вход по аккаунту


Congenital muscular dystrophy with primary laminin 2 (merosin) deficiency presenting as inflammatory myopathy.

код для вставкиСкачать
Larnrmn a2
Elena Pegoraro, MD,* Pedro Mancias, M D , t Steven H. Swerdlow, MD,$ Radmila B. Raikow, PhD,$
Carlos Garcia, MD,$ Harold Marks, MD,S Thomas Crawford, MD,# Virginia Carver, PhD,**
Brad Di Cianno,* and Eric I). Hoffman, PhD*
Ten laminin 1x2-deficient patients were identified by both immunofluorescence and immunoblotting (30% of congenital
muscular dystrophy patients tested). Three of the laminin 1x2-deficient patients were carrying a diagnosis of infantile
polymyositis prior to immunostaining studies. The clinical features in the 10 merosin-deficient patients were homogeneous, with severe floppiness at birth, delay in achievement of motor milestones, and magnetic resonance imaging
findings of white matter changes with normal intelligence. The 10-kb laminin 1x2-coding sequence was screened for
causative mutations by reverse transcriptase-polymerase chain reaction/single-strandedconformational polymorphism
analysis in muscle biopsy specimens from 5 patients, followed by automatic sequencing of aberrant conformers. Clear
loss-of-function deletion mutations were identified in both alleles of 1 patient. Muscle histopathology in this patient
showed a striking inflammatory infiltrate of T cells and B cells. Reexamination of biopsy specimens from other laminin
a2-deficient patients showed minor signs of inflammation in each. Based on these findings and the histological and
clinical picture suggesting failure of muscle regeneration, a pathogenesis model for this major subset of congenital
muscular dystrophy is proposed. Our data show that muscle histopathology showing a neonatal inflammatory process
should be considered consistent with congenital muscular dystrophy.
Pegoraro E, Mancias P, Swerdlow SH, Raikow RB, Garcia C, Marks H, Crawford T, Carver V, Di Cianno B,
Hoffman EP. Congenital muscular dystrophy with primary laminin a 2 (merosin) deficiency
presenting as inflammatory myopathy. Ann Neurol 1996;40:782-79 1
Congenital muscular dystrophy (CMD) is a muscle
disorder with symptoms of floppiness and profound
muscle weakness a t birth. Contractures or joint deformities, and a variable involvement of the central nervous system (CNS) are frequently present [ l , 21. While
the muscle involvement is similar in the different forms
of CMD, at least four autosomal recessive forms can
be identified based on differential involvement of the
CNS. Fukuyama-type congenital muscular dystrophy
(FCMD) [3, 41 (chromosome 9q31-q33) [ 5 ] , muscleeye-brain disease (MEB) or Santavuori disease [6, 71,
and Walker-Warburg-type CMD [S] show dramatic
clinical signs of CNS involvement. The last form, occidental CMD, is characterized by the exclusive involvement of the skeletal muscle in the absence of clinically
evident CNS involvement. In the early 1980s a number of reports suggested that an additional subset of
CMD patients had an underlying inflammatory process
as the cause of their disease [9-12]. More recently, the
clinical condition characterized by antenatal or neonatal onset of muscle weakness, with or without arthrogryposis or CNS involvement, and inflammatory findings in muscle biopsy specimens has been called
congenital inflammatory myopathy [ 131.
A common feature of these clinically heterogeneous
CMD syndromes is a muscle histopathology characterized by a marked increase in connective tissue, suggesting that abnormalities of one of the components
of the extracellular matrix might be involved in the
pathogenesis of the disease [ 141. Recently, a large oligo-
From the *Departments of Molecular Genetics and Biochemistry,
Human Genetics, Neurology, and Pediatrics, University of Pittsburgh, School of Medicine, Pittsburgh; the tnepartment of Pediatrics, University of Arkansas Medical Sciences, Little Rock, AK; the
$Department of Pathology, University of Pittsburgh Medical Center, Pittsbiirgh, PA; rhe $School of Medicine in New Orleans, Lo&
siana Srate CJniversity Medical Center, New Orleans, L.A; the SNeurology Departnxnt, A. I. dul’ont Institute, Wilmington, DE; the
#Department of Neurology, John Hopkins CJniversity, Baltimore,
MU; the **University of Miami, Mixni, FL.
Received Feb 6, 1936, and in revised form Jun 6. Accepted for
publication Jun 12, 1936.
Address reprint requests to Dr Hoffman, Department of Molecular
Genetics and Biochemistry, Biomedical Science Tower W12 11,
University of Pittsburgh, School of Medicine, Pittsburgh, PA
Copyright 0 1996 by the American Neurological Association
meric complex of plasma membrane glycoproteins,
dystrophin-associated glycoproteins (DAGs) [15-181,
was shown to provide o n e of many probable connections between the intracellular membrane cytoskeleton
(including dystrophin) a n d the basal lamina [ 19, 201.
Laminin is a major component of all basal laminae a n d
is a cross-shaped heterotrimer of o n e heavy chain a n d
two light chains. The laminin a 2 chain (merosin) is a
400-kd isoform found in muscle, in some areas of the
CNS, a n d in the Schwann cells in the peripheral nervous system (PNS) [21]. T h i s heavy-chain isoform of
the laminin heterotrimer is posttranslationally cleaved
into 300-kd a n d 80-kd components that remain associated by disulfide bonds [21]. The two light chains
(220 kd) a n d yl (200 kd), are
found in muscle,
n o t tissue-specific [22, 231. T h e complete merosin sequence a n d its chromosome localization (6q22-q23)
are known [24].
