close

Вход

Забыли?

вход по аккаунту

?

Anovel D104G mutation in the adenine nucleotide translocator 1 gene in autosomal dominant progressive external ophthalmoplegia patients with mitochondrial DNA with multiple deletions.

код для вставкиСкачать
BRIEF COMMUNICATIONS
Role of Parkin Mutations in
111 Community-Based
Patients with Early-Onset
Parkinsonism
Martin Kann, BS,1,2 Helfried Jacobs, MD,1,3
Kathrin Mohrmann, BS,1,2 Kirsten Schumacher, BS,1,2
Katja Hedrich, MA,1,2 Jennifer Garrels, BS,1,2
Karin Wiegers, BS,1,2 Eberhard Schwinger, MD,2
Peter P. Pramstaller, MD,4 Xandra O. Breakefield, PhD,5
Laurie J. Ozelius, PhD,6 Peter Vieregge, MD,1,7
and Christine Klein, MD1,2
Early-onset parkinsonism is frequently reported in connection with mutations in the parkin gene. In this study,
we present the results of extensive genetic screening for
parkin mutations in 111 community-derived early-onset
parkinsonism patients (age of onset <50 years) from
Germany with an overall mutation rate of 9.0%. Gene
dosage alterations represented 67% of the mutations
found, underlining the importance of quantitative analyses of parkin. In summary, parkin mutations accounted
for a low but significant percentage of early-onset parkinsonism patients in a community-derived sample.
Ann Neurol 2002;51:621– 625
Mutations in the parkin gene have been found in many
cases of early-onset Parkinsonism (EOP), especially in
patients with onset at an early age and a positive family
history.1–3 However, not all mutation carriers present
with the additional clinical features that are commonly
reported in parkin-related EOP, such as diurnal fluctuations of symptoms, sleep benefit, foot dystonia, early
3–5
L-dopa-induced dyskinesia, and hyperreflexia.
Some
patients have presented with parkinsonism indistinguishable from idiopathic Parkinson’s disease (PD).4
From the 1Department of Neurology, Medical University of Lübeck, Lübeck, Germany; 2Department of Human Genetics, Medical
University of Lübeck, Lübeck, Germany; 3Department of Neurology, Hospital Rothenburg/Wümme, Rothenburg, Germany; 4Department of Neurology, General Regional Hospital Bolzano, Bolzano, Italy; 5Molecular Neurogenetics Unit, Neurology Department,
Massachusetts General Hospital and Department of Neurology,
Harvard Medical School, Boston, MA; 6Albert Einstein College of
Medicine, Molecular Genetics Department, Bronx, NY; and 7LippeLemgo Hospital, Lemgo, Germany.
Received Oct 17, 2001, and in revised form Dec 26. Accepted for
publication Jan 4, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10179
Address correspondence to Dr Klein, Klinik für Neurologie der
Medizinischen Universität zu Lübeck, Ratzeburger Allee 160
D-23538 Lübeck, Germany. E-mail: klein_ch@neuro.mu-luebeck.de
However, phenotypic overlap of some EOP cases early
in the course of the disease has been described with
early-onset torsion dystonia6 and dopa-responsive dystonia.7 Interestingly, some patients with clinically typical dopa-responsive dystonia have recently been identified as carriers of parkin mutations instead of the
expected mutations in the GTP cyclohydrolase I gene.7
The parkin gene contains 12 exons, in which a large
variety of mutations has been found. Gene dosage alterations account for the majority of these mutations.3,5 The parkin gene encodes an E3 ligase that has
an important role in the degradation of parkin itself
and several other target proteins.8 –12
Although knowledge of the molecular function of
parkin has been expanding, little light has so far been
shed on the extent of parkin mutations in communityderived EOP patients. Most studies have selectively
screened familial cases consistent with an autosomal recessive mode of inheritance.2,13,14 Only one investigation included 100 cases of exclusively sporadic EOP;
this was also one of only two studies to systematically
screen for gene dosage alterations.3,5
In this study, we analyzed 111 community-based
EOP patients from Germany who were selected only
on the basis of age of onset. All patients were screened
for mutations in the parkin gene by means of conventional and quantitative polymerase chain reaction and
were evaluated for potential genotype–phenotype associations.
Subjects and Methods
Patients were recruited from membership lists of the German
Parkinson’s Association, the largest self-support German organization, and from general practitioners and neurologists
in northwestern Germany.15 Details of patient ascertainment
have been previously reported.15 Having given informed consent, all patients underwent a detailed neurological examination by the same physician (H.J.) to verify an a priori diagnosis of PD. The examination included assessments based on
the Unified Parkinson’s Disease Rating Scale (UPDRS), the
Hoehn-Yahr stage, the Mini-Mental State Examination
score, and the Schwab-England Activities of Daily Living
score. PD was defined according to the United Kingdom
Parkinson’s Disease Brain Bank criteria.16 Possible secondary
parkinsonism was ruled out by a review of the medical
records and by history, physical examination, and laboratory
and neuroimaging tests. Patients whose age at the onset of
parkinsonian symptoms was greater than 50 years or who
had Mini-Mental State Examination scores of less than 24 or
evidence of either secondary parkinsonism or overt drugassociated psychosis were excluded from the study.15
Genetic screening was performed by single-strand conformation polymorphism analysis, sequencing, and a quantitative duplex polymerase chain reaction assay to test for gene
dosage alterations, as previously described.5 One hundred
fifty chromosomes of ethnically matched, normal individuals
were evaluated for each missense mutation. Forty patients
with dystonic features as presenting symptoms were screened
© 2002 Wiley-Liss, Inc.
621
for the GAG deletion in the early-onset torsion dystonia
gene17 and for mutations in the GCH1 gene by single-strand
conformation polymorphism analysis.18 Unfortunately, neither RNA nor parents or relatives were available for further
analysis. Because all patients are still alive, no postmortem
analysis was performed to test for the presence of neurofibrillary tangle pathology.
Results
One hundred eleven unrelated patients with a mean
age at onset of 37.0 ⫾ 6.7 years, 18% of whom had a
positive family history of PD, were included in the
study. In the overall sample, the mean UPDRS3 score
was 25.4 ⫾ 13.7, the mean UPDRS4 score was 3.5 ⫾
3.3, and the mean Hoehn-Yahr stage was 2.4 ⫾ 0.9.
Dystonic features were found in 57% of all patients,
and morning and permanent dyskinesia was present in
14 and 39% of patients, respectively. A mean sleep
benefit of 1.51 ⫾ 1.48, on a visional analogue scale of
0 to 5, was reported. According to mutational status,
patients were divided into three subgroups: (1) patients
without mutations, (2) patients with two or three mutations (compound heterozygous), and (3) patients
with one mutated allele (heterozygous). When these
subgroups were analyzed separately, a positive family
history was found in 18 (no mutation), 25 (compound
heterozygous), and 17% (heterozygous), and the mean
ages at onset were 37.4, 28.3, and 36.5 years, respectively. The respective mean Hoehn-Yahr stages were
2.3 ⫾ 0.9, 3.3 ⫾ 0.5, and 2.0 ⫾ 0.9; the respective
mean UPDRS3 scores were 25.7 ⫾ 13.9, 27.0 ⫾ 15.3,
and 18.7 ⫾ 9.5; and the respective mean UPDRS4
scores were 3.5 ⫾ 3.2, 7.0 ⫾ 4.7, and 2.3 ⫾ 2.7.
Dystonic features were found in 55, 100, and 67%,
and the mean sleep benefit measurements were 1.56 ⫾
1.50, 2.00 ⫾ 1.41, and 0.33 ⫾ 0.82, respectively. Detailed clinical features and score assessments are summarized in the Table. A tendency toward decreased age
at onset, increased prevalence of dystonia, and positive
family histories may be suspected when a growing
number of parkin mutations is observed.
Genetic analysis revealed a total of 15 mutations in
10 patients (Fig). In 4 patients (40%), mutations
were found in the compound heterozygous state, with
1 patient carrying a total of three different mutations.
Six patients (60%) had heterozygous parkin mutations. Gene dosage alterations in this study accounted
for 67% of all mutations (10 of 15). The GAG deletion in the DYT1 gene was excluded in patients
with dystonic symptoms, as were mutations in the
GTP cyclohydrolase I gene. Statistical evaluation was
kept at a descriptive level only because the power of
the appropriate tests was too low on account of large
differences in the sizes of the subgroups with respect
to mutational status.
622
Annals of Neurology
Vol 51
No 5
May 2002
Discussion
In this investigation, we extensively screened 111 EOP
patients for mutations in the parkin gene by conventional and quantitative polymerase chain reaction and
found an overall mutation rate of 9.0% (3.6% compound heterozygotes and 5.4% heterozygotes).
This study is unique for several reasons. First, patients were recruited in a community-based fashion
and not from tertiary referral centers that specialized
in diagnosing and curing movement disorders. Second, a single investigator (H.J.) assessed clinical characteristics and scores for all patients, minimizing
the probability of investigator-associated variability.
Third, a very detailed differentiation between parkinsonian and dystonic features was applied, providing
an extensive clinical description of the patient sample.
Finally, the genetic screening performed in this investigation included conventional and quantitative analyses of all exons of the parkin gene and was extended
to genes responsible for phenotypically similar disorders as appropriate, representing what we believe is
the most complete genetic analysis of EOP patients to
date.
This study provides evidence that parkin-related
EOP was a significant condition in a communityderived sample of patients less than 50 years of age at
the onset of parkinsonism. Other studies published to
date recruited their patients from specialty centers,2,3,5
which may have resulted in a bias toward an increased
percentage of patients with a positive family history19
and, consequently, an increased mutation rate for the
respective samples.3 The low prevalence of mutations
prevented statistical comparison between subgroups
with respect to mutational status and symptoms. However, with an increasing number of mutated alleles per
patient, some interesting preliminary observations can
be made, such as apparent decreased age at onset, increased occurrence of dystonic features, and positive
family histories. This is in agreement with previous
data showing a markedly reduced overall prevalence of
mutations in sporadic cases (18%) versus familial cases
(49%) with an age at onset of less than 45 years.3 The
remarkably low mutation rate cannot be explained by
the fact that we used an age cutoff of less than 50
years, rather than less than 45 years, as in the study by
Lücking and colleagues,3 as only 4 of our patients were
older than 45 years of age at onset. In contrast, one
would expect an even higher rate of mutations in our
patients in comparison with the 100 sporadic cases reported by Lücking and colleagues3 because we did not
exclude patients with a positive family history of PD.
Another reason for the low mutation rate might be
that our sample did not contain patients from genetic
isolates.5 Similarly, founder effects could be demonstrated for several recurrent mutations in the previously
described multiethnic EOP cohort.3,20
Table. Clinical Data of the Sample, Separated According to Mutation Statusa
Variable
No. of patients
General characteristics
Gender
Age at onset (yr)
Disease duration (yr)
Family history positive for PD
Cardinal symptoms
Bradykinesia
Rigidity
Resting tremor
Postural instability
Scores
UPDRS3 score
UPDRS4 score
Hoehn-Yahr stage
Schwab-England Activities of
Daily Living score
MMSE score
Initial symptoms
Parkinsonian symptoms (overall)
Hypomimia
Micrographia
Muscle stiffness
Painful joints
Trembling
Unstable walking
Autonomic dysfunction (overall)
Hyperhidrosis
Hypersalivation
Sleep disorder
Dystonic features (overall)
Foot dystonia
Writer’s cramp
L-Dopa–associated features and
typical signs of EOP
L-Dopa dose at examination
Duration of L-dopa medication (mo)
Wearing-off
Freezing
On-off fluctuations
Morning dyskinesia
Permanent dyskinesia
Sleep benefit (from 0 to 5 on
a visional analogue scale)
Patients without
Mutation
All Patients
111
Mean/%
100%
SD/n/n
101
Mean/%
Compound
Heterozygous Cases
91.0%
SD/n/n
4
Mean/%
Heterozygous Cases
3.6%
SD/n/n
6
Mean/%
5.4%
SD/n/n
6.4
4.9
18/101
25% F
75% M
28.3
16.3
25%
12.2
13.3
1/4
50% F
50% M
36.5
6.3
17%
2.4
3.7
1/6
59% F
41% M
37.0
7.6
18%
6.7
5.6
20/111
60% F
40% M
37.4
7.2
18%
100%
97%
49%
45%
111/111
108/111
54/111
50/111
100%
97%
48%
46%
101/101
98/101
48/101
46/101
100%
100%
50%
75%
4/4
4/4
2/4
3/4
100%
100%
67%
17%
6/6
6/6
4/6
1/6
25.4
3.5
2.4
84.1
13.7
3.3
0.9
9.8
25.7
3.5
2.3
84.0
13.9
3.2
0.9
9.6
27.0
7.0
3.3
77.5
15.3
4.7
0.5
12.6
18.7
2.3
2.0
88.3
9.5
2.7
0.9
9.8
28.4
1.5
28.4
1.5
27.5
1.7
29.0
0.6
100%
111/111
100%
101/101
100%
4/4
100%
6/6
31%
51%
64%
42%
61%
35%
59%
34/111
57/111
71/111
47/111
68/111
39/111
66/111
31%
55%
66%
46%
61%
34%
63%
31/101
56/101
67/101
46/101
62/101
34/101
63/101
50%
0%
50%
25%
75%
100%
25%
2/4
0/4
2/4
1/4
3/4
4/4
1/4
17%
17%
33%
0%
50%
17%
33%
1/6
1/6
2/6
0/6
3/6
1/6
2/6
50%
27%
36%
57%
16%
23%
56/111
30/111
40/111
63/111
18/111
26/111
53%
29%
87%
55%
13%
26%
54/101
29/101
38/101
55/101
13/101
26/101
25%
25%
0%
100%
100%
0%
1/4
1/4
0/4
4/4
4/4
0/4
17%
0%
33%
67%
17%
0%
1/6
0/6
2/6
4/6
1/6
0/6
224
36.7
306
84.0
136
87.5
365
38.6
216
28.0
379
50.6
41%
25%
43%
14%
39%
1.51
220
39.3
383
49.8
45/111
28/111
48/111
15/111
43/111
1.48
41%
24%
45%
13%
39%
1.56
41/101
24/101
45/101
13/101
39/101
1.50
75%
75%
50%
50%
50%
2.00
3/4
3/4
2/4
2/4
2/4
1.41
17%
17%
17%
0%
33%
0.33
1/6
1/6
1/6
0/6
2/6
0.82
a
The first columns represent the sample as a whole, whereas the following provide data for each subgroup separately. Mean values and standard
deviations are given where applicable.
SD ⫽ standard deviation; n/n ⫽ cases/group; PD ⫽ Parkinson’s disease; UPDRS ⫽ Unified Parkinson’s Disease Rating Scale; MMSE ⫽
Mini-Mental State Examination; EOP ⫽ early-onset parkinsonism.
Insufficient mutation screening methods resulting in
decreased sensitivity are also unlikely to have affected
our study because we validated our single-strand conformation polymorphism assay in double-blind fashion
with restriction fragment length polymorphism and
fluorescence resonance energy-transfer techniques, with
known polymorphisms as positive controls (data not
shown). An assay that we recently developed for use in
gene dosage studies5 also proved to be a reliable means
of screening for such alterations. In accordance with
previous findings,3,5 gene dosage alterations accounted
for the majority of the mutations in our sample. As for
the distribution of mutations in the gene, we confirmed the results of former studies, which indicated
that mutations and especially gene dosage alterations in
exons 10 to 12 are rare.3,5 The majority of mutations
detected in this study affected the ubiquitin domain
and its 3⬘-adjacent regions, which probably resulted in
a loss of function of the parkin protein.8,9 Several of
these mutations were also considered causative for EOP
in the homozygous and compound heterozygous states
in former studies.3,5 The role of heterozygous parkin
Kann et al: Parkin Mutations in Parkinsonism
623
mutations as potential PD susceptibility factors remains to be determined.
In conclusion, parkin mutations were found in a minority of EOP patients in a community-derived sample
for which patients were selected neither for special features of the disease nor for a positive family history.
However, because the parkin-positive patients represented a significant percentage of the EOP patients, additional studies involving similar patient cohorts and
even later-onset PD patients may be warranted to further evaluate the role of parkin in PD and to eventually
develop guidelines for genetic testing.
This work was supported by the Deutsche Forschungsgemeinschaft
(Kl-1134/2-1, K.H., P.V., and C.K.), the National Institutes of
Health (NINDS NS3872, X.O.B), the Parkinson’s Disease Foundation (C.K.), and the Deutsche Parkinsonvereinigung (P.V.
and H.J.).
We thank all the patients who participated in this study and Dr
Friedrich of the Institute for Medical Biometry and Statistics at the
Medical University of Lübeck for statistical assistance.
References
Fig. (A) Schematic representation of the parkin gene with its
12 exons. Gene dosage alterations are summarized above the
scheme, and missense mutations are given below the scheme.
Dashed lines indicate heterozygous deletions, straight lines indicate heterozygous duplications, and the double line indicates
triplication. See the table for an overview of the patients and
their mutations. (B) Gene dosage results for Patient L-324,
who carried a total of three mutations, including a heterozygous deletion of exons 3 and 4 and heterozygous duplications
of exons 7 and 9. (C) Gene dosage results for Patient L-389,
who carried a heterozygous triplication of exon 2 and a heterozygous deletion of exons 4 and 5. Because no RNA was
available for further analysis, it could not be tested whether
the triplication may also have been a homozygous duplication.
624
Annals of Neurology
Vol 51
No 5
May 2002
1. Kitada T, Asakawa S, Hattori N, et al. Mutations in the parkin
gene cause autosomal recessive juvenile parkinsonism. Nature
1998;392:605– 608.
2. Hattori N, Kitada T, Matsumine H, et al. Molecular genetic
analysis of a novel parkin gene in Japanese families with autosomal recessive juvenile parkinsonism: evidence for variable homozygous deletions in the parkin gene in affected individuals.
Ann Neurol 1998;44:935–941.
3. Lücking CB, Dürr A, Bonifati V, et al, for the French Parkinson’s Disease Genetics Study Group. Association between earlyonset Parkinson’s disease and mutations in the parkin gene.
N Engl J Med 2000;342:1560 –1567.
4. Klein C, Pramstaller PP, Kis B, et al. Parkin deletions in a
family with adult-onset, tremor-dominant parkinsonism: expanding the phenotype. Ann Neurol 2000;48:65–71.
5. Hedrich K, Kann M, Lanthaler AJ, et al. The importance of
gene dosage studies: mutational analysis of the parkin gene in
early-onset parkinsonism. Hum Mol Genet 2001;10:1649 –
1656.
6. Leung JC, Klein C, Friedman J, et al. Novel mutation in the
TOR1A (DYT1) gene in atypical early onset dystonia and polymorphisms in dystonia and early onset parkinsonism. Neurogenetics 2001;3:133–143.
7. Tassin J, Dürr A, Bonnet AM, et al. Levodopa-responsive
dystonia: GTP cyclohydrolase I or parkin mutations? Brain
2000;123:1112–1121.
8. Imai Y, Soda M, Takahashi R. Parkin suppresses unfolded protein stress-induced cell death through its E3 ubiquitin-protein
ligase activity. J Biol Chem 2000;275:35661–35664.
9. Shimura H, Hattori N, Kubo S, et al. Familial Parkinson disease gene product, parkin, is a ubiquitin-protein ligase. Nat
Genet 2000;25:302–305.
10. Zhang Y, Gao J, Chung KK, et al. Parkin functions as an E2dependent ubiquitin-protein ligase and promotes the degradation of the synaptic vesicle-associated protein, CDCrel-1. Proc
Natl Acad Sci U S A 2000;97:13354 –13359.
11. Shimura H, Schlossmacher MG, Hattori N, et al. Ubiquitination of a new form of alpha-synuclein by parkin from human
brain: implications for Parkinson’s disease. Science 2001;293:
263–269.
12. Imai Y, Soda M, Inoue H, et al. An unfolded putative transmembrane polypeptide, which can lead to endoplasmic reticulum stress, is a substrate of parkin. Cell 2001;105:891–
902.
13. Nisipeanu P, Inzelberg R, Blumen SC, et al. Autosomalrecessive juvenile parkinsonism in a Jewish Yemenite kindred:
mutation of parkin gene. Neurology 1999;53:1602–1604.
14. Abbas N, Lücking CB, Ricard S, et al, for the French Parkinson’s Disease Genetics Study Group and the European Consortium on Genetic Susceptibility in Parkinson’s Disease. A wide
variety of mutations in the parkin gene are responsible for autosomal recessive parkinsonism in Europe. Hum Mol Genet
1999;8:567–574.
15. Jacobs H, Heberlein I, Vieregge A, Vieregge P. Personality
traits in young patients with Parkinson’s disease. Acta Neurol
Scand 2001;103:82– 87.
16. Gibb WR, Lees AJ. The relevance of the Lewy body to the
pathogenesis of idiopathic Parkinson’s disease. J Neurol Neurosurg Psychiatry 1988;51:745–752.
17. Klein C, Friedman J, Bressman S, et al. Genetic testing for
early-onset torsion dystonia (DYT1): introduction of a simple
screening method, experiences from testing of a large patient
cohort, and ethical aspects. Genet Test 1999;3:323–328.
18. Bandmann O, Valente EM, Holmans P, et al. Dopa-responsive
dystonia: a clinical and molecular genetic study. Ann Neurol
1998;44:649 – 656.
19. Marder K, Tang MX, Mejia H, et al. Risk of Parkinson’s disease among first-degree relatives: a community-based study.
Neurology 1996;47:155–160.
20. Periquet M, Lücking C, Vaughan J, et al. Origin of the mutations in the parkin gene in Europe: exon rearrangements are independent recurrent events, whereas point mutations may result
from Founder effects. Am J Hum Genet 2001;68:617– 626.
Paraneoplastic Chorea
Associated with CRMP-5
Neuronal Antibody and
Lung Carcinoma
Steven Vernino, MD, PhD,1 Paul Tuite, MD,4
Charles H. Adler, MD, PhD,2 James F. Meschia, MD,3
Bradley F. Boeve, MD,1 Peter Boasberg, MD,7
Joseph E. Parisi, MD,6 and Vanda A. Lennon, MD, PhD1,5,6
Paraneoplastic chorea is described in 16 patients: 11 with
limited small-cell carcinoma, 2 with lung cancer revealed
by imaging, 1 with renal cell carcinoma, and 1 with lymphoma. All had CRMP-5-IgG; 6 also had ANNA-1 (antiHu), including 1 without evident cancer. Chorea was the
initial and most prominent symptom in 11 patients,
asymmetric or unilateral in 5 patients, and part of a multifocal syndrome in 14 patients. Basal ganglia abnormalities were revealed by magnetic resonance imaging and at
autopsy (as perivascular inflammation and microglial activation). Four patients improved with chemotherapy,
and 2 improved with intravenous methylprednisolone.
