. Physiological exploration with the microelectrode involved both recording of neuronal activity and stimulation at microampere current levels. Cells responding to somatic sensory stimulation (sensory cells) were characterized by their response to stimulation of cutaneous or deep structures. Cutaneous sensory cells responded to touch or light pressure applied to the skin. Deep sensory cells responded to joint movement or to squeezing of muscle or tendon, or both, but they did not respond to stimulation of the skin deformed by these stimuli. A reproducible response to repeated application of a stimulus in one part of the body was required to identify a neuronal receptive field (RF).

Data Analysis To demonstrate the unique distribution of sensory cells, we reviewed a consecutive series of essential tremor (ET) patients undergoing stereotactic surgery (70 patients, 237 trajectories) as a control population. None of these patients was found to have abnormalities on detailed somatic sensory examination. Deep or cutaneous RFs were used to identify the body part represented by a cell. Separate body parts were defined as intraoral structures, face, upper extremity, or lower extremity, based on the distribution of sensory cells.13 Differences in proportions were tested statistically using a ␹2 or Fisher’s exact test, as appropriate.

Results In the present case, sensory cells were recorded along each of three trajectories in the 14mm lateral parasagittal plane. There were 19 sensory cells (9 deep and 10 cutaneous sensory cells) among a total of 83 cells recorded in this patient (see Fig, B). In the 70 ET patients, a total of 4,088 cells were identified along 237 trajectories. The location of recordings in thalamus was verified by the presence of thalamic neuronal bursting firing pattern for cells.14 It was not possible to distinguish the discharge patterns in this patient from those in the controls given this small number of cells in our patient. We first examined the location of deep sensory cells relative to the AC–PC line. The distribution of deep sensory cells above the AC–PC line in the present case was 5 of 9 (56%); significantly different ( p ⫽ 0.04) from that in the ET patients 476 of 563 (85%). The proportion of all cells that were located above the AC–PC line was not different ( p ⫽ 0.09) between our case (62 of 83; 74%) and the ET patients (3361 of 4088; 82%). Therefore, the proportion of deep sensory cells above the AC–PC line was significantly lower in the present case than in controls. Along the most posterior trajectory in the case (see Fig, B), 11 sensory cells were identified; the RFs of these cells together included all four body parts (ie, intraoral structures, face, upper extremity, and lower extremity). Along 237 trajectories in 70 control patients with ET, however, 47% (112 of 237), 35% (84 of 237), and 8% (20 of 237) of trajectories had one, two, and three of four body parts represented in the RFs of deep and cutaneous sensory cells recorded along a trajectory. Trajectories with no RFs comprised 9% (21 of 237), but trajectories with RFs including all four body parts were never found in this sample. Therefore, the proportion of trajectories with four body parts represented was significantly higher in the stroke patient (1 of 3) than in control ET patients (0 of 237; p ⫽ 0.01). Discussion The present case of thalamic stroke demonstrates reorganization of thalamic maps based on comparison with maps in 70 control patients with ET. RFs for cells located along one trajectory in the present case covered all of four body parts, which was not observed in a large control sample of ET patients (237 trajectories, 70 patients). The normal consistency of RFs along one trajectory is due to the normal somatotopy of cutaneous and deep cells. Cells with cutaneous RFs are arranged into parasagittal sheets of cells, so that a trajectory in the parasagittal plane will usually encounter cellular RFs located on one or two different parts of the body.13 The representation of deep structures in any part of the body parallels the cutaneous representation of those parts (see Fig 5 in Lenz et al13). Thus, the representation of four body parts along a single tra-

Ohara and Lenz: Reorganization of Human Thalamus After Stroke

801

802

Annals of Neurology

Vol 50

No 6

December 2001

Š

Fig. (A) Axial T2-weighted magnetic resonance image shows an ischemic lesion in the right posterior thalamus. (B) Positions of the trajectories (oblique lines) relative to the anterior commissure–posterior commissure (AC–PC) line (horizontal line). Locations of cells and stimulation sites are indicated by tick marks to the right and left of each trajectory, respectively. Open circles ⫽ cutaneous sensory cells; closed circles ⫽ deep sensory cells. (C) Projected and receptive fields for the sites identified in B. Each site where a cell was recorded or stimulation was carried out is indicated by the same number in B and C. Threshold (in ␮A) is indicated below the projected field figurines. Bar in A is in centimeters.

jectory in the parasagittal plane (see Fig) is a dramatic departure from normal thalamic somatotopy. These results also suggest that the location of cells with deep RFs has been shifted to a position inferior to the usual location of such cells. These results could result from physical distortion of the thalamus resulting from the stroke. However, it is more likely that the somatic sensory nucleus and the nuclei anterior to it have reorganized as a result of the stroke. To our knowledge, this is the first report of single neuron analysis demonstrating functional reorganization in the primate thalamus after a thalamic stroke. It has been reported that thalamic sensory organization is modified by interruption of sensory input.14 –17 The thalamic reorganization resulting from the thalamic damage has been previously reported in only 2 patients without either a description of lesions or a microelectrode analysis in the region of the stroke.8 The relationship between thalamic reorganization and cortical reorganization is inextricable due to reciprocal connections of these structures. Studies using positron emission tomography in patients with a subcortical stroke assume that the function in the subcortical structures is taken over by related cortical areas.5,6 Recovery from motor stroke due to striatal–capsular infarction is associated with bilateral activation of the cortical motor system, with use of ipsilateral pathways and recruitment of additional cortical sensorimotor areas.5 The hemiplegia after ischemic infarction in the basal ganglia and the thalamus is reported to recover in parallel with the activation in the ipsilateral sensorimotor cortex.6 Therefore, cortical reorganization is an important aspect of recovery from subcortical strokes. However, the present case demonstrates that thalamic reorganization may also play a role in recovery from thalamic strokes. This work was supported by the National Institutes of Health (NS38493 and NS40059 to F.A.L.). We thank L. H. Rowland for excellent technical assistance.

References

2. Nudo RJ, Wise BM, SiFuentes F, et al. Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science 1996;272:1791–1794. 3. Nudo RJ, Milliken GW. Reorganization of movement representations in primary motor cortex following focal ischemic infarcts in adult squirrel monkeys. J Neurophysiol 1996;75: 2144 –2149. 4. Xerri C, Merzenich MM, Peterson BE, et al. Plasticity of primary somatosensory cortex paralleling sensorimotor skill recovery from stroke in adult monkeys. J Neurophysiol 1998;79: 2119 –2148. 5. Weiller C, Chollet F, Friston KJ, et al. Functional reorganization of the brain in recovery from striatocapsular infarction in man. Ann Neurol 1992;31:463– 472. 6. Weder B, Knorr U, Herzog H, et al. Tactile exploration of shape after subcortical ischaemic infarction studied with PET. Brain 1994;117:593– 605. 7. Dejerine J, Roussy G. La syndrome thalamique. Rev Neurol 1906;14:521–532. 8. Ohye C, Shibazaki T, Hirai T, et al. Plastic change of thalamic organization in patients with tremor after stroke. Appl Neurophysiol 1985;48:288 –292. 9. Kim JS. Delayed onset hand tremor caused by cerebral infarction. Stroke 1992;23:292–294. 10. Ferbert A, Gerwig M. Tremor due to stroke. Mov Disord 1993;8:179 –182. 11. Miwa H, Hatori K, Kondo T, et al. Thalamic tremor: case reports and implications of the tremor-generating mechanism. Neurology 1996;46:75–79. 12. Lenz FA, Normand SL, Kwan HC, et al. Statistical prediction of the optimal lesion site for thalamotomy in parkinsonian tremor. Mov Disord 1995;10:318 –328. 13. Lenz FA, Dostrovsky JO, Tasker RR, et al. Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J Neurophysiol 1988;59:299 –316. 14. Lenz FA, Kwan HC, Martin R, et al. Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J Neurophysiol 1994;72:1570 – 1587. 15. Garraghty PE, Kaas JH. Functional reorganization in adult monkey thalamus after peripheral nerve injury. Neuroreport 1991;2:747–750. 16. Rasmusson DD. Changes in the response properties of neurons in the ventroposterior lateral thalamic nucleus of the raccoon after peripheral deafferentation. J Neurophysiol 1996;75: 2441–2450. 17. Lenz FA, Zirh AT, Garonzik IM, et al. Neuronal activity in the region of the principle sensory nucleus of human thalamus (ventralis caudalis) in patients with pain following amputations. Neuroscience 1998;86:1065–1081.

1. Castro-Alamancos MA, Borrel J. Functional recovery of forelimb response capacity after forelimb primary motor cortex damage in the rat is due to the reorganization of adjacent areas of cortex. Neuroscience 1995;68:793– 805.

Ohara and Lenz: Reorganization of Human Thalamus After Stroke

803

Evidence for Postjunctional Serotonin (5-HT1) Receptors in the Trigeminocervical Complex Peter J. Goadsby, MD, DSc, Simon Akerman, MSc, and R. James Storer, PhD

Units linked to stimulation of the superior sagittal sinus were identified and recorded from in the trigeminocervical complex of the anesthetized cat. Iontophoresis of glutamate NMDA receptor agonists increased the baselinefiring rate of these neurons. Coejection of sumatriptan, 4991W93, or ergometrine resulted in a significant reduction in NMDA agonist-induced increases in firing. These data establish the existence of triptan-sensitive (5-HT1) receptors on postsynaptic central trigeminal neurones. Ann Neurol 2001;50:804 – 807

The pain in primary headache syndromes, such as migraine and cluster headache, is transmitted via the ophthalmic (first) division of the trigeminal nerve from pain-producing intracranial structures, such as the dura mater and large vessels, projecting to the trigeminocervical complex.1 Modulation of pain transmission by triptans, 5-HT1B/1D receptor agonists, formed an important advance in treatment,2 offering clarification of potential pharmacological mechanisms involved in terminating these attacks.3 A key question in understanding the mechanism of action of these compounds is their site of action. One proposed site of action for triptans is within the trigeminocervical complex.3 It has been assumed, based on observations of inhibition of plasma protein extravasation,4 and because of the localization of 5-HT1D mRNA in the trigeminal ganglion,5 that triptans block trigeminal transmission by a prejunctional mechanism. However, this has never been tested. It has been shown that some component of transmission across the trigeminal nucleus involves glutamatergic mechanisms.6 – 8 In the current study we exploited the fact that microiontophoretic application of NMDA receptor agonists can excite trigeminal neurons locally and

From the Headache Group, Institute of Neurology, The National Hospital for Neurology and Neurosurgery, Queen Square, London, UK Received Mar 27, 2001, and in revised form Sep 4. Accepted for publication Sep 4, 2001. Published online Nov 5, 2001; DOI: 10.1002/ana.0274R Address correspondence to Dr Goadsby, Institute of Neurology, Queen Square, London WC1N 3BG UK. E-mail: [email protected]

804

© 2001 Wiley-Liss, Inc.

thus directly override any prejunctional inhibitory effect of triptans, allowing direct examination of the functional localization of 5HT1 receptors in the trigeminocervical complex. Materials and Methods All studies reported were conducted and terminated under general anesthesia in accordance with a project license issued by the Home Office of the United Kingdom under the Animals (Scientific Procedures) Act 1986. Five cats weighing 3.3 ⫾ 0.4kg (mean SD) were anesthetized with ␣-chloralose (60mg/kg ip; Sigma, St. Louis, MO) and prepared for physiological monitoring. Halothane (May & Baker-Rhone-Poulenc, Dagenham, UK/Fluothane, Zeneca, Macclesfield, UK) (0.5–3% in a 40% oxygen/air mixture) was administered during surgical procedures and then discontinued during experimental protocols.9 Core temperature was monitored and maintained using a rectal thermistor probe and a homeothermic heater blanket system (Harvard Apparatus, Holliston, MA). Cats were intubated and ventilated with a 40% oxygen/air mixture (Harvard Apparatus); end-tidal CO2 and expired oxygen were continuously monitored (Datex-Ohmeda, Helsinki, Finland). The depth of anesthesia was examined periodically by testing for sympathetic (pupillary and cardiovascular) responses to noxious stimulation and withdrawal reflexes in the absence of neuromuscular blockade.