T o m i a n d coauthors [25] recently reported a series
of patients with CMD of the occidental type (nonFCMD type) in which a component of the basal lamina, laminin a 2 (merosin), was completely absent from
muscle biopsy specimens. Later, linkage analysis a n d
homozygosity mapping in 4 rnerosin-negative families
mapped the CMD gene o n a 16-cM region of chromosome 6 q 2 , where the merosin gene was located, suggesting that merosin deficiency is likely to be the primary defect in these families [26]. A recent mutation
study of the laminin a 2 gene showed a splice site and
a nonsense mutation in 2 families, presumably resulting in a truncated laminin a 2 protein [27]; however, n o biopsy studies were reported.
Here w e study histological, biochemical, molecular,
a n d clinical features of a series of congenital dystrophy
patients. W e found merosin deficiency in about 30%
of biopsy samples studied, and identified three novel
merosin gene mutations. A n inflammatory myopathy
was an early histological presentation in 30% (3/10)
of the patients studied. Interestingly, the inflammatory
response was transient in these 3 patients a n d n o t present in subsequent muscle biopsy specimens.
Materials and Methods
Patient SeLection
O u r 2,400 muscle biopsy database was screened for patients
meeting one or more of the following criteria: (1) early onset
of muscle weakness; (2) reported contractures, joint deformities, arthrogryposis, hip dislocation, or scoliosis since birth;
(3) association of muscle weakness and CNS involvement
(magnetic resonance imaging [MRI] abnormalities, macro/
microcephaly, eye abnormalities, hydrocephalus, seizures, abnormal electroencephalogram [EEG]); (4) early-onset muscle histopathology consistent with a dystrophic process; or
(5) diagnosis of C M D by the referring physician. All patients
showed normal immunofluorescence and/or immunoblotting for dystrophin.
Case Histo9
Patient 2 was first seen at the age of 4 months for hypotonia.
During the workup a creatine kinase (CK) level of 9,500
IU/liter was noted (normal value, <200 IU/liter). Results of
an electromyography (EMG)/iierve conduction study were
normal. The muscle biopsy findings were consistent with a
marked inflammatory process. Dystrophin imniunofluorescence and dystrophin immunoblotting were normal. Brain
computed tomography (CT) scans were normal. A diagnosis
of infantile polymyositis was made and the patient was
placed on monthly intravenous immunoglobulin (IVIG)
treatment (total dose, 1,000 mg/kg/visit). The patient made
no improvement after three treatments, prompting a second
muscle biopsy at the age of 9 months. The repeat muscle
biopsy sample showed an early-stage muscular dystrophy
with fibrosis and fatty replacement, and scattered degenerating and regenerating fibers. Laminin 012 immunostaining
showed complete absence of immunoreactivity in all muscle
fibers in both muscle samples, and a diagnosis of C M D with
laminin 012 (merosin) deficiency was established. At the age
of 14 months, the patient remained severely hypotonic and
very weak: He had a weak cry but with no overt facial weakness, and he was barely able to bring his arm to his face and
kicked very weakly. H e was not rolling over and was not
attempting to sit. There was very poor head control with
head lag. Despite the profound weakness the patient looked
very alert and appeared cognitively normal.
Moleczdar Studies
For laminin a2 immunofluorescence, 4-pm-thick cryostat
sections were incubated for 1 hour with mouse anti-human
merosin ( a 2 chain) monoclonal antibody at the dilution of
1 : 5,000 (Chemicon) as previously described [28].
Laminin a 2 immunoblots were done as previously described for dystrophin using anti-laminin 012 antibody at
the dilution of 1 :500,000 [28, 291.
For DNA analysis, blood was collected in ethylenediaminetetraacetic acid (EDTA) tubes from the proband’s (Patient 2’s) parents. DNA was isolated from peripheral blood
as previously described [30].DNA was also isolated from
cryosections of the probands muscle biopsy specimen as previously described [31 1.
Isolation of total RNA was done as previously described
[32]. Briefly, 10 to 100 mg of frozen muscle biopsy tissue
was directly solubilized in guanidine isothiocyanate using a
Brinkman Polytron homogenizer. Homogenates were clarified by centrifugation, and then layered on cesium chloride
(CsCI) gradients to pellet RNA. Tubes were cut below the
CsCl line, quickly rinsed with 70% cold ethanol, and immediately resuspended in Tris-EDTA (TE) and 1% sodium dodecyl sulfate (SDS); salt was added, and then 100% ethanol
to precipitate. RNA was stored as a precipitate at -80°C.
RNA integrity was verified by agarose gel electrophoresis.
About 1 to 2 pg of total RNA was reverse transcribed using
oligo(dT) primer as previously described [33].The complementary DNA (cDNA) was boiled to denature RNAlDNA
heteroduplexes and to inactivate reverse transcriptase. Primers were designed to cover the entire 9,534-bp of the merosin
cDNA sequence [Primer Table submitted to National Auxiliary Publications Service (NAPS), no. 053361.
Pegoraro et a\: Merosin Gene Mutations 783
Table I . Antibodies Used o n
Frozen and Parafin Sectionsfroni Patient 2
Antibodies Used on the
Frozen Sections
Antibodies Used on the Paraffin Sections
CD8/ Leu2a
CD5/Leu 1
CD57 / Leu7
s 100/polycl.
B 1)
I :20
I :20
Neo Markers
1 : 100
1 : 120
By hand
CD56/ 123C3.D5
1 :30
1 :30
1 :30
1 :30
CD 57/ Leu7
C D 45 R 0/ O P D 4
CD21/ IF8
1 :20
1 :30
1 :30
1 :70
1 :70
1 :400
Biotes t
1 : 1,600
1 : 500
M; prot. 2-4'
Prot. 1-12'
Prot. XXIV-5'"
Prot. XXIV-5'^
Prot. 1-4'
XXIV (Sigma) ar 30 rng/dl.
