Prenatal diagnosis in laminin a2 chain (merosin)-deficient congenital

muscular dystrophy: A collective experience of five international centers

Mariz Vainzofa,*, Pascale Richardb, Ralf Herrmannc, Cecilia Jimenez-Mallebrerad, Beril Talime,

Lydia U. Yamamotoa, Celine Ledeuilb, Rachael Meinf, Stephen Abbsf, Martin Brockingtond,

Norma B. Romerog, Mayana Zatza, Haluk Topaloglue, Thomas Voitc, Caroline Sewryd,h,

Francesco Muntonid, Pascale Guicheneyg, Fernando M.S. Tomeg

aDepartment of Genetics Biology, Human Genome Research Center, IB-USP, R. Matao, 106, Cidade Universitaria, Sao Paulo, SP-CEP 05508-900, BrazilbUnite Fonctionnelle de Cardiogenetique et Myogenetique, Service de Biochimie B, and IFR 14, Groupe Hospitalier Pitie-Salpetriere, Paris, France

cDepartment of Pediatrics and Pediatric Neurology, University Hospital Essen, Hufelandstrasse 55, D-45122 Essen, GermanydDepartment of Pediatrics, Hammersmith Hospital, London W12 ONN, UK

eDepartment of Pediatrics, Hacettepe University Children’s Hospital, 06100 Ankara, TurkeyfGenetics Centre, Guy’s and St Thomas NHS Foundation Trust, London SE1 9RT, UK

gInserm U582, Institut de Myologie, IFR14, Groupe Hospitalier Pitie-Salpetriere, Universite Pierre et Marie Curie (UPMC), Paris, FrancehDepartment of Histopathology, Robert Jones and Agnes Hunt Orthopaedic Hospital, Oswestry SY10 7AG, UK

Received 2 February 2005; received in revised form 30 March 2005; accepted 22 April 2005

Abstract

The congenital muscular dystrophies (CMD) are clinically and genetically heterogeneous. The merosin (laminin a2 chain) deficient form

(MDC1A), is characterized clinically by neonatal hypotonia, delayed motor milestones and associated contractures. It is caused by deficiency

in the basal lamina of muscle fibers of the a2 chain of laminins 2 and 4 (LAMA2 gene at 6q22–23). Laminin a2 chain is also expressed in fetal

trophoblast, which provides a suitable tissue for prenatal diagnosis in families where the index case has total deficiency of the protein. This

article reports the collective experience of five centers over the past 10 years in 114 prenatal diagnostic studies using either protein analysis of

the chorionic villus (CV) of the trophoblast plus DNA molecular studies with markers flanking the 6q22–23 region and intragenic

polymorphisms (nZ58), or using only DNA (nZ44) or only protein (nZ12) approaches. Of the 102 fetuses studied by molecular genetics,

27 (26%) were predicted to be affected while 75 (74%) were considered as unaffected, with 52 (51%) being heterozygous, thus conforming

closely to an autosomal recessive inheritance. In 18 of the 27 affected fetuses, the trophoblast was studied by immunocytochemistry and there

was a total or only traces deficiency of the protein in CV basement membrane in all. In 10 cases material from the presumably affected fetus

was available for analysis after termination of the pregnancy and immunohistochemical study confirmed the diagnosis in all of them. Prenatal

studies of ‘at risk’ pregnancies in the five centers produced neither false negative (merosin-deficiency in CVs in a normal fetus), nor false

positive (normal merosin expression in CVs and affected child), indicating the reliability of the technique, when all the necessary controls are

done. Our experience suggests that protein and DNA analysis can be used either independently or combined, according to the facilities of

each center, to provide accurate prenatal diagnosis of the MDC1A, and have an essential role in genetic counseling.

q 2005 Elsevier B.V. All rights reserved.

Keywords: MDC1A; a2-laminin; Merosin; Prenatal diagnosis

0960-8966/$ - see front matter q 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.nmd.2005.04.009

* Corresponding author. Tel.: C55 11 3091 7966x226; fax: C55 11 3091

7966x229.