Stimulation and Recording The superior sagittal sinus (SSS) was isolated for stimulation and supramaximally stimulated (20 –28V, 250 ␮sec, 0.1–1.0Hz) after neuromuscular blockade with gallamine triethiodide (Concord, Essex, UK) (initially 10 –15mg/kg IV maintained with 5–10mg/kg/hr).10 Extracellular recordings were made using a microiontophoretic combination electrode consisting of seven-barreled glass pipette incorporating a central tungsten-recording electrode with an exposed recording tip length of approximately 12␮m.11 The electrode was advanced or retracted in the cord substance using a microdrive (Kopf Instruments, Tujunga, CA, or Burleigh Instruments, NY). Signal from the recording electrode was fed via a preamplifier (1,000) (Neurolog, Digitimer, Herts, UK) through a 50Hz noise eliminator (Humbug, Quest Scientific, North Vancouver, BC) and neurolog filter (bandwidth 300Hz to 10kHz) to a second stage variable amplifier then via a window discriminator and A/D converter to a microcomputer where the signal was processed. The receptive fields of the units were characterized as nociceptive specific (NS) if they responded to noxious mechanical stimuli (such as pinch or pricking with a needle) or wide dynamic range (WDR) if they responded to both.12 A microiontophoresis current generator (Dagan 6400, Dagan Corporation, Minneapolis, MN) provided the current for ejecting test substances from the barrels. Retaining and balancing currents were used routinely.13 At the end of the experiment selected sites were marked with an electrolytic lesion made by passing a 20 to 25␮A cathodal current through the recording electrode for 20 to 30 seconds.

Fig 1. Inhibition of DL-H excitation by sumatriptan. DL-H was applied continuously at 4nA. Cell firing was suppressed during pulses of sumatriptan at ⫹50nA (bars). Firing rate (ordinate, Hz) is plotted in 1-second bins on the abscissa (time, seconds).

Drugs The 4991W93, 4S[3-(trans-3-dimethylamino-cyclobutyl)1H-indol-5-yl methyl] oxazolidin-2-one, and sumatriptan were obtained from GlaxoWellcome Research and Development (Stevenage, UK). Micropipette barrels used for iontophoresis of test substances were filled with 1.0M DLhomocysteate (DL-H), pH 8.0 (ICN, Pharmaceuticals); 1.0 M L-glutamate, monosodium, pH 8.0 (Sigma), 50mM ergonovine (ergometrine) maleate, pH 4.0 (Sigma), sumatriptan, and 4991W93; saline (as a control).

Statistical Analysis Summary data are presented as the mean standard error of the mean. Neuronal responses to the test compound in each animal were compared with baseline firing rates using the critical ratio test.14 Population effects of compounds were tested using the Mann-Whitney U test using a level of significance of p ⬍ 0.05 (SPSS version 9.0).

Results Animals from which data are reported had cardiorespiratory parameters that were normal for the anesthetized cat. Blood gas parameters were measured at intervals throughout the experiment and were within normal limits: arterial blood pH 7.38 ⫾ 0.04 and pCO2 3.40 ⫾ 0.89kPa. Neuronal Characteristics Cells were located 0mm to ⫺4mm caudal to the midpoint C2 rootlets, 0 to 300␮m lateral to the dorsal root entry zone at a depth of ⫺700␮m to ⫺2,000␮m below the (dorsal) cord surface. Cells responded to superior sagittal sinus (SSS) stimulation with latencies consistent with 〈␦ fibers (8 –10ms). All neurones reported were classified as wide dynamic range (WDR). Baseline Characteristics and Effect of Excitatory Amino Acids Neurones responding to electrical stimulation of the SSS were identified. They responded with a firing rate

of 15 spikes/50 stimuli. Microiontophoresis of DL-H increased firing of neurones linked to SSS from a baseline of 13.6 ⫾ 1.4/sec to 50.2 ⫾ 5.6/sec (n ⫽ 10; p ⬍ 0.01). Microiontophoresis of L-glutamate increased firing of neurones linked to SSS from a baseline of 0.48 ⫾ 0.06/sec to 27.9 ⫾ 2.4/sec (n ⫽ 5; p ⬍ 0.01). Effect of 5HT1B/1D Agonists Iontophoresis of sumatriptan during continued ejection of DL-H resulted in a reduction in firing rate from 54.3 ⫾ 2.9 to 29.8 ⫾ 4.2/sec (n ⫽ 5; p ⬍ 0.05; Fig 1). Similarly, ejection of 4991W93 (Fig 2) produced a prompt reduction in firing of trigeminal neurones. Ejection of the ergot alkaloid ergometrine resulted in a dose-dependent reduction in firing from a control level of 46.1 ⫾ 4.6/sec to 30.8 ⫾ 4.5/sec at a current of 180nA (n ⫽ 5; p ⬍ 0.05; Fig 3). Discussion This study demonstrates that microiontophoresis of the glutamate receptor agonists DL-homocysteate (DL-H) or L-glutamate into the region of the trigeminal neurones can activate cells also linked to stimulation of the superior sagittal sinus. Furthermore, while being activated by glutamate receptor agonists, ejection of ergometrine, or the triptans, sumatriptan or 4991W93, inhibited neuronal activation. Given that DL-H and L-glutamate were exogenously supplied, and can thus override any prejunctional blockade of transmitter release, it can be suggested that each of ergometrine, sumatriptan, and 4991W93 may have postsynaptic actions within the trigeminal nucleus. If this is correct then postsynaptic trigeminal neurones are likely to contain one or more of 5-HT1B, 5-HT1D or 5-HT1F receptors to mediate this action. Triptans, which are serotonin 5HT1B/1D receptor agonists, inhibit evoked trigeminovascular nociceptive activity in the trigeminocervical complex. Each of

Goadsby et al: Postjunctional 5-HT Receptors

805

Fig 2. Inhibition of DL-H excitation by 4991W93. DL-H was applied continuously at 4nA. Cell firing was suppressed during pulses of 4991W93 microiontophoresis at ⫹50nA (bars). Firing rate (ordinate, Hz) is plotted in 1-second bins on the abscissa (time, seconds).

sumatriptan, after blood-brain barrier disruption, eletriptan, naratriptan, rizatriptan, 4991W93, and zolmitriptan, block evoked trigeminal nucleus activity.3 Furthermore, serotonin itself inhibits trigeminovascularly evoked nociceptive responses in the caudal trigeminal nucleus.15 Triptans and serotonin are effective acute antimigraine treatments.2 It has been assumed that triptans act on a prejunctional 5HT1 receptor whose definitive characteristics are yet to be determined. However, the cited intravenous studies cannot define the localization of the putative triptan receptor. Iontophoretic application of a triptan16 can localize the effect to the second order trigeminal synapse. However, for spontaneously active and sagittal sinus-evoked responses, it cannot differentiate pre- and postsynaptic blockade. The new data effectively eliminate the possibility of a presynaptic effect by supplying the transmitter exogenously. This suggests that some component of the inhibition is likely to be postsynaptic, although it remains a distinct possibility that both pre- and

postsynaptic modulation is taking place. This conclusion is analogous to the prejunctional arguments made for the peripheral action of triptans in which it was shown that sumatriptan would not block the effects of exogenously supplied substance P,17 and thus was most likely to be acting prejunctionally to block neurogenic plasma protein extravasation. Experiments using direct distension of the superior sagittal sinus18 similarly suggested that sumatriptan could block peripheral transmission without necessarily constricting blood vessels.19 It remains possible that the observed effect in our study involves modulation of some other third site, such as a modulatory neuron with facilitatory effects at this synapse. No such neuron has been thus described, and because the effects of microiontophoresis are limited by diffusion, the neuron at least would have to be modulating the second order synapse in the trigeminal nucleus. In summary, data are presented that demonstrate activation of trigeminal neurons in the most caudal tri-

Fig 3. Inhibition of L-glutamate excitation by ergometrine. L-glutamate was applied in 10-second pulses at 50nA followed by 20-second periods of retaining current. Cell firing was suppressed during ergometrine microiontophoresis (bars). The suppression was dose dependent to the maximum inhibition at ⫹180nA. Firing rate (ordinate, Hz) is plotted in 1-second bins on the abscissa (time, seconds).

806

Annals of Neurology

Vol 50

No 6

December 2001

geminal nucleus caudalis with nociceptive vascular input, from structures known to be pain-producing in humans, by activation of glutamate receptors. Coiontophoresis of glutamate agonists, DL-homocysteate or L-glutamate, with either of the 5HT1B/1D agonists, sumatriptan and 4991W93, or the ergot alkaloid, ergometrine, resulted in inhibition of neuronal activation. These data are consistent with the view that there are postsynaptic triptan-sensitive receptors in the trigeminal nucleus, while not excluding the presence of presynaptic modulation. These receptors may be a target for antimigraine compounds, and their identity and presence needs to be considered in any formulation of the mechanism of action of triptans. This work was supported by GlaxoWellcome and The Wellcome Trust. PJG is a Wellcome Trust Senior Research Fellow. We thank Paul Hammond and Michele Lasalandra for their excellent technical assistance. This material was presented in abstract form at the Tenth International Headache Congress, June 29 to July 2, 2001, in New York.20

13. Bloom FE. To spritz or not to spritz: the doubtful value of aimless iontophoresis. Life Sci 1974;14:1819 –1834. 14. Nagler J, Conforti N, Feldman S. Alterations produced by cortisol in the spontaneous activity and responsiveness to sensory stimuli of single cells in the tuberal hypothalamus of the rat. Neuroendocrinology 1973;12:52– 66. 15. Goadsby PJ, Hoskin KL. Serotonin inhibits trigeminal nucleus activity evoked by craniovascular stimulation through a 5-HT1B/1D receptor: a central action in migraine? Ann Neurol 1998;43:711–718. 16. Storer RJ, Goadsby PJ. Microiontophoretic application of serotonin (5HT)1B/1D agonists inhibits trigeminal cell firing in the cat. Brain 1997;120:2171–2177. 17. Buzzi MG, Moskowitz MA. The antimigraine drug, sumatriptan (GR43175), selectively blocks neurogenic plasma extravasation from blood vessels in dura mater. Br J Pharmacol 1990;99:202–206. 18. Kaube H, Hoskin KL, Goadsby PJ. Activation of the trigeminovascular system by mechanical distension of the superior sagittal sinus in the cat. Cephalalgia 1992;12:133–136. 19. Hoskin KL, Kaube H, Goadsby PJ. Sumatriptan can inhibit trigeminal afferents by an exclusively neural mechanism. Brain 1996;119:1419 –1428. 20. Goadsby PJ, Akerman S, Storer RJ. Triptans can act postsynaptically in the trigeminal nucleus: a microiontophoretic study. Cephalalgia 2001;21:285–286.