BD = Becton-Dickinson, Mountain View, CA; Dako = Dako, Carpinteria, CA; Neo Markers = Neo Markers, Fremont, CA; Riotesr =
Biotest. Denville, N]; M = microwave pretreatment in 0.1 M citrate buffer, pH 6, for 2-5-minute intervals: Pror. = Venrana protease 1
or 2 for length of incubation in minutes a t 37°C
Polymerase chain reactions (PCRs) contained about 50 ng
of cDNA, 50 ng of each primer, 100 nM each deoxynucleo-
side triphosphate (dNTP), and 0.2 pCi of a-"P deoxyadenosine triphosphate (dATP) in 12.5-p1 total volume. Reaction
conditions were 3 minutes at 94°C to denature: 30 seconds
at 9 4 ° C 30 seconds at 65"C, and 30 seconds a t 72°C for
35 cycles with an extension of 10 minutes at 72°C for all
primer sets except primer set 10F/533R and 8951F/9375R
for which 1 minute at 94"C, 1 minute at 5 5 ° C and 2 minutes at 72°C for 35 cycles were used.
Three different single-stranded conformational polymorphism (SSCP) conditions were used for screening for merosin
mutation as previously described [34]. Conformers were reamplified using the original amplification primer and PCR
conditions and directly sequenced on an automatic sequencer
as previously described [34].
Muscle Injltrate Chrzracterization
A panel of antibodies was used to characterize the muscle
infiltrates using immunoperoxidase and both frozen and
paraffin-embedded muscle tissue as follows.
Frozen muscle biopsy specimens were
sectioned onto Superfrost Plus (Fisher Scientific) slides,
dipped in cold acetone for 5 to 10 minutes, and dried in a
container with anhydrous calcium sulfate. The dried tissue
sections were rehydrated in 0.1 M Tris buffer, p H 7.6, which
was also the diluznt and wash for all subsequent steps, then
blocked for 10 minutes with Shandon's Universal Blocker.
Sections were reacted wirh primary antibody for 2 hours a t
room temperature and biotinylated secondary antibody (Vector Laboratories, Burlingame, CA) for 30 minutes. The dilution and sources of the primary antibodies are listed in Table
1. The reaction was developed using the avidin-biotin comFROZEN SECTIONS.
784 Annals of Neuroloby
Vol 40
No 5
November 1996
plex (ABC) method (Vector) and diaminobenzidine (DAB).
The sections were then counterstained with Gill's hematoxylin.
PARAFFIN SECTIONS. Deparaffinized and rehydrated 4pm-thick sections on Superfrost Plus slides were blocked for
endogenous pcroxidase with 0.7% hydrogen peroxide in
70% methanol for 40 minutes. 'The dilution and sources of
the primary antibodies are listed in Table 1. The slides were
reacted with the primary aiiribodies and biotinylated secondary antibody using automated immunostainer except in the
case of CD56, which required overnight exposure to primary
antibody and was done by hand. The ABC method (Vector)
with the Shandon Immunostainer or by hand, and the Ventana ES immunostainer with Ventana secondary reagents
(Ventana Medical System, Tucson, AZ), were used for development as indicated in Table 1. The chromogen was DAB.
The slides were counterstained with Meyer's hematoxylin.
Biochemical, Histological, and Clinical Feattires of
10 Laminin a2-Deficient Patients
Thirty-three patients were selected for the diagnosis of
CMD from a database search of 2,400 muscle biopsies,
a n d the 33 biopsy samples were tested with larninin a2
irnrnunofluorescence a n d immunoblot. Thirty percent
(10/33) of the patients were completely deficienr for
laminin a 2 ('Table 2 , Figs 1, 2 ) . T h e male-female ratio
was 1:0.8, a n d the ages ranged between 10 months
a n d 11 years (mean, 5.26 2 3.76 years). All patients
were severely weak at birth, b u t n o n e needed rnechanical ventilation or feeding. T h e most c o m m o n presenta-
Age at
Sex Biopsy Age
4 mo
5 nio
10 m o
5 1110
1.9 v r
6 mo
7 nio
7 mo
5 Y'
Best Motor
Creatine Kin'ise
Electromyography Fasciculations First
Able to move
arms and
legs against
Sit iinsupporred N A
8,000 at 4 nio N A
9,890 ar 5 nio 1,574 at 10
2,000 at 5 mo N A
M yopathic
M yo path ic
8.3 y r Sit unsupported
5.6 yr' Sit unsupported
Stand in stand7 yr
ing frame
1 Y'
11 yr
Sit unsupported
M 7
1 Y'
1.6 yr
2.5 yr
9.6 yr
9.6 yr Sit unsupported M yopathic
Sit unsupported Normal
Abnormal white
matter signal
Abnormal white
matter signal
4,360 at 5 [no 322 at 8 yr
4,000 at 7 mo 2,020 at 1.2 yr N 1)
Abnormal white
2,200 at 7 mo N A
matter signal
6,000 at 1 nio N A
N 1)
Abnormal white
mat rer signal
Abriorrnal white
4,000 at 1.6 yr NA
matter signal
Abnormal white
6.192 at 6 m o 314 at 9 yr
matter signdl
= not
available; N D = not done.
tion was a floppy infant unable to move the limbs
against gravity. Facial weakness was never present. In
all patients, achievement of motor milestones was extremely delayed: T h e average age for sitting unsupported was 2.06 5 0.8 years (range, 14-36 months),
and only 1 patient was able to stand using a standing
frame at the age of 6 years (Patient 6 ) . Despite the
gradual attainment of motor milestones, most patients
never achieved full muscle and limb function and none
were able to walk. Contractures were never present at
birth, but developed afterward and involved the elbows, hips, knees, and ankles and could become disabling. Episodes of respiratory insufficiency were present in a t least 70% of the patients. All patients were
cognitively normal for age. CK measuremenrs were invariably markedly increased in the early stage of the
disease (in 8 patients the mean was 5,056 5 2,755
IU/liter between 0.1 and 1.6 years [normal value,
<200 IU/liter]), but tended to decrease to near-normal
range over time. EMG studies showed a myopathic
pattern in 70% of the patients. MRI of the brain in
all patients studied showed an abnormal white matter
signal in the cortex, brainstem, and spinal cord (see
Table 2).
