X‐linked vacuolar myopathies: Two separate loci and refined genetic mapping

Share Embed


Descripción

BRIEF COMMUNICATIONS

Glutamate Uptake Is Decreased in Platelets from Alzheimer’s Disease Patients Carlo Ferrarese, MD, PhD,* Barbara Begni, PhD,* Carla Canevari, MD,* Chiara Zoia, PhD,† Roberto Piolti, MD,* Maura Frigo, MD,* Ildebrando Appollonio, MD,* and Lodovico Frattola, MD*

Because excitotoxicity may be involved in neurodegeneration in Alzheimer’s disease, we investigated possible modifications of platelet glutamate uptake in AD patients. High-affinity glutamate uptake was studied in platelets from 35 Alzheimer’s disease patients, 10 multi-infarct dementia patients, and 35 age-matched normal controls; it was decreased by 40% in platelets from Alzheimer’s disease patients compared with controls and with multi-infarct dementia patients. Platelet glutamate uptake could be used as peripheral marker of glutamatergic involvement and as adjunctive diagnostic tool in Alzheimer’s disease patients. Ferrarese C, Begni B, Canevari C, Zoia C, Piolti R, Frigo M, Appollonio I, Frattola L. Glutamate uptake is decreased in platelets from Alzheimer’s disease patients. Ann Neurol 2000;47:641– 643

Although the main clinical manifestation is progressive dementia, Alzheimer’s disease (AD) appears today as a systemic disorder. Alterations in amyloid precursor protein (APP) metabolism and of various enzymes have been described not only in brain but also in peripheral tissues from AD patients (for review, see Gasparini and colleagues1). Abnormalities in APP isoform expression, indicating altered APP processing, have been widely shown in platelets,2,3 which are the main source of amyloid in plasma, and may contribute to ␤-amyloid deposition in brain and in blood vessels.4 Moreover, partial defects in mitochondrial enzymes, such as cytochrome c oxidase (COX), which may facilitate oxidative stress and energy failure, have been shown not only in brain5 but also in platelets from AD patients6; experiments with cybrids suggested that these defects arise from mutations of mitochondrial DNA.7 Altered amyloid processing, oxidative stress, and energy failure

may inhibit glutamate uptake, thereby contributing to excitotoxicity and neuronal death in AD. In fact, decreased glutamate uptake, correlated to neurodegeneration, has been shown in brain from AD patients.8 However, these studies have been performed on brain samples at autopsy; for this reason, they might represent end-stage phenomena and have no utility as markers of disease progression or as diagnostic tools. Platelets possess also a high affinity glutamate uptake, similar to that described in brain synaptosomes,9 which could be affected by the reported alterations in APP isoforms and in mitochondrial enzymes. For these reasons, in the present study we investigated possible modifications of glutamate uptake in platelets from AD patients as a putative peripheral marker of glutamatergic dysfunction occurring in the central nervous system. Patients and Methods Patients Thirty-five probable AD patients (16 men and 19 women) were selected for this study from the Department of Neurology of the University of Milano-Bicocca, at the San Gerardo Hospital, Monza, Italy. The diagnosis of probable AD was based both on DSM-IV and the National Institute of Neurological Disorders and Stroke—Alzheimer Disease and Related Disorders Association criteria.10 Patients were followed up for at least 1 year before being included in the study. Patient evaluation included medical history, physical and neurological examinations, laboratory blood tests to exclude metabolic causes of dementia (thyroid hormones, vitamin B12, and erythrocyte sedimentation rate), and a neuroimaging study (computed tomography or magnetic resonance of the brain, or both). In addition, all patients received neuropsychological tests including the Mini-Mental State Examination (MMSE)11 and the Global Deterioration Scale (GDS).12 Patients with neoplastic or hematological disorders, recent infections or surgery, severe hepatic or renal insufficiency, myocardial infarction in the previous 6 months, or cranial trauma in the previous 6 months or who had undergone antiplatelet, anti-inflammatory, antineoplastic, corticosteroid, or immunosuppressive drug treatments were excluded. We also included 35 normal controls (16 men and 19 women) who were sex and age matched to patients, and 10 patients (5 men and 5 women) with multi-infarct dementia (MID) according to Hachinski’s criteria.13 Clinical characteristics of patients and controls are shown in Table 1.

Platelet Preparation From the *Department of Neurology, University of Milano-Bicocca, Ospedale San Gerardo, Monza, and †Department of Anatomy, University of Milan, Milan, Italy. Received Jun 29, 1999, and in revised form Dec 16. Accepted for publication Dec 18, 1999. Address correspondence to Prof Ferrarese, Department of Neurology, Universita` di Milano-Bicocca, Ospedale S. Gerardo, Via Donizetti 106, 20052 Monza, Italy.

After overnight fasting and obtaining of informed consent, 15 ml of blood was collected from the antecubital vein, placed in plastic tubes containing K2-EDTA, and immediately processed for platelet preparation. Platelet-rich plasma (PRP) was separated by centrifugation at 390g for 15 minutes at 15°C. The top three-fourths of the PRP was removed and centrifuged at 3,600g for 15 minutes to obtain a platelet pellet. Platelet pellets were stored in 1 ml 0.32 M sucrose containing 5% dimethylsulfoxide (Sigma-

Copyright © 2000 by the American Neurological Association

641

Table 1. Clinical Characteristics of AD Patients, MID Patients, and Normal Controls

AD patients MID patients Normal controls

n

Male/Female Ratio

Age, yr (range/mean ⫾ SEM)

Duration of Disease, yr (range/mean ⫾ SEM)

MMSE Score (range/mean ⫾ SEM)

GDS Score (range/mean ⫾ SEM)

35 10 35

16/19 5/5 16/19

55–89/71.5 ⫾ 1.4 52–93/67.2 ⫾ 4.7 55–90/71.3 ⫾ 2.7

1–10/4 ⫾ 0.3 2–5/4.1 ⫾ 0.3 —

27–0/8.7 ⫾ 1.4 10–0/4.3 ⫾ 1.3 —

2–7/5.4 ⫾ 0.2 5–7/6.4 ⫾ 0.2 —

AD ⫽ Alzheimer’s disease; MID ⫽ multi-infarct dementia; MMSE ⫽ Mini-Mental State Examination; GDS ⫽ Global Deterioration Scale.

Aldrich, St Louis, MO) to maintain membrane integrity and were maintained at ⫺80°C until use for biochemical studies; cytological studies with tripan blue and May-Grunwald Giemsa demonstrated 80% vital platelets with intact organelles after thawing.

Glutamate Uptake Studies To investigate the sodium-dependent uptake of glutamate in platelets, we followed the original method of Mangano and Schwarcz,9 with minor modifications. We performed preliminary experiments in platelets from control subjects to assess the time course and the pH, sodium, and temperature dependence and for the kinetic analysis of glutamate uptake, with final glutamate concentrations ranging from 0.2 to 120 ␮M. These results were in line with the original method. Specific inhibitors of glutamate uptake (L-aspartic acid; L[⫺]threo-3-hydroxyaspartic acid) completely blocked the sodiumdependent high-affinity uptake. We also compared glutamate uptake in fresh platelet preparations and in platelets after freezing and thawing with the described procedure, and similar results were obtained. For routine experiments on patients, platelets were incubated with 25 ␮l [3H]glutamate (specific activity ⫽ 57 Ci/mMol; DuPont-NEN, Boston, MA) to obtain a final saturating concentration of 60 ␮M. Uptake was stopped after 30 minutes by the addition of 3 ml ice cold Tris buffer containing 1 mM cold glutamate. Net high-affinity uptake was determined by subtracting no-sodium blanks from the uptake in the presence of sodium and expressed as pmol glutamate/mg protein/30 min, as previously described.14

Statistical Analysis All results are expressed as mean ⫾ SEM. One-way analysis of variance (ANOVA), followed by the Dunnett test for independent samples, was used to assess the significance of differences between means of control and patient groups. Statistical analysis of correlations was performed with Pearson’s correlation test.

Results The Figure shows the distribution of glutamate uptake values in platelets from normal controls, AD patients, and MID patients. A greater variability of the values was observed in AD and MID patients than in the controls. However, most of the AD patients had uptake values below the mean of controls. A partial overlap was present between glutamate uptake values of AD patients and controls in the range between 50 to 130 pmol/mg of proteins (see Fig), with the exception of 3 patients with very high uptake values. One of these had recent (1 year) onset of memory deficit and gait disturbances, with initial diagnosis of normal pres-

642

Annals of Neurology

Vol 47

No 5

May 2000

Fig. High-affinity glutamate uptake in platelets from Alzheimer’s disease (AD) patients, multi-infarct dementia (MID) patients, and normal controls (NC) . *p ⬍ 0.0005 vs NC and MID patients. 30⬘ ⫽ 30 minutes.

sure hydrocephalus, for peculiar ventricular enlargement. Subsequent imaging studies excluded normal pressure hydrocephalus, however, and the patient is now considered probable AD. In another patient the presence of ischemic area in parietal cortex rendered the diagnosis of mixed dementia more likely. The third patient showed clinical condition typical of AD. Statistical analysis confirmed an extremely significant ( p ⬍ 0.0005 at ANOVA; p ⬍ 0.01 at Dunnett) 36% decrease of mean uptake values in AD patients compared with controls, while no significant modification was observed between MID patients and controls(Table 2). Platelet glutamate uptake was also significantly decreased by about 40% ( p ⬍ 0.0005 at ANOVA;

Table 2. Glutamate Uptake in AD Patients, MID Patients, and Normal Controls

AD patients MID patients Normal controls

n

Glutamate Uptake (pmol/mg protein/30 min)a

35 10 35

83 ⫾ 9.5b 158 ⫾ 24 130 ⫾ 8.4

Values are expressed as mean ⫾ SEM. p ⬍ 0.0005 vs normal controls and MID patients.

a

b

AD ⫽ Alzheimer’s disease; MID ⫽ multi-infarct dementia.

p ⬍ 0.01 at Dunnett) in AD patients compared with MID patients (see Table 2). Glutamate uptake values were significantly correlated with neither the severity of cognitive alteration, determined by the MMSE and GDS scores, nor the duration of the disease (data not shown). Discussion In the present study we demonstrated a highly significant 40% reduction of glutamate uptake, specific for platelets from AD patients, while no significant uptake modification was described in platelets from MID patients. A similar decrease in glutamate transporter sites, correlated to neurodegeneration, has been previously described in brain from AD patients.8 Our data indicate that the impairment of glutamate uptake might be systemic in AD and that platelets can be used as a peripheral marker of a central glutamatergic dysfunction, which has the advantage of being both independent from secondary or postmortem changes and suitable for clinical studies. Several mechanisms may contribute to a systemic alteration in glutamate uptake in AD. Abnormal APP metabolism, which has been shown in brain as well as in platelets and fibroblasts, might be involved in inhibition of glutamate transporters, through membrane damage and lipid peroxidation products. Specific transporter alterations, linked to abnormal APP expression, have been described in AD brains.15 Platelets express the three major glutamate transporters,16 and we are currently investigating their possible modifications in AD patients. Nevertheless the soluble form of APP displays a neuroprotective effect, and recent data indicate its direct role in facilitating glutamate uptake.17 An alternative hypothesis is that depletion of adenosine triphosphate (ATP) energy stores might impair the energy-dependent glutamate uptake; mitochondrial enzyme alterations have been described in previous studies in brain5 and in various peripheral tissues, including platelets6 and fibroblasts,18 and could lead to ATP depletion. Other possibilities are that the abnormality of mitochondrial oxidative metabolism results in increased production of free radicals, which may contribute to glutamate uptake impairment,19 or in decreased glutamate dehydrogenase activity, with increased intracellular glutamate and “reversed uptake.” In this study, we showed that glutamate uptake is correlated with neither duration of the disease nor severity of dementia, expressed by MMSE and GDS scores. These data could indicate that glutamate uptake impairment and the resulting excitotoxicity may be considered early events in AD, possibly involved as a pathogenic mechanism in cell death, which ultimately results in loss of synapses and clinical appearance of dementia.20 If this interpretation is confirmed by other studies on larger patient populations, on patients at disease on-

set and at follow-up, and possibly in presymptomatic patients carrying AD-specific mutations, this study can be proposed as an early peripheral marker of glutamatergic involvement in the progression to dementia. References 1. Gasparini L, Racchi M, Binetti G, et al. Peripheral markers in testing pathophysiological hypotheses and diagnosing Alzheimer’s disease. FASEB J 1998;12:17–34 2. Di Luca M, Pastorino L, Bianchetti A, et al. Differential level of platelet amyloid ␤ precursor protein isoforms: an early marker for Alzheimer disease. Arch Neurol 1998;55:1195–1200 3. Rosenberg RN, Baskin F, Fosmire JA, et al. Altered amyloid protein processing in platelets of patients with Alzheimer disease. Arch Neurol 1997;54:139 –144 4. Bush AI, Martins RN, Rumble B, et al. The amyloid precursor protein of Alzheimer’s disease is released by human platelets. J Biol Chem 1990;265:15977–15983 5. Parker WD, Parks J, Filley CM, Kleinschmidt-DeMasters BK. Electron transport chain defects in Alzheimer’s disease brain. Neurology 1994;44:1090 –1096 6. Parker WD Jr, Filley CM, Parks JK. Cytochrome oxidase deficiency in Alzheimer’s disease. Neurology 1990;40:1302–1303 7. Swerdlow RH, Parks JK, Cassarino DS, et al. Cybrids in Alzheimer’s disease: a cellular model of the disease? Neurology 1997;49:918 –925 8. Masliah E, Alford M, De Teresa R, et al. Deficient glutamate transport is associated with neurodegeneration in Alzheimer’s disease. Ann Neurol 1996;40:759 –766 9. Mangano RM, Schwarcz R. The human platelet as a model for the glutamatergic neuron: platelet uptake of L-glutamate. J Neurochem 1981;36:1067–1076 10. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984;34:939 –944 11. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatric Res 1975;12:189 –198 12. Reisberg B, Ferris SH, De Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 1982;139:1136 –1139 13. Hachinski VC, Iliff LD, Zilhka E, et al. Cerebral blood flow in dementia. Arch Neurol 1975;32:632– 637 14. Ferrarese C, Zoia C, Pecora N, et al. Reduced platelet glutamate uptake in Parkinson’s disease. J Neural Transm 1999;106: 685– 692 15. Li S, Mallory M, Alford M, et al. Glutamate transporter alterations in Alzheimer disease are possibly associated with abnormal APP expression. J Neuropathol Exp Neurol 1997;56:901–911 16. Ferrarese C, Zoia C, Begni B, et al. Platelet glutamate transporters and uptake in neurodegenerative disorders. Soc Neurosci Abstr 1998;24:178.18 (Abstract) 17. Masliah E, Raber J, Alford M, et al. Amyloid protein precursor stimulates excitatory amino acid transport. J Biol Chem 1998; 273:12548 –12554 18. Gibson G, Martins R, Blass J, Gandy S. Altered oxidation and signal transduction systems in fibroblasts from Alzheimer patients. Life Sci 1996;59:477– 489 19. Volterra A, Trotti D, Tromba C, et al. Glutamate uptake inhibition by oxygen free radicals in rat cortical astrocytes. J Neurosci 1994;14:2924 –2932 20. Szatkowski M, Barbour B, Attwell D. Non-vesicular release of glutamate from glial cells by reversed electrogenic glutamate uptake. Nature 1990;348:443– 446

Brief Communication: Ferrarese et al: Platelet Glutamate Uptake in AD

643

Benzodiazepine Receptor Binding in Huntington’s Disease: [11C]Flumazenil Uptake Measured Using Positron Emission Tomography Gabriella Ku¨nig, MD,*† Klaus L. Leenders, MD,*‡ Rosario Sanchez-Pernaute, MD,§ Angelo Antonini, MD,储 Peter Vontobel, PhD,* Aalt Verhagen, PhD,¶ and Ilonka Gu¨nther, PhD#

We used positron emission tomography and [11C]flumazenil to analyze the benzodiazepine receptor binding in symptomatic and asymptomatic carriers of the Huntington’s disease gene. We found an inverse relationship between [11C]flumazenil and [11C]raclopride binding in the putamen of symptomatic patients, and interpret this result as GABA receptor upregulation. Ku¨nig G, Leenders KL, Sanchez-Pernaute R, Antonini A, Vontobel P, Verhagen A, Gu¨nther I. Benzodiazepine receptor binding in Huntington’s disease: [11C]flumazenil uptake measured using positron emission tomography. Ann Neurol 2000;47:644 – 648

Huntington’s disease (HD) is a genetically determined neurodegenerative disease with autosomal dominant inheritance and complete penetrance. The predominant pathological alterations consist of diffuse loss of neurons and astrogliosis in the caudate nucleus and putamen. The degeneration concerns particularly the striatal medium spiny neurons.1 The earliest neurodegeneration occurs in the caudate nucleus, following a mediolateral gradient with progression of the disease. The pathological stage concurs with the clinical stage of the disease.2 Besides the striatal pathology, some involvement of cor-

From the *PSI, Paul Scherrer Institut, PET Program, Villigen, †Department of Neurology, University Hospital Zu¨rich, Zu¨rich, and ¶VanTx Research Ag Center for Clinical Pharmacology and Development, Basel, Switzerland; ‡Department of Neurology, Groningen University Hospital, Groningen, The Netherlands; §Department of Neurology, Fundacion JimE`nez Dı`az, Madrid, Spain; 储Center for Parkinson’s Disease and Movement Disorders, Istituti Clinici di Perfezionamento, Milan, Italy; and #Service Hospital Frederic Joliot 4, Orsay, France. Received Feb 23, 1999, and in revised form Dec 15. Accepted for publication Dec 18, 1999. Address correspondence to Dr Leenders, Department of Neurology, Groningen University Hospital, Hanzeplein 1, PO Box 30.001, 9700 Groningen, The Netherlands.