Ann Neurol 2002;51:625– 630
A diverse spectrum of subacute central or peripheral
nervous system disorders can occur with small-cell lung
carcinoma (SCLC). Neuron-specific autoantibodies aid
the serological diagnosis of these autoimmune syndromes. Peripheral neuropathy and limbic encephalitis
are common presentations. Disorders of the cerebellum, brainstem, spinal cord, nerve roots, and neuromuscular junction are also recognized. Paraneoplastic
movement disorders are rare. Four cases of chorea have
been reported with SCLC,1– 4 2 have been reported
with lymphoma,5,6 and 1 has been reported with renal
cell carcinoma.7
From the 1Department of Neurology, Mayo Clinic, Rochester,
MN; 2Department of Neurology, Mayo Clinic, Scottsdale, AZ;
3
Department of Neurology, Mayo Clinic, Jacksonville, FL; 4Department of Neurology, University of Minnesota, Minneapolis,
MN; 5Department of Immunology, Mayo Graduate School of
Medicine, Rochester, MN; 6Department of Laboratory Medicine
and Pathology, Mayo Clinic, Rochester, MN; and 7John Wayne
Cancer Institute, Santa Monica, CA.
Received Oct 22, 2001, and in revised form Jan 2, 2002. Accepted
for publication Jan 4, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10178
Address correspondence to Dr Vernino or Dr Lennon, Neuroimmunology Laboratory, Department of Neurology, Mayo Clinic, 200
First Street SW, Rochester, MN 55905. E-mail: verns@mayo.edu or
lennon.vanda@mayo.edu
© 2002 Wiley-Liss, Inc.
625
Autoimmune forms of chorea are recognized in children and young adults. Sydenham’s chorea characteristically follows childhood streptococcal infection and is
associated with antistreptolysin antibodies.8 Chorea is
also a rare manifestation of systemic lupus erythematosus and antiphospholipid antibody syndrome.9
Immunoglobuin G (IgG) specific for CRMP-5, a
neuronal cytoplasmic protein related to the collapsin
response-mediator protein family, is a marker of paraneoplastic autoimmunity related to SCLC or thymoma.10 In the first report of CRMP-5-IgG, movement
disorders were noted in 15% of seropositive patients,
and subacute chorea was noted in 11%. Chorea has
not been reported with anti-CV2 antibody,11 a name
initially assigned to an antibody with oligodendrocyterestricted immunoreactivity in the adult brain (see Table 2 in the letter of Lennon and colleagues12). Our
laboratory has proposed that anti-CV2 is the same entity as CRMP-5-IgG.10,12 In this article, we describe
16 patients with paraneoplastic chorea as a syndromic
manifestation of lung cancer-related neurological autoimmunity associated with CRMP-5-IgG.
Patients and Methods
Patients were identified prospectively by the Neuroimmunology Laboratory at the Mayo Clinic through serological evaluation for paraneoplastic autoantibodies. Neuronal nuclear
and cytoplasmic antibodies (including CRMP-5-IgG) in serum and spinal fluid were detected with a standardized immunofluorescence assay.10,13,14 CRMP-5-IgG was confirmed
in Western blot analysis with recombinant human CRMP5.10 Neuronal voltage-gated calcium and potassium channel
antibodies were detected in radioimmunoprecipitation assays.15,16 Seven patients were evaluated at the Mayo Clinic,
and 9 were investigated at other institutions.
Results
Clinical Presentation
Sixteen Caucasian patients (10 women, 6 men; mean
age, 69 years) presented with chorea in a paraneoplastic
context. All were smokers. Clinical and serological data
are summarized in the Table. In 11 patients, involuntary movements were the initial or most prominent
symptoms. In 5, chorea was asymmetrical. It affected
the face in 13 patients and was associated with hyperkinetic dysarthria in 9 patients. Dystonic posturing
was noted in 4 patients.
Fourteen patients had other neurological manifestations in addition to chorea, including subacute vision
loss (5 patients), progressive peripheral neuropathy (5),
limbic encephalitis (5), cerebellar ataxia (3), LambertEaton myasthenic syndrome (1), and myelitis (1). Four
reported an abrupt loss of taste and smell. Ten patients
(of 12 tested) had spinal fluid abnormalities, 9 had
protein elevation (68 –143mg/dl), and 5 had mild lymphocytic pleocytosis (10 –26 cells/hpf).
626
Annals of Neurology
Vol 51
No 5
May 2002
In 9 of 15 patients followed longitudinally, chorea
improved with symptomatic therapy (4/9), chemotherapy for SCLC (4/6), or intravenous methylprednisolone (2/3). Chorea did not improve in either of the
2 patients who underwent plasma exchange.
Neuroimaging Findings
Cranial magnetic resonance imaging (MRI) images
were reviewed in 8 patients. In 5 patients, scans performed within 3 months of chorea onset showed increased T2 and fluid-attenuated inverse recovery signals
involving the caudate and putamen (Fig 1A and C).
Resolution of MRI signal abnormalities was documented in 2 patients. In the 3 patients without definitive basal ganglia abnormalities, the initial brain MRI
was performed 4 to 14 months after chorea onset (see
Fig 1E). Diffusion-weighted MRI images showed no
abnormalities in 3 patients. In 1 patient, positron emission tomography of the brain revealed hypometabolism
in the caudate nucleus bilaterally, similar to the pattern
seen in Huntington’s chorea.
Neuropathological Findings
Patients 4 and 16 both had chorea and limbic encephalitis at death. Neither was receiving immunosuppressant therapy. At autopsy, in both patients the basal
ganglia showed marked neuronal loss, gliosis, microglial activation, perivascular lymphocytic infiltration,
and scant microglial nodules consistent with paraneoplastic brain inflammation (Fig 2).17 Similar neuropathological lesions were found in the mesial temporal
lobes in both patients.
Serological Findings
All 16 patients had CRMP-5-IgG. The serum titers
(range, 120 – 61,440) did not correlate with the severity of chorea or the severity and nature of accompanying neurological manifestations. Of 7 patients tested,
all had CRMP-5-IgG in their cerebrospinal fluid. The
cerebrospinal fluid titer exceeded that of the serum in 3
patients. Coexisting neuronal autoantibodies were
found in 50% of patients, regardless of a cancer diagnosis (type 1 anti-neuronal nuclear antibody [ANNA1], 6 patients), calcium channel antibodies (4), and potassium channel antibody (1).
Oncological Associations
SCLC was proven histologically in 11 patients and limited to the chest in all instances. Two patients with
SCLC had a concurrent diagnosis of adenocarcinoma
(of the lung and breast, respectively). Cancer diagnosis
preceded the onset of chorea in only 2 patients. In 1
patient, SCLC was diagnosed more than 2 years after
chorea onset, and in another, a small primary bronchial
SCLC remained undiscovered until autopsy. Positron
emission tomography led to a diagnosis of SCLC in 3
Table. Neurological, Oncological, and Serological Characteristics of Paraneoplastic Chorea in 16 Patients
Patient
Age/Gender
Movement Disorder
1
69/M
2
64/F
3
80/F
Right hemichorea and dystonia;
hyperkinetic dysarthria
Generalized chorea, most prominent on right; hyperkinetic
dysarthria
Chorea of face and limbs, prominent in left leg
4
66/M
5
68/M
6
66/M
7
67/F
8
66/F
9
72/F
10
73/M
Other Paraneoplastic
Neurological
Manifestations
Vision loss
SCLC
Limbic encephalitis
SCLC
Anosmia/ageusia, sensorimotor neuropathy
SCLC and
Lung
ACA
SCLC
Choreoathetosis of face and limbs Limbic encephalitis,
myelitis
Generalized chorea; hyperkinetic
dysarthria
Chorea of mouth, tongue, and
Limbic encephalitis, vision
hands, hyperkinetic dysarthria
loss, anosmia/ageusia
Choreoathetosis of face and limbs Sensory neuropathy
Generalized chorea and ballismus;
hyperkinetic dysarthria
Choreoathetosis of face; dystonia
of left foot; hyperkinetic dysarthria
Chorea of arms and trunk; akithesia
Choreodystonia of face and neck
Cancer
SCLC
SCLC
CRMP-5-IgG
Titerb
120 serum
512 CSF
15,360 serum VGCC-N 83pM
ANNA-1 15,360
120 serum
256 CSF
15,360 serum
4,096 CSF
120 serum
256 CSF
30,720 serum
1,024 CSF
7,680 serum
Limbic encephalitis
Lung mass
1,920 serum
Vision loss, ataxia, gastroparesis, anosmia/ageusia
SCLC
3,840 serum
128 CSF
Vision loss, painful neuropathy
Sensory neuropathy
Renal cell
61,440 serum
2,048 CSF
Lymphoma 7,680 serum
Generalized chorea; hyperkinetic
dysarthria
Chorea of hands and face
Ataxia, anosmia/ageusia
SCLC
Vision loss
11
75/F
12
71/F
13
71/M
14
74/F
Left hemiballismus and facial grimacing
15
65/F
Chorea of face and limbs; hyperkinetic dysarthria
16
62/F
Generalized chorea and dystonic
Limbic encephalitis, ataxia
neck turning, hyperkinetic dysarthria
SCLC
Treatment/
Chorea
Responsec
Survival
after Chorea
Onset (mo)
IVMP/Improved
2, dead
Chemotherapy
and IVIG/
Improved
d
Risperidone/
Improved
17, alive
Chemotherapy/
Improved
Chemotherapy/
Improved
VGCC-N 157pM Chemotherapy/
ANNA-1 61,440
No benefit
VGCC-N 77pM Clonazepam/
ANNA-1 3,840
Minimal benefit
Chloropromazine/Improved
ANNA-1 960
Chemotherapy/
Improved
37, alive
ANNA-1 3,840
24, dead
ANNA-1 7,680
13, alive
IVMP/Improved
3,840 serum
Lung massa 960 serum
SCLC and
Breast
ACA
Lambert-Eaton myasthenic SCLC
syndrome
Other Neuronal
Autoantibodies
VGKC 350pM
IVMP/No benefit
Chemotherapy/
No benefit
d
Haloperidol/
Improved
22, dead
38, alive
14, alive
27, alive
22, dead
42, alive
10, alive
12, dead
15,360 serum
4, alive
30,720 serum VGCC-P/Q
128 CSF
79pM
VGCC-N 62pM
61,440 serum
Unknown
6, dead
a
History of asbestos exposure.
Inverse of highest dilution of serum or CSF yielding a positive result.
c
Response refers to changes in symptoms and signs of chorea associated with treatment.
d
Patients 3 and 13 were treated with plasma exchange without benefit.
b
SCLC ⫽ small-cell lung carcinoma; ACA ⫽ adenocarcinoma; CSF ⫽ cerebrospinal fluid; VGCC ⫽ voltage-gated calcium channel antibody,
P/Q-type or N-type (normal value, ⬍20pmol/L); VGKC ⫽ voltage-gated potassium channel antibody (normal value, ⬍50pmol/L); IVMP ⫽
intravenous methylprednisolone; IVIG ⫽ intravenous immunoglobulin G.
patients when computed tomography of the chest was
inconclusive. Of 5 patients without proven SCLC, 1
had renal cell carcinoma, 1 had non-Hodgkin’s lymphoma, 2 had radiographic evidence of a lung mass,
and 1 had ANNA-1 as evidence of occult SCLC.
Discussion
This case series defines paraneoplastic chorea as a distinctive manifestation of neurological autoimmunity
related to SCLC. CRMP-5-IgG serves as a serological
marker. Prompt recognition of this disorder is critical
for two reasons. First, it affords the opportunity to diagnose and treat the cancer early. Second, our experience indicates that cancer treatment and immunomodulatory therapy may improve the neurological
outcome.
Paraneoplastic chorea is likely underrecognized.
There are many recognized causes of acquired chorea
in adults. A subacute onset is atypical of senile chorea,
Huntington’s disease, and other degenerative and in-
herited neurological diseases and should raise the possibility of a paraneoplastic cause. Toxic and metabolic
causes, such as side effects of dopamine agonists and
antagonists, hyperthyroidism, hyperglycemia, and Wilson’s disease, can generally be ruled out if a careful
patient history is obtained and appropriate laboratory
tests are conducted. Nonparaneoplastic autoimmune
chorea, such as Sydenham’s chorea and antiphospholipid antibody syndrome, rarely present for the first
time in the elderly.
Vision loss, peripheral neuropathy, limbic encephalitis, and abrupt loss of smell and taste were frequent
accompaniments of paraneoplastic chorea in our study,
although 2 of our patients presented with chorea alone.
Encephalopathy,1– 4,7 vision loss,4 ataxia, and neuropathy2 were noted in previously reported cases of paraneoplastic chorea. When chorea is associated with cognitive and psychiatric changes of limbic encephalitis,
the clinical presentation may mimic Huntington’s disease, Wilson’s disease, or CNS vasculitis.
Vernino et al: Paraneoplastic Chorea
627
Fig 1. Magnetic resonance imaging (MRI)
findings in patients with paraneoplastic
chorea. (A) In Patient 1, 1 week after the
onset of hemichorea, fluid-attenuated inverse recovery (FLAIR) MRI revealed
asymmetric signal hyperintensity in the
caudate and putamen. (B) In Patient 2,
cranial MRI performed to evaluate forgetfulness (4 months before the onset of chorea) was normal except for minimal
FLAIR hyperintensity of the caudate head
and left putamen. (C,D) In Patient 2, 1
month after the onset of chorea, cranial
MRI showed marked FLAIR hyperintensity
in the caudate head and anterior putamen
bilaterally as well as in the mesial temporal lobes. Chorea improved dramatically
after chemotherapy for small-cell lung carcinoma. (E) In Patient 2, clinical improvement was accompanied by the resolution of FLAIR signal abnormalities in the
basal ganglia. (F) In Patient 3, cranial
MRI 9 months after the onset of chorea
was unremarkable except for low T2 signals in the caudate and putamen.
Autoimmune hyperkinetic movement disorders presumably result from immune-mediated damage to the
striatum. This was evident radiographically in 5 of our
patients and pathologically in another 2 patients. The
MRI abnormalities (nonenhancing T2 hyperintensity
in the caudate and anterior putamen) were identical to
those reported in 3 other cases of paraneoplastic chorea3–5 and similar to those reported for Sydenham’s
chorea.18 These findings can be distinguished from the
basal ganglia abnormalities seen in Creutzfeldt-Jakob
disease by a lack of diffusion imaging abnormalities
and a lack of thalamic or cortical involvement. Neuro-
628
Annals of Neurology
Vol 51
No 5
May 2002
imaging abnormalities in the basal ganglia in paraneoplastic chorea may be transient (see Fig 1). In several
patients, brain MRI performed more than 3 months
after chorea onset was unremarkable.
We consider it unlikely that neuronal cytoplasmic
and nuclear autoantibodies are effectors of neuronal cytotoxicity; rather, they are surrogate markers for the activation of antigen-specific cytotoxic T lymphocytes.10
All 16 patients with paraneoplastic chorea had CRMP5-IgG. In this setting, a selective attack on the basal
ganglia by cytotoxic T cells could occur if major histocompatibility complex class I molecules on those
Fig 2. Neuropathological findings associated with paraneoplastic chorea. Patient 4 was a 66-year-old man who presented
with paraneoplastic limbic encephalitis, myelopathy, and chorea and died of recurrent metastatic small-cell lung carcinoma.
At autopsy, inflammatory changes were found in the basal
ganglia, spinal cord, and mesial temporal lobes. (A) A photomicrograph of the right caudate (⫻100 magnification before
52% reduction) shows perivascular lymphocyte infiltration,
astrocytosis, and microglial activation. The perivascular lymphocytes were shown to be T cells by immunohistochemical
staining (CD3 positive, not shown). (B) A photomicrograph of
the left hippocampus (⫻200 magnification before 52% reduction) shows a microglial nodule and gliosis.
neurons, upregulated by the local cytokine milieu, preferentially displayed peptides derived from cytoplasmic
CRMP-5. Two previous case reports described unclassified neuronal antibodies that recognized 68kDa3 and
76kDa4 proteins. Both were almost certainly CRMP-5.
The amino acid sequence of human CRMP-5 predicts
a molecule of 62kDa, but native CRMP-5 sometimes
migrates at 66 to 70kDa under denaturing gel electrophoresis conditions. Among neuronal nuclear and cytoplasmic autoantibody markers of autoimmunity related to SCLC, CRMP-5-IgG is second in frequency
only to ANNA-1 (anti-Hu).10 ANNA-1 is found in
20% of patients who are positive for CRMP-5-IgG.19
Of our 16 patients, 6 (37%) had coexisting ANNA-1.
However, chorea was not noted in another 600
ANNA-1-positive patients identified by our Clinical
Neuroimmunology Laboratory (V.A.L., unpublished
data). ANNA-1 was reported in a previous case of chorea with SCLC,1 but CRMP-5-IgG was not evaluated.
A case of chorea with renal cell carcinoma7 lacked antiHu, but no other serological testing was done. That
patient, like ours with renal cell carcinoma, had a
smoking history and likely had occult SCLC. Our laboratory has previously noted that renal cell carcinoma
is relatively common as a second neoplasm accompanying SCLC in an autoimmune context.20
On the basis of the observations reported in this article, we recommend that adult patients with a recent
onset of unexplained chorea have computed tomography of the chest and serological evaluation for markers
of lung cancer-related autoimmunity,13,14 including
CRMP-5-IgG. An evaluation of cerebrospinal fluid for
paraneoplastic neuronal antibodies is sometimes more
sensitive than serum evaluation. Because neoplasms related to CRMP-5-IgG are usually limited in spread and
difficult to find,10 positron emission tomography imaging in seropositive patients with known risk factors
for lung cancer may be diagnostic when chest computed tomography is indeterminate. Optimal treatment
of chorea requires prompt and effective therapy directed at the underlying cancer, the putative source of
neuronal immunogens, and may be ideally combined
with immunomodulatory therapy.
This work was supported by the Mayo Clinic Cancer Center
(S.V., V.A.L.).
References
1. Heckmann JG, Lang CJ, Druschky A, et al. Chorea resulting
from paraneoplastic encephalitis. Mov Disord 1997;12:
464 – 466.
2. Albin RL, Bromberg MB, Penney JB, Knapp R. Chorea and
dystonia: a remote effect of carcinoma. Mov Disord 1988;3:
162–169.
3. Tani T, Piao Y-S, Mori S, et al. Chorea resulting from paraneoplastic striatal encephalitis. J Neurol Neurosurg Psychiatry
2000;69:512–515.
4. Croteau D, Owainati A, Dalmau J, Rogers LR. Response to
cancer therapy in a patient with a paraneoplastic choreiform
disorder. Neurology 2001;57:719 –722.
5. Batchelor TT, Platten M, Palmer-Toy DE, et al. Chorea as a
paraneoplastic complication of Hodgkin’s disease. J Neurooncol
1998;36:185–190.
6. Nuti A, Ceravolo R, Salvetti S, et al. Paraneoplastic choreic
syndrome during non-Hodgkin’s lymphoma. Mov Disord
2000;15:350 –352.
7. Kujawa KA, Niemi VR, Tomasi MA, et al. Ballistic-choreic
movements as the presenting feature of renal cancer. Arch Neurol 2001;58:1133–1135.
8. Special Writing Group of the Committee on Rheumatic Fever,
Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young of the American Heart Association. Guidelines for the diagnosis of rheumatic fever. Jones criteria, 1992 update. JAMA 1992;268:2069 –2073.
9. Cervera R, Asheron R, Font J, et al. Chorea in the antiphospholipid syndrome: clinical, radiologic, and immunologic characteristics of 50 patients from our clinics and the recent literature. Medicine 1997;76:203–212.
10. Yu Z, Kryzer TJ, Griesmann GE, et al. CRMP-5 neuronal
autoantibody: marker of lung cancer and thymoma-related autoimmunity. Ann Neurol 2001;49:146 –154.
11. Rogemond V, Honnorat J. Anti-CV2 autoantibodies and paraneoplastic neurological syndromes. Clin Rev Allergy Immunol
2000;19:51–59.
Vernino et al: Paraneoplastic Chorea
629
12. Lennon V, Yu Z, Kryzer T, Griesmann G. Are the “newly discovered” paraneoplastic anti-CRMP5 antibodies simply antiCV2 antibodies? Ann Neurol 2001;50:690 – 691 (reply to the
editor).
13. Lennon VA. The case for a descriptive generic nomenclature:
clarification of immunostaining criteria for PCA-1, ANNA-1,
and ANNA-2 autoantibodies. Neurology 1994;44:2412–2415
(letter).
14. Vernino S, Lennon VA. New Purkinje cell antibody (PCA-2):
marker of lung cancer-related neurological autoimmunity. Ann
Neurol 2000;47:297–305.
15. Vernino S, Auger RG, Emslie-Smith AM, et al. Myasthenia,
thymoma, presynaptic antibodies, and a continuum of
neuromuscular hyperexcitability. Neurology 1999;53:1233–
1239.
16. Lennon VA, Kryzer TJ, Griesmann GE, et al. Calcium-channel
antibodies in the Lambert-Eaton syndrome and other paraneoplastic syndromes. N Engl J Med 1995;332:1467–1474.
17. Bakheit A, Kennedy P, Behan P. Paraneoplastic limbic
encephalitis: clinico-pathological correlations. J Neurol Neurosurg Psychiatry 1990;53:1084 –1088.
18. Emery ES, Vieco PT. Sydenham chorea: magnetic resonance
imaging reveals permanent basal ganglia injury. Neurology
1997;48:531–533.
19. Antoine J, Honnorat J, Camdessanche J, et al. Paraneoplastic
anti-CV2 antibodies react with peripheral nerve and are associated with a mixed axonal and demyelinating periperal neuropathy. Ann Neurol 2001;49:211–214.
20. Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and
oncologic profiles of patients seropositive for type 1 antineuronal nuclear autoantibodies. Neurology 1998;50:652– 657.
Mitochondrial DNA
C4171A/ND1 Is a Novel
Primary Causative Mutation
of Leber’s Hereditary Optic
Neuropathy with a
Good Prognosis
Ji Yeon Kim, MD,1 Jeong-Min Hwang, MD, PhD,2
and Sung Sup Park, MD, PhD1
A novel mitochondrial DNA C4171A mutation in the
ND1 gene in two Korean families with Leber’s hereditary
optic neuropathy is described. All affected patients recovered spontaneously after suffering months to years of initial visual loss. This mutation replaces leucine with methionine in a conserved extramembrane loop of the ND1
gene and was absent in 514 normal controls and in 63
Leber’s hereditary optic neuropathy lineages harboring
the primary mutations. We consider mitochondrial DNA
C4171A/ND1 a primary causative mutation of Leber’s
hereditary optic neuropathy with a good prognosis.
Ann Neurol 2002;51:630 – 634
Leber’s hereditary optic neuropathy (LHON; MIM
535000) is a maternally inherited disease characterized
by acute visual loss with varying degrees of recovery.1
G11778A/ND4, G3460A/ND1, and T14484C/ND6
are the main contributors to the development of blindness in LHON and account for approximately 95% of
all Caucasian LHON patients.2,3
A significant number of individuals suspected of
having LHON do not harbor any of the three primary
mitochondrial DNA (mtDNA) mutations: G11778A,
G3460A, and T14484C. Recently, new mtDNA missense mutations have been reported to cause LHON.4,5
This finding suggests that there is further genetic heterogeneity of LHON.