References 1. Goadsby PJ. The pathophysiology of headache. In: Silberstein SD, Lipton RB, Solomon S, eds. Wolff’s headache and other head pain. 7th ed. Oxford: Oxford University Press, 2001:57–72. 2. Ferrari MD. Migraine. The Lancet 1998;351:1043–1051. 3. Goadsby PJ. The pharmacology of headache. Prog Neurobiol 2000;62:509 –525. 4. Moskowitz MA, Cutrer FM. SUMATRIPTAN: a receptortargeted treatment for migraine. Annu Rev Med 1993;44:145–154. 5. Rebeck GW, Maynard KI, Hyman BT, Moskowitz MA. Selective 5 HT1D␣ serotonin receptor gene expression in trigeminal ganglion: implications for antimigraine drug development. Proc Natl Acad Sci USA 1994;91:3666 –3669. 6. Storer RJ, Goadsby PJ. Trigeminovascular nociceptive transmission involves N-methyl-D-aspartate and non-N-methyl-Daspartate glutamate receptors. Neuroscience 1999;90:1371–1376. 7. Mitsikostas DD, Sanchez del Rio M, Waeber C, et al. The NMDA receptor antagonist MK-801 reduces capsaicin-induced c-fos expression within rat trigeminal nucleus caudalis. Pain 1998;76:239 –248. 8. Mitsikostas DD, Sanchez del Rio M, Waeber C, et al. NonNMDA glutamate receptors modulate capsaicin induced c-fos expression within trigeminal nucleus caudalis. Br J Pharmacol 1999;127:623– 630. 9. Storer RJ, Butler P, Hoskin KL, Goadsby PJ. A simple method, using 2-hydroxypropyl-␤-cyclodextrin, of administering ␣-chloralose at room temperature. J Neurosci Meth 1997;77:49 –53. 10. Storer RJ, Akerman S, Connor HE, Goadsby PJ. 4991W93, a potent blocker of neurogenic plasma protein extravasation, inhibits trigeminal neurons at 5-hydroxytryptamine (5-HT1B/1D) agonist doses. Neuropharmacology 2001;40:911–917. 11. Hellier M, Boers P, Lambert GA. Fabrication of a metal-cored multi-barrelled microiontophoresis assembly. J Neurosci Meth 1990;32:55– 61. 12. Hu JW, Dostrovsky JO, Sessle BJ. Functional properties of neurons in the cat trigeminal subnucleus caudalis (medullary dorsal horn). I. Responses to oro-facial noxious and nonnoxious stimuli and projections to thalamus and subnucleus oralis. J Neurophysiol 1981;45:173–192.

Increased HTLV-I Proviral Load and Preferential Expansion of HTLV-I TaxSpecific CD8⫹ T Cells in Cerebrospinal Fluid from Patients with HAM/TSP Masahiro Nagai, MD, PhD, Yoshihisa Yamano, MD, PhD, Meghan B. Brennan, BA, Carlos A. Mora, MD, and Steven Jacobson, PhD

To date, high human T-cell lymphotropic virus type I proviral load in patients with human T-cell lymphotropic virus type I-associated myelopathy/tropical spastic para-

From the Viral Immunology Section, Neuroimmunology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Rockville Pike, Bethesda, MD 20892. Received Apr 13, 2001, and in revised form Sep 4, 2001. Accepted for publication Sep 4, 2001. Published online Nov 2, 2001; DOI 10.1002/ana.10065 Address correspondence to Dr Jacobson, Viral Immunology Section, Neuroimmunology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Building 10, Room 5B-16, 9000 Rockville Pike, Bethesda, MD. E-mail: [email protected]

© 2001 Wiley-Liss, Inc.

807

paresis has been reported and is thought to be related to the pathogenesis of human T-cell lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis. However, the proviral load in cerebrospinal fluid has not been well investigated. We measured human T-cell lymphotropic virus type I proviral load in cerebrospinal fluid cells from human T-cell lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis patients using real-time quantitative polymerase chain reaction (TaqMan). Human T-cell lymphotropic virus type I proviral load in cerebrospinal fluid cells were significantly higher than that of the matched peripheral blood mononuclear cells, and a high ratio of human T-cell lymphotropic virus type I proviral load in cerebrospinal fluid cells to peripheral blood mononuclear cells were observed in patients with short duration of illness. Human T-cell lymphotropic virus type I Tax-specific CD8ⴙ T cells, as detected by peptide-loaded HLA tetramers, accumulated in cerebrospinal fluid compared with that in peripheral blood mononuclear cells, while the frequency of cytomegalovirus-specific CD8ⴙ T cells in cerebrospinal fluid was reduced. These observations suggest that accumulation of both human T-cell lymphotropic virus type I-infected cells and preferential expansion of human T-cell lymphotropic virus type I-specific CD8ⴙ cells in cerebrospinal fluid may play a role in the pathogenesis of human T-cell lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis. Ann Neurol 2001;50:807– 812

Human T-cell lymphotropic virus type I (HTLV-I) is an exogenous human retrovirus responsible for the progressive neurological disease, HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP).1,2 The vast majority of people infected with HTLV-I are clinically asymptomatic, and less than 5% of those infected develop HAM/TSP. It has been suggested that virus–host interactions play a role in this inflammatory neurological disease. Newly adapted real-time quantitative polymerase chain reaction (PCR) approaches have supported the observations that HTLV-I proviral loads in peripheral blood mononuclear cells (PBMC) from HAM/TSP patients were higher than those of HTLV-I asymptomatic carriers; thus, these high proviral loads have been related to the pathogenesis of this disorder.3– 6 However, the HTLV-I proviral load in cerebrospinal fluid (CSF) from HAM/TSP patients remains to be determined. Experimental observations in the CSF are thought to better reflect events in the central nervous system (CNS) than can be achieved by analysis of peripheral blood.7 HTLV-I virus,8 anti-HTLV-I antibodies,1,2 and HTLV-I-specific CD8⫹ cytotoxic T lymphocytes9,10 have been detected in CSF from HAM/TSP patients and have provided insights into the viral immunological events in the CNS of affected patients. It

808

Annals of Neurology

Vol 50

No 6

December 2001

was therefore of interest to determine the levels of HTLV-I proviral loads in HAM/TSP CSF. In addition, immunological methods that measure HLArestricted virus-specific CD8⫹ T-cell responses to immunodominant epitopes of HTLV-I (tetramers)10 –12 were also used to determine whether these cells could be detected in the CSF of patients with HAM/TSP. We report that the level of HTLV-I proviral (tax) DNA in CSF cells from HAM/TSP patients using realtime quantitative PCR techniques (TaqMan) was elevated in the CSF of HAM/TSP patients compared with PBMC. Moreover, in patients from whom sufficient CSF cells could be obtained, the frequency of HTLV-I-specific CD8⫹ cells were also elevated in the CSF and appear to represent a virus-specific expansion. Patients and Methods Subjects and Sample Preparation Peripheral blood and CSF were obtained from patients with HAM/TSP by informed consent. The diagnosis of HAM/ TSP was assessed according to the World Health Organization guidelines.13 HTLV-I antibodies were present in the serum and in the CSF of all patients. Lumber punctures and blood samples were performed after obtaining informed consent as part of a clinical protocol reviewed and approved by the National Institutes of Health institutional review panel. The characteristics of patients with HAM/TSP are summarized in Table 1. Peripheral blood was taken on the same day that the CSF was collected. PBMC were isolated from peripheral blood samples on a density gradient, and CSF cells were collected after CSF was spun at 1,500 rpm for 10 minutes. DNA was extracted from cells using spin columns DNA extraction kit (Qiagen, Valencia, CA). Sufficient CSF cells were obtained from HAM/TSP patients 7, 8, and 9, permitting the identification of HTLV-I Tax11-19-specific CD8⫹T-cell and CMV-pp65-specific CD8⫹ T cells as described below.

Quantitative Polymerase Chain Reaction HTLV-I proviral load was measured using real-time quantitative PCR (TaqMan) as described previously.12 The amount of HTLV-I proviral DNA was calculated by the following formula: copy number of HTLV-I (pX) per 100 cells ⫽ [(copy number of HTLV-I)/(copy number of ␤-actin/2)] ⫻ 100.

Identification of Tax11-19-Specific CD8⫹T-Cell and CMV-pp65-Specific CD8⫹ T Cells

Analysis of antigen-specific CD8⫹ cells was performed using a phycoerythrin-conjugated Tax11-19 peptide (LLFGYPVYV) loaded HLA-A*0201 tetramer (provided by NIAID MHC Tetramer Core Facility, Atlanta, GA, and National Institutes of Health AIDS Research and Reference Reagent Program) or CMV-pp65(NLVPMVATV)-loaded HLA-A*0201 tetramer (kindly provided from Dr. Mats Eng-

Table 1. Clinical Characteristics and HTLV-I Proviral Load of HAM/TSP Patients

Subject

Race/Origin

Age (yr)

Sex

EDSSa

Duration of Illness (yr)

HAM-1 HAM-2 HAM-3 HAM-4 HAM-5 HAM-6 HAM-7 HAM-8 HAM-9

B/USA W/USA B/USA B/USA B/African H/USA B/Jamaican B/Jamaican W/USA

56 59 64 45 39 57 55 14 45

F M M F F F F F M

6 7 6 6 6.5 6.5 7.5 6.5 2.0

13 14 12 3 1 2 26 4 2

HTLV-I Proviral Loadb CSF

PBMC

Ratio

47.95 44.56 40.82 49.67 33.90 85.81 NT NT NT

30.00 24.41 22.93 24.19 10.49 20.16 26.47 27.67 24.12

1.6 1.8 1.8 2.1 3.2 4.3 — — —

EDSS ⫽ expanded disability status scale. HTLV-I proviral load is expressed as the copy number of HTLV-I DNA per 100 cells. The ratio was calculated by (HTLV-I proviral load in CSF)/(HTLV-I proviral load in PBMC). a

b

HTLV-I ⫽ human T-cell lymphotropic virus type I; HAM/TSP ⫽ HTLV-I-associated myelopathy/tropical spastic paraparesis; CSF ⫽ cerebrospinal fluid; PBMC ⫽ peripheral blood mononuclear cells; W ⫽ Caucasian; B ⫽ black; H ⫽ Hispanic; NT ⫽ not tested.

strand, Uppsala University, Sweden). The cells were washed and incubated with HLA-A2 tetramer, anti-CD4-FITC (Caltag Laboratories, Burlingame, CA), and anti-CD8-TRICOLOR (Caltag Laboratories) monoclonal antibodies for 30 minutes at 4°C. The labeled cells were washed, followed by flow cytometric analysis. HLA-DR, CD25 expression on tetramer-positive cells was also measured as previously described.12

Results and Discussion Table 1 presents the clinical characteristics of HAM/ TSP patients and HTLV-I proviral loads of matching PBMC and CSF cells. In all patients, the HTLV-I proviral load in CSF cells was higher than that in PBMC. The mean ⫾SEM and median copy numbers in CSF were 50.45 ⫾ 7.44 (median 46.26), and those in PBMC were 22.03 ⫾ 2.65 (median 23.56). This difference was statistically significant ( p ⫽ 0.0277, Wilcoxon signed-ranks test). This result is consistent with previously published data reporting that the proportion of HTLV-I Tax protein-expressing cells in the CSF from HAM/TSP patients was higher than that in PBMC.14 However, Puccioni-Sohler and colleagues15 measured higher proviral loads in the PBMC of HAM/ TSP patients in Brazil, as compared with those in the CSF. These conflicting data may result from technical differences since the Brazilian study measured HTLV-I proviral loads using semiquantitative PCR methods without normalization with a housekeeping gene. In this present study, the ratio of HTLV-I proviral load in the CSF compared with PBMC ranged from 1.6 to 4.3. Of interest is the observation that patients who had relatively high ratios (⬎2) were all cases with a short duration of illness and expanded disability status scale scores of ⬎6 (Table 1). This relative augmentation of HTLV-I proviral load in the CSF may therefore relate to disease pathogenesis in the CNS of