Muscle biopsy findings in all patients were consistent with a dystrophic process. Common features included marked fiber size variation, increased central
nuclei, hypercontracted fibers, degeneration and regeneration, and a marked increase in the endomysial and
perimysial connective tissue with fatty infiltration. In-
flammatory cell infiltrate was highly variable, with
specimens from older patients generally showing much
less inflammation than those from very young patients.
Inflammatory cells were occasionally seen around blood
vessels, and rarely associated with individual muscle
fibers. T h e biopsy sample from one of the youngest
patients studied (Patient 2, age 5 months) showed very
dramatic infiltrate, which is described in more detail
below. We attempted to distinguish between T cellmediated cytotoxicity (invasion of viable myofibers)
and necrotic fibers with infiltrating macrophages. Both
macrophages and T cells were associated with myofibers; however, the scarcity of infiltrated fibers, and the
very small size of the neonatal myofibers made it impossible to accurately distinguish between cytotoxicity
and necrosis. Abnormal spindles such as a thickening
of the major blood vessel wall were often observed.
In 3 patients (Patients 2, 4, and 10) a large infiltrate
of mononuclear cells surrounding single muscle fibers
and invading the endomysial and perimysial spaces
prompted a diagnosis of infantile polymyositis prior to
laminin a2 testing. Unfortunately, the earlier muscle
samples showing mononuclear cell infiltration in Patients 4 and 10 were not available for further testing.
Identijcation of Laminin a2 Gene Mutations
Gaming Congenitrrl Muscular Dystrophy
Biopsy specimens from 5 merosin-deficient patients
and 5 control subjects were selected for mutation
screening based on adequate muscle tissue.
Pegoraro et al: Merosin G e n e Mutations
Fig 2. Ininiiirtoblot analysis of muscle biopy specimens f r o m
8 nierosin-dejcient patients showing complete luminin a2
dejicieriry. Laminin a2 inimunoblot analysis of Patients 7
(lane I), 6 (lane 2), 9 (lane 3), 3 (lane 4), 8 ( h e 5)>2
(lane 6). I0 (hne 71, and 4 (lane 81, and control muscle
( C ) is shown. The lower panel is the correspondingposttransj2r gel stained with Coomas& blue to visualize the myosin
heq-chain protein used for normalizing myojiber protein content o f samples.
Fig 1. Ear(v inflanmatoy changes in primary bminin a2
dejciency (A, B) Lnminin a2 staining in a muscle biopsy
specimen fiom a control and a patient having laminin a2
gene mutation (Patient 2). (C- E) Hematoyliti-eosin staining
of the same patiemti muscle biopsy specimen tnken at the age
of 5 months. (F) Hematosylin-eosin staining of a second muscle biopsy specimen taken at 9 months. (C) A severe itiflammatoy myopathy with muscle jiber size variation, scattered
necratic jibers, mild increase in endomysiul and greater
increase in perimysial connective tissues, fical .fit9 injiltrution, and a massive perivasrular and interstitial injtmnmtitory
nttiscle jibers caw be seen undergoing phagoryreaction. A PUJ
tosis. (0)lri,h?ammatoy cells nrrariged into a prima y - p p e $1licle. Single muscle jibers are surrounded by T and B cells
(see Fig 7)3biitfiiu show overt necrosis. (E) A d$%ent
region of the same muscle biopsy specimen with only fiatures
of an early-stage musrzilar dvstroplly with .fiber size variation,
fhu degenerating jibers, mild increase in perimysial coiinertive
tissue, and initial fit9 itijltration. (F) At 9 months, an
early-stage muscular dystrophy with fpw scattered mottonticlear
celh in the interstitium.
Patient 2 showed a unique conformer identified with
primer set 3062F-3587R, and a second conformer
with primer set 7463F-8073R, which was also shared
with 2 control subjects and 2 patients (Patients 1 and
5) (Fig 3 ) . Direct sequencing of the conformer detected
with primer set 3062F-3587R revealed a single base
deletion at nucleotide position 3264 of the merosin
coding sequence. This was predicted to result in a
frame shift with a stop codon at nucleotides 3270 to
3272 of the merosin cDNA (Fig 4, Table 3). SSCP
and sequence analysis suggested that the proband was
homozygous or hemizygous for the G deletion.
Thc conformer detected with primer set 7469F8073R showed three base-pair changes: a T -+ C at
786 Annals of Neurology Vol 40
No 5
November 1996
Fig 3. Screening of the merosin coding sequence for potential
mutations by single-stranded conformational po(vmoiphism
(SSCP). RNA was isolated fiom the muscle biopsy specimen
of 4 control subjects (nterosin-positive, he-onset mvopathy
patients) and the proband. The complete coding sequence of
the merosin gene iuas amplified using 20 sets oj’overlappirig
primers and screened for possible mutations by SSCP. Shown
is an example of SSCP analysis o f reverse tranm+tasepolymerase chain reaction products for primer sets 3062F3587R and 7469F-8073R. In the left panel (primer set
3062F-3587R) is shown a unique cor$rmer (arrows) in the
proband rind not in the controls ( I and 2);in the righk panel
(7469F-8073R) is d o w n a conformer (arrows) containing
polymoiphisms in t l ~ eproland and Control Subjects I and 3.
Fig 4. Nirrleotide sequence of laminin a2 romplementay
D N A (cDNA) showing a single base-pair deletion at rtucleotide 3264 resulting in a stop codon at nticleotide 3270.
Shown are the Applied Biosystems D N A sequencer traces from
the unique cot$rmer detected with primer set 3062F3587R. The single base-pair deletion (A G) results in a fiame
shift with a stop codon a j&cJbase pairs douvistream.