This study reports two LHON families who pre-
From the 1Department of Clinical Pathology, Seoul National University College of Medicine and Seoul National University Hospital
Clinical Research Institute; and 2Department of Ophthalmology,
Seoul National University College of Medicine and Seoul Municipal
Boramae Hospital, Seoul, Korea.
Received Nov 6, 2001, and in revised form Jan 2, 2002. Accepted
for publication Jan 4, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10177
Address correspondence to Dr Park, Department of Clinical Pathology, Seoul National University Hospital, Seoul 110-744, Korea.
E-mail: sparkle@plaza.snu.ac.kr
630
© 2002 Wiley-Liss, Inc.
sented with a good visual recovery but did not harbor
the three primary mtDNA mutations. A sequence analysis of their mtDNA showed a pathogenic mutation in
the ND1 gene within a highly conserved hydrophilic
loop that caused LHON in these families.
Patients and Methods
Case Reports
FAMILY S (FIG 1)
Family Member IV:3 (Proband). A 17-year-old boy noted a
painless and progressive deterioration of vision, first in his
left eye in October 2000 and then 2 weeks later in his right
eye. In December 2000, the best corrected visual acuity was
Fig 1. Pedigrees of the 2 Korean Leber’s hereditary optic neuropathy (LHON) families with the C4171A mutation in the
ND1 gene. Black symbols indicate subjects suffering from
LHON; open symbols denote asymptomatic subjects. Probands
are indicated by arrows. A question mark indicates visual loss
of an unknown origin. In Family K, family member III:1 is
afflicted with Coats disease of the right eye.
a finger count in both eyes. A color test using Ishihara plates
showed 0/14 in both eyes (OU). The fundus findings included a slight blurring of the upper margin of the optic disc
in his right eye and the beginnings of temporal paleness of
the disc in his left. A Goldmann visual-field examination
showed a large cecocentral scotoma in each eye. The results
of a brain computed tomography scan were normal. In September 2001, visual acuity improved to 20/200 in the right
eye (OD) and 20/30 in the left eye (OS). A Goldmann
visual-field examination showed a smaller cecocentral scotoma in each eye.
Family Member II:2. The grandmother of the proband was
a 67-year-old woman who had suffered from visual loss in
both eyes at the age of 27 years. Almost 3 years after her
initial visual loss, her visual acuity improved over a period of
months. In February 2001, the best corrected visual acuity
was 20/40 OD and 20/25 OS. The optic disc showed increased cupping and a temporal pallor OU. A Humphrey
visual-field examination showed a bilateral central depression.
Family Member III:1. The uncle of the proband was a 46year-old man who had suffered from sudden visual loss in
both eyes at 19 years of age. Six months after the initial
visual loss, his visual acuity improved over a period of weeks.
In February 2001, the best corrected visual acuity was 20/20
OU. The optic disc showed increased cupping and a temporal pallor OU. A Goldmann visual-field examination showed
a bilateral central scotoma.
FAMILY K (SEE FIG 1)
Family Member III:2 (Proband). An 8-year-old boy complained of a painless and progressive deterioration of vision
first in his right eye in April 1999 and 4 months later in his
left eye. In October 1999, the best corrected visual acuity
was 20/400 OU. A color test using the Ishihara plates
showed 2/14 OD and 0/14 OS. The fundus findings included hyperemic optic discs with peripapillary microangiopathy with telangiectatic capillaries and the beginnings of
temporal paleness of the disc in each eye. A Goldmann
visual-field examination showed a small cecocentral scotoma
in each eye. A bilateral elongation of the implicit time was
detected on the pattern visual evoked potential. The results
of magnetic resonance imaging of the brain were normal. In
August 2001, the visual acuity improved to 20/20 OD and
20/40 OS. A color test using Ishihara plates showed 14/14
OU. The fundus findings included a mild temporal paleness
of the disc in each eye. A Humphrey visual-field examination
showed an almost normalized visual field in each eye.
Molecular Genetic Analysis
Venous blood samples were obtained from each family member with informed consent. Total DNA was extracted from
leukocytes by standard methods. We performed direct sequencing of the entire mitochondrial genome from the proband (IV:3) in Family S according to a modification of a
protocol reported previously.6 Screening for the C4171 mutation was performed with polymerase chain reaction restriction fragment length polymorphism analysis in 514 normal
controls, 63 LHON patients harboring the three primary
Kim et al: mtDNA C4171A Mutation in LHON
631
mutations, and 22 LHON patients without the three primary mutations. In this procedure, the proband in Family K
with the C4171A mutation was identified.
The percentage of mutant mtDNA in the heteroplasmic
sample was estimated with a Diana I densitometer (Raytest,
Straubenhardt, Germany). The ND1 amino acid sequence
was downloaded from MITOMAP (http://infinity.gen.
emory.edu/mitomap.html) and aligned with Pfam 6.5
(http://pfam.wustl.edu) and a previous report.7 The structural analysis of the protein was carried out with the TMpred
program provided by the Baylor College of Medicine (http://
searchlauncher.bcm.tmc.edu/seq-search/struc-predict.html).
Results
Sequencing analysis of the entire mtDNA of the proband (IV:3) in Family S showed a total of 29 sequence
changes relative to the revised Cambridge Reference
Sequence (Table).8,9
No other previously identified pathogenic mtDNA
mutations were detected. On the basis of extensive surveys of the available Web-based database and the published literature, the C4171A mutation was the only
novel mutation among the sequence changes (http://
infinity.gen.emory.edu/mitomap.html).8,9 This mutation was homoplasmic in all the family members af-
flicted with LHON in the 2 families, whereas it was
heteroplasmic (88%) in the asymptomatic proband’s
mother (II:3) in Family K (data not shown). Families S
and K clearly belonged to Asian haplogroup A. The
D-loops of both mtDNAs were shown to have 13 sequence changes from the revised Cambridge Reference
Sequence, all of which were the same in the 2 probands (see Table). These findings showed that Families
S and K were unrelated in their family histories but
genetically had the same mtDNA background.
A new C-to-A transversion at np 4171 was not detected in the 514 normal control subjects and 63
LHON patients harboring the three primary mutations. This mutation leads to a replacement of a
leucine by a methionine residue at codon 289 of the
ND1 gene (Leu289Met). The Leu289 residue is a
highly conserved amino acid in mammalian species.
The hydropathy profile of the ND1 protein indicates
that this Leu3 Met occurred at a highly conserved region belonging to the predicted extramembrane loop
located between the G and H transmembrane domains
(Fig 2). Both leucine and methionine are large, hydrophobic, and nonpolar amino acids. Therefore, this
Table. Nucleotide Changes from the Revised Cambridge Reference Sequence9 for the Proband of Family S
Gene
Nucleotide
Position
Nucleotide
Change
Amino Acid
Change
Known Sequence
Alteration
663
750
1438
1736
2706
4171
4248
4769
4824
7028
8794
8860
11719
12705
14766
15326
73
152
235
263
303
311
501
515
516
16223
16290
16319
16362
A3G
A3G
A3G
A3G
A3G
C3A
T3C
A3G
A3G
C3T
C3T
A3G
G3A
C3T
C3T
A3G
A3G
T3C
A3G
A3G
Ins C
Ins C
C3T
Del A
Del C
C3T
C3T
G3A
T3C
NA
NA
NA
NA
NA
Leu289Met
Ile314/silent
Met100/silent
Thr119Ala
Ala375/silent
His90Tyr
Thr112Ala
Gly320/silent
Ile123/silent
Thr7Ile
Thr194Ala
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
⫹
⫹
⫹
⫹
⫹
⫺
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫺
⫹
⫺
⫹
⫹
⫹
⫹
12S rRNA
16S rRNA
ND1
ND2
CO1
ATP6
ND4
ND5
CYB
D-loop
NA ⫽ not applicable.
632
Annals of Neurology
Vol 51
No 5
May 2002
Fig 2. Hydropathy profile of the human mitochondrial ND1.
The hydropathic index was calculated with the Kyte–Doolittle
method with a TMpred. The larger arrow at right indicates
the Leu289 residue in the conserved hydrophilic loop between
the G and H transmembrane domains. Small arrows indicate
the other pathogenic ND1 mutations at the Leu285 and
Ala52 residues.
L289M substitution slightly changes the Kyte–
Doolittle hydropathy profile of the ND1 protein.
Discussion
More than 95% of LHON cases are the result of one
of the three primary DNA mutations that affect the
mtDNA complex I genes.3 The ND1 subunit plays an
important role in the molecular enzymology of complex I. In the transmembrane helical structure of the
ND1 protein, the most conserved regions belong to
predicted extramembrane loops. Almost all the reported ND1 mutations, including the G3460A and
C4160T mutations, lie in these loops.10 The G3460A
mutation (Ala52Thr), the second most severe primary
LHON mutation,11 induces a strong deficiency in
complex I activity with a 60 to 80% reduction and a
less severe reduction (30%) in respiration.10,12
The T4160C mutation (Leu285Pro), which had the
copresence of the T14484C mutation and presented
neurological and ophthalmological defects, displayed a
marked deficiency in complex I activity and a resistance to rotenone.13–15 The C4171A mutation changes
an Leu289 residue in the same extramembrane loop in
which the Leu285 residue (C4160T mutation site) is
located. This loop is distal to the quinone reductase
site, which is blocked by a rotenone.13,16 Therefore,
the effects that the Leu285 and Leu289 substitutions
have on complex I function may be indirect, possibly
allosterically deranging the quinone reductase site.13
The C4160T mutation replaces the large, nonpolar,
and hydrophobic leucine with the small and uncharged
but hydrophilic proline, whereas the C4171A mutation
does not alter the side-chain hydrophobicity and the
volume, except for its aliphaticity. It is suspected that
the C4171A mutation (Leu289Met) produces a clinical
phenotype of LHON with a good visual prognosis because of the slight changes in the amino acid characteristics. This region of the ND1 subunit, around np
4160 and 4171 of mtDNA, appears to be important,
but the clinical abnormalities produced by mutations
may differ. Making a comparison with the C4171A
mutation with LHON, Howell17 reported that the
T4160C mutation might cause neurological abnormalities, and the optic neuropathy was caused by the
T14484C mutation in a Queensland family who harbored C4160T and T14484C mutations.
Interestingly, the C4171A mutation showed a high
frequency of vision recovery. All patients (100%) with
this mutation demonstrated an improvement in visual
acuity of up to 20/40 or better in at least one eye. This
finding is in marked contrast to results for patients
with the G11778A mutation, with only 5 (4%) of 136
showing significant visual recovery.18 This is also a better outcome than that of the patients with the G3460A
and T14484C mutations.19 The visual prognosis with
the C4171A mutation might be the best ever reported.
In summary, a novel point mutation of the mtDNA
C4171A/ND1 is suggested as a primary cause of
LHON in these patients, according to the following
observations: (1) this mutation was identified in clinically typical LHON patients with definite maternal
transmission, (2) all the known primary and secondary
LHON mutations were excluded, (3) this mutation
was only a novel mutation, (4) this mutation occurred
at a evolutionarily conserved amino acid (Leu289Met)
in the ND1 subunit, and (5) it was not detected in
514 normal controls and 63 other LHON lineages.
References
1. Riordan-Eva P, Harding AE. Leber’s hereditary optic
neuropathy: the clinical relevance of different mitochondrial
DNA mutations. J Med Genet 1995;32:81– 87.
2. Wallace DC, Lott MT, Brown MD, Kerstann K. Mitochondria
and neuro-ophthalmologic diseases. In: Scrive CR, Beaudet AL,
Sly WS, Valle D, Childs B, Kinzler KW, Vogelstein B, eds.
The metabolic and molecular bases of inherited disease. New
York: McGraw-Hill, 2001:2425–2509.
3. Mackey DA, Oostra RJ, Rosenberg T, et al. Primary pathogenic
mtDNA mutations in multigeneration pedigrees with Leber hereditary optic neuropathy. Am J Hum Genet 1996;59:
481– 485.
4. Besch D, Leo-Kottler B, Zrenner E, Wissinger B. Leber’s hereditary optic neuropathy: clinical and molecular genetic findings in a patient with a new mutation in the ND6 gene.
Graefes Arch Clin Exp Ophthalmol 1999;237:745–752.
5. Chinnery PF, Brown DT, Andrews RM, et al. The mitochondrial ND6 gene is a hot spot for mutations that cause Leber’s
hereditary optic neuropathy. Brain 2001;124:209 –218.
6. Polyak K, Li Y, Zhu H, et al. Somatic mutations of the mitochondrial genome in human colorectal tumours. Nat Genet
1998;20:291–293.
Kim et al: mtDNA C4171A Mutation in LHON
633
7. Fearnley IM, Walker JE. Conservation of sequences of subunits
of mitochondrial complex I and their relationships with other
proteins. Biochim Biophys Acta 1992;1140:105–134.
8. Anderson S, Bankier AT, Barrell BG, et al. Sequence and organization of the human mitochondrial genome. Nature 1981;
290:457– 465.
9. Andrews RM, Kubacka I, Chinnery PF, et al. Reanalysis and
revision of the Cambridge reference sequence for human mitochondrial DNA. Nat Genet 1999;23:147.
10. Carelli V, Ghelli A, Ratta M, et al. Leber’s hereditary optic
neuropathy: biochemical effect of 11778/ND4 and 3460/ND1
mutations and correlation with the mitochondrial genotype.
Neurology 1997;48:1623–1632.
11. Huoponen K, Vilkki J, Aula P, et al. A new mtDNA mutation
associated with Leber hereditary optic neuroretinopathy. Am J
Hum Genet 1991;48:1147–1153.
12. Brown MD, Trounce IA, Jun AS, et al. Functional analysis of
lymphoblast and cybrid mitochondria containing the 3460,
11778, or 14484 Leber’s hereditary optic neuropathy mitochondrial DNA mutation. J Biol Chem 2000;275:39831–
39836.
13. Howell N, Kubacka I, Xu M, McCullough DA. Leber hereditary optic neuropathy: involvement of the mitochondrial ND1
gene and evidence for an intragenic suppressor mutation. Am J
Hum Genet 1991;48:935–942.
14. Parker WD Jr, Oley CA, Parks JK. A defect in mitochondrial
electron-transport activity (NADH-coenzyme Q oxidoreductase) in Leber’s hereditary optic neuropathy. N Engl J Med
1989;320:1331–1333.
15. Carelli V, Ghelli A, Bucchi L, et al. Biochemical features of
mtDNA 14484 (ND6/M64V) point mutation associated with
Leber’s hereditary optic neuropathy. Ann Neurol 1999;45:
320 –328.
16. Friedrich T, Strohdeicher M, Hofhaus G, et al. The same domain motif for ubiquinone reduction in mitochondrial or chloroplast NADH dehydrogenase and bacterial glucose dehydrogenase. FEBS Lett 1990;265:37– 40.
17. Howell N. Primary LHON mutations: trying to separate
“fruyt” from “chaf.” Clin Neurosci 1994;2:130 –137.
18. Stone EM, Newman NJ, Miller NR, et al. Visual recovery in
patients with Leber’s hereditary optic neuropathy and the
11778 mutation. J Clin Neuroophthalmol 1992;12:10 –14.
19. Johns DR, Heher KL, Miller NR, Smith KH. Leber’s hereditary optic neuropathy. Clinical manifestations of the 14484
mutation. Arch Ophthalmol 1993;111:495– 498.
Deficiency of TetralinoleoylCardiolipin in
Barth Syndrome
Michael Schlame, MD,1 Jeffrey A. Towbin, MD,2
Paul M. Heerdt, MD,3 Roswitha Jehle,3
Salvatore DiMauro, MD,4 and Thomas J. J. Blanck, MD1
Barth syndrome is an X-linked cardiac and skeletal mitochondrial myopathy. Barth syndrome may be due to
lipid alterations because the product of the mutated gene
is homologous to phospholipid acyltransferases. Here we
document that a single mitochondrial phospholipid species, tetralinoleoyl-cardiolipin, was lacking in the skeletal
muscle (n ⴝ 2), right ventricle (n ⴝ 2), left ventricle
(n ⴝ 2), and platelets (n ⴝ 6) of 8 children with Barth
syndrome. Tetralinoleoyl-cardiolipin is specifically enriched in normal skeletal muscle and the normal heart.
These findings support the notion that Barth syndrome is
caused by alterations of mitochondrial lipids.
Ann Neurol 2002;51:634 – 637
Barth syndrome (BTHS; MIM 302060) is an X-linked
cardiac and skeletal myopathy associated with cyclic
neutropenia and 3-methylglutaconic aciduria. The
most common presentation is seen in boys with dilated
cardiomyopathy (DCM), proximal muscle weakness,
short stature, and increased susceptibility to infections.1–3 The gene responsible for BTHS, G4.5, can
express up to 10 different proteins, termed tafazzins,
due to alternative splicing and usage of different initiation sites.4,5 The precise pathogenesis of BTHS has
remained elusive, although alterations of the mitochondrial ultrastructure and the respiratory chain have long
been recognized.6,7
Tafazzins share several conserved regions with phospholipid acyltransferases of diverse organisms, suggesting that BTHS may be caused by a defect or defects in
From the 1Department of Anesthesiology, Hospital for Special Surgery, New York, NY; 2Department of Pediatrics (Cardiology), Baylor College of Medicine, Houston, TX; 3Department of Anesthesiology, New York Presbyterian Hospital, Weill Medical College of
Cornell University, New York, NY; and 4Department of Neurology,
College of Physicians and Surgeons, Columbia University, New
York, NY.
Received Aug 30, 2001, and in revised form Jan 7, 2002. Accepted
for publication Jan 7, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10176
Address correspondence to Drs Schlame and Blanck, Department of
Anesthesiology, New York University School of Medicine, 550 First
Avenue, New York, NY 10016. E-mail: schlame@hotmail.com and
thomas.blanck@med.nyu.edu
634
© 2002 Wiley-Liss, Inc.
lipid acyl transfer.8 In agreement with this hypothesis,
Vreken and colleagues reported reduced content of cardiolipin and reduced incorporation of linoleic acid into
polyglycerophospholipids in cultured fibroblasts from
BTHS patients.9 Cardiolipin is the only phospholipid
exclusively localized in mitochondria, where it is intimately associated with the protein complexes of the respiratory chain.10 Therefore, changes in cardiolipin
may drastically affect mitochondrial function. In this
article, we document cardiolipin deficiency in several
tissues from BTHS patients and show that the deficiency specifically affects tetralinoleoyl-cardiolipin (L4CL), a cardiolipin species in which all four acyl positions are substituted by linoleic acid. The formation of
L4-CL is thought to involve specific acyltransferase activity because L4-CL is derived from other cardiolipin
species by sequential cycles of deacylation and reacylation.11
Patients and Methods
We studied 8 patients with BTHS (ages 1–18 years), all of
whom had DCM and an established mutation in the G4.5
gene. In addition, every child had two or more of the following noncardiac symptoms: cyclic neutropenia, growth retardation, proximal muscle weakness, and 3-methylglutaconic
aciduria. We obtained left and right heart ventricle tissue
from 2 patients, skeletal muscle and platelets from 2 patients,
and platelets from 4 patients. Skeletal muscle biopsies were
also obtained from 5 children with Prader-Willi syndrome
(ages 2–15 years) and from 2 infants (ages 1–3 months), 1
presenting with DCM and the other with left ventricular
noncompaction. Control skeletal muscle biopsies were from
9 pediatric patients (ages 1–18 years) and 4 adult patients
(ages 28 – 47 years) in whom diagnostic biopsies had shown
minimal nonspecific changes. Furthermore, cardiac tissue
samples (paired samples of left and right ventricles) were obtained from 5 adult patients (ages 59 – 67 years) with ischemic cardiomyopathy and from 12 patients with DCM (3
children, ages 11–19 years, and 9 adults, ages 37– 66 years),
all of whom underwent heart transplantation. None of the
DCM patients displayed features of BTHS. Control cardiac
tissue was obtained from 6 adult donor hearts that were not
transplantable. Platelets were isolated from 14 healthy volunteers, including 8 adults (ages 23–56 years), 3 children (ages
8 –15 years), and 3 female carriers of BTHS (ages 42– 46
years). Placenta tissue and fibroblasts were obtained from
healthy individuals. The protocols used for harvesting human
specimens were approved by the Institutional Review Boards
of New York Presbyterian Hospital, New York, NY (heart,
skeletal muscle, placenta, and fibroblasts); the Hospital for
Special Surgery, New York, NY (platelets and skeletal muscle); and Texas Children’s Hospital, Houston, TX (heart).
Patients or their guardians gave written informed consent in
all cases. Tissues (heart, skeletal muscle, liver, kidney, esophagus, and thymus) were harvested from adult animals, including Sprague-Dawley rats, beagles, and C3H/HeN mice.
Animals were sacrificed by protocols approved by the Animal
Care and Use Committee of the Hospital for Special Surgery. Human fibroblasts, human Jurkat lymphoblasts, and
rat H9c2 myocytes were grown in standard cell culture media. Fibroblasts were cultured from skin biopsies. Lymphoblast and myocyte cell lines were purchased from American
Type Culture Collection (Rockville, MD). Cardiolipin was
analyzed as previously described.12 Standard procedures were
used for the analysis of fatty acids and phospholipids and for
the measurement of protein content.
Results
Cardiolipin concentration was decreased in skeletal muscle,
heart muscle, and platelets from patients with BTHS. This
was primarily due to L4-CL deficiency, as demonstrated by
chromatographic resolution of molecular species of cardiolipin (Fig 1). The reduction in total cardiolipin was about
80% in platelets and skeletal muscle but only about 20% in
cardiac tissue. In the heart, the loss of L4-CL was partly offset by the emergence of novel molecular species (peaks
A–D). The new species were identified by fatty acid analysis
as palmitoyl-palmitoleoyl-oleoyl-linoleoyl-cardiolipin (peak
A), palmitoyl-linoleoyl-dioleoyl-cardiolipin (peak B), palmitoyleicosadienoyl-dioleoyl-cardiolipin (peak C), and palmitoylpalmitoleoyl-oleoyl-stearoyl-cardiolipin (peak D). Other
phospholipids in platelets of BTHS patients showed normal
concentration and normal overall fatty acid patterns (data
not shown). The absence of L4-CL was consistently observed
Fig 1. High-performance liquid chromatography of cardiolipin
from control subjects and Barth syndrome (BTHS) patients.
