161st ENMC International Workshop on nemaline myopathy and related disorders, Newcastle upon Tyne, 2008

1. Introduction 

Twenty doctors and scientists from 7 countries gathered in Newcastle upon Tyne on the 28th and 29th September 2008, for the seventh ENMC Workshop on nemaline myopathy (NM) and related disorders. A representative of the patient group “A Foundation Building Strength” gave a presentation at the start of the Workshop.

The main themes of the Workshop were continued gene discovery; better definition of clinical entities; animal and tissue culture models of NM; preliminary experimental investigation in model systems of possible therapeutic strategies; locus-specific mutation databases; establishing databases in preparation for anticipated future clinical trials and finally discussion on better methods for molecular analysis of the known NM and related disorders disease genes, especially nebulin. Expected major efforts in the near future include further analysis of possible therapies, development of additional animal models in which to investigate therapies and the implementation of the databases in order to be ready for any future clinical trials. The Workshop participants recognise that a multinational approach to these rare disorders is essential, especially in relation to enrolling patients in any future clinical trials.

2. Clinical news and genotype–phenotype correlations 

Dr. Nigel Clarke (Sydney, Australia) summarised his observations on congenital fibre type disproportion (CFTD) versus NM [1]. He reported evidence that congenital fibre type disproportion and NM due to mutation of the TPM3 gene appear to form a continuous disease spectrum for both clinical and histological features, differing only in the presence or absence of nemaline bodies. Some mutations (such as p.Arg168His) can be associated with either histological pattern while others seem only to cause CFTD (such as p.Leu100Met) or NM (p.Met9Arg). Although only recently reported, already TPM3 mutations associated with CFTD are more numerous than NM, suggesting that the majority of TPM3 mutations do not or rarely induce rod formation. One ‘hotspot’ has emerged for dominant TPM3 mutations. Four different missense mutations have involved amino acids Arg168 and Lys169 and p.Arg168His is a common recurrent mutation, identified so far in five unrelated families. Clinical features appear relatively consistent amongst patients with dominant TPM3 mutations although the overall severity varies even within the same family. The best diagnostic indicators are marked type 1 fibre hypotrophy on muscle biopsy, mild bilateral ptosis, prominent neck flexor weakness, early respiratory muscle involvement and a relatively benign course. In contrast, for the ACTA1 gene there are early indications of important differences between mutations associated with CFTD and NM. Although clinical features are indistinguishable, mutations associated with CFTD cluster on the tropomyosin-interacting surface of actin and are associated with near-normal sarcomeric structure on electron microscopy while ACTA1 mutations associated with NM are scattered throughout the molecular structure and are often associated with sarcomeric disruption.

Dr. Alan Beggs (Boston, USA) gave an update on the work done by the Boston group, including their studies of TPM3 in NM and in CFTD. The Boston group has a patient registry of 512 patients with congenital or undefined myopathy 220 of whom have a diagnosis of NM or related disorder. It is this cohort that the Boston group has been investigating for causative genes. Of 64 probands whose TPM3 gene was screened, five of 12 CFTD cases, and one of 18 NM patients, had mutations, while there were none among 34 cases of non-specific myopathy. Similar to the Sydney group’s experience, p.Arg168His was seen in one case of CFTD and one of NM. They found that in their CFTD patients with TPM3 mutations the type 1 muscle fibres were 40% the size of type 2 fibres, a much larger reduction than the recognised criteria for CFTD, while pathological re-evaluation of the “CFTD cases” (reported by referring pathologists) revealed that several with normal TPM3 genes had atypical findings inconsistent with current definitions of CFTD. Combining their data with published data from Dr. Clarke suggests that more than 30% of CFTD cases will have TPM3 mutations as their cause.