Table 3. Molecular Data in 5 Merosin-Dejcient Patients
(Primer Set)
2626F-3 138R
A3 148T
T7809C (V2587A) C7879G
del 3264
T7809C (V2587A) del >3264-7894
T7809C (V2587A) del 1939-1943
nucleotide 7809, resulting in a valine to alanine amino
acid change (V2587A); and two benign polymorphisms, a C
G at nucleotide 7879 (V2610V) and
A at nucleotide 7894 (P2615P). Patient 2 appeared homozygous or hemizygous for one allele of
each of these putative polymorphisms. C7873G caused
a gain of a Sau3AI restriction site, while G7894A
caused a loss of a Hinff restriction site.
The parental origin of the G nucleotide deletion in
Patient 2 was determined using genomic DNA extracted from peripheral blood of the proband‘s parents
and from the proband’s muscle biopsy (Fig 5). Patient
2 appeared either homozygous or hemizygous for the
single base-pair deletion, as previously suggested by the
cDNA sequence. His father was heterozygous for this
single base-pair deletion; however, his mother appeared
homozygous for the normal-size allele (see Fig 5). The
patient clearly inherited the deleted allele from the paternal chromosome and apparently showed “noninheritance” from the mother. To rule out nonmaternity or
isodisomy, seven polymorphic CA repeats flanking the
laminin a 2 gene on the long arm of chromosome 6
were typed in the proband and his parents, each of
which showed the expected diparental inheritance (data
not shown). These data ruled out both nonmaternity
and isodisomy, and suggested that Patient 2 inherited
a large deletion mutation from his mother.
Analysis of the C7879G and the G7894A polymorphisms we identified in Patient 2 and his parents
showed a large deletion spanning more than nucleotide
3264 (G deletion) to nucleotide 7894 (G7894A polymorphism) (Fig 6). This deletion covers at least 50%
of the laminin a2 coding sequence.
In Patient 5 , sequencing of the conformer detected
with primer set 1675F-2234R showed a heterozygous
genomic DNA
Fig 5. Paternal inheritance of the single base deletion and
maternal inheritance of a large deletion in the 3237F3293R region of the laminin a2 coding sequence. Shown ic
autoradiography of the polymerase chain reaction product
encompassing the single base-pair deletion at nucleotide 3264
of tbe merosin coding sequence (primer set 3237F-3293R) in
the proband (Patient 2 ) (cDNA and genomic DNA) and
in the proband? parents’genomic DNA. The proband: father
is heterozygous for the normal-size allele and for the deleted
allele (arrow), while the proband? motber appears homozygous for the normal-size allele. This analysis suggests that the
proband is hemizygous for the paternal deleted allele, and
inherits a deletion mutation .fiorn the mother.
deletion of 5 bp (data not shown) (see Table 3). Primers were designed directly flanking this deletion mutation. The deletion was confirmed by multiple primer
sets in Patient 5’s cDNA, but inheritance could not be
tested as the deletion appeared immediately adjacent
to an introdexon boundary and could not be amplified from genomic DNA. Thus, only one putative mutation was detected in this patient.
Histopathological Study of the Proband j. Muscle
Biopsy Specimens
Patient 2 had the first muscle biopsy performed at the
age of 5 months. The biopsy sample showed a very
striking inflammatory process with numerous mononuclear cells infiltrating the muscle. The mononuclear
cells, either scattered or organized in large aggregates,
were localized in the interstitium and in the perivascular space (see Fig 1).
Immunohistochemical staining was performed on
frozen and paraffin-embedded tissue sections using antibodies directed against T cell-, B cell-, and mononuclear phagocyte-associated antigens (Table 4). The first
muscle biopsy specimen showed large dense lymphoid
aggregates and scattered lymphocytes singly or in clusters within the muscle bundles. Most of these infiltrating mononuclear cells were positive for T-cell markers
(see Table 4). In the large aggregates, which were found
only in the frozen sample, CD4+ cells were clearly
more numerous than CD8’ cells (Fig 7 ) .These areas
had numerous T cells and only a few histiocytes, based
Pegoraro et al: Merosin Gene Mutations 787
nective tissues, and initial fatty infiltration were the
main features (see Fig 1). A tentative diagnosis of infantile polymyositis was made.
Because of the lack of response to IVIG therapy,
Patient 2 had a second muscle biopsy performed at the
age of 9 months. In the second muscle biopsy sample,
only features of an early-stage muscular dystrophy were
present. Serial sections revealed scattered mononuclear
cells and small aggregates in the interstitium. T h e discrepancy between the apparent muscle recovery and the
poor clinical improvement prompted laminin ct2 staining in the muscle biopsy specimen that showed coniplete laminin ct2 (merosin) deficiency.
Fig 6 Restriction enzyme analysis of tiovel polymorphisms
shou?c maternal inheritance o f a merosin gene deleted for
50% of the coding sequence in the probmd (Patient 2).
Shown is a summary o f t h e restriction digestion of polymerase
chain reaction products encompassing tuio polymorphisrns
(C7879G m d G7894A) in tbe proband (Plxtient 2) nnd his
parents. Tbis nn+is shoua that the pvoband inherits CI large
deletioii mutatioii j o i n tbe mother from niicleotide 3264 (G
deletion) to mtrcleotide 7894 (G7894A po!yrtiorphisni).
on CD2, CD3, and CD68 reactivity. Although the
number of cells could not be counted accurately in the
frozen sections, the CD8/CD3 ratio was calculated in
a smaller paraffin sample. O n e thousand five hundred
CD8' lymphocytes, and on serial section, 1,900 CD3'
lymphocytes were counted, making the CD8'/CD3+
ratio 0.8.
In the frozen sample, €3 cells were found in aggregates in association with the T-cell aggregates (see Fig
7). The paraffin sample contained only one mediumsize B-cell aggregate, which appeared to represent a follicular structure based on the presence of C1>21t dendritic cells.
