Nebulin expression in patients with nemaline myopathy

Juliana Gurgel-Giannettia, Umbertina Reeda, Marie-Louise Bangb, Katarina Pelinc, Kati Donnerc,Sueli K. Mariea, Mary Carvalhoa, Moacir A.T. Firemand, Edmar Zanotelid, Acary S.B. Oliveirad,

Mayana Zatze, Carina Wallgren-Petterssonc, Siegfried Labeitb, Mariz Vainzofa,e,*

aDepartment of Neurology, LIM 15, School of Medicine, University of SaÄo Paulo, SaÄo Paulo, SP, BrazilbEMBL, Meyerhofstrasse 1, 69012 Heidelberg, Germany

cThe FolkhaÈlsan Institute of Medical Genetics and Department of Medical Genetics, University of Helsinki, Helsinki, FinlanddDepartment of Neurology, UNIFESP-EPM, SaÄo Paulo, SP, Brazil

eCentro de Estudos do Genoma Humano, IB ± University of SaÄo Paulo, R. do MataÄo, 277 SaÄo Paulo, SP-CEP 05508-900, Brazil

Received 28 March 2000; received in revised form 3 July 2000; accepted 24 July 2000

Abstract

Nemaline myopathy is a structural congenital myopathy which may show both autosomal dominant and autosomal recessive inheritance

patterns. Mutations in three different genes have been identi®ed as the cause of nemaline myopathy: the gene for slow a-tropomyosin 3

(TPM3) at 1q22±23, the nebulin gene (NEB) at 2q21.1±q22, and the actin gene (ACTA1) at 1q42. The typical autosomal recessive form

appears to be the most common one and is caused by mutations in the nebulin gene. We have studied the pattern of nebulin labeling, in

patients with the typical congenital form (ten patients), the severe congenital form (two patients) or the mild, childhood-onset form (one

patient), using antibodies against three different domains of nebulin. A qualitative and quantitative nebulin analysis in muscle tissue showed

the presence of nebulin in myo®bers from all patients. Some differences relating to the rod structure were observed. The majority of the

largest subsarcolemmal rods were not labeled with the N2 nebulin antibody (I-band epitope) and showed an indistinct pattern with the two

antibodies directed to the Z-band portion of nebulin (epitopes M176±181 and serine-rich domain). Diffuse rods were not revealed using the

three antibodies. A discordant pattern of nebulin N2 epitope labeling was found in two affected sisters with a mutation in the nebulin gene,

suggesting that modi®cations in nebulin distribution inside the rods might occur with the progression of the disease. Western blot analysis

showed no direct correlation with immuno¯uorescence data. In nine patients, the band had a molecular weight comparable to the normal

control, while in one patient, it was detected with a higher molecular weight. Our results suggest that presence/absence of speci®c nebulin Z-

band epitopes in rod structures is variable and could depend on the degree of rod organization. q 2001 Elsevier Science B.V. All rights

reserved.

Keywords: Nemaline myopathy; Nebulin; Congenital myopathy

1. Introduction

Nemaline myopathy (NM) is a non-progressive muscular

disorder associated with the presence of rod-like structures

(nemaline bodies) inside the muscle ®bers and with an esti-

mated incidence of 2 per 100 000 live births [1].

The presence of `rods' or nemaline bodies in the muscle

®bers is the pathological hallmark of this disorder. Within

the ®bers, the rods can be present as compact sub-sarco-

meric forms, as ®ne diffuse structures, or both [1±5].

Usually, the rods are present in the type I ®bers, that are

often predominant. But the proportion of ®bers containing

rods can also vary among individuals and among different

types of muscles [1,4±9].

Clinically, NM is characterized by the presence of hypo-

tonia as well as proximal and facial weakness associated

with skeletal deformities. According to the degree of muscle

weakness, severity and age at onset, and based on correla-

tions from the international database on nemaline myopa-

thy, the ENMC Nemaline Consortium at its workshop in

June 1999 [10] suggested the following classi®cation: (1)

severe congenital form, with contractures and no sponta-

neous movements or no respiration at birth; (2) the typical

form, with onset in early childhood with weakness espe-

cially pronounced in the facial, bulbar and respiratory

muscles and in the neck ¯exors; proximal more than distal

weakness, milestones delayed but reached, slow or non-

progressive course; (3) intermediate congenital form, with

Neuromuscular Disorders 11 (2001) 154±162

0960-8966/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.

PII: S0960-8966(00)00177-2

www.elsevier.com/locate/nmd

* Corresponding author. Tel.: 155-11-3818-7563; fax: 155-11-3818-

7419.

E-mail address: mvainzof@usp.br (M. Vainzof).

infantile onset, breathing and moving at birth, but later in

childhood unable to achieve respiratory independence,

sitting or walking, or use of wheelchair before the age of

11 years; (4) mild childhood or juvenile-onset, with no

facial weakness and no foot drop; (5) adult form with

onset of symptoms in adult age; (6) other forms, associated

with cardiomyopathy, ophthalmoplegia, unusual distribu-

tion of weakness or intranuclear nemaline bodies [1,4±21].