HAM/TSP patients, although it remains unclear whether this high ratio is also observed in the early stages of HAM/TSP with low expanded disability status scale and short duration of disease. HTLV-I primarily infects CD4⫹ T cells, although CD8⫹ T cells may also be an important in vivo reservoir.16,17 To exclude the possibility that the difference in HTLV-I proviral load between CSF cells and PBMC was merely attributable to proportional changes of CD4⫹ T cells, the percentage of CD4⫹ and CD8⫹ cells in CSF and PBMC were evaluated by flow cytometry. The CD4⫹/CD8⫹ ratio in CSF did not differ significantly compared with that of PBMC (data not shown) as previously reported.18,19 As demonstrated in HAM-6, the percentages of CD4⫹ and CD8⫹ cells in CSF were 46.7% and 25.8%, respectively. The percentages in PBMC were 33.2% and 19.5%, respectively. Therefore, a change in the proportion of T-cell subsets would not be sufficient to account for the approximate fourfold increase in HTLV-I proviral load. This result suggests that HTLV-I-infected lymphocytes may preferentially migrate into the CSF from peripheral blood, or HTLV-I-infected lymphocytes may selectively expand in this compartment. These two possibilities are not mutually exclusive. It has been demonstrated that HTLV-I can cross the blood–brain barrier by migration of infected lymphocytes in vivo, which has been associated with clonal expansion of HTLV-Iinfected cells in the CSF of patients with HAM/TSP.20 It has been suggested that CSF cells may reflect immunological events in CNS tissue.7 The expansion of HTLV-I Tax11-19-specific CD8⫹ T cells in the CSF of HAM/TSP patients9,10 have supported the hypothesis that these cells may be directly involved in the pathogenesis of this disorder. However, it is unclear whether this increase in HTLV-I Tax-specific CD8⫹

Nagai et al: HTLV-1 Infection in CSF of HAM/TSP

809

810

Annals of Neurology

Vol 50

No 6

December 2001

Š

Fig. Accumulation of HTLV-I Tax11-19-specific CD8⫹ T cells in CSF. CSF and peripheral blood were collected from each HAM/ TSP patient on the same day. Isolated cells were stained with HTLV-I Tax11-19-Tetramer or CMV-pp65-Tetramer, and antiCD8 monoclonal antibody. The percentage represents tetramer-positive CD8⫹ cells in total CD8⫹ cells in PBMC or CSF cells. (A) HAM-7. (B) HAM-8.

cells in the CSF of HAM/TSP patients was specific for HTLV-I. This important question of virus-specific T-cell expansion in the CSF was directly addressed by comparing the frequency of HTLV-I Tax-specific CD8⫹ T cells to the common CMV-specific CD8⫹ T-cell response in both CSF and PBMC from two HAM/TSP patients from which sufficient CSF cells could be obtained (HAM-7 and HAM-8). Both patients were HLA-A*0201 and had detectable CMVspecific CD8⫹ T cells as determined by HLA-A*0201/ CMV-pp65 peptide tetramers.21 There was an expansion in HTLV-I tax-specific CD8⫹ cells in the CSF compared with PBMC from both HAM/TSP patients tested (Fig). In PBMC, 2.83% and 1.78% of CD8⫹ cells were HTLV-I Tax11-19-specific from HAM-7 and HAM-8, respectively, and were consistent with previous reports on tetramer-reactive cells in the peripheral blood of HAM/TSP patients.10 –12 In the CSF of HAM-7 and HAM-8, this population of HTLV-I tax11-19-specific CD8⫹ cells increased to 16.2% and 8.82%, respectively. Importantly, no such expansion of CMV-specific CD8⫹ T cells was observed in the CSF of these HAM/TSP patients (Fig 1A, B). CMV-pp65 reactive CD8⫹ cells were either absent or minimally detected in this compartment. An additional HAM/ TSP patient (HAM-9) showed a similar tetramer profile. The frequency of HTLV-I tax11-19-specific CD8⫹ cells in PBMC was 2.95% and increased to 12.84% in CSF cells while the frequency of CMVspecific CD8⫹ T cells in PBMC was 0.49% and that in CSF cells was 0.44%. It has been demonstrated that activated T cells migrate into the CNS, irrespective of antigen specificity; however, only CNS antigen-specific cells are retained.22 To determine the proportion of activated cells within these each virus-specific CD8⫹ T populations, HLA-DR Table 2. Expression of Activated T-Cell Marker on Antigen-Specific CD8⫹ T Cells HAM-7

Tax-Tetramer⫹ CMV-Tetramer⫹ Total CD8⫹ cells

HAM-8

HLA-DR

CD25

HLA-DR

CD25

64.08 51.09 38.38

6.12 3.11 2.15

75.66 11.97 14.53

10.34 3.7 1.24

Data are percentages of expression. Tax-Tetramer⫹ ⫽ HTLV-I Tax11-19-specific CD8⫹ T cells. CMV-Tetramer⫹ ⫽ CMV-pp65specific CD8⫹ T cells. CMV ⫽ cytomegalovirus; HTLV-I ⫽ human T-cell lymphotropic virus type I; HAM ⫽ HTLV-I–associated myelopathy.

and CD25 expression were evaluated as markers of T-cell activation. The percentage expression of both HLA-DR and CD25 on HTLV-I Tax-specific CD8⫹ T cells in PBMC was higher than that on CMV-specific CD8⫹ T cells in PBMC (Table 2). The increased expression of markers of T-cell activation on HTLV-I Tax-specific CD8⫹ cells in the peripheral blood is consistent with the hypothesis that these cells may have a selective advantage to migrate into the CSF of HAM/ TSP patients. Preferential expansion in the CSF may be associated with the recognition of HTLV-I infected cells in this compartment or the CNS and may therefore contribute to the pathogenesis of this disease. In this report, greater HTLV-I proviral loads in CSF compared with those in PBMC from HAM/TSP patients, particularly in cases with short duration of illness, have been determined by real-time quantitative PCR. The preferential expansion of HTLV-I Tax1119-specific CD8⫹ T cells in the CSF is consistent with the hypothesis that HAM/TSP is an immunopathologically mediated disease associated with high HTV-I proviral loads driving virus-specific immune responses, which may be amenable to immunotherapeutic strategies that target the virus and/or HTLV-I antigenspecific T cells. References 1. Gessain A, Barin F, Vernant JC, et al. Antibodies to human T-lymphotropic virus type-I in patients with tropical spastic paraparesis. Lancet 1985;2:407– 410. 2. Osame M, Usuku K, Izumo S, et al. HTLV-I associated myelopathy, a new clinical entity [letter]. Lancet 1986;1:1031–1032. 3. Yoshida M, Osame M, Kawai H, et al. Increased replication of HTLV-I in HTLV-I-associated myelopathy. Ann Neurol 1989; 26:331–335. 4. Gessain A, Saal F, Gout O, et al. High human T-cell lymphotropic virus type I proviral DNA load with polyclonal integration in peripheral blood mononuclear cells of French West Indian, Guianese, and African patients with tropical spastic paraparesis. Blood 1990;75:428 – 433. 5. Kira J, Koyanagi Y, Yamada T, et al. Increased HTLV-I proviral DNA in HTLV-I-associated myelopathy: a quantitative polymerase chain reaction study [published erratum appears in Ann Neurol 1991;29:363]. Ann Neurol 1991;29:194 –201. 6. Nagai M, Usuku K, Matsumoto W, et al. Analysis of HTLV-I proviral load in 202 HAM/TSP patients and 243 asymptomatic HTLV-I carriers: high proviral load strongly predisposes to HAM/TSP. J Neurovirol 1998;4:586 –593. 7. Mor F, Cohen IR. T cells in the lesion of experimental autoimmune encephalomyelitis. Enrichment for reactivities to myelin basic protein and to heat shock proteins. J Clin Invest 1992; 90:2447–2455. 8. Jacobson S, Raine CS, Mingioli ES, McFarlin DE. Isolation of an HTLV-1-like retrovirus from patients with tropical spastic paraparesis. Nature 1988;331:540 –543.

Nagai et al: HTLV-1 Infection in CSF of HAM/TSP

811

9. Elovaara I, Koenig S, Brewah AY, et al. High human T cell lymphotropic virus type 1 (HTLV-1)-specific precursor cytotoxic T lymphocyte frequencies in patients with HTLV-1associated neurological disease. J Exp Med 1993;177: 1567–1573. 10. Greten TF, Slansky JE, Kubota R, et al. Direct visualization of antigen-specific T cells: HTLV-1 Tax11–19-specific CD8(⫹) T cells are activated in peripheral blood and accumulate in cerebrospinal fluid from HAM/TSP patients. Proc Natl Acad Sci U S A 1998;95:7568 –7573. 11. Bieganowska K, Hollsberg P, Buckle GJ, et al. Direct analysis of viral-specific CD8⫹ T cells with soluble HLA- A2/Tax11–19 tetramer complexes in patients with human T cell lymphotropic virus-associated myelopathy. J Immunol 1999;162:1765–1771. 12. Nagai M, Kubota R, Greten TF, et al. Increased activated human t cell lymphotropic virus type I (HTLV-I) Tax11–19specific memory and effector CD8⫹ cells in patients with HTLV- I-associated myelopathy/tropical spastic paraparesis: correlation with HTLV-I provirus load. J Infect Dis 2001;183: 197–205. 13. Osame M. Review of WHO kagoshima meeting and diagnostic guidelines for HAM/TSP. In: Blattner W, ed. Human retrovirology HTLV. New York: Raven; 1990:191–197. 14. Moritoyo T, Izumo S, Moritoyo H, et al. Detection of human T-lymphotropic virus type I p40tax protein in cerebrospinal fluid cells from patients with human T-lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis. J Neurovirol 1999;5:241–248. 15. Puccioni-Sohler M, Rios M, Bianco C, et al. An inverse correlation of HTLV-I viral load in CSF and intrathecal synthesis of HTLV-I antibodies in TSP/HAM. Neurology 1999;53: 1335–1339. 16. Hanon E, Stinchcombe JC, Saito M, et al. Fratricide among CD8⫹ T lymphocytes naturally infected with human T cell lymphotropic virus type I. Immunity 2000;13:657– 664. 17. Nagai M, Brennan MB, Sakai JA, et al. CD8⫹ T cells are an in vivo reservoir for human T-cell lymphotropic virus type I. Blood 2001;98:1506 –1511. 18. Mori M, Kinoshita K, Ban N, et al. Activated T-lymphocytes with polyclonal gammopathy in patients with human T-lymphotropic virus type I-associated myelopathy. Ann Neurol 1988;24:280 –282. 19. Ijichi S, Eiraku N, Osame M, et al. Activated T lymphocytes in cerebrospinal fluid of patients with HTLV-I-associated myelopathy (HAM/TSP). J Neuroimmunol 1989;25:251–254. 20. Cavrois M, Gessain A, Gout O, et al. Common human T cell leukemia virus type 1 (HTLV-1) integration sites in cerebrospinal fluid and blood lymphocytes of patients with HTLV-1associated myelopathy/tropical spastic paraparesis indicate that HTLV-1 crosses the blood-brain barrier via clonal HTLV-1infected cells. J Infect Dis 2000;182:1044 –1050. 21. Solache A, Morgan CL, Dodi AI, et al. Identification of three HLA-A*0201-restricted cytotoxic T cell epitopes in the cytomegalovirus protein pp65 that are conserved between eight strains of the virus. J Immunol 1999;163:5512–5518. 22. Hickey WF, Hsu BL, Kimura H. T-lymphocyte entry into the central nervous system. J Neurosci Res 1991;28:254 –260. 23. Kurtzke J. 1983. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 33:1444 –1452.