Cells expressing 0 5 7 were not detected in either
the frozen or the paraffin sample. T h e paraffin section
was stained for 0 5 6 , and it revealed a few scattered
positive mononuclear cells.
Interestingly, the mononuclear cell infiltrates were
strictly focal: T w o other different fragments of the
same muscle biopsy sample were also cut, but fragments showed only scattered mononuclear cells. M U cle histopathology was consistent with an early-stage
muscular dystrophy: Fiber size variation, few central
nuclei, scattered degenerating and regenerating muscle
fibers, mild increase in endoniysial and perimysial con-
788 Annals of Neurology
Vol 40 No 5
November 1996
The basal lamina of cells is critical for development,
specialization of cell function, homeostasis, and tissue
remodeling or repair. An indispensable component of
all basal laminae is laminin, a large polar hererotrimer.
Despite the importance and ubiquitous nature of laminin, and the multiple genes contributing to this molecule, only a single disease has previously been associated with any of the eight genes. The rare skin disorder
junctional epidermolysis bullosa (OMIM 150292) has
been shown to be caused by mutations of the cwo subunits of laminin 5 (kalinin) [35].Here we describe
three novel gene mutations found in an a chain of
laminin, namely, laminin a2 (merosin). Laminin a2
is known to associate with different p chains to create
two distinct laminin molecules: laminin 2 ( P I , yl), the
major component of the rnyofiber basal lamina, and
laminin 4 (also S-laminin; p2, I), which is critical for
the formation of the neuroniuscular junction [36].
Laminin 2 is also found expressed by Schwann cells in
peripheral nerve, and in some regions of the CNS [24,
Here we report 10 patients with biochemically defined complete laminin a 2 deficiency. Five were tested
for mutations. O n e of these patients was a compound
heterozygote for clear loss-of-function mutations (1-bp
deletion, and a large deletion encompassing >50% of
the gene). We identified nine base changes in coding
sequence that did not change amino acids, and were
clear polymorphisms. In addition, we identified a
frame-shifting deletion (del 1937-1943) and a base
change causing an amino acid alteration (V2587A).
Due to the unknown exon/intron structure of the
gene, we were unable to investigate these changes in
parents and control subjects, and were therefore uncertain as to whether these represent disease-causing mutations. As the V2587A change also appears in Patient
2, who has two well-documented loss-of-function mutations, we believe this change is also a polymorphism
(see Table 3). The del 1937-1943 is a 5-bp deletion
that shifts the reading frame and is likely pathogenic;
however, we were unable to identif). a second mutation
Table 4. Stains Using T-cell, B-cell, und Phigocyte Markers in the Proband; First M11sc1eBiopsy Specimen
Usual Reactivity
CD3/ Leu4
CD4 / Leu3a&b
CD8/ Leu2
CD2O / L26
NK, T-cell subset
T cells
B cells
CD2 1/ 1F8
B cells
B cells
CD681PG-M 1
CD56/ 12C3.D5
Mononuclear phagocytes
Myeloid, mononuclear phagocytes
Interdigitating reticulum, Langerhans cells
B cells, mononuclear phagocytes, activated T cells
Numerous positive
Numerous positive
Numerous positive
Numerous positive
Numerous positive
Numerous positive
Moderate no. of positive cells (less than
with CD4)
Lymphocytes negative
Similar to CD3
Perivascular aggregates and moderate no. of
scattered positive cells
Similar to CD20
Similar to CD20
Positive dendritic cells where CD20 aggregate of cells is
Scattered positive histiocytes
Scattered positive histiocytes
Sniall no. of scattered posirive cells
B cell-rich areas and histiocytes positive
Very few small scattered cells
Helperlinducer T cells and macrophages
Cytotoxic/suppressor T cells
natural killer; FDC = follicular
dendriric cells.
consistent with presumed recessive inheritance. In 2
patients we did not identify any SSCP conformers or
base changes. There are two possible explanations for
this finding: Either these patients show laminin a2 deficiency as a secondary consequence of some other primary biochemical defect, or our assay system was not
sensitive enough to detect the mutations of the laminin
a 2 gene in these patients. T h e laminin 012 gene is very
large and contains many polymorphisms, features that
greatly complicate molecular analysis. O n the other
hand, there are many examples of secondary deficiencies with similar clinical pictures as for primary deficiencies. For example, only a small minority of patients
showing immunostaining abnormalities of a-sarcoglycan (adhalin) show mutations in the a-sarcoglycan
gene; many others show mutations in the dystrophin,
P-sarcoglycan, or y-sarcoglycan genes [34].Despite this
clear genetic heterogeneity, mutations of all four genes
show a similar clinical and histopathological picture.
Further studies are needed to define the specificity of
the a2 laminin immunostaining abnormalities for mutations in the corresponding gene.
T h e clinical findings in our patients with laminin a2
deficiency are consistent with the expression patterns of
the laminin a2 gene. The a1 and a2 isoforms are
coexpressed in muscle throughout much of fetal life;
however, in late gestational age, a1 expression is repressed [24, 36, 38-40]. W e hypothesize that CMD
patient muscle may be protected from pathology by
expression in early fecal life, but then subjected to
destruction by the immune system as a1 is downregulated.
Of particular interest is the consistently abnormal
MFU findings of white matter changes, which would
suggest clinical evidence of CNS involvement. To the
contrary, patients appear mentally normal and d o not
show clinical signs of CNS involvement. There is precedence for highly abnormal MKI findings in the
absence of clinical symptoms: Female carriers of the
severe dysmyelinating disorder Pelizaeus-Merzbacher
disease show highly abnormal MRIs but are asymptomatic [41].Merosin-deficient patients also show a mild
demyelinating neuropathy consistent with the documented expression of laminin a 2 in Schwann cells in
the PNS [42].