Genetically, NM may show both autosomal dominant and

autosomal recessive inheritance patterns [1,5,6,10,21±23].

The dominant and recessive forms are clinically and histo-

logically similar.

Up to now, mutations in three different genes have been

identi®ed as the cause of NM: The ®rst, NEM1 form,

mapped to 1q22±23, was identi®ed by Laing et al. [22]. A

missense point mutation (Met9Arg) in codon 9 of the slow

a-tropomyosin 3 (TPM3) was found in a large Australian

autosomal dominant kindred with late childhood onset [24].

Subsequently, a homozygous missense mutation in the same

gene was found in a patient with presumed recessive nema-

line myopathy and a severe phenotype [25].

The second, NEM2 form, was identi®ed by Wallgren-

Pettersson et al. through linkage analysis, at 2q21.1±q22,

in seven European families [23,26,27]. The candidate

gene in this region was the nebulin gene, and recently,

pathogenic mutations were found [28]. To date, 11 different

mutations have been identi®ed in 11 families of different

ethnic origins. The patients are compound heterozygous for

the mutations in six of the families, and homozygous in ®ve.

The majority of the mutations are small deletions or inser-

tions causing frameshifts, or base substitutions causing stop

codons. The clinical spectrum in these patients showed that

nine had the typical form, three had the severe form, two had

the mild form and one the intermediate form [10].

Very recently, mutations in a third gene, the actin gene at

1q42, were found in patients with either congenital myopa-

thy with excess of thin ®laments (n � 3), severe NM

(n � 11) or mild NM (n � 4). Both autosomal dominant

and autosomal recessive inheritance were described for

this form [29].

The direct correlation between the different mutations

and the expression of the proteins in the muscle is still

under investigation for these three genes.

The main objective of this study was to analyze the loca-

lization and quantity of nebulin in muscle biopsies from

patients with NM, using antibodies for three different

domains of nebulin, and to correlate both with the clinical

variability.

2. Patients and methods

A total of 13 patients (from 12 unrelated families) with a

diagnosis of NM were included in this investigation.

The diagnosis was based on clinical examination, course

of the disease (using the protocol of the ENMC International

Nemaline Myopathy Consortium [10], family history,

serum creatine-kinase levels, electromyography and muscle

biopsy.

Muscle samples were obtained from biceps or deltoid

biopsies, frozen in liquid nitrogen immediately after

removal and stored at 2708C until use. Routine histological

and histochemical procedures were done, with staining for

HE, modi®ed Gomori trichrome, NADH, ATPase 9.4, 4.3

and acid and alkaline phosphatase [9].

The ®ber typing was determined by counting 1000 ®bers

from each patient in ATPase 9.4 and 4.3 reactions, and by

calculating the percentage of type I and type II ®bers. We

considered a predominance of type I ®bers to be present

when there were more than 75% of ®ber type I.

For the analysis of the proportion of ®bers with rods, 800±

1000 ®bers of each patient were analyzed and the percen-

tage of ®bers with and without rods was calculated. In addi-

tion, a classi®cation of the rods and their pattern of

distribution was done as described below: (a) subsarcolem-

mal, when they were localized in a compact manner close to

the ®ber membrane; (b) diffuse, when several small rods

were distributed across the whole ®ber.

The subsarcolemmal rods were classi®ed in large and

small, based on the measurement and calculation of their

relative size inside the ®bers: The diameter and area of 100

®bers and the diameter of their respective rods were

measured, using a speci®c software (KS300, Zeiss). We

also calculated the proportional area that the rods occupy

inside the ®ber and the number and proportion of ®bers with

a large rod (when it occupied more than 11% of the ®ber)

and a small rod (less than 10% of the ®ber) were calculated.

In addition, the average of rod size for each patient was

calculated (Table 2).

Immunohistochemical staining of frozen sections were

done through single and double labeling reactions [30],

using a rabbit polyclonal antibody for a-actinin 2 (kindly

provided by Dr A. Beggs) diluted 1/100, as a marker for rod

structure. Three different antibodies for nebulin were used: a

mouse monoclonal antibody directed to an I-band epitope

near the N2 line region, diluted 1/200 (Sigma [31]). In addi-

tion, two rabbit polyclonal antibodies were used which had

been raised to the expressed serine-rich domain, and to the

M176±181 domains. Both polyclonal antibodies were

diluted 1/10 [32,33]. As second antibodies, anti-rabbit and

anti-mouse IgG antibodies both FITC and CY3 conjugated

were used.

For the analysis of nebulin distribution in muscle ®bers,

the following data were considered: the pattern of labeling

in the entire ®ber and the pattern of labeling of the rod,

identi®ed through labeling with an antibody for ACTN2.

According to the intensity of labeling inside the rods, they

were considered positive (strong labeling), equal (indistin-

guishable from the remaining ®ber cross-sectional area), or

negative (no labeling at all).