Each box shows the chromatogram of cardiolipin obtained
from the indicated specimen and type of patient. The fluorescence yield was recorded versus the retention time (RT). The
molecular species are as follows: L4 ⫽ tetralinoleoyl-cardiolipin
(RT ⫽ 19.3 min); L3O ⫽ trilinoleoyl-oleoyl-cardiolipin
(RT ⫽ 21.9 min); A ⫽ palmitoyl-palmitoleoyl-oleoyl-linoleoylcardiolipin (RT ⫽ 25.2 min); B ⫽ palmitoyl-linoleoyldioleoyl-cardiolipin (RT ⫽ 27.0 min); C ⫽ palmitoyleicosadienoyl-dioleoyl-cardiolipin (RT ⫽ 28.5 min); D ⫽
palmitoyl-palmitoleoyl-oleoyl-stearoyl-cardiolipin (RT ⫽ 30.0
min); and S ⫽ the internal standard tristearoyl-oleoylcardiolipin (RT ⫽ 42.6 min).
Schlame et al: Cardiolipin in Barth Syndrome
635
in all BTHS specimens (Fig 2). In contrast, L4-CL was the
predominant cardiolipin species in skeletal muscle, the right
ventricle, the left ventricle, and platelets from healthy controls. Normal levels of L4-CL were also found in patients
with cardiac and skeletal muscle diseases other than BTHS,
including children with Prader-Willi syndrome, infants with
DCM or left ventricular noncompaction, children and adults
with DCM, and adults with ischemic cardiomyopathy. Female carriers of BTHS had normal concentrations of L4-CL
in their platelets. We compared levels of L4-CL in several
human and animal tissues (Table). L4-CL levels were highest
(70 – 80% of total cardiolipin) in the heart and skeletal muscle; intermediate (20 –50% of total cardiolipin) in the liver,
kidney, esophagus, platelets, thymus, and placenta; and lowest (⬍10% of total cardiolipin) in the brain and in cell cultures, including fibroblasts, myocytes, and lymphoblasts.
calization of L4-CL10 agrees with the mitochondrial abnormalities found in BTHS. Second, tissues with the highest
abundance of L4-CL, skeletal muscle and heart tissue, are
also those most affected by the disease. However, we cannot
rule out that additional biochemical changes may occur in
BTHS, either within phospholipids or in other molecules.
The variable expression of the three cardinal manifestations of BTHS—cardiopathy, myopathy, and leukopenia—
may blur the diagnosis of BTHS in some cases. In addition,
mutations in mitochondrial DNA may cause syndromes that
mimic BTHS.13 Our data show that analysis of cardiolipin
in platelets is a new and relatively noninvasive method of
differentiating BTHS from related diseases. Among various
DCMs with infantile, juvenile, and adult onset, we did not
Table. Abundance of Tetralinoleoyl-Cardiolipin in
Mammalian Tissues
Discussion
The data presented in this article support the hypothesis of
Neuwald8 that BTHS is due to alterations of mitochondrial
phospholipid composition. We found that L4-CL was virtually undetectable in platelets, heart tissue, and skeletal muscle
from affected boys, whereas values were normal in female
carriers. This indicates a defect in linoleoyl transfer, in keeping with Neuwald’s suggestion that the mutated proteins in
BTHS are acyltransferases.8 A linoleoyl-specific acyltransferase has been implicated in the formation of L4-CL, a process that most likely involves sequential deacylation and
reacylation of newly synthesized cardiolipin.11 Inhibition of
this process had been suggested by decreased linoleoyl incorporation into cardiolipin in BTHS fibroblasts.9 The novel
cardiolipin species found in heart tissue from BTHS patients
(peaks A–D; see Fig 1) may represent precursors that accumulate due to inhibition of the remodeling pathway leading
to L4-CL. Although the fatty acid patterns of these species,
featuring saturated and monounsaturated residues, are consistent with de novo biosynthesis, more work is needed to
understand their origins and their potential roles in the
pathogenesis of BTHS.
Is L4-CL deficiency a crucial pathogenetic factor or merely
an epiphenomenon of BTHS? Two notions argue in favor of
a critical role for L4-CL. First, the strictly mitochondrial lo-
Tissue/Cell
Species
L4-CL
(mol %)
n
Heart ventricle
Human
Dog
Rat
Human
Rat
Human
Dog
Rat
Rat
Rat
Rat
Human
Mouse
Human
Human
Rat
Dog
Rat
Human
80 ⫾ 2
77 ⫾ 2
77 ⫾ 5
79 ⫾ 2
73 ⫾ 2
74 ⫾ 1
70 ⫾ 1
41 ⫾ 3
51 ⫾ 5
50 ⫾ 1
46 ⫾ 6
42 ⫾ 2
36 ⫾ 1
20 ⫾ 2
7⫾1
5⫾1
2⫾1
5⫾4
1 ⫾ 0.1
6
3
3
13
6
6
2
3
4
2
3
7
2
4
4
2
4
2
2
Skeletal muscle
Heart atrium
Liver
Kidney
Esophagus
Platelet
Thymus
Placenta
Fibroblast
H9c2 myocyte
Brain
Lymphoblast
L4-CL is expressed as percentages of total cardiolipin. Data are means
plus or minus the standard error of the mean. L4-CL ⫽ tetralinoleoylcardiolipin; n ⫽ number of independent determinations.
Fig 2. Content of tetralinoleoyl-cardiolipin (L4-CL) in skeletal muscle, cardiac tissue, and platelets. Specimens were obtained from
controls and various groups of patients, as indicated along the bottom of the figure. Squares represent male subjects, circles represent
female subjects, gray symbols represent adults, and black symbols represent children. Barth syndrome (BTHS) carriers were mothers of
male children with confirmed BTHS. CM ⫽ cardiomyopathy; DCM ⫽ dilated cardiomyopathy; ICM ⫽ ischemic cardiomyopathy.
636
Annals of Neurology
Vol 51
No 5
May 2002
find the characteristic L4-CL deficiency associated with
BTHS.
This study was supported by the Texas Children’s Hospital Foundation Chair in Pediatric Cardiovascular Research and the John
Patrick Albright Foundation (J.A.T.) and grants from the Muscular
Dystrophy Association (S.M.) and the NIH (J.A.T., S.M.;
GM50686, T.J.J.B.)
We thank Daniel Burkhoff, College of Physicians and Surgeons,
Columbia University, for obtaining tissue samples from human
hearts. We are also grateful to all members of the Barth Syndrome
Foundation who actively supported the work presented in this article.
References
1. Barth PG, Wanders RJA, Vreken P, et al. X-linked cardioskeletal myopathy and neutropenia (Barth syndrome) (MIM
302060). J Inherit Metab Dis 1999;22:555–567.
2. Barth PG, Wanders RJA, Vreken P. X-linked cardioskeletal myopathy and neutropenia (Barth syndrome)—MIM 302060.
J Pediatr 1999;135:273–276.
3. Kelley RI, Cheatham JP, Clark BJ, et al. X-linked dilated
cardiomyopathy with neutropenia, growth retardation, and
3-methylglutaconic aciduria. J Pediatr 1991;119:738 –747.
4. Bione S, D’Adamo P, Maestrini E, et al. A novel X-linked gene,
G4.5, is responsible for Barth syndrome. Nat Genet 1996;12:
385–389.
5. D’Adamo P, Fassone L, Gedeon A, et al. The X-linked gene
G4.5 is responsible for different infantile dilated cardiomyopathies. Am J Hum Genet 1997;61:862– 867.
6. Neustein HB, Lurie PR, Dahms B, et al. An X-linked recessive
cardiomyopathy with abnormal mitochondria. Pediatrics 1979;
64:24 –29.
7. Barth PG, Scholte HR, Berden JA, et al. An X-linked mitochondrial disease affecting cardiac muscle, skeletal muscle and
neutrophil leukocytes. J Neurol Sci 1983;62:327–355.
8. Neuwald AF. Barth syndrome may be due to an acyltransferase
deficiency. Curr Biol 1997;7:R465–R466.
9. Vreken P, Valianpour F, Nijtmans LG, et al. Defective remodeling of cardiolipin and phosphatidylglycerol in Barth syndrome. Biochem Biophys Res Commun 2000;279:378 –382.
10. Schlame M, Rua D, Greenberg ML. The biosynthesis and functional role of cardiolipin. Prog Lipid Res 2000;39:257–288.
11. Schlame M, Rustow B. Lysocardiolipin formation and reacylation in isolated rat liver mitochondria. Biochem J 1990;272:
589 –595.
12. Schlame M, Shanske S, Doty S, et al. Microanalysis of cardiolipin in small biopsies including skeletal muscle from patients
with mitochondrial disease. J Lipid Res 1999;40:1585–1592.
13. De Kremer RD, Paschini-Capra A, Bacman S, et al. Barth’s
syndrome-like disorder: a new phenotype with a maternally inherited A3243G substitution of mitochondrial DNA (MELAS
mutation). Am J Med Genet 2001;99:83–93.
Brain Biopterin and
Tyrosine Hydroxylase in
Asymptomatic DopaResponsive Dystonia
Yoshiaki Furukawa, MD,1 Gregory Kapatos, PhD,3
John W. Haycock, PhD,4 Julian Worsley, MSc,2
Henry Wong, MSc,1 Stephen J. Kish, PhD,2
and Torbjoern G. Nygaard, MD5,6
It is assumed that brain biopterin and dopamine loss
should not be as severe in asymptomatic dopa-responsive
dystonia caused by GCH1 mutations as it is in symptomatic dopa-responsive dystonia. However, the actual status
of dopaminergic systems in asymptomatic cases is unknown. In the autopsied putamen of an asymptomatic
GCH1 mutation carrier, we found that brain biopterin
loss (ⴚ82%) paralleled that reported in dopa-responsive
dystonia patients (ⴚ84%). However, tyrosine hydroxylase
protein and dopamine levels (ⴚ52 and ⴚ44%, respectively) were not as severely affected as in symptomatic
patients (exceeding ⴚ97 and ⴚ88%, respectively). Our
data suggest that the extent of striatal tyrosine hydroxylase protein loss may be critical in determining doparesponsive dystonia symptomatology.
Ann Neurol 2002;51:637– 641
Dopa-responsive dystonia (DRD) is a syndrome characterized by childhood-onset dystonia and a dramatic
and sustained response to low doses of L-dopa.1,2 Autosomal dominant (AD) DRD can be caused by mutations in the GCH1 gene, which encodes GTP cyclohydrolase I (GTPCH), the first enzyme in the
biosynthetic pathway for tetrahydrobiopterin (BH4;
the essential cofactor for tyrosine hydroxylase [TH]).3,4
Some patients with autosomal recessive DRD have mu-
From the 1Movement Disorders Research Laboratory and 2Human
Neurochemical Pathology Laboratory, Centre for Addiction and
Mental Health–Clarke Division, Toronto, Ontario, Canada; 3Department of Psychiatry and Behavioral Neurosciences, Wayne State
University School of Medicine, Detroit, MI; 4Department of Biochemistry and Molecular Biology, Louisiana State University Health
Sciences Center, New Orleans, LA; 5Department of Neurology, East
Orange Veterans Affairs Medical Center, East Orange, NJ; and 6Department of Neurosciences, University of Medicine and Dentistry–
New Jersey Medical School, Newark, NJ.
Received Nov 8, 2001. Accepted for publication Jan 11, 2002.
Published online Apr 1, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10175
Address correspondence to Dr Furukawa, Movement Disorders Research Laboratory (R 211), Centre for Addiction and Mental
Health–Clarke Division, 250 College Street, Toronto, Ontario,
Canada M5T 1R8. E-mail: yoshiaki_furukawa@camh.net
© 2002 Wiley-Liss, Inc.
637
tations in the TH gene, and so the two abnormal gene
products identified so far in DRD are related to the
enzyme TH.5,6
Although the loss of enzyme protein was considered
to be limited to GTPCH in the AD form of DRD, we
found previously that levels not only of total biopterin
(BP; most exists as BH4), total neopterin (NP; the byproducts of the GTPCH reaction), and dopamine
(DA) but also of TH protein in the striatum were reduced in autopsied subjects with GTPCH-deficient
DRD.7,8 Other DA nerve terminal markers (including
dopa decarboxylase [DDC] protein) were preserved in
these symptomatic cases. However, the actual status of
brain dopaminergic involvement in asymptomatic ADDRD cases due to incomplete penetrance of GCH1
mutations9 is unknown. To evaluate possible factors
influencing the penetrance, we measured levels of BP,
NP, TH, and DDC proteins as well as DA in the autopsied brain of an asymptomatic GCH1 mutation carrier.
Case Report
Family Report
Clinical details of this English-American family with
AD-DRD have been reported previously.10 –12 Our
subject was a 55-year-old woman who was an asymptomatic carrier of a mutation linked to DRD in this
pedigree.11 Her mother had adult-onset benign parkinsonism (a phenotypic expression of DRD), and her
granddaughter developed typical DRD. The asymptomatic carrier had pulmonary fibrosis after chemotherapy and a bone marrow transplant for multiple
myeloma. She died after voluntary ventilator termination. A neuropathological investigation of this asymptomatic carrier demonstrated no Lewy bodies and a
normal population of cells with reduced melanin in the
substantia nigra. There were no degenerative changes
in any areas of the brain.
Molecular Genetic and Neurochemical Analysis
This study was approved by the Institutional Review
Boards of the East Orange Veterans Affairs Medical
Center and the Centre for Addiction and Mental
Health. We conducted direct sequencing of GCH1, using genomic DNA from brain tissue.8 Brain BP and
NP levels were determined by high-performance liquid
chromatography with fluorescence detection according
to the method described by Fukushima and Nixon,13
with some modifications (using the standards from
Schircks Laboratories, Jona, Switzerland). BP includes
BH4, quinonoid dihydrobiopterin, and 7,8-dihydrobiopterin. NP consists of degradation products of dihydroneopterin triphosphate, which is synthesized
from GTP by GTPCH and then converted into the
second intermediate in the biosynthesis of BH4 by
638
Annals of Neurology
Vol 51
No 5
May 2002
6-pyruvoyltetrahydropterin synthase. NP is generally
considered to reflect GTPCH activity. Concentrations
of TH and DDC proteins in the striatum were measured by Western blot analysis with affinity-purified
rabbit antirat TH and antibovine DDC antibodies.8
The amount of TH or DDC was calculated by interpolation from a tissue standard curve run on the same
gel. Striatal subdivisions were defined anatomically as
reported previously.14 Levels of DA were determined
in the intermediate portions of the rostral, intermediate, and caudal subdivisions of the putamen and of the
caudate nucleus (see Kish and colleagues14 for details)
by high-performance liquid chromatography with electrochemical detection.8 For each neurochemical measurement, the ages and postmortem times of 4 to 8
neurologically and psychiatrically normal control subjects were matched closely to those of the asymptomatic carrier. Because there is an influence of immune
status on NP concentration,8 we did not include any
control individuals with systemic infectious diseases,
conditions in which brain NP levels might have been
elevated.
Results
On one allele in the asymptomatic carrier, we identified a G-to-A transition in exon 1 of GCH1 (at nucleotide position 323), resulting in a glycine-to-aspartic
acid substitution (Gly 108 Asp). This missense mutation was previously reported in a compound heterozygote for GCH1 mutations; the compound heterozygote
developed relatively severe symptoms (dystonia with
motor delay).15 No other mutations in either the coding region or the splice sites of GCH1 were found in
the asymptomatic carrier.
Concentrations of brain BP were substantially decreased in the GCH1 mutation carrier compared with
those of the age-matched controls (putamen, ⫺82%;
caudate nucleus, ⫺82%; frontal cortex, ⫺57%; Table
1). Brain NP levels were also substantially reduced (putamen, ⫺57%; caudate nucleus, ⫺45%; frontal cortex,
⫺68%). Compared with the controls, the asymptomatic carrier had moderate TH protein loss in the putamen (⫺52%); the magnitude of loss in the caudate
nucleus (⫺30%) was less than that in the putamen.
Striatal DDC protein levels were normal in the asymptomatic carrier. Subregional DA concentrations in the
striatum of our asymptomatic carrier were within the
range of the controls with the exception of those in the
intermediate (⫺43%) and caudal (⫺44%) subdivisions
of the putamen (Table 2).
Discussion
To our knowledge, this is the first report of neurochemical findings in the brain of an asymptomatic
GCH1 mutation carrier.
The amino acid substitution that resulted from the
Table 1. Total Biopterin and Neopterin, Tyrosine Hydroxylase Protein, and Dopa Decarboxylase Protein Levels in the Striatum
and Frontal Cortex of an Asymptomatic GCH1 Mutation Carrier and of Normal Control Subjectsa
No. of
Subjects
Variable
Putamen
Caudate
Nucleus
Frontal Cortex
BP (pmol/g wet weight)
Asymptomatic carrier
Normal control subjectsb
1
4c
267
1471 ⫾ 90
(1218–1628)
198
1101 ⫾ 153
(686–1404)
36
84 ⫾ 12
(65–119)
NP (pmol/g wet weight)
Asymptomatic carrier
Normal control subjectsb
1
4c
29
67 ⫾ 8
(49–90)
22
40 ⫾ 4
(30–48)
10
31 ⫾ 10
(17–59)
TH protein (␮g tissue standard/10␮g protein)
Asymptomatic carrier
Normal control subjectsb
1
4d
6.73
13.91 ⫾ 1.67
(9.43–17.48)
10.14
14.44 ⫾ 0.72
(12.62–16.14)
NE
NE
DDC protein (␮g tissue standard/10␮g protein)
Asymptomatic carrier
Normal control subjectsb
1
4d
15.74
13.58 ⫾ 1.36
(9.52–15.02)
16.41
17.01 ⫾ 1.17
(14.12–19.43)
NE
NE
a
BP, NP, TH protein, and DDC protein levels were measured in the caudal subregion of the putamen and in the intermediate subregion of
the caudate nucleus. The age and postmortem time of the asymptomatic carrier were 55 years and 6 hours, respectively.
b
Values are expressed as mean ⫾ standard error (range).
c
For BP and NP measurements, the ages and postmortem times of the 4 normal control subjects (4 men) were 55 ⫾ 3 (46 –59) years and 6 ⫾
1 (4 – 8) hours (mean values ⫾ standard error [range]).
d
For TH protein and DDC protein measurements, the ages and postmortem times of the 4 normal control subjects (4 men) were 55 ⫾ 3
(46 –59) years and 7 ⫾ 1 (5– 8) hours (mean values ⫾ standard error [range]).
BP ⫽ total biopterin; NP ⫽ total neopterin; TH ⫽ tyrosine hydroxylase; DDC ⫽ dopa decarboxylase; NE ⫽ not examined.
Table 2. Subregional Levels of Dopamine in the Striatum of an Asymptomatic GCH1 Mutation Carrier and
of Normal Control Subjectsa
DA (ng/mg wet weight)
No. of
Subjects
Asymptomatic carrier
Normal control subjects
1
8b
Putamen
Caudate Nucleus
Rostral
Intermediate
Caudal
Rostral
Intermediate
Caudal
4.19
5.83 ⫾ 0.63
(3.61–8.57)
3.37
5.87 ⫾ 0.57
(4.72–9.39)
4.31
7.69 ⫾ 0.75
(4.76–11.41)
4.16
3.78 ⫾ 0.35
(2.26–4.92)
5.04
5.21 ⫾ 0.61
(1.77–7.06)
4.16
5.01 ⫾ 0.55
(3.48–8.44)
a
Subregions in the striatum were defined anatomically as reported previously.14 The age and postmortem time of the asymptomatic carrier were
55 years and 6 hours, respectively, and those of the 8 normal control subjects (1 woman, 7 men) were 57 ⫾ 3 (46 –70) years and 9 ⫾ 2 (4 –18)
hours (mean values ⫾ standard error [range]).
b
Values are expressed as mean ⫾ standard error (range).
DA ⫽ dopamine.
GCH1 mutation present in the asymptomatic carrier
(and linked to DRD in this family11) affects a highly
conserved amino acid residue across species.15 The mutant allele that has the nonconservative amino acid
change most likely produces dysfunctional GTPCH
and results in reduced BP and NP. Although it has
been assumed that brain GTPCH activity levels are decreased in asymptomatic GCH1 mutation carriers but
are still higher than those in GTPCH-deficient DRD
patients, our BP and NP data both in the striatum and
in the frontal cortex did not distinguish the asymptomatic case (see Table 1) from two symptomatic cases reported previously8 (BP and NP: putamen, ⫺84 and
⫺62% [mean values]; caudate nucleus, ⫺86% and exceeding ⫺62%; frontal cortex, ⫺62% and exceeding
⫺64%). Such brain pterin results are in agreement
with findings that low levels of cerebrospinal fluid
BP12 and lymphoblast NP16 in asymptomatic carriers
were indistinguishable from those in patients with ADDRD. Ichinose and colleagues3 reported that
GTPCH activity levels in phytohemagglutininstimulated mononuclear blood cells were higher in
asymptomatic cases (n ⫽ 2) than in symptomatic cases
(n ⫽ 7). However, using cultured lymphoblasts, Bezin
and colleagues16 suggested that the phytohemagglutinin induction alone misrepresents the actual status of
GTPCH activity.
Although striatal DDC protein levels were normal in
our asymptomatic carrier and in the reported patients
with GTPCH-deficient DRD (consistent with fluoro-
Furukawa et al: Asymptomatic Dopa-Responsive Dystonia
639
dopa positron emission tomography studies10,12), TH
protein concentrations in the putamen were moderately (⫺52%) and severely (exceeding ⫺97%) reduced
in the asymptomatic and symptomatic DRD cases,8 respectively (Fig). These human brain findings are compatible with TH protein loss but preserved DDC activity in brains of BH4-deficient mice, that is, GTPCHdeficient hph-1 mutants and 6-pyruvoyltetrahydropterin
synthase gene (PTS) null mutants.17,18 These data suggest that striatal DA reduction in GTPCH-deficient
DRD is caused not only by decreased TH activity resulting from a low cofactor level but also by actual loss
of TH protein without nerve terminal loss. We previously speculated that low levels of TH protein in the
striatum, especially in the putamen, of GTPCHdeficient DRD were caused by a diminished regulatory
effect of its cofactor, BH4, on the steady-state level
(stability/expression) of this enzyme protein.2,8 In fact,
gene-transfer experiments have suggested that the coexpression of GTPCH with TH stabilizes TH protein in
vivo.19 Because TH protein concentrations in the substantia nigra (where striatal TH molecules are synthesized) were normal in the GTPCH-deficient DRD patients, BH4 could control stability rather than the
expression of TH molecules.2,8 This is supported by a
recent report showing loss of TH protein but not of
TH mRNA in the brains of BH4-deficient PTS knockout mice.18 Alternatively, there might be a dysfunction
Fig. Levels of total biopterin (BP) and neopterin (NP), tyrosine hydroxylase (TH) protein, and dopamine (DA) in the
putamen of an asymptomatic GCH1 mutation carrier (AS
carrier), as well as mean levels in 2 symptomatic cases with
GTP cyclohydrolase I-deficient dopa-responsive dystonia
(DRD), expressed as percentages of age-matched control mean
values. For the 2 DRD cases, striatal neurochemical data are
from Furukawa and colleagues.8 Concentrations of BP, NP,
TH protein, and DA were determined in the caudal subregion
of the putamen, with the exception of BP and NP levels in
the reported DRD cases, which were measured in the intermediate subregion. In the DRD cases, TH protein levels in the
caudal subdivision of the putamen were less than 3% of the
mean values for controls.