644

tical areas has been shown.3 Chorea-free subjects at risk for HD may show reduced striatal D2 receptor binding as well as a decrease in striatal glucose consumption if the underlying pathology has progressed sufficiently. These changes are inexorably progressive.4,5 In postmortem studies of patients with HD, a decreased GABA receptor density in the caudate nucleus and putamen as well as, to a minor degree, in the frontal cortex has been reported.6,7 In a [11C]flumazenil positron emission tomography (PET) study,8 reduced [11C]flumazenil binding in the caudate nucleus of HD patients was detected. The aim of the present study was to investigate [11C]flumazenil binding to GABA receptors in the striatal regions of patients with HD and subjects at risk for HD with (RD) and without (ND) pathologically reduced striatal [11C]raclopride binding to D2 receptors. Additionally, [18F]fluorodeoxyglucose and [11C]raclopride PET scanning was performed in all patients, and correlations between [11C]raclopride binding, glucose consumption, and [11C]flumazenil binding were made in the separate groups. Patients and Methods Patients We investigated 10 patients with manifest HD (mean age, 41 ⫾ 8.8 years) and 13 asymptomatic gene carriers. The asymptomatic subjects were subcategorized into 8 subjects with a pathological reduction of [11C]raclopride binding to striatal D2 receptors on at least one side of the basal ganglia (RD; mean age, 39.5 ⫾ 6 years), and 5 asymptomatic gene carriers without pathological [11C]raclopride PET results (ND; mean age, 29 ⫾ 4.5 years). With the exception of 2 HD patients in whom raclopride was not studied, all subjects were studied with 3 tracers: [11C]raclopride, [11C]flumazenil, and [18F]fluorodeoxyglucose. The PET data were compared with those of separate groups of healthy controls for each tracer: 11 controls for [11C]flumazenil (mean age, 34 ⫾ 5 years), 15 for [11C]raclopride (mean age, 34.5 ⫾ 9 years), and 20 for [18F]fluorodeoxyglucose (mean age, 36 ⫾ 8 years).

Scanning Procedure The scanning procedure and the data acquisition for [11C]raclopride and [18F]fluorodeoxyglucose were identical to previous work, using the binding index for [11C]raclopride binding and the regional glucose index (GMI) for [18F]fluorodeoxyglucose.5 The injected activity was on average 600 MBq for [11C]flumazenil, 264 MBq for [11C]raclopride, and 237 MBq for [18F]fluorodeoxyglucose. The scans with the different tracers were performed on separate days, with an interval of 1 day to 4 weeks.

[11C]Flumazenil The quantification of benzodiazepine receptor binding was studied with a single injection protocol and kinetic analysis using a measured arterial input function. Assuming rapid equilibration of free, nonspecific, and specific bound contributions in tissue compared with blood-brain barrier trans-

Copyright © 2000 by the American Neurological Association

port rate constants, [11C]flumazenil kinetics can be analyzed with a single reversible tissue compartment model. This allows the estimation of the apparent volume of distribution (Vd) of the summed contribution of free, nonspecific, and specific bound [11C]flumazenil, which was shown to be proportional to the regional binding potential Bmax/Kd of the receptor.9 The Vds were finally expressed as ratios of individual regions of interest (ROIs) Vd to global ROI Vd in order to eliminate effects of different specific [11C]flumazenil activities between subjects. The ROIs were placed over the putamen and caudate nucleus. The ROI size of the basal ganglia was 62.5 mm2 for the caudate nucleus and 250 mm2 for the putamen.

Statistics Differences in tracer binding between the groups (control, ND, RD, HD) were calculated according to the nonpara-

metric one-way analysis of variance of Kruskal Wallis, which can be applied for small groups. For the paired comparison, a nonconservative adjustment factor (⌬R(crit) of Schaich and Hamerle) was used. Correlation analysis was performed by Spearman rank correlation and regression analysis ([11C]flumazenil vs [11C]raclopride and GMI). The [11C]flumazenil binding values for each patient are presented in Table 1. Table 2 indicates mean values of all tracers for patient groups and controls.

Results [11C]Flumazenil binding was reduced in the caudate nucleus of the HD patients when compared with the control and ND group ( p ⬍ 0.05). The values of the RD group were in between those of the control and HD groups and were, as such, not significantly differ-

Table 1. [11C]Flumazenil Binding in Patients and Controls

Patient No. HD 1 2 3 4 5 6 7 8 9 10 Mean SD RD 11 12 13 14 15 16 17 18 Mean SD ND 19 20 21 22 23 Mean SD Controls (n ⫽ 11) Mean SD

Vd

Sex

Age (yr)

Disease Duration (yr)

F M M M F F M M M M

50 44 44 48 46 41 33 36 33 44 41.9 6

4 2 2 8 5 1 2 2 0.5 5.0 3.2 2.31

M F M M F F F M

33 33 51 34 39 39 45 42 39.5 6.4

46 45 43 43 45 41 43 42 43.5 1.7

0.44 0.40 0.42 0.40 0.68 0.45 0.42 0.46 0.46 0.09

0.53 0.58 0.55 0.74 0.69 0.58 0.52 0.64 0.60 0.08

M M F M M

27 27 29 24 36 28.6 4.5

43 45 42 45 42 43.4 1.5

0.47 0.49 0.55 0.61 0.51 0.52 0.06

0.75 0.52 0.56 0.48 0.70 0.60 0.12

0.48 0.09

0.57 0.07

34 5

CAG Repeat Length

Caudate

Putamen

50 42 42 44 44 44 49 45 43 46 44.9 2.7

0.31 0.37 0.31 0.33 0.39 0.48 0.46 0.38 0.49 0.56 0.41 0.09

0.43 0.45 0.51 0.52 0.62 0.53 0.64 0.58 0.47 0.74 0.55 0.10

HD ⫽ patients with manifest Huntington’s disease; RD ⫽ asymptomatic gene carriers with decreased [11C]raclopride binding to striatal D2 receptors; ND ⫽ asymptomatic gene carriers with normal raclopride binding to striatal D2 receptors; Vd ⫽ [11C]flumazenil binding as volume of distribution; SD ⫽ standard deviation; the Vd values of left and right: caudate and putamen are pooled.

Brief Communication: Ku¨nig et al: GABA Receptors in Huntington’s Disease

645

Table 2. [11C]Raclopride Binding, [11C]Flumazenil Binding, and [18F]Fluorodeoxyglucose (FDG) Metabolism in the Caudate Nucleus and Putamen of Patients with Manifest Huntington’s Disease (HD), Subjects at Risk for Huntington’s Disease, and Separate Control Groups

Caudate nucleus Controls (n ⫽ 11, 15, 20) ND (n ⫽ 5) RD (n ⫽ 10, 8, 10) HD (n ⫽ 10) Putamen Controls (n ⫽ 11, 15, 20) ND (n ⫽ 5) RD (n ⫽ 10, 8, 10) HD (n ⫽ 10)

Flumazenil

Raclopride

FDG

0.48a⫾ 0.09 0.52a⫾ 0.06 0.46 ⫾ 0.09 0.41a⫾ 0.09

2.63b⫾ 0.48 2.79 ⫾ 0.25 1.76 ⫾ 0.26 1.20b⫾ 0.32

1.21c⫾ 0.07 1.11 ⫾ 0.03 1.04 ⫾ 0.12 0.85c⫾ 0.10

0.57 ⫾ 0.07 0.60 ⫾ 0.12 0.60 ⫾ 0.08 0.55 ⫾ 0.10

2.61b⫾ 0.50 2.51 ⫾ 0.38 1.7 ⫾ 0.2 1.09b⫾ 0.28

1.28c⫾ 0.06 1.19 ⫾ 0.05 1.08 ⫾ 0.12 0.92c⫾ 0.09

Values are means ⫾ SD of [11C]flumazenil volume of distribution (Vd), [11C]raclopride binding index, and FDG metabolic index (GMI). a 11 [ C]flumazenil binding was reduced in the caudate nucleus of the HD patients when compared with the control and ND groups ( p ⬍ 0.05). b 11 [ C]raclopride binding and cGMI were significantly reduced in putamen ( p ⬍ 0.0001) and caudate nucleus ( p ⬍ 0.001) of the HD patients. RD ⫽ asymptomatic gene carriers with reduced [11C]raclopride binding to striatal D2 receptors; ND ⫽ gene carriers with normal [11C]raclopride binding to striatal D2 receptors; n ⫽ number of subjects investigated with the different tracers.

ent from either. In the putamen, we did not find differences between the groups. However, [11C]raclopride binding and GMI were significantly reduced in the putamen ( p ⬍ 0.0001) and caudate nucleus ( p ⬍ 0.001) of the HD patients. There was a (negative) correlation between [11C]raclopride and [11C]flumazenil binding in the putamen of HD patients (n ⫽ 8; Spearman rank, r ⫽ ⫺0.97, p ⬍ 0.01, two-tailed test; regression, p ⫽ 0.0004) (Fig, A). This correlation was lacking in the RD group (n ⫽ 8; Spearman rank, r ⫽ ⫺0.33, p ⫽ NS; regression, p ⫽ 0.42). Furthermore, we found a positive correlation between GMI and [11C]flumazenil binding in the caudate nucleus of the entire patient group (HD ⫹ RD ⫹ ND) (n ⫽ 23; Spearman rank, r ⫽ 0.61, p ⬍ 0.01, two-tailed test; regression, p ⫽ 0.0027) and of the HD patients (n ⫽ 10; Spearman rank, r ⫽ 0.73, p ⬍ 0.05, two-tailed test; regression, p ⫽ 0.01) (see Fig, B). Further correlations between [11C]flumazenil and [11C]raclopride binding index or GMI were not found. GMI and [11C]raclopride binding correlated positively in putamen (n ⫽ 21; Spearman rank, r ⫽ 0.730, p ⬍ 0.01, two-tailed test; regression, p ⫽ 0.0003) and caudate nucleus (n ⫽ 21; Spearman rank, r ⫽ 0.735, p ⬍ 0.01, two-tailed test; regression, p ⫽ 0.0016) of the whole patient group (HD ⫹ RD ⫹ ND), as well as in the putamen of the HD group (n ⫽ 8; Spearman rank, r ⫽ 0.74, p ⬍ 0.05, two-tailed test; regression, p ⫽ 0.01). Discussion Most neostriatal neurons are medium-sized spiny projection neurons.10 These neurons have been shown to degenerate in HD, but signs of regeneration have also been demonstrated.1 Histological postmortem investigations of HD brains revealed a stage-dependent neuronal cell loss reaching 90% in the most advanced

646

Annals of Neurology

Vol 47

No 5

May 2000

stages of the disease.11 Cortical and intrastriatal stimulation experiments revealed a GABA response in these neurons; these responses were induced disynaptically by (1) corticostriatal glutamatergic projection neurons and (2) striatal GABAergic interneurons.12 It is well known that the dopaminergic nigrostriatal projections modulate these GABA responses.13 The GABA receptor complex is composed of several subunits—the GABA receptor itself, the benzodiazepine binding site (benzodiazepine receptor), and several modulatory subunits for the ion channel. Because benzodiazepines change the chloride conductance of the neuronal membrane, [11C]flumazenil, as a benzodiazepine receptor antagonist, can be used for in vivo quantification of GABA receptors.8,14,15 Like Holthoff and colleagues,8 we found reduced 11 [ C]flumazenil binding in the caudate nucleus of the clinically manifest HD patients and normal values in the putamen, whereas the [18F]fluorodeoxyglucose metabolism was reduced in both regions. Additionally, the [11C]raclopride PET showed a reduced binding to D2 receptors in these regions. The correlation between GMI and [11C]flumazenil was significantly positive only in the caudate nucleus of clinically manifest patients and the entire patient group (the ND group was too small to reach significant correlations). We suggest that the reduced [11C]flumazenil binding in the caudate nucleus of HD patients reflects the loss of projection neurons, which probably is marked enough only in clinically manifest subjects to result in a decrease in GABA receptor density. In turn, this decrease correlates with reduced glucose metabolism. The correlation in the entire group might, however, be due to the low values of the HD group because it disappears if these persons are excluded (data not shown). The reduced [11C]raclopride binding and glucose metabolism in the

Fig. (A) Scatterplot of [11C]raclopride and [11C]flumazenil binding in the putamen and caudate nucleus of 8 patients with Huntington’s disease. (B) Scatterplot of [18F]fluorodeoxyglucose and [11C]flumazenil binding binding in the caudate nucleus and putamen of 10 patients with Huntington’s disease. Index ⫽ [11C]raclopride binding index; GMI ⫽ glucose metabolic index; Vd ⫽ [11C]flumazenil volume of distribution.

putamen of the HD group indicates neuronal cell loss. Therefore, the normal [11C]flumazenil binding in the putamen despite these pathological findings suggests compensatory receptor upregulation in the pallidal projection areas of striatal GABAergic medium spiny neurons. Postmortem studies of patients with HD have been performed by means of quantitative receptor autoradiography. Penney and Joung16 reported GABA benzodiazepine receptor decrease in the caudate nucleus and putamen and increase in the pallidum of postmortem brain tissue of a single HD patient. Walker and colleagues17 confirmed these results in a group of 2 early and 4 advanced cases of HD. Since the PET scanner used did not allow accurate differentiation between putamen and globus pallidus, this benzodiazepine receptor decrease in the putamen may be overruled by receptor upregulation in the pallidum of our HD sub-

jects. Furtado and associates18 reported a linear correlation between CAG repeat length and the degree of neuropathological severity in the striatum of patients with HD. A further neuropathological analysis and an in vivo [11C]raclopride PET study in HD gene carriers imply that striatal damage in HD is a linear function of the CAG repeat length times age.19,20 Because we found a positive correlation between GMI and [11C]raclopride binding in the putamen of the entire patient group (HD ⫹ RD ⫹ ND) and of the HD group, we interpret the negative correlation between putaminal [11C]raclopride and [11C]flumazenil binding in our HD population as GABA receptor upregulation, which might be proportional to D2 receptor loss. In the caudate nucleus of HD patients, this correlation may have been lost as a consequence of advanced neuronal cell loss. In our group of preclinical asymptomatic gene

Brief Communication: Ku¨nig et al: GABA Receptors in Huntington’s Disease

647

carriers (RD), this correlation was lacking, perhaps because the neurodegeneration is still “too low” to induce a compensatory GABA receptor upregulation. In conclusion, the [11C]flumazenil PET study presented here shows a compensatory striatal GABA receptor upregulation in patients with manifest HD that is lacking in preclinical gene carriers. This indicates that compensatory processes develop when clinical symptoms appear. We are grateful to Dr W. Hunkeler of Hoffmann-La Roche (Basel, Switzerland) for providing flumazenil and norflumazenil.