Western blot was done with 6% sodium dodecyl sulfate±

polyacrylamide gel electrophoresis gels, and transfer at 150

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162 155

V for 1 h [32]. The blots were incubated and revealed ®rst

with antibody for nebulin (Sigma), and subsequently, with

antibody against ACTN2. The incubations with primary

antibodies were done overnight, and the detection was

done using alkaline phosphatase-conjugated secondary anti-

body.

To determine the relative protein content of the samples,

a relative quantitative densitometric analysis was done for

each patient, comparing nebulin quantity to ACTN2 band

concentration. In addition, this proportion was assessed in

normal control muscle, at different concentrations of muscle

extract (100, 50, 25%).

In 11 patients, the TPM3 gene was analyzed for presence

or absence of the Met9Arg mutation (cases 1, 3, 4±13)

according to the methodology described by Laing et al.

[24]. Screening for mutations in the nebulin gene is ongoing

in patients 2, 7 and 8 [28], and 26 exons, starting from the 3 0

end encoding the Z-disc part of this enormous protein, have

been analyzed to date.

3. Results

3.1. Clinical evaluation

Data on clinical assessment are shown in Table 1.

Among the 13 patients (from 12 unrelated families), two

showed the severe congenital form (patients 3 and 4), ten the

typical form (patients 1, 2, 5±12) and one the mild child-

hood or juvenile-onset form (patient 13). The age at biopsy

varied between 10 months and 28 years. The patients with

the severe congenital form have respiratory insuf®ciency

and still require mechanical ventilation. Patient 3 is unable

to walk and patient 4 walks with assistance.

Among the patients classi®ed with the typical form, all

are ambulant. Three (1, 2, 5) suffered some complications

during the ®rst year of life. Patient 1 had swallowing dif®-

culties, necessitating a gastrostomy. Patients 2 and 5

presented with recurrent pneumonia. Their muscle power

improved with time, and they became able to walk with

assistance or independently. Bone deformities (such as

kyphoscoliosis, high arched palate, pes cavus and thoracic

deformities) were observed in nine patients, and facial

weakness was present in eleven patients. Patient 12 died

suddenly from an unknown cause.

Family history suggested an autosomal recessive inheri-

tance in three patients (from two pedigrees), while the

remaining patients were sporadic cases.

3.2. DNA analysis

DNA analysis did not reveal the Met9Arg mutation in the

11 patients in whom the study was done.

Screening for mutations in about 14% of the nebulin gene

in patients 2, 7 and 8 detected a mutation in one allele in

both affected sisters: a 2 bp deletion in exon 173, which

causes a stop at codon 6154, and was not detected in 284

control chromosomes.

3.3. Histological and histochemical analysis

Predominance of type I ®bers was observed in almost all

patients and varied between 100% (in eight patients) and

60% (patient 7) (Table 2).

All patients showed rods in muscle ®bers, with a propor-

tion between 100 and 41% of ®bers (Table 2).

The distribution of the rods inside the ®bers showed the

following patterns: it was predominantly diffuse in two

patients (patients 5 and 12) and sub-sarcolemmal in 11

patients. The subsarcolemmal rods were classi®ed as

being predominantly large in seven patients (1, 2, 4, 7±9,

11) and small in four patients (3, 6, 10, 13). In seven of these

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162156

Table 1

Clinical data

Patient no. Age of onset Age at biopsy Clinical

forma

Respiratory

insuf®ciency

Gastrostomy Facial

dysmorphism

Bone

deformities

Inheritanceb Maximal motor

ability

1 Birth 14 years TF No Yes Yes S Walks assisted

2 Birth 28 years TF No No Yes Yes S Walks assisted

3 Birth 2 years SCF Yes No Yes Yes S Non-ambulant

4 Birth 8 years SCF Yes Yes Yes Yes S Walks assisted

5 Birth 6 years TF No No Yes Yes S Walks

6 Birth 3 years TF No No Yes Yes S Walks

7c Birth 10 months TF No No Yes No AR Walks

8c Birth 3 years TF No No Yes No AR Walks

9 7±8 months 13 years TF No No Yes Yes S Walks

10 Birth 5 years TF No No Yes AR? Walks

11 4 months 4 years TF No No No Yes S Walks

12 Birth 9 years TF No No Yes Yes S Walks

13 ? 25 years MCF No No No No S Walks

a SCF, severe congenital form; TF, typical form; MCF, mild childhood form.b S, sporadic; AR, autosomal recessive.c Patients 7 and 8 are siblings.

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162 157

Tab

le2

His

toch

emic

alan

dIm

mu

no

his

toch

emic

alan

aly

sisa

Pat

ien

tn

o.

%T

yp

eI/

II

®b

ers

%O

f®

ber

s

wit

hro

ds

Pre

dom

inan

tro

dpat

tern

Rod

size

pro

port

ion

of

®ber

s

(%occ

upie

din

the

®ber

)

Anti

body

N2:

pat

tern

of

rod

label

ing

on

neb

uli

nst

ain

Anti

body

M101-1

02

Anti

body

M176±181

0±10%

.11%

Mea

nsi

ze

(mm

)

11

�2

WB

mol.

wt.