640
Annals of Neurology
Vol 51
No 5
May 2002
of TH protein transport from the substantia nigra to
the striatum due to congenital partial GTPCH deficiency. The exact mechanism by which striatal TH
protein decreases in the AD form of DRD is unclear.
However, the different degrees of TH protein loss in
the putamen between asymptomatic and symptomatic
GTPCH-deficient DRD cases with the same magnitude of striatal BP and NP reduction (see Fig) suggest
that there are additional genetic or environmental factors, or both, that may modulate a regulatory effect of
BH4 on the steady-state level of TH molecules.
Finally, consistent with other postmortem brain data
suggesting that greater than 60 to 80% of striatal DA
loss is necessary for clinically overt motor symptoms to
occur,20 the maximal 44% reduction of DA in the striatum of our GCH1 mutation carrier (see Table 2) was
not sufficient to produce any DRD symptoms. In the
asymptomatic carrier, the DA reduction in the caudal
subregion of the putamen (⫺44%), which is most affected by DA loss in patients with Parkinson’s disease,14,20 was much milder than that (⫺88%) in the
reported DRD patients8 (see Fig).
In conclusion, although our findings in a single case
require replication in a representative number of subjects, only a modest reduction of TH protein despite a
marked reduction of BP in the putamen of our asymptomatic GCH1 mutation carrier suggests that the extent of striatal TH protein loss may contribute to
gender-related incomplete penetrance of GCH1 mutations in the AD form of DRD.
This study was supported in part by the Centre for Addiction and
Mental Health Foundation.
We thank Linda DiStefano for her technical assistance.
References
1. Nygaard TG. Dopa-responsive dystonia: delineation of the clinical syndrome and clues to pathogenesis. Adv Neurol 1993;60:
577–585.
2. Furukawa Y, Kish SJ. Dopa-responsive dystonia: recent advances and remaining issues to be addressed. Mov Disord 1999;
14:709 –715.
3. Ichinose H, Ohye T, Takahashi E, et al. Hereditary progressive
dystonia with marked diurnal fluctuation caused by mutations
in the GTP cyclohydrolase I gene. Nat Genet 1994;8:236 –242.
4. Furukawa Y, Guttman M, Sparagana SP, et al. Dopa-responsive
dystonia due to a large deletion in the GTP cyclohydrolase I
gene. Ann Neurol 2000;47:517–520.
5. Lüdecke B, Dworniczak B, Bartholomé K. A point mutation in
the tyrosine hydroxylase gene associated with Segawa’s syndrome. Hum Genet 1995;95:123–125.
6. Furukawa Y, Graf WD, Wong H, et al. Dopa-responsive dystonia simulating spastic paraplegia due to tyrosine hydroxylase
(TH) gene mutations. Neurology 2001;56:260 –263.
7. Rajput AH, Gibb WRG, Zhong XH, et al. Dopa-responsive
dystonia: pathological and biochemical observations in a case.
Ann Neurol 1994;35:396 – 402.
8. Furukawa Y, Nygaard TG, Gütlich M, et al. Striatal biopterin
and tyrosine hydroxylase protein reduction in dopa-responsive
dystonia. Neurology 1999;53:1032–1041.
9. Furukawa Y, Lang AE, Trugman JM, et al. Gender-related penetrance and de novo GTP-cyclohydrolase I gene mutations in
dopa-responsive dystonia. Neurology 1998;50:1015–1020.
10. Nygaard TG, Takahashi H, Heiman GA, et al. Long-term
treatment response and fluorodopa positron emission tomographic scanning of parkinsonism in a family with doparesponsive dystonia. Ann Neurol 1992;32:603– 608.
11. Nygaard TG, Wilhelmsen KC, Risch NJ, et al. Linkage mapping of dopa-responsive dystonia (DRD) to chromosome 14q.
Nat Genet 1993;5:386 –391.
12. Kishore A, Nygaard TG, de la Fuente-Fernandez R, et al. Striatal D2 receptors in symptomatic and asymptomatic carriers of
dopa-responsive dystonia measured with [11C]-raclopride and
positron-emission tomography. Neurology 1998;50:1028 –1032.
13. Fukushima T, Nixon JC. Analysis of reduced forms of biopterin in biological tissues and fluids. Anal Biochem 1980;102:
176 –188.
14. Kish SJ, Shannak K, Hornykiewicz O. Uneven pattern of dopamine loss in the striatum of patients with idiopathic Parkinson’s disease: pathophysiologic and clinical implications.
N Engl J Med 1988;318:876 – 880.
15. Furukawa Y, Kish SJ, Bebin EM, et al. Dystonia with motor
delay in compound heterozygotes for GTP-cyclohydrolase I
gene mutations. Ann Neurol 1998;44:10 –16.
16. Bezin L, Nygaard TG, Neville JD, et al. Reduced lymphoblast
neopterin detects GTP cyclohydrolase dysfunction in doparesponsive dystonia. Neurology 1998;50:1021–1027.
17. Hyland K, Gunasekera RS, Engle T, Arnold LA. Tetrahydrobiopterin and biogenic amine metabolism in the hph-1 mouse.
J Neurochem 1996;67:752–759.
18. Sumi-Ichinose C, Urano F, Kuroda R, et al. Catecholamines
and serotonin are differently regulated by tetrahydrobiopterin: a
study from 6-pyruvoyltetrahydropterin synthase knockout mice.
J Biol Chem 2001;276:41150 – 41160.
19. Leff SE, Rendahl KG, Spratt SK, et al. In vivo L-DOPA production by genetically modified primary rat fibroblast or 9L
gliosarcoma cell grafts via coexpression of GTP cyclohydrolase I
with tyrosine hydroxylase. Exp Neurol 1998;151:249 –264.
20. Hornykiewicz O. Biochemical aspects of Parkinson’s disease.
Neurology 1998;51(Suppl 2):S2–S9.
Seizure-Associated
Hippocampal Volume Loss:
A Longitudinal Magnetic
Resonance Study of
Temporal Lobe Epilepsy
Regula S. Briellmann, MD,
Samuel F. Berkovic, MD, FRACP, Ari Syngeniotis, MD,
Mark A. King, FRACP, and
Graeme D. Jackson, MD, FRACP
This longitudinal quantitative magnetic resonance imaging study of 24 patients with mild temporal lobe epilepsy
shows an ipsilateral hippocampal volume decrease of 9%
(range, ⴚ30 to ⴙ0.5%; p ⴝ 0.002, paired t test) over a
period of 3.5 ⴞ 0.7 years. The hippocampal volume loss
was correlated to the number of generalized seizures between the scans ( p ⴝ 0.0007, r ⴝ 0.6), suggesting
seizure-associated hippocampal damage.
Ann Neurol 2002;51:641– 644
There is controversy over whether seizures cause damage to the brain. Animal studies strongly suggest that
even single seizures are harmful.1 In humans, seizureassociated damage is difficult to prove because the controlled conditions of an animal experiment are not fulfilled. Seizure-associated damage may be reflected in
volume loss. Volumetric changes have been extensively
studied in temporal lobe epilepsy (TLE). In TLE, magnetic resonance imaging (MRI) and pathological studies suggest that hippocampal volume (HCV) loss may
be the result of two potential sources of damage: first,
that induced by an early event such as a prolonged febrile convulsion, and second, damage associated with
seizures themselves.2,3 Some retrospective quantitative
MRI studies have suggested progression of hippocampal atrophy, correlating with the frequency of generalized seizures,3,4 whereas other studies have not.5
To disentangle the effects of an initial injury from
damage secondary to seizures themselves, prospective
From the Brain Research Institute, Department of Neurology, Austin and Repatriation Medical Centre, University of Melbourne, Australia.
Received Oct 30, 2001, and in revised form Jan 15, 2002. Accepted
for publication Jan 16, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10171
Address correspondence to Dr Jackson, Brain Research Institute,
Austin and Repatriation Medical Centre, Heidelberg West, Victoria
3084, Australia. E-mail: g.jackson@brain.org.au
© 2002 Wiley-Liss, Inc.
641
studies are needed. Prospective MR studies have documented the development of hippocampal sclerosis
(HS)6 – 8 after a severe initial injury. Seizure-associated
damage has only rarely been investigated prospectively.9
We prospectively followed newly diagnosed adult
TLE patients with normal MRI findings on their first
scan. These patients rarely progress to intractable TLE
with HS,10 but they uniquely allow us to study the
effects of seizures on an MRI normal hippocampus.
Statistics
Patients and Methods
Patients
Results
Clinical Findings
The 24 patients (13 females and 11 males) were 30
⫾14 years at the epilepsy onset. Seven had right-sided
TLE, and 17 had left-sided TLE. Seizure onset was between 1 month and 17 years before the first scan
(mean, 3.6 years). These early seizures consisted mainly
of unrecognized simple partial seizures. At the time of
the first scan, they had experienced between 0 and 5
(mean, 1.5) GTCSs. Fifteen patients were treated with
carbamazepine, 3 received other drugs, and the remaining 6 patients were untreated.
The second MRI was performed 3.5 ⫾ 0.7 years after the first. During this period, the patients had a further 0 to 8 (mean, 1.7) GTCSs. Fifteen patients had a
mild course (0 –1 GTCSs), whereas 9 had two or more
seizures (Table). At the time of the second scan, 10 of
the 24 patients were treated with carbamazepine, 5 received other drugs, 7 were untreated, and 2 more patients were on a combination therapy of two antiepileptic drugs.
Thirty-four consecutive patients with newly diagnosed lateralized TLE, evaluated at our First Seizure Clinic, were enrolled in the study. Diagnosis was based on clinical history,
seizure description, and electroencephalogram (EEG) findings.11 All patients had a routine EEG within 24 hours and,
when it was normal, a sleep-deprived EEG.12 A special effort
was made to get a seizure description by an eyewitness. All
patients had at least two recurrent seizures within 4 months.
The diagnosis of TLE was based on typical temporal auras
and/or EEG discharges with a maximum over the temporal
lobes. Lateralization was based on lateralized EEG discharges,
with or without lateralized seizure features. Epileptiform discharges were diagnosed in the presence of focal spikes or
sharp waves followed by slow waves. Only patients with normal MRI findings reported on their clinical study were included. Most patients were seen every 2 to 6 months during
the follow-up period (M.A.K.).
After 3 to 4 years, these 34 patients were invited for a
second MRI. Five did not wish to participate, 4 would have
required sedation for scanning, and 1 suffered from another
intercurrent disease; therefore, 24 patients were rescanned.
Dates and circumstances of each generalized tonic clonic seizure (GTCS) and antiepileptic treatment between the two
scans were obtained from clinical notes and confirmed by
interview. We believed that our counts of complex partial
seizures were not accurate enough to be included as a variable.
Magnetic Resonance Examination
The first MRI was performed on a 1.5T Siemens SP
Magnetom scanner, and the second was performed on a
1.5T GE scanner. The hippocampal structure was assessed
with the same protocol at both scanners.13 For HCV measurements, three-dimensional coronal acquisition was used
(Magnetization-Prepared Rapid Acquisition Gradient Echo
for the first MRI and Fast Spoiled Gradient Echo for the
second).
The measurement of HCV was based on our established
protocol.13 Measurements were performed twice (by R.S.B.)
in separate sessions, without an awareness of the number of
GTCSs. The variability between measurements of the same
scan was 3% (SD, 6). Limits of agreement [mean difference ⫾ (2 ⫻ standard deviation)]14 were between ⫹357 and
⫺248mm3. All HCVs were corrected for the total of both
hemicranial volumes13 without gender correction. The mean
of the two corrected measurements was used.
642
Annals of Neurology
Vol 51
No 5
May 2002
A paired t test was used for the individual volume change
between the scans. Simple regression analysis was used for
the correlation between the volume change (expressed as a
percentage) and the number of GTCSs between the scans
(the correlation coefficient is indicated). A post hoc analysis
on possible predictors for a mild course (less than two
GTCSs between scans) was performed with post hoc analysis
of variance tests corrected for multiple comparisons (Bonferroni correction). The level of significance was set at 5%.
Magnetic Resonance Imaging Findings
In 1 of the 24 patients, qualitative and quantitative assessment showed normal findings at the initial investigation and unilateral HS (qualitatively defined as volume loss, increased signal, and disturbed internal
architecture) at the second investigation.15 In the other
23 patients, the qualitative assessment of the hippocampi was normal on both scans.
The mean ipsilateral HCV was 3,234mm3 (SD,
455) at the first scan and 2,967mm3 (SD, 539) at the
second scan. The mean contralateral HCV was
3,297mm3 (SD, 466) at the first scan and 3,140mm3
(SD, 473) at the second scan. Between the first and
second scans, the corrected, individual ipsilateral HCV
decreased by 9% or 267mm3 (range, decrease of 30%
to increase of 0.5%; p ⫽ 0.002, paired t test). The
contralateral HCV decreased by 5% or 157mm3
(range, decrease of 17% to increase of 6%, not significant).
The ipsilateral HCV loss correlated with the number
of GTCSs between the scans ( p ⫽ 0.0007, r ⫽ 0.6;
Fig). After exclusion of the HS patient, the remaining
patients still showed a correlation between GTCSs and
Table. Number of GTCSs between Scans
Less than two
(n ⫽ 15)
Age at first MRI
Male/female
Left/right focus side
Significant antecedents
Simple, short febrile convulsions
Positive family history of epilepsy
Onset epilepsy (yr)
Before first MRI
Duration of epilepsy (yr)
Number of GTCSs (average, range)
Antiepileptic treatment at first MRI
Nil
Carbamazepine
Other
Between first and second MRI
Duration follow-up
Auras present
Number of GTCSs (average, range)
Antiepileptic treatment at second MRI:
Nil
Carbamazepine
One other drug
Combination of two drugs
a
35
6
13
0
0
3
34
(⫾11)
(40%)/9 (60%)
(87%)/2 (13%)
(20%)
(⫾15)
Two or more
(n ⫽ 9)
25
5
4
0
1
4
21
(⫾7)
(55%)/4 (45%)
(45%)/5 (55%)
(11%)
(45%)
(⫾9)a
2.5 (⫾2.6)
1.4 (0–5)
4.6 (⫾6)
1.9 (1–4)
5 (33%)
9 (60%)
1 (7%)
1 (11%)
6 (66%)
2 (22%)
3.7 (⫾0.7)
12 (80%)
0.3 (0–1)
3.2 (⫾0.9)
9 (100%)
3.8 (2–8)
4 (27%)
6 (40%)
5 (33%)
0
3 (33%)
4 (45%)
0
2 (22%)
p ⫽ 0.02 (post hoc analysis of variance, Bonferroni corrected).
GTCSs ⫽ generalized tonic clonic seizure; MRI ⫽ magnetic resonance imaging.
HCV loss of less than 15%. These patients were older
at the onset of their epilepsy (mean, 34; SD, 15 years)
than those with a more severe course (mean, 21; SD, 9
years; p ⫽ 0.02; see Table). Treatment with antiepileptic drugs was not different between those with mild
or more severe courses. However, a combination of
two drugs was only used in 2 patients with a more
severe course (one with three GTCSs and an HS patient with six GTCSs). The correlation between
GTCSs and ipsilateral HCV loss was still significant
after the exclusion of these 2 patients ( p ⫽ 0.01, r ⫽
0.5).
Fig. Correlation between the number of generalized tonic
clonic seizures (GTCSs) experienced between the two scans and
the loss of ipsilateral hippocampal volume (HCV). The patient
with hippocampal sclerosis is highlighted ( filled symbol). HCV
loss ⫽ 5.5 ⫹ 2.3 ⫻ (number of GTCSs); r ⫽ 0.6; p ⫽
0.0007; F ⫽ 15.4.
ipsilateral HCV loss ( p ⫽ 0.009, r ⫽ 0.5). Ipsilateral
HCV loss was not associated with the number of
GTCSs before the first scan, age at onset, age at investigation, gender, or a positive family history of epilepsy.
All patients with a mild course showed ipsilateral
Discussion
This prospective, longitudinal quantitative MR study
shows that patients with a diagnosis of lateralized TLE
as adults have a small ipsilateral HCV decrease associated with the number of GTCSs experienced between
the scans. Only 1 of 24 patients developed HS during
the observation period.15 The rarity of progression into
HS in adults is in agreement with previous reports.10
The majority of adult, newly diagnosed TLE patients have a mild course with rare seizures.16 In this
patient group, the exact number of GTCSs can be determined relatively easily, and the disease is recognized
not long before the patient is first investigated. The
follow-up period of 3.5 years on average may be at the
lower limit that allows a sufficient number of seizures.
Briellmann et al: Progressive Hippocampal Damage in TLE
643
However, confounding effects may increase with the
duration of the follow-up. The imaging protocol was
not changed, but a new scanner had been installed during the follow-up. The two scanners were calibrated to
the same volumetric phantom, and the individual intracranial volume was used for normalization. Furthermore, we have previously reported results of a set of
controls investigated at different scanners,17 and the
small interscanner variability observed was in keeping
with previous studies.18 The observed HCV change
was different for the two sides; a systematic scannerrelated change would be expected to be randomly distributed.17 Therefore, the use of different scanners cannot explain our findings.
There is theoretical, experimental, and clinical evidence of secondary damage caused by seizures. Observations from kainic acid-induced seizures or kindling
models have demonstrated that seizures induce hippocampal alterations that may lead to increased excitability and HCV loss.1,19,20 The progression of brain
volume loss has been inferred by retrospective clinical
studies, mainly of refractory TLE.3,4,13 In our series, a
subtle HCV loss was observed in a mild type of TLE.
In a comparable series of 36 patients,10 an analysis
used cutoff values to define abnormality. Subtle HCV
changes, as in our series, were not assessed. It is possible that the ipsilateral HCV change is a marker for a
more widespread volume abnormality also including
neighboring structures such as the ipsilateral amygdala.
Our study does not prove that the hippocampus is the
seizure focus in all of our patients; however, our longitudinal series supports the argument for a harmful
effect of even a few GTCSs on the human brain.
This research was supported by a grant from the National Health
and Medical Research Council of Australia and was supported by
the Brain Imaging Research Foundation in Australia (R.S.B.).
We thank Dr M. Newton and the First Seizure Clinic for providing
us with their patients.
References
1. Bengzon J, Kokala Z, Elmer E, et al. Apoptosis and proliferation of dentate gyrus neurons after single and intermittent limbic seizures. Proc Natl Acad Sci U S A 1997;94:10432–10437.
2. Mathern GW, Babb TL, Vickrey BG, et al. The clinicalpathogenic mechanisms of hippocampal neuron loss and surgical outcomes in temporal lobe epilepsy. Brain 1995;118:
105–118.
3. Tasch E, Cendes F, Li LM, et al. Neuroimaging evidence of
progressive neuronal loss and dysfunction in temporal lobe epilepsy. Ann Neurol 1999;45:568 –576.
4. Kälviäinen R, Salmenperä T, Partanen K, et al. Recurrent seizures may cause hippocampal damage in temporal lobe epilepsy.
Neurology 1998;50:1377–1382.
5. Trenerry MR, Jack CR, Sharbrough FW, et al. Quantitative
MRI hippocampal volumes: association with onset and duration
of epilepsy, and febrile convulsions in temporal lobectomy patients. Epilepsy Res 1993;15:247–252.
644
Annals of Neurology
Vol 51
No 5
May 2002
6. Nohira V, Lee N, Tien RD, et al. Magnetic resonance imaging
evidence of hippocampal sclerosis in progression: a case report.
Epilepsia 1994;35:1332–1336.
7. VanLandingham KE, Heinz ER, Cavazos JE, Lewis DV. Magnetic resonance imaging evidence of hippocampal injury after
prolonged focal febrile convulsions. Ann Neurol 1998;43:
413– 426.
8. Jackson GD, Chambers BR, Berkovic SF. Hippocampal sclerosis:
development in adult life. Dev Neurosci 1999;21:207–214.
9. O’Brien TJ, So EL, Meyer FB, et al. Progressive hippocampal
atrophy in chronic intractable temporal lobe epilepsy. Ann
Neurol 1999;45:526 –529.
10. Van Paesschen W, Duncan JS, Stevens JM, Connelly A. Longitudinal quantitative hippocampal magnetic resonance imaging
study of adults with newly diagnosed partial seizures: one-year
fellow-up results. Epilepsia 1998;39:633– 639.
11. Commission on Classification and Terminology of the International League against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30:
389 –399.
12. King MA, Newton MR, Jackson GD, et al. Epileptology of the
first-seizure presentation: a clinical, electroencephalographic,
and magnetic resonance imaging study of 300 consecutive patients. Lancet 1998;352:1007–1011.
13. Briellmann RS, Berkovic SF, Jackson GD. Men may be more
vulnerable to seizure associated brain damage. Neurology 2000;
55:1479 –1485.
14. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet
1986;8476:307–310.
15. Briellmann RS, Newton MR, Wellard RM, Jackson GD. Hippocampal sclerosis following brief generalized seizures. Neurology 2001;57:315–317.
16. Aguglia U, Gambardella A, LePiane E, et al. Mild nonlesional temporal lobe epilepsy. A common, unrecognized disorder with onset in adulthood. Can J Neurol Sci 1998;25:
282–286.
17. Briellmann RS, Syngeniotis A, Jackson GD. Comparison of
hippocampal volumetry at 1.5 tesla and at 3 tesla. Epilepsia
2001;42:1021–1024.
18. Bartzokis G, Mintz J, Marx P, et al. Reliability of in vivo volume measures of hippocampus and other brain structures using
MRI. Magn Res Imaging 1993;11:993–1006.
19. Sutula TP. Experimental models of temporal lobe epilepsy: new
insights from the study of kindling and synaptic reorganization.
Epilepsia 1990;31:S45–S54.
20. Hellier JL, Patrylo PR, Buckmaster PS. Recurrent spontaneous
motor seizures after repeated low-dose systemic treatment with
kainate: assessment of a rat model of temporal lobe epilepsy.
Epilepsy Res 1998;31:73– 84.
A Novel D104G Mutation
in the Adenine Nucleotide
Translocator 1 Gene in
Autosomal Dominant
Progressive External
Ophthalmoplegia Patients
with Mitochondrial DNA
with Multiple Deletions
Hirofumi Komaki, MD,1 Toshiyuki Fukazawa, MD,3
Hideki Houzen, MD,3 Kazuto Yoshida, MD,4
Ikuya Nonaka, MD,2 and Yu-ichi Goto, MD1
Autosomal dominant progressive external ophthalmoplegia is a mitochondrial disorder characterized by multiple
large deletions of mitochondrial DNA. A recent study
showed pathogenic heterozygous missense mutations in
the heart/skeletal muscle isoform of the adenine nucleotide translocator 1 gene in autosomal dominant progressive external ophthalmoplegia patients. In one Japanese
autosomal dominant progressive external ophthalmoplegia family, we found a novel A-to-G heterozygous mutation at nucleotide 311 of the adenine nucleotide translocator 1 gene, which segregated with affected individuals
and could not be detected in the genomic DNA sequence
of 120 normal controls. This mutation converted a
highly conserved aspartic acid into a glycine at codon
104. Polymerase chain reaction analysis of single muscle
fibers showed the presence of one type of deletion in each
fiber, suggesting clonal expansion of mitochondrial DNA
with deletions. These findings support the pathogenesis
of the adenine nucleotide translocator 1 gene mutation in
human disease.
