Neuromuscular Disorders
Volume 8, Issue 7 , Pages 521-525, October 1998

58th ENMC Workshop: Myotubular Myopathy, 20–22 March, 1998, Naarden, The Netherlands

Department of Medical Genetics, University of Helsinki and The Folkhälsan Department of Medical Genetics, Helsinki, Finland

Article Outline

 

The 4th Workshop of the ENMC International Consortium on Myotubular Myopathy brought together clinicians and basic scientists from eight countries (list of participants below). The main topics were the mutations found in the myotubularin (MTM1) gene and their correlation with the clinical picture in boys affected by X-linked myotubular myopathy (XLMTM), recent findings shedding light on the pathogenetic mechanisms involved, current recommendations towards a diagnostic protocol, and the various directions for the future collaborative work of the Consortium.

Previous workshop reports have dealt with diagnostic criteria for XLMTM [1], refinement of the linkage region and the identification of candidate genes [2], and the cloning and characterization of the MTM1 gene [3].

Back to Article Outline

1. Clinical follow-up study 

Peter Barth (Amsterdam, The Netherlands) presented his long-term follow-up study, conducted together with Victor Dubowitz, of two Dutch families with XLMTM, representing the first and third reports, respectively, in the literature on this disease 4, 5. The follow-up period extended over more than two decades and the oldest living patient was 54 years old [6]. In one of the families, the clinical course was typical, i.e. neonatally very severe, and the mutation found in the MTM1 gene was a nonsense mutation. In the other family, the course was much more variable, with some family members only mildly affected, and the mutation was a missense mutation.

Back to Article Outline

2. Immunocytochemical studies in autosomal versus X-linked myotubular myopathy 

Caroline Sewry (London, UK) reviewed the histochemical and immunocytochemical data of the eight most recent cases from Hammersmith Hospital and one from Sardinia. All cases, with the exception of the Sardinian case, had been investigated prior to the identification of the MTM1 gene so they were divided on the basis of family history, severity and sex. Six severely affected males whose age at biopsy ranged from 6 days to 6 weeks were compared with one male and two females that presented in infancy and were biopsied at 14–23 months.

In the older, milder cases myopathic features such as whorled fibres were noted. The pale halo to fibres seen with oxidative enzymes was seen both in severe and mild cases, and was not a uniform feature of all milder cases. Immunocytochemistry showed that most fibres in the very young neonates express fetal myosin but this is not specific to myotubular myopathy and can occur in neonates with a variety of problems. Severe cases biopsied later at a few months of age show a decline in the number of fibres with fetal myosin, including fibres with central nuclei, and at 9 months of age there is very little fetal myosin. Similarly, milder cases showing fibres with central nuclei do not express fetal myosin. Thus these fibres do not resemble fetal myotubes and maturation with regard to myosin isoforms does occur [7]. Co-expression of fast and slow myosin isoforms is rare in mild and severe cases. Type 1 hypotrophy is an early feature and this was evident even in a severe case at 6 days of age. High levels of desmin and vimentin, suggestive of immaturity, may occur in some fibres with central nuclei but was not a consistent feature of all such fibres, and does not occur in all cases. High levels of desmin can occur in type 1 hypertrophic fibres in milder cases but this is not specific for myotubular myopathy and can occur in other congenital myopathies. Dr. Sewry emphasized the need to examine immunocytochemical aspects in more cases, and, if possible to do follow-up biopsy studies of severely affected neonates who survive.

Back to Article Outline

3. Canine model 

Stéphane Blot (Paris, France) presented his studies on a canine inherited neuromuscular disorder displaying features of autosomal recessive myotubular (centronuclear) myopathy. The Labrador Retriever dogs studied show muscle weakness, abnormal posture and muscle atrophy. Serum levels of creatine kinase are normal. Electromyographic findings include spontaneous activity, but nerve conduction velocities are normal. Muscle histology shows variation in fibre size with smallness and predominance of type 1 fibres, and centrally located nuclei. Peripheral nerves and spinal cord appear normal. The Workshop participants agreed that this canine disease resembles myotubular (centronuclear) myopathy in man.