11

00

/09

8S

ub

sarc

ole

mm

alla

rge

32

68

14.2

02%

98%

N�

V�

V

21

00

/09

7S

ub

sarc

ole

mm

alla

rge

1dif

fuse

b43

57

12.6

2%

±98%

"�

�V

38

0/2

09

0S

ub

sarc

ole

mm

alsm

all

1dif

fuse

b53

47

10.9

Pre

sentc

pre

sent

N�

Mosa

ic

41

00

/01

00

Su

bsa

rco

lem

mal

larg

e8

92

20.3

030%

70%

N�

�5

10

0/0

10

0D

iffu

seN

oR

evea

lfa

ult

yN

��

61

00

/09

2S

ub

sarc

ole

mm

alsm

all

1dif

fuse

b71

29

8.9

42%

30%

28%

N�

�7

60

/40

98

Su

bsa

rco

lem

mal

larg

e1

dif

fuse

b5

95

27.9

Pre

sentc

Pre

sent

N�

�M

osa

ic

89

0/1

01

00

Su

bsa

rco

lem

mal

larg

e1

dif

fuse

b9.5

90.5

20.9

020%

80%

N�

�V

91

00

/07

6S

ub

sarc

ole

mm

alsm

all

and

larg

e46

54

20.7

2%

fault

y51%

fault

y47%

fault

yN

�F

ault

y�

Fau

lty

10

95

/54

4S

ub

sarc

ole

mm

alsm

all

94

65.8

77%

23%

0N

��

11

10

0/0

99

.5S

ub

sarc

ole

mm

alla

rge

1dif

fuse

b42

58

13.3

4%

57%

fault

y39%

N�

�1

27

0/3

04

0.6

Dif

fuse

No

Rev

eal

N�

�M

osa

ic

13

10

09

5S

ub

sarc

ole

mm

alsm

all

1dif

fuse

b95

55.0

0100%

0N

��

aW

B,

Wes

tern

blo

t;N

,as

no

rmal

con

tro

l;",

hig

her

than

norm

alco

ntr

ol;

V,

rods

not

posi

tive

but

stru

ctura

lly

vis

ible

;fa

ult

y,

smal

lar

eas

wit

hno

label

ing

insi

de

the

®ber

s.b

Mo

reth

an3

0%

of

the

®b

ers

con

tain

ing

dif

fuse

rod

sal

so.

cA

nal

ysi

sim

pai

red

du

eto

the

smal

lsi

zeo

fth

e®

ber

.

11 patients, diffuse rods were also present in association

with the subsarcolemmal rods.

3.4. Protein studies

3.4.1. Immunohistochemical analysis in ®ber cross sections

In control muscles, all the three nebulin antibodies

showed the typical cross-striation pattern of sarcomeric

proteins throughout the ®bers. Antibody M176±181 also

showed the mosaic pattern of type I/II ®bers, that was not

observed with the antibodies directed to the N2-line and Z-

line epitopes, with slow type I ®bers more intensely labeled

than fast type II ®bers.

None of the patients showed a complete absence of nebu-

lin in muscle ®bers with the three antibodies for nebulin. In

three patients (5, 9, 10), a faulty pattern was observed, with

small areas with no labeling inside the ®bers. It was more

intense in patient 9, using the N2 antibody (Fig. 1). In

patient 9 only, this faulty pattern was also observed with

the other two nebulin antibodies.

A ®ber-typing mosaic pattern with antibody M176±181

was observed in three patients, which was compatible with

the proportion of type I/II on ATPase staining.

3.4.2. Rod labeling pattern

The proportion of subsarcolemmal rods that were

revealed using antibodies for the ACTN2 and N2 domains

of nebulin was similar in all patients, taking into account

both negative and positive images of these structures.

Immunohistochemical analysis of the rod structure, using

the N2 nebulin antibody, showed that a positive labeling

pattern was predominant in the small rods, while a negative

staining of nebulin was observed in the majority of the

largest sub-sarcolemmal rods. (Table 2 and Fig. 1). With

antibodies from the serine-rich domain (M101±102) and

M176±181, the rods were equally labeled, although the

structure of the rods was visible in some patients (Fig. 2).

The diffuse rods were not revealed by the nebulin reaction

with any of the three antibodies.

3.5. Western blot analysis

Western blot (WB) analysis was possible only with anti-

body N2. In three patients there was not enough protein in

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162158

Fig. 1. Examples of double immuno¯uorescence patterns for nebulin (N2

antibody) and ACTN2 staining of the rods, showing a positive nebulin

pattern (1), a negative pattern (2), a pattern indistinguishable from that

seen in the areas of the ®ber not occupied by rods (�) and a faulty pattern in

the cross-sectional area of the ®ber.

Fig. 2. Schematic structure of the nebulin gene and immuno¯uorescence analysis of muscle biopsy from patient 1, showing negatively labeled rods with

antibody from the N2 line (left), and indistinguishable labeling with antibodies from the M176±181 (middle) and serine-rich domain (right) antibodies.