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

1. Introduction

The congenital muscular dystrophies (CMD) are clini-

cally heterogeneous disorders characterized by the presence

of hypotonia from infancy, generalized muscle weakness

and joint contractures to variable degrees, as well as a

possible association with central nervous system and eye

abnormalities. Electromyography reveals a myopathic

Neuromuscular Disorders 15 (2005) 588–594

www.elsevier.com/locate/nmd

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594 589

pattern; serum creatine-kinase is usually elevated in the

early phases of the disease and muscle biopsy shows

pathological changes consistent with a dystrophic process

[1–4].

To date, 10 genetically different forms have been

identified [4]. The most common form, MDC1A, accounts

for about 40% of cases, and is due to mutations in the

LAMA2 gene on chromosome 6q22–23 [5,6], which codes

for the a2 chain of laminin 2 (merosin) and laminin 4 (s-

merosin) [7]. The LAMA2 gene is composed of 65 exons and

at least 90 different mutations have been described in

affected patients (http://www.dmd.nl/), but, as there are

neither common mutations nor hot spots for mutations, the

identification of causative genetic abnormality is laborious,

expensive and time consuming. In this respect, the

immunohistological analysis of the protein in the muscle

biopsies is a powerful diagnostic tool, since the great

majority of patients with complete laminin a2 chain

deficiency have shown gene mutations when they have

been searched for systematically [8–10].

Laminins are glycoproteins that are an integral com-

ponent of basement membranes. They are composed of

three subunits: one heavy and two light chains. The main

heterotrimers of mature skeletal muscle fibres are a2–b1–g1

(laminin 2) and a2–b2–g1 (laminin 4), both of which

contain the a2 chain [7,11]. Laminin a2 chain binds to adystroglycan, linking the dystrophin–glycoprotein complex

to the extracellular matrix [12,13]. In addition, it binds to

integrin a7b1D [13].

The laminin a2 chain is expressed in several tissues

in addition to muscle, including skin and trophoblast

[14–18]. Thus, immunohistochemical detection of laminin

a2 chain can be used for prenatal assessment of

trophoblast samples from fetuses at risk for MCD1A,

using specific antibodies.

Additionally, genotype analysis and determination of ‘at-

risk’ haplotypes, using chromosome 6q22–23 microsatellite

markers flanking the LAMA2 gene and various intragenic

polymorphisms, have also been used for prenatal diagnosis

[17,19–23]. The reliability of this diagnosis depends on the

information yielded by the markers and the accuracy of the

diagnosis in the index patients.

The first prenatal diagnoses through the study of the

protein in the trophoblast were done in Germany in 1994 by

Voit [15,24] and shortly afterwards in UK [16] and in

France (Fardeau and Tome, in January 1995, unpublished

observation), following the suggestion made by Voit et al.

[15] that direct assessment of merosin from chorionic

villous (CV) might provide a rapid and straightforward

approach to determine laminin a2 chain expression in a

fetus ‘at risk’. Since then, several laboratories have

performed prenatal diagnosis of MDC1A, and the purpose

of this report is to summarize the data of 114 prenatal tests

performed in five reference neuromuscular centers

worldwide.

2. Patients and methods

The chorionic villus (112 cases) or amniotic fluid

samples (two cases) of 114 ‘at risk’ fetuses of MDC1A

were collected for prenatal diagnosis at 9–12 weeks of

gestation, according to routine procedures (Table 1). The

criteria for doing this were an unequivocal diagnosis of

MDC1A in at least one member of the family, that was

based on the clinical features (hypotonia in infancy,

presence of white matter hypodensity determined by brain

magnetic resonance imaging) and muscle biopsy showing

absence or drastic reduction of laminin a2 chain by

immunocytochemistry. A muscle sample from each index

patient was studied generally using at least two antibodies

against laminin a2 chain. The antibodies were directed

against the N-terminal (300 kDa) and C-terminal (80 kDa)

epitopes.