Back to Article Outline

4. Characterization of the myotubularin phosphatase gene family, from yeast to human 

Jocelyn Laporte (of Jean-Louis Mandel's laboratory in Strasbourg, France) presented an overview of the gene family. The Strasbourg group has shown that myotubularin exhibits tyrosine phosphatase activity in vitro on a synthetic substrate, suggesting that the putative phosphatase function is real. Recent studies suggest that myotubularin could be a dual-specificity phosphatase [8]. XLMTM may be due to loss of this enzymatic activity.

Using database screening and direct cDNA screening, the Strasbourg group has been able to isolate several homologous genes from the MTM gene family [9]. This family is the largest tyrosine phosphatase family described to date, with genes in human, mouse, zebrafish, drosophila, C. elegans, S. pombe and S. cerevisiae. There is only one such gene present in S. cerevisiae. Twenty genes within this family have been partially characterized.

The Strasbourg group has also performed expression analysis and chromosomal mapping of seven of the eight human myotubularin-related genes known to date (Laporte et al., in preparation). They are ubiquitously expressed, all except one, which is brain-specific. Their successful chromosomal assignment provides candidate genes for other human genetic diseases; in this context, it will be worth finding out whether myotubularin-related genes are causative in the autosomal forms of myotubular (centronuclear) myopathies.

A protein sequence comparison over 200 amino acids around the tyrosine phosphatase active site allowed the construction of a phylogenetic tree and reveals conserved residues, which may be essential for the function of these proteins. All the missense mutations found to date in XLMTM patients affect residues conserved in the Drosophila gene.

Some myotubularin-related genes do not contain the consensus active site and should thus have a function different from that of myotubularin. This was also recently suggested by Cui and co-workers [8].

The research groups in Strasbourg and Berne have determined the complete genomic structure of the human myotubularin gene [10]; the corresponding sequences are available in GenBank under the accession numbers AF020663 to AF020676. This information may be useful for direct screening of patient DNA for mutations. Primers and conditions used for SSCP screening of the MTM1 gene can already be found in a recent paper [11]. A new polymorphic marker in the promoter region of the MTM1 gene, called DXS9929, is also detailed in the GDB.

Back to Article Outline

5. Mutation update and review of described MTMI mutations 

Stephan M. Tanner (Berne, Switzerland), in collaboration with Jocelyn Laporte and Sabina Liechti-Gallati, reviewed the mutations found to date in the MTM1 gene.

The distribution and nature of mutations in the MTM1 gene responsible for XLMTM is of great importance to develop a strategy for reliable diagnostics and phenotype-genotype studies. To provide an overview about the current knowledge, the authors have compiled all published mutations 9, 11, 12, 13, 14, 15, 16and their own unpublished results on a total of 137 independent XLMTM cases into a common figure (Laporte et al., in preparation).

The distribution pattern is very heterogeneous; except for exon 1, all exons show pathogenic aberrations. Exons 4, 8, 9, 11, and 12 clearly seem to be more sensitive for mutational events, or as an alternative explanation, they may encode important functional domains of the protein myotubularin. This is suggested by the nearly exclusive appearance of missense mutations in these exons.

Eleven unique deletions were found, covering either part of, or the complete coding sequence of MTM1. Small mutations consisting of 26 nonsense, 31 frameshift, 17 splice-site, and 52 in-frame mutational events (40 missense, plus one amino acid deletion and 11 times the same insertion of three amino acids) were found in 126 cases. Most mutations are unique; only eight different mutations were detected in more than two unrelated cases. Furthermore, the more frequent mutations were generally found to have occurred independently, implying a rule of `one family- one mutation'.

Most mutations may be explained by the well-described methylation-mediated DNA repair mechanism working on CpG dinucleotides, or by DNA polymerase slippage on repeated sequence elements during replication.

Unfortunately, the observed heterogeneity of XLMTM on the molecular level complicates molecular diagnosis and phenotype-genotype correlations. On the other hand, the frequently detected missense mutations may help to elucidate the function(s) of myotubularin and its underlying gene MTM1.