Asterisks denote identifying rods.

the muscle extract for analysis (patients 3, 4, 13) while in the

remaining ten patients the presence of nebulin was

con®rmed. In nine of these 10 patients, the band had a

molecular weight comparable to that of the normal control,

while in one patient (patient 2), a band with a higher mole-

cular weight was detected (Fig. 3). The relative proportion

of nebulin to ACTN2 in normal muscle varied between 37

and 52%, and did not depend on the different muscle protein

concentrations. In the 13 patients, nebulin quanti®cation on

blots was compromised in 7 patients due to muscle extract

degradation. In the remaining six, there was apparently no

correlation with the immunohistochemical data: patients

with undetectable nebulin labeling of diffuse rods (patients

5 and 12) as well as those with negative large subsarcolem-

mal rod labeling on IF (patient 8; Table 2 and Fig. 3) showed

strong bands on WB.

3.6. Intrafamilial correlation

Histochemical study in the two sisters showed a higher

proportion of type 2 ®bers in the younger one (40%) when

compared with the older sister (10%). Both of them,

however, showed the same predominant pattern of large

subsarcolemmal rods.

Immuno¯uorescence (IF) study with the N2 nebulin anti-

body showed different patterns of labeling: the younger, 10-

month-old girl (patient 7) showed some large sub-sarcolem-

mal nebulin-de®cient rods, other positively-labeled rods and

a global mosaic-like nebulin pattern on muscle ®bers. Her

older sister (aged 3 years) (patient 8) showed a homoge-

neous pattern of labeling of ®bers, with predominance of

large sub-sarcolemmal rods that were de®cient for nebulin

(Fig. 4). With the two other nebulin antibodies, the rod

structures were indistinct inside the ®ber. A mosaic of

®ber typing was observed with the M176±181 antibody in

the younger sister (Fig. 4).

4. Discussion

Nebulin is a giant protein (600±800 kDa) found in skele-

tal muscles, which accounts for 3±4% of the total myo®-

brillar proteins [33]. The C-terminal region of nebulin is

located at the Z lines, whereas its N-terminal end is at the

pointed end of the thin ®lament [31,33±35]. Nebulin acts as

a molecular ruler for the thin ®lament and is important for

the assembly and integration of Z discs with the sarcomere

[28]. Mutations in the nebulin gene were recently found in

patients affected by the typical congenital form of NM [28].

However, the correlation between the mutations and the

expression of the protein in the muscle is still under inves-

tigation.

According to the clinical course, ten patients from the

present study could be candidates for the nebulin-related

typical NEM2 form [1,6,10]. All presented with clinical

features compatible with the typical congenital form

although not all had facial weakness or skeletal deformities.

At muscle biopsy, in addition to the presence of rods, a

predominance of type I ®bers was observed. Analysis of

the proportion between type I and type II ®bers in the two

affected sisters showed a more preserved mosaic pattern

(60/40 type I/II) in the younger 10-month-old girl when

compared with her 3-year-old sister (90/10 type I/II).

These data are in accordance with previous pathological

studies which suggested a progressive transformation of

type II to type I ®bers [4,8].

4.1. Nebulin in muscle ®bers

The IF analysis for nebulin using three antibodies against

different domains of this protein showed a positive pattern

in the muscle ®bers of all patients. In addition, the presence

of a normal molecular weight nebulin band, using the N2

nebulin antibody, in nine out of ten patients, suggests that

the primary defect in these patients would neither be related

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162 159

Fig. 3. Double Western blot analysis for nebulin and ACTN2 showing a normal control lane and those of the two sets of NM patients. Myosin content is each

muscle extract is shown in the Ponceau pre-stained blot. Observe the higher molecular weight nebulin band in patient 2, detected in two independent

experiments.

to a total de®ciency of the protein nor to an alteration in

protein size.

Comparing patients with the severe congenital form with

those who had the typical form, no difference in nebulin

pattern appears to occur.

The observation of a higher molecular weight band in one

patient, a ®nding to our knowledge not reported previously,

suggests a mutation associated with the expression of a

larger protein, e.g. an insertion. There is another possibility

also, relating to differential splicing of the nebulin gene.

Several differentially expressed exons have been identi®ed

in the region encoding the Z-disc part of nebulin [10,28,34]

In addition, ®ve novel nebulin exons have been detected in

the mRNA of one Australian patient with a severe form of

NM. The physiological role of these exons remains

unknown [10]. Therefore, although the ®nding of a nebulin

band with a higher molecular weight in one of our patients

warrants further studies to con®rm a primary nebulin muta-

tion, other possible explanations relate to these newly

described phenomena in this gene are alternative splicing

and additional exons. On the other hand, we cannot discard

the possibility of altered migration due to artifact in this

muscle extract.