Spinocerebellar Ataxia Type 2 Presenting as Familial Levodopa-Responsive Parkinsonism Din-E Shan, MD, PhD,1,3 Bing-Wen Soong, MD, PhD,1,3 Chen-Ming Sun, MD,2 Shwn-Jen Lee, PhD, PT,4 Kwong-Kum Liao, MD,1,3 and Ren-Shyan Liu, MD2

A genetic analysis identified 2 patients, approximately one-tenth of our patients with familial parkinsonism, who had expanded trinucleotide repeats in SCA2 genes. The reduction of 18F-dopa distribution in both the putamen and caudate nuclei confirmed that the nigrostriatal dopaminergic system was involved in parkinsonian patients with SCA2 mutation. Ann Neurol 2001;50:812– 815

Familial parkinsonism occurs in about 2.8% of Chinese patients.1 Specific genes responsible for familial parkinsonism include ␣-synuclein, parkin, and ␶ genes.2 The mode of inheritance related to the ␣-synuclein gene mutation is autosomal-dominant, but it is now known to be a rare cause of familial parkinsonism. Genetic anticipation occurs in some families with parkinsonism, suggesting involvement of an unstable trinucleotide (CAG) repeat.3 Spinocerebellar ataxia type 3 (SCA3) may present with L-dopa-responsive parkinsonism.4 Noncerebellar symptoms are usually mild in SCA6, but include parkinsonism.5 Recently, a large family from northern Taiwan with a mutation at the SCA2 locus was reported; some present with parkinsonism.6 SCA2 has expanded CAG repeats within the range of 34 to 59 in chromosome 12q. In Taiwan, the SCA2 mutation accounts for about 10.8% of autosomal-dominant cerebellar ataxia.7 The SCA2 phenotype differs from those of SCA1 and SCA3, with its higher frequencies of slowed ocular movements, postural and action tremor, myoclonus, and hyporeflexia.8 Parkinsonism with minor cerebellar deficits occurs.9 Patients may present with either a typical L-dopa-

From the 1Neurological Institute, and 2National PET/Cyclotron Center, Taipei Veterans General Hospital, and 3Departments of Neurology and 4Physical Therapy, National Yang-Ming University, Taipei, Taiwan, ROC. Received July 10, 2001, and in revised form Sep 7, 2001. Accepted for publication Sep 8, 2001. Published online Nov 1, 2001; DOI: 10.1002/ana.10055 Address correspondence to: Dr Shan, Neurological Institute, Taipei Veterans General Hospital, Taipei, Taiwan 11217, Republic of China. E-mail: [email protected]

812

© 2001 Wiley-Liss, Inc.

Table. Clinical Features, Gait Characteristics, and Simple Ratios of Regional 18F-Dopa Uptake of Patients 1 and 2 Pt 1 Age at onset Duration of illness Symptoms at onset First UPDRS subscale IIIa Last UPDRS subscale III Rest tremor of legs Rest tremor of hands Postural tremor of hands Action tremor of hands Rigidity Brisk tendon reflexes Plantar reflexes Ataxia Dysmetria Pursuit Saccades

50 6 yr Tremor of left leg 7

Pt 2 50 9 yr Tremor of both legs 14

45

23

⫹⫹⫹ ⫹⫹ ⫹⫹

⫹⫹⫹ ⫹ ⫺

⫺ ⫹⫹⫹ Left knee Flexor responses ⫺ ⫺ Intact Slow and hypometric Nystagmus ⫺ Autonomic dysfunction ⫺ Speed/off (cm/sec) 67.0 ⫾ 11.0 Speed/on (cm/sec) 107.9 ⫾ 7.6 Stride length/off (cm) 65.3 ⫾ 5.0 Stride length/on (cm) 100.8 ⫾ 6.1 Simple ratio 2.01 (left caudate)b 2.18 Simple ratio (right caudate)b 1.53 Simple ratio (left putamen)b 1.54 Simple ratio (right putamen)b

⫺ ⫹⫹ Left biceps/knee Flexor responses ⫺ ⫺ Intact Slow and hypometric ⫹ ⫺ 110.7 ⫾ 6.8 126.1 ⫾ 4.4 123.9 ⫾ 7.5 127.4 ⫾ 4.0 2.02 1.96 1.85 1.75

a

The first UPDRS assessments were performed 1 year after onset in both patients. b The simple ratios of the regional 18F-dopa uptake of the left and right caudate nucleus and the left and right putamen to that of the occipital cortex in the control subject are 2.75, 2.63, 2.76, and 2.52, respectively. UPDRS ⫽ Unified Parkinson’s Disease Rating Scale. ⫺ ⫽ absent; ⫹ ⫽ mild; ⫹⫹ ⫽ moderate; ⫹⫹⫹ ⫽ severe.

three cardinal features—rest tremor, rigidity, and bradykinesia–to be diagnosed as having parkinsonism. Genomic DNA was isolated from the peripheral leukocytes, and a polymerase chain reaction (PCR) was performed using primers F-1 and R-1 for SCA2, as previously described.10,11 Patients with a proven SCA2 gene expansion were selected for assessments by the Unified Parkinson’s Disease Rating Scale (UPDRS),12 by gait analyses,13 and by 18F-dopa-positron emission tomography (PET) scanning. Gait analyses were assessed both before and after a single challenge dose of 200/50mg of L-dopa/benserazide. The data were compared with those from 12 sporadic parkinsonian patients of comparable age (mean 60.3 ⫾ 4.6 years) and duration of illness (mean 9.5 ⫾ 3.1 years). For PET scanning, antiparkinson medication was stopped for at least 12 hours and 100mg of carbidopa was given 90 minutes before injection. Static 18F-dopa-PET scanning was performed on Scanditronix PC4096-15WB whole-body PET scanner (Scanditronix, Sweden) 120 minutes after injection of 185MBq of 18F-dopa. Images converted to ANALYZE format (Biomedical Imaging Resources, Mayo Foundation, Rochester, MN) by MEDX 3.3 (Sensor Systems, Sterling, VA) were transformed into Talairach and Tournoux space14 by the SPM99 (Wellcome Department of Cognitive Neurology, London, UK) under Matlab (Mathworks, Natic, MA), using a 12-parameter linear affine normalization and a further nonlinear iteration algorithms with a template image generated from a 45-year-old healthy man. This template image was the control subject in this study. The region of interest (ROI) technique was applied to the transaxial slice at the AC/PC line. A standardized template of ROIs was made for defining bilateral caudate nuclei, putamina, and gray matter of occipital lobes in this transaxial slice, according to the corresponding Talairach coordinates. This standardized template of ROIs was then superimposed on the images to compute mean regional radioactivity. Semiquantitative analysis using simple ratios of regional 18 F-dopa uptakes was conducted, which were calculated from the mean regional radioactivity for bilateral caudate nuclei Fig 1. Family pedigrees of Patient 1 (A) and Patient 2 (B).

responsive parkinsonism or an atypical parkinsonism, including pictures of ataxia or progressive supranuclear palsy (PSP).6 To characterize these parkinsonian patients further, in relation to SCA2 gene mutations, we conducted the following study. Patients and Methods Genetic screening for expanded CAG repeats in the SCA2 gene was performed in 23 parkinsonian patients from 19 families with familial parkinsonism, and in 2 patients with postural tremor but with family members affected with parkinsonism. All patients were recruited from the approximately 800 patients with parkinsonism who were registered in the Special Clinic for Movement Disorders of Taipei Veterans General Hospital. Each patient must have two of the

Shan et al: SCA2 and Familial PD

813

Fig 2. Transaxial view of 18F-dopa positron emission tomography (PET) images in Patient 1 (A), Patient 2 (B), a healthy subject (C), and a standardized template of regions of interest (ROIs) superimposed on the healthy subject (D). Profound loss of putaminal 18 F-dopa uptake and, to a lesser extent, of caudate 18F-dopa uptake, occurred in both patients. and putamina, respectively, divided by the gray matter of the occipital cortex.

Results Two patients with parkinsonian features were identified as having an expanded SCA2 gene with 36 and 37 CAG repeat numbers, respectively. Their clinical features are summarized in Table. Patient 1, an ethnic Taiwanese woman, had tremors of the left leg. Her mother had presented with a trihexyphenidyl-responsive leg tremor at 54 years of age, followed by gait disturbance, and expired at the age of 73 (Fig 1A). The tremor in Patient 1 spread to the right leg 3 years after onset; she would occasionally feel hesitant when starting to walk. Her baseline UPDRS subscale III, walking speed, and stride length were comparable with those of controls (31.9, 68.8cm/sec, 76.2cm, respectively). A single dose of L-dopa/benserazide produced a 27% improvement in UPDRS subscale III (vs. 44% in controls), a 65% improvement in walking speed (vs. 26% in controls, p ⬍ 0.01 by Wilcoxon rank-sum test), and a 53% improvement in stride length (vs. 19% in controls, p ⬍ 0.01), confirming that she was responsive to L-dopa. Patient 2, an ethnic Taiwanese man, had tremors of both legs. His mother had presented with sinemetresponsive tremors at the age of 68, followed by gait disturbance, and expired at 79 years old (see Fig 1B). The leg tremor of Patient 2 was not suppressed by propranolol, but it disappeared after taking alcohol, trihexyphenidyl, or primidone. He had occasional shuffling and falling three years after the onset. His baseline walking speed and stride length were better than those of controls ( p ⬍ 0.01, by unpaired t test), confirming his slower progression of motor disability. A single dose of L-dopa/benserazide produced a 48% improvement in UPDRS subscale III, a 14% improvement in walking speed, and a 3% improvement in stride length. Surface electromyograms (EMGs) disclosed in the right vastus medialis of both patients a 4-Hz rest

814

Annals of Neurology

Vol 50

No 6

December 2001

tremor, which disappeared on voluntarily changing of the leg position. Motor conduction time was measured by transcranial magnetic stimulation and showed a slightly prolonged conduction time from the cerebral cortex to the lumbar cord (20.3msec, normal ⬍17msec) in Patient 1, but normal conduction peripherally. The sympathetic skin response showed an absent response in both palm and sole, and the normal pattern of R-R intervals variation was lost in patient 2. Magnetic resonance imaging showed slight atrophy of the cerebral cortex in Patient 2. PET studies showed decreased 18F-dopa distribution in the bilateral striatum of both patients (Fig 2). The simple ratios of the regional 18F-dopa uptake of the caudate nucleus or the putamen, respectively, to that of the occipital cortex in both patients were lower than those of the control (see Table). Discussion The results of this study confirm previous findings that patients with a mutation at the SCA2 locus could present with typical L-dopa-responsive parkinsonism and with minimal cerebellar deficits.6,9 We identified 2 patients, approximately one-tenth of our population with familial parkinsonism, who had expanded CAG repeats with low repeat numbers. This is consistent with the finding that parkinsonism occurs in patients with a relatively low repeat number in the expanded SCA2 genes.6 None of our patients presented with myoclonus, dystonia, or myokymia, which usually appears in patients with a relatively high repeat number.15 Genetic anticipation occurred in both patients. The classic 4 to 6 Hz rest tremor of Parkinson’s disease (PD) occurs mostly in both hands and is temporarily suppressed when the extremity is voluntarily activated.16 The pattern of the tremor in both of our patients was similar to that of typical parkinsonian tremor, except for the predominant leg distribution and the good response to primidone and alcohol. The-

oretically, patients with lesions involving both the nigrostriatal dopaminergic system and the cerebellothalamic system may present with Holmes’ tremor, a lowfrequency rest and intention tremor.16 None of our patients demonstrated the existence of intention tremor. However, it is possible that, with the progression of their cerebellar deficit, their tremor pattern might change. We measured the 18F-dopa metabolism using the striato-occipital ratio method, which is simple but as sensitive as the method that measures the decarboxylation coefficient.17 The results confirmed that the nigrostriatal dopaminergic system was involved in both patients with SCA2 mutation. Unlike previous PET reports in patients with sporadic PD, the caudate nucleus was severely involved in both cases.18 Their pattern of 18 F-dopa distribution appeared similar to that found in patients with PSP. This explained that some of the affected family members might present with a clinical picture indistinguishable from PSP,6 and that our patients had start hesitation and falling early in the course, features characteristic of PSP. Although our case number was limited, some characteristic features appeared: (1) both had onset in their fifties; (2) both started with a prominent 4-Hz rest tremor of the legs; (3) the progression of motor disability was relatively slow; (4) both maintained a good L-dopa-responsiveness; and (5) both showed a reduced 18 F-dopa distribution, not only in the putamen but in the caudate nucleus as well. It would be difficult to identify our patients among patients with sporadic PD, based on the clinical features alone. Gwinn-Hardy and colleagues6 described that their parkinsonian patients could be distinguished by the absence of a rest tremor, and the mildly broad-based gait; none of these features existed in our patients. They could have had an asymmetric onset or an asymmetric 18F-dopa distribution, just as sporadic PD patients had. Neither presented with overt signs, indicating cerebellar dysfunction, except mild gaze-directed nystagmus in Patient 2. The only hints were an unusual alcohol-responsive leg tremor, and a mild slowing of the ocular saccades or gait hesitation suggestive of PSP.