References 1. Graveland GA, Williams RS, DiFiglia M. Evidence for degenerative and regenerative changes in neostriatal spiny neurons in Huntington’s disease. Nature 1985;227:770 –773 2. Vonsattel JP, Myers RH, Stevens TJ, et al. Neuropathological classification of Huntington’s disease. J Neuropathol Exp Neurol 1985;44:559 –577 3. Forno LS, Jose C. Huntington’s chorea: a pathological study. Adv Neurol 1973;1:453– 470 4. Grafton ST, Mazziotta JC, Pahl JJ, et al. Serial changes of cerebral glucose metabolism and caudate size in persons at risk for Huntington’s disease. Arch Neurol 1992;49:1161–1167 5. Antonini A, Leenders KL, Spiegel R, et al. Striatal glucose metabolism and dopamine D2 receptor binding in asymptomatic gene carriers and patients with Huntington’s disease. Brain 1996;119:2085–2095 6. Olsen RW, Biochemistry of GABA receptor proteins in mammal brain. In: Yamamura HI, Olsen RW, Usdin E, eds. Psychopharmacology and biochemistry of neurotransmitter receptors. New York: Elsevier/North Holland, 1980:537–550 7. Lloyd KG, Beaumont K, Ziegler M. Assessment of GABA receptors in different CNS diseases by means of radioligand assays. In: Yamamura HI, Olsen RW, Usdin E, eds. Psychopharmacology and biochemistry of neurotransmitter receptors. New York: Elsevier/North Holland, 1980:551–568 8. Holthoff VA, Koeppe RA, Frey KA, et al. Positron emission tomography measures of benzodiazepine receptors in Huntington’s disease. Ann Neurol 1993;34:76 – 81 9. Koeppe RA, Holthoff VA, Frey KA, et al. Compartmental analysis of [11C]flumazenil kinetics for the estimation of ligand transport rate and receptor distribution using positron emission tomography. J Cereb Blood Flow Metab 1991;11:735–744 10. Kemp JM, Powell TPS. The structure of the caudate nucleus of the cat: light and electron microscopy. Phil Trans R Soc Lond 1971;B262:383– 401 11. Myers RH, Vonsattel JP, Paskevich PA, et al. Decreased neuronal and increased oligodendroglial densities in Huntington’s disease caudate nucleus. J Neuropathol Exp Neurol 1991;50: 729 –742 12. Kita H. Glutamatergic and gabaergic postsynaptic responses of striatal spiny neurons to intrastriatal and cortical stimulation recorded in slice preparations. Neuroscience 1995;70:925–940 13. Alexander GE, Crutcher MD, DeLong MR. Basal gangliathalamocortical circuits: paralles substrates for motor, oculomotor, “prefrontal” and “limbic” functions. Prog Brain Res 1990; 85:119 –146 14. Lassen NA, Bartenstein PA, Lammertsma AA, et al. Benzodiazepine receptor quantification in vivo in humans using [11C]flumazenil and PET: application of the steady-state principle. J Cereb Blood Flow Metab 1995;15:152–165

648

15. Delforge J, Spelle L, Bendriem B, et al. Parametric images of benzodiazepine receptor concentration using a partial-saturation injection. J Cereb Blood Flow Metab 1997;17:343–355 16. Penney JB Jr, Joung AB. Quantitative autoradiography of neurotransmitter receptors in Huntington disease. Neurology 1982; 32:1391–1395 17. Walker FO, Young AB, Penney JB, et al. Benzodiazepine and GABA receptors in early Huntington’s disease. Neurology 1984;4:1237–1240 18. Furtado S, Suchowersky O, Rewcastle B, et al. Relationship between trinucleotide repeats and neuropathological changes in Huntington’s disease. Ann Neurol 1996;39:132–136 19. Penney JB, Vonsattel JP, MacDonald ME, et al. CAG repeat number governs the development rate of pathology in Huntington’s disease. Ann Neurol 1997;41:689 – 692 20. Antonini A, Leenders KL, Eidelberg D. [11C]Raclopride-PET studies of the Huntington’s disease rate of progression: relevance of the trinucleotide repeat length. Ann Neurol 1998;43: 253–255

Serum Elastase Activity Is Elevated in Migraine Christophe Tzourio, MD, PhD,*† Mohamed El Amrani, MD,† Ladislas Robert, MD, PhD,‡ and Annick Alpe´rovitch, MD, MSc*

Migraine has been associated with diseases considered to be related to extracellular matrix disorders—in particular, cervical artery dissection. In this population-based study, we found a highly significant association between migraine and the activity of serum elastase, a metalloendopeptidase degrading specific elastin-type amino acid sequences. Such enzymes are involved in matrix degradation. This association was seen in both sexes and was stronger for migraine with aura. These findings could help in the understanding of why patients with migraine are at higher risk of stroke. Further study is needed to establish whether extracellular matrix abnormalities play a broader role in the pathophysiology of migraine. Tzourio C, El Amrani M, Robert L, Alpe´rovitch A. Serum elastase activity is elevated in migraine. Ann Neurol 2000;47:648 – 651

Migraine is usually considered as a neurovascular disorder, although its mechanisms, especially the involve-

From the *INSERM U360, Hoˆpital de la Salpeˆtrie`re, †Service de Neurologie, Hoˆpital Lariboisie`re, and ‡Laboratoire Universitaire de Recherche sur les Therapeutiques Substitutives en Ophtalmologie, Universite´ Paris VI, Hoˆtel-Dieu, Paris, France. Received Aug 24, 1999, and in revised form Dec 28. Accepted for publication Dec 28, 1999. Address correspondence to Dr Tzourio, INSERM U360, Hoˆpital de la Salpeˆtrie`re, 75651 Paris Cedex 13, France.

Copyright © 2000 by the American Neurological Association

ment of vascular factors, remain largely unknown. It has been hypothesized that this vascular involvement may explain the predisposition to ischemic stroke of patients with migraine.1– 4 In cervical artery dissection, a type of stroke for which the association with migraine has been specifically demonstrated,5 abnormalities of the extracellular matrix are thought to play a major role.6,7 It could be hypothesized that the modification of stability and elasticity of the vessel wall induced by extracellular matrix abnormalities may play a role in migraine. Elastases, enzymes capable of degrading elastic fibers, are major regulating enzymes of the extracellular matrix, and human serum has been shown to contain an elastase-type endopeptidase activity, commonly designated as serum elastase activity (SEA).8 We therefore studied the relationship between migraine and SEA. Subjects and Methods Study Participants The Epidemiology of Vascular Aging (EVA) study is a longitudinal study on cognitive and vascular aging of men and women born between 1922 and 1932 recruited from the electoral rolls of the city of Nantes, France.8 During the baseline visit between June 1991 and July 1993, 1,389 participants were recruited. All participants were administered a standardized questionnaire that gave information about demographic background, occupation, medical history, drug use, and personal habits such as cigarette and alcohol consumption. Body mass index (BMI) was computed as weight (in kilograms) divided by height (in meters), squared. Seated blood pressure was measured on two occasions with a digital electronic tensiometer using an oscillometric method (SP 9 Spengler).

Migraine Assessment The migraine study was performed during the second follow-up visit, which took place 4 years after the baseline visit and which involved 1,188 participants. Diagnosis of migraine was made according to the International Headache Society (IHS) criteria.9 Questions on headaches were put during a face-to-face interview by trained lay interviewers using a structured questionnaire with a high reproducibility.1 Special attention was paid to obtaining a lifetime history of migraine by asking participants questions about headaches during early adulthood. A replicate interview with the same structured questionnaire was performed by a neurologist specializing in headaches in 119 of the 238 participants who declared having or having had recurrent headaches. Participants were considered as migraineurs after the neurologist made the diagnosis of migraine (77% of the cases of migraine) or, in those not interviewed by the neurologist, through the lay interviewers’ questionnaire. During the interview, the neurologist could also diagnose the type of migraine, with and without aura, according to the IHS classification.9

Serum Elastase Activity Measurement Blood samples were taken from fasting participants between 8 and 9 AM. SEA was measured according to a procedure

described previously, using a synthetic substrate, N-Suc ala3 pNa.8 The optical density was read at 410 nm in a Dynatech Elisa plaque reader (Dynex, Rungis, France). Crystalline porcine pancreatic elastase (type IV 100 U/mg, Sigma, St Louis, MO) was used to plot the calibration curve. Determinations and standard curves were carried out in duplicate. Determination of SEA was possible in 96% of the participants, and results were expressed as units per milliliter of serum (U/ml). SEA had a mean value (SD) of 0.46 U/ml (0.55 U/ml) in the whole sample; its coefficient of variation was 119.

Statistical Analysis The relationship between migraine and SEA was investigated by analysis of variance, after logarithmic transformation of SEA for statistical testing.8 Potential confounders were identified in univariate analyses and, when necessary, included in multivariate models. In multivariate analyses, adjusted mean SEA (SE) was estimated and compared in participants with and without migraine. Results were presented for the whole population and after stratification by sex. We also analyzed the distribution of migraine prevalence by tertiles of SEA.

Results Migraine status could be determined in 1,185 participants (695 women, 490 men; 99.7% of the cohort). Migraine was diagnosed in 100 participants (85 women), of whom 43% (43/99) complained of persistent attacks at the time of interview. Of the 77 participants with migraine who were interviewed by the neurologist, 16% (12/77) had migraine with aura and 84% (65/77) without; 21% (16/76) declared having had preventive treatment for migraine during their life; and migraine started before the age of 40 years for 89% (68/76). Overall, mean SEA (SD) was higher in participants with migraine than without (0.61 [0.87] vs 0.45 [0.51] U/ml; p ⫽ 0.024). Table 1 presents the relations between migraine and potential confounders of the relationship between migraine and SEA. Participants with migraine were less likely to drink alcohol and had a lower systolic blood pressure than nonmigraineurs, although none of the differences was significant. Age, BMI, blood glucose, and triglycerides, which have been shown to be independently related to SEA,8 were not associated with migraine (see Table 1). We therefore used two different models in multivariate analysis: one simple model including age, sex, and BMI (model 1), and a more general model including all variables that have been shown to be associated with SEA or migraine—age, sex, BMI, alcohol consumption, systolic blood pressure, blood glucose, and triglycerides (model 2). In both models, the difference in mean SEA between participants with and without migraine remained highly significant in the whole sample (0.69 vs 0.44 U/ml; p ⬍ 0.001, in model 2) and in both sexes (Table 2). Analyses by types of migraine showed that mean

Brief Communication: Tzourio et al: Elastase in Migraine

649

Table 1. Baseline Characteristics of Participants with Migraine, by Sex Women No Migraine (n ⫽ 610) Age (yr) BMIa (kg/m2) Tobacco consumption (%) Never Exsmoker Present Mean daily alcohol consumption (%) Never ⬍20 ml ⱖ20 ml Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Blood glucose (mmol/L) Total cholesterol (mmol/L) Triglycerides (mmol/L) Serum elastase activity (U/ml)

Men

Migraine (n ⫽ 85)

p

No Migraine (n ⫽ 475)

Migraine (n ⫽ 15)

p

65.0 (2.9) 25.4 (4.1)

64.7 (3.1) 25.4 (3.9)

0.34 0.95

65.1 (3.0) 26.7 (3.5)

64.6 (2.6) 25.5 (2.6)

0.49 0.22

83 12 5

82 15 4

0.64

24 61 15

7 87 7

0.13

37 49 14 132.2 (17.5) 76.0 (10.4) 5.3 (0.6) 6.5 (1.0) 1.2 (0.5) 0.40 (0.42)

49 43 7 130.8 (17.9) 77.8 (10.8) 5.3 (0.6) 6.5 (1.0) 1.1 (0.5) 0.61 (0.92)

0.06 0.48 0.14 0.78 0.95 0.52 0.023

13 32 56 141.8 (20.2) 79.7 (10.9) 5.7 (1.0) 2.4 (0.4) 1.4 (0.9) 0.51 (0.59)

13 40 47 135.3 (12.7) 77.4 (11.7) 5.2 (1.0) 2.4 (0.3) 1.4 (0.6) 0.64 (0.48)

0.76 0.07 0.42 0.05 0.99 0.93 0.07

Standard deviations are given in parentheses. a

Body mass index computed as weight (in kilograms) divided by the height squared (in meters).

Table 2. Multivariate Analysis of the Relationship Between Migraine and Serum Elastase Activity (SEA)

0.020). The same pattern was observed in both sexes (data not shown).

Mean SEA (SEM)

Women Model 1a Model 2b Men Model 1a Model 2b Both sexes Model 1a Model 2b

No Migraine

Migraine

p

0.40 (0.02) 0.40 (0.02)

0.61 (0.06) 0.66 (0.06)

0.022 0.001

0.51 (0.03) 0.50 (0.03)

0.66 (0.16) 0.71 (0.15)

0.039 0.010

0.44 (0.02) 0.44 (0.02)

0.65 (0.06) 0.69 (0.06)

0.005 ⬍0.001

a

Adjusted for age, sex, and BMI. Adjusted for age, sex, BMI, alcohol consumption, systolic blood pressure, triglycerides, and blood glucose. b

BMI ⫽ body mass index; SEM ⫽ standard error of the mean.

SEA (SE) was much higher in migraine with aura (1.05 [0.24] U/ml) than in migraine without aura (0.66 [0.07] U/ml) and in participants without migraine (0.44 [0.02] U/ml; 3 group analysis of variance, p ⬍ 0.001, in model 2). When migraine types were directly compared, SEA was found to be significantly higher in migraine with aura than in migraine without (1.05 [0.24] vs 0.62 [0.11] U/ml; 2 group analysis of variance, p ⫽ 0.038, in model 2). Finally, we analyzed the distribution of the prevalence of migraine by tertiles of SEA. There was a regular rise in the prevalence of migraine with increasing tertiles of SEA (6.2%, 8.3%, and 10.7%; p for trend ⫽

650

Annals of Neurology

Vol 47

No 5

May 2000

Discussion In this population-based study in elderly participants, we identified a positive and significant association between migraine and SEA. To our knowledge, this is the first time that an association between migraine and an extracellular matrix–regulating enzyme has been described. This association may point to one of the mechanisms underlying the relationship between migraine and stroke.1– 4 If patients with migraine had a higher level of extracellular matrix degradation, they would therefore be more exposed than nonmigraineurs to the risk of cervical artery dissection5 or to vessel atherosclerotic lesions,10,11 leading to an ischemic stroke. This association could also help to understand better the association observed between migraine and disorders for which an extracellular matrix involvement has been demonstrated or is suspected, such as mitral valve prolapse12,13 or patent foramen ovale.14 A potential limitation of the study is that it is based on an elderly population. However, the participants of the EVA study had high cognitive scores,15 and it is unlikely that a painful and recurrent condition such as migraine could have been missed by the detailed questionnaire on headaches that was used. Instructions were also given to the interviewers to emphasize headaches that occurred during early adulthood. The major strength of the present study is its design: a populationbased sample with a high participation rate, a detailed structured questionnaire on migraine based on the IHS

criteria, and blinding of SEA measurement with regard to migraine status. This design renders highly unlikely the biases usually considered for less-representative studies (eg, recruitment and selection biases, exposure measurement bias, selective recall bias). It is also unlikely that this finding is due to chance alone: the association was strong despite the limited number of participants with migraine, and it was consistent in both sexes. Furthermore, SEA was much higher in migraine with aura than without, which is in agreement with the relationship observed between migraine and stroke.1,2,4 The regular increase in migraine frequency by tertiles of SEA in the whole sample and in both sexes is also reassuring. Moreover, we verified that this association was not related to any of the following: (1) late-onset migraine: those whose migraine began after 40 years of age had a lower SEA (0.20) than the others (0.65); (2) preventive treatment: participants declaring having had a preventive treatment had a lower SEA (0.44) than those without treatment (0.62); and (3) active migraine: participants still subject to attacks had a similar SEA (0.66) as those without attacks (0.71). These last results reinforce the view that this association was not related to atypical forms of migraine or to treatment. These findings, if confirmed in other samples—particularly in patients with cervical artery dissection— may improve understanding of the association between migraine and ischemic stroke. More broadly, it has recently been shown that elastin peptides produced by degradation of elastin by elastase16 can induce a nitric oxide-dependent vasodilatation.17 There is also recent evidence that nitric oxide triggers migraine attacks.18,19 The potential involvement of extracellular matrix modifications in the pathophysiology of migraine remains to be investigated.