Dr. Beggs also updated the status of studies on the CFL2 (cofilin-2 gene) p.Ala35Thr mutation found in two sisters with NM and minicores [2]. Expressing the mutant cofilin protein, revealed that it could not easily be purified, yet it could be extracted with urea, suggesting it was insoluble, at least in bacterial preparations. The Boston group is currently generating knock-in and knock- out cofilin mouse models.

Dr. Carina Wallgren-Pettersson (Helsinki, Finland) gave a clinical update on muscle disorders caused by mutations in the nebulin gene (NEB), including NM and distal nebulin myopathy, a disorder clinically resembling Laing distal myopathy. She also reviewed the phenomena of CFTD and arthrogryposis in relation to the genes known to cause NM. Mutations in all but the cofilin-2 gene, i.e. mutations in ACTA1, NEB, TPM2, TPM3 and TNNT1 have been found to cause CFTD. Arthrogryposis multiplex congenita has been caused by mutations in at least ACTA1, NEB and TPM2, while distal contractures were rare in patients with NM. A few patients with NM caused by mutations in NEB or ACTA1 had distal contractures, while no patients with distal contractures were found among those with mutations in the other known genes.

Dr. Norma Romero (Paris, France) reported a 27-year-old man from a non-consanguineous family affected by core–rod myopathy caused by mutations in NEB. The patient was born at 33weeks of pregnancy and presented with severe hypotonia, foetal distress and respiratory insufficiency from birth. Generalised weakness was noted but facial movements were relatively preserved. As a child at 2years, he required tracheotomy and permanent mechanical ventilation. No ocular or cardiac involvement was noted. EMG revealed a myopathic pattern and normal nerve conduction velocities. Walking with equipment was possible at the age of 4years, but ambulation was lost at 7years. Muscles biopsies performed at the ages of 2 and 7years showed numerous fibres with distinctive clusters of rods and central well-delineated cores, associated with type 1 fibre predominance. Electron microscopy revealed rods with the characteristic striation and well-defined structured cores with some focal sarcomeric disorganisation and depletion of mitochondria. Molecular genetic analysis excluded mutations in ACTA1 and the hot-spot regions of RYR1. Subsequently, two mutations were identified in NEB by the Helsinki group.

The patient reported here is the first known case of core–rod myopathy associated with mutations in NEB. This observation enlarges the spectrum of genes involved in core–rod myopathies.

Prof. Baziel van Engelen (Nijmegen, The Netherlands), discussed sporadic late onset NM (SLONM). This is a disorder that normally presents after age 40 and progresses subacutely. Limb-girdle weakness and atrophy dominate the clinical picture. Distal weakness, head drop, respiratory insufficiency, and dysphagia can also occur. Hence, SLONM may be misdiagnosed as a motor neuron disorder. SLONM is sporadically associated with a monoclonal gammopathy, which portends an unfavourable outcome. He presented a patient with SLONM associated with monoclonal gammopathy who improved significantly (clinically and histologically) after treatment with melphalan and autologous bone marrow transplantation. This is the first report of successful treatment of SLONM associated with gammopathy [3] and was confirmed by a second group [4], [5].

Vilma-Lotta Lehtokari (Helsinki, Finland) summarised the Helsinki group’s molecular genetic work on NM and cap disease. Attempting to identify mutations in known and novel NM genes in the Helsinki sample cohort, the group has performed linkage analyses for 20 genes (including the known NM genes) and sequenced those showing positive linkage. Hitherto, the group has not identified novel NM-causing genes among the analysed candidate genes. However, since the previous workshop the group has found approximately 50 novel mutations in NEB using dHPLC, RT-PCR and sequencing in 40 NM families and one core–rod myopathy family (presented by Norma Romero), TPM2 mutations have been identified in four probands with NM, distal arthrogryposis or cap disease [6], and TPM3 mutations in four probands with NM. Novel findings in the TPM2 gene were presented and discussed further. These findings comprise two recurrent and one novel mutation underlying cap disease, NM, and distal arthrogryposis identified by the group. In addition to these mutations, three novel cap disease mutations identified by other groups were presented.