Where a pregnancy was terminated because the fetus

carried the at risk haplotypes and/or the CVs showed

absence of laminin a2 chain, the fetal muscle, when it could

be obtained (10 cases), was studied for laminin a2 chain

expression.

3. Protein studies

CV samples were frozen in isopentane cooled in liquid

nitrogen. Frozen sections were immunolabelled using

routine methodologies, having as secondary antibodies

anti-mouse labeled with fluorescein, biotin–streptavidin-

Texas-Red or Alexa 594, according to the methodologies

used in each laboratory [5,25,26].

Immunohistochemical analyses were performed as

previously described using the following antibodies: against

the laminin a2 chain 80 kDa fragment (Gibco; mAb1922

Chemicon), Mer3/22B2, kindly provided by Dr L.V.B.

Anderson and commercially available from Novocastra, and

4H8 (Alexis). The two antibodies, Mer3/22B2 and 4H8,

detect mild reduction of the protein more easily than the

mAb1922, although they react with different domains of the

protein: the C terminal globular domain (Mer3/22B2) and

the N-terminal globular IVa domain (4H8), as reported by

He et al. [27]. The epitope of the Novocastra antibody is not

known but it behaves like the Alexis antibody. Additional

monoclonal antibodies against the a5, b1 and g1 laminin

chains (Gibco; Chemicon) were also used as positive

markers for the CV sample labeling. Control CV samples

from fetuses not ‘at risk’ for MDC1A were studied in

parallel.

4. Genetic studies

DNA was extracted from the CV and more recently from

20 ml of amniotic fluid cells (two cases) according to

standard techniques. In a few cases with known LAMA2

Table 1

Prenatal diagnoses

Cases studied DNA studies Protein analysis of CV

Total cases/

families

Affected Normal Total Affected Normal Total Absent or

traces

Normal

France 58/41 13 45 57 13 44 (29 heteroz.) 18 4 14

UK 29/21 8 21 29 8 21 (17 heteroz.) 25 8 17

Germany 14/13 6 8 11 6 5 (3 heteroz.) 14 6 8

Turkey 11/10 0 11 4 0 4 (2 heteroz.) 11 0 11

Brazil 2/2 0 2 1 0 1 (1 heteroz.) 2 0 2

Total 114/87 27 87 102 27 75 (52 heteroz.) 70 18 52

Chorionic villus (CV); Heterozygote carriers (heteroz.).

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594590

gene mutations a direct diagnosis was performed by

sequencing the mutated exons. In the corresponding

families the paternity was verified by microsatellite

analysis. In all other cases the genotype analysis was

performed to identify the ‘at-risk’ haplotypes. The

chromosome 6q22–23 microsatellite markers D6S1715,

D6S407, upstream of the LAMA2 gene and D6S1620,

D6S1705, D6S1572, D6S262, downstream of the LAMA2

gene, and the intragenic polymorphisms G1905A (exon 12),

A2848G (exon 19), and G5515A, A5551G and C5579A

(exon 37) and G6286A (exon 42) were used for this purpose,

according to the methodologies described by Guicheney

et al. [22] and Helbling-Leclerc and Guicheney [28]. The

microsatellite markers (amplified by PCR using fluorescent-

labeled primers, or revealed by peroxidase-labeled CA-

repeat probe) were first tested in parents and the affected

child to establish if they were informative. If all the markers

were heterozygous in parents, the three closest to the gene

were kept for the prenatal diagnosis (D6S407–D6S1620–

D6S1705). At least one informative intragenic polymorph-

ism should be informative in both parents. In the case that

one of the parents showed homozygosity for all the markers,

an indirect genetic prenatal diagnosis could not be

accurately made and the study of the trophoblast by

immunocytochemistry with appropriate antibodies

remained as the only informative strategy.