6. Brandt T, Hausser I, Orberk E, et al. Ultrastructural connective tissue abnormalities in patients with spontaneous cervicocerebral artery dissections. Ann Neurol 1998;44:281–285 7. Tzourio C, Cohen A, Lamisse N, et al. Aortic root dilatation in patients with spontaneous cervical artery dissection. Circulation 1997;95:2351–2353 8. Bizbiz L, Bonithon-Kopp C, Ducimetie`re P, et al. Relation of serum elastase activity to ultrasonographically assessed carotid artery wall lesions and cardiovascular risk factors: the EVA study. Atherosclerosis 1996;120:47–55 9. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988; 8(Suppl 7):1–96 10. Mohacsi A, Fulop TJ, Kozlovszky B, et al. Sera and leukocyte elastase-type protease and antiprotease activity in healthy and atherosclerotic subjects of various ages. J Gerontol 1992;47: B154 –B158 11. Robert L, Robert AM, Jacotot B. Elastin-elastase-atherosclerosis revisited. Atherosclerosis 1998;140:281–295 12. Spence JD, Wong DG, Melendez LJ, et al. Increased prevalence of mitral valve prolapse in patients with migraine. Can Med Assoc J 1984;131:1457–1460 13. Heck AF. Neurologic aspects of mitral valve prolapse. Angiology 1989;40:743–751 14. Anzola GP, Magoni M, Guindani M, et al. Potential source of cerebral embolism in migraine with aura: a transcranial Doppler study. Neurology 1999;52:1622–1625 15. Berr C, Dufouil C, Brousseau T, et al. Early effect of ApoE-⑀4 allele on cognitive results in a group of highly performing subjects: the EVA study. Neurosci Lett 1996;218:9 –12 16. Fulop TJ, Wei SM, Robert L, Jacob MP. Determination of elastin peptides in normal and arteriosclerotic human sera by ELISA. Clin Physiol Biochem 1990;8:273–282 17. Faury G, Ristori MT, Verdetti J, et al. Effect of elastin peptides on vascular tone. J Vasc Res 1995;32:112–119 18. Thomsen LL, Olesen J. Nitric oxide theory of migraine. Clin Neurosci 1998;5:28 –33 19. Martelletti P, D’Alo S, Stirparo G, et al. Modulation of nitric oxide synthase by nitric oxide donor compounds in migraine. Int J Clin Lab Res 1998;28:135–139

The EVA study is supported by the Merck, Sharp and DohmeChibret Laboratories (West Point, PA) and the Eisai Company (France).

References 1. Tzourio C, Tehindrazanarivelo A, Iglesias S, et al. Case-control study of migraine and risk of ischaemic stroke in young women. Br Med J 1995;310:830 – 833 2. Carolei A, Marini C, Dematteis G, et al. History of migraine and risk of cerebral ischemia in young adults. Lancet 1996;347: 1503–1506 3. Buring JE, Hebert P, Romero J, et al. Migraine and subsequent risk of stroke in the physicians’ health study. Arch Neurol 1995;52:129 –134 4. Chang CL, Donaghy M, Poulter N. Migraine and stroke in young women: case-control study. World Health Organisation Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception. Br Med J 1999;318:13–18 5. d’Anglejean-Chatillon J, Ribeiro V, Mas JL, et al. Migraine: a risk factor for dissection of cervical arteries. Headache 1989;29: 560 –561

Brief Communication: Tzourio et al: Elastase in Migraine

651

Evidence for Infection with Chlamydia pneumoniae in a Subgroup of Patients with Multiple Sclerosis Gerlinde Layh-Schmitt, PhD,* Claudia Bendl,† Ulrike Hildt,* MD, Tuan Dong-Si,† Eric Ju¨ttler,† Paul Schnitzler, PhD,* Caspar Grond-Ginsbach, PhD,† and Armin J Grau, MD†

In a pilot study, we identified Chlamydia pneumoniae in the cerebrospinal fluid by polymerase chain reaction in 5 of 10 patients with definite multiple sclerosis (MS). In a second series, 2 of 20 patients with definite MS and 3 of 17 patients with possible/probable MS or MS variants, but none of 56 patients with other neurological, diseases were polymerase chain reaction–positive. We confirm that C pneumoniae can be found in the cerebrospinal fluid of MS patients, but our rate of positive results is lower than in a recent report. Layh-Schmitt G, Bendl C, Hildt U, Dong-Si T, Ju¨ttler E, Schnitzler P, Grond-Ginsbach C, Grau AJ. Evidence for infection with Chlamydia pneumoniae in a subgroup of patients with multiple sclerosis. Ann Neurol 2000;47:652– 655

The etiology of multiple sclerosis (MS) is unknown. Epidemiological, genetic, and clinical evidence suggests that infectious disease may play a pathogenetic role. However, causative microbial agents could not be identified so far. A recent study detected Chlamydia pneumoniae in the cerebrospinal fluid (CSF) of most patients with MS but only in few control subjects.1,2 We investigated whether C pneumoniae can be identified in the CSF of patients with MS. Subjects and Methods In a pilot study, we investigated 10 patients (age, 33 ⫾ 14 years, mean ⫾ SD) with definite MS according to the Poser criteria (Table 1) and a 36-year-old male patient with cerebral ischemia. Patients were admitted between March and June 1998. In a second series started in October 1998, we studied 20 patients (age, 38 ⫾ 9 years) with definite MS (see Table 1) and 17 patients (age, 39 ⫾ 11 years) with probable or possible MS (n ⫽ 9) or supposed MS variants—myelitis

From the Departments of *Microbiology and †Neurology, University of Heidelberg, Germany. Received Oct 26, 1999, and in revised form Dec 29. Accepted for publication Dec 29, 1999. Address correspondence to Dr Grau, Department of Neurology, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany.

652

of unknown origin (n ⫽ 5), neuromyelitis optica (n ⫽ 1), or acute demyelinating encephalomyelitis (n ⫽ 2) (Table 2). Fifty-six patients (28 women) under the age of 65 years (42 ⫾ 12 years) served as the control group. Exclusion criteria for the control group were presumed immunological or noninfectious inflammatory diseases of the central nervous system. Control patients had the following diagnoses: cerebral ischemia (n ⫽ 9), cephalgia (n ⫽ 8), meningoencephalitis of viral or unknown origin (n ⫽ 7), neuroborreliosis (n ⫽ 2), probable tuberculous meningitis (n ⫽ 1), brain abscess (n ⫽ 1), polyradiculitis/polyneuritis (n ⫽ 6), brain or spinal cord tumors (n ⫽ 5), cranial nerve palsy (n ⫽ 6), epileptic seizures (n ⫽ 3), and others (n ⫽ 8). Twenty-six control patients had CSF pleocytosis (n ⫽ 18; range 6 –340 cells/␮l), pathological protein analysis (n ⫽ 22), or both. Patients gave informed consent. The study protocol was approved by the local Ethics Committee. DNA was extracted from 1 ml of frozen (⫺20°C) CSF using the Qiagen (Hilden, Germany) blood kit. The ompA gene of C pneumoniae was used as target for a nested polymerase chain reaction (PCR) using the method by Tong and Sillis3 (external primers: sense, 5⬘-TTACAAGCCTTGCCTGTAGG-3⬘; antisense, 5⬘-GCGATCCCAAATGTTTAAGGC-3⬘; “touchdown” PCR with 32 cycles; internal primers: sense, 5⬘-TTATTAATTGATGGTACAAT-3⬘; antisense, 5⬘-ATCTACGGCAGTAGTATAGT-3⬘; 40 cycles). To test whether sensitivity can be improved, an additional amplification (primers: sense, 5⬘-GAACCCTTCTGATCCAAGCT3⬘; antisense, 5⬘-AATGAAGCCTGCATTAGTGA-3⬘; 40 cycles, annealing temperature 60°C, denaturation temperature 94°C, extension temperature 72°C, holding time 1 minute) was later introduced between both initial amplification steps. Amplification products were analyzed by agarose (3%) gel electrophoresis with ethidium bromide staining. C pneumoniae TW-183 was used as positive control in PCR and cell culture. Each PCR was repeated at least twice. Pyruvate dehydrogenase gene was amplified to detect an inhibition of PCR amplification.4 A subgroup of CSF samples was investigated using the method by Sriram and associates2 except from detecting the probe with the ECL luminescence system (Amersham-Pharmacia, Little Chalfont, UK). To analyze the sensitivity of our PCR, we amplified DNA from C pneumoniae TW-183 with external primers and estimated the concentration of the amplified DNA by comparison with a quantified standard DNA fragment after agarose gel electrophoresis and ethidium bromide staining. Serial dilutions resulted in the detectability of about 10 molecules of template DNA in our PCR. Using the forward internal primer, positions 120 to 286 of the ompA gene were analyzed in PCR products applying the 310 ABI-Prism genetic analyzer (Perkin Elmer, Weiterstadt, Germany). For culture of C pneumoniae, human lung carcinoma cells were grown on cover slips (12 mm in diameter) in 24-well cell culture plates containing Dulbecco’s modified Eagle’s medium (DMEM) (Life Technologies, Germany) with 10% fetal calf serum, nonessential amino acids, glutamine, 50␮g/ml gentamycin, and 100 ␮g/ml vancomycin. Within 2 hours after lumbar puncture, CSF (300 ␮l in 700 ␮l DMEM) was centrifuged (1 hour, 800g, 37°C) on humanlung carcinoma cells. After incubation for 1 additional hour (37°C, 5% CO2), the medium was changed to DMEM con-

Copyright © 2000 by the American Neurological Association

Table 1. Patients with Definite Multiple Sclerosis Subdivided According to Results of Polymerase Chain Reaction (PCR) for Chlamydia pneumoniae Patients with Positive PCR for Chlamydia pneumoniae

No. of patients Age, yr (mean ⫾ SD) Female gender (n) Duration of disease, yr (mean ⫾ SD) Course of disease Relapsing-remitting (n) Secondary chronic progressive (n) Primary chronic progressive (n) Cerebrospinal fluid Cells, per ␮l (mean ⫾ SD) Oligoclonal bands (n) Intrathecal IgG production (n)

Patients with Negative PCR for Chlamydia pneumoniae

1st Series

2nd Series

Total

1st Series

2nd Series

Total

5 38 ⫾ 20 3 3.3 ⫾ 3.4

2 37 1 1.5

7 37 ⫾ 16 4 (57%) 2.8 ⫾ 2.9

5 27 ⫾ 3 1 3.5 ⫾ 4.0

18 38 ⫾ 9 9 6.8 ⫾ 8.3

23 36 ⫾ 9 10 (43%) 6.1 ⫾ 7.6

4 1 0

1 0 1

5 (71%) 1 (14%) 1 (14%)

13 4 1

17 (74%) 5 (22%) 1 (4%)

7⫾5 3 4

4 2 2

6⫾5 5 (71%) 6 (86%)

14 ⫾ 12 16 11

14 ⫾ 12 21 (91%) 16 (70%)

4 1 0 14 ⫾ 6 5 5

Percentages in parentheses are related to the total number of subjects with positive or negative PCR results, respectively. Differences between all patients with positive PCR results and all patients with negative PCR results were not statistically significant by Mann-Whitney U test (continuous variables) or Fisher’s exact test (categorial variables) ( p ⬎ 0.10).

taining cycloheximide (0.6 ␮g/ml). Immunofluorescence tests were carried out after 3, 6, and 10 days using a direct immunofluorescence assay with genus-specific fluorescein isothiocyanate conjugated monoclonal mouse antibodies (Progen, Heidelberg, Germany). Cell cultures were judged blinded to PCR results. In subgroups of patients, serological analysis was performed by S. P. Wang and J. Thomas Grayston, Seattle, WA, using a microimmunofluorescence test (MIF).5 We compared continuous variables by Mann Whitney U test and categorical variables by Fisher’s exact test.

Results In the pilot study, PCR for C pneumoniae was positive in 5 of 10 patients with definite MS and negative in the patient with cerebral ischemia (Fig). In the second series, PCR was positive in 1 patient with neuromyelitis optica but negative in all other subjects, including all patients with definite MS. Searching for more sensitive methods, we thereafter performed a third amplification step in all 104 samples. In such modified PCR,

Table 2. Patients with Neuromyelitis Optica, Myelitis, Acute Demyelinating Encephalomyelitis (ADEM) and Probable or Possible Multiple Sclerosis (MS)

Diagnosis

No. of Patients

Neuromyelitis optica Recurrent myelitis

1 2

Myelitis

3

ADEM

2

Probable MS

2

Possible MS

7

Cerebrospinal Fluid

Chlamydia pneumoniae PCR

Age yr/Sex

Course of Disease

Leukocytes (per ␮l)

Oligoclonal Bands

Intrathecal IgG prod.

⫹ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺

58/M 40/M 49/M 30/F 32/F 21/M 48/F 34/M 38/F 42/F 18/F 35/F 44/F 58/F 30/M 40/M 45/M

RR RR RR 1 epi. 1 epi. 1 epi. 1 epi. 1 epi. 1 epi. RR 1 epi. 1 epi. 1 epi. 1 epi. 1 epi. 1 epi. 1 epi.

53 16 3 11 9 12 11 13 7 15 1 4 8 10 2 1 1

⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹

PCR ⫽ polymerase chain reaction; prod. ⫽ production; RR ⫽ relapsing remitting; epi. ⫽ episode.

Brief Communication: Layh-Schmitt et al: Chlamydia pneumoniae and MS

653

Fig. Ethidium bromide–stained agarose gel of polymerase chain reaction (PCR) products (length, 207 base pairs) after amplification of the ompA gene of Chlamydia pneumoniae out of cerebrospinal fluid samples of patients with definite multiple sclerosis (lane 2 and lanes 4 –12; positive PCR in lanes 2, 5, 9, 10 and 12) and a patient with cerebral ischemia (lane 3) (lane 1, standards; lane 13, negative control; lane 14, positive control).

results from the pilot study remained unchanged, but 2 previously negative patients with definite MS from the second series and 2 patients with probable/possible MS were now positive. All control subjects remained negative. In patients with definite MS, positive PCR was more common than in the control group (7/30 vs 0/56; p ⬍ 0.001) and more common in the first (5/10) than in the second series (2/20; p ⫽ 0.047). Age, duration of disease, and CSF analyses were not different between definite MS patients with positive (n ⫽ 7) versus negative PCR (n ⫽ 23) (see Table 1). Pyruvate dehydrogenase gene amplification always resulted in positive findings. The method of Sriram and colleagues2 did not yield positive results in our hands in 10 PCR-negative patients with definite MS. We sequenced the amplicons in 6 patients. In 5 patients, the sequence was identical to the one published by Carter and co-workers.6 This confirms the specificity of the PCR. An aberration at one position (265 C3 T) was present in 1 patient. C pneumonia–specific IgG was detectable in low titers in CSF of 2 patients (both PCR-negative) and 2 control subjects. CSF IgA and IgM and serum IgM were always negative. Serum titers were not different either between patients with definite MS (n ⫽ 19; IgG: median 1:128, range 0 –1:512; IgA: median 0, range 0 –1:256) and control subjects (n ⫽ 42; IgG: median 1:128, range 0 –1:1024; IgA: median 0, range 0 –1:512) or between patients with positive (n ⫽ 6) and negative (n ⫽ 13) PCR. Culture of CSF specimens for detection of C pneumoniae was performed in 8 patients, 4 with positive PCR. In 2 patients—both of whom were PCR-positive—there was evidence for C pneumoniae elementary bodies but typical inclusion bodies and, thus, cell division was not detected.