The recurrent mutations in both tropomyosin genes known to cause NM, TPM2 and TPM3, were discussed. Notable was that the same TPM2 or TPM3 mutation in different individuals can cause variable phenotypes and histology, raising the question of the importance of the age of the patient when the biopsy is taken, of the modifying genome and of standardising clinical evaluation and diagnostic criteria.

Prof. Francesco Muntoni (London, UK) briefly discussed the clinical features of three children in whom mutations of the ACTA1 gene were found in London, as part of the UK National Diagnostic service for congenital muscular dystrophies and myopathies. In the course of the last year, three previously reported heterozygous mutations were identified: c.109G>T (p.Val37Leu), c.821C>A (p.Ala274Glu) and c.124C>T (p.His42Tyr). This latter mutation was identified in a female patient, from a consanguineous family, who presented in the first few months of life with axial hypotonia and feeding difficulties but apparently relatively preserved power in the proximal muscles of the arms and legs. She developed aspiration pneumonia at around the age of six months and was ventilated but eventually died after a short hospitalisation. Intranuclear rods were identified in the previous two patients identified with this mutation, but not in this patient. The same mutation has been used by Dr. Hardeman to make a knock-in mouse model of ACTA1 NM/intranuclear rod myopathy (see below).

Prof. Caroline Sewry (Oswestry, UK) had had the opportunity to review the slides of the case of zebra body myopathy published in 1975 by Lake and Wilson [7]. As rods were reported in both this case and in the only other published case of zebra myopathy, and she had observed several zebra bodies in the skeletal actin null patients, she had suggested that the ACTA1 gene might be a candidate for zebra body myopathy.

The patient had a clinical history compatible with a congenital myopathy and was floppy at birth. He presented at 7years of age and remained weak. The original publication was based on a biopsy performed when he was about 14years of age. A second biopsy at the age of 29years showed pronounced pathology with marked variation in fibre size with atrophic and hypertrophic fibres of both types. Type 1 fibres were predominant but there was also a suggestion of fibre type grouping, but some of this may have been due to fibre splitting. There were several bizzare multiple split fibres which contained rod-like structures, and eosinophilic material. Internal nuclei and endomysial connective tissue were increased, and there were several vacuoles which contained eosinophilic material. Some hypertrophic fibres showed accumulations of dark green material. NADH-TR staining showed large core-like areas resembling “wiped-out” areas. Electron microscopy showed rods in several fibres, and accumulation of filaments resembling actin. Occasional cytoplasmic bodies surrounded by prominent triads were present and also ring fibres. The peripheral myofibrillar band of the rings was unusual in that no Z-line was present. Only a few subsarcolemmal zebra bodies were present, in contrast to the first biopsy which gave its name to the myopathy. The second biopsy therefore had features in common with those from cases with an ACTA1 mutation (rods, accumulation of actin), but also some features consistent with a myofibrillar myopathy. Ethical approval and the patient’s permission are being sought for genetic studies.

3. Molecular diagnosis 

Dr. Katarina Pelin (Helsinki, Finland) presented diagnostic methods for the analysis of NEB. NEB has 183 exons, most NM patients are compound heterozygotes for two different mutations, and no mutation hotspots are evident. This makes mutation analysis of NEB time-consuming and expensive. Denaturing high-performance liquid chromatography (DHPLC) is currently the most effective method for mutation detection in NEB. This method does not, however, detect large deletions or duplications. Multiplex ligation-dependent probe amplification (MLPA) is a potential method for detection of large copy number changes in NEB. Synthetic MLPA probes have to be designed and optimised for each NEB exon, which has turned out to be time-consuming and expensive. The current aim is, therefore, to replace the DHPLC and MLPA methods with a microarray-based method for mutation analysis of NEB.