Ann Neurol 2002;51:645– 648
From the 1Department of Mental Retardation and Birth Defect Research and 2Department of Ultrastructural Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry,
Tokyo, Japan; 3Department of Neurology, Hokuyukai Hospital,
Hokkaido, Japan; and 4Department of Neurology, Asahikawa Red
Cross Hospital, Hokkaido, Japan.
Received Oct 18, 2001, and in revised form Jan 10, 2002. Accepted
for publication Jan 16, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10172
Address correspondence to Dr Goto, Department of Mental Retardation and Birth Defect Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4-1-1
Ogawahigashi, Kodaira, Tokyo 187-8502, Japan.
E-mail: goto@ncnp.go.jp
Mitochondrial diseases comprise a wide spectrum of
clinical phenotypes as a result of diverse mutations.
Many pathogenic point mutations and rearrangements
(deletions and duplications) in mitochondrial DNA
(mtDNA) have been identified as the causes of the diseases. The representative pattern of the inheritance of
mtDNA mutation is maternal transmission, although
another is Mendelian inheritance, which is suggestive
of a defect in nuclear DNA affecting mtDNA.1
Autosomal dominant progressive external ophthalmoplegia (adPEO) is a mitochondrial disorder characterized clinically by ptosis and progressive muscle
weakness, especially of the external eye muscles. The
disease onset is usually in early adulthood. The typical
histological findings are ragged-red fibers (RRFs) with
focal cytochrome c oxidase deficiency in patients’ skeletal muscle. The patients have multiple large-scale deletions of mtDNA in the affected tissue. The patterns
of inheritance indicate a nuclear gene defect predisposing them to secondary mtDNA deletions,2,3 and there
are three distinct autosomal loci for this disorder on
chromosomes 4q34-35,4,5 10q24,6,7 and 15q22-26.8
Within these loci, mutations of adenine nucleotide
translocator 1 (ANT1), C10orf2 (encoding Twinkle),
and mtDNA polymerase ␥ genes have been associated
with adPEO, respectively.
Human adenine nucleotide translocator exists as
three isoforms that exhibit tissue-specific gene expression.9 Because of its predominant expression in heart
and skeletal muscle, ANT1 is designated as a heart/
muscle isoform.10 Its main function is a translocation
of adenosine diphosphate and adenosine triphosphate
across the inner mitochondrial membrane. Adenine
nucleotide translocator also plays a central role in apoptosis as a central structural element of the permeability transition pore.11
Recently, two heterozygous missense mutations of
ANT1 gene were identified in 5 4q-adPEO families
and 1 sporadic patient.5 We herein report a novel heterozygous mutation of the ANT1 gene in a Japanese
adPEO family.
Patients and Methods
Autosomal Dominant Progressive External
Ophthalmoplegia Pedigree
All affected family members had progressive ptosis and external ophthalmoplegia (Fig 1a). Patient II-7 first noticed
mild bilateral progressive ptosis at age 48, and at age 55 she
began to develop generalized muscle weakness. A physical examination at age 70 showed a very thin woman (weight,
28kg) with severe bilateral ptosis, ophthalmoplegia, generalized muscle weakness, and atrophy. Routine laboratory analyses, creatine kinase, lactate, pyruvate in both blood and spinal fluid, and electrocardiograms were normal. Brain
magnetic resonance imaging showed diffuse abnormal signals
in periventricular white matter. She developed heart failure
at age 70 and died at age 71.
© 2002 Wiley-Liss, Inc.
645
Fig 1. (a) Pedigree of the family with autosomal dominant
progressive external ophthalmoplegia. Solid symbols represents
affected individuals, and open symbols represent unaffected
individuals. Arrows indicate the individuals subjected to mutation analysis. (b) Multiple mitochondrial DNA (mtDNA)
deletions in muscle DNA with Southern blot analysis. Total
DNA from a control subject (C) and Patients II-7 and III-7
were digested by restriction enzyme PvuII to linealize the
mtDNA, and Southern blot analysis was performed with
mtDNA and 18S ribosomal DNA-specific probes.
Patient III-7 was healthy until she started to show bilateral
ptosis at age 33. A physical examination at age 41 showed
bilateral ptosis and ophthalmoplegia. She had no sign of
muscle weakness except for ophthalmoplegia. Routine laboratory analyses, creatine kinase, lactate, pyruvate in blood
and spinal fluid, electrocardiograms, and brain computed tomography were normal. Histological studies on skeletal muscle biopsies for both II-7 and III-7 exhibited RRFs and cytochrome c oxidase deficiency in a few percent of the fibers
and moderate type 2 fiber atrophy. Other affected members
did not have definite symptoms associated with muscular
and neurological diseases, including psychiatric manifestations, pigmentary retinopathy, hypogonadism, and respiratory insufficiency, except for ptosis and external ophthalmoplegia.
We also performed sequence analysis of the ANT1 gene in
32 sporadic progressive external ophthalmoplegia patients
with multiple mtDNA deletions.
CCGAACGCCG; exon 2, AAATCTAGGAAGTGCAAACC
and AGTTTATTTCAGTAGAGGAC; exon 3, CAGTGGCCTCTCTCCCTCCA and TAGCTTTTGCACCCAGGCTC; and exon 4, AGTCTCTTTCCTCCAGCGTT and
AAGCGTGCATTAAGTGGTCT. Amplification products
were directly sequenced as described elsewhere.14
The A-to-G mutation at nucleotide 311 was detected with
the restriction enzyme Alw I (loss of a restriction site; Fig
2b). The Alw I restriction fragments were separated through
4% low melting agarose gel, stained with ethidium bromide,
and photographed.
Single-muscle-fiber PCR was performed to detect
mtDNA with large-scale deletion in each fiber with the
previously described method.15 The muscle fibers were isolated from 30␮m-thick cryostat cross sections stained for
succinate dehydrogenase activity to identify RRFs. DNA
was extracted from the dissected normal and RRFs with
alkaline extraction. Aliquots of the resulting 10␮l solutions
were subdivided into 10 PCR tubes, and PCR was carried
out with 10 sets of primers to detect mtDNA with deletions with the previously described method (Fig 3).16 The
breakpoints of the deletions were detected by direct sequence as described previously.
Results
Southern blot analysis demonstrated full length, and
the additional multiple fragments shorter than expected
indicated that both II-7 and III-7 had mtDNA with
Molecular Studies
Total DNA was extracted by the standard phenol/chloroform
method from muscle specimens (II-7 and III-7) and blood
(II-7, III-1, III-3, III-5, III-7, and III-13). Southern blot
analysis was performed to detect the mtDNA rearrangements
in muscle DNA. Total DNA was digested by restriction enzyme PvuII to linealize the circular mtDNA, and genomic
blots of restriction digests of total DNA were hybridized
with mtDNA and nuclear-encoded 18S ribosomal DNA
gene probes.12,13
The four exons of the ANT1 gene were amplified by polymerase chain reaction (PCR) with the following ANT1specific intronic primers (forward/reverse, 5⬘ to 3⬘): exon1,
TAAGGGGGAGCTGCGGGCCA and ATATAGACAC-
646
Annals of Neurology
Vol 51
No 5
May 2002
Fig 2. (a) Segregation of the mutation in the family. Genomic
DNA were amplified by polymerase chain reaction (PCR)
with a set of specific intronic primers for exon 2, and the
products were digested restriction enzyme Alw I. When the
775bp of the exon 2 PCR products were digested by Alw I,
the following product sizes were generated—320, 149, 132,
89, 62, 13, and 10bp in the wild genome and 409, 149,
132, 62, 13, and 10bp in the mutant genome—and the restriction site was lost. (b) Sequence conservation of ANT1.
D104 (arrows) is strictly conserved among species.
mal ones. With a PCR-restriction fragment length
polymorphism method, this mutation was detected in
4 affected family members (II-7, III-1, III-5, and III-7)
but not in 2 unaffected family members (III-3 and III13) or 120 normal individuals (see Fig 2a). No mutations were detected in 32 sporadic progressive external
ophthalmoplegia patients.
Single-muscle-fiber PCR from nonRRFs showed
only expected fragments (see Fig 3). Single-musclefiber PCR from three different RRFs showed some
fragments shorter than expected. The identification of
the breakpoints of the deletions by the direct sequence
of the fragments confirmed that there was only one
different type of deletion in each RRF. Fragments c
and e had direct repeats flanking the breakpoints.
Fig 3. Single-muscle-fiber polymerase chain reaction (PCR)
analysis of mitochondrial DNA (mtDNA). In the upper part
of the figure, the lines flanked by arrows illustrate the primer
pairs for PCR analysis as previously described.15 The sequences
of the primers are numbered according to the Cambridge
mtDNA sequence. The lanes at the right and left show DNA
size markers (M1, ␭/Hind III, EcoRI double digest marker;
M2, 100bp ladder marker). Primer pairs A and B were used
to amplify wild-type mtDNA. Primer pairs C to M were used
to amplify mtDNA with deletion. (a) PCR analysis from
about 100 muscle fibers shows many amplified fragments indicating multiple mtDNA deletions. (b) Single-muscle-fiber PCR
from nonragged-red fibers (nonRRFs) shows only expected normal fragments in lanes A and B. (c–e) Single-muscle-fiber
PCR from three different ragged-red fibers (RRFs) shows some
fragments shorter than expected. The identification of the
breakpoints of the deletions by direct sequence of the fragments
confirmed only one deletion in each RRF (right, breakpoints
and deletion sizes).
deletions of different sizes showing multiple mtDNA
deletions (see Fig 1b). To quantify the relative amount
of mtDNA to that of nuclear DNA, we used the 18S
ribosomal DNA genes as a probe.13 With the signal as
an internal control, mtDNA depletion was not detected in muscles of the 2 patients.
By the sequence analysis of the ANT1 gene, we
found a novel mutation, A-to-G heterozygote mutation
at nucleotide 311, which converted an aspartic acid
into a glycine at codon 104 in exon 2 (data not
shown). Other sequences were quite identical to nor-
Discussion
We identified a novel A-to-G heterozygous mutation
in exon 2 of the ANT1 gene in one Japanese adPEO
family. We concluded that this mutation was potentially pathogenic for the following reasons. First, affected and unaffected family members demonstrated
segregation with the mutation, which was not detected
in normal individuals. Second, this mutation converted
a strictly evolutionary conserved aspartic acid into a
glycine, that is, an acidic into a nonpolar in a side
chain (see Fig 2b). As mentioned previously, Kaukonen
and colleagues5 identified an ANT1 mutation, A114P,
in 5 of 41 families of adPEO patients and V289M in
1 of 13 sporadic patients. The former mutation was
located near the mutation we found, and the surrounding amino acids were relatively conserved, indicating
that the region is functionally important.
Single-muscle-fiber analysis showed there was one
type of deletion, and the type of deletion varied with
each RRF. These findings suggest each mtDNA with
deletion originated in a mitochondrion and expanded
in a postmiotic cell, as discussed elsewhere.16,17 Therefore, it is conceivable that a deletion is a relatively rare
event. Secondary accumulations of multiple mtDNA
deletions by nuclear gene defects have been observed in
some mitochondrial diseases, including adPEO caused
by mutated ANT1,5 Twinkle (probably functioning as
a mtDNA helicase),7 and polymerase ␥8 genes and mitochondrial neurogastrointestinal encephalomyopathy
caused by a mutated thymidine phosphorylase gene.18
The pathomechanism of multiple mtDNA deletions is
far from understood because the known functions of
each defective gene vary.
This study was supported in part by Health Sciences Research
Grants for Research on Brain Science (to Y.G. and I.N.) and Research Grants for Psychiatric and Neurological Diseases (to Y.G.)
from the Ministry of Health, Labor and Welfare of Japan.
We thank Kumiko Murayama (National Center of Neurology and
Psychiatry) for her technical assistance and Dr James Sylvester
Komaki et al: Novel ANT1 Mutation in adPEO Patients
647
(Hahnemann University, Philadelphia, PA) for providing the ribosomal DNA clone.
18. Nishino I, Spinazzola A, Hirano M. Thymidine phosphorylase
gene mutations in MNGIE, a human mitochondrial disorder.
Science 1999;283:689 – 692.
References
1. Chinnery PF, Turnbull DM. Mitochondrial DNA and disease.
Lancet 1999;354:SI17–SI21.
2. Servidei S, Zeviani M, Manfredi G, et al. Dominantly inherited
mitochondrial myopathy with multiple deletions of mitochondrial DNA: clinical, morphologic, and biochemical studies.
Neurology 1991;41:1053–1059.
3. Suomalainen A, Majander A, Wallin M, et al. Autosomal dominant progressive external ophthalmoplegia with multiple deletions of mtDNA: clinical, biochemical, and molecular genetic
features of the 10q-linked disease. Neurology 1997;48:
1244 –1253.
4. Kaukonen J, Zeviani M, Comi GP, et al. A third locus predisposing to multiple deletions of mtDNA in autosomal dominant
progressive external ophthalmoplegia. Am J Hum Genet 1999;
65:256 –261.
5. Kaukonen J, Juselius JK, Tiranti V, et al. Role of adenine nucleotide translocator 1 in mtDNA maintenance. Science 2000;
289:782–785.
6. Suomalainen A, Kaukonen J, Amati P, et al. An autosomal locus predisposing to deletions of mitochondrial DNA. Nat
Genet 1995;9:146 –151.
7. Spelbrink JN, Li FY, Tiranti V, et al. Human mitochondrial
DNA deletions associated with mutations in the gene encoding
Twinkle, a phage T7 gene 4-like protein localized in mitochondria. Nat Genet 2001;28:223–231.
8. Van Goethem G, Dermaut B, Lofgren A, et al. Mutation of
POLG is associated with progressive external ophthalmoplegia
characterized by mtDNA deletions. Nat Genet 2001;28:
211–212.
9. Stepien G, Torroni A, Chung AB, et al. Differential expression
of adenine nucleotide translocator isoforms in mammalian tissues and during muscle cell differentiation. J Biol Chem 1992;
267:14592–14597.
10. Li K, Warner CK, Hodge JA, et al. A human muscle adenine
nucleotide translocator gene has four exons, is located on chromosome 4, and is differentially expressed. J Biol Chem 1989;
264:13998 –14004.
11. Gross A, McDonnell JM, Korsmeyer SJ. BCL-2 family members and the mitochondria in apoptosis. Genes Dev 1999;13:
1899 –1911.
12. Nishino I, Kobayashi O, Goto Y, et al. A new congenital muscular dystrophy with mitochondrial structural abnormalities.
Muscle Nerve 1998;21:40 – 47.
13. Moraes CT, Shanske S, Tritschler HJ, et al. mtDNA depletion
with variable tissue expression: a novel genetic abnormality in
mitochondrial diseases. Am J Hum Genet 1991;48:492–501.
14. Akanuma J, Muraki K, Komaki H, et al. Two pathogenic
point mutations exist in the authentic mitochondrial genome,
not in the nuclear pseudogene. J Hum Genet 2000;45:
337–341.
15. Moraes CT, Schon EA. Detection and analysis of mitochondrial DNA and RNA in muscle by in situ hybridization
and single-fiber PCR. Methods Enzymol 1996;264:522–540.
16. Moslemi AR, Melberg A, Holme E, Oldfors A. Clonal expansion of mitochondrial DNA with multiple deletions in autosomal dominant progressive external ophthalmoplegia. Ann Neurol 1996;40:707–713.
17. Oldfors A, Moslemi AR, Fyhr IM, et al. Mitochondrial DNA
deletions in muscle fibers in inclusion body myositis. J Neuropathol Exp Neurol 1995;54:581–558.
648
Annals of Neurology
Vol 51
No 5
May 2002
Electrophysiological Findings
in X-Linked Myopathy with
Excessive Autophagy
Satu K. Jääskeläinen, MD, PhD,1 Vern C. Juel, MD,2
Bjarne Udd, MD, PhD,3 Marcello Villanova, MD, PhD,4
Rocco Liguori, MD,5 Berge A. Minassian, MD,6
Björn Falck, MD, PhD,1 Pekka Niemi, MD, PhD,7
and Hannu Kalimo, MD, PhD8
We report electrophysiological features and magnetic resonance imaging muscle findings in 4 patients and 1 female carrier of X-linked myopathy with excessive autophagy. Motor units were polyphasic with high mean
amplitude and normal duration. The thigh muscles were
most severely involved, but myotonic discharges were
abundant in both clinically affected and unaffected muscles. Along with the clinicopathological features, these
electrophysiological findings distinguish X-linked myopathy with excessive autophagy from other limb-girdle myopathies.
Ann Neurol 2002;51:648 – 652
X-linked myopathy with excessive autophagy (XMEA)
is a hereditary myopathy originally described in a Finnish kindred in 1988.1–3 The onset is in childhood,
manifesting as mild muscle weakness. Proximal lower
limb muscles are most severely affected. The progression is slow, and life expectancy is normal. The main
pathological findings are abundant sarcoplasmic vacu-
From the 1Department of Clinical Neurophysiology, University
Central Hospital, Turku, Finland; 2Department of Neurology, University of Virginia School of Medicine, Charlottesville, VA; 3Department of Neurology, Vaasa Central Hospital, Vaasa, Finland;
4
Institute of Neurological Sciences, University of Siena, Siena, Italy;
5
Institute of Clinical Neurology, University of Bologna, Bologna,
Italy; 6Division of Neurology, Department of Pediatrics and Department of Genetics, Hospital for Sick Children and University of
Toronto, Toronto, Canada; 7Department of Radiology, University
Central Hospital, Turku; and 8Department of Pathology, University
Central Hospital, Turku, Finland.
Received Sep 10, 2001, and in revised form Jan 10, 2002. Accepted
for publication Jan 16, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10173
Address correspondence to Dr Jääskeläinen, Department of Clinical
Neurophysiology, Turku University Central Hospital, PL 52, FI20521 Turku, Finland. E-mail: satu.jaaskelainen@tyks.fi
oles, immunopositive for dystrophin and laminin,
which contain debris and lysosomal enzymes.1,2,4 To
our knowledge, XMEA had been identified in 20 families at the time this article was written.5 It is distinct
from another X-linked vacuolar myopathy caused by a
deficiency of lysosome-associated membrane protein-26
known as Danon disease,7 which is associated with cardiomyopathy and mental retardation.8,9 The XMEA
gene locus has been mapped to the most telomeric 10
cM of chromosome Xq28.9 –11
We examined the electrophysiological features and
distribution of XMEA by conducting electroneuromyography examinations and magnetic resonance imaging
(MRI) of the muscles.
Patients and Methods
Clinical, electroneuromyography, and MRI examinations
were performed on 4 male patients with XMEA and 1 female carrier. Demographical and clinical data are summarized in Table 1. Diagnostic criteria for XMEA were confirmed at an international workshop.5 Muscle biopsies were
available from all patients except the younger Finnish patient. All patients exhibited the typical histopathological
findings of XMEA,1,2,4 and genetic analyses demonstrated a
linkage to Xq28. The Finnish family was examined at the
Turku University Central Hospital, the Italian patient was
examined at the Institute of Clinical Neurology of the University of Bologna, and the American patient was examined
at the University of Virginia Medical Center.
Electroneuromyography examinations were performed
with standard commercial electromyography (EMG) equipment (Keypoint, Medtronic, Skovlunde, Denmark; Advantage, Advantage Medical, Waterloo, Ontario, Canada; and
TECA Synergy, Oxford Instruments, Surrey, United Kingdom). The age-, gender-, and length-matched Z-scores of the
electrophysiological variables, based on reference values used
at the Finnish laboratory or reference values from the literature,12 were used in the interpretation of the results.
Motor nerve conduction with F responses were performed
with surface electrodes in the median and peroneal nerves.
Bilateral tibial H-reflexes were also evaluated. Sensory nerve
conduction studies were performed with bipolar surface electrodes or with a near-nerve technique in the superficial peroneal and radial or sural and median nerves. Neuromuscular
transmission was assessed in 2 patients with 2Hz repetitive
nerve stimulation at rest and after 40 seconds of muscle activation. Recordings were made from thenar and extensor
digitorum brevis muscles of the Italian patient and hypothenar and trapezius muscles of the American patient.
Electromyographic examinations of lower and upper limb
and bulbar muscles were performed with concentric needle
electrodes (Technomed, Beek, The Netherlands, Pirouette,
40mm, B64-408, with an active recording surface of
0.07mm2; Maxxim Medical, Clearwater, FL, 30mm, 0.4mm
in diameter, with a 0.021mm2 active recording surface).
Spontaneous activity during rest was evaluated. Twenty motor unit potentials (MUPs) from several muscles (Table 2)
were quantitatively analyzed with regard to the amplitude,
duration, and percentage of polyphasic MUPs. Quantitative
analysis of the interference pattern was performed in the
same muscles with the turns and amplitude technique.
Macro-MUP analysis and single-fiber EMG analysis were
performed on the brachial biceps and lateral vastus muscles
of the Italian patient.
MRI scans of muscle were obtained with a 1.5T clinical
imager (Signa LX, General Electric Medical Systems, Milwaukee, WI). Axial T1-weighted sequences for the assessment of fatty degenerative changes and T2-weighted sequences with maximal fat suppression for the detection of
interstitial edema were performed.
Results
The nerve conduction studies, including H-reflexes and
repetitive nerve stimulation, were normal in all subjects. On needle EMG examination, the most prominent findings were frequent myotonic and highfrequency discharges in all muscles, including clinically
unaffected facial, bulbar, and distal upper limb mus-
Table 1. Clinical and Demographic Data of the Finnish, Italian, and American XMEA Patients and the Finnish Carrier
Age
(yr)
Age at
Onset (yr)
Patient 1
Finnish
34
5–6
1419U/L (⬍285)
Patient 2
Finnish
Patient 3
Italian
16
6
1632U/L (⬍285)
32
10
Patient 4
American
37
3
355U/L (⬍204)
Carrier
Finnish
55
—
84U/L (⬍165)
Patient/Family
Serum CK (upper
normal limit)
376mmol/L (⬍180)
Clinical Weakness
Muscle Wasting
Tendon Reflexes
Proximal lower
limbs, abdominal/
trunk
Proximal lower
limbs
Proximal and distal
lower limbs, abdominal/trunk
Proximal lower and
upper limbs
Thigh, shoulder
girdle
Patellar trace, otherwise normal
Thigh, shoulder
girdle
Thigh
Normal
Thigh, shoulder
girdle,
humeral
None
Trace
Slight difficulty in
walking on the
heels
Trace
Normal
XMEA ⫽ X-linked myopathy with excessive autophagy; CK ⫽ creatine kinase.