654

Annals of Neurology

Vol 47

No 5

May 2000

Discussion Using an established nested PCR, we initially detected C pneumoniae in a high percentage of patients. Such high prevalence was not confirmed in a second series of patients. A third amplification yielded positive results in few additional patients. This could indicate that the two initially used primer pairs—although very sensitive in our positive control–may not be ideally suited for detection of C pneumoniae in all patients’ samples. Similar to our findings, Gieffers and associates7 identified C pneumoniae in CSF of 22% (9/41) of patients with definite or suspected MS by nested PCR. In contrast, Sriram and colleagues2 detected C pneumoniae in 97% of MS patients. In neither our study nor the study by Gieffers and co-workers7 were PCR results supported by the MIF test. Using an enzyme-linked immunosorbent assay, Sriram and associates found increased CSF antibody titers against C pneumoniae in MS patients.2 Recently, low or undetectable MIF titers against C pneumoniae were detected in CSF of patients with meningoencephalitis and acute C pneumoniae infection.8 Thus, the MIF test, although most sensitive in serum, may not be ideally suited for CSF analysis. C pneumoniae infection was previously shown in few patients with meningoencephalitis and polyradiculitis.8,9 One may assume that the presence of C pneumoniae in CSF may be associated with cellular inflammation of any type. However, negative results in all our 18 control subjects with CSF pleocytosis do not support such hypothesis. Our results confirm that C pneumoniae can be detected in MS patients. However, the rate of positive results was lower than previously reported, and prevalences in our two series were at variance. The role of C pneumoniae in MS requires further investigation, and it appears urgent to establish the most appropriate method to detect C pneumoniae in CSF. We thank San-pin Wang and J. Thomas Grayston, University of Seattle, WA, for performing the serological analyses and Wolfgang Stille, Frankfurt am Main, Germany, for his helpful suggestions during our work.

References 1. Sriram S, Mitchell W, Stratton C. Multiple sclerosis associated with Chlamydia pneumoniae infection of the CNS. Neurology 1998;50:571–572 2. Sriram S, Stratton CW, Yao SY, et al. Chlamydia pneumoniae infection of the central nervous system in multiple sclerosis. Ann Neurol 1999;46:6 –14 3. Tong CYW, Sillis M. Detection of Chlamydia pneumoniae and Chlamydia psittaci in sputum samples by PCR. J Clin Pathol 1993;46:313–317 4. Ho L, Javed AA, Pepin RA, et al. Identification of a cDNA clone for the ␤-subunit of the pyruvate dehydrogenase component of human pyruvate dehydrogenase complex. Biochem Biophys Res Commun 1988;150:904 –908 5. Wang SP, Grayston JT. Immunologic relationship between gen-

6.

7.

8.

9.

ital TRIC, lymphogranuloma venereum, and related organisms in a new microtiter indirect immunofluorescence test. Am J Ophthalmol 1970;70:367–374 Carter MW, Al-Mahdawi SAH, Giles IG, et al. Nucleotide sequence and taxonomic value of the major outer membrane protein gene of Chlamydia pneumoniae IOL-207. J Gen Microbiol 1991;137:465– 475 Gieffers J, Pohl D, Korenke G, et al. Chlamydia pneumoniae infections in patients with multiple sclerosis and related pathologies. Proceedings of the 99th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, September 26 –29, 1999:718 (Abstract) Koskiniemi M, Gencay M, Salonen O, et al. Chlamydia pneumoniae associated with central nervous system infections. Eur Neurol 1996;36:160 –163 Korman TM, Turnidge JD, Grayson ML. Neurological complications of chlamydial infections: case report and review. Clin Infect Dis 1997;25:847– 851 Levodopa-Induced Dyskinesias in Parkinson’s Disease: Is Sensitization Reversible?

Levodopa-Induced Dyskinesias in Parkinson’s Disease: Is Sensitization Reversible? Boulos-Paul Bejjani, MD,*† Isabelle Arnulf, MD,* Sophie Demeret, MD,* Philippe Damier, MD, PhD,* Anne-Marie Bonnet, MD,* Jean-Luc Houeto, MD,* and Yves Agid, MD, PhD*

Levodopa-induced dyskinesias (LIDs) in patients with Parkinson’s disease are considered to result from the severity of dopaminergic denervation in the striatum, which is an irrevocable phenomenon, and sensitization induced by long-term intermittent administration of levodopa. Taking advantage of the 64% reduction of levodopa treatment allowed in 12 Parkinson’s disease patients by continuous high-frequency stimulation of the subthalamic nucleus, we evaluated the severity of parkinsonian motor disability and LIDs during two levodopa challenges performed before the surgical implantation of the stimulation electrodes and after 8.8 months of continuous bilateral subthalamic nucleus stimulation that was interrupted 2 hours before the levodopa test. Motor disability during

From the *Centre d’Investigation Clinique, Fe´de´ration de Neurologie and INSERM U289, Groupe-Hospitalier Pitie´-Salpeˆtrie`re, Paris, France; and †Unite´ des troubles du mouvement, Hoˆpital NotreDame des Secous, Byblos, Lebanon. Received May 6, 1999, and in revised form Jan 4, 2000. Accepted for publication Jan 4, 2000. Address correspondence to Dr Arnulf, Centre d’Investigation Clinique, Groupe-Hospitalier Pitie´-Salpeˆtrie`re, 47-83 Boulevard de l’Hoˆpital, 75013 Paris, France.

the “off” and “on” drug periods was unchanged. The severity of LIDs during the “on” period and dystonia during the “off” period decreased by 54% and 62%, respectively. The reduced severity of LIDs in the absence of subthalamic nucleus stimulation demonstrates that the sensitization phenomenon resulting from long-term intermittent levodopa administration is partially reversible. Bejjani B-B, Arnulf I, Demeret S, Damier P, Bonnet A-M, Houeto J-L, Agid Y. Levodopainduced dyskinesias in Parkinson’s disease: is sensitization reversible? Ann Neurol 2000;47:655– 658

Levodopa-induced dyskinesias (LIDs) are disabling complications of long-term levodopa treatment in patients with Parkinson’s disease (PD). They are considered to result from the conjunction of at least two factors: the severity of dopaminergic brain lesions, as they are not observed in normal individuals receiving therapeutic doses of levodopa, and chronic intermittent administration of levodopa, as dyskinesias are not observed in untreated PD patients.1 There is no known way of intervening in the neurodegenerative process, which irrevocably worsens with time. The only factor that can be controlled is the administration of antiparkinsonian treatment. Various experimental studies have shown that the administration of levodopa and related compounds in animals with dopaminergic lesions causes the sensitization (ie, brain changes, including hypersensitivity of neurotransmitter receptors2) responsible for the early appearance and increased severity of LIDs. Whether the capacity of levodopa to generate LIDs in PD patients is reversible constitutes the key question to determine whether sensitization is reversible. This would presumably be the case if LIDs are abolished or if their severity is reduced after long-term withdrawal of antiparkinsonian treatment. Conversely, if LIDs persist with the same severity despite a marked reduction or interruption of levodopa treatment, this would suggest that levodopa triggers an irreversible dysfunction of the basal ganglia responsible for the sensitization phenomenon contributing to the long-term development of LIDs during the course of the disease. To determine whether the capacity of levodopa to induce LIDs is reduced or abolished in PD patients in whom chronic intermittent levodopa treatment has been diminished or withdrawn for several months, long-term reduction or withdrawal of antiparkinsonian treatment was made possible by the reduction in parkinsonian disability achieved by continuous bilateral high-frequency stimulation of the subthalamic nucleus (STN).

Copyright © 2000 by the American Neurological Association

655

Table 1. Characteristics of the Patients and Antiparkinsonian Treatment before and after Surgery Preoperative

Postoperative

Patient No.

Age/Sex (yr)

Disease/Treatment Duration (yr)

Levodopa (mg/day)

Intake (no./day)

Agonist (mg/day)

Duration (mo)

Levodopa (mg/day)

Intake (no./day)

Agonist (mg/day)

1 2 3 4 5 6 7 8 9 10 11 12

39/M 45/M 74/F 64/F 50/M 54/M 65/M 46/F 59/M 61/F 46/M 56/M

16/16 11/8 25/20 28/28 10/7 17/12 11/10 15/13 14/13 8/6 11/10 16/16

850 1,400 350 900 950 450 1,050 1,000 1,200 800 750 1,600

5 7 6 4 8 6 8 5 7 7 6 7

40.0 15.0 0 25.0 60.0 6.0 45.0 50.0 37.5 95.0 50.0 30.0

25 14 12 11 10 10 9 5 4 3 2 1

700 350 0 300 200 250 550 200 1,000 500 100 0

7 4 0 5 4 5 5 4 8 5 4 0

35.0 7.5 0 0 20.0 0 0 30.0 0 0 0 0

Mean ⫾ SE

55.0 ⫾ 2.9

15.1 ⫾ 1.7/ 13.25 ⫾ 1.8

950 ⫾ 107

6.3 ⫾ 0.4

37.8 ⫾ 7.4

8.8 ⫾ 1.9

345.8 ⫾ 85.6a

4.25 ⫾ 0.7a

7.7 ⫾ 3.8a

a

p ⬍ 0.05 when comparing postoperative with preoperative levodopa, and number of doses and agonist doses (equivalent dose of bromocriptine) per day.

Patients and Methods Patients

STN stimulation was stopped 2 hours before the postoperative levodopa challenge.

Twelve patients being treated with continuous bilateral highfrequency stimulation of the STN delivered through surgically implanted electrodes were enrolled in this study (Table 1), which was approved by the local Ethics Committee. All had a severe form of idiopathic PD with a greater than 50% improvement of symptoms under levodopa treatment (Unified Parkinson’s Disease Rating Scale Part III [UPDRS-III]3) and presented with disabling motor fluctuations and LIDs. The neurosurgical procedure and target determination were performed as previously described.4 All patients included in the study had good therapeutic benefit from STN stimulation as demonstrated at the time of the postoperative study under stimulation by a 64% reduction in motor disability (UPDRS-III levodopa treatment withdrawn, p ⬍ 0.01); an 89% and 93% reduction in LIDs and motor fluctuations (UPDRS-IV a and b, p ⬍ 0.01), respectively (not shown); and a 64% ( p ⬍ 0.01), 80% ( p ⬍ 0.01), and 32.5% ( p ⬍ 0.01) reduction in the mean total quantity of levodopa and dopaminergic agonists and number of doses administered per day, respectively (Patients 3 and 12 were no longer receiving medication) compared with preoperative study values (see Table 1).

Study Design Patients underwent two videotaped levodopa challenges. The first was performed in the month before surgery while patients were receiving chronic levodopa treatment (see Table 1). The second took place 8.8 ⫾ 1.9 months (range, 1–25 months) after the start of continuous bilateral STN stimulation. Antiparkinsonian treatment was kept stable for at least 1 week before the two evaluations. Levodopa challenges were performed between 9:00 and 10:00 AM while patients were fasting and after a 12-hour period of drug withdrawal, using a single administration of a suprathreshold levodopa dose (ie, 50 mg more than the usual preoperative morning dose). For each patient, the dose of levodopa was identical in both tests.

656

Annals of Neurology

Vol 47

No 5

May 2000

Outcome Measures Patients were scored at baseline and 1 hour after levodopa intake at a time of maximal motor improvement. The severity of motor disability was assessed using the UPDRS-III motor score. LIDs were assessed by the same observer on each occasion using a global five-point scale (0 ⫽ nil, 1 ⫽ minimal severity, 2 ⫽ moderate severity, 3 ⫽ severe, 4 ⫽ violent movements resulting in major disability) performed with the patients in a resting condition.4 Two sets of LIDs were considered: “on” LIDs were observed at the time of maximal clinical improvement, and “off” dystonia was observed at baseline preceding levodopa administration.

Results During the levodopa challenge, the baseline (“off ”) and treated (“on”) motor scores (UPDRS-III) and the percentage of motor improvement were not significantly different before and after surgery (Table 2). The severity of dystonia during the “off ” period and LIDs during the “on” period was reduced by 62% ( p ⬍ 0.05) and 54% ( p ⬍ 0.05), respectively (Wilcoxon paired test; Fig). Although no quantification was performed, the type of LIDs observed during the two levodopa challenges did not differ. No significant correlation was found between the severity of LIDs and either the duration of STN stimulation, the percentage of reduction in daily doses of levodopa, or a composite score combining both. In Patients 3 and 12, in whom antiparkinsonian medication was totally withdrawn, the improvement in the severity of LIDs during the “on” period (⫺50% and ⫺33%, respectively) was not different from that obtained in other patients.

Table 2. Parkinsonian Motor Disability (UPDRS Part III) in the “Off ” and “On” Drug States Obtained during Levodopa Challenges Performed before and after Surgery Parkinsonian Motor Score Baseline (“off ”) Treated (“on”) p for comparison “on” vs “off ” drug scores

Preoperative

Postoperativea

p for Comparison Preoperative vs Postoperative States

50.8 ⫾ 4.8b 15.9 ⫾ 2.0b 0.001

46.4 ⫾ 3.7b 13.0 ⫾ 2.0b 0.001

NS NS

a

Continuous subthalamic high-frequency stimulation was stopped 2 hours before the postoperative levodopa challenge. Results are mean ⫾ SE.

b

NS ⫽ not significant.

peated administration of levodopa doses during the day; however, striatal dopaminergic stimulation is constantly increased because of the increased daily dose of levodopa.8 In contrast to these two methods, the paradigm used in our study had the advantage of markedly reducing the levodopa daily dose during periods of time ranging from 1 to 25 months. Using this paradigm, we report that the long-term reduction in daily intake and number of doses of levodopa in PD patients treated by continuous STN stimulation resulted in a decreased capacity of levodopa to generate LIDs during an acute challenge. These results should be interpreted with caution for the following reasons: Fig. Effects of a single dose of levodopa on the severity of levodopa-induced dyskinesias in parkinsonian patients during two levodopa challenges: before surgery and after marked antiparkinsonian drug reduction made possible by continuous high-frequency stimulation of the subthalamic nucleus (test performed 2 hours after stimulation was switched off). Results are the mean ⫾ SEM obtained in 12 patients. Columns with dashed lines represent preoperative scores, and blank columns represent postoperative scores. *p ⬍ 0.05 (paired Wilcoxon test).

Discussion The possibility that a decrease in long-term levodopa treatment in PD patients reduces the severity of LIDs has already been suggested. Previous evaluations of drug washout (“drug holidays”) showed a transient5 to significant reduction6 in the severity of LIDs after antiparkinsonian treatment had been interrupted for 2 to 10 days. A possible explanation for the lack of efficacy of drug holidays in reducing the severity of LIDs is the short duration of the drug withdrawal as a result of the potentially dangerous exacerbation of motor disability.7 Replacing repeated administration of the drug by continuous infusion of levodopa8,9 or agonists10 yielded inhomogeneous results, because LIDs were either decreased,9 increased,10 or unchanged.8 Whatever the clinical benefits of continuous intravenous administration, such treatment has the advantage of avoiding re-

1. The drug washout was not complete in most patients despite the marked reduction in the daily doses of levodopa and dopaminergic agonists. Nevertheless, the reduction in the severity of LIDs was similar whether the drug washout was complete (2 patients) or incomplete (10 patients). 2. Although the period of time between the two levodopa tests was not correlated with the decreased severity of LIDs, we cannot rule out the possibility that the disappearance of LIDs may require a complete drug washout over a longer period. 3. STN stimulation was stopped 2 hours before the second levodopa test, as patients did not tolerate the marked aggravation of motor disability for a longer period of time. This might have been insufficient to eliminate a residual effect of the stimulation procedure on the pattern of LIDs. A residual effect of the long-term continuous stimulation or an effect resulting from a permanent microlesion in the STN on the pattern of LIDs seems unlikely: the preoperative and postoperative “off” (baseline) and “on” (treated) motor scores after the interruption of STN stimulation did not differ; an increase rather than a decrease in the intensity of LIDs would have been expected, as it has been claimed that lesions11 or high-

Brief Communication: Bejjani et al: Reversibility of LID

657

frequency stimulation12 of the STN induces or aggravates LIDs, especially those of the ballic type. The fact that LIDs were still observed in patients no longer receiving chronic treatment with levodopa strongly suggests that they are unavoidable in patients with severe degeneration of the nigrostriatal pathway irreversibly worsening with time. The loss of dopaminergic neurons that is likely to have occurred in the interval between the two levodopa challenges, ranging from moderate (25 months) to almost no loss (1 month), might be expected to have aggravated the severity of LIDs in our patients. In fact, the contrary was found, strongly suggesting that the observed reduction in dyskinesia severity was related to changes in the level and mode of levodopa administration. Three different aspects of levodopa administration contribute to LIDs: LIDs are more frequent in patients treated with levodopa early in the course of the disease,13 they are more severe in patients treated with high doses of levodopa,2 and experimental14 and clinical15 studies have stressed the role of intermittent pulsatile administration of levodopa in inducing and aggravating LIDs. The use of STN stimulation in our patients allowed the latter factors, namely, high doses and pulsatile administration of levodopa and dopaminergic agonists, to be either significantly reduced or totally eliminated. The re-establishment and stabilization of near-normal functioning of the basal ganglia through chronic STN stimulation was such that patients were like those in the early stages of the disease. In conclusion, a decline in the capacity of levodopa to induce LIDs was achieved in this study by reducing or suppressing the long-term administration of levodopa. This partial reversibility of LIDs can be interpreted as being the result of both the stabilized functioning of the basal ganglia induced by chronic STN stimulation and the decrease or suppression of the sensitization phenomenon caused by the chronic intermittent administration of high doses of levodopa. It is not possible to reduce LIDs, and presumably all motor complications, by influencing the disease itself. In contrast, the manner of the administration of levodopa and related compounds can be adapted, with a view to reducing motor complications and thus improving the well-being of patients, through dose reduction and increased fractionation of daily levodopa intake. We thank Drs Philippe Cornu, Didier Dormont, Bernard Pidoux, and Leon Tremblay for their participation in this study and the staff of the Clinical Investigation Center for their support.