Previous immunohistochemical studies of nebulin,

performed in patients from families in which linkage results

were compatible with linkage to the 2q locus, did not show

any staining differences between muscles of patients and

normal controls [1,6,36]. In 1999, Pelin et al. [28]

performed immunocytochemical reactions using antibodies

speci®c for the C-terminal SH3 domain of nebulin and for

nebulin single repeats M176±181, and correlated the results

with the different nebulin mutations found in the same

patients. In this analysis, all patients showed presence of

nebulin, and no speci®c differences were observed in the

pattern of staining within the rod structures. All the cases

showed normal labeling with the M176±181 antibody when

compared with controls. Only one patient, with a mutation

in M185, showed no reaction with the SH3 antibody and an

uneven labeling in the larger ®bers, with the M176±181

antibody [28]. In the present study, we used three different

antibodies for nebulin and observed that the pattern of the

entire ®ber reaction was abnormal in one patient (patient 9),

with all three antibodies. This abnormality consisted of

small areas with no reaction for nebulin. The possibility

of frozen artifact could be excluded because these altera-

tions were not observed on the histological staining and

ACTN2 labeling. As this patient had a higher proportion

of `moth-eaten' ®bers, this could be the cause of the pattern

observed. Although more evident with the N2 nebulin anti-

body, the faulty pattern was not correlated to the absence of

a speci®c nebulin epitope. It is possible, however, that it

could represent a particular abnormality of NM, and in

this case, our patient could be a candidate for a mutation

in the nebulin gene.

A ®ber-type mosaic pattern has been described in muscle

from normal controls, with the antibody M176±181 for

nebulin, with slow type I ®bers more intensely labeled

than fast type II ®bers [28]. We observed a mosaic pattern

with the M176±181 antibody in three patients, who had a

higher proportion of type II ®bers (patients 3, 7, 12). The N2

and serine-rich domain antibodies did not show a mosaic

pattern in any of the normal or pathological muscles studied.

With the N2 nebulin antibody a mosaic-like nebulin pattern

was observed in the same 10-month-old patient, which

could be a re¯ection of the positively and negatively labeled

rods observed in this patient. This was surprising, mainly

because it was not seen in her older sister, suggesting an

alteration related to the earliest stages of the disease.

4.2. Nebulin distribution in the rods

We analyzed nebulin labeling using antibodies against

three different domains of the protein. Two of them, anti-

bodies M176±181 and serine-rich domain (M101±102),

were against epitopes localized in the Z bands, and the

third N2 nebulin antibody, against an epitope localized

near to N-terminal region [31], within the I band of the

sarcomere.

Immuno¯uorescence reaction with the two Z-disc antibo-

dies showed that all the rods, independently of their size,

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162160

Fig. 4. Immuno¯uorescence analysis for nebulin, double reaction nebulin

(green) and ACTN2 (red) and the Gomori reaction showing the pattern of

the rods in the two affected sisters. The red labeling in the rods in patient 8

is indicating the presence of only ACTN2 while the orange color in patient

7 is indicating the presence and superposition of nebulin and ACTN2.

were labeled with the same intensity as the remaining ®ber.

The presence of some of the largest, dense subsarcolemmal

rods could be identi®ed because of differences in the stria-

tion pattern. However, with the nebulin antibody from the I

band, it was observed that many of the largest subsarcolem-

mal rods were predominantly negative, while small sarco-

lemmal rods were mostly positive. Additionally, the diffuse

rods were not detectable by nebulin antibody labeling.

Therefore, the presence of speci®c epitopes of nebulin in

these structures may be related to the size and the degree of

organization of the rods. It is tempting to suggest that the

large subsarcolemmal rod would present a demarcated

structure, with a compaction of proteins predominantly

from the Z disc, while the diffuse and small rods, positively

labeled with all antibodies, would be formed by a wider

mixture of proteins.

Nebulin analysis in the rods from the two sisters gives

further support to the hypothesis that as the disease

progresses, modi®cations in nebulin presence and/or distri-

bution inside these structures could occur.

The mechanism of rod formation is still unknown. A

primary defect in the nebulin gene could lead to abnormal-

ities in the nebulin structure causing an altered connection

of this protein at the Z disc and, as a consequence, sarco-

mere instability and rod formation. Perhaps through similar

mechanisms, mutations in other genes for the proteins of the

thin ®lament and Z disc, such as the tropomyosin and actin

genes, also lead to rod formation. Our results suggest that

the presence/absence of the nebulin in rod structures is vari-

able and could depend on the degree organization and

compaction of rods. Alternatively, the immunohistochem-

ical visualization of nebulin at rod structures could be

impaired by masking of the epitopes by other sarcomeric

proteins related to the Z disc, as already suggested for other

proteins [37,38].

It is known that WB studies are unable to detect small

alterations in protein quantity. However, the lack of correla-

tion between WB data and IF pattern of nebulin in the rods

provides further evidences for the hypothesis that the

presence or absence of nebulin in the rod structures is not

a re¯ection of quantitative alterations but a consequence of

reorganization of proteins.