5. Results

The examination of 112 CVs and of two samples of cells

from the amniotic fluid of ‘at-risk’ fetuses of laminin a2

chain deficiency, using DNA molecular studies using

markers flanking the 6q22–23 region and intragenic

polymorphisms (nZ44) or protein analysis (nZ12) or

both methods (nZ58), allowed the identification of 27

affected fetuses, while the remaining 87 were considered as

unaffected (Table 1).

Among the 70 CV samples assessed for the presence of

the laminin a2 chain, 18 showed total deficiency or only

traces of this protein in the CV basement membrane, and 52

showed an expression of the laminin a2 chain identical to

that in controls (Fig. 1, Table 1). Labeling of CVs for a5, b1

and g1 laminins chains was similar in all samples and

showed no abnormalities, indicating that there was good

preservation of the basement membrane, and that these

chains were normal in the trophoblasts of affected fetuses

(Fig. 1).

DNA haplotyping with microsatellite markers, per-

formed in 102 samples, identified the ‘at risk’ chromosomes

inherited from each parent in 27 (26.4%) fetuses, while 52

(50.9%) were heterozygous for one of the at-risk alleles.

Twenty-three (22.5%) were normal for both alleles.

In the French group a genetic evaluation was performed

in all cases but one (Table 1), either by an indirect approach

using microsatellite and intragenic marker analysis (47

cases) or by a direct approach where the mutations of the

index case were known (10 cases). Analysis of the parental

haplotypes was informative in most cases. In 17 cases

protein study was performed concomitantly with the genetic

study. A crossing-over between one of the closest markers

and the gene were detected in three fetuses out of 47 indirect

diagnoses. Nevertheless, in all three cases an accurate

diagnosis was possible due to information revealed by the

intragenic markers (Fig. 2). From all those seeking advice in

France, in only five families were the markers not

sufficiently informative in one of the parents, and genetic

evaluation was declined.

In the British group, in most cases the genetic analysis

was performed with only microsatellite markers, but

immunocytochemical studies of trophoblast were concomi-

tantly done in most cases (25 out of 29). In the German,

Turkish and Brazilian groups an approach similar to that of

the British was used and trophoblast immunostaining was

performed in all cases (Table 1).

In the French, British, German and Turkish groups more

than one prenatal diagnosis (2–4) was done in the same

family.

There was total concordance between the results of the

protein analysis and DNA studies in CV samples in cases

studied by both techniques.

A total of 10 (five in Essen, three in London and two in

Paris) fetal muscle samples could be studied after

termination of the pregnancy, owing to the presence of the

at-risk haplotype in the fetus and lack of laminin a2 chain

expression in the CVs sample. In nine, the fetal muscle

Fig. 1. Immunocytochemical study of normal control chorionic villus (CVN), known laminin a2 chain-deficient (CVS1), chorionic villus understudy (CVS2)

and fetal muscle (Fetus) obtained after interruption of pregnancy and corresponding to CVS2. In normal CV the laminin chains a5, a2, b1 and g1 are expressed

in CV basement membrane. In the known laminin a2 chain-deficient CVS1 and in the chorionic villus understudy (CVS2) there is complete lack of expression

of laminin a2 chain for both the C-terminal (Chemicon mAb 1922) and the N-terminal (Alexis 4H8) antibodies while a5, b1 and g1 laminin chains are

expressed in both CVS1 and CVS2 basement membranes. As predicted the fetal muscle tissue corresponding to the fetal trophoblast CVS2 also shows complete

lack of laminin a2 chain expression for both the C- and N-terminal antibodies but preserved expression of a5, b1 and g1 laminin chains.

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594 591

showed absence of laminin a2 chain expression and in one

case studied in London there were traces of this protein.

6. Discussion

The data presented here show that immunohistochemical

analysis of CV samples, substantiated in most cases by

molecular analysis, provided the correct diagnosis in all 70

fetuses ‘at risk’ of laminin a2 chain deficiency tested for

protein deficiency. The method is therefore very reliable. It

is essential, however, that the diagnostic criteria for

MDC1A in the proband are fulfilled, since the trophoblasts

from cases with partial laminin a2 chain deficiency have not

been studied by this method and partial reduction of laminin

a2 chain may also occur as a secondary feature in other

forms of CMD, such as MDC1B, 1C, 1D, FCMD, MEB or

WWS (reviewed in [4]).