References 1. Agid Y. Levodopa: is toxicity a myth? Neurology 1998;50:858 – 863 2. Agid Y, Bonnet AM, Ruberg M, Javoy-Agid F. Pathophysiology of L-dopa–induced abnormal involuntary movements. In: Ca-

658

Annals of Neurology

Vol 47

No 5

May 2000

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

sey D, Chase T, Christensen A, Gerlach J, eds. Dyskinesia. Berlin: Springer, 1985:145–149 Fahn S, Elton R. Unified Parkinson’s Disease Rating Scale. In: Fahn S, Marsden CD, Calne D, Goldstein M, eds. Recent developments in Parkinson’s disease. 2nd ed. New Jersey: MacMillan Healthcare Information, 1987:153–163 Bejjani B, Damier P, Arnulf I, et al. Pallidal stimulation for Parkinson’s disease: two targets? Neurology 1997;49:1564 – 1569 Koller WC, Weiner WJ, Perlik S, et al. Complications of chronic levodopa therapy: long-term efficacy of drug holiday. Neurology 1981;31:473– 476 Feldman RG, Kaye JA, Lannon MC. Parkinson’s disease: follow-up after “drug holiday.” J Clin Pharmacol 1986;26:662– 667 Mayeux R, Stern Y, Mulvey K, Cote L. Reappraisal of temporary levodopa withdrawal (“drug holiday”) in Parkinson’s disease. N Engl J Med 1985;313:724 –728 Nutt JG, Carter JH, Lea ES, Woodward WR. Motor fluctuations during continuous levodopa infusions in patients with Parkinson’s disease. Mov Disord 1997;12:285–292 Schuh LA, Bennett JP, Jr. Suppression of LIDs in advanced Parkinson’s disease. I. Continuous intravenous levodopa shifts dose response for production of LIDs but not for relief of parkinsonism in patients with advanced Parkinson’s disease. Neurology 1993;43:1545–1550 Stocchi F, Patsalos PN, Berardelli A, et al. Clinical implications of sustained dopaminergic stimulation. Clin Neuropharmacol 1994;17(suppl 2):S7–S13 Lee MS, Marsden CD. Movement disorders following lesions of the thalamus or subthalamic region. Mov Disord 1994;9: 493–507 Limousin P, Pollak P, Hoffmann D, et al. Abnormal involuntary movements induced by subthalamic nucleus stimulation in parkinsonian patients. Mov Disord 1996;11:231–235 Caraceni T, Scigliano G, Musicco M. The occurrence of motor fluctuations in parkinsonian patients treated long-term with levodopa: role of early treatment and disease progression. Neurology 1991;41:380 –384 Juncos JL, Engber TM, Raisman R, et al. Continuous and intermittent levodopa differentially affects basal ganglia function. Ann Neurol 1989;25:473– 478 Mouradian MM, Heuser IJ, Baronti F, Chase TN. Modification of central dopaminergic mechanisms by continuous levodopa therapy for advanced Parkinson’s disease. Ann Neurol 1990;27:18 –23

serum.10 To test the hypothesis that FRDA is a disease of abnormal intracellular iron distribution, we determined serum transferrin receptor (sTfR) concentrations in 17 patients with FRDA.

Increased Serum Transferrin Receptor Concentrations in Friedreich Ataxia

Materials and Methods

Robert B. Wilson, MD, PhD,* David R. Lynch, MD, PhD,† Jennifer M. Farmer, MS,‡ David G. Brooks, MD, PhD,‡ and Kenneth H. Fischbeck, MD§

Mitochondrial iron accumulation is thought to underlie the pathophysiology of Friedreich ataxia and may occur at the expense of cytosolic iron. Decreases in cytosolic iron induce expression of the transferrin receptor, some of which is released into the serum. Here, we demonstrate that serum transferrin receptor concentrations are increased in patients with Friedreich ataxia, which supports the hypothesis that it is a disease of abnormal intracellular iron distribution. Wilson RB, Lynch DR, Farmer JM, Brooks DG, Fischbeck KH. Increased serum transferrin receptor concentrations in Friedreich ataxia. Ann Neurol 2000;47:659 – 661

Friedreich ataxia (FRDA) is an autosomal recessive degenerative disorder affecting approximately 1 in 50,000 individuals. Clinical features typically include ataxia, areflexia, dysarthria, extensor plantar responses, sensory loss, leg weakness, and cardiomyopathy.1 Most patients with FRDA have expansions of a GAA repeat in the first intron of both alleles of FRDA, a nuclear gene that encodes the mitochondrial protein frataxin.2–5 Reduced expression of the yeast frataxin homologue, YFH1p, is associated with mitochondrial iron accumulation at the expense of cytosolic iron6,7; the increase in mitochondrial iron results in oxidative damage and loss of respiratory capacity, and the decrease in cytosolic iron induces expression of high-affinity iron-uptake proteins.6 –9 The primary high-affinity iron-uptake protein in human cells is the transferrin receptor. Decreases in cytosolic iron induce expression of the transferrin receptor, some of which is released into the

From the Departments of *Pathology and Laboratory Medicine, †Neurology, and ‡Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA; and §Neurogenetics Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD. Received Jun 21, 1999, and in revised form Jan 4, 2000. Accepted for publication Jan 4, 2000. Address correspondence to Dr Wilson, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Room 509A, Stellar-Chance Laboratories, 422 Curie Boulevard, Philadelphia, PA 19104.

sTfR concentrations were determined using a commercially available kit (Quantikine IVD; R&D Systems, Minneapolis, MN) based on a microplate sandwich enzyme immunoassay. Screening for the C282Y and H63D mutations in the HFE gene, which are associated with hereditary hemochromatosis, was performed by restriction fragment length polymorphism analysis in the Molecular Pathology Laboratory of the Hospital of the University of Pennsylvania, Philadelphia. Reference values for sTfR concentrations are normally distributed; thus, a standard Student’s t test was used to compare means. The correlation between sTfR concentration and the size of the smaller GAA repeat expansion was determined using standard linear regression analysis. The experimental protocol received prior approval by the Committee on Studies Involving Human Beings of the University of Pennsylvania, and informed consent was obtained from each subject.

Results We determined sTfR concentrations in 17 patients with FRDA from 15 different families, representing a broad range of clinical severity. The ages of the patients ranged from 5 to 49 years, and the ages of disease onset ranged from 4 to 25 years. Thirteen of the patients were confirmed to carry GAA repeat expansions in both alleles of the FRDA gene. The expansions ranged from 280 to 1,400 GAA repeats; the smaller of the two expansion alleles in each patient ranged from 280 to 1,025 GAA repeats. All 13 patients confirmed to carry GAA repeat expansions in both alleles, including 1 sibling who was not tested directly for repeat expansions, met the diagnostic criteria of Harding1 for typical FRDA. Four of the 17 patients with FRDA in our study were demonstrated to carry one allele with a GAA repeat expansion in the FRDA gene and one allele with a point mutation. Three of these 4 patients carry a missense mutation (G130V, W154R, and R165C, respectively), and the fourth carries a point mutation in a splicing consensus sequence. All 4 of these patients have atypical disease by the criteria of Harding.1 The mean sTfR concentration in the 17 FRDA patients in our study was 25.2 nmol/L ⫾ 5.1 (SD), which is significantly higher than the normal mean of 18.4 nmol/L ⫾ 4.9 ( p ⬍ 0.0001).11 One of the patients in our study, who had the lowest sTfR concentration of those patients with two GAA repeat expansion alleles, also had the highest serum ferritin concentration (data not shown).12 He tested negative for the C282Y and H63D mutations in the HFE gene, making concomitant hereditary hemochromatosis unlikely. Before this study, he indicated a strong inter-

Copyright © 2000 by the American Neurological Association

659

est in trying androstenedione, an over-the-counter anabolic steroid, for his muscle weakness. He was discouraged from doing so, but we are unable to rule this out completely as an explanation for his highly atypical results. His known medications are digoxin, which has been shown to have a stimulatory effect on erythropoietic progenitors,13 and vitamin E (1,000 IU/day), a lipid-soluble antioxidant that may indirectly affect intracellular iron transport through its protective effect on the inner mitochondrial membrane. The GAA repeat expansions in the FRDA gene reduce frataxin expression by reducing mRNA transcription.14 Because the size of the GAA repeat expansion correlates inversely with expression, the FRDA allele with the smaller of the two expansions would be expected to express most of the residual frataxin in patients with two GAA repeat expansion alleles. Consistent with this expectation, the size of the smaller expansion correlates inversely with age of onset and directly with rate of disease progression.3 Using the values that we obtained from the 12 patients in our study who have two GAA repeat expansion alleles and no indication of confounding factors, we found a direct correlation between sTfR concentration and the size of the smaller GAA repeat expansion (r ⫽ 0.65; Fig). Fig. Correlation between increased serum transferrin receptor (sTf R) concentrations and the size of the smaller GAA repeat expansion in the first intron of the FRDA gene. The graph includes data from the 12 patients in our study who have two GAA repeat expansion alleles and no indication of confounding factors. The graph also includes data from 1 normal control (in whom the sTfR concentration equals the normal mean of 18.4 nmol/L and the smaller GAA repeat length is 22) to illustrate the continuity of the correlation into the normal range of repeat lengths. Linear regression analysis demonstrated a sample correlation coefficient of 0.65. The dotted lines indicate the 95% confidence interval curves for the line of fit.

660

Annals of Neurology

Vol 47

No 5

May 2000

These data suggest a causal relation between increased sTfR concentrations in FRDA patients and mutations in the FRDA gene. Virtanen and co-workers15 recently found that sTfR concentrations in 35 normal 11- and 12-year-old children were approximately 20% higher than in normal adults, which is equivalent to a calculated mean of 22.2 nmol/L ⫾ 3.5 (SD). Among the patients in our study were 5 children who were 11 to 13 years of age at the time their sera were tested. The mean sTfR concentration in these 5 children was 29.0 nmol/L ⫾ 4.6 (SD), which is significantly higher than the calculated normal mean of 22.2 nmol/L ⫾ 3.5 (SD) ( p ⬍ 0.03). The mean sTfR concentration in the 11 adults in our study was 23.9 nmol/L ⫾ 4.4 (SD), which is significantly higher than the normal mean of 18.4 nmol/L ⫾ 4.9 (SD) ( p ⬍ 0.002).11 Discussion Yeast lacking YFH1p as well as cultured fibroblasts from patients with FRDA accumulate mitochondrial iron.6,7,16,17 These and other data led to the hypothesis that the primary defect in FRDA is in iron metabolism and that iron accumulation in mitochondria results in oxidant damage and consequent loss of mitochondrial respiratory capacity. The iron accumulation in the mitochondria of yeast lacking YFH1p apparently occurs at the expense of cytosolic iron.6 High-affinity ironuptake proteins in yeast are normally induced only when cells are grown in low-iron media. In yeast lacking YFH1p, high-affinity iron-uptake proteins are highly expressed even in iron-replete media.6 Reintroduction of YFH1p results in the export of accumulated mitochondrial iron into the cytosol and reduced expression of high-affinity iron-uptake proteins, suggesting that YFH1p may be involved in mitochondrial iron efflux.18 A primary regulator of cellular iron metabolism in vertebrates is the mRNA-binding protein IRP-1. When cellular iron is scarce, IRP-1 blocks translation of ferritin mRNA and stabilizes transferrin receptor mRNA. When cellular iron is in excess, IRP-1 incorporates a 4Fe-4S iron-sulfur cluster, exhibits aconitase activity, and allows translation of ferritin mRNA and degradation of transferrin receptor mRNA.19 The transferrin receptor is the primary high-affinity iron-uptake protein, and small amounts are released into the systemic circulation. Thus, an increase in sTfR concentration is consistent with a decrease in cytosolic iron. We demonstrated previously that serum iron and ferritin concentrations in FRDA patients are within normal limits, with means below the fiftieth percentile for each reference range.12 We demonstrate here that sTfR concentrations in FRDA patients are increased. If the parallel between FRDA and the yeast model system holds, these results might be explained as follows. Af-

fected cells in FRDA patients may accumulate mitochondrial iron at the expense of cytosolic iron; if this is the case, ferritin synthesis would decrease and transferrin receptor synthesis would increase. In this view, FRDA would be considered a disease of abnormal intracellular iron distribution. Affected cells would release less ferritin to the systemic circulation, but a decrease in serum ferritin concentrations would be obscured by the contribution of unaffected cells. Affected cells would also release more transferrin receptor to the systemic circulation; as is the case in iron deficiency states, the increase in sTfR concentrations would be readily detectable because of the relatively small contribution of unaffected cells, which have normal cytosolic iron. Our results suggest that systemic iron chelation therapy for FRDA may be problematic. Of particular concern is whether the cytosolic iron concentration in affected cells is low, as is suggested by our data and the finding of Rotig and colleagues20 that cytosolic aconitase activity is reduced; if this is the case, affected cells may compete more avidly for limited tissue iron supplies than unaffected cells. It is possible that iron chelation therapy may benefit FRDA patients by producing a more favorable iron concentration gradient between the inside and outside of mitochondria in affected cells. Even if such therapy limits iron availability to unaffected cells, the net effect may still be beneficial. Our results suggest, however, that a great deal of caution should be exercised. Tissue iron supplies are of particular importance for developing erythrocytes, and anemia can exacerbate the effects of cardiomyopathy. Hopefully, the elucidation of the exact function of frataxin will lead to more directed therapeutic approaches. This work was supported by a research grant from the Muscular Dystrophy Association and Seek A Miracle.

7. Foury F, Cazzalini O. Deletion of the yeast homologue of the human gene associated with Friedreich’s ataxia elicits iron accumulation in mitochondria. FEBS Lett 1997;411:373–377 8. Koutnikova H, Campuzano V, Foury F, et al. Studies of human, mouse, and yeast homologues indicate a mitochondrial function for frataxin. Nat Genet 1997;16:345–351 9. Wilson RB, Roof DM. Respiratory deficiency due to loss of mitochondrial DNA in yeast lacking the frataxin homologue. Nat Genet 1997;16:352–357 10. Cook JD, Skikne BS, Baynes RD. Serum transferrin receptor. Annu Rev Med 1993;44:63–74 11. Punnonen K, Irjala K, Rajamaki A. Serum transferrin receptor and its ratio to serum ferritin in the diagnosis of iron deficiency. Blood 1997;89:1052–1057 12. Wilson RB, Lynch DR, Fischbeck KH. Normal serum iron and ferritin concentrations in patients with Friedreich’s ataxia. Ann Neurol 1998;44:132–134 13. Gallicchio VS, Chen MG, Murphy MJ. Modulation of murine in vitro erythroid and granulopoietic colony formation by ouabain, digoxin and theophylline. Exp Hematol 1982;10:682– 688 14. Ohshima K, Montermini L, Wells RD, et al. Inhibitory effects of expanded GAA-TTC triplet repeats from intron I of the Friedreich ataxia gene on transcription and replication in vivo. J Biol Chem 1998;273:14588 –14595 15. Virtanen MA, Viinikka LU, Virtanen MKG, et al. Higher concentrations of serum transferrin receptor in children than in adults. Am J Clin Nutr 1999;69:256 –260 16. Wong A, Yang J, Cavadini P, et al. The Friedreich’s ataxia mutation confers cellular sensitivity to oxidant stress which is rescued by chelators of iron and calcium and inhibitors of apoptosis. Hum Mol Genet 1999;8:425– 430 17. Delatycki MB, Camakaris J, Brooks H, et al. Direct evidence that mitochondrial iron accumulation occurs in Friedreich ataxia. Ann Neurol 1999;45:673– 675 18. Radisky DC, Babcock MC, Kaplan JK. The yeast frataxin homologue mediates mitochondrial iron efflux: evidence for a mitochondrial iron cycle. J Biol Chem 1999;274:4497– 4499 19. Paraskeva E, Hentze MW. Iron-sulphur clusters as genetic regulatory switches: the bifunctional iron regulatory protein-1. FEBS Lett 1996;389:40 – 43 20. Rotig A, de Lonlay P, Chretien D, et al. Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich’s ataxia. Nat Genet 1997;17:215–217

We thank A. Behrman for critically reading the manuscript.