3. Payami H, Bernard S, Larsen K, et al. Genetic anticipation in Parkinson’s disease. Neurology 1995;45:135–138. 4. Tuite PJ, Rogaeva EA, St George-Hyslop PH, Lang AE. Doparesponsive parkinsonism phenotype of Machado-Joseph disease: confirmation of 14q CAG expansion. Ann Neurol 1995;38: 684 – 687. 5. Schols L, Kruger R, Amoiridis G, et al. Spinocerebellar ataxia type 6: genotype and phenotype in German kindreds. J Neurol Neurosurg Psychiatry 1998;64:67–73. 6. Gwinn-Hardy K, Chen JY, Liu H, et al. Spinocerebellar ataxia type 2 with parkinsonism in ethnic Chinese. Neurology 2000; 55:800 – 805. 7. Soong BW, Lu YC, Chao KB, Lee HY: Frequency analysis of autosomal dominant cerebellar ataxias in Taiwanese patients and clinical and molecular characterization of spinocerebellar ataxia type 6. Arch Neurol 2001;58:1105–1109. 8. Schols L, Gispert S, Vorgerd M, et al. Spinocerebellar ataxia type 2. Genotype and phenotype in German kindreds. Arch Neurol 1997;54:1073–1080. 9. Sasaki H, Fukazawa T, Wakisaka A, et al. Central phenotype and related varieties of spinocerebellar ataxia 2 (SCA2): a clinical and genetic study with a pedigree in the Japanese. J Neurol Sci 1996;144:176 –181. 10. Pulst SM, Nechiporuk A, Nechiporuk T, et al. Moderate expansion of a normally biallelic trinucleotide repeat in spinocerebellar ataxia type 2. Nature Genet 1996;14:269 –276. 11. Soong B, Cheng C, Liu R, Shan D: Machado-Joseph disease: clinical, molecular, and metabolic characterization in Chinese kindreds. Ann Neurol 1997;41:446 – 452. 12. Lang AE, Fahn S. Assessment of Parkinson’s disease. In: Munsant TL, ed. Quantification of neurological deficit. Woburn, MA: Butterworths, 1989:285–309. 13. Shan DE, Lee SJ, Chao LY, Yeh SI: Gait analysis in advanced Parkinson’s disease—effect of levodopa and tolcapone. Can J Neurol Sci 2001;28:70 –75. 14. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain 3-dimensional proportional system: an approach to cerebral imaging. New York: Thieme Medical, 1988. 15. Cancel G, Durr A, Didierjean O, et al. Molecular and clinical correlations in spinocerebellar ataxia 2: a study of 32 families. Hum Mol Genet 1997;6:709 –715. 16. Deuschl G, Raethjen J, Lindemann M, Krack P: The pathophysiology of tremor. Muscle Nerve 2001;24:716 –735. 17. Hoshi H, Kuwabara H, Leger G, et al. 6-[18F]fluoro-L-dopa metabolism in living human brain: a comparison of six analytical methods. J Cereb Blood Flow Metab 1993;13:57– 69. 18. Brooks DJ, Ibanez V, Sawle GV, et al. Differing patterns of striatal 18F-dopa uptake in Parkinson’s disease, multiple system atrophy, and progressive supranuclear palsy. Ann Neurol 1990; 28:547–555.

This work was supported by the National Science Council (NSC90-2314-B-075-059) from the ROC and Taipei Veterans General Hospital (VGH-90-372). We thank Ms Win-Yung Sheng for assistance in the statistical analysis.

References 1. Chia LG, Liu LH. Parkinson’s disease in Taiwan: an analysis of 215 patients. Neuroepidemiology 1992;11:113–120. 2. Vaughan JR, Davis MB, Wood NW: Genetics of parkinsonism: a review. Ann Hum Genet 2001;65:111–126.

Shan et al: SCA2 and Familial PD

815

Complete Allele Information in the Diagnosis of Facioscapulohumeral Muscular Dystrophy by Triple DNA Analysis Richard J. L. F. Lemmers, BSc,1 Peggy de Kievit,1 Michel van Geel, PhD,3 Michiel J.R. van der Wielen,1 Egbert Bakker, PhD,1 George W. Padberg, MD, PhD,2 Rune R. Frants, PhD,1 and Silve`re M. van der Maarel, PhD1

Facioscapulohumeral muscular dystrophy is caused by partial deletion of the D4Z4 repeat array on chromosome 4q35. Genetic diagnosis is based on sizing of this repeat array, which is complicated by cross-hybridization of a homologous polymorphic repeat array on chromosome 10 and by the frequent exchanges between these chromosomal regions. The restriction enzyme XapI optimizes the diagnosis of facioscapulohumeral muscular dystrophy by uniquely digesting 4-derived repeat units and leaving 10derived repeat units undigested, thus complementing BlnI, which uniquely digests 10-derived repeat units. A triple analysis with EcoRI, EcoRI/BlnI, and XapI unequivocally allows characterization of each of the four alleles, whether homogeneous or hybrid. This is particularly useful in the case of identical sized 4-derived and 10-derived arrays, in situations of suspected facioscapulohumeral muscular dystrophy with nonstandard allele configurations, and for assignment of hybrid fragments to their original alleles. Ann Neurol 2001;50:816 – 819

Facioscapulohumeral muscular dystrophy (FSHD, MIM 158900) is characterized by progressive weakness and atrophy of facial and shoulder girdle muscles and a gradual spread, variable in time, to abdominal and foot-extensor muscles, followed by involvement of up-

1

From the Center for Human and Clinical Genetics, Leiden University Medical Center, Leiden; 2Department of Neurology, University Medical Center Nijmegen, Nijmegen, The Netherlands; and 3 Department of Cancer Genetics, Roswell Park Cancer Institute, Buffalo, NY. Current address for Dr van Geel: Department of Dermatology, University Medical Center Nijmegen, P.O Box 9101, 6500 HB, The Netherlands. Received May 21, 2001, and in revised form Sep 10, 2001. Accepted for publication Sep 10, 2001. Published online Nov 5, 2001; DOI 10.1002/ana.10057 Address correspondence to Dr van der Maarel, Leiden University Medical Center, Department of Human and Clinical Genetics, Wassenaarseweg 72, 2333 AL Leiden, The Netherlands. E-mail: [email protected]

816

© 2001 Wiley-Liss, Inc.

per arm and pelvic girdle muscles. Recently, attention has been drawn to the absence of facial weakness and pelvic girdle onset.1–3 Clinical uncertainties in some patients and the high frequency of new mutations make reliable DNA diagnosis mandatory for genetic confirmation and counseling. FSHD is associated with partial deletions of the polymorphic D4Z4 repeat array on 4q35. This array is visualized in EcoRI-digested DNA by hybridization with probe p13E-11 (D4F104S1), which recognizes the region proximal to the D4Z4 repeat (Fig 1). In normal people, the array varies between 11 to 150 units; patients carry an allele of 1 to 10 units (10 to 38kb).4,5 DNA diagnosis is compromised by crosshybridization of p13E-11 to a highly homologous polymorphic array on 10q26.6,7 At least 10% of chromosome 10-derived fragments in the population are ⬍38kb and do not cause disease. The restriction enzyme BlnI specifically recognizes chromosome 10derived units, but not chromosome 4-derived units.8 Therefore, BlnI-insensitive chromosome 4 alleles can be discriminated from BlnI-sensitive chromosome 10 alleles by EcoRI/BlnI double digestion (Fig 2B; see Fig 1). Pulsed-field gel electrophoresis (PFGE), allowing discrimination of fragments ⬎50kb and separation of both 4 and 10 alleles, demonstrated the existence of 4-derived arrays on chromosome 10 or 10-derived arrays on chromosome 4.9 Hybrid repeat arrays consisting of mixtures of 4- and 10-derived units have also been identified, adding to the complexity of the DNA diagnosis. Especially in patients with a complex repeat array composition, in patients in whom the probe region itself is also deleted, or in patients who display somatic mosaicism for the repeat array deletion, PFGE has proved very useful.9 –12 These complicated situations underscore the need for simple and reliable tests without use of PFGE, particularly when DNA testing is increasingly wanted not only as a confirmatory but also as an exclusion test. Patients and Methods Subjects Control Subjects 1 and 2 are from a random Dutch population. Subject 3 is a de novo FSHD patient. Subject 4 is a 57-year-old man who was seen by his local neurologist for shoulder pain. When he claimed that his cousin had FSHD, DNA testing was performed showing a shortened EcoRI fragment, indicating that he a gene carrier with a probability of greater than 98%. He was referred for a second opinion. Apart from pain, he reported progressive difficulties raising his arms above his head for a period of 10 years. The physical examination showed mild m. infraspinatus atrophy and weakness (MRC grade 4), as well as bilateral deltoid, biceps, and triceps weakness (grade 4). Painful elevation of the arms limited to 145 degrees was present bilaterally. Exorotation of

ence of a novel consistent difference between both units. A G-to-C transition (position 7515 in accession no. AF117653) results in a single XapI restriction site in the 4-derived unit. The 10-derived unit is completely resistant to XapI. Although XapI cuts the D4F104S1 locus, resulting in weaker hybridization signals, probe p13E-11 is still applicable. A XapI restric-

Fig 1. Restriction map of the EcoRI fragments recognized by p13E-11 (shaded) on chromosomes 4 (solid) and 10 (open) based on one D4Z4 unit in sequence AF117653. Restriction enzymes are EcoRI (E), BlnI (B) and XapI (X) and restriction sizes are in kilobases (kb). XapI cuts frequently around the repeat array, therefore only relevant restriction fragments are indicated.

the upper arms was restricted. He had no facial weakness or other signs of FSHD. Creatine kinase was normal and an extensive electromyelogram showed polyphasic potentials of normal duration and amplitude in the left biceps muscle only. Because the reduction of the EcoRI fragment after BlnI digestion was greater than 3kb, additional genetic studies were requested. A cousin of Subject 4 was 65 years old and had a painful right shoulder for many years. A minor stroke 2 years before the examination had left the patient with a left-sided central facial paresis and a restricted endorotation of the right arm. The brother of this cousin was a 57-year-old left-handed man with shoulder pain of 25 years duration. He had an elevated left shoulder but no paresis or atrophy, and no signs compatible with FSHD. Ten years earlier, an orthopedic surgeon had diagnosed his condition as bilateral rotator-cuff syndrome. Electromyelography at that time was done because of bilateral supra- and infraspinatus atrophy, indicating mild neurogenic and myopathic features, which his neurologist considered suggestive of FSHD.

Fig 2. (A) Pulsed-field gel electrophoresis (PFGE) analysis of the repeat arrays on chromosomes 4 and 10 in the DNA of control Subjects 1 and 2 after digestion with EcoRI/HindIII (E), EcoRI/BlnI (B) or XapI (X) and hybridization with probe p13E-11. In PFGE analyses, HindIII is added to the EcoRI restriction since this enzyme increases the resolution within the 20- to 50-kb range and does not cut within the repeat array itself. Individual 1 carries a standard allele configuration of two 4-derived (BlnI-resistant) alleles of 65kb and 40kb, and two 10-derived (XapI-resistant) alleles of 100kb and 50kb. Individual 2 carries co-migrating 4-derived and 10-derived alleles of 120kb based on the resistance to both enzymes. (B) Schematic presentation of the D4Z4 locus (solid) and its homologue on chromosome 10 (open). The repeat units are represented by an arrow and may vary in number (n). The region recognized by probe p13E-11 is shaded. The restriction sites for EcoRI (E), BlnI (B) and XapI (X) are indicated. Above the chromosome, the restriction fragments are visualized by solid bars, after digestion with the proper enzymes for chromosome 4qter. Below the chromosome, the restriction fragments for 10qter are visualized. The allele sizes of Subject 1 are indicated on the proper restriction fragments.