In summary, based on IF and WB analysis, our studies

show the presence of nebulin in muscle ®bers of all the

patients studied with congenital NM. Abnormalities of

nebulin distribution within the rods could be a consequence

of reorganization of sarcomeric proteins.. WB analysis may

in some cases help to reveal a nebulin abnormality.

Acknowledgements

Antibody for ACTN2 was kindly given by Dr Alan

Beggs, to whom we are very grateful. We also wish to

thank Marta Canovas and Fabiana da Silva, for the excellent

technical assistance, Valdir Caldeiras and the Departament

of Biology, IB±USP, for the support on the confocal micro-

scopy and all the physicians who referred affected patients.

M.L.B. was supported by a Marie-Curie Fellowship from

the European Union. K.P. and K.D. were supported by

grants to C.W.-P. from the Association Francaise contre

les Myopathies, the Swedish Cultural Foundation of

Finland, the Finska LaÈkaresaÈllskapet, and the Medicinska

understoÈdsfoÈreningen Liv och HaÈlsa. For organizational and

®nancial support to the ENMC International Consortium on

Nemaline Myopathy, we are indebted to the European

Neuromuscular Centre (ENMC) and its main sponsors:

Association Francaise contre les Myopathies (France),

Italian Telethon Committee (Italy), Muscular Dystrophy

Group of Great Britain and Northern Ireland (UK), Vereni-

ging Spierziekten Nederland (The Netherlands) and

Deutsche Gesellschaft fuÈr Muskelkranke (Germany),

Schweizerische Stiftung fuÈr die Erforschung der Muskelk-

rankheiten (Switzerland), Prinses Beatrix Fonds (The Neth-

erlands), Verein zur Erforschung von Muskelkrankheiten

bei Kindern (Austria) and Muskelsvindfonden (Denmark)

as well as its associate members: Unione Italiana Lotta

alla Distro®a Muscolare and Muscular Dystrophy Associa-

tion of Finland. This work was supported by grants from

FundacËaÄo de Amparo aÁ Pesquisa do Estado de SaÄo Paulo

(FAPESP), FundacËaÄo Faculdade de Medicina, PRONEX,

CNPq and AssociacËaÄo Brasileira de Distro®a Muscular

(ABDIM).

References

[1] North KN, Laing NG, Wallgreen-Pettersson CJ, ENMC International

Consortium on Nemaline Myopathy. Nemaline myopathy: current

concepts. Med Genet 1997;34:705±713.

[2] Shy GM, Engel WK, Somers JE, Wanko T. Nemaline myopathy. A

new congenital myopathy. Brain 1963;86:793±810.

[3] Goebel HH. Congenital myopathies. Acta Paediatr Jpn 1991;33:247±

255.

[4] Wallgreen-Pettersson C, Rapola J, Donner M. Pathology of congeni-

tal nemaline myopathy. A follow-up study. J Neurol Sci

1988;83:243±257.

[5] Wallgren-Pettersson C, Laing NG. 40th ENMC International Work-

shop: Nemaline Myopathy. Neuromusc Disord 1996;5:389±391.

[6] Wallgren-Pettersson C, Beggs AH, Laing NG. 51st ENMC Interna-

tional Workshop: Nemaline Myopathy. Neuromusc Disord

1998;8:53±56.

[7] Shimomura C, Nonaka I. Nemaline myopathy: comparative muscle

histochemistry in the severe neonatal, moderate congenital and adult-

onset forms. Pediatr Neurol 1988;5:25±31.

[8] Nonaka I, Ishiura S, Arahata K, Ishibashi-Ueda H, Maruyama A, Li

K. Progression in nemaline myopathy. Acta Neuropathol

1989;78:484±491.

[9] Dubowitz V. Muscle disorders in childhood, 2nd ed. London: Saun-

ders, 1995. pp. 134±177.

[10] Wallgren-Pettersson C, Laing NG. Report of the 70th ENMC Inter-

national Workshop: Nemaline Myopathy. Neuromusc Disord

2000;313±320.

[11] Schmalbruch H, Kamieniekka Z, Arroe M. Early fatal nemaline

myopathy: case report and review. Dev Med Child Neurol

1987;29:800±804.

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162 161

[12] McMenamin JB, Curry B, Taylor GP, Becker LE, Murphy EG. Fatal

nemaline myopathy in infancy. Can J Neurol Sci 1984;11:305±309.

[13] Norton P, Ellison P, Sulaiman AR, Harb J. Nemaline myopathy in the

neonate. Neurology 1983;33:351±356.

[14] Oferil CJ, Castilho LJ, Arguelles PP, Torrez VI, Alvarez FE. Severe

and mild forms of nemaline myopathy: a report of three cases. Na Esp

Pediatr 1993;39:517±520.

[15] Hefferman LP, Rewcastle NB, Humphy MD. The spectrum of myopa-

thies. Arch Neurol 1968;18:529±542.

[16] Kuitunen P, Rapola J, Noponen AL, Donner M. Nemaline myopathy.

Report of four cases and review of the literature. Acta Paediatr Scand

1972;61:353±361.