References 1. Harding AE. Friedreich’s ataxia: a clinical and genetic study of 90 families with an analysis of early diagnostic criteria and intrafamilial clustering of clinical features. Brain 1981;104:589 – 620 2. Campuzano V, Montermini L, Molto MD, et al. Friedreich’s ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion. Science 1996;271:1423–1427 3. Durr A, Cossee M, Agid Y, et al. Clinical and genetic abnormalities in patients with Friedreich’s ataxia. N Engl J Med 1996;335:1169 –1175 4. Campuzano V, Montermini L, Lutz Y, et al. Frataxin is reduced in Friedreich ataxia patients and is associated with mitochondrial membranes. Hum Mol Genet 1997;6:1771–1780 5. Priller J, Scherzer CR, Faber PW, et al. Frataxin gene of Friedreich’s ataxia is targeted to mitochondria. Ann Neurol 1997;42:265–269 6. Babcock M, de Silva D, Oaks R, et al. Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin. Science 1997;276:1709 –1712

Brief Communication: Wilson et al: sTfR Concentrations in FRDA

661

Motor Benefit from Levodopa in Spastic Quadriplegic Cerebral Palsy Janice E. Brunstrom, MD,*†‡§储 Amy J. Bastian, PhD, PT,¶#** Michael Wong, MD, PhD,*†储 and Jonathan W. Mink, MD, PhD*†¶#储

We report on a 16-year-old girl with spastic quadriplegic cerebral palsy associated with premature birth and typical periventricular leukomalacia, who had a dramatic improvement in motor function after treatment with carbidopa/levodopa. Kinematic and electromyographic analyses of reaching movements demonstrate that levodopa decreased muscle co-contraction, decreased unwanted movements, and improved her ability to maintain a steady arm posture. These findings suggest that levodopa be considered as an adjunct therapy for the treatment of spastic quadriplegic cerebral palsy. Brunstrom JE, Bastian AJ, Wong M, Mink JW. Motor benefit from levodopa in spastic quadriplegic cerebral palsy. Ann Neurol 2000;47:662– 665

Cerebral palsy (CP) has multiple etiologies and is a significant cause of motor disability in children and adults. Although the clinical manifestations of CP are variable, virtually every classification scheme distinguishes “spastic” CP from “extrapyramidal” (nonspastic) CP. Crothers and Paine1 presented strong evidence that this is a valid and useful distinction, despite the fact that spastic CP often involves some extrapyramidal damage and extrapyramidal CP often involves some pyramidal tract dysfunction. Extrapyramidal motor signs such as dystonia are common in spastic quadriplegic CP and provide a rational basis for the trial of medications that have been shown to be effective in basal ganglia disorders. We report here on an adolescent with typical spastic quadriplegic CP, who had significant symptomatic benefit from treatment with levodopa.

Patient and Methods Clinical History The patient is a 16-year-old girl with spastic quadriplegic CP, who was born at 28 weeks of gestation and weighing 950 g. She had significant motor delay and never walked independently but had normal cognitive development. She is currently an eleventh grade honor student, who writes and operates a keyboard with her left hand. On examination, she has normal mental status, marked asymmetrical spasticity, weakness in all extremities (left arm least affected), truncal dystonia at rest that increases with action, and pronounced action dystonia of the right upper extremity. Magnetic resonance imaging of the brain revealed typical periventricular leukomalacia (PVL) and parietal lobe atrophy with no obvious basal ganglia abnormalities. Treatment was initiated with carbidopa/levodopa (100 mg of levodopa per day), and after 2 weeks, she reported a decrease in unwanted movements (body arching) and an improved ability to reach with either arm. When the dose was increased to 200 mg of levodopa per day, she was able to pull up onto her hands and knees and to reach with both upper extremities. Higher doses of levodopa were associated with decreased benefit. We used kinematics and electromyography (EMG) to evaluate the effects of levodopa on her ability to reach a target.

Kinematic and EMG Methods The subject was studied as she reached with her right arm to a ball target both before and 45 minutes after taking her morning dose of carbidopa/levodopa (18.75 mg/75.0 mg). She sat in her wheelchair with her back supported and her arm positioned with the shoulder neutral (vertical upper arm), her elbow flexed to 80 to 100 degrees, and her forearm in neutral pronation-supination. Infrared emitting diodes were placed over the first metacarpophalangeal joint of the index finger, ulnar styloid (wrist joint), lateral epicondyle (elbow joint), and tip of acromion process (shoulder joint). Bipolar surface EMG electrodes were placed over the anterior deltoid, posterior deltoid, biceps, brachioradialis, triceps, and wrist flexors. A 40-mm ball target was suspended from a flexible wire 580 mm in front of her shoulder. She was instructed to “move as fast as possible to touch and hold on any site on the target” on a verbal “go” signal. For each 6-second trial, the three-dimensional positions of all infrared emitting diodes were sampled at 100 Hz using an OPTOTRAK motion measurement system (Northern Digital, Ontario, CA). EMG signals were amplified to obtain a strong signal and were collected at 1 kHz.

Analysis From the Departments of *Neurology, †Pediatrics, ‡Cell Biology, and ¶Anatomy and Neurobiology, #Movement Disorders Center, and **Program in Physical Therapy, Washington University School of Medicine, and §Pediatric Neurology Cerebral Palsy Center and 储Department of Neurology, St Louis Children’s Hospital, St Louis, MO. Received Sep 30, 1999, and in revised form Jan 5, 2000. Accepted for publication Jan 7, 2000. Address correspondence to Dr Brunstrom, Pediatric Neurology Cerebral Palsy Center, St Louis Children’s Hospital, One Children’s Place, Room 12E25, St Louis, MO 63110.

662

We analyzed three representative reaches for each condition (before and after levodopa). The wrist marker position was numerically differentiated to obtain wrist linear velocity. EMG was rectified and low-pass filtered at 70 Hz. The “start” of the movement was the time and position at which the wrist linear velocity exceeded 5% of its peak. The “end” of the first portion of the movement was the time and position at which the wrist linear velocity reached its first steady minimum. The “movement” phase of the reach was defined as the time from the start to the end of the reach.

Copyright © 2000 by the American Neurological Association

Fig 1. Sagittal view of wrist paths for three trials before (A) and 45 minutes after (B) levodopa administration. Bold lines indicate the movement phase of the reach, and thin lines indicate the hold phase of the reach. All trials are drawn from the start of the reach to the end of the trial (⬃5.75 seconds). Single trials of rectified EMG before (C) and 45 minutes after (D) levodopa administration. EMG is shown for the first reaching trial for each condition. Trials are aligned on the start of the reach (time 0), and vertical lines indicate the end of the movement phase. Trials were matched for peak velocity. AD ⫽ anterior deltoid; PD ⫽ posterior deltoid; BI ⫽ biceps; BR ⫽ brachioradialis; TRI ⫽ triceps; WF ⫽ wrist flexors.

The “hold” phase of the reach was defined as the time from the end of the reach to the end of the trial. Measurements of interest for the movement phase included (1) wrist path ratio (curvature), (2) peak wrist velocity, (3) movement time, (4) end point error (overshoot or undershoot), and (5) integrated EMG for all muscles. The wrist path ratio was the ratio of the length that the wrist actually traveled to a straight line between the start and end positions. A path ratio of 1 represents a straight (normal) path; path ratios greater than 1 represent paths with increasing curvature. End point error was the distance between the

tip of the index finger and the target at the end of the first phase of the movement. EMG was integrated over the movement time to assess the level of muscle activity. Because movement times were different from trial to trial, all integrated EMG values were normalized to time. Measurements of interest for the hold phase included (1) distance traveled by the wrist and (2) integrated EMG for all muscles. The distance traveled by the wrist was used to represent how well the subject held a steady position. Integrated EMG was again normalized to time to adjust for the different hold times from trial to trial.

Brief Communication: Brunstrom et al: Levodopa in Spastic CP

663

Table. Kinematic Measures Before Levodopa

After Levodopa

Movement phase Wrist path ratio (curvature) 1.65 ⫾ 0.13 1.54 ⫾ 0.13 Peak wrist velocity (mm/sec) 629 ⫾ 114 639 ⫾ 111 Movement time (sec) 1.7 ⫾ 0.3 2.4 ⫾ 0.3 End point error (mm) ⫺115 ⫾ 9 ⫺142 ⫾ 40 Hold phase Wrist path length (mm) 1,017 ⫾ 189 319 ⫾ 46a p ⬍ 0.05.

a

Paired t tests were used to compare all measures before and after levodopa.

Results After taking levodopa, the subject showed marked improvement in her ability to maintain a static arm position during the hold phase. We also found significant decreases in EMG activity (co-contraction) during both movement and hold phases; however, kinematic measures during the movement phase remained unchanged. Figure 1A and B shows the wrist path for all trials before and after levodopa. Both the movement phase (bold line) and hold phase (thin line) are shown. No difference was found in the wrist path ratio (curvature) during the movement phase (Table); however, there was a striking effect of levodopa on the distance traveled by the wrist during the hold phase. Before levodopa, the subject produced large-amplitude slow movements (⬍1 Hz) when she held her arm outstretched and attempted to touch the target (see Fig 1A). This was primarily a result of flexion-extension movements at the elbow and shoulder joints (accompanied by pro-

nation and shoulder internal rotation). After levodopa, she was able to maintain her arm in a steady outstretched position as she attempted to touch the target (see Fig 1B). Other kinematic measures during the movement phase were not significantly different after levodopa (see Table). Peak wrist velocities were nearly identical, although movement times were longer after taking levodopa. The subject also tended to undershoot the target both before and after taking levodopa. Figure 1C and D shows EMG from six muscles for single reaching trials before and after levodopa. After levodopa, EMG activity decreased in all muscles during the movement and hold phases. We also noted that EMG activity became more tonic during the hold phase. Figure 2 shows the integrated EMG level (normalized for time; see Kinematic and EMG Methods section) before and after levodopa for both the movement and hold phases. All muscles showed substantial decreases in activity during both phases, although not all such decreases were statistically significant. Specifically, a significant decrease in activity was seen in three of six muscles during the movement phase and in five of six muscles during the hold phase. Discussion Our patient showed significant improvement in her functional abilities as a result of levodopa therapy even though her clinical presentation and magnetic resonance imaging findings of PVL are consistent with thediagnosis of spastic quadriplegia. Dystonic posturing and flexion deformities of the wrist are common in patients with spastic quadriplegia,1,2 and their presence here does not suggest the diagnosis of dopa-responsive dystonia.3,4

Fig 2. Average integrated EMG before (open bars) and 45 minutes after (solid bars) levodopa for the movement phase (A) and hold phase (B) of the reach. All integrated values were normalized for time. *p ⬍ 0.05, which is significant.

664

Annals of Neurology

Vol 47

No 5

May 2000

It is unlikely that pyramidal tract injury alone is responsible for the motor disorder of spastic CP.1,5,6 PVL, a necrosis of the developing white matter involving the corticospinal tract (pyramidal) pathways, is uniformly present in survivors of prematurity who have spastic CP and is regarded as the major cause of motor dysfunction in these patients.7,8 PVL is also present in a significant percentage of survivors of prematurity who do not have CP.9 Neurons of the developing basal ganglia are also vulnerable to hypoxic ischemic injury. Moreover, the germinal matrix hemorrhage that accompanies many cases of PVL occurs in the proliferative region overlying the head of the developing caudate nucleus.7 PVL may also disrupt the cortical inputs to the basal ganglia or the thalamocortical outputs from the basal ganglia, thereby causing dystonia without direct injury to the basal ganglia themselves.10 Clinically, many patients with spastic CP experience stiffness caused by unwanted co-contraction of agonist and antagonist muscles, and these patients have an impaired ability to isolate specific muscle groups and to focus muscle activity. They also have an inability to suppress primitive and unwanted movements.11 Similar features are present in basal ganglia disorders, especially dystonia.10,12 Recent evidence suggests that the pathophysiology of dystonia includes abnormalities in nigrostriatal dopamine neurotransmission.3,13–15 Our analysis shows that levodopa improved our subject’s ability to reach by decreasing muscle co-contraction, decreasing unwanted movements, and allowing her to better maintain a stable posture of her outstretched arm. We propose that these improvements result from the action of levodopa on basal ganglia circuits to reduce her dystonia rather than a reduction of her spasticity. We suggest that levodopa should be considered as an adjunctive therapy in patients with spastic quadriplegic CP who have prominent function-limiting dystonia.

5.

6. 7. 8.

9.

10.

11. 12. 13.

14.

15.

nia presenting as spastic cerebral palsy. Pediatrics 1988;82:137– 138 (Letter) Kuypers HGJM. Anatomy of the descending pathways. In: Brooks VB, ed. Handbook of physiology. Bethesda: American Physiological Society, 1981:597– 648 Tower SS. Pyramidal lesions in the monkey. Brain 1940;63: 36 –90 Volpe JJ. Neurology of the newborn. 3rd ed. Philadelphia: WB Saunders, 1995 Krageloh-Mann I, Petersen D, Hagberg G, et al. Bilateral spastic cerebral palsy—MRI pathology and origin: analysis from a representative series of 56 cases. Dev Med Child Neurol 1995; 37:379 –397 Olsen P, Paakko E, Vainionpaa L, et al. Magnetic resonance imaging of periventricular leukomalacia and its clinical correlation in children. Ann Neurol 1997;41:754 –761 Mink JW. The basal ganglia: focused selection and inhibition of competing motor programs. Prog Neurobiol 1996;50:381– 425 Bobath K. A neurophysiological basis for the treatment of cerebral palsy. Clin Dev Med 1980;75:28 – 44 Berardelli A, Rothwell JC, Hallett M, et al. The pathophysiology of primary dystonia. Brain 1998;121:1195–1212 Perlmutter JS, Stambuk MK, Markham J, et al. Decreased [18F]spiperone binding in putamen in idiopathic focal dystonia. J Neurosci 1997;17:843– 850 Perlmutter JS, Tempel LW, Black KJ, et al. MPTP induces dystonia and parkinsonism. Clues to the pathophysiology of dystonia. Neurology 1997;49:1432–1438 Crossman AR, Brotchie JM. Pathophysiology of dystonia. Adv Neurol 1998;78:19 –25

This work was supported by the National Institutes of Health, (grant NS01856 to J.E.B., grant NS01199 to A.J.B., and grant NS01808 to J.W.M.), the McDonnell Center for Higher Brain Function, and the Greater St Louis Chapter of the American Parkinson Disease Association. We thank Dr Shirley Sahrmann for helpful discussions regarding this work.