The LAMA2 gene codes for a protein of 390 kDa, and

under reducing conditions, the laminin a2 chain migrates as

an N-terminal fragment of approximately 300 kDa and

a C-terminal fragment of 80 kDa. Immunohistochemical

studies have shown that partial deficiencies of laminin a2

chain are more easily detected with the 300 kDa antibody

than with the antibody against the 80 kDa fragment [18,29].

The use of more than one antibody is therefore

recommended for diagnostic purposes.

In case of absence of labeling with antibodies against the

laminin a2 chain, an adequate control for the integrity and

identification of proteins from the basement membrane is

necessary. All laminin variants are heterotrimers with an a,

b, and g chain and each chain can be studied with specific

antibodies. Secondary changes in other chains occur in

muscle when laminin a2 chain is absent [5,25,30] and also

in other disorders [31]. The basement membrane of normal

muscle fibers and of normal CVs contains a5, b1 and g1

chains. Antibodies against these chains can be used to

control the good preservation of the basement membrane in

CV samples and muscle fibers of fetuses affected by laminin

a2 chain deficient CMD.

The reliability of prenatal diagnosis by microsatellite

analysis is dependent on the absence of locus heterogeneity

17 15 17 54 34 3

4 9 4 2 G G A A

A G A G1 57 5

A A A A 2 5 7 7

D6S1639 17 4 15 5D6S1715 4 12 9 3DS407 4 5 2 5Ex 12 G A G AEx 19 A A A GEx 37 A A A GD6S1620 2 1 5 5D6S1705 7 7 7 5

10 5 10 5 G A G A A A A A G A G A G A G A 2 2 2 2 7 7 7 7 4 2 4 2

DS407 10 2 10 5Ex 12 G G G AEx 19 A G A AEx 37 G G G AEx 42 G G G AD6S1620 2 1 2 2 NID6S1705 7 7 7 7 NID6S1572 4 9 2 2 NI

2 2 3 2A A A AG G G GG G G G4 4 4 45 5 5 4

DS407 2 3 2 5Ex 12 A G A GEx 37 G G G AEx 42 G A G AD6S1620 4 4 4 3D6S1572 5 1 5 4

Pedigree A Pedigree B Pedigree C

Fig. 2. Importance of intragenic markers for the interpretation of indirect genetic diagnosis. Pedigree A shows a molecular diagnosis done on the basis of single

nucleotide polymorphisms (SNP) since the distal markers were non-informative in the mother. Pedigrees B and C show families with crossing-overs (arrows)

occurring between the microsatellite markers and the LAMA2 gene. The intragenic SNP were used to determine on which haplotype and where the

recombinations occurred. B: crossing-overs (arrows) occurred on both paternal and maternal alleles. SNP analysis allowed determining that the fetus was

carrier of the two parental mutations. C: crossing-over (arrow) occurred on the paternal allele. SNP analysis allowed determining that no mutation was

transmitted to the fetus.

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594592

for the disease [28]. In families with a single child having

complete merosin deficient CMD and white matter

hypodensity determined by brain imaging, microsatellite

analysis can be used for prenatal diagnosis with confidence.

The issue is more complex for patients with partial laminin

a2 deficiency, as it can be primary and secondary [3,4]. The

prenatal diagnosis by microsatellite markers in cases of

partial merosin deficiency therefore needs to be interpreted

very cautiously, as results in cases of secondary protein

deficiencies can be misleading.

Prenatal diagnosis can be made by direct DNA analysis

of CV material, as first reported by Guicheney et al. [22]. In

the present study, indirect DNA analysis was used in the

majority of the cases from France, Germany and UK.