© 2002 Wiley-Liss, Inc.
649
Table 2. Results of the Electromyographic Motor Unit Potential Analysis in the XMEA Familiesa
Patient/Muscle
Patient 1
Iliopsoas
Vastus lateralis
Deltoideus
Tibialis anterior
Patient 2
Vastus lateralis
Deltoideus
Tibialis anterior
Patient 3
Vastus lateralis
Deltoideus
Tibialis anterior
Patient 4
Vastus lateralis
Deltoideus
Tibialis anterior
Carrier
Vastus lateralis
Deltoideus
Tibialis anteriorb
Mean MUP Amplitude
Mean MUP Duration
% of Polyphasic MUPs
Myotonic Discharges
1392␮V
2.6; 1338␮V (2362␮V)
1.8; 610␮V (655␮V)
3.9; 1658␮V (1078␮V)
10.5ms
⫺0.1
0.6
0.2
40%
26% (38%)
4% (15%)
0% (10%)
3⫹ (max. 4⫹)
3⫹
2⫹
2⫹
2.3
1.8
2.4
0.4
⫺0.7
1.4
4%
15%
33%
3⫹
2⫹
4⫹
719␮V
909␮V
806␮V
11.9ms
11.3ms
13.9ms
35%
40%
40%
3⫹
3⫹
4⫹
919␮V
1087␮V
1684␮V
11.2ms
9.7ms
12.1ms
75%
40%
35%
3⫹
3⫹
2⫹
10%
5%
23%
0
0
0
1.4
2.1
2.6
1.9
0.9
0.0
Individual Z-scores (normal ⫽ ⫾ 2.0) or the mean values (␮V ⫽ microvolt, ms ⫽ millisecond) are shown; the upper normal limit of
polyphasic MUPs is 15%. For Patient 1, the results of quantitative MUP analysis done 9 years earlier are shown in parentheses for comparison.
b
Findings due to mild bilateral chronic L5-radioculopathy.
a
XMEA ⫽ X-linked myopathy with excessive autophagy; MUP ⫽ motor unit potential.
cles. These spontaneous discharges were so profuse that
only scattered fibrillation potentials could be observed.
The carrier had no abnormal EMG discharges. Results
of the motor unit potential analysis are summarized in
Table 2. Increased occurrence of polyphasic MUPs,
with increased mean amplitude, and early MUP recruitment were found in the proximal lower limb muscles. The MUPs of the distal lower limb and proximal
upper limb muscles were less affected, and hand muscles were normal. The mean duration of the MUPs was
normal in all muscles. Macro-MUP analysis showed
normal median and individual MUP amplitude.
Single-fiber EMG analysis showed increased jitter and
fiber density but no blocking.
The interference pattern analyses were within normal limits in leg muscles and upper limb muscles,
whereas a pattern compatible with myopathy was
found in thigh muscles. These changes were more
prominent in the older patients and most severe in hip
flexor muscles.
Results of an electroneuromyography examination
performed 9 years earlier on the oldest Finnish patient
were available for comparison (see Table 2). The
changes were in opposite directions in the distal and
proximal lower limb muscles, with the distal muscle
showing a definite increase and the proximal muscle
showing a decrease in mean MUP amplitude over
time.
Muscle MRI of the older Finnish brother performed
650
Annals of Neurology
Vol 51
No 5
May 2002
when he was 25 years of age showed mild fatty degeneration in the thigh muscles (Fig, a). Nine years later,
the lesions were significantly more abundant. Fatty degeneration was marked in all thigh muscles, with relative sparing of the rectus femoris, biceps, and semitendinosus muscles (see Fig, b). More diffuse fatty
degeneration was seen in the leg muscles; this was more
apparent in the posterior compartment. In the younger
brother, fatty degeneration was minimal in the thigh
muscles, with the same pattern of relative sparing of
the rectus femoris muscle. Fat suppression sequences
revealed no significantly increased signal. The carrier
had normal findings on MRI of muscle.
Discussion
XMEA is a rare, although probably underdiagnosed,
myopathy. Exceptional histopathological features distinguish XMEA from most other vacuolated myopathies. We now report that XMEA also exhibits highly
characteristic neurophysiological findings.
The electrophysiological hallmarks of XMEA are
abundant myotonic discharges on needle EMG in all
muscles. Despite widespread electrophysiological myotonia, no patient exhibited clinical myotonia. Electrophysiological myotonia may be observed in several myopathies without clinical myotonia, including myotonic
dystrophy type 2, adult-onset acid maltase deficiency
myopathy, myositis, and thyroid myopathy.13 The reason for electrophysiological myotonia in XMEA is un-
Fig. Magnetic resonance imaging transverse section of the midthigh performed at 25 years of age in the older Finnish patient (Patient 1 in Tables 1 and 2). Mild, nonselective fatty
degeneration was observed in all muscle groups (a). Nine years
later, when the patient was 34 years of age, fatty degeneration
was marked in the quadriceps, the adductor group, and the
semimembranosus and gracilis muscles, with relative sparing of
rectus femoris, biceps, and semitendinosus muscles (b).
known. The multilayered folding of the basal lamina
around affected myofibers observed in XMEA could
give rise to dysfunction of various cell membrane structures. Of these, several ion channels are known to be
related to the generation of myotonic discharges in
congenital myotonia (chloride channel), paramyotonia
congenita and hyperkalemic periodic paralysis (adulttype sodium channel), and myotonic dystrophy (abnormalities found in chloride, sodium, and potassium
channel conductance).14,15 Elevated serum creatine kinase values in XMEA may also reflect muscle membrane instability. Another potential explanation is the
high calcium concentration observed just outside the
sarcolemma, which may alter the excitability of the affected myofibers. This could explain the difference
from Danon disease, for which neither myotonic discharges nor calcium deposition has been described.6,7
The quantitative EMG analyses were also distinctive,
with no increase in the amount of small, short MUPs
often observed in myopathy. Instead, the relative mean
MUP amplitudes were increased in clinically affected
muscles. This is compatible with XMEA biopsy findings of marked fiber size variation with hypertrophic
fibers,1 as MUP amplitude correlates to the size of
muscle fibers nearest the recording needle tip.16 MUP
polyphasia was another prominent finding in our series
of XMEA patients. Polyphasic MUPs typically reflect
increased fiber size variation and split fibers,16 and
both of these histological findings are present in
XMEA. The normal MUP duration helps to differentiate XMEA from a neuropathic condition because
normal MUP duration suggests that the number of fibers within a motor unit territory is not increased.
Normal results of the macro-MUP analysis are in accordance with this. Quantitative interference pattern
analyses and early recruitment patterns were also helpful in verifying the myopathic nature of the electrophysiological findings in XMEA.
The history, clinical distribution, and severity of the
muscle weakness and wasting were very similar in our
patients from different parts of the world. The different evaluation techniques revealed a consistent distribution of myopathic involvement. In particular, the
MRI findings were in good agreement with the EMG
findings; both revealed abnormalities, especially in the
anterior thigh and posterior leg muscles. However, because of the slicing level, MRI did not visualize the
electrophysiologically most severely affected iliopsoas
muscle.
Without needle EMG examination and muscle biopsy, XMEA patients may be misdiagnosed; for example, one of our patients was originally diagnosed with
juvenile spinal muscle atrophy. Along with the clinicopathological features, the characteristic electrophysiological findings permit differentiation from other myotonic and vacuolar myopathies, including myotonic
dystrophy types 1 and 2,17 adult-onset acid maltase deficiency myopathy,18 and Danon disease.6 – 8 Adultonset acid maltase deficiency myopathy may be difficult to differentiate from XMEA because vacuoles
containing glycogen19 may be found in XMEA and
myotonic discharges20 sometimes occur in adult-onset
acid maltase deficiency myopathy. Vacuoles in Danon
disease6 – 8 are very similar to those in XMEA, but multiplication of basal lamina and myotonia are not features of Danon disease. XMEA may also resemble
chronic myositis, but the abundant myotonic discharges and muscle histopathology should lead to the
correct diagnosis.
At least 100 different genes map to the XMEA gene
region. We hope that accurate electrodiagnostics will
lead to the discovery of new families, eventually expediting identification of the causative gene defect in
XMEA.
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.
Jääskeläinen et al: Electrophysiological Findings in XMEA
651
3. Online Mendelian inheritance in man, OMIM™. Baltimore,
MD: McKusic-Nathans Institute for Genetic Medicine, Johns
Hopkins University; and Bethesda, MD: National Center for
Biotechnology Information, National Library of Medicine,
2000. Available at: http://www.ncbi.nlm.nih.gov/omim. Accessed June 11, 2001.
4. Louboutin JP, Villanova M, Lucas-Hèron 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. Kalimo H, for the European Neuro Muscular Centre. X-linked
myopathy with excessive autophagy (XMEA) or X-linked vacuolated myopathy (XVM). Paper presented at: 77th ENMC International Workshop; March 10 –11, 2000; Naarden, The
Netherlands. Available at: http://www.enmc.org/workshops/
reports.html. Accessed June 11, 2001.
6. Nishino I, Fu J, Tanji K, et al. Primary LAMP-2 deficiency
causes X-linked vacuolar cardiomyopathy and myopathy
(Danon disease). Nature 2000;24;406:906 –910.
7. Danon MJ, Oh SJ, DiMauro S, et al. Lysosomal glycogen storage disease with normal acid maltase. Neurology 1981;31:
51–57.
8. Muntoni F, Catani G, Mateddu A, et al. Familial cardiomyopathy, mental retardation and myopathy associated with desmintype intermediate filaments. Neuromuscul Disord 1994;4:
233–242.
9. Auranen M, Villanova M, Muntoni F, et al. X-linked vacuolar
myopathies: two separate loci and refined genetic mapping.
Ann Neurol 2000;47:666 – 669.
10. Saviranta P, Lindlöf 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.
11. Villard L, des Portes V, Levy N, et al. Linkage of X-linked
myopathy with excessive autophagy (XMEA) to Xq28. Eur J
Hum Genet 2000;8:125–129.
12. Rosenfalck P, Rosenfalck A. Electromyography, sensory and
motor conduction: findings in normal subjects. Copenhagen:
Laboratory of Clinical Neurophysiology, Rigshospitalet, 1975:
1– 49.
13. Moxley R, Udd B, Ricker K. Proximal myotonic myopathy
(PROMM) and other proximal myotonic syndromes. Neuromuscul Disord 1998;8:508 – 518.
14. Harper PS, Rüdel R. Myotonic dystrophy. In: Engel AG,
Franzini-Armstrong C, eds. Myology. Vol 2. 2nd ed. New
York: McGraw-Hill, 1994:1192–1219.
15. Lehmann-Horn F, Engel AG, Ricker K, Rüdel R. The periodic
paralyses and paramyotonia congenita. In: Engel AG, FranziniArmstrong C, eds. Myology. Vol 2. 2nd ed. New York:
McGraw-Hill, 1994:1303–1334.
16. Stålberg E, Nandedkaar SD, Sanders DB, Falck B. Quantitative
motor unit potential analysis. J Clin Neurophysiol 1996;13:
401– 422.
17. Thornton C. The myotonic dystrophies. Semin Neurol 1999;
19:25–33.
18. De Bleecker JL, Engel AG, Winkelmann JC. Localization of
dystrophin and beta-spectrin in vacuolar myopathies. Am J
Pathol 1993;143:1200 –1208.
19. Amato AA. Acid maltase deficiency and related myopathies.
Neurol Clin 2000;18:151–165.
20. Barohn RJ, Mc Vey AL, DiMauro S. Adult acid maltase deficiency. Muscle Nerve 1993;16:672– 676.
652
© 2002 Wiley-Liss, Inc.
Diffuse Signal Abnormalities
in the Spinal Cord in
Multiple Sclerosis: Direct
Postmortem In Situ
Magnetic Resonance
Imaging Correlated with In
Vitro High-Resolution
Magnetic Resonance
Imaging and Histopathology
Elisabeth Bergers, MD,1 Joost C. J. Bot, MD,1
Paul van der Valk, MD,2 Jonas A. Castelijns, MD,1
Geert J. Lycklama a Nijeholt, MD,1
Wouter Kamphorst, MD,2 Chris H. Polman, MD,3
Erwin L. A. Blezer, MSc,4 Klaas Nicolay, PhD,4
Rivka Ravid, PhD,5 and Frederik Barkhof, MD1
In this study, we compared direct postmortem in situ
(whole-corpse) sagittal spinal cord magnetic resonance
imaging (1.5T) of 7 multiple sclerosis cases with targeted
high-resolution in vitro axial magnetic resonance imaging
(4.7T) and histopathology. On sagittal in situ magnetic
resonance imaging, 1 case had a normal spinal cord, 2
had only focal lesions, 3 had a combination of focal and
diffuse abnormalities, and 1 had only diffuse abnormalities. All spinal cords showed abnormalities on highresolution magnetic resonance imaging and histopathology, confirming the existence of diffuse cord changes as
genuine multiple sclerosis-related abnormalities while
highlighting the limited resolution of in vivo magnetic
resonance imaging.
Ann Neurol 2002;51:652– 656
Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system that affects the brain
and spinal cord, causing demyelination and axonal loss.
From the Departments of 1Radiology, 2Pathology, and 3Neurology,
VU Medical Center, Amsterdam, The Netherlands; 4Image Science
Institute, University of Utrecht, Utrecht, The Netherlands; and
5
Netherlands Brain Bank, Amsterdam, The Netherlands.
Received Aug 21, 2001, and in revised form Jan 10, 2002. Accepted
for publication Jan 16, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10170
Address correspondence to Dr Bergers, MS-MRI Centre, Department of Radiology, Academic Hospital, Vrije Universiteit, De Boelelaan 1117, P.O. Box 7057, 1007 MB Amsterdam, The Netherlands. E-mail: e.bergers@vumc.nl
Research has mainly been focused on brain abnormalities of MS patients. However, correlations between the
number of focal brain lesions as shown by magnetic
resonance imaging (MRI), on the one hand, and clinical course as reflected in the expanded disability status
scale and MS type, on the other hand, are disappointingly low. Now that other parameters that may correlate with clinical disability are being explored, there is
an increasing interest in the role of the spinal cord in
MS, especially because current MRI studies clearly visualize spinal cord abnormalities in a high percentage
of patients.1– 4 Correlations between clinical features
and spinal cord abnormalities have shown promising
results.2,4 – 8 Most studies have focused on focal spinal
cord lesions, defined as sharply delineated, hyperintense abnormalities seen on both proton density (PD)and T2-weighted MRI scans.1,6,9 Recently, Nijeholt
and colleagues2 described a new imaging finding in
MS: diffuse abnormalities of the spinal cord, best appreciated on PD-weighted images as poorly delineated
areas of increased signal intensity (SI) compared with
surrounding cerebrospinal fluid of the spinal cord.
These diffuse cord changes are difficult to detect on
heavily T2-weighted MRI scans because the cord still
has a lower SI than surrounding cerebrospinal fluid.
Despite the important clinical association of diffuse
spinal cord abnormalities with a (primary) progressive
disease phase, the entity of diffuse abnormality cannot
be fully understood without postmortem confirmation.
Postmortem in situ MRI of the complete spinal cord
offers a unique tool for imaging spinal cord abnormalities at the time of worst disability status without artifacts (flow and movement) and allows targeted histopathological examination of different types of MS
lesions in the spinal cord.
The aim of this study was to perform, for what we
believe was the first time, direct postmortem conventional sagittal in situ MRI of the complete spinal cord
to identify focal and diffuse spinal cord abnormalities.
The in situ findings were compared with targeted highresolution in vitro axial MRI and histopathology to
confirm diffuse SI increases in the spinal cord as genuine MS-related pathology and to explore the correlations between these MR modalities in combination
with demyelination on histopathology.
Materials and Methods
This study concerned 7 subjects with clinically diagnosed
MS who had registered while living, having given informed
consent, with the Netherlands Brain Bank (Coordinator, Dr
R. Ravid) for donation of their brains and spinal cords
through autopsy. Age at death ranged from 45 to 83 years,
and MS duration ranged from 14 to 42 years. Three subjects
had a secondary progressive disease course, and 3 had a primary progressive disease course; in 1 subject, the MS type
could not be determined.
Magnetic Resonance Imaging Procedures and Tissue
Handling
All subjects underwent MRI at 1.5T within 6 hours after
death, immediately before autopsy. MRI included sagittal in
situ PD/T2-weighted conventional spin-echo imaging (2200/
20 – 80/1 repetition time/echo time/excitations, 3mm-thick
slices, 1mm pixels) of the spinal cord. At autopsy, 5cm-long
cord specimens (6 cervical, 1 thoracic) were removed, fixed
in formaldehyde, imaged in an axial plane at 4.7T (3000/
15/8, contiguous 1mm-thick slices, 80␮m pixels), lamellated
in the axial direction into 6 equal parts, and embedded in
paraffin. From each block, 5␮m-thick sections were cut and
stained with hematoxylin-eosin and Luxol fast blue (myelin
stain).
Analysis
The appearance of the spinal cord on sagittal in situ MRI
was classified as normal or focally, diffusely, or combined
focally/diffusely abnormal, blinded to high-resolution MR
and histopathology. Focal lesions were defined as areas of
hyperintensity on PD- and T2-weighted MRI scans, sharply
demarcated from surrounding tissue and typically less than 1
vertebra in length. Diffuse abnormalities were defined as areas of mildly increased SI, best observed on PD-weighted
MRI as areas with high SI compared with surrounding cerebrospinal fluid, poorly demarcated, at least several vertebrae
in length, and often involving the entire spinal cord.
High-resolution in vitro axial MRI scans were analyzed for
the presence of abnormalities in lateral, posterior, and anterior columns; abnormalities were classified as intermediate,
high, or combined intermediate/high SI lesions. Intermediate
SI lesions were defined as isointense or hypointense with respect to gray matter, and high SI lesions were defined as
hyperintense with respect to gray matter. In all, analyses of
164 columns of 7 MS cases were included.
Histopathological sections were matched with highresolution axial MRI scans according to the shape and size of
the spinal cord, the configuration of gray matter, and the
presence and configuration of lesions. The histology sections
were then scored by a neuropathologist blinded to the MRI
findings for myelin; each column was classified as normal or
as partially, combined partially/totally, or totally demyelinated.
Associations were evaluated with a ␹2 test.
Results
In all cases, in situ MRI of the total spinal cord yielded
high-quality images without artifacts normally present
through movement, breathing, and blood flow. On
sagittal in situ MRI scans, 1 subject had a normal spinal cord (only a single focal medulla oblongata lesion),
2 subjects had multiple focal lesions (but no diffuse
signal changes), 3 subjects had combined focal/diffuse
abnormalities, and 1 subject had only diffuse abnormalities. Axial high-resolution MRI scans of the excised
specimens showed lesions in all subjects, and 4 of 7
subjects showed lesions in all (lateral and posterior and
anterior) columns. The MS case with only diffusely abnormal cord on in situ MRI was found to have lesions
Bergers et al: Diffuse Signal Abnormalities in the Spinal Cord
653
with intermediate, combined intermediate/high, and
high SI on axial high-resolution images (Fig 1). The
same diversity of lesions was found in MS cases with in
situ combined focal/diffuse (Fig 2) or only focal abnormalities. In the MS case with a normal sagittal MR
scan (a single focal medulla oblongata lesion), the specimen obtained at the fifth thoracic level showed two
lesions with combined intermediate/high SI in the lateral column on axial high-resolution MRI. Both lesions
were confirmed by histology showing partial and total
demyelination.
Including all columns of all subjects, SI on axial
high-resolution images correlated very strongly (␹2 ⫽
59, p ⬍ 0.0001) with myelin score. Totally demyelinated areas corresponded most often to high SI lesions,
combined partial/total demyelination corresponded to
combined intermediate/high SI lesions, partial demyelination corresponded to intermediate SI lesions, and a
normal myelin score corresponded to normal SI.
Correlating in situ spinal cord appearance with SI on
axial high-resolution images and myelin score (as in the
table) showed that for the in situ normal thoracic specimen, the vast majority of columns were normal. For
in situ focally abnormal cord specimens, roughly onethird of the columns were normal with regard to SI
and myelin score, in contrast to the absence of normal
columns for both variables for the in situ diffuse abnormal cord specimens. Furthermore, the in situ diffusely abnormal cord specimens showed the highest
percentage of columns with combined intermediate/
high (54%) and high SI lesions (42%) and total demyelination (58%).
Discussion
The quality of the direct postmortem in situ MRI of
the complete spinal cord was high, without artifacts
and with excellent visibility of focal and diffuse abnormalities. This novel application of MRI in MS cases, in
combination with high-resolution in vitro imaging and
targeted histopathology, allowed confirmation of the
existence of diffuse SI increases, as described by Lycklama a Nijeholt and colleagues,2 as genuine MS-related
pathology. Specimens of in situ diffusely abnormal
cords showed increased SI on axial high-resolution
MRI and demyelination histopathologically. Because
this was a postmortem study, one might consider the
Fig 1. Illustration of sagittal in situ spinal cord magnetic resonance imaging (MRI) of a subject with secondary progressive multiple
sclerosis. A diffuse signal intensity (SI) increase was observed on a proton density MRI (A and C; arrows), but abnormal SI was
not evident on heavily T2-weighted MRI (B and D) in combination with corresponding high-resolution axial MRI of the cervical
specimen at the level of C5 (E–H) and a histological section (G) stained with Luxol fast blue (I). Note that there are no focal lesions on the sagittal in situ image or in the cervical region. The axial high-resolution MRI shows abnormal SI of almost the entire
axial cord image. Lesions with mainly high SI are present in both lateral columns and in posterior and anterior columns with distortion of the gray matter. All areas with abnormal SI were confirmed on histopathology, showing partially or totally demyelinated
areas.
654
Annals of Neurology
Vol 51
No 5
May 2002
Fig 2. Illustration of sagittal in situ spinal cord proton density (PD)- and T2-weighted magnetic resonance imaging (MRI; A and
B) of a subject with secondary progressive multiple sclerosis with an in situ combined focally/diffusely abnormal cord in combination
with high-resolution axial MRI at the level of C5 (C–F) of the cervical specimen and a histological section (F) stained with Luxol
fast blue (G). On the PD-weighted image (A), a diffusely increased signal intensity (SI) throughout the entire cord was observed
(open arrows) in combination with some focal lesions (solid arrows). Focal lesions were also observed on the in situ heavily T2weighted sagittal MRI as sharply demarcated areas (B; solid arrows). The diffuse SI increase throughout the entire cord was very
difficult to observe on the T2-weighted MRI. The axial high-resolution MRI (C–F) shows clear lesions with high and intermediate
increased SI in both lateral columns and in posterior and anterior columns with distortion of the gray matter. The lesions were
confirmed by the histology section as areas with partial or total demyelination.