Dr. Ann Curtis (Newcastle upon Tyne, UK) and Dr Alan Beggs (Boston, USA) summarised the possible benefits of new generation sequence capture [8] and sequencing techniques for molecular analyses of large genes. These hold the promise of being able to sequence the entire nebulin gene of 183 exons [9] in multiple patients simultaneously, whereas the current methods are extremely time-consuming and laborious.

A discussion on collaborative efforts regarding molecular genetic diagnosis and collection of clinical and molecular data for international genotype–phenotype correlations was held.

4. Pathophysiology, animal models, tissue culture models and in-vitro analysis 

Dr. Edna Hardeman (Sydney, Australia) reported on the successful generation and characterisation of a mouse model for a severe form of NM in which a point mutation found in patients, Acta1(His42Tyr), was “knocked into” the mouse alpha-skeletal actin gene [10]. Cytoplasmic and intranuclear rods characteristic of the disease were detected in the muscles of the mice. Severe muscle weakness was apparent as evidenced by a shortened lifespan in male mice, severe kyphosis and an altered gait. Fore-arm weakness and decreased mobility was detected by 15days after birth. Since fibre hypertrophy can associate with milder forms of the disease in patients [11] and animal models [12], two hypertrophy-promoting factors were trialled on these mice to determine if they could alleviate clinical features in the mice.

The two hypertrophy-promoting proteins four-and-a-half LIM domain protein 1 (FHL1) and insulin-like growth factor 1 (IGF-1) resulted in increased body weight, alleviated impaired mobility, improved fore-arm weakness and reduced rod pathology in fibres. Recently, l-tyrosine supplementation has been implicated in improving some clinical features in NM patients [13]. To investigate this further, 25mg of l-tyrosine was orally administered daily to the Acta1(His42Tyr) mice. Within 24h, impaired mobility was alleviated. After 4weeks of treatment body weight increased, fore-arm strength and muscle force increased in male mice, and rod pathology decreased. The data suggests that hypertrophy and administration of the amino acid l-tyrosine may provide potential therapeutic strategies for NM.

Prof. John Sparrow (York, UK) described the development of the fruitfly, Drosophila melanogaster, and its flight muscles, as a useful model genetic system with which to study NM mutations. This system involves making transgenic fly lines carrying human ACTA1 mutations in the homologous residues of the fly Actin88F gene (93% amino acid homology with ACTA1) and studying their effects on muscle structure and function. He showed that all six ACTA1 mutations his lab had studied disrupted muscle structure and function in the flight muscles (where the Act88F gene is specifically expressed). All the mutations caused specific changes in the muscles cells with novel subcellular structures similar, if not homologous, to those seen in human NM biopsies. In addition, he showed previous studies from his lab where flies homozygous for nemaline-related actin mutants had shown that mutant actins could lead to either non-assembly of myofibrils or a case where the muscle developed normally, young flies flew, but a progressive myopathy led to use-dependent muscle disintegration and a ‘nemaline’ phenotype. The defect in this case was due to the mutation reducing actin binding to α-actinin, leading to progressive myofibril damage at the Z disk. A 10-fold reduction in the strength of actin binding to α-actinin has been reported for the ACTA1 p.Lys336Glu mutation associated with both nemaline myopathy and cardiomyopathy [14].

His laboratory had introduced an extra wild-type gene copy to flies heterozygous for the six NM mutations and shown a dramatic recovery of flight and muscle structure, though this had not been found to be the case in his earlier studies with more severe alleles.

As a result of all these studies he proposed that Drosophila flight muscle and the Act88F actin gene provide a very good genetic model for the study of human ACTA1 actin mutations. While flies cannot recapitulate all aspects of the human disease, since they lack an adaptive immune system and ability to regenerate damaged muscle, this could be seen as a positive aspect since it allowed study of the molecular, cellular and developmental aspects of the basic molecular lesion without complications of these response systems.