Among the 102 ‘at-risk’ fetuses tested, 27 affected were

identified. This diagnosis was confirmed, after the interrup-

tion of the pregnancy, by immunohistochemical analysis of

the muscle in all 10 affected fetuses that could be examined.

Considering the total of DNA studies reported here the

transmission of the mutations is in accordance with

Mendelian inheritance: 26.4% affected fetuses, 50.9%

heterozygous carriers and 22.5% fetuses with normal alleles

(Table 1). A heterozygote advantage has been reported in a

study of 29 informative families with merosin deficient

CMD done in UK [32] but such abnormal transmission was

not confirmed in the present larger data set.

A potential risk of error through double crossing-over, if

haplotyping alone is made, has been suggested [3].

Nevertheless, by using intragenic markers, single crossings

were observed in three fetuses in this series and an accurate

interpretation was always possible.

In families with a high degree of consanguinity, not only

did the parents show the same ‘at-risk’ haplotype, but the

examination of the fetal DNA samples may reveal the same

haplotypes as those of the mother. In such cases, to exclude

the risk of misinterpretation owing to a maternal DNA

contamination, a paternity testing has to be performed (four

in the French experience) to verify the presence of paternal

alleles at different chromosomal loci in fetal DNA.

Differences in the transcriptional control of the

expression of laminin a2 chain between the skeletal muscle

and peripheral nerve have been observed [33,34] but no

difference in the expression pattern of laminin a2 chain was

reported between skeletal muscle and trophoblast. In our

five different centers, no discordance was observed between

the results obtained by study of trophoblast by immuno-

cytochemistry and those provided by molecular genetic

techniques. In addition, when only the immunocytochemi-

cal study of the trophoblast was done and showed normal

presence of laminin a2 chain in CV basement membrane,

the children have always been healthy.

In the Brazilian family studied by molecular genetics, the

majority of the microstellite markers were informative,

allowing the identification of the ‘at-risk’ haplotype

inherited from each parent [35]. However, the known

LAMA2 gene polymorphisms located in exons 12, 19, 37

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594 593

and 42, which have been described as highly polymorphic in

the European population [28] were not informative in

Brazilian families, suggesting that appropriate intragenic

markers, such as new microsatellites, should be developed

for each population.

On such rare occasions (five in the French experience)

when the markers are not sufficiently informative in one of

the parents, the genetic evaluation alone could not be used,

and immunocytochemical analysis of the trophoblast should

be favored.

In conclusion, haplotyping and immunohistochemistry,

independently or combined (to avoid potential methodo-

logical pitfalls), are powerful tools which provide accurate

prenatal diagnosis of this form of congenital muscular

dystrophy (MDC1A) and play an essential role in genetic

counseling, but accurate diagnosis of the index patient is

mandatory for such indirect approaches.

Acknowledgements

Special thanks to Profs. Victor Dubowitz and Michel

Fardeau, whose initiatives lead the way to a better

understanding of congenital muscular dystrophies and to

the studies here reported. The collaboration of the

following persons is gratefully acknowledged: Thomas

R. Gollop, Nadyr F. Naccache, Dra. Rita de Cassia,

M. Pavanello, Prof. Dr G. Gillessen-Kaesbach, Dr D.

Wahl, Dr J. Colomer and Dr B. Cormand for samples

collection. We would also like to thank very much the

following researchers, who kindly provided us with

specific antibodies: Dr L.V.B. Anderson, Newcastle

upon Tyne, Prof. Dr U. Wewer, and Dr J. Chamberlain.

This work was supported by MDC (Muscular Dystrophy

Campaign) and NSCAG (support to FM), the Hammer-

smith Hospital Neuromuscular Unit, and grants from

FAPESP-CEPID, PRONEX and CNPq, by IMPULS e.V.

and the Alfried-Krupp-von-Bohlen-und-Halbach Foun-

dation (to TV), and by INSERM, Universite Paris VI,

Association Francaise contre les Myopathies (AFM),

Assistance Publique-Hopitaux de Paris (AP-HP), GIS

Institut des Maladies Rares (to PG, PR and NBR).