Table. Results of SI on Axial High-Resolution MRI and Myelin Score on Histological Sections Classified for In Situ Normal,
Focally Abnormal, Combined Focally/Diffusely Abnormal, and Diffusely Abnormal Spinal Cordsa
SI on Axial In Vitro High-Resolution MRI
In Situ
Appearance on
Sagittal MRI
Normalb
Focally abnormal
Diffusely abnormal
Combined focally/
diffusely
Myelin Score on Histology Section
Normal
Intermediate
Intermediate/
High
High
Normal
Partially
Demyelinated
Partially/Totally
Demyelinated
Totally
Demyelinated
18 (75%)
14 (29%)
0 (0%)
3 (13%)
8 (17%)
1 (4%)
0 (0%)
21 (44%)
33 (54%)
3 (13%)
5 (10%)
10 (42%)
20 (83%)
13 (27%)
0 (0%)
1 (4%)
14 (29%)
8 (33%)
1 (4%)
14 (29%)
2 (8%)
2 (8%)
7 (15%)
14 (58%)
11 (16%)
14 (21%)
22 (32%)
21 (31%)
12 (31%)
17 (25%)
20 (29%)
19 (28%)
Diffusely vs focally abnormal: ␹2 ⫽ 17, p ⫽ 0.001, for SI, and ␹2 ⫽ 20, p ⬍ 0.0001, for myelin score; diffusely vs combined focally/diffusely
abnormal: ␹2 ⫽ 10, p ⫽ 0.02, for SI, and ␹2 ⫽ 13, p ⫽ 0.006, for myelin score; focally vs combined focally/diffusely abnormal: ␹2 ⫽ 9, p ⫽
0.03, for SI, and ␹2 ⫽ 4, p ⫽ 0.3, for myelin score.
b
The in situ appearance of the spinal cord at the location of the removed specimen was normal; in the medulla oblongata, a single focal lesion
was found.
a
SI ⫽ signal intensity; MRI ⫽ magnetic resonance imaging.
possibility that a postmortem artifact was related to the
diffusely increased SI on in situ MRI. However, because this study included two in situ focally abnormal
cord MRI examinations with no diffuse abnormalities
(normal cord areas in between lesions) and 1 case with
a completely normal cord (only a single focal lesion in
the medulla oblongata), it is highly unlikely that a
postmortem artifact was present.
Bergers et al: Diffuse Signal Abnormalities in the Spinal Cord
655
The differences between in situ diffuse cord abnormalities and focal cord lesions demonstrated by highresolution axial MRI and myelin score on histopathological sections were not absolute, but quantitative: in
situ diffusely abnormal cords less often had normal columns, and lesions were slightly more severe. Therefore,
in situ diffuse abnormalities (eg, focal cord lesions) appeared to arise primarily in the spinal cord and were
not secondary to brain lesions. Correlations between
diffuse abnormalities on sagittal MRI and intermediate
SI lesions on axial high-resolution MRI were suggested
by Lycklama a Nijeholt and colleagues10; in this study,
the number of subjects was too low to confirm this
association.
The results of this study show that in situ sagittal
MRI underestimates the number and extensiveness of
spinal cord lesions and underlying focal components of
diffuse abnormalities. The fact that in vivo sagittal
MRI is also hampered by artifacts (movement and
flow) indicates that spinal cord damage by MS is even
more underrated than our postmortem data suggest.
Currently used in vivo sagittal and axial MRI techniques clearly lack sufficient spatial or contrast resolution or both and a sufficient signal-to-noise ratio.
In conclusion, direct postmortem imaging allowed
the confirmation of diffuse spinal cord abnormalities as
genuine MS-related spinal pathology in our study. The
extensiveness of cord abnormalities detected on highresolution MRI and histopathology was underestimated
by sagittal images, indicating that in vivo MRI lacks a
sufficient signal-to-noise ratio as well as sufficient spatial or contrast resolution or both.
References
Annals of Neurology
Clinical Variability in 3Hydroxy-2-MethylbutyrylCoA Dehydrogenase
Deficiency
Regina Ensenauer, MD,1 Helmut Niederhoff, MD,1
Jos P. N. Ruiter,2 Ronald J. A. Wanders, PhD,2
K. Otfried Schwab, MD,1 Matthias Brandis, MD,1
and Willy Lehnert, PhD1
We report the identification of two new 7-year-old patients with 3-hydroxy-2-methylbutyryl-CoA dehydrogenase deficiency, a recently described inborn error of isoleucine metabolism. The defect is localized one step
above 3-ketothiolase, resulting in a urinary metabolite
pattern similar to that seen for deficiency of the latter.
One patient has progressive neurodegenerative symptoms, whereas the clinical phenotype of the other patient
is characterized by psychomotor retardation without loss
of developmental milestones. A short-term biochemical
response to an isoleucine-restricted diet was observed in
both children.
Ann Neurol 2002;51:656 – 659
1. Kidd D, Thorpe JW, Kendall BE, et al. MRI dynamics of brain
and spinal cord in progressive multiple sclerosis. J Neurol Neurosurg Psychiatry 1996;60:15–19.
2. Nijeholt GJ, Barkhof F, Scheltens P, et al. MR of the spinal
cord in multiple sclerosis: relation to clinical subtype and disability. AJNR 1997;18:1041–1048.
3. Nijeholt GJ, Van Walderveen MAA, Castelijns JA, et al. Brain
and spinal cord abnormalities in multiple sclerosis. Correlation
between MRI parameters, clinical subtypes and symptoms.
Brain 1998;121:687– 697.
4. Thielen KR, Miller GM. Multiple sclerosis of the spinal cord:
magnetic resonance appearance. J Comp Assist Tomogr 1996;
20:434 – 438.
5. Stevenson VL, Miller DH, Rovaris M, et al. Primary and transitional progressive MS: a clinical and MRI cross-sectional
study. Neurology 1999;52:839 – 845.
6. Thorpe JW, Kidd D, Moseley IF, et al. Spinal MRI in patients
with suspected multiple sclerosis and negative brain MRI. Brain
1996;119:709 –714.
7. Filippi M, Campi A, Colombo B, et al. A spinal cord MRI
study of benign and secondary progressive multiple sclerosis.
J Neurol 1996;243:502–505.
8. Filippi M, Campi A, Martinelli V, et al. Brain and spinal cord
MR in benign multiple sclerosis: a follow-up study. J Neurol
Sci 1996;143:143–149.
656
9. Tartaglino LM, Friedman DP, Flanders AE, et al. Multiple sclerosis in the spinal cord: MR appearance and correlation
with clinical parameters. Radiology 1995;195:725–732.
10. Nijeholt GJ, Bergers E, et al. Postmortem high resolution MRI
of the spinal cord in multiple sclerosis: a correlative study with
conventional MRI, histopathology and clinical phenotype.
Brain 2001;124:154 –166.
Vol 51
No 5
May 2002
3-Hydroxy-2-methylbutyryl-CoA is an intermediary
product in the degradation pathway of isoleucine. Rat
liver 3-hydroxy-2-methylbutyryl-CoA dehydrogenase
(3H2MBD, EC 1.1.1.178) has been purified and characterized.1 This enzyme dehydrogenates 3-hydroxy-2methylbutyryl-CoA into 2-methylacetoacetyl-CoA one
step upstream of 3-ketothiolase. Although several pa-
From the 1Metabolic Unit, University Children’s Hospital,
Freiburg, Germany, and 2Laboratory of Genetic Metabolic Diseases,
Department of Pediatrics and Clinical Chemistry, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands.
Received Jun 20, 2001, and in revised form Jan 8, 2002. Accepted
for publication Jan 16, 2002.
Published online Apr 23, 2002 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.10169
Address correspondence to Dr Ensenauer, Department of Medical
Genetics, Mayo Clinic, 200 First Street SW, Rochester, MN 55905.
E-mail: ensenauer.regina@mayo.edu
tients have been reported with a urinary metabolite
pattern indicative of 3-ketothiolase deficiency and normal enzyme activity,2,3 3H2MBD deficiency was proven
only recently in a patient presenting with progressive
neurodegeneration in early childhood.4 His clinical
symptoms differ markedly from those of patients with
3-ketothiolase deficiency, who experience recurrent episodes of ketoacidosis and who usually develop normally.5 The pathogenesis of the neurodegenerative processes
in 3H2MBD deficiency has not been understood so far.
Here we report two additional patients with inherited 3H2MBD deficiency, both 7 years of age, who
present with phenotypes differing considerably from
each other, thereby widening the clinical spectrum of
this newly recognized metabolic disease.
Patients and Methods
Patient 1
Born in 1993, this girl was the third child of nonconsanguineous and healthy parents; she was delivered after 38
weeks of gestation without complications. The birth weight,
length, and head circumference were close to the 50th percentile. Growth was normal for age. She could, however, not
sit alone before the age of 15 months and could not walk
alone until 24 months. At that age, she spoke only one word.
An episode of lethargy and enteritis occurred after a proteinrich and fatty meal. Psychomotor and speech development
are delayed despite the slow but continuous acquisition of
developmental milestones. At 7 years of age, funduscopy and
magnetic resonance imaging of the brain were normal, with
the exception of mild frontoparietal cortical atrophy. An
electroencephalogram showed diffuse ␤-activity without signs
of hypersynchronous activity.
The initial diagnostic workup showed elevated excretions of 3-hydroxy-2-methylbutyric acid (3H2MBA),
2-ethylhydracrylic acid (2EHA), and tiglylglycine (TG; Table
1), indicating a possible defect of isoleucine catabolism. After
informed parental consent was obtained, an oral isoleucine
loading test (100mg/kg of body weight) was performed,
demonstrating a marked increase in urinary 3H2MBA and
TG excretion (Fig). 2-Methylacetoacetic acid was not detected. Over a period of 5 years, the biochemical phenotype
remained constant.
In cultured fibroblasts, the activity of 3-ketothiolase was
normal, as was 3H2MBD activity measured initially in a research laboratory. After the first report of a patient with
3H2MBD deficiency,4 fibroblasts were reexamined in the
same laboratory that identified the enzyme defect in the patient described by Zschocke and colleagues.4 The results
shown in Table 1 are consistent with a diagnosis of
3H2MBD deficiency.
Table 1. Relevant Clinical and Biochemical Data of Patients 1 and 2 as Compared with the Patient Described
by Zschocke and Colleagues
Patient 1 (girl)
Patient 2 (boy)
Index Patient (boy)
Age at diagnosis (yr)
Metabolic decompensation
Clinical course
7
Possibly one mild episode
7
Not occurred
2
2nd day of life
Psychomotor retardation
Developmental regression, dystonia, blindness, epilepsy
Ocular fundus
Normal
Brain MRI
Discrete external volume
reduction frontoparietally
Normal
Pigmentary retinal degeneration, optic atrophy
Profound atrophy, occipital
infarctions
Developmental regression, choreoathetosis, near blindness,
epilepsy
Nonpigmentary retinal
degeneration
Slight frontotemporal
atrophy
Echocardiography
Urinary organic acids
Blood lactate (controls 1–2mmol/L)
CSF lactate (controls
⬍1.7mmol/L)
Acylcarnitine profile
3H2MBD activity
(controls 7.10 ⫾
0.83nmol/min ⫻
mg)
1 3H2MBA, 1 2EHA,
1 TG
1.7
Concentric hypertrophy of left
ventricle
11 3H2MBA, 11 2EHA,
11 TG, 11 3HIBA
1.8–4
Not reported
11 3H2MBA,
11 TG
2.1
Not measured
1.96
2.1
C5:1-carnitine (1)
1.01 ⫾ 0.50
C5:1-carnitine (1)
0.57 ⫾ 0.17
C5:1-carnitine (1)
0.39 ⫾ 0.19
3H2MBD activity was measured in a reverse mode following reduction of 2-methylacetoacetyl-CoA to 3-hydroxy-2-methylbutyryl-CoA by
decrease of NADH absorbance at 340nm. Residual enzyme activity obtained by this method was higher in the index patient than previously
reported.4
3H2MBA ⫽ 3-hydroxy-2-methylbutyric acid; 2EHA ⫽ 2-ethylhydracrylic acid; 3HIBA ⫽ 3-hydroxyisobutyric acid; TG ⫽ tiglylglycine; CSF
⫽ cerebrospinal fluid; 3H2MBD ⫽ 3-hydroxy-2-methylbutyryl-CoA dehydrogenase.
© 2002 Wiley-Liss, Inc.
657
Fig. Oral isoleucine challenge: urinary concentrations of
3-hydroxy-2-methylbutyric acid (3H2MBA) and tiglylglycine
(TG) in Patient 1 after the administration of 100mg of isoleucine/kg of body weight. Urine was collected before and at 0
to 6, 6 to 12, 12 to 18, and 18 to 24 hours after isoleucine
loading. In a control person, an only slightly increased excretion of both metabolites was measured in the urine collected
within the first 6 hours after isoleucine loading (48mmol of
3H2MBA/mol of creatinine; 34mmol of TG of creatinine),
declining gradually to baseline concentrations thereafter (7 and
9mmol/mol of creatinine, respectively).
Regardless of the first set of normal enzyme activities, the
natural protein intake was reduced to 1 to 1.5g/kg each day
as a precaution, a regimen similar to the treatment of patients with 3-ketothiolase deficiency. After the diagnosis was
made, isoleucine intake was further reduced to 30mg/kg each
day, equivalent to a daily natural protein intake of 0.7g/kg. An
isoleucine-free amino acid mixture (ISO-AM 2; SHS, Heilbronn, Germany) was added to increase the total daily protein
intake to 1.4g/kg each day. Within several weeks, a reduction
of 3H2MBA and TG excretion was observed (Table 2). Clinically, the patient appears to be more alert and active.
Patient 2
This boy, born in 1993, was the product of an uncomplicated 38-week gestation, the first child of nonconsanguineous and healthy parents. The birth weight, length, and head
circumference were on the 75th, 90th, and 25th percentiles,
respectively. Developmental milestones were reached adequately during the 1st year of life. At 14 months of age, the
parents noticed developmental regression, including loss of
vocalization and deterioration of visual function. His appetite
decreased, and he started to vomit frequently.
At that time, the weight and head circumference were on
the 3rd percentile, and the length was on the 50th percentile.
Ataxia was evident. He could stand only with support, and
targeted grasping was impossible. He was not able to focus
on objects or follow them. Ophthalmologic investigation
showed a pigmentary retinopathy-like fundus appearance and
normal optic discs. Brain magnetic resonance imaging, electroencephalograms, auditory-evoked potentials, and nerve
conduction studies were normal. Urinary organic acid analysis showed mild lactic aciduria and elevated excretions of
3H2MBA, 3-hydroxyisobutyric acid, 2EHA and TG, which
were suggestive of a defect in the isoleucine degradation
pathway (see Table 1). 2-Methylacetoacetate was not detected. Informed consent was obtained for all investigations.
Protein restriction to 1g/kg of body weight per day was initiated. In cultured fibroblasts, normal 3-ketothiolase and
3H2MBD activities were obtained.
Over the following years, the retinal and neurological manifestations were progressive. Funduscopy showed steadily increasing signs of complex pigmentary retinal degeneration and
optic atrophy bilaterally. Progressive loss of motor and mental
skills became evident. At the age of 41⁄4 years, brain magnetic
resonance imaging showed severe cerebral atrophy and occipital infarctions. Lactate levels intermittently increased (see Table 1), and Leigh syndrome was suspected. In muscle, a mild
reduction of respiratory chain complex I activity was measured
(0.09U/g, normal 0.35 ⫾ 0.14; 0.01U/U relative to citrate
synthase, normal 0.023 ⫾ 0.009). Protein restriction was discontinued. At the age of 5 years, 8 months, he developed epileptic seizures that became difficult to control.
Because of the persistent urinary metabolite pattern, a reevaluation of enzyme activity in fibroblasts was performed at
the age of 6 years, 10 months, yielding now clear-cut
3H2MBD deficiency (see Table 1).
The boy is now 7 years old and suffers from severe mental
and physical retardation with rarely any spontaneous movements and profound muscle hypotonia. He reacts poorly to
external stimuli. Echocardiographic studies showed marked
concentric hypertrophy of the left ventricle with still normal
systolic function. After only a few weeks of an isoleucinerestricted diet (30mg/kg each day), no changes in symptoms
or behavior were noticed. The amount of urinary 3H2MBA,
however, decreased (see Table 2).
Urine organic acid analyses were performed by gas
Table 2. Urinary Concentrations of 3H2MBA and TG in Patients 1 and 2 before and Several Weeks after Beginning of an
Isoleucine-Restricted Dietary Regimen
Patient 1
Untreated
Treated
p
Patient 2
3H2MBA
TG
3H2MBA
TG
21–110 (84)
15–48 (28)
0.002
23–78 (53)
18–62 (41)
0.02
37–427 (85)
20–53 (30)
0.007
20–316 (54)
49–125 (87)
Not significant
Data are presented as ranges; medians are given in parentheses. Unit: mmol/mol creatinine.
3H2MBA ⫽ 3-hydroxy-2-methylbutyric acid; TG ⫽ tiglylglycine.
658
Annals of Neurology
Vol 51
No 5
May 2002
chromatography-mass spectrometry, and plasma acylcarnitine
analyses were performed by tandem mass spectrometry. The
assay of 3H2MBD activity in cultured fibroblasts was first
done in a research laboratory and was later repeated in the
Laboratory for Genetic Metabolic Disease at the University
of Amsterdam, The Netherlands. This assay was performed
spectrophotometrically at 37°C in the reverse direction after
the reduction of 2-methylacetoacetyl-CoA to 3-hydroxy-2methylbutyryl-CoA with a decrease in NADH absorbance at
340nm. The reaction medium contained the following components: 100mM MES buffer (pH 6.5), 0.1mM NADH,
0.1% (w/v) Triton X-100, 0.1mg of fibroblast protein/ml,
and 100␮M 2-methylacetoacetyl-CoA.
Discussion
After the first description of an affected boy,4 two new
patients with 3H2MBD deficiency have been diagnosed and are presented here. The clinical phenotype
of each patient differs considerably. Despite mental retardation and developmental delay, the disease in the
girl (Patient 1) is not characterized by progressive neurodegeneration, which, however, is a prominent finding in the boy (Patient 2) and also in the index patient.
No loss of milestones has occurred, her vision is not
impaired, and there are no signs of retinal degeneration
or optic atrophy. Although the boy developed profound morphological brain damage within the first 4
years of life, magnetic resonance imaging studies of the
girl at 7 years of age showed an only discrete external
volume reduction frontoparietally. No other organ pathology, particularly with respect to the heart muscle,
has been documented in her so far. This is in contrast
to the finding of progressive concentric left ventricular
hypertrophy in the boy, a finding that has not been
reported for the index patient.
The cause of the considerable phenotypical variability
in 3H2MBD deficiency is unknown. The girl’s higher
residual enzyme activity might explain her milder clinical picture. It is not understood, however, how the
quantitative differences in residual enzyme activity lead
to such qualitative differences in clinical phenotype.
Whether future molecular genetic analysis, once the
gene has been identified, will explain the marked variability in clinical symptoms remains an open question.
Toxic and disease-causing metabolites have been
identified in neither 3H2MBD nor 3-ketothiolase deficiency.6 In the latter disease, even higher amounts of
3H2MBA and TG are usually observed. It may, however, be that 3H2MBD is not only involved in isoleucine breakdown but also in the degradation of other
2-methyl branched-chain fatty acids. The accumulation
of such compounds in the central nervous system may
explain why 3H2MBD deficiency results in a more severe phenotype than a deficiency of 3-ketothiolase.
Several patients have been described in the literature
who had variably increased urinary excretions of
3H2MBA and TG2,3 but had normal 3-ketothiolase activity. These patients should be reevaluated. In the two
patients described here, diagnosis was delayed for years
because of the initial findings of normal 3H2MBD activity measured in a research laboratory. Additional diagnostic tools such as in vitro oxidation of 14C-labeled
isoleucine in cultured fibroblasts7 were not available at
that time.
The mildly reduced activity of respiratory chain
complex I measured in muscle of Patient 2 is thought
to be secondary to the interference of accumulating
metabolites with mitochondrial energy production.
Unlike the index patient, who presented with extremely elevated lactate levels, hyperammonemia, hypoglycemia, and metabolic acidosis on the second day
of life, severe episodes of metabolic decompensation
have never occurred in our patients.
Both children responded biochemically to isoleucine
restriction. Although urinary 3H2MBA decreased significantly in both patients, there was less reduction of
TG in the girl, and there was no reduction in the boy.
At present, it is still too early to draw any conclusions
about clinical effects of the diet. In the much younger
index patient for whom treatment was started at the
age of 2 years, restlessness improved and the frequency
of seizures decreased on the same dietary regimen over
a period of 7 months. Further studies of affected patients will show if treatment initiated at earlier stages
will protect them from deterioration and improve the
outcome of the disease.
References
1. Luo MJ, Mao LF, Schulz H. Short-chain 3-hydroxy-2methylacyl-CoA dehydrogenase from rat liver: purification and
characterization of a novel enzyme of isoleucine metabolism.
Arch Biochem Biophys 1995;321:214 –220.
2. Iden P, Middleton B, Robinson BH, et al. 3-Oxothiolase activities and [14C]-2-methylbutanoic acid incorporation in cultured
fibroblasts from 13 cases of suspected 3-oxothiolase deficiency.
Pediatr Res 1990;28:518 –522.
3. Aramaki S, Lehotay D, Sweetman L, et al. Urinary excretion of
2-methylacetoacetate, 2-methyl-3-hydroxybutyrate and tiglylglycine after isoleucine loading in the diagnosis of
2-methylacetoacetyl-CoA thiolase deficiency. J Inherited Metab
Dis 1991;14:63–74.
4. Zschocke J, Ruiter JPN, Brand J, et al. Progressive infantile neurodegeneration caused by 2-methyl-3-hydroxybutyryl-CoA dehydrogenase deficiency: a novel inborn error of branched-chain fatty
acid and isoleucine metabolism. Pediatr Res 2000;48:852– 855.
5. Fukao T, Scriver CR, Kondo N. The clinical phenotype and outcome of mitochondrial acetoacetyl-CoA thiolase deficiency (betaketothiolase or T2 deficiency) in 26 enzymatically proved and
mutation-defined patients. Mol Genet Metab 2001;172:109 –114.
6. Ozand PT, Rashed M, Gascon GG, et al. 3-Ketothiolase
deficiency: a review and four new patients with neurologic symptoms. Brain Dev 1994;16:38 – 45.
7. Gibson KM, Burlingame TG, Hogema B, et al. 2-Methylbutyrylconenzyme A dehydrogenase deficiency: a new inborn error of
L-isoleucine metabolism. Pediatr Res 2000;47:830 – 833.
Ensenauer et al: Dehydrogenase Deficiency
659
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 762 Кб
Теги
autosomal, d104g, progressive, patients, nucleotide, translocator, dominantly, adenine, dna, ophthalmoplegia, external, anovel, mutation, genes, multiple, deletion, mitochondria
1/--страниц
Пожаловаться на содержимое документа