Dr. Elizabeth Busch-Nentwich (Cambridge, UK) described her group’s experience of possible zebrafish models of NM and related disorders. The group screens mutated zebrafish with motility defects for linkage to known muscle genes and then analyses the candidate genes for mutations. They identified one zebrafish line with phalloidin aggregates whose linkage region included the nebulin gene and which shows nonsense-mediated decay of the nebulin mRNA. Another mutant, lacking capz function, showed destabilisation of the Z-line and α-actinin accumulations. Morpholino antisense oligonucleotide knock-down of tropomodulin 4, which is only expressed in fast muscle fibres, gave results suggesting tropomodulin was a good candidate gene for nemaline myopathy in human patients where the genetic mutation had not been identified.

Prof. Kathryn North (Sydney, Australia) reported her group’s studies on the mechanisms of rod formation in inherited myopathies. Protein aggregates or rods are the primary pathological feature in NM. Mutations in the gene encoding skeletal muscle α-actin (ACTA1) are responsible for about 20% of NM cases associated with cytoplasmic rods, as well as cases of intranuclear rod myopathy. “Nemaline” rods also occur as a secondary feature in some mitochondrial disorders, in particular in association with complex I deficiency. The mechanisms of rod formation are not well understood – particularly when they can occur in diverse disorders with very different structural and metabolic defects. Therefore the group sought to determine whether there is a common mechanism of rod formation associated with abnormalities in structural components of the thin filament (different ACTA1 mutations modelled on tissue cultures from patients with nemaline and intranuclear rod myopathy) and rods induced by metabolic cell stressors such as ATP depletion.

The group has developed a tissue culture model in which to study mutations in ACTA1 identified in patients. The expression of mutant actin-EGFP leads to formation of cytoplasmic or intranuclear rod-like structures in muscle cells, similar to those observed in patient muscle [15], [16]. The group is also able to induce rod formation in cells transfected with wild-type actin-EGFP by depleting ATP, as a model of oxidative stress (and mitochondrial myopathy) [16]. They have determined that these different rods have altered biochemical properties, varying in their protein composition and actin conformation. Using Fluorescence Recovery After Photobleaching (FRAP), they have also seen a difference in actin turnover in different rods suggesting that they have very different actin dynamics.

In summary, not all rods are the same, and each may contain a unique fingerprint of actin-binding proteins that are specific to the mutation, specific to the cellular location of the rods (intranuclear versus cytoplasmic) and/or specific to the underlying pathological process (i.e. mutant actin or ATP depletion). In addition, this study confirms a link between structural and metabolic pathologies in the formation of rods. Characterisation of these different rods will ultimately help in understanding the mechanism of their formation and also the impact they have on cellular function in disease.

Prof. Steve Marston (London, UK) reported on his group’s in vitro studies, using the sliding filament and other systems, of mutations in TPM and ACTA1. They have found consistent differences in calcium sensitivity and cross-bridge cycling for mutations in sarcomeric proteins causing dilated and hypertrophic cardiomyopathy in that mutations that cause dilated cardiomyopathy decrease calcium sensitivity and usually decrease cross-bridge turnover, while those that cause hypertrophic cardiomyopathy increase calcium sensitivity and usually increase cross-bridge turnover. These techniques have now been applied to skeletal myopathy mutations and the consistency seen in dilated and hypertrophic cardiomyopathy mutations is not replicated with the ACTA1 mutations causing the various congenital myopathy phenotypes. In a review of 30 congenital myopathy-causing ACTA1 mutations that have been studied using a range of biochemical and in vitro approaches, diverse molecular defects were found but there is no obvious pattern seen in mutations resulting in the same histopathology [17]. Moreover defects in contractile properties were rarely seen, suggesting that the disease-causing defects act at the more fundamental level of myofilament assembly. The conclusion is that in spite of decades of investigation, there is still too little known of the structure – function relationships of actin.