References 1. Crothers B, Paine RS. The natural history of cerebral palsy. Cambridge: Harvard University Press, 1959 2. Nelson KB, Swaiman KF, Russman BS. Cerebral palsy. In: Swaiman KF, ed. Pediatric neurology principles and practice, vol 1. St Louis: Mosby, 1994:471– 488 3. Nygaard TG, Waran SP, Levine RA, et al. Dopa-responsive dystonia simulating cerebral palsy. Pediatr Neurol 1994;11: 236 –240 4. Fink JK, Filling-Katz MR, Barton NW, et al. Treatable dysto-

Brief Communication: Brunstrom et al: Levodopa in Spastic CP

665

X-Linked Vacuolar Myopathies: Two Separate Loci and Refined Genetic Mapping Mari Auranen, MD,* Marcello Villanova, MD, PhD,† Francesco Muntoni, MD,‡ Michel Fardeau, MD,§ Stephen W. Scherer, PhD,㛳 Hannu Kalino, MD,¶ and Berge A. Minassian, MD㛳 **

X-linked vacuolar myopathies can be divided into two forms: one that is associated with cardiomyopathy and mental retardation (XVCM-MR) and a second form, termed X-linked myopathy with excessive autophagy (XMEA), that spares cardiac muscle and has no central nervous system involvement. In this article, we demonstrate linkage between XMEA and markers on chromosome Xq28 and assign the XMEA gene locus to the most telomeric 10.5 cM of chromosome X. We also show that XVCM-MR is not allelic to XMEA. Auranen M, Villanova M, Muntoni F, Fardeau M, Scherer SW, Kalimo H, Minassian BA. X-linked vacuolar myopathies: two separate loci and refined genetic mapping. Ann Neurol 2000;47:666 – 669

An X-linked myopathy characterized clinically by slowly progressive muscle weakness, mainly in the proximal muscles of the lower limbs, and histopathologically by a peculiar vacuolation of myofibers, was described in 1988 by Kalimo and colleagues.1 There was no central nervous system or cardiac involvement. The term X-linked myopathy with excessive autophagy (XMEA) was applied, because the vacuoles contained lysosomal enzymes and cellular debris and appeared to exocytose the debris between multiplied layers of basement membrane (BM). In 1995, Villanova and associates2 reported a second

From the *National Public Health Institute, Department of Human Molecular Genetics, Helsinki, Finland; †Laboratorio di Patologia Neuromuscolare, Istituto Ortopedico “Rizzoli,” Bologna, Italy; ‡Department of Paediatrics, Imperial College School of Medicine, Hammersmith Hospital, London UK; §Institute of Myology, INSERM U 523, Hoˆpital de la Salpetrie`re, Paris, France; **Division of Neurology, Departments of Paediatrics and 㛳Genetics, Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada; and ¶Department of Pathology, Turku University Hospital, Turku, Finland. Received Dec 2, 1999, and in revised form Jan 20, 2000. Accepted for publication Jan 21, 2000. Address correspondence to Dr Minassian, Division of Neurology, Department of Paediatrics and Genetics, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1XG, Canada.

666

family with this disease and further characterized the pathological features. They demonstrated that most vacuolar membranes were immunopositive for dystrophin, ␤-spectrin, and laminin, suggesting that they may have arisen by invagination of the sarcolemma and BM.2 They also found deposition of complement membrane attack complex (MAC) on the myofiber surface and elevated serum levels of complement components C5 and C9.3,4 These findings suggested complement-mediated damage as a pathogenetic component of this genetic myopathy. Using linkage analysis, Saviranta and co-workers5 excluded linkage between most of the X chromosome and XMEA. They obtained a positive but not significant LOD score (z ⫽ 0.9) at the most distal band, Xq28. Several families with an X-linked vacuolar myopathy with additional features of cardiomyopathy (marked biventricular concentric hypertrophy) and mild to moderate mental retardation (XVCM-MR) have also been described. Muscle biopsy specimens show subsarcolemmal and cytoplasmic vacuoles with immunocytochemical features similar to those of XMEA.6 –10 In this study, we confirm linkage of XMEA to Xq28 and narrow the critical region to less than 10.5 cM. We also show that XMEA and XVCM-MR are not allelic disorders. Patients and Methods Patients Six XMEA families with 13 affected males were studied. Detailed clinical and pathological descriptions of 2 of the families have been reported previously.1,2 Proximal lower limb weakness started in childhood and was slowly progressive. Most adults over the age of 30 years required assistive devices for ambulation. Upper limb and distal lower limb muscles were minimally involved in advanced cases. No evidence of central or peripheral nervous system involvement could be detected by clinical, neurophysiological, or neuroradiological testing. Electrocardiograms (ECGs) and echocardiograms were normal. Creatine kinase (CK) levels were elevated from early childhood and ranged from 1.5 to 15 times the upper limit of normal. The youngest male in this study is now 16 years of age, and affected or unaffected status of all males could therefore be determined unambiguously. One unpublished family with XVCM-MR was available for study. The first child was first investigated at the age of 10 years because of developmental delay and easy fatigability, symptoms that had been noticed since the first years of life. Currently, at age 19, he has microcephaly, mild mental retardation, and proximal and axial weakness. Brain magnetic resonance imaging, nerve conduction velocities, and evoked potentials are normal. ECG shows evidence of right ventricular hypertrophy, and an echocardiogram showed left ventricular dilatation, reduced ejection fraction, and diffuse hypokinesia, especially of the diaphragmatic and septal walls. The patient’s 15-year-old brother is similarly affected, and both have serum CK levels elevated 10 times above the upper limit of normal.

Copyright © 2000 by the American Neurological Association

Pathology In muscle specimens from at least 1 member of each XMEA family and the 2 XVCM-MR males, light microscopy showed increased variation of fiber size, scattered atrophic fibers, and fiber splitting. Many fibers contained vacuoles that stained positive with antibodies against dystrophin, ␤-spectrin, and laminin. On electron microscopy, vacuoles were found to be membrane-bound, and many were lined on the inner aspect by BM. Several subsarcolemmal vacuoles appeared to communicate with the extracellular space. Additional features present in the XMEA but not the XVCM-MR specimens were the acid phosphatase reactivity of the vacuoles, the strong MAC deposition on the sarcolemma of vacuolated fibers, and the accumulation of debris between multiplied layers of BM.

Genetic Analysis Marker sequences were obtained from the Genome Database (http://www.gdb.org). After polymerase chain reaction, fluorescence-labeled fragments were separated on an automated laser fluorescence DNA sequencer ABI 377XL (Perkin Elmer, Foster City, CA) with Genescan v.2.1 fragment analysis software. The alleles were identified by the Genotyper program (Perkin Elmer). Marker order and distances were based on the genetic maps of Genethon (http://www .genethon.fr/) and the Genetic Location Database (http:// cedar.genetics.soton.ac.uk/public_html/index.html). Two-

point linkage analyses were performed using the MLINK program of the LINKAGE package.11

Results Ten microsatellite markers (Fig 1) from Xq26-Xq28 were genotyped in all XMEA families. The highest maximum two-point LOD scores were obtained for Xq28 markers DXS15 and DXS8061 at 4.13 and 3.67 (␪ ⫽ 0.0), respectively. To detect informative recombinations, haplotypes were carefully analyzed. Part A of the Figure shows the most informative recombination. In family XMEA-2, affected males 2.7 and 2.8 share the same Xq27-28 haplotype telomeric to DXS8013. Their niece, 2.2, inherited a recombinant X chromosome from her deceased affected father that contained only part of the haplotype common to the affected uncles. This recombination narrows the XMEA locus to the last 10.5 cM12 of the chromosome telomeric to DXS1193. In an attempt to establish a gene locus in our XVCM-MR family, we determined the haplotype of the entire chromosome with 30 regularly spaced markers in addition to the markers used in the XMEA families. The 2 affected XVCM-MR brothers share the same haplotype for the whole of chromosome X cen-

Fig. (A and B) Xq26-28 haplotypes in a family with X-linked myopathy with excessive autophagy (XMEA) and a family with X-linked vacuolar myopathy with cardiomyopathy and mental retardation (XVCMMR). Scores for each marker in each individual are aligned with the name of the marker. Boxed haplotype indicates the informative recombination in XMEA-2, which narrows the XMEA locus telomeric to DXS1193 (a 10.5-cM region). Open rectangles indicate that the XVCM-MR gene in this family has to be centromeric to DXS8091, thereby excluding the remaining 11.8-cM region. (C) The 1.3-cM region between DXS8091 and DXS1193 has been completely sequenced and contains 805,000 bp and the large FMR2 gene (http:// www.hgsc.bcm.tmc.edu/seq_data_old/cgi-bin/ bcm-web-regions.cgi?region⫽Xq28). The critical regions for XVCM-MR and XMEA are shown by dashed lines. No gene overlaps both regions. Cen ⫽ centromere; tel ⫽ telomere.

Brief Communication: Auranen et al: X-Linked Vacuolar Myopathies

667

tromeric to marker DXS8091. Their haplotypes differ from marker DXS8091 to the telomere, thereby excluding the last 11.8 cM of the chromosome as a possible locus for their disease (see Fig, B). Discussion Toward the goal of identifying the genetic cause of XMEA, we have now proved linkage between this disease and markers in the most distal 10.5-cM segment of chromosome X. Two pathogenetic hypotheses have been proposed to explain the formation of vacuoles in XMEA. Based on the lysosomal acid phosphatase content of the vacuoles and the apparent extrusion of their contents at the sarcolemma, Kalimo and associates1 suggested that the vacuoles are phagosomes clearing debris from a sublethal breakdown of the myofiber. Villanova and colleagues,2 based on the MAC, dystrophin, ␤-spectrin, and laminin positivities of the vacuoles and the sarcolemma, proposed that the vacuoles form as a result of invaginations of segments of sarcolemma and BM after complement deposition. Because no ␭- or ␬-chain deposition was observed, they concluded that complement activation is likely not caused by immune complexes but rather by the alternative pathway.2 Spuler and Engel13 demonstrated sarcolemmal MAC deposits in different types of muscular dystrophies. This raises the possibility that the MAC deposits in XMEA may be nonspecific reactions to structural changes in the sarcolemma. MAC deposition, however, has not been reported in other vacuolar myopathies. Vacuole formation is a feature of several other disorders, including hypokalemic periodic paralysis,14 acid maltase deficiency,15 myophosphorylase deficiency,16 inclusion body myositis,17 lysosomal glycogen storage disease with normal acid maltase,18 oculopharyngeal dystrophy,19 familial distal myopathy,20 and chloroquine myopathy (in which autophagic vacuolation closely resembles XMEA).21 Clinical and pathological features and inheritance patterns clearly distinguish XMEA from these conditions. Several features separate XMEA from XVCM-MR. XMEA lacks cardiac or central nervous system involvement, and XVCM-MR (at least in the family studied here) does not have the MAC positivity or the accumulation of debris between multiplied layers of BM characteristic of XMEA. Nonetheless, we investigated the possibility that these two X-linked slowly progressive vacuolar myopathies may be caused by different mutations of the same gene or of different transcripts of the same gene. Haplotype analysis showed that the XVCM-MR gene in our family is centromeric to locus DXS8091, whereas the XMEA gene is telomeric to the next marker, DXS1193 (see Fig). The genes for these two disorders therefore localize to distinct nonoverlapping regions. There remained the remote possibility

668

Annals of Neurology

Vol 47

No 5

May 2000

that a large gene spanning the 805-kb distance between DXS8091 and DXS1193 and extending beyond each marker into each critical region might be involved in both diseases (see Fig, C). FMR2, the gene responsible for the FRAXE form of fragile-X mental retardation,22 occupies 520 kb of this region and encompasses DXS8091 in its first intron, but it does not extend as far as DXS1193 located 285 kb beyond its telomeric end. FMR2 is therefore outside the XMEA critical region and is not a candidate gene for this disease. Band Xq28 is one of the most gene-rich regions of the genome. The genes for three muscle diseases are located in the XMEA region: Emery-Dreifuss muscular dystrophy, myotubular myopathy, and X-linked dilated cardioskeletal myopathy (Barth syndrome). All three have markedly different clinical pictures, muscle pathology, or both, making them unlikely allelic diseases to XMEA. At present, we are undertaking a candidate gene approach to screen genes from the region for mutations in our XMEA patients. We have completed sequencing all exons of our first candidate gene, DNL1L, a gene expressed almost exclusively in muscle,23 without finding mutations. Meanwhile, efforts are ongoing to recruit more XMEA families to identify new recombinations and narrow the gene locus. Note Added in Proof A second group also proved linkage of XMEA to Xq28 (Villard L et al). Their article appears in the April 2000 issue of the European Journal of Genetics.

We thank Drs G. Israelian, G. Karpati, S. Carpenter, C. Sewry, and P. Montagna for referring patients, Drs B. Banwell and A. Boright for helpful discussion and critical review of the manuscript, and Roxanna Irani for technical support.

References 1. Kalimo H, Savontaus M-L, Lang H, et al. X-linked myopathy with excessive autophagy: a new hereditary muscle disease. Ann Neurol 1988;23:258 –265 2. Villanova M, Louboutin JP, Chateau D, et al. X-linked vacuolated myopathy: complement membrane attack complex on surface membrane of injured muscle fibers. Ann Neurol 1995;37: 637– 645 3. Louboutin JP, Villanova M, Ulrich G, et al. Elevated levels of complement components C5 and C9 and decreased antitrypsin activity in the serum of patients with X-linked myopathy. Muscle Nerve 1996;19:1144 –1147 4. Louboutin JP, Villanova M, Lucas-Heron B, Fardeau M. X-linked vacuolated myopathy: membrane attack complex deposition on muscle fiber membranes with calcium accumulation on sarcolemma. Ann Neurol 1997;41:117–120 5. Saviranta P, Lindlof M, Lehesjoki AE, et al. Linkage studies in a new X-linked myopathy, suggesting exclusion of DMD locus and tentative assignment to distal Xq. Am J Hum Genet 1988; 42:84 – 88

6. Bergia B, Sybers HD, Butler IJ. Familial lethal cardiomyopathy with mental retardation and scapuloperoneal muscular dystrophy. J Neurol Neurosurg Psychiatry 1986;49:1423–1426 7. Kashio N, Usuki F, Akamine T, et al. Cardiomyopathy, mental retardation and autophagic vacuolar myopathy: abnormal MRI findings in the head. J Neurol Sci 1991;105:1–5 8. Muntoni F, Catani G, Mateddu A, et al. Familial cardiomyopathy, mental retardation and myopathy associated with desmintype intermediate filaments. Neuromusc Disord 1994;4:233– 242 9. Murakami N, Goto Y, Itoh M, et al. Sarcolemmal indentation in cardiomyopathy with mental retardation and vacuolar myopathy. Neuromusc Disord 1995;5:149 –155 10. Garcia-Acuna JM, Fernandez-Vazquez F, Fernandez-Lopez JA, Forteza-J. Myocardiopathy, mental retardation and autophagic vacuolar myopathy. Med Clin (Barc) 1995;104:197–198 11. Lathrop GM, Lalouel J-M, Julier CA, Ott J. Strategies for multilocus linkage analysis in humans. Proc Natl Acad Sci USA 1984;81:3443–3446 12. Nagaraja R, Macmillan S, Kere J, et al. X chromosome map at 75-kb TS resolution, revealing extremes of recombination and GC content. Genome Res 1997;7:210 –222 13. Spuler S, Engel AG. Unexpected sarcolemmal complement membrane attack complex deposits on nonnecrotic muscle fibers in muscular dystrophies. Neurology 1998;50:41– 46 14. Engel AG. Evolution and content of vacuoles in primary hy-

15.

16.

17.

18.

19. 20.

21. 22. 23.

pokalemic periodic paralysis. Mayo Clin Proc 1970;45:774 – 814 Engel AG. Acid maltase deficiency in adults: studies in fourcases of a syndrome which may mimic muscular dystrophy or other myopathies. Brain 1970;93:599 – 616 De Bleecker JL, Engel AG, Winkelmann JC. Localization of dystrophin and ␤-spectrin in vacuolar myopathies. Am J Pathol 1993;143:1200 –1208 Carpenter S, Karpati G, Heller I, Eisen A. Inclusion body myositis: a distinct variety of idiopathic inflammatory myopathy. Neurology 1978;28:8 –17 Danon MJ, Oh SJ, DiMauro S, et al. Lysosomal glycogen storage disease with normal acid maltase. Neurology 1981;31: 51–57 Tome FMS, Fardeau M. Nuclear inclusion in oculopharyngeal dystrophy. Acta Neuropathol (Berl) 1980;49:85– 87 Markesbery WR, Griggs RC, Herr B. Distal myopathy: electron microscopic and histochemical studies. Neurology 1977;27: 727–735 Hughes JT, Esiri M, Oxbury JM, Whitty CW. Chloroquine myopathy. Q J Med 1971;40:85–93 Gecz J, Mulley JC. Characterisation and expression of a large, 13.7 kb FMR2 isoform. Eur J Hum Genet 1999;7:157–162 Parrish JE, Ciccodicola A, Wehnert M, et al. A muscle-specific DNase I-like gene in human Xq28. Hum Mol Genet 1995;4: 1557–1564

Brief Communication: Auranen et al: X-Linked Vacuolar Myopathies

669

Lihat lebih banyak...

Comentarios

Copyright © 2017 DATOSPDF Inc.