DNA Isolation For linear gel electrophoresis, DNA was isolated from peripheral blood lymphocytes essentially as described.13 For PFGE, peripheral blood lymphocytes were embedded in agarose plugs at a concentration of 7.5 ⫻ 105 cells per plug.

DNA Diagnosis DNA digestion, separation by linear gel electrophoresis or PFGE, Southern blotting, and hybridization were carried out as described previously.9,14 XapI was purchased from Fermentas. Blots were exposed for 16 to 24 hours to phosphorimager screens and analyzed with the ImageQuant software program (MBI Fermentas, St. Leon-Rot, Germany).

Results Sequence comparison of eight (4-derived) D4Z4 units and seven 10-derived units from GenBank and sequences generated in our laboratory showed the pres-

Lemmers et al: Genetic Diagnosis of FSHD

817

tion site distal to the repeat array obviates a double digest (see Figs 1 and 2B). We tested this difference in 50 healthy controls and 20 FSHD patients who had been examined previously for their repeat array constitution. No inconsistencies were identified; upon EcoRI/BlnI digestion, 4-derived arrays were visualized, while upon XapI digestion, only 10-derived repeat arrays were seen. As demonstrated in Figure 2A, XapI not only facilitates the proper assignment of each allele to its respective chromosome, as exemplified in control 1, but also assists in the identification of equal-size 4-derived and 10-derived alleles, as shown in control 2. This individual carries comigrating 4-derived and 10-derived alleles of 120kb, as inferred from its resistance to BlnI and XapI. Patients 3 and 4 (Fig 3A) were referred for the diagnosis of FSHD. Routine DNA diagnosis includes digestion of DNA with EcoRI and double digestion with EcoRI/BlnI prior to separation by linear gel electrophoresis. As shown in Figure 3A, Patient 3 carries a standard FSHD 4-derived allele of 17kb, based on its BlnI resistance and XapI sensitivity, confirming the clinical diagnosis of FSHD. Patient 4 carries a 30-kb fragment upon digestion with EcoRI, while EcoRI/BlnI double digestion shows a fragment of 17kb, possibly derived from the 30-kb EcoRI fragment. Based on these two enzymes, he was previously considered gene carrier. XapI demonstrates that the 30-kb EcoRI fragment is a homogeneous 10derived array and that, therefore, the 17-kb EcoRI/BlnI fragment cannot be derived from the 30-kb EcoRI fragment. Further PFGE analysis (see Fig 3B) reveals the original allele from which the 17-kb EcoRI/BlnI fragment is derived. It shows a 4-derived (BlnI-resistant) allele of 90kb, two 10-derived (XapI-resistant) alleles of 80kb and 30kb, respectively, and a BlnI- and XapIsensitive allele of 250kb. Since the 90-, 80-, and 30-kb fragments contain homogeneous arrays, the 17-kb EcoRI/BlnI fragment has to come from the hybrid 250-kb allele, ruling out FSHD. Subsequent DNA analysis of his cousins supported this interpretation because it provided no evidence for repeat array fragments of less than 38kb (data not shown). Discussion In recent years, unusual manifestations of FSHD have gained considerable attention in the literature, emphasizing the need for a reliable genetic test for FSHD.2,3,15–19 In a clinically highly selected population, we reached a sensitivity and specificity of the test of more than 95%.14 This led to a liberal use of the test for exclusion purposes, and to a potential rise in the number of false-positive results. The molecular diagnosis of FSHD is complicated by the co-hybridizing homologous repeat array on chromosome 10qter, which may vary between 1 to 150

818

Annals of Neurology

Vol 50

No 6

December 2001

Fig 3. (A) Linear gel analysis of two potential facioscapulohumeral muscular dystrophy (FSHD) patients for genetic diagnosis of FSHD. Fragment sizes in the marker lane (M) are indicated in the middle. The cross-hybridizing 9.4-kb chromosome Y fragment is indicated on the right. DNA of these individuals was digested with EcoRI (E), EcoRI/BlnI (B) and with XapI (X) and separated in adjacent lanes. The new informative XapI fragment is indicated with an arrowhead. Patient 3 carries a standard FSHD allele of 17kb. The fragment is resistant to BlnI digestion and becomes reduced to 14kb due to the BlnI site 3kb distal to the proximal EcoRI site; therefore, it contains chromosome 4-derived repeats and is sensitive to XapI digestion. In another hospital, Patient 4, clinically diagnosed as possibly having FSHD, was originally genetically diagnosed with FSHD with an hybrid allele of 30kb decreasing to 17kb after EcoRI/BlnI double digestion. Further analysis with XapI showed that the 30-kb fragment is a homogeneous 10-derived array, based on its XapI resistance. Consequently, the 17-kb EcoRI/BlnI fragment cannot be derived from the 30-kb array. (B) Pulsed-field gel electrophoresis (PFGE) analysis of Patient 5 carrying a potential hybrid repeat array of 30kb. Combinatory analysis with EcoRI/ HindIII, BlnI and XapI shows that the 250-kb allele is the only allele that is sensitive to both discriminating enzymes. The other three alleles are homogeneous arrays, since they are resistant to BlnI or XapI. It is concluded that the 17-kb EcoRI/BlnI hybrid fragment (arrowhead) is not derived from the 30-kb allele, but rather from the 250-kb allele (arrow). FSHD is therefore excluded in this patient. (C) Schematic representation of the hybrid allele of Patient 4. Chromosome 4-derived units are solid, 10-derived units are open. The restriction sites for EcoRI (E), BlnI (B) and XapI (X) are indicated. Fragment sizes of the hybrid allele in Patient 4 are indicated below the chromosome.

units without pathological consequences, and by the presence of translocated or hybrid repeat arrays in some 21% of the Dutch population.9,11,20 In contrast to an earlier report,9 recent improvements in genetic diagno-

sis showed that only homogeneous or hybrid 4-derived arrays of less than 38kb on chromosome 4 have been identified in FSHD. Short homogeneous 10-derived arrays on chromosome 4 or, small repeat arrays on chromosome 10, irrespective of their composition, have never been associated with disease. Unfortunately, the current techniques do not permit determination of the original EcoRI fragment from which hybrid EcoRI/BlnI fragments are derived. Introduction of the restriction enzyme XapI refines FSHD diagnosis. XapI has opposite characteristics to BlnI: it recognizes a single restriction site within the 4-derived repeat unit, while the 10-derived repeat unit is XapI-resistant. Therefore, digestion of DNA with EcoRI, with EcoRI/BlnI, and with XapI, allows unequivocal determination of chromosome 4-derived and 10-derived alleles. XapI is particularly useful when repeat arrays are composed of clusters of 4-derived and 10-derived repeat units, in which it provides complete allele information. Patient 4 (see Fig 3A–C) exemplifies the power of this procedure, since a potential hybrid short repeat array of four 4-derived repeat units (17kb), followed by three 10-derived units (13kb) in a telomeric direction, would have explained the 30-kb EcoRI fragment, in a clinically atypical FSHD patient. However, the XapI digest indicates that the 30-kb EcoRI fragment is a homogeneous 10-derived repeat and that, therefore, the 17-kb EcoRI/BlnI hybrid fragment cannot have been derived from the 30-kb EcoRI fragment. Indeed, PFGE analysis demonstrates the presence of a 250-kb EcoRI fragment sensitive to BlnI and XapI from which the 17-kb EcoRI/BlnI fragment must be derived. This was supported by the finding that his supposedly affected cousins did not carry alleles ⬍38kb. Since we have not observed a small fragment with BlnI-sensitive repeats, residing on chromosome 4 and responsible for FSHD, there is no genetic support for the diagnosis of FSHD. Although such complicated allele constitutions are probably rare, the increasing use of the DNA test as exclusion criterion necessitates the additional use of XapI to obtain complete allele information. This work is funded by the Prinses Beatrix Fonds, the Dutch FSHD Foundation, the Muscular Dystrophy Association (USA), the FSH Society (USA), and the Association Franc¸aise Contre les Myopathies. S.M.vd.M. is a Gisela Thier Fellow of the Leiden University Medical Center.

References 1. Padberg GW. Facioscapulohumeral disease. Thesis. Leiden University, 1982. 2. Felice KJ, North WA, Moore SA, Mathews KD. FSH dystrophy 4q35 deletion in patients presenting with facial-sparing scapular myopathy. Neurology 2000;54:1927–1931.

3. van der Kooi AJ, Visser MC, Rosenberg N, et al. Extension of the clinical range of facioscapulohumeral dystrophy: report of six cases. J Neurol Neurosurg Psychiatry 2000;69:114 –116. 4. Wijmenga C, Hewitt JE, Sandkuijl LA, et al. Chromosome 4q DNA rearrangements associated with facioscapulohumeral muscular dystrophy. Nature Genet 1992;2:26 –30. 5. van Deutekom JC, Wijmenga C, van Tienhoven EA, et al. FSHD associated DNA rearrangements are due to deletions of integral copies of a 3.2 kb tandemly repeated unit. Hum Mol Genet 1993;2:2037–2042. 6. Bakker E, Wijmenga C, Vossen RH, et al. The FSHD-linked locus D4F104S1 (p13E-11) on 4q35 has a homologue on 10qter. Muscle Nerve 1995;2:S39 – 44. 7. Deidda G, Cacurri S, Grisanti P, et al. Physical mapping evidence for a duplicated region on chromosome 10qter showing high homology with the facioscapulohumeral muscular dystrophy locus on chromosome 4qter. Eur J Hum Genet 1995;3: 155–167. 8. Deidda G, Cacurri S, Piazzo N, Felicetti L. Direct detection of 4q35 rearrangements implicated in facioscapulohumeral muscular dystrophy (FSHD). J Med Genet 1996;33:361–365. 9. van Deutekom JC, Bakker E, Lemmers RJ, et al. Evidence for subtelomeric exchange of 3.3 kb tandemly repeated units between chromosomes 4q35 and 10q26: implications for genetic counselling and etiology of FSHD1. Hum Mol Genet 1996;5: 1997–2003. 10. Cacurri S, Piazzo N, Deidda G, et al. Sequence homology between 4qter and 10qter loci facilitates the instability of subtelomeric KpnI repeat units implicated in facioscapulohumeral muscular dystrophy. Am J Hum Genet 1998;63:181–190. 11. Lemmers RJLF, van der Maarel SM, van Deutekom JCT, et al. Inter- and intrachromosomal subtelomeric rearrangements on 4q35: implications for facioscapulohumeral muscular dystrophy (FSHD) aetiology and diagnosis. Hum Mol Genet 1998;7: 1207–1214. 12. van der Maarel SM, Deidda G, Lemmers RJ, et al. De novo facioscapulohumeral muscular dystrophy: frequent somatic mosaicism, sex-dependent phenotype, and the role of mitotic transchromosomal repeat interaction between chromosomes 4 and 10. Am J Hum Genet 2000;66:26 –35. 13. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 14. Bakker E, van der Wielen MJ, Voorhoeve E, et al. Diagnostic, predictive, and prenatal testing for facioscapulohumeral muscular dystrophy: diagnostic approach for sporadic and familial cases. J Med Genet 1996;33:29 –35. 15. Brouwer OF, Padberg GW, Ruys CJ, et al. Hearing loss in facioscapulohumeral muscular dystrophy. Neurology 1991;41: 1878 –1881. 16. Brouwer OF, Padberg GW, Bakker E, et al. Early onset facioscapulohumeral muscular dystrophy. Muscle Nerve 1995;2: S67–S72. 17. Funakoshi M, Goto K, Arahata K. Epilepsy and mental retardation in a subset of early onset 4q35⫺ facioscapulohumeral muscular dystrophy. Neurology 1998;50:1791–1794. 18. Miura K, Kumagai T, Matsumoto A, et al. Two cases of chromosome 4q35-linked early onset facioscapulohumeral muscular dystrophy with mental retardation and epilepsy. Neuropediatrics 1998;29:239 –241. 19. Felice KJ, Moore SA. Unusual clinical presentations in patients harboring the facioscapulohumeral dystrophy 4q35 deletion. Muscle Nerve 2001;24:352–356. 20. van Overveld PG, Lemmers RJ, Deidda G, et al. Interchromosomal repeat array interactions between chromosomes 4 and 10: a model for subtelomeric plasticity. Hum Mol Genet 2000;9: 2879 –2884.