5. Experimental therapies 

Dr. Katarina Pelin presented two potential strategies for development of RNA-based therapies for NM caused by mutations in the nebulin gene (NEB). About 30% of the mutations in NEB cause aberrant splicing, most often exon skipping. Correction of aberrant splicing can be achieved using modified antisense U7 small nuclear ribonucleoproteins (snRNPs), carrying an antisense sequence complementary to the aberrantly spliced exon, and an exonic splicing enhancer (ESE) tail for improvement of exon recognition [18]. Development of the technique for correction of aberrant NEB pre-mRNA splicing requires finding the optimal antisense target and the most effective exonic splicing enhancer.

The majority of the mutations identified in NEB are nonsense, frameshift or missense mutations located within exons. The mutated exons could be replaced with wild-type exons at the pre-mRNA level through spliceosome-mediated RNA trans-splicing, i.e. splicing between two separate pre-mRNA molecules. Targeted spliceosome-mediated RNA trans-splicing requires three components: the spliceosome, a target pre-mRNA, and a pre-trans-splicing RNA molecule (PTM) [19]. The spliceosome and target pre-mRNA are provided by the cells whereas the PTM RNA molecule is produced from an expression vector transfected into the cells. PTMs can be designed to carry out one of three forms of trans-splicing; 3′exon replacement, 5′exon replacement and internal exon replacement, depending on the trans-splicing domain, i.e. the domain responsible for recognition and splicing of the target pre-mRNA. Internal exon replacement would be the method of choice for correction of the majority of the NEB mutations, as only 6% of the published mutations are located in the last four exons, and none in the first four exons of NEB. Development of the technique is technically challenging and requires a thorough knowledge of the regulation of NEB pre-mRNA splicing.

Dr. Kristen Nowak (Perth, Western Australia), outlined studies investigating the possibility of upregulating or reactivating the expression of cardiac actin in post-natal skeletal muscle as a therapy for skeletal muscle actin disease. Various lines of evidence have suggested that mutant actin protein produced from dominant skeletal muscle actin (ACTA1) mutations might be able to be diluted by a wild-type actin protein. The West Australian group hypothesised that cardiac actin would be the best protein to try to harness to do this.

Cardiac actin is expressed in skeletal muscle during development and is 99% identical to skeletal muscle actin at the amino acid level. By birth, the expression of cardiac actin is usually down-regulated in skeletal muscle and skeletal muscle actin then becomes the predominant striated actin isoform [20], [21]. Additionally, it was shown that the disease severity of recessive nemaline myopathy patients who are homozygous for ACTA1 null mutations (who have no skeletal muscle actin and severe myopathy), is moderated by the level of cardiac actin retained in their skeletal muscles [22]. Although even the least affected of the patients is still significantly affected, this indicates that cardiac actin can function at least to some extent in post-natal skeletal muscle.

To determine whether the expression levels of cardiac actin could be sufficiently modulated to functionally compensate for the absence of skeletal muscle actin (and in the future perhaps overcome the effects of the presence of mutant skeletal muscle actin), cardiac actin transgenic mice were bred with skeletal actin knock-out mice (which all die by 9days postnatal [23]).

Dr Nowak outlined how the mice lacking skeletal actin, but expressing cardiac actin in their skeletal muscles, survived past 9days postnatal. Indeed the majority of mice live to adulthood and are capable of producing offspring. The mice that reach adulthood have only minor functional deficits of their skeletal muscle. These findings suggest that cardiac actin is sufficiently functional for virtually normal life in place of skeletal muscle actin. This is a promising step towards a possible therapy for those patients without skeletal muscle actin. Additionally, these results give hope that cardiac actin may be able to be utilised to dilute out the effects of dominant skeletal muscle actin mutations, which are the type most commonly identified.

A drug screen, using drugs already approved for use in humans by the American Food and Drug Administration (FDA), is being performed to establish whether any are able to reactivate cardiac actin expression in mature skeletal muscle cultures [24] as an off-label effect.