References

[1] Dubowitz V. 22nd ENMC sponsored workshop on congenital

muscular dystrophy held in Baarn, The Netherlands, 14–16 May

1993. Neuromuscul Disord 1994;4:75–81.

[2] Dubowitz V. Workshop report: 68th ENMC-International workshop

on congenital muscular dystrophy. 9–11 April 1999, Naarden, The

Netherlands. Neuromuscul Disord 1999;9:446–54.

[3] Voit T, Tome FMS. The congenital muscular dystrophies. In: Engel A

G, Franzini-Armstrong C, editors. Myology, 3rd ed. New York:

McGraw-Hill; 2004. p. 1203–38 [chapter 44].

[4] Muntoni F, Voit T. The congenital muscular dystrophy in 2004: a

century of exciting progress. Neuromuscul Disord 2004;4:635–49.

[5] Tome FMS, Evangelista T, Leclerc A, et al. Congenital muscular

dystrophy with merosin deficiency. C R Acad Sci III 1994;317:351–7.

[6] Helbling-Leclerc A, Topaloglu H, Tome FM, et al. Readjusting the

localization of merosin (laminin alpha 2-chain) deficient congenital

muscular dystrophy locus on chromosome 6q2. C R Acad Sci III

1995;318:1245–52.

[7] Burgeson RE, Chiquet M, Deutzmann R, et al. A new nomenclature

for the laminins. Matrix Biol 1994;14:209–11.

[8] Pegoraro E, Mancias P, Swerdlow SH, et al. Congenital muscular

dystrophy with primary laminin alpha2 (merosin) deficiency

presenting as inflammatory myopathy. Ann Neurol 1996;40:782–91.

[9] Pegoraro E, Marks H, Garcia CA, et al. Laminin alpha-2 muscular

dystrophy: genotype/ phenotype studies of 22 patients. Neurology

1998;51:101–10.

[10] Allamand V, Guicheney P. Merosin-deficient congenital muscular

dystrophy, autosomal recessive (MDC1A, MIM#156225, LAMA2

gene coding for alpha2 chain of laminin). Eur J Hum Genet 2002;10:

91–4.

[11] Engvall E. Laminin variants: why, where and when? Kidney Int 1993;

43:2–6.

[12] Ervasti JM, Campbell KP. A role for the dystrophin–glycoprotein

complex as a transmembrane linker between laminin and actin. J Cell

Biol 1993;122:809–23.

[13] Colognato H, Yurchenco PD. Form and function: the laminin family

of heterotrimers. Dev Dyn 2000;218:213–34.

[14] Leivo I, Engvall E. Merosin, a protein specific for basement

membranes of Schwann cells, striated muscle, and trophoblast, is

expressed late in nerve and muscle development. Proc Natl Acad Sci

USA 1988;85:1544–8.

[15] Voit T, Fardeau M, Tome FMS. Prenatal detection of merosin

expression in human placenta. Neuropediatrics 1994;25:332–3.

[16] Muntoni F, Sewry C, Wilson L, et al. Prenatal diagnosis in congenital

muscular dystrophy. Lancet 1995;345:591.

[17] Naom I, D’Alessandro M, Sewry C, et al. The role of immunocyto-

chemistry and linkage analysis in the prenatal diagnosis of merosin-

deficient congenital muscular dystrophy. Hum Genet 1997;99:

535–40.

[18] Sewry CA, Naom I, D’Alessandro M, et al. Variable clinical

phenotype in merosin-deficient congenital muscular dystrophy

associated with differential immunolabelling of two fragments of

the laminin a2 chain. Neuromuscul Disord 1997;7:169–75.

[19] Helbling-Leclerc A, Zhang X, Topaloglu H, et al. Mutations in the

laminin a2-chain gene (LAMA2) cause merosin-deficient congenital

muscular dystrophy. Nat Genet 1995;11:216–8.