Back to Article Outline

6. Clinical database 

Angus Clarke and Iain Fenton (Cardiff, UK) presented the database they have established in Cardiff for collecting clinical and histological data on patients with XLMTM. The work is ongoing and full data analysis awaits the collection of additional information. A scoring system of clinical features will be designed for correlation with genotypes, to serve as the basis for a summary paper on genotype-phenotype correlations. After acceptance of the paper, the database will be made available on the Internet.

Back to Article Outline

7. Mutational database 

Wolfram Kress (Würzburg, Germany) presented his plans together with Tiemo Grimm for a mutational consortium database in Würzburg. The main questions to be addressed are the origin of mutations and the existence of mosaicism [17]. After publication in due course, the mutational database also will be made available on the Internet.

Back to Article Outline

8. Molecular genetic diagnostic procedure of XLMTM 

The clinical and histological diagnosis of XLMTM has to be confirmed by molecular analysis of the MTM1 gene. This is important because phenocopies exist, the most obvious one being the severe neonatal form of myotonic dystrophy. The role of autosomal genes in causing a similar phenotype also remains to be clarified [17].

Stephan M. Tanner, working in the laboratory of Sabina Liechti-Gallati, presented a suggestion for consensus molecular genetic diagnostic procedure of XLMTM. This two-stage protocol was generally supported by the workshop participants, an alternative approach being direct sequencing.

8.1. Stage 1 

All 15 individual exons of the MTM1 gene are amplified by PCR using primers and conditions published [11]. It is worth beginning with the analysis of exons 8, 9, 4, 11, 12, and 5, in the order mentioned. These six exons contain about two thirds of all mutations characterized to date. The second step includes exons 13, 14, 6, 3, 7, and 10. Exons 1, 2, and 15 should be analyzed last, since they have shown only very few pathogenic mutations (Laporte et al., in preparation). The obtained PCR-amplicons are submitted to single strand conformation polymorphism and heteroduplex analysis (SSCP and HD) on 12% acrylamide gels containing 7.25% glycerol using a two-buffer gel system. DNA is visualized by silver staining. Gel migration variants are purified using spin columns and sequenced on automated sequencing systems (Tanner et al., in preparation).

If no mutation has been found in stage 1, the possibility remains that the patient has a disease-causing mutation in an intron or in one of the regulatory sequences. Therefore, a second stage is recommended as a back-up procedure.

8.2. Stage 2 

Total RNA is isolated from peripheral white blood cells, muscle biopsies, fibroblasts or lymphoblastoid cell lines. This is feasible due to the ubiquitous transcription of MTM1 with the same open reading-frame in all kinds of cells. Reverse transcription (RT) of the full length coding sequence is performed and subsequent PCR-based analysis of the cDNA may follow in two approaches:

Firstly, reverse transcription gel analysis (RTG) covers the coding sequence in three overlapping PCR amplicons (~600–750 bp). These products are analyzed on 10% acrylamide gels containing 7.25% glycerol. Exon skipping, deletions, or splice defects causing gain or loss of mRNA sequences can readily be identified. In addition, direct sequencing of these products allows the detection of missed exonic mutations. This method has been successfully applied for the detection and characterization of four different mutations, one of which was initially missed in the SSCP and HD analysis (Tanner et al., in preparation).

Secondly, reverse transcription sequence analysis (RTS) may be an alternative for RT-PCR-based diagnosis. With this approach the coding sequence is split in six shorter (~300–400 bp), overlapping amplicons which are analyzed by SSCP and HD or directly sequenced. This has been shown to be useful for precise localization of splice defects (Tanner et al., in preparation).

In conclusion, a fast two-step molecular genetic diagnostic procedure is proposed for putative XLMTM cases, solely based on PCR, and starting with screening of genomic DNA. If the first step is not conclusive and adequate material is available, the next step is transcript analysis. The complete analysis may be performed in 2 weeks for up to 10 patients. Using this strategy, the Berne group has not missed any mutation among those known in the MTM1 gene, which implies a detection rate for mutations of more than 98%.

Thus, when a mutation has been detected in the proband, female relatives can be offered investigation of their carrier status, and those found to be carriers will have the option of prenatal diagnosis.