6. International databases and clinical trial infrastructure 

Nigel Laing (Perth, Australia) discussed the locus-specific database he is developing for skeletal muscle α-actin gene (ACTA1) mutations.

Prof. Hanns Lochmüller (Newcastle, UK) gave a short introduction on the network of excellence TREAT-NMD (www.treat-nmd.eu) and on patient registries for rare neuromuscular disorders which have been the topic of a recent ENMC workshop [25]. The TREAT-NMD ultimate goal is better treatments for patients with neuromuscular disorders. This requires testing of new treatments in clinical trials. One of the essential infrastructures for this are databases, or patient registries listing patients, their phenotypes, precise mutations, clinical contacts, etc. Developing experimental therapies for neuromuscular disorders depends on the precise mutations in the individual patients. Neuromuscular disease patients benefit from the registries in various ways, including: feedback on standards of care and research developments, belonging to a broader community, not being left behind as clinical trials develop and having a link to the research community. The benefits for developing effective therapies are facilitated access to appropriate patient cohorts, and improved feasibility and planning of clinical trials. The TREAT-NMD patient registries for Duchenne muscular dystrophy and spinal muscular atrophy are nationally based, with national curators that feed into global databases. The TREAT-NMD registry effort will expand to include rarer neuromuscular diseases, such as nemaline myopathy and related disorders.

In this session, the international plans for joint databases in similar formats for all the congenital myopathies were discussed. Dr. Heinz Jungbluth (London, UK) presented a draft data sheet for this purpose. It was agreed that the Consortium on Nemaline Myopathy and Related Disorders should support and join this venture, which is being further developed on a collaborative basis.

Dr. Jan Kirschner (Freiburg, Germany) reviewed the TREAT-NMD (www.treat-nmd.eu) clinical trials coordination centre (CTCC). The aim of the CTCC is to assist investigators and industry partners in the set-up of clinical trials in neuromuscular disorders. The CTCC offers all services necessary to conduct a clinical trial from expert counselling, protocol development, regulatory affairs to complete trial management including trial site selection, data management, monitoring, and statistical evaluation. The best way to scientifically prove drug efficacy in humans is a placebo-controlled trial. If a food supplementation such as tyrosine is used in a clinical trial, it will be considered as a medicinal product with all the regulatory requirements that might vary between different countries. It was agreed that a controlled pilot study with tyrosine should be done before proceeding with a larger multi-centre approach.

Prof. Kathryn North summarised her experience of dietary supplementation with tyrosine in NM [13]. Facial and bulbar weakness in NM causes chewing and swallowing difficulties, recurrent aspiration and poor control of oral secretions. Five patients (four infants and one adolescent) with NM received dietary supplementation with l-tyrosine (250–3000mg/day). All four infants were reported to have an initial decrease in sialorrhoea and increase in energy levels. The adolescent showed improved strength and exercise tolerance. No adverse effects of treatment were observed. Dietary tyrosine supplementation may improve bulbar function, activity levels and exercise tolerance in NM.

7. Participants 

Alan Beggs, Boston, USA

Elizabeth Busch-Nentwich, Cambridge, UK

Nigel Clarke, Sydney, Australia

Ann Curtis, Newcastle upon Tyne, UK

Baziel van Engelen, Nijmegen, NL

Edna Hardeman, Sydney, Austalia

Heinz Jungbluth, London, UK

Janbernd Kirschner, Freiburg, Germany

Nigel Laing, Perth, Australia

Vilma Lehtokari, Helsinki, Finland

Hanns Lochmuller, Newcastle upon Tyne, UK

Steve Marston, London, UK

Francesco Muntoni, London, UK

Joel Lunardi, Grenoble, France

Kathryn North, Sydney, Australia

Kristen Nowak, Perth, Australia

Katarina Pelin, Helsinki, Finland

Norma Romero, Paris, France

Caroline Sewry, Oswestry, UK

John Sparrow, York, UK

Carina Wallgren-Pettersson, Helsinki, Finland