An additional novel finding in our study is the dramatic inflammatory cell content of muscle taken at a
young age from laminin a2-deficient patients. Indeed,
3 of the 10 patients studied were carrying a possible
diagnosis of infantile polymyositis, a diagnosis that implies an autoimmune mechanism. Interestingly, the patients reported here appear very similar to those reported as having “congenital inflammatory myopathy”
by Shevell and associates [ 131. While merosin staining
was not reported for the patients in this earlier report,
2 of the 3 patients described showed clinical symptoms
and histopathology findings consistent with the merosin-deficient patients in our series. Here we report
Pegoraro et a]: Merosin Gene Mutations 789
Fig 7.Injammatoy injiltrate in a congenital muscular dystrophy patient with laminin a2 gene mutations sttggets a role of
injammatory cells in the pathogenesis of laminin a2 dejciency. (a) A fiozen-section immunohistochemical stain for CD22/Leu14
(B cells) demonstrating B-cell agpegates (upper left) as well as scattered positive cells. (b I , 62) Parafin-section immunohistochernical stains f;)r CD3/Leu4 (T cells) and CD20 (marker for B cells) showing the follicle-like B-cell organization and numerous T
cells. (c, d) Immunohistochemical strzins for CD4 (helper/inducer T cells) and CD8 (cytotoxic/suppressor 7’ cells): The heher/
are more numerous thdn the rytotoxic/suppressor T cells (d) in these areas.
inducer T cells (i)
the first characterization of the inflammatory cell population in serial biopsy specimens from a patient having
loss-of-function mutations of the laminin a2 gene. We
found extensive but focal infiltration of T cells (both
CD4’ and CD8’) and B cells in perimysial, endomysial, and perivascular spaces in the muscle sample taken
at the age of 5 months, but not in a later biopsy sample
obtained at 9 months. The infiltrating cells occasionally
formed foci with characteristics of primary follicles of
lymphoid tissue. Infiltrating cells were occasionally associated with individual myofibers; however, the scarcity of these fibers and their small diameter made it
impossible to distinguish between T cell-mediated cytotoxicity or necrosis. It is difficult to understand the
focal nature and temporal specificity of this inflammatory response. There is evidence for considerable interaction between the basal lamina and the immune system: Antibodies against laminin are found in normal
humans [43],and are thought to be pathogenic in a
number of disorders (systemic sclerosis, idiopathic pulmonary fibrosis, idiopathic myelofibrosis) [44]. A
pathogenesis model that we are currently testing is that
790 Annals of Neurology
Vol 40
No 5
November 1996
laminin a2 deficiency results in an immune-mediated
attack of myofibers due to improper assembly of postnatal basal lamina. Consequent myofiber regeneration
is compromised as a result of a defective adult-type
myofiber basal lamina. Regardless of the precise mechanism of myofiber death, our results suggest that many
patients carrying the diagnosis of congenital inflammatory myopathy may in fact have an underlying laminin a2 disorder.
Supported by a grant from the National Institute of Neurological
Disorders and Srroke (NS-28403). Dr Hoffman is an Established
Investigaror of the Ameiican Heart Asociation.
The authors thank David Duggan and Hisashi Kobayashi for
thoughtful suggestions and help.
1. Dubowitr V. A eolour atlas of muscle disorders in childhood.
London: Wolfe Medical, 1389
2. Ranker BQ. The congeniral muscular dystrophies. In: Engel
AG, Franzini-Armstrong C , eds. Myology, vol 1. New York:
McGraw-Hill Book, 1994:1275-1289
3. Fukuyama Y, Kawazura M, Haruna H. A peculiar form of
congenital progressive muscular dystrophy. Pediarr Univ Tokyo
4. Fukuyama Y, Osawa M, Suzuki H. Congenital progressive
muscular dysrrophy of the Fukuyama rype-clinical, genetic
and parhological considerations. Brain Drv 198 1;13:1-9
5. Toda T, Segawa M, Nomura Y, et al. Localization of a gene for
Fukuyania rype congenital muscular dystrophy to chromosome
9q31-q33. Nature Genet 1993;5:283-286
6. Sanravuori P, Leisti J , Kruus S. Muscle, eye and brain disease:
a new syndrome. Neuropediatrics 1977;8:553
7. Sanravuori P. Somer H . Sainio K, er al. Muscle-eyr-brain disease (MEB). Brain Dev 1989;11:147-153
8. Dobyns WB, Pagon RA, Arnisrrong D , et al. Diagnostic criteria for Walker-Warburg syndrome. Am J Med Genet 1989;32:
9. Misugi N. Light and electron microscopic studies of congenital
muscular dystrophy. Brain Dev 1980;2: 191- 199
10. Olney KK, M i l l e r RG. Inflammatory infilrration in Fukuyania
type congeniral muscular dystrophy. Muscle Nerve 1983;6:7577
11. Kohrman M H , Picchietri DL, Wollman R, Chelmicka-Schorr
EE. A variant of Fukuyama congenital muscular dystrophy in
a non-Japanese child. Pediarr Neurol 1986;2:290-293
12. Kinoshita M, Mishina M, Koya N . Ten years follow-up study
of steroid therapy for congenital encephalomyoparhy. Brain
Dev 1986;8:280-284
13. Shevell M, Rosenblatt B, Silver K, cr al. Congenital inflammatory rnyopathy. Neurology 1990;40: 11 11- 1 1 14
14. Fidzianska A, Goebel H H , Lenard H G , Heckman C. Congenital muscular dystrophy (CMD)-a collagen formative disease?
J Neurol Sci 1982;55:79-90
15. Campbell KP, Kahl SD. Association of dystrophin and integral
membrane glycoprotein. Nature 1989;338:259-262
16. Ervasri JM, Campbell KP. Membrane organization of the
dysrrophin-glycoprorein complex. Cell 1991;66:1121-1131
17 Ibraghimov-Beskrovnaya 0, Ewasri JM, Leveille CJ, et al.
Primary structure of dysrrophin-associared glycoproreins
linking dysrrophin to extracellular matrix. Nature 1992;355:
18 Suzuki A, Yoshida M, Yaniamoto H, Ozawa E. Glycoproteinbinding site of dystrophin is confined to the cysteine-rich domain and the first half of the carboxyl-terminal domain. FEBS
Lerr 1992;308:154- 160
19. Tinsley J M , Blake DJ, Zuelling RA, Davies KE. Increasing
complexity of the dystrophin-associated protein complex. Proc
Natl Acad Sci USA 1994;91:8307-8313
20. Campbell KP. Three muscular dystrophies: loss of cytoskeleton-exrracellular matrix linkage. Cell 1995;80:675-679
21. Ehrig K, Leivo I, Argraves WS, er al. Merosin, a tissue-specific
basement membrane protein, is a laininin like prorein. Proc
Narl Acad Sci USA 1990;87:3264-3268
22. Tinipl R, Rohde H , Gehron Robey P, er al. Laminin a glycoprotein from basement membranes. l Biol Chem 1979;254:
23. Cooper AR. Kurkinen A, Taylor A, Hogan BLM. Studies on
the biosynrhesis of laminin by murine parietal endoderm cells.
Eur J Biochem 198l;l l9:189-197
24. Voulreenaho R. Nissinen M, Sainio K, er al. Human laminin
M chain (nierosin): complete primary structure, chromosomal
assignment, and expression of the M and A chain in human
feral tissues. J Cell Biol 1194;124:381-394
25. Tom; FMS. Evangelista T, Leclerc A, et al. Congeniral muscular dystrophy with merosin deficiency. C R Acad Sci Paris
1994;317:35 1-357
26. Hillaire D,Ixclerc A, Faurt S, er al. Localization of mcrosin-
negarive congenital muscular dystrophy to chromosome 6q2
by honiozygosity mapping. Hum Mol Genet l994;3: 16571661
27. Helbling-Leclerc A, Zhang X, Topaloglu H , er al. Mutations in
the laminin 1x2-chain gene (LAMA2) cause nierosin-deficienr
congenital muscular dysrrophy. Nature Genet 1995;11:216-218
28. Hoffman EP, Brown RH, Kunkel LM. Dystrophin: the prorein
product of rhe Duchenne muscular dystrophy locus. Cell 1987;
29. Hoffman EP, Kunkel LM, Angelini C, et al. Improved diagnosis of Becker muscular dystrophy by dystrophin resting. Nrurology 1 9 8 ~ ; 3 ~ : 1 0 l 1 - 1 0 1 7
30. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic
Acids Res 1988;16:1215
31. Pegoraro E, Schimke RN, Garcia C, et al. Genetic and biochemical normalizarion in female carriers of Duchenne muscular dystrophy: evidence for failure of dystrophin production in
dystrophin comperenr myonuclei. Neurology 1995;45:677690
32. Glisin VR, Crkvenjadov R, Byus C. Ribonucleic acid isolated
by cesium chloride centrifugation. Biochemistry 1974;13:
33. Zhou J , Hoffman EP. Parhophysiology of sodium channelopathies: studies of sodium channel expression by quanritative
multiplex fluorescence polymerase chain reaction. J Biol Chem
34. Duggan DJ, Fanin M, Pegoraro E, er al. Adhalinoparhies: complete biochemical deficiency patients are 5% of childhoodonset dysrrophin-normal muscular dystrophy and most partial
deficiency patients do not have gene mutations. J Neurol Sci
(in press)
35. Aberdani D, Galliano MF, Vailly J, et al. Herlirz’s junctional
epidermolysis bullosa is linked ro mutations in the gene
(LAMC2) for rhe gamma-2 subunit on niceinlkalinin (laminin-5). Narure Genet 1994;6:299-304
36. Sanes JR, Engvall E, Burkowski R, Hunter DD. Molecular
heterogeneity of basal laminae: isoform of laminin and collagen
IV ar the neuromuscular junction and elsewhere. J Cell Biol
37. Engvall E. Laminin varianrs: why, where and when? Kidney
Inr 1993;43:2-6
38. Leivo I, Engvall E. Merosin, a protein specific for basement
membrane of Schwann cells, striated muscle and trophoblast,
is expressed late in nerve and muscle development. Proc Natl
Acad Sci USA 1990;85:1544-1548
39. Scwry CA, Chevallay M, Tome FMS. Expression of laminin
subunits in human feral skeletal muscle. Histochem J 1995;27:
40. Engvall E, Eanvicker D, Haaparanra T, er al. Distribution and
isolarion of four laminin variants; tissue restricted disrriburion
of heterorrimers assembled from five different subunits. Cell
Regul 1990;1:731-740
41. Silverstein AM, Hirsh DK, Trobe JD, Gebarski SS. M R imaging of the brain in five members of family wirh PaelizeusMerzbacher disease. Am J Neuroradiol 1990;11:495-499
42. Shorer Z, Philpot J, Munroni F, et al. Dcmyelinating peripheral neuroparhy in merosin-deficienr congenital muscular dystrophy. J Child Neurol 1995;10:472-475
43. Bernard A, Lauwerys R, Mahieu P, Foidart JM. Anti basement
membrane antibodies in the serum of healthy subjects. N Engl
J Med 1986;314:1456-1457
44. Gabrielli A, Leoni P, Ilanieli G , er al. Antibodies against galacrosy1 (a1 + 3) galactose in connective tissue diseases. Arthritis
Rheum 1991;34:375-376
Pegoraro et al: Merosin Gene Mutarions
Без категории
Размер файла
1 127 Кб
myopathy, deficiency, primary, inflammatory, presenting, congenital, laminin, muscular, merosin, dystrophy
Пожаловаться на содержимое документа