Back to Article Outline

9. Diagnosis by detection of myotubularin 

Jocelyn Laporte reported that the Strasbourg group have raised polyclonal and monoclonal antibodies against peptides derived from the myotubularin sequence or against full length myotubularin. The four polyclonal and nine monoclonal antibodies recognizing myotubularin are under characterization. As a first step to detect the endogenous myotubularin from normal and patient cell lines (lymphoblast, fibroblast and myoblast), the group performed immunoprecipitation experiments with a monoclonal antibody directed against the N-terminus of the protein.

Endogenous myotubularin was detected in each normal cell line but was not detected in cell lines from patients with truncating mutations. Thus, the detection of myotubularin by immunoprecipitation appears promising as a technique for diagnosis if adequate material is available (Laporte et al., in preparation). The sensitivity and specificity of this technique, however, remain to be tested further.

Back to Article Outline

10. Future prospects 

Plans for collaborative efforts include searching for the autosomal genes causing myotubular (centronuclear) myopathy. In this context, samples from further families with autosomal recessive or autosomal dominant myotubular (centronuclear) myopathy are urgently needed [18].

A collaboration was also set up for studying X-inactivation in females belonging to families with XLMTM. Further histological studies will be done to elucidate the spectrum of muscle biopsy features of affected boys in relation to age and type of mutation. The canine model will be studied further, both immunocytochemically, and genetically, in an attempt to localise the gene causing myotubular/centronuclear myopathy in the dog.

In order to isolate substrate(s) of myotubularin and eventually of MTMR gene products, the Strasbourg group have launched two-hybrid and substrate-trap experiments. To study the physiological role of myotubularin, they are disrupting the mouse gene in the hope of creating a mouse model for the disease. The group is also setting up a cellular model.

The next workshop is planned for September, 1999. The Consortium hopes then to be able to present a complete diagnostic protocol, including a protein method, and to be in a position to answer many of the remaining questions outlined above, to the benefit of families with myotubular myopathy.

Back to Article Outline

11. Workshop participants 

Professor Peter Barth, Amsterdam, The Netherlands

Dr. Enrico Bertini, Rome, Italy

Dr. Stephane Blot, Paris, France

Dr. Angus Clarke, Cardiff, UK

Professor Victor Dubowitz, London, UK

Mr. Iain Fenton, Cardiff, UK

Dr. Wolfram Kress, Würzburg, Germany

Dr. Jocelyn Laporte, Strasbourg, France

Prof. Sabina Liechti-Gallati, Berne, Switzerland

Dr. Bernard van Oost, Utrecht, The Netherlands

Dr. Karen Helene Orstavik, Oslo, Norway

Dr. Caroline Sewry, London, UK

Dr. Stephan M. Tanner, Berne, Switzerland

Dr. Carina Wallgren-Pettersson, Helsinki, Finland

Back to Article Outline

Acknowledgements 

This workshop was made possible thanks to the financial support of 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), Vereniging Spierziekten Nederland (The Netherlands) and Deutsche Gesellschaft für Muskelkranke (Germany), Schweizerische Stiftung für die Erforschung der Muskelkrankheiten (Switzerland), Prinses Beatrix Fonds (The Netherlands), Verein zur Erforschung von Muskelkrankheiten bei Kindern (Austria) and Muskelsvindfonden (Denmark) as well as its associate members: Unione Italiana Lotta alla Distrofia Muscolare and Muscular Dystrophy Association of Finland. We are grateful to Professor Alan E.H. Emery for scientific advice and encouragement and to Michael Rutgers, M.Sc., and Janine de Vries for their always reliable organisational support.