Lemmers et al: Genetic Diagnosis of FSHD

819

So Near Yet So Far: Neglect in Far or Near Space Depends on Tool Use Alan J. Pegna, PhD, Leila Petit, MA, Anne-Sarah Caldara-Schnetzer, MA, Asaid Khateb, PhD, Jean-Marie Annoni, MD, Roman Sztajzel, MD, and Theodor Landis, MD

The study of unilateral spatial neglect has shown that space can be dissociated on a peripersonal versus extrapersonal basis. We report a novel type of dissociation based on tool use in a patient suffering from left neglect. Line bisection was carried out in near and far space, using a stick and a laser pointer. A rightward bias was always found for the former, but not for the latter. Neglect thus appears to be contingent not only on distance, but also on the motor action required by the task. Ann Neurol 2001;50:820 – 822

right parietal infarct. Our investigation witnessed yet another pattern of results.

Case Report Patient C.M. is a right-handed 87-year-old French-speaking man who had longstanding hypertension. When a sudden weakness in his left upper limb was noted, he was brought to the hospital. On admission, neurological examination indicated temporal disorientation, dysarthria with hypophonia, a left homonymous hemianopia, a left facial droop, and left hypesthesia. He was completely hemiplegic on the left side and showed diminished sensory functions for touch, pain, temperature, and vibration on the left side. A left USN was noted. EchoDoppler of the cervical arteries showed a 40% stenosis of the right internal carotid artery. A computed tomography scan of the brain performed 24 hours after the onset of symptoms showed an ancient infarct located in the right posterior watershed region. Ten days later, a computed tomography scan of the brain showed acute infarcts with contrast enhancement involving the right pre- and postrolandic regions, as well as the posterior arm of the internal capsule.

Neuropsychological Assessment Recent studies of patients suffering from unilateral spatial neglect (USN) have produced evidence suggesting that near (peripersonal) and far (extrapersonal) space are not coded by identical brain networks. Indeed, brain injury can affect either representation selectively. Halligan and Marshall1 first reported a patient who exhibited USN in a line-bisection task carried out within grasping space, but these investigators failed to show any deviations when the lines were situated in extrapersonal space and bisection carried out with a light pen. Later evidence2–5 showed the opposite dissociation with USN present in far space, but not near space. Recently, Berti and Frassinetti6 reported that the limit between far and near space could be modulated through the use of a tool. Indeed, their patient showed peripersonal, but not extrapersonal, USN. Nevertheless, a rightward bias in line bisection appeared when the stimuli were placed beyond grasping distance, when the task was carried out with a stick. These investigators concluded that the use of a stick produced an extension of body space resulting in a recoding of “far” space as “near” by the relevant brain structures, consistent with suggestions from research in monkeys.7,8 Considering the importance of these findings, we sought to investigate them further in a patient who showed left USN restricted to peripersonal space after he sustained a

From the Neurology Clinic, Geneva University Hospitals, Geneva, Switzerland. Received Jun 21, 2001, and in revised form Sep 10, 2001. Accepted for publication Sep 10, 2001. Published online Nov 2, 2001; DOI: 10.1002/ana.10058 Address correspondence to Dr Pegna, Neurology Clinic, Geneva University Hospitals, 24 Rue Micheli-du-Crest, 1211 Geneva 14, Switzerland. E-mail: [email protected]

820

© 2001 Wiley-Liss, Inc.

A comprehensive neuropsychological examination was carried out at 8 and 10 day after admission. On both days he was cooperative and slow and presented attentional fluctuations. He complained spontaneously of hemiplegia and of speech and memory disturbances. Speech was slurred, and his voice was weak and monotone. There was no aphasia. The patient displayed spatial dysgraphia when writing short sentences; he also omitted words on the left side of the text when reading. Similarly, drawings were incomplete in the left visual half-field. Line bisection tested clinically showed a clear rightward bias, as well as the omission of lines situated on the left side of the page. There was no representational neglect. Visual recognition in the right visual half-field was assessed clinically and did not show any signs of visual agnosia. A number of motor perseverations were also noted on graphic productions. A mild memory impairment was noted on the Rey auditory–verbal learning task.

Method The patient was tested on a controlled line-bisection task in which the size of the lines and the distance from the stimuli are systematically varied. Three male control subjects (aged 87, 86, and 80) hospitalized for non-central nervous system diseases served as controls in the same line bisection task. The patient and the control subjects gave informed consent to participate in the test. The experiment was carried out with the subjects comfortably seated in a quiet room. The distances were selected such that three were within grasping distance and three were in extrapersonal space. Line bisection was thus carried out at 30, 55, 80, 110, 140, and 170cm from the patient’s eyes, with the first three distances corresponding to near space and the latter three to far space. The sizes of the lines presented were established to correspond to visual angles of 15, 30, and 45 degrees at each of the six possible distances. Lines were drawn with black ink

Table. Determination of Extent to Which Deviations Should Be Considered Pathological

CM Controls Percentile limits (5th–95th)a

Reaching Near

Pointing Near

Reaching Far

Pointing Far

15.4 (14.3–17.5) 1.4 ⫺1.6–4.2

2.3 (1.6–3.3) 0.5 ⫺2.3–2.5

7.7 (1.4–8.0) 2.0 ⫺1.1–3.5

0.8 (⫺6.2–4.3) 0.2 ⫺2.1–1.4

Median displacement during line bisection in far and near space, using laser light pen (pointing) or pencil/stick (reaching) of patient C.M. (with upper and lower quartiles given in parentheses) and controls. Values are percentage displacement errors (positive values indicate a rightward displacement). a

Obtained from controls.

on white paper and were placed on a cardboard for testing. The lines were positioned at eye level, with the center of each line corresponding to the midline. The complete set of lines was bisected first using a pencil (attached to a 150cm pole for distances of ⬎80cm), and then again using a projection light pen that was placed in the right hand and held close to the sternum. Thirty-six measures were thus obtained for every subject (3 line sizes ⫻ 6 distances ⫻ 2 bisection conditions). The deviations from the center produced by the subjects, measured in degrees of visual angle, were then transformed into percentages with respect to the size of the lines. A value of 0 indicated that the bisection was situated correctly in the center. Deviation of the subject’s response to the right of the objective center (corresponding to left neglect) was indicated by a positive value. Deviations to the left were indicated by negative values.

Results For the patient, the median percentage displacement from the center, along with the upper and lower quartile range, are shown in the Table for each of the 2 ⫻ 2 conditions: near space versus far space and pointing (light pen) versus reaching (pencil and pole). The differences between conditions are statistically significant, as determined by Kruskal-Wallis one-way analysis of variance by ranks test [H(3) ⫽ 19.3; N ⫽ 36; p ⫽ 0.0002]. As is obvious from the data, deviations are greatest for near space in the reaching condition, followed by bisection in extrapersonal space with a stick. By contrast, line bisection in near and far space with a light pen shows only a minimal deviation. To determine the extent to which the deviations should be considered pathological, the values obtained from the controls were used to compute the 5th to 95th percentile limits (values generally used as a boundaries for “normal” function in neuropsychological testing) in each of the four conditions (see Table). The Figure presents the patient’s results plotted along with the normal limits produced by controls. The deviation in near space with a pencil is pathological. Although less pronounced, a similar statement can be made when bisection is carried out in extrapersonal space by means of a pole. When using a light pen, however, the patient does not produce any signs of neglect, whether in near or in far space.

Discussion The key finding in this study was that USN was not observed when the patient bisected the lines with a light projection pen whether the task was carried out in near or in far space. By contrast, the use of the pencil and stick produced a rightward bias, whatever the distance. One possible explanation is that the representation of space may be modified depending on the action to be carried out. In other words, the representation of far or near space could be influenced by the motor response required when measuring neglect. This hypothesis has been repeatedly addressed in the literature. To circumvent motor responses, for example, USN has been evaluated in near and far space, using optical illusions though no dissociation was observed between distances.9 However, dissociations could result from the effect of lesions on motor programs, rather than on a more perceptual level of neglect.3,7 Proof of a differential effect of response mode on the intensity of USN comes from a number of earlier reports. For example, USN has been found to be less marked in a detection task using visuospatial stimuli than verbal.10 Furthermore, neglect patients show a greater number of errors in a pointing task when using the right hand, rather than the left hand,11 suggesting that some interaction takes place between the motor act and the processing of information. Finally, a patient has been reported in whom perceptual tasks (eg, reading) were less markedly impaired than when a motor component was present (eg, line bisection).5 These observations, along with our data, suggest that during the course of information processing, the action required by the task determines which representation of space should be activated. Interestingly, when using a stick, USN in far space is less pronounced than in near space. Similarly, the use of a light pen shows a slightly greater rightward shift in near than far space. Thus, an explanation given strictly in terms of the “action” hypothesis would also appear insufficient. Rather, our findings demonstrate that not only is USN dissociable in terms of far and near space, but also what is considered to be near or far by the brain is at least partially determined by the type of action required with respect to the object. These findings

Pegna et al: Tool-Dependent Hemineglect

821

Fig. Percentage deviation of patient C.M. (—䊐—) in near (A) and far (B) space, using a pencil/stick (reaching condition) or a light pen (pointing condition). Positive values indicate a rightward displacement. Dashed lines represent 5 to 95 percentile limits produced by controls.

point to a need to test USN while varying the motor involvements of the tasks, as has recently been suggested.3 This research was supported by the Swiss National Science Foundation (3100-056782.99).

References 1. Halligan PW, Marshall JC. Left neglect for near but not far space in man. Nature 1991;350:498 –500. 2. Cowey A, Small M, Ellis S. Left visuospatial neglect can be worse in far than in near space. Neuropsychologia 1994;32: 1059 –1066. 3. Cowey A, Small M, Ellis S. No abrupt change in visual hemineglect from near to far space. Neuropsychologia 1999;37:1– 6. 4. Barrett AM, Schwartz RL, Crucian GP, et al. Attentional grasp in far extrapersonal space after thalamic infarction. Neuropsychologia 2000;38:778 –784.

822

Annals of Neurology

Vol 50

No 6

December 2001

5. Vuilleumier P, Valenza N, Mayer E, et al. Near and far visual space in unilateral neglect. Ann Neurol 1998;43:406 – 410. 6. Berti A, Frassinetti F. When far becomes near: remapping of space by tool use. J Cognitive Neurosci 2000;12:415– 420. 7. Iriki A, Tanaka M, Iwamura Y. Coding of modified body schema during tool use by macaque postcentral neurones. Neuroreport 1996;7:2325–2330. 8. Colby CL. Action-oriented spatial reference frames in cortex. Neuron 1998;20:15–24. 9. Pizzamiglio L, Cappa S, Vallar G, et al. Visual neglect for far and near extra-personal space in humans. Cortex 1989;25: 471– 477. 10. Heilman KM, Watson RT. Changes in the symptoms of neglect induced by changing task strategy. Arch Neurol 1978;35: 47– 49. 11. Joanette Y, Brouchon M, Gauthier L, Samson M. Pointing with left vs right hand in left visual field neglect. Neuropsychologia 1986;24:391–396.