[17] Martinez BA, Lane BD. Childhood nemaline myopathy: a review of

clinical presentation in relation to prognosis. Dev Med Child Neurol

1987;29:815±820.

[18] Meier C, Voellny W, Gertsch M, Zimmermann A, Geissbuthler J.

Nemaline myopathy appearing in adults as cardiomyopathy. A clin-

icopathologic study. Arch Neurol 1984;41:443±445.

[19] Palmucci L, Dorigiuzzi C, Morgini T, Chiado-Piat L. Adult onset

nemaline myopathy: a distinct nosology entity. Clin Neuropathol

1993;12:153±155.

[20] Shahar E, Tervo RC, Murphy EG. Heterogeneity of nemaline myopa-

thy. A follow-up study of 13 cases. Pediatr Neurosci 1988;14:236±

240.

[21] Wallgren-Pettersson C, KaÈaÈriaÈinen H, Rapola J, Salmi T, JaÈaÈskelaÈi-

nen J, Donner M. Genetics of congenital nemaline myopathy: a study

of ten families. J Med Genet 1990;27:480±487.

[22] Laing NG, Majda BT, Akkari PA, et al. Assignment of a gene

(NEM1) for autosomal dominant nemaline myopathy to chromosome

1. Am J Hum Genet 1992;50:576583.

[23] Wallgren-Pettersson C, Avela K, Marchand S, et al. A gene for auto-

somal recessive nemaline myopathy assigned to chromosome 2q by

linkage analyses. Neuromusc Disord 1995;5:441±443.

[24] Laing NG, Wilton SD, Akkari PA, et al. A mutation in the alpha

tropomyosin gene TPM3 associated with autosomal dominant nema-

line myopathy. Nat Genet 1995;9:75±79.

[25] Tan P, Briner J, Boltshauser E, et al. Homozygosity for a nonsense

mutation in the a-tropomyosin gene TPM3 in a patient with severe

nemaline myopathy. Neuromusc Disord 1999;9:573±579.

[26] Wallgren-Pettersson C, Pelin K, HilpelaÈ P, et al. Clinical and genetic

heterogeneity in autosomal recessive nemaline myopathy. Neuromusc

Disord 1999;9:564±572.

[27] Wallgren-Pettersson C. Genetics of nemaline myopathies and

myotubular myopathies. Neuromusc Disord 1998;8:401±404.

[28] Pelin K, HilpelaÈ P, Donner K, et al. Mutations in the nebulin gene

associated with autosomal recessive nemaline myopathy. Proc Natl

Acad Sci USA 1999;96:2305±2310.

[29] Nowak KJ, Wattanasirichaigoon D, Goebel NH, et al. Mutations in

the skeletal muscle alpha actin gene in patients with actin myopathy

and nemaline myopathy. Nat Genet 1999;23:208±212.

[30] Vainzof M, Zubrzycka-Gaarn EE, Rapaport D, et al. Immuno¯uores-

cence dystrophin study in Duchenne dystrophy through the concomi-

tant use of two antibodies directed against the carboxy-terminal and

the amino- terminal region of the protein. J Neurol Sci 1991;101:141±

147.

[31] FuÈrst DO, Osborn M, Nave R, Weber K. The organization of titin

®laments in the half-sarcomere revealed by monoclonal antibodies I.

Immunoelectron microscopy: a map of ten nonrepetitive epitopes

starting at Z line extends close to the M line. J Cell Biol

1988;106:1563±1572.

[32] Ho-Kim M-A, Bedard A, Vincent M, Rogers PA. Dystrophin: a sensi-

tive and reliable immunochemical assay in tissue and cell culture

homogenates. Biochem Biophys Res Commun 1991;181:1164±

1172.

[33] Wang K, Wright J. Architecture of the sarcomere matrix of skeletal

muscle: immunoelectron microscopic evidence that suggests a set of

parallel inextensible nebulin ®laments anchored at Z line. J Cell Biol

1988;6:2199±2212.

[34] Millevoi S, Trombitas K, Lolmerer B, et al. Characterization of nebul-

ette and nebulin and emerging concepts of their roles for Z-discs. J

Mol Biol 1998;282:111±123.

[35] Labeit S, Lolmerer B. The complete primary structure of human

nebulin and its correlation to muscle structure. J Mol Biol

1995;248:308±315.

[36] Sewry CA. The role of immunocytochemistry in congenital myopa-

thies. Neuromusc Disord 1998;8:394±400.

[37] Jockush BM, Veldman H, Grif®ths GW, Van Oost BA, Jennekens

FGI. Immuno¯uorescence microscopy of a myopathy. a-Actinin is

a major constituent of nemaline rods. Exp Cell Res 1980;127:409±

420.

[38] Yamagushi M, Robson RM, Stromer MH, Dahl DS, Oda T. Actin

®laments form the backbone of nemaline myopathy rods. Nature

1978;271(5642):265±267.

J. Gurgel-Giannetti et al. / Neuromuscular Disorders 11 (2001) 154±162162