Back to Article Outline

References 

  1. Wallgren-Pettersson C, Thomas N. Report on the 20th ENMC sponsored international workshop: myotubular/centronuclear myopathy. Neuromusc Disord. 1994;4:71–74
  2. Thomas N, Wallgren-Pettersson C. Report of the 33rd ENMC sponsored international workshop: X-linked myotubular myopathy. Neuromusc Disord. 1996;6:129–132
  3. Wallgren-Pettersson C, Thomas N. 45th ENMC International Workshop: Myotubular Myopathy, 13–15th September 1996, Naarden, The Netherlands. Neuromusc Disord. 1997;7:268–271
  4. van Wijngaarden GK, Fleury P, Bethlem J, Meijer H. Familial `myotubular' myopathy. Neurology. 1969;19:901–908
  5. Barth PG, Van Wijngaarden GK, Bethlem J. X-linked myotubular myopathy with fatal neonatal asphyxia. Neurology. 1975;25:531–536
  6. Barth PG, Dubowitz V. X-linked myotubular myopathy – a long-term follow-up study. Eur J Paed Neurol. 1998;1:49–56
  7. Sewry CA. The role of immunocytochemistry in the congenital myopathies. Neuromusc Disord 1998;in press.
  8. Cui X, De Vivo I, Slany R, Miyamoto A, Firestein R, Cleary ML. Association of SET domain and myotubularin-related proteins modulates growth control. Nat Genet. 1998;18:331–337
  9. Group 1: Laporte J, Hu L-J, Kretz C, Mandel J-L. Group 2: Kioschis P, Coy JF, Klauck SM, Poustka A. Group 3: Dahl N. A gene mutated in X-linked myotubular myopathy defines a new putative tyrosine phosphatase family conserved in yeast. Nat Genet 1996;13:175–182.
  10. Laporte J, Guiraud-Chaumeil C, Tanner SM, et al. Genomic organization of the MTM1 gene implicated in X-linked myotubular myopathy. Eur J Hum Genet 1998;6:325–330.
  11. Group 1: Laporte J, Guiraud-Chaumeil C, Vincent M-C, Mandel J-L. Group 2: Tanner SM, Liechti-Gallati S. Group 3: Wallgren-Pettersson C, Dahl N, Kress W, Bolhuis PA, Fardeau M, Samson F, Bertini E. and members of the ENMC International Consortium on Myotubular Myopathy. Mutations in the MTM1 gene implicated in X-linked myotubular myopathy. Hum Mol Genet 1997;6:1505–1511.
  12. de Gouyon B, Zhao W, Laporte J, Mandel JL, Metzenberg A, Herman G. Characterization of mutations in the recently identified myotubularin gene in twenty-six patients with X-linked myotubular myopathy. Hum Mol Genet. 1997;6:1499–1504
  13. Guiraud-Chaumeil C, Vincent MC, Laporte J, Fardeau M, Samson F, Mandel J-L. A mutation in the MTM1 gene invalidates a previous suggestion of non-allelic heterogeneity in X-linked myotubular myopathy. Am J Hum Genet. 1997;60:1544–1548
  14. Nishino I, Goto Y, Arahata K, Nonaka I. MTM1 gene mutations in Japanese patients with X-linked myotubular myopathy. World Muscle Society, 2nd International Congress, Tunis 16th–18th October, 1997. Neuromusc Disord. 1997;7:428–429
  15. Tanner SM, Laporte J, Guiraud-Chaumeil C, Liechti-Gallati S. Confirmation of prenatal diagnosis results of X-linked recessive myotubular myopathy by mutational screening, and description of three new mutations in the MTM1 gene. Hum Mutat. 1998;11:62–68
  16. Hu L-J, Laporte J, Kress W, et al.  Deletions in Xq28 in two boys with myotubular myopathy and abnormal genital development define a new contiguous gene syndrome in a 430 kb region. Hum Mol Genet. 1996;5:1108–1115
  17. Vincent MC, Guiraud-Chaumeil C, Laporte J, Manouvrier-Hanu S, Mandel JL. Extensive germinal mosaicism in X-linked myotubular myopathy simulates genetic heterogeneity. J Med Genet. 1998;35:241–243
  18. Wallgren-Pettersson C, Clarke A, Samson F, et al.  The myotubular myopathies: differential diagnosis of the X linked recessive, autosomal dominant, and autosomal recessive forms and present state of DNA studies. J Med Genet. 1995;32:673–679

PII: S0960-8966(98)00076-5

Neuromuscular Disorders
Volume 8, Issue 7 , Pages 521-525, October 1998

Article Tools