[20] Naom I, D’Alessandro M, Sewry C, et al. Refinement of the laminin

alpha 2-chain locus to human chromosome 6q2 in severe and mild

merosin deficient congenital muscular dystrophy. J Med Genet 1997;

34:99–104.

[21] Naom I, Sewry C, D’Alessandro M, et al. Prenatal diagnosis in

merosin-deficient congenital muscular dystrophy. Neuromuscul

Disord 1997;7:176–9.

[22] Guicheney P, Vignier N, Helbling-Leclerc A, et al. Genetics of

laminin a 2-chain (or merosina) deficient congenital muscular

dystrophy: from identification of mutations to prenatal diagnosis.

Neuromuscul Disord 1997;7:180–6.

[23] Guicheney P, Vignier N, Zhang X, et al. PCR based mutation

screening of the laminin a2-chain (LAMA2): application to prenatal

diagnosis and search for founder effects in congenital muscular

dystrophy. J Med Genet 1998;35:211–7.

[24] Dubowitz V. 41st ENMC international workshop on congenital

muscular dystrophy 8–10 March 1996, Naarden, The Netherlands.

Neuromuscul Disord 1996;6:295–306.

[25] Sewry CA, Philpot J, Mahony D, et al. Expression of laminin subunits

in congenital muscular dystrophy. Neuromuscul Disord 1995;5:

307–16.

M. Vainzof et al. / Neuromuscular Disorders 15 (2005) 588–594594

[26] Vainzof M, Marie SK, Reed UC, et al. Deficiency of merosin (laminin

M or alpha-2) in congenital muscular dystrophy associated with

cerebral white matter alterations. Neuropediatrics 1995;26:293–7.

[27] He Y, Jones KJ, Vignier N, et al. Congenital muscular dystrophy with

primary partial laminin alpha2 chain deficiency: molecular study.

Neurology 2001;57:1319–22.

[28] Helbling-Leclerc A, Guicheney P. Analysis of LAMA2 gene in

Merosin-deficient congenital muscular dystrophy. In: Bushby K,

Anderson LVB, editors. Muscular dystrophy: methods and protocols.

Totowa, NJ: Humana Press; 2001. p. 199–218.

[29] Sewry CA, Anderson LVB, Bushby K, Muntoni F. Comparison of

3 antibodies to laminin a2 for the detection of partial expression

in congenital muscular dystrophy. Muscle Nerve 1998;suppl 7:

S109 [Abstracts: IX International Congress on Neuromuscular

Diseases].

[30] Cohn RD, Herrmann R, Sorokin L, Wewer UM, Voit T. Laminin alpha2

chain-deficient congenital muscular dystrophy: variable epitope

expression in severe and mild cases. Neurology 1998;51:94–100.

[31] Taylor J, Muntoni F, Dubowitz V, Sewry CA. Early onset autosomal

dominant myopathy; a role for laminin b1? Neuromuscul Disord

1997;7:211–6.

[32] D’Alessandro M, Naom I, Ferlini A, Sewry C, Dubowitz V,

Muntoni F. Is there selection in favour of heterozygotes in families

with merosin-deficient congenital muscular dystrophy? Hum Genet

1999;105:308–13.

[33] Sewry CA, Uziyel Y, Torelli S, Buchanan S, Sorokin L, Cohen J, et al.

Differential labelling of laminin alpha 2 in muscle and neural tissue of

dy/dy mice: are there isoforms of the laminin alpha 2 chain?

Neuropathol Appl Neurobiol 1998;24:66–72 [Erratum in: Neuro-

pathol Appl Neurobiol 1998;24:166].

[34] Di Muzio A, De Angelis MV, Di Fulvio P, et al. Dysmyelinating

sensory-motor neuropathy in merosin-deficient congenital muscular

dystrophy. Muscle Nerve 2003;27:500–6.

[35] Yamamoto LU, Gollop TR, Naccach NF, Protein NA, et al. analysis

for the pre-natal diagnosis of a2-laminin-deficient congenital

muscular dystrophy. Diag Mol Pathol 2004;13:167–71.